CD45 Controls Interleukin-4-mediated IgE Class Switch Recombination in Human B Cells through Its Function as a Janus Kinase Phosphatase*

CD45 plays a critical regulatory role in receptor signaling through its protein tyrosine phosphatase and Janus kinase (JAK) phosphatase activities. To investigate whether CD45 also plays a regulatory role in Ig class switching in human B cells, we examined the effects of CD45 triggering on Ig class switching to IgE and its relationship with CD45 JAK phosphatase activity. Anti-CD45 triggering of CD45 significantly inhibited interleukin-4 + anti-CD40-induced switch recombination in a switch recombination vector assay in stably transfected Ramos 2G6 human B cells, as well as Ig ε germ-line transcription and Sμ-Sε switch recombination in primary human B cells. These negative regulatory effects on Ig class switching were concomitant with the ability of CD45 to dephosphorylate the induced phosphorylation of JAK1, JAK3, and signal transducer and activator of transcription 6, but not on stress-activated/mitogen-activated protein kinases. We also showed that phosphorylated JAK1 and JAK3 were directly dephosphorylated by recombinant CD45 in vitro. These results indicate that CD45 is able to function as JAK phosphatase in human B cells and that this activity is directly associated with the negative regulation of the class switch recombination to IgE. CD45 may be an appropriate target drug for modulating IgE in allergic diseases.

CD45 plays a critical regulatory role in receptor signaling through its protein tyrosine phosphatase and Janus kinase (JAK) phosphatase activities. To investigate whether CD45 also plays a regulatory role in Ig class switching in human B cells, we examined the effects of CD45 triggering on Ig class switching to IgE and its relationship with CD45 JAK phosphatase activity. Anti-CD45 triggering of CD45 significantly inhibited interleukin-4 ؉ anti-CD40-induced switch recombination in a switch recombination vector assay in stably transfected Ramos 2G6 human B cells, as well as Ig ⑀ germ-line transcription and S-S⑀ switch recombination in primary human B cells. These negative regulatory effects on Ig class switching were concomitant with the ability of CD45 to dephosphorylate the induced phosphorylation of JAK1, JAK3, and signal transducer and activator of transcription 6, but not on stress-activated/mitogenactivated protein kinases. We also showed that phosphorylated JAK1 and JAK3 were directly dephosphorylated by recombinant CD45 in vitro. These results indicate that CD45 is able to function as JAK phosphatase in human B cells and that this activity is directly associated with the negative regulation of the class switch recombination to IgE. CD45 may be an appropriate target drug for modulating IgE in allergic diseases.
CD45 is a type I transmembrane molecule and a prototypic transmembrane protein tyrosine phosphatase that has been shown to play a critical role in controlling the activation and development of lymphocytes. Anti-CD45 antibody (␣CD45) 1 has been reported to alter signaling in B cells (1)(2)(3), T cells (4,5), basophils (6), and microglial cells (7). Anti-CD45RB is a potent immunosuppressive agent that can prevent transplant rejection in animals (8). CD45 is well known to couple to and directly regulate the activity of Src family tyrosine kinases (9).
It has been suspected that CD45 might have broader biological effects by acting on additional substrates (10). Indeed, it has been reported recently that (a) CD45 negatively regulates cytokine receptor signaling as a Janus kinase (JAK) phosphatase in hematopoietic cells (11) and (b) ␣CD45 can control cytokine-mediated signal transducers and activators of transcription (STAT3 and -5) to inhibit cytokine-driven lymphocyte proliferation at an early activation stage (12). These discoveries suggest that CD45 plays a wider role in cytokine-and JAK kinase-mediated B cell function, prompting us to investigate the effect of ␣CD45 on IL-4 signaling and Ig class switch recombination (CSR) in human B cells.
IL-4 is among the most important factors in determining Ig class switching in humans and, in particular, in the production of IgE (13,14). The ␣ chain of the IL-4 receptor activates the JAK family members JAK1 and JAK3 via induction of tyrosine phosphorylation, a process that is required for the subsequent tyrosine phosphorylation and activation of STAT 6 (15)(16)(17)(18). Phosphorylated STAT6 demerits and is translocated to the nucleus, where it binds to the STAT6 consensus sequences in the IgH ⑀ germ-line promoter and activates ⑀ germ-line transcription with production of ⑀ germ-line transcripts (⑀GTs). Both the process of germ-line transcription and GTs themselves are felt to be important for switching, although the exact roles of each in Ig CSR remain to be determined. For optimal ⑀ germ-line transcription and ⑀GTs production, synergy between IL-4 stimulation and a second signal through CD40 is required (19).
In this study, we examined whether activation of CD45 could alter isotype switching in human B cells. Switch recombination was quantified employing a recently established switch vector assay (20), S-S⑀ recombination in primary B cells was measured by digestion-circularization-PCR (DC-PCR), and ⑀GTs were assessed by RT-PCR. We also examined the effect of ␣CD45 on IL-4-and ␣CD40-driven phosphorylation and activation of JAK1, JAK3, and STAT6, as well as phosphorylation of c-Jun N-terminal kinase (JNK), p38 mitogen-activated protein kinase (p38 MAPK), and extracellular-signal related kinase (ERK). Biotechnology (Santa Cruz, CA). Recombinant CD45 was obtained from Calbiochem (San Diego, CA). Restriction endonucleases, ligase, and mung bean nuclease used for construction of switch vector came from Promega (Madison, WI) and New England Biolabs (Beverly, MA).
Cells, Cell Lines, and Cell Culture-The human B lymphoma cell line Ramos 2G6 (ATCC) was maintained and cultured in complete RPMI 1640. For transfection, 10 g of plasmid DNA that had been predigested with AseI was mixed with 1 million cells in 0.2 ml and then subjected to electroporation (200 V, 0.975 millifarads). Selection of stable transfected cell lines was achieved by Geneticin (Invitrogen) selection beginning 2 days later with concentration being increased over a period of 4 weeks to 1.5 mg/ml. The switch construct XF-5a has been described previously (20).
Peripheral blood mononuclear cells were isolated from healthy volunteers by centrifugation on Ficoll-Hypaque. Human B cells were purified from peripheral blood mononuclear cells by T cell depletion after monocytes/macrophages and NK cells were removed. Human B cells were cultured in RPMI 1640 medium supplemented with 2 mM glutamine, 100 units/ml penicillin, 100 g/ml streptomycin, and 10% fetal calf serum (Omega, Tarzana, CA).
RNA Extraction and RT-PCR-Total mRNA was obtained from stimulated and unstimulated cells using Trizol reagent (Invitrogen). RNA suspended in 0.1% diethyl pyrocarbonate-treated water was digested with DNase I (Sigma) to remove possible contaminating DNA and then extracted with phenol/chloroform followed by precipitation in ethanol. Total RNA (1 g) was reverse-transcribed to cDNA as described previously (20). All polymerase chain reaction (PCR) assays were done in 50-l reaction volumes containing 50 mM KCl, 20 mM Tris-HCl (pH 8.4), 2.5 mM MgCl 2 , each primer at 1 M, and 2.5 units of Taq polymerase (Promega). For detection of IgH ⑀GTs and glyceraldehyde-3-phosphate dehydrogenase, PCR was conducted with 40 cycles of 94°C for 1 min, 60°C for 1min, and 72°C for 1min. The primers GM3 (5Ј-AGCTGTC-CAGGAACCCGACAGGGAG-3Ј) and C⑀2B (5Ј-GTTGATAGTCCCT-GGGGTGTA-3Ј) were used to amplify ⑀GTs. This set of primers amplifies ⑀GTs as a 518-bp PCR product.
Immunoprecipitation-Ramos 2G6 cells (1 ϫ 10 6 /ml) were collected by centrifugation and lysed in Triton lysis buffer (2% Triton X-100, 0.15 M NaCl, 5 mM EDTA, 100 M Na 3 VO 4 , 10 g/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride, and 50 mM Tris/HCl, pH 7.5). The lysates were clarified and incubated with excess of protein A-Sepharose 4B (50% slurry). The cleared samples were immunoprecipitated with the appropriate antibodies (2 g) and protein A-Sepharose 4B (30 l) at 4°C for 1 h followed by washing three times with the lysis buffer. The immune complexes were subject to Western blot analysis as described previously (21).
Gel Electrophoresis and Western Blot Analysis-Samples containing equal amount of protein were boiled with electrophoresis sample buffer for 3 min and separated using SDS-PAGE. The separated proteins were transferred electrophoretically to membranes (Millipore, Bedford, MA). The membranes were blocked at room temperature for 1 h in phosphate-buffered saline, pH 7.4, with 1% bovine serum albumin, incubated with primary Abs for 1 h at room temperature, and washed followed by incubation with horseradish peroxidase-labeled secondary Abs for 1 h. The blots were developed using enhanced chemiluminescence reagents (ECL, Amersham Biosciences) and exposed to BioMax film (from Eastman Kodak Co.).
Phosphatase Assay-Tyrosine phosphatase assays using immunoprecipitated JAK1 and JAK3 from stimulated cells as substrates were performed as described (11,22). Recombinant CD45 proteins were diluted in phosphatase buffer (50 mM Tris/HCl, pH 7.2, 1 mM EDTA, 0.1% 2-mercaptoethanol), and mixed with immunoprecipitates including phosphorylated JAK1 or phosphorylated JAK3 proteins at 30°C for 20 min. After the reaction was stopped by boiling the samples with electrophoresis buffer for 3 min, samples were electrophoresed and transferred to polyvinylidene difluoride membrane. Tyrosine phosphorylation was monitored by anti-phosphotyrosine Ab.
Digestion-Circularization (DC)-PCR for Human S-S⑀ recombination-DC-PCR was used to quantify the levels of S-S⑀ recombination in stimulated primary human B cells. Genomic DNA was digested with BglII followed by ligation under conditions favoring self-ligation. The resultant ligated DNA was precipitated and the appropriate amounts of DNA was used as templates for PCR with primer a (5Ј-GATATGCTG-TTTGCACAAACTAG-3Ј) and b (5Ј-AACAACCCTCATGACCACCAGC-T-3Ј). Amplification for 40 cycles was performed at 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min. The size of the expected PCR product was 222 bp.
To verify the amount of ligated DNA between different groups and the efficiency for digestion and ligation of the input DNA sample, the human activation-induced cytidine deaminase (AID) gene was used as an unrelated control gene for the DC-PCR assay. BglII digestion would generate a 4578-bp fragment from the human AID gene (EMBL/Gen-Bank TM accession No. AB040430). Primer pairs AID5 (5Ј-CCATGGTA-CAAATCTCAGGACGAATC-3Ј) and AID6 (5Ј-AGATGGTGAAACCCC-GTCTCTATTAA-3Ј) were used. This pair of primer would amplify a 238-bp product. PCR was conducted using 20 ng of ligated DNA as templates at 94°C for 1 min, 62°C for 1 min, and 72°C for 1 min for 40 cycles.
FACS Analysis-The expression of green fluorescence protein (GFP) in cell lines stably transfected with the switch vector after the stimulated or unstimulated culture conditions was measured by either singleor dual-color flow cytometry (FACS Core laboratory, UCLA) as described (20). FACS data were analyzed with FCS expression software (Deno Novo software Inc., Thornhill, Ontario, Canada).
Statistical Analyses-Statistical analysis was performed using Mann-Whitney's U test in levels of Ig class switch recombination. Macintosh computers (StatView software, Abacus Concepts, Berkeley, CA) were used for all statistical analyses.

Triggering of CD45 Inhibits Switch Recombination Activity
in Switching Constructs-We tested the ability of different ␣CD45 reagents, ␣CD45 (HI30), ␣CD45RA (HI100), and ␣CD45RB (MT4), to affect IL-4 ϩ ␣CD40-driven human Ig isotype switching. We have previously established a switch recombination assay by using GFP as an indicator for SSR in the Ramos 2G6 human B cell line (20). More than 95% of these Ramos cells express readily detectable levels of CD45, CD45RA, and CD45RB as determined by flow cytometry (data not shown). All three ␣CD45 attenuated IL-4 ϩ ␣CD40-induced SSR (as indicated by GFP expression) from the switch construct (XF-5a) with ␣CD45 HI30 showing the greatest inhibition (Fig. 1A). The optimal concentration of all three mAbs was 5-10 g/ml (data not shown). Thus, ␣CD45 HI30 was used in most of the following experiments.
As shown in Fig. 1B, IL-4 ϩ ␣CD40-induced SSR was inhibited by ␣CD45 in a dose-dependent fashion with the maximal inhibition generally being reached at 5 g/ml or higher. At these concentrations, ␣CD45 alone did not significantly alter cell viability nor did it alter viability in IL-4 ϩ ␣CD40-stimulated cultures (data not shown).
It should be noted that GFP expression measures both deletional and inversional recombination events in the constructs (Fig. 1C) (20). Thus, ␣CD45-mediated inhibition of the GFP expression in the SSR assay reflects the suppression of recombination activity including both deletional and inversional recombination.
CD45 Suppresses S-S⑀ Switch Recombination in Primary B Cells-As ␣CD 45 was able to inhibit SSR in the human B cell line, we determined the effects of CD45 stimulation on S-S⑀ switch recombination in human primary B cells as assessed through a quantitative DC-PCR assay. As shown in Fig. 2, S-S⑀ recombination was readily detectable in IL4 ϩ ␣CD40stimulated primary B cells (lane 2) but was not detectable in unstimulated cells (lane 1). The induction of S-S⑀ recombination was inhibited by ␣CD45 in a dose-dependent manner, with significant inhibition being achieved at 5 g/ml (lanes 3-6). It was decreased 85% as measured by semiquantitative analysis using densitometry.
CD45 Negatively Regulates ⑀ Germ-line Transcripts-To investigate the mechanism by which CD45 triggering interferes with Ig CSR, we examined the effects of ␣CD45 (HI130) on IL4 ϩ ␣CD40 induction of ⑀GTs in primary B cells and human B cell lines. Expression of IgH germ-line transcripts from Ig heavy-chain loci precedes the occurrence of isotype switching and has been shown to be important in Ig CSR (23). ␣CD45 inhibited IL4 ϩ ␣CD40-induced ⑀GTs in a dose-dependent manner in both primary human B cells and two human B cell lines, Ramos 2G6 and 2C4/F3 (Fig. 3, A and B). The maximum inhibitory effect was achieved at concentrations at or above 5 g/ml. Based on these results, the subsequent experiments were performed with 5 g/ml ␣CD45 unless noted otherwise. To estimate the sensitivity of our PCR assay for the detection of GTs, we serially diluted cDNA as the PCR template and subjected the products to amplification via RT-PCR (Fig. 3C). The sensitivity of the technique was determined to be 50 pg of total cellular RNA. Semiquantitative analysis by comparison with the data shown in Fig. 3A demonstrated that ␣CD45 reduced the production of ⑀GTs about 90% compared with the level seen with IL-4 ϩ ␣CD40 stimulation.
CD45 Attenuates IL-4-induced JAK1 and JAK3 Phosphorylation-IL-4 is known to activate some members of the JAK family protein kinases through phosphorylation (15)(16)(17), whereas CD45 was recently shown to be able to function as a JAK phosphatase (11). Therefore we investigated whether ␣CD45-mediated inhibition of the ⑀GT induction and S-S⑀ switch recombination was mediated via regulation of JAK1 and JAK3 phosphorylation. IL-4 ϩ ␣CD40-stimulated Ramos 2G6 cells were treated with ␣CD45, and anti-phosphotyrosine immunoprecipitates of cell lysates were subjected to immunoblotting with anti-JAK1 Ab or anti-JAK3 Ab.
IL-4 ϩ ␣CD40 induced a rapid and sustained tyrosine phosphorylation of JAK1 and JAK3 (Fig. 4, A and B). On the other hand, IL-4 ϩ ␣CD40-induced tyrosine phosphorylation of JAK1 and JAK 3 did not occur following ␣CD45 treatment. Immunoblot analysis with anti-JAK1 Ab and anti-JAK3 Ab revealed comparable total amounts of JAK1 and JAK3 in ␣CD45-treated or untreated cells, indicating that IL-4 ϩ ␣CD40-induced tyrosine phosphorylation of JAK1 and JAK3 was blocked by ␣CD45.
Anti-CD45 Leads to a Decrease in IL-4-induced STAT6 Phosphorylation-Because activation of STAT6 by JAK kinase is essential in mediating the IL-4 response (18) and plays an important role in both expression of ⑀GTs and isotype switching to IgE (24), we investigated whether IL-4-induced STAT6 activation was also regulated by ␣CD45. Samples from appropriately stimulated cells were subjected to Western blot analysis with specific anti-phosphorylated STAT6. As shown in Fig.  5, phosphorylated STAT6 was readily detected after the stimulation. Within 5 min of IL-4 ϩ ␣CD40 stimulation, STAT6 phosphorylation increased and remained elevated throughout the 20-min incubation. However, after ␣CD45 treatment, IL-4 ϩ ␣CD40 failed to induce STAT6 phosphorylation, indicating that triggering of ␣CD45 strongly inhibited IL-4 ϩ ␣CD40-induced STAT6 phosphorylation and activation.
Recombinant CD45 Dephosphorylates JAK1 and JAK3-The inhibition of STAT6 phosphorylation by ␣CD45 was likely mediated via its effect on JAK kinases. To confirm that CD45

FIG. 1. CD45 signals negatively control SSR in Ramos B cells.
A, flow cytometric analysis of SSR. Ramos 2G6 cells (1 ϫ 10 5 cells/ml) permanently transfected with the switch vector (shown schematically in panel C) were treated with ␣CD45 (HI30), ␣CD45RA (HI100), or ␣CD45RB (MT4) (5 g/ml) and then stimulated with IL-4 (5 ng/ml) plus ␣CD40 (0.1 g/ml). After 4 days, SSR was measured based on expression of GFP from vectors undergoing switch recombination. The frequency of GFP-positive cells is shown. These data represent one of four similar experiments. B, dose response of ␣CD45-driven inhibition of switch recombination. Ramos 2G6 cells containing the switch vector were cultured for 4 days in the presence of media and IL-4 (5 ng/ml) and ␣CD40 (0.1 g/ml) plus the concentrations of ␣CD45 indicated. The resulting percent of GFP-positive cells is shown. The numbers represent the mean value Ϯ standard deviation from four independent experiments. *, p Ͻ 0.05. C, schematic diagram of the switch construct (XF5a) permanently transfected in Ramos 2G6 cells is shown at the top (18). During switch recombination, S joins to S␥2 with the intervening DNA between being looped out and excised. The excised DNA ends are joined to form the circular DNA, and GFP expression from the IRES-GFP unit is driven by the pRc/RSV-LTR (lower right). In case of inversion between S and S␥2 (lower left), the IRES-GFP expression unit, now in the correct orientation, is under the transcription control of the cytomegalovirus promoter (pCMV), which leads to GFP expression (closed star). Each open arrow indicates the promoter transcriptional site and direction. IRES, internal ribosomal entry site; pSV, SV40 promoter; RSV LTR, Rous sarcoma virus long-term repeat; Sd, splicing donor site; Sa, splicing acceptor site.

FIG. 2. Effects of ␣CD45 on S-S⑀ recombination determined by DC-PCR.
A, primary B cells (5 ϫ 10 5 cells/ml) were treated with various concentrations of ␣CD45 (HI30) and cultured with or without IL-4 (5 ng/ml) plus ␣CD40 (0.1 g/ml) for 5 days. DNA was then prepared, digested with BglII, and ligated as described under "Experimental Procedures" to yield self-ligation products. DC-PCR for S-S⑀ recombination products was amplified for 40 cycles. DC-PCR for the AID gene was used as an internal control for efficiency of digestion, ligation, and PCR amplification. B, diagram of DC-PCR strategy, primer positions, and orientations for quantifying human S-S⑀ recombination. The S and S⑀ segments in the top line of the map are intact prior to isotype switch. BglII sites were represented by small closed squares. After S-S⑀ rearrangement, a composite S-S⑀ region of indeterminate size lies on a single BglII fragment with ends close to the oligonucleotide sequences marked 5Ј primer (a) and 3Ј⑀ primer (b) (arrows). Following digestion, this BglII fragment was circularized by ligation and detected by PCR using the 5Ј and 3Ј⑀ oligonucleotides as primers. By doing this, all recombination events generate a uniform PCR fragment so that the frequency of S-S⑀ recombination can be quantified.
functions as a JAK phosphatase to negatively regulate cytokine receptor signaling in our system, we tested the phosphatase activity of recombinant CD45 in a phosphatase assay for the immunoprecipitated JAK kinases from stimulated Ramos 2G6 cells. Recombinant CD45, which contains the intracellular phosphatase domain but lacked the extracellular portion of CD45, directly dephosphorylated JAK 1 and JAK3 in a dosedependent manner (Fig. 6). The addition of vanadate, a potent inhibitor of protein tyrosine phosphatases, blocked the ability of recombinant CD45 to dephosphorylate both JAK1 and JAK3 (data not shown).
The effect of ␣CD45 on Phosphorylation of JNK, p38 MAPK, and ERK-We used IL-4 ϩ ␣CD40 to stimulate human B cells in order to test the effects of CD45 on ⑀GTs induction, S-S⑀ switch recombination, and the relationship of these outcomes to JAK phosphatase activity. CD40 cross-linking in B cells also induces JNK, p38 MAPK, and ERK (25)(26)(27)(28). p38 MAPK has been reported as potentially playing an important role in ⑀GTs (29). Therefore we also examined whether ␣CD45 plays a role in regulating the induced activation of these kinases. Thus we measured the effects of ␣CD45 on IL-4 ϩ ␣CD40-induced phosphorylation of JNK, p38 MAPK, and ERK using anti-phosphorylated JNK Ab, anti-phosphorylated p38 MAPK Ab, and antiphosphorylated ERK Ab.
JNK was activated by IL-4 ϩ ␣CD40, resulting in increased JNK phosphorylation on Thr-183 and Tyr-185 (Fig. 7A). Induced JNK phosphorylation was not inhibited by the same concentration of ␣CD45 that strongly inhibited JAK phosphorylation (Figs. 4 and 7A). Similarly, the activation of p38 MAPK phosphorylated on Tyr-182 was not significantly altered by ␣CD45 treatment (Fig. 7B). ERK was only modestly phosphorylated on Tyr-204 in response to IL-4 ϩ ␣CD40 and was also unaffected by ␣CD45 treatment (Fig. 7C). ␣CD45 itself had no effect on activating these three kinases.  6. Recombinant CD45 (rCD45) dephosphorylates JAK1 and JAK3. Ramos 2G6 cells were stimulated with IL-4 plus ␣CD40 for 5 min and lysed, and the supernatants were immunoprecipitated with anti-JAK1 or anti-JAK3 Abs. Recombinant CD45 (0, 10, 50, and 100 ng, respectively) was added to the immunoprecipitated complex containing phosphorylated JAK1 or phosphorylated JAK3 proteins, and the mixtures were incubated at 30°C for 20 min. The reaction mixtures were then blotted and detected with anti-phosphotyrosine Ab. The same samples were probed with anti-JAK1 Ab or anti-JAK3 Ab as input protein controls.

DISCUSSION
In the present study, we tested the ability of CD45 stimulation to directly alter B cells isotype switching to ⑀ and examined the mechanisms underlying this effect. CD45 was able to suppress IL-4 ϩ ␣CD40-induced expression of ⑀GTs and isotype switching to ⑀ in primary human B cells and B cell lines. CD45 signaling decreased the phosphorylation of JAK1, JAK3, and STAT6, molecules that play crucial roles in IL-4-induced isotype switching (24,30). We also demonstrated that recombinant CD45 directly dephosphorylated JAK1 and JAK3. At the same time, we did not detect any effect of ␣CD45 on CD40driven activation of JNK, p38 MAPK, and ERK. These findings suggest that CD45 controls isotype switching in human B cells primarily through its function as a JAK phosphatase.
The biologic responses induced by IL-4 involve a complex interaction of signaling pathways including the activation of JAK1 and JAK3 and STAT6. Although phosphatases, including SHIP and SHP-1, may regulate IL-4 signaling (31,32), the molecular mechanisms responsible for the dephosphorylation of key signaling intermediates and negative modulation of different parts of the IL-4 cascade remain to be elucidated. Irie-Sasaki et al. (11) recently reported in murine system that CD45 negatively regulated cytokine receptor signaling as a hematopoietic JAK phosphatase. In this study, we demonstrated that CD45 directly dephosphorylated JAK1 and JAK3 in human B cell. Thus, CD45 regulates IL-4 signaling, along with suppressor of cytokine signaling (SOCS) family members, which have been shown to suppress IL-4 signaling through the inhibition of JAKs (33), and B cell lymphoma gene-6 (BCL-6), which appears to regulate a subset of IL-4-induced genes (34).
CD40 plays an important role in B cell proliferation, survival, memory, and Ig CSR. Although our initial results demonstrated that CD45 could play a negative role in controlling the induced IgE class switching, they did not prove that the inhibition was exerted only through IL-4 signaling rather than CD40 signaling. This is particularly relevant, as CD40 signaling can activate multiple kinases and signal pathways (26 -28).
Recently, it has been reported that p38 MAPK plays an important role in CD40-induction of ⑀GTs (29), and we have also found that p38 MAPK is related to Ig CSR in our system. 2 Thus, we also investigated the effects of CD45 stimulation on IL-4 ϩ ␣CD40-driven JNK, p38 MAPK, and ERK activation to determine whether CD45 was exerting its effects via altered CD40 signaling. We demonstrated that induction of JNK, p38 MAPK, and ERK phosphorylation was not significantly altered by CD45 triggering, indicating that CD40-dependent activation of these specific signal pathways was not affected by CD45 triggering. These results, together with the fact that CD45 triggering suppressed IL-4 ϩ ␣CD40-induced JAK1, JAK3, and STAT6 phosphorylation, provide strong evidence that CD45 functions, at least in part, as a JAK phosphatase in human B cells and thereby inhibits IL-4-dependent class switching to IgE. It remains possible that CD45 may alter other CD40-activated or -dependent signaling pathways, e.g. CD40involved JAK3 and Lyn (35,36), that are not known to be involved in class switching processes.
Isotype switching is a key event in generation of humoral immunity. IL-4 and CD40 play critical roles in this process. The notion that CD45 might function as a JAK phosphatase, as reported recently (11), was further supported by our results that recombinant CD45 directly dephosphorylated the phosphorylated-JAK1 and -JAK3 in vitro. Our findings showed that IL-4 ϩ ␣CD40-induced IgE CSR could be suppressed strongly by CD45 engagement, suggesting that CD45-mediated JAK phosphatase activity is responsible for the suppression, because STAT6 activation by phospho-JAK1 and -JAK3 is required for IL-4-dependent ⑀GTs production and subsequent S-S⑀ recombination. However, inhibition of S-S⑀ recombination might not be the sole consequence of inhibition of the IL-4-dependent ⑀ germ-line transcription, although the latter may well participate in the inhibition of S-S⑀ recombination. For example, CD45 might inhibit S-S⑀ recombination through other steps in CSR, e.g. by altering the activation and/or induction of the putative switch recombinase activity, which is also dependent on cytokine activation (37).
CD45 has been suggested to be an important gatekeeper in determining early intracellular signaling events by influencing phosphorylation or dephosphorylation of proteins organized in lipid microdomains (rafts) (38). It is expressed throughout B cell development and differentiation with the exception of terminally differentiated plasma cells. CD45 has been shown to play an important role in modulating the signal that is transduced via the B cell antigen receptor by regulating Lyn (39,40). However, the role of CD45, specifically in regulating B cell class switching, which we have studied, has not been investigated in detail previously. By showing that CD45 negatively regulates the phosphorylation of JAK1 and STAT6, signal transduction molecules known to be critical for IL-4-induced isotype switching, we concluded that CD45 suppressed IgE isotype switching through its JAK phosphatase activity. At the same time, we cannot exclude the possible effects of CD45 on unknown or other molecules including Lyn, for which involvement in the class switching process has not been elucidated. Although recombinant CD45 strongly dephosphorylated JAK1 and JAK3, it is still not clear how ␣CD45 modulates the protein tyrosine phosphatase activity of CD45 so as to decrease IL-4-induced phosphorylation of JAK1 and JAK3. Whether this is a conformational effect and/or requires cross-linking is unknown. Even though we showed that the intracellular portion of CD45 possessed JAK phosphatase activity in vitro, how CD45 acts as a JAK phosphatase in vivo is also unknown. Finally, developmentally regulated alternative splicing of the single CD45 gene results in multiple isoforms with differences in their extracel- FIG. 7. The effect of ␣CD45 on the phosphorylation of JNK, p38 MAPK, and ERK. Ramos 2G6 cells were cultured for 10 min. under the same conditions with or without stimulation as described in the legend for Fig. 4. Equal amounts of cell lysates were subjected to SDS-PAGE followed by blotting with anti-phosphorylated JNK Ab (A), anti-phosphorylated p38 MAPK Ab (B), and anti-phosphorylated ERK Ab (C). The same samples were probed with anti-JNK, anti-p38 MAPK, and anti-ERK, as shown in the lower panels (A-C), to measure the total amount of protein kinase present in the samples. lular portions and these distinct isoforms are likely associated with differential functions.
It has become increasing evident that CD45 can have both positive and negative effects in regulating receptor thresholds and in the resulting biologic outcomes. Thus it is not surprising that the role of CD45 may vary according to cell lineage and developmental stage. Indeed, CD45 activation has been implicated in such disparate events as transplant rejection (8), cell adhesion-specific signaling events (41), the pathogenesis of systemic lupus erythematosus (42)(43)(44), and Alzheimer's disease (45,46). Only by careful analysis in different cell types and at different developmental stages will the central role of CD45 as a regulator in specific diseases be determined. Our data shows that CD45 signaling via its JAK phosphatase activity can regulate IL-4 ϩ ␣CD40-induced Ig class switching in human B cells, an event that plays a central role in the humoral immune reactivity associated with the TH2 responses seen in allergic disease and raises the possibility of CD45 as a drug target.