CA2086377A1 - Inducible lymphoid specific protein tyrosine phosphatase - Google Patents

Abstract

Abstract of the Disclosure Tyrosine phosphorylation and dephosphorylating event have been shown to be central to the process of growth regulation and signal transduction. The invention is the identification of a new gene with a tyrosine phosphatase domain (EC 3.1.3.48) which is expressed exclusively in thymus and spleen. A cDNA of 2760 bp encodes a 339 amino acid, intracellular, single domain tyrosine phosphatase. When expressed as a glutathionine-S-transferase fusion protein, efficient lysis of p-nitrophenyl phosphate is noted, indicating in vitro enzymatic activity of the cloned gene product. Normal mouse lymphocytes increase mRNA expression 10-15 fold upon stimulation with phytohemaglutinin, concanavalian A, lipopolysaccharide or anti-CD3 monoclonal antibody. This new hematopoietic tyrosine phosphatase, (HePTPase), may play a role in the regulation of T and B lymphocyte developement and signal transduction.

Description

The inv ntion is the isolation and identification of a new mammalian gene with a tyrosine phosphatase domain.

The importance of tyrosine phosphorylation in cellular growth and differentiation is becoming understood. Increased phosphorylation on tyrosine of regulatory substrates can be mediated through surface receptors and lead to increased cell division and when unregulated, to transformatio~. It is becoming clear that this regulation of proliferation may be a balance between the activities of tyrosine kinases and a growing class of molecules having tyrosine phosphatase activity.
While many genes for tyrosine kinases (PTK) have bsen identified, few with tyrosine phosphatase activity (PTPase) have been isolated, and those only recently.
Presently, the exact relationship between PTK and PTPase activity is unknown. The recent description of molecules with protein tyrosine phosphatase activity and substrate preferences suggests that regulatory tyrosine phosphorylation is dependent on the interaction of both PTK and PTPases. It is possible that some members of the PTPases might nonspecifically maintain basal tyrosine dephosphorylation. Under this paradigm, PTK molecules would provide the only regulatory activities determining the level of tyrosine phosphorylation. Conversely, if certain PTPases possessed substrate specificity, they could allow an additional layer of control over critical phosphorylation events.

In addition to growth regulation, tyrosine dephosphorylation is now regarded as an important mechanism that regulates signal transduction. One such experimental system in which balanced tyrosine phosphorylation and specific dephosphorylation may be crucial is in the regulation of T lymphocyte activation upon antigen recognition by the T cell receptor (1).
Tyrosine phosphorylation of a defined set of cellular 7 ~

substrates appears to be the earliest sign of this activation, occurring within 5 seconds oi- receptor occupancy (2). At least two of the src family tyrosine kinases, ~y~ (33 and lck (4), are expressed in T lymphocytes and may mediate these early reactions, although specific substrates for each have not yet been clearly dafined (1). Both of these kinases are associated with the T cell receptor complex, lck through the T ly~phocyte accessory molecules CD4 and CD8 (5), and fyn possibly being directly associated with CD3 (6).
These tyrosine kinases, with the T cell receptor, have been suggested to directly or indirectly interact with a ~eceptor tyrosine phosphatase, CD45 (Leukocyte Common Antigen) which may be required for early phosphorylation events (7). This paradox might be explained if CD45 acts by increasing the activities of the src family kinases by dephosphorylation of the inactivating carboxy terminal phosphotyrosines of these molecules. Although e~perimental evidence e~cists for possible in vitro CD45 dephosphorylation of lck (8), this concept has not yet been extended to ~y~ or other lymphoid tyrosine kinases.
It is likely, therefore, that other leukocyte specific PTPases exist, possibly to function by regulating phosphorylation events associated with lymphocyte activation.

The invention provides a nucleic acid sequence of a mammalian gene encoding a tyrosine phosphorylation domain. The e~pression of thls gene in humans is seen in both resting T, and B lymphocytes, granulocytes, but not in non-hematopoietic tissues. Additionally, m~NA levels of this gene are increased 10-15 fold upon stimulation of normal mouse lymphocytes with PHA, LPS, Con A, or with a mAb raised to the CD3 component of the T lymphocyte antigenic receptor. This gene may have a regulatory role in lymphocyte tyrosine phosphorylation, in leukemic blast proliferation, or in both.

-3- ~g ~ &3 ~7 Figure 1 shows a complete sequence of 2760 bp of HePTP (SEQ ID NO:l~. The deduced peptide sequence (SEQ
ID NO:2~ is shown below the open reading frame base pair sequence (SEQ ID NO:3) using single letter amino acid ~;ymbo1s.

Figure 2 shows the peptide alignment of 7 tyrosine phosphatase domains (bold type). HePTP = Hematopoietic Protein Tyrosine Phosphatase. PTP lB = Protein Tyrosine Phosphatase lB. LAR DOM 1, LAR DOM 2 = 5' and 3' domains of LAR, respectively. CD45 DOM 1, CD45 DOM 2 = 5' and 3' domains of CD45, respectively.

Figure 3 is a Northern analysis using 20 ~ug of total RNA from Balb/c murine brain, thymus, liver, spleen, kidney, and ovary.

Fi~ure 4a is a Northern analysis using 20 ug of total RNA from normal murine spleen lymphocytes stimulated for the indicated times with PHA. Increased expression of HePTP mRNA is seen at 24 h of incubation but is maximal by 72 h.

Figure 4b is a Northern analysis using 20 ug of total RNA from normal murine spleen lymphocytes stimulated for the indicated times with LPS, Con A and various dilutions of anti-CD3 hybridoma SN for 24 h.
Increased expression of the HePTP mRNA is seen with each stimulant compared with incubation with medium alone.
The response to anti-CD3 antibody is shown to be concentration dependent implying a graded response.

Figure 5 is a plot of 500 ng and 100 ng of HePTPase/GST fusion protein-containing bacterial lysate were compared to 500 ng of GST-containing lysate for the _4_ 2~ 7 ability to lyse p-nitropAenyl phosphate. Values are expressed a~ a percentage of the maximally observed amount of lysis.

MATERIALS ~ND METHODS

PCR gsne fragm~nt am~lification Template DNA waæ prepared by isolation from a cDNA
library constructed with RNA from pokeweed mitogen (PWM) stimulated human peripheral T lymphocytes (9). Anchored PCR (10) was performed using a tyrosine phosphatase consensus sequence specific degenerate oligonucleotide suite (GA)A(CT)TA(CT)AT(CTA)AA(CT)GC, with an ECoR1 recognition site at the 5' end. The second oligonucleotide corresponded to the AgtlO forward sequencing primer, CTTTTGAGCAAGTTCAGCCTGGTTAAG (Promega, Madison WI). A 50 ~ul reaction was performed using 8 ng/~l of each primer, 0.08 mM of each dNTP, 5~1 of PCR
reaction buffer (Perkin Elmer Cetus, Norwalk CT), 1 ul oE
cDNA library template DNA, and 1 ~1 of recombinant Taq polymerase (50u/~ul, Perkin Elmer Cetus). The mixture was denatured at 94 for 4 minutes, followed by 40 cycles with 1 minute annealing at 55, 4 minute extension at 72, and 1 minute denaturation at ~4. A 0.4 KB product was then digested with ECoRI restriction enzyme, and cloned into pBluescript KS plasmid (Stratagene, La Jolla CA) using T4 DNA ligase (BRL, Bethesda ML). Several clones were examined by DNA sequencing and compared to published PTPase sequences for similarity. One such segment was identified. This 0.4-kb gene fragment was radiolabeled using [y-P~2]dCTP and the multiprime DNA labeling system (Amersham Int., Amersham, GB). This product was used to probe the amplified cDNA library that was used as template DNA in the initial amplification step. Forty positive phage clones were isolated. One such clone which contained an insert of 2760 bp was inserted into pBluescript KS plasmid and further analyzed.

DNA 5eguenca Analyels Closed circular plasmid DNA was purified using the Qiagen system tQiagen, Studio City CA). 1 ug was denatured with all~ali, and the single strand products were hybridized with synthetic deoxynucleotide primers.
The sequences were determined by the chain termination method of Sanger (11). Eighteen sequencing primers were constructed allowing overlapping sequence determination of the uninterrupted 2760-bp clone independently in both directions.

Northern and 80uthern Analysls Total RNA was prepared according to the method of Chomczynski (12). Northern and Southern gel analysis was performed as described previously (13).

Isolation and stlmulation of murlne lymphocytes The spleens from ten adult BALB/c mice were gently di~rupted through a fine wire mesh and incubated in Dulbecco's modified Eagle Medium (Gibco, Grand Island, NY) containing 10% FCS (HyClone, Logan, UT) at a cell density of 1 x 6/ml, LPS (25 ,ug/ml; Difco, Detroit, MI), Con A 5 ~ug/ml; Difco), PHA (90 ~g/ml; Wellcome Diagnostics, Dartford, GB) or anti-CD3 hamster hybridoma SN at a dilution of 1:10, 1:100, and 1:1000 (a generous gift of Dr. J. Bleustone, University of Chicago, USA) were added to individual samples. At 4 and 24 h 1 x 107 cells were harvested for RNA extraction as described.

Prokaryotic expres~ion o~ ~ePTP

The complete coding region of HePTP was amplified -6- 2~3~7 using the PCR. Oligonucleotides TTTGGGGGCAGCCATGACCCAG, corresponding to position 143 to 164, found 13 hp up~tream of the initiation codon, and GCAGGGGTCAGGGGCTGG, corresponding as antisense to position 1182 to 1164, found 8 bp beyond the termination codon, were utilized.
Both were synthesiæed with 5' Bam Hl recognition sequences. The 1039-bp digested product was inserted into the prokaryotic expression vector pGEX 3X
~Pharmacia, Piscata~ay, NY) digested with Bam HI, and expressed as part of a 68-kDA GST fusion protein in E.coli strain DH5a (BRL).

In vltro phos~hata~e acti~ity Cell pellets from 1.5 ml of induced cultures of bacteria containing pGEX 3X specified GST protein, or containing the HePTPase/GST fusion protein were incubated in 200 ~l of a lysis buffer containing 25 mM Tris-HC1 (pH
7.4). 200 mM NaC1, 1~ hen egg lysozyme (Pharmacia), and 1~ NP40 (BDH Chemicals, Poole, GB) for 10 min at 23C.
Gentle sonication was performed. Protein contents of -the two SN were determined with the Bio-Rad assay (Bio Rad Laboratories, Richmond, CA) and were each adjusted to 100 ng/~l with lysis buffer. A further 1:5 dilution of an aliquot of the HePTPase/GST-containing lysate was performed. Five microliters of each of the three samples was incubated in a reaction mixture of 5 mM of p-nitrophenyl phosphate (Sigma, St. Louis, MO), 80 mM

2-[N-morpholino]ethane sulfonic acid (pH 5.5). 10 mM
EDT~, and 10 mM DTT. After 5, 10, and 15 min of incubation at 37C the reactions were stopped by the addition of 1 ml of 0.2 N NaOH. The absorbance was measured at 410 nm and expressed for each value as a percentage of that maximally observed. All experiments were performed independently and in triplicate.

`' 2~ 77 RESULTS

Isolation of and analysi~ o~ a human ~DNA showing protein tyros~ne phosphata~e ~a~le~c~ sim11arity:

Using anchored PCR and an oligonucleotide to the PTPase consensus sequence (see Materials and Methods), a new member of the protein tyrosine phosphatase family has been identified. The nucleotide sequence of this 2760 bp cDNA revealed an initiation AUG codon at position 157 with an inframe upstream stop codon. The open reading frame terminate~ at position 1173 followed by a TG~
termination codon. This transcribes a polypeptide of 339 amino acids. A 1587 bp 3' untranslated tail terminates with an AAUAAA polyadenylation signal at position 2737 -2742 and a poly A tail beyond position 2760 (Fig. 1).
hydroaffinity analysis of the deduced polypeptide sequence shows no obvious hydrophobic transmembrane region (data not shown). The sequence would seem to code for a single domain cytoplasmic phosphatase that is similar in structure to T Cell PTP, and PTPlB. A
distinct open reading frame spanning the entire single tyrosine phosphatase domain indicates a functional protein, and characteristic amino acid motifs belonging to the family of tyrosine phosphatases are seen. The expected DYINA motif used in construction of the specific oligonucleotide, which was used to amplify this sequence was observed. The highly conserved sequence VHCXAGXXR, which is thought to form part of the dephosphorylation catalytic site, is seen in this gene as in all PTPases thus far identified (Fig. 2). Further comparison of the peptide sequences of selected phosphatase domains shows similar degrees of similarity between HePTP and the catalytic regions of CD45 (39~), LAR (38%), T Cell PTP
(38%), and PTP lB (36%) with an eight amino acid deletion seen in HePTP. For CD45 and LAR, the catalytic domains are considered to be the 3' members of their tandem ~' 7~

repeats (labeled domain 1 in Fig. 2).

Th~ cDNA saquence i8 r~stricted to lymphoid Qlement8:

A m~use counterpa~t to this gene (unpublished data) was used to determine the tissue distribution of HePTP
mRNA. A northern blot analysis was performed using total RNA isolated from the brain, liver, spleen, kidney, thymus and ovary of a Blab/C mouse and the spleen of a Balb/C nu/nu mouse. Expression was seen in the thymus and spleen samples (Fig. 3) and in nude spleen samples (data not shown). This suggests relatively more expression in T lymphocyte~ than in B lymphocytes. No expression was detected in brain, liver, kidney, and ovary. A Southern analysis of human genomic DNA digested with EcoR 1 showed hybridizing bands with molecular 15 weights 15.6 kb, 2.6 and 2.8 kb (data not shown), implying on the basis of number and size distribution that a single genomic copy of HePTP is present.

Ex~reaeion is increa8ed a~ter PHA, LPS, Con A or antl-CD3 antibody ~timulation To investigate the inducibility of HePTPase gene expression in normal murine lymphocytes after exposure to lectin and stimulation with anti-CD3 antibody, a timed incubation with PHA, Con A, LPS and anti-CD3 hybridoma SN
was performed. Normal spleen lymphocytes were incubated independently with each stimulant as described above, and harvested at various time intervals up to 72 h. A 10-15 old increase in mRNA level of murine HePTP was seen after each exposure (Fig. 4a and b). Stimulation through the antigenic receptor with anti-CD3 antibody produced a pattern of inducibility similar to that seen with each lectin. Maximal increase was seen at 24-72 h of incubation in each case.

:

9 2G~377 ~pre~ion and in ~tro pho~pha~as~ act~ity HePTPase was expressed as a 68~kDa fusion protein incorporating the complete 339-amino acid protein sequence as a GST fusion protein. Equal concentrations of protein from bacterial lysates containing the HePTP/GST fusicn protein or from lysates containing the expressed GST protein alone, were compared in their ability to lyse the phosphotyrosine analogue, p-nitrophenyl phosphate (Fig. 5). HePTPase/GST-containing lysate showed marked activity under the assayconditions compared with the control lysate. A 1:5 dilution of the HePTPase/GST lysate displayed approximately 1/5 of the enzymatic activity.

Reversible tyrosine phosphorylation is known to play critical roles in diverse series of processes involved in the regulation of cell division, differentiation, and signal transduction. The degree of tyrosine phosphorylation ~s thought to be a balance between tyrosine kinases and the recently recognized class of tyrosine phosphatases. An emerging concept is that both tyrosine kinases and phosphatases share specific substrates and balance each other's actlvity. This simple relationship is complicated, however, by the possibility that certain tyrosine phosphata~es can increase the activities of some tyrosine kinases by dephosphorylating their carboxy terminal tyrosines, and thus enhance their functions (14).

Few physiologic roles of any tyrosine phosphatase have been assigned, although the broad tissue distribution of most would imply a general role in growth and signal transduction. However, some such as CD45 and HePTP are restricted in terms of expression and would -lo- 203~3~7 su~est ~ore specialized roles. For example, CD45 (Leukocyte Co~non Antigen), which is expressed primarily in lymphocytes, has been shown to be indispensable for T
cell activation through the antigen receptor (7~.
Tyrosine kinase activity also is regarded to be essential for this process since its inhibition by chemical agents pre~ents T cell activation (15). Thus, it has been postulated that T lymphocyte src family kinases are activated by CD45 through dephosphorylation of the carboxy terminal tyrosine residues (8), since phosphorylation at these sites is known to decrease kina~e activity. This model of T cell activation requires that CD45 recognize the kinases themselves rather than their phosphorylated products. Such a model of T lymphocyte stimulation requires that this tyrosine phosphatase recognizes, by virtue of substrate specificity or location, a small number of targets such as the kinases themselves, ignoring the products of tyrosine kinase phosphorylation. The growing number of identified tyrosine phosphatase enzymes, some with characteristic tissue distribution such as HePTP, would also imply that specific biochemical roles and substrates will be identified for individual enzymes.

HePTP appears to be a single domain PTPase with in vitro activity demonstrated to the phosphatase subYtrate, p-nitrophenyl phosphate. Peptide similarity to the other cytoplasmic single domain molecules, PTPlB, T Cell PTP, and each of the 2 catalytic domains of LAR and CD45 is noted (Fig. 2). Unlike what is observed in the other single domain PTPases, PTP lB and T Cell PTPase, strict specificity in the pattern of expression is seen and restriction to lymphoid elements implies a discrete role in the biology of these tissues.

Stimulation of T lymphocytes with nonspecific lectins or through the TcR complex with anti-CD3 results $ 3 7 7 in increased HePTP mRNA expression noticeable at 24 h with maximal stimulation at 72h. This pattern of mRNA
inducibility after T lymphocyte antigenic activation is similar to that o~ IL 2R~, IL 2, IL6, and the tyrosine kinase lck, which become maximally induced by 24-48 h.
Induction by LPS of HePTPase mRNA in a mixed population of lymphocytes also implies expression of this gene is induced in B cells upon stimulation. In both lymphocyte populations, HePTPase might function during sustained stimulation, modulating the immune response through dephosphorylation of specific residues, though its specific role remains to be defined.

The human sequence HePTP has been submitted to genbank under the accession number M 64322.

-12- 2~377 R~F~RENC~5 1. Klausner, Ro D., Samelsun, L. E. (1991). T cell antigen receptor activation pathways: The tyrosine kînase connection. Cell 64:B75-87.

2. Hsi, ~. D., Siegel, J. N., ~inami, Y., Luon~, E. T., Xlausner, R~ D., Samelaon, L. E. (1989). T cell activation induces rapid tyrosine phosphorylation of a limited number of cellular substrates. J Biol Chem 264:10836-10842.

3. Cooke, M. P., Abraham, K. M., Forbu~h, K. A., Perlmutter, R. M. (1991). Regulation of T-Cell Receptor Signalin~ by a src family Protein-Tyrosine Kinase (p59fyn). Cell 65: 281-291.

4. Perlmutter, R. M. (1989). T Cell Signaling.
Science 245: 344-348.

5. Yelllette, A., Bookman, M. A., Horak, E. M., Samel~on, L. ~., Bolen, J. B. (1989). Signal transduction through the CD4 receptor involves the activation of the internal membrane tyrosine protein kinase p56 lck. Nature 338:257-259.

6. 8nmelson, L.E ., Phlllips, A. F., Luong, E. T., Xlaucner, R. D. (1990). Association of the protein tyrosine kinase with the T cell antigen receptor.
Proc.Natl.Acad.Sci.USA 87: 4358-4362.

7. Pin~el, J. T., Thomas, M. T. (1989). Evidence that the leukocyte common antigen is required for antigen induced T lymphocyte proliferation. Cell 58:1055-1065.

8. Nu8telin, T., Altman, A., (1990). Dephosphorylation and activation of the T cell tyrosine kinase pp56 lch by ' :

-13- 2~377 the leukocyte common antigen (CD45). Oncogene 5:809-814.

9. Ximuxa, N. (1986). Sequences and diversity of human T cell receptor ~ chain variable region genes. J. Exp.
Med. 739-750-10. Loh, E. Y.~ Elliott, ~. F., Cieirla~ S., ~a~ier, L.L., Da~is, M. (1989). PCR with single sided specificity:
analysis of T cell receptor ô chain. Science 243:217-220.

11. S~lger, F., Nicklen, S., Coulson, A. R. (1977). DNA
sequencing with chain termination inhibitor. Proc. Natl.
Acad. Sci. USA. 74:5463-6467.

12. Chomczyn~ki, P., Sacchi, N. (1987). Single step method o~ RNA isolation by acid guanidinium thiocyanate phenol chloroform extraction. Analytical Biochem.
162:156-159.

13. Yanagi, Y., Yoshlkal, Y., Leggett, K., Clark, S. P., Alekcander, I., Mak, T.W. (1984). A human T cell specific cDNA clone encodes a protein having extensive homology to immunoglobulin chains. Nature 308:145-148.

14. C~ntley, L. C., Auger, X. R., Carpenter, C., Duckworth, B., Grazianl, A., Kapeller, R., Soltoff, 8.
(1991). Oncogenes and signal transduction. Cell 164:281-302.

15. Mu~telln, T., Coggeshall, K. M., I~akov, N., Altman, A. (1990). T cell antigen receptor mediated activation of phospholipase c require~ tyrosine phosphorylation.
Science 247:1584-1587).

Claims (8)

1. An isolated nucleotide sequence encoding a mammalian inducible lymphoid-specific protein tyrosine phosphatase.

2. An isolated nucleotide sequence encoding the hematopoietic tyrosine phosphatase, HePTPase.

3. The nucleotide sequence, SEQ ID NO: 1.

4. The nucleotide sequence, SEQ ID NO: 3, of the cDNA
encoding HePTPase.

5. An isolated mammalian inducible lymphoid specific protein tyrosine phosphatase.

6. The amino acid sequence, SEQ ID NO:2, of HePTPase.

7. An isolated nucleotide sequence having at least 60%
homology with the sequence of claim 3 over a span of at least 50 nucleotide bases as measured by hyhridization therewith.

8. An isolated nucleotide sequence having sufficient homology with the sequence of claim 3 to enable it to function as a probe.