A Novel Receptor Tyrosine Phosphatase-a That Is Highly Expressed in the Nervous System*

A novel transmembrane receptor protein tyrosine phosphatase-a (RPTP-a) was cloned from a rat brain stem cDNA library. The extracellular segment of one form of RPTP-a contains 824 amino acids and is com- posed of three immunoglobulin-like and five fibronectin type I11 (FNII1)-like repeats. The 627-amino acid cytoplasmic region of RPTP-a consists of two catalytic domains oriented in tandem. Northern blot analyses indicate that RPTP-a is highly expressed in the brain 88 two major transcripts of 5.7 and 6.9 kilobases (kb). The 5.7-kb transcript is expressed exclusively in the brain while the 6.9-kb species can be detected in the lung and heart, but at significantly lower levels. In situ hybridization studies confirm that RPTP-u is lo-calized predominately in the nervous system and can be detected in the rat as early as embryonic day 12. During RPTP-u is expressed extensively in the central and peripheral nervous sys- tems, including the trigeminal and dorsal root ganglia as as the In expression is restricted primarily to the cerebellum, and hippocampus. Within the structure, RPTP-u is cleotidyl primers for PCR, RNA-PCR, for sequencing, oligonucleotide probe for in situ hybridization, an automated DNA synthesizer Biosystems,

A novel transmembrane receptor protein tyrosine phosphatase-a (RPTP-a) was cloned from a rat brain stem cDNA library. The extracellular segment of one form of RPTP-a contains 824 amino acids and is composed of three immunoglobulin-like and five fibronectin type I11 (FNII1)-like repeats. The 627-amino acid cytoplasmic region of RPTP-a consists of two catalytic domains oriented in tandem. Northern blot analyses indicate that RPTP-a is highly expressed in the brain 88 two major transcripts of 5.7 and 6.9 kilobases (kb). The 5.7-kb transcript is expressed exclusively in the brain while the 6.9-kb species can be detected in the lung and heart, but at significantly lower levels. In situ hybridization studies confirm that RPTP-u is localized predominately in the nervous system and can be detected in the rat as early as embryonic day 12. During embryonic development, RPTP-u is expressed extensively in the central and peripheral nervous systems, including the trigeminal and dorsal root ganglia as well as the retina. In adult rat brain, expression is restricted primarily to the olfactory tubercule, cerebellum, and hippocampus. Within the latter structure, RPTP-u is present in the pyramidal cell layer and granular layer of the dentate gyrus. Transfection of RPTP-a cDNA into human embryonic kidney 293 cells results in the synthesis of a protein with an apparent molecular mass of 200 kDa as detected by immunoprecipitation and immunoblot analyses using polyclonal antibodies against the FNIII-like repeats present in the extracellular domain of RPTP-a. The gene for RPTPa has been mapped to distal chromosome 17 in the mouse.
Protein tyrosine phosphorylation is critical for regulation of normal cell growth, proliferation, and differentiation. The extent of phosphorylation of tyrosine residues on cellular proteins is controlled by both protein tyrosine kinases (1) and protein tyrosine phosphatases (PTPs)' (2). While a variety * This work was supported by Sugen Inc. (to J. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
ll Recipient of Grant MH-17785. protein tyrosine kinases have been identified and well characterized ( l ) , relatively little was known about protein tyrosine phosphatases until tyrosine phosphatase 1B from human placenta was purified and subsequently cloned (3,4). To date, more than 26 PTPs have been identified ( 5 , 6 ) , and the known enzymes can be divided into two groups; a cytosolic form and a receptor-type form with a single transmembrane domain (RPTP). Most RPTPs contain two conserved catalytic tyrosine phosphatase (PTP) domains and each encompasses a segment of 240 amino acid residues ( 5 , 6). The catalytic domain contains a highly conserved motif of 11-amino acid residues, (I/V)HCXAGXXR(S/T)G, in which the cysteine residue was shown to be essential for protein tyrosine phosphatase activity (7)(8)(9). The RPTPs can be subclassified into five types based upon the amino acid sequence diversity of their extracellular segments (2, 5 , 6,10,11). The presence of both Ig-like and fibronectin type I11 (FNII1)-like repeats in the extracellular domain is characteristic of type I1 receptor tyrosine phosphatases, which includes LAR (12)(13)(14), PTPp (15), PTPK (16), and Drosophila DPTP (14). These structural features resemble the N-CAM family of cell adhesion molecules (17)(18)(19)(20), suggesting type I1 receptor tyrosine phosphatases may function as cell adhesion molecules.
Tyrosine phosphorylation initiated by several receptor tyrosine kinases has been implicated in neuronal cell proliferation and differentiation (21). These include the receptors for neurotrophic factors belonging to the tropomyosin-receptor kinase (trk) family of receptor tyrosine kinases, as well as platelet-derived growth factor and fibroblast growth factor receptor tyrosine kinases (22). Virtually nothing is known about the control of tyrosine phosphorylation by protein tyrosine phosphatases in the mammalian nervous system. The present report describes the cloning and characterization of a type I1 receptor tyrosine phosphatase, RPTP-a, which is highly expressed both in the developing and mature central and peripheral nervous systems of the rat. RPTP-a is a receptor protein having a single transmembrane, an extracellular segment consisting of three Ig-like and five FNIII-like repeats, and a cytoplasmic segment having two tandem catalytic domains. Transfection of RPTP-a cDNA into human embryonic kidney 293 cells results in the synthesis of a protein with an apparent molecular mass of 200 kDa. RPTP-u is a new member of the type I1 receptor tyrosine phosphatases (2, 5 ) that is most homologous to rat LAR (13), followed by human PTP-6 (23) and Drosophila LAR (14). The RPTP-a gene had been mapped to distal chromosome 17 of the mouse.

EXPERIMENTAL PROCEDURES
Materials-Chemical reagents were purchased from Sigma. The Xgtll rat brain stem cDNA library was obtained from Clontech.  stretches DYINAS and VCHSAG of known PTP domains, was used in the RNA-PCR to specifically amplify novel sequences of PTP domain from 1 pg of PC12 total RNA (following procedures recommended by Perkin-Elmer Cetus). PCR products of expected size (about 450 bp) were isolated by agarose gel electrophoresis, purified, and then cloned into a pBluescript vector (Stratagene). The cDNA inserts were sequenced by the dideoxy chain termination method using the Sequenase Version 2.0 kit. Sequencing of about 100 individual clones led to the identification of three PTP domains whose predicted sequences were homologous to, but distinct from, corresponding sequences of all known PTPs.
Library Screening and Isolation of cDNA Clones-The cDNA insert from one of the above PTP domains, referred to as the a PTP domain, was amplified and labeled using a random priming labeling kit. The labeled cDNA insert was then used as a probe to screen a rat brain stem cDNA library (24). About 600,000 recombinants were screened, from which 17 positive clones were purified after the first round of screening. The most 5' end of the longest cDNA insert isolated following the first round of screening was used in successive rounds of screening until overlapping cDNA clones that contained the entire coding sequence for RPTP-a were obtained. Both strands of several overlapping cDNA clones were sequenced using a Pharmacia DNA sequenator. Generation of Antibodies against Glutathidne S-Transferase Fusion Protein-A cDNA fragment coding for amino acids 512-835 of RPTP-a (corresponding to the FNIII region of the extracellular domain) was amplified by PCR and subcloned into a pGEX bacterial expression vector (Pharmacia LKB Biotechnology Inc.) to construct plasmid pGEX-FNIII. This plasmid was transformed into Escherichia coli strain DH5a. After isopropyl-1-thio-8-D-galactopyranoside induction, the insoluble glutathione S-transferase fusion protein was isolated by SDS-polyacrylamide gel electrophoresis. Two rabbits were immunized with the fusion protein to produce polyclonal antisera (FNIII Ab) against RPTP-a.
RNA Isolation and Northern Blot Analysis-Total RNA was isolated from different tissues of 6-day-old rats using RNA isolation kits obtained from Stratagene. Samples containing 30 pg of total RNA were resolved in a formaldehyde/agarose-gel(24) and then transferred to Nytran membrane (Schleicher & Schuell). Probes corresponding to different cDNA regions of RPTP-u were amplified by PCR, purified, and labeled using a random priming kit. Hybridization and subsequent washing were carried out essentially as described elsewhere (25) with an additional wash carried out in 0.1 X SSC and 0.1% SDS at 60 "C for 15 min.
In Situ Hybridyzation-Probe-1, 5"TAGACCACAATGGAACC-ATCGTTGTCAGGCTTTGGGGCGACACTAGGCTT-3', was synthesized in an Applied Biosystems 380A DNA synthesizer and then purified. Probe-1 is antisense and complementary to the cDNA of RPTP-a from nucleotide 2918-2967; a region that encodes amino acid stretch 694KPSVAPKPDNDGSIVVY710. This probe was chosen because it is least homologous to the corresponding sequences in rat LAR and human PTP-6. Probe-1 was labeled at 3'-end with dATP (using a terminal deoxynucleotidyl transferase kit) and then purified by Sephadex G-25 column chromatography. The specific DNA.
activities of the labeled probe ranged from 1 to 4 X IO8 cpmlpg of Adult male and timed-pregnant female Spragne-Dawley rats and their litters were used for the study. The day of successful copulation was determined by the presence of sperm in vaginal smears and was recorded as embryonic day 0 (EO). The day of birth was considered postnatal day 0 (PO). Whole bodies of E12 embryos were fixed in 4% paraformaldehyde in 0.1 M sodium phosphate, pH 7.4, for 4 h followed by overnight infiltration with 15% sucrose in 0.1 M sodium phosphate, pH 7.4. Adult rats were sacrificed by decapitation, and then the brains were rapidly removed and frozen on dry ice. 20-pm sections obtained with a cryostat microtome were postfixed for 30 min and then washed three times in 0.1 M sodium phosphate, pH 7.4. The sections were then dehydrated and stored at -20 "C.
Prehybridization and hybridization were carried out as described elsewhere (11,26). Sections were mounted on the slides and incubated with 1 x IO6 cpm of labeled probe-1 in 10 mM dithiothreitol. The specificity of hybridization was determined by incubating an adjacent section with labeled probe-1 in the presence of a 70-fold excess of unlabeled probe-1. The slides were washed twice in 2 X SSC at room temperature (30 min), 1 X SSC at 50 "C (30 min), 0.5 X SSC at 50 "c (30 min), and finally in 0.5 X SSC at room temperature (10 rnin). Sections were dehydrated and exposed to X-Omat film for 10 days.
Transfection, Immunoprecipitation, and Immunoblot Analyses-A cDNA insert containing the entire open reading frame of RPTP-u was subcloned into an eukaryotic expression vector to generate RKSigma and RKSigmaR, in which the cDNA insert was oriented in the sense or antisense direction, respectively. Human embryonic kidney 293 cells (27), grown to 20% confluence on fibronectin coated dishes, were transiently transfected with appropriate plasmids by the calcium phosphate precipitation method (25). The transfected cells were harvested after 48 h and lysed in lysis buffer (137 mM NaC1, 10% glycerol, 1% Triton X-100, 0.5% deoxycholate, 0.1% SDS, 2 mM EDTA, 20 mM Tris, pH 7.3) containing proteinase inhibitors (aprotinin and leupeptin (1 pglml) and phenylmethylsulfonyl fluoride (100 pg/ml)). After centrifugation (15,000 X g, 15 min at 4 'C), the supernatants were analyzed by either first immunoprecipitated with FNIII antibody followed by immunoblot analysis or directly by immunoblot analysis. The proteins resolved by SDS-polyacrylamide gel electrophoresis were transferred onto Nytran membrane, immunoprobed with FNIII antibody, and visualized with 1251-protein A after exposure to x-ray film.
Analysis of Tyrosine Phosphatase Actiuity-The synthetic peptide poly-Glu-Tyr was phosphorylated by treatment with purified epidermal growth factor receptor kinase domain in the presence of [y-"P] ATP and Mn2+. The lysates from either untransfected or transfected human 293 cells were immunoprecipitated with anti-RPTP-a FNIII antibody (see above). The immunocomplex was suspended in 50 p1 of buffer P (25 mM Hepes, pH 7.0,5 mM EDTA, 10 mM P-mercaptoethanol) and incubated with phosphorylated substrate at 37 "C for 30 min with shaking (28). Reactions were terminated by addition of an acidic charcoal mixture. PTP activity was monitored by following the appearance of free 32P in the reaction supernatants.
Mapping the Mouse RPTP-u Gene (Table I)-Inbred and recombinant inbred (RI) mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Genomic DNA was isolated from liver, digested with restriction enzyme TaqI, and then analyzed by Southern blotting with probe pGEX-FNIII (see above) as described previously (29). To establish genetic linkages, the distribution patterns observed for RPTP-a in the RI strains were compared to approximately 1360 other markers maintained in a database at New York University? Significance of matches was assessed using the Bayloc algorithm (30).

Molecular
Cloning of the Full-length cDNA for RPTP-a-Three cDNA fragments coding for novel PTP domains were initially identified in the total RNA isolated from PC12 cells by reverse PCR with the primers for the conserved regions of other known PTPs. One of these PTP domains, termed the u PTP domain, was expressed at high levels in the brain and only to a limited extent in the lung and intestine. Based on these observations, a random and oligo(dT)-primed rat brain stem cDNA library (Clontech Inc.) was screened for the full-length cDNA of RPTP-a using the PCR fragment initially cloned from the RNA of PC12 cells. After three rounds of screening, several overlapping clones were isolated. They were sequenced in both directions. As shown in Fig. 1, the 5.7-kb full-length cDNA for RPTP-a was constructed by joining the two longest overlapping clones (H-29 and B-10) at a unique MluI site. The 5.7-kb cDNA has an open reading frame of 1501 amino acid residues starting from the ATG at P. D'Eustachio, unpublished results. bp 839. The ATG was preceded by stop codons in all three reading frames and meets the requirements for the translation initiation site (31). There is an 838-bp 5"untranslated sequence and a 353-bp 3"untranslated sequence containing a poly(A) tail.
Characterization of RPTP-a-The deduced amino acid sequence for RPTP-a is shown in Fig. 2A. It contains 23 hydrophobic amino acids likely to function as a signal peptide (32), followed by an 824-amino acid extracellular domain and a 24-hydrophobic amino acids transmembrane domain. The 627-amino acid cytoplasmic region contains two tandem PTP domains. The overall sequence of RPTP-a demonstrates a high degree of similarity to rat LAR (13), having 71% sequence identity within the entire coding region. RPTPa is also 75% identical to partially sequenced human PTP-6 (23).
The extracellular segment of RPTP-a consists of three Iglike domains (residue 33-315) (33) and 5 FNIII-like repeats (residue 320-786) (34,35). Search in database revealed significant homology to the extracellular segments of rat LAR (63% identity). Also, the homology between the Ig-like domains of RPTP-a and LAR (75% identity) is greater than that of the FNIII-like repeats (54% identity). There was significant homology also between the FNIII-like repeats of RPTP-a and human PTP-8. In addition, the extracellular sequence of RPTP-a is homologous to neural cell adhesion molecules in terms of their overall structure and sequence. The highest homology was detected with L1 (26% identity) (36), followed by N-CAM (22% identity) (18) and neuroglian (22%) (37). Fig. 2B shows alignment of three Ig domains of RPTP-a to each other and to the first Ig domain of rat LAR and mouse N-CAM, respectively. Within each Ig domain, two cysteines are thought to be involved in intradomain disulfide binding and are conserved (Fig. 2B, asterisks). The alignment of the five FNIII-like repeats of RPTP-a to each other and to the first FNIII-like repeat of rat LAR and type I11 repeat (domain 7) of human fibronectin is shown in Fig. 2C. The fifth repeat of RPTP-a exhibits greater divergence from consensus than the other four. Also, the extracellular region of RPTP-a does not contain the RGD sequence that is required for fibronectin binding (38).
The cytoplasmic segment of RPTP-a, like most receptor tyrosine phosphatases, contains two conserved PTP domains in tandem.
PTP domain I shows 47% sequence identity with PTP domain I1 (Fig. 2 A ) . The cysteine residues, which have been implicated to be essential for phosphatase activity, are conserved in both PTP domains. The overall cytoplasmic segment of RPTP-a has a striking sequence homology to that of rat LAR (84% identity) and human PTP-6 (86% identity). Furthermore, comparison of the cytoplasmic segments of RPTP-a and rat LAR indicates that a higher degree of identity is present between domain I1 (93%) than domain I (84%). A schematic diagram showing the multiple domain structure of RPTP-a is presented in Fig.  20.
Although RPTP-a is highly homologous to rat LAR, especially in the cytoplasmic region, the diversity of their extracellular domains, particularly in the FNIII-type repeats, demonstrates that RPTP-a is a new member of the type I1 receptor tyrosine phosphatases.
Distribution of RPTP-a mRNA-Northern blot analyses of total RNA isolated from various 6-day-old rat tissues are presented in Fig. 3A. The blot depicted in this figure was probed with a cDNA fragment coding for amino acid residues 408-521 of RPTP-a. Identical results were obtained with a cDNA probe specific for a cytoplasmic segment of RPTP-a and with oligonucleotide probe-1 (data not shown). RPTP-a is highly expressed in the brain as compared to other tissues. Of the two major RPTP-a transcripts (5.7 and 6.9 kb) expression of the 5.7-kb species is detected exclusively in brain while the 6.9-kb transcript, presented also in the brain, is expressed to a significantly lower level in lung and heart. Upon longer exposure, the 6.9-kb species can also be detected in kidney and intestine. The cDNA that was sequenced and described in this paper probably corresponded to the 5.7-kb mRNA. 3 The nature of these various mRNA species is not clear. Fig.  3B shows a short exposure of same Northern blot indicating RPTP-a is highly expressed in the brain. Fig. 3C is the gel stained with ethidium bromide revealing the integrity of RNA.
In Situ Hybridization-Since RPTP-a is highly expressed in the brain as detected by Northern blot analysis, its expression at specific sites within whole embryos and adult rat brain was then identified by in situ hybridization. Using labeled probe-1, an antisense oligomer, the expression of RPTP-a was detected in brain, spinal cord, and dorsal root ganglia as early as embryonic day E12 (Fig. 4A). No

0.
S l g r n a l l !   30 of total RNA prepared from 6-day-old rat tissues, as indicated, were used for Northern blot analysis. The blot was hybridized with a cDNA probe encoding amino acid residue 408-521 of the extracellular domain of RPTP-a. The sizes of RNA markers (Life Technologies, Inc.) are in kilobases (kb). A , gel exposed for 18 h; B, for 3 h. C, RNA gel stained with ethidium bromide.

FIG. 4. I n situ hybridization analysis of RPTP-a expression in rat.
A , sagittal section of whole rat at embryonic day 12 (E12) was hybridized with labeled probe-1 (see "Experimental Procedures"). cx, cortical neuroepithelium; 4V, fourth ventricle; LV, lateral ventricle; SC, spinal cord; DRG, dorsal root ganglia. C, sagittal section of adult rat brain was hybridized with labeled probe-1. Cb, cerebellum; Py, pyramidal cell layer; GrDG, granular layer of dentate gyrus; OB, olfactory bulb. B and D, sections adjacent to A and C, respectively, were hybridized with labeled probe-1 in the presence of a 70-fold excess of unlabeled oligomer to demonstrate the specificity of in situ hybridization. specific since they were completely chased by a 70-fold excess of unlabeled probe-1 (Fig. 4B). In addition, a relatively low level of the expression of RPTP-a was detected in the lung at embryonic day 18 (EM), and the intensity of RPTP-a expression overall gradually decreased during embryonic development (data not shown). In the adult rat brain, expression of RPTP-u was confined to specific regions of the brain that included the olfactory bulb, cerebellum and hippocampus (Fig.  4C). Within the latter structure, the pyramidal cell layer and granular layer of dentate gyrus were labeled specifically (Fig.   4, C and D).
Transient Expression of RPTP-a-Human embryonic kidney 293 cells (27), commonly used for overexpressing exogenous proteins, were transfected transiently with a mammalian expression vector that directs the synthesis of RPTP-a in order to investigate its biochemical properties. The transfected cells were lysed in lysis buffer, and the supernatants were then subjected to immunoprecipitation with polyclonal FNIII antibody. This was followed by immunoblot analysis using the same antibodies. The results shown in Fig. 5A revealed that FNIII antibody can specifically detect a protein of 200 kDa in supernatants from cells transfected with a plasmid containing RPTP-a cDNA (RKSigma) (lane 1 ) but not from those transfected with a plasmid containing control DNA (RKSigmaR) (lane 2 ) . No band corresponding to the 200-kDa protein was detected in control immunoprecipitation experiments using preimmune serum (data not shown). The apparent molecular mass of the expressed protein after SDSpolyacrylamide gel electrophoresis is in good agreement with the expected size for RPTP-a. The bands detected in lane 2 of Fig. 5, A and B, were nonspecific since they were also observed after incubation with preimmune serum.
Several members of the type I1 receptor tyrosine phosphatases, such as LAR (13, 40) and RPTP-K (16), undergo proteolytic processing after synthesis. T o investigate whether similar processing occurs with RPTP-a, a new member of this group, the supernatants of lysates from the transfected cells were subjected directly to immunoblot analysis using FNIII antibody or preimmune serum. The results presented in Fig.  5B show that in addition to the 200-kDa protein previously identified, an additional protein with an apparent molecular mass of 100 kDa was detected in lysates of cells transfected with plasmid RKSigma (lane 1 ), but not with RKSigmaR in immunoblot analyses of supernatants from lysates of transfected COS-1 cells (data not shown). These results suggest that RPTP-o also undergoes proteolytic processing and the 100-kDa protein may represent a cleavage product.
PTP Actiuity of RPTP-o-To demonstrate that RPTPo has the intrinsic tyrosine phosphatase activity, lysates from human 293 cells transfected with RKSigma or RKSigmaR, respectively, were incubated with anti RPTP-a antibodies. The immunocomplexes were incubated with radiolabeled peptide. The tyrosine phosphatase activity of RPTP-o was assayed by measuring the amount of free 32P released. As shown in Fig. 6, approximately 7-fold increase in tyrosine phosphatase activity was observed in the immunocomplex from cells transfected with RKSigma as compared to those with RKSigmaR.
Mapping the Mouse RPTP-a Gene-When TaqI-digested genomic DNA from various inbred strains was analyzed by Southern blotting using an RPTP-u probe (pGEX-FNIII), two allelic forms of the gene were detected (Table IA). The pattern of inheritance of these alleles in recombinant inbred strains of mice defined a single genetic locus (Table IB). Comparison of the inheritance pattern for this locus with these previously established for approximately 1360 markers distributed over the entire mouse genome allowed us to map the locus for RPTP-o to distal mouse chromosome 17.

DISCUSSION
We report, presently, the molecular cloning, chromosomal localization, and expression profile of RPTP-o. A data base search revealed that RPTP-a is structurally similar to rat LAR (13) with an overall sequence identity of 71%. Similarity among the cytoplasmic domains (84%) is greater than that of the extracellular domains (63%). RPTP-a was mapped to distal chromosome 17 of the mouse, which corresponds in part to human chromosomes 6 and 19. Hence, we conclude that RPTP-o is a novel member of the type I1 receptor tyrosine phosphatases (2). RPTP-o is highly expressed in the brain as two major transcripts (5.7 and 6.9 kb). Northern blot analysis and in situ hybridyzation demonstrated that the 5.7-kb transcript is expressed exclusively in the nervous system. The 6.9-kb transcript can be detected in lung, heart, kidney, and intestine, but at significantly lower levels than that found in brain. The compared to ones for 436 other markers. All concordances whose odds of chance occurrence are less than 0.05 are shown. For each such match, the marker's name and chromosomal location are shown, together with the observed recombination fraction (R/N), odds of observing that fraction or a smaller one by chance (30), and, conditional on the existence of linkage, the estimated distance in cM between the two loci (53), and the 95% binomial confidence limits for that estimate (39). This suggests that RPTP-a plays a broad role in development and maintenance of the nervous system. The major difference between RPTP-o and LAR, which share a high degree of sequence homology, is their expression pattern in various tissues. LAR has a broad tissue distribution, being detected in a number of epithelia as well as in smooth and cardiac muscle (40). In contrast, the 5.7-kb transcript of RPTP-U, the most abundant species, is expressed exclusively in nervous tissues. Also, DLAR (14), the proposed counterpart of LAR in Drosophila, was found to be expressed predominately in the axons in the central nervous system of Drosophila embryos (41).
Unlike rat LAR which has 6 predicted N-glycosylation sites, no such sites have been found in the extracellular domain of RPTP-o. However, this domain is rich in serine and threonine residues (16%) and likely to provide multiple attachment sites for 0-linked glycosylation. Moreover, RPTP-o has only five FNIII repeats while LAR contains eight. It is possible that larger alternatively spliced transcript (6.9 kb) encodes a protein containing eight FNIII-like repeats?
Like other type I1 receptor tyrosine phosphatases, such as PTP-p, RPTP-K, Drosophila LAR, and Drosophila PTP (14), the extracellular segment of RPTP-a consists of both Ig-and FNIII-like repeats. This structural feature also has been identified in neural cell adhesion molecules such as N-CAM, neuron-glia (Ng-CAM), and Ng-CAM-related (Nr-CAM) (19,20). It has been demonstrated that the extracellular domain, especially the Ig-like repeats of these neural adhesion mole-cules, is crucial for mediating homophilic and heterophilic binding when expressed in transfected cells (42, 43). Since there is significant homology among the extracellular domains of RPTP-a and L1, N-CAM and neuroglial (26,22, and 22% identity, respectively), it is possible that the extracellular domain of RPTP-u also mediate such binding which, in turn, couples the tyrosine phosphatase activity to some intracellular signaling pathway.
Immunoblot analyses suggest that RPTP-a, like LAR and RPTP-K, undergoes proteolytic processing. Sequence analysis indicates that RPTP-a has several potential cleavage motifs, such as RK at residues 731-732, RR at 762-763, and RHSR a t residues 766-769; all located within the fifth FNIII repeat of the extracellular domain. These sites may serve as specific cleavage signals for proteolytic processing (44).
Numerous studies have indicated that CD45 plays a crucial role in coupling the T-cell antigen receptor to a intracellular signalingpathway (45-47). Experiments have suggested CD45 can activate Ick or fYn protein tyrosine kinases by dephosphorylation at inhibiting sites at the carboxyl terminus (48, 49). In the mammalian nervous system several tyrosine kinases such as c-yes (50), and the neuron-specific form of c-src (51) have been shown to be expressed at high levels in the brain. Moreover, protein tyrosine kinase inhibitors have been shown to block long term potentiation in the hippocampus (52). Furthermore, it has been demonstrated that several receptor tyrosine kinases, such as nerve growth factor and fibroblast growth factor receptors (22), play a central role in neural development. The molecular cloning of RPTP-u, which is highly expressed in both the central and peripheral nervous systems of embryo and adults, will enable us to design experiments for determining the role of tyrosine phosphatases in mammalian neurogenesis and maintenance.