Netrin receptors

Abstract

The invention provides methods and compositions relating to vertebrate UNC-5 proteins which function as receptor proteins for netrins, a family of cell guidance proteins. The proteins may be produced recombinantly from transformed host cells from the disclosed vertebrate UNC-5 encoding nucleic acid or purified from human cells. The invention provides specific hybridization probes and primers capable of specifically hybridizing with the disclosed vertebrate unc-5 gene, vertebrate UNC-5-specific binding agents such as specific antibodies, and methods of making and using the subject compositions in diagnosis, therapy and in the biopharmaceutical industry.

Description

FIELD OF THE INVENTION

The field of this invention is proteins which regulate vertebrate cell guidance.

BACKGROUND

In the developing nervous system, migrating cells and axons are guided to their targets by cues in the extracellular environment. The netrins are a family of phylogenetically-conserved guidance cues that can function as diffusible attractants and repellents for different classes of cells and axons^1-10. Recent studies in vertebrates, insects and nematodes have implicated members of the DCC subfamily of the immunoglobulin (Ig) superfamily as receptors involved in migrations toward netrin sources^{6, 11-13}. The mechanisms that direct migrations away from netrin sources (presumed repulsions) are less well understood. In Caenorhabditis elegans, loss of unc-5 (which encodes the transmembrane protein UNC-5¹⁴) function causes defects in these migrations^{15, 16}, and ectopic expression of unc-5 in some neurons can redirect their axons away from a netrin source¹⁷. However, the relationship between UNC-5 and the netrins has not been defined. We disclose herein vertebrate homologues of the C. elegans UNC-5, which define a novel subfamily of the Ig superfamily, and whose mRNAs show prominent expression in various classes of differentiating neurons and we disclose that these vertebrate UNC-5 homologues are vertebrate netrin-binding proteins.

SUMMARY OF THE INVENTION

The invention provides methods and compositions relating to vertebrate UNC-5 proteins, related nucleic acids, and protein domains thereof having vertebrate UNC-5-specific activity. The proteins may be produced recombinantly from transfected host cells from the subject vertebrate UNC-5 encoding nucleic acids or purified from vertebrate cells. The invention provides isolated vertebrate unc-5 hybridization probes and primers capable of specifically hybridizing with the disclosed vertebrate unc-5 genes, vertebrate UNC-5-specific binding agents such as specific antibodies, and methods of making and using the subject compositions in diagnosis (e.g. genetic hybridization screens for vertebrate unc-5 transcripts), therapy (e.g. gene therapy to modulate vertebrate unc-5 gene expression) and in the biopharmaceutical industry (e.g. as immunogens, reagents for modulating cell guidance, reagents for screening chemical libraries for lead pharmacological agents, etc.).

DETAILED DESCRIPTION OF THE INVENTION

The nucleotide sequences of natural unc5h-1 cDNAs from rat and human are shown as SEQ ID NOS:1 and 2, respectively; and the conceptual translates are shown as SEQ ID NOS: 5 and 6, respectively. The nucleotide sequences of natural unc5h-2 cDNAs from rat and human are shown as SEQ ID NOS:3 and 4, respectively; and the conceptual translates are shown as SEQ ID NOS:7 and 8, respectively. The vertebrate UNC-5 proteins of the invention include incomplete translates of SEQ ID NOS:1, 2, 3 and 4 and deletion mutants of SEQ ID NOS:5, 6, 7 and 8, which translates and deletion mutants have vertebrate UNC-5-specific amino acid sequence and assay-discernable vertebrate UNC-5-specific binding specificity or function. Such active vertebrate UNC-5 deletion mutants, vertebrate UNC-5 peptides or protein domains comprise at least about 8, preferably at least about 12, more preferably at least about 24 consecutive residues of SEQ ID NO:5, 6, 7 or 8. For examples, vertebrate UNC-5 protein domains identified below are shown to provide protein-binding domains which are identified in and find use, inter alia, in solid-phase binding assays as described below.

Vertebrate UNC-5-specific activity or function may be determined by convenient in vitro, cell-based, or in vivo assays: e.g. in vitro binding assays, cell culture assays, in animals (e.g. gene therapy, transgenics, etc.), etc. Binding assays encompass any assay where the molecular interaction of a vertebrate UNC-5 protein with a binding target is evaluated. The binding target may be a natural extracellular binding target such as a netrin protein, or other regulator that directly modulates vertebrate UNC-5 activity or its localization; or non-natural binding target such a specific immune protein such as an antibody, or an vertebrate UNC-5 specific agent such as those identified in screening assays such as described below. Vertebrate UNC-5-binding specificity may assayed by binding equilibrium constants (usually at least about 10⁷ M⁻¹, preferably at least about 10⁸ M⁻¹, more preferably at least about 10⁹ M⁻¹), by the ability of the subject protein to function as negative mutants in vertebrate UNC-5-expressing cells, to elicit vertebrate UNC-5 specific antibody in a heterologous mammalian host (e.g a rodent or rabbit), etc. In any event, the vertebrate UNC-5 binding specificity of the subject vertebrate UNC-5 proteins necessarily distinguishes C. elegans UNC-5.

The claimed vertebrate UNC-5 proteins are isolated or pure: an “isolated” protein is unaccompanied by at least some of the material with which it is associated in its natural state, preferably constituting at least about 0.5%, and more preferably at least about 5% by weight of the total protein in a given sample and a pure protein constitutes at least about 90%, and preferably at least about 99% by weight of the total protein in a given sample. The vertebrate UNC-5 proteins and protein domains may be synthesized, produced by recombinant technology, or purified from mammalian, preferably human cells. A wide variety of molecular and biochemical methods are available for biochemical synthesis, molecular expression and purification of the subject compositions, see e.g. Molecular Cloning, A Laboratory Manual (Sambrook, et al. Cold Spring Harbor Laboratory), Current Protocols in Molecular Biology (Eds. Ausubel, et al., Greene Publ. Assoc., Wiley-Interscience, NY) or that are otherwise known in the art.

The invention provides natural and non-natural vertebrate UNC-5-specific binding agents, methods of identifying and making such agents, and their use in diagnosis, therapy and pharmaceutical development. For example, vertebrate UNC-5-specific agents are useful in a variety of diagnostic and therapeutic applications. Vertebrate UNC-5-specific binding agents include vertebrate UNC-5-specific ligands, such as netrins, and somatically recombined protein receptors like specific antibodies or T-cell antigen receptors (see, e.g Harlow and Lane (1988) Antibodies, A Laboratory Manual, Cold Spring Harbor Laboratory) and other natural binding agents identified with assays such as one-, two- and three-hybrid screens, non-natural binding agents identified in screens of chemical libraries such as described below, etc. For diagnostic uses, the binding agents are frequently labeled, such as with fluorescent, radioactive, chemiluminescent, or other easily detectable molecules, either conjugated directly to the binding agent or conjugated to a probe specific for the binding agent. Agents of particular interest modulate vertebrate UNC-5 function, e.g. vertebrate UNC-5-dependent cell guidance; for exampated cells, whole tissues, or individuals may be treated with a vertebrate UNC-5 binding agent to activate, inhibit, or alter vertebrate UNC-5-dependent cell guidance or function.

The invention provides UNC-5 related nucleic acids, which find a wide variety of applications including use as translatable transcripts, hybridization probes, PCR primers, diagnostic nucleic acids, etc.; use in detecting the presence of unc-5 genes and gene transcripts and in detecting or amplifying nucleic acids encoding additional unc-5 homologs and UNC-5 structural analogs. The subject nucleic acids are of synthetic/non-natural sequences and/or are isolated, i.e. unaccompanied by at least some of the material with which it is associated in its natural state, preferably constituting at least about 0.5%, preferably at least about 5% by weight of total nucleic acid present in a given fraction, and usually recombinant, meaning they comprise a non-natural sequence or a natural sequence joined to nucleotide(s) other than that which it is joined to on a natural chromosome. Nucleic acids comprising the nucleotide sequence of SEQ ID NO:1, 2, 3 or 4 or fragments thereof, contain such sequence or fragment at a terminus, immediately flanked by a sequence other than that which it is joined to on a natural chromosome, or flanked by a native flanking region fewer than 10 kb, preferably fewer than 2 kb, which is at a terminus or is immediately flanked by a sequence other than that which it is joined to on a natural chromosome. While the nucleic acids are usually RNA or DNA, it is often advantageous to use nucleic acids comprising other bases or nucleotide analogs to provide modified stability, etc.

The amino acid sequences of the disclosed vertebrate UNC-5 proteins are used to back-translate vertebrate UNC-5 protein-encoding nucleic acids optimized for selected expression systems (Holler et al. (1993) Gene 136, 323-328; Martin et al. (1995) Gene 154, 150-166) or used to generate degenerate oligonucleotide primers and probes for use in the isolation of natural vertebrate UNC-5-encoding nucleic acid sequences (“GCG” software, Genetics Computer Group, Inc, Madison Wis.). vertebrate UNC-5-encoding nucleic acids used in vertebrate UNC-5-expression vectors and incorporated into recombinant host cells, e.g. for expression and screening, transgenic animals, e.g. for functional studies such as the efficacy of candidate drugs for disease associated with vertebrate UNC-5-modulated transcription, etc.

The invention also provides nucleic acid hybridization probes and replication/amplification primers having a vertebrate UNC-5 cDNA specific sequence contained in SEQ ID NO:1, 2, 3 or 4 and sufficient to effect specific hybridization thereto (i.e. specifically hybridize with the corresponding SEQ ID NO:1, 2, 3 or 4 in the presence of C. elegans unc-5 cDNA). Such primers or probes are at least 12, preferably at least 24, more preferably at least 36 and most preferably at least 96 bases in length. Demonstrating specific hybridization generally requires stringent conditions, for example, hybridizing in a buffer comprising 30% formamide in 5×SSPE (0.18 M NaCl, 0.01 M NaPO₄, pH7.7, 0.001 M EDTA) buffer at a temperature of 42° C. and remaining bound when subject to washing at 42° C. with 0.2×SSPE; preferably hybridizing in a buffer comprising 50% formamide in 5×SSPE buffer at a temperature of 42° C. and remaining bound when subject to washing at 42° C. with 0.2×SSPE buffer at 42° C. vertebrate LNC-5 cDNA homologs can also be distinguished from other protein using alignment algorithms, such as BLASTX (Altschul et al. (1990) Basic Local Alignment Search Tool, J Mol Biol 215, 403-410).

Vertebrate unc-5 hybridization probes find use in identifying wild-type and mutant vertebrate unc-5 alleles in clinical and laboratory samples. Mutant alleles are used to generate allele-specific oligonucleotide (ASO) probes for high-throughput clinical diagnoses. Therapeutic vertebrate UNC-5 nucleic acids are used to modulate cellular expression or intracellular concentration or availability of active vertebrate UNC-5. For example, vertebrate UNC-5 nucleic acids are also used to modulate cellular expression or intracellular concentration or availability of active vertebrate UNC-5 protein. Vertebrate UNC-5 inhibitory nucleic acids are typically antisense: single-stranded sequences comprising complements of the disclosed natural vertebrate UNC-5 coding sequences. Antisense modulation of the expression of a given vertebrate UNC-5 protein may employ antisense nucleic acids operably linked to gene regulatory sequences. Cells are transfected with a vector comprising a vertebrate UNC-5 sequence with a promoter sequence oriented such that transcription of the gene yields an antisense transcript capable of binding to endogenous vertebrate UNC-5 encoding mRNA. Transcription of the antisense nucleic acid may be constitutive or inducible and the vector may provide for stable extrachromosomal maintenance or integration. Alternatively, single-stranded antisense nucleic acids that bind to genomic DNA or mRNA encoding a given vertebrate UNC-5 protein may be administered to the target cell, in or temporarily isolated from a host, at a concentration that results in a substantial reduction in expression of the targeted protein. An enhancement in vertebrate UNC-5 expression is effected by introducing into the targeted cell type vertebrate UNC-5 nucleic acids which increase the functional expression of the corresponding gene products. Such nucleic acids may be vertebrate UNC-5 expression vectors, vectors which upregulate the functional expression of an endogenous allele, or replacement vectors for targeted correction of mutant alleles. Techniques for introducing the nucleic acids into viable cells are known in the art and include retroviral-based transfection, viral coat protein-liposome mediated transfection, etc.

The invention provides efficient methods of identifying agents, compounds or lead compounds for agents active at the level of a vertebrate UNC-5 modulatable cellular function. Generally, these screening methods involve assaying for compounds which modulate vertebrate UNC-5 interaction with a natural vertebrate UNC-5 binding target. A wide variety of assays for binding agents are provided including labeled in vitro protein-protein binding assays, immunoassays, cell based assays, animal based assay, etc. Preferred methods are amenable to automated, cost-effective high throughput screening of chemical libraries for lead compounds. Such libraries encompass candidate agents of numerous chemical classes, though typically they are organic compounds; preferably small organic compounds and are obtained from a wide variety of sources including libraries of synthetic or natural compounds. Identified agents find use in the pharmaceutical industries for animal and human trials; for example, the agents may be derivatized and rescreened in in vitro and in vivo assays to optimize activity and minimize toxicity for pharmaceutical development.

In vitro binding assays employ a mixture of components including vertebrate UNC-5 protein, which may be part of a fusion product with another peptide or polypeptide, e.g. a tag for detection or anchoring, etc. The assay mixtures comprise a natural extracellular vertebrate UNC-5 binding target, such as a netrin. While native binding targets may be used, it is frequently preferred to use portions (e.g. peptides) thereof so long as the portion provides binding affinity and avidity to the subject vertebrate UNC-5 protein conveniently measurable in the assay. The assay mixture also comprises a candidate pharmacological agent and typically, a variety of other reagents such as salts, buffers, neutral proteins, e.g. albumin, detergents, protease inhibitors, nuclease inhibitors, antimicrobial agents, etc. The mixture components can be added in any order that provides for the requisite bindings and incubations may be performed at any temperature which facilitates optimal binding. The mixture is then incubated under conditions whereby, but for the presence of the candidate pharmacological agent, the vertebrate UNC-5 protein specifically binds the cellular binding target, portion or analog with a reference binding affinity. Incubation periods are likewise selected for optimal binding but also minimized to facilitate rapid, high-throughput screening.

After incubation, the agent-biased binding between the vertebrate UNC-5 protein and one or more binding targets is detected. A separation step is often initially used to separate bound from unbound components. Separation may be effected by precipitation (e.g. TCA precipitation, immunoprecipitation, etc.), immobilization (e.g on a solid substrate), etc., followed by washing by, for examples, membrane filtration, gel chromatography (e.g. gel filtration, affinity, etc.). One of the components usually comprises or is coupled to a label. The label may provide for direct detection such as radioactivity, luminescence, optical or electron density, etc. or indirect detection such as an epitope tag, an enzyme, etc. A variety of methods may be used to detect the label depending on the nature of the label and other assay components, e.g. through optical or electron density, radiative emissions, nonradiative energy transfers, etc. or indirectly detected with antibody conjugates, etc. A difference in the binding affinity of the vertebrate UNC-5 protein to the target in the absence of the agent as compared with the binding affinity in the presence of the agent indicates that the agent modulates the binding of the vertebrate UNC-5 protein to the vertebrate UNC-5 binding target. Analogously, in the cell-based transcription assay also described below, a difference in the vertebrate UNC-5 transcriptional induction in the presence and absence of an agent indicates the agent modulates vertebrate UNC-5-induced transcription. A difference, as used herein, is statistically significant and preferably represents at least a 50%, more preferably at least a 90% difference.

The following experimental section and examples are offered by way of illustration and not by way of limitation.

Experimental

cDNAs (SEQ ID NOS:1 and 3) encoding two rat homologues of UNC-5, termed UNC5H-1 and UNC5H-2, SEQ ID NOS:5 and 7, respectively, were isolated from an E18 rat brain cDNA library (see Methods). The predicted proteins show sequence similarity with UNC-5 over their entire lengths, but are more similar to one another (52% identity) than to UNC-5 (28% identity in each case). Like UNC-5¹⁴, both possess two predicted Ig-like domains and two predicted thrombospondin type-1 repeats in their extracellular domains (residues 1-332 of SEQ ID NO:5 and 1-334 of SEQ ID NO:7), a predicted membrane spanning region, and a large intracellular domain. The UNC5H proteins also each possess a signal sequence which, curiously, is lacking in UNC-5¹⁴. The predicted topology of the UNC5H proteins in cell membranes was verified using recombinant versions of the proteins expressed in transfected cells and antibodies directed against the extracellular and intracellular domains (see Methods). The cytoplasmic domains of the two UNC5H proteins do not contain obvious signaling motifs, but do possess a small region of homology to Zona Occludens-1 (ZO-1), a protein that localizes to adherens junctions and is implicated in junction formation^{18, 19}. ZO-1 contains PDZ-domains^{18, 19}, structures implicated in protein clustering²⁰, but the region of homology with UNC-5 homologues corresponds to a unique sequence at the carboxy terminus of ZO-1. The homology between ZO-1 and C. elegans UNC-5 is less pronounced (and is not detected by computer BLAST search), but is nonetheless apparent when all four sequences are aligned.

To determine whether the UNC-5 homologues are candidates for receptors involved in neuronal migration or axon guidance, we first examined the sites of expression of Unc5h-1 and Unc5h-2 by RNA in situ hybridization in rat embryos. Unc5h-1 transcripts are detected at early stages of neural tube development in the ventral spinal cord. At embryonic day 11 (E11), when motoneurons are beginning to differentiate in that region²¹ , transcripts are present throughout the ventral spinal cord, excluding the midline floor plate region, but are most intense in the ventricular zone and at the lateral edges. At E12, prominent expression is observed in the motor columns, but also extends more dorsally, and is now becoming excluded from the ventricular zone. This more dorsal expression appears transient, as expression by E13 is confined to postmitotic cells in the ventral spinal cord, apparently including the motoneurons. Unc5h-2 transcripts are not detected at significant levels in the spinal cord until E14, when they are found in the roof plate region. Unc5h-2 transcripts are, however, detected in developing sensory ganglia that flank the spinal cord, at low levels at E12, and at higher levels by E14. The expression of these two genes is thus observed in regions where differentiating neurons are undergoing axonogenesis, consistent with a possible role in this process.

Expression of these genes is also observed at higher axial levels of the nervous system, as well as in non-neural structures. At E13, Unc5h-1 is expressed in the basal plate (ventral neural tube) in the hindbrain and midbrain, in the developing hypothalamus and thalamus, and in the pallidum. Unc5h-2 expression at this stage is detected in the dorsal aspect of the developing optic cup, the nasal pits, apical ridge of the limb bud, urogenital tubercle, and in restricted regions of the midbrain and caudal diencephalon. By E16, Unc5h-1 mRNA is also detected at high levels in the entorhinal cortex and at lower levels throughout the cortex. Unc5h-2 is also detected at this stage at low levels in the cortex, and at high levels in hypertrophic chondrocytes. Expression of the two homologues persists postnatally, with, at postnatal day 10 (P 10), continued expression of both at low levels throughout the cortex, expression of both in distinct patterns in the septal area, and high level expression of Unc5h-1 in the developing hippocampus and entorhinal cortex. In addition, a prominent site of postnatal expression of both genes is in the cerebellum. Both are expressed in the inner granule cell layer, and Unc5h-2 is in addition expressed in the inner aspect of the external germinal layer, where granule cell precursors differentiate prior to migrating to their final destination in the inner granule cell layer^{22, 23}. Thus, expression of Unc5h-2 in this region is associated with a prominent cell migration event in the developing cerebellum.

Although the expression patterns of the two UNC5H proteins were suggestive of potential roles in cell or axon migration, to obtain more direct evidence implicating them in mediating responses to netrins we tested whether netrin-1 can bind cells expressing these proteins. Transfected monkey kidney COS-1 cells or human embryonic kidney 293 cells expressing either UNC5H-1 or UNC5H-2 showed significant binding of netrin-1 protein above background, as is also observed for transfected cells expressing the netrin receptors DCC and neogenin, but not for transfected cells expressing TAG-1 or L 1, two other members of the Ig superfamily¹³. In these experiments, binding was performed in the presence of soluble heparin, which eliminates non-specific binding of netrin-1 to the cells¹³ but does not evidently prevent binding to the UNC5 homologues. To verify, in the case of UNC5H-2, that exogenously added heparin is not required for the interaction, we generated a soluble protein comprising the extracellular domain of UNC5H-2 fused to the constant region (Fc) of a human immunogloblin molecule. This UNC5H-2-Fc fusion protein bound transfected 293 cells expressing netrin-1 (some of which remains associated with the surface of these cells^{3, 10}) in the absence of added heparin but did not show binding to non-transfected cells, nor to cells expressing UNC5H-2 itself, DCC, or neogenin. The UNC5H-2-Fc fusion also did not bind transfected cells expressing F-spondin, an adhesive extracellular matrix protein made by floor plate cells²⁴, or Semaphorin III, a chemorepellent for sensory axons at the stages that Unc5h-2 is expressed in sensory ganglia²⁵. Both of these proteins, like netrin-1, are secreted but partition between cell surfaces and the soluble fraction^{24, 26}. Thus, the interaction between netrin- 1 and UNC5H-2 appears specific, and does not require heparin nor reflect a generalized interaction with proteins that associate non-specifically with cell surfaces.

The affinity of UNC-5 homologues for netrin-1 was estimated in equilibrium binding experiments using netrin(VIoV)-Fc, a fusion of the amino terminal two-thirds of netrin-1 to the constant portion of human IgG¹³. This netrin-1 derivative is bioactive but, unlike netrin-1, does not aggregate at high concentrations, and it binds DCC with a Kd comparable to that of full length netrin-1 ¹³. Specific binding of netrin (VIoV)-Fc to each of the three UNC5 homologues showed saturation and the binding curves were fitted to the Hill equation, yielding Kd values of 19±0.8 nM and 3.4±1.0 nM for UNC5H1 and UNC5H2 respectively. These values are comparable to the Kd for the DCC-netrin (VIoV-Fc) interaction (˜5 nM), and are consistent with the effective dose for the axon outgrowth promoting effects of netrin-1^{2, 13}.

Establishing the involvement of these vertebrate UNC5H proteins in cell migration and axon guidance will require perturbing their functions in vivo. In the meantime, however, our results are at least consistent with such an involvement, as these homologues are expressed by some populations of cells that are undergoing migrations or extending axons. For example, Unc5h1 is expressed by spinal motoneurons, whose axons are repelled in vitro by floor plate cells²⁷, and whose outgrowth in vitro can be suppressed by netrin-1. It is also expressed in the region of trochlear motoneurons, which can be repelled by netrin-1⁴. Both Unc5h genes are also expressed in the developing cerebellum, which is a site of extensive cell migration.

Although the in vivo functions of the UNC-5 homologues described here remain to be determined, our evidence that vertebrate UNC5H proteins bind netrin-1 provides direct support for the idea that members of this new subfamily of the Ig superfamily are netrin receptors. This idea was first proposed for C. elegans UNC-5, based on the findings that unc-5 is required cell-autonomously for dorsal migrations that require the function of the netrin UNC-6¹⁴, and that ectopic expression of unc-5 in neurons that normally project longitudinally or ventrally can steer their axons dorsally¹⁷. Although consistent with the possibility that UNC-5 is an UNC-6 receptor, these results are also consistent with a role for UNC-5 in modifying the function of a distinct UNC-6 receptor. The possibility of a modifier function was made more plausible by evidence that the DCC homologue UNC-40, which is a putative UNC-6 receptor involved in ventral migrations¹¹, is expressed by axons that project dorsally and is required for those projections^{11, 15, 16}, suggesting that UNC-5 might function by switching an attractive netrin receptor (UNC-40) into a repulsive netrin receptor. However, our results suggest that UNC-5 also functions directly as a netrin receptor. A model in which UNC-40 and UNC-5 can form a receptor complex but UNC-5 can also function alone in transducing the UNC-6 netrin signal provides an explanation for the observation that loss of unc-40 function results in a much less severe phenotype for dorsal migrations than do either loss of unc-5 or loss of unc-6 function^{15, 16}.

Recent studies have demonstrated a remarkable phylogenetic conservation in function of netrin proteins in guiding axons towards a source of netrin at the midline of the nervous systems of nematodes, flies and vertebrates^1,7,8,9, as well as a conserved role for members of the DCC subfamily of the Ig superfamily in mediating the axonal responses that underlie those guidance events^{11, 12, 13} The identification of vertebrate homologues of UNC-5, and the evidence that they are netrin-binding proteins, suggests that the signaling mechanisms through which netrins elicit repulsive responses are also conserved.

Isolation of rat UNC-5 homologues, and in situ hybridization. A search of the human expressed sequence tag (EST) databases revealed a small sequence (Genbank accession number R11880) with distant similarity to the carboxy-terminal portion of UNC-5. The corresponding cDNA fragment, amplified by polymerase chain reaction from an embryonic human brain cDNA library (Stratagene), was used to screen the library, resulting in the isolation of a 3.8 kB cDNA clone comprising all but the first 440 nt of the coding region of the human homologue of UNC5H1. Non-overlapping probes from this cDNA were used to screen an E18 rat brain library (gift of S. Nakanishi), leading to isolation of seven partial and one full length UNC5H1 cDNA and one full length UNC5H2 cDNA. Additional screens of E13 rat dorsal and ventral spinal cord libraries resulted in isolation of a second full length UNC5H2 cDNA as well as a nearly full length UNC5H1 cDNA. Sequencing was performed on a Licor (L4000) automated sequencer as well as by³³P cycle sequencing. Genbank accession numbers are U87305 and U87306 for rUNC5H1 and rUNC5H2 respectively. RNA in situ hybridization was performed as described¹³.

Antibodies, expression constructs and immunohistochemistry. Rabbit polyclonal antisera were raised to a peptide corresponding to a sequence (YLRKNFEQEPLAKE, SEQ ID NO:7, residues 148-161) in the extracellular domain of UNC5H-2 that is almost completely conserved in UNC5H-1 (one amino acid substitution), and to peptides corresponding to unique sequences in the cytoplasmic domains of UNC5H-1 (GEPSPDSWSLRLKKQ, SEQ ID NO:5, residues 580-594) and UNC5H-2 (EARQQDDGDLNSLASA, SEQ ID NO:7, residues 909-924). Antisera were affinity-purified on the respective peptides (Quality Controlled Biochemicals). cDNAs for the various constructs were subcloned into the COS cell expression vector pMT21 and the 293-EBNA cell expression vector pCEP4 (Invitrogen), and transiently transfected into those cells using lipofectamine. The antiserum to the extracellular peptide can detect both UNC5H proteins expressed in transfected cells without cell permeabilization, whereas the antisera directed against the cytoplasmic domain peptides detected their respective proteins after cell permeabilization. Netrin-1 protein (SEQ ID NO:9) was produced, purified, used and visualized in binding assays as described¹³, except that a monoclonal antibody (9E10)²⁹ directed to a C-terminal myc-epitope tag was used to detect recombinant netrin-1, and heparin was used at 1μg/ml. A 293-EBNA cell line stably expressing the UNC5H-2-Fc fusion was derived and maintained as described^{10, 13}. The fusion protein was purified from serum-free medium conditioned for seven days by affinity chromatography on protein A agarose. The 293 cell line expressing netrin-1 was as described¹³. Binding of the UNC5H-2-Fc fusion to this line was visualized using a Cy3-conjugated secondary antibody (Jackson Immunoresearch) directed against human Fc.

References

1. Ishii, N., et al., Neuron 9, 873-81 (1992).

2. Serafini, T. et al. Cell 78, 409-24 (1994).

3. Kennedy, T. E., Serafini, T., de la Torre, J. R. & Tessier-Lavigne, M. Cell 78, 425-35 (1994).

4. Colamarino, S. A. & Tessier-Lavigne, M. Cell 81, 621-9 (1995).

5. Shirasaki, R., Tamada, A., Katsumata, R. & Murakami, F. Neuron 14, 961-72 (1995).

6. Wadsworth, W. G., Bhatt, H. & Hedgecock, E. M. Neuron 16, 35-46 (1996).

7. Mitchell, K. J., et al., Neuron 17, 203 (1996)

8. Harris, R., Sabatelli, L. M. & Seeger, M. A. Neuron 17, 217-228 (1996).

9. Serafini, T., et al., Cell in press.

10. Shirasaki, R., Mirzayan, C., Tessier-Lavigne, M. & Murakami, F. Neuron in press, (1996).

11. ChanS. S.-Y., et al., Cell 87, 187-196 (1996).

12. Kolodziej, P. A., et al., Cell 87, 197-204 (1996)

13. Keino-Masu, K, et al., Cell 87, 175-185 (1996).

14. Leung-Hagesteijn, C. et al. Cell 71, 289-99 (1992).

15. Hedgecock, E. M., Culotti, J. G. & Hall, D. H. Neuron 4, 61-85 (1990).

16. McIntire, S. L., et al., Neuron 8, 307-22 (1992).

17. Hamelin, M., Zhou, Y., Su, M. W., Scott, I. M. & Culotti, J. G. Nature 364, 327-30 (1993).

18. Willoff E; et al., Proc. Natl Acad. Sci 90, 7834-8 (1993).

19. Itoh M; et al., J. Cell Biol. 121, 491-502 (1993).

20. Sheng, M. Neuron 17, 575-578 (1996).

21. Altman, J., & Bayer, S. A. Adv. Anat. Embryol. Cell Biol. 85, 1-166 (1984).

22. Ramon y Cajal, S. Histologie du Systeme Nerveux de l'Home et des Vertebres, Vol. 2 (1911).

23. Rakic, P. J. Comp. Neurol. 141, 283-312 (1971).

24. Klar A; Baldassare M; & Jessell TM. Cell 69, 95-110. (1992).

25. Messersmith, E. K. et al. Neuron 14, 949-59 (1995).

26. Luo, Y., Raible, D. & Raper, J. A. Cell 75, 217-27 (1993).

27. Guthrie, S. & Pini, A. Neuron 14, 1117-30 (1995).

28. Evan, G. I., Lewis, G. K., Ramsey, G., & Bishop, J. M. Mol. Cell. Biol. 5, 3610-3616 (1985).

EXAMPLES

1. Protocol for high throughput vertebrate UNC-5—netrin binding assay.

A. Reagents:

• • Neutralite Avidin: 20 μg/ml in PBS.
  • Blocking buffer: 5% BSA, 0.5% Tween 20 in PBS; 1 hour at room temperature.
  • Assay Buffer: 100 mM KCl, 20 mM HEPES pH 7.6, 1 mM MgCl₂, 1% glycerol, 0.5% NP-40, 50 mM b-mercaptoethanol, 1 mg/ml BSA, cocktail of protease inhibitors.

³³P vertebrate UNC-5 protein 10× stock: 10⁻⁸-10⁻⁶ M “cold” vertebrate UNC-5 supplemented with 200,000-250,000 cpm of labeled vertebrate LNC-51 (Beckman counter). Place in the 4° C. microfridge during screening.

• • Protease inhibitor cocktail (1000×): 10 mg Trypsin Inhibitor (BMB # 109894), 10 mg Aprotinin (BMB # 236624), 25 mg Benzamidine (Sigma # B-6506), 25 mg Leupeptin (BMB # 1017128), 10 mg APMSF (BMB # 917575), and 2 mM NaVo₃ (Sigma # S-6508) in 10 ml of PBS.
  • nerin-1: 10⁻⁷-10⁻⁵ M biotinylated netrin-1 in PBS.

B. Preparation of assay plates:

• • Coat with 120 μl of stock N-Avidin per well overnight at 4° C.
  • Wash 2 times with 200 μl PBS.
  • Block with 150 μl of blocking buffer.
  • Wash 2 times with 200 μl PBS.

C. Assay:

• • Add 40 μl assay buffer/well.
  • Add 10 μl compound or extract.
  • Add 10 μl ³³P-UNC-5 (20-25,000 cpm/0.1-10 pmoles/well=10⁻⁹-10⁻⁷ M final conc).
  • Shake at 25° C. for 15 minutes.
  • Incubate additional 45 minutes at 25° C.
  • Add 40 μM biotinylated netrin-1 (0.1-10 pmoles/40 μl in assay buffer)
  • Incubate 1 hour at room temperature.
  • Stop the reaction by washing 4 times with 200 μM PBS.
  • Add 150 μM scintillation cocktail.
  • Count in Topcount.

D. Controls for all assays (located on each plate):

• • a. Non-specific binding
  • b. Soluble (non-biotinylated netrin-1) at 80% inhibition.

All publications and patent applications cited in this specification are herein incorporated by reference as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference. Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims.

Claims

1. An isolated vertebrate UNC-5 protein comprising SEQ ID NO: 5, 6, 7 or, 8, or a fragment thereof having vertebrate UNC-5-specific activity.

2. An isolated protein according to claim 1, wherein said protein specifically binds a natural netrin protein.

3. A recombinant nucleic acid encoding a protein according to claim 1.

4. A cell comprising a nucleic acid according to claim 3.

5. A method of making an isolated vertebrate UNC-5 protein, comprising steps: introducing a nucleic acid according to claim 3 into a host cell or cellular extract, incubating said host cell or extract under conditions whereby said nucleic acid is expressed as a transcript and said transcript is expressed as a translation product comprising said protein, and isolating said translation product.

6. An isolated vertebrate UNC-5 protein made by the method of claim 5.

7. An isolated vertebrate unc-5 nucleic acid comprising SEQ ID NO: 1 ,2, 3, or 4, or a fragment thereof having at least 24 consecutive bases of SEQ ID NO: 1, 2, 3, or 4 and sufficient to specifically hybridize with a nucleic acid having the sequence of the corresponding SEQ ID NO: 1, 2, 3, or 4 in the presence of natural C. elegans unc-5 cDNA.

8. A method of screening for an agent which modulates the binding of a vertebrate UNC-5 protein to a binding target, said method comprising the steps of: incubating a mixture comprising: an isolated protein according to claim 1, a binding target of said protein, and a candidate agent; under conditions whereby, but for the presence of said agent, said protein specifically binds said binding target at a reference affinity; detecting the binding affinity of said protein to said binding target to determine an agent-biased affinity, wherein a difference between the agent-biased affinity and the reference affinity indicates that said agent modulates the binding of said protein to said binding target.

9. A method according to claim 8, wherein said binding target is a natural netrin protein.