SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Transcript profiling—microarray studies
  5. Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets
  6. Proteomics provides complementary view
  7. Concluding remarks
  8. Acknowledgment
  9. References

Current knowledge of helper T cell differentiation largely relies on data generated from mouse studies. To develop therapeutical strategies combating human diseases, understanding the molecular mechanisms how human naïve T cells differentiate to functionally distinct T helper (Th) subsets as well as studies on human differentiated Th cell subsets is particularly valuable. Systems biology approaches provide a holistic view of the processes of T helper differentiation, enable discovery of new factors and pathways involved and generation of new hypotheses to be tested to improve our understanding of human Th cell differentiation and immune-mediated diseases. Here, we summarize studies where high-throughput systems biology approaches have been exploited to human primary T cells. These studies reveal new factors and signalling pathways influencing T cell differentiation towards distinct subsets, important for immune regulation. Such information provides new insights into T cell biology and into targeting immune system for therapeutic interventions.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Transcript profiling—microarray studies
  5. Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets
  6. Proteomics provides complementary view
  7. Concluding remarks
  8. Acknowledgment
  9. References

Distinct subsets of CD4+ T cells that differ in their cytokine production patterns and effector functions were first discovered in mouse for 27 years ago, rapidly followed by similar findings in human [1]. Studies from both mouse and human indicated that Th1 cells produce interferon gamma (IFN-γ), a cytokine important for host defence against intracellular pathogens, whereas another subset of helper T cells, designated Th2 cells, secrete interleukin (IL) 4, a cytokine essential for activating B cells as well as for immune response against helminth infection. These two major subsets of helper T cells, Th1 and Th2 cells, differentiated from naive CD4 T cells, dominated the field for close to two decades. However, with time, it has become clear that the Th1/Th2 dichotomy model reflects only a part of the complexity of T cell responses. Since 2003, several studies suggested the existence of other Th subsets, namely Th17, naturally occurring regulatory T (nTreg) cells and induced Tregs (iTregs) [2]. However, it has been controversial whether human CD4+ T cells, when cultured under forkhead box P3 (FOXP3)-inducing conditions leading to development of mouse iTreg cells, demonstrate similar suppressive activity [3].

After the CD4+ cells have been developed in the thymus, they migrate to the peripheral lymphoid organs where the different cytokine-producing CD4+ T lymphocyte subsets are derived from the same naive CD4+ T cell precursors. Differentiation of naïve CD4+ T lymphocytes into functionally distinct effector cell subsets is influenced by a number of factors. In simplistic terms, the presentation of priming antigen induces activation of T cells and the cytokines present in the priming microenvironment determines to which subset the cells will polarize. As a basic model cytokines, IL-12 and IL-4 are the key cytokines inducing the differentiation of IFN-γ producing Th1 or IL-4 producing Th2 cells, respectively [4-6]. Moreover, in this system transcription factors, signal-transducer and activator of transcription protein (STAT) 4 and T-box 21 (TBX21, also known as T-bet) or STAT6 and GATA-binding protein 3 (GATA3) are essential for differentiation and cytokine expression of Th1 or Th2, respectively [4, 5]. In spite of the fact that these factors have been known for almost two decades, it remains incompletely understood, particularly in human, how these key factors trigger signalling networks and further influence intracellular events leading naive T cells to initiate differentiation towards distinct subsets. Hence, the role and interactions of a range of players in these processes need to be further characterized and understood. Systems biology approaches provide a holistic view of the processes, such as T helper differentiation, enable discovery of new factors and pathways involved and promote generation of novel hypotheses to be tested to improve our understanding of human Th cell differentiation and immune-mediated diseases.

Molecular mechanisms controlling helper T cell differentiation have been thoroughly described and summarized in a recent review [1]. Here, we will summarize studies where high-throughput systems biology approaches have been exploited to identify new players and signalling pathways important for helper T cell activation and differentiation. We focus on human primary T cell studies, where defining factors and signalling pathways affecting T cell differentiation towards distinct subsets not only provides new insights into T cell biology but also for developing new therapies targeting immune system for human diseases. Importantly, all clinical studies must be based on in depth understanding of human immune responses.

Transcript profiling—microarray studies

  1. Top of page
  2. Abstract
  3. Introduction
  4. Transcript profiling—microarray studies
  5. Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets
  6. Proteomics provides complementary view
  7. Concluding remarks
  8. Acknowledgment
  9. References

DNA microarray technology was developed to monitor the expression of large numbers of genes simultaneously in a quantitative manner and to identify differentially expressed genes by comparing expression profiles from different samples [7, 8]. This technology was the beginning of revolution that has changed the experimental approaches from one gene or protein at a time to building holistic views of biological processes. The technology was applied early on to study gene expression profiles in human Th1 and Th2 cells by two groups having early access to Affymetrix technology. In both studies, in addition to the known Th1/Th2 signature genes such as IFNγ, IL4, IL5, IL13 and interleukin 12 receptor, beta 2 (IL12RB2), several genes new in this context were found to be differentially expressed [9, 10]. Despite the differences of experiment settings as well as the relatively small number of genes included on microarrays, the encouraging results immediately led to identification of new players involved in human Th1 and Th2 differentiation. Meanwhile, with the sequencing of human genome and further development of microarray technology, rapidly increasing number of genes/probe sets were included in array designs.

To study the initiation of Th1 and Th2 differentiation in human CD4+ T cells and to identify the key molecules essential for fate decision of Th1/Th2 subset, Lund et al. used oligonucleotide arrays [11, 12]. In addition to genes previously implicated in the process, many novel players not previously associated with Th1/2 polarization were identified. Consistent with the kinetics of activation-induced IL12RB2 expression, effects of IL-12-induced gene expression were observed a bit later [12]. In addition to IL-12, type I IFNs also favour Th1 differentiation, inducing IFN-γ production in human CD4+ T cells [13]. Microarray studies revealed a set of genes regulated in a similar manner by IL-12, IL-18 and IFN-α in human helper T cells and resulted in the discovery of a new Th1-promoting transcription factor activating transcription factor 3 (ATF3) [14].

DNA microarrays were utilized to study mouse Th1, Th2 as well as cytotoxic T (Tc)1 and Tc2 cells [15]. Studies on mice deficient for STAT4 or STAT6 clearly demonstrated the essential role of STAT4 and STAT6 for directing Th1 and Th2 differentiation, respectively [16-20]. DNA microarrays were applied to identify genes regulated by STAT4 or STAT6 during differentiation of Th1 or Th2 cells, as STAT4/STAT6 targets are candidates likely to regulate the entire process [21, 22]. The results indicated that Ifng was the first gene clearly upregulated at early stage of Th1 differentiation in a STAT4-dependent manner. However, a number of genes were quickly upregulated in response to Th1 polarizing conditions independent of STAT4, suggesting that in addition to IL-12/STAT4, other signalling pathways such as IFN-γ/STAT1/TBX21 (T-bet) were also required for Th1 induction [22]. Studies on early IL-4-induced Th2 cell differentiation revealed several rapidly induced novel IL-4/STAT6-regulated genes suggesting their importance for the differentiation process [21]. Our excitement on the results defining the STAT6-regulated genes in the mouse was, however, compromised by the fact that only a small subset of them was similarly regulated during the early stages of human Th cell differentiation pointing out the emergency and importance of human studies.

Application of microarray technology in human T cell research quickly resulted in the discovery of new players involved in Th1/Th2 differentiation and regulated via STAT4 or STAT6. Their roles in Th cell subsets differentiation have been shown in further functional studies. For example, ATF3 was identified as a transcript upregulated by IFN-α, IL-12 and IL-18. Overexpression of ATF3 in CD4+ T cells enhanced the production of IFN-γ, whereas siRNA knock-down of ATF3 reduced IFN-γ production. ATF3 transcription factor was found to be recruited to and transactivate the IFNG promoter in a complex with jun proto-oncogene (JUN) during early Th1 differentiation [14]. The expression of paired basic amino acid cleaving enzyme (Furin) was first found to be induced by IL-12 in STAT4-dependent manner [22]. Further study exploiting conditional deletion of Furin in T cells indicated that as a proprotein convertase Furin is indispensable in maintaining peripheral tolerance, perhaps by controlling the levels of active form of transforming growth factor, beta 1 (TGFβ-1) [23]. Deletion of Furin in T cells results in loss of peripheral tolerance characterized by activated T cells that overproduce both Th1 and Th2 type cytokines, circulating autoantibodies and development of inflammatory bowel disease.

Transcriptional profiling of a more recent Th17 lineage, became publically available only recently in human. A detailed kinetics of gene expression at very early stages of human Th17 differentiation was studied by genome-wide microarrays [24]. During the first three days of human Th17 cell differentiation, altogether >1000 genes were differentially expressed including many genes previously not associated with Th17 polarization. Together with previously known regulators, it is now possible to start to construct gene regulatory network leading to initiation of human Th17 differentiation and to elucidate how this process can be modulated.

Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets

  1. Top of page
  2. Abstract
  3. Introduction
  4. Transcript profiling—microarray studies
  5. Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets
  6. Proteomics provides complementary view
  7. Concluding remarks
  8. Acknowledgment
  9. References

GATA3 and T-bet have been accepted as key transcription factors for Th1/Th2 lineage commitment. Chromatin immunoprecipitation followed by microarray (ChIP-chip) technology has been applied to identify which genes are regulated by these two factors in Th1/Th2 cells. Using differentiated Th1/Th2 from human naïve T cells, Jenner et al. [25] first mapped GATA3 and T-bet binding sites in human Th1 and Th2 cells. Interestingly, the study shows that a set of genes are bound by GATA3 in both Th1 and Th2 cells and those genes are also targeted by T-bet in Th1 cells, including IL4 and IFNG which are activated by one transcription factor and inhibited by another one, supporting their oppositing regulatory roles in Th1/Th2 fate decision.

For over a decade, microarrays have been the preferred genome-scale analysis method. Next-generation sequencing (NGS) has developed fast and is replacing microarray platforms in many applications. For example, ChIP-chip has been largely replaced by ChIP accompanied by high-throughput massive parallel sequencing (ChIP-Seq) to elucidate binding sites for transcription factors or histone modification marks in a genome-wide scale. In recent years, it has been noted that the phenotypes of these functionally distinct subsets of helper T cells are not stable. For example, in certain situations, Foxp3+ T cells can produce IL-17 [26], IL-17-producing Th17 cells can produce IFN-γ [27] and even Th2 cells can turn to be IFN-γ-producing cells [28]. Even in humans, double or multiple signature cytokine–producing helper T cells have also been detected [29], indicating the flexibility of helper T cell phenotypes. ChIP-Seq technology provides a powerful tool to explore the mechanisms controlling plasticity of Th subsets phenotypes by mapping the binding sites of lineage-specific transcription factors and the histone modification marks.

Several genome-wide studies to investigate where transcription factors important for Th cell differentiation bind in murine T cell subsets have been reported in recent years [30-36]. Studies in both mouse and human have demonstrated STAT4/TBX21, STAT6/GATA3, STAT3/RAR-related orphan receptor gamma (RORC), STAT5/FOXP3 as key transcription factors regulating Th1, Th2, Th17 and iTreg differentiation, respectively [1]. Therefore, investigating where transcription factors bind across the genome during the process of T subsets differentiation will provide information needed to further understand transcriptional regulation in helper T cells. As previously indicated, cytokine microenvironment is crucial for fate decision of helper T cell lineage commitment and many cytokine signals are transmitted into cells through STAT family of proteins, which further induce target gene transcription. ChIP-Seq studies for STAT4, STAT6, STAT3 and STAT5 have been performed in response to IL-12, IL-4, IL-6 and/or IL-21, IL-2 stimulation using WT and STAT deficient T cells accordingly [30-33, 35, 37]. While it is well established that the STATs are important for cytokine signalling and initiation of differentiation processes, it was interesting to find out that even in differentiated Th1, Th2, Th17 or iTreg cells, genome-wide mapping revealed large amount of STAT binding sites in addition to those previously established to control the expression of signature cytokine loci. However, identification of binding sites does not alone indicate the functional impact of such binding. To address this question, further maps of transcription factor binding have been combined with detection of chromatin modification markers such as histone H3 lysine 4 (H3K4) and lysine 27 (H3K27) trimethylation, enriched in promoters or repressors, respectively, as well as expression levels of neighbouring genes [30, 31, 34]. For example, H3K4me3 and STAT4, STAT6 or STAT3 are found to be associated with Ifng, Il4 or Il17, respectively, in differentiated Th1, Th2 and Th17 lineages [30-32]. Interestingly, STAT proteins appear to induce gene expression in one lineage and inhibition of associated active epigenetic marks in another lineage [32, 38]. Using WT and STAT4 or STAT6 deficient mice combined with ChIP-seq for STAT4, STAT6 and epigenetic marks as well as microarrays, Wei et al. [32] identified an important set of genes which both STAT4 and STAT6 bind but act in an opposing manner to modulate epigenetic changes and associating gene expression. Surprisingly, STAT proteins not only bind to promoter regions directly controlling transcription of target genes, but a large amount of STAT binding sites are also identified in introns and intergenic regions, where the exact role of STAT binding is poorly understood. Furthermore, studies on key regulators Gata3 and Tbx21 suggested these factors to regulate both expression of Th2 or Th1 lineage-specific genes as well as their active and repressive histone modifications. For example, in Th2 cells, GATA3-binding sites identified in the Il4 and Il13 genes, overlapped with H3K4me1, H3K4me2 and/or H3K4me3, but H3K27me3 was not detected in genomic region surrounding the GATA3-binding sites. On the other hand, GATA3-binding sites were identified also in the Tbx21 and Ifng genes, whose expression were inhibited by GATA3. However, these sites overlapped with H3K27me3, a marker of repressive histone modification [34]. This study suggests that GATA3 can activate Th2 cytokine locus by facilitating methylation of H3K4 of enhancers, which is correlated with transcriptional activation; on the other hand, GATA3 can also repress the expression of genes critical for Th1 differentiation such as Tbx21 and Ifng by modulating methylation of H3K27 which is correlated with transcription repression. ChIP-seq study from differentiated human Th1/Th2 cells also suggests that GATA3 and T-bet can bind to distal regulatory elements. Moreover, enrichment of both enhancer mark (H3K4me1) and insulator CCCTC-binding factor (zinc finger protein) (CTCF) were found in these distal GATA3 and T-bet binding sites [39]. Therefore, such transcription factors can either directly influence transcriptional machinery and/or affect loci encoding critical cytokines, cytokine receptors, signalling molecules as well as transcription factors that are involved in the regulation of Th1 and Th2 differentiation [32, 34, 36, 39].

By mapping active enhancer landscapes in differentiated Th1 and Th2 cells, Vahedi et al. showed that Th1-specific enhancers are enriched for the consensus motifs for STAT1 and STAT4 but are relatively devoid of the STAT6 motif; on the other hand, Th2-specific enhancers exhibit enrichment of STAT6 motifs. They discovered that binding of these STAT proteins can activate lineage-specific enhancers and suppress enhancers associated with alternative cell fates [40]. However, the study was carried out in murine T cells. To investigate how early cell fate commitment is regulated, the first human genome-wide maps of histone modifications revealed enhancer elements during the initiation of polarization towards Th1 and Th2 lineages. The analysis indicated that even at this very early stage, cell-specific gene regulation and enhancers are at work directing lineage commitment. Further examination of lineage-specific enhancers identified known and novel transcription factors as putative drivers of lineage-specific gene expression. Lastly, an integrative analysis of chromatin state maps, transcription factor-binding sites (TFBS) motif analysis and modelling of enhancer-gene pairs combined with autoimmune Genome-wide association studies (GWASs) on single-nucleotide polymorphisms (SNPs) associated with immunopathogenesis suggest a role for distal regulatory elements in the disease aetiology. In several instances, these variants occurred more frequently in known motifs for transcription factors previously shown to be important in T helper cell differentiation. Functional validation of a panel of enhancer motifs demonstrated that these SNPs altered binding of transcription factors involved in Th cell differentiation. These results provide a novel mechanistic insight into dysregulation of Th cell differentiation in human immune-mediated diseases [41].

In spite of established importance of Janus kinase (JAK)-STAT pathways in Th cell differentiation, studies reporting genome-wide mapping of TF binding sites for STATs and other key transcription factors in human are very limited. Given the importance of regulation of Th cell responses in a range of human diseases and conditions, it is likely that individual variation in binding sites of transcription factors that are key regulators of such responses provide important insights into immune regulation underlying human diseases. The first study to identify genome-wide targets of STAT6 in human was reported by Elo et al. [38]. Genes regulated by STAT6 in primary human CD4+ T cells induced to differentiate to Th2 lineage were detected using siRNA-mediated downregulation of STAT6 followed by genome-wide transcriptomics. Direct targets of STAT6 were identified through genome-wide STAT6 ChIP-seq. Combining the results revealed that up to 80% of IL-4-induced genes are regulated by STAT6 during the early stages of human Th2 differentiation and 30% of these are direct targets of STAT6. This strategy resulted in the discovery of a range of genes likely to contribute to the early linage specification and transcriptional programme leading to the Th2 phenotype. A panel of novel candidates for future studies aiming at modulation of Th2 response was identified [38].

Special AT-rich sequence-binding protein 1 (SATB1) was identified preferentially expressed in Th2 cells from microarray studies [11]. Studies demonstrate that expression of SATB1 is rapidly upregulated in polarizing Th2 cells, its induction is dependent on IL-4/STAT6 but not GATA3 [42]. It was shown that the induced SATB1 protein formed a unique transcriptionally active chromatin structure where chromatin is folded into numerous small loops, all anchored to SATB1 at their base. SATB1 is required not only for compacting chromatin into dense loops cross Th2 cytokine locus but also for regulating a set of genes, inducing Il4, Il5, Il13, GATA3 and v-maf musculoaponeurotic fibrosarcoma oncogene homologue (avian) (c-Maf) expression [42, 43]. SATB1 binds to IL5 promoter and inhibits IL5 expression. SATB1 ChIP-on-Chip study in human T cells indicated that SATB1 recruits beta-catenin and p300 acetyltransferase on GATA-3 promoter in differentiating Th2 cells. The SATB1: beta-catenin complex activates a number of SATB1-regulated genes [42]. These results demonstrate that SATB1 orchestrates Th2 lineage commitment by regulating chromatin structure as well as gene expression. In a recent study, genome-wide microarray analysis showed that human natural Tregs express lower level of SATB1 than conventional T cells and this repression of SATB1 was needed for the phenotype and suppressive function of Treg cells [44]. siRNA downregulation of Foxp3 in human Treg cells resulted in increased expression of SATB1 and loss of suppressive function. Combining Foxp3 ChIP-on-chip with in silico analysis, the Foxp3-binding regions were identified in the promoter and in the genomic locus of SATB1, which was further confirmed by ChIP-qPCR. Furthermore, overexpression of SATB1 in human Treg cells results in reprogramming of human Treg cells into T effector cells and this is supported by microarray study showing induction of many genes involved in effector T cell differentiation in SATB1 overexpressing Treg cells. Application of systems biology approaches in this study revealed that inhibition of SATB1-mediated global chromatin remodelling is required for maintaining of Treg phenotype and cell function [44].

One of the most striking findings from these global histone epigenomic studies is in Th lineages, the colocalization of both activating and repressive epigenetic modifications in the same area in loci of master regulators. A typical example is that the H3K4me3 and H3K27me3 modifications coexist in Tbx21 locus in Th2, Th17 and Treg cells suggesting the gene is poised for expression [30]. Indeed, studies have shown that Th2, Th17 or Treg cells can be reprogrammed to IFN-γ-producing cells indicating plasticity of T helper cells. Such a plasticity in human cells was reported almost for two decades ago when human Th2 cell clones could be modulated to produce IFN-γ through particular costimulatory signals [45].

Proteomics provides complementary view

  1. Top of page
  2. Abstract
  3. Introduction
  4. Transcript profiling—microarray studies
  5. Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets
  6. Proteomics provides complementary view
  7. Concluding remarks
  8. Acknowledgment
  9. References

In comparison with transcriptomics, the high-throughput methods for protein analysis have been more limited in scope and sensitivity. These limitations largely result from fundamental differences in biochemical properties of the analytes. In particular, for nucleic acids, Watson–Crick base-pairing can be exploited in identification, enzymatic manipulation and amplification of target molecules, whereas proteins do not exhibit similar regularity, effectively limiting the system-wide analytical methods to direct biochemical characterization by electrophoresis, mass spectrometry and targeted antibody-based assays. Accordingly, gene expression is commonly measured on the level of mRNAs, representing a useful surrogate of protein abundances and providing a highly informative snapshot of the state of cellular differentiation [46]. Nevertheless, mRNA levels do not universally predict levels of the corresponding protein products [47], and proteins are subject to modification, translocation, binding and degradation events beyond the scope of transcriptomics. Hence, direct investigation of proteins can provide biologically significant information not available by any other approaches.

System-wide study of proteins, later to be referred as proteomics, was initially enabled by introduction of two-dimensional electrophoresis (2-DE), combining isoelectric focusing with SDS-PAGE separation [48, 49]. The breakthrough in protein identification by mass spectrometry followed in the early 90s [50]. Subsequently, these methods were applied in proteomic characterization of Jurkat T cell line [51], as well as primary CD4+ T cells from adult human peripheral blood [52]. In both studies, in the order of 2000 protein spots were visualized. By mass spectrometry, 62 proteins were identified from the Jurkat cells [51], and 91 from primary T cells [52], representing mainly proteins highly abundant in these cell types. When performed with sufficient reproducibility, 2-DE provides a useful tool for identifying quantitative differences between experimental conditions, such as cytokine-induced T helper cell differentiation programmes. With human T cells, this strategy was successfully used for detecting protein-level responses for IFN-alpha [53], IL-4 [54, 55], IL-12 [55, 56], as well as costimulation through CD28 [57].

Despite its utility, the 2-DE method has several shortcomings. Due to the limited sensitivity of protein staining methods, only relatively large differences are detected, and the identification can be complicated by the presence of multiple proteins within the differentially stained gel spots. Furthermore, the approach is particularly poorly suited for the analysis of hydrophobic, membrane-associated proteins [58, 59]. As many of these proteins are crucial for signal transduction, this limitation has been addressed by targeted analyses of membrane fractions. Using in vitro stimulation combined with SILAC-based labelling and quantification, Loyet et al. [60] compared the membrane proteomes of fully differentiated human primary Th1 and Th2 cells, resulting in identification of T cell receptor associated transmembrane adaptor 1 (TRAT1, also known as TRIM) and bone marrow stromal cell antigen 2 (BST2) as proteins preferentially expressed on Th1 cells despite similar mRNA expression levels. Furthermore, in context of T cell activation, the specific microenvironment of the T cell receptor, often referred to as lipid rafts, along with the associated proteome, has been targeted in several studies [61-64].

A number of conceptually related investigations of other T cell subproteomes have been reported, targeting cellular compartments or organelles such as the cytoplasm [65], mitochondria [66], microsomal fractions [67, 68], nucleus [69, 70], nucleolus [71], as well as proteins specifically associated with chromatin or nuclear matrix [72]. In addition to the advantage of reduction in sample complexity, such targeted approaches can provide valuable information of a biological process in consideration. For example, the study by Moulder et al. [70] resulted in detection of IL-4-induced responses in levels of nuclear factors IKAROS family zinc finger 1 (IKZF1), SATB1, Y box binding protein 1 (YBX1, also known as YB1), suggesting involvement in the regulation of Th2 differentiation. Similarly, the analysis of human Th cell microsomal fractions lead to the identification of GTPase, IMAP family member (GIMAP) 1 and GIMAP4 as factors downregulated by IL-4 and induced by Th1 cytokines, potentially important for development of human Th1 responses [68].

In parallel with quantification of protein expression levels, proteomics provides a powerful tool for characterization of post-translational modifications (PTMs), such as phosphorylation, acetylation, ubiquitylation, methylation and modification by o-linked beta-N-acetylglucosamine [73]. In the context of the development of effector T helper cell populations, the phosphorylation-mediated signalling pathways initiated by TCR ligation have been intensively studied using predominantly the Jurkat T cell line [74]. In the global phosphoproteomic studies, the focus has typically been in the early time points of activation, using immunoprecipitation with antiphosphotyrosine antibodies or immobilized metal affinity chromatography for enrichment of phosphorylated proteins or peptides [74-79, 81]. With primary human T cells, only few corresponding investigations have been published [81, 82, 84]. Nevertheless, some noteworthy differences between Jurkat cells and primary T cells exist. In particular, Jurkat cells are deficient for lipid phosphatases phosphatase and tensin homologue (PTEN) and inositol polyphosphate-5-phosphatase, 145kDa (INPP5D, also known as SHIP), decreasing sensitivity to apoptosis, elevating intracellular levels of D3 phospholipids, and leading to constitutive membrane association of IL2-inducible T cell kinase (ITK), and activation of v-akt murine thymoma viral oncogene homologue 1 (AKT1) [84, 85, 87]. Accordingly, phosphorylation levels of protein tyrosine kinase 2 beta (PTK2B, also known as PYK2), phospholipase C, gamma 1 (PLCgamma1), vav 1 guanine nucleotide exchange factor (VAV1) and mitogen-activated protein kinase 3 (MAPK3, also known as ERK1)/mitogen-activated protein kinase 1 (MAPK1, also known as ERK2), as well as levels of Ca2+ flux are higher in Jurkat cells than in primary T cells [88].

In contrast to pathways engaged by T cell activation, cytokine signalling leading to T helper cell differentiation has not been systematically investigated by global phosphoproteomics, the only exception being recent description of tyrosine-phosphoproteome proximal to IL-2 receptor [89]. Furthermore, PTMs important for T helper cell differentiation might not be limited to phosphorylation. For example, recent reports have highlighted the role of arginine methylation in T cell activation and development of Th2 responses [90, 91]. The full scope and importance of post-translational regulation in T helper cell differentiation, but also in general, remains to be uncovered.

Concluding remarks

  1. Top of page
  2. Abstract
  3. Introduction
  4. Transcript profiling—microarray studies
  5. Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets
  6. Proteomics provides complementary view
  7. Concluding remarks
  8. Acknowledgment
  9. References

Next-generation sequencing technology has been applied to study epigenomics in mouse helper T cell subsets and will be certainly used in human primary T cell differentiation studies. With the development of this technology to selectively enrich and sequence specific parts of the genome at high coverage, it can also be used to study RNA splicing events during T cell subsets differentiation. Identification of genetic variants such as SNPs associated with human diseases is another major application of this technology. In a recent study, using next-generation exome sequencing, in a cohort of 76 patients with T cell large granular lymphocytic leukaemia, mutations in STAT3 were identified in 40% patients with this disease. All the found mutations were located in exon 21, encoding the SH2 domain. The findings suggest involvement of STAT3 signalling in the pathogenesis of this disease [92].

In addition to the Th cell subsets discussed above, Th9, Th22, follicular helper T (Tfh) cells have also been reported as distinct subsets of helper T cells [92, 93, 95]. It is still discussed whether they exist in vivo or whether they represent a distinct lineage as some of these cells express several cytokines [95, 97, 98]. Importantly, comprehensive genome-wide analysis of these subsets in human would be important to improve our understanding their role and regulation in human health and disease.

Recently developed multiparameter mass cytometry, called the CyTOF, offers single-cell analysis of theoretically up to 100 simultaneous parameters without fluorescent agents or interference from spectral overlap [99]. The technology couples features from flow cytometry with mass spectrometry, specifically the workflow is the same as flow cytometry but using metal-conjugated antibody to detect protein or phosphylated protein instead of using fluorescence-conjugated antibodies. The development of this technology brings flow cytometry into post-fluorescence era which makes it possible to characterize multiple cell subsets in precious samples such as clinical samples. The method can also be used to study intracellular signalling events in polarizing or TCR-activated T cells.

In the past decade, application of systems biology approaches has already greatly improved our understanding of Th cells differentiation. Many molecules, signalling pathways and even transcriptional regulatory networks have been identified in a more efficient manner. Our current understanding of molecular mechanisms of Th cell differentiation is not only limited to few molecules like 10 years ago but is now complemented with the studies exploiting systems biology strategies, Fig. 1. The development of systems biology approaches is accompanied with the development for computational biology and bioinformatics. The increasing amount of data requires efficient analysis and statistical approaches and tools. Exploitation and further development of methods in computational biology is essential for efficient data analysis and interpretation. Integration of data generated from different platforms and different regulatory levels provides the basis for constructing regulatory networks of signalling and transcription during the process of lineage commitment. This in turn can be used to identify and validate points of modulating the process. In a recent study, using systems biology approach, Ciofani et al. [100] integrated data from ChIP-seq study of several transcription factors, gene expression profiling of cells with mutated transcription factors and kinetic gene expression profiling to construct a transcriptional regulatory network of murine Th17 differentiation. Such a strategy is likely to be increasingly used in the future.

image

Figure 1. Systems biology studies on human Th subsets differentiation.

Download figure to PowerPoint

Current understanding on T cell activation and subset differentiation largely relies on data generated from mouse studies. As a convenient tool, mouse cells are more homogeneous and mouse lines are easy to manipulate by inbreeding. Although large amount of studies conducted in mice provide insight into immune responses, mouse immune system is not identical to the one in humans, and the fidelity by which mouse models replicate immunopathogenic mechanisms in humans can be questioned, especially when many genetic polymorphisms identified linked with human diseases may not even occur in the mouse genome. Therefore, to complement the mouse studies, it is important to gain further understanding on human immune responses which can be used for combating human diseases. It has been unclear for a long time how similar or different human T cell responses are compared to those in mouse. How to apply knowledge obtained from mouse studies to further understand or treat human diseases? Therefore, studies comparing gene expression profile and epigenetic modification of human and mouse T helper subsets are highly desired. Data generated from these studies may help us to answer these questions.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Transcript profiling—microarray studies
  5. Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets
  6. Proteomics provides complementary view
  7. Concluding remarks
  8. Acknowledgment
  9. References

This work was supported by The Academy of Finland, European Union FP7 Grant ‘Systems Biology of T cell activation in health and disease’ (SYBILLA), the Academy of Finland (Centre of Excellence in Molecular Systems Immunology and Physiology Research, 2012–2017, Decision 250114), The Sigrid Jusélius Foundation, The National Technology Agency of Finland (TEKES).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Transcript profiling—microarray studies
  5. Genome-wide identification of target genes regulated by transcription factors important for differentiation of T helper subsets
  6. Proteomics provides complementary view
  7. Concluding remarks
  8. Acknowledgment
  9. References