Semaphorin gene family Number:7,153,936 from the United States Patent and Trademark Office (PTO) owispatent

Title: Semaphorin gene family

Abstract: A novel class of proteins, semaphorins, nucleic acids encoding semaphorins, semaphorin peptides, and methods of using semaphorins and semaphorin-encoding nucleic acids are disclosed. Semaphorin peptides and receptor agonists and antagonists provide potent modulators of nerve cell growth and regeneration. The invention provides pharmaceutical compositions, methods for screening chemical libraries for regulators of cell growth/differentiation; semaphorin gene-derived nucleic acids for use in genetic mapping, as probes for related genes, and as diagnostic reagents for genetic neurological disease; specific cellular and animal systems for the development of neurological disease therapy.

Patent Number: 7,153,936 Issued on 12/26/2006 to Goodman,   et al.

---

Inventors:	Goodman; Corey S. (Berkeley, CA), Kolodkin; Alex L. (Berkeley, CA), Matthes; David (Berkeley, CA), Bentley; David R. (Berkeley, CA), O'Connor; Timothy (Berkeley, CA)
Assignee:	The Regents of the University of California (Oakland, CA)
Appl. No.:	10/067,632
Filed:	February 4, 2002

---

Related U.S. Patent Documents

---

	Application Number	Filing Date	Patent Number	Issue Date
	09060610	Apr., 1998	6344544
	08835268	Apr., 1997	5807826
	08121713	Sep., 1993	5639856

---

Current U.S. Class:	530/350 ; 435/69.1; 435/70.1; 530/399
Current International Class:	C07K 1/00 (20060101); C07K 14/00 (20060101); A61K 38/00 (20060101)

---

References Cited [Referenced By]

---

U.S. Patent Documents

	5416197		May 1995		Raper et al.
	5639856		June 1997		Goodman et al.
	5807826		September 1998		Goodman et al.
	6013781		January 2000		Goodman et al.

Foreign Patent Documents

	93/00365		Jan., 1993		WO

Other References

Waterson et al., Nature Genetics 1(2): 114-23, May 1992. cited by other . Kolodkin et al. Neuron, 9:831-45, 1992. cited by other . Genbank Accession No. Mg1187. cited by other . Genbank Accession No. X57318. cited by other . Genbank Accession No. M35707. cited by other . Genbank Accession No. X62984 (Harbitz et al). cited by other . Hames et al, Nucleic acid hybridisation, a practical approach, published 1985 by IRL Press (Oxford) pp. 81, 82, 93-95. cited by other . Aguado et al, J. Gen. Virol., vol. 73, issued Nov. 1992, pp. 2887-2902. cited by other . Smith et al, J. Gen. Virol., vol. 72, issued 1991, pp. 1349-1376. cited by other . Goodman and Shatz, "Developmental Mechanisms that Generate Precise Patterns of Neuronal Connectivity", Cell, vol. 72/Neuron, vol. 10 (Suppl.), 77-98 (1993). cited by other . Raper, "Repulsive Cues in Axonal Guidance", an abstract presented at the GIBCO-BRL Symposium on Genes and Development/Function of Brain on Jul. 26, 1993. cited by other . Luo et al., Cell, 75:217-227 (1993). cited by other . Raper and Kapfhammer, Neuron, 2:21-29 (1990). cited by other . Schnell and Schwab, Nature, 343:269-272 (1990). cited by other . Schwab and Caroni, J. Neuroscience, 8(7):2381-2393 (1988). cited by other . Kolodkin et al., Cell, vol. 75 (1993) 1389-1399. cited by other . Jessell et al., Principles of Neural Science (3rd edition)(1991) chapter 18. cited by other . Saneto et al., Neuro chemistry (Chap. 2) (1987) pp. 27-63. cited by other . Patel, Biochem. Soc. Trans. (1989) 17(5) 931. cited by other . McMartin, Biochem. Soc. Trans (1989) 17(5) pp. 931-934. cited by other . Bundgaard et al., Biochem. Soc. Trans., (1989) 17(5) pp. 947-949. cited by other.

Primary Examiner: Andres; Janet L.
Assistant Examiner: Wang; Chang-Yu
Attorney, Agent or Firm: Townsend and Townsend and Crew LLP

---

Government Interests

---

The research carried out in the subject application was supported in part by grants from the National Institutes of Health. The government may have rights in any patent issuing on this application.

---

Parent Case Text

---

This application is a continuation of U.S. application Ser. No. 09/060,610, filed Apr. 15, 1998 now U.S. Pat. No. 6,344,544, which is a continuation of U.S. application Ser. No. 08/835,268, filed Apr. 8, 1997, now U.S. Pat. No. 5,807,826, which is a division of U.S. application Ser. No. 08/121,713, filed Sep. 13, 1993, now U.S. Pat. No. 5,639,856.

---

Claims

---

What is claimed is:

1. A composition comprising an isolated polypeptide comprising the amino acid sequence of SEQ ID NO:54, and a physiologically acceptable carrier.

---

Description

---

INTRODUCTION

1. Technical Field

The technical field of this invention concerns peptides, polypeptides, and polynucleotides involved in nerve cell growth.

2. Background

The specificity of the wiring of the nervous system--the complex pattern of specific synaptic connections--begins to unfold during development as the growing tips of neurons--the growth cones--traverse long distances to find their correct targets. Along their journey, they are confronted by and correctly navigate a series of choice points in a remarkably unerring way to ultimately contact and recognize their correct target.

The identification of growth cone guidance cues is to a large extent, the holy grail of neurobiology. These are the compounds that tell neurons when to grow, where to grow, and when to stop growing. The medical applications of such compounds and their antagonists are enormous and include modulating neuronal growth regenerative capacity, treating neurodegenerative disease, and mapping (e.g. diagnosing) genetic neurological defects.

Over decades of concentrated research, various hypotheses of chemo-attractants and repellant, labeled pathways, cell adhesion molecules, etc. have been evoked to explain guidance. Recently, several recent lines of experiments suggest repulsion may play an important role in neuron guidance and two apparently unrelated factors ("Neurite Growth Inhibitor" and "Collapsin") capable of inhibiting or collapsing growth cones have been reported.

Relevant Literature

For a recent review of much of the literature in this field, see Goodman and Shatz (1993) Cell 72/Neuron 10, 77 98. A description of grasshopper fasciclin IV (now called G-Semaphorin I) appears in Kolodkin et al. (1992) Neuron 9, 831 845. Recent reports on Collapsin and Neurite Growth Inhibitor include Raper and Kapfhammer (1990) Neuron 4, 21 29, an abstract presented by Raper at the GIBCO-BRL Symposium on "Genes and Development/Function of Brain" on Jul. 26, 1993 and Schwab and Caroni (1988) J Neurosci 8, 2381 and Schnell and Schwab (1990) Nature 343, 269, respectively.

SUMMARY OF THE INVENTION

A novel class of proteins, semaphorins, nucleic acids encoding semaphorins, and methods of using semaphorins and semaphorin-encoding nucleic acids are disclosed. Semaphorins include the first known family of human proteins which function as growth cone inhibitors and a family of proteins involved in viral, particularly pox viral, pathogenesis and oncogenesis. Families of semaphorin-specific receptors, including receptors found on nerve growth cones and immune cells are also disclosed.

The invention provides agents, including semaphorin peptides, which specifically bind semaphorin receptors and agents, including semaphorin receptor peptides, which specifically bind semaphorins. These agents provide potent modulators of nerve cell growth, immune responsiveness and viral pathogenesis and find use in the treatment and diagnosis of neurological disease and neuro-regeneration, immune modulation including hypersensitivity and graft-rejection, and diagnosis and treatment of viral and oncological infection/diseases.

Semaphorins, semaphorin receptors, semaphorin-encoding nucleic acids, and unique portions thereof also find use variously in screening chemical libraries for regulators of semaphorin or semaphorin receptor-mediated cell activity, in genetic mapping, as probes for related genes, as diagnostic reagents for genetic neurological, immunological and oncological disease and in the production of specific cellular and animal systems for the development of neurological, immunological, oncological and viral disease therapy.

DESCRIPTION OF SPECIFIC EMBODIMENTS

The present invention discloses novel families of proteins important in nerve and immune cell function: the semaphorins and the semaphorin receptors. The invention provides agents, including semaphorin peptides, which specifically bind semaphorin receptors and agents, including semaphorin receptor peptides, which specifically bind semaphorins. These agents find a wide variety of clinical, therapeutic and research uses, especially agents which modulate nerve and/or immune cell function by specifically mimicing or interfering with semaphorin-receptor binding. For example, selected semaphorin peptides shown to act as semaphorin receptor antagonists are effective by competitively inhibiting native semaphorin association with cellular receptors. Thus, depending on the targeted receptor, these agents can be used to block semaphorin mediated neural cell growth cone repulsion or contact inhibition. Such agents find broad clinical application where nerve cell growth is indicated, e.g. traumatic injury to nerve cells, neurodegenerative disease, etc. A wide variety of semaphorin- and semaphorin receptor-specific binding agents and methods for identifying, making and using the same are described below.

Binding agents of particular interest are semaphorin peptides which specifically bind and antagonize a semaphorin receptor and semaphorin receptor peptides which specifically bind a semaphorin and prevent binding to a native receptor. While exemplified primarily with semaphorin peptides, much of the following description applies analogously to semaphorin receptor peptides.

The semaphorin peptides of the invention comprise a unique portion of a semaphorin and have semaphorin binding specificity. A "unique portion" of a semaphorin has an amino acid sequence unique to that disclosed in that it is not found in any previously known protein. Thus a unique portion has an amino acid sequence length at least long enough to define a novel peptide. Unique semaphorin portions are found to vary from about 5 to about 25 residues, preferably from 5 to 10 residues in length, depending on the particular amino acid sequence. Unique semaphorin portions are readily identified by comparing the subject semaphorin portion sequences with known peptide/protein sequence data bases. Preferred unique portions derive from the semaphorin domains (which exclude the Ig-like, intracellular and transmembrane domains as well as the signal sequences) of the disclosed semaphorin sequences, especially regions that bind the semaphorin receptor, especially that of the human varieties. Preferred semaphorin receptor unique portions derive from the semaphorin binding domains, especially regions with residues which contact the semaphorin ligand, especially that of the human varieties. Particular preferred peptides are further described herein.

The subject peptides may be free or coupled to other atoms or molecules. Frequently the peptides are present as a portion of a larger polypeptide comprising the subject peptide where the remainder of the polypeptide need not be semaphorin- or semaphorin receptor-derived. Alternatively, the subject peptide may be present as a portion of a "substantially full-length" semaphorin domain or semaphorin receptor sequence which comprises or encodes at least about 200, preferably at least about 250, more preferably at least about 300 amino acids of a disclosed semaphorin/receptor sequence. Thus the invention also provides polypeptides comprising a sequence substantially similar to that of a substantially full-length semaphorin domain or a semaphorin receptor. "Substantially similar" sequences share at least about 40%, more preferably at least about 60%, and most preferably at least about 80% sequence identity. Where the sequences diverge, the differences are generally point insertions/deletions or conservative substitutions, i.e. a cysteine/threonine or serine substitution, an acidic/acidic or hydrophobic/hydrophobic amino acid substitution, etc.

The subject semaphorin peptides/polypeptides are "isolated", meaning unaccompanied by at least some of the material with which they are associated in their natural state. Generally, an isolated peptide/polypeptide constitutes at least about 1%, preferably at least about 10%, and more preferably at least about 50% by weight of the total peptide/protein in a given sample. By pure peptide/polypeptide is intended at least about 90%, preferably at least 95%, and more preferably at least about 99% by weight of total peptide/protein. Included in the subject peptide/polypeptide weight are any atoms, molecules, groups, or polymers covalently coupled to the subject semaphorin/receptor peptide/polypeptide, especially peptides, proteins, detectable labels, glycosylations, phosphorylations, etc.

The subject peptides/polypeptides may be isolated or purified in a variety of ways known to those skilled in the art depending on what other components are present in the sample and to what, if anything, the peptide/polypeptide is covalently linked. Purification methods include electrophoretic, molecular, immunological and chromatographic techniques, especially affinity chromatography and RP-HPLC in the case peptides. For general guidance in suitable purification techniques, see Scopes, R., Protein Purification, Springer-Verlag, NY (1982).

The subject peptides/polypeptides generally comprise naturally occurring amino acids but D-amino acids or amino acid mimetics coupled by peptide bonds or peptide bond mimetics may also be used. Amino acid mimetics are other than naturally occurring amino acids that conformationally mimic the amino acid for the purpose of the requisite semaphorin/receptor binding specificity. Suitable mimetics are known to those of ordinary skill in the art and include .beta.-.gamma.-.delta. amino and imino acids, cyclohexylalanine, adamantylacetic acid, etc., modifications of the amide nitrogen, the .alpha.-carbon, amide carbonyl, backbone modifications, etc. See, generally, Morgan and Gainor (1989) Ann. Repts. Med. Chem 24, 243 252; Spatola (1983) Chemistry and Biochemistry of Amino Acids, Peptides and Proteins, Vol VII (Weinstein) and Cho et. al (1993) Science 261, 1303 1305 for the synthesis and screening of oligocarbamates.

The subject semaphorin peptides/polypeptides have a "semaphorin binding specificity" meaning that the subject peptide/polypeptide retains a molecular conformation specific to one or more of the disclosed semaphorins and specifically recognizable by a semaphorin-specific receptor, antibody, etc. As such, a semaphorin binding specificity may be provided by a semaphorin-specific immunological epitope, lectin binding site, etc., and preferably, a receptor binding site. Analogously, the semaphorin receptor peptides/polypeptides have a "semaphorin receptor binding specificity" meaning that these peptides/polypeptides retain a molecular conformation specific to one or more of the disclosed semaphorin receptors and specifically recognizable by a semaphorin, a receptor-specific antibody, etc.

"Specific binding" is empirically determined by contacting, for example a semaphorin-derived peptide with a mixture of components and identifying those components that preferentially bind the semaphorin. Specific binding is most conveniently shown by competition with labeled ligand using recombinant semaphorin peptide either in vitro or in cellular expression systems as disclosed herein. Generally, specific binding of the subject semaphorin has binding affinity of 10.sup.-6M, preferably 10.sup.-8M, more preferably 10.sup.-10M, under in vitro conditions as exemplified below.

The peptides/polypeptides may be modified or joined to other compounds using physical, chemical, and molecular techniques disclosed or cited herein or otherwise known to those skilled in the relevant art to affect their semaphorin binding specificity or other properties such as solubility, membrane transportability, stability, binding specificity and affinity, chemical reactivity, toxicity, bioavailability, localization, detectability, in vivo half-life, etc. as assayed by methods disclosed herein or otherwise known to those of ordinary skill in the art. For example, point mutations are introduced by site directed mutagenesis of nucleotides in the DNA encoding the disclosed semaphorin polypeptides or in the course of in vitro peptide synthesis.

Other modifications to further modulate binding specificity/affinity include chemical/enzymatic intervention (e.g. fatty acid-acylation, proteolysis, glycosylation) and especially where the peptide/polypeptide is integrated into a larger polypeptide, selection of a particular expression host, etc. In particular, many of the disclosed semaphorin peptides contain serine and threonine residues which are phosphorylated or dephosphorylated. See e.g. methods disclosed in Roberts et al. (1991) Science 253, 1022 1026 and in Wegner et al. (1992) Science 256, 370 373. Amino and/or carboxyl termini may be functionalized e.g., for the amino group, acylation or alkylation, and for the carboxyl group, esterification or amidification, or the like. Many of the disclosed semaphorin peptides/polypeptides also contain glycosylation sites and patterns which may disrupted or modified, e.g. by enzymes like glycosidases or used to purify/identify the receptor, e.g. with lectins. For instance, N or O-linked glycosylation sites of the disclosed semaphorin peptides may be deleted or substituted for by another basic amino acid such as Lys or His for N-linked glycosylation alterations, or deletions or polar substitutions are introduced at Ser and Thr residues for modulating O-linked glycosylation. Glycosylation variants are also produced by selecting appropriate host cells, e.g. yeast, insect, or various mammalian cells, or by in vitro methods such as neuraminidase digestion. Useful expression systems include COS-7, 293, BHK, CHO, TM4, CV1, VERO-76, HELA, MDCK, BRL 3A, W138, Hep G2, MMT 060562, TRI cells, baculovirus systems, for examples. Other covalent modifications of the disclosed semaphorin peptides/polypeptides may be introduced by reacting the targeted amino acid residues with an organic derivatizing (e.g. methyl-3-[(p-azido-phenyl)dithio] propioimidate) or crosslinking agent (e.g. 1,1-bis(diazoacetyl)-2-phenylethane) capable of reacting with selected side chains or termini. For therapeutic and diagnostic localization, semaphorins and peptides thereof may be labeled directly (radioisotopes, fluorescers, etc.) or indirectly with an agent capable of providing a detectable signal, for example, a heart muscle kinase labeling site.

The following are 14 classes of preferred semaphorin peptides where bracketed positions may be occupied by any one of the residues contained in the brackets and "X" signifies that the position may be occupied by any one of the 20 naturally encoded amino acids (see, Table 1). These enumerated peptides maintain highly conserved structures which provide important semaphorin binding specificities; (a) [AspGlu]Cys[GlnLysArgAlaAsn]Asn[TyrPheVal]Ile (SEQ ID NO:1) Cys[GlnLysArgAlaAsn]Asn[TyrPheVal]Ile[ArgLysGlnThr] (SEQ ID NO:2) (b) CysGlyThr[AsnGly][AlaSerAsn][TyrPheHisGly][LysArgHisAsnGln] (SEQ ID NO:3) CysGlyThr[AsnGly][AlaSerAsn]XaaXaaPro (SEQ ID NO:4) CysGlyThr[AsnGly]XaaXaaXaaProXaa[CysAsp] (SEQ ID NO:5) CysGlyThrXaaXaaXaaXaaProXaa[CysAsp]XaaXaa[TyrIle] (SEQ ID NO:6) (c) [ArgIleGlnVal][GlyAla][LeuValLys][CysSer]Pro[PheTyr][AspAsn] (SEQ ID NO:7) [CysSer]Pro[PheTyr][AspAsn]Pro[AspGluArgLys][HisLeuAsp] (SEQ ID NO:8) GlyXaa[GlyAla]Xaa[CysSer]ProTyr[AspAsn]Pro (SEQ ID NO:9) (d) Leu[PheTyr]Ser[GlyAla]Thr[ValAsnAla]Ala (SEQ ID NO:10) Leu[PheTyr]SerXaaThrXaaAla[AspGlu][PheTyr] (SEQ ID NO:11) [PheTyr]Ser[GlyAla]Thr[ValAsnAla]Ala[AspGlu][PheTyr] (SEQ ID NO:12) (e) Leu[AsnAsp][AlaLys]ProAsnPheVal (SEQ ID NO:13) (f) PhePhePheArgGlu (SEQ ID NO:14) PhePhe[PheTyr]ArgGlu[ThrAsn] (SEQ ID NO:15) PhePheArgGlu[ThrAsn]Ala (SEQ ID NO:16) Phe[PheTyr]ArgGlu[ThrAsn]Ala (SEQ ID NO:17) TyrPhePhe[PheTyr]ArgGlu (SEQ ID NO:18) [PheTyr]PhePhe[PheTyr]ArgGlu (SEQ ID NO:19) [PheTyr][PheTyr][PheTyr]ArgGlu[ThrAsn]Ala (SEQ ID NO:20) [IleVal][PhrTyr]Phr[PhrTyr][PheTyr]ArgGlu (SEQ ID NO:21) Asp[LysPheTyr]Val[PheTyr][PheTyIleLeu][PheTyrIleLeu][PheTyr](SEQ ID NO:22) [ValIle][PheTyr][PheTyrIleLeu][PheTyrIleLeu]Phe[ArgThr]Xaa[ThrAsn]- (SEQ ID NO:23) [ValIle][PheTyr][PheTyrIleLeu][PheTyrIleLeu][PheTyr][ArgThr][GluAspVal][T- hrAsn] (SEQ ID NO:24) (g) Glu[PheTyr]IleAsn[CysSer]GlyLys (SEQ ID NO:25) [PheTyr]IleAsnCysGlyLys[AlaValIle] (SEQ ID NO:26) (h) Arg[ValIle][AlaGly][ArgGln][ValIle]CysLys (SEQ ID NO:27) Arg[ValIle]Xaa[ArgGln][ValIle]CysXaaXaaAsp (SEQ ID NO:28) GlyLys[ValAlalIle]XaaXaaXaaArg[ValAlalIle]XaaXaaXaaCysLys (SEQ ID NO:29) (i) [ArgLysAsn]Trp[ThrAlaSer][ThrAlaSer][PheTyrLeu]Leu[LysArg] (SEQ ID NO:30) [PheTyr]Leu[LysArg][AlaSer]ArgLeu[AsnIle]Cys (SEQ ID NO:3 1) [AsnlIle]CysSer[IleVal][ProSer]Gly (SEQ ID NO:32) Trp[ThrAlaSer][ThrAlaSer][PheTyrLeu]LeuLys[AlaSerValIleLeu]XaaLeu (SEQ ID NO:33) Trp[ThrAlaSer][ThrAlaSer]XaaLeuLysXaaXaaLeuXaaCys (SEQ ID NO:34) TrpXaa[ThrSer]XaaLeuLysXaaXaaLeuXaaCys (SEQ ID NO:35) (j) [PheTyr][PheTyr][AsnAsp]GluIleGlnSer (SEQ ID NO:36) [PheTyr]Pro[PheTyr][PheTyr][PheTyr][AsnAsp]Glu (SEQ ID NO:37) (k) GlySerAla[ValIleLeu]CysXaa[PheTyr] (SEQ ID NO:38) SerAla[ValIleLeu]CysXaa[PheTyr]XaaMet (SEQ ID NO:39) (l) AsnSer[AsnAla]TrpLeu[ProAla]Val (SEQ ID NO:40) (m) [ValLeulIle]Pro[GluAspTyrSerPhe]ProArgProGly (SEQ ID NO:41) [ValLeuIle]ProXaaPro[ArgAla]ProGlyXaaCys (SEQ ID NO:42) Pro[GluAspTyrSerPhe]ProArgProGly[ThrGlnSer]Cys (SEQ ID NO:43) (n) AspPro[HisPheTyr]Cys[AlaGly]Trp (SEQ ID NO:44) Pro[HisPheTyr]Cys[AlaGly]TrpAsp (SEQ ID NO:45) AspProXaaCys[AlaGly]TrpAsp (SEQ ID NO:46) CysXaaXaaXaaXaaAspProXaaCysXaaTrpAsp (SEQ ID NO:47) CysXaaXaaXaaAspProXaaCysXaaTrpAsp (SEQ ID NO:48) CysXaaXaaAspProXaaCysXaaTrpAsp (SEQ ID NO:49) CysXaaXaaCysXaaXaaXaaXaaAspXaaXaaCysXaaTrpAsp (SEQ ID NO:50) CysXaaXaaCysXaaXaaXaaAspXaaXaaCysXaaTrpAsp (SEQ ID NO:51) CysXaaXaaCysXaaXaaAspXaaXaaCysXaaTrpAsp (SEQ ID NO:52).

The following peptides represent particularly preferred members of each class: (a) AspCysGlnAsnTyrIle (SEQ ID NO:67) (b) CysGlyThr[AsnGly][AlaSer]XaaXaaPro (SEQ ID NO:68) (c) GlyXaa[SerCys]ProTyrAspPro (SEQ ID NO:69) (d) LeuTyrSerGlyThr[ValAsnAla]Ala (SEQ ID NO:70) (e) LeuAsnAlaProAsnPheVal (SEQ ID NO:71) (f) [PheTyr]PhePhe[PheTyr]ArgGlu (SEQ ID NO:19) (g) Glu[PheTyr]IleAsn[CysSer]GlyLys (SEQ ID NO:25) (h) Arg[ValIle]AlaArgValCysLys (SEQ ID NO:72) (i) Trp[ThrAla][ThrSer][PheTyr]LeuLys[AlaSer]ArgLeu (SEQ ID NO:73) (j) ProPheTyrPhe[AsnAsp]GluleGlnSer (SEQ ID NO:74) (k) GlySerAlaValCysXaa[PheTyr] (SEQ ID NO:75) (l) AsnSerAsnTrpLeu[ProAla]Val (SEQ ID NO:76) (m) Pro[GluAsp]ProArgProGly[ThrGlnSer]Cys (SEQ ID NO:77) (n) AspProTyrCys[AlaGly]TrpAsp (SEQ ID NO:78).

The following 14 classes are preferred peptides which exclude semaphorin peptides encoded in open reading frames of Variola major or Vaccinia viruses. (a) [AspGlu]Cys[GlnLysArgAlaAsn]Asn[TyrPheVal]Ile (SEQ ID NO:01) Cys[GlnLysArgAlaAsn]Asn[TyrPheVal]Ile[ArgLysGlnThr] (SEQ ID NO:02) (b) CysGlyThr[AsnGly][AlaSer][TyrPheHisGly][LysArgHisAsnGln] (SEQ ID NO:79) CysGlyThr[AsnGly][AlaSerAsn][TyrPheHis][LysArgHisAsnGln] (SEQ ID NO:80) CysGlyThr[AsnGly][AlaSer]XaaXaaPro (SEQ ID NO:81) (c)[ArgleGlnVal][GlyAla][LeuValLys][CysSer]Pro[PheTyr][AspAsn] (SEQ ID NO:07) [CysSer]Pro[PheTyr][AspAsn]Pro[AspGluArgLys][HisLeuAsp] (SEQ ID NO:08) GlyXaa[GlyAla]Xaa[CysSer]ProTyr[AspAsn]Pro (SEQ ID NO:09) (d) Leu[PheTyr]Ser[GlyAla]Thr[ValAsnAla]Ala (SEQ ID NO:10) Leu[PheTyr]SerXaaThrXaaAla[AspGlu][PheTyr] (SEQ ID NO:11) [PheTyr]Ser[GlyAla]Thr[ValAsnAla]Ala[AspGlu][PheTyr] (SEQ ID NO:12) (e) Leu[AsnAsp][AlaLys]ProAsnPheVal (SEQ ID NO:13) (f) PhePhePheArgGlu (SEQ ID NO:14) PhePhe[PheTyr]ArgGlu[ThrAsn] (SEQ ID NO:15) PhePheArgGlu[ThrAsn]Ala (SEQ ID NO:16) Phe[PheTyr]ArgGlu[ThrAsn]Ala (SEQ ID NO:17) TyrPhePhe[PheTyr]ArgGlu (SEQ ID NO:18) [PheTyr]PhePhe[PheTyr]ArgGlu (SEQ ID NO:19) [PheTyr][PheTyr][PheTyr]ArgGlu[ThrAsn]Ala (SEQ ID NO:20) [IleVal][PheTyr]Phe[PheTyr][PheTyr]ArgGlu (SEQ ID NO :21) Asp [LysPheTyr]Val [PheTyr][PheTyrLeu][PheTyrIleLeu][PheTyr] (SEQ ID NO:22) Asp[LysPheTyr]Val[PheTyr][PheTyrIleLeu][PheTyrIle][PheTyr] (SEQ ID NO:82) [ValIle][PheTyr][PheTyrLeu][PheTyrIleLeu]Phe[ArgThr]Xaa[ThrAsn] (SEQ ID NO:83) [ValIle][PheTyr][PheTyrIleLeu][PheTyrIle]Phe[ArgThr]Xaa[ThrAsn] (SEQ ID NO:84) [ValIle][PheTyr][PheTyrIleLeu][PheTyrIleLeu]PheArgXaa[ThrAsn] (SEQ ID NO:85) [ValIle][PheTyr][PheTyrLeu][PheTyrIleLeu][PheTyr][ArgThr][GluAspVa- l][ThrAsn] (SEQ ID NO: 86) (g) Glu[PheTyr]IleAsn[CysSer]GlyLys (SEQ ID NO:25) [PheTyr]IleAsnCysGlyLys[AlaValIle] (SEQ ID NO:26) (h) Arg[ValIle][AlaGly][ArgGln][ValIle]CysLys (SEQ ID NO:27) Arg[ValIle]Xaa[ArgGln][ValIle]CysXaaXaaAsp (SEQ ID NO:28) GlyLys[ValAlalIle]XaaXaaXaaArg[ValAlalIle]XaaXaaXaaCysLys (SEQ ID NO:29) (i) [ArgLysAsn]Trp[ThrAla][ThrAlaSer][PheTyrLeu]Leu[LysArg] (SEQ ID NO:87) [PheTyr]Leu[LysArg][AlaSer]ArgLeu[AsnIle]Cys (SEQ ID NO:31) [AsnIle]CysSer[IleVal][ProSer]Gly (SEQ ID NO:32) Trp[ThrAla][ThrAlaSer][PheTyrLeu]LeuLys[AlaSerValIleLeu]XaaLeu (SEQ ID NO:88) Trp[ThrAlaSer][ThrAlaSer][PheTyrLeu]LeuLys[AlaSerIleLeu]XaaLeu (SEQ ID NO:89) Trp[ThrAla][ThrAlaSer]XaaLeuLysXaaXaaLeuXaaCys (SEQ ID NO:90) (j) [PheTyr][PheTyr][AsnAsp]GluIleGlnSer (SEQ ID NO:36) [PheTyr]Pro[PheTyr][PheTyr][PheTyr][AsnAsp]Glu (SEQ ID NO:37) (k) GlySerAla[ValIleLeu]CysXaa[PheTyr] (SEQ ID NO:38) SerAla[ValIle]CysXaa[PheTyr]XaaMet (SEQ ID NO:39) (1) AsnSer[AsnAla]TrpLeu[ProAla]Val (SEQ ID NO:40) (m) [ValLeuIle]Pro[GluAspTyrSerPhe]ProArgProGly (SEQ ID NO:41) [ValLeuIle]ProXaaProArgProGlyXaaCys (SEQ ID NO:91) Pro[GluAspTyrSerPhe]ProArgProGly[ThrGlnSer]Cys (SEQ ID NO:43) (n) AspPro[HisPheTyr]Cys[AlaGly]Trp (SEQ ID NO:44) Pro[HisPheTyr]Cys[AlaGly]TrpAsp (SEQ ID NO:45) AspProXaaCys[AlaGly]TrpAsp (SEQ ID NO:46) CysXaaXaaXaaXaaAspProXaaCysXaaTrpAsp (SEQ ID NO:47) CysXaaXaaXaaAspProXaaCysXaaTrpAsp (SEQ ID NO:48) CysXaaXaaAspProXaaCysXaaTrpAsp (SEQ ID NO:49) CysXaaXaaCysXaaXaaXaaXaaAspXaaXaaCysXaaTrpAsp (SEQ ID NO:50) CysXaaXaaCysXaaXaaXaaAspXaaXaaCysXaaTrpAsp (SEQ ID NO :51) CysXaaXaaCysXaaXaaAspXaaXaaCysXaaTrpAsp (SEQ ID NO:52).

The following 2 classes are preferred peptides which exclude semaphorin peptides encoded in open reading frames of Variola major or Vaccinia viruses Grasshopper Semaphorin I. (f) TyrPhePhe[PheTyr]ArgGlu (SEQ ID NO:18) Asp[LysTyr]Val[PheTyr][PheTyrLeu][PheTyrIleLeu][PheTyr] (SEQ ID NO:92) Asp[LysTyr]Val[PheTyr][PheTyrIleLeu][PheTyrIle][PheTyr] (SEQ ID NO:93) [ValIle]Tyr[PheTyrLeu][PheTyrIleLeu]Phe[ArgThr]Xaa[ThrAsn] (SEQ ID NO:94) [ValIle]Tyr[PheTyrIleLeu][PheTyrIle]Phe[ArgThr]Xaa[ThrAsn] (SEQ ID NO:95) [ValIle]Tyr[PheTyrIleLeu][PheTyrIleLeu]PheArgXaa[ThrAsn] (SEQ ID NO:96) Val[PheTyr][PheTyrLeu][PheTyrIleLeu][PheTyr][ArgThr][GluAspVal][Th- rAsn] (SEQ ID NO: 97) Val[PheTyr][PheTyrIleLeu][PheTyrIle][PheTyr][ArgThr][GluAspVal][ThrAsn] (SEQ ID NO: 98) Val[PheTyr][PheTyrIleLeu][PheTyrIleLeu][PheTyr]Arg[GluAspVal][ThrAsn] (SEQ ID NO: 99) (n) CysXaaXaaXaAspProXaaCysXaaTrpAsp (SEQ ID NO:48) CysXaaXaaAspProXaaCysXaaTrpAsp (SEQ ID NO:49) CysXaaXaaCysXaaXaaXaaAspXaaXaaCysXaaTrpAsp (SEQ ID NO:51) CysXaaXaaCysXaaXaaAspXaaXaaCysXaaTrpAsp (SEQ ID NO:52).

The following 5 classes include peptides which encompass peptides encoded in open reading frames of Variola major or Vaccinia viruses. Accordingly, in the event that these viral peptides are not novel per se, the present invention discloses a hitherto unforseen and unforseeable utility for these peptides as immunosuppressants and targets of anti-viral therapy. (b) CysGlyThr[AsnGly][AlaSerAsn][TyrPheHisGly][LysArgHisAsnGln] (SEQ ID NO:03) CysGlyThr[AsnGly][AlaSerAsn]XaaXaaPro (SEQ ID NO:04) CysGlyThr[AsnGly]XaaXaaXaaProXaa[CysAsp] (SEQ ID NO:05) CysGlyThrXaaXaaXaaXaaProXaa[CysAsp]XaaXaa[TyrIle] (SEQ ID NO:06) (f) Asp[LysPheTyr]Val[PheTyr][PheTyrIleLeu][PheTyrIleLeu][PheTyr] (SEQ ID NO:22) [ValIle][PheTyr][PheTyrIleLeu][PheTyrIleLeu]Phe[ArgThr]Xaa[ThrAsn] (SEQ ID NO:23) Val[PheTyr][PheTyrIleLeu][PheTyrIleLeu][PheTyr][ArgThr][GluAspVal][ThrAsn- ] (SEQ ID NO: 100) (i) [ArgLysAsn]Trp[ThrAlaSer][ThrAlaSer][PheTyrLeu]Leu[LysArg] (SEQ ID NO:30) Trp[ThrAlaSer][ThrAlaSer][PheTyrLeu]LeuLys[AlaSerValIleLeu]XaaLeu (SEQ ID NO: 33) Trp[ThrAlaSer][ThrAlaSer]XaaLeuLysXaaXaaLeuXaaCys (SEQ ID NO:34) TrpXaa[ThrSer]XaaLeuLysXaaXaaLeuXaaCys (SEQ ID NO:35) (k) SerAla[ValIleLeu]CysXaa[PheTyr]XaaMet (SEQ ID NO:39) (m) [ValLeuIle]ProXaaPro[ArgAla]ProGlyXaaCys (SEQ ID NO:42).

The disclosed semaphorin sequence data are used to define a wide variety of other semaphorin- and semaphorin receptor-specific binding agents using immunologic, chromatographic or synthetic methods available to those skilled in the art.

Of particular significance are peptides comprising unique portions of semaphorin-specific receptors and polypeptides comprising a sequence substantially similar to that of a substantially full-length semaphorin receptor. Using semaphorin peptides, these receptors are identified by a variety of techniques known to those skilled in the art where a ligand to the target receptor is known, including expression cloning as set out in the exemplification below. For other examples of receptor isolation with known ligand using expression cloning, see, Staunton et al (1989) Nature 339, 61; Davis et al (1991) Science 253, 59; Lain et al (1992) Cell 68, 775; Gearing et al (1989) EMBO 8, 3667; Aruffo and Seed (1987) PNAS 84, 8573 and references therein. Generally, COS cells are transfected to express a cDNA library or PCR product and cells producing peptides/polypeptides which bind a semaphorin/receptor peptide/polypeptide are isolated. For neurosemaphorin receptors, fetal brain cDNA libraries are preferred; for immunosemaphorin receptors, libraries derived from activated lymphoid or myeloid cell lines or tissue derived from sites of inflammation or delayed-type hypersensitivity are preferred; and for semaphorin and semaphorin receptor variants used by tumor cells to evade immune surveillance or suppress an immune response (oncosemaphorins), libraries derived from cancerous tissue or tumor cell lines resistant to the host immune system are preferred. Alternatively, PCR primers based upon known semaphorin/receptor sequences such as those disclosed herein are used to amplify PCR product from such tissues/cells. Other receptor/ligand isolation methods using immobilized ligand or antibody are known to those skilled in the art.

Semaphorin receptor peptides with receptor binding specificity are identified by a variety of ways including having conserved consensus sequences with other semaphorin receptors, by crosslinking to ligand or receptor-specific antibody, or preferably, by screening such peptides for semaphorin binding or disruption of semaphorin-receptor binding. Methods for identifying semaphorin receptor peptides with the requisite binding activity are described herein or otherwise known to those skilled in the art. By analogous methods, semaphorin receptor peptides are used to define additional semaphorin peptides with semaphorin binding specificity, particularly receptor specificity.

The various semaphorin and semaphorin receptor peptides are used to define functional domains of semaphorins, identify compounds that associate with semaphorins, design compounds capable of modulating semaphorin-mediated nerve and immune cell function, and define additional semaphorin and semaphorin receptor-specific binding agents. For example, semaphorin mutants, including deletion mutants are generated from the disclosed semaphorin sequences and used to identify regions important for specific protein-ligand or protein-protein interactions, for example, by assaying for the ability to mediate repulsion or preclude aggregation in cell-based assays as described herein. Further, x-ray crystallographic data of the disclosed protein are used to rationally design binding molecules of determined structure or complementarity for modulating growth cone growth and guidance.

Additional semaphorin- and receptor-specific agents include specific antibodies that can be modified to a monovalent form, such as Fab, Fab', or Fv, specifically binding oligopeptides or oligonucleotides and most preferably, small molecular weight organic receptor antagonists. For example, the disclosed semaphorin and receptor peptides are used as immunogens to generate semaphorin- and receptor-specific polyclonal or monoclonal antibodies. See, Harlow and Lane (1988) Antibodies, A Laboratory Manual, Cold Spring Harbor Laboratory, for general methods. Anti-idiotypic antibody, especially internal imaging anti-ids are also prepared using the disclosures herein.

In addition to semaphorin and semaphorin-receptor derived polypeptides and peptides, other prospective agents are screened from large libraries of synthetic or natural compounds. For example, numerous means are available for random and directed synthesis of saccharide, peptide, and nucleic acid based compounds. Alternatively, libraries of natural compounds in the form of bacterial, fungal, plant and animal extracts are available or readily producible. Additionally, natural and synthetically produced libraries and compounds are readily modified through conventional chemical, physical, and biochemical means. See, e.g. Houghten et al. and Lam et al (1991) Nature 354, 84 and 81, respectively and Blake and Litzi-Davis (1992), Bioconjugate Chem 3, 510.

Useful agents are identified with a range of assays employing a compound comprising the subject peptides or encoding nucleic acids. A wide variety of in vitro, cell-free binding assays, especially assays for specific binding to immobilized compounds comprising semaphorin or semaphorin receptor peptide find convenient use. While less preferred, cell-based assays may be used to determine specific effects of prospective agents on semaphorin-receptor binding may be assayed. Optionally, the intracellular C-terminal domain is substituted with a sequence encoding a oligopeptide or polypeptide domain that provides a detectable intracellular signal upon ligand binding different from the natural receptor. Useful intracellular domains include those of the human insulin receptor and the TCR, especially domains with kinase activity and domains capable of triggering calcium influx which is conveniently detected by fluorimetry by preloading the host cells with Fura-2. More preferred assays involve simple cell-free in vitro binding of candidate agents to immobilized semaphorin or receptor peptides, or vice versa. See, e.g. Fodor et al (1991) Science 251, 767 for light directed parallel synthesis method. Such assays are amenable to scale-up, high throughput usage suitable for volume drug screening.

Useful agents are typically those that bind to a semaphorin or disrupt the association of a semaphorin with its receptor. Preferred agents are semaphorin-specific and do not cross react with other neural or lymphoid cell membrane proteins. Useful agents may be found within numerous chemical classes, though typically they are organic compounds; preferably small organic compounds. Small organic compounds have a molecular weight of more than 150 yet less than about 4,500, preferably less than about 1500, more preferably, less than about 500. Exemplary classes include peptides, saccharides, steroids, heterocyclics, polycyclics, substituted aromatic compounds, and the like.

Selected agents may be modified to enhance efficacy, stability, pharmaceutical compatibility, and the like. Structural identification of an agent may be used to identify, generate, or screen additional agents. For example, where peptide agents are identified, they may be modified in a variety of ways as described above, e.g. to enhance their proteolytic stability. Other methods of stabilization may include encapsulation, for example, in liposomes, etc.

The subject binding agents may be prepared in a variety of ways known to those skilled in the art. For example, peptides under about 60 amino acids can be readily synthesized today using conventional commercially available automatic synthesizers. Alternatively, DNA sequences may be prepared encoding the desired peptide and inserted into an appropriate expression vector for expression in a prokaryotic or eukaryotic host. A wide variety of expression vectors are available today and may be used in conventional ways for transformation of a competent host for expression and isolation. If desired, the open reading frame encoding the desired peptide may be joined to a signal sequence for secretion, so as to permit isolation from the culture medium. Methods for preparing the desired sequence, inserting the sequence into an expression vector, transforming a competent host, and growing the host in culture for production of the product may be found in U.S. Pat. Nos. 4,710,473, 4,711,843 and 4,713,339.

For therapeutic uses, the compositions and agents disclosed herein may be administered by any convenient way, preferably parenterally, conveniently in a pharmaceutically or physiologically acceptable carrier, e.g., phosphate buffered saline, saline, deionized water, or the like. Typically, the compositions are added to a retained physiological fluid such as blood or synovial fluid. For CNS administration, a variety of techniques are available for promoting transfer of the therapeutic across the blood brain barrier including disruption by surgery or injection, drugs which transciently open adhesion contact between CNS vasculature endothelial cells, and compounds which fascilitate translocation through such cells. As examples, many of the disclosed therapeutics are amenable to directly injected or infused, contained within implants e.g. osmotic pumps, grafts comprising appropriately transformed cells. Generally, the amount administered will be empirically determined, typically in the range of about 10 to 1000 .mu.g/kg of the recipient. For peptide agents, the concentration will generally be in the range of about 50 to 500 .mu.g/ml in the dose administered. Other additives may be included, such as stabilizers, bactericides, etc. These additives will be present in conventional amounts.

The invention provides isolated nucleic acid sequences encoding the disclosed semaphorin and semaphorin receptor peptides and polypeptides, including sequences substantially identical to sequences encoding such polypeptides. An "isolated" nucleic acid sequence is present as other than a naturally occurring chromosome or transcript in its natural state and typically is removed from at least some of the nucleotide sequences with which it is normally associated with on a natural chromosome. A complementary sequence hybridizes to a unique portion of the disclosed semaphorin sequence under low stringency conditions, for example, at 50.degree. C. and SSC (0.9 M saline/0.09 M sodium citrate) and that remains bound when subject to washing at 55.degree. C with SSC. Regions of non-identity of complementary nucleic acids are preferably or in the case of homologous nucleic acids, a nucleotide change providing a redundant codon. A partially pure nucleotide sequence constitutes at least about 5%, preferably at least about 30%, and more preferably at least about 90% by weight of total nucleic acid present in a given fraction.

Unique portions of the disclosed nucleic acid sequence are of length sufficient to distinguish previously known nucleic acid sequences. Thus, a unique portion has a nucleotide sequence at least long enough to define a novel oligonucleotide. Preferred nucleic acid portions encode a unique semaphorin peptide. The nucleic acids of the invention and portions thereof, other than those used as PCR primers, are usually at least about 60 bp and usually less than about 60 kb in length. PCR primers are generally between about 15 and 100 nucleotides in length.

Nucleotide (cDNA) sequences encoding several full length semaphorins are disclosed herein. The invention also provides for the disclosed sequences modified by transitions, transversions, deletions, insertions, or other modifications such as alternative splicing and also provides for genomic semaphorin sequences, and gene flanking sequences, including regulatory sequences; included are DNA and RNA sequences, sense and antisense. Preferred DNA sequence portions include portions encoding the preferred amino acid sequence portions disclosed above. For antisense applications where the inhibition of semaphorin expression is indicated, especially useful oligonucleotides are between about 10 and 30 nucleotides in length and include sequences surrounding the disclosed ATG start site, especially the oligonucleotides defined by the disclosed sequence beginning about 5 nucleotides before the start site and ending about 10 nucleotides after the disclosed start site. Other especially useful semaphorin mutants involve deletion or substitution modifications of the disclosed cytoplasmic C-termini of transmembrane semaphorins. Accordingly, semaphorin mutants with semaphorin binding affinities but with altered intracellular signal transduction capacities are produced.

For modified semaphorin-encoding sequences or related sequences encoding proteins with semaphorin-like functions, there will generally be substantial sequence identity between at least a segment thereof and a segment encoding at least a portion of the disclosed semaphorin sequence, preferably at least about 60%, more preferably at least 80%, most preferably at least 90% identity. Homologous segments are particularly within semaphorin domain-encoding regions and regions encoding protein domains involved in protein-protein, particularly semaphorin-receptor interactions and differences within such segments are particularly conservative substitutions.

Typically, the invention's semaphorin peptide encoding polynucleotides are associated with heterologous sequences. Examples of such heterologous sequences include regulatory sequences such as promoters, enhancers, response elements, signal sequences, polyadenylation sequences, etc., introns, 5' and 3' noncoding regions, etc. Other useful heterologous sequences are known to those skilled in the art or otherwise disclosed references cited herein. According to a particular embodiment of the invention, portions of the semaphorin encoding sequence are spliced with heterologous sequences to produce soluble, secreted fusion proteins, using appropriate signal sequences and optionally, a fusion partner such as .beta.-Gal.

The disclosed sequences are also used to identify and isolate other natural semaphorins and analogs. In particular, the disclosed nucleic acid sequences are used as hybridization probes under low-stringency or PCR primers, e.g. oligonucleotides encoding functional semaphorin domains are .sup.32P-labeled and used to screen .lamda.cDNA libraries at low stringency to identify similar cDNAs that encode proteins with related functional domains. Additionally, nucleic acids encoding at least a portion of the disclosed semaphorin are used to characterize tissue specific expression of semaphorin as well as changes of expression over time, particularly during organismal development or cellular differentiation.

The semaphorin encoding nucleic acids can be subject to alternative purification, synthesis, modification, sequencing, expression, transfection, administration or other use by methods disclosed in standard manuals such as Molecular Cloning, A Laboratory Manual (2nd Ed., Sambrook, Fritsch and Maniatis, Cold Spring Harbor), Current Protocols in Molecular Biology (Eds. Aufubel, Brent, Kingston, More, Feidman, Smith and Stuhl, Greene Publ. Assoc., Wiley-Interscience, NY, N.Y., 1992) or that are otherwise known in the art. For example, the nucleic acids can be modified to alter stability, solubility, binding affinity and specificity, etc. semaphorin-encoding sequences can be selectively methylated, etc. The nucleic acid sequences of the present invention may also be modified with a label capable of providing a detectable signal, either directly or indirectly. Exemplary labels include radioisotopes, fluorescers, biotinylation, etc.

The invention also provides vectors comprising nucleic acids encoding semaphorin peptides, polypeptides or analogs. A large number of vectors, including plasmid and viral vectors, have been described for expression in a variety of eukaryotic and prokaryotic hosts. Advantageously, vectors may also include a promotor operably linked to the semaphorin-encoding portion. Vectors will often include one or more replication systems for cloning or expression, one or more markers for selection in the host, e.g. antibiotic resistance. The inserted semaphorin coding sequences may be synthesized, isolated from natural sources, prepared as hybrids, etc. Suitable host cells may be transformed/transfected/infected by any suitable method including electroporation, CaCl.sub.2 mediated DNA uptake, viral infection, microinjection, microprojectile, or other methods.

Appropriate host cells include bacteria, archebacteria, fungi, especially yeast, and plant and animal cells, especially mammalian cells. Of particular interest are E. coli, B. subtilis, Saccharomyces cerevisiae, SF9 cells, C129 cells, 293 cells, Neurospora, and CHO, COS, HeLa cells, immortalized mammalian myeloid and lymphoid cell lines, and pluripotent cells, especially mammalian ES cells and zygotes. Preferred replication systems include M13, ColE1, SV40, baculovirus, lambda, adenovirus, AAV, BPV, etc. A large number of transcription initiation and termination regulatory regions have been isolated and shown to be effective in the transcription and translation of heterologous proteins in the various hosts. Examples of these regions, methods of isolation, manner of manipulation, etc. are known in the art. Under appropriate expression conditions, host cells can be used as a source of recombinantly produced semaphorins or analogs.

For the production of stably transformed cells and transgenic animals, nucleic acids encoding the disclosed semaphorins may be integrated into a host genome by recombination events. For example, such a sequence can be microinjected into a cell, and thereby effect homologous recombination at the site of an endogenous gene, an analog or pseudogene thereof, or a sequence with substantial identity to an semaphorin-encoding gene. Other recombination-based methods such as nonhomologous recombinations, deletion of endogenous gene by homologous recombination, especially in pluripotent cells, etc., provide additional applications. Preferred transgenics and stable transformants over-express the disclosed receptor gene and find use in drug development and as a disease model. Alternatively, knock-out cells and animals find use in development and functional studies. Methods for making transgenic animals, usually rodents, from ES cells or zygotes are known to those skilled in the art.

The compositions and methods disclosed herein may be used to effect gene therapy. See, e.g. Zhu et al. (1993) Science 261, 209 211; Gutierrez et al. (1992) Lancet 339, 715 721. For example, cells are transfected with semaphorin sequences operably linked to gene regulatory sequences capable of effecting altered semaphorin expression or regulation. To modulate semaphorin translation, cells may be transfected with complementary antisense polynucleotides. For gene therapy involving the transfusion of semaphorin transfected cells, administration will depend on a number of variables that are ascertained empirically. For example, the number of cells will vary depending on the stability of the transfused cells. Transfusion media is typically a buffered saline solution or other pharmacologically acceptable solution. Similarly the amount of other administered compositions, e.g. transfected nucleic acid, protein, etc., will depend on the manner of administration, purpose of the therapy, and the like.

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

EXAMPLES

I. Isolation and Characterization of Grasshopper Semaphorin I (SEQ ID NOS: 57 and 58(Previously Referred to as Fasciclin IV)

In order to identify cell surface molecules that function in selective fasciculation, a series of monoclonal antibody (MAb) screens was conducted. The immunogen used for most of these screens was membranes from the longitudinal connectives (the collection of longitudinal axons) between adjacent segmental ganglia of the nervous system of the larval grasshopper. From these screens, MAb 3B11 and 8C6 were used to purify and characterize two surface glycoproteins, fasciclin I and fasciclin II, see, Bastiani et al., 1987; the genes encoding both were subsequently cloned, see, Snow et al. 1989, Zinn et al. 1988, and Harrelson and Goodman, 1988.

Another MAb isolated during these screens, MAb 6F8, was chosen for the present study because, just as with fasciclin I and fasciclin II, the antigen recognized by this MAb is expressed on a different but overlapping subset of axon pathways in the developing CNS. The 6F8 antigen appears to be localized on the outside of cell surfaces, as indicated by MAb binding when incubated both in live preparations, and in fixed preparations in which no detergents have been added. Because the 6F8 antigen is a surface glycoprotein expressed on a subset of axon fascicles (see below), we call it fasciclin IV.

Fasciclin IV expression begins early in embryonic development before axonogenesis. At 29% of development, expression is seen on the surface of the midline mesectodermal cells and around 5 7 neuroblasts and associated ectodermal cells per hemisegment. This expression is reminiscent of the mesectodermal and neuroblast-associated expression observed with both fasciclin I and fasciclin II; however, in each case, the pattern resolves into a different subset of neuroblasts and associated ectodermal cells.

At 32% of development, shortly after the onset of axonogenesis in the CNS, fasciclin IV expression is seen on the surface of the axons and cell bodies of the three pairs of MP4, MP5, and MP6 midline progeny, the three U motoneurons, and on several unidentified neurons in close proximity to the U's. This is in contrast to fasciclin II, which at this stage is expressed on the MP1 and dMP2 neurons, and fasciclin I, which is expressed on the U neurons but not on any midline precursor progeny.

The expression of fasciclin IV on a subset of axon pathways is best observed around 40% of development, after the establishment of the first longitudinal and commissural axon pathways. At this stage, the protein is expressed on two longitudinal axon fascicles, a subset of commissural axon fascicles, a tract extending anteriorly along the midline, and a subset of fascicles in the segmental nerve (SN) and intersegmental nerve (ISN) roots.

Specifically, fasciclin IV is expressed on the U fascicle, a longitudinal pathway (between adjacent segmental neuromeres) pioneered in part by the U neurons, and on the A/P longitudinal fascicle (in part an extension of the U fascicle within each segmental neuromere). In addition, fasciclin IV is also expressed on a second narrower, medial, and more ventral longitudinal pathway. The U axons turn and exit the CNS as they pioneer the ISN; the U's and many other axons within the ISN express fasciclin IV. The continuation of the U fascicle posterior to the ISN junction is also fasciclin IV-positive. The specificity of fasciclin IV for distinct subsets of longitudinal pathways can be seen by comparing fasciclin IV and fasciclin II expression in the same embryo; fasciclin IV is expressed on the U and A/P pathways whereas fasciclin II is expressed on the MP1 pathway.

The axons in the median fiber tract (MFT) also express fasciclin IV. The MFT is pioneered by the three pairs of progeny of the midline precursors MP4, MP5, and MP6. The MFT actually contains three separate fascicles. The axons of the two MP4 progeny pioneer the dorsal MFT fascicle and then bifurcate at the posterior end of the anterior commissure; whereas the axons of the two MP6 progeny pioneer the ventral MFT fascicle and then bifurcate at the anterior end of the posterior commissure. Fasciclin IV is expressed on the cell bodies of the six MP4, MP5, and MP6 neurons, and on their growth cones and axons as they extend anteriorly in the MFT and bifurcate in one of the two commissures. However, this expression is regional in that once these axons bifurcate and begin to extend laterally across the longitudinal pathways and towards the peripheral nerve roots, their expression of fasciclin IV greatly decreases. Thus, fasciclin IV is a label for the axons in the MFT and their initial bifurcations in both the anterior and posterior commissures. It appears to be expressed on other commissural fascicles as well. However, the commissural expression of fasciclin IV is distinct from the transient expression of fasciclin II along the posterior edge of the posterior commissure, or the expression of fasciclin I on several different commissural axon fascicles in both the anterior and posterior commissure (Bastiani et al., 1987; Harrelson and Goodman, 1988).

Fasciclin IV is also expressed on a subset of motor axons exiting the CNS in the SN. The SN splits into two major branches, one anterior and the other posterior, as it exits the CNS. Two large bundles of motoneuron axons in the anterior branch express fasciclin IV at high levels; one narrow bundle of motoneuron axons in the posterior branch expresses the protein at much lower levels. Fasciclin IV is also expressed on many of the axons in the ISN.

The CNS and nerve root expression patterns of fasciclin IV, fasciclin I, and fasciclin II around 40% of embryonic development are summarized below. Although there is some overlap in their patterns (e.g., both fasciclin IV and fasciclin I label the U axons), these three surface glycoproteins label distinct subsets of axon pathways in the developing CNS.

Fasciclin IV is Expressed on Epithelial Bands in the Developing Limb Bud

Fasciclin IV is expressed on the developing limb bud epithelium in circumferential bands; at 34.5% of development these bands can be localized with respect to constrictions in the epithelium that mark presumptive segment boundaries. In addition to a band just distal to the trochanter/coxa segment boundary, bands are also found in the tibia, femur, coxa, and later in development a fifth band is found in the tarsus. Fasciclin IV is also expressed in the nascent chordotonal organ in the dorsal aspect of the femur. The bands in the tibia, trochanter, and coxa completely encircle the limb. However, the femoral band is incomplete, containing a gap on the anterior epithelia of this segment.

The position of the Ti1 axon pathway with respect to these bands of fasciclin IV-positive epithelia suggests a potential role for fasciclin IV in guiding the Ti1 growth cones. First, the band of fasciclin IV expression in the trochanter, which is approximately three epithelial cell diameters in width when encountered by the Ti1 growth cones, is the axial location where the growth cones reorient from proximal migration to circumferential branch extension. The Tr1 cell, which marks the location of the turn, lies within this band, usually over the central or the proximal cell tier. Secondly, although there is a more distal fasciclin IV expressing band in the femur, where a change in Ti1 growth is not observed, there exists a gap in this band such that fasciclin IV expressing cells are not traversed by the Ti1 growth cones. The Ti1 axons also may encounter a fasciclin IV expressing region within the coxa, where interactions between the growth cones, the epithelial cells, and the Cx1 guidepost cells have not yet been investigated. In addition to its expression over the surface of bands of epithelial cells, fasciclin IV protein, as visualized with MAb 6F8, is also found on the basal surface of these cells in a punctate pattern. This punctate staining is not an artifact of the HRP immunocytochemistry since fluorescent visualization of MAb 6F8 is also punctate. The non-neuronal expression of fasciclin IV is not restricted to limb buds. Circumferential epithelial bands of fasciclin IV expression are also seen on subesophageal mandibular structures and on the developing antennae.

MAb Directed Against Fasciclin IV can Alter the Formation of the Ti1 Axon Pathway in the Limb Bud

The expression of fasciclin IV on an epithelial band at a key choice point in the formation of the Ti1 axon pathway led us to ask whether this protein is involved in growth cone guidance at this location. To answer this question, we cultured embryos, or epithelial fillets (e.g., O'Connor et al., 1990), during the 5% of development necessary for normal pathway formation, either in the presence or absence of MAb 6F8 or 6F8 Fab fragments. Under the culture conditions used for these experiments, defective Ti1 pathways are observed in 14% of limbs (Chang et al., 1992); this defines the baseline of abnormalities observed using these conditions. For controls we used other MAbs and their Fab fragments that either bind to the surfaces of these neurons and epithelial cells (MAb 3B11 against the surface protein fasciclin I) or do not (MAb 4D9 against the nuclear protein engrailed; Patel et al., 1989). To assess the impact of MAb 6F8 on Ti1 pathway formation, we compared the percentage of aberrant pathways observed following treatment with MAb 6F8 to that observed with MAbs 3B11 and 4D9. Our cultures began at 32% of development when the Ti1 growth cones have not yet reached the epithelium just distal to the trochanter/coxa boundary and therefore have not encountered epithelial cells expressing fasciclin IV. Following approximately 30 hours in culture (.about.4% of development), embryos were fixed and immunostained with antibodies to HRP in order to visualize the Ti1 axons and other neurons in the limb bud. Criteria for scoring the Ti1 pathway, and the definition of "aberrant", are described in detail in the Experimental Procedures.

Although MAb 6F8 does not arrest pathway formation, several types of distinctive, abnormal pathways are observed. These defects generally begin where growth cones first contact the fasciclin IV expressing cells in the trochanter. Normally, the Ti1 neurons each have a single axon, and the axons of the two cells are fasciculated in that portion of the pathway within the trochanter. Following treatment with MAb 6F8, multiple long axon branches are observed within, and proximal to, the trochanter. Two major classes of pathways are taken by these branches; in 36% of aberrant limbs, multiple, long axon branches extend ventrally in the region distal to the Cx1 cells which contains the band of fasciclin IV expressing epithelial cells. In the ventral region of the trochanter, these branches often independently turn proximally to contact the Cx1 cells, and thus complete the pathway in this region.

In the second major class of pathway defect, seen in 47% of aberrant limbs, axon branches leave the trochanter at abnormal, dorsal locations, and extend proximally across the trochanter/coxa boundary. These axons then veer ventrally, often contacting the Cx1 neurons. The remaining 17% of defects include defasciculation distal to the trochanter, axon branches that fail to turn proximally in the ventral trochanter and continue into the posterior compartment of the limb, and axon branches which cross the trochanter/coxa boundary and continue to extend proximally without a ventral turn.

When cultured in the presence of MAb 6F8, 43% of limbs exhibited malformed Ti1 pathways (n=381) as compared to 11% with MAb 3B11 (n=230) and 5% with MAb 4D9 (n=20). These percentages are pooled from treatments with MAbs concentrated from hybridoma supernatant, IgGs isolated from these supernatants, and Fab fragments isolated from these IgG preparations (see Experimental Procedures). The frequency of malformed Ti1 pathways and the types of defects observed showed no significant variation regardless of the method of antibody preparation or type of antibody used. Since Fabs show similar results as IgGs, the effects of MAb 6F8 are not due to cross linking by the bivalent IgG.

In summary, following treatment with MAb 6F8, the Ti1 pathway typically exhibits abnormal morphology beginning just distal to the trochanter and at t