Eur. J. Immunol. 1999. 29: 823–837 SF2 regulates CD45 splicing during T cell activation 823 SF2 and SRp55 regulation of CD45 exon 4 skipping during T cell activation Raphael Lemaire, Annabelle Winne, Madathia Sarkissian and Robert Lafyatis Boston University School of Medicine, The Arthritis Center, Boston, USA CD45 is an alternatively spliced membrane phosphatase required for T cell activation. Exons 4, 5 and 6 can be included or skipped from spliced mRNA resulting in several protein isoforms that include or exclude epitopes referred to as RA, RB or RC, respectively. T cells reciprocally express CD45RA or CD45RO (lacking all three exons), corresponding to naive versus memory T cells. Overexpression of the alternative splicing regulators, SF2 or SWAP, induces skipping of CD45 exon 4 in transfected COS cells. We show here that the arginine/ serine-rich domain of SWAP and the RNA recognition motifs of SF2 are required for skipping of CD45 exon 4. Unlike SWAP, SF2 specifically regulated alternative splicing of CD45 exon 4, having no effect on a non-regulated exon of CD45 (exon 9). Like SF2 and SWAP, the SR proteins SC35, SRp40 and SRp75, but not SRp55 also induced CD45 exon 4 skipping. In contrast, antisense inhibition of SRp55 induced exon 4 skipping. SF2 and SRp55 proteins were not detectable or expressed at a very low level in freshly isolated CD45RA+ and CD45RO+ T cells. Activation of CD45RA+ T cells shifted CD45 expression from CD45RA to CD45RO, and induced a large increase in expression of both SF2 and SRp55. Thus, SF2 at least in part determines splicing of CD45 exon 4 during T cell activation. SRp55, SR protein phosphorylation, or other splicing factors likely regulate CD45 splicing in CD45RO+ memory T cells. The different SR proteins expressed by memory and activated T cells emphasize the different phenotypes of these cell types that both express CD45RO. Key words: T lymphocyte / Cell activation / Memory / SR protein / mRNA splicing 1 Introduction The leukocyte common antigen (CD45) is a transmembrane tyrosine phosphatase expressed by cells of the lymphopoietic lineage [1]. CD45 is required for normal TCR-mediated signaling [2] and regulates the basal activity of the Fyn and Lck protein tyrosine kinases by dephosphorylation of tyrosine residues [3, 4]. Experiments using chimeric proteins containing the cytoplasmic domain of CD45 indicate that the extracellular domain of CD45 is not necessary for TCR signaling [5, 6]. However, engagement of the external domains of CD45 can regulate the differentiation of immature CD4+CD8+ thymocytes into mature T cells [7], suggesting that unidentified ligands modulate the activity of CD45 [8]. [I 18816] Abbreviations: RRM: RNA recognition motif RT: Reverse transcription © WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999 Received Revised Accepted 25/8/98 26/11/98 30/11/98 Alternative splicing of CD45 pre-mRNA generates several isoforms in the N terminus of the extracellular domain. Three of the 33 exons that comprise the CD45 gene are alternatively spliced. These regulated exons (exons 4, 5 and 6) can be spliced out of the pre-mRNA independently or in combination [8]. Antibodies have been used to define the corresponding polypeptide regions as RA, RB and RC. In humans, seven isoforms have been identified: an isoform containing all three exons (reacting with antibodies specific for RA, RB, and RC), isoforms containing two exons – either 4 and 5, or 5 and 6 (reacting with antibodies specific for RA and RB, or RB and RC); isoforms containing only one exon (reacting with antibodies specific for RA, RB or RC) and an isoform containing none of these exons (reacting with an antibody specificity referred to as RO) [9]. Expression of the CD45 isoforms depends on the activation and differentiation states of lymphopoietic cells. In particular, the expression of CD45 isoforms containing exon 4 (CD45RA+) correlates with the stage of T cell maturation: it is abundant in immature CD3–CD4–CD8– thymocytes, decreased in intermediate CD4+CD8+ thymocytes, and then increased in mature CD4+CD8– or CD4–CD8+ thymo0014-2980/99/0303-823$17.50 + .50/0 824 R. Lemaire et al. cytes [10]. Naive peripheral T cells that express CD45RA isoforms switch to expression of the CD45RO isoform upon activation. This shift to the CD45RO isoform has been associated with T cell memory [11–14]. These different isoforms differentially regulate intracellular signaling pathways [15, 16]. In addition, CD45RO transgenepositive CD45–/– mice, but not CD45ABC transgenepositive CD45–/– mice, develop peripheral T cells able to generate cytotoxic T cell responses and neutralizing, Th cell-dependent IgG antibodies after viral infections. These results suggest that expression of CD45RO has a functional role in T cell differentiation [17]. Understanding the factors regulating alternative splicing of CD45 exon 4 should provide insight into the intracellular signals that are activated during development of T cell memory. Assembly on the pre-mRNA of a multicomponent riboprotein called the spliceosome determines both constitutive and regulated mRNA splicing. The spliceosome consists of small ribonucleoprotein (snRNP) particles U1, U2 and U4/U5/U6 as well as non-snRNP proteins also essential for pre-mRNA splicing [18–20]. Several proteins regulate alternative splicing of pre-mRNA. Members of the SR protein family are the best characterized mammalian splicing regulators. These proteins function as both constitutive and regulating splicing factors [21–23]. The SR protein family is remarkably conserved, with one or two N-terminal RNA recognition motifs (RRM) and a C-terminal region consisting of an arginine/ serine-rich (R/S) domain (hence the designation SR proteins) [22, 24, 25]. SR proteins likely differentially activate splice sites due to different affinity interaction with either 5’ splice sites or exon enhancers [26] and may similarly affect 3’ splice site selection [27]. To date, eight human geens encoding SR proteins have been cloned: SRp20 [24], SF2/ASF (SRp30a) [28, 29], SC35 (SRp30b) [30], SRp30c [31], SRp40 [32], SRp55 [31, 33], SRp75 [25] and 9G8 [34]. Several other proteins with R/S motifs also regulate alternative mRNA splicing, including the Drosophila Transformer (Tra), Transformer-2 (Tra-2) and DmSWAP proteins. Tra and Tra-2 are part of a cascade of splicing regulators that determine sex in Drosophila [35]. The structure of Tra-2 is quite similar to the SR protein family: it has an RRM between two R/S motifs. On the other hand, Tra contains only one R/S motif without other defined RNA binding motifs. DmSWAP was originally identified because it regulates splicing of the mutant white-apricot allele, thereby affecting Drosophila eye color [36]. It was subsequently shown to regulate splicing of its own premRNA [37]. SWAP also contains a C-terminal R/S motif. Recently, we showed that HsSWAP, the human homologue of DmSWAP [38], regulates splicing of several mammalian genes [39]. SWAP and the SR protein, SF2, Eur. J. Immunol. 1999. 29: 823–837 stimulated CD45 exon 4 skipping [39], suggesting that these factors may be important in regulating CD45 splicing during T cell maturation. We show here that the R/S domain of SWAP and the RRM of SF2 are responsible for the alternative splicing activity of SWAP and SF2 on CD45 pre-mRNA. We also report that, like SF2, other SR proteins, namely SC35, SRp40 and SRp75, enhance skipping of CD45 exon 4. In contrast, SRp55 enhanced CD45 exon 4 inclusion. Freshly isolated CD45RA+ (naive) and CD45RO+ (memory) T cells expressed very low or undetectable levels of both SF2 and SRp55. Activation of CD45RA+ T cells led to greatly increased expression of both SF2 and SRp55, strongly implicating SF2, but not SRp55, in CD45 exon skipping during T cell activation. 2 Results 2.1 The R/S domain of SWAP and the RRM region of SF2 are required for alternative splicing of CD45 exon 4 To precisely determine functional domains of SWAP and SF2 required for regulation of CD45 exon 4 alternative splicing, deletion mutants of SWAP and SF2 cDNA were cloned into an expression vector (diagrammed in Fig. 1A). Resulting deletion mutants were assessed for CD45 splicing activity in a cell transfection system developed previously [39]. Plasmids designed to overexpress SF2 or SWAP proteins lacking various regions were cotransfected into COS cells with a CD45 minigene construct, pSV-mini-LCA18 [40]. pSV-mini-LCA18 contains the genomic region of CD45 between exon 2 and 8, of which regions including exons 3, 5 and 7, and adjacent introns have been deleted (diagrammed in Fig. 1B). This construct reproduces exon 4 inclusion and exclusion in B cell and thymocyte cell lines, respectively [40]. The effect of overexpression of SWAP and SF2 mutants on splicing of RNA transcribed from the co-transfected CD45 construct was assayed by reverse transcription (RT)-PCR. Transfection of pSV-mini-LCA18 with control expression plasmid into COS cells resulted in production primarily of the CD45 isoform containing exon 4 (RA) and approximately 10–15 % of the CD45 isoform excluding exon 4 (form RO, Fig. 2A and B: pSG5). On the basis of previous work showing that specific SWAP deletion mutants (SWAP ¿ BH and SWAP ¿ B) regulate CD45 exon 4 splicing, but SWAP ¿ P and SWAP ¿ H (see Fig. 1) do not, we suggested that splicing regulation of CD45 exon 4 requires the R/S region of SWAP [39]. Overexpression of additional SWAP mutants lacking the R/S region ( ¿ BB, SWAP ¿ S and SWAP ¿ ST) had no sig- Eur. J. Immunol. 1999. 29: 823–837 SF2 regulates CD45 splicing during T cell activation 825 Figure 1. Deletion and hybrid mutant expression vectors for SWAP and SF2. (A) Full-length SWAP cDNA and deletion mutants SWAP ¿ S, SWAP ¿ ST, SWAP ¿ H, SWAP ¿ P, SWAP ¿ BH, SWAP ¿ B and SWAPRS are diagrammed. Five highly conserved regions of SWAP, including a long N terminus homology ( ), the two repeated surp homologies ( ), the R/S-rich region ( ) and a fifth homology ( ) are indicated. Functional domains of SF2 are diagrammed including 2 RRM indicated by and , and an R/S motif indicated by . The SF2/SWAPRS hybrid mutant was generated by ligation of SF2 ¿ RS and the SWAP R/S region. The SF2 hybrid mutant ¿ BB/SF2RS was generated from the SWAP N-terminal region and the SF2 R/S region. (B) CD45 mini-gene constructs are diagrammed. Thin lines and small boxes indicate CD45 introns and exons, respectively. 826 R. Lemaire et al. Eur. J. Immunol. 1999. 29: 823–837 Figure 2. The R/S domain of SWAP and the RRM region of SF2 are required for alternative splicing of CD45 exon 4. (A) RT-PCR products and (B) graphical representation of alternatively spliced CD45 mRNA from COS cells 2 days after transient cotransfection with 0.5 ? g pSV-mini-LCA18, and 3 ? g of pSG5 (control), full-length SWAP expression plasmid (SWAP), mutant SWAP expression plasmids ( ¿ BB, SWAP ¿ S, SWAP ¿ ST, and SWAPRS) or hybrid SF3/SWAPRS expression plasmid (SF2/ SWAPRS). (C) RT-PCR products and (D) graphical representation of alternatively spliced CD45 mRNA from COS cells 2 days after transient co-transfection with 0.5 ? g pSV-mini-LCA18 and 3 ? g pSG5 (control), full-length SF2 (SF2), mutant SWAP expression plasmid ( ¿ BB), mutant SF2 ( ¿ BB/SF2RS and SF2 ¿ RS) or hybrid SF2/SWAP (SF2/SWAPRS) expression plasmids. nificant effect on CD45 splicing as compared to control transfection. In particular, construct SWAP ¿ ST expresses the entire SWAP protein except the Cterminal R/S motif, but had no effect on CD45 exon 4 splicing. However, a mutant lacking most of the N terminus but retaining specifically the R/S region domain of SWAP (SW APRS) shifted CD45 splicing to the CD45RO isoform similar to full-length SWAP (Fig. 2A and B). This current work confirmed that the R/S region, but not conserved domains in the N-terminal and central part of the protein, is required for SWAP regulation of CD45 exon 4 splicing. Expression mutants of SF2 in which the R/S domain is missing or isolated showed that a different region of SF2 is responsible for stimulating CD45 exon 4 skipping. Overexpression of the SF2 mutant ¿ BB/SF2RS, in which the R/S region of SF2 was ligated to the inactive N terminus of SWAP, had little effect on CD45 exon 4 splicing as compared to full-length SF2 (Fig. 2C and D). In contrast, an SF2 mutant lacking the R/S region but retaining the RRM (SF2 ¿ RS) induced a shift to the RO isoform similar to the full-length protein (Fig. 2C and D). These data show that the RRM, but not the R/S region, are required for SF2 regulation of CD45 exon 4 splicing. To test whether the R/S region of SWAP and the RRM of SF2 could cooperatively affect alternative splicing of CD45 exon 4, a hybrid mutant composed of the SWAP R/S region and SF2 N-terminal RRM was generated (SF2/SWAPRS, Fig. 1A). The hybrid SF2/SWAPRS protein caused a marked shift to CD45RO splicing. This shift was much greater than the effects of full-length SWAP (Fig. 2A and B) or SF2 (Fig. 2C and D), and was dose dependent (Fig. 3A). To better define why this hybrid had such a marked effect on splicing of CD45 exon 4, COS cells transfected with SWAP, SF2 or SF2/SWAPRS were analyzed by immunoblotting using either an anti-SF2 mAb ( § SF2) or a rabbit antiserum to SWAP. SF2 expression changed only slightly in cells transfected with 3 ? g SF2 expression plasmid compared to cells transfected with 3 ? g control plasmid (Fig. 3B, left panel). In contrast, SWAP was highly expressed in COS cells transfected with 3 ? g SWAP expression plasmid (Fig. 3B, right panel) Eur. J. Immunol. 1999. 29: 823–837 SF2 regulates CD45 splicing during T cell activation 827 Figure 3. High level SF2/SWAPRS hybrid expression is required for high level CD45 and 4 skipping. (A) Graphical representation of alternatively spliced CD45 mRNA from COS cells 2 days after transient co-transfection with 0.5 ? g pSV-mini-LCA18 and 3 ? g pSG5 (control), SF2 ¿ RS, SF2, SWAP, or increasing concentrations (0.1, 1.0 and 3.0 ? g) of SF2/SWAPRS hybrid expression vector; and (B) expression of SWAP, SF2 and SF2/SWAPRS in COS cells. COS cells transfected in parallel with the experiment in (A) with pSG5 (control), SF2 ¿ RS, SF2, SWAP, or various concentrations of SF2/SWAPRS hybrid expression vector were immunoblotted using antibodies against SF2 ( § SF2, left panel) or SWAP ( § SWAP, right panel). compared to undetectable levels in control cells (not shown, [38]). Similar to SWAP, the SF2/SWAP hybrid protein was highly expressed in COS cells when 3 ? g of plasmid was transfected (Fig. 3B). The molecular mass of the hybrid protein was approximately 55 kDa (calculated molecular mass of the hybrid protein = 37.7 kDa), consistent with our previous observations showing that SWAP and fragments of SWAP have larger observed than predicted Mr [39]. As expected, both Ab to SF2 (Fig. 3B) and antisera to SWAP (Fig. 3B) detected the hybrid protein. As the amount of transfected SF2/ SW APRS plasmid was decreased, expression of the hybrid protein decreased. Low levels of hybrid expression (transfection of 0.1 ? g) approximately equivalent to maximum SF2 expression (transfection of 3.0 ? g pSF2) produced similar effects on the splicing of CD45 exon 4 (Fig. 3A and B). The lack of reaction of § SF2 against the SF2 ¿ RS deletion mutant may be because this antibody requires the R/S region for binding (the epitope targeted by this Ab has not been described), or because of low level expression. 2.2 SF2 specifically induces alternative splicing of CD45 exon 4 To determine the specificity of SWAP and SF2 for regulating the splicing of CD45 exon 4, we assessed the activity of SWAP and SF2 on a non-regulated exon of CD45, exon 9. CD45 minigene constructs pSV-miniLCA2, containing the regulated exon 4, and pSV-miniLCA4, containing the non-regulated exon 9 (diagrammed in Fig. 1B), were cotransfected with SWAP, SF2 or control plasmid, and splicing of RNA assayed by RT-PCR. These CD45 minigene constructs are identical except for the exon and adjacent intronic sequences inserted between exon 3 and 8 of CD45. As for pSV-mini-LCA18, splicing of exon 4 in pSV-mini-LCA2 is regulated like endogenously spliced CD45 exon 4 in B cells and thymocytes [40]. Exon 9 of pSV-mini-LCA4, however, is not regulated and is included in spliced mRNA from both B cells and thymocytes. Basal exon 4 skipping was higher from pSVLCA-4 than pSV-mini-LCA18, likely due to the altered context of the exon. Both SWAP and SF2 induced CD45 exon 4 skipping from pSV-mini-LCA2, as previously shown for pSV-mini-LCA18. However, SWAP, but not SF2, also induced skipping of CD45 exon 9 (Fig. 4A and B). These results show that the effect of SF2 is specific for CD45 exon 4, but that SWAP induces exon skipping 828 R. Lemaire et al. Eur. J. Immunol. 1999. 29: 823–837 Figure 4. SF2, but not SWAP, selectively stimulates skipping of CD45 exon 4. (A) Graphical representation of alternatively spliced CD45 mRNA from COS cells 2 days after transient co-transfection with 0.5 ? g pSV-mini-LCA2 or pSV-mini-LCA4, and 3 ? g pSG5 control plasmid or SWAP (+ SWAP). (B) Graphical representation of alternatively spliced CD45 mRNA from COS cells 2 days after transient co-transfection with 0.5 ? g pSV-mini-LCA2 or pSV-mini-LCA4, and 3 ? g pSG5 control plasmid or SF2 (+ SF2) expression vectors. of both a normally regulated exon (exon 4), and a nonregulated exon (exon 9). 2.3 SRp55 specifically inhibits alternative splicing of CD45 exon 4 The potential for other SR proteins to regulate alternative splicing of CD45 exon 4 was assessed by overexpression of corresponding cDNA in COS cells. PCRamplified cDNA of SC35, SRp40, SRp55 and SRp75 were cloned into the pSG5 eukaryotic expression vector, and each of them were transiently co-transfected into COS cells with the CD45 minigene construct, pSV-miniLCA18. Overexpression of SF2, SRp40, SRp75, and to a lesser degreee SC45, increased exon 4 skipping (Fig. 5A). In contrast, overexpression of SRp55 did not increase exon 4 skipping. The lack of effect of SRp55 was not related to a failure in SRp55 expression from the transfected construct, since Western blotting analysis using mAb104 showed overexpression of exogenous SRp55 in SRp55-transfected COS cells (Fig. 5C, right and left panels: SRp55 versus pSG5). mAb104 reacts against the R/S-rich motif of all SR proteins [41]. Similarly, SC35-transfected COS cells showed increased expression of SC35 protein (Fig. 5C, left panel, SC35 versus pSG5). Exogenous expression of SRp40 and SRp75 from their respective expression vector were not clearly increased in COS cells. SRp40 and SRp75 levels after transfection were similar to basal levels (Fig. 5C, right panel: SRp40 and SRp75 versus pSG5). This suggests that these vectors produce relatively modest amounts of protein and that, like SF2, stimulation of exon 4 skipping by SRp40 and SRp75 does not require large increases in SR protein expression. To try to inhibit SR protein expression, cDNA of SF2, SC35, SRp40 and SRp55 were cloned into pSG5 in an antisense orientation and co-transfected into COS cells with pSV-mini-LCA18. Overexpression of antisense SRp55 mRNA increased exon 4 skipping (35 % compared to 15 % in control transfections; Fig. 5A), indicating that endogenous SRp55 inhibits alternative splicing of CD45 exon 4 skipping in COS cells. Antisense cDNA of SF2, SC35, SRp40 had no effect on CD45 splicing (Fig. 5A). The lack of effect of these antisense constructs may be because they failed to down-regulate expression of the targeted mRNA or because these SR proteins do not regulate basal CD45 exon 4 splicing in COS cells. To assess SRp55 specificity on CD45 exon 4, pSV-miniLCA2 and pSV-mini-LCA4 minigene constructs (diagrammed in Fig. 1B) were co-transfected with antisense SRp55. Inhibition of SRp55 induced skipping of exon 4 (regulated exon) but not exon 9 (non-regulated exon, Fig. 5B), indicating that the effect of SRp55 is specific for the regulated exon 4 of CD45. 2.4 Activated CD45RA+ T lymphocytes overexpress both SF2 and SRp55 proteins CD45RO and CD45RA are reciprocally expressed on freshly isolated human peripheral blood T cells and define naive and memory T cell populations in humans. CD45 expression shifts from the CD45RA isoforms to the CD45RO isoform upon T cell activation [11, 12]. Potential involvement of SF2 and SRp55 in the shift from CD45RA to CD45RO isoform expression occurring during T cell activation was investigated. CD45RO+ and CD45RO– T cell populations were separated from human blood using an anti-CD45RO Ab and were activated for 3 days with PHA. Eur. J. Immunol. 1999. 29: 823–837 SF2 regulates CD45 splicing during T cell activation 829 Figure 5. SF2, SRp40 and SRp75 stimulate and SRp55 inhibits CD45 exon 4 skipping. (A) Graphical representation of alternatively spliced CD45 mRNA from COS cells 2 days after transient co-transfection with 0.5 ? g pSV-mini-LCA18 and 3 ? g control (pSG5), sense (SF2, SC35, SRp40, SRp55 or SRp75) or antisense ( § SF2, § SC35, § SRp40 or § SRp55) expression vectors. (B) Graphical representation of alternatively spliced CD45 mRNA from COS cells 2 days after transient co-transfection with 0.5 ? g pSV-mini-LCA2 or pSV-mini-LCA4, and 3 ? g pSG5 control plasmid or antisense SRp55 ( § SRp55) expression vector. (C) Expression of SC35, SRp40, SRp55 and SRp75 in COS cells. In parallel with the experiment in (A), COS cells transfected with pSG5 (control), SC35, SRp40, SRp55 or SRp75 expression vectors were immunoblotted using mAb104. Expression of the various CD45 isoforms was determined in both CD45RO+ and CD45RO– T cell populations, before and after activation, using RT-PCR. Freshly isolated CD45RO+ T cells expressed primarily the CD45RO splice variant lacking exons 4, 5 and 6, as well as the CD45RB variant lacking exons 4 and 6. CD45 variants containing exon 4 were expressed at very low levels (Fig. 6A: RO, day 0), and may represent low level contamination from B cells for CD45RA+ T cells. Activation of CD45RO+ T cells did not induce any major change in CD45 isoform expression (Fig. 6A, RO, day 3). Freshly isolated CD45RO– T cells (referred to as CD45RA+) primarily expressed the CD45ABC splice variant containing all three exons 4, 5 and 6, as well as forms containing exons 4 and 5, and 5 and 6. The CD45RO variant was not expressed by these cells (Fig. 6A: RA, day 0). Upon activation CD45RA+ T cells skipped CD45 exon 4 as expression shifted from CD45ABC (and CD45AB and CD45BC) to CD45RO and CD45RB isoforms (Fig. 6 A: RA, day 3). SR protein and SWAP expression upon T cell activation were then assessed in both CD45RO+ and CD45RA+ T cell populations using immunoblotting and RT-PCR. SF2 was not detectable in freshly isolated CD45RO+ T cells, and detectable at only a low level in freshly isolated CD45RA+ T cells (Fig. 6B, left panel). Activation of separated CD45RA+ T cells using PHA led to greatly increased expression of SF2 associated with the shift in splicing to exclusion of CD45 exon 4. Acivated CD45RO+ T cells also began to express low levels of SF2 (Fig. 6B). SF2 mRNA was similarly up-regulated. Basal SF2 mRNA expression in freshly isolated CD45RO+ and CD45RA+ T cells was very low (Fig. 7A and B). PHA activation induced a 50-fold increase in SF2 mRNA by CD45RA+ T cells, but only a modest change in SF2 expression by CD45RO+ T cells (Fig. 7A and B). Results were similar in cells from three other subjects. SR protein levels were detected using mAb104 directed against the common R/S rich motif found in all SR pro- 830 R. Lemaire et al. Eur. J. Immunol. 1999. 29: 823–837 isoform potentially generates a 210-amino acid truncated protein lacking the second RRM and the R/S region. The sequence after the deletion is out of frame with full-length SRp55. The SRp55 ¿ cDNA encodes a protein with a different 73-amino acid C terminus (Fig. 7E). In addition to increased expression of both SRp55 mRNA, activation of CD45RA+ T cells induced a fivefold shift from the alternatively spliced to the fulllength form of SRp55 (Fig. 7C and D). 3 Discussion We had shown previously that SWAP and SF2 induce skipping of exon 4 from CD45 pre-mRNA [39]. Here, we define the functional domains required for these effects. Using deletion mutants, we show that the R/S domain of SWAP is requisite and sufficient for alternative splicing of CD45 exon 4. In sharp contrast, SF2 specifically required its RRM but not its R/S domain to induce alternative CD45 exon 4 splicing. These data are consistent with previous mutational analysis of SF2 establishing a requirement for both R/S and RRM domains in constitutive splicing, but only the RRM for alternative splicing [42, 43]. Figure 6. Activated CD45RA+ T lymphocytes overexpress both SF2 and SRp55 proteins. (A) RT-PCR products of alternatively spliced CD45 mRNA from CD45RO+ and CD45RA+ T cells, before (Day 0) or after 3 day activation with 10 ? g/ml PHA (Day 3). (B) Cell lysates from the same cells shown in (A) were analyzed by immunoblot using antibodies to SF2 ( § SF2, left panel) or the R/S region (mAb104, right panel). teins [41]. SC35 and SRp40 (Fig. 6B) and SWAP (data not shown) proteins were not detectable in unstimulated or activated T cells. Both SRp55 and SRp75 were undetectable in unstimulated CD45RA+ and CD45RO+ T cells; however, they were highly expressed after activation of CD45RA+, but not CD45RO+ T cells (Fig. 6B, right panel). SRp55 mRNA expression paralleled protein levels (Fig. 7, panels C and D). Using PCR primers that encompassed the entire coding region of SRp55, a PCR fragment of the expected size (1129 bp) was generated. However, an additional 499-bp band was also generated. A radiolabeled fragment of SRp55 hybridized to both bands on Southern blot analysis (Fig. 7C). Cloning and sequencing of this novel SRp55 isoform (SRp55 ¿ ) showed a deletion of the region spanning bp 450 to 1080 of the previously cloned full-length SRp55 cDNA [31]. This shorter SRp55 We show that SF2, unlike SWAP, specifically regulates alternative splicing of CD45 exon 4. This observation supports previous work suggesting that individual SR proteins can show substrate specificity [21, 22, 44–47]. SF2 has previously been shown to promote exon inclusion [48, 49] by binding to purine-rich regions referred to as splicing enhancers within regulated exons [45, 50, 51]. SF2 and other SR proteins bind to these splicing enhancers by their RRM, which appear to bind similar or related sequences [51, 52]. Specificity likely arises from differences in sequence-specific RNA binding by RRM [45]. As SF2 specifically induced skipping of exon 4, but not exon 9 from the CD45 pre-mRNA, cis-acting elements within exon 4 are likely targets for SF2 regulation. Although SF2 and SWAP both contain an R/S domain, only SWAP regulated CD45 splicing through this domain. The role of R/S domains in constitutive and regulated splicing remains unclear. The first identified activity of these domains was to target proteins to a subnuclear “speckled” compartment [53]. However, it has been shown recently, that the R/S domain of SF2 is not required for subnuclear localization of SF2 [54]. Other data have shown that R/S regions from different proteins can interact, suggesting that by binding to U1 70K and U2AF, this motif may be important in bridging 5’ and 3’ splice sites for exon or intron definition [27, 55]. The importance of these R/S motifs in protein-protein inter- Eur. J. Immunol. 1999. 29: 823–837 SF2 regulates CD45 splicing during T cell activation 831 Figure 7. Activated CD45RA+ T lymphocytes overexpress both SF2 and SRp55 mRNAs. (A and C) mRNAs from CD45RO+ and CD45RA+ T cells, before (Day 0) or after 3 day activation with 10 ? g/ml PHA (Day 3) were amplified by RT-PCR using primers to SF2 (A) or SRp55 (C), and coamplification using primers to 18S rRNA. PCR products were separated by agarose gel electrophoresis, blotted onto nylon, and hybridized to a 32P-labeled cDNA probe for either SF2 or SRp55. After stripping, membranes were hybridized to a 32P-labeled 18S rRNA cDNA; (B and D) graphical representation of mRNA expression of SF2 (B) and SRp55 (D). (E) Diagram of the 344-amino acid full-length (SRp55) and 188-amino acid alternatively spliced (SRp55 ¿ ) forms of SRp55. Vertical numbers define the 210-amino acid region missing in SRp55 ¿ . The novel 73-amino acid sequence in the C-terminal region of SRp55 ¿ is shown as a black box. actions has been best studied for Tra regulation of doublesex (dsx) splicing. Tra appears to enhance the formation of distinct protein complexes that include Tra-2 and specific SR proteins on dsx repeat elements and the purine-rich enhancer element [56]. However, the R/S regions of U2AF, Tra and SWAP can also bind RNA, suggesting alternative functions of these motifs [56, 57] (R. Lafyatis, unpublished). Indeed, positively charged arginine residues in other proteins bind nucleic acids, presumably through ionic interactions [58]. In contrast to the sequence-specific binding of RRM though aromatic residues and hydrogen bonds [59], binding by R/S motifs is not likely sequence specific. How these non-sequencespecific interactions might affect splicing has been hypothesized for U2AF. U2AF65 is a constitutive splicing protein that binds to the branchpoint/polypyrimidine tract through interactions with its R/S motif and its RRM motif, respectively. The interaction of the R/S motif is necessary for binding of the U2 snRNP to the branchpoint. Valcarcel et al. [57] that propose the R/S domain promotes U2 snRNP binding by neutralizing negatively charged residues of the branchpoint. Our observations of non-selective stimulation of exon skipping by the SWAP R/S domain could be consistent with either a 832 R. Lemaire et al. direct non-specific interaction with pre-mRNA or interaction with other R/S proteins. The very high levels of SWAP expression required to see these effects in COS cells are not seen in any cell lines or tissues we have examined to date, suggesting that SWAP does not normally play the dominant role in regulation of CD45 exon 4 splicing in vivo. However, it may participate in cooperative binding of other sequence-specific splicing factors to regulate CD45 exon 4 splicing. Our initial hypothesis to explain the marked switch in splicing induced by the SF2/SWAPRS hybrid protein was that RRM and R/S motifs can cooperate to activate splicing. However, we found that SWAP and SF2/ SWAPRS expression are much higher than SF2 expression in transfected COS cells. These results suggested that the effect of the SWAP R/S motif on the hybrid molecule is to stabilize the hybrid, leading to much higher expression of the active region of SF2 (the RRM motifs), rather than due to a cooperative effect between the SF2 RRM and SWAP R/S. The dose response of CD45 exon 4 splicing to the hybrid protein supported this interpretation. As hybrid protein expression was decreased to levels of SF2 expressed in SF2-transfected cells, the effect of the hybrid on CD45 exon 4 splicing similarly declined to match the effect of SF2. Further supporting this interpretation, increasing the amount of transfected SF2 increases the shift in CD45 exon 4 skipping [39]. This curve never plateaus, but is limited instead by the amount of DNA that can be transfected [39]. In summary, these data suggest that CD45 exon 4 splicing is highly regulated by SF2 through its RRM, and that the degree of regulation continues to increase as SF2 expression increases beyond that achievable by the SF2 expression vector in COS cells. The R/S region of SWAP appears to confer increased protein or mRNA stability compared to SF2 with its native R/S domain. Previous studies have suggested that SR proteins coordinately regulate splicing patterns in cells through changes in the relative concentrations of the different SR proteins [22, 24]. We show that the shift from CD45RA to CD45RO isoforms upon T cell activation is potentially regulated by such a complex network of splicing factors. Not only SF2 and SWAP, but also SRp40, SRp75, and to a lesser degree SC35 induced skipping of exon 4 from the CD45 pre-mRNA. Since CD45 splicing is likely regulated by multiple SR proteins in the cell, understanding the contribution of individual SR proteins on the net splicing patterns cannot be completely assessed through overexpression of each factor. Further insight into which SR proteins are controlling splicing in COS or other cell types requires the development of reagents or techniques to specifi- Eur. J. Immunol. 1999. 29: 823–837 cally inhibit the function of individual SR proteins. Overexpression of antisense SRp55 stimulated CD45 exon 4 skipping, indicating that SRp55 promotes exon 4 inclusion. The effect of SRp55 on CD45 exon 4 splicing was specific, as antisense SRp55 did not inhibit splicing of CD45 exon 9. SRp55 similarly regulates inclusion of the alternative spliced exon 5 of the cardiac troponin T premRNA. This regulation requires an exonic splicing enhancer that binds to SRp55 [51]. The Drosophila SRp55 gene, B52, also binds RNA with sequence specificity [60], and is required for viability of flies [33, 61]. The marked increase in SF2 protein and mRNA expression seen upon T cell activation are similar in magnitude to the difference between basal SF2 expression and SF2/SW APRS hybrid protein expression in transfected COS cells. In COS cells, high level hybrid protein expression markedly shifts splicing from CD45 exon 4 inclusion to skipping. As discussed above, this is likely due to the SF2 RRM. These data strongly suggest that SF2 is responsible for the shift in splicing of CD45 during T cell activation. Unlike SC35 and SRp40, SRp75 levels also increase and may contribute to exon 4 skipping in activated T cells. Since our transfection assays indicated that SRp55 promotes CD45 exon inclusion, increased expression of SRp55 is not likely determining the pattern of CD45 exon 4 splicing during T cell activation. Our data also show that memory and activated T cells express dramatically different SR proteins despite splicing CD45 similarly. As the CD45RO mRNA has a relatively short t1/2 estimated at 3.5 h [13], constant expression of this isoform in CD45RA-CD45RO+ memory T cells requires continuing expression of factors regulating CD45 splicing. The fact that SF2 is expressed only at a low level in freshly isolated CD45RO+ T cells suggests that SF2 regulates CD45 exon 4 splicing only during T cell activation. Distinct splicing factors possibly regulate CD45 exon 4 splicing in memory T cells. Low level SRp55 expression may contribute to CD45 exon 4 skipping in these cells, but this SR protein is also expressed at low levels in CD45RA+ naive T cells. Other splicing factors and SR protein phosphorylation likely contribute to the difference in CD45 splicing in memory versus naive T cells. The changes in SF2, SRp55 and SRp72 during T cell activation may also be important in regulating other alternatively spliced genes, including Fas, IL-2 and IL-4 [62–65]. We report a new alternatively spliced form of SRp55 that lacks one RRM and its C-terminal R/S motif. This alternatively spliced form is nearly as prevalent as the fulllength form in resting T cells. Splicing is shifted to predominantly the full-length form during T cell activation. This form would not be identified using the only Ab to the Eur. J. Immunol. 1999. 29: 823–837 SR proteins available, since its lacks the R/S motifs recognized by mAb104. Further work will be required to identify its role in regulating splicing. However, its structure suggests that it might regulate the activity of the fulllength SRp55 through a dominant negative effect. During a typical primary immune response, a rapid expansion of antigen-specific T cells is followed by elimination of most of these cells through programmed cell death [66]. T cell memory is apparently due to the survival of a small fraction of these cells. Why certain cells are destined for death versus memory is not known [66]. Although development of T cell memory has been best studied in various mouse models, the pattern of CD45 splicing associated with T cell activation and memory is quite different between mouse and humans [9]. Thus, it is difficult to extrapolate the observation in mice that changes in CD45RB expression do not correlate closely with the development of memory [67]. Recent studies of human T cells continue to support the association between CD45 isoform expression and T cell memory, although a subpopulation of CD45RA+CD27– cells within the CD8+ subset can be distinguished that has cytotoxic effector functions [14]. Our studies reinforce the idea that activated and memory populations in humans are distinct. Despite the observation of CD45RO expression by both these cell types, the factors regulating expression of CD45 in each case are dramatically different. Defining how CD45 splicing is regulated in memory cells may help define the molecular basis for persistence of these cells following an immune response. SF2 regulates CD45 splicing during T cell activation 833 using the same pair of primers for sense and antisense forms, cloning the amplified fragments into the pCRTMII vector (TA cloning® kit, Invitrogen), digesting the fragments with EcoRI, and ligating the fragments into the EcoRI site of pSG5 in sense or antisense orientation. All constructs were sequenced to ensure that the proper cDNA had been amplified. SWAP expression deletion mutants pSWAP ¿ S, pSWAP ¿ ST and p ¿ BB were obtained by restriction enzyme digestion of pSWAP12exp with SacI; StuI and BamHI, respectively, or BstEII and BGlII. Resulting fragments were gel purified and ligated directly (pSWAP ¿ S), or for construction of pSWAP ¿ ST and p ¿ BB, digested plasmids were blunted using Klenow enzyme prior to ligation. The SWAP expression deletion mutant pSWAPRS was obtained by PCR amplification of the SWAP R/S region from pSWAP12exp using a 5’ oligonucleotide tailed with BstEII (5’SWAPRS). The amplified fragment was digested with BstEII and ligated into the BstEII site of pSWAP12exp. The SF2 expression deletion mutant pSF2 ¿ RS was obtained by restriction enzyme digestion of pSF2 with BglII, and religation of the plasmid. SF2 and SWAP hybrid mutants ¿ BB/SF2RS and SF2/ SWAPRS were obtained by amplifying the R/S region from pSF2 and pSWAP12exp by PCR using oligonucleotides tailed with BamHI (5’SF2RS and 3’SF2RS) for SF2, and with BglII (5’SWAPRS) for SWAP, digesting the amplified fragments with BamHI and BglII, respectively, and ligating the SF2 R/S region fragment into the BamHI site of p ¿ BB, and the SWAP R/S region fragment into the BglII site of pSF2 ¿ RS. 4 Materials and methods 4.2 Oligonucleotide sequences 4.1 Plasmid constructs 5’SF2:5’-TGAATTCGTCACCGCCATGTCG-3’; 3’SF2:5’-AT GGATCCAATCATCTTATGTACGAGA-3’; 5’ § SF2: 5’-CTGG ATCCGTCACCGCCATGTCGG-3’; 3’ § SF2:5’-TGAATTCAAT CATCTTATGTACGAG-3’; 5’SF2RS: 5’-CTGGATCCAGGT TACCATATGACCTATGCAGTTCG-3’; 5’SWAPRS: 5’-CTAG ATCTGGTTACCCTTCAAACAGATGCAGAG-3’; 3’SWAPRS: 5’-GGCATGAGATCCTGAGTT-3’; 5’SC35: 5’-TGAATTCAG AGCTATGAGCTACG-3’; 3’SC35: 5’-CTGGATCCGATGGA CTATGTGGTCC-3’; 5’ § SC35: 5’-CTGGATCCAGAGCTATG AGCTACG-3’; 3’ § SC35: 5’-TGAATTCGATGGACTATGTGG TC-3’; 5’SRp75: 5’-TGAATTCAGCCATCACTGCCGTTG-3’; 3’SRp75: 5’-CTGGATCCAGCTGTGGCCATAGC-3’; 5’SRp 55: 5’-ATGCCGCGCGTCTACATA-3’; 3’SRp55: 5’-ATTC CTCTTGCAGAAGAG-3’; 5’SRp40: 5’-CGTCCGGAAGTAC TAGC-3’; 3’SRp40: 5’GTGAGCAGAAGGCTTATC-3’; 45X8: 5’-CACATGTTGGCTTAGATGG-3’; LCA-MCS; 5’-TGCAGG TCGACTCTAGAG-3’. The underlined regions of these oligonucleotides are restriction enzyme sites. Plasmid pSWAP12exp encoding the full-length SWAP cDNA was prepared by cloning the EcoRI fragment of pSWAP12 (full-length mouse SWAP insert [38]) into pSG5 (Stratagene). Plasmids pSF2, pSC35 and SRp75 encoding, respectively, full-length SF2, SC35 and SRp75 were prepared by amplifying corresponding regions from human HeLa cell cDNA by PCR using 5’ oligonucleotides tailed with EcoRI (5’SF2, 5’SC35 and 5’SRp75, respectively), and 3’ oligonucleotides tailed with BamHI (3’SF2, 3’SC35 and 3’SRp75, respectively), digesting the amplified fragments with EcoRI and BamHI, purifying the fragments after separation on an agarose gel (GeneClean Bio-101), and ligating the fragments into the EcoRI/BamHI site of pSG5. Plasmids encoding antisense forms of SF2 and SC35 were prepared similarly, except that 5’ oligonucleotides were tailed with BamHI (5’ § SF2 and 5’ § SC35, respectively) and 3’ oligonucleotides with EcoRI (3’ § SF2 and 3’ § SC35) sites. Plasmids encoding sense and antisense SRp55 and SRp40 were obtained by amplifying SRp55 and SRp40 from human HeLa cell cDNA 834 R. Lemaire et al. 4.3 Cell culture, transfection and RNA extraction COS-7 cells were obtained from the American type Culture Collection (Rockville, MD), and cultured in DMEM supplemented with 10 % FBS, penicillin/streptomycin and glutamine (culture medium). Cells were transfected at 70–80 % confluence by adding 18 ? l of lipofectamine (Gibco-BRL) diluted in 82 ? l Opti-MEM mixed with 3 ? g DNA diluted in 82 ? l Opti-MEM. After 5 h, the medium was changed to culture medium. Total RNA was prepared from cells after 2 days by direct lysis of the cells by the addition of 400 ? l RT-lysis buffer and purification according to the procedure described (RNeasy Total RNA kit, Qiagen) for isolation of total RNA. 4.4 cDNA preparation and PCR For preparation of cDNA, 3 ? l RNA (0.5–1.0 ? g), 1 ? l oligo(dT)12–18 (0.5 ? g/ ? l) and 8 ? l water were heated to 70 °C for 10 min, and cooled on ice for 1 min. To this mixture 4 ? l 5X RT buffer, 2 ? l 100 mM DTT, 1 ? l 10 mM nucleotide triphosphates (Pharmacia), and 1 ? l Superscript II RNase Reverse Transcriptase (200 U/ ? l, Gibco-BRL) were added. After a 1-h- incubation at 42 °C, the cDNA was heated to 95 °C for 5 min. For analysis of CD45 exon 4, reverse-transcribed cDNA were amplified using a 5’ oligonucleotide in the second exon (LCA-MCS) and a 3’ primer in the eighth exon (45X8). PCR reactions (25 ? l) were composed of 2.5 ? l 10X PCR buffer, 25 ng of each oligonucleotide, 2.5 ? l 1.25 mM nucleotide triphosphates, 2 ? l 50 mM MgCl2, 0.2 U Taq polymerase (Gibco-BRL) and 1 ? l template cDNA. Amplification conditions were 96 °C for 45 s; 59 °C for 1 min; 30 cycles. PCR amplification of SR proteins was performed similarly to that of CD45. All experiments shown of COS cell transfection and PCR are representative of at leat three experiments showing similar results. Data were quantified by scanning of photographs and densitometric analysis using NIH image 1.5.2. software. 4.5 Purification of CD45RO+ and CD45RO– T lymphocytes Human blood from healthy volunteers was collected in heparinized tubes (50 ml) in accordance with an approved protocol. Mononuclear cells were purified by separation on FicollPaque (Pharmacia). The resulting cells were washed two times in PBS supplemented with 1.0 mg/ml BSA, resuspended in 13 ml DMEM supplemented with 10 % FBS (complete medium) and incubated in a tissue culture flask. After 1 h, nonadherent cells were removed, recovered by centrifugation, resuspended in 0.8 ml cold complete medium with 200 ? l mouse anti-human CD40 (Serotec clone B-B20), and incubated for 30 min on a rotator at 4 °C. Following this incubation the cells were washed with PBS and resuspended in 1.0 ml complete medium. Goat anti-mouse-coated Dyna- Eur. J. Immunol. 1999. 29: 823–837 beads (200 ? l; M-450, approximately 4 × 108 beads/ml) were washed in PBS and added to the cell suspension in 1.0 ml cold complete medium. Cells and beads were incubated for 30 min on a rotator at 4 °C, and then B cell/bead complexes removed with a magnet and discarded. The remaining cells were collected by centrifugation, resuspended in 0.5 ml complete medium with 20 ? l anti-human CD45RO (Dako, clone UCHL1) and incubated for 30 min on a rotator at 4 °C. Following this incubation, the cells were washed with PBS, resuspended in 0.5 ml complete medium, 200 ? l goat antimouse-coated Dynabeads were added and incubation continued as above. CD45RO+ T cell/bead complexes were removed with a magnet and both CD45RO+ and CD45RO– cells either lysed immediately for PCR and Western blot analysis or cultured in complete medium supplemented with 10 ? g/ml PHA. 4.6 SDS-PAGE and immunoblotting Samples for SDS-PAGE were prepared from cell cultures by direct lysis in 500 ? l 1.2 X PAGE sample buffer. Samples were heated to 95 °C for 5 min, loaded on an SDS-PAGE (10 % polyacrylamide) gel, and run at 40 mA for 3 h [68]. For immunoblotting, proteins were transferred to nitrocellulose (Transblot, Bio-Rad: transfer buffer: 25 mM Tris, 190 mM glycine, 20 % methanol) overnight at 15 mA. Nitrocellulose filters were blocked in TNT buffer (10 mM Tris pH 8.0, 0.15 M NaCl, 0.05 % Tween-20) plus 5 % low-fat milk for 1–3 h at 4 °C. Blotted proteins were incubated with the primary antibody diluted in TNT buffer plus 0.5 % low-fat milk. Primary antibody consisted of a 1:200 dilution of rabbit antisera to SWAP, a 1:5 dilution of mAb104 partially purified by NH4SO4 precipitation from hybridoma culture supernatants (hybridoma obtained from ATCC: mAb104 recognizes a phosphorylated epitope within the R/S domain [41]), or a 1:10 dilution of hybridoma tissue culture supernatant of § SF2 (kindly provided by A. Krainer). After 1 h, blots were washed with TNT buffer 5 times for 5 min, and incubated with the secondary antibody (1:2000 dilution of alkaline phosphataseconjugated goat anti-rabbit for antisera to SWAP, or either alkaline phosphatase- or horseradish peroxidaseconjugated rabbit anti-mouse for mAb104 or § SF2). After 45 min incubation with the secondary antibody, blots were washed 5 times for 5 min, and developed. Alkaline phosphatase reaction was developed with 5-bromo-4-chloro-3indolyl phosphate-toluidine salt/p-nitro blue tetrazolium chloride in developing buffer (100 mM NaCl, 5 mM MgCl2, 100 mM Tris, pH 9.5). Horseradish peroxidase reaction was developed using enhanced chemiluminescence (ECL) Detection reagents (Amersham) followed by autoradiography. Acknowledgments: We thank Dr. A. Krainer for generously providing mAb § SF2. We thank all the members of the Arthritis Section for critical review of the work in progress. We thank Drs. A. Marshak-Rothstein and F. Denhez for Eur. J. Immunol. 1999. 29: 823–837 review of the manuscript. Support for this work in this manuscript was provided by grants to R. Lafyatis from the Arthritis Foundation and from the National Institutes of Health (Shannon Award), and to R. Lemaire from Delegation a la Recherche (CHRU, Lille, France). 5 References 1 Charbonneau, H., Tonks, N. K., Walsh, K. A. and Fischer, E. H., The leucocyte common antigen (CD45): a putative receptor-linked protein tyrosine phosphatase. Proc. Natl. Acad. Sci. USA 1988. 85: 7182–7186. 2 Pingel, J. T. and Thomas, M. L., Evidence that the leukocyte-common antigen is required for antigeninduced T-lymphocyte proliferation. Cell 1989. 58: 1055–1065. 3 Hurley, T. R., Hyman, R. and Sefton, B. M., Differential effects of expression of the CD45 tyrosine protein phosphatase on the tyrosine phosphorylation of the lck, fyn, and C-src tyrosine protein kinases. Mol. Cell. Biol. 1993. 13: 1651–1656. 4 McFarland, E. D. C., Hurley, T. R., Pingel, J. T., Sefton, B. M., Shaw, A. and Thomas, M., Correlation between Src family member regulation by the protein-tyrosinephosphatase CD45 and transmembrane signaling through the T-cell receptor. Proc. Natl. Acad. Sci. USA 1993. 90: 1402–1406. 5 Hovis, R., Donovan, J. A., Musci, M. A., Motto, D. G., Goldman, F. D., Ross, S. E. and Koretzky, G. A., Rescue of signaling by a chimeric protein containing the cytoplasmic domain of CD45. Nature 1993. 260: 544–546. 6 Volarevic, S., Niklinska, B., Burns, C. M., June, C. H., Weissman, A. W. and Ashwell, J. D., Regulation of TCR signaling by CD45 lacking transmembrane extracellular domains. Nature 1993. 260: 541–544. 7 Benveniste, P., Takahama, Y., Wiest, D., Nakayama, T., Sharrow, S. O. and Singer, A., Engagement of the external domains of CD45 tyrosine phosphatase can regulate the differentiation of immature CD4+CD8+ into mature T cells. Proc. Natl. Acad. Sci. USA 1994. 91: 6933–6937. 8 Trowbridge, I. S. and Thomas, M. L., CD45: An emerging role as a protein tyrosine phosphatase required for lymphocyte activation and development. Annu. Rev. Immunol. 1994. 12: 85–116. 9 Rogers, P. R., Pilapil, S., Hayakawa, K., Romain, P. L. and Parker, D. C., CD45 alternative exon expression in murine and human CD4+ T cell subsets. J. Immunol. 1992. 148: 4054–4065. 10 Ratech, H., Denning, S. and Kaufman, R., An analysis of alternatively spliced CD45 mRNA transcripts during T SF2 regulates CD45 splicing during T cell activation 835 cell maturation in humans. Cell. Immunol. 1997. 177: 109–118. 11 Akbar, A. N., Terry, L., Timms, A., Beverly, P. C. L. and Janossy, G., Loss of CD45R and gain of UCHL1 is a feature of primed T cells. J. Immunol. 1988. 140: 2171–2178. 12 Birkeland, M. L., Johnson, P., Trowbridge, I. S. and Pure, E., Changes in CD45 isoform accompany antigeninduced murine T-cell activation. Proc. Natl. Acad. Sci. USA 1989. 86: 6734–6738. 13 Deans, J. P., Serra, H. M., Shaw, J., Shen, Y. J., Torres, R. M. and Pilarski, L., Transient accumulation and subsequent rapid loss of messenger RNA encoding high molecular mass CD45 isoforms after T-cell activation. J. Immunol. 1992. 148: 1898–1905. 14 Hamann, D., Baars, P. A., Rep, M. H., Hooibrink, B., Kerkhof-Guarde, S. R., Klein, M. R. and van Lier, R. A. W., Phenotypic and functional separation of memory and effector human T-cells. J. Exp. Med. 1997. 186: 1407–1418. 15 McKenney, D. W., Onodera, H., Gorman, L., Mimura, T. and Rothstein, D. M., Distinct isoforms of the CD45 protein-tyrosine phosphatase differentially regulate interleukin 2 secretion and activation signal pathways involving Vav in T cells. J. Biol. Chem. 1995. 270: 24949–24954. 16 Chui, D., Ong, C. J., Johnson, P., Teh, H.-S. and Marth, J. D., Specific CD45 isoforms differentially regulate T cell receptor signaling. EMBO J. 1994. 13: 798–807. 17 Kozieradzki, I., Kundig, T., Kishihara, K., Ong, C. J., Chiu, D., Wallace, V. A., Kawai, K., Timms, E., Ionescu, J., Ohashi, P., Marth, J. D., Mak, T. W. and Penninger, J. M., T cell development in mice expressing splice variants of the protein tyrosine kinase phosphatase CD45. J. Immunol. 1997. 158: 3130–3139. 18 Green, M. R., Biochemical mechanisms of constitutive and regulated pre-mRNA splicing. Annu. Rev. Cell. Biol. 1991. 7: 559–599. 19 Guthrie, C., Messenger RNA splicing in yeast: clues to why the spliceosome is a ribonucleoprotein. Science 1991. 253: 157–163. 20 Lamm, G. M. and Lamond, A. I., Non-snRNP protein splicing factors. Biochim. Biophys. Acta 1993. 1173: 247–265. 21 Fu, X.-D., Specific commitment of different pre-mRNAs to splicing by single SR proteins. Nature 1993. 365: 82–85. 22 Zahler, A. M., Neugebauer, K. M., Lane, W. S. and Roth, M. B., Distinct functions of SR proteins in alternative pre-mRNA splicing. Science 1993. 260: 219–222. 836 R. Lemaire et al. 23 Zahler, A. M. and Roth, M. B., Distinct functions of SR proteins in recruitment of U1 small nuclear ribonucleoprotein to alternative 5’ splice sites. Proc. Natl. Acad. Sci. USA 1995. 92: 2642–2646. 24 Zahler, A. M., Lane, W. S., Stolk, J. A. and Roth, M. B., SR proteins: a conserved family of pre-mRNA splicing factors. Genes Dev. 1992. 6: 837–847. Eur. J. Immunol. 1999. 29: 823–837 36 Chou, T.-B., Zachar, Z. and Bingham, P. M., Developmental expression of a regulatory gene is programmed at the level of splicing. EMBO J. 1987. 6: 4095–4104. 37 Zachar, Z., Chou, T.-B. and Bingham, P. M., Evidence that a regulatory gene autoregulates splicing of its transcript. EMBO J. 1987. 6: 4105–4111. 25 Zahler, A., Neugebauer, K. M., Stolk, J. A. and Roth, M. B., Human SR proteins and isolation of a cDNA encoding SRp75. Mol. Cell. Biol. 1993. 13: 4023–4028. 38 Denhez, F. and Lafyatis, R., Conservation of regulated alternative splicing and identification of functional domains in vertebrate homologs to the drosophila splicing regulator, suppressor-of-white-apricot. J. Biol. Chem. 1994. 269: 16170–16179. 26 Tacke, R. and Manley, J. L., The human splicing factors ASF/SF2 and SC35 possess distinct, functionally significant RNA binding specificities. EMBO J. 1995. 14: 3540–3551. 39 Sarkissian, M., Winne, A. and Lafyatis, R., Mammalian SWAP regulation of alternative mRNA splicing of CD45 exon 4 and fibronectin IIICS. J. Biol. Chem. 1996. 271: 31106–31114. 27 Kohtz, J. D., Jamison, S. F., Will, C. L., Zuo, P., Lurhmann, R., Garcia-Blanco, M. A. and Manley, J. L., Protein-protein interactions and 5’ splice site recognition in mammalian mRNA precursors. Nature 1994. 368: 119–124. 28 Ge, H., Zuo, P. and Manley, J. L., Primary structure of the human splicing factor ASF reveals similarities with Drosophila regulators. Cell 1991. 66: 373–382. 29 Krainer, A. R., Mayeda, A., Kozak, D. and Binns, G., Functional expression of cloned human splicing factor SF2: homology to RNA-binding proteins, U1 70K, and drosophila splicing regulators. Cell 1991. 66: 383–394. 30 Fu, X.-D. and Maniatis, T., Isolation of a complementary DNA that encodes the mammalian splicing factor SC35. Science 1992. 256: 535–538. 31 Screaton, G. R., Caceres, J. F., Mayeda, A., Bell, M. V., Plebanski, M., Jackson, D. G., Bell, J. I. and Krainer, A. R., Identification and characterization of three members of the human SR family of pre-mRNA splicing factors. EMBO J. 1995. 14: 4336–4349. 32 Diamond, R. H., Du, K., Lee, V. M., Mohn, K. L., Harber, B. A., Tewari, D. S. and Taub, R., Novel delayedearly and highly insulin-induced growth response gene. Identification of HR, a potential regulator of alternative splicing. J. Biol. Chem. 1993. 268: 15185–15192. 33 Ring, H. Z. and Lis, J. T., The SR protein B52/SRp55 is essential for drosophila development. Mol. Cell. Biol. 1994. 14: 7499–7506. 34 Cavaloc, Y., Popielarz, M., Fuchs, J. P., Gattoni, R. and Stevenin, J., Characterization and cloning of the human splicing factor 9G8: a novel 35 KDa factor of the serine/arginine protein family. EMBO J. 1994. 13: 2639–2649. 35 Baker, B. S., Sex in flies: the splice of life. Nature 1989. 340: 521–524. 40 Streuli, M. and Saito, H., Regulation of tissue-specific alternative splicing: exon-specific cis-elements govern the splicing of leukocyte common antigen pre-mRNA. EMBO J. 1989. 8: 787–796. 41 Roth, M. B., Murphy, C. and Gall, J. G., A conserved family of nuclear phosphoproteins localized to sites of polymerase transcription. J. Cell. Biol. 1991. 115: 587–596. 42 Caceres, J. F. and Krainer, A. R., Functional analysis of pre-mRNA splicing factor SF2/ASF structural domain. EMBO J. 1993. 12: 4715–4726. 43 Zuo, P. and Manley, J. L., Functional domains of human splicing factor ASF/SF2. EMBO J. 1993. 1993: 4727–4737. 44 Kim, Y. G., Zuo, P., Manley, J. L. and Baker, B. S., The drosophila RNA-binding protein RBP1 is localised to transcriptionally active sites of chromosomes and shows a functional similarity to human splicing factor. Genes Dev. 1992. 6: 2569–2579. 45 Sun, Q., Mayeda, A., Hampson, R. K., Krainer, A. R. and Rottman, F. M., General splicing factor SF2/ASF promotes alternative splicing by binding to an exonic splicing enhancer. Genes Dev. 1993. 7: 2598–2608. 46 Tian, M. and Maniatis, T., A splicing enhancer complex controls alternative splicing of doublesex pre-mRNA. Cell 1993. 74: 105–114. 47 Staknis, D. and Reed, R., SR proteins promote the first specific recognition of pre-mRNA and are present together with the U1 small nuclear ribonucleoprotein particle in a general splicing enhancer complex. Mol. Cell. Biol. 1994. 14: 7670–7682. 48 Mayeda, A., Helfman, D. M. and Krainer, A. R., Modulation of exon skipping and inclusion by heterogenous nuclear ribonucleoprotein A1 and pre-mRNA splicing factor SF2/ASF. Mol. Cell. Biol. 1993. 13: 2993–3001. 49 Caceres, J. F., Stamm, S., Helfman, D. M. and Krainer, A. R., Regulation of alternative splicing in vivo by overexpression of antagonistic splicing factors. Science 1994. 265: 1706–1709. Eur. J. Immunol. 1999. 29: 823–837 50 Lavigueur, A., La Branche, H., Kornblihtt, A. R. and Chabot, B., A splicing enhancer in the human fibronectin alternate ED1 exon interacts with SR proteins and stimulates U2 snRNP binding. Genes Dev. 1993. 7: 2405–2417. 51 Ramchatesingh, J., Zahler, A. M., Neugebauer, K. M., Roth, M. B. and Cooper, T. A., A subset of SR proteins activates splicing of the cardiac troponin T alternative exon by direct interactions with an exonic enhancer. Mol. Cell. Biol. 1995. 15: 4898–4907. 52 Tanaka, K., Watakabe, A. and Shimura, Y., Polypurine sequences within a downstream exon function as a splicing enhancer. Mol. Cell. Biol. 1994. 14: 1347–1354. 53 Li, H. and Bingham, P. M., Arginine/serine-rich domains of the su(wa) and tra processing regulators target proteins to a subnuclear compartment implicated in splicing. Cell 1991. 67: 335–342. 54 Caceres, J. F., Misteli, T., Screaton, G., Spector, D. L. and Krainer, A. R., Role of the modular domains of SR proteins in subnuclear localization and alternative splicing specificity. J. Cell. Biol. 1997. 138: 225–238. 55 Wu, J. Y. and Maniatis, T., Specific interactions between proteins implicated in splice site selection and regulated alternative splicing. Cell 1993. 75: 1061–1070. 56 Lynch, K. W. and Maniatis, T., Assembly of specific SR protein complexes on distinct regulatory elements of the Drosophila doublesex splicing enhancer. Genes Dev. 1996. 10: 2089–2101. 57 Valcarcel, J., Gaur, R. K., Singh, R. and Green, M. R., Interaction of U2AF65 RS region with the pre-mRNA of branch point and promotion of base pairing with U2 snRNA. Science 1996. 273: 1706–1709. 58 Burd, C. G. and Dreyfuss, G., Conserved structures and diversity of functions of RNA-binding proteins. Science 1994. 265: 615–621. 59 Oubridge, C., Ito, N., Evans, P. R., Teo, C. H. and Nagai, K., Crystal structure at 1.92 A resolution of the RNA-binding domain of the U1A spliceosomal protein complexed with an RNA hairpin. Nature 1994. 372: 432–438. SF2 regulates CD45 splicing during T cell activation 837 60 Shi, H., Hoffman, B. E. and Lis, J. T., A specific RNA hairpin loop structure binds the RNA recognition motifs of the drosophila SR protein B52. Mol. Cell. Biol. 1997. 17: 2649–2657. 61 Kraus, M. E. and Lis, J. T., The concentration of B52, an essential splicing factor and regulator of splice choice in vitro, is critical for Drosophila development. Mol. Cell. Biol. 1994. 14: 5360–5370. 62 Cheng, J., Zhou, T., Liu, C., Shapiro, J. P., Brauer, M. J., Kiefer, M. C., Barr, P. J. and Mountz, J. D., Protection from Fas-mediated apoptosis by a soluble form of the Fas. Science 1994. 263: 1759–1762. 63 Liu, C., Cheng, J. and Mountz, J. D., Differential expression of human Fas mRNA species upon peripheral blood mononuclear cell activation. Biochem. J. 1995. 310: 957–963. 64 Tsytsikov, V. N., Yurovsky, V. V., Atamas, S. P., Alms, W. J. and White, B., Identification and characterization of two alternative splice variants of human interleukin-2. J. Biol. Chem. 1996. 271: 23055–23060. 65 Alms, W. J., Atamas, S. P., Yurovsky, V. V. and White, B., Generation of variant human interleukin-4 by alternative splicing. Mol. Immunol. 1996. 33: 361–370. 66 Sprent, J., Immunological memory. Curr. Opin. Immunol. 1997. 9: 371–379. 67 Zimmermann, C., Brduscha-Riem, K., Blaser, C., Zinkernagel, R. M. and Pircher, H., Visualization, characterization, and turnover of CD8+ memory T cells in virusinfected host. J. Exp. Med. 1996. 183: 1367–1375. 68 O’Farrell, P. H., High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 1975. 250: 4007–4021. Correspondence: Robert Lafyatis, Boston University School of Medicine, Arthritis Center K-5, 80 E. Concord street, Boston, MA 02118, USA Fax: +1-617-638 5226 e-mail: rlafyatis — med-med1.bu.edu