Eur. J. Immunol. 1991. 21: 913-919

Mary A. Valentineo, Michael B. Widme#, Jeffrey A. LedbetteP, Fran Pinadto, Robert Voice*, Edward A. Clarko, Byron Gallis* and David L. Brautigan+ Department of Microbiologyo SC-42, University of Washington, Oncogen Corporation*, lmmunex Corporation*, Seattle and Section of Biochemistry+, Division of Biology and Medicine, Brown University, Rhode Island

IL 2-induced serine phosphorylation of CD45

Interleukin 2 stimulates serine phosphorylation of CD45 in CTLL-2.4 cells* Ligation of interleukin 2 (IL2) is known to regulate both protein tyrosine and serinehhreonine phosphorylation. A family of leukocyte transmembrane proteins whose cytoplasmic domain exhibits intrinsic protein tyrosine phosphatase activity is collectively called CD45 and is identified by a set of common cell surface epitopes. Although CD45 is known to be a phosphoprotein, it is not known how phosphorylation specifically regulates its function. We therefore identified a cell line, the IL4-dependent line CTLL-2.4, in which CD45 could be phosphorylated in response to addition of IL2. These cells are a variant of an IL 2-dependent murine cell line which were selected for long-term growth on IL 4 but which retain the ability to proliferate on exposure to IL2. Incubation of CTLL-2.4 in low serum concentrations followed by stimulation with IL 2 caused a three- to fivefold increase in the phosphorylation of CD45 in a time- and concentration-dependent manner. CD45 in non-stimulated cells contained one major tryptic phosphopeptide, whereas, after exposure of the cells to IL2, two new phosphopeptides were present in CD45. The pattern of IL2-induced phosphorylation was different from that found following addition of phorbol 12myristate 13-acetate (PMA) to the cells. Although IL2 induced rapid and potent tyrosine phosphorylation in CTLL-2.4 cells, all of the basal and cytokine-activated phosphorylation of CD45 occurred on serine residues. The IL 2-stimulated phosphorylation caused no change in the amount of cell surface CD45 and no alteration of its catalytic activity using an artificial tyrosine phosphorylated substrate-RCM-lysozyme. We speculate that the increase in phosphorylation of CD45 may modify its association with potential substrates. The differences in the phosphorylation patterns induced by IL 2 and PMA further suggest that more than one kinase can use CD45 as substrate and that IL2 activates a protein serinekhreonine kinase different from protein kinase C.

1 Introduction Phosphorylation of cellular proteins is an important means of regulating protein activity. The signaling pathways that control the kinases and phosphatases in lymphocytes are beginning to be defined and these pathways can be triggered by ligation of IL2 to its receptor onTcells [l-41. One of the Tcell proteins involved inTcell phosphorylation is CD45, a member of the phosphotyrosil-protein phosphatase (PTPase) family.The PTPases have numerous characteristics which distinguish them from all other known protein phosphatases. They are inhibited by micromolar concentrations of Zn2+[5,6] andNa3V04[7], but activated by the protein serine phosphatase inhibitors, EDTA and NaF [ 5 , 6 ] .They are essentially inactive toward phosphoserine-containing substrates [8], and phosphothreonine peptides [9], but exhibit a high rate of activity and submicromolar K, values toward phosphotyrosine-containing substrates including insulin and growth factor

[I 90871

*

913

This workwas supported by National Institutes of Health Grants GM 42508, GM 37905, GM 35266, and by the Immunex and Oncogen Corporations.

Correspondence: Mary A.Valentine, Department of Microbiology SC-42, University of Washington, Seattle, WA 98195, USA Abbreviations: PTPase: Protein tyrosine phosphatase TCA: Trichloroacetic acid 0 V C H Verlagsgesellschaft mbH, D-6940 Weinheim, 1991

receptors [lo]. Other characteristics of these enzymes have been described in a recent review [ ll] . Purification of PTPases from placenta [12, 131 has provided protein sequence information concerning the cytosolic form of PTPase, which has about 30% identity in sequence to the tandem repeats in the C-terminal domains of the leukocyte common antigen, CD45 [14]. It is not surprising, therefore, that CD45 also possesses PTPase activity [15]. Further sequence analysis [ 161 and isolation of a cDNA clone [171 from a peripheral Tcell library establishes the PTPases as a unique family of proteins. The CD45 molecule is an integral membrane protein found in hematopoietic cells [18, 191. Different isoforms of CD45, ranging in molecular mass from 180-220 kDa, are produced in various cell types by alternative mRNA splicing [20-221. Primary structures deduced from cDNA sequences coding for CD45 reveal it as a protein with an extracellular domain ranging from 400-550 residues, a 22-amino acid transmembrane segment, and a cytoplasmic segment of approximately 700 amino acids, which itself has two highly similar domains of approximately 300 residues [22,23]. Progress has been made in understanding the function of CD45. Early studies have shown that the mAb to CD45 may inhibit Tand B cell proliferation [24-261, or may augment the PHA-stimulated Tcell proliferation [27]. Cross-linking CD45 alone or cross-linking CD45 with cell surface proteins which can trigger increases in intracellular free Ca2+[Ca2+]i in T and B cells, blocks the increase in [Ca2+]i and cell proliferation [28,29]. CD45 may regulate the activity of a 0014-2980/91/0404-0913$3.50+ .25/0

914

M. A.Valentine, M. B. Widmer, J. A. Ledbetter et al.

tyrosine protein kinase, pp56ICk,by dephosphorylation of tyrosine 505, a putative negative regulatory site [30,31]. Details of CD45 structure and function can be found in recent reviews [32-341. CD45 is present in murine cells as a phosphoprotein containing only phosphoserine [35, 361. Here we show that in murine CTLL-2.4 cells, CD45 is constitutively phosphorylated in a single major site but that addition of IL2 induces phosphorylation in two additional sites of CD45. This pattern of phosphorylation differs from that induced by addition of PMA, indicating that ligation of the IL2R activates a kinase(s) different from PKC.

2 Materials and methods 2.1 Cell lines A variant of the IL2-dependent CTLL-2 [37] cell line, CTLL-2.4, was selected for long-term growth in IL4. Briefly, CTLL-2 cells were maintained in continuous culture at 37°C in DMEM containing 5% FBS, 50 pM 2-ME, additional amino acids [38] and 5 ng/ml human or murine rIL2. The cells were subcultured three times per week. For culture with IL4, 1 x 105 CTLL-2 cells were washed free of IL2 and plated in DMEM containing 10 ng/ml IL4. The vast majority of cells died within 2 days upon first exposure to IL4 in the absence of IL2. However, viable proliferating cells emerged in the cultures after several weeks and these were further subcultured in IL 4.The line is referred to as CTLL-2.4.This cell line has been maintained by continuous culture in IL4 for more than 1 year and proliferates equally well in the presence of IL2 or IL4.

2.2 Radiolabeling and immunoprecipitation of CD45

Eur. J. Immunol. 1991. 21: 913-919 Francisco, CA) GS-300 scanning densitometer and a Waters (Milford, MA) 740 data module integrator.

2.3 Anti-phosphotyrosine Western blotting of CTLL-2.4 proteins CTLL-2.4 cells were grown in IL4, washed twice with PBS, and resuspended in medium without IL4 and containing 0.2% serum. Cells were incubated at 5 x 106 cells/ml for 2.5 h. For a zero time point, 5 x 106 cells were removed prior to addition of IL2 at a final concentration of 1 nM. Cells were removed at 10, 20 and 30 min after incubation with IL2, centrifuged for 1 min in a microfuge, and boiled in 200 p1 of Laemmli buffer containing DTT. Fifty microliters of this cell lysate was fractionated by SDS-PAGE and transferred to PVDF (Irnmobilon, Millipore Corp., Bedford, MA).The proteins were reacted with 0.25 pg/ml of a mAb to phosphotyrosine [42] and then with 1251-labeled protein A (ICN Biomedicals, Costa Mesa, CA).The paper was subjected to autoradiography at -70°C with an intensifying screen.

2.4 Preparation of 32P-TpRCM-lysozyme Egg lysozyme was denatured in guanidinium chloride, reduced and alkylated with iodoacetamide and then modified with maleic anhydride in accordance with the procedure described by Tonks et al. [12]. Phosphorylation on tyrosine was carried out in a 0.5-ml reaction mixture composed of the following solutions added in the order listed: 50 p1 of 500 mM Hepes, pH 7.0, 50 pl of 100 mM MgC12, 10 mM MnC12, 1 mM EDTA; 5 p1 of 1.0% Brij 35; 20 yl of 5 M NaCl, 1 pl of 2-ME; 129 p1 of H 2 0; 100 pl of 1 mM ATP containing 250pCi [ Y ~ ~ATP P ] (5000 dpndpmol) and 20 pl of tyrosine kinase (Oncogene Sci.). Final concentrations were 50 mM Hepes, 10 mM MgC12, 1 mM MnC12, 0.1 mM EDTA, 200 mM NaCl, 30 mM 2-ME and 100 PM ATP.

CTLL-2.4 cells were grown in IL4, washed twice with After overnight incubation at 30°C, the reaction was normal saline, and resuspended in DMEM (phosphate terminated by addition of 80 p1of 100% trichloroacetic acid free, Gibco, Grand Island, NY) containing 0.2% FCS.The (TCA) and chilled on ice 20 min. Precipitated protein was cells were incubated in a flask at 1 x 106/ml and collected by centrifugation at 13000 x g for 5 min and 1 mCi/ml = 37 MBq/ml 32PO:- for 2.5 h. At this point, washed with 10% TCA twice by suspension and centrifudiluent (normal saline), or human rIL2 (Immunex Corpo- gation. The tube was then gently rinsed without resuspenration, Seattle, WA) was added to the cultures at one or sion of the pellet with 1 ml of -20°C acetone. The pellet more concentrations for the indicated periods of time. was dissolved in 100 pl of 0.5 M Hepes, pH 7.0, then diluted CTLL-2.4 cells were centrifuged for 1 min in a microfuge, to 1 ml by addition of 5 pl of 0.2 M EDTA, 1 p1 of 2-ME and the SN was decanted, and the pellet was lysed in a buffer 894 yl of H20. A 5-yl sample was counted to determine the [39] containing protease inhibitors and phosphatase inhibi- [32P]tyrosineconcentration and substrate then diluted to tors. The phosphatase inhibitors included 100 pM Na3V04, 2 p~ with 50 mM Hepes, pH 7.0, 1mM EDTA, 15 mM 50 mM NaF, 50 mM Na2HP04, and 50 mM P-glycerophos- 2-ME and stored at -20°C. phate. The lysates were centrifuged, precleared with protein G-Sepharose, and immunoprecipitated with protein G-Sepharose (Pharmacia, Uppsala, Sweden) as de- 2.5 Assay for CD45 PTPase activity scribed [40] and rat mAb to CD45 (30Fll; [41]). Control immunoprecipitations employed a rat isotype-matched CI'LL-2.4 cells were incubated in 0.2% serum for 2.5 h at antibody, IgGzb, to murine granulocyte MO-CSF (Immun- 1 x 106/10ml and then treated with various concentrations ex). Three to five micrograms of respective antibody were of IL2 for the indicated times. One million cells were used for each immunoprecipitation. Immunoprecipitates pelleted, lysed in 1 ml of a buffer containing 10 p~ soybean were boiled in Laemmli sample buffer, separated by trypsin inhibitor, 10 pM leupeptin, 50 mM P-glycerophosSDS-PAGE, and autoradiographed as described [39]. The phate, 50 mM NaF, 50 mM Na2HP04, 150 mM NaCl, 1 mM densities of CD45 bands were quantified with a Hoefer (San EDTA, 1% NP40, 5 mM DTT, pH 7.0. The lysate was

IL 2-induced serine phosphorylation of CD45

Eur. J. Immunol. 1991. 21: 913-919

centrifuged for 5 min in a microfuge and the SN was precleared with a control antibody and a protein ASepharose [39]. After preclearing, 100 pl of lysate (the equivalent of 100OOO cells) was incubated with 200 p1 of lysis buffer, 3 pg of 30Fll anti-CD45 mAb, and 65 p1 of a 20% suspension of protein G-Sepharose for 2 h, 4"C, on a rocker. The immune pellet was washed four times with 700 pl of lysis buffer and once with 1 ml of reaction buffer (150 mM NaCl, 1 mM EDTA, 25 mM Hepes, pH 7.0,5 mM D m ) . Reactions were initiated by addition of 30 p1 of [32P]RCM-lysozyme to 70 p1 of reaction buffer (final concentration of substrate was 840 nM). Reactions were for 3 min at 30 "C and were linear with respect to time and input cell protein (data not shown). Typically, 10%-15% of the cpm were released from RCM-lysozyme. During the reactions, immune pellets were vortexed gently at 1, 2 and 2.5 min to maintain a suspension of Sepharose. The reactions were diluted by addition of 400 p1 of 25 mg/ml BSA, centrifuged briefly in a microfuge to pellet the Sepharose, and 500 pl of SN was removed and added to 120 p1 of 45% TCA.TCA precipitates were maintained on ice for 10 min, centrifuged for 10 min in a rnicrofuge, and 500 pl of the SN fluid was counted in 3 ml of ECOSCINT (National Diagnostics, Manville, NJ). 2.6 Phosphopeptide mapping and phosphoamino acid analysis Phosphopeptide mapping was performed as previously described [43]. Briefly, the CD45 band was excised and washed extensively prior to digestion overnight with 150 pg of L-1-tosylamido-2-phenylethylchloromethyl ketonetrypsin. Digested phosphopeptides were lyophilized, dissolved in 10% pyridine, 0.5% acetic acid (pH6.5) and electrophoresed prior to ascending chromatography in n-butylalcohoVacetic acid/water/pyridine (15 : 3 : 12 : 10). Phosphopeptides were visualized' by autoradiography. Phosphoamino acid analysis of CD45 was done using CD45 blotted onto Immobilon membranes [44]. The membrane was cut into small pieces and the attached CD45 hydrolyzed in 200 p1 of 5 N constant-boiling HCI for 1h (110°C). The SN fluid was transferred to a new tube, dried in a speed vac and the amino acids resuspended in formic acidacetic acid buffer (pH 1.9) prior to spotting onto a glass-backed silica plate. Electrophoresis proceeded as described and unlabeled phosphoamino acids were included as internal markers [45]. 2.7 FCM analysis CD45 surface expression was measured using FITC-conjugated 30Fll mAb and an EpicsC (Coulter Electronics, Hialeah, FL) flow cytometer. Thy-1.2 antigen expression was measured as a control using 30-H12, an isotypematched mAb.

3 Results

A

915

B C D

-200 -116 -97

-67

-43

Figure 1. Enhancement of CD45 phosphorylation by IL2. CTLL2.4 cells were incubated in phosphate-free DMEM and 0.2% serum with 3zPO:- for 2.5 h. Equal aliquots of cells were incubated with two different concentrations of IL2 or diluent for 10 min. The CTLL-2.4 cells were lysed and immunoprecipitated with control antibody (lane A) or a mAb (30Fll) to CD45 (lanes B-D). The immunoprecipitates were fractionated by SDS-PAGE and an autoradiograph of the gel is shown. Cells were treated for 10 min with (A) 1 n M IL2, (B) no IL2, (C) 0.1 n M IL2 and (D) 1.0 n M IL2. A densitometric scan of the CD45 band yielded densities of (A) 0.11, (B) 0.71, (C) 1.71 and (D) 2.52 (see text).

did not stimulate phosphorylation of CD45 in these cells after they were starved for IL4 (data not shown), IL2 did cause an increase in CD45 phosphorylation (Fig. l).Therefore cells were routinely grown in IL4 and then incubated in low serum concentrations without exogenous cytokine prior to addition of IL2. CTLL-2.4 cells grown in IL4 were incubated in phosphatefree DMEM and 0.2% FCS with 32POi- for 2.5 h. Aliquots of cells were incubated with two different concentrations of IL2 or diluent for 10 min.The CTLL-2.4 cells were lysed and CD45 was immunoprecipitated with anti-CD45 mAb (30Fll; [39]) or cell lysates were immunoprecipitated with a control isotype-matched IgGZb mAb against granulocyte-M@ CSF. An autoradiogram of the immunoprecipitates is shown in Fig. 1. Although CD45 was phosphorylated in the absence of IL2 (Fig. 1, lane B), the phosphorylation level was increased by incubation of the cells with 0.1 nM IL2 or 1.0 nM IL2 for 10 min (Fig. 1, lanes C and D). Incubation of the cells with 1 nM IL2 followed by immunoprecipitation with isotype-matched control antibody showed that no 180-kDa protein or higher molecular mass isomers of CD45 were immunoprecipitated (Fig. 1, lane A). A densitometric scan of CD45 showed that 0.1 nM and 1.0 nM IL2 increased the phosphorylation of CD45 by 2.5- and 3.5-fold, respectively, above the basal level of phosphorylation.

3.1 IL %dependentphosphorylation of CD45

3.2 The phosphorylation of CD45 after addition of IL2 to CTLL-2.4 cells is time dependent

The CTLL-2 cells used in this study were ILCdependent CTLL-2.4. Preliminary studies showed that whereas IL 4

CD45 was also phosphorylated in a time-dependent manner after addition of IL2 to CTLL-2.4 cells. Cells were

916

M. A.Valentine, M. B. Widmer, J. A. Ledbetter et al.

Eur. J. Immunol. 1991. 21: 913-919

A B C D

-200 -116 - 97 -67

SE R TH R TVR

-43

Figure 2. Time course of phosphorylation of CD45 in response to 1L2. CTLL-2.4 cells were labeled with 32PO:- and immunoprecipitated as described in the legend to Fig. 1. Cells were immunoprecipitated with control antibody in lane A and mAb 30Fll to CD45 in lanes C-D. Cells were incubated with 1 nM IL2 for (A) 10 min, (B) 0 min, (C) 10 rnin and (D) 30 min. A densitometric scan of the CD45 band yielded densities of (A) 0.12, (B) 0.28, (C) 1.10 and (D) 1.37 (see text).

grown on IL4 and preincubated in low serum concentrations and 32POi- as described above. Immunoprecipitation of CD45 from CTLL-2.4cells at 10 min (Fig. 2,lane C) and 30 min (Fig. 2, lane D) after addition of IL2 showed that the level of phosphorylation increased with increasing time of exposure to IL2 (Fig. 2, lane B). A densitometric scan of the autoradiogram of immunoprecipitated 32Plabeled CD45 revealed that the level of phosphorylation rose by almost threefold at 10 rnin and nearly fivefold at 30 rnin after addition of IL2. The increases occurred even after subtraction of the higher background densities in lanes C and D of Fig. 2. N o protein in this area of the autoradiogram was observed after immunoprecipitation with a control isotype-matched antibody (Fig. 2, lane A).

Figure 3. Phosphopeptide and phosphoamino acid maps of CD45 in the absence and presence of IL 2. CD45 was immunoprecipitated from CTLL-2.4 cells, and fractionated by SDS-PAGE as described in the legend of Fig. 1.The 32P-labeledCD45 bands were localized by autoradiography and excised from the gel as described in Sect. 2.6. (A) and (B) are 2-D tryptic peptide maps of 32P-labeled CD45. (C) and (D) are phosphoamino acid maps of 32P-labeled CD45. CD45 was extracted from cells treated with diluent (A and C) or 0.1 nM I L 2 for 10 rnin (B and D).

ly phosphorylated peptide (Fig. 3B, light arrow) either remained the same in intensity or decreased. Phosphoamino acid analysis of CD45 from CTLL-2.4 cells incubated in the absence (Fig. 3C) or in the presence of IL2 (Fig. 3D) shows that the entire increase in 32POi- is in phosphoserine. No phosphotyrosine or phosphothreonine was detected in CD45.

Addition of phorbol esters (100 ng PMA/ml, 5 min) resulted in a different tryptic map from that seen following addition of IL2. PMA increased phosphorylation of only the cathodal peptide (Fig. 4) and did not affect either of the two anodal peptides. These results suggest that the site on CD45 which is phosphorylated by PKC is most likely on the cathodal peptide and that IL2 can activate a kinase other than PKC.

3.3 Phosphopeptide and phosphoamino acid maps of CD45 after incubation of CTLL-2.4 cells with or without IL 2

CTLL-2.4 cells were incubated for 2.5 h in 0.2% serum in the presence of 32PO:- and treated with diluent or 0.1 nM IL2 for 10 min. Immunoprecipitated CD45 was isolated for digestion with trypsin and phosphopeptides of CD45 were separated by high voltage electrophoresis in one dimension and thin layer chromatography in a second dimension [43]. The tryptic digest of constitutively phosphorylated CD45 revealed a single major phosphopeptide migrating toward the anode (Fig. 3A). Addition of IL2 to CTLL-2.4 cells resulted in the appearance of two new phosphopeptides in addition to the peptide phosphorylated in the absence of IL2 (Fig. 3B). This increase is consistent with the 2- t o 3-fold increase measured by densitometry and these patterns were reproduced in five separate experiments. In replicate experiments, phosphorylation of the constitutive-

+

0

Figure 4 . Phosphopeptide map of CD45 from PMA-stimulated CTLL-2.4 cells. CTLL-2.4 cells were incubated for 5 min with 100 nglml PMA and CD45 was immunoprecipitated and subjected to phosphopeptide mapping as in Fig. 3.

Eur. J. Immunol. 1991. 21: 913-919

IL2-induced serine phosphorylation of CD45

917

3.4 Surface expression of CD45 in cell streated with IL2 It was possible that the phosphorylation of CD45 induced by IL2 was correlated with regulation of the level of cell surface CD45. Therefore, CTLL-2.4 cells were incubated with 0.2% serum for 2.5 h and then treated with IL2 for 15 min or 60 min. The cells were then stained with FITCanti-CD45 and the mean fluorescence intensity was determined (Fig. 5). There was no change in the mean fluorescence intensity of cells treated with IL2 for 15 min or 1 h, compared to the untreated cells (Fig. 5). CI'LL-2.4 cells incubated in low serum concentrations were also treated with IL4 for 15 min and 60 min and no change in the surface levels of CD45 was observed (data not shown).

Y2-U00

2

Fluorescence Intensity (log,,) Figure5. Surface expression of CD45 on CTLL-2.4 cells in the absence or presence of IL2. CTLL-2.4 were incubated for 2.5 h in 0.2% FCS. Cells were untreated (-) or treated for 15 min (. * .) or 60 min (-.-.-) with 1 nM IL2.The cells were then incubated with the FITC-conjugated anti-CD45 antibody 30Fll and CD45 surface immunofluorescence was measured by FCM with a Coulter Epics C. Autofluorescence of unstained cells (---) is also shown.

3.5 PTPase activity of CD45 from CTLL-2.4 cells in the absence and presence of IL2 CTLL-2.4 cells were grown in IL4 and then incubated in 0.2% serum for 2.5 h. The cells were then incubated with diluent or 10 nM IL2 for 2 , 5 , 10 and 20 min.The cells were lysed and CD45 was immunoprecipitated and assayed for catalytic activity in the immune complex with the substrate RCM-lysozyme. No change in activity was observed throughout the time course of incubation with IL2 (Fig. 6A).The enzyme was assayed under conditions where the release of phosphate from RCM-lysozyme was linear with time and protein input (data not shown). In all experiments where CD45 was measured, immunoprecipitates of ILZtreated cells were performed in parallel with IgG2b isotype-matched control antibody. With these precipitations no 32Pwas released from the RCM-lysozyme substrate. Immunoprecipitation of cells with two other, irrelevant mAb (to the IL 1R and IL4R) also did not cause release of 32P from the substrate. Thus, the 32Preleased from the immune precipitates with anti-CD45 mAb are specific for the presence of CD45. The level of catalytic activity of CD45 was also measured after incubation of CTLL-2.4 cells for 10 min with concen-

20

T h e (mid

0.1

1

10

I

I

I

I

0.5

1

5

10

IL 2 (nM)

Figure 6. Time course and dose-response curve of PTPase activity after addition of IL2 to CTLL-2.4 cells. (A) CTLL-2.4 cells were incubated in 0.2% serum for 2.5 hand then saline (open circles) or 10 nM IL2 (closed squares) was added for 2,5,10 and 20 min. Cell extracts were prepared and CD45 was immunoprecipitated with 30F11 mAb to CD45 and parallel samples done with isotype control antibody. The immunoprecipitates were incubated for 3 min at 30°C with [32P]RCM-lysozyme(total cpdsamples = 150000) and processed as described in Sect. 2.5. No activity was observed in reactions with isotype-matched control antibody; all samples were corrected for acid-soluble 32P in RCM-lysozyme (blank = 850 cpm). Values presented are the mean of two independent experiments done 1 month apart, each value being the average of duplicate determinations. (B) CTLL-2.4 cells were incubated in 0.2% serum for 2.5 h and then diluent (saline) or IL2 at a final concentration of 100 pM, 500 pM, 1, 5 or 10 nM was added for 10min. CD45 was immunoprecipitated for PTPase assay as described in (A), above. N o activity was observed in precipitates with control antibody; substrate contained 105OOO cpdsample and blank 830 cpm.

trations of IL2 ranging from 100 pM to 10 nM (Fig. 6B). N o change in CD45 activity was observed.

3.6 IL %induced tyrosine phosphorylation in CTLL-2.4 cells The increase in phosphate content in CD45 in response to IL2 was solely in phosphoserine, even though it has been shown that IL 2 stimulates increases in tyrosine phosphorylation in cells treated with IL2 [l]. The effect of IL2 on tyrosine phosphorylation in IL 4-dependent CTLL-2.4 cells was examined by immunoblotting of whole cell lysates with an anti-phosphotyrosine antibody. The cells were starved for IL2 as above and treated with 1 nM IL2 for 10,20 and 30 min, or left untreated. Major substrates of 95, 53 and 46 kDa were tyrosine phosphorylated (Fig. 7). Additional

918

M . A.Valentine, M. B. Widmer, J. A . Ledbetter et al.

CTLL 1L2 0 10 20 30 219100-

67-

42.7-

27.4-

18Figure 7. Tyrosine phosphorylation in CTLL-2.4 cells by addition of IL2.Whole cell lysates of CTLL-2.4cells,either unstimulated or stimulated with 1 nM IL 2 for 10,20 or 30 min were electrophoresed on 10% polyacrylamide gels and then transferred to PVDF paper followed by immunoblotting with hapten (pheny1phosphate)purified anti-phosphotyrosine mAb and '251-labeled protein A. There was no reactivity of the antibody in the presence of 40 mM phenylphosphate (data not shown).

tyrosine phosphorylation of proteins of 120,82,40,37 and 32 kDa appeared after IL2 stimulation, but these were minor substrates. Notably, no protein of 180 kDa was tyrosine phosphorylated that could represent CD45. Immunoprecipitations of CD45 with 30Fl1, followed by immunoblotting with anti-phosphotyrosine antibody also did not provide any evidence that CD45 was phosphorylated on tyrosine after IL2 stimulation. These combined data show that addition of IL2 can increase both tyrosine and serine phosphorylation in these cells and that CD45 is selectively phosphorylated only in serine residues by more than one kinase.

4 Discussion This work demonstrates that IL 2 stimulates both protein tyrosine and serine phosphorylation in JL 2-responsive CTLL-2.4 cells. These cells are a variant of an IL2dependent cell line and were selected for long-term growth in IL4 but remain responsive to IL2. As the cells had not been exposed to IL2 for 1 year prior to these experiments, they offered a model which permitted distinction between

Eur. J. Immunol. 1991. 21: 913-919

levels of CD45 phosphorylation in the basal and IL2induced states. As shown by others [35, 361, CD45 is constitutively phosphorylated on serine. CD45 in untreated CTLL-2.4 cells contains one major phosphopeptide labeled in serine, but addition of IL2 to CTLL-2.4 cells enhances serine phosphorylation in two additional phosphopeptides. The increased phosphorylation of CD45 is dependent upon time of incubation with IL2 and IL2 concentration. IL 2-enhanced phosphorylation of CD45 is not correlated with decreased surface expression or with an alteration of the enzymatic activity of CD45. IL2 stimulates phosphorylation of proteins on both tyrosine [ 11 and serine and threonine residues [2-41 in responsive cells. The IL2 intermediate-affinity p75 receptor is required for enhancement of phosphorylation of proteins [46-48, but the IL2R apparently is not a protein kinase because it lacks a sequence characteristic of this family of enzymes 1491. IL2 induces translocation 1501 and activation 1511 of PKC, but PKC is not required for IL2-driven cell proliferation [52-541. Although addition of PMA can enhance serine phosphorylation of CD45 1551, tryptic maps show that the phosphorylation pattern is distinct from that seen following addition of IL2. Therefore, other protein serine kinases, such as those activated by growth factors in a putative cascade mechanism 156, 571, can contribute to phosphorylation of CD45. The increased phosphorylation of certain proteins on tyrosine residues as detected by Western blotting of CTLL-2.4 cells may be a consequence of diminished dephosphorylation by CD45 involving changes in affinity rather than velocity (see below). Antigen or other factors which stimulate T cells do not cause alteration of cell surface levels of CD45. We also did not find a change in the amount of surface CD45 correlated with its IL 2-induced phosphorylation. In neutrophils, formylmethionylleucylphenylalanineor the calcium ionophore A23187, increase degranulation and the cell surface level of CD45 increases by mobilization of the protein from intracellular granules 1581. Ligation of CD45 on human T cell blasts, however, does have functional consequence. Cross-linking of surface CD45 on T cells reduces the proliferation in response to IL 2 by 50% -70%, demonstrating a regulatory interaction between CD45 and the IL2 signaling mechanism [59]. These results could be interpreted that CD45 in part modulates the IL2R signal transduction, or that CD45 antagonizes IL2 stimulation, a process enhanced by ligation and clustering of CD45. We were unable to suppress the IL2 response with anti-CD45 30Fll mAb. This may not be surprising, however, since CTLL-2.4 cells have less CD45 than normal lymphocytes, and have been selected for IL2 or IL4 hyperresponsiveness over many generations. Therefore, the balance of CD45 vs. IL 2 in CTLL-2.4 cells is quite different and may account for enhanced phosphorylation but lack of antagonism by 30Fll. Phosphorylation of serine residues in CD45 in response to IL2 did not seem to change the phosphatase activity of CD45, at least when using an artificial substrate with immunoprecipitated CD45. Thus, the apparent catalytic velocity of CD45 is not regulated by IL2-induced phosphorylation. However, CD45 is reported to form a stable complex with the cytoskeleton, in particular with the non-erythroid spectrin, fodrin 1601. Phosphorylation may alter the affinity of CD45 for fodrin, or for other proteins that may be associated either with CD45 in the plasma membrane and/or serve as its physiologic substrates.

Eur. J. Immunol. 1991. 21: 913-919

There are examples of serinehhreonine phosphatases in which binding of a regulatory subunit alters phosphatase reactivity with substrates [61] and phosphorylation of the regulatory subunit leads to dissociation of the catalytic subunit [62]. Phosphorylation of CD45 in response to I L 2 can be expected to regulate its activity, albeit in a subtle manner not readily detected with an in virro activity measurement. Dissociation from intracellular proteins and redirection of its catalytic activity away from some substrates are potential ways for controlling CD45 activity. Demonstration of these mechanisms should clarify the regulatory role of CD45 phosphorylation. Received November 26, 1990.

5 References 1 Saltzmann, E. M., Thom, R. R. and Casnellie, J. E., J. Biol. Chem. 1988. 263: 695. 2 Ishii, T., Takeshita, T., Numata, N. and Sagamura, K., J. Immunol. 1980. 141: 174. 3 Evans, S. W. and Farrar, W. L., J. Biol. Chem. 1987. 262: 4624. 4 Ishii, T., Sagamura, K., Nakamura, M. asnd Hinuma, Y., Biochem. Biophys. Res. Commun. 1986. 135: 487. 5 Brautigan, D. L., Bornstein, P. and Gallis, B., J. Biol. Chem. 1981. 256: 6519. 6 Gallis, B., Bornstein, €! and Brautigan, D. L., Proc. Natl. Acad. Sci. USA 1981. 78: 6689. 7 Swarup, G., Cohen, S. and Garbers, D. C., Biochem. Biophys. Res. Commun. 1982. 107: 1104. 8 Horlein, D., Gallis, B., Brautigan, D. L. and Bornstein, €!, Biochemistry 1982. 21: 5577. 9 Sparks, J. W. and Brautigan, D. L., J. Biol. Chem. 1985. 260: 2042. 10 Roome, J., OHare, T., Pilch, P. F. and Brautigan, D. L., Biochem. J. 1988. 256: 493. 11 Lau, K.-H. W., Farley, J. R. and Baylink, D. J., Biochem. J. 1989. 257: 23. 12 Tonks, N. K., Diltz, C. D. and Fischer, E. H., J. Biol. Chem. 1988. 263: 6722. 13 Tonks, N. K., Diltz, C. D. and Fischer, E. H., J. Biol. Chem. 1988. 263: 6731. 14 Charbonneau, H., Tonks, N. K., Walsh, K. A. and Fischer, E. H., Proc. Natl. Acad. Sci. USA 1988. 85: 7182. 15 Tonks, N. K., Charbonneau, H., Diltz, C. D., Fischer, E. H. and Walsh, K. A., Biochemistry 1988. 27: 8695. 16 Charbonneau, H., Tonks, N. K., Kumar, S., Diltz, C. D., Harrylock, M., Cool, D. E., Krebs, E. G., Fischer, E. H. and Walsh, K. A., Proc. Natl. Acad. Sci. USA 1989. 86: 5252. 17 Cool, D. E., Tonks, N. K., Charbonneau, H., Walsh, K. A., Fischer, E. H. and Krebs, E. G., Proc. Natl. Acad. Sci. USA 1989. 86: 5257. 18 Trowbridge, I. S., Ralph, P. and Bevan, M., Proc. Natl. Acad. Sci. USA 1975. 72: 157. 19 Omary, M. B.,Trowbridge, I. S. and Battifora, H. A., J. Exp. Med. 1980. 152: 842. 20 Raschke, N. C., Proc. Natl. Acad. Sci. USA 1987. 84: 161. 21 Saga,Y.,Tung,J.-S., Shen, F.-W. and Boyse, E. A., Proc. Natl. Acad. Sci. USA 1986. 83: 6940. 22 Streuli, M., Hall, L. R., Saga,Y., Schlossman, S. F. and Saito, H., J. Exp. Med. 1987. 166: 1548. 23 Hall, L. R., Streuli, M., Schlossman, S. F. and Saito, H., J. Immunol. 1988. 140: 2781. 24 Harp, J. A., Davis, B. S. and Ewald, S. J., J. Immunol. 1984. 133: 10.

IL 2-induced serine phosphorylation of CD45

919

25 Bemabeu, C., Carrera, A. C., De Lanazuri, M. 0. and SBnchez-Madrid, F., Eur. J. Immunol. 1987. 17: 1461. 26 Mittler, R. S., Greenfield, R. S., Schachter, B. Z., Richard, N. E and Hoffman, M. K., J. Immunol. 1987. 138: 3159. 27 Ledbetter, J. A., Rose, L. M., Spooner, C. E., Beatty, P. G., Martin, I? J. and Clark, E. A., J. Immunol. 1985. 135: 1819. 28 Kiener, F? A. and Mittler, R. S., J. Immunol. 1989. 143: 22. 29 Ledbetter, J. A.,Tonks, N. K., Fischer, E. H. and Clark, E. A,, Proc. Natl. Acad. Sci. USA 1988. 85: 8628. 30 Mustelin,T., Coggeshall, K. M. and Altman, A., Proc. Natl. Acad. Sci. USA 1989. 86: 6302. 31 Ostergaard, H. L., Shackelford, D. A. K., Hurley, T. R., Johnson, P., Hyman, R., Sefton, B. M. and Trowbridge, I. S . , Proc. Natl. Acad. Sci. USA 1989. 86: 8959. 32 Clark, E. A. and Ledbetter, J. A., Immunol. Today 1989. 10: 225. 33 Hunter, T., Cell 1989. 58: 1013. 34 Thomas, M. L., Annu. Rev. Immunol. 1989. 7: 339. 35 Omary, M. B. and Trowbridge, I. S., J. Biol. Chem. 1980.255: 1662. 36 Watson, A. J., Immunogenetics 1982. 16: 485. 37 Gillis, S. and Smith, K. A., Nature 1977. 268: 154. 38 Cerottini, J. C., Engers, H. D., MacDonald, H. R. and Brunner, K. T., J. Exp. Med. 1974. 140: 703. 39 Acres, R. B., Conlon, P. J., Mochizuki, D.Y. and Gallis, B., J. Biol. Chem. 1986. 261: 16210. 40 Gallis, B., Prickett, K. S., Jackson, J., Slack, J., Schooley, K., Sims, J. E. and Dower, S. K., J. Immunol. 1989. 143: 3235. 41 Ledbetter, J. A. and Herzenberg, L. A., Immunol. Rev. 1979. 47: 63. 42 Kamps, M. P. and Sefton, B. M., Oncogene 1988. 2: 305. 43 Valentine, M. A., Meier, K. E., Rossie, S. and Clark, E. A., J. Biol. Chem. 1989. 264: 11282. 44 Kamps, M. P. and Sefton, B. M., Anal. Biochem. 1989. 176: 22. 45 Cooper, J. A., Sefton, B. M. and Hunter,T., Methods Enzymol. 1983. 99: 387. 46 Saltzman, E. M., Lukowsky, S. M. and Casnellie, J. E., J. Biol. Chem. 1989. 264: 19979. 47 Farrar, W. L. and Ferris, D. K., J. Biol. Chem. 1989. 264: 12 562. 48 Ferris, D. K.,Willette-Brown, J., Ortaldo, J. R. and Farrar,W. L., J. Immunol. 1989. 143: 870. 49 Hatakeyama, M.,Tsudo, M., Minamoto, S., Kono,T., Doi,T., Miyata,T., Miyasaka, M. and Taniguchi,T., Science 1989.244: 551. 50 Farrar, W. L. and Anderson, W. B., Nature 1985. 315: 233. 51 Friedrich, B. and Gullberg, M., Eur. J. Immunol. 1988. 18: 489. 52 Mills, G. B., Girard, I?, Grinstein, S. and Gelfand, E.W., Cell 1988. 55: 91. 53 Valge,V. E., Wong, J. G. P., Datlof, B. M., Sinskey, A. J. and Rao, A., Cell 1988. 55: 101. 54 Friedrich, C. B., Cantrell, D. A. and Gullberg, M., Eur. J. Biochem. 1989. 19: 111. 55 Shackelford, D. A. andTrowbridge, I. S., J. Biol. Chem. 1986. 261: 8334. 56 Olivier, A. R., Ballou, L. M. and Thomas, G., Proc. Natl. Acad. Sci. USA 1988. 85: 4720. 57 Sturgill,T.W., Ray, L. B., Erickson, E. and Maller, J. L., Nature 1988. 334: 715. 58 Lacal, I?, Pulido, R., SBnchez-Madrid,F. and Mollinedo, F., J. Biol. Chem. 1988. 263: 9946. 59 Gilliland, L. K., Schieven, G. L., Grosmaire, L. S., Damle, N. K. and Ledbetter, J. A,, Tissue Antigens 1990. 35: 128. 60 Bourguignon, L. Y. W., Suchard, S. J., Nagpal, M. L. and Glennei J. R., J. Cell. Biol. 1985. 101: 477.-61 Chen, S. C., Kramer, G. and Hardesty, B., J. Biol. Chem. 1989. 264: 7267. 62 Hiraga, A. and Cohen, I?, Eur. J. Biochem. 1986. 161: 763.