Eur. J. Immunol. 1990. 20: 1655-1660
Akira Yamadaa, Michel Streuli, Haruo Saito, David M. Rothstein, Stuart F. Schlossman and Chikao Morimoto Division of Tumor Immunology, Dana-Farber Cancer Institute, Harvard Medical School, Boston
PKC activation and CD45 expression
1655
Effect of activation of protein kinase C on CD45 isoform expression and CD45 protein tyrosine phosphatase activity in T cells* The T200Aeukocyte common antigen (CD45) is a family of at least five large-molecular weight glycoproteins, which are differentially expressed on Tcell subsets. The CD45 antigen consists of a variable heavily glycosylated exterior domain, a single membrane-spanning region, and a large cytoplasmic domain that has protein tyrosine phosphatase (PTPase) activity. In this study,we examined the effects of activation of protein kinase C (PKC) on the phosphorylation and expression of CD45 isoforms and PTPase activity in human T cells. After activation of PKC by phorbol 12-myristate 13-acetate (PMA), CD45RA expression rapidly increased within the first 24 h, whereas CD45RO expression did not change within this time. However by 48 h, expression of CD45RO also began to increase. Metabolic labeling showed that the rapid increment in CD45RA expression observed after PMA stimulation is primarily due to increased de novo synthesis of the 205-kDa and not the 220-kDa molecule. PMA treatment resulted in the phosphorylation of each CD45 isoform to a degree corresponding to its relative surface expression. Significantly,we found that the phosphorylation of CD45 by PKC activation down-regulated CD45 PTPase activity.
CD45 cDNA [2, 31. These isoforms have the same 707amino acid cytoplasmic domain, 22-amino acid transmemThe T200/leukocyte common antigen (LCA; CD45) is a brane region, 23-amino acid signal peptide and 8-amino family of lymphocyte cell surface glycoproteins ranging in acid amino-proximal sequence, but differ from one another molecular mass from 180 kDa to 220 kDa that is abundantly in that they contain extracellular domains of 552, 504,486, expressed on lymphocytes and other hematopoietic cells 438, or 391 amino acids depending on which combination of [l]. The five different CD45 isoforms are generated by the alternatively spliced exons A, B and C are used. alternative splicing of three exons of a single gene, and Although the functional role of CD45 is unclear, studies distinct isoforms are differentially expressed on T cell using anti-CD45 antibodies have implicated these molesubsets [2-41. For example, anti-2H4 and other CD45RA cules in several immunologic functions including, the mAb recognize the two highest molecular mass isoforms of blocking of the induction of suppressor activity [9-111 and CD45 (220 kDa and 205 kDa) and define a subset of CD4 the inhibition of cytotoxic activity and NK function cells (CD4+CD45RA+) which function as inducers of [ 12-15]. Additionally, anti-CD45 antibodies have been CD8+ suppressor cells [4,5]. In contrast, the mAb UCHLl reported to augment T cell proliferation by PHA and (CD45RO) binds to the lowest molecular mass isoform of autologous MLR (AMLR) [16-181. Ledbetter et al. [19] CD45 (180 kDa) and defines a subset of CD4 cells have suggested a regulatory role for CD45 based on their (CD4+CD45RO+) which provides maximal B cell helper observation that heterologous antibody-mediated crossfunction and responds maximally to recall antigens such as linking of CD3 and CD45 on the surface of T cells prevents tetanus toxoid [6, 71. In addition, these markers are anti-CD3 from transducing an activation signal. Recently, believed by many t o delineate maturational status, with the Charbonneau et al. [20] have shown that the two homoloCD45RA subset containing “naive cells” and the CD45RO gous domains of the intracytoplasmic segment of CD45 share homology with a soluble protein tyrosine phosphasubset containing “memory cells” [8]. tase (PTPase) isolated from human placenta. The tyrosine The primary structures of five human CD45 isoforms have phosphatase activity of CD45 was subsequently demonbeen predicted based on the nucleotide sequences of cloned strated by Tonks et al. [21].Thus, CD45 may be considered the prototype of a family of transmembrane receptors which functions in a specific signal transduction pathway via tyrosine dephosphorylation. This dephosphorylation pathway might be directly linked to the CD4--associated [I 81891 tyrosine kinase p56ICk,providing a possible mechanism for * This work was supported by NIH grants AI-12069, AI-26598 and the regulatory role of CD45 in CD3-CD4 cross-linking and AR-33713. activation.
1 Introduction
A
Present address: Department of Immunology, Kurume University School of Medicine, Kurume, Fukuoka 830, Japan.
Correspondence: Chikao Morimoto, Division of Tumor Immunology, Dana-Farber Cancer Institute, 44 Binney St, Boston, MA 02115. USA Abbreviations: PTPase: Protein tyrosine phosphatase AMLR: Autologous MLR 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
Tumor-promoting phorbol esters such as PMA serve as poorly hydrolyzable diacetylglycerol analogues that can permeate biological membranes and induce sustained activation of PKC [22, 231. CD4, CD8, p561ck as well as CD45, all undergo serine phosphorylation when lymphoid cells are treated with PMA [24-261. Although activation of PKC has emerged as a major common signal-transducing 0014-2980/90/0808-1655$3.50+ .25/0
1656
Eur. J. Immunol. 1990.20: 1655-1660
A. Yamada, M. Streuli, H. Saito et al.
mechanism for Tcell activation [27], the functional consequence of the phosphorylation induced by PKC activation is not well understood. Since CD45 is known to be a substrate for PKC [28], we have studied the effects of the activation of PKC on CD45 isoform expression and PTPase activity.
2 Materials and methods 2.1 Preparation of cells Human PBMC were isolated from leukocyte-enriched preparations from healthy volunteer donors by FicollHypaque (Pharmacia Fine Chemicals, Piscataway, NJ) density gradient centrifugation. The PBMC obtained were further separated into E rosette-positive (E+) and E rosette-negative (E-) populations with 5% SRBC as previously described [29]. E+ cells were used as T cells. These cells were routinely > 95% CD2+ and > 90% CD3+. 2.2 Antibodies
The mAb GAP 8.3 (anti-CD45) reacting with an epitope common to all LCA isoforms (180 kDa, 190 kDa, 200 kDa and 220 kDa; [30]) was secreted by the hybridoma clone obtained from the American Type Culture Collection (Rockville, MD). The mAb 2H4 (anti-CD45RA) (reacting with the 205-kDa and 220-kDa isoforms) was made as previously described [4].The mAb UCHLl (anti-CD45RO; recognizing the 180-kDa isoform; [7]) was obtained from Dr. I? Beverley, U.C.H. Medical School, London. Other mAb, including T3 (anti-CD3), T4 (anti-CD4), T8 (antiCD8), 4B4 (anti-CDw29) and anti-IL2R (Tac, CD25),were also used in this study. Their production and characterization have been described elsewhere [5, 29-31]. The IgG fraction of a polyclonal goat anti-mouse (IgG IgM) antibody and fluorescein-conjugated F(ab')z goat antimouse Ig were purchased from Tago, Burlingame, CA.
+
2.3 F C M analysis of lymphocyte populations
T cells were stained by indirect immunofluorescence using fluorescein-conjugated F(ab')z goat anti-mouse Ig and then analyzed on an EPICS V cell sorter (Coulter Electronics, Hialeah, FL) as previously described [32]. 2.4 Biosynthetic pulse labeling of cells
To analyze de novo protein synthesis, cells (5 x lo6cells/ml) were preincubated at least 1 h in methionine- and cysteinefree RF'MI 1640 medium supplemented with 1% dialyzed FCS, 20mM Hepes and 50 pg/ml of gentamycin (Cys-Met-RPMI). Cells were resuspended in this same medium at 1 x loh cells/ml and then stimulated with 10 ng/ml of PMA (Sigma, St. Louis, MO) at 37°C for various time periods. After incubation, the cells were labeled with 1mCi/ml [35S]methionine and of [35S]cysteine (DuPont/NEN, Boston, MA) for 20 min at 37°C. After washing, in order to allow for maturation of glycosylated CD45 molecules, cells were incubated at 37°C for an additional 60 min,with cysteine and methionine-rich (1 mM
each) RPMI 1640 supplemented with 10% FCS, and then the cells (5 X lo7)were lysed in 1 ml of lysis buffer A (20 mM Tris-HCI, pH 8.0, containing 1% NP40, 150 mM NaCI, 1 mM PMSF and 1 mM EDTA). The cell lysates were then used for further analysis by immunoprecipitation. In the case of 18-h treatment with PMA, the first incubation with PMA was performed using normal RPMI 1640 medium (containing cysteine and methionine) supplemented with 10% human AB serum. Two hours before the end of the incubation period, cells were washed twice with Cys-/ Met-RPMI and then incubated in this same medium as detailed above. 2.5 Phosphorylation study
For phosphorylation studies, the cells were washed three times with Hepes-buffered saline containing 135 mM NaCl, 5 mM KCI, 0.6 mM CaC12,2mM glucose and adjusted to 5 x lo6 cells/ml. After preincubation for 30 min at 37"C, carrier-free 32P-phosphatewas added to a final concentration of 250 pCi/ml= 9.25 MBq and the cells were incubated at 37°C for 3 h. After incubation, the cells were washed twice with Hepes-buffered saline and stimulated with PMA at 37 "C for 10 to 60 min. The cells (1 x lo7) were lysed in 1 ml of lysis (buffer B, 20 mM Tris-HCI, pH 8.0, 50 mM NaCl, 1% NP40,l mM PMSF, 1mM EDTA, 5 mM NaF, 10 mM sodium pyrophosphate) on ice. 2.6 Surface labeling of cells and immunoprecipitation assay Cells (1 x lo7) were labeled with 1 mCi Na1251(Amersham Int., Amersham, GB) by lactoperoxidase-catalyzediodination [33] and then lysed in 1 ml of the lysis buffer A. Cell lysates were precleared overnight at 4 "C with formalinfixed Staphylococcus aureus Cowan I organisms (Sigma). The precleared lysate was then incubated with 5-20 pg of purified anti-T200 (GAP 8.3) antibody at 4°C for 6 h. After incubation, 20 pl of protein A-Sepharose (50% suspension, Pharmacia, Uppsala, Sweden) was added to each tube and further incubated for 2-3 h at 4°C. Precipitates were then washed once in lysis buffer containing 0.5% NaCI, once in lysis buffer containing 0.5% deoxycholate and once in lysis buffer alone. Immunoprecipitates were solubilized with SDS-PAGE sample buffer and analyzed by SDS-PAGE on a 6% gel under reducing conditions. Gels were then fixed, dried and subjected to autoradiography using Kodak X-AR5 film and light intensifying screens (DuPont, Wilmington, DE). For densitometric analysis an LKB2222-010 Ultra Scan XL (LKB, Bromma, Sweden) was used. 2.7 PTPase assay
PTPase assays were performed as previously described [34]. Cells (either unstimulated, or after 30 min of stimulation with PMA) were washed with PBS and lysed in lysis buffer C (20 mM Tns-HC1, pH 8.0, 150 mM NaCl, 1% Triton X-100, 1 mM PMSF, 1 mM EDTA and 2 x Denhardt's solution) at a concentration of 2.5 X lo7 cells/ml. The lysates were mixed with appropriate preformed immune complex, and incubated at 4°C for 12 h. Immune complexes were sedimented by centrifugation, washed and
PKC activation and CD45 expression
Eur. J. Immunol. 1990. 20: 1655-1660
suspended in 50 p1 of PBS containing 2 x Denhardt's solution. The PTPase assay was performed at 37 "C for 60 min. The radioactivity released by 5 U of calf intestinal alkaline phosphatase (BFU, Gaithersburg, MD) was taken as 100% release. The [32P]Tyr-substratewas prepared as described previously [34], using the synthetic peptide Tyr(P) Raytide and tyrosine kinase (Oncogene Science, Manhasset, NY).
3 Results and discussion 3.1 The expression of CD45 isoforms on T cells after PMA treatment
After exposure of T cells to PMA, the expression of a number of cell surface molecules is modified. For example, the surface expression of both the CD3 and CD4 antigens is decreased after PMA treatment [24, 35-37], while expression of lymphokine receptors, such as the IL 2R (CD25), is increased [38]. However, the expression of CD45 isoforms on Tcells after PMA exposure is unknown. Therefore, to determine the effect of PMA on the expression of different CD45 isoforms, we cultured purified T cells from healthy donors in the presence of PMA (10 ng/ml). After various incubation periods, cells were harvested and CD45 isoforms were analyzed by FCM using 2H4(anti-CD45RA) and UCHLl (CD45RO), as well as GAP 8.3 Ab (antiCD45) which recognizes an epitope common to all CD45 isoforms. As shown inTable 1, the percentage of CD4+ cells markedly decreased within first 24 h of PMA treatment, but then attained its initial value. The percentage of CD8+ cells was slightly increased. The percentage of CD3+ cells and the density of CD3 antigen onTcells (data not shown) were also decreased by PMA treatment. On the other hand, IL 2R (Tac, p55) expression increased. As seen inTable 1, the common CD45 epitope GAP 8.3 was expressed on 98% of resting Tcells. After PMA treatment, the percent of positive cells remained stable, but a gradual
1657
Table 2. Effect of PMA on the expression of cell surface molecules on CD4 cellsa) Culture mAb period Anti-CD4SRA Anti-1L 2R Anti-CD4 (h) 0 1
2 IS 24 48
530/0 55% 57% 63% 66YO 70%
7% 8YO 8% 42% 15% 37%
94Y"
92YO 89% 73% 56% 71%
AntLCD8 3Yo 3% 3% 3YU 4 Yo 5 YU
a) Purified CD4 cells were cultured with 10 nglml PMA. At various times cells were stained by indirect immunofluorescence on an Epics V cell sorter. b) Data are a representative of two experiments performed.
increase in density of expression was observed (data not shown). The percentage of CD45RA+ cells rapidly increased after PMA treatment, with initial changes noted within 2-4 h. By 24 h, the expression of CD45RA had increased from 50% to 70%. The density of CD45RA expression remained essentially unchanged until 48 h. By 72 h, the majority of CD45RA+ cells expressed a high density of CD45RA with an increase in mean channel fluorescence from 78 to 104 (data not shown). In contrast, CD45R0, which is expressed in high amount on the reciprocal subset of resting T cells, did not change within the first 24 h after PMA treatment. However, by 48 h, the percentage of CD45RO+ cells also began to increase. Again, by 96 h of culture, both CD45RA and CD45RO expression were elevated by 29% and 24%, respectively. These phenomena were observed over a wide range of PMA concentrations, ranging between 1ng/ml to 100ng/ml (data not shown). In all of the following experiments, 10 ng/ml of PMA was used unless otherwise stated. To determine whether the increase in CD45RA antigen
-
-
Table 1. Effect of PMA on the expression of surface molecules on Tcellsa) Culture period (h 1
0
Anti-CD4SRA
so f
6h)
0 2
0 8* 2
4
16+ 4 1Y+ 7
8 24 18 72 96
20f
7 9 24 f 12 2 Y f 8 27+
Anti-CD4SRO
63 k S 0
-1 f 3 3 f S 4 f 3 2+s 11+1 21 f 2 24 f 1
Anti-CD4S
mAb Anti-CD3
Anti-CD1
Original percentage of positive cells 98 f 2 93 f 2 M-t_ 6 Increment in percentagc of positive cells') 0 n 0 I f 1 Of 1 -12+ 6 I f 1 -3s f 7 1f1 -58rf: 2 I f 1 -12f s -27 f 10 -4s 12 I f 1 -18f 6 1+1 -28f 9 -2 f s -12+ 8 0+1 3+ 3 1+1 0+ 1
+
AntLCD8
Anti-IL 2K
25 f 8
3.5 f 3
0 3f0 4 f l h f 3
7 f 3 Xf3 10 4 11 2 7
*
a) Tcells were cultured with 10 nglml of PMA. At various times after initiation of culture, cells were washed, stained by indirect immunofluorescence, and analyzed on an EPICS V cell sorter. b) Values show the mean f SD of the three different experiments. c) Increment in percentage of positive cells = (percentage of positive cells at indicated culture period) - (percentagc of positive cells at time 0). Values show the mean f SD of three different experiments.
1658
A.Yamada, M. Streuli, H. Saito et al.
Eur. J. Immunol. 1990. 20: 1655-1660
expression observed occurred among CD4 cells, CD8 cells or both, changes in CD45RA expression on CD4 cells after PMA treatment was examined. As shown in Table 2, after PMA treatment of CD4 cells, the percentage of CD45RA+ cells rapidly increased as seen in the unseparated T cell population. In addition, the CD4 molecule on CD4 cells was down-regulated after PMA treatment as described previously [24, 371. Moreover, an increase in CD45RA antigen expression is also observed on purified CD8 cells after PMA treatment (data not shown). Thus increase of CD45RA antigen expression was observed in both CD4 and CD8 T cell subsets. Mammalian cells are known to increase cell surface expression of various molecules by at least two different mechanisms. One mechanism is via de novo protein synthesis, as exemplified by the increased expression of the IL 2R receptor after cell activation [39]. The other mechanism is the release of pre-synthesized proteins. For example, the Mol antigen (CR3) on monocytes is stored in large quantities in the cytoplasm of resting cells [40]. To determine whether de novo protein synthesis was required for the rapid increment in CD45RA expression observed, cycloheximide, an inhibitor of de novo protein synthesis, was used.T cells pretreated with 20 PM of cycloheximide for 1 h prior to stimulation with PMA failed to increase their expression of CD45RA, CD45RO or CD25,8 h after PMA stimulation, as measured by FCM (data not shown). These results indicated that de novo protein synthesis is essential for increased expression of the CD45 isoforms after PMA treatment.
3.2 Effect of PMA on de novo synthesis of CD45 molecules To determine which of the two CD45 isoforms recognized by anti-CD45RA is synthesized de novo immediately after PMA stimulation, PMA-treated T cells were pulse labeled with [35S]methionineand [35S]cysteineand CD45 molecules were immunoprecipitated using anti-CD45 (CAP 8.3) mAb and analyzed by SDS-PAGE (Fig. 1A). The results of densitometric analysis of the autoradiogram shown in Fig. 1A are shown in Fig. 1B. A control CD45 immunoprecipitation from lZ5Isurface-labeled T cells is shown in lane a (Fig. 1A). In unstimulated T cells, the lower three CD45 isoforms (180 kDa, 190kDa and 205 kDa) are constitutively synthesized de novo, whereas the 220-kDa protein was not detected (lane b). After PMA treatment, the incorporation of ["Slamino acids into the CD45 180-kDa, 190-kDa and 205-kDa isoforms, but not 220-kDa CD45 isoforms, rapidly increased, reaching a maximum after 3 h (lane d). By 18 h of PMA stimulation (lane f), de novo synthesis of each protein returned to the same level as that seen in resting T cells (lane b). Even though the 180-kDa, 190-kDa and 205-kDa CD45 isoforms were being maximally synthesized at 3 h, de novo synthesis of the 220-kDa protein could still not be detected (Fig. 2 b, lane d).These results suggest that the rapid increment in CD45RA expression observed after PMA stimulation is primarily due to increased de novo synthesis of the 205-kDa and not the 220-kDa CD45 molecule. This finding is in general agreement with results obtained from CD4+ and CD8+ clones [41] and long-term CD4+ cell lines expressing CD45RA [42], in which the 205-kDa protein is present while the 220-kDa protein has disappeared.
Figure I. Effect of PMA on de n o w synthesis of CD45 molecules. T cells were cultured in cysteine/methionine-free medium with 10 nglrnl of PMA. After various incubation times, cells were pulse labeled with [3sS]cysteine and [3sS]methioninefor 20 min. Cells were then chased in complete medium for 1 h at 37°C and lysed. Lysates were immunoprecipitated with mAb GAP 8.3 and subjected to SDS-PAGE (6% acrylamide). Autoradiograph (A) and its densitometric tracings (B) of a resultant immunoprecipitation are shown. Lane (a) shows control GAP 8.3 immunoprecipitation from 1251-surface-labeled Tcells, lanes (b-f) from 3sSpulse-labeled Tcells. Lane (b): unstimulated Tcells; lane (c):Tcells stimulated with PMA for 1 h; lane (d): stimulated with PMA for 3 h; lane (e): stimulated with PMA for 6 h; lane (f): stimulated with PMA for 18 h. Arrows indicate the positions of the 220-kDa, 205-kDa, 190-kDa and 180-kDa CD45 isoforms.
3.3 Effect of PMA on CD45 phophorylation and CD45-linked PTPase activity
CD45 molecules are known to be substrates of PKC both in vivo [25] and in vitro [28].The common cytoplasmic portion of CD45 molecules has two putative PKC phosphorylation sites [20]. Although it has been reported that PMA stimulation increases the phosphorylation of the CD45 antigen [25], it is not clear whether all CD45 isoforms are equally phosphorylated. In order to establish which CD45 isoforms are phosphorylated following PMA stimulation, we incubated T cells with carrier-free 32P-phosphate and then stimulated with 100ng/ml of PMA. After stimulation, cells were lysed and the incorporation of 32P into CD45 isoforms was analyzed by immunoprecipitation using antiCD45 mAb GAP8.3 (Fig. 2 A). In unstimulated Tcells, all CD45 isoforms are constitutively but weakly phosphorylated (Fig. 2 A, lanes b, d and f). Densitometric analysis of the autoradiogram (Fig. 2 A) is shown in Fig. 2 B. After a 10-min stimulation with PMA, all CD45 isoforms, but particularly the lower-molecular mass isoforms (180-kDa and 190-kDa proteins) were phosphorylated (Fig. 2 A, B, lane c). Increased phosphorylation of all CD45 isoforms is also seen after 30 or 60 min of PMA stimulation (Fig. 2 A and B, lanes e and g). Furthermore, 32Pincorporation into each CD45 isoform seems to correspond to the density of expression as determined by 1251-surfacelabeling (Fig. 2 B, lane a). However, the relationship between the phospho-
Eur. J. Immunol. 1990. 20: 1655-1660
PKC activation and CD45 expression
1659
(A)
rylation of Ser/Thr residues of CD45 molecules and changes in surface expression of these molecules after PMA activation is unclear. Phosphorylation of the CD45 molecules following PKC activation may result in a change in the surface expression of these molecules and/or an alteration of CD45 function. For example, phosphorylation could alter the binding affinity of CD45 for its putative ligand or change the CD45 PTPase activity. In this regard, we have previously reported that the density of CD45RA antigen expression is increased after 2 days of Con A or 7 days AMLR activation, respectively, and that the level of CD45RA antigen expression correlates with suppressor function of Con A and AMLR-activated T cells [9-111. Furthermore, the anti-CD45RA (2H4) antibody modulates the suppressor/inducer and suppressor/effector function of CD4+ and CD8+ cells [9-11].2H4 and other anti-CD45RA antibodies also augment T cell proliferation by PHA and AMLR [16-181. These findings suggest that the CD45 molecules are functionally important molecules that play an important role in modifying T cell activation signals. Therefore, experiments were next performed to determine whether the phosphorylation of the CD45 molecules following PMA treatment altered the PTPase activity of the cytoplasmic domains of CD45 antigen. CD45 antigens were partially purified by immunoprecipitation from cell lysates using an anti-CD45 antibody (GAP 8.3) and the level of PTPase activity in the immunoprecipitates was determined. As shown in Table 2, the CD45 molecules precipitated with mAb GAP 8.3 from resting T cells released 23.3% and 15.2% of the phosphate from a phosphotyrosine substrate in two experiments (Table 3). As a control, class I MHC molecules were precipitated with the W6/32 mAb, and demonstrated no PTPase activity (no release of free phosphate). Under the same conditions, the PTPase activity of CD45 molecules from PMA-activated T cells (activated with 100 ng/ml of PMA for 30 min) was decreased by almost 50% relative to that of restingTcells (from 23.3% to 12.5% and from 15.2% to 9.7%, respectively).
Figure 2. PMA-induced phosphorylation of CD45 molecules. 32P-phosphate-loadedTcells were stimulated with or without 100 ng/ml of PMA at 37 "C.After various incubation periods, cells were washed, lysed, and then the lysates were subjected to SDS-PAGE on 6% acylamide gels. Autoradiograph (A) and its densitometric tracings (B) of a resultant immunoprecipitation assay are shown. Lane (a) shows control GAP 8.3 immunoprecipitation from 1251-surface-labeledT cells, lanes (b-g), show GAP 8.3 immunoprecipitation from 32P-phosphate-loadedTcells. Lanes (b, d and f) are fromTcells incubated without PMA for 10rnin (lane b), 30 min (lane d) or 60 rnin (lane f ) . Lanes (c, e and g) are from Tcells stimulated with PMA for 10 rnin (lane c), 30 rnin (lane e) or 60 min (lane g).
These results suggest that the activation of PKC downregulates CD45 PTPase activity. However, it is unclear whether this inhibition of CD45-linked PTPase activity by PKC would up- or down-regulate T cell activation. For example, the association of p56ICkwith the cytoplasmic portion of both CD4 and CD8 has been reported [43-451. p561Ckcontains a tyrosine residue at position 505 which is highly phosphorylated in vivo [46]. Since a mutation of this
tyrosine residue to phenylalanine increases tyrosine kinase activity [47,48] (analogous tov-src), one might suspect that dephosphorylation of TyrSoSshould increase tyrosine kinase activity. Indeed, isolated CD45 can directly activate p56ICk in vitro [49] and moreover, the intracellular domain of CD45 has recently been shown t o dephosphorylate p56ICkat the regulatory TySo5 site in an in vitro transfection study
18h (lane 1 )
0.3
Relative Distance
Table 3. Effect of PMA on CD45-linked PTPase activitya)
Exp.
1
Lysate from Tcells treated withb)
Immune complex
Untreated
W6l32 GAP8.3 W6/32 GAP8.3 Alkaline phosphatased)
280 20 560 174 11046 88 103
0.3 23.3 0.2 12.5 100
W6132 GAP8.3 W6l32 GAP8.3 Alkaline phosphatase
235 18566 27 11975 122498
0.2 15.2 0.0 9.7 100
PMA -
2
Untreated PMA -
Free phosphate cDrnC)
released (YO
a) PTPase assay is described in Sect. 2.7. b) Tcells were incubated with or without PMA (100 nglml) for 30 min at 37 "C. After incubation, cells were washed and lysed. For each assay, a lysate made from 2.5 X lo7 Tcells was immunoprecipitated. c) cpm of the blank sample (200-500 cpm) was already subtracted. Values show the mean cpm of triplicate assay. d) Alkaline phosphatase was used for determination of 100% release value.
1660
A.Yamada, M. Streuli, H. Saito et al.
[50].These results suggest that CD45-linked PTPase could either up- or down-regulate T cell activation via tyrosine dephosphorylation. Further studies including identification of the physiologic substrates for this PTPase or the physiologic ligands for the different extracellular domains of CD45 isoforms are required to clarify these points. The authors would like to thank Mr. Herb Levine and Mr. John Daley for technical assistance in flow cytometric analysis. We would also like to thank Ms. Gayathri Ranganathan for her excellent secretarial assistance. Received December 15, 1989; in revised form March 22, 1990.
4 References 1 Thomas, M. L., Annu. Rev. Immunol. 1989. 7: 399. 2 Streuli, M., Hall, L. R., Saga,Y., Schlossman, S. F. and Saito, H., J. Exp. Med. 1987. 166: 1548. 3 Ralph, S. J.,Thomas, M. L., Morton, C. C. and Trowbridge, I. S., EMBO J. 1987. 6: 1251. 4 Rudd, C. E., Morimoto, C. ,Wong, L. L. and Schlossman, S. F., J. Exp. Med. 1987. 166: 1758. 5 Morimoto. C., Letvin. N. L.. Distaso. J. A.. A1drich.W. R. and Schlossman, S. F., J. Immunol. 1985. 134: 1508. 6 Smith, S. H., Brown, M. H., Rowe, D., Callard, R. E. and Beverley, F? C. L., Immunology 1986. 58: 63. 7 Terry, L. A., Brown, M. H. and Beverley, F? C. L., Immunology 1988. 64: 331. 8 Tedder, T. F., Clement, L. T. and Cooper, M. D., J. Immunol. 1985. 134: 2983. 9 Takeuchi, T., Schlossman, S. F. and Morimoto, C., Cell. Immunol. 1987. 107: 107. 10 Takeuchi,T., Rudd, C. E., Schlossman, S. F. and Morimoto, C., Eur. J. Immunol. 1987. 17: 97. 11 Morimoto, C., Letvin, N. L., Rudd, C. E., Hagan, M., Takeuchi, T. and Schlossman, S. F., J. Immunol. 1986. 137: 3247. 12 Lefrancois, L. and Bevan, M. J., Nature 1985. 315: 449. 13 Nakayama, E . , Shiku, H., Stockart, E., Oettgen, H. F. and Old, L. J., Proc. Natl. Acad. Sci. USA 1979. 76: 1977. 14 Seaman, W. E.. Talal, N., Herzenberg, L. A. and Ledbetter, J. A., J. Immunol. 1981. 127: 982. 15 Newman, W., Proc. Natl. Acad. Sci. USA 1982. 79: 3858. 16 Cobold, S . , Hall, G. and Waldmann, H. in McMichael, A. J. et al. (Eds.) Leukocyte Typing 111, Oxford University Press, Oxford 1987, p. 788. 17 Ledbetter, J. A., Rose, L. M., Spooner, C. E., Beatty, F? G., Martin, F? J. and Clark, E. A., J. Immunol. 1985. 135: 1819. 18 Martorell, J.,Vilella, R., Borche, L., Rojo, I. and Vives, J., Eur. J. Immunol. 1987. 17: 1447. 19 Ledbetter, J. A.,Tonks, N. K., Fischer, E. H. and Clark, E. A., Proc. Natl. Acad. Sci. USA 1988. 85: 8628. 20 Charbonneau, H.,Tonks, N. K., Walsh, K. A. and Fischer, E. H., Proc. Natl. Acad. Sci. USA 1988. 85: 7182. 21 Tonks, N. K., Charbonneau, H., Diltz, C. D., Fischer, E. H. and Walsh, K. A., Biochemistry 1988. 27: 8695. 22 Nishizuka, Y , Nature 1984. 308: 603.
Eur. J. Immunol. 1990. 20: 1655-1660
23 Kraft, A., Anderson, W., Cooper, L. and Sando, J. J., J. Biol. Chem. 1982. 257: 13198. 24 Blue, M. L., Hafler, D. A., Craig, K. A., Levine, H. and Schlossman, S. F., J. Immunol. 1987. 139: 3949. 25 Autero, M. and Gahmberg, C. G., Eur. J. Immunol. 1987.17: 1503. 26 Veillette, A., Horak, I. D. and Bolen, J. B., Oncogene Res. 1988. 2: 385. 27 Alcover, A., Ramarli, D., Richardson, N. E., Chang, H. C. and Reinherz, E. L., Immunol. Rev. 1987. 95: 5. 28 Shackelford, D. A . and Trowbridge, I.S., J. Biol. Chem. 1986. 261: 8334. 29 Morimoto, C., Letvin, N. L., Boyed, A.W., Hagan, M., Brown, H. M., Kornacki, M. A. and Schlossman, S. F., J. Immunol. 1985. 134: 3762. 30 Berger, A. E., Davis, J. E. and Cresswell, F?, Human Immunol. 1981. 3: 231. 31 Reinherz, E. L. and Schlossman, S. F., Cell 1980. 19: 821. 32 Matsuyama,T.,Yamada, A., Kay, J.,Yamada, K. M., Akiyama, S. K., Schlossman, S. F. and Morimoto, C., J. Exp. Med. 1989. 170: 1133. 33 Hubbard, A. L. and Cohn, Z. A. in Maddy, A . H. (Ed.) Biochemical analysis of membranes. Champman and Hall, London 1976, p. 427. 34 Streuli, M., Krueger, N. X.,Tsai, A.Y. M. and Saito, H., Proc. Natl. Acad. Sci. USA 1989. 86: 8698. 35 Cantrell, D. A., Davis, A. A. and Crumpton, M. J., Proc. Natl. Acad. Sci. USA 1985. 82: 8158. 36 Solbach, W., J. Exp. Med. 1982. 156: 1250. 37 Hoxie, J. A., Matthews, D. M. and Callahan, K. J., J. Immunol. 1986. 137: 1194. 38 Shackelford, D. A. and Trowbridge, 1. S., J. Biol. Chern. 1984. 259: 11706. 39 Smith, K. A., Adv. Immunol. 1988. 42: 165. 40 Springer,T. A., Dustin, M. L., Kishimoto,T. K. and Marlin, S. D., Annu. Rev. Immunol. 1987. 5: 223. 41 Brod, S. A., Rudd, C. E., Purvee, M. and Hafler, D. A., J. Exp. Med. in press. 42 Rothstein, D. M., Sohen, S., Daley, J. F., Schlossman, S. F. and Morimoto, C., Cell Immunol. 1990, in press. 43 Rudd, C. E. ,Trevillyan, J. M., Dasgupta, J. D. ,Wong, L. L. and Schlossman, S. F., Proc. Natl. Acad. Sci. USA 1988. 85: 5190. 44 Barber, E. K., Dasgupta, J. D., Schlossman, S. F. and Trevillyan, J. M., Proc. Natl. Acad. Sci. USA 1989. 86: 3277. 45 Veillette, A., Bookman. M. A., Horak, E. M. and Bolen, J. B., Cell 1988. 55: 301. 46 Veillette, A., Horak, I. D., Horak, E. M., Bookman, M. A. and Bolen, J. B., Mol. Cell. Biol. 1988. 8: 4353. 47 Marth, J. D., Cooper, J. A., King, C. S., Ziegler, S. F.,Tinker, D. A., Overall, R.W., Krebs, E. G. and Perlmutter, R., Mol. Cell. Biol. 1988. 8: 540. 48 Amrein, K. E. and Sefton, B. M., Proc. Natl. Acad. Sci. USA 1988. 85: 4247. 49 Mustelin, T., Coggeshall, K. M. and Altman, A., Proc. Natl. Acad. Sci. USA 1989. 86: 6302. 50 Ostergaard, H., Schackelford, D. A., Hurley, T. R., Johnson, F?, Hyman, R., Safton, B. M. and Trowbridge, I. S . , Proc. Natl. Acad. Sci. USA 1989. 86: 8959.
Akira Yamadaa, Michel Streuli, Haruo Saito, David M. Rothstein, Stuart F. Schlossman and Chikao Morimoto Division of Tumor Immunology, Dana-Farber Cancer Institute, Harvard Medical School, Boston
PKC activation and CD45 expression
1655
Effect of activation of protein kinase C on CD45 isoform expression and CD45 protein tyrosine phosphatase activity in T cells* The T200Aeukocyte common antigen (CD45) is a family of at least five large-molecular weight glycoproteins, which are differentially expressed on Tcell subsets. The CD45 antigen consists of a variable heavily glycosylated exterior domain, a single membrane-spanning region, and a large cytoplasmic domain that has protein tyrosine phosphatase (PTPase) activity. In this study,we examined the effects of activation of protein kinase C (PKC) on the phosphorylation and expression of CD45 isoforms and PTPase activity in human T cells. After activation of PKC by phorbol 12-myristate 13-acetate (PMA), CD45RA expression rapidly increased within the first 24 h, whereas CD45RO expression did not change within this time. However by 48 h, expression of CD45RO also began to increase. Metabolic labeling showed that the rapid increment in CD45RA expression observed after PMA stimulation is primarily due to increased de novo synthesis of the 205-kDa and not the 220-kDa molecule. PMA treatment resulted in the phosphorylation of each CD45 isoform to a degree corresponding to its relative surface expression. Significantly,we found that the phosphorylation of CD45 by PKC activation down-regulated CD45 PTPase activity.
CD45 cDNA [2, 31. These isoforms have the same 707amino acid cytoplasmic domain, 22-amino acid transmemThe T200/leukocyte common antigen (LCA; CD45) is a brane region, 23-amino acid signal peptide and 8-amino family of lymphocyte cell surface glycoproteins ranging in acid amino-proximal sequence, but differ from one another molecular mass from 180 kDa to 220 kDa that is abundantly in that they contain extracellular domains of 552, 504,486, expressed on lymphocytes and other hematopoietic cells 438, or 391 amino acids depending on which combination of [l]. The five different CD45 isoforms are generated by the alternatively spliced exons A, B and C are used. alternative splicing of three exons of a single gene, and Although the functional role of CD45 is unclear, studies distinct isoforms are differentially expressed on T cell using anti-CD45 antibodies have implicated these molesubsets [2-41. For example, anti-2H4 and other CD45RA cules in several immunologic functions including, the mAb recognize the two highest molecular mass isoforms of blocking of the induction of suppressor activity [9-111 and CD45 (220 kDa and 205 kDa) and define a subset of CD4 the inhibition of cytotoxic activity and NK function cells (CD4+CD45RA+) which function as inducers of [ 12-15]. Additionally, anti-CD45 antibodies have been CD8+ suppressor cells [4,5]. In contrast, the mAb UCHLl reported to augment T cell proliferation by PHA and (CD45RO) binds to the lowest molecular mass isoform of autologous MLR (AMLR) [16-181. Ledbetter et al. [19] CD45 (180 kDa) and defines a subset of CD4 cells have suggested a regulatory role for CD45 based on their (CD4+CD45RO+) which provides maximal B cell helper observation that heterologous antibody-mediated crossfunction and responds maximally to recall antigens such as linking of CD3 and CD45 on the surface of T cells prevents tetanus toxoid [6, 71. In addition, these markers are anti-CD3 from transducing an activation signal. Recently, believed by many t o delineate maturational status, with the Charbonneau et al. [20] have shown that the two homoloCD45RA subset containing “naive cells” and the CD45RO gous domains of the intracytoplasmic segment of CD45 share homology with a soluble protein tyrosine phosphasubset containing “memory cells” [8]. tase (PTPase) isolated from human placenta. The tyrosine The primary structures of five human CD45 isoforms have phosphatase activity of CD45 was subsequently demonbeen predicted based on the nucleotide sequences of cloned strated by Tonks et al. [21].Thus, CD45 may be considered the prototype of a family of transmembrane receptors which functions in a specific signal transduction pathway via tyrosine dephosphorylation. This dephosphorylation pathway might be directly linked to the CD4--associated [I 81891 tyrosine kinase p56ICk,providing a possible mechanism for * This work was supported by NIH grants AI-12069, AI-26598 and the regulatory role of CD45 in CD3-CD4 cross-linking and AR-33713. activation.
1 Introduction
A
Present address: Department of Immunology, Kurume University School of Medicine, Kurume, Fukuoka 830, Japan.
Correspondence: Chikao Morimoto, Division of Tumor Immunology, Dana-Farber Cancer Institute, 44 Binney St, Boston, MA 02115. USA Abbreviations: PTPase: Protein tyrosine phosphatase AMLR: Autologous MLR 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
Tumor-promoting phorbol esters such as PMA serve as poorly hydrolyzable diacetylglycerol analogues that can permeate biological membranes and induce sustained activation of PKC [22, 231. CD4, CD8, p561ck as well as CD45, all undergo serine phosphorylation when lymphoid cells are treated with PMA [24-261. Although activation of PKC has emerged as a major common signal-transducing 0014-2980/90/0808-1655$3.50+ .25/0
1656
Eur. J. Immunol. 1990.20: 1655-1660
A. Yamada, M. Streuli, H. Saito et al.
mechanism for Tcell activation [27], the functional consequence of the phosphorylation induced by PKC activation is not well understood. Since CD45 is known to be a substrate for PKC [28], we have studied the effects of the activation of PKC on CD45 isoform expression and PTPase activity.
2 Materials and methods 2.1 Preparation of cells Human PBMC were isolated from leukocyte-enriched preparations from healthy volunteer donors by FicollHypaque (Pharmacia Fine Chemicals, Piscataway, NJ) density gradient centrifugation. The PBMC obtained were further separated into E rosette-positive (E+) and E rosette-negative (E-) populations with 5% SRBC as previously described [29]. E+ cells were used as T cells. These cells were routinely > 95% CD2+ and > 90% CD3+. 2.2 Antibodies
The mAb GAP 8.3 (anti-CD45) reacting with an epitope common to all LCA isoforms (180 kDa, 190 kDa, 200 kDa and 220 kDa; [30]) was secreted by the hybridoma clone obtained from the American Type Culture Collection (Rockville, MD). The mAb 2H4 (anti-CD45RA) (reacting with the 205-kDa and 220-kDa isoforms) was made as previously described [4].The mAb UCHLl (anti-CD45RO; recognizing the 180-kDa isoform; [7]) was obtained from Dr. I? Beverley, U.C.H. Medical School, London. Other mAb, including T3 (anti-CD3), T4 (anti-CD4), T8 (antiCD8), 4B4 (anti-CDw29) and anti-IL2R (Tac, CD25),were also used in this study. Their production and characterization have been described elsewhere [5, 29-31]. The IgG fraction of a polyclonal goat anti-mouse (IgG IgM) antibody and fluorescein-conjugated F(ab')z goat antimouse Ig were purchased from Tago, Burlingame, CA.
+
2.3 F C M analysis of lymphocyte populations
T cells were stained by indirect immunofluorescence using fluorescein-conjugated F(ab')z goat anti-mouse Ig and then analyzed on an EPICS V cell sorter (Coulter Electronics, Hialeah, FL) as previously described [32]. 2.4 Biosynthetic pulse labeling of cells
To analyze de novo protein synthesis, cells (5 x lo6cells/ml) were preincubated at least 1 h in methionine- and cysteinefree RF'MI 1640 medium supplemented with 1% dialyzed FCS, 20mM Hepes and 50 pg/ml of gentamycin (Cys-Met-RPMI). Cells were resuspended in this same medium at 1 x loh cells/ml and then stimulated with 10 ng/ml of PMA (Sigma, St. Louis, MO) at 37°C for various time periods. After incubation, the cells were labeled with 1mCi/ml [35S]methionine and of [35S]cysteine (DuPont/NEN, Boston, MA) for 20 min at 37°C. After washing, in order to allow for maturation of glycosylated CD45 molecules, cells were incubated at 37°C for an additional 60 min,with cysteine and methionine-rich (1 mM
each) RPMI 1640 supplemented with 10% FCS, and then the cells (5 X lo7)were lysed in 1 ml of lysis buffer A (20 mM Tris-HCI, pH 8.0, containing 1% NP40, 150 mM NaCI, 1 mM PMSF and 1 mM EDTA). The cell lysates were then used for further analysis by immunoprecipitation. In the case of 18-h treatment with PMA, the first incubation with PMA was performed using normal RPMI 1640 medium (containing cysteine and methionine) supplemented with 10% human AB serum. Two hours before the end of the incubation period, cells were washed twice with Cys-/ Met-RPMI and then incubated in this same medium as detailed above. 2.5 Phosphorylation study
For phosphorylation studies, the cells were washed three times with Hepes-buffered saline containing 135 mM NaCl, 5 mM KCI, 0.6 mM CaC12,2mM glucose and adjusted to 5 x lo6 cells/ml. After preincubation for 30 min at 37"C, carrier-free 32P-phosphatewas added to a final concentration of 250 pCi/ml= 9.25 MBq and the cells were incubated at 37°C for 3 h. After incubation, the cells were washed twice with Hepes-buffered saline and stimulated with PMA at 37 "C for 10 to 60 min. The cells (1 x lo7) were lysed in 1 ml of lysis (buffer B, 20 mM Tris-HCI, pH 8.0, 50 mM NaCl, 1% NP40,l mM PMSF, 1mM EDTA, 5 mM NaF, 10 mM sodium pyrophosphate) on ice. 2.6 Surface labeling of cells and immunoprecipitation assay Cells (1 x lo7) were labeled with 1 mCi Na1251(Amersham Int., Amersham, GB) by lactoperoxidase-catalyzediodination [33] and then lysed in 1 ml of the lysis buffer A. Cell lysates were precleared overnight at 4 "C with formalinfixed Staphylococcus aureus Cowan I organisms (Sigma). The precleared lysate was then incubated with 5-20 pg of purified anti-T200 (GAP 8.3) antibody at 4°C for 6 h. After incubation, 20 pl of protein A-Sepharose (50% suspension, Pharmacia, Uppsala, Sweden) was added to each tube and further incubated for 2-3 h at 4°C. Precipitates were then washed once in lysis buffer containing 0.5% NaCI, once in lysis buffer containing 0.5% deoxycholate and once in lysis buffer alone. Immunoprecipitates were solubilized with SDS-PAGE sample buffer and analyzed by SDS-PAGE on a 6% gel under reducing conditions. Gels were then fixed, dried and subjected to autoradiography using Kodak X-AR5 film and light intensifying screens (DuPont, Wilmington, DE). For densitometric analysis an LKB2222-010 Ultra Scan XL (LKB, Bromma, Sweden) was used. 2.7 PTPase assay
PTPase assays were performed as previously described [34]. Cells (either unstimulated, or after 30 min of stimulation with PMA) were washed with PBS and lysed in lysis buffer C (20 mM Tns-HC1, pH 8.0, 150 mM NaCl, 1% Triton X-100, 1 mM PMSF, 1 mM EDTA and 2 x Denhardt's solution) at a concentration of 2.5 X lo7 cells/ml. The lysates were mixed with appropriate preformed immune complex, and incubated at 4°C for 12 h. Immune complexes were sedimented by centrifugation, washed and
PKC activation and CD45 expression
Eur. J. Immunol. 1990. 20: 1655-1660
suspended in 50 p1 of PBS containing 2 x Denhardt's solution. The PTPase assay was performed at 37 "C for 60 min. The radioactivity released by 5 U of calf intestinal alkaline phosphatase (BFU, Gaithersburg, MD) was taken as 100% release. The [32P]Tyr-substratewas prepared as described previously [34], using the synthetic peptide Tyr(P) Raytide and tyrosine kinase (Oncogene Science, Manhasset, NY).
3 Results and discussion 3.1 The expression of CD45 isoforms on T cells after PMA treatment
After exposure of T cells to PMA, the expression of a number of cell surface molecules is modified. For example, the surface expression of both the CD3 and CD4 antigens is decreased after PMA treatment [24, 35-37], while expression of lymphokine receptors, such as the IL 2R (CD25), is increased [38]. However, the expression of CD45 isoforms on Tcells after PMA exposure is unknown. Therefore, to determine the effect of PMA on the expression of different CD45 isoforms, we cultured purified T cells from healthy donors in the presence of PMA (10 ng/ml). After various incubation periods, cells were harvested and CD45 isoforms were analyzed by FCM using 2H4(anti-CD45RA) and UCHLl (CD45RO), as well as GAP 8.3 Ab (antiCD45) which recognizes an epitope common to all CD45 isoforms. As shown inTable 1, the percentage of CD4+ cells markedly decreased within first 24 h of PMA treatment, but then attained its initial value. The percentage of CD8+ cells was slightly increased. The percentage of CD3+ cells and the density of CD3 antigen onTcells (data not shown) were also decreased by PMA treatment. On the other hand, IL 2R (Tac, p55) expression increased. As seen inTable 1, the common CD45 epitope GAP 8.3 was expressed on 98% of resting Tcells. After PMA treatment, the percent of positive cells remained stable, but a gradual
1657
Table 2. Effect of PMA on the expression of cell surface molecules on CD4 cellsa) Culture mAb period Anti-CD4SRA Anti-1L 2R Anti-CD4 (h) 0 1
2 IS 24 48
530/0 55% 57% 63% 66YO 70%
7% 8YO 8% 42% 15% 37%
94Y"
92YO 89% 73% 56% 71%
AntLCD8 3Yo 3% 3% 3YU 4 Yo 5 YU
a) Purified CD4 cells were cultured with 10 nglml PMA. At various times cells were stained by indirect immunofluorescence on an Epics V cell sorter. b) Data are a representative of two experiments performed.
increase in density of expression was observed (data not shown). The percentage of CD45RA+ cells rapidly increased after PMA treatment, with initial changes noted within 2-4 h. By 24 h, the expression of CD45RA had increased from 50% to 70%. The density of CD45RA expression remained essentially unchanged until 48 h. By 72 h, the majority of CD45RA+ cells expressed a high density of CD45RA with an increase in mean channel fluorescence from 78 to 104 (data not shown). In contrast, CD45R0, which is expressed in high amount on the reciprocal subset of resting T cells, did not change within the first 24 h after PMA treatment. However, by 48 h, the percentage of CD45RO+ cells also began to increase. Again, by 96 h of culture, both CD45RA and CD45RO expression were elevated by 29% and 24%, respectively. These phenomena were observed over a wide range of PMA concentrations, ranging between 1ng/ml to 100ng/ml (data not shown). In all of the following experiments, 10 ng/ml of PMA was used unless otherwise stated. To determine whether the increase in CD45RA antigen
-
-
Table 1. Effect of PMA on the expression of surface molecules on Tcellsa) Culture period (h 1
0
Anti-CD4SRA
so f
6h)
0 2
0 8* 2
4
16+ 4 1Y+ 7
8 24 18 72 96
20f
7 9 24 f 12 2 Y f 8 27+
Anti-CD4SRO
63 k S 0
-1 f 3 3 f S 4 f 3 2+s 11+1 21 f 2 24 f 1
Anti-CD4S
mAb Anti-CD3
Anti-CD1
Original percentage of positive cells 98 f 2 93 f 2 M-t_ 6 Increment in percentagc of positive cells') 0 n 0 I f 1 Of 1 -12+ 6 I f 1 -3s f 7 1f1 -58rf: 2 I f 1 -12f s -27 f 10 -4s 12 I f 1 -18f 6 1+1 -28f 9 -2 f s -12+ 8 0+1 3+ 3 1+1 0+ 1
+
AntLCD8
Anti-IL 2K
25 f 8
3.5 f 3
0 3f0 4 f l h f 3
7 f 3 Xf3 10 4 11 2 7
*
a) Tcells were cultured with 10 nglml of PMA. At various times after initiation of culture, cells were washed, stained by indirect immunofluorescence, and analyzed on an EPICS V cell sorter. b) Values show the mean f SD of the three different experiments. c) Increment in percentage of positive cells = (percentage of positive cells at indicated culture period) - (percentagc of positive cells at time 0). Values show the mean f SD of three different experiments.
1658
A.Yamada, M. Streuli, H. Saito et al.
Eur. J. Immunol. 1990. 20: 1655-1660
expression observed occurred among CD4 cells, CD8 cells or both, changes in CD45RA expression on CD4 cells after PMA treatment was examined. As shown in Table 2, after PMA treatment of CD4 cells, the percentage of CD45RA+ cells rapidly increased as seen in the unseparated T cell population. In addition, the CD4 molecule on CD4 cells was down-regulated after PMA treatment as described previously [24, 371. Moreover, an increase in CD45RA antigen expression is also observed on purified CD8 cells after PMA treatment (data not shown). Thus increase of CD45RA antigen expression was observed in both CD4 and CD8 T cell subsets. Mammalian cells are known to increase cell surface expression of various molecules by at least two different mechanisms. One mechanism is via de novo protein synthesis, as exemplified by the increased expression of the IL 2R receptor after cell activation [39]. The other mechanism is the release of pre-synthesized proteins. For example, the Mol antigen (CR3) on monocytes is stored in large quantities in the cytoplasm of resting cells [40]. To determine whether de novo protein synthesis was required for the rapid increment in CD45RA expression observed, cycloheximide, an inhibitor of de novo protein synthesis, was used.T cells pretreated with 20 PM of cycloheximide for 1 h prior to stimulation with PMA failed to increase their expression of CD45RA, CD45RO or CD25,8 h after PMA stimulation, as measured by FCM (data not shown). These results indicated that de novo protein synthesis is essential for increased expression of the CD45 isoforms after PMA treatment.
3.2 Effect of PMA on de novo synthesis of CD45 molecules To determine which of the two CD45 isoforms recognized by anti-CD45RA is synthesized de novo immediately after PMA stimulation, PMA-treated T cells were pulse labeled with [35S]methionineand [35S]cysteineand CD45 molecules were immunoprecipitated using anti-CD45 (CAP 8.3) mAb and analyzed by SDS-PAGE (Fig. 1A). The results of densitometric analysis of the autoradiogram shown in Fig. 1A are shown in Fig. 1B. A control CD45 immunoprecipitation from lZ5Isurface-labeled T cells is shown in lane a (Fig. 1A). In unstimulated T cells, the lower three CD45 isoforms (180 kDa, 190kDa and 205 kDa) are constitutively synthesized de novo, whereas the 220-kDa protein was not detected (lane b). After PMA treatment, the incorporation of ["Slamino acids into the CD45 180-kDa, 190-kDa and 205-kDa isoforms, but not 220-kDa CD45 isoforms, rapidly increased, reaching a maximum after 3 h (lane d). By 18 h of PMA stimulation (lane f), de novo synthesis of each protein returned to the same level as that seen in resting T cells (lane b). Even though the 180-kDa, 190-kDa and 205-kDa CD45 isoforms were being maximally synthesized at 3 h, de novo synthesis of the 220-kDa protein could still not be detected (Fig. 2 b, lane d).These results suggest that the rapid increment in CD45RA expression observed after PMA stimulation is primarily due to increased de novo synthesis of the 205-kDa and not the 220-kDa CD45 molecule. This finding is in general agreement with results obtained from CD4+ and CD8+ clones [41] and long-term CD4+ cell lines expressing CD45RA [42], in which the 205-kDa protein is present while the 220-kDa protein has disappeared.
Figure I. Effect of PMA on de n o w synthesis of CD45 molecules. T cells were cultured in cysteine/methionine-free medium with 10 nglrnl of PMA. After various incubation times, cells were pulse labeled with [3sS]cysteine and [3sS]methioninefor 20 min. Cells were then chased in complete medium for 1 h at 37°C and lysed. Lysates were immunoprecipitated with mAb GAP 8.3 and subjected to SDS-PAGE (6% acrylamide). Autoradiograph (A) and its densitometric tracings (B) of a resultant immunoprecipitation are shown. Lane (a) shows control GAP 8.3 immunoprecipitation from 1251-surface-labeled Tcells, lanes (b-f) from 3sSpulse-labeled Tcells. Lane (b): unstimulated Tcells; lane (c):Tcells stimulated with PMA for 1 h; lane (d): stimulated with PMA for 3 h; lane (e): stimulated with PMA for 6 h; lane (f): stimulated with PMA for 18 h. Arrows indicate the positions of the 220-kDa, 205-kDa, 190-kDa and 180-kDa CD45 isoforms.
3.3 Effect of PMA on CD45 phophorylation and CD45-linked PTPase activity
CD45 molecules are known to be substrates of PKC both in vivo [25] and in vitro [28].The common cytoplasmic portion of CD45 molecules has two putative PKC phosphorylation sites [20]. Although it has been reported that PMA stimulation increases the phosphorylation of the CD45 antigen [25], it is not clear whether all CD45 isoforms are equally phosphorylated. In order to establish which CD45 isoforms are phosphorylated following PMA stimulation, we incubated T cells with carrier-free 32P-phosphate and then stimulated with 100ng/ml of PMA. After stimulation, cells were lysed and the incorporation of 32P into CD45 isoforms was analyzed by immunoprecipitation using antiCD45 mAb GAP8.3 (Fig. 2 A). In unstimulated Tcells, all CD45 isoforms are constitutively but weakly phosphorylated (Fig. 2 A, lanes b, d and f). Densitometric analysis of the autoradiogram (Fig. 2 A) is shown in Fig. 2 B. After a 10-min stimulation with PMA, all CD45 isoforms, but particularly the lower-molecular mass isoforms (180-kDa and 190-kDa proteins) were phosphorylated (Fig. 2 A, B, lane c). Increased phosphorylation of all CD45 isoforms is also seen after 30 or 60 min of PMA stimulation (Fig. 2 A and B, lanes e and g). Furthermore, 32Pincorporation into each CD45 isoform seems to correspond to the density of expression as determined by 1251-surfacelabeling (Fig. 2 B, lane a). However, the relationship between the phospho-
Eur. J. Immunol. 1990. 20: 1655-1660
PKC activation and CD45 expression
1659
(A)
rylation of Ser/Thr residues of CD45 molecules and changes in surface expression of these molecules after PMA activation is unclear. Phosphorylation of the CD45 molecules following PKC activation may result in a change in the surface expression of these molecules and/or an alteration of CD45 function. For example, phosphorylation could alter the binding affinity of CD45 for its putative ligand or change the CD45 PTPase activity. In this regard, we have previously reported that the density of CD45RA antigen expression is increased after 2 days of Con A or 7 days AMLR activation, respectively, and that the level of CD45RA antigen expression correlates with suppressor function of Con A and AMLR-activated T cells [9-111. Furthermore, the anti-CD45RA (2H4) antibody modulates the suppressor/inducer and suppressor/effector function of CD4+ and CD8+ cells [9-11].2H4 and other anti-CD45RA antibodies also augment T cell proliferation by PHA and AMLR [16-181. These findings suggest that the CD45 molecules are functionally important molecules that play an important role in modifying T cell activation signals. Therefore, experiments were next performed to determine whether the phosphorylation of the CD45 molecules following PMA treatment altered the PTPase activity of the cytoplasmic domains of CD45 antigen. CD45 antigens were partially purified by immunoprecipitation from cell lysates using an anti-CD45 antibody (GAP 8.3) and the level of PTPase activity in the immunoprecipitates was determined. As shown in Table 2, the CD45 molecules precipitated with mAb GAP 8.3 from resting T cells released 23.3% and 15.2% of the phosphate from a phosphotyrosine substrate in two experiments (Table 3). As a control, class I MHC molecules were precipitated with the W6/32 mAb, and demonstrated no PTPase activity (no release of free phosphate). Under the same conditions, the PTPase activity of CD45 molecules from PMA-activated T cells (activated with 100 ng/ml of PMA for 30 min) was decreased by almost 50% relative to that of restingTcells (from 23.3% to 12.5% and from 15.2% to 9.7%, respectively).
Figure 2. PMA-induced phosphorylation of CD45 molecules. 32P-phosphate-loadedTcells were stimulated with or without 100 ng/ml of PMA at 37 "C.After various incubation periods, cells were washed, lysed, and then the lysates were subjected to SDS-PAGE on 6% acylamide gels. Autoradiograph (A) and its densitometric tracings (B) of a resultant immunoprecipitation assay are shown. Lane (a) shows control GAP 8.3 immunoprecipitation from 1251-surface-labeledT cells, lanes (b-g), show GAP 8.3 immunoprecipitation from 32P-phosphate-loadedTcells. Lanes (b, d and f) are fromTcells incubated without PMA for 10rnin (lane b), 30 min (lane d) or 60 rnin (lane f ) . Lanes (c, e and g) are from Tcells stimulated with PMA for 10 rnin (lane c), 30 rnin (lane e) or 60 min (lane g).
These results suggest that the activation of PKC downregulates CD45 PTPase activity. However, it is unclear whether this inhibition of CD45-linked PTPase activity by PKC would up- or down-regulate T cell activation. For example, the association of p56ICkwith the cytoplasmic portion of both CD4 and CD8 has been reported [43-451. p561Ckcontains a tyrosine residue at position 505 which is highly phosphorylated in vivo [46]. Since a mutation of this
tyrosine residue to phenylalanine increases tyrosine kinase activity [47,48] (analogous tov-src), one might suspect that dephosphorylation of TyrSoSshould increase tyrosine kinase activity. Indeed, isolated CD45 can directly activate p56ICk in vitro [49] and moreover, the intracellular domain of CD45 has recently been shown t o dephosphorylate p56ICkat the regulatory TySo5 site in an in vitro transfection study
18h (lane 1 )
0.3
Relative Distance
Table 3. Effect of PMA on CD45-linked PTPase activitya)
Exp.
1
Lysate from Tcells treated withb)
Immune complex
Untreated
W6l32 GAP8.3 W6/32 GAP8.3 Alkaline phosphatased)
280 20 560 174 11046 88 103
0.3 23.3 0.2 12.5 100
W6132 GAP8.3 W6l32 GAP8.3 Alkaline phosphatase
235 18566 27 11975 122498
0.2 15.2 0.0 9.7 100
PMA -
2
Untreated PMA -
Free phosphate cDrnC)
released (YO
a) PTPase assay is described in Sect. 2.7. b) Tcells were incubated with or without PMA (100 nglml) for 30 min at 37 "C. After incubation, cells were washed and lysed. For each assay, a lysate made from 2.5 X lo7 Tcells was immunoprecipitated. c) cpm of the blank sample (200-500 cpm) was already subtracted. Values show the mean cpm of triplicate assay. d) Alkaline phosphatase was used for determination of 100% release value.
1660
A.Yamada, M. Streuli, H. Saito et al.
[50].These results suggest that CD45-linked PTPase could either up- or down-regulate T cell activation via tyrosine dephosphorylation. Further studies including identification of the physiologic substrates for this PTPase or the physiologic ligands for the different extracellular domains of CD45 isoforms are required to clarify these points. The authors would like to thank Mr. Herb Levine and Mr. John Daley for technical assistance in flow cytometric analysis. We would also like to thank Ms. Gayathri Ranganathan for her excellent secretarial assistance. Received December 15, 1989; in revised form March 22, 1990.
4 References 1 Thomas, M. L., Annu. Rev. Immunol. 1989. 7: 399. 2 Streuli, M., Hall, L. R., Saga,Y., Schlossman, S. F. and Saito, H., J. Exp. Med. 1987. 166: 1548. 3 Ralph, S. J.,Thomas, M. L., Morton, C. C. and Trowbridge, I. S., EMBO J. 1987. 6: 1251. 4 Rudd, C. E., Morimoto, C. ,Wong, L. L. and Schlossman, S. F., J. Exp. Med. 1987. 166: 1758. 5 Morimoto. C., Letvin. N. L.. Distaso. J. A.. A1drich.W. R. and Schlossman, S. F., J. Immunol. 1985. 134: 1508. 6 Smith, S. H., Brown, M. H., Rowe, D., Callard, R. E. and Beverley, F? C. L., Immunology 1986. 58: 63. 7 Terry, L. A., Brown, M. H. and Beverley, F? C. L., Immunology 1988. 64: 331. 8 Tedder, T. F., Clement, L. T. and Cooper, M. D., J. Immunol. 1985. 134: 2983. 9 Takeuchi, T., Schlossman, S. F. and Morimoto, C., Cell. Immunol. 1987. 107: 107. 10 Takeuchi,T., Rudd, C. E., Schlossman, S. F. and Morimoto, C., Eur. J. Immunol. 1987. 17: 97. 11 Morimoto, C., Letvin, N. L., Rudd, C. E., Hagan, M., Takeuchi, T. and Schlossman, S. F., J. Immunol. 1986. 137: 3247. 12 Lefrancois, L. and Bevan, M. J., Nature 1985. 315: 449. 13 Nakayama, E . , Shiku, H., Stockart, E., Oettgen, H. F. and Old, L. J., Proc. Natl. Acad. Sci. USA 1979. 76: 1977. 14 Seaman, W. E.. Talal, N., Herzenberg, L. A. and Ledbetter, J. A., J. Immunol. 1981. 127: 982. 15 Newman, W., Proc. Natl. Acad. Sci. USA 1982. 79: 3858. 16 Cobold, S . , Hall, G. and Waldmann, H. in McMichael, A. J. et al. (Eds.) Leukocyte Typing 111, Oxford University Press, Oxford 1987, p. 788. 17 Ledbetter, J. A., Rose, L. M., Spooner, C. E., Beatty, F? G., Martin, F? J. and Clark, E. A., J. Immunol. 1985. 135: 1819. 18 Martorell, J.,Vilella, R., Borche, L., Rojo, I. and Vives, J., Eur. J. Immunol. 1987. 17: 1447. 19 Ledbetter, J. A.,Tonks, N. K., Fischer, E. H. and Clark, E. A., Proc. Natl. Acad. Sci. USA 1988. 85: 8628. 20 Charbonneau, H.,Tonks, N. K., Walsh, K. A. and Fischer, E. H., Proc. Natl. Acad. Sci. USA 1988. 85: 7182. 21 Tonks, N. K., Charbonneau, H., Diltz, C. D., Fischer, E. H. and Walsh, K. A., Biochemistry 1988. 27: 8695. 22 Nishizuka, Y , Nature 1984. 308: 603.
Eur. J. Immunol. 1990. 20: 1655-1660
23 Kraft, A., Anderson, W., Cooper, L. and Sando, J. J., J. Biol. Chem. 1982. 257: 13198. 24 Blue, M. L., Hafler, D. A., Craig, K. A., Levine, H. and Schlossman, S. F., J. Immunol. 1987. 139: 3949. 25 Autero, M. and Gahmberg, C. G., Eur. J. Immunol. 1987.17: 1503. 26 Veillette, A., Horak, I. D. and Bolen, J. B., Oncogene Res. 1988. 2: 385. 27 Alcover, A., Ramarli, D., Richardson, N. E., Chang, H. C. and Reinherz, E. L., Immunol. Rev. 1987. 95: 5. 28 Shackelford, D. A . and Trowbridge, I.S., J. Biol. Chem. 1986. 261: 8334. 29 Morimoto, C., Letvin, N. L., Boyed, A.W., Hagan, M., Brown, H. M., Kornacki, M. A. and Schlossman, S. F., J. Immunol. 1985. 134: 3762. 30 Berger, A. E., Davis, J. E. and Cresswell, F?, Human Immunol. 1981. 3: 231. 31 Reinherz, E. L. and Schlossman, S. F., Cell 1980. 19: 821. 32 Matsuyama,T.,Yamada, A., Kay, J.,Yamada, K. M., Akiyama, S. K., Schlossman, S. F. and Morimoto, C., J. Exp. Med. 1989. 170: 1133. 33 Hubbard, A. L. and Cohn, Z. A. in Maddy, A . H. (Ed.) Biochemical analysis of membranes. Champman and Hall, London 1976, p. 427. 34 Streuli, M., Krueger, N. X.,Tsai, A.Y. M. and Saito, H., Proc. Natl. Acad. Sci. USA 1989. 86: 8698. 35 Cantrell, D. A., Davis, A. A. and Crumpton, M. J., Proc. Natl. Acad. Sci. USA 1985. 82: 8158. 36 Solbach, W., J. Exp. Med. 1982. 156: 1250. 37 Hoxie, J. A., Matthews, D. M. and Callahan, K. J., J. Immunol. 1986. 137: 1194. 38 Shackelford, D. A. and Trowbridge, 1. S., J. Biol. Chern. 1984. 259: 11706. 39 Smith, K. A., Adv. Immunol. 1988. 42: 165. 40 Springer,T. A., Dustin, M. L., Kishimoto,T. K. and Marlin, S. D., Annu. Rev. Immunol. 1987. 5: 223. 41 Brod, S. A., Rudd, C. E., Purvee, M. and Hafler, D. A., J. Exp. Med. in press. 42 Rothstein, D. M., Sohen, S., Daley, J. F., Schlossman, S. F. and Morimoto, C., Cell Immunol. 1990, in press. 43 Rudd, C. E. ,Trevillyan, J. M., Dasgupta, J. D. ,Wong, L. L. and Schlossman, S. F., Proc. Natl. Acad. Sci. USA 1988. 85: 5190. 44 Barber, E. K., Dasgupta, J. D., Schlossman, S. F. and Trevillyan, J. M., Proc. Natl. Acad. Sci. USA 1989. 86: 3277. 45 Veillette, A., Bookman. M. A., Horak, E. M. and Bolen, J. B., Cell 1988. 55: 301. 46 Veillette, A., Horak, I. D., Horak, E. M., Bookman, M. A. and Bolen, J. B., Mol. Cell. Biol. 1988. 8: 4353. 47 Marth, J. D., Cooper, J. A., King, C. S., Ziegler, S. F.,Tinker, D. A., Overall, R.W., Krebs, E. G. and Perlmutter, R., Mol. Cell. Biol. 1988. 8: 540. 48 Amrein, K. E. and Sefton, B. M., Proc. Natl. Acad. Sci. USA 1988. 85: 4247. 49 Mustelin, T., Coggeshall, K. M. and Altman, A., Proc. Natl. Acad. Sci. USA 1989. 86: 6302. 50 Ostergaard, H., Schackelford, D. A., Hurley, T. R., Johnson, F?, Hyman, R., Safton, B. M. and Trowbridge, I. S . , Proc. Natl. Acad. Sci. USA 1989. 86: 8959.
