WO1998015172A2 - Transgenic or mutated animal as model for a neuron deficit - Google Patents

Abstract

The present invention relates to a transgenic or mutated animal, wherein the expression of a gene involved in a neuron, for example a motoneuron, survival or activity is modulated and/or altered, which induces a nervous and more particularly a neuromascular deficiency, wherein said animal is characterized in that an administration by intramuscular, intravenous, oral or any other route, of an active dose of a molecule or a composition of active molecules, as for example neurotrophic factors or cytokines, is capable to compensate the said deficiency in vivo. The invention also relates to the use of said transgenic or mutated animal for the screening of active molecules.

Description

TRANSGENIC OR MUTATED ANIMAL AS MODEL FOR A NEURON DEFICIT

BACKGROUND OF THE INVENTION

This invention relates to transgenic or mutated animals and their use in screening molecules capable of compensating for neuron deficit or functional deficiency.

A functional nervous system relies on precise neuronal circuits between neurons and neuronal, muscular, or endocrine targets. Although ongoing modifications occur during adulthood, the neuronal connectivity pattern at developmental stages follows a highly invariant program, whereby neurons from a particular area branch to specific targets .

Retrograde labeling and electrophysiological experiments have revealed that a single skeletal muscle is innervated by a cluster of longitudinally organized spinal motoneurons (MNs), called a motor pool (Romanes, 1951; Romanes, 1964; Landmesser, 1978a; also review in Udin and Fawcett, 1988) . Muscles of the axial skeleton are supplied by MNs from the medial motor column (MMC) , whereas forelimb and hindli b muscles are innervated by MNs located in an enlarged structure of the spinal cord called the lateral motor column (LMC) .

Tissues that are innervated by the autonomic nervous system receive sympathetic projections from the autonomic ganglia, which are formed from neural-crest-derived neurons. These neurons are innervated by preganglionic MNs, which form the motor column of Terni (CT) , situated alongside the central lumen of the spinal cord (review in Lumsden, 1995) .

Precise topographic maps of the peripheral targets within the central nervous system have been drawn. For example, MNs that innervate forelimb muscles are contained in a rostro-caudal (R-C) area within the LMC, extending from the 5th cervical to the 1st thoracic neural segments (C5- Ti) . MNs from 05-5 supply shoulder and arm muscles (proximal musculature) , whereas finger movements are controlled by MNs from C7 and 03, Ti for fine movements (distal musculature) .

Anatomical and functional studies of embryos in various mammals have allowed for further subdivision of the LMC into lateral (LMC1) and medial (LMC ) subclasses, each of which contains MNs that specifically innervate extensor (dorsal) or flexor (ventral) muscles (Landmesser, 1978a; Landmesser, 1978b; Hollyday, 1980; Curfs, et al., 1993 and ref. therein) . Therefore, there is a fine correlation between the R-C and transverse coordinates of a MN within the spinal cord, and the proximo-distal and dorso-ventral position of its peripheral target.

It is now well established that the specific connection between a MN and its target depends, in part, on the ability of the growth cone to differentially recognize guidance cues along its pathway (reviewed in Goodman and Shatz , 1993). Numerous studies in Drosophila melanogaster, chick, and zebrafish have demonstrated that early in development, MNs become committed in a cell-autonomous fashion (see review in Eisen, 1994). Spinal cord manipulations along the R-C axis in the chick embryo have revealed that MNs acquire this identity at stage 15, as soon as they leave the cell cycle (Matise and Lance-Jones, 1996) . These data provide evidence that MNs acquire unique regional specification early in development; however, little is known about the genetic factors involved in this regionalization. A direct correlation between the specificity of target matching and expression of a unique repertoire of LIM- homeobox genes within groups of MNs has been described for the chick and zebrafish embryos. In the chick, distinct longitudinal columns express a unique combination of LIM- homeobox genes, suggesting that these transcription factors could provide positional information to MNs.

The first LIM-homeobox gene to be expressed is Islet-1 (Isll) , and it is detected in all MNs and only Mns immediately after the last cell division of precursor neural cells (Ericson, et al., 1992). Expression of Isl2, Lim3 , and Liml follows, but is restricted to MN subtypes and precedes the formation of motor columns (Tsuchida, et al., 1994). In the zebrafish embryo, distinct axial muscles initially receive projections from primary MNs that occupy discrete positions within the spinal cord (Eisen, et al . , 1986), and express a specific combination of LIM-homeobox genes (Appel, et al., 1995; Tokumoto, et al . , 1995). Grafting of a primary MN to an ectopic position three hours before axogenesis results in the expression by this MN of a new combination of LIM-homeobox genes followed by a projection pattern corresponding to its new position (Appel, et al., 1995) . Altogether, these data suggest that LIM-homeobox gene expression can confer positional information to MNs. However, within each column, MNs from different R-C levels express the same combination. Therefore, these genes are not sufficient to explain the fine level of spatial organization observed within the spinal cord.

In vertebrates, other homeobox-containing genes, the Hox genes, have been shown to specify R-C positional information (review in McGinnis and Krumlauf, 1992; Krumlauf, 1994; Burke, et al., 1995). The earliest expression of Hox genes in ectodermal derivatives is detected in the neurectoderm, and later in the ependymal layer which contains neuronal precursors.

The thirty eight Hox genes are grouped into four clusters on different chromosomes. Within each cluster, the different anterior limits of expression of individual homeotic genes follow a positional and temporal colinearity. This results in the expression of unique combinations of Hox transcripts in cells at distinct R-C positions.

Defects in axonal pathfinding for some MNs in the segmented hindbrain were reported after manipulation of Hox gene expression, as demonstrated in the Hoxa-1 null mutant (Carpenter, et al . , 1993), or by modulation of Hox gene expression in retinoic acid-treated embryos (Marshall, et al., 1992; Kessel, 1993). These results suggest that, in addition to their important role in the hindbrain formation, Hox genes could encode positional information for the navigating growth cone.

SUMMARY OF THE INVENTION

The invention relates to the field of the therapy of diseases affecting the neurons, as for example the motoneurons. The invention concerns also the identification of active molecules capable of compensating the deficit and the functional deficiency of the neurons, and particularly the motoneurons, by using a transgenic animal.

Usual techniques for the screening of active compounds in the pharmaceutical field involved a first step which is generally an in vitro assay. In the present invention, although an in vitro assay can be used for screening genetic defects located in the regulatory region of the Hoxc-8 gene, for example, the identification of the new active molecules necessary for the compensation of the deficit or the functional deficiency of the neurons can be done on an in vivo model. This identification is based particularly on new tests characterized by using a physiopathological animal model, which is a transgenic animal. The transgenic animal is defined as carrying a mutant phenotype characterized by a nervous defect as for example a defect in prehension capacity compared to the wild type phenotype. The wild type phenotype is defined as a prehension capacity, as for example the ability to cling with forepaws to the grid of the animal cage for several seconds, and at least 5 seconds.

A physiopathological model has been described in

European Patent Application No. 0 717 105. It can be distinguished from the present invention by the modification of a part of neuronal receptor. No modification affects the neuro-neuronal transmission in the central nervous system in the mutant mice described in that European patent application. In addition, in the present invention the transgenic or mutated animal can be easily observed and used in standard tests of behavioral performance.

The mutant phenotype according to the invention is visible allowing the selection at different stages of the development from embryonic stage to the adult stage. Also a mutated animal has to be considered as a natural mutated animal or an animal transformed, for example, by homologous recombination. The transgenic or mutated animal (i.e., non- human) of the present invention belongs to the vertebrate and is defined as a mouse, a chick, a fish, or any other animal usually implicated in the study of nervous or neuro uscular trouble or used as a model in neurogenesis.

The deficiency in the present invention is defined as a defect in the recognition or in the innervation between the neuron and its target: Ex: a neuro-neuronal synapse, a neuro- uscular synapse, a neuro-endocrine synapse. The synapse is unable to function. The function is defined as a synaptic connection, which is established in a continuously stable and functional way (electrical or chemical signal on the target) between the neuron and its target (the innervation is incorrect in the mutant) .

In one preferred embodiment of the invention, the gene involved is Hoxc-8 for murine (which corresponds to HOXC-8 for human) . In other species, equivalent genes do exist and are defined as the homeobox genes in the guidebook, 1994, edited by Denis Duboule, a Sambrook and Tooze publication, at Oxford University Press, England. "Gene involved" means that when mutated, the gene induces a deficiency.

A loss-of-function mutation for the homeotic gene, Hoxc-8 , has been generated in the mouse by substituting a part of the Hoxc-8 coding region with the lacZ reporter gene in embryonic stem cells. Cells that would have normally expressed the endogenous Hoxc-8 can, therefore, be followed by monitoring β-galactosidase activity (Le Mouellic, et al.,

1992). Based on anatomical criteria, anterior homeotic transformations of axial skeletal segments have been described throughout the Hoxc-8 expression domain, and the frequency of transformation correlated with both the level of expression of Hoxc-8 in cells and the density of Hoxc-8 expressing cells in each segment (Le Mouellic, et al . , 1992; Tiret, et al., 1993). Hoxc-8 represents a good candidate gene for specifying R-C coordinates during development. Accordingly, this invention provides a transgenic or mutated animal, wherein the expression of a gene involved in a neuron, for example a motoneuron, survival or activity is modulated and/or altered, which induces a nervous and more particularly a neuromuscular deficiency. The animal is characterized in that an administration by intramuscular, intravenous, oral or any other route, of an active dose of a molecule or a composition of active molecules, as for example, neurotrophic factors or cytokines, is capable of compensating the deficiency in vivo.

In another embodiment, this invention provides a transgenic or mutated animal, wherein the expression of a gene involved in a neuron, for example a motoneuron, survival or activity is modulated and/or altered, which induces a nervous, and more particularly a neuromuscular deficiency. The invention is characterized by administration by intramuscular, intravenous, oral, or any other route, of cells or expression vectors producing an active molecule, as for example neurotrophic factors or cytokines.

This invention also provides a process for screening an active molecule capable of compensating the pathological death of neurons or deficient activity in view of stimulating or stabilizing the survival of neurons. Further, this invention provides a process for screening an active molecule capable of compensating the pathological death of neurons or deficient activity or any other deficiency related to nervous diseases, and more particularly, to degenerative neuromuscular diseases. The active molecule is tested on a transgenic animal carrying a nervous, for example neuromuscular, deficiency phenotype.

In another embodiment, this invention provides a process for screening an active molecule capable of compensating the pathological death of neurons, or a deficient activity, or any other deficiency related to nervous diseases, and more particularly to degenerative neuromuscular disease. The active molecule is tested on a transgenic animal carrying the mutated Hoxc-8 gene or any sequence having at least 50% of homology with the Hoxc-8 coding sequence or 29% of identical amino acids with at least 70% of homology with the homeodo ain of Hoxc-8, or any nucleotide sequence capable of restoring the wild-type phenotype, wherein said gene is mutated.

For instance, the Hoxc-9 gene and product follow these homology criterions. Sequence alignment of the whole Hoxc-8 and Hoxc-9 protein is depicted in Figure 7 and demonstrates a 29% homology. Figure 8 shows a 73% homology of the amino acid sequence between homeodomains of Hoxc-8 and Hoxc-9 proteins.

When the Hoxc-9 product is substituted for the Hoxc-8 protein, the Hoxc-8 deficient mice phenotype is partially reverted. Alternatively, the active molecule is tested on a cell line prepared from primary or immortalized cells of the transgenic animal carrying the mutated Hoxc-8 gene or an equivalent gene as defined above. This invention further provides a process for screening an active molecule capable of compensating the pathological death of neurons or deficient activity or any other deficiency related to nervous disease, and more particularly to degenerative neuromuscular disease. The process comprises primary or immortalized cell cultures from a transgenic animal containing in its genome an hybrid nucleotide sequence comprising a part of the Hoxc-8 gene fused with a reporter gene or any sequence having at least 50% of homology with the Hoxc-8 coding sequence or 29% of identical amino acids with at least 70% of homology with the homeodomain of Hoxc-8 , or any nucleotide sequence capable of restoring the wild-type phenotype.

Still further, this invention provides an active molecule capable of restoring the wild-type phenotype in a transgenic animal by preventing neurons, for example motoneurons, from pathological death or capable of stimulating neuronal precursor proliferation or maintaining the stable innervation of the target cell. The active molecule can be used as reagent for the methods of the invention and for the tests on a transgenic animal of the invention. The selected active molecules are able to compensate the neuronal deficit in vivo.

In another embodiment, this invention provides a process for screening an active molecule capable of compensating for a prehension deficiency, a pathological death of neurons, or any other deficiency related to nervous diseases, and more particularly to degenerative neuromuscular diseases.

An active molecule capable of restoring a wild-type phenotype to an Hoxc-8 ' mutant is also provided. The invention further provides a pharmaceutical composition containing active molecules capable of restoring the survival of at least 40% of deficient neurons, for example motoneurons in vivo or capable of restoring, in the transgenic animal of the invention, a neuronal activity, for example motor activity equivalent to the wild-type animal.

Still further the invention provides a process of treatment of vertebrate and more particularly of mammals affected by a neuronal, for example motoneural deficiency, wherein the administration of an active molecule or a pharmaceutical composition is capable of inducing modifications in the behavioral and/or functional properties in the patient or the mutated animal.

The invention also provides a process of activity measurement on a mutated animal of an active molecule, said process being characterized by the use of behavioral and functional tests compared to the measure of the performances of the wild-type animal and of those of the untreated transgenic or mutated animal of the invention.

The following definitions apply to terms used herein. A "compensation" is helping in the contribution against cell death. Examples of active molecules are cytokines, neurotrophins secreted at the target cell level (as BDNF, NT-4 , CT-1) , or survival factors secreted at the central nervous system level (CNTF, IGF) . Any screened molecule inducing a compensation with a visible phenotype is able to restore the deficit for this model .

To "compensate" means that the pathological deficit of neurons is reduced by at least 40% (77% of deficit with CNTF on pmn/pmn mice [Sagot et al., 1995]).

"Active molecules" comprise any screened product as chemical or biological material.

"Motoneuron deficit" means a loss in the range of at least 20% to 40% of neuron number innervating a defined target.

We have to distinguish between "pathological cell death" as, involving for example, necrosis or apoptosis (Patent Appln. No. WO 92/17193 and two articles: Newell et al., Nature 347, p. 286-289 [1990] and Janssen et al., Immunology 146, p. 35-39 [1991] relative to murine and human adult T cell can undergo apoptosis in particular conditions) and "normal" programmed cell death of neurons, which is a physiological phenomenon. The pathological one is an augmentation of the naturally occurring cell death (Oppenheim R.W. (1991)).

This model can be useful in the establishment of therapies during early embryonic stages, when identities of neural cells are determined, or in the perinatal or the adult life when the plasticity of connectivity patterns allows a relative remodeling of axonal branching. A screening of pharmaceutical treatments can be achieved alone or in combination with various genetic therapies. The latter involves the transfer of biological materials (proteins, nucleic acids, viral particles, wild-type or genetically modified cells) .

Some examples of activity tests of molecules used for the process of screening follow:

(1) In vitro:

The reduction of the neuron deficit following the treatment with active molecules could be tested in vitro on primary culture. Spinal cord from 13.5 embryos are dissected, incubated for 5 in. at 37°C with 0.05 trypsin in calcium and magnesium- free HANK ' s balanced salt solution. After removal of the trypsin solution, the spinal cord is washed with DMEM (GIBCO-BRL) containing 10% heat inactivated serum and is gently triturated with a fine-polished Pasteur pipette to give a single cell suspension. The cells are plated on 24-well plates, precoated with polyornithine (0.5 mg/ml, overnight) and laminin (20 μg/ml, 4 hours). The

neurons are incubated at 37°C in a humidified 5% CO2 incubated in HAMS' F14 medium supplemented with 2 mM gluta ine, 0.35% BSA, 60 ng/ml progesterone, 16 mg/ml putrescine, 400 ng/ml L-thyroxine, 38 ng/ml sodium selenite,

340 ng/ml triido-thyronine, 60 μg/ml penicillin and 100

μg/ml streptomycin, and the active molecule at various concentrations (e.g. NT.3 for dorsal root ganglia 10 ng/ml [Elshamy and Ernfors, 1996]). In order to identify the motoneurons with the primary cell culture as obtained, Islet 1 antibodies can be used. The motoneuron Islet 1 antibodies complex is visualized by ELISA technique. In addition, the respective survival of motoneurons from Hoxc-8 ' mice in the presence or absence of the active molecule, can be assessed after 24 hours by counting the neurons, recognized by their bipolar morphology under phase-contrast optics, in a predetermined area of 1.5 mm 2 in the center of the well.

The potency of the active molecules can also be determined by assessing the survival of neurons in the presence of increasing concentrations of the active molecules; a dose-response curve after 6 days of culture in neurobasal medium is obtained (GIBCO-BRL), [Pennica et al., 1996] .

(2) In vivo at embryonic stages: The compensation of the deficiency by active molecules during the embryonic stages can be obtained by mating heterozygous Hoxc-8 mutated mice, 12 a.m. of the following day (plug day) is taken as day 0.5. The active molecule can be administered to the mother by a single or a multiple oral gauge of, for instance, sesame oil containing the active liposoluble molecule [e.g. retinoic acid 10 mg/kg, Kessel and Gruss, 1991] or by implanting in the peritoneal cavity of the mother with a mini-osmotic pump (ALZET) delivering the active molecule [ESCARY, 1993] on day 12 before the onset of cell death, or later when synapses need to be stabilized. The efficiency of the molecule is assessed by the ability of Hoxc-8 ^/ pups to cling to a grid with forepaws at 3 weeks after birth. (3) In vivo at post-natal stages:

Compensation of the neuromuscular deficiency could be tested in vivo by a technique involving encapsulated genetically engineered cells that produce the active molecule or a composition of active molecules. The active molecule gene is inserted into a dihydrofolate reductase (DHFR) - based expression vector named pNUT [Baetge et al. 1986]. Transfection of baby hamster kidney (BHK) fibroblasts with this vector are performed using calcium phosphate precipitation or electroporation. The transfected cells are then exposed to ethotrexate solution (200 μm) for 8 weeks - selected cells are harvested using dissociation medium (Sigma) and suspended in DMEM medium with 10% FCS (GIBCO) and 1% penicillin Streptomycin (GIBCO) . The cell suspension was mixed in a 1:1 ratio with a solution of collagen (Vibrogen, Celtrix or agarose (FMC Bioproducts, Rockland, ME) and slowly injected into a polypropylene fiber

[OD: 550 μm; ID: 350 μm] . Capsules a few mm in length are cut and their extremities heat-sealed [e.g. CNTF producing cells in pmn/pmn mice, Sagot et al. 1995]. Capsules containing 5.10 4 cells are then implanted by surgery into the target or into the central nervous system of newborn or adult Hoxc-8 ' mice. Mice are then tested every 3 days for their motor ability using, for instance, a selection of behavioral tests adopted from Altman and Sudarshan.

For Hoxc-8 ' treated animals, a test termed "climbing onto a fixed barrier", which consists in measuring the time required to climb onto a 3 mm width barrier, is particularly adapted. The principle of this functional restoration assay can be extended to any model of animal displaying a identifiable and reversible nervous phenotype. These results show that in the Hoxc-8 deficient mice, some MNs have lost their regional specifications and this leads to a less precise and non-functional connectivity pattern.

BRIEF DESCRIPTION OF THE DRAWINGS This invention will be described in greater detail with reference to the following drawings.

Figure 1. Nervous System Phenotype and Expression of Hoxc-8 In the Neural Tube.

(A) A 4 month old 129/Sv null mutant male (-/-) , showing a hunched profile with an erected tail. The forelimb phenotype is displayed in the inset photograph, in which the fingers of the forepaw are clenched. (B) He i-sagittal section from a whole-mount of a null mutant newborn, stained with X-Gal (blue) and counterstained with alizarin (bones are red) . C3 refers to the spinal cord domain delimited by the 2 (Ax) and 3 cervical vertebrae, and so on up to Cg . T]_ is between the 1 and the 2 thoracic vertebra. The domain of highest expression of Hoxc-8 is at C7-T1. Figures 1C and ID, as well as Figures 4, 5 and 6 are photographs from transverse sections at the level of Cg , and are represented by a dark line in (B) .

(C and D) Transverse sections through an E13.5 embryos. In the posterior cervical region, lacZ-expressing cells are found in the intermediate part of the neural tube and in the lateral motor columns (arrow heads) within the ventral horns. Null mutant cells have two copies of the lacZ gene, therefore staining of cells in the homozygotes is twice as strong as in the heterozygotes. However, the gross pattern of expression of the reporter gene is not observed to be significantly modified by the mutation, as previously described (Le Mouellic et al., 1992).

Ax, axis; SC, spinal cord; sg, spinal ganglion; vb, vertebral body; vr, ventral root.

Scale bar, 50 μm.

Figure 2. Distribution of Motor Pools: Projecting to Distal Forepa Muscles in Adults and ne borns . The R-C and transverse distributions of motor pools innervating the forelimb distal muscles of adults and newborns were observed on horizontal (A and C) and transverse (B) sections of spinal cords, after a retrograde labeling of Dil (A and C) and CTB-HRP (B) (see Experimental Procedures) . (B) Composite illustration of adult spinal cord transversed sections, showing positions of MN somata projecting to flexor muscles (FLEX-MNs; red points) or extensor muscles (EXT-MNs, green points). Because the number of labeled MNs depends on the number of muscle fibers into which the CT3-HRP is injected, this technique has no quantitative significance; the number of points is not the total number of MNs within the pool. Results obtained for +/+ and +/- animals were identical, and were therefore pooled and called +/-. (+/+ and +/-: n=7 , -/-: n=4). (A and left part of B) Control animals, in which the MNs that supply the forelimb distal muscle's are contained in C7 and CQ . FLEX-MNS are found medially to EXT-MNs, defining two distinct longitudinal columns.

(C and right part of B) Null mutant animals, in which MNs are detected in a wide R-C domain extending from C5 to Ti, with a few cell bodies observed as rostrally as C3.4 in some, but not all, animals studied. The specific transverse positions of longitudinal columns are no longer respected since the two populations of FLEX- and EXT-MNS observed are mixed. Figure 3. R-C Distribution of Motor Pools Projecting to Distal Forelimb Muscles at E12.5 and E14.5.

The R-C distribution of motor pools was determined on horizontal sections, after retrograde diffusion of Dil from muscular target cells to MN cell bodies.

LacZ (A) An E12.5 myf-5 whole-mount embryo stained with X-Gal, showing the forelimb distal premuscle masses in which crystals of Dil are implanted (star) . The plane of sectioning of (B-F) is represented by a dark line.

(B-F) Horizontal Sections showing the neural tube (NT) , the neural segments (Cj , Tj and the spinal ganglia (SGj under bright field (B) or green light excitation (C-F) after retrograde labeling of distal forelimb muscles with Dil. The results obtained from wild-type (+/+) or Hoxc-8 ' (+/-) embryos were identical, grouped together and called (+/-)•

(C and D) E12.5 embryos, in which MNs that supply the forelimb distal premuscle masses are observed in a R-C area extending from C5 to T2. In some but not all embryos, MNs are also observed in the most anterior segments C₃--4.

(Upper part of C) . The intensity of the signal depends on the time of fixation and diffusion of the dye, therefore this technique has no quantitative significance.

Differences in the R-C distribution of MNs within individual pools, or in the number of labeled MNs, are frequently noticed even between the right and the left part of the same embryo . (E and F) E14.5 embryos, in which MNs projecting to forelimb distal muscles are exclusively found in C-J -Q in the control embryos, as was observed for the newborn and adult animals. For all the null Hoxc-8 mutant animals, MNs were not found to be restricted to C₇_g, but were observed in C5- T . Sensory neurons from adjacent R-C segments are known to project to the same cutaneous territories (Kandel, et al . , 1991). Thus, the number of labeled spinal ganglia is prone to a slight variability. E12.5 embryos (+/+ and +/- n=8 , -/- : n=5) ; E14.5 embryos (+/+ and +/-: n=10, -/-: n=8.

E12.5 myf-5^{lacZ +} embryos {+/- n=3 , -/- : n=3); E14.5 myf-5 ' embryos (+/-: n=2 , -/-: n=3)

NT, neural tube; SG, spinal ganglion. Scale bar, 100 μm.

Figure 4. MN Deficits in Hoxc-8 Mutants.

The histology of motor columns was observed on Nissl- stained (B, C, B and F) or Hemalun-Eosin-Safran-stained (G- N) transverse sections at the level of Cg , where the gene is highly expressed in MNs. (B-C and E-N) Transverse sections of spinal cords at the level of Cs at various postnatal and embryonic stages. The outline of the ventral horn (B or the motor column (E, G, and K) of the heterozygote was translated by caique onto the photograph of the corresponding mutant animal to help visualize morphological alterations. (A) A representative example of the width reduction of the spinal cord, in the region of C₇-T1, that was observed in adult mutant animals.

(B and C) An adult mutant animal, for which the reduction in the width of the spinal cord is a direct consequence of a size reduction in the ventral grey matter.

(D) The mean number of MNs per Hemi-section of 12μm was determined in C₇ , Cg and Tτ_ of +/-(n=3) and -/- (n=3) newborn animals, as described in Experimental Procedures. This number was significantly lower in homozygotes compared to heterozygotes (P = 0.0016). The variation was significant among the three -/- animals (P = 0.0006), but was not significant among the three +/-animals (P 0.1356).

(E and F) The reduced number of Nissl-stained cells at birth results in a disappearance of the ventral enlargement of the spinal cord.

(G, H and K, L) The size reduction to the motor column is observed at embryonic stages E16.5 and E14.5.

(I and J) High magnification of a section corresponding to (G) and (H) , respectively. At E16.5, it is possible to distinguish the MNs from neighboring cells, and their number is reduced at this stage.

(M and N) High magnification of a section corresponding to (K) and (L) , respectively. At E14.5, although maturation of MNs is incomplete, a reduction was observed in the number of cells with a larger nucleus, surrounded by a shadowed eosin-stained cytoplasm which were considered to be MNs.

Scale bar, 50 μm.

Figure 5. MN Deficits Result from Apoptosis.

Picnosis and DNA fragmentation were observed in situ on transverse sections at the level of Cg , where MN deficits had been observed.

(A and B) In heterozygotes and homozygotes, apoptotic bodies (arrow heads) are observed within the motor column (MNs are brown, see Figure 6) , on sections counterstained with hemalun. The number of apoptotic cells is greatly increased in the mutant. The X-Gal staining has a spotty granular aspect, as previously noticed in CNS cell populations (Brustle, et al . , 1995).

(C and D) Apoptosis within the motor columns. DNA fragmentation is revealed by the TUNEL method. Apoptotic cells are stained green; MNs are stained red, using the pan- antibody that recognizes ISL1 and ISL2 , and a TRITC- conjugated secondary antibody. Apoptotic cells are exclusively detected in the motor columns, and their number is again significantly increased in the mutant.

(E and F) A triple-labeling experiment analyzed by confocal scanning microscopy (see Experimental Procedures) . Apoptotic cells are green, MNs red and lacZ-expressing cells blue. LacZ expression patterns, revealed by either i munohistochemistry or X-gal staining, are identical. The proportion of β-gal positive cells is higher in the mutant, as all MNs in the field express the lacZ gene (see also Figure 6D) . In the heterozygote, not all MNs express the lacZ gene (see also Figure 6C) . Apoptosis is certainly accompanied by digestion of proteins, resulting in a decrease or absence of antigenicity . Nonetheless, it is possible to observe cells that emit the three excitation wavelengths (arrow head) , therefore some MNs that still express the Hoxc-8 protein are in the process of dying.

(G and H) Confocal images of motor columns detected with the antibody to ISL 1/2. The number of Mns is lower in the mutant, therefore the two pools of Mns are difficult to distinguish; the columnar organization is also disrupted.

(I and J) Same technique as in (C and D) . Some DNA fragmentation is observed in the heterozygote at E14.5, whereas a significantly greater number of apoptotic cells (arrow heads) is detected in motor columns of the mutant.

(K and L) Immunodetection of ISL1 and ISL2 at E15.5. The number of Mns in the mutant is lower when compared to the heterozygote.

E13.5 embryos, +/- : n=4 , -/- : n=3 ; E14.5 embryos, +/- : n=2 , -/- : n=2 ; E15.5 embryos, +/- : n=2 , -/- : n=l.

Scale bar, 20 μm. Figure 6. LacZ Expression in Mns.

The expression of the lacZ reporter gene in Mns is detected by visualizing the β-galactosidase activity (blue) on immunostained transverse sections, using a polyclonal antibody to both ISL1 and ISL2 MN markers (brown) .

(A and B) At the level of C , expression of lacZ is detected in motor columns in both +/- (A) and -/- embryos (B) .

(C and D) High power magnification of (A and B) , respectively. Two different groups of Mns are observed. In the heterozygote, the most dorsal group contains a mixed population of MNS, in which some express the reporter gene while others do not. A more ventral population contains MNS that are mainly β-gal(-). In the homozygote, the vast majority of Mns express the lacZ gene, regardless of the position they occupy along the dorso-ventral axis. Moreover, the overall number of Mns is lower in the mutant (see Figures 5G and 5H) .

(E and F) The altered expression of the lacZ gene in Mns at Cg is also observed at E12.5. At this stage, although the number of Mns is not yet reduced in the homozygous mutant embryos, the proportion of lacZ-expressing Mns is higher.

E13.5 embryos, +/- : n=4 , -/- : n=5 ; E12.5 embryos, +/- : n=2 , -/- : n=2.

Scale bar, 50 μm. Figure 7: Sequence alignment of the whole Hoxc-8 and Hoxc-9 protein is depicted in Figure 7 and demonstrates a 29% homology. (ALIGN; gap penalties: -12/-2 ; GAS 235)

Figure 8: Figure 8 shows a 73% homology of the amino acid sequence between homeodomains of Hoxc-8 and Hoxc-9 proteins. (ALIGN; gap penalties: -12/-2 ; GAS 256)

DETAILED DESCRIPTION OF PREFERRED EMBODIMENT

A functional nervous system relies on precise neuronal circuits between neurons and neuronal or muscular targets. During embryonic development, neurons acquire positional genetic information to correctly connect with their final target throughout lifespan.

A loss-of-function mutation for the homeotic gene Hoxc- 8 generated in the mouse resulted in morphological alterations of axial skeletal segments, revealing a genetic mechanism of regionalization of the mammalian embryo (Le Mouellic, et al., 1992). In addition, some Hoxc-8 deficient mice survived and showed a fully penetrant phenotype from the end of gestation, namely a prehension deficiency of the forepaws, which was recently attributed to a nervous deficiency. Hoxc-8 is highly expressed in motoneurons within spinal cord segments C7 and Tj. At embryonic day E12.5 for Embryon at date 12.5, after mating, the initial distribution of motor pools that project to the forelimb distal flexor and extensor muscles spreads from C5 to T2. By E13.5, these pools are restricted to C7 and Cg . In Hoxc- 8 null mutant embryos, the distribution of the same pools extends from C5 to Tl. Concomitant with synaptogenesis, an increased apoptosis also occurs at E13.5 and E14.5 in motoneurons from C7-T1. Thus, inactivation of Hoxc-8 leads to spatially restricted motoneuron deficits and modified somatotopic maps of forelimb muscles within the spinal cord, leading to the prehension deficiency (Tiret, et al . , submitted) .

This transgenic mouse has the interesting peculiarity to display a locally restricted deficit of motoneurons, i.e. minus 30-40% in the neck spinal cord segments known to innervate the forelimb distal muscles. As a consequence of an altered innervation, a phenotype compatible with survival of animals is easily identified by the permanent clenching of their fingers. Thus, it can be considered to visibly restore the loss of movement control by the screening of any conceivable treatment at different life periods.

The Forelimb Phenotype of Deficient Mice for Hoxc-8 Correlates with its Expression in Cervial Motor Columns

Most of the inbred 129/Sv and more than 90% of the outbred C57Bl/6xDBA/ ; 2 Hoxc-8 mutants died within two days after birth (Le Mouellic, et al., 1992), emphasizing the influence of the genetic background (Tiret, et al . , 1993). Survivors shared a reproducible phenotype: their fingers were clenched (Figure IA) . The mutational strategy chosen resulted in the replacement of the Hoxc-8 gene by the E. Coli lacZ reporter gene. It was, therefore, possible in heterozygotes (+/-) and null mutants (-/-) to observe cells that express or should have expressed the Hoxc-8 endogenous gene by monitoring the β-galactosidase activity in embryonic cells (Le Mouellic, et al., 1992). In heterozygotes, the lacZ expression pattern was identical to the Hoxc-8 expression pattern previously studied by in situ hybridization (Uset, et al., 1987; Breier, et al., 1988; Le Mouellic, et al., 1988) and im unostaining (Awgulewitsch and Jacobs, 1990) . This result indicated that the fine function of cis-regulatory sequences had not been modified by the insertion of the transgene. Hoxc-8 expression at embryonic day 8.5 (E8.5) was detected in the allantois, the segmental plate mesoderm and the neurectoderm. From E9.5 onwards, stable expression in the neural tube was detected posteriorly to the 8th pair of somites, in the intermedio- lateral zone and in the ventral horns. Boundaries of lacZ expression domains were identical in heterozygotes (+/-) and homozygotes (-/-) . The sclerotome from the 8th somite will form the 3rd cervical vertebra. β-galactosidase activity was observed in more rostral segments as well; blue cells were visualized posteriorly to the second cervical vertebrae in newborn animals (Figure IB) . Observation of transverse sections at the level of the anterior cervical region revealed that between the second and the fourth cervical vertebrae, a small number of blue cells expressed the gene at a low level. A staining in this R-C region was also observed at E13.5, E14.5 and E15.5 (data not shown), and confirmed that the initial anterior limit of expression did not change during development. An increased sensitivity of the enzymatic reaction could account for the detection of expressing cells in the most rostral cervical segments. Posterior to the fourth cervical vertebra, Hoxc-8 expression increased and was maximal in the thoraco-cervical region C5- Ti, and the most intense expression was detected at the level of C7-T1. A similar pattern of Hoxc-8 or lacZ- expressing cells was observed in +/- embryos (data not shown) . Hoxc-8 expression was particularly observed in the lateral motor columns that, in the C₇-T₁ region, contain MNs that supply forelimb distal muscles. We, therefore, hypothesized that the prehension phenotype in Hoxc-8 mutant animals may be directly related to the normally strong expression of Hoxc-8 in the R-C segments of the developing neural tube that innervate the forelimb.

The Spatial Organization of Motor Pools Innervating Forelimb Distal Muscles is Altered in Hoxc-8—^— Mice Segments C₇-T1 contain MNs that innervate muscles controlling finger movements. Hoxc-8 ^/ animals cannot execute fine movements with their hands, leading to a partial loss of the prehension function (Le Mouellic, et al., 1992; see also Figure IA) . They can hold a food croquette clutched in their limbs, but are unable to cling to a grid. This phenotype has a complete penetrance, and suggests that innervation of the forelimb is somewhat defective. We, therefore, undertook an analysis of the innervation of the muscle flexor digitorum profundus and the muscle extensor digitorum communis. These muscles, which are responsible for finger movements, receive, in mammals, projections from MNs that are found in the LMC throughout the 7th and 8th cervical segments (C7_ ) (Baulac and Meininger, 1980; Scarisbrick, et al . , 1990; Curfs, et al., 1993 and ref therein) . The identity of a MN is given by the target it innevates. Retrograde labeling of MNs were thus performed in wild-type and mutant adult mice, using the non- toxic subunit B of cholera-toxin conjugated to horseradish peroxidase (CTB-HRP) , or the 1 , 1 ' -dioctadecyl-3 , 3 , 3 ' , 3 ' - tetramethylindocarbocyanine perchlorate (Dil; see Exerperimental Procedures) . A neuronal distribution similar to what has been described in the rat was observed in spinal cords of wild-type and heretozygous adult and newborn mice (Figures 2A and left part of Figure 2B) . The innervation pattern of the same muscles was subsequently investigated in the Hoxc-8 null mutants. Staining by these two techniques clearly showed that MNs projecting to distal muscles were localized in a wider R-C domain compared to wild-type, extending from C5 to Ti (Figure 2C) . In two of the mutants, MNs were also labeled in C4 or C₃ (see the right part of Figure 2B) . The observation that muscles received projections from MNs from C3-₄ in some but not all animals could be explained by a variable expressivity of the mutation, as previously described for morphological skeletal abnormalities (Le Mouellic, et al., 1992; Tiret, et al., 1993) . By a retrograde CTB-HRP study on adult control animals, we confirmed the existence of a discrete subdivision in the LMC of the mouse. For this purpose, the CTB-HRP was injected separately into either flexor or extensor muscles, and the MNs supplying these muscles have been symbolized in drawings by green (FLEX-MNs) and red (EXT-MNs) points for the respective muscle groups (Figure 2B) . As in the rat, EXT-MNs occupied a constant position at the border of the white and grey matter of the spinal cord whereas FLEX-MNs were found more medially (Curfs, et al . , 1993). A similar analysis in the Hoxc-8 mutants revealed an altered transverse-localization of MN somata. Some FLEX-MNs and EXT-MNs no longer respected their specific position, and some displayed a reverse transverse location with respect to their target muscles. These ectopic MNs were observed in all the mutants analyzed, but their number and position tended to vary, which again suggested a differential expressivity of the mutation. In contrast, the sensory innervation pattern was not detectably altered by the mutation.

The Mutation Interferes with the R-C

Refinement of the Early Embryonic Innervation Pattern

Data obtained from adults (Figure 2) and newborn mutant animals indicated that the abnormal pattern of innervation of distal forelimb muscles was already established at birth. Moreover, it was possible to observe the clenching of the fingers in young mice born at E18.5 after Caesarean, suggesting that the neuro-muscular disorder had occurred during embryogenesis. These observations led to an analysis of the innervation pattern of forelimb muscles during embryogenesis using retrograde labeling experiments. The quality of the retrograde labeling at these early embryonic stages depended upon the size and site of deposit of the Dil crystal. For this reason, we performed some experiments in embryos carrying the lacZ gene in the myf-5 locus, which encodes the earliest marker expressed in premuscular tissue (Tajbakhsh and Buckingham, 1994). As Hoxc-8 and myf-5 genes were not located on the same chromosome, we had bred these two mutants to obtain Hoxc-8+/+, +/- or -/- embryos in a myf-5 lacZ '/+ embryos were identical, showing that the myf-5 mutation had not interfered with our experiments. At E12.5,

MNs supplying distal forelimb muscles were localized in a

C5-T₂ domain in both controls and mutant embryos (Lower part of Figure 3C and Figure 3D) . In some embryos, MNs were labeled in the most rostral segments C4 and C3 (Upper part of Figure 3C) . A subsequent R-C restriction was noticed at E14.5 in wild-type embryos, in that only MNs included in the C7-g segment were retrogradely labeled (Figure 3E) . In contrast, the labeled MNs of Hoxc-8 mutants remained included in a wide C₅-T1 area (Figure 3F) with individual variations observed for MNs included in C3_4 segments. Heterozygotes had a normal viability, did not display any phenotype prehension; the organization of motor pools was not modified. Therefore, heterozygotes were assimilated to control animals, and provided, in addition, the lacZ reporter gene as an easily detectable cellular marker. The same retrograde labeling experiments were performed in E13.5 embryos. This stage corresponded to the first day at which the precise regognition of individual distal forelimb muscles was possible. In control embryos, distal muscles were subdivided into two distinct groups. The extensor carpi radialis received axons from C5-C6 MNs while all the other distal muscles were innervated by C7~g MNs. By contrast, these latter distal muscles of mutant E13.5 embryos were innervated by C₅-T₁ MNs (data not shown) .

The mature pattern of innervation of muscles was therefore established in wild-type embryos prior to the apoptotic process. The Hoxc-8 mutation interferred with the establishement of this proper connection pattern.

Spatially Restricted Motoneuron Deficits in Hoxc-8 Mutants

A constant reduction of width was observed in the posterior cervical segments of spinal cords dissected from

Hoxc-8 ' adult and newborn animals (e.g. in Figure 4A) .

Because Hoxc-8 is strongly transcribed in MNs from segments

C7 to Ti, we focused our analysis on this R-C region.

Analyses of transverse sections revealed a size reduction in the ventral grey matter (Figures 4B and 4C) , and this morphological alteration was already apparent at the time of birth (Figures 4E and 4F) . The relative number of MNs within the lateral motor column were then compared between control and null mutant newborn mice (see Experimental Procedures) . The mutation resulted in a MN deficit (Figure 4D) , with the average percentage of missing MNs increasing from C7 to Ti (C7, 26%; Cg , 34% and Ti, 44%). Moreover, a significant variation in the number of MNs was detected among the three mutant animals examined (see legend of Figure 4D) , presumably reflecting a variable expressivity of the mutation. The morphology of the motor column and the number of MNs in C₇-T1 was further assessed at earlier stages of embryogenesis. At E16.5, and atrophy of the motor column could be observed on HES-stained sections (Figures 4G and 4H) and resulted from a reduction in the number of MNs (Figures 41 and 4J) . At E14.5, a size difference of the motor column was detectable between heterozygous and homozygous embryos (Figures 4K-N) , and this suggested that a reduction in the number of MNs had already occurred by E14.5.

MN Deficits Result from Increased Cell Death between E13.5 and E14.5

During embryogensis, neurons pass through a regressive step of cell death (Oppenheim, 1991) which, in the nervous system, involves different degeneration mechanisms such as apoptosis or necrosis (Clarke, 1990). Analysis of frozen sections stained with hemotoxylin revealed the presence of masses of dense chromatin exclusively localized within the motor columns at E13.5 and E14.5. This morphology is indicative of nuclear fragmentation and was also observed in +/- embryos, but to a much higher extent in -/- embryos (Figures 5A and 5B) . Picnotic nuclei is one prominent feature of apoptosis, therefore, we looked for DNA fragmentation in situ using the terminal deoxynucleotidyl transferase dUTP nick and labeling technique (TUNEL) . Cells positive for DNA fragmentation were visualized specifically in motor columns of E13.5 and E14.5 +/- and -/- embryos, and the number of fragmented nuclei detected by this technique was much higher in the -/- versus +/- control embryos (Figures 5C, 5D, 51 and 5J) . The topographic localization suggested that among the apoptotic cells were MNs. Moreover, confocal scanning microscopy allowed for co- detection of fragmented DNA on ISL 1/2-immunostained sections, and thereby demonstrated DNA fragmentation within MNs (Figure 5E and 5F) . However, cell death did not strictly result from the complete loss of the endogenous gene product, as apoptotic lacZ-positive MNs were visualized in heterozygotes as well (arrow head in Figure 5E) . As a result of the significant number of dying MNs, the shape of the motor columns was modified, and dorsal and ventral MNs group boundaries were difficult to distinguish (Figures 5G and 5H) . In contrast, no fragmentation could be observed in MNs at E12.5 (data not shown), and this correlated with the absence of apoptotic bodies in cells within motor columns at this stage. Therefore, the onset of cell death was not modified by the mutation, but the degree of apoptosis in mutants was significantly increased between E13.5 and E14.5, resulting in a marked reduction of MNs by E15.5 posteriorly to C7 (Figures 5K and 5L) . LacZ Expression in Motoneurons

Given the Hoxc-8 expression in motor columns (Figure 1) , we sought to determine its precise expression in MNs along the R-C axis during neurogenesis . In the chick, all subpopulations of MNs express Isll and/or Isl2 (Tsuchida, et al . , 1994). Therefore, lacZ-positive cells that express Isll/2 in the ventral horns are MNs that express, or should have expressed, Hoxc-8. Co-detection experiments for the β- galactosidase and ISL 1/2 markers were performed at E13.5, at which stage precise recognition of neural segments in the vertebrae was possible. In the most rostral segments of the heterozygote, from C₃ to CQ , Hoxc-8 was detected in cells of the neural tube (Figure IB), but not in MNs. β-galatosidase activity in non-MNs cells indicated that other neurons and/or glial cells express Hoxc-8. Posteriorly from C7 , lacZ was expressed in non-MNs cells, as well as MNs.

Expression of Hoxc-8 has been previously observed in MNs by immunostaining at E19.5 (Awgulewitsch and Jacobs, 1990).

The number of MNs that expressed the lacZ gene increased from C₇ to i (data not shown) , and at the level of C8 , both lacZ-expressing and non-expressing MNs made up the motor column (Figure 6C) . This mosaic was also observed at E12.5

(Figure 6E) . In the null mutants, the anterior limit of

Hoxc-8 expression in MNs was also C₇. Although the overall number of MNs posterior to C₇ was reduced at E13.5 (Figure

6B, also Figure 5) , it was readily observed that the proportion of MNs that expressed lacZ was dramatically increased (Figure 6D) . This observation was confirmed at E12.5, in that nearly all MNs posterior to Cg expressed the lacZ gene in -/- embryos (Figure 6F) .

Immunostaining of transverse sections from E13.5 heterozygous embryos using a monoclonal antibody raised against Hoxc-8 demonstrated that all MNs expressed Hoxc-8 in C7_;Cg and Tl segments (data not shown); therefore, the lacZ was a faithfull reporter gene in the homozygotes, but not in the heterozygotes in which the number of lacZ copy seemed to be a limitant parameter for the detection of the β- galactosidase activity in motoneurons.

Hox Genes and Neural Precursor Behavior

Through fine analysis of β-galactosidase activity, we have confirmed that the R-C expression domains of Hoxc-8 in the wild-type or recombined alleles of both heterozygote and homozygote animals are identical (Utset, et al., 1987; Le Mouellic, et al . , 1988; Breier, et al., 1988; Awgulewitsch and Jacobs, 1990; Le Mouellic, et al . , 1992). The expression of the lacZ gene is primarily located in the ventral and intermediate region of the spinal cord and extends rostrally to C2 • However, a co-detection with the MN marker, ISL1/2, revealed that MNs expressing the Hoxc- 8/lacZ reporter gene are not found anteriorly to C7. Lineage experiments in the chick have revealed that among multipotential precursors within the ependymal layer, only 15% give birth to MNs, in addition to others neurons and glial cells (Leber, et al., 1990). Thus, the absence of lacZ-expressing MNs from C2-6 iⁿ homozygotes is not due to the absence of a functional Hoxc-8 gene and emphasizes the point that discrete subtypes of progenitor cells, which should have expressed Hoxc-8, have not been modified by the mutation at any R-C level.

The absence of the Hoxc-8 product appears to interfere with the identity of progenitor cells.

MN Identity and Cell Death in the Refinement of Connectivity Patterns

An important criterion used to identify MNs is the muscle that they innervate (Eisen, 1994). Our earliest retrograde labeling experiments were performed at E12.5, when the number and position of MNs are manifestly identical in wild-type and mutant animals (Figures 6E and 6F) . At this stage, the innervation of forelimb distal muscles is wide-ranging and imprecise; MNs from all the segments of the brachial region C5 to T₂ are recruited (Figure 3C) . In control embryos, this connectivity pattern undergoes a R-C refinement between E12.5 and E14.5 at which time axon terminals arising from C₅ and T₂ are lost, leaving the definitive innervation to occur at C₇ and Cg (Figure 3E) . Motor pools that spread over a wider R-C domain at early embryonic stages have been described for hindlimb innervation in the Xenopus embryo and for forelimb innervation in the chick embryo (Lamb, 1976; Lamb, 1977; Pettigrew, et al., 1979). A R-C gradient of transgene expression has revealed fundamental differences among myoblasts, and supports the idea that mechanisms of MN selection involve matching coordinates (Donoghue, et al . , 1992) . We have shown that the innervation pattern was corrected at E13.5 in wild-type embryos. We have demonstrated that in the mutants a modification in the positional values in the MNs of C7_g abolishes their ability to be specifically selected in response to recognition cues from muscle target fibers which, in forelimbs, do not express Hoxc-8. Thus, the Hoxc-8 mutation has a detectable effect on the establishment of the initial connectivity pattern. During their entire lifespan, forelimb distal muscles in the mutant animals will continuously receive projections from C₅-T₁ MNs (Figure 2). Because Hoxc-8 is not normally expressed in C5-.6 MNs, a change of their identify is unlikely, although a putative role of non-MN neighboring cells cannot be ruled out. The loss of information specified by Hoxc-8 would produce a reduced specific affinity of C7_g MNs, resulting in a broader R-C population of MNs that are equally competitive in the selection of persistent synapses.

The elimination of some nerve terminals from a developing muscle may involve withdrawal and degeneration of some branches of an axon, withdrawal and relocation of the axon itself, or complete degeneration of the axon and its MN. Electrophysiological studies of chick biceps muscle innervation suggests that MN projections do not invade several targets (Pettigrew, et al., 1979). Rather, during the period of refinement, naturally occurring cell death removes about 50% of the spinal MNs; however, the physiological significance of this regressive event is still not completely understood (Hamburger, 1975; Oppenheim, 1991) . We observed cell death at comparable times in the motor columns of control mouse embryos. Apoptotic cells were detected within the motor column at E13.5 and, to a lesser extent at E14.5 (Figure 5). Strikingly, the mutation of Hoxc-8 gene enhanced MN death from C₇ to Ti , and the ratio of eliminated cells correlated with the level of Hoxc- 8 expression in each segment. An indirect link between Hox gene expression and cell death has been previously suggested by experiments in transgenic mice; ectopic expression of Hoxc-8 at the level of anterior cervical segments (Ci) prevented the first dorsal root ganglia from degenerating (Charite, et al . , 1994). Furthermore, the number of surviving MNs has been shown to depend upon the limiting amounts of neurotrophic factors delivered by target cells (Oppenheim, 1991) . Because Hoxc-8 is not normally expressed in the distal forelimb, its inactivation probably does not affect the production of available neurotrophic molecules; however, the retrograde uptake of such molecules by C7_ MNs might be reduced due to a deficient mutual recognition within muscles.

MNs Specification and Combinatorial Molecular Determinants

Forelimb muscles receive their motor projections from the brachial LMC. This column is subdivided into MCi and LMCrn which innervate extensor (dorsal) and flexor (ventral) muscles, respectively (reviewed by Lumsden, 1995). In the Hoxc-8 null mutants, some MNs supplying extensor muscles are found in a medial position, whereas others that supply flexor muscles are observed in a lateral position, which reveals a morphological disruption of the columnar organization in the mutant animals. Precise connectivity patterns result from the selection of distinct pathways by the growth cone in response to environmental guidance cues (Lance-Jones and Landmesser, 1981; Ferguson, 1983; Tosney and Landmesser, 1984; Tosney, 1987; Phelan and Hollyday, 1990; Lance-Jones and Dias, 1991; see review in Goodman and Shatz , 1993). The alteration of a specific recognition between the navigating growth cone and the guidance cues spread within the mesenchy e suggests a change of MN identities in the Hoxc-8 mutants. In this regard, the role of the proximal limb mesenchyme, in which Hoxc-8 is expressed, could be further explored by generating tissue- specific mutations for Hoxc-8. The discovery of the LIM- family homeobox genes has provided new insights into putative molecular determinants involved in motor column organization. Distinct combinations of these genes differentially mark motor columns in chick embryos, or primary MNs in zebrafish embryos (Tsuchida, et al., 1994; Appel, et al . , 1995; Tokumoto, et al., 1995). The combinatorial expression of LIM genes is uniform in motor columns, and is not specific for individual motor pools along the R-C axis. Inactivation of both IslI and LimI results in developmental defects leading to early embryonic lethality (Shawlot and Behringer, 1995; Pfaff, et al., 1996) . Null mutant mouse embryos for the IslI gene are devoid of MNs and die at E10.5. Therefore, the investigation of a putative involvement of either IslI or LimI in specifying MN identity will require conditional inactivation of these genes such that embryos overcome these early phenotypes . Spinal cord reversals have shown that MNs acquire their positional identities along the R-C axis between stages 14 and 15 (Matise and Lance-Jones, 1996), before the first MNs leave the cell cycle (Hollyday and Hamburger, 1977). Hox genes are expressed in the primitive neural tube, whereas LIM-homeobox genes are sequentially expressed by stage 15 (Ericson, et al . , 1992; Tsuchida, et al., 1994). Thus, it is reasonable to think that presumptive MNs acquire their R-C and medio-lateral coordinates, at least in part, by the successive expression of Hox and LIM genes.

Hox Genes and Functional Refinement in Vertebrates

In all of the Hoxc-8 mutant animals, whether they survive or die soon after birth, we observed the abnormal clenching of fingers. This phenotype, together with the associated motor innervation defects, reveals a complete penetrance of the mutation. Coordination of movements results from a precise connectivity pattern between different integrative levels of the central nervous system. Multiple phenomena, whether they be mutually exclusive or synergistic, could account for the motor dysfunction observed in mutant animals. Firstly, an impaired myotatic reflex of forepaw muscles may result from a topographic mismatch between the unmodified sensory projections and MNs from abnormal R-C locations. Furthermore, the Hoxc-8 mutation may alter identities of interneurons from the intermediate zone of the spinal cord (Figures 1C and ID) , which could amplify the dysfunction of more elaborated spinal reflexes. Secondly, MNs that are located in C5--6 and Ti segments may not receive appropriate afferent inputs from descending tracts of the encephalon. Finally, the inversion of some MNs that innervate flexor and extensor muscles could cause them to transmit contradictory commands. Determining the importance of these different mechanisms will necessitate sophisticated studies of neuronal circuits. Nonetheless, our experiments confirm that subtle movements depend on the invariant restriction of motor pools, and demonstrate that the establishment of topographic maps depends on positional in formations encoded by Hox genes. Improper innervation of neck and back axial muscles by MNs of the MMC that normally express Hoxc-8 may similarly lead to abnormal function of these muscles and could explain the hunched profile and erect tail observed in some mutant animals (Figure IA) . Moreover, homozygotes do not ingest much milk, and the few that survive show growth disorders (Le Mouellic, et al., 1992). Some adult mutant males suffer from priapism. These two observations imply an imbalance in the control of vegetative functions by parasympathetic and sympathetic systems. In the thoracic region, Hoxc-8 is expressed in the column of Terni (data not shown) and in paravertebral ganglia (personal observations and Breier, et al., 1988). A similar change in the identity of sympathetic neurons would be sufficient to explain these serious physiological disorders. The presence of ectopic innervating MNs in C3--₄ (Figures 2 and 3) and the mean number of surviving MNs (Figure 4D) among mutants tend to vary. Therefore, the death or survival of mutant animals is probably a consequence of the degree of expressivity of the mutation.

The presence of Hox genes is a common feature of all metazoans, including diblastic animals that do not have a mesoderm layer (e.g. Schummer, et al., 1992), and this could possibly reflect stronger evolutionary constraints on the development of neurectoderm than on mesoderm. In the absence of the positional information encoded by Hoxc-8 , paraxial mesodermal cells interpret the remaining information to follow an alternative program of skeletal morphogenesis. In contrast, in the nervous system, Hoxc-8 inactivation induces an important increase in MN death. Likewise, in the segmented hindbrain, mutation of the Hoxa-1 gene induces the deletion, rather than the homeotic transformation, of rhombomere 5 (Carpenter, et al., 1993). The lack of substitution program in the nervous system demonstrates that, in restricted domains of the neural tube, cells can not be rescued by compensatory positional information provided by other Hox genes. Handicapping phenotypes of Hox mutant mice reveal sharp domains of expression in the neural tube, where the patterning role of each Hox gene is essential. Furthermore, the spinal cord is the predominant site of Hoxc-8 expression throughout adult life, and this suggests a continuous need for the positional information provided by Hoxc-8 for learning and the maintenance of movement control.

EXPERIMENTAL PROCEDURES Mouse Breeding and Determination of Genotypes

Embryos were obtained by crossing heterozygotes (+/-) from the inbred 129/Sv or the outbred C57Bl/6xDBA/2 strain, or +/- females with a -/- male from the outbred strain. The genotype was established for each animal by southern blotting of DNA from yolk sacs or tail biopsies as described (Le Mouellic, et al . , 1992), or by PCR. PCR primers 31AS (5'-CGTAGCCATAGAATTGGAG; SEQ ID N°l) and GED31 (5¹- GAGCTCCTACTTCGTCAAC ; SEQ ID N°2), 3 IAS and NEO (5'- CAGCAGAAACATACAAGCTG ; SEQ ID N ^β 3 ) were used to identify the endogenous and targeted loci, respectively. The amplification conditions were: denaturation at 94°C for 15

s, annealing at 61°C for 20 s and extension at 72°C for 30 s for 30 cycles. Taq polymerase and buffers were obtained from GIBCO BRL.

Staining for β-Galactosidase Activity X-Gal staining on whole-mount embryos was performed as previously described (Le Mouellic, et al., 1992). For X-Gal staining on transverse sections, embryos were fixed in 4% PFA for 20-30 min depending on the embryonic stage, washed in ice-cold PBS for 5 min, then immersed in a cryoprotectivc buffer, consisting of 30% sucrose, in PBS for few hours, embedded in OCT medium (Miles Laboratory) and cut into 10 μm sections on a cryotome. Sections were stained in X-Gal buffer for 12 hours at 35°C, then mounted in 1:1 PBS/Glycerol and examined under a Reichert-Jung microscope.

Retrograde Labeling Experiments

The CTB-HRP experiments and drawings were done as previously described (Curfs, et al., 1993; Dederen, et al., 1994) . When Dil was used, a solution of 5% DiICχg

(Molecular Probes # 282) in dimethylformamide was prepared.

For living newborn and adult animals, 0.3 μl of the solution was injected into the muscles. The time of diffusion was three day's for newborn animals. Because the dye covered greater distances in adults, the time of diffusion was one week. Animals were then anesthesized, transcardially perfused with an ice-cold 4% PFA solution for 15 minutes, and the spinal cords were then dissected. For retrograde labeling of MNs on fixed animals, a crystal of Dil was precisely introduced into the muscle mass of the distal part of the forelimb. When Hoxc-8^{+/ ~} - Myf-5^lacZ/+

—/— lacZ/- or Hoxc-8 ' - Myf-5 ' embryos were used, the deposit of the crystal was preceded by a 2 hour incubation in X-Gal buffer, which allowed for the detection of individual muscles. Embryos were then incubated at 37°C, either in 4%

PFA or in 0.1M PBS containing ImM NaN₃. Axonal diffusion was checked every two days under a microscope equipped with an epifluorescent unit. The time of diffusion was about 2 weeks for embryos stored in 4% PFA, and few days for embryos conserved in 0.1M PBS containing ImM NaN3. The materials from both adult and newborn animals were then embedded and cut into 60 μm sections on a freezing microtome. These were then mounted onto gelatin-coated slides and air dried. Sections obtained from embryos were prepared in the same way but were 10 μm thick. All sections were examined under a microscope equipped with an epifluorescent unit and the appropriate filter for TRITC was used.

Histology and MN Counts

Adult animals were deeply anesthesized using 0.020 ml of a 3% averting solution per gram of body weight, transcardiacally perfused with 20 ml of ice-cold PBS, followed by 50 ml of a 4% paraformaldehyde fixative solution. A bilateral laminectomy allowed for dissection of segments C4-T2 of the spinal cord. Dorsal and ventral nerve roots were preserved and used as indicators of the R-C level of each section. An overnight post-fixation in the same fixative was followed by washes in PBS. Newborn animals were deeply anesthesized and killed, the skin was removed and animals were immersed in a large volume of Bouin fixative for 4-5 days, then rinsed in 70% alcohol. E14.5 and E16.5 embryos were fixed 6 hours in Bouin fixative and rinsed in 70% alcohol. The following steps were identical for adult spinal cord, whole-mount newborns and embryos. An alcoholic dehydratation was followed by a paraffin embedding. Embryos were then cut into sections of 7μm. Sections were rehydrated and stained with Hemalun-Eosin-Safran (HES) .

Adult spinal cords and newborns were serially cut at 12μm, mounted on slides and stained with a 0.25% Thionin solution (Sigma#T3387 ) for 5 minutes, followed by a 30 second differentiation step in 70% alcohol, which allowed for the selective visualization of Nissl bodies. MNs within the left and right lateral motor columns from C₇T₁ were counted independently on every other section to eliminate the risk of counting the same MN on two adjacent sections. Nissl- stained cells with a large nucleus, a nucleolus and a cellular size of >20μm were considered as MNs. Data analyses were carried out using the General Linear Procedure (Analyses of Variance) in the SAS software (SAS Institute Inc. 1989). Differences among numbers of MNs were considered significant when P < 0.05. Numerical results of Figure 4D were expressed as mean +/- S.E.M.

Labeling of Apoptotic Cells

Frozen sections of embryos were thawed, fixed in 4% PFA for 10 minutes and washed in PBS. The primary polyclonal 40 antibody that recognizes both ISLI and ISL2 (K5, see

Tsuchida, et al., 1994) was applied overnight at 4°C at a dilution of 1:2700. Sections were then processed according to manufacturer's instructions (In situ Cell Death Detection kit; Boehringer Mannheim #1684795) for in situ detection of DNA fragmentation. A TRITC-conjugated anti-rabbit serum (Jackson ImmunoResearch Laboratories, Inc. #111-025-003) was then used at a dilution of 1:100. Stained sections were examined under a microscope equipped with an epifluorescent unit containing the appropriate filter for FITC band TRITC.

For Figures 5E and 5F, a mouse monoclonal anti β- galactosidase antibody (Boehringer Mannheim #1083104) was added along with the primary α-ISLI/2 antibody at a dilution of 1:25. A Cy-5-conjugated anti-rabbit serum (Jackson ImmunoResearch Laboratories, Inc. #111-075-003) and a TRITC- conjugated anti-mouse serum (Sigma #T5393) were then used at dilutions of 1:25 and 1:100, respectively. Confocal microscopy of samples labeled with fluorophores was performed with a Confocal Laser Scanning microscope (Leica Instruments, Heidelberg., Germany), which uses an argon- krypton laser operating in multi-line mode.

Immunohistochemistry and Codetection with β-galactosidase

Embryos were dissected from uterin horns, rinsed for 10 min in ice-cold PBS, fixed 20 to 40 minutes in PBS containing 4% paraformaldehyde at 4°C. They were then rinsed in PBS and immersed in PBS containing 30% sucrose at 4°C overnight. They were then cut into 10 μm sections with a cryostat. Sections were allowed to dry and stored at - 80°C. Sections to be analyzed were thawed, fixed 5 minutes

in acetone, washed in PBS and incubated overnight at 4°C with the primary antibody, (K5, 1:5000) diluted in PBS containing 0.1% BSA and 1% inactivated sheep serum (Sigma #G9023). Sections were then rinsed in PBS 0.1M. Endogenous background was reduced by using an avidin/biotin blocking kit for biotin (Vector #5P-2001) . A secondary biotin- conjugated anti-rabbit antibody (Tebu #'L43015) was then applied for 1 hour at RT at a dilution of 1:1000. Sections were rinsed with PBS 0.1M. Endogenous peroxidase activity was further blocked by a short 5 minute incubation in methanol containing 0.3% H₂O₂, followed by a 1 hour incubation of sections in ExtAvidin^~ Peroxidase (Sigma #E2886) . Staining was performed by using the AEC detection kit (Vector #SK-4200) . When co-analysis of β-galactosidase activity was also desired, the sections were incubated at 35°C for 3 hours in X -Gal buffer after the first antibody step.

* * *

From gastrulation until adulthood, Hox genes are transcribed in overlapping domains of neuroectodermal derivatives; Hoxc-8 is highly expressed in motoneurons within spinal cord segments C7 to Ti. At embryonic day E12.5, the initial distribution of motor pools that project to the forelimb distal flexor and extensor muscles spreads from C5 to T2 • By E13.5, these pools are restricted to C7 and Cg in wild-type embryos but not in mutants. In Hoxc-8 null mutant embryos, concomitant with synaptogenesis, an increased apoptosis occurs at E13.5 in motoneurons from C7- Tl. Thus, inactivation of Hoxc-8 leads to spatially restricted motoneuron deficits and modified somototopic maps of forelimb muscles within the spinal cord, leading to a prehension deficiency. Our results demonstrate that Hox genes play an essential role in the establishment of a functional nervous system by providing positional specifications to motoneurons precursors.

Our results have shown that mutation disrupts various innervation processes of the mouse's forearm muscles.

Moreover, the inactivation of the Hoxc-8 gene results in motoneuron death, particularly in segments C7 , C8 and Tl, and in the embryonic stages E13.5 and E14.5. This is caused by a 20% to 40% lack of MNs which leads to the deficient operation of the hand, which is a simple and visible phenotype. Therefore, these animals are excellent models for neuromyopathies . Indeed, one can develop genetic therapies (cellular and/or pharmacological) that are simple and general so as to offset motoneuron death. A large specificity can be searched because neuronal death exists only in certain segments and within a specific time frame. Restoring the hand's grasping function is a simple test that allows people to evaluate whether or not a therapy is effective.

An initial approach could be based on the over- expression of various neurotrophic factors, either after the injection of producer cells, or directly from the factors themselves. These animals or embryos can also be used to cull active pharmacological molecules. In terms of cells, the aim would be either to prevent the motoneuron apoptosis or to stimulate proliferation of the MNs or certain neuron precursors.

A plasmid pGMD-DV has been deposited at the CNCM under the number 1-1774, October 9, 1996. This plasmid contained in E. coli, is described in Le Mouellic et al., 1992.

The following references are cited herein. Each reference is relied upon in its entirety and incorporated by reference herein.

Appel, B. , Korzh, V., Glasgow, E., Thor, S., Edlund, T. , Dawid, I. B. and Eisen, J. S. (1995). Motoneuron fate specification revealed by patterned LIM homeobox gene expression in embryonic zebrafish. Development. 121, 4117- 4125.

Awgulewitsch, A., Bieberich, C. , Bogarad, L. , Shashikant, C. , and Ruddle, F. H. (1990). Structural analysis of the Hox-3. I transcription unit and the Hox-3.2-Hox-3.1 intergenic region. Proc. Natl. Acad. Sci. USA. 87, 6428- 6432.

Awgulewitsch, A. and Jacobs, D. (1990). Differential expression of Hox-3. I protein in subregions of the embryonic and adult spinal chord. Development. 108, 411-420.

Baetge, E.E., Suh, Y.H., and Joh, T.H. (1986). Complete nucleotide and deduced amino acid sequence of bovine phenylethanolamine N-methyltransferase: partial amino acid homology with rat tyrosine hydroxylase. Proc. Natl. Acad. Sci. USA. 83, 5454-5458. Baulac, M. and Meininger, V. (1980). Topographical arrangement of motoneurons from the ventral muscles of the arm in rat. Neurosci. Lett. (Suppl.). 5, S87.

Blatt, C, Aberdam, D. , Schwartz, L. and Sachs, L. (1988). DNA rearrangement of a homeobox gene in myeloid leukaemic cells. EMBO J. 7, 4283-4290.

Breier, G., Dressier, G. R. and Gruss, P. (1988). Primary structure and developmental expression pattern of Hox-3.1, a member of the murine Hox 3 homeobox cluster. Embo J. 7, 1329-1336.

Brύstle, O. , Maskos, U. and McKay, R. D. G. (1995). Host- guided migration allows targeted introduction of neurons into the embryonic brain. Neuron. 15, 1275-1285.

Burke, A., C, Nelson, C, E., Morgan, B. , A. and Tabin, C. (1995). Hox genes and the evolution of vertebrate axial morphology. Development. 121, 333-346.

Carpenter, E. M. , Goddard, J. M. , Chisaka, 0., Manley, N. R. and Cappecchi, M. R. (1993). Loss of Hoxa-I (Hox-1.6) function results in the reorganization of the murine hindbrain. Development. 118, 1063-1075.

Charite, J., de Graaf, W. , Shen, S. and Deschamps, J. (1994) . Ectopic expression of Hoxb-8 causes duplication of the ZPA in the forelimb and homeotic transformation of axial structures. Cell. 78, 589-601.

Clarke, P. G. H. (1990) . Developmental cell death: Morphological diversity and multiple mechanisms. Anat. Embryol. 181, 195-213.

Curfs, M. H. J. M. , Gribnau, A. A. M. and Dederen, P. J. W.

C. (1993). Postnatal maturation of the dendritic fields of motoneuron pools supplying flexor and extensor muscles of the distal forelimb in the rat. Development. 17, 535-541.

DeChiara, T. M. , Vejsada, R. , Poueymirou, W. T., Acheson,

A., Suri, C, Conover, J. C, Friedman, B. , McClain, J., Pan, L. , Stahl , N., Ip, N. Y., Kato, A., and Yancopoulos, G.

D. (1995) . Mice Lacking the CNTF Receptor, Unlike Mice Lacking CNTF, Exhibit Profound Motor Neuron Deficits at Birth. Cell. 83, 313-322.

Dederen, P. J. W. C. , Gribnau, A. A. M. and Curfs, M. H. J. M. (1994). Retrograde neuronal tracing with cholera toxin B subunit: comparison of three different visualization methods. Histochemical Journal. 26, 850-862.

Donoghue, M. J., Morris-Valero, R. , Johnson, Y. R. , Merlie, J. P. and Sanes, J. R. (1992). Mammalian muscle cells bear a cell-autonomous, heritable memory of their rostrocaudal position. Cell. 69, 67-77. Duboule, D. (1994). Guidebook to the Homeobox Genes; edited by Denis Duboule, University of Geneva, Geneva, Switzerland, page 41, 151, and 152.

Eisen, J. S. (1994). Development of motoneuron phenotype. Annu. Rev. Neurosc. 17, 1-30.

Eisen, J. S.; Myers, P. Z.; Westerfield, M. (1986). Pathway selection by growth cones of identified motoneurones in live Zebrafish embryos. Nature. 320. 269-271.

Elshamy W.M. and Ernfors P. (1996). A local action of neurotrophin-3 prevents the death of proliferating sensory neuron precursor cells. Neuron, 16, 963-972.

Ericson, J., Thor, S., Edlund, T., T., J. and Yamada, T. (1992). Early stages of motor neuron differentiation revealed by expression of homeobox gene Islet-1. Science. 256, 1555-1560.

Ernfors, P., Lee, K. , and Jaenisch, R. (1994). Mice lacking brain-derived neurotrophic factor develop with sensory deficits. Nature. 368, 147.

Escary, J. L. , Perreau, J., Dumenil, D. , Ezine, S., and Brulet, P. (1993). Leukaemia inhibitory factor is necessary for maintenance of haematopoietic stem cells and thymocyte stimulation. Nature. 363, 361.

Ferguson, B. (1983). Development of motor innervation of the chick following dorso-ventral limb bud rotations. J. Neurosci. 3, 1760-1772.

Goodman, C. S. and Shatz, C. J. (1993). Developmental mechanisms that generate precise patterns of neuronal connectivity. Cell 72/Neuron 10 (Suppl.). 139-149.

Hamburger, V. (1975). Cell death in the development of the lateral motor column of the chick embryo. J. Comp. Neurol . 160, 535-546.

Hollyday, M. (1980). Organization of motor pools in the chick lumbar lateral motor column. J. Comp. Neurol. 194, 143-170.

Hollyday, M. and Hamburger, V. (1977). An autoradiographic study of the formation of the lateral motor column in the chick embryo. Brain Res. 132, 197-208.

Houenou, L. J., Turner, P. L. , Li, L. , Oppenheim, R. W. , and Festoff, B. W. (1995). A serine protease inhibitor, protease nexin I, rescues motoneurons from naturally occurring and axotomy-induced cell death. Proc. Natl. Acad. Sci. USA, 92, 895-899. Janssen et al . (1991). Immunology, 146, 35-39 .

Johnson, J. E., Qin-Wei, Y., Prevette, D. , and Oppenheim, R.W. (1995). Brain-Derived Proteins that Rescue Spinal Motoneurons from Cell Death in the Chick Embryo: Comparisons with Target-Derived and Recombinant Factors. J. Neurobiol., 27, 573-589.

Kandel, E.R., Schwartz, J. H. and Jessel, T.M. (1991).

Principles of neural sciences 3rd edn . Elsevier Science Publishing, New York.

Kessel , M. and Gruss, P. (1991). Homeotic Transformations of murine vertebrae and concomitant alteration of Hox Codes induces by retinoic acid. Cell, 67, 89-104.

Kessel, M. (1993). Reversal of axonal pathways from rhombomeres 3 correlates with extra Hox expression domains. Neuron. 10, 379-393.

Krumlauf, R. (1994). Hox genes in vertebrate development. Cell. 78, 589-601.

Lamb, A. H. (1976). The projection pattern of the ventral horn to the hind limb during development. Dev. Biol. 54, 82-99. Lamb, A. H. (1977). Neuronal death in the development of the somatotopic projections of the ventral horn in Xenopus. Brain. Res. 134, 145-150.

Lance-Jones, C. and Dias, M. (1991) . The influence of presumptive limb connective tissue on motoneuron axon guidance. Dev. Biol. 143, 93-110.

Lance-Jones, C. and Landmesser, L. (1981). Pathway selection by chick lumbosacral motoneurons during normal development. Proc. R. Soc. London Ser. B. 214, 1-18.

Land esser, L. (1978a). The development of motor projection patterns in the chick hind limb. J. Physiol. 284, 391-414.

Landmesser, L. (1978b). The distribution of motoneurons supplying chick hind limb muscles. J. Physiol. 284. 371- 389.

Le Mouellic, H. , Condamine, H. and Brulet, P. (1988). Pattern of transcription of the homeogene Hox.3. I in the mouse embryo. Genes. Dev. 2, 125-135.

Le Mouellic, H. , Lallemand, Y. and Brulet, P. (1992). Homeosis in the mouse induced by a null mutation in the Hox- 3.1 gene. Cell. 69, 251-264. Leber, S. M. , Breeedlove, S. M. and Sanes, J. R. (1990). Lineage, arrangement, and death of clonally related motoneurons in chick spinal card. J. Neurosci. 10, 2451- 2462.

Li, M. , Sendtner, M. , and Smith, A. (1995). Essential function of LIF receptor in motor neurons. Nature, 378, 724.

Lu sden, A. (1995). A LIM code for motoneurons? Current Biology. 5, 491-495.

Marshall, H., Nonchev, S., Sham, M. H., Muchamore, I., Lumsden, A. and Krumlauf, R. (1992). Retinoic acid alters hindbrain Hox code and induces transformation of rhombomere 2/3 into a 4/5 identity. Nature. 360, 737-741.

Martinou, J., Dubois-Dauphin, Michel, Staple, J. K. , Rodriquez, I., Frankowski, H., Missotten, M. , Albertini, P., Talabot, D. , Catsicas, S., Pietra, C. , and Huarte, J. (1994). Overexpression of BCL-2 in Transgenic Mice Protects Neurons from Naturally Occurring Cell Death and Experimental Ischemia. Neuron, 133, 1017-1030.

Matise, M. P. and Lance-Jones, C. (1996). A critical period for the specification of motor pools in the chick lumbosacral spinal cord. Development. 122, 659-669. Maulbecker, C. C. and Gruss, P. (1993) . The oncogenic potential of deregulated homeobox genes. Cell growth & differentiation. 4. 431-441.

McGinnis, W. and Krumlauf, R. (1992) . Homeobox genes and axial patterning. Cell. 68, 283-302.

Michaelidis, T.M. , Sendtner, M. , Cooper, J.D., Airaksinen, M.S., Hotmann, B. , Meyer, M. , and Thoenen, H. (1996). Inactivation of bcl-2 Results in Progressive Degeneration of Motoneurons, Sympathetic and Sensory Neurons during Early Postnatal Development. Neuron. 17, 75-89.

Middleton, G. , Nunez, G. , and Davis, A. M. (1996). Bax promotes neuronal survival and antagonizes the survival effects of neurotrophic factors. Development. 122, 695- 701.

Newell, et al . Nature, 347, 286-289 (1990).

Oppenheim, R. W. (1991). Cell death during development of the nervous system. Annu. Rev. Neurosci. 14, 453-501.

Oppenheim, R. W. (1991). Neurotrophic Survival Molecules for Motoneurons: An Embarrassment of Riches. Neuron, 17, 195-197. Pennica, D. , Arce, V., Swanson, T.A., Vejsada, R. , Pollock, R. A., Armanini, M. , Dudley, K. , Phillips, H.S., Rosenthal, A., Kato, A.C., and Henderson, C.E. (1996). Cardiotrophin- 1, a Cytokine Present in Embryonic Muscle, Supports Long- Term Survival of Spinal Motoneurons. Neuron. 17, 53-74.

Pettigrew, A., Lindeman, R. and Bennett, M. R (1979). Development of segmental innervation of the chick forelimb. J. Embryol. Exp. Morphol . 49, 115-137.

Pfaff, S. IL. , Mendelsohn, M. , Stewart, C. L. , Edlund, T. and Jessell, T. M. (1996). Requirement for LIM homeobox gene IslI in motor neuron generation reveals a motor neurondependent step in interneuron differentiation. Cell. 84, 309-320.

Phelan, K. A. and Hollyday, M. (1990). Axon guidance in muscleless chick wings: the role of muscle cells in motoneuronal pathway selection and muscle nerve formation. J. Neurosci. 10, 2699-2716.

Romanes, G. J. (1951) . The motor cell columns of the lumbosacral spinal cord of the cat. J. Comp. Neurol. 94, 313- 363.

Romanes , G . J . ( 1964 ) . The motor pools of the spinal cord . Prog . Brain Res . II , 93 -119 . Sagot, Y., Tan, S.A., Baetge, E., Schmalbruch, H. , Kato, A.C., and Aebischer, P. (1995). Polymer Encapsulated Cell Lines Genetically Engineered to Release Ciliary Neurotrophic Factor Can Slow Down Progressive Motor Neuronopathy in the Mouse. Euro. J. of Neurosci. 7, 1313-1322.

Scarisbrick, I. A., Haase, P. and Hrycyshyn, A. W. (1990). The arrangement of forearm motoneurons in young and adult rats and the possibility of naturally occurring motoneuron death. J. Anat. 171, 57-67.

Schummer, M. , Scheurlen, I., Schaller, C. and Galliot, B. (1992). HOM/HOX homeobox genes are present in hydra (Chlorhydra viridissi a) and are differentially expressed during regeneration. The EMBO Journal. 11, 1815-1823.

Sendtner, M. , Schmalbruch, H., Stockli, A., Carroll, P., Kreutzberg, G. W. , and Thoenen, H. (1992) Ciliary neurotrophic factor prevents degeneration of motor neurons in mouse mutant progressive motor neuronopathy. Nature, 358, 502.

Shashikant, C. S., Bieberich, C. J., Belting, H. G. , Wang, J. C, Borbely, M. A. and Ruddle, F. H. (1995). Regulation of Hoxc-8 during mouse embryonic development: identification and characterization of critical elements involved in early neural tube expression. Development. 121, 4339-4447. Shawlot, W. and Behringer, R. R. (1995) . Requirement for LIM1 in head-organizer function. Nature. 374, 425-430.

Simeone, A., Acumpora, D. , Nigro, V., Foiella, A., D'Esposito, M. , Stornajuolo, A., Mavilio, F. , and Boncinelli, E. (1991). Differential regulation by retinoic acid of the homeobox genes of the four HOX loci in human embryonal carcinoma cells. Mech. Dev. 33, 215-228.

Tajbakhsh, S. and Buckingham, M. E. (1994). Mouse limb muscle is determined in the absence of the earliest myogenic factor myf-5. Proc. Natl. Acad. Sci. USA. 91, 747-751.

Tiret, L. , Le Mouellic, H., Lallemand, Y., Maury, M. and Brulet, P. (1993). Altering the spatial determinations in the mouse embryos by manipulating the Hox genes. C. R. Acad. Sci. Paris. 316, 1017-1024.

Tokumoto, M. , Gong, Z., Tsubokawa, T. , Hew, C. L. , Uyemura, K. , Hotta, Y. and Okamoto, H. (1995). Molecular heterogeneity among primary motoneurons and within myotomes revealed by the differential mRNA expression of novel, Islet-I homologs in embryonic zebrafish. Developmental biology. 171, 578-589.

Tosney, K. W. (1981). Proximal tissues and patterned neurite outgrowth at the lombosacral level of the chick embryo: deletion of the dermamyotome. Dev. Biol. 122, 540- 588.

Tosney, K. W. and Landmesser, L. T. (1984) . Pattern and specificity of axonal outgrowth allowing varying degrees of chick limb bud ablation. J. Neurosci. 10, 2518-2527.

Tsuchida, T. , Ensini, M. , Morton, S. B. , Baldassare, M. , Edlund, T. , Jessel, T. M. and Pfaff S. L. (1994). Topographic organization of embryonic motor neurons defined by expression of LIM homeobox genes. Cell. 79, 957-970.

Udin, S. B. and Fawcett, J. W. (1988). Formation of topographic maps. Ann. Rev. Neurosci. II, 289-327.

Utset, M. F. Awgulewitsch, A. , Ruddle, F. H. and McGinnis, W. (1987). Region-specific expression of two mouse homeo box genes. Science. 235, 1379-1382.

EPO Patent Application No. EP 0 717 105 A2 , Changeux, et al.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(l) APPLICANT:

(A) NAME: INSTITUT PASTEUR

(B) STREET: 25-28 RUE DU DOCTEUR ROUX

(C) CITY: PARIS (E) COUNTRY: FRANCE

(F) POSTAL CODE (ZIP) : F 75724

(ll) TITLE OF INVENTION: TRANSGENIC OR MUTATED ANIMAL AND ITS USE IN SCREENING MOLECULES CAPABLE OF COMPENSATING FOR A NEURON DEFICIT OR A FUNCTIONAL DEFICIENCY

(ill) NUMBER OF SEQUENCES: 7

(iv) COMPUTER READABLE FORM: (A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: PatentIn Release #1.0, Version #1.30 (EPO)

(2) INFORMATION FOR SEQ ID NO: 1:

(l) SEQUENCE CHARACTERISTICS: (A) LENGTH: 19 base pairs (B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ll) MOLECULE TYPE: DNA (ill) HYPOTHETICAL: NO

(iv) SEQUENCE DESCRIPTION: SEQ ID NO: 1: CGTAGCCATA GAATTGGAG 19

(2) INFORMATION FOR SEQ ID NO: 2:

(l) SEQUENCE CHARACTERISTICS: (A) LENGTH: 19 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear ( i) MOLECULE TYPE: DNA (m) HYPOTHETICAL: NO (iv) SEQUENCE DESCRIPTION: SEQ ID NO: 2: GAGCTCCTAC TTCGTCAAC 19 (2) INFORMATION FOR SEQ ID NO: 3: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA

(iii) HYPOTHETICAL: NO (iv) SEQUENCE DESCRIPTION: SEQ ID NO: 3:

CAGCAGAAAC ATACAAGCTG 20

(2) INFORMATION FOR SEQ ID NO: 4: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 243 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein (iii) HYPOTHETICAL: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

Met Ser Ser Tyr Phe Val Asn Pro Leu Phe Ser Lys Tyr Lys Gly Gly 1 5 10 15

Glu Ser Leu Glu Pro Ala Tyr Tyr Asp Cys Arg Phe Pro Gin Ser Val 20 25 30

Gly Arg Ser His Ala Leu Val Tyr Gly Pro Gly Gly Ser Ala Pro Gly 35 40 45

Phe Gin His Ala Ser His His Val Gin Asp Phe Phe His His Gly Thr 50 55 60

Ser Gly lie Ser Asn Ser Gly Tyr Gin Gin Asn Pro Cys Ser Leu Ser 65 70 75 80

Cys His Gly Asp Ala Ser Lys Phe Tyr Gly Tyr Glu Ala Leu Pro Arg 85 90 95

Gin Ser Leu Tyr Gly Ala Gin Gin Glu Ala Ser Val Val Gin Tyr Pro 100 105 110 Asp Cys Lys Ser Ser Ala Asn Thr Asn Ser Ser Glu Gly Gin Gly His

115 120 125 Leu Asn Gin Asn Ser Ser Pro Ser Leu Met Phe Pro Trp Met Arg Pro 130 135 140

His Ala Pro Gly Arg Arg Ser Gly Arg Gin Thr Tyr Ser Arg Tyr Gin 145 150 155 160

Thr Leu Glu Leu Glu Lys Glu Phe Leu Phe Asn Pro Tyr Leu Thr Arg 165 170 175 Lys Arg Arg lie Glu Val Ser His Ala Leu Gly Leu Thr Glu Arg Gin

180 185 190

Val Lys lie Trp Phe Gin Asn Arg Arg Met Lys Trp Lys Lys Glu Asn 195 200 205

Asn Lys Asp Lys Leu Pro Gly Ala Arg Asp Glu Glu Lys Val Glu Glu 210 215 220

Glu Gly Asn Glu Glu Glu Glu Lys Glu Glu Glu Glu Lys Glu Glu Asn 225 230 235 240

Lys Asp Gly

(2) INFORMATION FOR SEQ ID NO: 5:

(l) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 261 ammo acids

(B) TYPE: ammo acid

(C) STRANDEDNESS: single (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

(m) HYPOTHETICAL: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5:

Met Ser Ala Thr Gly Pro lie Ser Asn Tyr Tyr Val Asp Ser Leu lie 1 5 10 15

Ser His Asp Asn Glu Asp Leu Leu Ala Ser Arg Phe Pro Ala Thr Gly 20 25 30

Ala His Pro Ala Ala Ala Arg Pro Ser Gly Leu Val Pro Asp Cys Ser 35 40 45

Asp Phe Pro Ser Cys Ser Phe Ala Pro Lys Pro Ala Val Phe Ser Thr 50 55 60

Ser Trp Ala Pro Val Pro Ser Gin Ser Ser Val Val Tyr His Pro Tyr

65 70 75 80

Gly Pro Gin Pro His Leu Gly Ala Asp Thr Arg Tyr Met Arg Thr Trp 85 90 95 Leu Glu Pro Leu Ser Gly Ala Val Ser Phe Pro Ser Phe Pro Ala Gly 100 105 110

Gly Arg His Tyr Ala Leu Lys Pro Asp Ala Tyr Pro Gly Arg Arg Ala

115 120 125

Asp Cys Gly Pro Gly Asp Gly Arg Ser Tyr Pro Asp Tyr Met Tyr Gly

130 135 140

Ser Pro Gly Glu Leu Arg Asp Arg Ala Pro Gin Thr Leu Pro Ser Pro 145 150 155 160

Glu Ala Asp Ala Leu Ala Gly Ser Lys His Lys Glu Glu Lys Ala Asp 165 170 175

Leu Asp Pro Ser Asn Pro Val Ala Asn Trp lie His Ala Arg Ser Thr 180 185 190

Arg Lys Lys Arg Cys Pro Tyr Thr Lys Tyr Gin Thr Leu Glu Leu Glu 195 200 205

Lys Glu Phe Leu Phe Asn Met Tyr Leu Thr Arg Asp Arg Arg Tyr Glu 210 215 220

Val Ala Arg Val Leu Asn Leu Thr Glu Arg Gin Val Lys lie Trp Phe 225 230 235 240

Gin Asn Arg Arg Met Lys Met Lys Lys Met Asn Lys Glu Lys Thr Asp 245 250 255

Lys Glu Gin Ser Glx 260

(2) INFORMATION FOR SEQ ID NO: 6:

(l) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 60 ammo acids

(B) TYPE: ammo acid

(C) STRANDEDNESS: single (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

(in) HYPOTHETICAL: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

Arg Arg Ser Gly Arg Gin Thr Tyr Ser Arg Tyr Gin Thr Leu Glu Leu

1 5 10 15

Glu Lys Glu Phe Leu Phe Asn Pro Tyr Leu Thr Arg Lys Arg Arg lie 20 25 30

Glu Val Ser His Ala Leu Gly Leu Thr Glu Arg Gin Val Lys lie Trp 35 40 45

Phe Gin Asn Arg Arg Met Lys Trp Lys Lys Glu Asn 50 55 60

(2) INFORMATION FOR SEQ ID NO: 7:

(l) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 60 ammo acids

(B) TYPE: ammo acid (C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein (m) HYPOTHETICAL: NO

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7:

Thr Arg Lys Lys Arg Cys Pro Tyr Thr Lys Tyr Gin Thr Leu Glu Leu 1 5 10 15

Glu Lys Glu Phe Leu Phe Asn Met Tyr Leu Thr Arg Asp Arg Arg Tyr 20 25 30

Glu Val Ala Arg Val Leu Asn Leu Thr Glu Arg Gin Val Lys lie Trp 35 40 45

Phe Gin Asn Arg Arg Met Lys Met Lys Lys Met Asn 50 55 60

Claims

What is claimed is: 1. Transgenic or mutated animal, wherein the expression of a gene involved in a neuron, for example a motoneuron, survival or activity is modulated and/or altered, which induces a nervous and more particularly a neuromuscular deficiency, wherein said animal is characterized in that an administration by intramuscular, intravenous, oral or any other route, of an active dose of a molecule or a composition of active molecules, as for example neurotrophic factors or cytokines, is capable to compensate the said deficiency in vivo.

2. Transgenic or mutated animal, wherein the expression of a gene involved in a neuron, for example a motoneuron, survival or activity is modulated and/or altered, which induces a nervous and more particularly a neuromuscular deficiency, wherein said animal is characterized in that an administration by intramuscular, intravenous, oral or any other route, of cells or expression vectors producing an active molecule or nucleotide sequence coding for an active molecule, as for example neurotrophic factors or cytokines is capable of compensating the said deficit or deficiency in vivo.

3. A process for screening an active molecule capable of compensating motoneuron survival or deficient activity characterized by testing said molecule on a transgenic animal as claimed in claims 1 or 2.

4. A process for screening an active molecule capable of compensating the pathological death of neurons or deficient activity or any other deficiency related to nervous disease, and more particularly to degenerative neuromuscular disease, said active molecule being tested on a transgenic animal carrying motoneuron deficiency phenotype.

5. A process for screening an active molecule capable of compensating the pathological death of neurons or deficient activity or any other deficiency related to nervous diseases, and more particularly to degenerative neuromuscular diseases, said active molecule being tested on a transgenic animal carrying the mutated Hoxc-8 gene or any sequence having at least 50% of homology with the Hoxc-8 coding sequence or 29% of identical amino acids with at least 70% of homology in the homeodomain of Hoxc-8, or any nucleotide sequence capable of restoring the phenotype, wherein said gene is mutated, or said active molecule being tested on a cell line prepared from primary or immortalized cells of the transgenic animal carrying the mutated Hoxc-8 gene or an equivalent gene as defined above.

6. A process for screening an active molecule capable of compensating the pathological death of neurons or deficient activity or any other deficiency related to nervous diseases, and more particularly to degenerative neuromuscular disease, characterized by primary or immortalized cell cultures from a transgenic animal containing in its genome hybrid nucleotide sequence comprising a part of the Hoxc-8 gene fused with a reporter gene or any sequence having at least 50% of homology with the Hoxc-8 coding sequence or 29% of identical amino acids with at least 70% of homology with the homeodomain of Hoxc- 8, or any nucleotide sequence capable of restoring the wild- type phenotype.

7. A process for screening according to claim 6, wherein said active molecule is capable of compensating the neuron deficit or functional deficiency.

8. Use of cells according to claims 6 or 7 for pharmaceutical screening of molecules for therapy of neurons, for example motoneurons, deficiency or survival.

9. Active molecule capable of restoring the wild-type phenotype in a transgenic animal by preventing neurons, for example motoneurons, from pathological death or capable of stimulating neuronal precursor proliferation or maintaining the stable innervation of the target cell said molecule used as a reagent for the methods as claimed in claims 6 or 7 or for the tests on transgenic animal as in claims 1 and 2.

10. A process for screening an active molecule capable of compensating for a prehension deficiency, motoneuron death, or any other deficiency related to nervous disease, and more particularly to degenerative neuromuscular disease, wherein said active molecule is tested on a transgenic or mutated animal of claim 1 or 2.

11. Active molecule capable of restoring a wild-type phenotype to an Hoxc-8 ' mutant in vivo.

12. A pharmaceutical composition containing active molecules capable of restoring the survival of at least 40% of deficient neurons in vivo or capable of restoring, in the transgenic animal of claim 1 or 2 , a neuronal activity equivalent to the wild-type animal.

13. A pharmaceutical composition containing active molecules capable of restoring the survival of at least 40% of deficient motoneurons in vivo or capable of restoring, in the transgenic animal of claim 1 or 2 , a motor activity equivalent to the wild-type animal.

14. A process of treatment of vertebrate and more particularly of mammals affected by a neuronal deficiency wherein the administration of an active molecule or a pharmaceutical composition which are capable of inducing modifications in the behavioral and/or functional properties in the patient or the mutated animal.

15. A process of treatment of vertebrate and more particularly of mammals affected by a motoneural deficiency wherein the administration of active molecules or a pharmaceutical composition which are capable of inducing modifications in the behavioral and/or functional properties in the patient or the mutated animal.

16. A process of activity measurement on a mutated animal of an active molecule, said process being characterized by the use of behavioral and functional tests compared to the measure of the performances of the wild-type animal and of those of the untreated transgenic or mutated animal as claimed in claim 1 or 2.

17. The process according to claims 3, 4, 5, 6, or 10 is characterized by adding an amount of active molecule or pharmaceutical composition on primary cell cultures and by counting the percentage of survival of the cells at various times compared to a culture of normal cells in the same condition.

18. The process according to claims 3, 4, 5, 6, or 10 is characterized by the measurement of the required time for the transgenic or mutated animal to climb onto a 3 mm width fixed barrier.