BMC Medical Genomics Biomed Central

BMC Medical Genomics BioMed Central

Research article Open Access A global transcriptional view of apoptosis in human T-cell activation Min Wang1, Dirk Windgassen2 and Eleftherios T Papoutsakis*1,3,4

Address: 1Interdepartmental Biological Sciences Program, Northwestern University, Evanston, IL, USA, 2Immunotherapy Development, Dendreon Corporation, Seattle, WA, USA, 3Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL, USA and 4Department of Chemical Engineering and the Delaware Biotechnology Institute, University of Delaware, Newark, DE, USA Email: Min Wang - [email protected]; Dirk Windgassen - [email protected]; Eleftherios T Papoutsakis* - [email protected] * Corresponding author

Published: 23 October 2008 Received: 17 June 2008 Accepted: 23 October 2008 BMC Medical Genomics 2008, 1:53 doi:10.1186/1755-8794-1-53 This article is available from: http://www.biomedcentral.com/1755-8794/1/53 © 2008 Wang et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract Background: T-cell activation is an essential step of immune response. The process of proper T- cell activation is strictly monitored and regulated by apoptosis signaling. Yet, regulation of apoptosis, an integral and crucial facet during the process of T-cell activation, is not well understood. Methods: In this study, a Gene-Ontology driven global gene expression analysis coupled with protein abundance and activity assays identified genes and pathways associated with regulation of apoptosis in primary human CD3+ T cells and separately CD4+ and CD8+ T cells. Results: We identified significantly regulated apoptotic genes in several protein families, such as BCL2 proteins, CASPASE proteins, and TNF receptors, and detailed their transcriptional kinetics during the T-cell activation process. Transcriptional patterns of a few select genes (BCL2A1, BBC3 and CASP3) were validated at the protein level. Many of these apoptotic genes are involved in NF- κB signaling pathway, including TNFRSF10A, TNFRSF10B, TRAF4, TRAF1, TRAF3, and TRAF6. Upregulation of NF-κB and IκB family genes (REL, RELA, and RELB, NFKBIA, NFKBIE and NFKB1) at 48 to 96 hours, supported by the increase of phosphorylated RELA (p65), suggests that the involvement of the NF-κB complex in the process of T-cell proliferation is not only regulated at the protein level but also at the transcriptional level. Examination of genes involved in MAP kinase signalling pathway, important in apoptosis, suggests an induction of p38 and ERK1 cascades in T- cell proliferation (at 48 to 96 hours), which was explored using phosphorylation assays for p38 (MAPK14) and ERK1 (MAPK3). An immediate and short-lived increase of AP-1 activity measured by DNA-binding activity suggests a rapid and transient activation of p38 and/or JNK cascades upon T-cell activation. Conclusion: This comparative genome-scale, transcriptional analysis of T-cell activation in the CD4+ and CD8+ subsets and the mixed CD3+ population identified many apoptosis genes not previously identified in the context of T-cell activation. Furthermore, it provided a comprehensive temporal analysis of the transcriptional program of apoptosis associated with T-cell activation.

Page 1 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

Background Flow cytometry The adaptive immune response starts with the activation The following monoclonal antibodies (Mabs) for flow of the naive CD4+ and CD8+ T cells in the peripheral cytometry were purchased from BD Biosciences (San Jose, immune system. Successful T-cell activation requires the CA) unless otherwise stated and included CD3 T-cell receptor complex (TCR) and the co-receptor CD28 (FITC+PE), active CASP3 PE, phospho-NF-κB-p65 PE, [1], the ligation of which leads to several downstream sig- phospho-p38 (MAPK14) PE, phospho-ERK1 (MAPK3) nalling events, including activation of protein kinases PE, PUMA (BBC3) (Cell Signaling Technology, Danvers, such as LCK and ZAP70, activation of MAP kinase cas- MA), BCL2A1 (Abcam, Cambridge, MA) and goat anti cades, and activation and nuclear localization of crucial rabbit IgG PE (Jackson ImmunoResearch Laboratories, transcription factors including AP-1, NFAT, and NF-κB West Grove, PA). Flow cytometry was carried out as [2]. In contrast, TCR signaling alone without CD28 co- described[8,9]. Briefly, all samples were gated on forward stimulation results in anergy and eventual cell death [3]. scatter and on propidium iodide negative (PI-) to elimi- nate debris and dead cells. For intracellular detections, Apoptosis has been extensively examined in T cells post cells were first stained with anti-CD3-FITC and then fixed, activation, such as activation-induced cell death (AICD), permeabilized, and stained as previously described [10]. due to its essential role in eliminating unwanted lym- Quantibrite beads (BD Biosciences Immunocytometry phocytes and maintaining the homeostasis after fighting Systems) labelled with different amounts of PE molecules infection and inflammation [4]. However, the regulation were used to quantify surface or intracellular protein lev- of apoptosis and the balance between the anti-apoptotic els and normalize measurements between timepoints. and pro-apoptotic signalling (which is an essential part of the surveillance machinery) during the process of T-cell Microarray experiments and data analysis activation have not been examined. Total RNA was extracted, RNA integrity was evaluated and microarray experiments and data analysis were carried out Genome-scale transcriptional analysis is a powerful tool as previously described [7]. Briefly, microarray data were for understanding complex processes such as T-cell activa- normalized and further analyzed (identification of signif- tion [5,6]. In a previous effort, using ontological analysis icant genes, hierarchical clustering, and Gene Ontology coupled with a comparative analysis of primary human T- assignment) with 'MultiExperiment Viewer (MeV)' from cell activation in the CD3+ T cells and the two subsets, The Institute for Genomic Research (TIGR) [11]. Raw and CD4+ and CD8+ T cells, we probed the common and normalized data were deposited in the Gene Expression potentially subset-specific immune response-associated Omnibus (GSE6607 (CD3+ T-cell experiment), GSE7571 transcriptome in T-cell activation [7]. In this study we (CD4+ T-cell experiment) and GSE7572 (CD8+ T-cell focus on the differentially expressed genes associated with experiment)) [12]. Within each population (three biolog- regulation of apoptosis, as well as essential apoptotic sig- ical replicates using cells from three different donors), nalling pathways: the NF-κB signalling pathway, and MAP multi-class SAM (Significance Analysis of Microarrays) kinase signalling. We identified several potentially impor- with a false discovery rate of < 1% was used to select genes tant apoptotic genes based on their patterns of expression that show statistically different expression between and examined the protein expression of a select set of groups. A SAM group is defined here as all the samples genes, most of which have not been previously discussed belonging to the same timepoint regardless of donor. in T-cell activation. Briefly, there were 5 groups (0 hour, 4, 10, 48 and 96 hours) in the set of CD3+ experiments and 6 groups (0 Methods hour, 6, 12, 24, 48 and 72 hours) in the set of CD4+ exper- Cells and culture system iments and CD8+ experiments. Gene expression at each CD3+, CD4+ and CD8+ T-cell cultures were set up as pre- time point was compared to that of 0 hour in each exper- viously described [7]. Briefly, negatively-selected T cells iment. Gene Ontology annotations, as curated by Euro- (CD3+, CD4+, and CD8+) were activated with anti-CD3/ pean Bioinformatics Institute, were retrieved from the anti-CD28 Mab conjugated to magnetic beads. Cell count- Gene Ontology Consortium website [13]. The EASE ing and sampling for flow cytometry and microarray anal- (Expression Analysis Systematic Explorer) score in onto- ysis were carried out at 0, 4, 10, 48 and 96 hours in the logical analysis is a modified Fisher Exact Probability p- CD3+ T-cell experiments, E1-E5, with cells from 5 inde- value [14] indicating the probability of finding by chance pendent healthy donors, and at 0, 6, 12, 24, 48 and 72 the same degree of enrichment on a Gene Ontology term hours in the CD4+ T-cell and CD8+ T-cell experiments, in a set of genes. The lower the EASE score, the more sig- E7-E11, with cells from 5 independent healthy donors. nificant is the enrichment, i.e. the less likely that degree of This study was approved by the Northwestern University enrichment can be found by chance. Hierarchical cluster- IRB. ing analysis was performed with the Euclidean distance metric. The list of genes associated with NF-κB signaling

Page 2 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

pathway was curated based on the information of Gene genes, associated with 'regulation of apoptosis' (EASE Ontology Consortium [13] per 'positive regulation of I- score: 7.65E-08), which shows that there is an active kappaB kinase/NF-kappaB cascade' and superarray [15] involvement of the apoptotic machinery during T-cell per 'NF-κB Signalling Pathway'. Information of superarray activation and proliferation. Made up of both anti- and enriched our list with members of the Rel, NF-κB, and IκB pro-apoptotic genes, these 125 genes shared well-pre- families. The list of NF-κB target genes was curated based served expression patterns among the three T-cell popula- on the information of website ([16]), a collective informa- tions (Figure 1A and Figure 1B). The mainly upregulated tion source of NF-κB research based on updated publica- cluster (Figure 1A) contains both anti- and pro-apoptotic tions. The gene list of MAP kinase signalling pathway was genes, and so does the mainly downregulated cluster (Fig- curated and sorted based on information of Kegg website ure 1B). This suggests an essential role of the balance [17] per 'MAPK signalling pathway', superarray [15] per between anti- apoptotic and pro- apoptotic signalling in 'MAP Kinase Signalling Pathway', and NCBI website [18]. T-cell activation.

AP-1 activity assay Members of the BCL2 family are key regulators of apopto- DNA-binding activity of AP-1 was assessed using the sis. The balance between pro- and anti-apoptotic BCL2 TransBinding AP-1 ELISA kit (Panomics; Fremont, CA) as family members determines the cellular fate in response described [19]. Briefly, nuclear extracts were incubated to survival cues and stress signals [23]. The functions and with biotinylated AP-1-consensus-binding-sequence oli- transcriptional regulation of these BCL2 family genes in T- gonucleotides and complexes were detected using a pri- cell activation remain largely unexplored. Alves et al. mary AP-1 antibody and a secondary antibody conjugated reported several significantly regulated BCL2 family genes to horseradish peroxidase. This assay is analogous to the in T-cell activation without discussion [24]. Our microar- traditional electrophoretic mobility shift assay in that it ray data identified a set of significantly regulated BCL2 measures the ability of a transcription factor from nuclear family genes reported by Alves [24], but with different lysates to bind to a consensus-binding sequence of that transcriptional patterns (Figure 1). These included contin- transcription factor, and has been extensively validated uously upregulated BCL2A1 (anti-apoptotic), early upreg- [20,21]. ulated MCL1 (myeloid cell leukemia sequence 1) (anti- apoptotic), BMF (BCL2 modifying factor) (pro-apoptotic) Results and PMAIP1 (pro-apoptotic), late (48 hours) downregu- Anti- and pro-apoptotic genes in T-cell activation lated BCL2 and dynamically (up-down-up) regulated As previously reported, within each population (CD3+, BBC3 (pro-apoptotic). Flow cytometric assays (sample CD4+ and CD8+ T cells), T cells from three independent flow cytometry histograms shown in Additional file 2 biological donors (three biological replicates) exhibited .)demonstrated the continuous upregulation of BCL2A1 overall similar phenotypic characteristics [7]. Briefly, the (Figure 2A) and upregulation of BBC3 at 0–10 hours and surface expression of the early T-cell activation marker 24–96 hours at the protein level (Figure 2C), both of CD69 and the middle activation marker CD25 (IL2RA) which are consistent with their transcriptional patterns. [22] were rapidly upregulated within 10 hours and 24 The BCL2A1 gene has been reported as a direct target of hours respectively; T-cell proliferation did not start until transcription factor NF-κB complex, p65/p50, in T cells 48 hours and cell numbers doubled by 96 hours following [25]. This indicates a continuous involvement of BCL2A1 T-cell activation. Accordingly, we divided our experimen- and a constant activity of the NF-κB (p65/p50) complex tal time course into early T-cell activation (0–10 hours), in T-cell activation. BCL2 has been hypothesized to be middle and late T-cell activation (10–48 hours), and T- able to block T-cell death [26]. The significant regulation cell proliferation (48–96 hours). Our microarray results of several genes of the BCL2 protein family and their inter- have been validated by Q-RT-PCR assays with a selection acting proteins, whose functions and transcription regula- of fifteen significant genes covering a broad range of tion have not been discussed in the context of T-cell expression patterns and intensities [7]. We have also activation, calls for attention to their role in T-cell activa- established the reproducibility of our genome scale tran- tion. These include upregulated early (anti-apoptotic scription data within each population (CD3+, CD4+ and BNIP2 (BCL2/adenovirus interacting protein 2)), at 10– CD8+ T cells) and across the three populations [7]. SAM 48 hours (anti-apoptotic BAG1 (BCL2-associated athano- analysis identified a total of 4167 unique, significant reg- gene)), and late BAG2, BCL2L12, and pro-apoptotic ulated genes in T-cell activation, with similar transcription BNIP3, and downregulated pro-apoptotic BNIP3L and patterns in three replicate biological experiments within pro-apoptotic MOAP1 (Figure 1). each population [7]. Caspases play a central role as executioners in most types Following SAM analysis, ontological analysis using the of apoptosis, including activation induced cell death MeV EASE module identified 125 significantly regulated (AICD). However they have not been discussed during T-

Page 3 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

ABCD3+ CD4+ CD8+ CD3+ CD4+ CD8+ GeneName Max/Min Median GeneName Max/Min Median 04h/0h 10h/0h 48h/0h 96h/0h 06h/0h 12h/0h 24h/0h 48h/0h 72h/0h 06h/0h 12h/0h 24h/0h 48h/0h 72h/0h 04h/0h 10h/0h 48h/0h 96h/0h 06h/0h 12h/0h 24h/0h 48h/0h 72h/0h 06h/0h 12h/0h 24h/0h 48h/0h 72h/0h IL2 91.3 23.1 BNIP3 3.5 1.2 HSPE1 29.1 14.6 CASP9 2.7 -1.0 HSPD1 18.9 6.4 HSPA5 3.1 -1.3 GZMB 37.7 8.1 TRAF6 1.4 1.0 IER3 11.0 7.6 BNIP2 -1.6 -1.1 TRAF4 7.5 2.5 SON -1.9 -1.1 BCL2A1 17.0 5.0 F2 -1.7 -1.1 TNFSF6 20.6 6.5 TNFRSF10A -2.3 -1.1 CFLAR 8.4 5.2 RELA -1.8 -1.0 UTP11L 4.7 3.6 MCL1 2.8 -1.1 TUBB2 4.6 3.5 BBC3 -4.1 -1.7 BRCA1 8.3 4.0 ANXA1 -9.2 -3.7 BIRC5 15.4 2.0 BTG1 -7.3 -3.6 MYBL2 5.7 1.1 TNFAIP3 -12.3 -3.5 IKIP 6.5 2.0 MAL -6.9 -4.2 GLO1 4.4 2.7 NALP1 -6.8 -4.4 CASP3 7.5 4.2 BNIP3L -8.0 -3.3 GSTP1 4.5 3.0 CASP1 -3.7 -1.9 TRAP1 3.4 2.6 CASP6 2.6 -1.2 PDCD5 4.1 2.8 PERP 2.3 -1.1 MIF 5.1 3.4 CTSB 2.2 -1.5 AIFM1 2.9 2.0 CASP8AP2 -2.9 -1.7 BARD1 3.2 1.6 MTL5 -3.9 -1.6 HMGB1 3.2 1.3 BCL2 -2.8 -1.3 NUDT2 3.6 1.3 F2R -3.0 -1.5 PRDX2 2.8 1.4 LCK -2.9 -1.5 FAIM -2.5 1.2 PCBP4 -2.1 -1.6 SIVA 2.0 1.2 MAP4K1 -1.9 -1.2 FAF1 -1.8 1.2 TNFRSF1A -2.0 1.0 CDK5 1.5 1.2 HDAC1 -1.8 -1.2 HSP90B1 2.0 1.2 PRODH -1.7 -1.4 NME6 1.7 1.2 HTT -2.1 -1.5 ERCC2 -2.0 1.0 TNFRSF25 -5.8 -1.8 CD38 4.4 2.1 S100B -3.7 -2.0 TNFRSF8 3.2 1.7 FEM1B -2.9 -2.4 STAT1 4.9 1.5 STK17B -5.1 -3.1 TNFRSF18 5.7 3.1 JMY -3.6 -2.5 EEF1E1 5.0 2.7 BIRC2 -2.5 -1.9 NME1 4.7 3.8 DDAH2 -3.8 -2.4 AATF 4.0 2.6 PIM1 -2.8 -1.9 PHB 3.1 2.3 TRADD -3.3 -2.7 SPHK1 3.0 2.1 PYCARD -4.1 -2.8 HSPA9 2.4 2.0 TRAF5 -3.3 -2.3 DHCR24 3.1 2.2 NME3 -2.8 -2.2 BAG2 4.2 2.0 BIRC1 -4.2 -3.0 TXNDC5 3.1 2.2 CD3G -3.9 -2.5 BCL2L12 2.5 1.8 FOXO1A -3.5 -2.7 EI24 3.0 1.7 MX1 -4.2 -2.2 AVEN 2.4 1.8 CASP4 -3.0 -2.0 DNAJA3 2.2 1.7 PTEN -3.0 -1.9 AIFM2 2.1 1.6 MOAP1 -1.9 -1.7 TNFSF5 5.2 -1.0 NLRP3 4.1 -1.2 BIRC3 9.9 1.9 -3 03 TNFRSF1B 4.4 2.2 TNFRSF9 3.6 2.1 PMAIP1 3.6 1.2 TNFRSF10B 2.9 1.9 C TP53BP2 4.2 1.6 BBC3 BNIP3 STK4 4.3 2.0 TRAF3 4.7 2.5 PMAIP1 BNIP3L TRAF1 4.4 2.5 BCL2 TNF 3.0 2.6 IFI16 3.0 1.8 BCL2A1 BNIP2 TEGT 2.6 1.9 BMF MOAP1 CYCS 4.2 2.0 TNFAIP8 3.0 1.5 BAG1 2.3 1.6 MCL1 BAG1 CUL1 1.7 1.3 TRIAP1 1.7 1.4 POGK 1.7 1.5 Apoptosis BMF 2.0 1.5 BAG2 BCL2L12 TNFRSF6 2.3 1.2 NFKB1 3.5 1.2 CDKN1A 5.7 1.5 MALT1 2.0 1.1

ExpressionFigure 1 profiles of genes associated with regulation of apoptosis Expression profiles of genes associated with regulation of apoptosis. Genes that were differentially expressed tempo- rally in T-cell activation of the three (CD3+, CD4+ and CD8+) populations were divided into two groups (A with mostly upregulated genes, and B with mostly downregulated genes) according to their distinct expression patterns based on hierarchical clustering using the Euclidean distance metric. Color denotes degree of differential expression compared to 0 hour (saturated red = 3-fold up-regulation, saturated green = 3-fold down-regulation, black = unchanged, gray = no data available). Expression data shown are averages from three independent biological experiments for each T-cell population. The median ratio, along with the maximum (for up-regulated genes) or minimum (for down-regulated genes) ratio of stimulated T cells at each timepoint (with respect to the expression of 0 hour) is provided (a negative value represents down-regulation). Pro-apoptotic genes names and descriptions are shown in red, anti-apoptotic genes are shown in green and genes with both pro- and anti-apoptosis roles are shown in blue. Genes with unknown functions in apoptosis are shown in black. (C) Schematic view of significantly regulated genes of BCL2 family. Green and red connections denote negative and positive regulation of apoptosis, respectively, based on information of NCBI website [18]. Regulation of gene transcription in CD3+ T cells, compared to 0 hour, is denoted by different color (green: downregulation, red: upregulation) at each timepoint in the sequence of 4, 10, 48 and 96 hours.

Page 4 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

cell activation. Our microarray data show that numerous (IAP) family, and genes of programmed cell death caspase genes were differentially expressed. Contrary to (PDCD) proteins. IAPs inhibit apoptosis by interfering the signalling mechanism whereby CASP9 becomes acti- with activation of caspase proteins [37]. The differentially vated first, and then in turn activates CASP3 and CASP6 expressed genes of the IAP family displayed different [4], at the transcriptional level, CASP3 was upregulated at expression patterns: BIRC1 and BIRC2 were downregu- 24–96 hours, earlier than CASP9 and CASP6, which were lated, BIRC5 was down-then-upregulated, and BIRC3 was only upregulated at 96 hours. The flow cytometric assay up-down-upregulated. Of note, pro-apoptotic PDCD5 specific for the active form of caspase 3 protein (sample and AIF1 (PDCD8) were upregulated at 10–96 hours. flow cytometry histograms shown in Additional file 2.) supported the involvement of CASP3 by 48 hours (Figure Non-caspase executor proteases, such as Granzyme B 2E). CASP8AP2, which is required in CASP8 mediating (GZMB) and CTSB (cathepsin B), were also significantly apoptosis [27], was downregulated at 4–10 hours and regulated. Besides its function in inducing target cell then upregulated to the same level of resting T cells at 48– apoptosis, intracellular degranulated GZMB has also been 96 hours. CASP1, as well as its adaptor PYCARD [28], and implicated in AICD in TH2 cells [38]. Our microarray data CASP4 displayed decreased expression throughout, sug- revealed that GZMB was continuously upregulated with gesting that CASP1 and CASP4 might play important roles similar transcription patterns in CD4+ T cell and CD8+ T in the homeostasis of resting T cells, but not in T-cell acti- cells, suggesting an involvement of GZMB in T-cell activa- vation. Caspase regulatory protein AVEN (reportedly an tion. CTSB has been reported to promote T-cell apoptosis inhibitor of CASP9 activation) [29] was upregulated at by immune-suppressive anti-T cell agents, mitogen 10–48 hours, concomitantly with the upregulation of antithymocyte globulins (ATGs) [39]. Transcription of CASP9 at 96 hours. CFLAR (CASP8 and FADD-like apop- CTSB was downregulated first (at 4–48 hours) then upreg- tosis regulator) was significantly and continuously upregulated at 96 hours, thus suggesting a role of CTSB in the ulated. CFLAR has different roles in T cells at different early stage of T-cell proliferation. stages. It has been reported that CFLAR is induced by res- timulaiton in activated T cells, inhibiting FAS-mediated The TNF receptor family plays important roles in extrinsi- apoptosis [30], while overexpression of CFLAR in naive T cally induced apoptotic pathways. Some TNF receptors cells decreased T-cell proliferation upon anti-CD3/anti- have death domains and are directly involved in apopto- CD28 stimulation [31]. The strong transcriptional upreg- sis. Transcription of several TNF receptors (TNFRSF6, -8, - ulation of CFLAR in T-cell activation, not previously 9, -18, -1A, -1B, -10A, -10B, and -25) was differentially reported, suggests an important role during the process of regulated. Of note, TNFRSF6 (FAS), the well-known AICD primary T-cell activation. receptor, was upregulated at 4–10 hours, but not the other two components of the death-inducing signalling com- Some heat shock proteins are involved in apoptosis, but plex (DISC): FADD and CASP8 [40]. FAS has also been none has been discussed in the context of T-cell activation. implicated in multiple pathways including NF-κB, extra- HSPE1 and HSPD1 shared continuously elevated expres- cellular signal-regulated protein kinase (ERK) 1 and -2, sion at 4–96 hours. It has been hypothesized that HSPE1 and p38; however, the function of FAS in early T-cell acti- and HSPD1 form a complex with and facilitate the activa- vation has not been reported [4]. Contrary to the reported tion of pro-caspase 3 [32]. Anti-apoptotic interacting heat induction of death-receptors TNFRSF1A and TNFRSF25 shock proteins HSPA9 and HSP90B1 [33,34] were also (DR3) in T-cell activation [41,42], our microarray data upregulated; their role remains unknown in T-cell activa- show that both are downregulated, together with TRADD tion. Of note, reportedly anti-apoptotic HSPA5 [35], dem- (TNFRSF1A-associated death domain protein), their com- onstrated distinct transcription patterns in the three mon adaptor protein. These data suggest that the apopto- populations: it was upregulated in CD3+ T cells especially sis pathways mediated by TNFRSF1A/TRADD and at 48 hours, but overall downregulated in both CD4+ and TNFRSF25/TRADD are suppressed upon T-cell activation. CD8+ T cells. HSPA5 has been reported to retain T-cell TNFRSF1B was upregulated and more significantly so in antigen receptor alpha chain (TCR-alpha) within the CD8+ T cells. The function of TNFRSF1B in T cells remains endoplasmic reticulum [36]. These different expression controversial, with both anti-apoptotic and pro-apoptotic patterns among the three T-cell populations indicate that functions reported [43,44]. Furthermore, TNFRSF1B has HSPA5 might be involved in the interaction between not been reported to be T-cell subset (CD4+ or CD8+) CD4+ and CD8+ T cells in T-cell activation. specific. BIRC3 (also known as IAP1), component of the TNFRSF1B signalling complexes inducing apoptosis [45], The transcriptional kinetics of other significantly regu- was more significantly upregulated in CD8+ T cells, simi- lated, and not-previously reported in T-cell activation larly to TNFRSF1B. This suggests that the TNFRSF1B sig- apoptotic genes are also worth discussing briefly. These nalling might have CD8+ specific functions. A few TNFSF include members of the inhibitor of apoptosis protein and TNFRSF related proteins, the function of which

Page 5 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

remain largely unknown, were significantly regulated. A B BCL2A1 BCL2A1 These included FAIM (FAS apoptotic inhibitory mole- 24 2424 E4 CD3+ cule), FAF1 (Fas associated factor 1), and SIVA (CD27 E5 CD4+ (TNFRSF7)-binding protein)), which were downregulated 18 1818 CD8+ first and then upregulated; and TNFAIP8, which was upregulated first and then downregulated. 12 1212 Numerous TNF receptors and related proteins involved in NF-κB signalling pathway were significantly regulated in 6 66 T-cell activation. TNFRSF8, reportedly able to promote apoptosis through inducing the activation of NF-κB com-

Protein expression Protein expression ratio (vs. 0h) 0 00 plex [46], was significantly upregulated at 48–96 hours.

Transcription expression expression ratio Transcription (vs. 0h) 09624 48 72 0 24 2448729648 72 96 0 24487296 TNFRSF9, reportedly able to promote apoptosis and sup- Timepoint (hours) Timepoint (hours) press the activation of NF-κB complex [47], was mainly upregulated. TNFRSF10A (TRAILR1) and TNFRSF10B C BBC3 D BBC3 4 66 (TRAILR2), which can interact with several members of E4 CD3+ TRAFs to activate NF-κB complex [48], together with sev- E5 CD4+ 3 33 eral members of the TRAF family (TRAF4, TRAF1, TRAF3, CD8+ and TRAF6) were upregulated. A detailed examination of the transcriptional orchestration of NF-κB signaling path- 2 00 way in T-cell activation is presented next. xpression ratio (vs. (vs. 0h) ratio xpression 1 n expression ratio (vs. 0h) -3-3 NF-κB signalling pathway e o o The transcription factor NF-κB complex, a collection of several homodimers or heterodimers of Rel proteins (REL,

Protein 0 -6-6 096024 2448729648 72 09624 48 72 RELA (p65), RELB, p50 and p52), plays a key role for the

Transcripti 0 24487296 Timepoint (hours) Timepoint (hours) regulation of T-cell activation by mediating the induction of various genes that control T-cell proliferation, activa- E F Active CASP3 Active CASP3 tion and survival [49]. A wide array of stimuli including 4 88 E4 IL1, and TNF as well as TCR stimulation lead to the onset E5 E1 of cascades that ultimately lead to NF-κB activation [2]. 3 66 Due to the broad range of the upstream signalings and the complexity of the dimers, our knowledge of the orches- 2 44 trated regulation of NF-κB signalling pathway and activity of the different NF-κB dimers in T-cell activation is far CD3+ from complete. The transcriptional regulation of signifi- 1 22 CD4+ cantly regulated genes associated with the NF-κB signal- CD8+ ling pathway is shown in Figure 3A and Figure 4A. NF-κB Protein expression Protein expression ratio (vs. 0h) 0 00 family genes (REL, RELA (p65), and RELB) and IκB family

09624 48 72 expression ratio Transcription (vs. 0h) 09624 48 72 0 24487296 0 24487296 genes (NFKBIA (the inhibitor of RELA), NFKBIE (inhibi- Timepoint (hours) Timepoint (hours) tor of REL) and NFKB1 (p105, precursor of p50)) shared ProteinactiveFigure CASP3 expression2 profiles of (A) BCL2A (C) BBC3 and (E) similar transcription patterns, an early upregulation at 4– Protein expression profiles of (A) BCL2A (C) BBC3 10 hours was followed by a decrease at 48–96 hours. IKIP and (E) active CASP3. Protein expression kinetics sup- (IKK interacting protein) was significantly upregulated at ports the transcriptional patterns of selected genes demon- 48–96 hours (Figure 3A). We also examined the intracel- strated by microarray analysis. CD3+ T cells were selected, lular protein expression of phosphorylated p65, the major stimulated (with anti-CD3/anti-CD28 antibodies), cultured active component of the NF-κB complex. Flow cytometric and harvested at the indicated timepoints of culture to ana- analysis (sample flow cytometry histograms shown in lyze the protein expression by flow cytometric assays. For Additional file 2.) demonstrated an increase of the phos- BCL2A and BBC3, data from two independent experiments, phorylated p65 at 48–96 hours (Figure 3B). This activa- E4 and E5, are shown; for CASP3, data from three independent experiments, E1, E4 and E5, are shown. (B), (D) and (F) tion delay is likely the result of the strong upregulation of demonstrated the transcriptional patterns of BCL2A1, BBC3 NFKBIA at 4 and 10 hours. It is also possible that p65 and CASP3 in all three populations (CD3+, CD4+ and might quickly become activated within 4 hours leading to CD8+). the transcriptional induction of NFKBIA, one of the target genes of the NF-κB complex [50]. Of note, not only the

Page 6 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

transcriptional regulation of RELB (maximum fold CD3+ CD4+ CD8+ A -3 0 3 change of 7.4) was more significant than that of RELA GeneName Max/Min Median 04h/0h 10h/0h 48h/0h 96h/0h 06h/0h 12h/0h 24h/0h 48h/0h 72h/0h 06h/0h 12h/0h 24h/0h 48h/0h 72h/0h FOS -149.3 -63.8 # # # # # # # # # # # # # # (maximum fold change of 6.5), but also RELB had higher JUN -159.5 -47.4 # # # # # # # # # # # # # # SCOTIN -4.3 -2.8 # # # # # # # # # # # # # # transcriptional levels than RELA. BCL3, a transcriptional APOL3 -3.6 -2.2 # # # # # # # # # # # # # # CASP1 -3.5 -2.1 # # # # # # # # # # # # # 0 F2R -3.0 -1.2 # # # # # # # 0 # # # 0 # # coactivator of NF-κB homodimer p50/p50 and p52/p52 TRAF5 -3.4 -2.2 # # # # # # # # # # # # # # TRADD -3.5 -2.7 # # # # # # # # # # # # # # FLNA -3.1 -1.7 # # # # # # # # # # # # # # [51], was upregulated at 4 and 96 hours. TNFRSF1A -2.0 1.0 # # # # # 0 0 0 0 # 0 0 0 0 TMEM9B -2.5 -1.4 # # # # # # # # # # # # # # ZDHHC17 -2.1 -1.4 0 # # # 0 # # # # # # # # # TLR1 -1.7 -1.5 0 # # # # # # # # # # # # # PPM1A -1.6 -1.3 TLR1 (toll-like receptor) and TICAM1 (toll-like receptor RHOH -2.6 -1.7 1 # # # 1 # # # # 0 # # # # BIRC2 -2.4 -1.8 1 # # # # # # # # # # # # # TNFAIP3 -10.5 -3.5 0 # # # # # # # # # # # # # adaptor), which promote activation of the NF-κB complex PLK2 -5.7 -3.9 1 # # # # # # # # # # # # # MAP3K7IP2 -2.7 -1.7 1 # # # 1 # # # # 0 # # # # SLC20A1 -2.1 -1.3 0 # 0 # 1 # # 0 # 1 # # # # [52], showed decreased expression at 48–96 hours, sug- RELA -1.9 -1.1 0 0 # # 0 0 # # # 1 0 0 # # PPM1A -1.6 -1.3 0 # # # 0 # # # # 0 # # # # gesting that this upstream activation cascade of the NF-κB TNFRSF10A -2.3 -1.1 1 0 # # 1 0 0 # # 1 # # # # TLR8 -3.1 -1.5 0 0 # # 2 # # # # # # # # # MAP3K14 1.7 1.3 0 1 # # 0 1 1 # # 0 1 0 # # complex might not be active during early T-cell prolifera- IL1B 1.7 -1.1 1 1 # # 0 # # # # # 0 0 0 # CARD9 3.1 1.6 1 1 0 # 0 2 1 1 0 1 1 1 0 # RFP2 2.7 1.5 # 1 # # 0 1 1 1 0 1 1 1 1 1 tion. Also involved in NF-κB complex activity regulation, STAT1 4.2 1.5 # 1 0 # 0 2 2 1 0 0 2 2 1 0 SEC61A1 3.7 1.7 members of the TNF receptor super family (TNFRSF6, -8, VAPA 1.7 1.0 1 1 1 # 0 0 0 0 # 0 0 0 # # TRAF6 1.4 1.0 1 0 0 # 0 0 0 # # # 0 # # # NDFIP1 2.4 1.2 1 1 0 # 1 1 0 0 # 1 1 0 0 # -9, -10A, and -10B), their associated proteins (TRAF1, -3, - TICAM1 2.2 1.5 1 1 0 0 1 1 1 1 1 1 1 1 0 0 TNFRSF6 2.3 1.2 1 1 # 1 0 0 # # # 0 0 0 0 0 MAP3K8 2.8 1.4 1 1 0 1 1 1 0 0 1 1 0 0 # 0 4 and -6, EDARADD (ectodysplasin A receptor-associ- EGR1 -14.3 -2.5 3 1 # 1 0 # # # # 0 # # # # NFKB1 3.5 1.2 2 1 0 0 1 1 0 # # 1 0 0 # # ated)) and TRAF interacting protein (TRAIP) showed LGALS1 6.2 -1.5 RHOC 1.7 -1.2 # # # 1 # # # # 1 # # # 0 0 TBK1 2.2 1.2 # # 0 0 # # 0 1 1 # 0 0 1 1 increased expression. TNFRSF-10A, -10B, TRAF1, -3, and - TRAF3IP2 -1.9 -1.0 # # # 1 0 0 # # 0 # 0 # 0 1 NFKBIL1 2.8 1.2 # # 0 1 # # 1 1 1 # # 0 0 1 ECT2 2.6 1.0 0 # 1 1 0 # # 0 0 # # # 0 0 6 were upregulated at 4–10 hours and TRAIP, which TFG 2.3 1.6 1 1 1 1 # 1 1 1 1 # 1 1 0 1 MALT1 2.4 1.2 1 0 0 1 1 # # # 0 1 # # # 0 κ FKBP1A 3.6 2.8 inhibits the TRAF-mediated NF- B activation [53], was TNFRSF9 3.6 2.1 1 1 1 1 0 1 1 1 1 2 2 2 2 2 UBE2N 2.8 2.0 upregulated at 48–96 hours. This orchestrated gene TNFRSF8 5.9 2.5 1 1 2 1 1 1 2 3 2 1 1 1 2 1 NFKBIL2 3.6 1.8 0 0 2 1 # 1 1 1 1 # 1 1 2 1 TRAIP 4.2 2.0 0 # 2 2 # 0 1 2 1 # 0 1 2 2 expression regulation suggests that the TRAF-mediated IKIP 6.5 2.2 BCL3 3.8 1.1 1 1 1 2 # 0 0 0 1 # # # # # TNFRSF10B 2.9 1.9 2 1 1 1 1 1 1 1 1 1 1 1 1 1 upstream signalling of NF-κB activation is active at 4–10 TRAF3 4.6 2.6 2 1 1 1 2 2 2 1 1 2 1 1 0 1 TNF 6.4 4.9 3 3 2 1 2 2 2 1 1 2 3 2 2 2 hours. Furthermore, the early upregulation of MAP3 NFKBIA 7.5 -1.9 3 3 1 2 # # # # # # # # # # RELB 7.4 3.0 3 2 1 1 2 2 2 1 1 2 2 2 1 0 REL 8.7 2.1 3 2 1 1 2 1 1 0 0 3 1 1 1 1 Kinases (MAP3K14, MAP3K8, MAP3K7IP2) supports the TNFSF6 22.0 6.8 2 2 2 1 4 4 3 3 3 3 2 2 3 3 TRAF1 4.4 2.4 2 2 1 1 1 1 1 1 1 1 1 1 1 1 EDARADD 6.0 4.0 1 2 2 2 1 3 3 2 2 2 2 2 2 2 early involvement of the TNF receptor pathway since CFLAR 8.4 5.3 2 2 2 2 3 2 3 2 2 3 3 3 2 3 NFKBIE 4.9 1.5 2 2 2 2 1 1 1 0 0 0 0 # # # MAP3 Kinases activate IκB kinases recruited by TRAFs to TRAF4 7.5 2.5 2 3 2 2 1 2 2 1 0 1 1 1 1 #

B o C RELA the TNF receptor complex [54]. In contrast, TNFRSF1A, Phosporylated p65 2 2 8 CD3+ TRADD (TNFRSF1A-associated via death domain) and E4 1 6 1 CD4+ E5 CD8+ TRAF5 were downregulated, suggesting that the reported 4 rati xpression 0 0 xpression ratio xpression TNFRSF1A-TRADD-TRAF2 cascade [55] and TRAF5 medi- 2 -1 0h) (vs. -1 (vs. 0h) ated cascade [56], regulating the activation of NF-κB com- 0 -2-2 0 24 48 72 96 0 24 48 72 96 Protein e 0Timepoint 24487296 (hours) 0Timepoint 24487296 (hours) plex, might not be active in T-cell activation. Transcription e SignificantlyFigure 3 regulated genes in NF-κB signalling pathway TCR specific signalling protein MALT1 (through the Significantly regulated genes in NF-κB signalling CARD11-BCL10-MALT1 complex) was recently reported pathway. (A) Expression profiles of genes involved in NF- to be required for optimal NF-κB activation through pro- κB signalling pathway. Color denotes degree of differential teolysis of the NF-κB inhibitor TNFAIP3 [57,58]. However expression compared to 0 hour (saturated red = 3-fold up- the transcriptional regulation of MALT1 and TNFAIP3 in regulation, saturated green = 3-fold down-regulation, black = T-cell activation has not been reported. Our microarray unchanged, gray = no data available). Expression data shown are averages from three independent biological experiments data demonstrated that MALT1 was upregulated at 4 and for each T-cell population. The median ratio, along with the 96 hours, and TNFAIP3 was upregulated at 4 hours and maximum (for up-regulated genes) or minimum (for down- downregulated at 10–96 hours. CARD11 and BCL10 were regulated genes) ratio of stimulated T cells at each timepoint not identified as significantly regulated, however. CARD9, (with respect to the expression of 0 hour) is provided (a neg- the equivalent gene of CARD11 in dendritic cells [59], was ative value represents down-regulation). Genes, whose tran- upregulated at 4 and 10 hours, suggesting its involvement scription can induce the activation of the NF-κB complex, in T-cell activation. identified in a large scale screening study [60] were shown in blue. (B) Protein expression profile of phosphorylated p65 Matsuda et al. identified genes whose transcription can of NF-κB complex. CD3+ T cells were selected, stimulated induce the activation of the NF-κB complex by introduc- (with anti-CD3/anti-CD28 antibodies), cultured and har- ing cDNA clones of full-length human cDNA libraries to vested at the indicated timepoints of culture to analyze the protein expression by flow cytometric assays. Data from two HEK 293 cells [60]. Our microarray data demonstrated independent experiments, E4 and E5, are shown. (C) The that several of these genes were significantly regulated transcriptional pattern of RELA in all three populations with different transcription patterns (gene names shown (CD3+, CD4+ and CD8+). in blue in Figure 3A). Some of these genes were upregulated, such as UBE2N, ECT2, TFG, and FKBP1A; while oth-

Page 7 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

A TLR1 TNFRSF6 TNFRSF8 TNFRSF9 Receptors TNFRSF10A TNFRSF10B TNFRSF1A Receptor TICAM1 TRAF3 TRAF1 TRAF4 EDARADD Associated Proteins TRAF5 TRAF6 TRAIP TRADD TRAF3IP2

MAP3 Kinases MAP3K14 MAP3K8 MAP3K7IP2

TCR Downstream MALT1 TNFAIP3 IKIP IκB Family NFKBIA NFKBIE NFKBIL1 NFKBIL2

NFκB Family RELA RELB REL NFKB1

B RAS Regulating RASGRP1 RASGRP2 Proteins GRAP1 ERBB1 ADRB2

RAS Proteins HRAS KRAS ADORA2B Phosphatases RAC Proteins RAC2 CDC42 DUSP5 MBIP JNK Upstream MAP4K2 MAP4K1 DUSP8 Kinases MAP4K4 MAP3K2 MAP3K12

DUSP1 p38/ERK MAPK14 MAPK3 Kinases MKNK1 MAP2K6 Other Kinases MAPK6 MAPK12

Transcription Factors FOS JUN

PathwayFigure 4 schematic of significantly regulated genes in (A) NF-κB signalling and (B) MAP kinase signalling Pathway schematic of significantly regulated genes in (A) NF-κB signalling and (B) MAP kinase signalling. Mem- bership was manually determined from the corresponding gene pages in NCBI [18] and references therein. The regulation of gene transcription in CD3+ T cells, compared to 0 hour, is denoted by different color (green: downregulation, red: upregulation) at each timepoint in the sequence of 4, 10, 48 and 96 hours.

Page 8 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

ers were downregulated, such as SCOTIN, APOL3 and Several kinases in the MAP kinase signalling pathway were FLNA. Two NF-κB inhibitor-like proteins (NFKBIL1 and differentially expressed. A few kinases upstream of the NFKBIL2), whose function has not been determined, were JNK cascade (MAP4K2 [71], MAP4K1 [72], MAP3K12 significantly upregulated at 96 hours and 48 hours respec- [73], MAP4K4 [74], MAP3K2 [75]) were downregulated, tively, suggesting that they are involved in early T-cell pro- and MBIP (MAP3K12 binding inhibitory protein), which liferation and late T-cell activation, respectively. inhibits the MAP3K12 mediated JNK activation [76], was upregulated, suggesting that the JNK cascade might not be Some genes (EGR1, NFKBIA and NFKBIE) had signifi- active at 4–96 hours. MAPK14 (p38), MAPK3 (ERK1), cantly different expression patterns in CD4+ and CD8+ T MKNK1 (the interacting protein of both p38 and ERK1 cells compared to CD3+ T cells. NFKBIA and NFKBIE are [77]) and MAP2K6 (p38 specific MAP kinase kinase [78]) inhibitors of REL proteins, and EGR1 has been reported to displayed similar regulation patterns: downregulated at inhibit the activity of RELA [61]. It is possible that the 4–10 hours and then upregulated at 48–96 hours. The communication between CD4+ and CD8+ T cells may flow cytometric assays (sample flow cytometry histograms affect the activation of the NF-κB complex. Activation of shown in Additional file 2.) of phosphorylated p38 and the NF-κB complex is further supported by the significant ERK1 confirmed their transcriptional patterns, namely regulation of many target genes, whose transcription is that there was no significant increase of phosphorylated regulated by NF-κB complex upon anti-CD3/anti-CD28 p38 until 24 hours and phosphorylated ERK1 until 48 stimulation (Additional file 1). Interestingly, some of hours, but large increases after that until 96 hours (Figure these target genes showed higher expression in resting T 5B and Figure 5D). Kinases involved in positive regulation cells. of activity of the NF-κB complex, MAP3K14 [79] and MAP3K8 [80], showed increased expression at 4–10 hours MAP kinase signalling which is consistent with expression patterns of several Mitogen-activated protein (MAP) kinases are important members of NF-κB and IκB family genes (Figure 3A). Two signalling mediators in regulation of apoptosis, including less examined kinases, MAPK6 and MAPK12, were mainly the anti-apoptotic ERKs, the anti-/pro-apoptotic c-Jun N- upregulated at 4–48 hours and at 10–96 hours respec- terminal kinases (JNKs), and anti-/pro-apoptotic p38- tively, suggesting that they are actively involved in T-cell MAPKs. Yet the mechanisms as to how these MAP kinases activation. regulate apoptosis remain controversial [62]. It has been suggested that the three main mammalian cell MAP Several MAP kinase regulating proteins were also differen- kinase cascades, JNK, p38 and ERK, are essential for T-cell tially expressed upon anti-CD3/anti-CD28 activation of T functions [63,64]. However, the temporal regulation of cells. Members of the dual specificity phosphatase family, these MAP kinase cascades in T-cell activation remains negatively regulating members of the MAP kinase family, largely unexplored. Thus, we focused on the significantly show different expression patterns. DUSP1, able to inacti- regulated genes involved in MAP kinase signalling path- vate ERK1, JNK and p38 [81], was significantly downreg- way (Figure 4B and Figure 5). A few genes of the RAS fam- ulated throughout. In contrast, DUSP5, specific inhibitor ily and RAS regulating protein were upregulated upon of ERK1 [82], was strongly upregulated at 4–10 hours and anti-CD3/anti-CD28 T-cell stimulation, including HRAS decreased thereafter in concert with the upregulation of (at 10–48 hours), KRAS (at 4 hours), NRAS (at 4–96 ERK1 at 48–96 hours at both the transcriptional and pro- hours) and ADORA2B (at 4–96 hours). HRAS, KRAS and tein level (Figure 5A and Figure 5C). DUSP8, of which lit- NRAS have been reported to be activated shortly after TCR tle is known, was significantly upregulated at 4 hours and stimulation [65], however the regulation of their expres- then downregulated at 48–96 hours, which is opposite to sion (either at the transcriptional or protein level) has not the transcriptional pattern of MAPK14 and MAPK3. Some been reported. ADORA2B reportedly regulates ERK and transcription factors regulated by MAP kinase pathway p38 MAP kinase cascades in mast cells [66], but its func- were downregulated, and most intensely so were FOS and tion in T cells is not known. Contrary to upregulated RAS JUN. FOS and JUN proteins are the main components of genes, several upstream RAS regulating proteins the transcription factor complex AP-1. AP-1 has been (RASGRP1, -2, ADRB2, GRAP, ERBB2) showed higher reported to be quickly activated in response to T-cell acti- expression in resting T cells (Figure 5A). The downregula- vation [83], which is contradictory to the significant tion of RASGRP1 and GRAP does not correlate with their downregulation of FOS and JUN. The temporal activity of reportedly positive roles in T-cell receptor signalling AP-1 in T-cell activation is not known. Thus, we examined [67,68]. Little is known about RASGRP2, ADRB2, and the DNA-binding activity of AP-1, which rapidly increased ERBB2 in the context of T-cell activation. Contrary to the within 4 hours and then rapidly decreased (Figure 5D). It reported activation of RAC proteins CDC42 and RAC2 is possible that FOS and JUN were immediately and tran- [69,70] in T-cell activation, here we found that their transiently regulated upon anti-CD3/anti-CD8 stimulation, scription was downregulated.

Page 9 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

CD3+ CD4+ CD8+ A B Phosporylated p38 C MAPK14 (p38) 1010 o 3 3 GeneName Max/Min Median E4 04h/0h 10h/0h 48h/0h 96h/0h 06h/0h 12h/0h 24h/0h 48h/0h 72h/0h 06h/0h 12h/0h 24h/0h 48h/0h 72h/0h RASGRP2 -60.9 -25.4 -2 -4 -5 -4 -2 -5 -5 -6 -5 -2 -5 -5 -5 -5 8 8 ADRB2 -4.9 -3.2 -1 -2 -2 -2 -1 -2 -1 -1 -1 -1 -2 -2 -2 -2 E5 1 1 GRAP -3.5 -2.0 -1 -2 -1 -2 -1 -1 -1 -1 -1 -1 -2 -1 -1 -1 CD3+ ERBB2 -1.8 -1.4 -0 -1 -1 -0 -1 -0 -0 -1 -1 -0 -0 -0 -0 -0 6 6 RASGRP1 -3.4 -2.2 -1 -1 -2 -1 -0 -1 -2 -1 -2 -0 -1 -1 -1 -1 CD4+ GPS14.3 1.4 -1 0 1 -1 1 2 2 2 2 -0 1 0 1 0 4 4 0h) (vs. CD8+ Proteins HRAS 3.1 2.1 0 1 1 -0 0 1 2 2 1 1 1 1 1 1 -1-1 KRAS 2.1 1.3 1 1 0 -1 1 1 1 0 0 1 0 0 0 -1 RAS Interacting ADORA2B 3.1 1.8 1 2 1 2 1 1 0 1 0 1 1 1 1 1 2 2 NRAS 2.8 1.8 1 1 1 1 1 1 1 1 1 1 1 0 0 0 CDC42 -2.5 -2.2 -0 -1 -1 -1 -0 -1 -1 -1 -1 -1 -1 -1 -1 -1 RAC 0 Transcription expression rati expression Transcription -3 RAC2-3.3 -1.4 -1 -2 -1 -0 -2 -1 -0 -0 0 -1 -0 -1 -0 -0 Protein expression0 ratio (vs.0h) -3 -1 -1 -1 -0 -1 -1 -1 -1 -1 -1 -1 -1 -0 -0 0 24 48 72 96 09624 48 72 MAP4K2 -2.4 -1.5 0Timepoint 24487296 (hours) 0Timepoint 24487296 (hours) MAP4K1 -1.9 -1.2 -1 -1 -0 -0 0 -0 0 0 0 -0 -0 -0 -0 -0 MAP3K12 -5.4 -2.2 -2 -2 -1 -1 -2 -1 -1 -1 -1 -2 -2 -1 -1 -1 MAP2K6 -3.6 -2.1 -1 -2 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 D Phosporylated ERK1 E MAPK3 (ERK1) MAP4K4 -2.8 -1.2 -0 -1 -2 -1 -0 0 -1 -1 -0 1 -0 -0 0 -0 424 o 2 2 MAP3K2 -2.2 -1.3 1 -0 -1 -1 1 0 -0 -1 -0 1 -0 -0 -1 -1 E1 MAP4K5 2.1 1.4 0 1 0 -1 0 1 1 0 -0 1 1 0 0 -0 0 1 -0 -0 0 1 1 -0 -0 0 1 0 -0 -1 E4 CD3+ MAP3K14 1.7 1.3 318 1 1 MAP3K8 2.8 1.4 1 1 0 1 1 1 0 0 1 1 0 0 -1 0 E5 CD4+ MBIP 2.3 1.9 1 1 1 1 1 1 1 1 1 0 0 0 0 1 Kinases 1 1 1 0 2 2 1 1 1 2 1 1 1 1 STK4 5.0 2.0 rati xpression CD8+ MAPK6 2.8 2.0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 212 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 MAPK12 2.5 1.8 0h) (vs. MAPKAPK3 2.5 2.0 -0 1 1 1 0 1 1 1 1 1 1 1 1 1 -0 -0 0 1 -1 -1 -0 -0 -0 -1 -1 -1 -1 -0 MAPK14 2.3 -1.4 16 -1 MAPK3 -1.4 -1.2 -1 -0 -0 0 -0 -0 -0 -0 -0 -0 -0 -1 -0 -0 -1 MKNK1 -3.3 -1.6 -1 -1 -0 0 -1 -1 -1 -0 -0 -2 -1 -1 -1 -0 DUSP1 -114.4 -64.6 -2 -4 -5 -5 -5 -6 -6 -7 -6 -5 -6 -7 -6 -6 SHC1-2.5 -1.3 -0 -0 -1 -1 -0 -0 -0 -1 -0 0 -0 -1 -1 -1 Protein expression0 ratio (vs.0h) 0 Transcription e -2-2 PPM1A -1.6 -1.3 0 -0 -0 -0 0 -0 -1 -1 -1 0 -0 -1 -1 -0 0 24 48 72 96 0 24 48 72 96 DUSP8 -5.4 -2.5 1 0 -1 -1 -1 -1 -2 -2 -2 -1 -2 -2 -2 -2 024487296Timepoint (hours) 024487296Timepoint (hours) PPP2CB 2.1 1.4 1 1 0 -1 1 0 0 0 0 1 1 1 0 1 MGST2 2.9 1.1 -1 -1 0 2 0 -0 -0 1 1 -0 -0 -0 0 1 F G ng Proteins GSTZ1 1.8 -1.1 -0 -0 0 1 -0 -0 -0 -0 -0 -0 0 -0 0 -0 AP-1 DNA-binding activity FOS 0 0 1 1 1 2 2 2 2 1 1 1 1 1 244 4040 PTPN7 4.6 2.0 CD3+ HSPA2 1.7 -1.2 1 0 0 0 0 -0 -0 -1 -1 -0 -0 -0 -0 -0 -0 1 2 1 0 2 2 2 2 -0 2 2 2 2 E4 CD4+ GSTP1 5.1 3.1 0 0 HSPA5 3.1 -1.2 0 1 2 0 -1 -0 0 -0 -1 Na-0 -0 -0 -0 183 CD8+ HSPA8 4.5 3.4 0 1 2 0 1 2 1 2 2 Na 2 2 2 2 E5 MAPK Regulati HSPA9B2.5 2.1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 -40-40 DUSP5 4.4 2.1 2 2 1 1 1 1 1 -1 -0 2 1 1 0 0 122 JUNB -8.5 -5.2 -0 -2 -2 -0 -2 -3 -3 -3 -2 -2 -2 -3 -3 -2 FOS -149.3 -63.8 -1 -4 -5 -5 -5 -6 -7 -7 -7 -4 -6 -7 -6 -6 -80-80

JUN -159.5 -47.4 -2 -3 -5 -4 -4 -6 -7 -7 -7 -4 -6 -6 -7 -6 (vs. 0h) FOSB -14.7 -11.8 0 -1 -2 -2 -3 -4 -4 -4 -4 -2 -4 -4 -4 -4 61 JUND -3.4 -2.2 0 -1 -0 -1 -1 -1 -1 -1 -1 -1 -2 -2 -2 -1 -120-120 MAX -2.6 -1.2 0 -0 -1 -1 -0 0 -0 -1 -1 0 0 -0 -0 -1 EGR3 25.3 5.7 4 3 2 3 4 2 2 1 2 5 3 2 2 3 Protein expression0 ratio (vs.0h) 0 ATF3 8.9 2.2 3 3 2 3 1 2 1 0 1 1 1 1 -0 1 -160-160 0 24 48 72 96 Transcription expression ratio 0 24 48 72 96 NR4A1 23.6 1.6 5 2 1 2 2 -1 -1 -1 -1 3 1 1 -0 1 3 2 1 -0 3 2 2 2 1 4 3 3 2 2 0Timepoint 24487296 (hours) 0 24487296 MYC 11.8 5.2 JUN NFATC1 8.6 2.7 2 1 2 1 3 1 2 2 2 3 1 1 1 1 H40 NFKB1 3.5 1.2 2 1 0 0 1 1 0 -0 -0 1 0 0 -0 -0 40 CD3+ 1 1 0 0 0 0 0 0 1 0 -0 -0 0 0 ATF4 1.9 1.2 CD4+ ETS2 3.0 2.2 1 1 0 0 1 2 1 1 1 2 1 1 0 1 0 0 EGR1 -14.3 -2.5 3 1 -1 1 0 -2 -3 -4 -3 0 -1 -2 -3 -3 CD8+ Regulated TranscriptionRegulated Factors CREBBP -1.9 -1.5 1 0 -1 0 0 -1 -1 -1 -1 -0 -1 -1 -1 -1 CDKN2D -3.0 -1.3 -0 -2 -1 0 -0 -1 -1 -1 0 -0 -1 -1 -0 1 -40-40 CCNB1 9.9 1.5 0 0 3 3 0 0 1 3 3 -1 0 0 2 3 CCNB2 9.5 1.2 -2 -2 3 3 -1 -1 0 3 3 -0 -0 0 2 3 2 3 3 3 3 3 4 4 4 3 4 4 4 4 -80-80

CCND2 15.3 10.5 (vs. 0h) CDKN1A 5.7 1.4 2 2 1 2 2 0 -0 0 1 0 -0 -0 -0 1 CDK4 9.1 5.2 1 2 2 1 2 3 3 3 3 2 2 2 3 2 -120-120 CCNA2 4.9 1.2 -0 -0 2 2 -0 0 0 2 2 -0 -0 0 2 2 CCNE1 2.7 1.5 0 1 1 1 -0 0 0 1 1 -0 0 1 1 1 -3 0 3 E2F1 4.4 2.2 -0 -0 2 1 -0 0 1 2 1 -0 1 2 2 1 -160-160 Regulated byRegulated ERK Transcription expression ratio Cell CycleCell Proteins RB1 2.8 1.6 1 0 1 1 0 0 0 1 1 0 1 1 1 1 0 24 48 72 96 CDK2 2.3 1.2 0 -0 1 1 -0 -0 0 1 1 -1 -0 0 1 1 024487296Timepoint (hours)

SignificantlyFigure 5 regulated genes involved in MAP kinase signalling Significantly regulated genes involved in MAP kinase signalling. (A) Expression profiles of genes that belong to the list of MAP kinase pathways. Color denotes degree of differential expression compared to 0 hour (saturated red = 3-fold up-regulation, saturated green = 3-fold down-regulation, black = unchanged, gray = no data available). Expression data shown are averages from three independent biological experiments for each T-cell population. The median ratio, along with the maximum (for up-regulated genes) or minimum (for down-regulated genes) ratio of stimulated T cells at each timepoint (with respect to the expression of 0 hour) is provided (a negative value represents down-regulation). (B) Protein expression profiles of phosphorylated p38 and of (D) phosphorylated ERK1 agree with the late transcription upregulation of MAPK8 and MAPK3. CD3+ T cells were selected, stimulated (with anti-CD3/anti-CD28 antibodies), cultured and harvested at the indicated timepoints of culture to analyse the protein expression by flow cytometric assays. Data from two independent experiments, E4 and E5, are shown. (F) DNA-binding activity profile of transcription factor AP-1 captured the immediate and transient activation of AP-1 in T-cell activation. CD3+ T cells were selected, stimulated (with anti-CD3/anti-CD28 antibodies), cultured and harvested at the indicated timepoints of culture. Data from three independent experiments, E1, E4 and E5, are shown. (C), (E), (G) and (H) demonstrated the transcriptional patterns of MAPK14 (p38), MAPK3 (ERK1), FOS and JUN (the two major components if AP-1) in all three populations (CD3+, CD4+ and CD8+).

Page 10 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

and that their upregulation was not captured by our first regulated by other proteins, such as HSPE1 and HSPD1 timepoint following T-cell activation. (Figure 1A), rather than CASP9. Surprisingly, the well- known AICD receptor, TNFRSF6 (FAS), was upregulated Discussion at 4–10 hours, suggesting the involvement of FAS imme- In this work, we sought to improve our understanding of diately upon T-cell activation. regulation of apoptosis in T-cell activation. We approached this problem by analysing global, temporal Our data show that numerous significantly regulated microarray data from ex vivo CD3+, CD4+ and CD8+ T- genes associated with apoptosis are involved in NF-κB sig- cell cultures, and leveraging Gene Ontology associations nalling pathway and MAP kinase signalling pathway. Sup- and prior knowledge. This approach extended our knowl- ported by the increase of phosphorylated p65 at 48–96 edge base in three ways: we presented detailed kinetic hours (Figure 3B), the simultaneous upregulation of NF- gene expression information on genes with previously κB family genes (REL, RELA, and RELB) and IκB family hypothesized or presumed important roles in apoptosis; genes (NFKBIA, NFKBIE and NFKB1) at 48–96 hours (Fig- we identified a new set of genes not previously associated ure 3A) suggests that multiple versions of the NF-κB dim- with T-cell activation; and, we integrated and connected mer complex are active during this time period in our the previous knowledge with temporal transcription pro- experiments. Lack of early detection of phosphorylated files to build a more comprehensive picture of regulation p65 that multiple versions of the NF-κB dimmer complex of apoptosis in T-cell activation. Gene expression data are active during this time period in our experiments were further explored by examining protein expression could be the result of the strong upregulation of NFKBIA (active form/phosphorylated form) and functional activ- at 4–10 hours. It is also possible that p65 might immedi- ity levels as a first assessment of their functional role. ately and transiently become activated, and then quickly deactivated within 4 hours, our first timepoint. Of note, Genome-scale transcription profiling provides the oppor- IKIP (IκB kinase interacting protein) was significantly tunity to more holistically evaluate the regulation of a upregulated at 48–96 hours (Figure 3A). To date, the func- group of genes, of the same family, or with analogous tion of IKIP remains unknown. Its transcriptional kinetics functions, or associated in specific pathways. Composed suggests that it might have a positive role in NF-κB activity of several members, the BCL2 family proteins are key regulation. players in regulation of apoptosis. BCL2, BAX, BAK have been reported to play important roles in apoptosis post Validated by the increase of phosphorylated p38 and activation in T cells [4]. Knockout of pro-apoptotic BBC3 ERK1 at 24–96 hours and 48–96 hours, respectively, as and PMAIP1 decreased DNA damage-induced apoptosis measured by flow cytometry assays (Figure 5B and Figure in mice fibroblasts, but only loss of BBC3 protected lym- 5C), the similar transcriptional patterns of MAPK14 phocytes from cell death [84]. However, the function and (p38), MAPK3 (ERK1), MKNK1 (the interacting protein transcriptional regulation of most of the BCL2 family of both MAPK14 and MAPK3 [77]) and MAP2K6 (p38 members and their regulatory proteins remain unex- specific MAP kinase kinase [78]) (downregulated at 4–10 plored in T-cell activation. Our data suggest the distinct hours and then upregulated at 48–96 hours) suggest that stages that BCL2 family members are involved in the proc- cascades of p38 and ERK1, but not JNK, are synergistically ess of T-cell activation, thus providing directions for future activated during late T-cell activation and early prolifera- studies. For instance, validated by protein abundance tion. Activity of transcription factor AP-1, mostly regu- assays (Figure 2A and 2B), the functions of continuously lated by JNK and p38 cascades [86], increased upregulated BCL2A1 and dynamically regulated BBC3 immediately, but only transiently, upon anti-CD3/anti- deserve further studies in T-cell activation. The synchro- CD28 stimulation, suggesting a potentially rapid but tran- nized early transcriptional upregulation of MCL1 and its sient activation of JNK and p38 cascades in T-cell activa- inhibitory interacting proteins PMAIP1 [24] and BMF [85] tion. Significantly, there was no activity increase of AP-1 at imply their involvement in T-cell activation quickly upon 48–96 hours, in contrast to the increase of phosphor- TCR ligation. ylated p65. It has been suggested that MAP kinases have different roles in CD4+ and CD8+ T cells [87]. However In typical apoptosis signalling, CASP9 is activated first, most of the significant regulated genes in MAP kinase sig- which leads to activation of downstream effectors CASP3, nalling pathway shared similar expression patterns and CASP6. Here, CASP3 was upregulated (at both the between the CD4+ and CD8+ subsets. It is possible that transcriptional and protein levels) by 48 hours, compared the MAP kinase signalling pathway is regulated similarly to the upregulation of CASP9 only at 96 hours. These find- between the two subsets in the context of T-cell activation. ings suggest that active CASP3 protein might have a differ- It is also possible that the MAP kinases might function dif- ent role in T-cell activation and thus its activity might be ferently due to regulation at protein level.

Page 11 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

T-cell activation using bead-conjugated-anti-CD3/anti- CD28 antibodies is an established state-of-the art method Additional file 2 [88] for both fundamental studies, such as gene-expres- Flow cytometry histograms, at 0 hour and 96 hours in CD3+ popula- sion profiling [6,89], as well as for clinical immuno- tion. Representative (CD3+ experiment, E4) flow cytometry histograms, at 0 hour and 96 hours, of BCL2A1, BBC3, active CASP3, phosphor- therapy applications [90]. It provides both the stimulatory ylated p65, phosphorylated p38 and phosphorylated ERK1. and costimulatory signals [91,92] and the physical con- Click here for file tact, in place of APCs [93], and as such it is assumed to [http://www.biomedcentral.com/content/supplementary/1755- mimic the in vivo situation. Consistent with that assump- 8794-1-53-S2.pdf] tion, and for well-established genes partaking in the T-cell activation process, our data are consistent with the large body of literature (a good fraction of which comes from in vivo studies) on T-cell activation. Acknowledgements This work was supported by National Institutes of Health grant (NIH R01- Conclusion GM065476). We thank Dr. Carlos Paredes and Dr. Peter Fuhrken for development of microarray and Q-RT-PCR analysis software. We acknowl- Our study captured novel temporal patterns of previously edge the use of instruments in the Keck Biophysics Facility, and the Center known but many novel, in the context of T-cell activation, for Genetic Medicine at Northwestern University. genes ontologically classified under the term 'regulation of apoptosis'. These patterns were reproducibly and References robustly identified as donor independent. Comprehen- 1. Rothstein DM, Sayegh MH: T-cell costimulatory pathways in sively integrating previous knowledge, we identified novel allograft rejection and tolerance. Immunol Rev 2003, 196:85-108. significant genes associated with regulation of apoptosis 2. Huang Y, Wange RL: T cell receptor signaling: beyond complex in T cells, as well as with NF-κB and MAPK signalling complexes. J Biol Chem 2004, 279(28):28827-28830. 3. Gimmi CD, Freeman GJ, Gribben JG, Gray G, Nadler LM: Human T- pathways. This study improves our understanding of the cell clonal anergy is induced by antigen presentation in the biology and the underlying regulation mechanisms of absence of B7 costimulation. Proc Natl Acad Sci USA 1993, apoptosis in T-cell activation in the natural CD3+ popula- 90(14):6586-6590. 4. Krammer PH, Arnold R, Lavrik IN: Life and death in peripheral T tion, as well as in the CD4+ and CD8+ subsets. cells. Nat Rev Immunol 2007, 7(7):532-542. 5. Diehn M, Alizadeh AA, Rando OJ, Liu CL, Stankunas K, Botstein D, Crabtree GR, Brown PO: Genomic expression programs and Competing interests the integration of the CD28 costimulatory signal in T cell The authors declare that they have no competing interests. activation. Proc Natl Acad Sci USA 2002, 99(18):11796-11801. 6. Riley JL, Mao M, Kobayashi S, Biery M, Burchard J, Cavet G, Gregson BP, June CH, Linsley PS: Modulation of TCR-induced transcrip- Authors' contributions tional profiles by ligation of CD28, ICOS, and CTLA-4 recep- MW and DW carried out the culture experiments and per- tors. Proc Natl Acad Sci USA 2002, 99(18):11790-11795. formed the microarray experiments. MW analysed the 7. Wang M, Windgassen D, Papoutsakis ET: Comparative analysis of transcriptional profiling of CD3+, CD4+ and CD8+ T cells microarray data and wrote the manuscript with assistance identifies novel immune response players in T-Cell activa- from EP. MW and EP conceived, designed, and coordi- tion. BMC Genomics 2008, 9(1):225. 8. Koester SK, Bolton WE: Intracellular markers. J Immunol Methods nated the study. 2000, 243(1–2):99-106. 9. Ramsborg CG, Papoutsakis ET: Global transcriptional analysis Additional material delineates the differential inflammatory response interleukin-15 elicits from cultured human T cells. Exp Hematol 2007, 35(3):454-464. 10. Ramsborg CG, Windgassen D, Fallon JK, Paredes CJ, Papoutsakis ET: Additional file 1 Molecular insights into the pleiotropic effects of plasma on Expression profile of NF-κB target genes in T-cell activation of the ex vivo-expanded T cells using DNA-microarray analysis. three (CD3+, CD4+ and CD8+) populations. Hierarchical clustering Exp Hematol 2004, 32(10):970-990. 11. Saeed AI, Sharov V, White J, Li J, Liang W, Bhagabati N, Braisted J, (Euclidean distance metric) shows the temporal expression. Color denotes Klapa M, Currier T, Thiagarajan M, et al.: TM4: a free, open-source degree of differential expression compared to 0 hour (saturated red = 3- system for microarray data management and analysis. Bio- fold up-regulation, saturated green = 3-fold down-regulation, black = techniques 2003, 34(2):374-378. unchanged, gray = no data available). Expression data shown are averages 12. The Gene Expression Omnibus [http://www.ncbi.nlm.nih.gov/ from three independent biological experiments for each T-cell population. geo/] The median ratio, along with the maximum (for up-regulated genes) or 13. The Gene Ontology Consortium Website [http://www.gene ontology.org/] minimum (for down-regulated genes) ratio of stimulated T cells at each 14. Hosack DA, Dennis G Jr, Sherman BT, Lane HC, Lempicki RA: Iden- timepoint (with respect to the expression of 0 hour) is provided (a negative tifying biological themes within lists of genes with EASE. value represents down-regulation). Genome Biol 2003, 4(10):R70. Click here for file 15. superarray bioscience [http://www.superarray.com/] [http://www.biomedcentral.com/content/supplementary/1755- 16. The Rel/NF-kappaB Signal Transduction Pathway [http:// 8794-1-53-S1.pdf] people.bu.edu/gilmore/nf-kb/] 17. Kyoto Encyclopedia of Genes and Genomes [http:// www.genome.jp/kegg/]

Page 12 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

18. The National Center for Biotechnology Information Web- 39. Michallet MC, Saltel F, Preville X, Flacher M, Revillard JP, Genestier L: site [http://www.ncbi.nlm.nih.gov/] Cathepsin-B-dependent apoptosis triggered by antithymo- 19. Fuhrken PG, Chen C, Miller WM, Papoutsakis ET: Comparative, cyte globulins: a novel mechanism of T-cell depletion. Blood genome-scale transcriptional analysis of CHRF-288-11 and 2003, 102(10):3719-3726. primary human megakaryocytic cell cultures provides novel 40. Kischkel FC, Hellbardt S, Behrmann I, Germer M, Pawlita M, Kram- insights into lineage-specific differentiation. Exp Hematol 2007, mer PH, Peter ME: Cytotoxicity-dependent APO-1 (Fas/ 35(3):476-489. CD95)-associated proteins form a death-inducing signaling 20. Jagelska E, Brazda V, Pospisilova S, Vojtesek B, Palecek E: New complex (DISC) with the receptor. Embo J 1995, ELISA technique for analysis of p53 protein/DNA binding 14(22):5579-5588. properties. J Immunol Methods 2002, 267(2):227-235. 41. Migone TS, Zhang J, Luo X, Zhuang L, Chen C, Hu B, Hong JS, Perry 21. Nareika A, He L, Game BA, Slate EH, Sanders JJ, London SD, Lopes- JW, Chen SF, Zhou JX, et al.: TL1A is a TNF-like ligand for DR3 Virella MF, Huang Y: Sodium lactate increases LPS-stimulated and TR6/DcR3 and functions as a T cell costimulator. Immu- MMP and cytokine expression in U937 histiocytes by enhanc- nity 2002, 16(3):479-492. ing AP-1 and NF-kappaB transcriptional activities. Am J Physiol 42. Ware CF, Crowe PD, Vanarsdale TL, Andrews JL, Grayson MH, Jerzy Endocrinol Metab 2005, 289(4):E534-542. R, Smith CA, Goodwin RG: Tumor necrosis factor (TNF) recep- 22. Rea IM, McNerlan SE, Alexander HD: CD69, CD25, and HLA-DR tor expression in T lymphocytes. Differential regulation of activation antigen expression on CD3+ lymphocytes and the type I TNF receptor during activation of resting and relationship to serum TNF-alpha, IFN-gamma, and sIL-2R effector T cells. J Immunol 1991, 147(12):4229-4238. levels in aging. Exp Gerontol 1999, 34(1):79-93. 43. Kim EY, Teh HS: Critical role of TNF receptor type-2 (p75) as 23. Cory S, Adams JM: The Bcl2 family: regulators of the cellular a costimulator for IL-2 induction and T cell survival: a func- life-or-death switch. Nat Rev Cancer 2002, 2(9):647-656. tional link to CD28. J Immunol 2004, 173(7):4500-4509. 24. Alves NL, Derks IA, Berk E, Spijker R, van Lier RA, Eldering E: The 44. Pimentel-Muinos FX, Seed B: Regulated commitment of TNF Noxa/Mcl-1 axis regulates susceptibility to apoptosis under receptor signaling: a molecular switch for death or activa- glucose limitation in dividing T cells. Immunity 2006, tion. Immunity 1999, 11(6):783-793. 24(6):703-716. 45. Li X, Yang Y, Ashwell JD: TNF-RII and c-IAP1 mediate ubiquiti- 25. Kim H, Kim YN, Kim H, Kim CW: Oxidative stress attenuates nation and degradation of TRAF2. Nature 2002, Fas-mediated apoptosis in Jurkat T cell line through Bfl-1 416(6878):345-347. induction. Oncogene 2005, 24(7):1252-1261. 46. Nishikori M, Ohno H, Haga H, Uchiyama T: Stimulation of CD30 26. Zhong F, Davis MC, McColl KS, Distelhorst CW: Bcl-2 differen- in anaplastic large cell lymphoma leads to production of tially regulates Ca2+ signals according to the strength of T nuclear factor-kappaB p52, which is associated with hyper- cell receptor activation. J Cell Biol 2006, 172(1):127-137. phosphorylated Bcl-3. Cancer Sci 2005, 96(8):487-497. 27. Imai Y, Kimura T, Murakami A, Yajima N, Sakamaki K, Yonehara S: 47. Ma BY, Mikolajczak SA, Danesh A, Hosiawa KA, Cameron CM, Taka- The CED-4-homologous protein FLASH is involved in Fas- ori-Kondo A, Uchiyama T, Kelvin DJ, Ochi A: The expression and mediated activation of caspase-8 during apoptosis. Nature the regulatory role of OX40 and 4-1BB heterodimer in acti- 1999, 398(6730):777-785. vated human T cells. Blood 2005, 106(6):2002-2010. 28. Thalappilly S, Sadasivam S, Radha V, Swarup G: Involvement of cas- 48. Trauzold A, Wermann H, Arlt A, Schutze S, Schafer H, Oestern S, pase 1 and its activator Ipaf upstream of mitochondrial Roder C, Ungefroren H, Lampe E, Heinrich M, et al.: CD95 and events in apoptosis. Febs J 2006, 273(12):2766-2778. TRAIL receptor-mediated activation of protein kinase C and 29. Chau BN, Cheng EH, Kerr DA, Hardwick JM: Aven, a novel inhib- NF-kappaB contributes to apoptosis resistance in ductal itor of caspase activation, binds Bcl-xL and Apaf-1. Mol Cell pancreatic adenocarcinoma cells. Oncogene 2001, 2000, 6(1):31-40. 20(31):4258-4269. 30. Kirchhoff S, Muller WW, Krueger A, Schmitz I, Krammer PH: TCR- 49. Ghosh S, May MJ, Kopp EB: NF-kappa B and Rel proteins: evolu- mediated up-regulation of c-FLIPshort correlates with tionarily conserved mediators of immune responses. Annu resistance toward CD95-mediated apoptosis by blocking Rev Immunol 1998, 16:225-260. death-inducing signaling complex activity. J Immunol 2000, 50. Brockman JA, Scherer DC, McKinsey TA, Hall SM, Qi X, Lee WY, Bal- 165(11):6293-6300. lard DW: Coupling of a signal response domain in I kappa B 31. Oehme I, Neumann F, Bosser S, Zornig M: Transgenic overex- alpha to multiple pathways for NF-kappa B activation. Mol pression of the Caspase-8 inhibitor FLIP(short) leads to Cell Biol 1995, 15(5):2809-2818. impaired T cell proliferation and an increased memory T cell 51. Dai R, Phillips RA, Ahmed SA: Despite inhibition of nuclear local- pool after staphylococcal enterotoxin B injection. Eur J Immu- ization of NF-kappa B p65, c-Rel, and RelB, 17-beta estradiol nol 2005, 35(4):1240-1249. up-regulates NF-kappa B signaling in mouse splenocytes: the 32. Samali A, Cai J, Zhivotovsky B, Jones DP, Orrenius S: Presence of a potential role of Bcl-3. J Immunol 2007, 179(3):1776-1783. pre-apoptotic complex of pro-caspase-3, Hsp60 and Hsp10 in 52. Shimizu T, Kida Y, Kuwano K: A dipalmitoylated lipoprotein the mitochondrial fraction of jurkat cells. Embo J 1999, from Mycoplasma pneumoniae activates NF-kappa B 18(8):2040-2048. through TLR1, TLR2, and TLR6. J Immunol 2005, 33. Bruneel A, Labas V, Mailloux A, Sharma S, Royer N, Vinh J, Pernet P, 175(7):4641-4646. Vaubourdolle M, Baudin B: Proteomics of human umbilical vein 53. Lee SY, Lee SY, Choi Y: TRAF-interacting protein (TRIP): a endothelial cells applied to etoposide-induced apoptosis. Pro- novel component of the tumor necrosis factor receptor teomics 2005, 5(15):3876-3884. (TNFR)- and CD30-TRAF signaling complexes that inhibits 34. Kaul SC, Taira K, Pereira-Smith OM, Wadhwa R: Mortalin: present TRAF2-mediated NF-kappaB activation. J Exp Med 1997, and prospective. Exp Gerontol 2002, 37(10–11):1157-1164. 185(7):1275-1285. 35. Reddy RK, Mao C, Baumeister P, Austin RC, Kaufman RJ, Lee AS: 54. Hatzoglou A, Roussel J, Bourgeade MF, Rogier E, Madry C, Inoue J, Endoplasmic reticulum chaperone protein GRP78 protects Devergne O, Tsapis A: TNF receptor family member BCMA (B cells from apoptosis induced by topoisomerase inhibitors: cell maturation) associates with TNF receptor-associated role of ATP binding site in suppression of caspase-7 activa- factor (TRAF) 1, TRAF2, and TRAF3 and activates NF-kappa tion. J Biol Chem 2003, 278(23):20915-20924. B, elk-1, c-Jun N-terminal kinase, and p38 mitogen-activated 36. Suzuki CK, Bonifacino JS, Lin AY, Davis MM, Klausner RD: Regulat- protein kinase. J Immunol 2000, 165(3):1322-1330. ing the retention of T-cell receptor alpha chain variants 55. Tsao DH, McDonagh T, Telliez JB, Hsu S, Malakian K, Xu GY, Lin LL: within the endoplasmic reticulum: Ca(2+)-dependent associ- Solution structure of N-TRADD and characterization of the ation with BiP. J Cell Biol 1991, 114(2):189-205. interaction of N-TRADD and C-TRAF2, a key step in the 37. Hunter AM, LaCasse EC, Korneluk RG: The inhibitors of apopto- TNFR1 signaling pathway. Mol Cell 2000, 5(6):1051-1057. sis (IAPs) as cancer targets. Apoptosis 2007, 12(9):1543-1568. 56. Mizushima S, Fujita M, Ishida T, Azuma S, Kato K, Hirai M, Otsuka M, 38. Devadas S, Das J, Liu C, Zhang L, Roberts AI, Pan Z, Moore PA, Das Yamamoto T, Inoue J: Cloning and characterization of a cDNA G, Shi Y: Granzyme B is critical for T cell receptor-induced encoding the human homolog of tumor necrosis factor cell death of type 2 helper T cells. Immunity 2006, receptor-associated factor 5 (TRAF5). Gene 1998, 25(2):237-247. 207(2):135-140.

Page 13 of 14 (page number not for citation purposes) BMC Medical Genomics 2008, 1:53 http://www.biomedcentral.com/1755-8794/1/53

57. Coornaert B, Baens M, Heyninck K, Bekaert T, Haegman M, Staal J, 77. Fukunaga R, Hunter T: MNK1, a new MAP kinase-activated pro- Sun L, Chen ZJ, Marynen P, Beyaert R: T cell antigen receptor tein kinase, isolated by a novel expression screening method stimulation induces MALT1 paracaspase-mediated cleavage for identifying protein kinase substrates. Embo J 1997, of the NF-kappaB inhibitor A20. Nat Immunol 2008. 16(8):1921-1933. 58. Sun L, Deng L, Ea CK, Xia ZP, Chen ZJ: The TRAF6 ubiquitin 78. Stein B, Brady H, Yang MX, Young DB, Barbosa MS: Cloning and ligase and TAK1 kinase mediate IKK activation by BCL10 characterization of MEK6, a novel member of the mitogen- and MALT1 in T lymphocytes. Mol Cell 2004, 14(3):289-301. activated protein kinase kinase cascade. J Biol Chem 1996, 59. Wegener E, Krappmann D: CARD-Bcl10-Malt1 signalosomes: 271(19):11427-11433. missing link to NF-kappaB. Sci STKE 2007, 2007(384):pe21. 79. Liao G, Zhang M, Harhaj EW, Sun SC: Regulation of the NF-kap- 60. Matsuda A, Suzuki Y, Honda G, Muramatsu S, Matsuzaki O, Nagano paB-inducing kinase by tumor necrosis factor receptor-asso- Y, Doi T, Shimotohno K, Harada T, Nishida E, et al.: Large-scale ciated factor 3-induced degradation. J Biol Chem 2004, identification and characterization of human genes that acti- 279(25):26243-26250. vate NF-kappaB and MAPK signaling pathways. Oncogene 80. Belich MP, Salmeron A, Johnston LH, Ley SC: TPL-2 kinase regu- 2003, 22(21):3307-3318. lates the proteolysis of the NF-kappaB-inhibitory protein 61. Chapman NR, Perkins ND: Inhibition of the RelA(p65) NF-kap- NF-kappaB1 p105. Nature 1999, 397(6717):363-368. paB subunit by Egr-1. J Biol Chem 2000, 275(7):4719-4725. 81. Slack DN, Seternes OM, Gabrielsen M, Keyse SM: Distinct binding 62. Wada T, Penninger JM: Mitogen-activated protein kinases in determinants for ERK2/p38alpha and JNK map kinases medi- apoptosis regulation. Oncogene 2004, 23(16):2838-2849. ate catalytic activation and substrate selectivity of map 63. Matsuda S, Miwa Y, Hirata Y, Minowa A, Tanaka J, Nishida E, Koyasu kinase phosphatase-1. J Biol Chem 2001, 276(19):16491-16500. S: Negative feedback loop in T-cell activation through 82. Kwak SP, Dixon JE: Multiple dual specificity protein tyrosine MAPK-catalyzed threonine phosphorylation of LAT. Embo J phosphatases are expressed and regulated differentially in 2004, 23(13):2577-2585. liver cell lines. J Biol Chem 1995, 270(3):1156-1160. 64. Roose JP, Mollenauer M, Gupta VA, Stone J, Weiss A: A diacylglyc- 83. Bandres JC, Ratner L: Human immunodeficiency virus type 1 erol-protein kinase C-RasGRP1 pathway directs Ras activa- Nef protein down-regulates transcription factors NF-kappa tion upon antigen receptor stimulation of T cells. Mol Cell Biol B and AP-1 in human T cells in vitro after T-cell receptor 2005, 25(11):4426-4441. stimulation. J Virol 1994, 68(5):3243-3249. 65. Perez de Castro I, Bivona TG, Philips MR, Pellicer A: Ras activation 84. Villunger A, Michalak EM, Coultas L, Mullauer F, Bock G, Ausserlech- in Jurkat T cells following low-grade stimulation of the T-cell ner MJ, Adams JM, Strasser A: p53- and drug-induced apoptotic receptor is specific to N-Ras and occurs only on the Golgi responses mediated by BH3-only proteins puma and noxa. apparatus. Mol Cell Biol 2004, 24(8):3485-3496. Science 2003, 302(5647):1036-1038. 66. Feoktistov I, Goldstein AE, Biaggioni I: Role of p38 mitogen-acti- 85. Kuwana T, Bouchier-Hayes L, Chipuk JE, Bonzon C, Sullivan BA, vated protein kinase and extracellular signal-regulated pro- Green DR, Newmeyer DD: BH3 domains of BH3-only proteins tein kinase kinase in adenosine A2B receptor-mediated differentially regulate Bax-mediated mitochondrial mem- interleukin-8 production in human mast cells. Mol Pharmacol brane permeabilization both directly and indirectly. Mol Cell 1999, 55(4):726-734. 2005, 17(4):525-535. 67. Ebinu JO, Stang SL, Teixeira C, Bottorff DA, Hooton J, Blumberg PM, 86. Chang L, Karin M: Mammalian MAP kinase signalling cascades. Barry M, Bleakley RC, Ostergaard HL, Stone JC: RasGRP links T- Nature 2001, 410(6824):37-40. cell receptor signaling to Ras. Blood 2000, 95(10):3199-3203. 87. Zhang YL, Dong C: MAP kinases in immune responses. Cell Mol 68. Trub T, Frantz JD, Miyazaki M, Band H, Shoelson SE: The role of a Immunol 2005, 2(1):20-27. lymphoid-restricted, Grb2-like SH3-SH2-SH3 protein in T 88. Kalamasz D, Long SA, Taniguchi R, Buckner JH, Berenson RJ, Bony- cell receptor signaling. J Biol Chem 1997, 272(2):894-902. hadi M: Optimization of human T-cell expansion ex vivo using 69. Nishihara H, Maeda M, Tsuda M, Makino Y, Sawa H, Nagashima K, magnetic beads conjugated with anti-CD3 and Anti-CD28 Tanaka S: DOCK2 mediates T cell receptor-induced activa- antibodies. J Immunother 2004, 27(5):405-418. tion of Rac2 and IL-2 transcription. Biochem Biophys Res Commun 89. Raghavan A, Ogilvie RL, Reilly C, Abelson ML, Raghavan S, Vasdewani 2002, 296(3):716-720. J, Krathwohl M, Bohjanen PR: Genome-wide analysis of mRNA 70. Salazar-Fontana LI, Barr V, Samelson LE, Bierer BE: CD28 engage- decay in resting and activated primary human T lym- ment promotes actin polymerization through the activation phocytes. Nucleic Acids Res 2002, 30(24):5529-5538. of the small Rho GTPase Cdc42 in human T cells. J Immunol 90. Onlamoon N, Hudson K, Bryan P, Mayne AE, Bonyhadi M, Berenson 2003, 171(5):2225-2232. R, Sundstrom BJ, Bostik P, Ansari AA, Villinger F: Optimization of 71. Pombo CM, Kehrl JH, Sanchez I, Katz P, Avruch J, Zon LI, Woodgett in vitro expansion of macaque CD4 T cells using anti-CD3 JR, Force T, Kyriakis JM: Activation of the SAPK pathway by the and co-stimulation for autotransfusion therapy. J Med Primatol human STE20 homologue germinal centre kinase. Nature 2006, 35:4-5. 1995, 377(6551):750-754. 91. Lenschow DJ, Walunas TL, Bluestone JA: CD28/B7 system of T 72. Hu MC, Qiu WR, Wang X, Meyer CF, Tan TH: Human HPK1, a cell costimulation. Annu Rev Immunol 1996, 14:233-258. novel human hematopoietic progenitor kinase that activates 92. Sloan-Lancaster J, Allen PM: Altered peptide ligand-induced par- the JNK/SAPK kinase cascade. Genes Dev 1996, tial T cell activation: molecular mechanisms and role in T 10(18):2251-2264. cell biology. Annu Rev Immunol 1996, 14:1-27. 73. Sakuma H, Ikeda A, Oka S, Kozutsumi Y, Zanetta JP, Kawasaki T: 93. Steinman RM: The dendritic cell system and its role in immu- Molecular cloning and functional expression of a cDNA nogenicity. Annu Rev Immunol 1991, 9:271-296. encoding a new member of mixed lineage protein kinase from human brain. J Biol Chem 1997, 272(45):28622-28629. 74. Machida N, Umikawa M, Takei K, Sakima N, Myagmar BE, Taira K, Pre-publication history Uezato H, Ogawa Y, Kariya K: Mitogen-activated protein kinase The pre-publication history for this paper can be accessed kinase kinase kinase 4 as a putative effector of Rap2 to acti- here: vate the c-Jun N-terminal kinase. J Biol Chem 2004, 279(16):15711-15714. 75. Blank JL, Gerwins P, Elliott EM, Sather S, Johnson GL: Molecular http://www.biomedcentral.com/1755-8794/1/53/prepub cloning of mitogen-activated protein/ERK kinase kinases (MEKK) 2 and 3. Regulation of sequential phosphorylation pathways involving mitogen-activated protein kinase and c- Jun kinase. J Biol Chem 1996, 271(10):5361-5368. 76. Fukuyama K, Yoshida M, Yamashita A, Deyama T, Baba M, Suzuki A, Mohri H, Ikezawa Z, Nakajima H, Hirai S, et al.: MAPK upstream kinase (MUK)-binding inhibitory protein, a negative regulator of MUK/dual leucine zipper-bearing kinase/leucine zipper protein kinase. J Biol Chem 2000, 275(28):21247-21254.

Page 14 of 14 (page number not for citation purposes)