peerj_json_files/PeerJ_Json_101.json

{
    "v1_Abstract": "Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/mutant mice. However, analysis of terminal neuronal differentiation revealed that the proportion of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. 23 24 25 26 27 28 29 30 31 32 33 34 35 36 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t",
    "v1_col_introduction": "introduction  : Perception of sensory inputs from both external and internal environments requires multiple\nlevels of organization in the nervous system. The dorsal spinal cord plays critical roles in\norganizing responses to sensory input, and contains neurons that relay somatosensory information\nfrom sensory neurons in the periphery to motor neurons located in the ventral horns and to higher\nbrain centers (for review: Helms & Johnson 2003). These functions reside in a large number of\ndistinct interneuron types that are arranged in an organized laminar structure in the dorsal horns\n(Rexed 1952; Brown 1981). Five parallel layers (laminae) have been defined in the murine spinal\ncord dorsal horn. These laminae are formed of unique combination of neurons, distinguished by\ntheir morphology and projections and by their gene expression profiles. The laminae receive\ndifferent sensory input, with tactile perception mediated by myelinated axon bundles projecting\nto internal dorsal laminae (III, IV, V), and pain and temperature conveyed through unmyelinated\naxons that project to more superficial laminae (I, II) (for review: Caspary & Anderson 2003).\nProprioception is mediated by sensory neurons that project through the dorsal spinal cord to an\nintermediate zone which in turn projects to the ventral spinal cord where a direct\nconnection is made with motoneurons (Brown, 1981; for review: Caspary &\nAnderson 2003).\nThere are six early-born (in the mouse, by embryonic day E10-12.5) dorsal neuron populations\ncalled dI1-dI6 and two late-born (E11-E13) populations called dILA and dILB, defined by\nexpression of specific homeodomain transcription factors (for review: Helms & Johnson 2003, Lewis\n2006). These neurons can be further classified by their dependance on roof plate signaling for\nformation: class A (dI1-dI3) neurons depend on, whereas class B (dI4-dI6, dILA/B) neurons are\nindependent of roof plate signals (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002; for\nreview: Helms & Johnson 2003). The dorsal interneuron subtypes dI1-3 migrate ventrally, whereas a\nsubset of dI4 and dI5 cells migrate laterally to populate the deep dorsal horn (laminae IV-V). The\ndILA/B subclasses migrate to the superficial dorsal laminae (I-III), and mediate pain and\ntemperature sensitive circuits (for review: Caspary & Anderson 2003).\nThe functional architecture of the mature dorsal horn is the result of developmental processes\nthat involve cell-type specification and differentiation, as well as cell migration. Several events\nthat control the specification of various neuronal subtypes in the spinal cord have been defined in\nrecent studies (for reviews: Lee & Jessell 1999; Briscoe & Ericson, 1999; Caspary & Anderson\n37\n38\n39\n40\n41\n42\n43\n44\n45\n46\n47\n48\n49\n50\n51\n52\n53\n54\n55\n56\n57\n58\n59\n60\n61\n62\n63\n64\n65\n66\n67\nPeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013)\nR ev ie w in g M an\nus cr ip t\n2003; Lewis 2006). These studies demonstrate that homeodomain transcription factors play a\ncentral role during development of neurons in the dorsal horn (for reviews: Goulding et al. 2002;\nHelms & Johnson 2003). Relatively few direct correlations have been made between dorsal\ninterneuron progenitor classes and terminally differentiated cell types. However, formation of the\nproprioceptor pathway, which projects through the dorsal horn to the ventrally located motor\nneurons (Brown 1981; Willis & Coggeshall 1991) was shown to be dependent on Math1\n(Bermingham et al. 2001; Gowan et al. 1991). Also, a dorsal horn-specific transcription factor, Drg11, is\nexpressed in late born cells derived from dl5 precursors and is required for correct afferent fiber projections of nociceptive\nsensory neurons and correct dorsal horn morphogenesis (Chen et al. 2001; Rebelo et al. 2010). Finally, the Lbx1 gene is\nrequired for maturation of several dorsal horn cell types which later populate laminae I-III, and is critical for the correct\nprojection of the nociceptive fibers into these laminae (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002).\nThe gene encoding the homeodomain factor Gbx1 is expressed broadly in the mantle zone\nof the spinal cord at E12.5-13.5 (Rhinn et al. 2004 ; Waters et al. 2003 ; John, Wildner & Britsch 2005). With the appearance of a discernable dorsal horn around E14, Gbx1 expression becomes more restricted, eventually defining a narrow layer in the dorsal horn around perinatal stages (John, Wildner & Britsch 2005). Recently, immunohistological analysis showed that at E12.5, only a subpopulation of the Lbx1-positive cells coexpress Gbx1 (John, Wildner & Britsch 2005). Lbx1 is a key\ndeterminant for the specification of class B neurons (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002), suggesting that Gbx1-positive cells could correspond to class B neuron precursors (John, Wildner & Britsch 2005). Late-born class B neurons comprise initially two neuron populations, dILA and dILB, which are born in an apparent salt and pepper pattern in the dorsal spinal cord. dILA neurons express Lbx1, Pax2, and Lhx1/5, whereas dILB cells express Lbx1, Lmx1b, and Tlx3 (M\u00fcller et al. 2002). At E12.5 and E14.5, Gbx1 neurons co-express the transcription factors Lhx1/5 and Pax2, but are negative for Lmx1b and Tlx3. This indicates that Gbx1 expression distinguishes a subpopulation of dILA neuronal cells (John, Wildner & Britsch 2005). Furthermore, these authors show that GABA or Gad67 expressing neurons coexpress Gbx1, suggesting that Gbx1-positive cells may differentiate into GABAergic neurons.\nTo investigate the function of Gbx1 during dorsal horn development, we have generated mice bearing a mutation that ablates\nGbx1 function. We report that Gbx1 knockout mice are viable and can reproduce as homozygous null mutants.\nHowever, the adult mutant mice display an altered gait during forward movement that specifically affects hindlimbs.\nThis abnormal gait was evaluated by a series of behavioral tests, which revealed that locomotion is impaired, but not muscle strength or motor coordination. We then analyzed the development of the spinal cord dorsal horn in Gbx1-/- mice. Despite the clear behavioral phenotype, we did not observe changes in the expression of homeodomain factors regulating dorsal spinal cord development, suggesting that development of the dorsal horn is not profoundly affected in Gbx1-/- mice. However, analysis of terminal neuronal differentiation revealed that expression of Gad67, a marker for GABAergic inhibitory interneurons, is diminished. Gbx1 is therefore required for the differentiation of inhibitory local circuit interneurons in the superficial dorsal horn, demonstrating a function for this transcription factor in the dorsal horn of the spinal cord.",
    "v2_Abstract": "Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/mutant mice. However, analysis of terminal neuronal differentiation revealed that the number of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. 22 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/mouse mutants",
    "v2_col_introduction": "introduction  : Perception of sensory inputs from both external and internal environments requires multiple\nlevels of organization in the nervous system. The dorsal spinal cord plays critical roles in\norganizing responses to sensory input, and contains neurons that relay somatosensory\ninformation from sensory neurons in the periphery to motor neurons located in the ventral\nhorns and to higher brain centers (for review: Helms & Johnson 2003). These functions reside in a\nlarge number of distinct interneuron types that are arranged in an organized laminar structure\nin the dorsal horns (Rexed 1952; Brown 1981). Five parallel layers (laminae) have been\ndefined in the murine spinal cord dorsal horn. These laminae are formed of unique\ncombination of neurons, distinguished by their morphology and projections and by their gene\nexpression profiles. The laminae receive different sensory input, with tactile perception\nmediated by myelinated axon bundles projecting to internal dorsal laminae (III, IV, V), and\npain and temperature conveyed through unmyelinated axons that project to more superficial\nlaminae (I, II) (for review: Caspary & Anderson 2003). There are six early-born (in the\nmouse, by embryonic day E10-12.5) dorsal neuron populations called dI1-dI6 and two\nlate-born (E11-E13) populations called dILA and dILB, defined by expression of specific\nhomeodomain transcription factors (for review: Helms & Johnson 2003, Lewis 2006). These\nneurons can be further classified by their dependance on roof plate signaling for formation:\nclass A (dI1-dI3) neurons depend on, whereas class B (dI4-dI6, dILA/B) neurons are\nindependent of roof plate signals (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002; for\nreview: Helms & Johnson 2003). The dorsal interneuron subtypes dI1-3 migrate ventrally,\nwhereas a subset of dI4 and dI5 cells migrate laterally to populate the deep dorsal horn\n(laminae IV-V). The dILA/B subclasses migrate to the superficial dorsal laminae (I-III), and\nmediate pain and temperature sensitive circuits (for review: Caspary & Anderson 2003).\nThe functional architecture of the mature dorsal horn is the result of developmental\nprocesses that involve cell-type specification and differentiation, as well as cell migration.\nSeveral events that control the specification of various neuronal subtypes in the spinal cord\nhave been defined in recent studies (for reviews: Lee & Jessell 1999; Briscoe & Ericson,\n1999; Caspary & Anderson 2003; Lewis 2006). These studies demonstrate that homeodomain\ntranscription factors play a central role during development of neurons in the dorsal horn (for\n33\nPeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013)\nR ev ie w in g M an\nus cr ip t\nPhenotypic analysis of Gbx1-/- mouse mutants\nreviews: Goulding et al. 2002; Helms & Johnson 2003). Relatively few direct correlations\nhave been made between dorsal interneuron progenitor classes and terminally differentiated\ncell types. However, formation of the proprioceptor pathway, which projects through the\ndorsal horn to the ventrally located motor neurons (Brown 1981; Willis & Coggeshall 1991)\nwas shown to be dependent on Math1 (Bermingham et al. 2001; Gowan et al. 1991). Also, a\ndorsal horn-specific transcription factor, Drg11, is expressed in late born cells derived from dl5 precursors and is\nrequired for correct afferent fiber projections of nociceptive sensory neurons and correct dorsal horn morphogenesis\n(Chen et al. 2001; Rebelo et al. 2010). Finally, the Lbx1 gene is required for maturation of several dorsal horn cell\ntypes which later populate laminae I-III, and is critical for the correct projection of the nociceptive fibers into these\nlaminae (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002).\nThe gene encoding the homeodomain factor Gbx1 is expressed broadly in the mantle\nzone of the spinal cord at E12.5-13.5 (Rhinn et al. 2004 ; Waters et al. 2003 ; John, Wildner & Britsch\n2005). With the appearance of a discernable dorsal horn around E14, Gbx1 expression becomes more restricted, eventually defining a narrow layer in the dorsal horn around perinatal stages (John, Wildner & Britsch 2005). Recently, immunohistological analysis showed that at E12.5, only a subpopulation of the Lbx1-positive cells coexpress Gbx1 (John, Wildner & Britsch 2005).\nLbx1 is a key determinant for the specification of class B neurons (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002), suggesting that Gbx1-positive cells could correspond to class B neuron precursors (John, Wildner & Britsch 2005). Late-born class B neurons comprise initially two neuron populations, dILA and dILB, which are born in an apparent salt and pepper pattern in the dorsal spinal cord. dILA neurons express Lbx1, Pax2, and Lhx1/5, whereas dILB cells express Lbx1, Lmx1b, and Tlx3 (M\u00fcller et al. 2002). At E12.5 and E14.5, Gbx1 neurons co-express the transcription factors Lhx1/5 and Pax2, but are negative for Lmx1b and Tlx3. This indicates that Gbx1 expression distinguishes a subpopulation of dILA neuronal cells (John, Wildner & Britsch 2005). Furthermore, these authors show that GABA or Gad67 expressing neurons coexpress Gbx1, suggesting that Gbx1-positive cells may differentiate into GABAergic neurons.\nTo investigate the function of Gbx1 during dorsal horn development, we have generated mice bearing a mutation that\nablates Gbx1 function. We report that Gbx1 knockout mice are viable and can reproduce as homozygous null\nmutants. However, the adult mutant mice display an altered gait during forward movement that specifically\naffects hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, which revealed that locomotion is impaired, but not muscle strength or motor coordination. We then analyzed the development of the spinal cord dorsal horn in Gbx1-/- mice. Despite the clear behavioral phenotype, we did not observe changes in the expression of homeodomain factors regulating dorsal spinal cord development, suggesting that development of the dorsal horn is not profoundly affected in Gbx1-/- mice. However, analysis of terminal neuronal differentiation revealed that expression of Gad67, a marker for GABAergic inhibitory interneurons, is diminished. Gbx1 is therefore required for the differentiation of inhibitory local circuit interneurons in the superficial dorsal horn, demonstrating a function for this transcription factor in the dorsal horn of the spinal cord.",
    "v3_Abstract": "Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/mutant mice. However, analysis of terminal neuronal differentiation revealed that the number of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. 22 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/mouse mutants",
    "v3_col_introduction": "introduction  : Perception of sensory inputs from both external and internal environments requires multiple\nlevels of organization in the nervous system. The dorsal spinal cord plays a critical role in\norganizing responses to sensory input, as it contains the neurons that integrate and relay\nsomatosensory information entering the spinal cord from sensory neurons in the periphery to\nthe motor neurons located in the ventral horns and to higher brain centers. These functions\nreside in a large number of distinct interneuron types that are arranged in an organized\nlaminar structure in the dorsal horns (Rexed 1952; Brown 1981). Five discrete parallel layers\n(laminae) have been defined anatomically in the murine spinal cord dorsal horn. Each lamina\nconsists of a unique combination of neurons that can be distinguished by morphology,\nprojections and, in some cases, by gene expression. Specific laminae receive different sensory\ninput: tactile perception is mediated by thick myelinated axon bundles that project to internal\ndorsal laminae (III, IV, V), whereas pain and temperature are conveyed through thinner,\nunmyelinated axon bundles that project to more superficial laminae (I, II) (for review:\nCaspary & Anderson 2003). There are six early-born (embryonic day E10-E12.5)\npostmitotic dorsal neuron populations called dI1-dI6 and two late-born (E11-E13) postmitotic\npopulations called dILA and dILB, defined by expression of specific homeodomain\ntranscription factors (for review: Lewis 2006). These neurons can be further classified by their\ndependance on roof plate signaling for formation: class A (dI1-dI3) neurons depend on,\nwhereas class B (dI4-dI6, dILA/B) neurons are independent of, roof plate signals (Gross,\nDottori & Goulding 2002; M\u00fcller et al. 2002). The dorsal interneuron (dI) subtypes dI1-3\nmigrate ventrally and a subset of dI4 and dI5 cells migrate laterally to populate the deep\ndorsal horn (laminae IV-V). The dILA/B subclasses migrate to the superficial dorsal laminae\n(I-III), and mediate pain and temperature sensitive circuits.\nThe functional architecture of the mature dorsal horn is the result of developmental\nprocesses that involve cell-type specification and differentiation, as well as cell migration.\nSeveral events that control the specification of various neuronal subtypes in the spinal cord\nhave been defined in recent studies (for reviews: Lee & Jessell 1999; Briscoe & Ericson,\n1999; Caspary & Anderson 2003; Lewis 2006). These studies demonstrate that homeodomain\ntranscription factors play a central role during development of neurons in the dorsal horn\n33\nPeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013)\nR ev ie w in g M an\nus cr ip t\nPhenotypic analysis of Gbx1-/- mouse mutants\n(reviews: Goulding et al. 2002; Helms & Johnson 2003). Relatively few direct correlations\nhave been made between dorsal interneuron progenitor classes and terminally differentiated\ncell types. However, formation of the proprioceptor pathway, which projects through the\ndorsal horn to the ventrally located motor neurons (Brown 1981; Willis & Coggeshall 1991)\nwas shown to be dependent on Math1 (Bermingham et al. 2001; Gowan et al. 1991). Also, a\ndorsal horn-specific transcription factor, Drg11, is expressed in late born cells derived from dl5 precursors and is\nrequired for correct afferent fiber projections of nociceptive sensory neurons and correct dorsal horn morphogenesis\n(Chen et al. 2001; Rebelo et al. 2010). Finally, the Lbx1 gene is required for maturation of several dorsal horn cell\ntypes which later populate laminae I-III, and is critical for the correct projection of the nociceptive fibers into these\nlaminae (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002).\nThe gene encoding the homeodomain factor Gbx1 is expressed broadly in the mantle\nzone of the spinal cord at E12.5-13.5 (Rhinn et al. 2004 ; Waters et al. 2004 ; John, Wildner & Britsch\n2005). With the appearance of a discernable dorsal horn around E14, Gbx1 expression becomes more restricted, eventually defining a narrow layer in the dorsal horn around perinatal stages (John, Wildner & Britsch 2005). Recently, immunohistological analysis showed that at E12.5, only a subpopulation of the Lbx1-positive cells coexpress Gbx1 (John, Wildner & Britsch 2005).\nLbx1 is a key determinant for the specification of class B neurons (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002), suggesting that Gbx1-positive cells could correspond to class B neuron precursors (John, Wildner & Britsch 2005). Late-born class B neurons comprise initially two neuron populations, dILA and dILB, which are born in an apparent salt and pepper pattern in the dorsal spinal cord. dILA neurons express Lbx1, Pax2, and Lhx1/5, whereas dILB cells express Lbx1, Lmx1b, and Tlx3 (M\u00fcller et al. 2002). At E12.5 and E14.5, Gbx1 neurons co-express the transcription factors Lhx1/5 and Pax2, but are negative for Lmx1b and Tlx3. This indicates that Gbx1 expression distinguishes a subpopulation of dILA neuronal cells (John, Wildner & Britsch 2005). Furthermore, these authors show that GABA or Gad67 expressing neurons coexpress Gbx1, suggesting that Gbx1-positive cells may differentiate into GABAergic neurons.\nTo investigate the function of Gbx1 during dorsal horn development, we have generated mice bearing a mutation that\nablates Gbx1 function. We report that Gbx1 knockout mice are viable and can reproduce as homozygous null\nmutants. However, the adult mutant mice display an altered gait during forward movement that specifically\naffects hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, which revealed that locomotion is impaired, but not muscle strength or motor coordination. We then analyzed the development of the spinal cord dorsal horn in Gbx1-/- mice. Despite the clear behavioral phenotype, we did not observe changes in the expression of homeodomain factors regulating dorsal spinal cord development, suggesting that development of the dorsal horn is not profoundly affected in Gbx1-/- mice. However, analysis of terminal neuronal differentiation revealed that expression of Gad67, a marker for GABAergic inhibitory interneurons, is diminished. Gbx1 is therefore required for the differentiation of inhibitory local circuit interneurons in the superficial dorsal horn, demonstrating a function for this transcription factor in the dorsal horn of the spinal cord.",
    "v4_Abstract": "Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/mutant mice. However, analysis of terminal neuronal differentiation revealed that the number of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. 22 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/mouse mutants",
    "v4_col_introduction": "introduction  : Perception of sensory inputs from both external and internal environments requires multiple\nlevels of organization in the nervous system. The dorsal spinal cord plays a critical role in\norganizing responses to sensory input, as it contains the neurons that integrate and relay\nsomatosensory information entering the spinal cord from sensory neurons in the periphery to\nthe motor neurons located in the ventral horns and to higher brain centers. These functions\nreside in a large number of distinct interneuron types that are arranged in an organized\nlaminar structure in the dorsal horns (Rexed 1952; Brown 1981). Five discrete parallel layers\n(laminae) have been defined anatomically in the murine spinal cord dorsal horn. Each lamina\nconsists of a unique combination of neurons that can be distinguished by morphology,\nprojections and, in some cases, by gene expression. Specific laminae receive different sensory\ninputs : tactile perception is mediated by thick myelinated axon bundles that project to\ninternal dorsal laminae (III, IV, V), whereas pain and temperature are conveyed through\nthinner, unmyelinated axon bundles that project to more superficial laminae (I, II) (for review:\nCaspary & Anderson 2003). There are six early-born (embryonic day E10-E12.5)\npostmitotic dorsal neuron populations called dI1-dI6 and two late-born (E11-E13) postmitotic\npopulations called dILA and dILB, defined by expression of specific homeodomain\ntranscription factors (for review: Lewis 2006). These neurons can be further classified by their\ndependance on roof plate signaling for formation: class A (dI1-dI3) neurons depend on,\nwhereas class B (dI4-dI6, dILA/B) neurons are independent of, roof plate signals (Gross,\nDottori & Goulding 2002; M\u00fcller et al. 2002). The dorsal interneuron (dI) subtypes dI1-3\nmigrate ventrally and a subset of dI4 and dI5 cells migrate laterally to populate the deep\ndorsal horn (laminae IV-V). The dILA/B subclasses migrate to the superficial dorsal laminae\n(I-III), and mediate pain and temperature sensitive circuits.\nThe functional architecture of the mature dorsal horn is the result of developmental\nprocesses that involve cell-type specification and differentiation, as well as cell migration.\nSeveral events that control the specification of various neuronal subtypes in the spinal cord\nhave been defined in recent studies (for reviews: Lee & Jessell 1999; Briscoe & Ericson,\n1999; Caspary & Anderson 2003; Lewis 2006). These studies demonstrate that homeodomain\ntranscription factors play a central role during development of neurons in the dorsal horn\n33 Pre Prin ts Pre Prin ts\nPhenotypic analysis of Gbx1-/- mouse mutants\n(reviews: Goulding et al. 2002; Helms & Johnson 2003). Relatively few direct correlations\nhave been made between dorsal interneuron progenitor classes and terminally differentiated\ncell types. However, formation of the proprioceptor pathway, which projects through the\ndorsal horn to the ventrally located motor neurons (Brown 1981; Willis & Coggeshall 1991)\nwas shown to be dependent on Math1 (Bermingham et al. 2001; Gowan et al. 1991). Also, a\ndorsal horn-specific transcription factor, Drg11, is expressed in late born cells derived from dl5 precursors and is\nrequired for correct afferent fiber projections of nociceptive sensory neurons and correct dorsal horn morphogenesis\n(Chen et al. 2001; Rebelo et al. 2010). Finally, the Lbx1 gene is required for maturation of several dorsal horn cell\ntypes which later populate laminae I-III, and is critical for the correct projection of the nociceptive fibers into these\nlaminae (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002).\nThe gene encoding the homeodomain factor Gbx1 is expressed broadly in the mantle\nzone of the spinal cord at E12.5-13.5 (Rhinn et al. 2004 ; Waters et al. 2004 ; John, Wildner & Britsch\n2005). With the appearance of a discernable dorsal horn around E14, Gbx1 expression becomes more restricted, eventually defining a narrow layer in the dorsal horn around perinatal stages (John, Wildner & Britsch 2005). Recently, immunohistological analysis showed that at E12.5, only a subpopulation of the Lbx1-positive cells coexpress Gbx1 (John, Wildner & Britsch 2005).\nLbx1 is a key determinant for the specification of class B neurons (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002), suggesting that Gbx1-positive cells could correspond to class B neuron precursors (John, Wildner & Britsch 2005). Late-born class B neurons comprise initially two neuron populations, dILA and dILB, which are born in an apparent salt and pepper pattern in the dorsal spinal cord. dILA neurons express Lbx1, Pax2, and Lhx1/5, whereas dILB cells express Lbx1, Lmx1b, and Tlx3 (M\u00fcller et al. 2002). At E12.5 and E14.5, Gbx1 neurons co-express the transcription factors Lhx1/5 and Pax2, but are negative for Lmx1b and Tlx3. This indicates that Gbx1 expression distinguishes a subpopulation of dILA neuronal cells (John, Wildner & Britsch 2005). Furthermore, these authors show that GABA or Gad67 expressing neurons coexpress Gbx1, suggesting that Gbx1-positive cells may differentiate into GABAergic neurons.\nTo investigate the function of Gbx1 during dorsal horn development, we have generated mice bearing a mutation that\nablates Gbx1 function. We report that Gbx1 knockout mice are viable and can reproduce as homozygous null\nmutants. However, the adult mutant mice display an altered gait during forward movement that specifically\naffects hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, which revealed that locomotion is impaired, but not muscle strength or motor coordination. We then analyzed the development of the spinal cord dorsal horn in Gbx1-/- mice. Despite the clear behavioral phenotype, we did not observe changes in the expression of homeodomain factors regulating dorsal spinal cord development, suggesting that development of the dorsal horn is not profoundly affected in Gbx1-/- mice. However, analysis of terminal neuronal differentiation revealed that expression of Gad67, a marker for GABAergic inhibitory interneurons, is diminished. Gbx1 is therefore required for the differentiation of inhibitory local circuit interneurons in the superficial dorsal horn, demonstrating a function for this transcription factor in the dorsal horn of the spinal cord.",
    "v5_Abstract": "Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/mutant mice. However, analysis of terminal neuronal differentiation revealed that the number of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1 mouse mutants",
    "v5_col_introduction": "introduction  : Perception of sensory inputs from both external and internal environments requires multiple levels of organization in the nervous system. The dorsal spinal cord plays a critical role in organizing responses to sensory input, as it contains the neurons that integrate and relay somatosensory information entering the spinal cord from sensory neurons in the periphery to the motor neurons located in the ventral horns and to higher brain centers. These functions reside in a large number of distinct interneuron types that are arranged in an organized laminar structure in the dorsal horns (Rexed 1952; Brown 1981). Five discrete parallel layers (laminae) have been defined anatomically in the murine spinal cord dorsal horn. Each lamina consists of an unique combination of neurons that can be distinguished by morphology, projections and, in some cases, by gene expression. Specific laminae receive different sensory inputs : tactile perception is mediated by thick myelinated axon bundles that project to internal dorsal laminae (III, IV, V), whereas pain and temperature are conveyed through thinner, unmyelinated axon bundles that project to more superficial laminae (I, II, III). There are six early-born (embryonic day E10-E12.5) postmitotic dorsal neuron populations called dI1-dI6 and two late-born (E11-E13) postmitotic populations called dILA and dILB, defined by expression of specific homeodomain transcription factors (for review: Lewis 2006). These neurons can be further classified by their dependance on roof plate signaling for formation: class A (dI1-dI3) neurons depend on, whereas class B (dI4-dI6, dILA/B) neurons are independent of, roof plate signals (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002). The dorsal interneuron (dI) subtypes dI1-3 migrate ventrally and a subset of dI4 and dI5 cells migrate laterally to populate the deep dorsal horn (laminae IV-V). The dILA/B subclasses migrate to the superficial dorsal laminae (I-III), and mediate pain and temperature sensitive circuits.\nThe functional architecture of the mature dorsal horn is the result of developmental processes that involve cell-type specification and differentiation, as well as cell migration. Several events that control the specification of various neuronal subtypes in the spinal cord have been defined in recent studies (for reviews: Lee & Jessell 1999; Briscoe & Ericson, 1999; Caspary & Anderson 2003; Lewis 2006). These studies demonstrate that homeodomain transcription factors play a central role during development of neurons in the dorsal horn (reviews: Goulding et al. 2002; Helms & Johnson 2003). Relatively few direct correlations\nPre Prin ts Pre Prin ts\n \u00a0 Phenotypic analysis of Gbx1-/- mouse mutants  \u00a0  \u00a0\n4 \u00a0\nhave been made between dorsal interneuron progenitor classes and terminally differentiated cell types. However, formation of the proprioceptor pathway, which projects through the dorsal horn to the ventrally located motor neurons (Brown 1981; Willis & Coggeshall 1991) was shown to be dependent on Math1 (Bermingham et al. 2001; Gowan et al. 1991). Also, a dorsal horn-specific transcription factor, Drg11, is expressed in late born cells derived from dl5 precursors and is required for correct afferent fiber projections of nociceptive sensory neurons and correct dorsal horn morphogenesis (Chen et al. 2001). Finally, the Lbx1 gene is required for maturation of several dorsal horn cell types which later populate laminae I-III, and is critical for the correct projection of the nociceptive fibers into these laminae (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002).\nThe gene encoding the homeodomain factor Gbx1 is expressed broadly in the mantle\nzone of the spinal cord at E12.5-13.5 (Rhinn et al. 2004 ; Waters et al. 2004 ; John, Wildner & Britsch 2005). With the appearance of a discernable dorsal horn around E14, Gbx1 expression becomes more restricted, eventually defining a narrow layer in the dorsal horn around perinatal stages (John, Wildner & Britsch 2005). Recently, immunohistological analysis showed that at E12.5, only a subpopulation of the Lbx1-positive cells coexpress Gbx1 (John, Wildner & Britsch 2005). Lbx1 is a key determinant for the specification of class B neurons (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002), suggesting that Gbx1positive cells could correspond to class B neuron precursors (John, Wildner & Britsch 2005). Late-born class B neurons comprise initially two neuron populations, dILA and dILB, which are born in an apparent salt and pepper pattern in the dorsal spinal cord. dILA neurons express Lbx1, Pax2, and Lhx1/5, whereas dILB cells express Lbx1, Lmx1b, and Tlx3 (M\u00fcller et al. 2002). At E12.5 and E14.5, Gbx1 neurons co-express the transcription factors Lhx1/5 and Pax2, but are negative for Lmx1b and Tlx3. This indicates that Gbx1 expression distinguishes a subpopulation of dILA neuronal cells (John, Wildner & Britsch 2005). Furthermore, these authors show that GABA or Gad67 expressing neurons coexpress Gbx1, suggesting that Gbx1-positive cells may differentiate into GABAergic neurons.\nTo investigate the function of Gbx1 during dorsal horn development, we have generated mice bearing a mutation that ablates Gbx1 function. We report that Gbx1 knockout mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement that specifically affects hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, which revealed that locomotion is impaired, but not muscle strength or motor coordination. We then analyzed the development\nPre Prin ts Pre Prin ts\n \u00a0 Phenotypic analysis of Gbx1-/- mouse mutants  \u00a0  \u00a0\n5 \u00a0\nof the spinal cord dorsal horn in Gbx1-/- mice. Despite the clear behavioral phenotype, we did not observe changes in the expression of homeodomain factors regulating dorsal spinal cord development, suggesting that development of the dorsal horn is not profoundly affected in Gbx1-/- mice. However, analysis of terminal neuronal differentiation revealed that expression of Gad67, a marker for GABAergic inhibitory interneurons, is diminished. Gbx1 is therefore required for the differentiation of inhibitory local circuit interneurons in the superficial dorsal horn, demonstrating a function for this transcription factor in the dorsal horn of the spinal cord (John, Wildner & Britsch 2005).",
    "v1_text": "materials and methods : Construction of a Gbx1 targeting vector Genomic sequences encompassing the mouse Gbx1 gene were isolated from a 129SV genomic phage library, using as a probe a 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Gbx1 cDNA fragment previously characterized (Rhinn et al. 2004). A Gbx1 loss of function mutation was produced by homologous recombination in embryonic stem cells (Ram\u00ecrez-Solis, Davis & Bradley 1993). The targeting vector contained a 5.4 kb XmnI fragment (upstream arm), ending 33 bp upstream of the homeodomain sequence located in Gbx1 second exon, and a 1.6 kb KpnI fragment (downstream arm), whose sequence started 91 bp downstream from the homeodomain. These fragments were excised from the recombinant phage and cloned in the mutagenesis pGN vector (Le Mouellic, Lallemand & Br\u00fblet 1990) to generate the pGNGbx1 targeting vector (Fig. 1A). In this vector, the fragments are inserted on each side of a lacZ reporter gene and a neomycin resistance gene, and their insertion by homologous recombinaton in the Gbx1 gene will generate a 313 bp deletion encompassing the entire homeodomain (Fig. 1A). Transfection of embryonic stem cells and selection of targeted clones HM-1 embryonic stem (ES) cells (Magin, Whir & Melton 1992) were cultured on neomycin-resistant mouse embryonic fibroblasts, as described in Robertson, 1987. Ten \u03bcg of the pGN-Gbx1 targeting vector were linearized by digestion of the unique NotI restriction site, and electroporated into 2\u00d7107 ES cells resuspended in 750 \u03bcl HeBS medium (20 mM Hepes pH 7.05, 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, 6 mM glucose), at 200 V, 960 \u03bcF. Positive selection was carried out for 11 days with 350 \u03bcg/ml G418. Resistant colonies were picked and DNA was extracted from a fraction (1/5) of the cells to perform Southern blot analysis to identify homologous recombination events. The probe used is an external fragment located immediately downstream to the targeting vector (Fig. 1A,B). Positive clones were expanded before freezing. The frequency of homologous recombination was 7 out of 350 clones analyzed. Generation and genotyping of chimeric and mutant mice After thawing, 10 to 15 ES cells were microinjected into blastocysts collected at E3.5 from C57BL/6 females mated with C57BL/6 males (for procedures: Nagy et al. 2003). Injected blastocysts were reimplanted in the uterine horn of pseudopregnant recipient females. Chimeric animals were back-crossed to C57BL/6J mice and germ-line transmission was scored by the presence of agouti coat pigmentation. Heterozygous offspring were identified by PCR genotyping. Tail tips were incubated in lysis buffer (50 mM Tris pH 8.0, 100 mM EDTA, 100 mM NaCl, 1% SDS, 0.6 mg/ml proteinase K) overnight at 55\u00b0C, phenol-chloroform extracted, ethanol precipitated and redissolved in 10 mM Tris-HCl, 1 mM EDTA pH 8.0 at a final concentration of 0.2-1.0 \u03bcg/\u03bcl. The presence of a wild-type or mutated allele was detected using three primers: a sense primer F1: 5\u2019- GGTGACAGCGAGGACAGCTTCCT-3\u2019, an antisense primer R1: 5\u2019-CCCAGAACGACTGCTCACATTGC-3\u2019, and an antisense primer LacZ R2: 5\u2019-GGCCTCTTCGCTATTACGCCA-3\u2019. The presence of a wild-type allele was detected using the F1/R1 primers which amplify a 354 bp fragment. The presence of a mutated allele was detected by using the F1/LacZ R2 primers which amplify a 269 bp fragment. Thirty cycles (denaturation: 1 min, 95\u00b0C, annealing : 1 min, 62\u00b0C; elongation : 30 s, 74\u00b0C) were performed, and the amplified products were separated by 2% agarose gel electrophoresis (Fig. 1C). Phenotypic and molecular analyses were performed after several generations of backcrosses (>5) to C57BL/6J mice, resulting in a nearly pure genetic background. Tissue collection and sample preparation Pregnant females obtained from natural matings (morning of vaginal plug was considered as E0.5) were sacrificed and fetuses were collected in phosphate-buffered saline (NaCl: 8.01 g/L, KCl: 0.2 g/L, Na2HPO4, 2H2O: 1.78 g/L, KH2PO4: 0.27 g/L, pH 7.5; hereafter abbreviated PBS 1x) after cesarean section. The specimens were dissected, fixed overnight in 4% paraformaldehyde (PFA) diluted in PBS 1x, pH 7.5, cryoprotected in 20% sucrose in PBS 1x, pH 7.5 and embedded in Shandon Cryomatrix (Thermo Electron Corperation) before freezing at -80\u00b0C. Cryosections (14 \u00b5m thickness, Leica CM3050S cryostat) sections were made in a coronal plane, collected on Superfrost slides, and stored at -80\u00b0C until use. 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t For whole-mount immunostaining or in situ hybridization, embryos were fixed overnight in 4% PFA, dried at room temperature, and stored at -20 \u00b0C in 100% methanol. results and discussion : Gbx1-deficient mice are viable, but display a typical duck-like gait A loss of function allele for the Gbx1 gene was generated by homologous recombination in murine embryonic stem cells (see Materials and Methods). The mutated Gbx1 allele is devoid of the entire homeodomain-coding sequence and ~100 adjacent nucleotides (Fig. 1A-C). After generation of germ-line transmitting chimeras, heterozygous mutant mice (Gbx1+/-) were found to be viable, fertile and apparently normal. After intercrossing Gbx1+/- mice, Gbx1-/- mutants (generated in a C57BL/6J genetic background) were born in the expected mendelian ratio. Immunohistochemistry performed with an anti-Gbx1 antibody confirmed the absence of detectable Gbx1 protein in the spinal cord of E18.5 Gbx1-/- mutants (Fig. 1D,E). We also checked the expression of Gbx2 from E12.5 to E18.5 (Fig. S1) to exclude a potential compensatory expression due to the loss of function of Gbx1. A subtle increase of Gbx2 mRNA expression might occur in spinal cord cells of Gbx1-/- mice at E12.5-14.5, however this increase was no longer detected at E16.5 or E18.5 (Supplementary material, Fig. S1). This subtle Gbx2 increase could partially compensate for the loss of Gbx1, leading to the absence of phenotypic abnormalities at early developmental stages. Interestingly, when observed by 10 weeks of age, most mutants displayed a typical, unevenness in walking (\"duck-like\") gait (Fig. 2 and Supplementary material: movie). Both male and female Gbx1-/- mice were fertile and had a normal life span. General health and sensorimotor abilities in adult Gbx1 mutants 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Gbx1-/- males and females had a normal body weight (Table 1) and a normal overall physical appearance. However, many of the Gbx1-/- mutants showed significantly abnormal gait (\u03c72 5.20, p<0.05)\u2265 . Indeed, 43% of Gbx1 mutant males and 63% of Gbx1 mutant females displayed lack of fluidity in movement, and limping related to hyper-flexion followed by hyper-extension of one or both hindpaws (Table 1; Fig. 2; Supplementary material: movie). Gbx1-/- males and females also showed significantly reduced short-term locomotor activity following immediate transfer for the modified SHIRPA test, as compared to WT counterparts (t 3.46, p<0.01) (Table 1). \u2265 The other features of general health and basic neurological reflexes were not affected in Gbx1 mutants. When tested for specific motor abilities, motor coordination performance measured in the rotarod test (t 1.29, NS)\u2264 and the muscle strength (grid grip) test (t 1.38, NS)\u2264 were not affected in Gbx1-/- males and females (Table 1). In the beam walking test, the latency to cross the beam was increased (t15=3.71, p<0.01 for females; non significant for males) and the number of slips was slightly increased (even if not significantly), especially in Gbx1-/- females (Fig. 3). In the open field test, there was a significant effect of genotype concerning locomotor activity [F(1,30)=6.51, p<0.05], reflecting reduced locomotion in all Gbx1-/- animals. When considering each gender separately, both Gbx1-/- males and females tended to have reduced locomotor activity over the testing period (although not statistically significant, p=0.09) (Fig. 4). The average speed during motion was also significantly lower in Gbx1-/- males and females than in WT (t 3.36, p<0.01) (Fig. 4). \u2265 The number of entries into, and the percentage of time spent in, the center of the arena also differed between genotypes [F(1,30)\u226514.48, p<0.001]. Both Gbx1-/- males and females had significantly decreased number of entries and spent less time in the center of the open field than WT counterparts (t \u22652.62, p<0.05) (Fig. 4), which might reflect increased anxiety in Gbx1-/- mutants. The reduced exploration of the center might also be due to the observed reduced locomotor activity of Gbx1-/- mutants. Altogether, these data show that Gbx1-/- mutant mice have a clear defect in locomotion, although this defect does not appear to result from a coordination problem or a muscle strength deficiency. To test whether ablation of Gbx1 could affect sensory response, we measured the response of Gbx1 mutant mice in a hot plate test. The withdrawal latency was higher in Gbx1-/- males (but not in females) than in WT (t15=2.10, p=0.05) (Table 1), suggesting reduced thermal pain sensitivity in Gbx1 mutant males. The consequence of Gbx1 inactivation on acoustic startle and pre-pulse inhibition of startle reflex was also evaluated. Regardless of gender, the startle reactivity was comparable between WT and Gbx1-/- mice for all the acoustic stimuli including the startling pulse [Genotype F(1,30)\u22640.83, Sex F(1,30)\u22640.85, Genotype*Sex F(1,30)\u22641.11, NS] (Fig. 5). When the startling pulse was preceded with prepulses with lower intensities, the PPI level was also comparable between genotypes [Genotype F(1,30)=0.55, Sex F(1,30)=0.32, Genotype*Sex F(1,30)=0.11, NS)] (Fig. 5). Furthermore, electromyography (EMG) measurements revealed that the sensory nerve conduction velocity (SNCV) differed significantly between genotypes [F(1,29)=7.31, p<0.05]; indeed, Gbx1-/- females had significantly increased SNCV (t14=2.83, p<0.05) (Table 2), as measured at the level of the caudal nerve. On the other hand, the latency and amplitude of the gastrocnemius muscle response evoked by sciatic nerve stimulation were comparable between genotypes [F(1,29)=1.63, NS]. In summary, we used a variety of behavioral and electrophysiological phenotyping tests to evaluate sensory and motor functions in Gbx1 mutant mice. Decreased exploratory behavior was found in the open field test and following immediate transfer during clinical observations. Exploration of the central part of the open field arena was significantly decreased in Gbx1-/- males and females, which might suggest increased anxiety in these mutants. However, this could also be due to the reduced locomotor activity of the mutants. Indeed, Gbx1-/- mice also showed decreased average speed with no significant effect on the distance travelled in the open field. Their altered gait during forward movement might explain the reduced speed and locomotor activity in the open field, which could not be attributed to defects in motor coordination or muscle strength. Abnormal gait may suggest proprioceptive-like deficits, as indicated by abnormal performance in beam walking, the test used for evaluation of proprioceptive functions, which was statistically significant only for Gbx1-/- females. Although no direct link can be clearly identified between motor deficits and sensory functions, we cannot exclude mutual interdependence of abnormal gait and sensory deficits indicated by reduced responses in hot plate test and increased SNCV, which were penetrant to a different degree in null-mutant males and females. Gbx1-/- mice do not show obvious hindbrain patterning defects Gbx genes are related to the Drosophila unplugged gene, which acts during development of the tracheal system, and for specification of neuroblast sublineages (Chiang et al. 1995; Cui & Doe 1995). There are two Gbx genes in amniote species (human, mouse and chicken), as well as in zebrafish (Lin et al. 1996; Bouillet et al. 1995; Shamim & Mason 1998; Niss & Leutz 1998; Rhinn et al. 2003). Previous studies showed that in mouse, Gbx2 is involved in early specification of the midbrain- hindrain boundary (MHB) organizer, a signaling center that will pattern the anterior hindbrain rhombomeres (Wassarman et al. 1997; Waters & Lewandoski 2006; for review: Rhinn & Brand 2001; Simeone 2000). In zebrafish it was shown 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t that gbx1 acts during early positioning of the MHB, whereas gbx2 functions at later stages, once the MHB is established (Rhinn et al. 2004; 2009; Burroughs-Garcia et al. 2011). In mouse, Gbx1 is not expressed at the MHB as is the case during early zebrafish development. Its expression starts at E7.75 in the prospective hindbrain, spanning rhombomeres 2 to 7 during the segmentation phase (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2003). This suggested that Gbx1 might be involved in early embryonic hindbrain patterning, which could underlie behavioral deficits associated with loss of Gbx1 function. To assess for possible rhombomeric abnormalities in Gbx1-/- mutants, we performed whole-mount in situ hybridizations at E9.5 with several markers, including Hoxb1 and Hoxa2. This analysis did not show any molecular or structural abnormality of the hindbrain rhombomeres in Gbx1-/- embryos (Supplementary material, Fig. S2). This suggests that Gbx1 is not required for early hindbrain patterning, in contrast to its mouse homologue Gbx2 (Wassarman et al. 1997; Waters & Lewandoski 2006). Analysis of hindbrain derivatives (brain stem and cerebellum) at E18.5 using Gad67 as a differentiation marker also did not reveal any difference in Gbx1-/- versus wild-type mice (Supplementary material, Fig. S3). Development of the spinal cord dorsal horn in Gbx1 mutant mice At E12.5, the expression domains of Gbx1 and Gbx2 overlap, both being expressed in the ventricular and mantle zones of the dorsal spinal cord (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2003). As Gbx2 expression is downregulated after E12.5, both genes are only transiently coexpressed in dorsal spinal cord progenitor cells, and Gbx1 is the only Gbx gene persistently expressed during later dorsal horn development (John, Wildner & Britsch 2005). Thus, the prominent expression of Gbx1 in the dorsal horn could be relevant for the abnormal gait phenotype of adult Gbx1 mutant mice, which led us to ask whether Gbx1 is required for the maturation and/or specification of neurons of the dorsal horn during development. Nissl staining of E18.5 spinal cord sections revealed no obvious difference between the dorsal horn of wild-type and Gbx1-/- animals at thoraco-lumbar levels (Fig. 6A,B). Despite the clear behavioral phenotype, we were unable to identify any consistent alteration in the expression of several molecular markers of dorsal spinal cord cell populations in Gbx1-/- embryos. These markers included the genes encoding the transcription factors Lbx1 (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002) (Fig. 6C,D), Lmx1b (Chen et al. 2001) (Fig. 6E,F) and the axon guidance molecule Netrin-1 (Leonardo et al. 1997) (Fig. 6G,H) analyzed at E12.5, 14.5, 16.5 and 18.5, and shown (Fig. 6) at E16.5. Projection pattern of primary sensory afferents in the dorsal horn of Gbx1-/- mutants We examined the development of primary sensory afferent projections to the dorsal horn, which are well defined at E18.5, in Gbx1 mutant mice. The projections of cutaneous nociceptive sensory neurons begin to invade the spinal grey matter by E12.5 (Ozaki & Snider 1997). Immunostaining with an anti-calbindin-28K antibody at E16.5 and E18.5 marks a subset of cutaneous neurons and their afferent fibers (Honda 1995; Chen et al. 2001). By E18.5, calbindin+ fibers have invaded the dorsal horn of wild-type and Gbx1 mutants (Fig. 7A-B'). The Drg11 gene is required for the projection of cutaneous sensory afferent fibers to the dorsal spinal cord (Chen et al. 2001). In Gbx1-/- mutant mice, Drg11 expression was not affected in the dorsal horn at E12.5, 14.5, 16.5 or 18.5 (Fig. 7C,D and data not shown). Altogether, these data suggest that there are no defects in patterning of sensory afferent fiber projections to the dorsal horn, which selectively affects cutaneous afferents, although the markers used cannot rule out other types of patterning differences (for instance from primary afferents that do not label with calbindin). 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t We further examined proprioceptive afferents at E18.5 by using antibodies to peripherin (Escurat et al. 1990). No consistent difference between wild-type and Gbx1-/- mice was observed at the level of proprioceptive fibers that extend toward motoneurons and interneurons in the deep dorsal horn, or at the level of fibers that enter into the spinal gray matter, at E18.5 (Fig. 7E,F) or E16.5 (Fig. 7G,H). During the revision of our manuscript, another Gbx1 mutant allele was described (Buckley et al. 2013). In contrast to our observations and at a comparable stage, those mutants show disorganized peripherin expression, together with a decrease of Islet1-expressing cells in the ventral horn of the lumbar spinal cord (Buckley et al. 2013). This led us to analyze Islet1-expressing cells in ventral motor neurons in our Gbx1 mutants. No differences in the number of Islet1+ cells within the lumbar ventral spinal cord were found at E14.5, E16.5 (Fig. S4) or E18.5 (data not shown). Thus, in contrast to the data of Buckley et al., our analysis does not suggest a defect in the assembly of the proprioceptive sensorimotor circuit. As the same Gbx1 exon (exon 2) was targeted in both loss of function alleles, the reason for the phenotypic discrepancy remains unclear, although it should be mentioned that the mice might have different genetic backgrounds. Reduced GABAergic neuronal differentiation in Gbx1-/- mutants Gbx1 is first expressed in the ventricular zone of the spinal cord at E11.5 (Rhinn et al. 2004 ; Waters, Wilson & Lewandoski 2003). Then at E12.5-E13.5, it is broadly expressed in the mantle zone of the dorsal spinal cord. At E14 with the appearance of a distinguishable dorsal horn, Gbx1 expression becomes more restricted. At E12.5, Gbx1 is coexpressed with Lbx1; thus Gbx1 cells correspond to class B neurons (John, Wildner & Britsch 2005). As described in the introduction, late-born class B neurons comprise initially two populations, dILA and dILB. Because Gbx1 neurons co- express Lhx1/5 and Pax2, but not Lmx1b and Tlx3, it has been suggested that these neurons correspond to the dILA neuronal subtype (John, Wildner & Britsch 2005). It has been shown that dILA neurons undergo GABAergic differentiation (Cheng et al. 2004), and as mature GABA+ neurons they continue expressing Gbx1 (John, Wildner & Britsch 2005). We therefore analyzed GABAergic neurons in the spinal cord of Gbx1-/- mutants, which we identified by expression of glutamic acid decarboxylase GAD67, an enzyme that regulates GABA synthesis. At E18.5, Gad67-expressing cells are found throughout the developing spinal cord of control mice (Somogyi et al. 1995). Importantly, Gad67 expression was reduced in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8A-D), i.e. there was a 16% decrease in the proportion of Gad67-expressing cells (Fig. 8I). This may reflect an abnormal development of GABAergic neurons, which in consequence coud lead to abnormal control of neuronal network in dorsal horn, possibly affecting inhibitory circuits throughout the spinal cord. This finding was strengthened by analysis of Pax2, another gene expressed in GABAergic cells in the spinal cord (Cheng et al. 2004), with cell countings corroborating the decrease in the proportion of GABAergic cells (Fig. 8E,F,I). Furthermore, it is know that during early post-natal period, the GABA pathway switches from excitatory to inhibitory in mouse (Ben-Ari et al. 2007). This shift was shown to occur within the two first weeks of age in hippocampal and spinal motor neurons in mouse (Stein et al. 2004) as well as in lamina I in rats (Sibilla & Ballerini 2009). Also, it was mentioned that the switch depends on the species, sex, brain structures and neuronal types (Ben-Ari et al. 2007) and it was shown using a model system of cultured hippocampal neurons that the switch is triggered by GABAergic activity itself (Ganguly et al. 2001). Interestingly, when Gbx1-/pups were checked visually around weaning every day in the morning (analysis done on 5 litters, 42 pups, 8 Gbx1-/- mutants), the locomotor deficits were first observed around post-natal days (P)16-17. Taking in account an eventual delay due to a diminished GABA activity, the appearance of the locomotion defect could coincide with the time point at which the GABA pathway switches from excitatory to inhibitory. We also observed that Gad67 expression was unchanged in the brain stem and cerebellum of E18.5 Gbx1-/- mutants (Fig. S3), arguing against an involvement of these structures in the observed phenotype. 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t We next addressed the question if the observed decrease of GABAergic cells is due to neuronal cell death or to a possible change of GABAergic to glutamatergic fate. TUNEL experiments were performed at various stages (E12.5, 14.5, 16.5, 18.5; Fig. 9 and data not shown). As expected, natural cell death occurs mainly in the developing spinal ganglia and ventral spinal cord (Fig. 9A,B; White et al. 1998) and natural cell death is suggested to be over by E15.5 (Fig. 9C,D; White et al. 1998). Our analysis showed no abnormal apoptosis in the dorsal spinal cord of Gbx1-/- mice (Fig. 9B,D). This finding would exclude that the decrease of Gad67-expressing cells is due to cell death, and suggest that Gbx1 is not required for cell survival. We then analyzed expression of Slc17a6, encoding VGLUT2, a vesicular glutamate transporter expressed in glutamatergic neurons (Kaneko et al. 2002). At E18.5, Slc17a6-expressing cells were increased in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8G,H). This finding suggests that part of the \"missing\" GABAergic cells may have differentiated into glutamatergic neurons. Glutamate and GABA are the main neurotransmitters for excitatory and inhibitory neurons, respectively, in the vertebrate brain. These neurotransmitters are usually expressed in a mutually exclusive manner (Bellocchio et al. 2000; Fremeau et al. 2001). In the dorsal horn of the spinal cord, most ascending projection neurons and a subset of local circuit interneurons are excitatory and are glutamatergic. These neurons are modulated by local inhibitory neurons, many of which are GABAergic (for reviews: Melzack & Wall 1965; Malcangio & Bowery 1996; Dickenson 2002). Thus, GABA may inhibit transmitter release from primary afferent fibers. The output neurons of the dorsal horn are projection neurons, relaying sensory information to several brain areas. However, the majority of dorsal horn neurons are local circuit interneurons that do not project outside of the spinal cord. The output of projection neurons is influenced by local excitatory and inhibitory neurons (Todd 2010; Larsson & Broman 2011; Guo et al. 2012). In Gbx1 mutants, the reduction in the proportion of GABAergic neurons, and the possible switch of some of these neurons to a glutamatergic identity, may disrupt neuronal circuitry, becoming phenotypically apparent at adult stages as measured by abnormal performance in several behavioral tests. Further electrophysiological studies will be necessary to link the decrease of GABAergic neurons to the abnormal gait observed in adult Gbx1 mutant mice. figure legends : Figure 1. Inactivation of the mouse Gbx1 gene by homologous recombination in embryonic stem (ES) cells. (A) The upper drawing shows the restriction map of the wild-type locus, boxes and lines corresponding to exons and introns, respectively. The homeodomain sequence is in red. In the targeting vector (below), two Gbx1 genomic fragments (between the dashed lines) flank a lacZ reporter gene and the neomycin resistance gene (grey box), transcribed in the same orientation (thin arrow) as Gbx1. In the recombined locus (lower drawing), 313 bp of Gbx1 exon 2 (including the homeodomain) are replaced by the lacZ neo sequence. The location of the 3' probe used for Southern blot analysis of ES cells is indicated in blue, and the PCR primers used to distinguish wild-type and recombined alleles for genotyping of animals (F1, R1, LacZ R2; see Materials and methods) are also indicated. (B) Southern blot analysis of a targeted cell line (+/-) in comparison to wild-type (+/+) HM-1 ES cells, using a probe external to the targeting vector 3' homology arm. (C) Genotyping of wild-type (+/+), heterozygous (+/-) or homozygous mutant (-/-) mice by PCR amplification of fragments specific for the wild-type (354 bp) or mutated allele (269 bp), using the F1, R1 and LacZ R2 primers. (D,E) Anti-Gbx1 immunostaining. At E18.5, Gbx1 protein is absent in the spinal cord of Gbx1-/- mice (E), compared to wild- type (D). Scale bars: 100 \u00b5m. Figure 2. Abnormal phenotype of a 10 week-old Gbx1-/- mouse when walking. Sequential pictures compare the normal gait of a wild-type mouse (A) and the abnormal gait (\"duck-like\" walk) of a Gbx1-/- mutant when walking (B). A movie of these mice is available (Supplementary Movie). Figure 3. Effects of Gbx1 mutation on the latency and number of slips in the beam walking test. ** p<0.01 vs WT; Student t-test. Figure 4. Open field performance of wild-type (WT) and Gbx1-/- mice. The distance traveled over the 20 min period of test reflects locomotor activity. The average speed was calculated during movement in the whole arena for the entire period of testing. Exploration of the central part of the open fied is expressed as the number of entries and percentage of time spent in the center. * p<0.05 and ** p<0.01 vs WT; Student t-test. Figure 5. Startle reactivity and pre-pulse inhibition in wild-type (WT) and Gbx1-/- mice. Startle reactivity to background noise (65 dB), or to 70, 80, 85, 90 dB acoustic stimulation, and startle reflex to a 110 dB stimulus, are presented. The percentage of pre-pulse inhibition of the startle response is displayed as a percentage of the pre-pulse intensity. WN, white noise; P, acoustic pulse intensity; ST, acoustic startle to 110 dB; PP, pre-pulse intensity. Figure 6. Absence of morphological and molecular abnormalities in the developing dorsal horn of Gbx1-/- mice. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/- (B,D,F,H) mice at E18.5 (A,B) and E16.5 (C-H) are shown. All sections are at the lumbar level. (A,B) Nisslstained sections. No differences are detectable between wild-type and mutants (n=3). In situ hybridizations for two transcription factor encoding genes, Lbx1 (C,D) and Lmx1b (E,F), and for the axon guidance molecule netrin-1 (G,H), are shown (n=3 for each marker). No differences are observed between wild-type and mutants. Scale bars: 100 \u00b5m. Figure 7. Developmental progression of afferent projections in the dorsal horn of Gbx1-/- mice. (A-B') Anti-calbindin-D28K antibody staining. At E18.5, calbindin fibers have already entered the spinal gray matter in wild-type (A,A') and Gbx1-/- specimens (B,B'; n=3). Panels A',B' are higher magnifications of the areas boxed in A,B. (C,D) Expression of Drg11 in wild-type at E18.5 (C) and Gbx1 mutant (D) mice. Mutant specimens were indistinguishable from wild-types (n=3). (E-H) Anti-peripherin antibody staining at E18.5 (E,F) and E16.5 (G,H). This staining reveals similar ingrowth of group IA muscle sensory afferents that grow to the ventral spinal cord (arrows) in wild-type (E,G) and mutant (F,H) (n=3 for each stage). Scale bars: 100 \u00b5m (A',B': 50 \u00b5m). 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t cells in WT; 34.85%\u00b11.84 in Gbx1-/- mice; Genotype F(1,4)=223.85, ***p<0.001, Sections F(2,8)=2.22, NS, Genotype*Sections F(2,8)=1.23, NS). Also, the proportion of Pax2+ cells is diminished by 14.7% in Gbx1-/- mice (58.57%\u00b14,03 Pax2+ cells in WT; 42.41%\u00b15.96 in Gbx1-/mice; Genotype F(1,4)=449.36, ***p<0.001, Sections F(2,8)=3.34, NS, Genotype*Sections F(2,8)=6.3, p<0.05). In contrast, countings revealed that the proportion of Slc17a6+ cells is increased by 14.4% in Gbx1-/- mice (50.96%\u00b11.84 Slc17a6+ cells in WT; 65.16%\u00b12.94 in Gbx1-/- mice; Genotype F(1,4)=688.84, ***p<0.001, Sections F(2,8)=1.004, NS, Genotype*Sections F(2,8)=4.73, p<0.05). Scale bars: 100 \u00b5m. Figure 9. Examples of TUNEL labeling of lumbar spinal cord sections of wild-type (A,C) and Gbx1-/- (B,D) mice. Sections are shown at E12.5 (A,B) (n=3) and E18.5 (C,D) (n=3). Some TUNEL-labelled cells are seen in the dorsal root ganglia (drg) and in the ventral spinal cord (magnified in upper insets) at E12.5, in both wild-type and mutant. (E) Apoptotic cells in the interdigital mesenchyme of an E13.5 forelimb are shown as a positive control. Scale bars: 100 \u00b5m. in situ hybridization : In situ hybridization was performed with digoxigenin-labeled probes as previously described (Chotteau-Leli\u00e8vre et al. 2006). Template DNAs were kindly provided by Drs K. Jagla (Lbx1), C. Birchmeier (Lmx1b), M. Tessier-Lavigne (Netrin), A.J. Tobin (Gad67) and P. Gruss (Pax2), P. Bouillet (Gbx2), B. Giros (Slc17a6), F. Chen (Drg11), R. Krumlauf (Hoxb1), and F. Rijli (Hoxa2). For all experiments 3 animals of each genotype, from 2 or more independent litters, were analyzed (except for Gbx2: Fig. S1A-D, n=2). Cell countings were performed in the dorsal horn (Gad67, Pax2, Slc17a6) or ventral horn (Islet1) on 3 transverse sections for each animal, at comparable levels of the lumbar spinal cord (all sections were collected serially, with section planes being separated by 112 \u00b5m). Three animals of each genotype were thus analyzed for each marker. All expression patterns were documented using a macroscope (Leica M420) or microscope (DM4000B, objective 10x), both connected to a Photometrics camera with the CoolSNAP (v. 1.2) imaging software (Roger Scientific, Chicago, IL). Cell counts were performed using the image J (NIH 1.45S) software. Blue labelled cells and unlabelled cells were counted manually with the cell counter plugin. Three sections separated by 112 \u00b5m in 3 independent embryos were counted for each condition, and statistical significance of cell counts was validated with a two way measures analysis of variance (ANOVA) with the first variable as fixed effect (i.e. genotype) and a second variable as random effect of repeat observations on the same individual. The level of significance was set at p<0.05. Graphs represent averages of counting values and SEM. conclusion : We have generated Gbx1-/- loss of function mutant mice, and investigated the development of the spinal cord dorsal horn in these mutants. Gbx1-/- mutants are viable and fertile, but display an altered gait during forward movement that specifically affects hindlimbs, beginning at post-natal days 16-17. This abnormal gait, documented by a series of behavioral tests, is not due to deficits in muscle strength or motor coordination. Although reduced performance of Gbx1-/- mice in beam walking, a test used in studies of proprioception, could potentially suggest proprioceptive deficits, such a hypothesis is not fully supported by at least two observations: (i) the incomplete penetrance of this phenotype because significant deficits were observed only in females, and (ii) by molecular data, which did not reveal deficits in the assembly of proprioceptive sensory afferents in the ventral or intermediate zone, described previously (Brown, 1981) as their target regions before contacting motoneurons. Some of the deficits, such as altered sensory nerve conduction velocity, are significantly altered in females, whereas significant difference in hot plate performance was identified only in males. Although such differences could reflect sexual dimorphism, it is difficult to draw such a conclusion as definitive for two major reasons: (i) in some tests where a significant difference in performance was observed for one gender, the opposite gender may display a similar tendency, which remained non statistically significant; (ii) if for example females would be more prone to effects of Gbx1 mutation we could expect to find them less performant in different tests; however, the gender effects were inconsistent and concerned males or females depending on the measured parameter. The spinal cord dorsal horn largely consists of inhibitory (GABAergic) and excitatory (glutamatergic) neurons that modulate somatosensory inputs from the periphery, including pain, temperature and mechanoception (Glasgow, 2005). Our analysis of major neuronal classes revealed a reduced proportion of GABAergic inhibitory interneurons expressing Gad67 in the superficial dorsal horn of Gbx1-/- mice. Gbx1 may therefore be functionally required for the differentiation of local inhibitory interneurons in the dorsal horn, corroborating a previous report of Gbx1 expression in a specific subset of GABAergic neurons in this region of the spinal cord (John, Wildner & Britsch 2005). Furthermore, 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t our findings suggest that Gbx1 functions as a gene that promotes GABAergic over glutamatergic differentiation in the dorsal horn. A disruption in the balance between inhibitory and excitatory neuronal activity could explain the phenotype observed in Gbx1 mutants. Indeed, the imbalance of inhibitory and excitatory activity may lead to altered signaling to second-order neurons in the intermediate zone, which through an excitatory polysynaptic chain excite motor neurons in ventral horn to initiate protective movements or abnormal proprioceptive behaviors. Such abnormal sensory processing is suggested at least for thermal stimuli, as Gbx1 mutant males displayed increased latency suggesting reduced pain in the hot plate test (thermosensory functions). Finally, considering that locomotor deficits become apparent at P16-17, we cannot exclude that abnormal gait may result from postnatal developmental or neurodegenerative events, which would need to be investigated. Despite the clear behavioral phenotype and reduced pool of GABAergic neurons in the dorsal horn, we did not observe any change in the expression of homeodomain factors involved in dorsal spinal cord patterning, or markers for primary sensory afferents, indicating that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutants. An explanation for these results\u2014and for the overall mild phenotype of the mutants\u2014is that Gbx1 and Gbx2 are coexpressed in dorsal spinal cord cells at early stages of embryogenesis : hence the presence of Gbx2 (and its subtle upregulation observed at E12.5-E14.5 in mutants) might compensate for Gbx1 loss of function with respect to early regulatory events. Generation of Gbx1;Gbx2 double mutants will be required to assess possible redundant functions, and the availability of a Gbx2 floxed (conditional) allele does allow strategies for a spinal cord-specific inactivation, which would alleviate the lethality of the Gbx2 null mutants (Wassarman et al. 1997). Despite the importance of dorsal spinal cord in normal sensory processing, our knowledge concerning the establishment of neuronal circuits remains limited (Graham, Brichta & Callister 2007; Todd 2010). In this regard, our work contributes to understand how transcription factors cooperate for regulating cell specification and eventual distribution of neuronal subtypes in the developing spinal cord, providing clues for further dissecting functional circuitry of the dorsal spinal cord. immunohistochemistry : After antigen unmasking in citrate buffer 0.01 M (pH 6) during 15 min in a microwave oven, sections were treated in H 202 3% in PBS 1x, pH 7.5 for 5 min, rinsed in PBS 1x, then blocked in PBS 1x, pH 7.5 containing 0.25% Triton-X100, 5% normal goat serum and incubated overnight at 4\u00b0C with rabbit antiGbx1 (kindly provided by Dr S. Britsch; 1:500), rabbit anti-calbindin D-28K (Chemicon, 1:1000), rabbit anti-Peripherin (Chemicon, 1:500), or mouse anti-Islet1 (40.2D6, concentrated, Developmental Studies Hybridoma Bank, Iowa City, IA, 1:100) in PBS 1x, pH 7.5 containing 0.25% Triton-X100, 5% normal goat serum followed by species-specific biotin-coupled secondary antibodies (1:400, Jackson Laboratories) diluted in PBS 1x, pH 7.5. Detection was performed using a Vectastain Elite ABC Kit, following the manufacturer's instructions. Nissl staining was performed by incubation in 0.5% cresyl violet in water for 15 min. TUNEL was performed using the APOPTAG\u00ae Peroxidase In Situ Apoptosis detection kit (Millipore). For all experiments 3 animals of each genotype were analyzed. abstract : Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/- mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutant mice. However, analysis of terminal neuronal differentiation revealed that the proportion of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. 23 24 25 26 27 28 29 30 31 32 33 34 35 36 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t acknowledgments : We thank B. Schuhbaur for excellent technical assistance. We are grateful to Dr. K. Niederreither for a critical reading of the manuscript, and to Drs. V. Brault, M. Paschaki and D. Demb\u00e9l\u00e9 for help with statistical analysis. We thank Drs. C. Birchmeier, S. Britsch, F. Chen, K. Jagla, P. Bouillet, B. Giros, P. Gruss, M. Tessier-Lavigne, R. Krumlauf and F. Rijli for the gift of reagents. 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t supplementary figure legends : Figure S1. Expression analysis of Gbx2 in the developing spinal cord of Gbx1 mutants. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/(B,D,F,H) mice are shown. All sections are at the lumbar level. In situ hybridizations for Gbx2 were performed at different developmental stages: E12.5 (A,B; n=2), E14.5 (C,D; n=2), E16.5 (E,F; n=3), and E18.5 (G,H; n=3). Scale bars: 100 \u00b5m. Figure S2. Analysis of rhombomeric markers in Gbx1-/- embryos. Whole-mount in situ hybridizations of E9.5 embryos with 2 markers of prospective rhombomeres: Hoxb1, which labels rhombomere 4 (A,B; n=3), and Hoxa2, which marks rhombomeres 2 to 6 and associated neural crest (C,D; n=3). Scale bars: 50 \u00b5m. Figure S3. In situ hybridization analysis of Gad67-expressing cells in the prenatal hindbrain. Sections are shown at various levels of the brain stem (A-F) and cerebellum (G,H) of wild-type (A,C,E,G; n=3) and Gbx1-/- (B,D,F,H; n=3) mice at E18.5. Scale bars: 100 \u00b5m. Figure S4. Analysis of developing spinal cord motor neurons in Gbx1 mutants. Expression of Islet1 in the lumbar spinal cord of wild-type (A,C) and Gbx1-/(B,D) mice at E14.5 (A,B; n=3) and E16.5 (C,D; n=3). (E) Countings revealed that the numbers of Islet1+ cells in the ventral horn are not significantly diminished in Gbx1-/- mice (at E14.5: 76\u00b17.33 Islet1+ cells in WT; 75.22\u00b13.13 in Gbx1-/- mice; Genotype F(1,4)=0.27, NS, Sections F(2,8)=0.18, NS, Genotype*Sections F(2,8)=0.27, NS; at E16.5: 22.38\u00b15.96 Islet1+ cells in WT; 25.33\u00b16.70 in Gbx1-/- mice; Genotype F(1,4)=3.03, NS, Sections F(2,8)=4.73, p<0.05, Genotype*Sections F(2,8)=5.46, p<0.05). Scale bars: 100 \u00b5m. Movie sequence showing 10 week-old Gbx1+/+ and Gbx1-/- mice. behavioral phenotyping procedures : Cohorts of 10 week-old male and female Gbx1-/- mice in a C57BL/6J genetic background (7 males and 8 females), with their wild-type (WT, 10 males and 9 females) counterparts, were used in this study. Mice were group housed and allowed 2 weeks acclimation in the phenotyping area with controlled temperature (21-22\u00b0C) under a 12-12 h light-dark cycle (lights on 7am-7pm), 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t with food and water available ad libitum. Testing started at 10 weeks of age, and all procedures were carried out in accordance with European institutional guidelines. Behavioral tests were performed successively for each cohort of mice, during the light phase of the circadian cycle, according to a pipeline established by the European Mouse Disease Clinic (EUMODIC pipeline 2), by trained experimenters familiar with observation of normal gait patterns in mice. Detailed procedures for each test are available at the URL: http://www.empress.har.mrc.ac.uk/viewempress/index.php? pipeline=EUMODIC+Pipeline+2. Neurological examination: General health and basic sensory motor functions were evaluated using a modified SHIRPA protocol (Brown, Chambon & Hrab\u00e9 de Angelis 2005; protocol at http://www.empress.har.mrc.ac.uk/viewempress/index.php? pipelineprocedure=EUMODIC+Pipeline+2~Modified+SHIRPA). This analysis is adapted from the procedure developed by Irwin (1968) and from the SHIRPA protocol (Hatcher et al. 2001). It provides an overview of physical appearance, body weight, neurological reflexes and sensory abilities. Rotarod test: This test evaluates motor coordination and balance by measuring the ability of animals to maintain balance on a rotating rod (Bioseb, Chaville, France). Mice were given three testing trials during which the rotation speed accelerated from 4 to 40 rounds per min (rpm) over 5 min. Trials were separated by 5-10 min intervals. The average latency (time to fall from the rotating rod) of the three trials was used as index of motor coordination performance. Grip test: This test measures the maximal muscle strength (g) using an isometric dynamometer connected to a grid (Bioseb). Mice were allowed to grip the grid either with the forepaws or with both the forepaws and hindpaws, then were pulled backwards until they released the grid. Each mouse was submitted to 3 consecutive trials immediately after the modified SHIRPA procedure. The maximal strength developed by the mouse before releasing the grid was recorded and the average value of the three trials was adjusted to body weight. Beam walking: This test is used to evaluate fine motor coordination and proprioceptive function. The apparatus used is a 2 cm diameter and 110 cm long wooden beam, elevated 50 cm above the ground. A goal box (12 x 12 x 14 cm) is attached at one extremity of the beam. Animals were first habituated to the goal box for 1 min. They were then submitted to 3 training trials during which they were placed at different points of the beam, with the head directed to the goal box, and allowed to walk the corresponding distance to enter the goal box. After training, animals were submitted to 3 testing trials during which they were placed at the extremity of the beam opposite to the goal box and allowed to walk the beam distance and enter the goal box. The latency to enter the goal box and the number of slips (when one or both hindpaws slipped laterally from the beam) were measured. Hot plate test: The mice were placed into a glass cylinder on a hot plate (Bioseb) adjusted to 52\u00b0C, and the latency of the first pain reaction of any hindlimb (licking, flinches) was recorded, with a maximum of 30 s testing. Electrophysiological measurements: Electrophysiological recordings were performed under ketamine-xylazine anesthesia (100 and 10 mg/kg body weight, respectively) using a Key Point electromyograph apparatus (Medtronic, France). Disposal scalp needle electrodes were used (ref 9013R0312, Medtronic). The body temperature was maintained at 37\u00b0C with a homeothermic blanket (Harvard, Paris, France). For measuring the sensory nerve conduction velocity (SNCV), recording electrodes were inserted at the proximal part of the tail and stimulating electrodes placed 20 mm from the recording needles towards the extremity of the tail. A ground needle electrode was inserted between the stimulating and recording electrodes. Caudal nerve was stimulated with a series of 20 pulses of 0.2 ms duration at a supramaximal intensity of 8 mA. The average response is included for statistical analysis. The compound muscle action potential (CMAP) was measured in gastrocnemius muscle after sciatic nerve stimulation. For this purpose, stimulating electrodes were placed at the level of the sciatic nerve at 1 cm from the vertebral column, and recording electrodes placed in the gastrocnemius muscle. A ground needle was inserted in the contralateral paw. The sciatic nerve was stimulated with a single 0.2 ms pulse at a supramaximal intensity of 8 mA. The amplitude (mV) and the distal latency of the responses (ms) were measured. 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Anxiety-related behavior - open field test: Mice were tested in automated open fields (Panlab, Barcelona, Spain), each virtually divided into central and peripheral regions. The open fields were placed in a room homogeneously illuminated at 150 Lux. Each mouse was placed in the periphery of the open field and allowed to explore freely the apparatus for 20 min, with the experimenter out of the animal\u2019s sight. The distance traveled, the number of rears, and time spent in the central and peripheral regions were recorded over the test session. The latency and number of crosses into as well as the percent time spent in center area are used as index of emotionality/anxiety. Sensorimotor gating - auditory startle reflex reactivity and pre-pulse inhibition (PPI): Acoustic startle reactivity and pre-pulse inhibition of startle were assessed in a single session using standard startle chambers (SR-Lab Startle Response System, San Diego Instruments). Ten different trial types were used: acoustic startle pulse alone (110 db), eight different prepulse trials in which either 70, 75, 85 or 90 dB stimuli were presented alone or preceding the pulse, and finally one trial (NOSTIM) in which only the background noise (65 dB) was presented to measure the baseline movement in the Plexiglas cylinder. In the startle pulse or prepulse alone trials, the startle reactivity was analyzed, and in the prepulse plus startle trials the amount of PPI was measured and expressed as percentage of the basal startle response. Statistical analyses: Data were analyzed using unpaired Student t-test, one way or repeated measures analysis of variance (ANOVA) with one between factor (genotype) and one within factor (time). Qualitative parameters (i.e. some of the clinical observations) were analyzed using \u03c72 test. The level of significance was set at p<0.05. animal ethics statement : Animal experimentation protocols were reviewed and approved by the Direction D\u00e9partementale des Services V\u00e9t\u00e9rinaires (agreement #67-172 to H.M., 67-189 to P.D., and institutional agreement #D67-218-5 for animal housing) and conformed to the NIH and European Union guidelines, provisions of the Guide for the Care and Use of Laboratory Animals, and the Animal Welfare Act. table captions : Table 1. Effects of Gbx1 mutation on body weight, basic neurological reflexes, specific motor abilities and pain sensitivity. Mice were analyzed at 10 weeks of age. Statistically different parameters in wild-type vs mutants appear in bold. * p<0.05 and **p<0.01 vs wild-type; Student t-test. 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t gastrocnemius : M-wave Latency (ms) 0.93 \u00b1 0.06 0.91 \u00b1 0.06 0.99 \u00b1 0.08 0.84 \u00b1 0.05 Amplitude (mV) 43.99 \u00b1 3.15 44.60 \u00b1 5.87 46.46 \u00b1 6.70 55.41 \u00b1 6.03 2 PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 1 Inactivation of the mouse Gbx1 gene by homologous recombination in embryonic stem (ES) cells. (A) The upper drawing shows the restriction map of the wild-type locus, boxes and lines corresponding to exons and introns, respectively. The homeodomain sequence is in red. In the targeting vector (below), two Gbx1 genomic fragments (between the dashed lines) flank a lacZ reporter gene and the neomycin resistance gene (grey box), transcribed in the same orientation (thin arrow) as Gbx1. In the recombined locus (lower drawing), 313 bp of Gbx1 exon 2 (including the homeodomain) are replaced by the lacZ neo sequence. The location of the 3' probe used for Southern blot analysis of ES cells is indicated in blue, and the PCR primers used to distinguish wild-type and recombined alleles for genotyping of animals (F1, R1, LacZ R2; see Materials and methods) are also indicated. (B) Southern blot analysis of a targeted cell line (+/-) in comparison to wild-type (+/+) HM- 1 ES cells, using a probe external to the targeting vector 3' homology arm. (C) Genotyping of wild- type (+/+), heterozygous (+/-) or homozygous mutant (-/-) mice by PCR amplification of fragments specific for the wild-type (354 bp) or mutated allele (269 bp), using the F1, R1 and LacZ R2 primers. (D,E) Anti-Gbx1 immunostaining. At E18.5, Gbx1 protein is absent in the spinal cord of Gbx1-/- mice (E), compared to wild- type (D). Scale bars: 100 \u03bcm. PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 2 Abnormal phenotype of a Gbx1-/- mouse when walking. Sequential pictures compare the normal gait of a wild-type mouse (A) and the abnormal gait (\"duck- like\" walk) of a Gbx1-/- mutant when walking (B). A movie of these mice is available (Supplementary movie) : PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 3 Effects of Gbx1 mutation on the latency and number of slips in the beam walking test. ** p<0.01 vs WT; Student t-test. PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 4 Open field performance of wild-type (WT) and Gbx1-/- mice. The distance traveled over the 20 min period of test reflects locomotor activity. The average speed was calculated during movement in the whole arena for the entire period of testing. Exploration of the central part of the open fied is expressed as the number of entries and percentage of time spent in the center. * p<0.05 and ** p<0.01 vs WT; Student t-test. PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 5 Startle reactivity and pre-pulse inhibition in wild-type (WT) and Gbx1-/- mice. Startle reactivity to background noise (65 dB), or to 70, 80, 85, 90 dB acoustic stimulation, and startle reflex to a 110 dB stimulus, are presented. The percentage of pre-pulse inhibition of the startle response is displayed as a percentage of the pre-pulse intensity. WN, white noise; P, acoustic pulse intensity; ST, acoustic startle to 110 dB; PP, pre-pulse intensity. PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 6 Absence of morphological and molecular abnormalities in the developing dorsal horn of Gbx1-/mice. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/- (B,D,F,H) mice at E18.5 (A,B) and E16.5 (C-H) are shown. All sections are at the lumbar level. (A,B) Nissl-stained sections. No differences are detectable between wild-type and mutants (n=3). In situ hybridizations for two transcription factor encoding genes, Lbx1 (C,D) and Lmx1b (E,F), and for the axon guidance molecule netrin-1 (G,H), are shown (n=3 for each marker). No differences are observed between wild-type and mutants. Scale bars: 100 \u03bcm. PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 7 Developmental progression of afferent projections in the dorsal horn of Gbx1-/- mice. (A-B') Anti-calbindin-D28K antibody staining. At E18.5, calbindin fibers have already entered the spinal gray matter in wild-type (A,A') and Gbx1-/- specimens (B,B'; n=3). Panels A',B' are higher magnifications of the areas boxed in A,B. (C,D) Expression of Drg11 in wild-type at E18.5 (C) and Gbx1 mutant (D) mice. Mutant specimens were indistinguishable from wild-types (n=3). (E-H) Anti- peripherin antibody staining at E18.5 (E,F) and E16.5 (G,H). This staining reveals similar ingrowth of group IA muscle sensory afferents that grow to the ventral spinal cord (arrows) in wild-type (E,G) and mutant (F,H) (n=3 for each stage). Scale bars: 100 \u03bcm (A',B': 50 mm). PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 8 Abnormal GABAergic differentiation in Gbx1-/- mice. Expression of Gad67 in wild-type (A,C) and Gbx1-/- (B,D) mice at E18.5 (n=3). Higher magnification views (C,D; areas boxed in A,B) show the dorsal horn, in which cell countings were performed. Expression of Pax2 in wild-type (E) and Gbx1-/- (F) mice at E18.5 (n=3). Expression of Slc17a6 in wild-type (G) and Gbx1-/- (H) mice at E18.5 (n=3). (I) Countings (percentages of labelled vs total cells) revealed that the proportion of Gad67+ cells is diminished by 16% in Gbx1-/- mice (50.89% \u00b12.61 Gad67+ cells in WT; 34.85%\u00b11.84 in Gbx1-/- mice; Genotype F(1,4)=223.85, ***p<0.001, Sections F(2,8)=2.22, NS, Genotype*Sections F(2,8)=1.23, NS). Also, the proportion of Pax2+ cells is diminished by 14.7% in Gbx1-/- mice (58.57%\u00b14,03 Pax2+ cells in WT; 42.41%\u00b15.96 in Gbx1-/- mice; Genotype F(1,4)=449.36, ***p<0.001, Sections F(2,8)=3.34, NS, Genotype*Sections F(2,8)=6.3, p<0.05). In contrast, countings revealed that the proportion of Slc17a6+ cells is increased by 14.4% in Gbx1-/- mice (50.96%\u00b11.84 Slc17a6+ cells in WT; 65.16%\u00b12.94 in Gbx1-/- mice; Genotype F(1,4)=688.84, ***p<0.001, Sections F(2,8)=1.004, NS, Genotype*Sections F(2,8)=4.73, p<0.05). Scale bars: 100 \u03bcm. PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t Figure 9 Examples of TUNEL labeling of lumbar spinal cord sections of wild-type (A,C) and Gbx1-/(B,D) mice. Sections are shown at E12.5 (A,B) (n=3) and E18.5 (C,D) (n=3). Some TUNEL-labelled cells are seen in the dorsal root ganglia (drg) and in the ventral spinal cord (magnified in upper insets) at E12.5, in both wild-type and mutant. (E) Apoptotic cells in the interdigital mesenchyme of an E13.5 forelimb are shown as a positive control. Scale bars: 100 \u03bcm. PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t PeerJ reviewing PDF | (v2012:11:45:4:2:ACCEPTED 13 Aug 2013) R ev ie w in g M an us cr ip t",
    "v2_text": "abstract : Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/- mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutant mice. However, analysis of terminal neuronal differentiation revealed that the number of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. 22 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants materials and methods : Construction of a Gbx1 targeting vector Genomic sequences encompassing the mouse Gbx1 gene were isolated from a 129SV genomic phage library, using as a probe a Gbx1 cDNA fragment previously characterized (Rhinn et al. 2004). A Gbx1 loss of function mutation was produced by 44 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants homologous recombination in embryonic stem cells (Ram\u00ecrez-Solis, Davis & Bradley 1993). The targeting vector contained a 5.4 kb XmnI fragment (upstream arm), ending 33 bp upstream of the homeodomain sequence located in Gbx1 second exon, and a 1.6 kb KpnI fragment (downstream arm), whose sequence started 91 bp downstream from the homeodomain. These fragments were excised from the recombinant phage and cloned in the mutagenesis pGN vector (Le Mouellic, Lallemand & Br\u00fblet 1990) to generate the pGN-Gbx1 targeting vector (Fig. 1A). In this vector, the fragments are inserted on each side of a lacZ reporter gene and a neomycin resistance gene, and their insertion by homologous recombinaton in the Gbx1 gene will generate a 313 bp deletion encompassing the entire homeodomain (Fig. 1A). Transfection of embryonic stem cells and selection of targeted clones HM-1 embryonic stem (ES) cells (Magin, Whir & Melton 1992) were cultured on neomycin-resistant mouse embryonic fibroblasts, as described in Robertson, 1987. Ten \u03bcg of the pGN-Gbx1 targeting vector were linearized by digestion of the unique NotI restriction site, and electroporated into 2\u00d7107 ES cells resuspended in 750 \u03bcl HeBS medium (20 mM Hepes pH 7.05, 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, 6 mM glucose), at 200 V, 960 \u03bcF. Positive selection was carried out for 11 days with 350 \u03bcg/ml G418. Resistant colonies were picked and DNA was extracted from a fraction (1/5) of the cells to perform Southern blot analysis to identify homologous recombination events. The probe used is an external fragment located immediately downstream to the targeting vector (Fig. 1A,B). Positive clones were expanded before freezing. The frequency of homologous recombination was 7 out of 350 clones analyzed. Generation and genotyping of chimeric and mutant mice After thawing, 10 to 15 ES cells were microinjected into blastocysts collected at E3.5 from C57BL/6 females mated with C57BL/6 males (for procedures: Nagy et al. 2003). Injected blastocysts were reimplanted in the uterine horn of pseudopregnant recipient females. Chimeric animals were back-crossed to C57BL/6J mice and germ-line transmission was scored by the presence of agouti coat pigmentation. Heterozygous offspring were identified by PCR genotyping. Tail tips were incubated in lysis buffer (50 mM Tris pH 8.0, 100 mM EDTA, 100 mM NaCl, 1% SDS, 0.6 mg/ml proteinase K) overnight at 55\u00b0C, phenol-chloroform extracted, ethanol precipitated and redissolved in 10 mM Tris-HCl, 1 mM EDTA pH 8.0 at a final concentration of 0.2-1.0 \u03bcg/\u03bcl. The presence of a wild-type or mutated allele was detected using three primers: a sense primer F1: 5\u2019-GGTGACAGCGAGGACAGCTTCCT-3\u2019, an antisense primer R1: 5\u2019-CCCAGAACGACTGCTCACATTGC-3\u2019, and an antisense primer LacZ R2: 5\u2019-GGCCTCTTCGCTATTACGCCA-3\u2019 The presence of a wild-type allele was detected using the F1/R1 primers which amplify a 354 bp fragment. The presence of a mutated allele was detected by using the F1/LacZ R2 primers which amplify a 269 bp fragment. Thirty cycles (denaturation : 1 min, 95\u00b0C, annealing : 1 min, 62\u00b0C; elongation : 30 s, 74\u00b0C) were performed, and the amplified products were separated by 2% agarose gel electrophoresis (Fig. 1C). Phenotypic and molecular analyses were performed after several generations of backcrosses (>5) to C57BL/6J mice, resulting in a nearly pure genetic background. Tissue collection and sample preparation Pregnant females obtained from natural matings (morning of vaginal plug was considered as E0.5) were sacrificed and fetuses were collected in phosphate-buffered saline (PBS 1x, pH 7.5) after cesarean section. The specimens were dissected, fixed overnight in 4% paraformaldehyde (PFA) diluted in PBS 1x, cryoprotected in 20% sucrose in PBS 1x, and embedded in Shandon Cryomatrix (Thermo Electron Corperation) before freezing at -80\u00b0C. Cryosections (14 \u00b5m thickness, Leica CM3050S cryostat) sections were made in a coronal plane, collected on Superfrost slides, and stored at -80\u00b0C until use. For whole-mount immunostaining or in situ hybridization, embryos were fixed overnight in 4% PFA, dried at room temperature, and stored 55 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants at -20 \u00b0C in 100% methanol. results and discussion : Gbx1-deficient mice are viable, but display a typical duck-like gait A loss of function allele for the Gbx1 gene was generated by homologous recombination in murine embryonic stem cells (see Materials and Methods). The mutated Gbx1 allele is devoid of the entire homeodomain-coding sequence and ~100 adjacent nucleotides (Fig. 1A-C). After generation of germ-line transmitting chimeras, heterozygous mutant mice (Gbx1+/-) were found to be viable, fertile and apparently normal. After intercrossing Gbx1+/- mice, Gbx1-/- mutants (generated in a C57BL/6J genetic background) were born in the expected mendelian ratio. Immunohistochemistry performed with an anti-Gbx1 antibody confirmed the absence of detectable Gbx1 protein in the spinal cord of E18.5 Gbx1-/- mutants (Fig. 1D,E). We also checked the expression of Gbx2 from E12.5 to E18.5 (Fig. S1) to exclude a potential compensatory expression due to the loss of function of Gbx1. A subtle increase of Gbx2 mRNA expression might occur in spinal cord cells of Gbx1-/- mice at E12.5-14.5, however this increase was no longer detected at E16.5 or E18.5 (Supplementary material, Fig. S1). This subtle Gbx2 increase could partially compensate for the loss of Gbx1, leading to the absence of phenotypic abnormalities at early developmental stages. Interestingly, when observed by 4-6 weeks of age, most mutants displayed a typical, unevenness in walking (\"duck-like\") gait (Fig. 2 and Supplementary material: movie). Both male and female Gbx1-/- mice were fertile and had a normal life span. General health and sensorimotor abilities in adult Gbx1 mutants Gbx1-/- males and females had a normal body weight (Table 1) and a normal overall physical appearance. However, many of the Gbx1-/mutants showed significantly abnormal gait (\u03c72 5.20, p<0.05)\u2265 . Indeed, 43% of Gbx1 mutant males and 63% of Gbx1 mutant females displayed lack of fluidity in movement, and limping related to hyper-flexion followed by hyper-extension of one or both hindpaws (Table 1; Fig. 2; Supplementary material: movie). Gbx1-/- males and females also showed significantly reduced short-term locomotor activity following immediate transfer for the modified SHIRPA test, as compared to WT counterparts (t 3.46, p<0.01) (Table 1). \u2265 The other features of general health and basic neurological reflexes were not affected in Gbx1 mutants. 88 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants When tested for specific motor abilities, motor coordination performance measured in the rotarod test (t 1.29, NS)\u2264 and the muscle strength (grid grip) test (t 1.38, NS)\u2264 were not affected in Gbx1-/- males and females (Table 1). In the beam walking test, the latency to cross the beam was increased (t15=3.71, p<0.01 for females; non significant for males) and the number of slips was slightly increased (even if not significantly), especially in Gbx1-/- females (Fig. 3). In the open field test, there was a significant effect of genotype concerning locomotor activity [F(1,30)=6.51, p<0.05], reflecting reduced locomotion in all Gbx1-/- animals. When considering each gender separately, both Gbx1-/- males and females tended to have reduced locomotor activity over the testing period (although not statistically significant, p=0.09) (Fig. 4). The average speed during motion was also significantly lower in Gbx1-/- males and females than in WT (t 3.36, p<0.01) (Fig. 4).\u2265 The number of entries into, and the percentage of time spent in, the center of the arena also differed between genotypes [F(1,30)\u226514.48, p<0.001]. Both Gbx1-/- males and females had significantly decreased number of entries and spent less time in the center of the open field than WT counterparts (t\u22652.62, p<0.05) (Fig. 4), which might reflect increased anxiety in Gbx1-/- mutants. The reduced exploration of the center might also be due to the observed reduced locomotor activity of Gbx1-/- mutants. Altogether, these data show that Gbx1-/- mutant mice have a clear defect in locomotion, although this defect does not appear to result in a coordination problem or a muscle strength deficiency. To test whether ablation of Gbx1 could affect sensory response, we measured the response of Gbx1 mutant mice in a hot plate test. The withdrawal latency was higher in Gbx1-/- males (but not in females) than in WT (t15=2.10, p=0.05) (Table 1), suggesting reduced thermal pain sensitivity in Gbx1 mutant males. The consequence of Gbx1 inactivation on acoustic startle and pre-pulse inhibition of startle reflex was also evaluated. Regardless of gender, the startle reactivity was comparable between WT and Gbx1-/- mice for all the acoustic stimuli including the startling pulse [Genotype F(1,30)\u22640.83, Sex F(1,30)\u22640.85, Genotype*Sex F(1,30)\u22641.11, NS] (Fig. 5). When the startling pulse was preceded with prepulses with lower intensities, the PPI level was also comparable between genotypes [Genotype F(1,30)=0.55, Sex F(1,30)=0.32, Genotype*Sex F(1,30)=0.11, NS)] (Fig. 5). Furthermore, electromyography (EMG) measurements revealed that the sensory nerve conduction velocity differed significantly between genotypes [F(1,29)=7.31, p<0.05]; indeed, Gbx1-/- females had significantly increased sensory nerve conduction velocity (t14=2.83, p<0.05) (Table 2), as measured at the level of the caudal nerve. On the other hand, the latency and amplitude of the gastrocnemius muscle response evoked by sciatic nerve stimulation were comparable between genotypes [F(1,29)=1.63, NS]. In summary, we used a variety of behavioral and electrophysiological phenotyping tests to evaluate sensory and motor functions in Gbx1 mutant mice. Decreased exploratory behavior was found in the open field test and following immediate transfer during clinical observations. Exploration of the central part of the open field arena was significantly decreased in Gbx1-/- males and females, which might suggest increased anxiety in these mutants. However, this could also be due to the reduced locomotor activity of the mutants. Indeed, Gbx1-/mice also showed decreased average speed with no significant effect on the distance travelled in the open field. Their altered gait during forward movement might explain the reduced speed and locomotor activity in the open field, which could not be attributed to defects in motor coordination or muscle strength. Importantly, abnormal gait resemble proprioceptive-like deficits, which was further suggested by deficits in beam-walking, the test used for evaluation of proprioceptive functions. Although no direct link can be clearly identified between motor deficits and sensory functions, we cannot exclude mutual interdependence of abnormal gait and sensory deficits indicated by reduced responses in hot plate test and reduced velocity of sensory nerve conduction, which were penetrant to a different degree in null-mutant males and females. Gbx1-/- mice do not show obvious hindbrain patterning defects Gbx genes are related to the Drosophila unplugged gene, which acts during development of the tracheal system, and for specification of neuroblast sublineages (Chiang et al. 1995; Cui & Doe 1995). There are two Gbx genes in amniote species (human, mouse and chicken), as well as in zebrafish (Lin et al. 1996; Bouillet et al. 1995; Shamim & Mason 1998; Niss & Leutz 1998; Rhinn et al. 2003). Previous studies showed that in mouse, Gbx2 is involved in early specification of the midbrain-hindrain boundary (MHB) organizer, a signaling center that will pattern the anterior hindbrain rhombomeres (Wassarman et al. 1997; Waters & Lewandoski 2006; for review: Rhinn & Brand 2001; Simeone 2000). In zebrafish it was shown that gbx1 acts during early positioning of the MHB, whereas gbx2 functions at later stages, once the MHB is established (Rhinn et al. 2004; 2009; Burroughs-Garcia et al. 2011). In mouse, Gbx1 is not expressed at the MHB as is the case during early zebrafish development. Its expression starts at E7.75 in the prospective hindbrain, spanning rhombomeres 2 to 7 during the segmentation phase (Rhinn et al. 2004; Waters, Wilson & Lewandoski 99 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants 2003). This suggested that Gbx1 might be involved in early embryonic hindbrain patterning, which could underlie behavioral deficits associated with loss of Gbx1 function. To assess for possible rhombomeric abnormalities in Gbx1-/- mutants, we performed whole-mount in situ hybridizations at E9.5 with several markers, including Hoxb1 and Hoxa2. This analysis did not show any molecular or structural abnormality of the hindbrain rhombomeres in Gbx1-/- embryos (Supplementary material, Fig. S2). This suggests that Gbx1 is not required for early hindbrain patterning, in contrast to its mouse homologue Gbx2 (Wassarman et al. 1997; Waters & Lewandoski 2006). Analysis of hindbrain derivatives (brain stem and cerebellum) at E18.5 using Gad67 as a differentiation marker also did not reveal any difference in Gbx1-/- versus wild-type mice (Supplementary material, Fig. S3). At E12.5, the expression domains of Gbx1 and Gbx2 overlap, both being expressed in the ventricular and mantle zones of the dorsal spinal cord (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2003). As Gbx2 expression is downregulated fter E12.5, both genes are only transiently coexpressed in dorsal spinal cord progenitor cells, and Gbx1 is the only Gbx gene persistently expressed during later dorsal horn development (John, Wildner & Britsch 2005). Development of the spinal cord dorsal horn in Gbx1 mutant mice The prominent expression of Gbx1 in the dorsal horn could be relevant for the abnormal gait phenotype of adult Gbx1 mutant mice, which led us to ask whether Gbx1 is required for the maturation and/or specification of neurons of the dorsal horn during development. Nissl staining of E18.5 spinal cord sections revealed no obvious difference between the dorsal horn of wild-type and Gbx1-/- animals at thoraco-lumbar levels (Fig. 6A,B). Despite the clear behavioral phenotype, we were unable to identify any consistent alteration in the expression of several molecular markers of dorsal spinal cord cell populations in Gbx1-/- embryos. These markers included the genes encoding the transcription factors Lbx1 (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002) (Fig. 6C,D), Lmx1b (Chen et al. 2001) (Fig. 6E,F) and the axon guidance molecule Netrin-1 (Leonardo et al. 1997) (Fig. 6G,H) analyzed at E12.5, 14.5, 16.5 and 18.5, and shown (Fig. 6) at E16.5. Projection pattern of primary sensory afferents in the dorsal horn of Gbx1-/- mutants We examined the development of primary sensory afferent projections to the dorsal horn, which are well defined at E18.5, in Gbx1 mutant mice. The projections of cutaneous nociceptive sensory neurons begin to invade the spinal grey matter by E12.5 (Ozaki & Snider 1997). Immunostaining with an anti-calbindin-28K antibody at E16.5 and E18.5 marks a subset of cutaneous neurons and their afferent fibers (Honda 1995; Chen et al. 2001). By E18.5, calbindin+ fibers have invaded the dorsal horn of wild-type and Gbx1 mutants (Fig. 7A-B'). The Drg11 gene is required for the projection of cutaneous sensory afferent fibers to the dorsal spinal cord (Chen et al. 2001). In Gbx1-/- mutant mice, Drg11 expression was not affected in the dorsal horn at E12.5, 14.5, 16.5 or 18.5 (Fig. 7C,D and data not shown). Altogether, these data suggest that there are no defects in patterning of sensory afferent fiber projections to the dorsal horn, which selectively affects cutaneous afferents, although the markers used cannot rule out other types of patterning differences (for instance from primary afferents that do not label with calbindin). We further examined proprioceptive afferents at E18.5 by using antibodies to peripherin (Escurat et al. 1990). No consistent difference between wild-type and Gbx1-/- mice was observed at the level of proprioceptive fibers that extend toward motoneurons and interneurons in the deep dorsal horn, or at the level of fibers that enter into the spinal gray matter, at E18.5 (Fig. 7E,F) or E16.5 (Fig. 7G,H). During the revision of our manuscript, another Gbx1 mutant allele was described (Buckley et al. 2013). In contrast to our observations, those mutants show disorganized peripherin expression, together with a decrease of Islet1-expressing cells in the ventral horn of the lumbar spinal 1010 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants cord (Buckley et al. 2013). This led us to analyze Islet1-expressing cells in ventral motor neurons in our Gbx1 mutants. No differences in the number of Islet1+ cells within the lumbar ventral spinal cord were found at E14.5, E16.5 (Fig. S4) or E18.5 (data not shown). Thus, in contrast to the data of Buckley et al., our analysis does not suggest a defect in the assembly of the proprioceptive sensorimotor circuit. As the same Gbx1 exon (exon 2) was targeted in both loss of function alleles, the reason for the phenotypic discrepancy remains unclear, although it should be mentioned that the mice might have different genetic backgrounds. Reduced GABAergic neuronal differentiation in Gbx1-/- mutants Gbx1 is first expressed in the ventricular zone of the spinal cord at E11.5 (Rhinn et al. 2004 ; Waters, Wilson & Lewandoski 2003). Then at E12.5-E13.5, it is broadly expressed in the mantle zone of the dorsal spinal cord. At E14 with the appearance of a distinguishable dorsal horn, Gbx1 expression becomes more restricted. At E12.5, Gbx1 is coexpressed with Lbx1; thus Gbx1 cells correspond to class B neurons (John, Wildner & Britsch 2005). As described in the introduction, late-born class B neurons comprise initially two populations, dILA and dILB. Because Gbx1 neurons co-express Lhx1/5 and Pax2, but not Lmx1b and Tlx3, it has been suggested that these neurons correspond to the dILA neuronal subtype (John, Wildner & Britsch 2005). It has been shown that dILA neurons undergo GABAergic differentiation (Cheng et al. 2004), and as mature GABA+ neurons they continue expressing Gbx1 (John, Wildner & Britsch 2005). We therefore analyzed GABAergic neurons in the spinal cord of Gbx1-/- mutants, which we identified by expression of glutamic acid decarboxylase GAD67, an enzyme that regulates GABA synthesis. At E18.5, Gad67-expressing cells are found throughout the developing spinal cord of control mice (Somogyi et al. 1995). Importantly, Gad67 expression was reduced in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8A-D), i.e. there was a 16% decrease in the proportion of Gad67-expressing cells (Fig. 8I). This may reflect an abnormal development or survival of GABAergic neurons, which in consequence coud lead to abnormal control of neuronal network in dorsal horn, possibly affecting inhibitory circuits throughout the spinal cord. This finding was strengthened by analysis of Pax2, another gene expressed in GABAergic cells in the spinal cord (Cheng et al. 2004), with cell countings corroborating the decrease in the proportion of GABAergic cells (Fig. 8E,F,I). Interestingly, when Gbx1-/- pups were checked visually around weaning every day in the morning (analysis done on 5 litters, 42 pups, 8 Gbx1-/- mutants), the locomotor deficits were first observed around post-natal days (P)16-17, which may coincide with the time point at which the GABA pathway switches from excitatory to inhibitory in mouse (Ben-Ari et al. 2007). We also observed that Gad67 expression was unchanged in the brain stem and cerebellum of E18.5 Gbx1-/- mutants (Fig. S3), arguing against an involvement of these structures in the observed phenotype. We next addressed the question if the observed decrease of GABAergic cells is due to neuronal cell death or to a possible change of GABAergic to glutamatergic fate. TUNEL experiments were performed at various stages (E12.5, 14.5, 16.5, 18.5) (Fig. 9, and data not shown). As expected, natural cell death occurs in the developing spinal ganglia (Fig. 9A, arrow; Paschaki et al. 2012) but not in the spinal cord (Fig. 9A,C). Our analysis showed no abnormal apoptosis in the spinal cord of Gbx1-/- mice (Fig. 9A-D). This finding would exclude that the decrease of Gad67-expressing cells is due to cell death, and suggest that Gbx1 is not required for cell survival. We then analyzed expression of Slc17a6, encoding VGLUT2, a vesicular glutamate transporter expressed in glutamatergic neurons (Kaneko et al. 2002). At E18.5, Slc17a6-expressing cells were increased in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8G,H). This finding suggests that part of the \"missing\" GABAergic cells may have differentiated into glutamatergic neurons. Glutamate and GABA are the main neurotransmitters for excitatory and inhibitory neurons, respectively, in the vertebrate brain. These neurotransmitters are usually expressed in a mutually exclusive manner (Bellocchio et al. 2000; Fremeau et al. 2001). In the dorsal horn of the spinal cord, most ascending projection neurons and a subset of local circuit interneurons are excitatory and are glutamatergic. These neurons are modulated by local inhibitory neurons, many of which are GABAergic (for reviews: Melzack & Wall 1965; Malcangio & Bowery 1996; Dickenson 2002). Thus, GABA may inhibit transmitter release from 1111 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants primary afferent fibers. The output neurons of the dorsal horn are projection neurons, relaying sensory information to several brain areas. However, the majority of dorsal horn neurons are local circuit interneurons that do not project outside of the spinal cord. The output of projection neurons is influenced by local excitatory and inhibitory neurons (Todd 2010; Larsson & Broman 2011; Guo et al. 2012). In Gbx1 mutants, the reduction of GABAergic neurons, and the possible switch of some of these neurons to a glutamatergic identity, may disrupt neuronal circuitry, becoming phenotypically apparent at adult stages as measured by abnormal performance in several behavioral tests. Further electrophysiological studies will be necessary to link the decrease of GABAergic neurons to the abnormal gait observed in adult Gbx1 mutant mice. figure legends : Figure 1. Inactivation of the mouse Gbx1 gene by homologous recombination. (A) The upper drawing shows the restriction map of the wild-type locus, boxes and lines corresponding to exons and introns, respectively. The homeodomain sequence is in red. In the targeting vector (below), two Gbx1 genomic fragments (between the dashed lines) flank a lacZ reporter gene and the neomycin resistance gene (grey box), transcribed in the same orientation (thin arrow) as Gbx1. In the recombined locus (lower drawing), 313 bp of Gbx1 exon 2 (including the homeodomain) are replaced by the lacZ neo sequence. The location of the 3' probe used for Southern blot analysis of ES cells is indicated in blue, and the PCR primers used to distinguish wild-type and recombined alleles for genotyping of animals (F1, R1, LacZ R2; see Materials and methods) are also indicated. (B) Southern blot analysis of a targeted cell line (+/-) in comparison to wild-type (+/+) HM-1 ES cells, using a probe external to the targeting vector 3' homology arm. (C) Genotyping of wild-type (+/+), heterozygous (+/\u2212 ) or homozygous mutant (-/-) mice by PCR amplification of fragments specific for the wild-type (354 bp) or mutated allele (269 bp), using the F1, R1 and LacZ R2 primers. (D,E) Anti-Gbx1 immunostaining. At E18.5, Gbx1 protein is absent in the spinal cord of Gbx1-/- mice (E), compared to wild- type (D). Scale bars: 100 \u00b5m. Figure 2. Abnormal phenotype of a Gbx1-/- mouse when walking. Sequential pictures compare the normal gait of a wild-type mouse (A) and the abnormal gait (\"duck-like\" walk) of a Gbx1-/- mutant when walking (B). A movie of these mice is available (Supplementary Material). Figure 3. Effects of Gbx1 mutation on the latency and number of slips in the beam walking test. ** p<0.01 vs WT; Student t-test. Figure 4. Open field performance of wild-type (WT) and Gbx1-/- mice. The distance traveled over the 20 min period of test reflects locomotor activity. The average speed was calculated during movement in the whole arena for the entire period of testing. Exploration of the central part of the open fied is expressed as the number of entries and percentage of time spent in the center. * p<0.05 and ** p<0.01 vs WT; Student t-test. Figure 5. Startle reactivity and pre-pulse inhibition in wild-type (WT) and Gbx1-/- mice. Startle reactivity to background noise (65 dB), or to 70, 80, 85, 90 dB acoustic stimulation, and startle reflex to a 110 dB stimulus, are presented. The percentage of pre-pulse inhibition of the startle response is displayed as a percentage of the pre-pulse intensity. BN, white noise; P, acoustic pulse intensity; ST, acoustic startle to 110 dB; PP, pre-pulse intensity. Figure 6. Absence of morphological and molecular abnormalities in the developing dorsal horn of Gbx1-/- mice. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/- (B,D,F,H) mice at E18.5 (A,B) and E16.5 (C-H) are shown. All sections are at the lumbar level. (A,B) Nissl-stained sections. No differences are detectable between wild-type and mutants (n=3). In situ hybridizations for two transcription factor encoding genes, Lbx1 (C,D) and Lmx1b (E,F), and for the axon guidance molecule netrin-1 (G,H), are shown (n=3 for each marker). No differences are observed between wild-type and mutants. Scale bars: 100 \u00b5m. Figure 7. Developmental progression of afferent projections in the dorsal horn of Gbx1-/- mice. (A-B') Anti-calbindin-D28K antibody staining. At E18.5, calbindin fibers have already entered the spinal gray matter in wild-type (A,A') and Gbx1-/- specimens (B,B'; n=3). Panels A',B' are higher magnifications of the areas boxed in A,B. ( C,D) Expression of Drg11 in wild-type at E18.5 (C) and Gbx1 mutant (D) mice. Mutant specimens were indistinguishable from wild-types (n=3). (E-H) Anti-peripherin antibody staining at E18.5 (E,F) and E16.5 (G,H). This staining reveals similar ingrowth of group IA muscle sensory afferents that grow to the ventral spinal cord (arrows) in wild-type (E,G) and mutant (F,H) (n=3 for each stage). Scale bars: 100 \u00b5m (A',B': 50 \u00b5m). Figure 8. Abnormal GABAergic differentiation in Gbx1-/- mice. Expression of Gad67 in wild-type (A,C) and Gbx1-/- (B,D) mice at E18.5 (n=3). Higher magnification views (C,D; areas boxed in A,B) show the dorsal horn, in which cell countings were performed. Expression of Pax2 in wild-type (E) and Gbx1-/- (F) mice at E18.5 (n=3). Expression of Slc17a6 in wild-type (G) and Gbx1-/- (H) mice at E18.5 (n=3). (I) Countings (percentages of labelled vs total cells) revealed that the proportion of Gad67+ cells is diminished by 16% in Gbx1-/- mice 1414 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants (50.89%\u00b12.61 Gad67+ cells in WT; 34.85%\u00b11.84 in Gbx1-/- mice; *** p<0.001; Student t-test). Also, the proportion of Pax2+ cells is diminished by 14.7% in Gbx1-/- mice (58.57%\u00b14,03 Pax2+ cells in WT; 42.41%\u00b15.96 in Gbx1-/- mice; *** p<0.001; Student t-test). In contrast, countings revealed that the proportion of Slc17a6+ cells is increased by 14.4% in Gbx1-/- mice (50.96%\u00b11.84 Slc17a6+ cells in WT; 65.16%\u00b12.94 in Gbx1-/- mice; *** p<0.001). Scale bars: 100 \u00b5m. Figure 9. Examples of TUNEL labeling of lumbar spinal cord sections of wild-type (A,C) and Gbx1-/- (B,D) mice. Sections are shown at E12.5 (A,B) (n=2) and E18.5 (C,D) (n=2). Some TUNEL-labelled cells are seen in the dorsal root ganglia (drg) at E12.5, in both wild-type and mutant. Scale bars: 100 \u00b5m. in situ hybridization : In situ hybridization was performed with digoxigenin-labeled probes as previously described (Chotteau-Leli\u00e8vre et al. 2006). Template DNAs were kindly provided by Drs K. Jagla (Lbx1), C. Birchmeier (Lmx1b), M. Tessier-Lavigne (Netrin), A.J. Tobin (Gad67) and P. Gruss (Pax2), P. Bouillet (Gbx2), B. Giros (Slc17a6), F. Chen (Drg11), R. Krumlauf (Hoxb1), and F. Rijli (Hoxa2). For all experiments 3 animals of each genotype, from 2 or more independent litters, were analyzed. Cell countings were performed in the dorsal horn (Gad67, Pax2, Slc17a6) or ventral horn (Islet1) on 3 transverse sections for each animal, at comparable levels of the lumbar spinal cord (all sections were collected serially, with section planes being separated by 112 \u00b5m). Three animals of each genotype were thus analyzed for each marker. All expression patterns were documented using a macroscope (Leica M420) or microscope (DM4000B, objective 10x), both connected to a Photometrics camera with the CoolSNAP (v. 1.2) imaging software (Roger Scientific, Chicago, IL). Cell counts were performed using the image J (NIH 1.45S) software. Blue labelled cells and unlabelled cells were counted manually with the cell counter plugin. Three sections separated by 112 \u00b5m in 3 independent embryos were counted for each condition, and statistical significance of cell counts was validated by Student's paired t-test. Graphs represent averages of counting values and SEM. conclusion : We have generated Gbx1-/- loss of function mutant mice, and investigated the development of the spinal cord dorsal horn in these mutants. Gbx1-/- mutants are viable and fertile, but display an altered gait during forward movement that specifically affects hindlimbs, beginning at post-natal days 16-17. This abnormal gait, documented by a series of behavioral tests, is not due to deficits in muscle strength or motor coordination, but may be related to proprioceptive deficits suggested by reduced performance in beam walking, a test used in studies of proprioception. Proprioceptive neuron afferents form two types of termination zones, one in the intermediate spinal cord and one in the ventral spinal cord where they are directly connected to motoneurons (Brown, 1981). Although our molecular analyses in mutant embryos do not reveal abnormal assembly of those proprioceptive sensory afferents, we cannot exclude abnormal functions of those afferents caused by reduced GABAergic cell number in superficial dorsal horn (see below), which is the transition and/or adjacent region for proprioceptive afferents. Such proprioceptive-like deficits may be also secondary to a defect in sensory perception or expression of such sensory response, as indicated by sensory deficits in Gbx1-/- mice and abnormal development of the dorsal horn, the main target of sensory afferents. Some of the deficits, such as altered sensory nerve conduction velocity, are significantly altered in females, whereas significant difference in hot plate performance was identified only in males. Although such differences could reflect sexual dimorphism, it is difficult to draw such a conclusion as definitive for two major reasons: (i) in some tests where a significant difference in performance was observed for one gender, the opposite gender may display a similar tendency, which remained non statistically significant; (ii) if for example females would be more prone to effects of Gbx1 mutation we could expect to find them less performant in different tests; however, the gender effects were inconsistent and concerned males of females depending on the measured parameter. The spinal cord dorsal horn largely consists of inhibitory (GABAergic) and excitatory (glutamatergic) neurons that modulate somatosensory inputs from the periphery, including pain, temperature and mechanoception (Glasgow, 2005). Our analysis of major neuronal classes revealed a reduced number of GABAergic inhibitory interneurons expressing Gad67 in the superficial dorsal horn of Gbx1-/- mice. Gbx1 may therefore be functionally required for the differentiation of local inhibitory interneurons in the dorsal horn, corroborating a previous report of Gbx1 expression in a specific subset of GABAergic neurons in this region of the spinal cord (John, Wildner & Britsch 2005). Furthermore, our findings suggest that Gbx1 functions as a gene that promotes GABAergic over glutamatergic differentiation in the dorsal horn. A disruption in the balance between inhibitory and excitatory neuronal activity could explain the phenotype observed in Gbx1 mutants. Indeed, the imbalance of inhibitory and excitatory activity may lead to altered signaling to second-order neurons in the intermediate zone, which through an excitatory polysynaptic chain excite motor neurons in ventral horn to initiate protective movements or abnormal proprioceptive behaviors. Such abnormal sensory processing is suggested at least for thermal stimuli, as Gbx1-/- mice displayed increased latency suggesting reduced pain in the hot plate test (thermosensory functions). Finally, considering that locomotor deficits become apparent at P16-17, we cannot exclude that abnormal gait may result from postnatal developmental or neurodegenerative events, which would need to be investigated. Despite the clear behavioral phenotype and reduced pool of GABAergic neurons in the dorsal horn, we did not observe any change in the expression of homeodomain factors involved in dorsal spinal cord patterning, or markers for primary sensory afferents, indicating that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutants. An explanation for these results\u2014and for the overall mild phenotype of the mutants\u2014is that Gbx1 and Gbx2 are coexpressed in dorsal spinal cord cells at early stages of embryogenesis : hence the 1212 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants presence of Gbx2 (and its subtle upregulation observed at E12.5-E14.5 in mutants) might compensate for Gbx1 loss of function with respect to early regulatory events. Generation of Gbx1;Gbx2 double mutants will be required to assess possible redundant functions, and the availability of a Gbx2 floxed (conditional) allele does allow strategies for a spinal cord-specific inactivation, which would alleviate the lethality of the Gbx2 null mutants (Wassarman et al. 1997). Despite the importance of dorsal spinal cord in normal sensory processing, our knowledge concerning the establishment of neuronal circuits remains limited (Graham, Brichta & Callister 2007; Todd 2010). In this regard, our work contributes to understand how transcription factors cooperate for regulating cell specification and eventual distribution of neuronal subtypes in the developing spinal cord, providing clues for further dissecting functional circuitry of the dorsal spinal cord. immunohistochemistry : After antigen unmasking in citrate 0.1 M (pH 6) during 15 min in a microwave oven, sections were treated in H 202 3% in PBS 1x for 5 min, rinsed in PBS 1x, then blocked in PBS 1x containing 0.25% Triton-X100, 5% normal goat serum and incubated overnight at 4\u00b0C with rabbit anti-Gbx1 (kindly provided by Dr S. Britsch; 1:500), rabbit anti-calbindin D-28K (Chemicon, 1:1000), rabbit anti-Peripherin (Chemicon, 1:500), and mouse anti-Islet1 (40.2D6, concentrated, Developmental Studies Hybridoma Bank, Iowa City, IA, 1:100) followed by species-specific biotin-coupled secondary antibodies (1:400, Jackson Laboratories) diluted in PBS 1x. Detection was performed using a Vectastain Elite ABC Kit, following the manufacturer's instructions. Nissl staining was performed by incubation in 0.5% cresyl violet in water for 15 min. TUNEL was performed using the APOPTAG\u00ae Peroxidase In Situ Apoptosis detection kit (Millipore). For all experiments 3 animals of each genotype were analyzed. funding statement : This work was supported by grants from the Agence Nationale de la Recherche (ANR Neurosciences 2007, ANR Blanc 2011), the Fondation pour la Recherche M\u00e9dicale (Equipe FRM 2007), the Deutsche Forschungsgemeinschaft (SFB-655 A3-Brand), the Italian Association for Cancer Research (AIRC), and by institutional funding from the Centre National de la Recherche Scientifique (CNRS), Institut National de la Sant\u00e9 et de la Recherche M\u00e9dicale (INSERM), and University of Strasbourg. Behavioral phenotyping was partly subsidized by the EUMODIC European Consortium and the Mouse Clinical Institute (MCI/ICS, Strasbourg). 1313 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants acknowledgments : We thank B. Schuhbaur for excellent technical assistance. We are grateful to Dr. K. Niederreither for a critical reading of the manuscript, and to Drs. M. Paschaki and D. Demb\u00e9l\u00e9 for help with statistical analysis. We thank Drs. C. Birchmeier, S. Britsch, F. Chen, K. Jagla, P. Bouillet, B. Giros, P. Gruss, M. Tessier-Lavigne, R. Krumlauf and F. Rijli for the gift of reagents. supplementary figure legends : Figure S1. Expression analysis of Gbx2 in the developing spinal cord of Gbx1 mutants. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/(B,D,F,H) mice are shown. All sections are at the lumbar level. In situ hybridizations for Gbx2 were performed at different developmental stages: E12.5 (A,B; n=2), E14.5 (C,D; n=2), E16.5 (E,F; n=3), and E18.5 (G,H; n=3). Scale bars: 100 \u00b5m. Figure S2. Analysis of rhombomeric markers in Gbx1-/- embryos. Whole-mount in situ hybridizations of E9.5 embryos with 2 markers of prospective rhombomeres: Hoxb1, which labels rhombomere 4 (A,B; n=3), and Hoxa2, which marks rhombomeres 2 to 6 and associated neural crest (C,D; n=3). Scale bars: 50 \u00b5m. Figure S3. In situ hybridization analysis of Gad67-expressing cells in the prenatal hindbrain. Sections are shown at various levels of the brain stem (A-F) and cerebellum (G,H) of wild-type (A,C,E,G) and Gbx1-/- (B,D,F,H) mice at E18.5. Scale bars: 100 \u00b5m. Figure S4. Analysis of developing spinal cord motor neurons in Gbx1 mutants. Expression of Islet1 in the lumbar spinal cord of wild-type (A,C) and Gbx1-/- (B,D) mice at E14.5 (A,B; n=2) and E16.5 (C,D; n=2). (E) Countings revealed that the numbers of Islet1+ cells in the ventral horn are not significantly diminished in Gbx1-/- mice (at E14.5: 76\u00b17.33 Islet1+ cells in WT; 75.22\u00b13.13 in Gbx1-/- mice; NS; at E16.5: 22.38\u00b15.96 Islet1+ cells in WT; 25.33\u00b16.70 in Gbx1-/- mice; NS). Scale bars: 100 \u00b5m. Movie sequence showing a Gbx1+/+ and a Gbx1-/- mouse. behavioral phenotyping procedures : Cohorts of 10 week-old male and female Gbx1-/- mice in a C57BL/6J genetic background (7 males and 8 females), with their wild-type (WT, 10 males and 9 females) counterparts, were used in this study. Mice were group housed and allowed 2 weeks acclimation in the phenotyping area with controlled temperature (21-22\u00b0C) under a 12-12 h light-dark cycle (lights on 7am-7pm), with food and water available ad libitum. Testing started at 10 weeks of age, and all procedures were carried out in accordance with European institutional guidelines. Behavioral tests were performed successively for each cohort of mice, during the light phase of the circadian cycle, according to a pipeline established by the European Mouse Disease Clinic 66 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants (EUMODIC pipeline 2), by trained experimenters familiar with observation of normal gait patterns in mice. Detailed procedures for each test are available at the URL: http://www.empress.har.mrc.ac.uk/viewempress/index.php?pipeline=EUMODIC+Pipeline+2. Neurological examination: General health and basic sensory motor functions were evaluated using a modified SHIRPA protocol (Brown, Chambon & Hrab\u00e9 de Angelis 2005; protocol at http://www.empress.har.mrc.ac.uk/viewempress/index.php?pipelineprocedure=EUMODIC+Pipeline+2~Modified+SHIRPA). This analysis is adapted from the procedure developed by Irwin (1968) and from the SHIRPA protocol (Hatcher et al. 2001). It provides an overview of physical appearance, body weight, neurological reflexes and sensory abilities. Rotarod test: This test evaluates motor coordination and balance by measuring the ability of animals to maintain balance on a rotating rod (Bioseb, Chaville, France). Mice were given three testing trials during which the rotation speed accelerated from 4 to 40 rounds per min (rpm) in 5 min. Trials were separated by 5-10 min intervals. The average latency (time to fall from the rotating rod) of the three trials was used as index of motor coordination performance. Grip test: This test measures the maximal muscle strength (g) using an isometric dynamometer connected to a grid (Bioseb). Mice were allowed to grip the grid either with the forepaws or with both the forepaws and hindpaws, then were pulled backwards until they released the grid. Each mouse was submitted to 3 consecutive trials immediately after the modified SHIRPA procedure. The maximal strength developed by the mouse before releasing the grid was recorded and the average value of the three trials was adjusted to body weight. Beam walking: This test is used to evaluate fine motor coordination and proprioceptive function. The apparatus used is a 2 cm diameter and 110 cm long wooden beam, elevated 50 cm above the ground. A goal box (12 x 12 x 14 cm) is attached at one extremity of the beam. Animals were first habituated to the goal box for 1 min. They were then submitted to 3 training trials during which they were placed at different points of the beam, with the head directed to the goal box, and allowed to walk the corresponding distance to enter the goal box. After training, animals were submitted to 3 testing trials during which they were placed at the extremity of the beam opposite to the goal box and allowed to walk the beam distance and enter the goal box. The latency to enter the goal box and the number of slips (when one or both hindpaws slipped laterally from the beam) were measured. Hot plate test: The mice were placed into a glass cylinder on a hot plate (Bioseb) adjusted to 52\u00b0C, and the latency of the first pain reaction of any hindlimb (licking, flinches) was recorded, with a maximum of 30 s testing. Electrophysiological measurements: Electrophysiological recordings were performed under ketamine-xylazine anesthesia (100 and 10 mg/kg body weight, respectively) using a Key Point electromyograph apparatus (Medronic, France). Disposal scalp needle electrodes were used (ref 9013R0312, Medtronic). The body temperature was maintained at 37\u00b0C with a homeothermic blanket (Harvard, Paris, France). For measuring the sensory nerve conduction velocity (SNCV), recording electrodes were inserted at the proximal part of the tail and stimulating electrodes placed 20 mm from the recording needles towards the extremity of the tail. A ground needle electrode was inserted between the stimulating and recording electrodes. Caudal nerve was stimulated with a series of 20 pulses of 0.2 ms duration at a supramaximal intensity of 8 mA. The average response is included for statistical analysis. The compound muscle action potential (CMAP) was measured in gastrocnemius muscle after sciatic nerve stimulation. For this purpose, stimulating electrodes were placed at the level of the sciatic nerve at 1 cm from the vertebral column, and recording electrodes placed in the gastrocnemius muscle. A ground needle was inserted in the contralateral paw. The sciatic nerve was stimulated with a single 0.2 ms pulse at a supramaximal intensity of 8 mA. The amplitude (mV) and the distal latency of the responses (ms) were measured. Anxiety-related behavior - open field test: Mice were tested in automated open fields (Panlab, Barcelona, Spain), each virtually divided into central and peripheral regions. The open fields were placed in a room homogeneously illuminated at 150 Lux. Each mouse was placed in the periphery of the open field and allowed to explore freely the apparatus for 20 min, with the experimenter out of the animal\u2019s sight. The distance traveled, the number of rears, and time spent in the central and 77 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants peripheral regions were recorded over the test session. The latency and number of crosses into as well as the percent time spent in center area are used as index of emotionality/anxiety. Sensorimotor gating - auditory startle reflex reactivity and pre-pulse inhibition (PPI): Acoustic startle reactivity and pre-pulse inhibition of startle were assessed in a single session using standard startle chambers (SR-Lab Startle Response System, San Diego Instruments). Ten different trial types were used: acoustic startle pulse alone (110 db), eight different prepulse trials in which either 70, 75, 85 or 90 dB stimuli were presented alone or preceding the pulse, and finally one trial (NOSTIM) in which only the background noise (65 dB) was presented to measure the baseline movement in the Plexiglas cylinder. In the startle pulse or prepulse alone trials, the startle reactivity was analyzed, and in the prepulse plus startle trials the amount of PPI was measured and expressed as percentage of the basal startle response. Statistical analyses: Data were analyzed using unpaired Student t-test, one way or repeated measures analysis of variance (ANOVA) with one between factor (genotype) and one within factor (time). Qualitative parameters (i.e. some of the clinical observations) were analyzed using \u03c72 test. The level of significance was set at p<0.05. animal ethics statement : Animal experimentation protocols were reviewed and approved by the Direction D\u00e9partementale des Services V\u00e9t\u00e9rinaires (agreement #67-172 to H.M., 67-189 to P.D., and institutional agreement #D67-218-5 for animal housing) and conformed to the NIH and European Union guidelines, provisions of the Guide for the Care and Use of Laboratory Animals, and the Animal Welfare Act. table captions : Table 1. Effects of Gbx1 mutation on body weight, basic neurological reflexes, specific motor abilities and pain sensitivity. Statistically different parameters in wild-type vs mutants appear in bold. * p<0.05 and **p<0.01 vs wild-type; Student t-test. Table 2. Effects of Gbx1 mutation on sensory nerve conduction velocity. The sensory nerve conduction velocity was measured at the level of the caudal nerve. The latency and the amplitude of gastrocnemius muscle response evoked by sciatic nerve stimulation were also recorded. * p<0.05 vs wild-type; Student t-test. grip strength : (adjusted to body weight) (g) 2 paws 3.97 \u00b1 0.18 4.00 \u00b1 0.24 3.75 \u00b1 0.17 3.38 \u00b1 0.26 4 paws 8.30 \u00b1 0.24 7.72 \u00b1 0.37 7.00 \u00b1 0.24 7.22 \u00b1 0.54 Hot plate (s) 13.43 \u00b1 1.26 17.13 \u00b1 1.08* 12.72 \u00b1 1.15 15.03 \u00b1 1.63 Males Females Wi d-type -/- Wild-type -/- PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Table 2: Sensory nerve conduction velocity (m/s) 70.33 \u00b1 1.70 72.20 \u00b1 1.53 63.93 \u00b1 2.35 71.81 \u00b1 1.50* Gastrocnemius M-wave Latency (ms) 0.93 \u00b1 0.06 0.91 \u00b1 0.06 0.99 \u00b1 0.08 0.84 \u00b1 0.05 amplitude (mv) : 43 \u00b1 3.15 : 44.60 \u00b1 5.87 46.46 \u00b1 6.70 55.41 \u00b1 6.03 Males Females Wi d-type -/- Wild-type -/- PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 1 Figure 1 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 2 Figure 2 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 3 Figure 3 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 4 Figure 4 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 5 Figure 7 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 6 Figure 6 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 7 Figure 7 PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 8 Figure 8: neuronal markers PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t Figure 9 Figure 9: apoptosis PeerJ reviewing PDF | (v2012:11:45:3:0:NEW 23 May 2013) R ev ie w in g M an us cr ip t",
    "v3_text": "results and discussion : Gbx1-deficient mice are viable, but display a typical duck-like gait A loss of function allele for the Gbx1 gene was generated by homologous recombination in murine embryonic stem cells (see Materials and Methods). The mutated Gbx1 allele is devoid of the entire homeodomain-coding sequence and ~100 adjacent nucleotides (Fig. 1A-C). After generation of germ-line transmitting chimeras, heterozygous mutant mice (Gbx1+/-) were found to be viable, fertile and apparently normal. After intercrossing Gbx1+/- mice, Gbx1-/- mutants (generated in a C57BL/6J genetic background) were born in the expected mendelian ratio. Immunohistochemistry performed with an anti-Gbx1 antibody confirmed the absence of detectable Gbx1 protein in the spinal cord of E18.5 Gbx1-/- mutants (Fig. 1D,E). We also checked the expression of Gbx2 from E12.5 onwards to exclude a potential compensatory expression due to the loss of function of Gbx1. A subtle increase of Gbx2 mRNA expression might occur in spinal cord cells of Gbx1-/- mice at E12.5-14.5, however this increase was no longer detected at E16.5 or E18.5 (Supplementary material, Fig. S1). Interestingly, when observed by 4-6 weeks of age, most mutants displayed a typical, unevenless in walking (\"duck-like\") gait (Fig. 2 and Supplementary material: movie). Both male and female Gbx1-/- mice were fertile and had a normal life span. General health and sensorimotor abilities in adult Gbx1 mutants Gbx1-/- males and females had a normal body weight (Table 1) and a normal overall physical appearance. However, many of the Gbx1-/mutants showed significantly abnormal gait (\u03c72 5.20, p<0.05)\u2265 . Indeed, 43% of Gbx1 mutant males and 63% of Gbx1 mutant females displayed lack of fluidity in movement, and limping related to hyper-flexion followed by hyper-extension of one or both hindpaws (Table 1; Fig. 2; Supplementary material: movie). Gbx1-/- males and females also showed significantly reduced short-term locomotor activity following immediate transfer for the modified SHIRPA test, as compared to WT counterparts (t 3.46, p<0.01) (Table 1). \u2265 The other features of general health and basic neurological reflexes were not affected in Gbx1 mutants. When tested for specific motor abilities, motor coordination performance measured in the rotarod test (t 1.29, NS)\u2264 and the muscle strength (grid grip) test (t 1.38, NS)\u2264 were not affected in Gbx1-/- males and females (Table 1). In the beam walking test, the latency to cross the beam was increased (t15=3.71, p<0.01 for females; non significant for males) and the number of slips was slightly increased (even if not significantly), especially in Gbx1-/- females (Fig. 3). In the open field test, there was a significant effect of genotype concerning locomotor activity [F(1,30)=6.51, p<0.05], reflecting reduced locomotion in all Gbx1-/- animals. When considering each gender separately, both Gbx1-/- males and females tended to have reduced locomotor activity over the testing period (although not statistically significant, p=0.09) (Fig. 4). The average speed during motion was also significantly lower in Gbx1-/- males and females than in WT (t 3.36, p<0.01) (Fig. 4).\u2265 The number of entries into, and the percentage of time spent in, the center of the arena also differed between genotypes [F(1,30)\u226514.48, p<0.001]. Both Gbx1-/- males and females had significantly decreased number of entries and spent less time in the center of the open field than WT counterparts (t\u22652.62, p<0.05) (Fig. 4), which might reflect increased anxiety in Gbx1-/- mutants. The reduced exploration of the center might also be due to the observed reduced locomotor activity of Gbx1-/- mutants. Altogether, these data show that Gbx1-/- mutant 88 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants mice have a clear defect in locomotion, although this defect does not appear to result in a coordination problem or a muscle strength deficiency. To test whether ablation of Gbx1 could affect sensory response, we measured the response of Gbx1 mutant mice in a hot plate test. The withdrawal latency was higher in Gbx1-/- males (but not in females) than in WT (t15=2.10, p=0.05) (Table 1), suggesting reduced thermal pain sensitivity in Gbx1 mutant males. The consequence of Gbx1 inactivation on acoustic startle and pre-pulse inhibition of startle reflex was also evaluated. Regardless of gender, the startle reactivity was comparable between WT and Gbx1-/- mice for all the acoustic stimuli including the startling pulse [Genotype F(1,30)\u22640.83, Sex F(1,30)\u22640.85, Genotype*Sex F(1,30)\u22641.11, NS] (Fig. 5). When the startling pulse was preceded with prepulses with lower intensities, the PPI level was also comparable between genotypes [Genotype F(1,30)=0.55, Sex F(1,30)=0.32, Genotype*Sex F(1,30)=0.11, NS)] (Fig. 5). Furthermore, electromyography (EMG) measurements revealed that the sensory nerve conduction velocity differed significantly between genotypes [F(1,29)=7.31, p<0.05]; indeed, Gbx1-/- females had significantly increased sensory nerve conduction velocity (t14=2.83, p<0.05) (Table 2), as measured at the level of the caudal nerve. On the other hand, the latency and amplitude of the gastrocnemius muscle response evoked by sciatic nerve stimulation were comparable between genotypes [F(1,29)=1.63, NS]. In summary, we used a variety of behavioral and electrophysiological phenotyping tests to evaluate sensory and motor functions in Gbx1 mutant mice. We found that both Gbx1-/- males and females show reduced locomotor activity in different situations. Decreased exploratory behavior was found in the open field test and following immediate transfer during clinical observations. Exploration of the central part of the open field arena was significantly decreased in Gbx1-/- males and females, which might suggest increased anxiety in these mutants. However, this could also be due to the reduced locomotor activity of the mutants. Indeed, Gbx1-/- mice also showed decreased average speed with no significant effect on the distance travelled in the open field. Their strong altered gait during forward movement might explain the reduced speed and locomotor activity in the open field. This is supported by data from the beam walking test showing that at least Gbx1-/females required longer time to cross the beam distance and tended to have higher number of slips (not statistically significant), suggesting inappropriate rear paw placement, and supporting altered proprioceptive sensitivity. Electrophysiological measurements showed that Gbx1-/females had increased sensory nerve conduction velocity, measured in the caudal nerve, supporting the altered sensory functions observed in the behavioral tests. Our results also show that Gbx1 gene disruption affects exploratory behavior with increased anxiety and decreased locomotor performance, but without motor coordination or muscle strength defects nor changes in central sensory motor gating. Altogether, the behavioral data revealed that the Gbx1-/- animals have apparent proprioceptive defects and/or altered sensory abilities. We cannot exclude that the proprioceptive defects may be secondary to a defect in sensory pathways (for example, caused by pain on movement), because mutant mice also display a significantly reduced response time in the hot plate (thermosensory) test. Gbx1-/- mice do not show obvious hindbrain patterning defects Gbx genes are related to the Drosophila unplugged gene, which acts during development of the tracheal system, and perhaps for specification of neuroblast sublineages (Chiang et al. 1995; Cui & Doe 1995). There are two Gbx genes in amniote species (human, mouse and chicken), as well as in zebrafish (Lin et al. 1996; Bouillet et al. 1995; Shamim & Mason 1998; Niss & Leutz 1998; Rhinn et al. 2003). Previous studies showed that in mouse, Gbx2 is involved in early specification of the midbrain-hindrain boundary (MHB) organizer, a signaling center that will pattern the anterior hindbrain rhombomeres (Wassarman et al. 1997; Waters & Lewandoski 2006; for review: Rhinn & Brand 2001; Simeone 2000). In zebrafish it was shown that gbx1 acts during early positioning of the MHB, whereas gbx2 functions at later stages, once the MHB is established (Rhinn et al. 2004; 2009; Burroughs-Garcia et al. 2011). In mouse, Gbx1 is not expressed at the MHB as is the case during early zebrafish development. Its expression starts at E7.75 in the prospective hindbrain, spanning rhombomeres 2 to 7 during the segmentation phase (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2004). This suggested that Gbx1 might be involved in early embryonic hindbrain patterning, which could underlie behavioral deficits associated with loss of Gbx1 function. To assess for possible rhombomeric abnormalities in Gbx1-/- mutants, we performed whole-mount in situ hybridizations at E9.5 with several markers, including Hoxb1 and Hoxa2. This analysis did not show any molecular or structural abnormality of the hindbrain 99 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants rhombomeres in Gbx1-/- embryos (Supplementary material, Fig. S2). This suggests that Gbx1 is not required for early hindbrain patterning, in contrast to its mouse homologue Gbx2 (Wassarman et al. 1997; Waters & Lewandoski 2006). Analysis of hindbrain derivatives (brain stem and cerebellum) at E18.5 using Gad67 as a differentiation marker also did not reveal any difference in Gbx1-/- versus wild-type mice (Supplementary material, Fig. S3). At E12.5, the expression domains of Gbx1 and Gbx2 overlap, both being expressed in the ventricular zone and the mantle zone of the entire dorsal spinal cord (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2004). After E12.5, however, Gbx2 expression is rapidly downregulated. Gbx1 and Gbx2 are thus transiently coexpressed in progenitor cells of the dorsal spinal cord, with Gbx1 being the only Gbx gene persistently expressed during later dorsal horn development. Development of the spinal cord dorsal horn in Gbx1 mutant mice The prominent expression of Gbx1 in the dorsal horn could be relevant for the abnormal gait phenotype of adult Gbx1 mutant mice, which led us to ask whether Gbx1 is required for the maturation and/or specification of neurons of the dorsal horn during development. Nissl staining of E18.5 spinal cord sections revealed no obvious difference between the dorsal horn of wild-type and Gbx1-/- animals at thoraco-lumbar levels (Fig. 6A,B). Despite the clear behavioral phenotype, we were unable to identify any consistent alteration in the expression of several molecular markers of dorsal spinal cord cell populations in Gbx1-/- embryos. These markers included the genes encoding the transcription factors Lbx1 (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002) (Fig. 6C,D) and Lmx1b (Chen et al. 2001) (Fig. 6E,F), analyzed at E12.5, 14.5, 16.5 and 18.5, and the axon guidance molecule Netrin-1 (Leonardo et al. 1997) (Fig. 6G,H) analyzed at E18.5. Projection pattern of primary sensory afferents in the dorsal horn of Gbx1-/- mutants Afferents sensing pain and temperature mainly project to laminae I/II. Afferents for sensing innocuous mechanoreceptor signals such as texture, shape, vibration, and pressure project predominantly to internal laminae (III, IV, V). Afferents for sensing proprioceptive signals project through the dorsal horn to the ventrally located motor neurons (Brown 1981; Willis & Coggeshall 1991). The formation of laminar-specific projections is a key event in the development of appropriate neuronal connections in many regions of the central nervous system. Collaterals from the different classes of sensory axons then penetrate the gray matter of the spinal cord sequentially. Each class of sensory axons projects directly to its target lamina and never branches into inappropriate laminae (for review: Caspary & Anderson 2003; Lewis 2006). We examined the development of primary sensory afferent projections to the dorsal horn, which are well defined at E18.5, in Gbx1 mutant mice (Ozaki & Snider 1997). The central projections of cutaneous nociceptive sensory neurons first arrive in the dorsal root entry zone and begin to invade the spinal grey matter by E12.5. Staining with an antibody to calbindin-28K at E16.5 and E18.5 marks a subset of cutaneous neurons and their afferent fibers (Honda 1995). By E18.5, calbindin+ fibers have invaded the dorsal horn of wild-type and Gbx1 mutants (Fig. 7A,B). Drg11 is required for the projection of cutaneous sensory afferent fibers to the dorsal spinal cord. Indeed, Drg11 mutant mice display abnormalities in the spatio-temporal patterning of cutaneous sensory afferent fiber projections to the dorsal, but not the ventral spinal cord, as well as defects in dorsal horn morphogenesis (Chen et al. 2001). In Gbx1-/- mutant mice, Drg11 expression was not affected in the dorsal horn at E12.5, 14.5, 16.5 or 18.5 (Fig. 7C,D). Altogether, these data suggest that there are no defects in patterning of sensory afferent fiber projections to the dorsal horn, which selectively affects cutaneous afferents. We further examined proprioceptive afferents at E18.5 by using antibodies to peripherin (Escurat et al. 1990). No consistent difference between wild-type and Gbx1-/- mice was observed at the level of proprioceptive fibers that extend toward motoneurons and interneurons in the deep dorsal horn, or at the level of fibers that enter into the spinal gray matter, at E16.5 or E18.5 (arrows in Fig. 7E,F). During the revision of our manuscript, another Gbx1 mutant allele was described (Buckley et al. 2013). In contrast to our observations, those mutants show disorganized peripherin expression, together with a decrease of Islet1-expressing cells in the ventral horn of the lumbar spinal 1010 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants cord (Buckley et al. 2013). This led us to analyze Islet1-expressing cells in ventral motor neurons in our Gbx1 mutants. No differences in the number of Islet1+ cells within the lumbar ventral spinal cord were found at E14.5, E16.5 or E18.5 (Fig. S4). Thus, in contrast to the data of Buckley et al., our analysis does not suggest a defect in the assembly of the proprioceptive sensorimotor circuit. As the same Gbx1 exon (exon 2) was targeted in both loss of function alleles, the reason for the phenotypic discrepancy remains unclear, although it should be mentioned that some of the analyses were not performed at the same developmental time points, and that the mice might have different genetic backgrounds. Reduced GABAergic neuronal differentiation in Gbx1-/- mutants Gbx1 is first expressed in the ventricular zone of the spinal cord at E11.5 (Rhinn et al. 2004 ; Waters, Wilson & Lewandoski 2004). Then at E12.5-E13.5, it is broadly expressed in the mantle zone of the dorsal spinal cord. At E14 with the appearance of a distinguishable dorsal horn, Gbx1 expression becomes more restricted. At E12.5, Gbx1 is coexpressed with Lbx1; thus Gbx1 cells correspond to class B neurons (John, Wildner & Britsch 2005). As described in the introduction, late-born class B neurons comprise initially two populations, dILA and dILB. Because Gbx1 neurons co-express Lhx1/5 and Pax2, but not Lmx1b and Tlx3, it has been suggested that these neurons correspond to the dILA neuronal subtype (John, Wildner & Britsch 2005). It has been shown that dILA neurons undergo GABAergic differentiation (Cheng et al. 2004), and as mature GABA+ neurons they continue expressing Gbx1 (John, Wildner & Britsch 2005). We therefore analyzed GABAergic neurons in the spinal cord of Gbx1-/- mutants, which we identified by expression of glutamic acid decarboxylase GAD67, an enzyme that regulates GABA synthesis. At E18.5, Gad67-expressing cells are found throughout the developing spinal cord of control mice (Somogyi et al. 1995). Importantly, Gad67 expression was reduced in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8A-D), i.e. there was a 16% decrease in the number of Gad67-expressing cells in comparison to WT mice (Fig. 8I). This may reflect an abnormal development or survival of GABAergic neurons, which in consequence coud lead to abnormal control of neuronal network in dorsal horn. This finding was strengthened by analysis of Pax2, another gene expressed in GABAergic cells in the spinal cord (Cheng et al. 2004), with cell countings corroborating the decrease in the number of GABAergic cells (Fig. 8E,F,I). Interestingly, when Gbx1-/- pups were checked visually around weaning, the locomotor deficits were first observed around post-natal days (P)16-17, which coincides with the time point at which the GABA pathway switches from excitatory to inhibitory (Ben-Ari et al. 2007). We also observed that Gad67 expression was unchanged in the brain stem and cerebellum of E18.5 Gbx1-/- mutants (Fig. S3), arguing against an involvement of these structures in the observed phenotype. We next addressed the question if the observed decrease of GABAergic cells is due to neuronal cell death or to a possible change of GABAergic to glutamatergic fate. TUNEL experiments performed at various stages (E12.5, 14.5, 16.5, 18.5) showed no abnormal apoptosis in the spinal cord of Gbx1-/- mice (Figure S5). We then analyzed expression of Slc17a6, encoding VGLUT2, a vesicular glutamate transporter expressed in glutamatergic neurons (Kaneko et al. 2002). At E18.5, Slc17a6-expressing cells were increased in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8G,H). This finding suggests that part of the \"missing\" GABAergic cells may have differentiated into glutamatergic neurons. Glutamate and GABA are the predominant neurotransmitters for excitatory and inhibitory neurons, respectively, in the vertebrate brain. These two neurotransmitters are typically expressed in a mutually exclusive manner (Bellocchio et al. 2000; Fremeau et al. 2001), thereby defining the major functional subdivision in neuronal cell types. In the dorsal horn of the spinal cord\u2014the major relay center for processing somatosensory information\u2014, most ascending projection neurons and a subset of local circuit interneurons are excitatory and use glutamate as their transmitter. These neurons are modulated by local inhibitory neurons, many of which are GABAergic (for reviews: Melzack & Wall 1965; Malcangio & Bowery 1996; Dickenson 2002). Thus, GABA may inhibit transmitter release from primary afferent fibers. The output neurons of the dorsal horn are projection neurons, which are concentrated in lamina I and scattered throughout laminae III/VI, and relay sensory information to several brain areas. The vast majority of neurons in the dorsal horn are local circuit interneurons that do not 1111 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants project outside of the spinal cord. The output of projection neurons is influenced by local excitatory and inhibitory neurons (Todd 2010; Larsson & Broman 2011; Guo et al. 2012). The balance between excitation and inhibition is crucial for maintaining normal sensory function (Basbaum et al. 2009; Costigan, Scholz & Woolf 2009; Ross et al. 2010; Takazawa & MacDermott 2010). In Gbx1 mutants, the reduction of GABAergic neurons, and the possible switch of some of these neurons to a glutamatergic identity, may disrupt neuronal circuitry, becoming phenotypically apparent at adult stages as measured by abnormal performance in several behavioral tests. Further electrophysiological studies will be necessary to link the decrease of GABAergic neurons to the abnormal gait observed in adult Gbx1 mutant mice. in situ hybridization : In situ hybridization was performed with digoxigenin-labeled probes as previously described (Chotteau-Leli\u00e8vre et al. 2006). Template DNAs were kindly provided by Drs K. Jagla (Lbx1), C. Birchmeier (Lmx1b), M. Tessier-Lavigne (Netrin), A.J. Tobin (Gad67) and P. Gruss (Pax2), P. Bouillet (Gbx2), B. Giros (Slc17a6), F. Chen (Drg11), R. Krumlauf (Hoxb1), and F. Rijli (Hoxa2). For all experiments 3 animals of each genotype, from 2 or more independent litters, were analyzed. Cell countings were performed in the dorsal horn (Gad67, Pax2, Slc17a6) or ventral horn (Islet1) on 3 transverse sections for each animal, at comparable levels of the lumbar spinal cord (all sections were collected serially, with section planes being separated by 112 \u00b5m). Three animals of each genotype were thus analyzed for each marker. All expression patterns were documented using a macroscope (Leica M420) or microscope (DM4000B, objective 10x), both connected to a Photometrics camera with the CoolSNAP (v. 1.2) imaging software (Roger Scientific, Chicago, IL). Cell countings were performed with the NIH imageJ software. abstract : Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/- mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutant mice. However, analysis of terminal neuronal differentiation revealed that the number of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. 22 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants materials and methods : Construction of a Gbx1 targeting vector Genomic sequences encompassing the mouse Gbx1 gene were isolated from a 129SV genomic phage library, using as a probe a Gbx1 cDNA fragment previously characterized (Rhinn et al. 2004). A Gbx1 loss of function mutation was produced by 44 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants homologous recombination in embryonic stem cells (Ram\u00ecrez-Solis, Davis & Bradley 1993). The targeting vector contained a 5.4 kb XmnI fragment (upstream arm), ending 33 bp upstream of the homeodomain sequence located in Gbx1 second exon, and a 1.6 kb KpnI fragment (downstream arm), whose sequence started 91 bp downstream from the homeodomain. These fragments were excised from the recombinant phage and cloned in the mutagenesis pGN vector (Le Mouellic, Lallemand & Br\u00fblet 1990) to generate the pGN-Gbx1 targeting vector (Fig. 1A). In this vector, the fragments are inserted on each side of a lacZ reporter gene and a neomycin resistance gene, and their insertion by homologous recombinaton in the Gbx1 gene will generate a 313 bp deletion encompassing the entire homeodomain (Fig. 1A). Transfection of embryonic stem cells and selection of targeted clones HM-1 embryonic stem (ES) cells (Magin, Whir & Melton 1992) were cultured on neomycin-resistant mouse embryonic fibroblasts, as described in Robertson, 1987. Ten \u03bcg of the pGN-Gbx1 targeting vector were linearized by digestion of the unique NotI restriction site, and electroporated into 2\u00d7107 ES cells resuspended in 750 \u03bcl HeBS medium (20 mM Hepes pH 7.05, 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, 6 mM glucose), at 200 V, 960 \u03bcF. Positive selection was carried out for 11 days with 350 \u03bcg/ml G418. Resistant colonies were picked and DNA was extracted from a fraction (1/5) of the cells to perform Southern blot analysis to identify homologous recombination events. The probe used is an external fragment located immediately downstream to the targeting vector (Fig. 1A,B). Positive clones were expanded before freezing. The frequency of homologous recombination was 7 out of 350 clones analyzed. Generation and genotyping of chimeric and mutant mice After thawing, 10 to 15 ES cells were microinjected into blastocysts collected at E3.5 from C57BL/6 females mated with C57BL/6 males (for procedures: Nagy et al. 2003). Injected blastocysts were reimplanted in the uterine horn of pseudopregnant recipient females. Chimeric animals were back-crossed to C57BL/6J mice and germ-line transmission was scored by the presence of agouti coat pigmentation. Heterozygous offspring were identified by PCR genotyping. Tail tips were incubated in lysis buffer (50 mM Tris pH 8.0, 100 mM EDTA, 100 mM NaCl, 1% SDS, 0.6 mg/ml proteinase K) overnight at 55\u00b0C, phenol-chloroform extracted, ethanol precipitated and redissolved in 10 mM Tris-HCl, 1 mM EDTA pH 8.0 at a final concentration of 0.2-1.0 \u03bcg/\u03bcl. The presence of a wild-type or mutated allele was detected using three primers: a sense primer F1: 5\u2019-GGTGACAGCGAGGACAGCTTCCT-3\u2019, an antisense primer R1: 5\u2019-CCCAGAACGACTGCTCACATTGC-3\u2019, and an antisense primer LacZ R2: 5\u2019-GGCCTCTTCGCTATTACGCCA-3\u2019 The presence of a wild-type allele was detected using the F1/R1 primers which amplify a 354 bp fragment. The presence of a mutated allele was detected by using the F1/LacZ R2 primers which amplify a 269 bp fragment. Thirty cycles (denaturation : 1 min, 95\u00b0C, annealing : 1 min, 62\u00b0C; elongation : 30 s, 74\u00b0C) were performed, and the amplified products were separated by 2% agarose gel electrophoresis (Fig. 1C). Phenotypic and molecular analyses were performed after several generations of backcrosses (>5) to C57BL/6J mice, resulting in a nearly pure genetic background. Tissue collection and sample preparation Pregnant females obtained from natural matings (morning of vaginal plug was considered as E0.5) were sacrificed and fetuses were collected in phosphate-buffered saline (PBS, pH 7.5) after cesarean section. The specimens were dissected, fixed overnight in 4% paraformaldehyde (PFA), cryoprotected in 20% sucrose, and embedded in Shandon Cryomatrix (Thermo Electron Corperation) before freezing at -80\u00b0C. Cryosections (14 \u00b5m thickness, Leica CM3050S cryostat) sections were made in a coronal plane, collected on Superfrost slides, and stored at -80\u00b0C until use. For whole-mount immunostaining or in situ hybridization, embryos were fixed overnight in 4% PFA, dehydrated, and stored at -20 \u00b0C in 100% methanol. 55 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants figure legends : Figure 1. Inactivation of the mouse Gbx1 gene by homologous recombination. (A) The upper drawing shows the restriction map of the wild-type locus, boxes and lines corresponding to exons and introns, respectively. The homeodomain sequence is in red. In the targeting vector (below), two Gbx1 genomic fragments (between the dashed lines) flank a lacZ reporter gene and the neomycin resistance gene (grey box), transcribed in the same orientation (thin arrow) as Gbx1. In the recombined locus (lower drawing), 313 bp of Gbx1 exon 2 (including the homeodomain) are replaced by the lacZ neo sequence. The location of the 3' probe used for Southern blot analysis of ES cells is indicated in blue, and the PCR primers used to distinguish wild-type and recombined alleles for genotyping of animals (F1, R1, LacZ R2; see Materials and methods) are also indicated. (B) Southern blot analysis of a targeted cell line (+/-) in comparison to wild-type (+/+) HM-1 ES cells, using a probe external to the targeting vector 3' homology arm. (C) Genotyping of wild-type (+/+), heterozygous (+/\u2212 ) or homozygous mutant (-/-) mice by PCR amplification of fragments specific for the wild-type (354 bp) or mutated allele (269 bp), using the F1, R1 and LacZ R2 primers. (D,E) Anti-Gbx1 immunostaining. At E18.5, Gbx1 protein is absent in the spinal cord of Gbx1-/- mice (E), compared to wild- type (D). Scale bars: 100 \u00b5m. Figure 2. Abnormal phenotype of a Gbx1-/- mouse when walking. Sequential pictures compare the normal gait of a wild-type mouse (A) and the abnormal gait (\"duck-like\" walk) of a Gbx1-/- mutant when walking (B). A movie of these mice is available (Supplementary Material). Figure 3. Effects of Gbx1 mutation on the latency and number of slips in the beam walking test. ** p<0.01 vs WT; Student t-test. Figure 4. Open field performance of wild-type (WT) and Gbx1-/- mice. The distance traveled over the 20 min period of test reflects locomotor activity. The average speed was calculated during movement in the whole arena for the entire period of testing. Exploration of the central part of the open fied is expressed as the number of entries and percentage of time spent in the center. * p<0.05 and ** p<0.01 vs WT; Student t-test. Figure 5. Startle reactivity and pre-pulse inhibition in wild-type (WT) and Gbx1-/- mice. Startle reactivity to background noise (65 dB), or to 70, 80, 85, 90 dB acoustic stimulation, and startle reflex to a 110 dB stimulus, are presented. The percentage of pre-pulse inhibition of the startle response is displayed as a percentage of the pre-pulse intensity. BN, white noise; P, acoustic pulse intensity; ST, acoustic startle to 110 dB; PP, pre-pulse intensity. Figure 6. Absence of morphological and molecular abnormalities in the developing dorsal horn of Gbx1-/- mice. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/- (B,D,F,H) mice at E18.5 (A,B) and E16.5 (C-H) are shown. All sections are at the lumbar level. (A,B) Nissl-stained sections. No differences are detectable between wild-type and mutants. In situ hybridizations for two transcription factor encoding genes, Lbx1 (C,D) and Lmx1b (E,F), and for the axon guidance molecule netrin-1 (G,H), are shown. No differences are observed between wild-type and mutants. Scale bars: 100 \u00b5m. Figure 7. Developmental progression of afferent projections in the dorsal horn of Gbx1-/- mice. (A,B) Anti-calbindin-D28K antibody staining. At E18.5, calbindin fibers have already entered the spinal gray matter in wild-type embryos (A). Homozygous Gbx1-/- specimens (B) are indistinguishable from wild-types. Insets show higher magnifications of the dorsal horn. (C,D) Expression of Drg11 in wild-type (C) and Gbx1 mutant (D) mice. Mutant specimens were indistinguishable from wild-types. (E,F) Anti-peripherin antibody staining. This staining reveals similar ingrowth of group IA muscle sensory afferents that grow to the ventral spinal cord (arrows) in wild-type (E) and mutant (F). Scale bars: 100 \u00b5m (insets in A,B: 50 \u00b5m). Figure 8. Abnormal GABAergic differentiation in Gbx1-/- mice. Expression of Gad67 in wild-type (A,C) and Gbx1-/- (B,D) mice at E18.5. Higher magnification views (C,D; areas boxed in A,B) show the dorsal horn, in which cell countings were performed. Expression of Pax2 in wild-type (E) and Gbx1-/- (F) mice at E18.5. Expression of Slc17a6 in wild-type (G) and Gbx1-/- (H) mice at E18.5. (I) Countings (percentages of labelled vs. total cells) revealed that the numbers of Gad67+ cells are diminished by 16% in Gbx1-/- mice (51.16\u00b12.25% 1414 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants Gad67+ cells in WT; 35.07\u00b11.81% in Gbx1-/- mice). Also, the numbers of Pax2+ cells are diminished by 14.7% in Gbx1-/- mice (57.60\u00b13.42% Pax2+ cells in WT; 42.91\u00b13.42% in Gbx1-/- mice). In contrast, countings revealed that the numbers of Slc17a6+ cells are increased by 14.4% in Gbx1-/- mice (51.10\u00b11.69% Slc17a6+ cells in WT; 65.53\u00b14.64% in Gbx1-/- mice). Scale bars: 100 \u00b5m. acknowledgments : We thank B. Schuhbaur for excellent technical assistance. We are grateful to Dr. K. Niederreither for a critical reading of the manuscript, and to Drs. C. Birchmeier, S. Britsch, F. Chen, K. Jagla, P. Bouillet, B. Giros, P. Gruss, M. Tessier-Lavigne, R. Krumlauf and F. Rijli for the gift of reagents. conclusion : We have generated Gbx1-/- loss of function mutant mice, and investigated the development of the spinal cord dorsal horn in these mutants. Gbx1-/- mutants are viable and fertile, but display an altered gait during forward movement that specifically affects hindlimbs, beginning at post-natal days 16-17. This abnormal gait, documented by a series of bevioral tests, is not due to deficits in muscle strength or motor coordination. Our marker analysis in mutant embryos does not reveal abnormal assembly of the proprioceptive sensory afferents. However, we cannot exclude that the proprioceptive defects may be caused by abnormal development of superficial dorsal horn (see below), the transition and/or adjacent region for proprioceptive afferents, or may be secondary to a defect in sensory pathways as mutant mice also display a significantly reduced response time in the hot plate (thermosensory) test. Some of the deficits, including for example altered sensory nerve conduction velocity, are significantly altered in females, whereas significant difference in hot plate performance was identified only in males. Although such differences could reflect sexual dimorphism, it is difficult to draw such a conclusion as definitive for two major reasons: (i) in some tests where a significant difference in performance was observed for one gender, the opposite gender may display a similar tendency, which remained non statistically significant; (ii) if for example females would be more prone to effects of Gbx1 mutation we could expect to find them less performant in different tests; however, the gender effects were inconsistent and concerned males of females depending on the measured parameter. The spinal cord dorsal horn largely consists of inhibitory (GABAergic) and excitatory (glutamatergic) neurons that modulate somatosensory inputs from the periphery, including pain, temperature and mechanoception. Our analysis of major neuronal classes revealed a reduced number of GABAergic inhibitory interneurons expressing Gad67 in the superficial dorsal horn of Gbx1-/- mice. Gbx1 may therefore be functionally required for the differentiation of local inhibitory interneurons in the dorsal horn, corroborating a previous report of Gbx1 expression in a specific subset of GABAergic neurons in this region of the spinal cord (John, Wildner & Britsch 2005). Furthermore, our findings suggest that Gbx1 functions as a gene that promotes GABAergic over glutamatergic differentiation in the dorsal horn. A disruption in the balance between inhibitory and excitatory neuronal activity could explain the phenotype observed in Gbx1 mutants. Indeed, the imbalance of inhibitory and excitatory activity may lead to altered signaling to second-order neurons in the intermediate zone, which through an excitatory polysynaptic chain excite motor neurons in ventral horn to initiate protective movements. Such abnormal sensory processing is suggested at least for thermal stimuli, as Gbx1-/- mice displayed significantly increased response time in the hot plate test (thermosensory functions). Finally, considering that locomotor deficits become apparent at P16-17, we cannot exclude that abnormal gait may result from postnatal developmental or neurodegenerative events, which would need to be investigated. Despite the clear behavioral phenotype and reduced pool of GABAergic neurons in the dorsal horn, we did not observe any change in the expression of homeodomain factors involved in dorsal spinal cord patterning, or markers for primary sensory afferents, indicating that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutants. An explanation for these results\u2014and for the overall mild phenotype of the mutants\u2014is that Gbx1 and Gbx2 are coexpressed in dorsal spinal cord cells at early stages of embryogenesis : hence the presence of Gbx2 (and its subtle upregulation observed at E12.5-E14.5 in mutants) might compensate for Gbx1 loss of function with respect to early regulatory events. Generation of Gbx1;Gbx2 double mutants will be required to assess possible redundant functions, and the availability of a Gbx2 floxed (conditional) allele does allow strategies for a spinal cord-specific inactivation, which would alleviate the lethality of the Gbx2 null mutants (Wassarman et al. 1997). 1212 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants Despite the importance of dorsal spinal cord in normal sensory processing, our knowledge concerning the establishment of neuronal circuits remains limited (Graham, Brichta & Callister 2007; Todd 2010). In this regard, our work contributes to understand how transcription factors cooperate for regulating cell specification and eventual distribution of neuronal subtypes in the developing spinal cord, providing clues for further dissecting functional circuitry of the dorsal spinal cord. immunohistochemistry : After antigen unmasking in citrate 0.1 M (pH 6) during 15 min in a microwave oven, sections were treated in H 202 3% for 5 min, rinsed in PBS1X then blocked in PBS containing 0.25% Triton-X100, 5% normal goat serum and incubated overnight at 4\u00b0C with rabbit anti-Gbx1 (kindly provided by Dr S. Britsch; 1:500), rabbit anti-calbindin D-28K (Chemicon, 1:1000), rabbit anti-Peripherin (Chemicon, 1:500), and mouse anti-Islet1 (40.2D6, Developmental Studies Hybridoma Bank, Iowa City, IA, 1:100) followed by species-specific biotin-coupled secondary antibodies (1:400, Jackson Laboratories) diluted in PBS1X. Detection was performed using a Vectastain Elite ABC Kit, following the manufacturer's instructions. Nissl staining was performed by incubation in 0.5% cresyl violet for 15 min. TUNEL was performed using the APOPTAG\u00ae Peroxidase In Situ Apoptosis detection kit (Millipore). For all experiments 3 animals of each genotype were analyzed. behavioral phenotyping procedures : Cohorts of 10 week-old male and female Gbx1-/- mice in a C57BL/6J genetic background (7 males and 8 females), with their wild-type (WT, 10 males and 9 females) counterparts, were used in this study. Mice were group housed and allowed 2 weeks acclimation in the phenotyping area with controlled temperature (21-22\u00b0C) under a 12-12 h light-dark cycle (lights on 7am-7pm), with food and water available ad libitum. Testing started at 10 weeks of age, and all procedures were carried out in accordance with European institutional guidelines. Behavioral tests were performed successively for each cohort of mice, during the light phase of the circadian cycle, according to a pipeline established by the European Mouse Disease Clinic (EUMODIC pipeline 2). Detailed procedures for each test are available at the URL: http://www.empress.har.mrc.ac.uk/viewempress/index.php?pipeline=EUMODIC+Pipeline+2. Neurological examination: General health and basic sensory motor functions were evaluated using a modified SHIRPA protocol (Brown, Chambon & Hrab\u00e9 de Angelis 2005; protocol at http://www.empress.har.mrc.ac.uk/viewempress/index.php?pipelineprocedure=EUMODIC+Pipeline+2~Modified+SHIRPA). This analysis is adapted from the procedure developed by Irwin (1968) and from the SHIRPA protocol (Hatcher et al. 2001). It provides an overview of physical appearance, body weight, neurological reflexes and sensory abilities. Rotarod test: This test evaluates motor coordination and balance by measuring the ability of animals to maintain balance on a rotating rod (Bioseb, Chaville, France). Mice were given three testing trials during which the rotation speed accelerated from 66 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants 4 to 40 rounds per min (rpm) in 5 min. Trials were separated by 5-10 min intervals. The average latency (time to fall from the rotating rod) of the three trials was used as index of motor coordination performance. Grip test: This test measures the maximal muscle strength (g) using an isometric dynamometer connected to a grid (Bioseb). Mice were allowed to grip the grid either with the forepaws or with both the forepaws and hindpaws, then were pulled backwards until they released the grid. Each mouse was submitted to 3 consecutive trials immediately after the modified SHIRPA procedure. The maximal strength developed by the mouse before releasing the grid was recorded and the average value of the three trials was adjusted to body weight. Beam walking: This test is used to evaluate fine motor coordination and proprioceptive function. The apparatus used is a 2 cm diameter and 110 cm long wooden beam, elevated 50 cm above the ground. A goal box (12 x 12 x 14 cm) is attached at one extremity of the beam. Animals were first habituated to the goal box for 1 min. They were then submitted to 3 training trials during which they were placed at different points of the beam, with the head directed to the goal box, and allowed to walk the corresponding distance to enter the goal box. After training, animals were submitted to 3 testing trials during which they were placed at the extremity of the beam opposite to the goal box and allowed to walk the beam distance and enter the goal box. The latency to enter the goal box and the number of slips (when one or both hindpaws slipped laterally from the beam) were measured. Hot plate test: The mice were placed into a glass cylinder on a hot plate (Bioseb) adjusted to 52\u00b0C, and the latency of the first pain reaction of any hindlimb (licking, flinches) was recorded, with a maximum of 30 s testing. Electrophysiological measurements: Electrophysiological recordings were performed under ketamine-xylazine anesthesia (100 and 10 mg/kg body weight, respectively) using a Key Point electromyograph apparatus (Metronic, France). The body temperature was maintained at 37\u00b0C with a homeothermic blanket (Harvard, Paris, France). For measuring the sensory nerve conduction velocity (SNCV), recording electrodes were inserted at the proximal part of the tail and stimulating electrodes placed 20 mm from the recording needles towards the extremity of the tail. A ground needle electrode was inserted between the stimulating and recording electrodes. Caudal nerve was stimulated with a series of 20 pulses of 0.2 ms duration at a supramaximal intensity of 8 mA. The average response is included for statistical analysis. The compound muscle action potential (CMAP) was measured in gastrocnemius muscle after sciatic nerve stimulation. For this purpose, stimulating electrodes were placed at the level of the sciatic nerve at 1 cm from the vertebral column, and recording electrodes placed in the gastrocnemius muscle. A ground needle was inserted in the contralateral paw. The sciatic nerve was stimulated with a single 0.2 ms pulse at a supramaximal intensity of 8 mA. The amplitude (mV) and the distal latency of the responses (ms) were measured. Anxiety-related behavior - open field test: Mice were tested in automated open fields (Panlab, Barcelona, Spain), each virtually divided into central and peripheral regions. The open fields were placed in a room homogeneously illuminated at 150 Lux. Each mouse was placed in the periphery of the open field and allowed to explore freely the apparatus for 20 min, with the experimenter out of the animal\u2019s sight. The distance traveled, the number of rears, and time spent in the central and peripheral regions were recorded over the test session. The latency and number of crosses into as well as the percent time spent in center area are used as index of emotionality/anxiety. Sensorimotor gating - auditory startle reflex reactivity and pre-pulse inhibition (PPI): Acoustic startle reactivity and pre-pulse inhibition of startle were assessed in a single session using standard startle chambers (SR-Lab Startle Response System, San Diego Instruments). Ten different trial types were used: acoustic startle pulse alone (110 db), eight different prepulse trials in which either 70, 75, 85 or 90 dB stimuli were presented alone or preceding the pulse, and finally one trial (NOSTIM) in which only the background noise (65 dB) was presented to measure the baseline movement in the Plexiglas cylinder. In the startle pulse or prepulse alone trials, the startle reactivity was analyzed, and in the prepulse plus startle trials the amount of PPI was measured and expressed as percentage of the basal startle response. 77 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants Statistical analyses: Data were analyzed using unpaired Student t-test, one way or repeated measures analysis of variance (ANOVA) with one between factor (genotype) and one within factor (time). Qualitative parameters (i.e. some of the clinical observations) were analyzed using \u03c72 test. The level of significance was set at p<0.05. animal ethics statement : Animal experimentation protocols were reviewed and approved by the Direction D\u00e9partementale des Services V\u00e9t\u00e9rinaires (agreement #67-172 to H.M., 67-189 to P.D., and institutional agreement #D67-218-5 for animal housing) and conformed to the NIH and European Union guidelines, provisions of the Guide for the Care and Use of Laboratory Animals, and the Animal Welfare Act. funding statement : This work was supported by grants from the Agence Nationale de la Recherche (ANR Neurosciences 2007, ANR Blanc 2011), the Fondation pour la Recherche M\u00e9dicale (Equipe FRM 2007), the Deutsche Forschungsgemeinschaft (SFB-655 A3-Brand), the Italian Association for Cancer Research (AIRC), and by institutional funding from the Centre National de la Recherche Scientifique (CNRS), Institut National de la Sant\u00e9 et de la Recherche M\u00e9dicale (INSERM), and University of Strasbourg. Behavioral phenotyping was partly subsidized by the EUMODIC European Consortium and the Mouse Clinical Institute (MCI/ICS, Strasbourg). 1313 PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Phenotypic analysis of Gbx1-/- mouse mutants supplementary figure legends : Figure S1. Expression analysis of Gbx2 in the developing spinal cord of Gbx1 mutants. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/(B,D,F,H) mice are shown. All sections are at the lumbar level. In situ hybridizations for Gbx2 were performed at different developmental stages: E12.5 (A,B), E14.5 (C,D), E16.5 (E,F), and E18.5 (G,H). Scale bars: 100 \u00b5m. Figure S2. Analysis of rhombomeric markers in Gbx1-/- embryos. Whole-mount in situ hybridizations of E9.5 embryos with 2 markers of prospective rhombomeres: Hoxb1, which labels rhombomere 4 (A,B), and Hoxa2, which marks rhombomeres 2 to 6 and associated neural crest (C,D). Scale bars: 50 \u00b5m. Figure S3. In situ hybridization analysis of Gad67-expressing cells in the prenatal hindbrain. Sections are shown at various levels of the brain stem (A-F) and cerebellum (G,H) of wild-type (A,C,E,G) and Gbx1-/- (B,D,F,H) mice at E18.5. Scale bars: 100 \u00b5m. Figure S4. Analysis of developing spinal cord motor neurons in Gbx1 mutants. Expression of Islet1 in the lumbar spinal cord of wild-type (A,C) and Gbx1-/- (B,D) mice at E18.5 (A,B) and E14.5 (C,D). (E) Countings revealed that the numbers of Islet1+ cells in the ventral horn are not significantly diminished in Gbx1-/- mice (at E18.5: 22.83\u00b16.165 Islet1+ cells in WT; 22.59\u00b17.33 in Gbx1-/- mice; at E14.5: 73.91\u00b10.8 Islet1+ cells in WT; 73.99\u00b16.36 in Gbx1-/- mice). Scale bars: 100 \u00b5m. Figure S5. Examples of TUNEL labeling of lumbar spinal cord sections of wild-type (A,C) and Gbx1-/- (B,D) mice. Sections are shown at E12.5 (A,B) and E18.5 (C,D). Some TUNEL-labelled cells are seen in the dorsal root ganglia (drg) at E12.5, in both wild-type and mutant. Scale bars: 100 \u00b5m. Movie sequence showing a Gbx1+/+ and a Gbx1-/- mouse. table captions : Table 1. Effects of Gbx1 mutation on body weight, basic neurological reflexes, specific motor abilities and pain sensitivity. Statistically different parameters in wild-type vs mutants appear in bold. * p<0.05 and **p<0.01 vs wild-type; Student t-test. Table 2. Effects of Gbx1 mutation on sensory nerve conduction velocity. The sensory nerve conduction velocity was measured at the level of the caudal nerve. The latency and the amplitude of gastrocnemius muscle response evoked by sciatic nerve stimulation were also recorded. * p<0.05 vs wild-type; Student t-test. grip strength : (adjusted to body weight) 2-paws 3.97 \u00b1 0.18 4.00 \u00b1 0.24 3.75 \u00b1 0.17 3.38 \u00b1 0.26 4-paws 8.30 \u00b1 0.24 7.72 \u00b1 0.37 7.00 \u00b1 0.24 7.22 \u00b1 0.54 Hot plate 13.43 \u00b1 1.26 17.13 \u00b1 1.08* 12.72 \u00b1 1.15 15.03 \u00b1 1.63 Males Females Wi d-type Gbx1-/Wild-type Gbx1-/- PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t Table 2: Sensory Nerve Conduction Velocity 70.33 \u00b1 1.70 72.20 \u00b1 1.53 63.93 \u00b1 2.35 71.81 \u00b1 1.50* Gastrocnemius M-wave Latency 0.93 \u00b1 0.06 0.91 \u00b1 0.06 0.99 \u00b1 0.08 0.84 \u00b1 0.05 amplitude : 43 \u00b1 3.15 : 44.60 \u00b1 5.87 46.46 \u00b1 6.70 55.41 \u00b1 6.03 Males Females Wi d-type Gbx1-/Wild-type Gbx1-/- PeerJ reviewing PDF | (v2012:11:45:2:0:NEW 10 Apr 2013) R ev ie w in g M an us cr ip t",
    "v4_text": "abstract : Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/- mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutant mice. However, analysis of terminal neuronal differentiation revealed that the number of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. 22 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants materials and methods : Construction of a Gbx1 targeting vector Genomic sequences encompassing the mouse Gbx1 gene were isolated from a 129SV genomic phage library, using as a probe a Gbx1 cDNA fragment previously characterized (Rhinn et al. 2004). A Gbx1 loss of function mutation was produced by 44 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants homologous recombination in embryonic stem cells (Ram\u00ecrez-Solis, Davis & Bradley 1993). The targeting vector contained a 5.4 kb XmnI fragment (upstream arm), ending 33 bp upstream of the homeodomain sequence located in Gbx1 second exon, and a 1.6 kb KpnI fragment (downstream arm), whose sequence started 91 bp downstream from the homeodomain. These fragments were excised from the recombinant phage and cloned in the mutagenesis pGN vector (Le Mouellic, Lallemand & Br\u00fblet 1990) to generate the pGN-Gbx1 targeting vector (Fig. 1A). In this vector, the fragments are inserted on each side of a lacZ reporter gene and a neomycin resistance gene, and their insertion by homologous recombinaton in the Gbx1 gene will generate a 313 bp deletion encompassing the entire homeodomain (Fig. 1A). Transfection of embryonic stem cells and selection of targeted clones HM-1 embryonic stem (ES) cells (Magin, Whir & Melton 1992) were cultured on neomycin-resistant mouse embryonic fibroblasts, as described in Robertson, 1987. Ten \u03bcg of the pGN-Gbx1 targeting vector were linearized by digestion of the unique NotI restriction site, and electroporated into 2\u00d7107 ES cells resuspended in 750 \u03bcl HeBS medium (20 mM Hepes pH 7.05, 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, 6 mM glucose), at 200 V, 960 \u03bcF. Positive selection was carried out for 11 days with 350 \u03bcg/ml G418. Resistant colonies were picked and DNA was extracted from a fraction (1/5) of the cells to perform Southern blot analysis to identify homologous recombination events. The probe used is an external fragment located immediately downstream to the targeting vector (Fig. 1A,B). Positive clones were expanded before freezing. The frequency of homologous recombination was 7 out of 350 clones analyzed. Generation and genotyping of chimeric and mutant mice After thawing, 10 to 15 ES cells were microinjected into blastocysts collected at E3.5 from C57BL/6 females mated with C57BL/6 males. Injected blastocysts were reimplanted in the uterine horn of pseudopregnant recipient females. Chimeric animals were back-crossed to C57BL/6J mice and germ-line transmission was scored by the presence of agouti coat pigmentation. Heterozygous offspring were identified by PCR genotyping. Tail tips were incubated in lysis buffer (50 mM Tris pH 8.0, 100 mM EDTA, 100 mM NaCl, 1% SDS, 0.6 mg/ml proteinase K) overnight at 55\u00b0C, phenol-chloroform extracted, ethanol precipitated and redissolved in 10 mM Tris-HCl, 1 mM EDTA pH 8.0 at a final concentration of 0.2-1.0 \u03bcg/\u03bcl. The presence of a wild-type or mutated allele was detected using three primers: a sense primer F1: 5\u2032-GGTGACAGCGAGGACAGCTTCCT-3\u2032, an antisense primer R1: 5\u2032-CCCAGAACGACTGCTCACATTGC -3\u2032 and an antisense primer LacZ R2: 5\u2019-GGCCTCTTCGCTATTACGCCA-3\u2019. The presence of a wild-type allele was detected using the F1/R1 primers which amplify a 354 bp fragment. The presence of a mutated allele was detected by using the F1/LacZ R2 primers which amplify a 269 bp fragment. Thirty cycles (denaturation : 1 min, 95\u00b0C, annealing : 1 min, 62\u00b0C; elongation : 30 s, 74\u00b0C) were performed, and the amplified products were separated by 2% agarose gel electrophoresis (Fig. 1C). Phenotypic and molecular analyses were performed after several generations of backcrosses (>5) to C57BL/6J mice, resulting in a nearly pure genetic background. Tissue collection and sample preparation Pregnant females obtained from natural matings (morning of vaginal plug was considered as E0.5) were sacrificed and fetuses were collected in phosphate-buffered saline (PBS, pH 7.5) after cesarean section. The specimens were dissected, fixed overnight in 4% paraformaldehyde (PFA), cryoprotected in 20% sucrose, and embedded in Shandon Cryomatrix (Thermo Electron Corperation) before freezing at -80\u00b0C. Cryosections (14 \u00b5m thickness, Leica CM3050S cryostat) sections were made in a coronal plane, collected on Superfrost slides, and stored at -80\u00b0C until use. For whole-mount immunostaining or in situ hybridization, embryos were fixed overnight in 4% PFA, dehydrated, and stored at -20 \u00b0C in 100% methanol. 55 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants results and discussion : Gbx1-deficient mice are viable, but display a typical duck-like gait A loss of function allele for the Gbx1 gene was generated by homologous recombination in murine embryonic stem cells (see Materials and Methods). The mutated Gbx1 allele is devoid of the entire homeodomain-coding sequence and ~100 adjacent nucleotides (Fig. 1A-C). After generation of germ-line transmitting chimeras, heterozygous mutant mice (Gbx1+/-) were found to be viable, fertile and apparently normal. After intercrossing Gbx1+/- mice, Gbx1-/- mutants (generated in a C57BL/6J genetic background) were born in the expected mendelian ratio. Immunohistochemistry performed with an anti-Gbx1 antibody confirmed the absence of detectable Gbx1 protein in the spinal cord of E18.5 Gbx1-/- mutants (Fig. 1D,E). We also checked the expression of Gbx2 from E12.5 onwards to exclude a potential compensatory expression due to the loss of function of Gbx1. No increase of Gbx2 expression was observed in Gbx1-/- mutants, except for a possible, subtle increase at E14.5 (Supplementary material, Fig. S1). When tested by 6 weeks of age, however, the mutants display a typical, unevenless in walking (\"duck-like\") gait (Fig. 2 and Supplementary material: movie). Both male and female Gbx1-/- mice were fertile and had a normal life span. General health and sensorimotor abilities in adult Gbx1 mutants Gbx1-/- males and females had a normal body weight (Table 1) and a normal overall physical appearance. However, many of the Gbx1-/mutants showed significantly abnormal gait (\u03c72 5.20, p<0.05)\u2265 . Indeed, 43% of Gbx1 mutant males and 63% of Gbx1 mutant females displayed lack of fluidity in movement, and limping related to hyper-flexion followed by hyper-extension of one or both hindpaws (Table 1; Fig. 2; Supplementary material: movie). Gbx1-/- males and females also showed significantly reduced short-term locomotor activity following immediate transfer for the modified SHIRPA test, as compared to WT counterparts (t 3.46, p<0.01) (Table 1). \u2265 The other features of general health and basic neurological reflexes were not affected in Gbx1 mutants. When tested for specific motor abilities, motor coordination performance measured in the rotarod test (t 1.29, NS)\u2264 and the muscle strength (grid grip) test (t 1.38, NS)\u2264 were not affected in Gbx1-/- males and females (Table 1). In the beam walking test, the latency to cross the beam was significantly increased (t15=3.71, p<0.01) and the number of slips was slightly increased (even if not significantly) in Gbx1-/females as compared to WT counterparts (Fig. 3). Males were less affected then females (Fig. 3). In the open field test, there was a significant effect of genotype concerning locomotor activity [F(1,30)=6.51, p<0.05], reflecting reduced locomotion in all Gbx1-/- animals. When considering each gender separately, both Gbx1-/- males and females tended to have reduced locomotor activity over the testing period (p=0.09) (Fig. 4). The average speed during motion was also significantly lower in Gbx1-/- males and females than in WT (t 3.36, p<0.01)\u2265 (Fig. 4). The number of entries into, and the percentage of time spent in, the center of the arena also differed between genotypes [F(1,30)\u226514.48, p<0.001]. Both Gbx1-/- males and females had significantly decreased number of entries and spent less time in the center of the open field than WT counterparts (t\u22652.62, p<0.05) (Fig. 4), which might reflect increased anxiety in Gbx1-/- mutants. The reduced exploration of the center might also be due to the observed reduced locomotor activity of Gbx1-/- mutants. Altogether, these data show that Gbx1-/- mutant mice have a clear defect in locomotion, although this defect does not appear to result in a coordination problem or a muscle strength deficiency. To test whether ablation of Gbx1 could affect sensory response, we measured the response of Gbx1 mutant mice in a hot plate test. The withdrawal latency was higher in Gbx1-/- males (but not in females) than in WT (t15=2.10, p=0.05) (Table 1), suggesting reduced thermal pain sensitivity in Gbx1 mutant males. The consequence of Gbx1 inactivation on acoustic startle and pre-pulse inhibition of startle reflex was also evaluated. Regardless of gender, the startle reactivity was comparable between WT and Gbx1-/- mice for all the acoustic stimuli including the startling pulse 88 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants [Genotype F(1,30)\u22640.83, Sex F(1,30)\u22640.85, Genotype*Sex F(1,30)\u22641.11, NS] (Fig. 5). When the startling pulse was preceded with prepulses with lower intensities, the PPI level was also comparable between genotypes [Genotype F(1,30)=0.55, Sex F(1,30)=0.32, Genotype*Sex F(1,30)=0.11, NS)] (Fig. 5). Furthermore, electromyography (EMG) measurements revealed that the sensory nerve conduction velocity differed significantly between genotypes [F(1,29)=7.31, p<0.05]; indeed, Gbx1-/- females had significantly increased sensory nerve conduction velocity (t14=2.83, p<0.05) (Table 2), as measured at the level of the caudal nerve. On the other hand, the latency and amplitude of the gastrocnemius muscle response evoked by sciatic nerve stimulation were comparable between genotypes [F(1,29)=1.63, NS]. In summary, we used a variety of behavioral and electrophysiological phenotyping tests to evaluate sensory and motor functions in Gbx1 mutant mice. We found that both Gbx1-/- males and females show reduced locomotor activity in different situations. Decreased exploratory behavior was found in the open field test and following immediate transfer during clinical observations. Exploration of the central part of the open field arena was significantly decreased in Gbx1-/- males and females, which might suggest increased anxiety in these mutants. However, this could also be due to the reduced locomotor activity of the mutants. Indeed, Gbx1-/- mice also showed decreased average speed with no significant effect on the distance travelled in the open field. Their strong altered gait during forward movement might explain the reduced speed and locomotor activity in the open field. This is supported by data from the beam walking test showing that at least Gbx1-/females required longer time to cross the beam distance and tended to have higher number of slips, suggesting inappropriate rear paw placement, and supporting altered proprioceptive sensitivity. Electrophysiological measurements showed that Gbx1-/- females had increased sensory nerve conduction velocity, measured in the caudal nerve, supporting the altered sensory functions observed in the behavioral tests. Our results also show that Gbx1 gene disruption affects exploratory behavior with increased anxiety and decreased locomotor performance, but without motor coordination or muscle strength defects nor changes in central sensory motor gating. Altogether, the behavioral data revealed that the Gbx1-/- animals have apparent proprioceptive defects and/or altered sensory abilities. We cannot exclude that the proprioceptive defects may be secondary to a defect in sensory pathways (for example, caused by pain on movement), because mutant mice also display a significantly reduced response time in the hot plate (thermosensory) test. Gbx1-/- mice do not show obvious hindbrain patterning defects Gbx genes are related to the Drosophila unplugged gene, which acts during development of the tracheal system, and perhaps for specification of neuroblast sublineages (Chiang et al. 1995; Cui & Doe 1995). There are two Gbx genes in amniote species (human, mouse and chicken), as well as in zebrafish (Lin et al. 1996; Bouillet et al. 1995; Shamim & Mason 1998; Niss & Leutz 1998; Rhinn et al. 2003). Previous studies showed that in mouse, Gbx2 is involved in early specification of the midbrain-hindrain boundary (MHB) organizer, a signaling center that will pattern the anterior hindbrain rhombomeres (Wassarman et al. 1997; Waters & Lewandoski 2006; for review: Rhinn & Brand 2001; Simeone 2000). In zebrafish it was shown that gbx1 acts during early positioning of the MHB, whereas gbx2 functions at later stages, once the MHB is established (Rhinn et al. 2004; 2009; Burroughs-Garcia et al. 2011). In mouse, Gbx1 is not expressed at the MHB as is the case during early zebrafish development. Its expression starts at E7.75 in the prospective hindbrain, spanning rhombomeres 2 to 7 during the segmentation phase (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2004). This suggested that Gbx1 might be involved in early embryonic hindbrain patterning, which could underlie behavioral deficits associated with loss of Gbx1 function. To assess for possible rhombomeric abnormalities in Gbx1-/- mutants, we performed whole-mount in situ hybridizations at E9.5 with several markers, including Hoxb1 and Hoxa2. This analysis did not show any molecular or structural abnormality of the hindbrain rhombomeres in Gbx1-/- embryos (Supplementary material, Fig. S2). This suggests that Gbx1 is not required for early hindbrain patterning, in contrast to its mouse homologue Gbx2 (Wassarman et al. 1997; Waters & Lewandoski 2006). Analysis of hindbrain derivatives (brain stem and cerebellum) at E18.5 using Gad67 as a differentiation marker also did not reveal any 99 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants difference in Gbx1-/- versus wild-type mice (Supplementary material, Fig. S3). At E12.5, the expression domains of Gbx1 and Gbx2 overlap, both being expressed in the ventricular zone and the mantle zone of the entire dorsal spinal cord (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2004). After E12.5, however, Gbx2 expression is rapidly downregulated. Gbx1 and Gbx2 are thus transiently coexpressed in progenitor cells of the dorsal spinal cord, with Gbx1 being the only Gbx gene persistently expressed during later dorsal horn development. Development of the spinal cord dorsal horn in Gbx1 mutant mice The prominent expression of Gbx1 in the dorsal horn could be relevant for the abnormal gait phenotype of adult Gbx1 mutant mice, which led us to ask whether Gbx1 is required for the maturation and/or specification of neurons of the dorsal horn during development. Nissl staining of E18.5 spinal cord sections revealed no obvious difference between the dorsal horn of wild-type and Gbx1-/- animals at thoraco-lumbar levels (Fig. 6A,B). Despite the clear behavioral phenotype, we were unable to identify any consistent alteration in the expression of several molecular markers of dorsal spinal cord cell populations in Gbx1-/- embryos. These markers included the genes encoding the transcription factors Lbx1 (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002) (Fig. 6C,D) and Lmx1b (Chen et al. 2001) (Fig. 6E,F), and the axon guidance molecule Netrin-1 (Leonardo et al. 1997) (Fig. 6G,H). Projection pattern of primary sensory afferents in the dorsal horn of Gbx1-/- mutants Afferents sensing pain and temperature mainly project to laminae I/II. Afferents for sensing innocuous mechanoreceptor signals such as texture, shape, vibration, and pressure project predominantly to internal laminae (III, IV, V). Afferents for sensing proprioceptive signals project through the dorsal horn to the ventrally located motor neurons (Brown 1981; Willis & Coggeshall 1991). The formation of laminar-specific projections is a key event in the development of appropriate neuronal connections in many regions of the central nervous system. Collaterals from the different classes of sensory axons then penetrate the gray matter of the spinal cord sequentially. Each class of sensory axons projects directly to its target lamina and never branches into inappropriate laminae (for review: Caspary & Anderson 2003; Lewis 2006). We examined the development of primary sensory afferent projections to the dorsal horn, which are well defined at E18.5, in Gbx1 mutant mice (Ozaki & Snider 1997). The central projections of cutaneous nociceptive sensory neurons first arrive in the dorsal root entry zone and begin to invade the spinal grey matter by E12.5. Staining with an antibody to calbindin-28K marks a subset of cutaneous neurons and their afferent fibers (Honda 1995). By E18.5, calbindin+ fibers have invaded the dorsal horn of wild-type and Gbx1 mutants (Fig.7A,B). Drg11 is required for the projection of cutaneous sensory afferent fibers to the dorsal spinal cord. Indeed, Drg11 mutant mice display abnormalities in the spatio-temporal patterning of cutaneous sensory afferent fiber projections to the dorsal, but not the ventral spinal cord, as well as defects in dorsal horn morphogenesis (Chen et al. 2001). In Gbx1-/- mutant mice, Drg11 expression was not affected in the dorsal horn at E18.5 (Fig.7C,D). Altogether, these data suggest that there are no defects in patterning of sensory afferent fiber projections to the dorsal horn, which selectively affects cutaneous afferents. Because primary cutaneous sensory afferent projections mediate (among other modalities) nociception, these results imply that Gbx1 mutant mice might not exhibit deficiencies in their behavioral responses to noxious stimuli. However, we observed by performing the hot plate test, that Gbx1-/- males show longer withdrawal latencies, suggesting reduced thermal pain sensitivity. We further examined proprioceptive afferents at E18.5 by using antibodies to peripherin (Escurat et al. 1990). No consistent difference between wild-type and Gbx1-/- mice was observed at the level of proprioceptive fibers that extend toward motoneurons and interneurons in the deep dorsal horn, or at the level of fibers that enter into the spinal gray matter (arrows in Fig. 7E,F). During the revision of our manuscript, another Gbx1 mutant allele was described (Buckley et al. 2013). In contrast to our observations, those mutants show disorganized peripherin expression, together with a decrease of Islet1-expressing cells in the ventral horn of the lumbar spinal cord (Buckley et al. 2013). This led us to analyze Islet1-expressing cells in ventral motor neurons in our Gbx1 mutants. No differences in the number of Islet1+ cells within the lumbar ventral spinal cord were found at E14.5 or E18.5 (Fig. S4). Thus, in contrast to the data of Buckley et al., our analysis does not suggest a defect in the assembly of the proprioceptive sensorimotor circuit. As the same Gbx1 exon (exon 2) was targeted 1010 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants in both loss of function alleles, the reason for the phenotypic discrepancy remains unclear, although it could be attributed to differences in the genetic background of the two mutant lines. Reduced GABAergic neuronal differentiation in Gbx1-/- mutants Gbx1 is first expressed in the ventricular zone of the spinal cord at E11.5 (Rhinn et al. 2004 ; Waters, Wilson & Lewandoski 2004). Then at E12.5-E13.5, it is broadly expressed in the mantle zone of the dorsal spinal cord. At E14 with the appearance of a distinguishable dorsal horn, Gbx1 expression becomes more restricted. At E12.5, Gbx1 is coexpressed with Lbx1; thus Gbx1 cells correspond to class B neurons (John, Wildner & Britsch 2005). As described in the introduction, late-born class B neurons comprise initially two populations, dILA and dILB. Because Gbx1 neurons co-express Lhx1/5 and Pax2, but not Lmx1b and Tlx3, it has been suggested that these neurons correspond to the dILA neuronal subtype (John, Wildner & Britsch 2005). It has been shown that dILA neurons undergo GABAergic differentiation (Cheng et al. 2004), and as mature GABA+ neurons they continue expressing Gbx1 (John, Wildner & Britsch 2005). We therefore analyzed GABAergic neurons in the spinal cord of Gbx1-/- mutants, which we identified by expression of glutamic acid decarboxylase GAD67, an enzyme that regulates GABA synthesis. At E18.5, Gad67-expressing cells are found throughout the developing spinal cord of control mice (Somogyi et al. 1995). Importantly, Gad67 expression was reduced in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8A-D), i.e. there was a 16% decrease in the number of Gad67-expressing cells in comparison to WT mice (Fig. 8I; P<0.001). This may reflect an abnormal development or survival of GABAergic neurons, which in consequence coud lead to abnormal control of neuronal network in dorsal horn. This finding was strengthened by analysis of Pax2, another gene expressed in GABAergic cells in the spinal cord (Cheng et al. 2004), with cell countings corroborating the decrease in the number of GABAergic cells (Fig. 8E,F,I; P<0,05). Interestingly, when Gbx1-/- pups were checked daily around weaning, the locomotor deficits were first observed around post-natal days (P)16-17, which coincides with the time point at which the GABA pathway switches from excitatory to inhibitory (Ben-Ari et al. 2007). We also observed that Gad67 expression was unchanged in the brain stem and cerebellum of E18.5 Gbx1-/- mutants (Fig. S3), arguing against an involvement of these structures in the observed phenotype. We next addressed the question if the observed decrease of GABAergic cells is due to neuronal cell death or to a possible change of GABAergic to glutamatergic fate. TUNEL experiments performed at various stages (E12.5 to E18.5) showed no abnormal apoptosis in the spinal cord of Gbx1-/- mice (data not shown). We then analyzed expression of Slc17a6, encoding VGLUT2, a vesicular glutamate transporter expressed in glutamatergic neurons (Kaneko et al. 2002). At E18.5, Slc17a6-expressing cells were increased in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8G,H; P<0.01). This finding suggests that part of the \"missing\" GABAergic cells may have differentiated into glutamatergic neurons. Glutamate and GABA are the predominant neurotransmitters for excitatory and inhibitory neurons, respectively, in the vertebrate brain. These two neurotransmitters are typically expressed in a mutually exclusive manner (Bellocchio et al. 2000; Fremeau et al. 2001), thereby defining the major functional subdivision in neuronal cell types. In the dorsal horn of the spinal cord\u2014the major relay center for processing somatosensory information\u2014, most ascending projection neurons and a subset of local circuit interneurons are excitatory and use glutamate as their transmitter. These neurons are modulated by local inhibitory neurons, many of which are GABAergic (for reviews: Melzack & Wall 1965; Malcangio & Bowery 1996; Dickenson 2002). Thus, GABA may inhibit transmitter release from primary afferent fibers. The output neurons of the dorsal horn are projection neurons, which are concentrated in lamina I and scattered throughout laminae III/VI, and relay sensory information to several brain areas. The vast majority of neurons in the dorsal horn are local circuit interneurons that do not project outside of the spinal cord. The output of projection neurons is influenced by local excitatory and inhibitory neurons (Todd 2010; Larsson & Broman 2011; Guo et al. 2012). The balance between excitation and inhibition is crucial for maintaining normal sensory function (Basbaum et al. 2009; Costigan, Scholz & Woolf 2009; Ross et al. 2010; Takazawa & 1111 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants MacDermott 2010). In Gbx1 mutants, the reduction of GABAergic neurons, and the possible switch of some of these neurons to a glutamatergic identity, may disrupt neuronal circuitry, becoming phenotypically apparent at adult stages as measured by abnormal performance in several behavioral tests. Further electrophysiological studies will be necessary to link the decrease of GABAergic neurons to the abnormal gait observed in adult Gbx1 mutant mice. conclusion : We have generated Gbx1-/- loss of function mutant mice, and investigated the development of the spinal cord dorsal horn in these mutants. Gbx1-/- mutants are viable and fertile, but display an altered gait during forward movement that specifically affects hindlimbs, beginning at post-natal days 16-17. This abnormal gait, documented by a series of bevioral tests, is not due to deficits in muscle strength or motor coordination. Our marker analysis in mutant embryos does not reveal abnormal assembly of the proprioceptive sensory afferents. However, we cannot exclude that the proprioceptive defects may be caused by abnormal development of superficial dorsal horn (see below), the transition and/or adjacent region for proprioceptive afferents, or may be secondary to a defect in sensory pathways as mutant mice also display a significantly reduced response time in the hot plate (thermosensory) test. Some of the deficits, including for example altered sensory nerve conduction velocity, are significantly altered in females, whereas significant difference in hot plate performance was identified only in males. Although such differences could reflect sexual dimorphism, it is difficult to draw such a conclusion as definitive for two major reasons: (i) in some tests where a significant difference in performance was observed for one gender, the opposite gender may display a similar tendency, which remained non statistically significant; (ii) if for example females would be more prone to effects of Gbx1 mutation we could expect to find them less performant in different tests; however, the gender effects were inconsistent and concerned males of females depending on the measured parameter. The spinal cord dorsal horn largely consists of inhibitory (GABAergic) and excitatory (glutamatergic) neurons that modulate somatosensory inputs from the periphery, including pain, temperature and mechanoception. Our analysis of major neuronal classes revealed a reduced number of GABAergic inhibitory interneurons expressing Gad67 in the superficial dorsal horn of Gbx1-/- mice. Gbx1 may therefore be functionally required for the differentiation of local inhibitory interneurons in the dorsal horn, corroborating a previous report of Gbx1 expression in a specific subset of GABAergic neurons in this region of the spinal cord (John, Wildner & Britsch 2005). Furthermore, our findings suggest that Gbx1 functions as a gene that promotes GABAergic over glutamatergic differentiation in the dorsal horn. A disruption in the balance between inhibitory and excitatory neuronal activity could explain the phenotype observed in Gbx1 mutants. Indeed, the imbalance of inhibitory and excitatory activity may lead to altered signaling to second-order neurons in the intermediate zone, which through an excitatory polysynaptic chain excite motor neurons in ventral horn to initiate protective movements. Such abnormal sensory processing is suggested at least for thermal stimuli, as Gbx1-/- mice displayed significantly increased response time in the hot plate test (thermosensory functions). Finally, considering that locomotor deficits become apparent at P16-17, we cannot exclude that abnormal gait may result from postnatal developmental or neurodegenerative events, which would need to be investigated. Despite the clear behavioral phenotype and reduced pool of GABAergic neurons in the dorsal horn, we did not observe any change in the expression of homeodomain factors involved in dorsal spinal cord patterning, or markers for primary sensory afferents, indicating that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutants. An explanation for these results\u2014and for the overall mild phenotype of the mutants\u2014is that Gbx1 and Gbx2 are coexpressed in dorsal spinal cord cells at early stages of embryogenesis (up to E12.5): hence the presence of Gbx2 might compensate for Gbx1 loss of function with respect to early regulatory events. Generation of Gbx1;Gbx2 double mutants will be required to assess possible redundant functions, and the availability of a Gbx2 floxed (conditional) allele does allow strategies for a spinal cord-specific inactivation, which would alleviate the lethality of the Gbx2 null mutants (Wassarman et al. 1997). Despite the importance of dorsal spinal cord in normal sensory processing, our knowledge concerning the establishment of neuronal circuits remains limited (Graham, Brichta & Callister 2007; Todd 2010). In this regard, our work contributes to understand how transcription factors cooperate for regulating cell specification and eventual distribution of neuronal subtypes in the developing spinal cord, providing clues for further dissecting functional circuitry of the dorsal spinal cord. 1212 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants figure legends : Figure 1. Inactivation of the mouse Gbx1 gene by homologous recombination. (A) The upper drawing shows the restriction map of the wild-type locus, boxes and lines corresponding to exons and introns, respectively. The homeodomain sequence is in red. In the targeting vector (below), two Gbx1 genomic fragments (between the dashed lines) flank a lacZ reporter gene and the neomycin resistance gene (grey box), transcribed in the same orientation (thin arrow) as Gbx1. In the recombined locus (lower drawing), 313 bp of Gbx1 exon 2 (including the homeodomain) are replaced by the lacZ neo sequence. The location of the 3' probe used for Southern blot analysis of ES cells is indicated in blue, and the PCR primers used to distinguish wild-type and recombined alleles for genotyping of animals (F1, R1, LacZ R2; see Materials and methods) are also indicated. (B) Southern blot analysis of a targeted cell line (+/-) in comparison to wild-type (+/+) HM-1 ES cells, using a probe external to the targeting vector 3' homology arm. (C) Genotyping of wild-type (+/+), heterozygous (+/\u2212 ) or homozygous mutant (-/-) mice by PCR amplification of fragments specific for the wild-type (354 bp) or mutated allele (269 bp), using the F1, R1 and LacZ R2 primers. (D,E) Anti-Gbx1 immunostaining. At E18.5, Gbx1 protein is absent in the spinal cord of Gbx1-/- mice (E), compared to wild- type (D). Scale bars: 100 \u00b5m. Figure 2. Abnormal phenotype of a Gbx1-/- mouse when walking. Sequential pictures compare the normal gait of a wild-type mouse (A) and the abnormal gait (\"duck-like\" walk) of a Gbx1-/- mutant when walking (B). A movie of these mice is available (Supplementary Material). Figure 3. Effects of Gbx1 mutation on the latency and number of slips in the beam walking test. ** p<0.01 vs WT; Student t-test. Figure 4. Open field performance of wild-type (WT) and Gbx1-/- mice. The distance traveled over the 20 min period of test reflects locomotor activity. The average speed was calculated during movement in the whole arena for the entire period of testing. Exploration of the central part of the open fied is expressed as the number of entries and percentage of time spent in the center. * p<0.05 and ** p<0.01 vs WT; Student t-test. Figure 5. Startle reactivity and pre-pulse inhibition in wild-type (WT) and Gbx1-/- mice. Startle reactivity to background noise (65 dB), or to 70, 80, 85, 90 dB acoustic stimulation, and startle reflex to a 110 dB stimulus, are presented. The percentage of pre-pulse inhibition of the startle response is displayed as a percentage of the pre-pulse intensity. BN, white noise; P, acoustic pulse intensity; ST, acoustic startle to 110 dB; PP, pre-pulse intensity. Figure 6. Absence of morphological and molecular abnormalities in the developing dorsal horn of Gbx1-/- mice. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/- (B,D,F,H) mice at E16.5 are shown. All sections are at the lumbar level. (A,B) Nissl-stained sections. No differences are detectable between wild-type and mutants. In situ hybridizations for two transcription factor encoding genes, Lbx1 (C,D) and Lmx1b (E,F), and for the axon guidance molecule netrin-1 (G,H), are shown. No differences are observed between wild-type and mutants. Scale bars: 100 \u00b5m. Figure 7. Developmental progression of afferent projections in the dorsal horn of Gbx1-/- mice. (A,B) Anti-calbindin-D28K antibody staining. At E18.5, calbindin fibers have already entered the spinal gray matter in wild-type embryos (A, arrow). Homozygous Gbx1-/- specimens (B) are indistinguishable from wild-types. (C,D) Expression of Drg11 in wild-type (C) and Gbx1 mutant (D) mice. Mutant specimens were indistinguishable from wild-types. (E,F) Anti-peripherin antibody staining. This staining reveals similar ingrowth of group IA muscle sensory afferents that grow to the ventral spinal cord (arrows) in wild-type (E) and mutant (F). Scale bars: 100 \u00b5m. Figure 8. Abnormal GABAergic differentiation in Gbx1-/- mice. Expression of Gad67 in wild-type (A,C) and Gbx1-/- (B,D) mice at E18.5. Higher magnification views (C,D; areas boxed in A,B) show the areas used for cell countings. Expression of Pax2 in wild-type (E) and Gbx1-/- (F) mice at E18.5. Expression of Slc17a6 in wild-type (G) and Gbx1-/- (H) mice at E18.5. (I) Countings revealed that the numbers of Gad67+ cells are diminished by 16% in Gbx1-/- mice (51.16\u00b12.25% Gad67+ cells in WT; 35.07\u00b11.81% in Gbx1-/- mice; *** p<0,001; 1414 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants Student t-test). Also, the numbers of Pax2+ cells are diminished by 14.69% in Gbx1-/- mice (57.60\u00b13.42% Pax2+ cells in WT; 42.91\u00b13.42% in Gbx1-/- mice; * p<0,05; Student t-test). In contrast, countings revealed that the numbers of Slc17a6+ cells are increased by 14,42% in Gbx1-/- mice (51.10\u00b11.69% Slc17a6+ cells in WT; 65.53\u00b14.64% in Gbx1-/- mice; ** p<0,01; Student t-test). Scale bars: 100 \u00b5m. in situ hybridization : In situ hybridization was performed with digoxigenin-labeled probes as previously described (Chotteau-Leli\u00e8vre et al. 2006). Template DNAs were kindly provided by Drs K. Jagla (Lbx1), C. Birchmeier (Lmx1b), M. Tessier-Lavigne (Netrin), A.J. Tobin (Gad67) and P. Gruss (Pax2), P. Bouillet (Gbx2), B. Giros (Slc17a6), F. Chen (Drg11), R. Krumlauf (Hoxb1), and F. Rijli (Hoxa2). For all experiments 3 animals of each genotype, from 2 or more independent litters, were analyzed. Cell countings were performed, for each animal, on 3 transverse sections at comparable levels of the lumbar spinal cord (all sections were collected serially, with section planes being separated by 112 \u00b5m). Three animals of each genotype were thus analyzed for each marker. All expression patterns were documented using a macroscope (Leica M420) or microscope (DM4000B), both connected to a Photometrics camera with the CoolSNAP (v. 1.2) imaging software (Roger Scientific, Chicago, IL). immunohistochemistry : After antigen unmasking in citrate 0.1 M (pH 6) during 15 min in a microwave oven, sections were treated in H 202 3% for 5 min, rinsed in PBS1X then blocked in PBS containing 0.25% Triton-X100, 5% normal goat serum and incubated overnight at 4\u00b0C with rabbit anti-Gbx1 (kindly provided by Dr S. Britsch ; 1:500), rabbit anti-calbindin D-28K (Chemicon, 1:1000), rabbit anti-Peripherin (Chemicon, 1:500), and mouse anti-Islet1 (40.2D6, Developmental Studies Hybridoma Bank, Iowa City, IA, 1:100) followed by species-specific biotin-coupled secondary antibodies (Jackson Laboratories). Detection was performed using Vectastain Elite ABC Kit following manufacturer instructions. Nissl staining was performed by incubation in 0.5% cresyl violet for 15 min. TUNEL was performed using the APOPTAG\u00ae Peroxidase In Situ Apoptosis detection kit (Millipore). For all experiments 3 animals of each genotype were analyzed. behavioral phenotyping procedures : Cohorts of 10 week-old male and female Gbx1-/- mice in a C57BL/6J genetic background (7 males and 8 females), with their wild-type (WT, 10 males and 9 females) counterparts, were used in this study. Mice were group housed and allowed 2 weeks acclimation in the phenotyping area with controlled temperature (21-22\u00b0C) under a 12-12 h light-dark cycle (lights on 7am-7pm), with food and water available ad libitum. Testing started at 10 weeks of age, and all procedures were carried out in accordance with European institutional guidelines. Behavioral tests were performed successively for each cohort of mice, during the light phase of the circadian cycle, according to a pipeline established by the European Mouse Disease Clinic (EUMODIC pipeline 2). Detailed procedures for each test are available at the URL: http://www.empress.har.mrc.ac.uk/viewempress/index.php?pipeline=EUMODIC+Pipeline+2. Neurological examination: General health and basic sensory motor functions were evaluated using a modified SHIRPA protocol (Brown, Chambon & Hrab\u00e9 de Angelis 2005; protocol at http://www.empress.har.mrc.ac.uk/viewempress/index.php?pipelineprocedure=EUMODIC+Pipeline+2~Modified+SHIRPA). This analysis is adapted from the procedure developed by Irwin (1968) and from the SHIRPA protocol (Hatcher et al. 2001). It provides an overview of physical appearance, body weight, neurological reflexes and sensory abilities. Rotarod test: This test evaluates motor coordination and balance by measuring the ability of animals to maintain balance on a rotating rod (Bioseb, Chaville, France). Mice were given three testing trials during which the rotation speed accelerated from 4 to 40 rounds per min (rpm) in 5 min. Trials were separated by 5-10 min intervals. The average latency of the three trials was used as index of motor coordination performance. 66 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants Grip test: This test measures the maximal muscle strength (g) using an isometric dynamometer connected to a grid (Bioseb). Mice were allowed to grip the grid either with the forepaws or with both the forepaws and hindpaws, then were pulled backwards until they released the grid. Each mouse was submitted to 3 consecutive trials immediately after the modified SHIRPA procedure. The maximal strength developed by the mouse before releasing the grid was recorded and the average value of the three trials was adjusted to body weight. Beam walking: This test is used to evaluate fine motor coordination and proprioceptive function. The apparatus used is a 2 cm diameter and 110 cm long wooden beam, elevated 50 cm above the ground. A goal box (12 x 12 x 14 cm) is attached at one extremity of the beam. Animals were first habituated to the goal box for 1 min. They were then submitted to 3 training trials during which they were placed at different points of the beam, with the head directed to the goal box, and allowed to walk the corresponding distance to enter the goal box. After training, animals were submitted to 3 testing trials during which they were placed at the extremity of the beam opposite to the goal box and allowed to walk the beam distance and enter the goal box. The latency to enter the goal box and the number of slips (when one or both hindpaws slipped laterally from the beam) were measured. Hot plate test: The mice were placed into a glass cylinder on a hot plate (Bioseb) adjusted to 52\u00b0C, and the latency of the first pain reaction of any hindlimb (licking, flinches) was recorded, with a maximum of 30 s testing. Electrophysiological measurements: Electrophysiological recordings were performed under ketamine-xylazine anesthesia using a Key Point electromyograph apparatus (Metronic, France). The body temperature was maintained at 37\u00b0C with a homeothermic blanket (Harvard, Paris, France). For measuring the sensory nerve conduction velocity (SNCV), recording electrodes were inserted at the proximal part of the tail and stimulating electrodes placed 20 mm from the recording needles towards the extremity of the tail. A ground needle electrode was inserted between the stimulating and recording electrodes. Caudal nerve was stimulated with a series of 20 pulses of 0.2 ms duration at a supramaximal intensity of 8 mA. The average response is included for statistical analysis. The compound muscle action potential (CMAP) was measured in gastrocnemius muscle after sciatic nerve stimulation. For this purpose, stimulating electrodes were placed at the level of the sciatic nerve and recording electrodes placed in the gastrocnemius muscle. A ground needle was inserted in the contralateral paw. The sciatic nerve was stimulated with a single 0.2 ms pulse at a supramaximal intensity of 8 mA. The amplitude (mV) and the distal latency of the responses (ms) were measured. Anxiety-related behavior - open field test: Mice were tested in automated open fields (Panlab, Barcelona, Spain), each virtually divided into central and peripheral regions. The open fields were placed in a room homogeneously illuminated at 150 Lux. Each mouse was placed in the periphery of the open field and allowed to explore freely the apparatus for 20 min, with the experimenter out of the animal\u2019s sight. The distance traveled, the number of rears, and time spent in the central and peripheral regions were recorded over the test session. The latency and number of crosses into as well as the percent time spent in center area are used as index of emotionality/anxiety. Sensorimotor gating - auditory startle reflex reactivity and pre-pulse inhibition (PPI): Acoustic startle reactivity and pre-pulse inhibition of startle were assessed in a single session using standard startle chambers (SR-Lab Startle Response System, San Diego Instruments). Ten different trial types were used: acoustic startle pulse alone (110 db), eight different prepulse trials in which either 70, 75, 85 or 90 dB stimuli were presented alone or preceding the pulse, and finally one trial (NOSTIM) in which only the background noise (65 dB) was presented to measure the baseline movement in the Plexiglas cylinder. In the startle pulse or prepulse alone trials, the startle reactivity was analyzed, and in the prepulse plus startle trials the amount of PPI was measured and expressed as percentage of the basal startle response. Statistical analyses: Data were analyzed using unpaired Student t-test, one way or repeated measures analysis of variance (ANOVA) with one between factor (genotype) and one within factor (time). Qualitative parameters (i.e. some of the clinical observations) were analyzed using \u03c72 test. The level of significance was set at p<0.05. Animal ethics statement 77 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants Animal experimentation protocols were reviewed and approved by the Direction D\u00e9partementale des Services V\u00e9t\u00e9rinaires (agreement #67-172 to H.M., 67-189 to P.D., and institutional agreement #D67-218-5 for animal housing) and conformed to the NIH and European Union guidelines, provisions of the Guide for the Care and Use of Laboratory Animals, and the Animal Welfare Act. acknowledgments : We thank B. Schuhbaur for excellent technical assistance. We are grateful to Drs. C. Birchmeier, S. Britsch, F. Chen, K. Jagla, P. Bouillet, B. Giros, P. Gruss, M. Tessier-Lavigne, R. Krumlauf and F. Rijli for the gift of reagents. funding statement : This work was supported by grants from the Agence Nationale de la Recherche (ANR Neurosciences 2007, ANR Blanc 2011), the Fondation pour la Recherche M\u00e9dicale (Equipe FRM 2007), the Deutsche Forschungsgemeinschaft (SFB-655 A3-Brand), the Italian Association for Cancer Research (AIRC), and by institutional funding from the Centre National de la Recherche Scientifique (CNRS), Institut National de la Sant\u00e9 et de la Recherche M\u00e9dicale (INSERM), and University of Strasbourg. Behavioral phenotyping was partly subsidized by the EUMODIC European Consortium and the Mouse Clinical Institute (MCI/ICS, Strasbourg). 1313 Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants supplementary figure legends : Figure S1. Expression analysis of Gbx2 in the developing spinal cord of Gbx1 mutants. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/(B,D,F,H) mice are shown. All sections are at the lumbar level. In situ hybridizations for Gbx2 were performed at different developmental stages: E12.5 (A,B), E14.5 (C,D), E16.5 (E,F), and E18.5 (G,H). Scale bars: 100 \u00b5m. Figure S2. Analysis of rhombomeric markers in Gbx1-/- embryos. Whole-mount in situ hybridizations of E9.5 embryos with 2 markers of prospective rhombomeres: Hoxb1, which labels rhombomere 4 (A,B), and Hoxa2, which marks rhombomeres 2 to 6 and associated neural crest (C,D). Scale bars: 50 \u00b5m. Figure S3. In situ hybridization analysis of Gad67-expressing cells in the prenatal hindbrain. Sections are shown at various levels of the brain stem (A-F) and cerebellum (G,H) of wild-type (A,C,E,G) and Gbx1-/- (B,D,F,H) mice at E18.5. Scale bars: 100 \u00b5m. Figure S4. Analysis of developing spinal cord motor neurons in Gbx1 mutants. Expression of Islet1 in the lumbar spinal cord of wild-type (A,C) and Gbx1-/- (B,D) mice at E18.5 (A,B) and E14.5 (C,D). (E) Countings revealed that the numbers of Islet1+ cells are not significantly diminished in Gbx1-/- mice (at E18.5: 22.83\u00b16.165 Islet1+ cells in WT; 22.59\u00b17.33 in Gbx1-/- mice; NS; at E14.5: 73.91\u00b10.8 Islet1+ cells in WT; 73.99\u00b16.36 in Gbx1-/- mice; NS; Student t-test). Scale bars: 100 \u00b5m. Movie sequence showing a Gbx1+/+ and a Gbx1-/- mouse. table captions : Table 1. Effects of Gbx1 mutation on body weight, basic neurological reflexes, specific motor abilities and pain sensitivity. Statistically different parameters in wild-type vs mutants appear in bold. * p<0.05 and **p<0.01 vs wild-type; Student t-test. Table 2. Effects of Gbx1 mutation on sensory nerve conduction velocity. The sensory nerve conduction velocity was measured at the level of the caudal nerve. The latency and the amplitude of gastrocnemius muscle response evoked by sciatic nerve stimulation were also recorded. * p<0.05 vs wild-type; Student t-test. grip strength : (adjusted to body weight) 2-paws 3.97 \u00b1 0.18 4.00 \u00b1 0.24 3.75 \u00b1 0.17 3.38 \u00b1 0.26 4-paws 8.30 \u00b1 0.24 7.72 \u00b1 0.37 7.00 \u00b1 0.24 7.22 \u00b1 0.54 Hot plate 13.43 \u00b1 1.26 17.13 \u00b1 1.08* 12.72 \u00b1 1.15 15.03 \u00b1 1.63 Males Females Wi d-type Gbx1-/Wild-type Gbx1-/- Pre Prin ts Pre Prin ts Table 2: Sensory Nerve Conduction Velocity 70.33 \u00b1 1.70 72.20 \u00b1 1.53 63.93 \u00b1 2.35 71.81 \u00b1 1.50* Gastrocnemius M-wave Latency 0.93 \u00b1 0.06 0.91 \u00b1 0.06 0.99 \u00b1 0.08 0.84 \u00b1 0.05 amplitude : 43 \u00b1 3.15 : 44.60 \u00b1 5.87 46.46 \u00b1 6.70 55.41 \u00b1 6.03 Males Females Wi d-type Gbx1-/Wild-type Gbx1-/- Pre Prin ts Pre Prin ts",
    "v5_text": "abstract : Dorsal horn neurons in the spinal cord integrate and relay sensory information to higher brain centers. These neurons are organized in specific laminae and different transcription factors are involved in their specification. The murine homeodomain Gbx1 protein is expressed in the mantle zone of the spinal cord at E12.5-13.5, correlating with the appearance of a discernable dorsal horn around E14 and eventually defining a narrow layer in the dorsal horn around perinatal stages. At postnatal stages, Gbx1 identifies a specific subpopulation of GABAergic neurons in the dorsal spinal cord. We have generated a loss of function mutation for Gbx1 and analyzed its consequences during spinal cord development. Gbx1-/- mice are viable and can reproduce as homozygous null mutants. However, the adult mutant mice display an altered gait during forward movement, that specifically affects the hindlimbs. This abnormal gait was evaluated by a series of behavioral tests, indicating that locomotion is impaired, but not muscle strength or motor coordination. Molecular analysis showed that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutant mice. However, analysis of terminal neuronal differentiation revealed that the number of GABAergic inhibitory interneurons in the superficial dorsal horn is diminished. Our study unveiled a role for Gbx1 in specifying a subset of GABAergic neurons in the dorsal horn of the spinal cord involved in the control of posterior limb movement. Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 3 materials and methods : Construction of a Gbx1 targeting vector Genomic sequences encompassing the mouse Gbx1 gene were isolated from a 129SV genomic phage library, using as a probe a Gbx1 cDNA fragment previously characterized (Rhinn et al. 2004). A Gbx1 loss of function mutation was produced by homologous recombination in embryonic stem cells (Ram\u00ecrez-Solis, Davis & Bradley 1993). The targeting vector contained a 5.4 kb XmnI fragment (upstream arm), ending 33 bp upstream of the homeodomain sequence located in Gbx1 second exon, and a 1.6 kb KpnI fragment (downstream arm), whose sequence started 91 bp downstream from the homeodomain. These fragments were excised from the recombinant phage and cloned in the mutagenesis pGN vector (Le Mouellic, Lallemand & Br\u00fblet 1990) to generate the pGN-Gbx1 targeting vector (Fig. 1A). In this vector, the fragments are inserted on each side of a lacZ reporter gene and a neomycin resistance gene, and their insertion by homologous recombinaton in the Gbx1 gene will generate a 313 bp deletion encompassing the entire homeodomain (Fig. 1A). Transfection of embryonic stem cells and selection of targeted clones HM-1 embryonic stem (ES) cells (Magin, Whir & Melton 1992) were cultured on neomycinresistant mouse embryonic fibroblasts, as described in Robertson, 1987. Ten \u03bcg of the pGNGbx1 targeting vector were linearized by digestion of the unique NotI restriction site, and electroporated into 2\u00d7107 ES cells resuspended in 750 \u03bcl HeBS medium (20 mM Hepes pH 7.05, 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, 6 mM glucose), at 200 V, 960 \u03bcF. Positive selection was carried out for 11 days with 350 \u03bcg/ml G418. Resistant colonies were Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 6 picked and DNA was extracted from a fraction (1/5) of the cells to perform Southern blot analysis to identify homologous recombination events. The probe used is an external fragment located immediately downstream to the targeting vector (Fig. 1A,B). Positive clones were expanded before freezing. The frequency of homologous recombination was 7 out of 350 clones analyzed. Generation and genotyping of chimeric and mutant mice After thawing, 10 to 15 ES cells were microinjected into C57BL/6 blastocysts. Injected blastocysts were reimplanted in the uterine horn of pseudopregnant recipient females. Chimeric animals were back-crossed to C57BL/6J mice and germ-line transmission was scored by the presence of agouti coat pigmentation. Heterozygous offspring were identified by PCR genotyping. Tail tips were incubated in lysis buffer (50 mM Tris pH 8.0, 100 mM EDTA, 100 mM NaCl, 1% SDS, 0.6 mg/ml proteinase K) overnight at 55\u00b0C, phenolchloroform extracted, ethanol precipitated and redissolved in 10 mM Tris-HCl, 1 mM EDTA pH 8.0 at a final concentration of 0.2-1.0 \u03bcg/\u03bcl. The presence of a wild-type or mutated allele was detected using three primers: a sense primer F1: 5\u2032- GGTGACAGCGAGGACAGCTTCCT-3\u2032, an antisense primer R1: 5\u2032- CCCAGAACGACTGCTCACATTGC -3\u2032 and an antisense primer LacZ R2: 5\u2019- GGCCTCTTCGCTATTACGCCA-3\u2019. The presence of a wild-type allele was detected using the F1/R1 primers which amplify a 354 bp fragment. The presence of a mutated allele was detected by using the F1/LacZ R2 primers which amplify a 269 bp fragment. Thirty cycles (denaturation : 1 min, 95\u00b0C, annealing : 1 min, 62\u00b0C; elongation : 30 s, 74\u00b0C) were performed, and the amplified products were separated by 2% agarose gel electrophoresis (Fig. 1C). Phenotypic and molecular analyses were performed after several generations of backcrosses (>5) to C57BL/6J mice, resulting in a nearly pure genetic background. Animal experimentation protocols were reviewed and approved by the Direction D\u00e9partementale des Services V\u00e9t\u00e9rinaires (agreement #67-172 to H.M., 67-189 to P.D., and institutional agreement #D67-218-5 for animal housing) and conformed to the NIH and European Union guidelines, provisions of the Guide for the Care and Use of Laboratory Animals, and the Animal Welfare Act. Tissue collection and sample preparation Pregnant females obtained from natural matings (morning of vaginal plug was considered as Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 7 E0.5) were sacrificed and fetuses were collected in phosphate-buffered saline (PBS, pH 7.5) after cesarean section. The specimens were dissected, fixed overnight in 4% paraformaldehyde (PFA), cryoprotected in 20% sucrose, and embedded in Shandon Cryomatrix (Thermo Electron Corperation) before freezing at -80\u00b0C. Cryosections (14 \u00b5m thickness, Leica CM3050S cryostat) sections were performed according to a coronal plane, collected on Superfrost slides, and stored at -80\u00b0C until use. For whole-mount immunostaining or in situ hybridization, embryos were fixed overnight in 4% PFA, dehydrated, and stored at -20 \u00b0C in 100% methanol. in situ hybridization : In situ hybridization was performed with digoxigenin-labeled probes as previously described (Chotteau-Leli\u00e8vre et al. 2006). Template DNAs were kindly provided by Drs K. Jagla (Lbx1), C. Birchmeier (Lmx1b), M. Tessier-Lavigne (Netrin), A.J. Tobin (Gad67) and F. Chen (Drg11). For cell countings, 9 sections were counted from each animal, and 3 animals of each genotype were analyzed. All expression patterns were documented using a macroscope (Leica M420) or microscope (DM4000B), both connected to a Photometrics camera with the CoolSNAP (v. 1.2) imaging software (Roger Scientific, Chicago, IL). immunohistochemistry : After antigen unmasking in citrate 0.1 M (pH 6) during 15 min in a microwave oven, sections were treated in H202 3% for 5 min, rinsed in PBS1X then blocked in PBS containing 0.25% Triton-X100, 5% normal goat serum and incubated overnight at 4\u00b0C with rabbit anti-Gbx1 (kindly provided by Dr S. Britsch ; 1:500), rabbit anti-calbindin D-28K (Chemicon, 1:1000), or rabbit anti-peripherin (Chemicon, 1:500) followed by species-specific biotin-coupled secondary antibodies (Jackson Laboratories). Detection was performed using Vectastain Elite ABC Kit following manufacturer instructions. Nissl staining was performed by incubation in 0.5% cresyl violet for 15 min. behavioral phenotyping procedures : Cohorts of 10 week-old male and female Gbx1-/- mice in a C57BL/6J genetic background (n=7-10), with their wild-type (WT) counterparts, were used in this study. Mice were group housed and allowed 2 weeks acclimation in the phenotyping area with controlled temperature Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 8 (21-22\u00b0C) under a 12-12 h light-dark cycle (lights on 7am-7pm), with food and water available ad libitum. Testing started at 10 weeks of age, and all procedures were carried out in accordance with European institutional guidelines. Detailed procedures are available on the EUMODIC/EMPREeSS website (http://www.empress.har.mrc.ac.uk/) Neurological examination: General health and basic sensory motor functions were evaluated using a modified SHIRPA protocol (Brown, Chambon & Hrab\u00e9 de Angelis 2005; protocol at http://www.empress.har.mrc.ac.uk/viewempress/index.php?pipelineprocedure=EUMODIC+P ipeline+2~Modified+SHIRPA). This analysis is adapted from the procedure developed by Irwin (1968) and from the SHIRPA protocol (Hatcher et al. 2001). It provides an overview of physical appearance, body weight, neurological reflexes and sensory abilities. Rotarod test: This test evaluates motor coordination and balance by measuring the ability of animals to maintain balance on a rotating rod (Bioseb, Chaville, France). Mice were given three testing trials during which the rotation speed accelerated from 4 to 40 rounds per min (rpm) in 5 min. Trials were separated by 5-10 min intervals. The average latency of the three trials was used as index of motor coordination performance. Grip test: This test measures the maximal muscle strength (g) using an isometric dynamometer connected to a grid (Bioseb). Mice were allowed to grip the grid either with the forepaws or with both the forepaws and hindpaws, then were pulled backwards until they released the grid. Each mouse was submitted to 3 consecutive trials immediately after the modified SHIRPA procedure. The maximal strength developed by the mouse before releasing the grid was recorded and the average value of the three trials was adjusted to body weight. Beam walking: This test is used to evaluate fine motor coordination and proprioceptive function. The apparatus used is a 2 cm diameter and 110 cm long wooden beam, elevated 50 cm above the ground. A goal box (12 x 12 x 14 cm) is attached at one extremity of the beam. Animals were first habituated to the goal box for 1 min. They were then submitted to 3 training trials during which they were placed at different points of the beam, with the head directed to the goal box, and allowed to walk the corresponding distance to enter the goal box. After training, animals were submitted to 3 testing trials during which they were placed at the extremity of the beam opposite to the goal box and allowed to walk the beam distance and enter the goal box. The latency to enter the goal box and the number of slips (when one or both hindpaws slipped laterally from the beam) were measured. Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 9 Hot plate test: The mice were placed into a glass cylinder on a hot plate (Bioseb) adjusted to 52\u00b0C, and the latency of the first pain reaction of any hindlimb (licking, flinches) was recorded, with a maximum of 30 s testing. Electromyography: Electrophysiological recordings were performed under ketamine-xylazine anesthesia using a Key Point electromyograph apparatus (Metronic, France). The body temperature was maintained at 37\u00b0C with a homeothermic blanket (Harvard, Paris, France). For measuring the sensitive nerve conduction velocity (SNCV), recording electrodes were inserted at the base of the tail and stimulating electrodes placed 20 mm from the recording needles towards the extremity of the tail. A ground needle electrode was inserted between the stimulating and recording electrodes. Caudal nerve was stimulated with a series of 20 pulses of 0.2 ms duration at a supramaximal intensity of 8 mA. The average response is included for statistical analysis. The compound muscle action potential (CMAP) was measured in gastrocnemius muscle after sciatic nerve stimulation. For this purpose, stimulating electrodes were placed at the level of the sciatic nerve and recording electrodes placed in the gastrocnemius muscle. A ground needle was inserted in the contralateral paw. The sciatic nerve was stimulated with a single 0.2 ms pulse at a supramaximal intensity of 8 mA. The amplitude (mV) and the distal latency of the responses (ms) were measured. Anxiety-related behavior - open field test: Mice were tested in automated open fields (Panlab, Barcelona, Spain), each virtually divided into central and peripheral regions. The open fields were placed in a room homogeneously illuminated at 150 Lux. Each mouse was placed in the periphery of the open field and allowed to explore freely the apparatus for 20 min, with the experimenter out of the animal\u2019s sight. The distance traveled, the number of rears, and time spent in the central and peripheral regions were recorded over the test session. The latency and number of crosses into as well as the percent time spent in center area are used as index of emotionality/anxiety. Sensorimotor gating - auditory startle reflex reactivity and pre-pulse inhibition (PPI): Acoustic startle reactivity and pre-pulse inhibition of startle were assessed in a single session using standard startle chambers (SR-Lab Startle Response System, San Diego Instruments). Ten different trial types were used: acoustic startle pulse alone (110 db), eight different prepulse trials in which either 70, 75, 85 or 90 dB stimuli were presented alone or preceding the pulse, and finally one trial (NOSTIM) in which only the background noise (65 dB) was presented to measure the baseline movement in the Plexiglas cylinder. In the startle pulse or Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 10 prepulse alone trials, the startle reactivity was analyzed, and in the prepulse plus startle trials the amount of PPI was measured and expressed as percentage of the basal startle response. Statistical analyses: Data were analyzed using unpaired Student t-test, one way or repeated measures analysis of variance (ANOVA) with one between factor (genotype) and one within factor (time). Qualitative parameters (i.e. some of the clinical observations) were analyzed using \u03c72 test. The level of significance was set at p<0.05. results and discussion : Gbx1-deficient mice are viable, but display a typical duck-like gait A loss of function allele for the Gbx1 gene was generated by homologous recombination in murine embryonic stem cells (see Materials and Methods). The mutated Gbx1 allele is devoid of the entire homeodomain-coding sequence and ~100 adjacent nucleotides (Fig. 1A-C). After generation of germ-line transmitting chimeras, heterozygous mutant mice (Gbx1+/-) were found to be viable, fertile and apparently normal. After intercrossing Gbx1+/- mice, Gbx1-/mutants (generated in a C57BL/6J genetic background) were born in the expected mendelian ratio. Immunohistochemistry performed with an anti-Gbx1 antibody confirmed the absence of detectable Gbx1 protein in the spinal cord of E18.5 Gbx1-/- mutants (Fig. 1D,E). When tested by 6 weeks of age, however, the mutants display a typical duck-like gait (Fig. 2 and Supplementary information, file S1). Both male and female Gbx1-/- mice were fertile and had a normal life span. General health and sensorimotor abilities in adult Gbx1 mutants Gbx1-/- males and females had a normal body weight (Table 1) and a normal overall physical appearance. However, many of the Gbx1-/- mutants showed significantly abnormal gait (\u03c72 \u22655.20, p<0.05). Indeed, 43% of Gbx1 mutant males and 63% of Gbx1 mutant females displayed lack of fluidity in movement (Table 1; Fig. 2; Supplementary information, file S1). Gbx1-/- males and females also showed significantly reduced short-term locomotor activity following immediate transfer for the modified SHIRPA test, as compared to WT counterparts (t\u22653.46, p<0.01) (Table 1). The other features of general health and basic neurological reflexes were not affected in Gbx1 mutants. When tested for specific motor abilities, motor coordination performance measured in the rotarod test (t\u22641.29, NS) and the muscle strength (grid grip) test (t\u22641.38, NS) were not Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 11 affected in Gbx1-/- males and females (Table 1). In the beam walking test, the latency to cross the beam was significantly increased (t15=3.71, p<0.01) and the number of slips was slightly increased (even if not significantly) in Gbx1-/- females as compared to WT counterparts (Fig. 3). Gbx1 mutant males also showed a tendency for higher latencies than WT, but the difference between genotypes was not statistically significant (t15=1.65, p=0.12) (Fig. 3). In the open field test, there was a significant effect of genotype concerning locomotor activity [F(1,30)=6.51, p<0.05], reflecting reduced locomotion in all Gbx1-/- animals. When considering each gender separately, both Gbx1-/- males and females tended to have reduced locomotor activity over the testing period (p=0.09) (Fig. 4). The average speed during motion was also significantly lower in Gbx1-/- males and females than in WT (t\u22653.36, p<0.01) (Fig. 4). The number of entries into, and the percentage of time spent in, the center of the arena also differed between genotypes [F(1,30)\u226514.48, p<0.001]. Both Gbx1-/- males and females had significantly decreased number of entries and spent less time in the center of the open field than WT counterparts (t\u22652.62, p<0.05) (Fig. 4), which might reflect increased anxiety in Gbx1-/- mutants. The reduced exploration of the center might also be due to the observed reduced locomotor activity of Gbx1-/- mutants. Altogether, these data show that Gbx1-/mutant mice have a clear defect in locomotion, although this defect does not appear to result in a coordination problem or a muscle strength deficiency. To test whether ablation of Gbx1 may affect sensory response, we measured the response of Gbx1 mutant mice in a hot plate test. The withdrawal latency was higher in Gbx1-/- males (but not in females) than in WT (t15=2.10, p=0.05) (Table 1), suggesting reduced thermal pain sensitivity in Gbx1 mutant males. The consequence of Gbx1 inactivation on acoustic startle and pre-pulse inhibition of startle reflex was also evaluated. Regardless of gender, the startle reactivity was comparable between WT and Gbx1-/- mice for all the acoustic stimuli including the startling pulse [Genotype F(1,30)\u22640.83, Sex F(1,30)\u22640.85, Genotype*Sex F(1,30)\u22641.11, NS] (Fig. 5). When the startling pulse was preceded with prepulses with lower intensities, the PPI level was also comparable between genotypes [Genotype F(1,30)=0.55, Sex F(1,30)=0.32, Genotype*Sex F(1,30)=0.11, NS)] (Fig. 5). Furthermore, electromyography (EMG) measurements revealed that the sensory nerve conduction velocity differed significantly between genotypes [F(1,29)=7.31, p<0.05]; indeed, Gbx1-/- females had significantly increased sensory nerve conduction velocity (t14=2.83, p<0.05) (Table 1), as measured at the level of the caudal nerve. Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 12 On the other hand, the latency and amplitude of the gastrocnemius muscle response evoked by sciatic nerve stimulation were comparable between genotypes [F(1,29)=1.63, NS]. In summary, we used a variety of behavioral and electrophysiological phenotyping tests to evaluate sensory and motor functions in Gbx1 mutant mice. We found that both Gbx1-/males and females show reduced locomotor activity in different situations. Decreased exploratory behavior was found in the open field test and following immediate transfer during clinical observations. Exploration of the central part of the open field arena was significantly decreased in Gbx1-/- males and females, which might suggest increased anxiety in mutants. However, this could also be explained by the reduced locomotor activity displayed by these mutants. Indeed, Gbx1-/- mice also showed decreased average speed with no significant effect on the distance travelled in the open field. Their strong altered gait during forward movement might explain the reduced speed and locomotor activity in the open field. This is supported by data from the beam walking test showing that at least Gbx1-/- females required longer time to cross the beam distance and tended to have higher number of slips, suggesting inappropriate rear paw placement, and supporting altered proprioceptive sensitivity. Likewise, Gbx1-/mutants displayed impaired rotarod behavior with even a stronger effect in females than in males. EMG measurements showed that Gbx1-/- females had increased sensory nerve conduction velocity, measured in the caudal nerve, supporting the altered sensory deficits observed in the behavioral tests. Our results also show that Gbx1 gene disruption affects exploratory behavior with increased anxiety and decreased locomotor performance, but without muscle strength defect nor changes in central sensory motor gating. Altogether, the behavioral data revealed that the Gbx1-/- animals have apparent proprioceptive defects and/or altered sensory abilities. We cannot exclude that the proprioceptive defects may be secondary to a defect in sensory pathways (for example, caused by pain on movement), because mutant mice also display a significantly reduced response time in the hot plate (thermosensory) test. Gbx1-/- mice do not show obvious hindbrain patterning defects Gbx genes are related to the Drosophila unplugged gene, which acts during development of the tracheal system, and perhaps for specification of neuroblast sublineages (Chiang et al. 1995; Cui & Doe 1995). There are two Gbx genes in amniote species (human, mouse and chicken), as well as in zebrafish (Lin et al. 1996; Bouillet et al. 1995; Shamim & Mason 1998; Niss & Leutz 1998; Rhinn et al. 2003). Previous studies showed that in mouse, Gbx2 is Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 13 involved in early specification of the midbrain-hindrain boundary (MHB) organizer, a signaling center that will pattern the anterior hindbrain rhombomeres (Wassarman et al. 1997; Waters & Lewandoski 2006; for review: Rhinn & Brand 2001; Simeone 2000). In zebrafish it was shown that gbx1 acts during early positioning of the MHB, whereas gbx2 functions at later stages, once the MHB is established (Rhinn et al. 2004; 2009 ; BurroughsGarcia et al. 2011). In mouse, Gbx1 is not expressed at the MHB as is the case during early zebrafish development. Its expression starts at E7.75 in the prospective hindbrain, spanning rhombomeres 2 to 7 during the segmentation phase (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2004). This suggested that Gbx1 might be involved in early embryonic hindbrain patterning, which could underlie behavioral deficits associated with loss of Gbx1 function. To assess for possible rhombomeric abnormalities in Gbx1-/- mutants, we performed wholemount in situ hybridizations at E9.5 with several markers including Hoxb1, Krox20, and Hoxa2. This analysis did not show any molecular or structural abnormality of the hindbrain rhombomeres in Gbx1-/- embryos (data not shown). This suggests that Gbx1 is not required for early hindbrain patterning, in contrast to its mouse homologue Gbx2 (Wassarman et al. 1997; Waters & Lewandoski 2006). At E12.5, the expression domains of Gbx1 and Gbx2 overlap, both being expressed in the ventricular zone and the mantle zone of the entire dorsal spinal cord (Rhinn et al. 2004; Waters, Wilson & Lewandoski 2004). After E12.5, however, Gbx2 expression is rapidly downregulated. Gbx1 and Gbx2 are thus transiently coexpressed in progenitor cells of the dorsal spinal cord, with Gbx1 being the only Gbx gene persistently expressed during later dorsal horn development. Development of the dorsal horn in Gbx1 mutant mice The prominent expression of Gbx1 in the dorsal horn could be relevant for the abnormal gait phenotype of adult Gbx1 mutant mice, which led us to ask whether Gbx1 is required for the maturation and/or specification of neurons of the dorsal horn during development. Nissl staining of E18.5 spinal cord sections revealed no obvious difference between the dorsal horn of wild-type and Gbx1-/- animals at thoraco-lumbar levels (Fig. 6A,B). Despite the clear behavioral phenotype, we were unable to identify any consistent alteration in the expression of several molecular markers of dorsal spinal cord cell populations in Gbx1-/- embryos. These markers included the genes encoding the transcription factors Lbx1 (Gross, Dottori & Goulding 2002; M\u00fcller et al. 2002) (Fig. 6C,D) and Lmx1b (Chen et al. 2001) (Fig. 6E,F), and the axon guidance molecule Netrin-1 (Leonardo et al. 1997) (Fig. 6G,H). Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 14 Projection pattern of primary sensory afferents in the dorsal horn of Gbx1-/- mutants Afferents sensing pain and temperature mainly project to laminae I/II. Afferents for sensing innocuous mechanoreceptor signals such as texture, shape, vibration, and pressure project predominantly to internal laminae (III, IV, V). Afferents for sensing proprioceptive signals project through the dorsal horn to the ventrally located motor neurons (Brown 1981; Willis & Coggeshall 1991). The formation of laminar-specific projections is a key event in the development of appropriate neuronal connections in many regions of the central nervous system. Collaterals from the different classes of sensory axons then penetrate the gray matter of the spinal cord sequentially. Each class of sensory axons projects directly to its target lamina and never branches into inappropriate laminae. Some cutaneous afferents traverse the entire width of the spinal cord to reach superficial laminae on the contralateral side, strictly avoiding both the ventral spinal cord and inappropriate laminae of the deep dorsal horn. We examined the development of primary sensory afferent projections to the dorsal horn, which are well defined at E18.5, in Gbx1 mutant mice (Ozaki & Snider 1997). The central projections of cutaneous nociceptive sensory neurons first arrive in the dorsal root entry zone and begin to invade the spinal grey matter by E12.5. Staining with an antibody to calbindin28K marks a subset of cutaneous neurons and their afferent fibers (Honda 1995). By E18.5, calbindin+ fibers have invaded the dorsal horn of wild-type and Gbx1 mutants (Fig.7A,B). Drg11 is required for the projection of cutaneous sensory afferent fibers to the dorsal spinal cord. Indeed, Drg11 mutant mice display abnormalities in the spatio-temporal patterning of cutaneous sensory afferent fiber projections to the dorsal, but not the ventral spinal cord, as well as defects in dorsal horn morphogenesis (Chen et al. 2001). In our Gbx1-/- mutant mice, Drg11 expression was not affected in the dorsal horn at E18.5 (Fig.7C,D). Altogether, these data suggest that there are no defects in patterning of sensory afferent fiber projections to the dorsal horn, which selectively affects cutaneous afferents. Because primary cutaneous sensory afferent projections mediate (among other modalities) nociception, these results imply that Gbx1 mutant mice might not exhibit deficiencies in their behavioral responses to noxious stimuli. Howewer, we observed by performing the hot plate test, that Gbx1-/- males show longer withdrawal latencies, suggesting a tendency to reduced thermal pain sensitivity. We further examined proprioceptive afferents at E18.5 by using antibodies to peripherin (Escurat et al. 1990). No differences between wild-type and Gbx1-/- mice were observed at Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 15 the level of proprioceptive fibers that extend toward motoneurons and interneurons in the deep dorsal horn, and at the level of fibers that enter into the spinal gray matter (arrows in Fig. 7E,F). Reduced GABAergic neuronal differentiation in Gbx1-/- mutants Gbx1 is first expressed in the ventricular zone of the spinal cord at E11.5 (Rhinn et al. 2004 ; Waters, Wilson & Lewandoski 2004). Then at E12.5-E13.5, it is broadly expressed in the mantle zone of the dorsal spinal cord. At E14 with the appearance of a distinguishable dorsal horn, Gbx1 expression becomes more restricted. At E12.5, Gbx1 is coexpressed with Lbx1; thus Gbx1 cells correspond to class B neurons (John, Wildner & Britsch 2005). As described in the introduction, late-born class B neurons comprise initially two populations, dILA and dILB. Because Gbx1 neurons co-express Lhx1/5 and Pax2, but not Lmx1b and Tlx3, it has been suggested that these neurons correspond to the dILA neuronal subtype (John, Wildner & Britsch 2005). It has been shown that dILA neurons undergo GABAergic differentiation (Cheng et al. 2004), and as mature GABA+ neurons they continue expressing Gbx1 (John, Wildner & Britsch 2005). We therefore analyzed GABAergic neurons in the spinal cord of Gbx1-/- mutants, which we identified by expression of glutamic acid decarboxylase GAD67, an enzyme that regulates GABA synthesis. At E18.5, Gad67-expressing cells are found throughout the developing spinal cord of control mice (Somogyi et al. 1995). Importantly, Gad67 expression was reduced in the dorsal spinal cord of Gbx1 mutant mice (Fig. 8A-D), i.e. there was a 20% decrease in the number of Gad67-expressing cells in comparison to WT mice (Fig. 8E). This may reflect an abnormal development or survival of GABAergic neurons, which in consequence coud lead to abnormal control of neuronal network in dorsal horn. Glutamate and GABA are the predominant neurotransmitters for excitatory and inhibitory neurons, respectively, in the vertebrate brain. These two neurotransmitters are typically expressed in a mutually exclusive manner (Bellocchio et al. 2000; Fremeau et al. 2001), thereby defining the major functional subdivision in neuronal cell types. In the dorsal horn of the spinal cord\u2014the major relay center for processing somatosensory information\u2014, most ascending projection neurons and a subset of local circuit interneurons are excitatory and use glutamate as their transmitter. These neurons are modulated by local inhibitory neurons, many of which are GABAergic (for reviews: Melzack & Wall 1965; Malcangio & Bowery 1996; Dickenson 2002). Thus, GABA may inhibit transmitter release from primary afferent fibers. The output neurons of the dorsal horn are projection neurons, which are concentrated in Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 16 lamina I and scattered throughout laminae III/VI, and relay sensory information to several brain areas. The vast majority of neurons in the dorsal horn are local circuit interneurons that do not project outside of the spinal cord. The output of projection neurons is influenced by local excitatory and inhibitory neurons (Todd 2010; Larsson & Broman 2011; Guo et al. 2012). The balance between excitation and inhibition is crucial for maintaining normal sensory function (Basbaum et al. 2009; Costigan, Scholz & Woolf 2009; Ross et al. 2010; Takazawa & MacDermott 2010). Blocking inhibitory transmission at spinal levels, for example, can lead to allydonia (pain following a normally non-painful tactile or thermal stimulus) (for review: Sandk\u00fchler 2009). For example, tactile allodynia may result from a loss of inhibition of excitatory interneurons that convey low-threshold mechanoreceptive inputs to lamina I projections neurons, leading to a mis-coding of the information by cells that normally only detect painful stimuli (Torsney & MacDermott 2006). In Gbx1 mutants, the reduction of GABAergic neurons may disrupt neuronal circuitry, which becomes apparent at adult stages as measured by abnormal performance in several behavioral tests. Further electrophysiological studies will be necessary to link the decrease of GABAergic neurons to the abnormal gait observed in adult Gbx1 mutant mice. conclusion : We have generated Gbx1-/- loss of function mutant mice, and investigated the development of the spinal cord dorsal horn in these mutants. Gbx1-/- mutants are viable and fertile, but display an altered gait during forward movement that specifically affects hindlimbs. Abnormal gait documented by a series of bevioral tests cannot be explained by deficits in muscle strength or motor coordination. However, we cannot exclude that proprioceptive deficits or allodynia could be at the origin of the Gbx1-/- phenotype. Our analysis of major neuronal classes revealed a reduced number of GABAergic inhibitory interneurons expressing Gad67 in the superficial dorsal horn of Gbx1-/- mice. Gbx1 may therefore be functionally required for the differentiation of local inhibitory interneurons in the dorsal horn, corroborating a previous report of Gbx1 expression in a specific subset of GABAergic neurons in this region of the spinal cord (John, Wildner & Britsch 2005). Despite the clear behavioral phenotype and reduced pool of GABAergic neurons in the dorsal horn, we did not observe any change in the expression of homeodomain factors involved in dorsal spinal cord patterning, or markers for primary sensory afferents, indicating that the development of the dorsal horn is not profoundly affected in Gbx1-/- mutants. An Pre Prin ts Pre Prin ts Phenotypic analysis of Gbx1-/- mouse mutants 17 explanation for these results\u2014and for the overall mild phenotype of the mutants\u2014is that Gbx1 and Gbx2 are coexpressed in dorsal spinal cord cells at early stages of embryogenesis (up to E12.5): hence the presence of Gbx2 might compensate for Gbx1 loss of function with respect to early regulatory events. Generation of Gbx1;Gbx2 double mutants will be required to assess possible redundant functions, and the availability of a Gbx2 floxed (conditional) allele does allow strategies for a spinal cord-specific inactivation, which would alleviate the lethality of the Gbx2 null mutants (Wassarman et al. 1997). Despite the importance of dorsal spinal cord in normal sensory processing, our knowledge concerning the establishment of the neuronal circuits remains limited (Graham, Brichta & Callister 2007; Todd 2010). In this regard, our work contributed to understand how transcription factors cooperate for regulating cell specification and eventual distribution of neuronal subtypes in the developing spinal cord, providing clues for further dissecting functional circuitry of the dorsal spinal cord. acknowledgments : We thank B. Schuhbaur for excellent technical assistance. funding statement : This work was supported by grants from the Agence Nationale de la Recherche (ANR Neurosciences 2007, ANR Blanc 2011), the Fondation pour la Recherche M\u00e9dicale (Equipe FRM 2007), the Deutsche Forschungsgemeinschaft (SFB-655 A3-Brand), the Italian Association for Cancer Research (AIRC), and by institutional funding from the Centre National de la Recherche Scientifique (CNRS), Institut National de la Sant\u00e9 et de la Recherche M\u00e9dicale (INSERM), and University of Strasbourg. Behavioral phenotyping was partly subsidized by the EUMODIC European Consortium and the Mouse Clinical Institute (MCI/ICS, Strasbourg). Pre Prin ts Pre Prin ts figure legends : Figure 2. Abnormal phenotype of a Gbx1-/- mouse when walking. Sequential pictures compare the normal gait of a wild-type mouse (A) and the abnormal gait (\"duck-like\" walk) of a Gbx1-/- mutant when walking (B). A movie of these mice is available (Supplementary information, file S1). Figure 3. Effects of Gbx1 mutation on the latency and number of slips in the beam walking test. ** p<0.01 vs WT; Student t-test. Figure 4. Open field performance of wild-type (WT) and Gbx1-/- mice. The distance traveled over the 20 min period of test reflects locomotor activity. The average speed was calculated during movement in the whole arena for the entire period of testing. Exploration of the central part of the open fied is expressed as the number of entries and percentage of time spent in the center. * p<0.05 and ** p<0.01 vs WT; Student t-test. Pre Prin ts Pre Prin ts Figure 6. Absence of morphological and molecular abnormalities in the developing dorsal horn of Gbx1-/- mice. Sections through the spinal cord of wild-type (A,C,E,G) and Gbx1-/(B,D,F,H) mice at E16.5 are shown. All sections are at the lumbar level. (A,B) Nissl-stained sections. No differences are detectable between wild-type and mutants. In situ hybridizations for two transcription factor encoding genes, Lbx1 (C,D) and Lmx1b (E,F), and for the axon guidance molecule netrin (G,H), are shown. No differences are observed between wild-type and mutants. Scale bars: 100 \u00b5m. Figure 7. Developmental progression of afferent projections in the dorsal horn of Gbx1-/mice. (A,B) Anti-calbindin-D28K antibody staining. At E18.5, calbindin fibers have already entered the spinal gray matter in wild-type embryos (A, arrow). Homozygous specimens (B) are indistinguishable from wild-types. (C,D) Expression of Drg11 in wild-type (C) and Gbx1 mutant (D) mice. Mutant specimens were indistinguishable from wild-type. (E,F) Antiperipherin antibody staining. Peripherin staining reveals no difference between mutant (E) and wild-type (F) in the ingrowth of group IA muscle sensory afferents that grow to the ventral spinal cord (arrows). Scale bars: 100 \u00b5m. Figure 8. Abnormal GABAergic differentiation in Gbx1-/- mice. Expression of Gad67 in wild-type (A,C) and Gbx1-/- (B,D) mice at E18.5. Higher magnification views (C,D; areas boxed in A,B) show the areas used for cell countings. (E) Countings revealed that the numbers of Gad67+ cells are diminished by 20% in Gbx1-/- mice (67.35\u00b15.56% Gad67+ cells in WT; 48.62\u00b15.84% in Gbx1-/- mice; *** p<0.001; Student t-test). Scale bars: 100 \u00b5m. Pre Prin ts Pre Prin ts table captions : Table 2. Effects of Gbx1 mutation on electromyography (EMG) measurements. The sensory nerve conduction velocity was measured at the level of the caudal nerve. The latency and the amplitude of gastrocnemius muscle response evoked by sciatic nerve stimulation were also recorded. * p<0.05 vs wild-type; Student t-test.",
    "url": "https://peerj.com/articles/143/reviews/",
    "review_1": "Rebecca Berdeaux \u00b7 Aug 5, 2013 \u00b7 Academic Editor\nACCEPT\nThank you for the revision.",
    "review_2": "Rebecca Berdeaux \u00b7 Aug 2, 2013 \u00b7 Academic Editor\nMINOR REVISIONS\nThank you for your revision. Although some changes to the manuscript were made and new data presented, the manuscript does not yet meet publication standards. I had previously requested that statements about the role of FXR-SHP be strongly modulated or removed. These statements are still too strong, as no data presented show specifically that FXR-SHP regulate bile acid levels in your model. Please respond to the following specific points in a revised version.\n\nMajor points:\n\n1. Page 5 \u201c\u2026 results clearly demonstrate that under physiological conditions, FXR-SHP regulation plays important roles in bile acid homeostasis in pregnant and lactating rats.\u201d This statement is not substantiated by the data. Page 13: The concluding sentence \u201cThe present study clearly demonstrates that in pregnant rats, FXR-SHP regulates bile acid synthesis enzyme genes to prevent the accumulation of bile acids in the liver, together with down-regulation of bile acid transporters\u2026\u201d is not substantiated by the data. An experiment showing that inhibition or activation of FXR-SHP causes alterations of bile acid synthesis genes in pregnant or lactating rats would be required. Statements about involvement of FXR-SHP should be substantially reduced and congruent with the level of analysis shown (correlation). At best, the data suggest a correlation.\n\nPage 10: Please include a citation for the statement \u201cIncreased FXR protein and SHP mRNA play an important role in bile acid homeostasis during pregnancy.\u201d If this is meant to refer to the present study, the sentence should be rewritten to only describe the correlation observed. No loss of function studies were presented to show that the alterations in mRNA or protein levels had an effect on bile acid homeostasis. Similarly, I do not agree that a correlation \u201cadds to our understanding of FXR-SHP regulation of bile acid synthesis and transport in rats during pregnancy and lactation.\u201d The statement \u201cestrogen and FXR interactions may not be evident in rats\u2026\u201d is not supported by QPCR data. Please omit.\n\n2. I cannot find anywhere where the n number of animals per condition is stated. Please state in the methods and the figure legends to go along with the discussion of statistical analysis. QPCR data should represent biological, not technical replicates and this should be clearly stated. For western blot data, single samples of each time point are shown; how were error bars generated and from how many additional experiments and replicates?\n\n3. The western blots in Fig 3 appear to be from two different gels. Loading controls should be from the same gel. Why does the tubulin blot smile considerably but Cyp7A1 does not, although these are similar predicted molecular weight proteins (57 kDa/ 55 kDa)? If these two blots were from the same gel, both sets of bands should have the same shape.\n\n4. In general, there seems to be an assumption that mRNA expression is reflective of protein abundance. For example, page 10 \u201cConsistent with hepatic bile acid concentrations, bile acid synthesis enzymes\u2026 were not increased during pregnancy.\u201d With the exception of CYP7A1, no protein data are shown for the other enzymes, so it is appropriate to specify that mRNA expression was not altered. As written, this sentence sounds like protein abundance was evaluated. Use of appropriate formatting for mRNA names will help make this more clear.\n\nIn the discussion, mRNA data are over-interpreted. The paragraph about BSEP, Mrp3 and Mrp4 sounds like the proteins are being discussed, when mRNA data was presented, which does not necessarily reflect the protein content. The discussion should be re-written to make these points clear to readers and to acknowledge the limitations of the study design.\n\n\nMinor revisions:\n\n\nThe authors were suggested to consult an English language editing service. There remain numerous typographical errors and mistakes with English usage. As a courtesy, I have listed several corrections, but the authors are urged to take advantage of professional editing services.\n\u03bf Abstract, line 4 \u201c \u2026 pregnant and lactating rats\u201d\n\u03bf Line 4: \u201ctimed pregnant\u201d\n\u03bf Page 3 \u201casymptotic\u201d should be \u201casymptomatic\u201d\n\u03bf Page 3 \u201cFXR \u2026 regulation mechanism \u201c delete \u201cmechanism\u201d \u2026 \u201cfor its ability\u201d change to \u201cin its ability\u201d\n\u03bf Page 3: ICP is a liver disease that\n\u03bf mRNA and protein names should follow appropriate nomenclature throughout: mRNA- italics with the first letter capitalized \u201cCyp7a1\u201d; protein- all caps. See MGI database website for assistance.\n\u03bf Page 9 \u201cwestern bolts\u201d should be \u201cwestern blots\u201d; \u201c\u2026the expressions of FXR protein\u2026\u201d should say \u201cthe expression\u201d\n\u03bf Page 9 \u201c..the expression of bile acid efflux\u2026 were decreased\u2026\u201d should be \u201cwas decreased\u201d\n\u03bf Page 9 \u201cwith slightly increase during lactation\u201d is incorrect \u201cwith slight increases during lactation\u201d",
    "review_3": "Reviewer 1 \u00b7 Jul 30, 2013\nBasic reporting\nNo further comment\nExperimental design\nNo further comment\nValidity of the findings\nNo further comment\nAdditional comments\nNo further comment\nCite this review as\nAnonymous Reviewer (2013) Peer Review #1 of \"Hepatic bile acids and bile acid-related gene expression in pregnant and lactating rats (v0.2)\". PeerJ https://doi.org/10.7287/peerj.143v0.2/reviews/1",
    "review_4": "Rebecca Berdeaux \u00b7 Jul 18, 2013 \u00b7 Academic Editor\nMAJOR REVISIONS\nTwo reviewers have evaluated the manuscript. While the study has merit, there are several issues to be addressed, some technical and many regarding the descriptions of the conclusions. Please respond to all reviewer comments. You may wish to consult with a language editing service. I agree with reviewer #1 that statements regarding the role of SHP are not supported by the data. The data show a possible correlation, but no loss of function studies are performed to demonstrate that SHP is in any responsible for the observed alterations in bile acids. The manuscript should be adjusted according to the comments of the reviewers.\n\nOne reviewer directs your attention to several previously published studies. Please consider how these may impact your work. You are not, however, required to cite any of these specific publications but may do so if you feel they are relevant. Inclusion of these specific citations will not be required for ultimate acceptance of a revision. I do, however, agree that more discussion and citations are warranted pertaining to the eitopathogenesis of ICP and the current knowledge about bile acid concentration in healthy pregnancy.",
    "pdf_1": "https://peerj.com/articles/143v0.3/submission",
    "pdf_2": "https://peerj.com/articles/143v0.2/submission",
    "review_5": "Reviewer 1 \u00b7 Jul 17, 2013\nBasic reporting\nIn this paper Zhu et al. have investigated the changes in the expression of genes involved in bile acid homeostasis during rat pregnancy and lactation. Although similar studies have been previously carried out the paper contain some interesting information mainly due to the fact that genes involved in metabolism and transport as well as key nuclear receptors, were included in the same study. Nevertheless, the manuscript contains important flaws that must be corrected before its publication could be recommended.\nExperimental design\nSee General Comments for the Authors\nValidity of the findings\nSee General Comments for the Authors\nAdditional comments\nIn this paper Zhu et al. have investigated the changes in the expression of genes involved in bile acid homeostasis during rat pregnancy and lactation. Although similar studies have been previously carried out the paper contain some interesting information mainly due to the fact that genes involved in metabolism and transport as well as key nuclear receptors, were included in the same study.\n\nMain points\n\n1. Throughout the text there is an important conceptual confusion among the terms homeostasis, metabolism, synthesis and transport. Homeostasis is maintained by several mechanisms, which include metabolism (synthesis and catabolism) and transport. Examples of misuse of terms:\nPage 9: \u201cregulation of homeostasis and transport\u201d\nPage 11: \u201c\u201dbile acid synthesis homeostasis\u201d\nAbstract: \u201cclosely related to bile acid metabolism\u201d do you mean \u201chomeostasis\u201d?\n\n2. ABSTRACT, Last sentence: \u201cincreased expression of bile acid transporters\u201d. This is not true for all of them. I would suggest \u201cchanges in the expression of bile acid transporters\u201d or \u201cincreased expression of major bile acid transporters\u201d.\n\nINTRODUCTION.\n3. The introduction is focused on ICP, which does not match with the title or the actual study carried out. This is a basic descriptive study carried out under physiological conditions. Although a mention to the relevance of these data to further understand ICP the first paragraph should be shortened and placed later in the Introduction section.\n\n4. If the etiopathogenesis of ICP is mentioned this should be done properly. The role of estrogens is mentioned (page 4) but not that of progesterone metabolites (Pascual et al. Clin Sci: PMID: 11980579).\n\n5. In the second paragraph of page 4: \u201clittle is known the\u201d. The sentence is wrong and not true.\n\nMETHODS\n\n6. A limitation of the present work is the way total bile acids have been measured. Although in rat, which does not have gallbladder, an important part of bile acid pool is in the liver in fasting conditions, this is only part of the total pool.\na. The authors do not indicate whether the rats were fasted overnight.\nb. Owing to their marked lipophilicity, the extraction of bile acids from liver tissue by centrifugation of homogenate diluted with saline is a very poor method.\nc. In Figure 1: Liver bile acids are given as \u00b5mol/L. What does this mean? Surprisingly control value is 100.\n\nDISCUSSION:\n7. Page 10, first paragraph: I disagree that serum bile acid concentrations are not increased during healthy pregnancy. There are several reports in this sense. See for instance the paper already mentioned above (Pascual et al., Clin Sci: PMID: 11980579).\n\n8. Page 12, first paragraph: There are important discrepancies between this study and that by Cao et al., 2001 that must be highlighted and commented.\n\n9. Page 12, Second paragraph: Mechanism of progesterone metabolites-induced impairment in bile secretion includes both NTCP inhibition and, probably more importantly, the inhibition of BSEP (Vallejo et al., J. Hepatol. PMID: 16458994).\n\n10. Page 12, Last of second paragraph: the sentence \u201cto avoid\u2026\u201d is highly speculative and not supported by or even related to the present study.\n\n11. Figures 5 and 6. It would make more sense to show all SLC transporters in one figure and all ABC pumps in the other., i.e., Exchange Ntcp with Abcg2. In addition, here and in the text use the updated nomenclature of rat Oatp transporters.\n\n\nMinor Points\n1. Abstract: \u201cH epatic bile acid homeostasis maintained\u201d correct to \u201cHepatic bile acid homeostasis is maintained\u201d.\n\n2. Introduction: Pag. 4, last line: \u201cin pregnant rats\u201d change to \u201cpregnant and lactating rats\u201d.\n\n3. Page 8: \u201c64.7%, 57.7% on GD10 and GD14\u201d change to \u201c64.7% and 57.7% on GD10 and GD14, respectively\u201d.\n\n4. Discussion: \u201cSHP would be responsible\u201d. This is not supported by the results and is probably wrong. I would suggest: \u201cSHP may play an important role in\u201d.\n\n5. Figure 5: \u201cNtcp\u201d label is to close to the Y-axis.\nCite this review as\nAnonymous Reviewer (2013) Peer Review #1 of \"Hepatic bile acids and bile acid-related gene expression in pregnant and lactating rats (v0.1)\". PeerJ https://doi.org/10.7287/peerj.143v0.1/reviews/1",
    "pdf_3": "https://peerj.com/articles/143v0.1/submission",
    "review 6": "Reviewer 2 \u00b7 Jun 26, 2013\nBasic reporting\nThis work is very interesting and enhances my understanding of hepatic bile acids and bile acid-related gene expression in pregnant and lactating rat.\nExperimental design\nThe authors researched bile acids synthesis by detecting bile acid synthesis enzymes, bile acids transporters and the regulation of bile acids homeostasis by detecting FXR-SHP gene expression in pregnant and lactation rat.\nValidity of the findings\nThe study was initiated to investigate FXR-SHP regulation of bile acid homeostasis and transport in rats during pregnancy and lactation.\nAdditional comments\nHowever, the authors should prepare a revised version of the paper, taking into account the following points\n1) I am wondering why the authors collected the test time on gestation days 10,14,19 and postnatal days 1,7,14 and 21.\n2) I am wondering whether the authors provide some protein expression proof for Cyp8b1, SHP and FXR.\n3) I think this is a typo \u201cPXR-SHP\u201d on page four. It is FXR-SHP, is right?\nCite this review as\nAnonymous Reviewer (2013) Peer Review #2 of \"Hepatic bile acids and bile acid-related gene expression in pregnant and lactating rats (v0.1)\". PeerJ https://doi.org/10.7287/peerj.143v0.1/reviews/2",
    "all_reviews": "Review 1: Rebecca Berdeaux \u00b7 Aug 5, 2013 \u00b7 Academic Editor\nACCEPT\nThank you for the revision.\nReview 2: Rebecca Berdeaux \u00b7 Aug 2, 2013 \u00b7 Academic Editor\nMINOR REVISIONS\nThank you for your revision. Although some changes to the manuscript were made and new data presented, the manuscript does not yet meet publication standards. I had previously requested that statements about the role of FXR-SHP be strongly modulated or removed. These statements are still too strong, as no data presented show specifically that FXR-SHP regulate bile acid levels in your model. Please respond to the following specific points in a revised version.\n\nMajor points:\n\n1. Page 5 \u201c\u2026 results clearly demonstrate that under physiological conditions, FXR-SHP regulation plays important roles in bile acid homeostasis in pregnant and lactating rats.\u201d This statement is not substantiated by the data. Page 13: The concluding sentence \u201cThe present study clearly demonstrates that in pregnant rats, FXR-SHP regulates bile acid synthesis enzyme genes to prevent the accumulation of bile acids in the liver, together with down-regulation of bile acid transporters\u2026\u201d is not substantiated by the data. An experiment showing that inhibition or activation of FXR-SHP causes alterations of bile acid synthesis genes in pregnant or lactating rats would be required. Statements about involvement of FXR-SHP should be substantially reduced and congruent with the level of analysis shown (correlation). At best, the data suggest a correlation.\n\nPage 10: Please include a citation for the statement \u201cIncreased FXR protein and SHP mRNA play an important role in bile acid homeostasis during pregnancy.\u201d If this is meant to refer to the present study, the sentence should be rewritten to only describe the correlation observed. No loss of function studies were presented to show that the alterations in mRNA or protein levels had an effect on bile acid homeostasis. Similarly, I do not agree that a correlation \u201cadds to our understanding of FXR-SHP regulation of bile acid synthesis and transport in rats during pregnancy and lactation.\u201d The statement \u201cestrogen and FXR interactions may not be evident in rats\u2026\u201d is not supported by QPCR data. Please omit.\n\n2. I cannot find anywhere where the n number of animals per condition is stated. Please state in the methods and the figure legends to go along with the discussion of statistical analysis. QPCR data should represent biological, not technical replicates and this should be clearly stated. For western blot data, single samples of each time point are shown; how were error bars generated and from how many additional experiments and replicates?\n\n3. The western blots in Fig 3 appear to be from two different gels. Loading controls should be from the same gel. Why does the tubulin blot smile considerably but Cyp7A1 does not, although these are similar predicted molecular weight proteins (57 kDa/ 55 kDa)? If these two blots were from the same gel, both sets of bands should have the same shape.\n\n4. In general, there seems to be an assumption that mRNA expression is reflective of protein abundance. For example, page 10 \u201cConsistent with hepatic bile acid concentrations, bile acid synthesis enzymes\u2026 were not increased during pregnancy.\u201d With the exception of CYP7A1, no protein data are shown for the other enzymes, so it is appropriate to specify that mRNA expression was not altered. As written, this sentence sounds like protein abundance was evaluated. Use of appropriate formatting for mRNA names will help make this more clear.\n\nIn the discussion, mRNA data are over-interpreted. The paragraph about BSEP, Mrp3 and Mrp4 sounds like the proteins are being discussed, when mRNA data was presented, which does not necessarily reflect the protein content. The discussion should be re-written to make these points clear to readers and to acknowledge the limitations of the study design.\n\n\nMinor revisions:\n\n\nThe authors were suggested to consult an English language editing service. There remain numerous typographical errors and mistakes with English usage. As a courtesy, I have listed several corrections, but the authors are urged to take advantage of professional editing services.\n\u03bf Abstract, line 4 \u201c \u2026 pregnant and lactating rats\u201d\n\u03bf Line 4: \u201ctimed pregnant\u201d\n\u03bf Page 3 \u201casymptotic\u201d should be \u201casymptomatic\u201d\n\u03bf Page 3 \u201cFXR \u2026 regulation mechanism \u201c delete \u201cmechanism\u201d \u2026 \u201cfor its ability\u201d change to \u201cin its ability\u201d\n\u03bf Page 3: ICP is a liver disease that\n\u03bf mRNA and protein names should follow appropriate nomenclature throughout: mRNA- italics with the first letter capitalized \u201cCyp7a1\u201d; protein- all caps. See MGI database website for assistance.\n\u03bf Page 9 \u201cwestern bolts\u201d should be \u201cwestern blots\u201d; \u201c\u2026the expressions of FXR protein\u2026\u201d should say \u201cthe expression\u201d\n\u03bf Page 9 \u201c..the expression of bile acid efflux\u2026 were decreased\u2026\u201d should be \u201cwas decreased\u201d\n\u03bf Page 9 \u201cwith slightly increase during lactation\u201d is incorrect \u201cwith slight increases during lactation\u201d\nReview 3: Reviewer 1 \u00b7 Jul 30, 2013\nBasic reporting\nNo further comment\nExperimental design\nNo further comment\nValidity of the findings\nNo further comment\nAdditional comments\nNo further comment\nCite this review as\nAnonymous Reviewer (2013) Peer Review #1 of \"Hepatic bile acids and bile acid-related gene expression in pregnant and lactating rats (v0.2)\". PeerJ https://doi.org/10.7287/peerj.143v0.2/reviews/1\nReview 4: Rebecca Berdeaux \u00b7 Jul 18, 2013 \u00b7 Academic Editor\nMAJOR REVISIONS\nTwo reviewers have evaluated the manuscript. While the study has merit, there are several issues to be addressed, some technical and many regarding the descriptions of the conclusions. Please respond to all reviewer comments. You may wish to consult with a language editing service. I agree with reviewer #1 that statements regarding the role of SHP are not supported by the data. The data show a possible correlation, but no loss of function studies are performed to demonstrate that SHP is in any responsible for the observed alterations in bile acids. The manuscript should be adjusted according to the comments of the reviewers.\n\nOne reviewer directs your attention to several previously published studies. Please consider how these may impact your work. You are not, however, required to cite any of these specific publications but may do so if you feel they are relevant. Inclusion of these specific citations will not be required for ultimate acceptance of a revision. I do, however, agree that more discussion and citations are warranted pertaining to the eitopathogenesis of ICP and the current knowledge about bile acid concentration in healthy pregnancy.\nReview 5: Reviewer 1 \u00b7 Jul 17, 2013\nBasic reporting\nIn this paper Zhu et al. have investigated the changes in the expression of genes involved in bile acid homeostasis during rat pregnancy and lactation. Although similar studies have been previously carried out the paper contain some interesting information mainly due to the fact that genes involved in metabolism and transport as well as key nuclear receptors, were included in the same study. Nevertheless, the manuscript contains important flaws that must be corrected before its publication could be recommended.\nExperimental design\nSee General Comments for the Authors\nValidity of the findings\nSee General Comments for the Authors\nAdditional comments\nIn this paper Zhu et al. have investigated the changes in the expression of genes involved in bile acid homeostasis during rat pregnancy and lactation. Although similar studies have been previously carried out the paper contain some interesting information mainly due to the fact that genes involved in metabolism and transport as well as key nuclear receptors, were included in the same study.\n\nMain points\n\n1. Throughout the text there is an important conceptual confusion among the terms homeostasis, metabolism, synthesis and transport. Homeostasis is maintained by several mechanisms, which include metabolism (synthesis and catabolism) and transport. Examples of misuse of terms:\nPage 9: \u201cregulation of homeostasis and transport\u201d\nPage 11: \u201c\u201dbile acid synthesis homeostasis\u201d\nAbstract: \u201cclosely related to bile acid metabolism\u201d do you mean \u201chomeostasis\u201d?\n\n2. ABSTRACT, Last sentence: \u201cincreased expression of bile acid transporters\u201d. This is not true for all of them. I would suggest \u201cchanges in the expression of bile acid transporters\u201d or \u201cincreased expression of major bile acid transporters\u201d.\n\nINTRODUCTION.\n3. The introduction is focused on ICP, which does not match with the title or the actual study carried out. This is a basic descriptive study carried out under physiological conditions. Although a mention to the relevance of these data to further understand ICP the first paragraph should be shortened and placed later in the Introduction section.\n\n4. If the etiopathogenesis of ICP is mentioned this should be done properly. The role of estrogens is mentioned (page 4) but not that of progesterone metabolites (Pascual et al. Clin Sci: PMID: 11980579).\n\n5. In the second paragraph of page 4: \u201clittle is known the\u201d. The sentence is wrong and not true.\n\nMETHODS\n\n6. A limitation of the present work is the way total bile acids have been measured. Although in rat, which does not have gallbladder, an important part of bile acid pool is in the liver in fasting conditions, this is only part of the total pool.\na. The authors do not indicate whether the rats were fasted overnight.\nb. Owing to their marked lipophilicity, the extraction of bile acids from liver tissue by centrifugation of homogenate diluted with saline is a very poor method.\nc. In Figure 1: Liver bile acids are given as \u00b5mol/L. What does this mean? Surprisingly control value is 100.\n\nDISCUSSION:\n7. Page 10, first paragraph: I disagree that serum bile acid concentrations are not increased during healthy pregnancy. There are several reports in this sense. See for instance the paper already mentioned above (Pascual et al., Clin Sci: PMID: 11980579).\n\n8. Page 12, first paragraph: There are important discrepancies between this study and that by Cao et al., 2001 that must be highlighted and commented.\n\n9. Page 12, Second paragraph: Mechanism of progesterone metabolites-induced impairment in bile secretion includes both NTCP inhibition and, probably more importantly, the inhibition of BSEP (Vallejo et al., J. Hepatol. PMID: 16458994).\n\n10. Page 12, Last of second paragraph: the sentence \u201cto avoid\u2026\u201d is highly speculative and not supported by or even related to the present study.\n\n11. Figures 5 and 6. It would make more sense to show all SLC transporters in one figure and all ABC pumps in the other., i.e., Exchange Ntcp with Abcg2. In addition, here and in the text use the updated nomenclature of rat Oatp transporters.\n\n\nMinor Points\n1. Abstract: \u201cH epatic bile acid homeostasis maintained\u201d correct to \u201cHepatic bile acid homeostasis is maintained\u201d.\n\n2. Introduction: Pag. 4, last line: \u201cin pregnant rats\u201d change to \u201cpregnant and lactating rats\u201d.\n\n3. Page 8: \u201c64.7%, 57.7% on GD10 and GD14\u201d change to \u201c64.7% and 57.7% on GD10 and GD14, respectively\u201d.\n\n4. Discussion: \u201cSHP would be responsible\u201d. This is not supported by the results and is probably wrong. I would suggest: \u201cSHP may play an important role in\u201d.\n\n5. Figure 5: \u201cNtcp\u201d label is to close to the Y-axis.\nCite this review as\nAnonymous Reviewer (2013) Peer Review #1 of \"Hepatic bile acids and bile acid-related gene expression in pregnant and lactating rats (v0.1)\". PeerJ https://doi.org/10.7287/peerj.143v0.1/reviews/1\nReview 6: Reviewer 2 \u00b7 Jun 26, 2013\nBasic reporting\nThis work is very interesting and enhances my understanding of hepatic bile acids and bile acid-related gene expression in pregnant and lactating rat.\nExperimental design\nThe authors researched bile acids synthesis by detecting bile acid synthesis enzymes, bile acids transporters and the regulation of bile acids homeostasis by detecting FXR-SHP gene expression in pregnant and lactation rat.\nValidity of the findings\nThe study was initiated to investigate FXR-SHP regulation of bile acid homeostasis and transport in rats during pregnancy and lactation.\nAdditional comments\nHowever, the authors should prepare a revised version of the paper, taking into account the following points\n1) I am wondering why the authors collected the test time on gestation days 10,14,19 and postnatal days 1,7,14 and 21.\n2) I am wondering whether the authors provide some protein expression proof for Cyp8b1, SHP and FXR.\n3) I think this is a typo \u201cPXR-SHP\u201d on page four. It is FXR-SHP, is right?\nCite this review as\nAnonymous Reviewer (2013) Peer Review #2 of \"Hepatic bile acids and bile acid-related gene expression in pregnant and lactating rats (v0.1)\". PeerJ https://doi.org/10.7287/peerj.143v0.1/reviews/2\n",
    "LLM_extracted_review": "['**Review 1:**\\n- The manuscript does not yet meet publication standards. Statements about the role of FXR-SHP are too strong and not substantiated by the data. A correlation is suggested, but no experiments show that FXR-SHP regulates bile acid levels.\\n- Specific points include the need to substantiate claims about FXR-SHP\\'s role in bile acid homeostasis and to clarify the correlation observed.\\n- The n number of animals per condition is not stated; this should be included in the methods and figure legends.\\n- Western blots appear to be from different gels, and loading controls should be from the same gel.\\n- There is an assumption that mRNA expression reflects protein abundance, which is not supported by the data presented.\\n- The discussion over-interprets mRNA data, and the limitations of the study design should be acknowledged.\\n\\n**Review 2:**\\n- No further comments.\\n\\n**Review 3:**\\n- The study shows a possible correlation regarding the role of SHP, but no loss of function studies are performed to demonstrate its responsibility for observed alterations in bile acids.\\n- More discussion and citations are warranted regarding the etiopathogenesis of ICP and bile acid concentration in healthy pregnancy.\\n\\n**Review 4:**\\n- The manuscript contains important flaws that must be corrected before publication can be recommended.\\n- There is conceptual confusion among terms like homeostasis, metabolism, synthesis, and transport.\\n- The introduction is overly focused on ICP, which does not match the study\\'s title or content.\\n- The method of measuring total bile acids is limited, and the extraction method is poor.\\n- Discrepancies between this study and previous studies must be highlighted and discussed.\\n\\n**Review 5:**\\n- The authors should clarify the test time points for gestation and postnatal days.\\n- There is a need for protein expression proof for certain enzymes and receptors.\\n- A typo regarding \"PXR-SHP\" should be corrected to \"FXR-SHP.\"']"
}