Regulation of Alternative Splicing of CD45 by Antagonistic Effects of SR Protein Splicing Factors

Transcription

1 This information is current as of August 23, References Subscription Permissions Alerts Regulation of Alternative Splicing of CD45 by Antagonistic Effects of SR Protein Splicing Factors Gerdy B. ten Dam, Christian F. Zilch, Diana Wallace, Bé Wieringa, Peter C. L. Beverley, Lambert G. Poels and Gavin R. Screaton J Immunol 2000; 164: ; ; doi: /jimmunol This article cites 67 articles, 42 of which you can access for free at: Why The JI? Submit online. Rapid Reviews! 30 days* from submission to initial decision No Triage! Every submission reviewed by practicing scientists Fast Publication! 4 weeks from acceptance to publication *average Information about subscribing to The Journal of Immunology is online at: Submit copyright permission requests at: Receive free -alerts when new articles cite this article. Sign up at: Downloaded from by guest on August 23, 2018 The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD Copyright 2000 by The American Association of Immunologists All rights reserved. Print ISSN: Online ISSN:

2 Regulation of Alternative Splicing of CD45 by Antagonistic Effects of SR Protein Splicing Factors 1 Gerdy B. ten Dam, 2 * Christian F. Zilch, Diana Wallace, Bé Wieringa,* Peter C. L. Beverley, Lambert G. Poels, and Gavin R. Screaton CD45 is a transmembrane glycoprotein possessing tyrosine phosphatase activity, which is involved in cell signaling. CD45 is expressed on the surface of most leukocytes and can be alternatively spliced by the inclusion or skipping of three variable exons (4, 5, and 6 or A, B, and C) to produce up to eight isoforms. In T cells, the splicing pattern of CD45 isoforms changes after activation; naive cells express high m.w. isoforms of CD45 which predominantly express exon A (CD45RA), whereas activated cells lose expression of exon A to form low m.w. isoforms of CD45 including CD45RO. Little is known about the specific factors controlling the switch in CD45 splicing which occurs on activation. In this study, we examined the influence of the SR family of splicing factors, which, like CD45, are expressed in tissue-specific patterns and have been shown to modulate the alternative splicing of a variety of transcripts. We show that specific SR proteins have antagonistic effects on CD45 splicing, leading either to exon inclusion or skipping. Furthermore, we were able to demonstrate specific changes in the SR protein expression pattern during T cell activation. The Journal of Immunology, 2000, 164: Intron removal is an essential step during eukaryotic gene expression. This process requires recognition of the 5 and 3 splice sites by the spliceosome: a multicomponent ribonucleoprotein complex. The major components of the spliceosome are the small ribonucleoprotein particles (snrnps) 3 U1, U2, and U4/U6 and the non-snrnp proteins including the family of SR proteins (reviewed by Green (3) and Kramer (4)). This family of closely related, highly conserved proteins is characterized by the presence of one or two N-terminal RNA recognition motifs (RRMs) and a C-terminal domain rich in arginines and serines (RS domain). SR proteins are essential for constitutive splicing (5 10) and alternative splicing (10 17). The family is highly conserved between diverse species and have been shown to be essential for cell survival (18, 19). SR proteins function at multiple steps during the splicing reaction. In early steps, they facilitate U1 snrnp binding to the 5 splice site (20, 21). Furthermore, they stabilize complex assembly at the 3 splice site by assisting the binding of U2AF (22) and by forming bridges to connect 5 and 3 splice sites (23 25). Substrate-specific effects of individual SR proteins have been demonstrated in constitutive as well as alternative splicing (7, 11, Departments of *Cell Biology and Anatomy, Faculty of Medical Sciences, University of Nijmegen, Nijmegen, The Netherlands; Imperial Cancer Research Fund Tumour Immunology Unit, University College London Medical School, London, United Kingdom; The Edward Jenner Institute for Vaccine Research, Compton, Newbury, United Kingdom; and Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford, United Kingdom Received for publication November 1, Accepted for publication March 1, The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by a European Molecular Biology Organization short-term fellowship (to G.B.t.D.), a Deutscher Akademischer Austauschdienstgrant (to C.F.Z.), the Medical Research Council and the Wellcome Trust (to G.R.S.), and the Imperial Cancer Research Fund. 2 Address correspondence and reprint requests to Dr. Gerdy B. ten Dam at her current address: Department of Biochemistry, Trigon Building, Faculty of Medical Sciences, University of Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. address:g.tendam@bioch.kun.nl 3 Abbreviations used in this paper: SR, serine(s)-arginine(r)-rich; RRM, RNA recognition motif; DAF, decay accelerating factor. 12, 14, 26, 27). SR proteins are expressed in tissue-specific patterns (8, 10, 14, 26) and can be regulated by phosphorylation (28). The effects of the SR proteins SF2/ASF can be antagonized by the heterogeneous ribonucleoproteins A1 and A2B1 (29, 30). In addition, a further level of control of SR protein function may be subserved by nucleocytoplasmic shuttling (31, 32). CD45 is a transmembrane glycoprotein expressed on leukocytes (reviewed by Thomas (1) and Trowbridge and Thomas (2)). Alternative splicing of three variable exons (4, 5, and 6 or A, B, and C) allows the production of eight possible isoforms. In rodent cells, all of these isoforms have been isolated (33 35), whereas in humans only five have been identified (36, 37). The splicing is cell-type specific and activation dependent. B cells express mainly the high m.w. isoform (ABC), whereas T cells express a panel of different isoform ranging form the smallest isoform (null), found also on thymocytes, to the largest isoform (ABC). T cell activation leads to a programmed shift from high to the low m.w. isoforms, i.e., down-regulation of CD45RA expression and concomitantly up-regulation of CD45RO expression. It is this change in cell surface phenotype that is used to differentiate naive from memory T cells (38). During this activation, there is also a decrease in expression of CD45RC which is higher than the small decrease observed for CD45RB but less prominent than the CD45RA decrease (39). In this study, we sought to identify factors involved in the regulation of alternative splicing of CD45. Members of the SR protein family were analyzed for their influence on splice site selection by cotransfection experiments with a CD45 minigene in COS-1 cells in vitro (40). Furthermore, we analyzed the SR protein profile in T cells before and after stimulation. Materials and Methods DNA cloning The human CD45 minigene psec-s-lca1-7 was constructed in the psec expression plasmid (41) and contains 1) the 5 CD45 structural gene region (exons 1 and 2) comprising the ATG codon and the signal peptide; 2) the genomic CD45 sequence of exon 3 to exon 7 with the alternative exons 4, 5, and 6; and 3) the decay accelerating factor (DAF) GPI anchor attachment signal to ensure membrane expression of the expressed proteins (Fig. 1A). Copyright 2000 by The American Association of Immunologists /00/$02.00

3 5288 REGULATION OF CD45 SPLICING FIGURE 1. CD45 alternative splicing. A, Diagrammatic representation of the CD45 minigene. Transcription is driven by the SV40 promoter, introns are shown as lines, numbered boxes are CD45 exons. DAF GPI attachment site. The position of the oligonucleotide primers LCA2 and LCA7 used for PCR are shown as arrows above exons 2 and 7, respectively. B, Diagram of the eight possible alternatively spliced CD45 transcripts and the predicted sizes of the PCR products after amplification with the LCA2 and LCA7 primers. The CD45 isoform expression pattern in the human and mouse is taken from Thomas (1). SR protein cdnas were cloned in the vector pcgt7 which carries an N-terminal tag from the bacteriophage T7 gene (31, 42). SR protein products can therefore be recognized by the T7 tag mab (Novagen, Madison, WI). SRp75 was cloned into the parental vector pcg, which lacks the T7 tag. The SR protein chimeras and deletion mutants are listed in Fig. 5. RRM or RS domain swaps or deletions were all constructed by PFU-PCR amplification of the desired fragments with specifically designed primers and subcloning of the fragments in the pcgt7 expression vector. (Nomenclature of the constructs is illustrated for SF2/ASF (30aRRM1 30aRRM2 30aRS) and SC35 (30bRRM1 30bRS), e.g., substitution of the RS domain of SF2/ASF by the RS domain of SC35 is called 30aRRM1 30aRRM2 30bRS, deletion of a domain, e.g., RS domain of SC35 is indicated by or, 30bRRM1-30bRS). To generate the SRp20-SC35 chimera s PCR products of SRp20-RRM (aa 1 88), SRp20-RS (aa ), SC35-RRM (aa 1 115), and SC35-RS (aa ) were subcloned in pcgt7 in the indicated combinations. The SF2-ASF-SRp40 chimeras and SF2/ASF deletion mutants were described by Caceres et al. (31, 32). In vivo analysis of alternative CD45 splicing COS-1 cells were cultured in DMEM (Life Technologies, Rockville, MD) supplemented with 10% FCS (Life Technologies) in 6-well plates. At a cell density of 30 40%, cells were (co)transfected with 1 g of minigene DNA and the indicated amount of SR protein construct using Lipofectin reagents according to the manufacturer (Life Technologies). Transfected cells were harvested h after transfection and RNA was isolated with the RNAzol B (Cinna/Biotecx Laboratories, Friendswood, TX) method and analyzed by RT-PCR. First-strand specifically primed (DAF reverse primer, 5 -CCTAAATGAAGAGCACAATTGCA- 3 ) cdna synthesized with superscript reverse Transcriptase (Life Technologies) from 200 ng of RNA was amplified for 22 cycles using a 5 end-labeled forward primer (LCA2, 5 -ATTGGATCCGCTGACTTCCA GATATGACC-3 )andanonlabeledreverseprimer(lca7,5 -CCGAGATC TTCAGAGGCATTAAGGTAGGC-3 ). PCR products were run on a 6% denaturing polyacrylamide gel and detection and quantitation were conducted by autoradiography and PhosphoImage analysis (Kodak X-OMAT S1, Molecular Analyst Bio-Rad, Hercules, CA), respectively. RNA from PHA-stimulated RO and RA T cells was reversed transcribed with the LCA9 primer (5 -GTAATCCACAGTGATGTTTGC-3 ) and amplified using LCA2 and LCA7 primers. PCR products were run on a 2% agarose gel. For immunofluorescence analysis, COS-1 cells were seeded on coverslips in 6-well plates and transfected with the CD45 minigene. Cells were stained 48 h after transfection at 4 C with the CD45 exon B-specific Ab PD7/26 (Dakopatts, Copenhagen, Denmark). SR proteins and immunoblotting To analyze SR protein products (SR proteins, chimeric constructs, and deletion mutants) expressed by the pcgt7 vector, transfected COS-1 cells were lysed in 5% SDS in 50 mm Tris-HCl (ph 6.8) and 20 mm EDTA, sonicated, and resolved by 12% SDS-polyacrylamide gels. Proteins were electroblotted and probed with the T7 tag Ab (Novagen) or the 104 mab (43). For SR protein detection in lymphocytes, human T cells were isolated from fresh buffy coats and separated into CD45RA and CD45RO populations by selection with the UCHL1 (CD45RO) and the SN130 (CD45RA) Abs. Both populations were stimulated with PHA-P for 0, 12, 36, 72, and 144 h and analyzed at indicated time points for CD45RA/RO expression by flow cytometry to observe changes in RA/RO expression. Subsequently, equal amounts of cells were lysed in 5% SDS lysis buffer and sonicated twice for 20 s. Proteins were resolved on 12% SDS-polyacrylamide gels, electroblotted, and probed with the 104 mab. Cell separation and stimulation of CD45RA and CD45RO T cells PBMC were isolated from buffy coats (North East London Blood Transfusion Service, London, U.K.) by Ficoll-Paque density gradient centrifugation (Pharmacia, St Albans, U.K.). Monocytes and macrophages were

4 The Journal of Immunology 5289 FIGURE 2. Alternative splicing of the CD45 minigene in COS-1 cells. A, Immunofluorescence analysis of nonfixed transfected COS-1 cells stained with the CD45RB Ab (PD7/26). B, Isoform expression pattern of the CD45 minigene after transient transfection in COS-1 cells. The left panel shows radioactive-labeled RT-PCR products run on a denaturing gel. CD45 isoforms are indicated on the left and isoform size (bp) on the right. M, CD45 marker (637, 295, and 154 bp). Phosphor image quantitation of the CD45 isoform expression (n 3). removed by adherence to plastic for 1 h at 37 C and these were saved and used as APC after mitomycin C treatment where required. CD45RA and CD45RO T cells were then negatively selected. Nonadherent cells were incubated at 4 C for 30 min with a mixture of mabs, BU12 (anti-cd19, a gift from N. Ling, University of Birmingham Medical School, Birmingham, U.K.), OKM1 (anti-cd11b; American Type Culture Collection), and R10 (antiglycophorin, a gift from P. Edwards, University of Cambridge, Cambridge, U.K.) along with SN130 (anti-cd45ra; Imperial Cancer Research Fund, London, U.K.) for removal of CD45RA T cells or UCHL1 (anti-cd45ro; Imperial Cancer Research Fund) for removal of CD45RO T cells. After washing, the cells were incubated with sheep anti mouse IgG-coated magnetic beads at 4 C for 20 min (Dynal, Bromborough, U.K.); labeled cells were then removed with a Dynal magnet. After five rounds of magnetic bead separation, purity was 98% by FACS analysis. For stimulation, CD45RA or CD45RO T cells were incubated in 24-well flat-bottom tissue culture plates (Becton Dickinson, Mountain View, CA) along with 5% mitomycin C-treated adherent cells and 1 g/ml PHA-P (Sigma, St. Louis, MO) for 0, 12, 36, 72, and 144 h. Cells were analyzed by flow cytometry at the time points indicated for alterations in CD45RA and CD45RO expression. Results Alternative splicing of CD45 occurs by inclusion or skipping of three alternative exons (4, 5, and 6 or A, B, and C) and results in a maximum of eight different isoforms, of which five are expressed in human hematopoietic cells (Fig. 1B). We studied alternative FIGURE 3. Role of SR proteins in regulation of CD45 alternative splicing. COS-1 cells were cotransfected with the CD45 minigene and the indicated SR protein constructs. Lane 1, CD45 marker (637, 295, and 154 bp); lane 2, control transfection with the CD45 minigene only; lane 3, control cotransfection with the pcgt7 lacking an insert; and lanes 4 11 cotransfection with the indicated SR proteins. splicing of CD45 with a recently designed human CD45 minigene (psec-s-lca1-7, Fig. 1A), which can be analyzed at both the RNA and protein levels (40). The minigene is driven by the SV40 promoter and contains exons 1 7 of CD45. A cdna fragment of exons 1 and 2 is fused to a genomic fragment of exons 3 7. Exon 7 is then fused to the GPI anchor sequence of DAF, which allows the truncated CD45 sequences to be expressed at the cell surface. COS-1 cells transfected with the CD45 minigene and stained with a CD45 exon-specific Ab showed a granular membrane staining pattern (Fig. 2A). Western blot analysis showed expression of the CD45 isoforms ABC, BC, AB, B, and null with high expression levels of the low m.w. isoforms (data not shown; Ref. 40). RT- PCR analysis of transfected COS-1 cells demonstrated that the expression of the CD45 null isoform was the most abundant (50%), followed by an equal expression of the BC and B isoforms (20 25%) and a very low ABC expression (1 3%) (Fig. 2B). SR proteins display antagonistic effects on the alternative splicing of CD45 The focus of this paper was to examine the role of SR proteins in CD45 alternative splicing. SR proteins are essential for constitutive splicing (5 10) and alternative splicing (10 16) and have demonstrated substrate-specific effects on splicing (7, 11, 12, 14, 26, 27). The SR proteins are thus candidates to control CD45 splicing. In COS-1 cells, the CD45 minigene is processed to give a mixture of isoforms (Fig. 2B), of which the smallest is the most abundant (CD45 null, which skips the alternative exons). The ability of SR proteins to modulate CD45 splicing was assessed by cotransfecting individual SR cdnas with the CD45 minigene (Fig. 3). Three groups of SR proteins with different specificities were identified. In the first group, SF2/ASF, SC35, SRp30c, SRp40, and SRp75 overexpression resulted in down-regulation of the high m.w. isoforms and up-regulation of the CD45 null isoform with the CD45 B isoform still detectable (Fig. 3, lanes 6 9 and 11). The second group, SRp20 and 9G8, showed the opposite (Fig. 3, lanes 4 and 5), with elevated expression of the largest ABC isoform and a decrease in expression of the smallest CD45 null isoform. SRp20 and 9G8 both have a single (canonical) RRM and the homology

5 5290 REGULATION OF CD45 SPLICING FIGURE 4. Antagonistic effects of SR proteins SRp20 and 9G8 vs SRp30c and SRp40 on the alternative splicing of CD45. A, Increasing amounts (0.1 5 g) of SRp20, 9G8, SRp30c, and SRp40 were cotransfected with a constant amount of the CD45 minigene. Results were analyzed by RT-PCR and radioactive products were resolved on denaturing polyacrylamide gels. B, Isoforms were quantitated by phosphor image analysis and the amounts are shown as a percentage of the sum. The graph was extracted from the cotransfection experiments using 3 g of the tested SR proteins shown in A. between the two proteins in this domain is high (79%) when compared with the canonical RRM of the other SR proteins (35 45%) (9). Finally, SRp55 showed no effect on the alternative splicing of CD45 (Fig. 3, lane 10). The effects of the antagonistic SR proteins SRp30c and SRp40 (exon skipping) vs SRp20 and 9G8 (exon inclusion) were examined in more detail using incremental amounts of the SR construct cotransfected with 1 g of the CD45 target (Fig. 4). The shift to exon inclusion induced by SRp20 was almost complete with only small amounts of isoforms BC and B still detectable. The effect of 9G8 was less dramatic, the ABC band increased to almost 20% of the total sum, but the smallest isoform did not decrease and kept steady at 50%. Interestingly, 9G8 promoted the appearance of the AB isoform, which is hardly detectable in the other transfections. Overexpression of SRp30c and SRp40 resulted in a dramatic increase of the smallest isoform skipping exons A, B, and C. The CD45 ABC and BC isoforms were hardly detectable anymore, whereas the expression level of the B isoform only decreased marginally or not at all. These experiments were repeated several times in both COS-1 and HeLa cells and small differences in the levels of CD45 isoform expression in the control lanes were seen, but changes induced by SR cotransfection were reproducible in all cases. SR protein expression (except SRp75) was analyzed and confirmed by Western blotting using the T7 epitope tag. Localization to nuclear speckles was also verified by immunofluorescence analysis (data not shown). Role of structural domains of SR proteins in splice site selection of the CD45 pre-mrna As detailed in the introduction, SR proteins consist of two or three modular domains, i.e., the RS domain and either one or two N- terminal RRMs. All SR proteins share the canonical RRM characterized by the conserved RNP-1 and RNP-2 submotifs. In addition, a subset of SR proteins also has a central atypical RRM which lacks the conserved residues in the RNP submotifs (17, 44). To determine which of these domains is responsible for the switch in CD45 splicing noted above, we examined the function of a series

6 The Journal of Immunology 5291 SRp20 with a switch to exon inclusion and production of higher m.w. isoforms (Fig. 6, B and D, lane 8). This is seen not only with the deletion of RRM2 in wild-type SF2/ASF but also in the chimera between the RS domain of SF2/ASF and the RRM of SC35 (Fig. 6B, lanes 5 and 8). Deletion of the RS domains of SF2 and SC35 has little effect on function, like the wild-type proteins, promoting exon skipping. (Fig. 6B, lane 6 and 6A, lane 6). All results of the experiments using chimeric SR proteins are tabulated in Fig. 5. FIGURE 5. Summary of the role of the SR proteins and their modular domains on the alternative splicing of CD45 in vivo. The SR protein constructs, deletion mutants, and domain swap constructs used in this work are shown schematically. Promotion of CD45 ABC or null splicing after overexpression of each construct is indicated by or. No effect is indicated by and a single indicates no promotion of ABC or null splicing but formation of intermediates. Overexpression of 9G8 promotes AB splicing in addition to ABC. of domain deletion and domain swap constructs containing fragments of SRp20, SF2/ASF, SC35, and SRp40. Schematic drawings of all SR proteins, domain deletion, and domain swap constructs are listed in Fig. 5. In each case, the expression and integrity of these constructs was tested by Western blotting using the T7 tag Ab (data not shown). To evaluate whether the RRM or RS domain of SRp20 was responsible for the change in CD45 alternative splicing, we constructed an RS domain deletion mutant (20 RS) and substituted the RRM of SRp20 with the corresponding domain of SC35 (30b 1-20RS, Fig. 6, A and C). Deletion of the RS domain of SRp20 promoted skipping of all alternative exons (Fig. 6A, lane 5). When the RRM of SRp20 was exchanged for the RRM of SC35 (30b 1-20 RS ), exon skipping was also seen giving a similar pattern to wild-type SC35 (Fig. 6A, lanes 4 and 7), implying a role of the RRM of SRp20 in the formation of the ABC isoform. Domain swap chimeras between SF2/ASF and SRp40 are shown in Fig. 7. Both the wild-type proteins promote exon skipping, as do many of the mutants except those in which RRM2 of SRp40 is fused to RRM1 of SF2/ASF (Fig. 7, lanes 6 and 7). RRM2 of SF2/ASF seems to play a specific role in splice site selection; when it is deleted, the splicing of SF2/ASF resembles Changes in SR protein expression following the switch in CD45 splicing Upon stimulation, naive T cells switch from CD45RA (exon inclusion, expressing the CD45 isoforms ABC, AB, BC, and B) to CD45RO (exon skipping, expressing the isoforms B and null), whereas mature memory CD45RO cells remain RO. Resting T cells were separated into RO and RA populations and activated by culture in the presence of PHA. After 6 days, 99% of the RA T cells had lost expression of the exon A epitope and become RO. RT-PCR revealed that at the RNA level in the CD45RA population, the CD45 ABC, AB, and BC isoforms were downregulated between 36 and 72 h and by 72 h there was a predominant expression of the CD45 null and B isoforms. The CD45RO T cell population expressed mainly the CD45 null and B isoforms and remained CD45 null- and B-positive upon stimulation (Fig. 8A). mab 104, which recognizes a phosphoepitope expressed by all SR proteins (43, 45), was used to assess SR protein expression in RA and RO cells over the time course of stimulation (Fig. 8B). Expression of SR proteins is low in resting cells and by h, when cells were proliferating well, SR protein levels were increased. Besides this total increase of SR protein expression, several changes occur specifically in CD45RA -stimulated T cells. SRp75 induction is greater and occurs later in RA -vsro - stimulated cells. Furthermore, in RA -stimulated T cells, a band appears in the 30-kDa cluster of SR proteins. This 30-kDa SR protein increased markedly after 72 h of stimulation and judging by its size it may represent SRp30c; the smallest SR protein in the 30-kDa group, however, direct identification is not possible, as a specific mab is not available. We have previously shown induction of SRp30c at the RNA level following T cell activation. Also apparent from these blots are differences in the ratios of some of the 30-kDa SR proteins. For instance, the third band possibly representing SC35 is induced more in RA-stimulated cells whereas the fourth band, possibly 9G8, is not up-regulated in the RA-stimulated cells. SRp55 and SRp20 were hardly detectable in these T cell populations. T cells from two individual were analyzed which demonstrated comparable results after Western blot analysis with the mab 104. Discussion Alternative splicing of CD45 occurs by inclusion or exclusion of three alternative exons (4, 5, and 6 or A, B, and C) in the N- terminal part of the molecule, resulting in a maximum of eight different isoforms. cis-acting sequences and trans-acting factors have been postulated to influence the alternative splicing pattern of CD45. Linker scanning analysis revealed that in exon A three segments (positions 8 10, 40 91, and , total length 198 bp) and in exon C one large segment (positions , total length 144 bp) were essential for tissue-specific alternative splicing (46, 47). No such segments were found in exon 5, suggesting that splicing of exon 5 is not regulated in a tissue-specific manner (46, 47).

7 5292 REGULATION OF CD45 SPLICING FIGURE 6. Role of modular domains of SRp20, SC35, and SF2/ASF in CD45 splicing. A, RS domain deletion mutants and domain swap constructs of SRp20 and SC35 were cotransfected with the CD45 minigene and analyzed by RT-PCR. Lane M, CD45 marker and lane 2, control transfection with the CD45 minigene only. B, The indicated deletion mutants of SF2/ASF and the construct 30b 1 030a RS were cotransfected with the CD45 minigene and analyzed by RT-PCR. Single transfection with the CD45 minigene is shown in lane 2. CD45 isoforms are indicated on the right, isoform size (bp) is indicated on the left. C and D, CD45 isoform expression patterns shown in A and B, respectively, were quantitated using phosphor imager (Bio- Rad molecular analyst) analysis and reflected in a graph. Relative amounts of CD45 isoforms are shown as a percentage of the sum. Isoforms are indicated on the right and also in gray, cotransfected constructs are indicated below the figure, and isoform expression level (%) is indicated on the left. Fusions between T and B cells retain the T cell phenotype producing the null isoform which led to the suggestion that exon inclusion is the default pattern, whereas skipping is regulated by dominant trans-acting factors in T cells (48). We have previously demonstrated that regulation of alternative splicing of this CD45 minigene is restricted to lymphoid cells. All nonlymphoid transgenic cells and transfected cell lines such as COS-1, HeLa, and 3T3 showed a specific and stable pattern of splicing to form the low m.w. isoforms of CD45. We have shown that SR proteins show antagonistic effects on alternative splicing of CD45. SRp20 promoted exon inclusion, leading to elevated levels of the CD45 ABC isoform and to a decrease in expression levels of the CD45 null, B, and BC isoforms. 9G8 promoted splicing of the ABC and also the AB isoform; however, only splicing of the BC isoform was decreased. Exon exclusion was promoted by SC35, SF2ASF, SRp30c, SRp40, and SRp75. All of these SR proteins promoted splicing of the CD45 null isoform with decreased levels of splicing of the ABC and BC isoforms. Generally, splicing of the CD45 B isoform was not affected. This could be explained by the fact that splicing of exon 5 is not regulated in a tissue-specific fashion (46, 47). Also, during the switch from CD45RA to CD45RO in activated T cells, the T cells do not lose expression of the CD45 B isoform (38, 49, 50). These results show that individual SR proteins are able to switch CD45 splicing; however, to induce a complete shift in CD45 splicing, additional factors like other SR proteins or non-sr factors might be involved. Regulation of CD45 splicing by SRp20 and 9G8 causing exon inclusion could be explained by the presence of exonic (or intronic) enhancer sequences in the CD45 pre-mrna. Several independent studies identified a SRp20-binding sequence with enhancer activity, all sharing the degenerate sequence CUC(U/ G)UC(C/T) (51 54). In addition, application of the so-called SELEX strategy yielded SRp20-specific sequences with the consensus CA/UA/UC (55). The CUC(U/G)UC(C/T) sequence is also found in exon 4 of the SRp20 gene, the splicing of which is autoregulated by SRp20 itself (56). In this study, it was suggested that recognition and splicing of exons with weak splice acceptor sites is a general function of SRp20. A perfect match to the SRp20 consensus sequence is not found in the exonic sequences of CD45; however, the sequence CACCACUGCAUUCUCACCC (nt in exon 4) is reminiscent of all of the identified SRp20 enhancer sequences. Overexpression of 9G8 promotes CD45 AB and ABC splicing. This shift is not as complete as the switch induced by SRp20, as the CD45 null, B, and BC splicing are still detectable. Appearance of the AB isoform, in combination with the ABC isoform, indicates that 9G8 is specifically involved in exon A splicing. No other SR protein was able to promote CD45 AB splicing. Recently, two very divergent 9G8 splicing enhancers have been described (54). One shows strong homology with the 9G8 consensus sequence AGAC(G/U)ACGAY isolated by the SELEX approach (55), whereas the other is pyrimidine rich and shows some sequence homology with the Drosophila double sex splicing enhancer (UCUUCAAUCAAACA) which can bind 9G8 specifically (52). SELEX, using a mutated form of 9G8 lacking the zincknuckle region, yields a pyrimidine-rich sequence (C(A/U)(A/ U)C) that resembles the SRp20 SELEX winner sequence (54, 55). Sequences closely resembling the 9G8 consensus are found in CD45 exon 4 but not in exons 5 and 6 (e.g., CD45 exon 4:

8 The Journal of Immunology 5293 FIGURE 7. Role of modular domains of SRp40 and SF2/ASF on the alternative splicing of CD45. A, Alteration in the splicing pattern of the CD45 minigene after cotransfection with the SRp40 deletion mutant and the SRp40-SF2/ASF swap constructs. RT-PCR analysis of the cotransfection experiments with the indicated constructs is shown. Control transfection with the CD45 minigene only is shown in lane. CD45 isoforms are indicated on the right, isoform size (bp) is indicated on the left. B, Graph of CD45 isoform expression pattern after cotransfection with the SRp40 deletion mutant and the SRp40/SRp30a swap constructs. CD45 isoforms are indicated on the right, constructs used for cotransfection are indicated below the figure, and percentage of isoform expression as a fraction of the total is indicated on the left. 306GACTGACTACA 316, nucleotide numbers are from the CD45ABC cdna), which may explain the ability of 9G8 to promote the use of CD45 isoforms containing exon 4. CD45 exon exclusion is promoted by SF2/ASF, SC35, SRp30c, SRp40, and SRp75. It has recently been shown using a CD45 minigene containing only alternative exon 4 that overexpression of hsswap, SF2/ASF, SC35, SRp40, and SRp75 all promoted exclusion of exon 4, which is in agreement with our findings (57, 58). Lemaire et al. (58) also demonstrated that regulation of CD45 exon 4 splicing is dependent on exon 4 itself and not affected by the presence of the 3 constitutive exon, exon 7. Information for exon exclusion must therefore be present in the exonic (or intronic) sequences of the alternative CD45 exons. Elements that repress splice sites have been identified in other systems (59 66). In Drosophila, binding of Sxl1 to the 3 splice site interferes with U2AF binding (67). In addition, a splicing silencer (IIIA) present in the adenovirus late region L1 mrna was demonstrated to bind SR proteins, which prevented recruitment of U2snRNP to the spliceosome. This silencer binds a number of SR proteins but is most FIGURE 8. Dynamic protein expression levels of SR proteins in stimulated CD45RA and CD45RO T cells. CD45RO- and RA-selected T cells, stimulated with PHA for the indicated times, were analyzed by RT-PCR (A) or Western blot analysis (B) using the SR protein-specific mab 104. Position of the SR proteins are indicated on the right of each panel. efficiently bound by SF2/ASF-SC35 (59). However, enhancer/silencer elements specific for the SR proteins causing CD45 exon exclusion have not been identified and will be the subject of our future studies. The results of the experiments using chimeric SR proteins are tabulated in Fig. 5. In summary, these data demonstrate that the atypical RRM of SF2/ASF determines the specificity for CD45 mrna splicing. All constructs that include this domain promote exon skipping, whereas all constructs lacking this domain promoted exon inclusion and behaved like SRp20. The results resemble those with adenovirus E1A where wild-type SF2/ASF favors the selection of the most proximal 13S site whereas the mutant lacking RRM2 switches splicing to the 12S site which is also promoted by SRp20 (31). During T cell activation, we demonstrated specific changes in SR protein expression levels. We show that SRp75 is up-regulated to a higher level (but later) in stimulated CD45RA vs CD45RO T cells. In addition, the smallest band of the 30-kDa SR proteins which might represent SRp30c is induced only in CD45RA cells. This is concomitant with the fact that both SRp30c and SRp75 promote splicing of the CD45 null isoform in transfected COS-1 cells. One band in the 30-kDa panel of SR proteins, possibly 9G8, is not up-regulated in stimulated CD45RA, which is in agreement with the fact that 9G8 promotes CD45 ABC and AB splicing and not CD45 null splicing.

9 5294 REGULATION OF CD45 SPLICING The expression level, phosphorylation, cellular location, and specific mix of SR/hnRNP proteins are believed to be one component in the regulation of specific splice site choices (10, 14, 26). This model presumes that SR protein expression patterns are tissue specific, developmentally regulated, and responsive to the metabolic state of the cells. We have demonstrated complex changes in the expression of SR proteins upon T cell activation, which occur in parallel with changes in the splicing of CD45. Furthermore, we have also demonstrated that members of the SR protein family can have dramatic and antagonistic effects on CD45 splicing by transfection in vivo. The SR protein family is thus a strong candidate for CD45 regulation in vivo. In future experiments, we hope to strengthen these observations using T cell transgenic mouse technology and also plan to map the SR-specific intronic and exonic enhancer/silencer sequences within the CD45 gene. Acknowledgments We thank Alison Cowper for her excellent technical assistance and Javier Caceres for the provision of SF2 mutant expression vectors. References 1. Thomas, M. L The leukocyte common antigen family. Annu. Rev. Immunol. 7: Trowbridge, I. S., and M. L. Thomas CD45: an emerging role as a protein tyrosine phosphatase required for lymphocyte activation and development. Annu. Rev. Immunol. 12: Green, M. R Biochemical mechanisms of constitutive and regulated premrna splicing. Annu. Rev. Cell Biol. 7: Kramer, A The structure and function of proteins involved in mammalian pre-mrna splicing. Annu. Rev. Biochem. 65: Krainer, A. R., G. C. Conway, and D. Kozak Purification and characterization of pre-mrna splicing factor SF2 from HeLa cells. Genes Dev. 4: Fu, X. D., and T. Maniatis Isolation of a complementary DNA that encodes the mammalian splicing factor SC35. Science 256: Kim, Y. J., P. Zuo, J. L. Manley, and B. S. Baker The Drosophila RNAbinding protein RBP1 is localized to transcriptionally active sites of chromosomes and shows a functional similarity to human splicing factor ASF/SF2. Genes Dev. 6: Zahler, A. M., W. S. Lane, J. A. Stolk, and M. B. Roth SR proteins: a conserved family of pre-mrna splicing factors. Genes Dev. 6: Cavaloc, Y., M. Popielarz, J. P. Fuchs, R. Gattoni, and J. Stevenin Characterization and cloning of the human splicing factor 9G8: a novel 35 kda factor of the serine/arginine protein family. EMBO J. 13: Screaton, G. R., J. F. Caceres, A. Mayeda, M. V. Bell, M. Plebanski, D. G. Jackson, J. I. Bell, and A. R. Krainer Identification and characterization of three members of the human SR family of pre-mrna splicing factors. EMBO J. 14: Krainer, A. R., G. C. Conway, and D. Kozak The essential pre-mrna splicing factor SF2 influences 5 splice site selection by activating proximal sites. Cell 62: Ge, H., and J. L. Manley A protein factor, ASF, controls cell-specific alternative splicing of SV40 early pre-mrna in vitro. Cell 62: Fu, X. D., A. Mayeda, T. Maniatis, and A. R. Krainer General splicing factors SF2 and SC35 have equivalent activities in vitro, and both affect alternative 5 and 3 splice site selection. Proc. Natl. Acad. Sci. USA 89: Zahler, A. M., K. M. Neugebauer, W. S. Lane, and M. B. Roth Distinct functions of SR proteins in alternative pre-mrna splicing. Science 260: Caceres, J. F., S. Stamm, D. M. Helfman, and A. R. Krainer Regulation of alternative splicing in vivo by overexpression of antagonistic splicing factors. Science 265: Wang, J., and J. L. Manley Overexpression of the SR proteins ASF/SF2 and SC35 influences alternative splicing in vivo in diverse ways. RNA 1: Manley, J. L., and R. Tacke SR proteins and splicing control. Genes Dev. 10: Wang, J., Y. Takagaki, and J. L. Manley Targeted disruption of an essential vertebrate gene: ASF/SF2 is required for cell viability. Genes Dev. 10: Wang, J., S. H. Xiao, and J. L. Manley Genetic analysis of the SR protein ASF/SF2: interchangeability of RS domains and negative control of splicing. Genes Dev. 12: Eperon, I. C., D. C. Ireland, R. A. Smith, A. Mayeda, and A. R. Krainer Pathways for selection of 5 splice sites by U1 snrnps and SF2/ASF. EMBO J. 12: Kohtz, J. D., S. F. Jamison, C. L. Will, P. Zuo, R. Luhrmann, M. Garcia Blanco, and J. L. Manley Protein-protein interactions and 5 -splice-site recognition in mammalian mrna precursors. Nature 368: Tarn, W. Y., and J. A. Steitz Modulation of 5 splice site choice in premessenger RNA by two distinct steps. Proc. Natl. Acad. Sci. USA 92: Fu, X. D., and T. Maniatis The 35-kDa mammalian splicing factor SC35 mediates specific interactions between U1 and U2 small nuclear ribonucleoprotein particles at the 3 splice site. Proc. Natl. Acad. Sci. USA 89: Wu, J. Y., and T. Maniatis Specific interactions between proteins implicated in splice site selection and regulated alternative splicing. Cell 75: Staknis, D., and R. Reed 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. 14: Fu, X. D Specific commitment of different pre-mrnas to splicing by single SR proteins. Nature 365: Mayeda, A., G. R. Screaton, S. D. Chandler, X. D. Fu, and A. R. Krainer Substrate specificities of SR proteins in constitutive splicing are determined by their RNA recognition motifs and composite pre-mrna exonic elements. Mol. Cell. Biol. 19: Xiao, S. H., and J. L. Manley Phosphorylation-dephosphorylation differentially affects activities of splicing factor ASF/SF2. EMBO J. 17: Mayeda, A., and A. R. Krainer Regulation of alternative pre-mrna splicing by hnrnp A1 and splicing factor SF2. Cell 68: Mayeda, A., S. H. Munroe, J. F. Caceres, and A. R. Krainer Function of conserved domains of hnrnp A1 and other hnrnp A/B proteins. EMBO J. 13: Caceres, J. F., T. Misteli, G. R. Screaton, D. L. Spector, and A. R. Krainer Role of the modular domains of SR proteins in subnuclear localization and alternative splicing specificity. J. Cell Biol. 138: Caceres, J. F., G. R. Screaton, and A. R. Krainer A specific subset of SR proteins shuttles continuously between the nucleus and the cytoplasm. Genes Dev. 12: Barclay, A. N., D. I. Jackson, A. C. Willis, and A. F. Williams Lymphocyte specific heterogeneity in the rat leucocyte common antigen (T200) is due to differences in polypeptide sequences near the NH2-terminus. EMBO J. 6: Saga, Y., J. S. Tung, F. W. Shen, and E. A. Boyse Alternative use of 5 exons in the specification of Ly-5 isoforms distinguishing hematopoietic cell lineages. Proc. Natl. Acad. Sci. USA 84: Saga, Y., J. S. Lee, C. Saraiya, and E. A. Boyse Regulation of alternative splicing in the generation of isoforms of the mouse Ly-5 (CD45) glycoprotein. Proc. Natl. Acad. Sci. USA 87: Ralph, S. J., M. L. Thomas, C. C. Morton, and I. S. Trowbridge Structural variants of human T200 glycoprotein (leukocyte-common antigen). EMBO J. 6: Streuli, M., L. R. Hall, Y. Saga, S. F. Schlossman, and H. Saito Differential usage of three exons generates at least five different mrnas encoding human leukocyte common antigens. J. Exp. Med. 166: Akbar, A. N., L. Terry, A. Timms, P. C. Beverley, and G. Janossy Loss of CD45R and gain of UCHL1 reactivity is a feature of primed T cells. J. Immunol. 140: Zapata, J. M., R. Pulido, A. Acevedo, F. Sanchez Madrid, and M. O. de Landazuri Human CD45RC specificity: a novel marker for T cells at different maturation and activation stages. J. Immunol. 152: ten Dam, G. B., Wieringa, B., and Poels, L. G Alternative splicing of CD45 pre-mrna is uniquely obedient to conditions in lymphoid cells. Biochim. Biophys. Acta 1446: ten Dam, G. B., Poels, L. G., and Wieringa, B Cell surface GPI-anchoring of CD45 Isoforms. Mol. Biol. Rep. 25: Wilson, A. C., M. G. Peterson, and W. Herr The HCF repeat is an unusual proteolytic cleavage signal. Genes Dev. 9: Roth, M. B., C. Murphy, and J. G. Gall A monoclonal antibody that recognizes a phosphorylated epitope stains lampbrush chromosome loops and small granules in the amphibian germinal vesicle. J. Cell Biol. 111: Fu, X. D The superfamily of arginine/serine-rich splicing factors. RNA 1: Roth, M. B., A. M. Zahler, and J. A. Stolk A conserved family of nuclear phosphoproteins localized to sites of polymerase II transcription. J. Cell Biol. 115: Streuli, M., and H. Saito Regulation of tissue-specific alternative splicing: exon- specific cis-elements govern the splicing of leukocyte common antigen pre-mrna. EMBO J. 8: Tsai, A. Y., M. Streuli, and H. Saito Integrity of the exon 6 sequence is essential for tissue-specific alternative splicing of human leukocyte common antigen pre-mrna. Mol. Cell. Biol. 9: Rothstein, D. M., H. Saito, M. Streuli, S. F. Schlossman, and C. Morimoto The alternative splicing of the CD45 tyrosine phosphatase is controlled by negative regulatory trans-acting splicing factors. J. Biol. Chem. 267: Deans, J. P., H. M. Serra, J. Shaw, Y. J. Shen, R. M. Torres, and L. M. Pilarski Transient accumulation and subsequent rapid loss of messenger RNA encoding high molecular mass CD45 isoforms after T cell activation. J. Immunol. 148: Smith, S. H., M. H. Brown, D. Rowe, R. E. Callard, and P. C. Beverley Functional subsets of human helper-inducer cells defined by a new monoclonal antibody, UCHL1. Immunology 58: Chiara, M. D., O. Gozani, M. Bennett, P. Champion Arnaud, L. Palandjian, and R. Reed Identification of proteins that interact with exon sequences, splice sites, and the branchpoint sequence during each stage of spliceosome assembly. Mol. Cell. Biol. 16:3317.

10 The Journal of Immunology Lynch, K. W., and T. Maniatis Assembly of specific SR protein complexes on distinct regulatory elements of the Drosophila doublesex splicing enhancer. Genes Dev. 10: Lou, H., K. M. Neugebauer, R. F. Gagel, and S. M. Berget Regulation of alternative polyadenylation by U1 snrnps and SRp20. Mol. Cell. Biol. 18: Schaal, T. D., and T. Maniatis Selection and characterization of premrna splicing enhancers: identification of novel SR protein-specific enhancer sequences. Mol. Cell. Biol. 19: Cavaloc, Y., C. F. Bourgeois, L. Kister, and J. Stevenin The splicing factors 9G8 and SRp20 transactivate splicing through different and specific enhancers. RNA 5: Jumaa, H., and P. J. Nielsen The splicing factor SRp20 modifies splicing of its own mrna and ASF/SF2 antagonizes this regulation. EMBO J. 16: Sarkissian, M., A. Winne, and R. Lafyatis Mammalian SWAP regulation of alternative mrna splicing of CD45 exon 4 and fibronectin IIICS. J. Biol. Chem. 271: Lemaire, R., A. Winne, M. Sarkissian, and R. Lafyatis SF2 and SRp55 regulation of CD45 exon 4 skipping during T cell activation. Eur. J. Immunol. 29: kanopka, A., O. Muhlemann, and G. akusjarvi Inhibition by SR proteins of splicing of a regulated adenovirus pre-mrna. Nature 381: Blanchette, M., and B. Chabot A highly stable duplex structure sequesters the 5 splice site region of hnrnp A1 alternative exon 7B. RNA 3: Blanchette, M., and B. Chabot Modulation of exon skipping by highaffinity hnrnp A1-binding sites and by intron elements that repress splice site utilization. EMBO J. 18: Si, Z. H., D. Rauch, and C. M. Stoltzfus The exon splicing silencer in human immunodeficiency virus type 1 Tat exon 3 is bipartite and acts early in spliceosome assembly. Mol. Cell. Biol. 18: Si, Z., B. A. Amendt, and C. M. Stoltzfus Splicing efficiency of human immunodeficiency virus type 1 tat RNA is determined by both a suboptimal 3 splice site and a 10 nucleotide exon splicing silencer element located within tat exon. Nucleic Acids Res. 25: Chan, R. C., and D. L. Black The polypyrimidine tract binding protein binds upstream of neural cell-specific c-src exon N1 to repress the splicing of the intron downstream. Mol. Cell. Biol. 17: Chan, R. C., and D. L. Black Conserved intron elements repress splicing of a neuron-specific c-src exon in vitro. Mol. Cell. Biol. 17: Muro, A. F., M. Caputi, R. Pariyarath, F. Pagani, E. Buratti, and F. E. Baralle Regulation of fibronectin EDA exon alternative splicing: possible role of RNA secondary structure for enhancer display. Mol. Cell. Biol. 19: Valcarcel, J., R. Singh, P. D. Zamore, and M. R. Green The protein Sexlethal antagonizes the splicing factor U2AF to regulate alternative splicing of transformer pre-mrna. Nature 362:171.