Genetic variation in ICOS regulates mRNA levels of ICOS and splicing isoforms of CTLA4

Molecular Immunology 44 (2007) 1644–1651

Genetic variation in ICOS regulates mRNA levels of ICOS and splicing isoforms of CTLA4 Tanja Kaartinen a,∗ , Jani Lappalainen b , Katri Haimila a , Matti Autero b , Jukka Partanen a a

b

Research and Development, Finnish Red Cross Blood Service, Kivihaantie 7, FI-00310 Helsinki, Finland Division of Biochemistry, Faculty of Biosciences, University of Helsinki, P.O. Box 56 (Viikinkaari 5), FI-00014 Helsinki, Finland Received 7 June 2006; accepted 4 August 2006 Available online 25 September 2006

Abstract Genetic and functional studies suggest that polymorphism in cytotoxic T lymphocyte-associated antigen-4 (CTLA4) and inducible costimulator (ICOS) genes, both reported to harbour autoimmune susceptibility loci, could regulate the immune activation through affecting their expression and splicing of CTLA4. To address this, we studied expression of CTLA4 and ICOS and the role of polymorphisms in the gene region by measuring the relative amounts of transcripts, including the soluble CTLA4 (sCTLA4) splicing isoform in healthy volunteers. We combined a physiologically relevant in vitro activation for human CD4+ T lymphocytes and a quantitative RT-PCR. The susceptibility allele CT60G in CTLA4 gene was confirmed to be associated with a decreased amount of sCTLA4, but only in resting cells. During the T cell activation two genetic variants in ICOS gene, IVS1+173T/C and c.1624C/T, affected expression of CTLA4 isoforms and ICOS, respectively. We could not confirm that the level of sCTLA4 is down-regulated following T lymphocyte activation, instead the levels of CTLA4 splicing isoforms correlated to each others. Our results indicate that genetic variation in this gene region regulates the expression of both CTLA4 and ICOS and not only the splicing of sCTLA4 as suggested earlier. © 2006 Elsevier Ltd. All rights reserved. Keywords: ICOS; CTLA4; Regulatory SNP; Functional polymorphism; T lymphocytes; Autoimmune susceptibility

1. Introduction Chromosomal region 2q33 harbouring the human CD28 cosignal receptor gene family, CD28, cytotoxic T lymphocyteassociated antigen-4 (CTLA4) and inducible costimulator (ICOS), has been suggested to carry a predisposing gene locus for several autoimmune diseases (Gough et al., 2005), such as type 1 diabetes, Graves’ disease and coeliac disease. The molecules have a crucial role during the T lymphocyte activation and its regulation: CTLA4 inhibits and ICOS stimulates the activation. Hence they are good functional candidates for autoimmunity and transplantation tolerance studies. Evidence based on linkage and genetic association studies has so far pointed to CT60 polymorphism near CTLA4 as the most promising susceptibility locus for type 1 diabetes and Graves’ disease (Ueda et al., 2003). However, we and others have shown that in

∗

Corresponding author. Tel.: +358 9 5801275; fax: +358 9 5801310. E-mail address: [email protected] (T. Kaartinen).

0161-5890/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.molimm.2006.08.010

coeliac disease (Haimila et al., 2004) and in multiple sclerosis (Bonetti et al., 2004) polymorphisms near ICOS might also be likely candidates. All single nucleotide polymorphisms (SNP) in this gene region that are sufficiently common for candidates are located outside of the sequences coding for amino acids on the final protein product. This means that the functional susceptibility polymorphism is most likely a regulatory SNP affecting the amount of the product. Because of the complex genetics behind autoimmune diseases and the strong linkage disequilibrium in this chromosomal region, it may not be easy to pinpoint the actual predisposing polymorphism(s). However, studies looking for functional effects of the polymorphisms could reveal some relevant candidates. First functional studies of this locus focused on CTLA4 and its non-synonymous exon 1+49A/G and promoter −318C/T SNP which appeared to have an influence on the amount of CTLA4 protein (Kouki et al., 2000; Ligers et al., 2001; Maurer et al., 2002; Wang et al., 2002; Anjos et al., 2002). More recently, it was demonstrated that a candidate susceptibility polymorphism might change the ratio of CTLA4 splicing isoforms (Ueda et al.,

T. Kaartinen et al. / Molecular Immunology 44 (2007) 1644–1651

2003; Atabani et al., 2005). However, conflicting results have also been reported (Anjos et al., 2005). The splicing isoform of CTLA4 from which the third exon coding for the transmembrane domain is missing (Magistrelli et al., 1999) is a soluble protein named sCTLA4. It binds the same B7 family ligands as the full-length isoform (flCTLA4) and may inhibit immune activation as demonstrated in mixed lymphocyte reaction (Oaks et al., 2000). Also, elevated levels of sCTLA4 are detected in autoimmune disorders (Pawlak et al., 2005). This suggests that sCTLA4 might be a regulator or an active component of immune response. There is also evidence that the levels of sCTLA4 decrease during immune activation (Magistrelli et al., 1999; Oaks et al., 2000), but not so much in systemic lupus erythematosus patients (Wong et al., 2005). Hence, it is possible to create a hypothesis that sCTLA4 regulates T cell immune activation and that its levels are regulated by genetic variation in the CD28 gene complex. This could provide us with a mechanistic explanation for the associated gene polymorphisms. We here addressed this question by studying CTLA4 and ICOS expression levels in human peripheral CD4+ T lymphocytes obtained from individuals with different genetic variants in the CTLA4 and ICOS genes. We focused on mRNA levels, as we can assume that differences in mRNA immediately after immune activation should reflect more directly genetic effects than measurements of protein levels or cellular activation, which are subject to more complex regulation. 2. Materials and methods 2.1. Samples and stimulation of CD4+ T cells CD4+ T cells were separated from fresh whole blood samples using the RosetteSepTM Human CD4+ T Cell Enrichment Cocktail (StemCell Technologies Inc., Vancouver, BC, Canada) immediately after drawing the blood. Enrichment is based on negative selection leaving the desired cells untouched. Separation was done according to the manufacturer’s instructions except the density gradient centrifugation which was done at 900 × g. Routinely >95% purity was achieved as checked by flow cytometry (CD4+ CD14− ). Activation of cells was done with anti-CD3 antibody (UCHT1) and CD80-muIg fusion protein (both from Ancell Corporation, Bayport, MN, USA). These biotin-conjugated proteins were attached on avidin (Pierce Biotechnology Inc., Rockford, IL, USA) coated 24-well plates (activation method is described more detailed elsewhere, Autero manuscript in preparation). Mouse IgG with irrelevant specificity was used as a negative control (Biotin Mouse IgG2a Isotype Control; eBioscience, San Diego, CA, USA). 2.5 × 106 CD4+ T cells were added per well in complete RPMI-1640 (10% fetal bovine serum, 100 IU/ml penicillin, 100 ␮g/ml streptomycin) and stimulated for different time periods from 1 to 24 h. The stimulations were done in triplicates whenever the cell number per sample was large enough (79% of all stimulations). All study subjects were healthy adults who voluntarily participated in this study which was approved by the ethical committee of the Hospital District of Helsinki and Uusimaa (HUS).

1645

2.2. Flow cytometry Basic protocols were used for surface staining of CD4+ T cells. The following FITC- or PE-labelled monoclonal antibodies were used: CD4 from eBioscience (San Diego, CA, USA), CD14 from BD PharMingen (San Diego, CA, USA), CD25 from Miltenyi Biotec (Bergisch Gladbach, Germany) and CD69 from ImmunoTools (Friesoythe, Germany). Appropriate isotype controls were included in all experiments. Flow cytometry analysis was performed on Becton Dickinson FACSCaliburTM and data were analyzed using the CellQuestTM software (BD Biosciences, San Jose, USA). 2.3. RNA extraction and cDNA synthesis Total RNA was extracted from frozen cell samples with RNeasy Mini Kit and DNase treatment was done with RNaseFree DNase Set (both from Qiagen, Hilden, Germany). Reverse transcription (RT) was performed with High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA, USA) according to the instructions provided by the manufacturer. RNA concentration in each RT reaction was 5 ng/␮l and RNase inhibitor (Applied Biosystems) was added at 1 U/␮l. Reactions were primed by random primers and synthesis was done by MultiScribe RT enzyme both provided with the cDNA kit. To control the absence of genomic DNA-RT reactions were performed on a few samples without MultiScribe RT enzyme. No amplification could be detected in these reactions by quantitative PCR. 2.4. Quantitative real-time PCR To quantitate relative amounts of mRNA molecules ABI PRISM 7000 Sequence Detection System and TaqMan® chemistry were employed (Applied Biosystems, Foster City, CA, USA). PCR reaction components and conditions were as recommended by Applied Biosystems. TaqMan® Gene Expression Assay was used for ICOS detection (Hs00359999 m1). The primers and probes for flCTLA4 and sCTLA4 (Ueda et al., 2003) were purchased from Sigma. 18S ribosomal RNA was used as an endogenous control gene (TaqMan Ribosomal RNA Control Reagents from Applied Biosystems) and it was multiplexed with the actual targets. All amplifications were done in triplicates. The same standard curve, diluted from total RNA from peripheral blood mononuclear cells, and non-template controls were included in every run. To calculate the amount of target mRNA in a sample the relative standard curve method (Applied Biosystems) was used. Briefly, input amounts were calculated using standard curves and relative amounts (from now on referred as the actual RNA level) were obtained by normalizing to the endogenous control. These results were proportioned to standards and thereby depicted as manifolds of them. Also activation-induced changes in the expressions were calculated: actual RNA level at time point of interest/actual level at 0 h = fold change from 0 h.

1646

T. Kaartinen et al. / Molecular Immunology 44 (2007) 1644–1651

2.5. Genotyping DNA samples from volunteers were extracted with standard methods and genotypes for following polymorphisms were typed as described elsewhere (Haimila et al., 2002, 2004): CTLA4−318 (rs5742909), CTLA4+49 (rs231775), CT60 (rs3087243), ICOS IVS1+173 (rs10932029), ICOSc.602 (rs10183087), ICOS c.1624 (rs10932037) and ICOSc.2373 (rs4675379). 2.6. Statistical analysis The data were analyzed using the SPSS software (SPSS Inc., Chicago, IL, USA, version 14.0) and nonparametric tests were used. The expression differences between genotypes were analyzed by Kruskal–Wallis test and correlations using Spearman’s correlation. Differences were considered to be statistically significant when p < 0.05. 3. Results To avoid overwhelming T lymphocyte activation that could obscure small but possibly relevant differences in the expression profiles, suboptimal activation conditions for human peripheral blood CD4+ T lymphocytes were first defined. We measured the

up-regulation of T cell activation markers CD69 and CD25 on the cell surface after 24 h stimulation using variable amounts of activating molecules (Fig. 1A). Suboptimal T cell activation was attained using 20 ng of anti-CD3 antibody (UCHT1) and 150 ng of CD80 fusion protein. Same amounts of stimulatory proteins were used in all experiments. The induction of ICOS and CTLA4 gene expression was delivered through the CD28 and T cell receptor (TCR) route, as no change from the starting level was detected using irrelevant IgG instead of anti-CD3 and CD80 (data not shown). No CD69 expression was detectable on fresh CD4+ T lymphocytes before activation (Fig. 1B), which indicated that the cells were in a resting state. A small subpopulation of resting CD4+ T cells expressing CD25 (Fig. 1A and B) can be explained by CD4+ CD25+ regulatory T cells. Samples from 17 apparently healthy individuals were tested using quantitative RT-PCR that could detect mRNA for ICOS and both isoforms of CTLA4 separately. To subtract possible sample specific variation before the activation step, each actual value at any time point was divided by the value measured at time point 0 h, that is, the value of 0 h was set to 1. About twofold increase in the expressions of ICOS and CTLA4 mRNAs was noted as soon as after 1 h stimulation (Fig. 2A–C). The mean peak values for increased expressions reached at different time points (1, 3 or 6 h, depending on the sample) were 2.7 for ICOS, 3.4 for flCTLA4 and 3.0 for sCTLA4. The elevated RNA

Fig. 1. Stimulation of CD4+ T cells was defined to be suboptimal as measured by induced CD69 surface expression when 20 ng of CD3 antibody (UCHT1) and 150 ng of CD80 were used (A). Also CD25 up-regulation was controlled. T cell activation marker CD69 was absent from cells before activation, but was up-regulated together with CD25 following suboptimal stimulation (B, n = 9 for CD69, n = 12 for CD25). Results are represented as percentage of positive cells among purified CD4+ T cells after 24 h stimulation.

T. Kaartinen et al. / Molecular Immunology 44 (2007) 1644–1651

1647

Fig. 2. Activation-induced changes in cosignal receptor gene expression. The actual mRNA levels for ICOS (A) and CTLA4 splicing isoforms (flCTLA4 in B and sCTLA4 in C) in activated CD4+ T cells from 17 healthy volunteers were measured using quantitative RT-PCR. Activation-induced change from 0 h was calculated by dividing the actual level at time point of interest by the actual level at 0 h. Also, activation-induced changes in s/flCTLA4 ratio are shown (D). Colours represent different donors and the same colour is used for the same individual in all panels. Values are means from triplicate stimulations. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of the article.)

production was still evident after 24 h activation, but then also the variation between individuals was substantially larger (data not shown). It was of note that following activation the expression of sCTLA4 was up-regulated rather than down-regulated in CD4+ T cells (Fig. 2C). As the proportional amounts of sCTLA4 and flCTLA4 can be speculated to regulate T cell activation, we calculated also the activation-induced change in their ratios at different time points. A slightly decreasing trend in the sCTLA4/flCTLA4 mRNA ratio was generally observed after activation, with a few exceptions in which the ratio increased to about two-fold (Fig. 2D). To study the correlation between the two CTLA4 isoforms further, we tested whether the actual levels of mRNAs for sCTLA4 and flCTLA4 at different time points, instead of the fold changes, showed significant correlation to each other. The correlation was strong at time points 1–24 h (p < 0.001, r > 0.78 for each time point: 1, 3, 6 and 24 h; Fig. 3), but not at time point 0 h (p = 0.74, r = 0.09). We next studied whether genetic variation in the CTLA4 or ICOS genes was associated with their mRNA expression levels during the activation. For a feature to be a relevant subject for a genetic study, it must be relatively constant between measurements between independent samples from the same individuals. To address this point, we replicated the activations in 5 of the 17 individuals using independently drawn samples. The replicates gave sufficiently consistent results to make the analysis of genotype–phenotype association meaningful (data not shown).

Seven polymorphisms located close to, or in the CTLA4 and ICOS genes were tested. We found an association between ICOS IVS1+173 polymorphism and CTLA4 mRNA expression (Fig. 4). Activation-induced increase, that is, the fold change from 0 h, in the expression of both CTLA4 splicing isoforms was higher in samples homozygous for the ICOS IVS1+173 T allele as compared to the TC heterozygotes. No CC homozygotes were available. The differences were statistically significant after acti-

Fig. 3. Correlation between the actual mRNA levels of CTLA4 splicing isoforms after T cell activation (p < 0.001, r > 0.78 for each time point: 1, 3, 6 and 24 h).

1648

T. Kaartinen et al. / Molecular Immunology 44 (2007) 1644–1651

Fig. 4. Expression of both CTLA4 splicing isoforms was dependent on genetic variation in ICOS gene (IVS1+173T/C). Activation-induced change in the expression of both sCTLA4 (A) and flCTLA4 (B) was higher in persons homozygous for T allele compared to heterozygotes. Horizontal lines in the left panels represent means of the genotype groups (TT n = 7, TC n = 8 in A and TT n = 6, TC n = 5 in B). In the right panels time points up to 6 h are shown and the colours are similar to the left panels.

vation for 3 and 6 h (p = 0.03 and 0.05 for sCTLA4, Fig. 4A; p = 0.05 and 0.01 for flCTLA4, Fig. 4B). Also, as shown in the right hand side panels in Fig. 4A and B showing individual samples by genotype, a trend was detectable nearly throughout the activation. There was no association between the other genotypes and expression levels as measured by the fold change from time point 0 h. Only the fold changes in the expression levels were studied above in relation to genetic polymorphisms. As the actual mRNA levels at time point 0 h, in fact, did differ between the samples (details not shown) and as there was no correlation in actual levels of sCTLA4 and flCTLA4 mRNAs at time point 0 h, we also wanted to analyze the association between the genotypes and actual mRNA levels. First, the association between ICOS IVS1+173T/C genotype and CTLA4 expression was re-evaluated. In the case of flCTLA4 also the actual levels of mRNA were statistically significantly higher in TT homozygotes (p = 0.03, time point 3 h). Hence

this finding was not restricted only to the activation-induced change from starting level. On the other hand, the actual levels of sCTLA4 did not differ between the ICOS IVS1+173T/C genotypes. We wanted to replicate earlier findings made by others that suggest that CD4+ T cells (Ueda et al., 2003) and CD4+ CD25+ regulatory T cell subpopulation (Atabani et al., 2005), expressed CTLA4 splicing isoforms differently depending on the CT60 genotypes. We found that at time point 0 h the CD4+ T cells from CT60 GG homozygous individuals expressed lower actual mRNA levels of sCTLA4 than those of other CT60 genotypes (p = 0.013, Fig. 5). Most interestingly, the association of CT60 genotypes with the amount of sCTLA4 was no longer evident after activation. There was no association between CT60 genotypes and flCTLA4 levels. In addition, genetic variation in ICOS gene was also associated with differences in ICOS expression (Fig. 6). Activated CD4+ T cells from ICOS c.1624 CC homozygous persons

T. Kaartinen et al. / Molecular Immunology 44 (2007) 1644–1651

1649

Fig. 5. Genetic variation (CT60) located 3 to CTLA4 gene was associated with differences in the actual mRNA levels of sCTLA4 in resting CD4+ T cells. Actual level of sCTLA4 was studied from 7 CT60 GG homozygotes and 10 with other genotypes (9 GA+1 AA). Horizontal lines in the left panel represent means of the genotype groups. In the right panel time points up to 6 h are shown and the colours are similar to the left panel.

Fig. 6. The actual level of activation-induced mRNA expression of ICOS was associated with genetic variation in the same gene. CC homozygotes (n = 11) at ICOS c.1624 had significantly more ICOS transcripts after 1 and 3 h activation compared to CT heterozygotes (n = 5). Horizontal lines in the left panel represent means of the genotype groups. In the right panel time points up to 6 h are shown and the colours are similar to the left panels.

(n = 11) had higher actual levels of ICOS mRNA than cells from c.1624 TC heterozygous persons (n = 5; p = 0.01, time point 1 h). Expression difference was still statistically significant after 3 h activation (p = 0.03), but was later dispersed. There were no TT homozygotes available in the study population. In this case the activation-induced changes were not statistically significant. The explanation for this inconsistency can be seen in Fig. 6 where the actual starting levels of ICOS were somewhat higher in the CC group comparing to TC, although this did not reach statistical significance. No association was observed between the mRNA levels of ICOS, flCTLA4 or sCTLA4 and other genotypes. 4. Discussion Our results demonstrate that genetic variation in the human ICOS gene is associated with variability in expression of T cell

cosignal receptor genes CTLA4 and ICOS. We found two ICOS polymorphisms which were associated with mRNA expression. The IVS1+173T/C polymorphism which is located in the first intron of ICOS had, interestingly, an effect on the expression of the closely linked CTLA4 gene. TT homozygotes had a higher activation-induced change in the expression of both sCTLA4 and flCTLA4 mRNAs as compared to those with other genotypes. The effect could not be detected in resting cells. This ICOS polymorphism is located in the linkage disequilibrium group which included the CTLA4 gene in our analysis of Finnish families (Haimila et al., 2004). Hence, linkage disequilibrium to some unknown SNP might explain how genetic variation in the ICOS gene regulates the CTLA4 gene. The ICOS c.1624C/T polymorphism in the 3 untranslated region of ICOS regulated the expression of ICOS mRNA as the mRNA levels were increased among the CC homozygotes compared to other genotypes. In both polymorphic sites the most

1650

T. Kaartinen et al. / Molecular Immunology 44 (2007) 1644–1651

frequent genotype in the population was associated with a higher level of gene expression. Furthermore, Haimila et al. (2004) reported that the ICOS IVS1+173 T allele, found to be associated with the increased expression of CTLA4 in the present report, was a susceptibility factor for coeliac disease, whereas the ICOS c.1624 T allele, increasing the ICOS expression, was protective. It is possible to speculate that the combination of the alleles regulates the expression equilibrium between these two cosignal molecules. Moreover, genetic variation in ICOS promoter can regulate human ICOS expression as shown by Shilling et al. (2005). There is increasing evidence for the existence of allele specific gene expression of human genes and the concomitant importance of regulatory SNPs behind complex disorders is coming more accepted (Knight, 2005). Allele specific differences seem to be very context dependent and appear only in moderate magnitude ranging generally from 1.5- to 2-fold (Yan et al., 2002; Pastinen et al., 2004). Here reported genotype associated differences in mRNA levels of ICOS and CTLA4 fit well to this picture. Similar to Ueda et al. (2003) we found an association of CTLA4 polymorphism CT60 with mRNA levels of sCTLA4 in resting T lymphocytes. However, when we further studied activated cells the association between a low production and CT60 GG homozygosity was no longer observed. Hence, this genetic variation could regulate the basic level of sCTLA4 expression. It is of note that the autoimmune susceptibility allele CT60G was found to produce lower levels of sCTLA4 mRNA, whereas in autoimmune diseases the serum concentration of sCTLA4 has been reported to be increased (Pawlak et al., 2005). On the other hand, no link between CT60 genotypes and serum sCTLA4 concentration was found in type 1 diabetic patients or controls in another study (Purohit et al., 2005). Thus, the nature of this association and the role of sCTLA4 in immune activation remain to be elucidated. Furthermore, the results by Anjos et al. (2005) did not support the association between CT60 genotype and sCTLA4 mRNA levels. There were, however, some possibly important methodological differences between the studies, such as the cell populations investigated and how the expressions were measured. To study further the possible differential regulation between sCTLA4 and flCTLA4 isoforms, we analyzed the correlation of their mRNA levels during the activation. We found a surprisingly strong and statistically significant correlation between their levels, indicating that during the immune activation through CD28 and TCR signalling the two CTLA4 isoforms are not differentially regulated. In a murine model of type 1 diabetes the expression of both CTLA4 and ICOS was found to be affected (Greve et al., 2004). Murine susceptibility locus Idd5.1, which maps to mouse equivalent to human chromosome 2q33, led to increased ICOS but decreased flCTLA4 expression as measured by the flow cytometry. This supports our observation suggesting that disease susceptibility locus on chromosome 2q33 might lead to expression differences in both genes. This type of genetic complexity can readily result in controversial results in genetic association studies.

The above genetic and functional evidence, however, clearly demonstrate that various functionally interesting genetic variations exists on the 2q33 gene region and they are likely to regulate the expression levels of these molecules. How these variations lead to increased susceptibility to autoimmune disorders is, however, not clear at the moment. Acknowledgements This study was supported by the Sigrid Juselius Foundation and EVO Medical Research Fund of Finnish Red Cross Blood Service. References Anjos, S., Nguyen, A., Ounissi-Benkalha, H., Tessier, M.C., Polychronakos, C., 2002. A common autoimmunity predisposing signal peptide variant of the cytotoxic T-lymphocyte antigen 4 results in inefficient glycosylation of the susceptibility allele. J. Biol. Chem. 277, 46478–46486. Anjos, S.M., Shao, W., Marchand, L., Polychronakos, C., 2005. Allelic effects on gene regulation at the autoimmunity-predisposing CTLA4 locus: a reevaluation of the 3 +6230G>A polymorphism. Genes Immun. 6, 305–311. Atabani, S.F., Thio, C.L., Divanovic, S., Trompette, A., Belkaid, Y., Thomas, D.L., Karp, C.L., 2005. Association of CTLA4 polymorphism with regulatory T cell frequency. Eur. J. Immunol. 35, 2157–2162. Bonetti, A., Reunanen, K., Finnila, S., Koivisto, K., Wikstrom, J., Sumelahti, M.L., Pirttila, T., Elovaara, I., Reunanen, M., Saarela, J., Peltonen, L., Rantamaki, T., Tienari, P.J., 2004. A two-stage study on multiple sclerosis susceptibility and chromosome 2q33. Genes Immun. 5, 142–146. Gough, S.C., Walker, L.S., Sansom, D.M., 2005. CTLA4 gene polymorphism and autoimmunity. Immunol. Rev. 204, 102–115. Greve, B., Vijayakrishnan, L., Kubal, A., Sobel, R.A., Peterson, L.B., Wicker, L.S., Kuchroo, V.K., 2004. The diabetes susceptibility locus Idd5, 1 on mouse chromosome 1 regulates ICOS expression and modulates murine experimental autoimmune encephalomyelitis. J. Immunol. 173, 157–163. Haimila, K., Smedberg, T., Mustalahti, K., Maki, M., Partanen, J., Holopainen, P., 2004. Genetic association of coeliac disease susceptibility to polymorphisms in the ICOS gene on chromosome 2q33. Genes Immun. 5, 85–92. Haimila, K.E., Partanen, J.A., Holopainen, P.M., 2002. Genetic polymorphism of the human ICOS gene. Immunogenetics 53, 1028–1032. Knight, J.C., 2005. Regulatory polymorphisms underlying complex disease traits. J. Mol. Med. 83, 97–109. Kouki, T., Sawai, Y., Gardine, C.A., Fisfalen, M.E., Alegre, M.L., DeGroot, L.J., 2000. CTLA-4 gene polymorphism at position 49 in exon 1 reduces the inhibitory function of CTLA-4 and contributes to the pathogenesis of Graves’ disease. J. Immunol. 165, 6606–6611. Ligers, A., Teleshova, N., Masterman, T., Huang, W.X., Hillert, J., 2001. CTLA4 gene expression is influenced by promoter and exon 1 polymorphisms. Genes Immun. 2, 145–152. Magistrelli, G., Jeannin, P., Herbault, N., Benoit, D.C., Gauchat, J.F., Bonnefoy, J.Y., Delneste, Y., 1999. A soluble form of CTLA-4 generated by alternative splicing is expressed by nonstimulated human T cells. Eur. J. Immunol. 29, 3596–3602. Maurer, M., Loserth, S., Kolb-Maurer, A., Ponath, A., Wiese, S., Kruse, N., Rieckmann, P., 2002. A polymorphism in the human cytotoxic T-lymphocyte antigen 4 (CTLA4) gene (exon 1+49) alters T-cell activation. Immunogenetics 54, 1–8. Oaks, M.K., Hallett, K.M., Penwell, R.T., Stauber, E.C., Warren, S.J., Tector, A.J., 2000. A native soluble form of CTLA-4. Cell. Immunol. 201, 144–153. Pastinen, T., Sladek, R., Gurd, S., Sammak, A., Ge, B., Lepage, P., Lavergne, K., Villeneuve, A., Gaudin, T., Brandstrom, H., Beck, A., Verner, A., Kingsley, J., Harmsen, E., Labuda, D., Morgan, K., Vohl, M.C., Naumova, A.K., Sinnett, D., Hudson, T.J., 2004. A survey of genetic and epigenetic variation affecting human gene expression. Physiol. Genomics 16, 184–193.

T. Kaartinen et al. / Molecular Immunology 44 (2007) 1644–1651 Pawlak, E., Kochanowska, I.E., Frydecka, I., Kielbinski, M., Potoczek, S., Bilinska, M., 2005. The soluble CTLA-4 receptor: a new marker in autoimmune diseases. Arch. Immunol. Ther. Exp. (Warsz.) 53, 336–341. Purohit, S., Podolsky, R., Collins, C., Zheng, W., Schatz, D., Muir, A., Hopkins, D., Huang, Y.H., She, J.X., 2005. Lack of correlation between the levels of soluble cytotoxic T-lymphocyte associated antigen-4 (CTLA-4) and the CT-60 genotypes. J. Autoimmune Dis. 2, 8. Shilling, R.A., Pinto, J.M., Decker, D.C., Schneider, D.H., Bandukwala, H.S., Schneider, J.R., Camoretti-Mercado, B., Ober, C., Sperling, A.I., 2005. Cutting edge: polymorphisms in the ICOS promoter region are associated with allergic sensitization and th2 cytokine production. J. Immunol. 175, 2061–2065. Ueda, H., Howson, J.M., Esposito, L., Heward, J., Snook, H., Chamberlain, G., Rainbow, D.B., Hunter, K.M., Smith, A.N., Di Genova, G., Herr, M.H., Dahlman, I., Payne, F., Smyth, D., Lowe, C., Twells, R.C., Howlett, S., Healy, B., Nutland, S., Rance, H.E., Everett, V., Smink, L.J., Lam, A.C., Cordell, H.J., Walker, N.M., Bordin, C., Hulme, J., Motzo, C., Cucca, F., Hess,

1651

J.F., Metzker, M.L., Rogers, J., Gregory, S., Allahabadia, A., Nithiyananthan, R., Tuomilehto-Wolf, E., Tuomilehto, J., Bingley, P., Gillespie, K.M., Undlien, D.E., Ronningen, K.S., Guja, C., Ionescu-Tirgoviste, C., Savage, D.A., Maxwell, A.P., Carson, D.J., Patterson, C.C., Franklyn, J.A., Clayton, D.G., Peterson, L.B., Wicker, L.S., Todd, J.A., Gough, S.C., 2003. Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature 423, 506–511. Wang, X.B., Zhao, X., Giscombe, R., Lefvert, A.K., 2002. A CTLA-4 gene polymorphism at position −318 in the promoter region affects the expression of protein. Genes Immun. 3, 233–234. Wong, C.K., Lit, L.C., Tam, L.S., Li, E.K., Lam, C.W., 2005. Aberrant production of soluble costimulatory molecules CTLA-4, CD28, CD80 and CD86 in patients with systemic lupus erythematosus. Rheumatology (Oxford) 44, 989–994. Yan, H., Yuan, W., Velculescu, V.E., Vogelstein, B., Kinzler, K.W., 2002. Allelic variation in human gene expression. Science 297, 1143.