CA2326952A1 - T cell protein tyrosine phosphatase - Google Patents

Description

T CELL PROTEIN TYROSINE PHOSPHATASE
Field of the Invention This invention relates to T cell protein tyrosine phosphatase and more particularly to its role in cell signalling.
Background of the Invention Intracellular signaling mediated by the phosphorylation of tyrosyl residues of proteins is regulated by two opposing families of enzymes: protein tyrosine kinases and protein tyrosine phosphatases. The phosphorylation of tyrosine is a key process in the regulation of cell growth and metabolism either due to the intrinsic ability of phosphorylation to cause activation of protein function or by its ability to form a basis for protein scaffolding to occur.
Protein tyrosine phosphatases (PTPases) form a large superfamily of enzymes that counteract the action of tyrosine kinases in cellular systems by catalyzing the hydrolysis of phosphorylated tyrosyl residues. Classically, tyrosine kinases have been seen as the main players in signal transduction, while the phosphatases have been relegated to a housekeeping role.
Increasing evidence from the literature suggests that the phosphatases in fact play a critical role in maintaining the dynamic nature of many signaling pathways. This counterplay between tyrosine kinases and phosphatases allows extracellular signals to regulate cellular physiology.
PTPases are characterized by the presence of a 240 amino acid catalytic domain containing the signature motif (I/V)HCXAGXXR(S/T). The cysteine in this motif is essential for the catalysis of phosphate ester hydrolysis. The current number of PTPases in the human genome is unknown, but is estimated at somewhere between 100 to 500, with about 48 full length sequences currently known (1, 2). This number makes the PTPase superfamily one of the largest in the mammalian genome. PTPases can be further divided into those enzymes that specifically recognize only phosphotyrosine and those enzymes, known as the dual-specificity phosphatases, that recognize phosphotyrosine as well as phosphoserine and

2 phosphothreonine residues. A subclass of phosphatases has been found which also hydrolyze phospholipid substrates.
PTPases can be further divided into transmembrane and intracellular members of this family. In addition, a number of PTPases contain other protein interaction motifs such as proline rich sequences, SH2 domains, PDZ
domains, and tyrosine phosphorylation sites themselves. PTPase function and substrate specificity seem to be regulated by a variety of interacting proteins as well as by post translation modifications.
The T cell protein tyrosine phosphatase (TCPTP) was first identified in humans as two splice isoforms of the same gene, a 45 kDa isoform named TCPTPa and a 48kDa isoform named TCPTPb. Northern blot analysis has shown that the transcript is ubiquitously expressed, but in substantially higher amounts in hematopoietic tissues. TCPTPa has a relatively simple structure (figure 1 ) with an N terminal PTPase domain and a C terminal nuclear localization signal. Both isoforms are identical except for the 3' end of the mRNA, which in TCPTPb encodes a hydrophobic stretch of amino acids at the C terminus which results in retention of TCPTPb in the endoplasmic reticulum. The TCPTPb isoform is found in substantially lower amounts in the cell than the TCPTPa isoform (3).
TCPTP has been knocked out in mice (4). These knock out animals display a severe defect in both erythro- and lymphopoiesis, and any mature lymphocytes that are produced are functionally incompetent. Interestingly, if bone marrow from knock out animals is transplanted into irradiated wild type mice (irradiation is used to destroy any wild type bone marrow present), the resulting mature lymphocytes develop normally, but remain incapable of responding to T and B cell mitogens. These results seem to indicate that two separate defects exist, a stromal cell defect resulting in an abnormal bone marrow microenvironment and a proliferative defect in lymphocytes. Although both defects seem distinct, they share the common aspect of cytokine production and response, both of which are essential for normal hematopoiesis and lymphocyte proliferation.

3 To elucidate the role of any phosphatase in intracellular signaling, it is essential to identify its putative substrates. Recent work has shown that mutation of key catalytic residues in the PTPase domain will abolish the phosphatase activity of the enzyme but will not alter the Km (5). This allows the generation of PTPase mutants, known as substrate traps, which will still bind to their substrates in vivo and allow purification of the complex by coimmunoprecipitation.
The mutation of the PTPase catalytic cysteine to serine (CS) in the signature motif abolishes activity and, in some instances, this mutant can form stable complexes with substrates. It has been determined that mutation of the catalytic aspartic acid to alanine (DA) results in the formation of a substrate trap that forms complexes that are more stable than those with the cysteine mutant, and are therefore more useful in the identification of substrates.
For TCPTP, a study carried out in epidermal growth factor (EGF)-treated COS1 cells showed that TCPTP DA exited the nucleus and acted specifically upon the EGF receptor and p52Sn° (6). This report is the only one to date to identify substrates for TCPTP. Although TCPTP is ubiquitously expressed, the identification of these substrates does not adequately explain the hematopoietic defect observed in the gene targeted animals.
Description of the Invention The inventors have shown that T cell protein tyrosine phosphatase (TCPTP) is involved in the T cell signalling pathway downstream of the IL-2 receptor. The inventors have also identified previously unrecognised substrates on which TCPTP acts, Janus kinase 1 (JAK1 ) and Janus kinase 3 (JAK3).
TCPTP's role in signalling downstream of the IL-2 receptor sheds light on the previously unexplained connection between knock out of the TCPTP
gene and the appearance of hematopoietic defects.

4 The invention enables a method for screening a candidate compound for its ability to affect the interaction of a T cell protein tyrosine phosphatase (TCPTP) with a member of the JAK family of kinases, comprising:
(a) providing a preparation containing a T cell protein tyrosine phosphatase, a JAK kinase and a candidate compound; and (b) determining whether the candidate compound affects the interaction of the T cell protein tyrosine phosphatase with the JAK kinase.
The T cell protein tyrosine phosphatase may be obtained from mammalian hematopoietic cells, including T cells, or may be prepared by recombinant expression.
The JAK kinase may be obtained from hematopoietic cells or prepared by recombinant expression. JAK kinases 1 and 3 are preferred.
The effect of a compound on the interaction of TCPTP and a JAK
kinase may be determined, for example, by looking at its effect on the binding of TCPTP and a JAK kinase. A suitable assay is conducted in which candidate compounds are tested for their ability to increase or decrease the binding of a JAK kinase and TCPTP. Suitable assays will be known to those of skill in the art; for example, the binding assay described herein may be employed.
Once identified by such a method, a candidate compound may be useful as a pharmaceutical or may serve as a lead compound in the design of further compounds which may be useful as pharmaceuticals.
Compounds which reduce TCPTP binding to a JAK kinase, and therefore will reduce TCPTP activity against JAK kinases, are potentially useful as pharmaceuticals for suppressing cytokine-stimulated hematopoietic cell proliferation, particularly IL2-stimulated proliferation. It appears that TCPTP by dephosphorylating a JAK kinase, turns on the JAK kinase. These compounds are therefore potentially useful to treat any disorder in which immune suppression is desirable, for example T cell and other hematopoietic cell malignancies, autoimmune disorders and in preparation for and after tissue or organ transplantation.

Compounds which increase TCPTP binding to a JAK kinase may increase TCPTP activity in relation to JAK kinases, and are potentially useful as pharmaceuticals to boost immune responses in any disorder in which cytokine-stimulated T cell or hematopoietic cell proliferation is deficient, for example, in patients with SCID or in cancer patients undergoing cancer treatments such as chemotherapy.
The invention further enables a method for screening a candidate compound for its ability to modulate the activity of T cell protein tyrosine phosphatase, comprising:
(a) providing an assay system in which an activity of T cell PTP may be measured;
(b) determining the effect of a candidate compound on the T cell PTP
activity in the assay system.
As used herein, "activity" of TCPTP includes any functional activity as well as any downstream effect of any such functional activity.
For example, the effect of a candidate compound on IL2-stimulated proliferation in vitro of a T cell culture may be examined in an assay system such as that described in the examples herein.
The effect of a candidate compound on the ability of a TCPTP
preparation to dephosphorylate a JAK kinase, preferably a JAK1 or JAK3 kinase may also be examined. Suitable methods are known to those in the art, for example as described in Tiganis et al. (8).
Additionally, the effect of a candidate compound on the activity of a JAK kinase, preferably a JAK 1 or JAK 3 kinase, may also be examined, for example by a method such as that described in Barber et al. (9).
The identification of JAK1 and JAK3 as substrates for TCPTP will permit the determination of the properties of a preferred compound to affect the interaction of TCPTP and JAK.
Recent evidence from the TCPTP knock out animals indicates that this phosphatase plays a critical role in regulating cellular proliferation in lymphocytes. This information, coupled with the identification of JAK 1 and 3 as substrates, lends further proof for TCPTP's role in proliferation.

Deregulation of TCPTP control of proliferative signaling pathways could be an important factor in hematologic malignancies.
Another interesting aspect of TCPTP's function would be in the area of inhibitor design. The ability to develop inhibitors specific for TCPTP may not only be of use in chemotherapy, but also in immune suppression in surgical transplant operations.
When PTP1 B, the closest family relative to TCPTP, is knocked out in mice, the animals are resistant to diet-induced obesity and diabetes and, more importantly, are completely viable. Potentially, inhibitors for PTP1 B
could reduce obesity and type 2 diabetes. The close identity of the TCPTP
and PTP1 B catalytic domains means, however, that potential small molecule inhibitors may inhibit both PTPases, with undesirable immune suppression due to TCPTP inhibition.
In accordance with a further embodiment of the invention, potential anti-diabetic and anti-obesity PTP1 B inhibitors may now be screened for TCPTP inhibition to rule out any with the potential for immune suppression through TCPTP inhibition, and permit selection of those which selectively inhibit PTP1 B without affecting TCPTP activity.
Conversely, compounds showing TCPTP inhibition and therefore potentially useful as immune suppressors, can be screened for inhibitory activity against PTP1 B to rule out any unwanted effects on carbohydrate metabolism.
EXAMPLES
Materials- Unless specified all materials were obtained from Sigma. CTLL-2 cells were from Dr. Dwayne Barber (Ontario Cancer Institute, Toronto, ON), RPMI and Fetal Bovine Serum were from Wisent, and recombinant murine IL-2 was from Roche. CTLL-2 cells were routinely cultured in RPMI
supplemented with 10% (v/v) Fetal Bovine Serum, 5 U/mL penicillin, 5 mg/mL
streptomycin sulfate, and 55 pM 2-mercaptoethanol (Gibco BRL) at 37°C
with

5% C02. Antibody 4610 against phosphotyrosine, anti-Jak3 antiserum (#06-342), and anti-PARP were from Upstate Biotechnology. Monoclonal antibody anti-Jak1 was from BD Transduction Laboratories. All PCR reactions were performed using the Vent DNA polymerase (New England Biolabs) with primers synthesized by the Sheldon Biotechnology Centre (McGill University) or ACGT Corporation (Toronto, ON).
Methods Plasmids and mutagenesis- All wild-type or mutant cDNAs were placed in the pEF-BOS expression vector (Dr. Gary Koretzky, University of Pennsylvania) in the CIaI/Xbal sites. The cDNAs for TCPTP-WT as well as the pGEX-2TK-TCPTP-WT, CS constructs were from Dr. Michel Tremblay (McGill University).
The TCPTP-DA mutant was constructed using overlap extension PCR with the following mutagenic primers: (+) 5~tgaaacgagaaccatatctcac3~, (-) 5~ctggaaccccaaaagctggcc3~ and the following terminal primers:
(+)5~taaatcgatccaccatgtcggcaaccatcg3~, (-) 5~gtgtctagattaggtgtctgtcaatcttgg3~.
The same terminal primers were used for pEF-TCPTP-WT/CS. The TCPTP-DA GST fusion construct was PCR cloned using the following primers: (+) 5~gaaggatccatgtcggcaaccatcgagcgg3~, (-) 5~ggcgaattcttaggtgtctgtcaatcttgg3~ and ligated into the BamHI/EcoRl sites of pGEX-2TK. The integrity of all constructs was verified by fully sequencing the cDNA (Hospital for Sick Children, DNA sequencing facility).
In vitro substrate trapping- E. coli were transformed with the various pGEX
constructs and grown for 16 hrs in LB medium with 100 ~g of Ampcillin (LBA).
The culture was diluted 1:2 with LBA containing 1 mM IPTG and grown for 2 hrs at 37°C. The bacteria were pelleted and resuspended in PBS with 1 Triton-X100 and lysed via sonication at 50% power for 3 x 10 sec. The lysate was cleared and the GST-fusion proteins were bound to glutathione-sepharose (Pharmacia) for 45 min, then washed three times in PBS with 0.1 Triton-X100. Approximately 2~g of fusion protein was used for each experiment as compared to BSA standards. CTLL-2 cells, at log phase, were washed two times with PBS, and starved of cytokine for four hours. Cells were collected and resuspended at 10 x 106 cells/mL in RPMI and stimulated with 100 U/mL of recombinant murine IL-2 for 10 min at 37°C. After stimulation, the cells were pelleted, washed once with ice cold PBS, and lysed in HNMETG lysis buffer (50 mM Hepes(pH 7.5), 150 mM NaCI, 1.5 mM
MgCl2, 1 mM EGTA, 10 mM NaF, 10 % (v/v) glycerol, 1 % (v/v) Triton-X100, Complete protease inhibitors(Roche)) supplemented with 5 mM iodoacetic acid (IAA), for 20 min with gentle rotation at 4°C. Unreacted IAA was removed by incubating cell lysate with 10 mM DTT for 20 min at 4°C. The lysate was clarified by centrifugation at 10,OOOg for 10 min at 4°C.
Protein concentration was determined via the Bradford Assay (Bio Rad), and 500~,g of protein in 1 mL was pre-absorbed with 2 ~g of GST bound glutathione-sepharose beads (GSH-beads). The cleared lysate was transferred to a tube containing 2wg GST, TCPTP-WT/CS/DA bound GSH-beads, and incubated for 90 min at 4°C. The beads were washed four times with HNMETG and eluted in 2 x SDS sample buffer. Protein complexes were resolved by 8% or 10% SDS-PAGE and transferred to PVDF membranes (Immobilon, Millipore).
For phosphotyrosine, membranes were blocked 60 min in 1 % BSA in TBST.
4610 was added at 1 ~g/mL in blocking buffer for 60 min at room temperature.
The membranes were washed for 10 min, 5 min, 5 min and the appropriate secondary was used for 45 min in blocking buffer. After repeating the above washes, the membranes were visualized via enhanced chemiluminescence (Amersham), and exposed to film (Kodak). For other antibodies, the blocking buffer used was 5% non-fat dried milk (Carnation) in TBST. Membranes were stripped in 62.5 Tris (pH 6.8), 2% SDS, 100 mM 2-mercaptoethanol, and washed extensively before use.
Transfection and in vivo substrate trapping- CTLL-2 cells, at log phase growth, were washed twice with RPMI and resuspended at 20 x 106 cells/400~L RPMI in a 0.4 cm electroporation cuvette (Bio Rad). Plasmid DNA was added at 20~g and incubated with the cells for 10 min at room temperature. Cells were pulsed at 960~F and 250 V then chilled on ice for 10 min. Cells were resuspended in 20 mL CTLL-2 culture medium and grown for 16-18 hours as above. Transfectants were then washed twice with PBS and starved of cytokine for four hours. Transfectants were stimulated with 100 U/mL IL-2 for 10 min at 37°C. The cells were quickly pelleted and washed with ice cold PBS and lysed in HNMETG for 20 min at 4°C with gentle rotation. Lysates were cleared for 10 min at 4°C and 10,OOOg and protein quantitated as above. 1 mg of lysate in 1 mL was pre-cleared with protein A-sepharose for 30 min at 4°C, then incubated with monoclonal Ab 6F3 anti-TCPTP and protein-G sepharose for 90 min at 4°C. The immune complexs were washed four times with HNMETG and eluted in to 2 x SDS sample buffer. SDS-PAGE, transfer, and Western Blotting were as above.
Cellular fractionation- The method used is similar to that of Andrews and Faller. CTLL-2 cells, starved and/or IL-2 stimulated, were washed twice with ice cold PBS and resuspended in hypotonic buffer (10 mM Hepes(pH 7.9), 1.5 mM MgCl2, 10 mM KCI, with Complete protease inhibitors) for 10 min on ice.
The cells were lysed with a dounce homogenizer (type A pestle) until ~95% of cells were disrupted as determined by Trypan Blue uptake. Nuclei and cellular debris were pelleted for 10 sec in a benchtop centrifuge and washed once in hypotonic buffer. Nuclei were extracted with high salt buffer (20 mM
Hepes(pH 7.5), 25% glycerol, 420 mM NaCI, 1.5 mM MgCl2, 0.2 mM EDTA, with Complete protease inhibitors) for 20 min on ice. The extract was cleared by centrifugation at 10,OOOg at 4°C for 5 min. 20~g of each fraction, along with 20p.g HNMETG lysate was resolved by SDS-PAGE. Immunoblotting was performed with 6F3 anti-TCPTP and anti-PARP to determine the validity of the fractionation procedure.
Example 1 The murine cytotoxic T cell line CTLL-2, which is dependent on IL-2 for growth, was used as a model system to identify substrates downstream of the IL-2 receptor. Putative substrates were identified by transfection of expression vectors encoding either the wild-type murine TCPTP or the aspartic to alanine substrate trapping mutant (TCPTP DA) into CTLL-2 cells, followed by stimulation with recombinant IL-2. Under these experimental conditions, the TCPTP DA trapping mutant formed a stable complex with two phosphotyrosine (pTyr)-containing proteins, of 135 and 125 kDa respectively, as shown in Figure 2b. This interaction appeared to be 5 quite specific as compared to anti-phosphotyrosine immunoblotting of starved versus IL-2 stimulated lysates (TCL) under the same conditions.
A similar experiment was performed in the murine T cell hybridoma D011.10, with ligation of the T cell receptor used to induce tyrosine phosphorylation of intracellular signaling molecules. The results are shown in 10 Figure 2a. Interestingly, stable complexes of the phosphatase and potential substrates were not isolated in TCPTP immunoprecitpitates from these transfectants. This indicates that the role of TCPTP is specific in signal transduction downstream of cytokine receptors, and it does not seem to play a role in T cell receptor signaling.
Example 2 To identify the potential substrates complexed with the TCPTP DA, an in vitro GST fusion protein mixing assay was used. In this assay, the TCPTP
wild-type or the CS or DA trapping mutants were immobilized to an affinity matrix, and either starved or IL-2 stimulated protein extracts were incubated with the recombinant proteins. As seen in Figure 3, top panel, two distinct tyrosine phosphorylated proteins formed complexes with the TCPTP DA
fusion protein in vitro. Since it had been previously reported that the Janus kinases (JAK) 1 and 3 are rapidly tyrosine phosphorylated after IL-2 stimulation, we attempted to identify these coprecipitating bands by immunoblotting with antibodies against JAK 1 and 3. As observed in figure 3, second panel from top, JAK 1 was coprecipitated inducibly upon IL-2 stimulation, while JAK3 was precipitated with the TCPTP DA matrix in both starved and stimulated cells (third panel from top). Coomassie staining of the membrane shows equal loading of the fusion proteins used in the assay (bottom panel).

Example 3 To determine if these observed in vitro interactions were physiologically relevant, we repeated the experiment in vivo as illustrated in Figure 4. Total cell lysates from TCPTP wild type transfectants (Figure 4b) showed that two main tyrosine phosphorylated proteins are dephosphorylated, as compared to the remainder of cellular proteins present. In Figure 4a, in TCPTP wild-type or DA transfectants, TCPTP was immunoprecipitated and the immune complexes resolved and blotted for phosphotyrosine-containing proteins. The open headed arrow indicates potential substrates for TCPTP downstream of the IL-2 receptor. Interestingly, along with the protein of 135 and 125 kDa, another 55 kDa protein was constitutively associated with TCPTP DA.
Probing the membrane with antibodies for JAK 1 and 3 demonstrated that indeed JAK3 was coimmunoprecipitated with the TCPTP DA trapping mutant.
As well, the known substrate p52Sn° was also trapped inducibly with the addition of IL-2. One reason for not observing an interaction with JAK1 may be the existing hierarchy of the JAKs in cytokine signaling. It has been shown that JAK3 is required for JAK1 phosphorylation and activation in JAK3 null B
cells (7). If TCPTP is acting on JAK3, this may not allow efficient enough phosphorylation of JAK1 to allow a phosphotyrosine dependent interaction with TCPTP.
Example 4 Since TCPTP contains a nuclear localization signal, and in fibroblasts and epithelial cells has been found to localize to the nucleus, we questioned whether TCPTP exited the nucleus upon IL-2 stimulation or if a cytoplasmic pool existed in T cells. Fractionation of CTLL-2 cells (Figure 5), either starved for cytokine or IL-2 stimulated, were fractionated into a cytosolic/membrane and nuclear fraction. Equal amounts of protein lysate from each fraction and condition were loaded and resolved. Immunoblotting with antibodies against TCPTP indicated that roughly 20% of the phosphatase localized to the cytosolic fraction, with the remainder present in the nucleus. Reprobing the membrane with antibodies against PARP, a known nuclear protein, proved the integrity of the nuclear fraction under our conditions. This suggests that a cytoplasmic pool of TCPTP may be present to act upon JAK1 and 3 in IL-2 receptor signaling. Finally, to further affirm the specificity of this interaction, TCPTP DA transfectants were immunoprecipitated with or without sodium orthovanadate, a known competitive inhibitor of tyrosine phosphatases (figure 4c). Anti-phosphotyrosine immunoblotting revealed that in the sample containing orthovanadate, all coimmunoprecipitating proteins were excluded from the active site of the trapping mutant. This indicates that these potential substrates do in fact interact with the enzyme active site and not another portion of the phosphatase.

References (1 ) Tonks N.K., Neel B.G. (1996) Cell. 87, 365-8.
(2) Li L., Dixon J.E. (2000) Semin. Immunol. 12, 75-84.
(3) Ibarra-Sanchez M.d., Simoncic P.D., Nestel F.R., Duplay P., Lapp W.S., Tremblay M.L. (2000) Semin. Immunol. 12, 379-86.
(4) You-Ten K.E., Muise E.S., Itie A., Michaliszyn E., Wagner J., Jothy S., Lapp W.S., Tremblay M.L. (1997) J. Exp. Med. 29, 683-93.
(5) Tiganis T., Bennett A.M., Ravichandran K.S., Tonks N.K. (1998) Mol. Cell.
Biol. 18, 1622-34.

(6) Flint A.J., Tiganis T., Barford D., Tonks N.K. (1997) Proc. Natl. Acad.
Sci.
U. S. A. 94, 1680-5.
Oakes S.A., Candotti F., Johnston J.A., Chen Y.Q., Ryan J.J., Taylor N., Liu X., Hennighausen L., Notarangelo L.D., Paul W.E., Blaese R.M.,

(7) O'Shea J.J. (1996) Immunity 5, 605-15.

(8) Tiganis et al.(1998), Mol. Cell Biol., v. 18, p. 1622.

(9) Barber et al. (1994), Mol. Cell Biol., v. 10, p. 6506.

Claims