cAMP responsive element modulator α promotes effector T cells in systemic autoimmune diseases

T lymphocytes play a crucial role in adaptive immunity. Dysregulation of T cell‐derived inflammatory cytokine expression and loss of self‐tolerance promote inflammation and tissue damage in several autoimmune/inflammatory diseases, including systemic lupus erythematosus (SLE) and psoriasis. The transcription factor cAMP responsive element modulator α (CREMα) plays a key role in the regulation of T cell homeostasis. Increased expression of CREMα is a hallmark of the T cell‐mediated inflammatory diseases SLE and psoriasis. Notably, CREMα regulates the expression of effector molecules through trans‐regulation and/or the co‐recruitment of epigenetic modifiers, including DNA methyltransferases (DNMT3a), histone‐methyltransferases (G9a) and histone acetyltransferases (p300). Thus, CREMα may be used as a biomarker for disease activity and/or target for future targeted therapeutic interventions.


IMMUNE DYSREGULATION IN SYSTEMIC AUTOIMMUNE/ INFLAMMATORY DISEASES
Immune responses are timely and spatially regulated to allow clearance of pathogens while limiting inflammationmediated tissue damage and subsequent organ failure.They are divided into innate and adaptive immune mechanisms.Innate immune responses (more or less) precisely directed against conserved molecular patterns shared between pathogens, such as components of bacterial walls (e.g., lipopolysaccharides, muramyl dipeptide, etc.) or viral components (e.g., cytoplasmic nucleic acids).Adaptive immune mechanisms can be 'learned' and target structures specific to individual pathogens (e.g., the SARS-CoV-2 spike protein) [1].Adaptive immune responses are mediated by lymphocytes, including B and T cells.Selfreactivity is prevented or controlled through complex mechanisms of selection during lymphocyte differentiation in primary lymphoid organs, and their suppression in secondary lymphoid tissues [2].Uncontrolled activation of self-reactive lymphocytes and increased effector functions can result in systemic inflammation, amplified tissue damage and the expression of autoimmune disease.Examples for systemic autoimmune/inflammatory diseases centrally driven by effector T cells include systemic lupus erythematosus (SLE) and psoriasis.
SLE is a complex systemic autoimmune/inflammatory disease that can affect any organ of the human body [3].It is characterized by systemic inflammation, tissue damage and resulting organ failure in the presence of high-titre autoantibodies and autoreactive lymphocytes [4].Despite recent advances, the exact molecular pathophysiology of SLE remains largely unknown.Female predominance after the onset of puberty, increased incidence and prevalence in non-white ethnicities, incomplete penetrance in genetically identical monozygotic twins in the presence of familial clusters suggest a complex interplay between genetic and environmental factors resulting in the dysregulation of inflammatory responses, breakdown of self-tolerance, tissue damage, autoantibody production, immune complex deposition and lastly organ failure [5][6][7][8][9][10][11].
Psoriasis is a systemic autoimmune/inflammatory disease that primarily affects the skin.It is characterized by increased keratinocyte proliferation and immune cell infiltration of the skin, including T lymphocytes.Effector CD4 + and CD8 + T cells centrally contribute to skin inflammation and tissue damage in psoriasis [12,13].As in SLE, the molecular pathophysiology of psoriasis is incompletely understood.A complex interplay between genetic and environmental factors has been proposed, resulting in autoimmunity and inflammation [14].
Both SLE and psoriasis feature effector T cell dysregulation as key pathomechanism and are characterized by imbalanced expression of immune-regulatory (including IL-2) and pro-inflammatory effector (including IL-17A) cytokines [15,16].Dysregulated cytokine expression in SLE and psoriasis, and potentially other inflammatory diseases driven by T cell dysregulation is underpinned by deregulated transcription factor networks, including the cAMP response element modulator (CREM) family [17].Specific focus has been given to the CREMα isoform that is upregulated in both SLE and psoriasis T cells [18].CREMα has been implicated in steering T cells towards a pro-inflammatory effector phenotype in controlled ex vivo and in vitro experiments [19][20][21][22][23][24][25].

EFFECTOR T CELLS IN AUTOIMMUNE/INFLAMMATORY DISEASES
The key role of self-reactive T cells, and their potential as therapeutic targets in autoimmune disease, including SLE [26], JIA and psoriasis [14,27] is beyond doubt and centrally contributes to systemic inflammation, tissue and organ damage.When exposed to a specific antigen, naïve CD4 + T cells are activated and differentiate into effector T cells, including (but not limited to) Th1, Th2, Th17 and follicular helper CD4 + T (TFH) cells (Figure 1).
In the effector T cell-driven systemic autoimmune diseases SLE and psoriasis, special attention has been paid to IL-17 producing T cells, and the IL-23/Th17 axis has been identified as an important factor in the pathophysiology, centrally contributing to inflammation and damage [15,16].IL-23 is a heterodimeric cytokine consisting of the p19 and p40 subunits secreted primarily by innate immune cells including dendritic cells (DCs) and macrophages.IL-23 receptor (IL-23R) activation causes phosphorylation of the protein kinases Jak2 and Tyk2, which then activate the transcription factors signal transducers and activators of transcription (STAT)3 and RARrelated orphan receptor (ROR)-γt, both of which promote Th17 differentiation and IL-17 expression [28,29].The effector cytokine IL-17A binds to and activates the IL-17 receptor (IL-17R), stimulating neutrophils, lymphocytes, monocytes, fibroblasts and keratinocytes and causing and amplifying inflammation [30][31][32].
In addition to the presence of transcription factors (including STAT3 and ROR-γt), Th17 lineage differentiation is controlled by modifications to chromatin accessibility, namely epigenetic modifications.Key factors include the cytokine IL-6, which in cooperation with TGF-β inhibits the generation of regulatory T cells and steers differentiation towards Th17 subpopulations [33,34].Although IL-6-deficient mice cannot produce Th17 responses, depletion of regulatory T cells, through treatment with inactivating anti-CD25 antibodies, in this strain restores development of Th17 cells [35].This indicates additional pathways influencing Th17 differentiation.IL-21, a member of the IL-2 family of cytokines was identified in the same study as contributing to Th17 differentiation independent of IL-6 [35].However, other studies provided conflicting reports on the relevance of IL-21 in this context [36,37].Th17 differentiation independent of TGF-β can also be achieved by IL-1β in combination with IL-6 and IL-23 [38].Although the necessity of IL-23 may seem counterintuitive, because naïve CD4 T cells lack the IL-23 receptor, the cytokine has been shown to play a key role in lineage stabilization [39,40] and induces the development of highly pathogenic Th17 cells in cooperation with IL-6 [41].
The immune regulatory cytokine IL-2 plays a key role in inhibiting Th17 differentiation [42].Upon cytokine receptor activation, transcription factor phosphorylation is mediated through Janus kinases (JAKs), for example, mediating activation of downstream STAT family transcription factors.While IL-6 and IL-23 signalling occurs through STAT3 activation, IL-2 signalling is mediated through STAT5.Autoimmune/inflammatory diseases frequently feature an imbalance between pro-and antiinflammatory cytokines, such as IL-2/IL-17A in SLE and psoriasis, contributing to and underscoring the effector T cell phenotype of the diseases [22,25,43].
In addition to helper CD4 + T cells, cytotoxic CD8 + T cells are dysregulated in SLE and psoriasis [44][45][46].Notably, in SLE, increased inflammatory responses are coupled with reduced cytotoxic capacity of CD8 + T cells which contributes to inflammation, but also the secondary immunodeficiency and increased susceptibility to infections [47].Notably, though downregulation of CD8 surface expression, CD8 + T cells can develop into CD3 + CD4 À CD8 À 'double negative' (DN) T cells that play a role in both SLE and psoriasisassociated tissue inflammation and damage through their ability to produce pro-inflammatory effector cytokines [19,20,[48][49][50][51].

TRANSCRIPTIONAL AND EPIGENETIC REGULATION OF EFFECTOR T CELLS
As mentioned above for Th17 lymphocytes, the presence and recruitment of transcriptional regulators define effector T cell differentiation and activation [52] (Figure 1).Transcription factors recruit to accessible regulatory regions of the genome where they induce or repress transcription (trans-activation/Àrepression) or mediate changes to chromatin structure through the induction of epigenetic remodelling [53].This is mediated by epigenetic modifications that enhance or reduce gene expression through modifications to chromatin accessibility that influence gene expression profiles and cellular functions.The two best studied epigenetic modifications, to date, are DNA methylation and posttranslational modifications to histone proteins [54,55].
DNA methylation is characterized by the addition of a methyl group to the 5 0 carbon position of cytosine within cytosine guanine dinucleotides that negatively affects the recruitment of transcription factors and RNA polymerases (Figure 2a).It is mediated by DNA methyltransferases that have historically been classified as either 'maintenance' DNA methyltransferases, which are responsible for re-methylation of the second DNA strand during cell division (e.g., DNMT1), or 'de novo' enzymes that mediate DNA methylation in previously unmethylated regions (e.g., DNMT3a) [6,56,57].DNA methylation can be reduced by 'loss of methylation' during cell division that can be the result of reduced DNMT1 activity or lack of methyl donors, or in an active process that involves the enzymatic removal of methyl groups utilizing molecules such as ten-eleven translocation (TET) family enzymes and thymine DNA glycosylase (TDG) [58].As DNA methylation is a relatively stable epigenetic mark, it plays a role in the maintenance of genomic stability, silencing of genes or entire chromosomes (e.g., the second X chromosome in women) [59].
The three-dimensional DNA structure is organized by nucleosomes that contain octamers of histone proteins and two loops of DNA (Figure 2b).Post-translational modification to histone proteins regulate their electric charge, thereby fine-tuning the distance between nucleosomes and resulting accessibility of DNA the transcriptional complex [60].Histone modifications include, but are not limited to, acetylation, methylation, phosphorylation and ubiquitination [60].Histone modifications, their interactions and effects on chromatin structure and gene expression are complex and incompletely understood.Histone modifications are mediated by a host of enzymes many of which are directed to chromatin through transcription factors, such as the DNA methyltransferase G9a or the transcriptional co-activator p300 that functions as histone acetyltransferase [61,62].Overall, histone marks are less stable when compared to DNA methylation, but generally follow the same 'code'.Indeed, DNA methylation and histone modifications are linked by molecular interactions that 'translate' between histone and DNA methylation codes [63].
Epigenetic modifications have a significant impact on T cell differentiation and function, including linage commitment to short-lived effectors, long-term memory T cells, T regulatory cells and other specific T cell populations [64].Alterations to epigenetic marks in lymphocytes and other immune cells have been linked with SLE, psoriasis and other autoimmune/inflammatory diseases [65].Indeed, in SLE, epigenetic marks have been proposed as 'missing links' between genetic predisposition and disease expression.Notably, DNA methylation varies significantly between genetically identical monozygotic twins discordant for SLE [66].The patterns of DNA methylation are complex and vary from gene to gene.While, in SLE, global hypomethylation of DNA has been reported in T cells [67], some (silenced) genes exhibit increased DNA methylation (including the immune regulatory cytokine encoding IL2 gene in CD4 + T cells and the CD8 cluster during the generation of DN T cells) [19,68].Reduced methylation of a series of genes encoding for pro-inflammatory cytokines and chemokines permit the increased expression of type I interferons [69] and IL-17A [21,23,68].
While less well-studied, epigenetic alterations have also been linked with the inflammatory phenotype of psoriasis.Increased methylation of the PDCD1 gene results in reduced expression of the regulatory co-receptor PD-1 on CD4 + T cells [18].Reduced methylation of the IFNG promoter in DN T cells allows for IFN-γ expression [48], and distinct DNA methylation patterns in CD8 + T cells allow distinguishing between patients with skin psoriasis, psoriatic arthritis and healthy individuals [13].

THE CREM TRANSCRIPTION FACTOR SUPERFAMILY
When first reported, three separate CREM protein-coding transcripts were described [70].However, over recent years it became apparent that a much larger number of distinct isoforms arise from the 14 exons of the CREM gene.Isoforms are the result of alternative splicing, outof-frame DNA-binding domains, and the presence of alternative promoters (Figure 3a).Thus, CREMα is only one of >50 isoforms within the CREM family of transcription factors (Figure 3b) that share a high degree of sequence homology, especially of their DNA binding domains [70,71].Important representatives of the CREM family, in addition to CREMα, include the inducible cAMP early repressor (ICER), cAMP responsive element binding protein (CREB) and the CREM/activating transcription factors (ATFs).Through their basic leucine zipper DNA binding domains, CREM regulatory proteins recruit to palindromic DNA sequences, so-called cAMP responsive elements (CRE), containing TGACGTCA sequences or CRE 'half-sites' (TGACG or CGTCA) [72].Because the human genome contains approximately 750 000 CREs [73], CREM transcription factor proteins have the potential to affect the expression of a multitude of genes.However, the vast majority of CREs are not accessible due to DNA methylation [74,75].
As suggested by their name, CREM family transcription factors are activated in response to intracellular cAMP which is regulated by a host of extracellular signals, such as growth factors or hormones that bind to transmembrane receptors, causing adenylate cyclase to produce large quantities of cAMP [76].In turn, cAMP After activation by serine 133 phosphorylation, CREB recruits to CREs and, upon recruitment of the transcriptional co-activators CREB-binding protein (CBP) and p300, induces trans-regulatory events [78].It (for the most part) promotes an anti-inflammatory response and is a key factor for maintaining immune homeostasis in T cells.However, CREB is also involved in promoting T helper cell activation and differentiation.For example, CREB plays a critical role during effector CD4 + Th17 differentiation, and selective deletion of CREB in T cells impairs Th17 cell differentiation [79] and inhibits the survival of TGF-β-induced regulatory T cells [79].Effector Th17 cells play an important role in resolving infections, and their dysregulation is a hallmark of several autoimmune diseases.Because CREB and CREM compete for the same binding, while having different (sometimes opposing) functions, their tight timely and temporally regulated balance is important for immune homeostasis and function.
Furthermore, in juvenile idiopathic arthritis (JIA), CREMα is overexpressed in synovial fluid T cells and incubation of healthy donor PBMCs with JIA-derived synovial fluid results in an upregulation of FoxP3 and IL-17 production [27].Because effects were reversed by treatment with CREM-specific siRNA, a role for CREMα in mediating inflammation in JIA is beyond doubt.

REGULATION OF CREMα IN T CELLS
The transcription of CREMα is induced following stimulation of the CD3 and CD28 co-receptors [84] and/or oestrogen receptor signalling [85].As its name indicates, CREM proteins are activated in response elevated cAMP levels, and phosphorylation of CREMα is mediated by protein kinases, including calcium/calmodulin-dependent protein kinase IV (CaMK4) [23].In T cells from patients with SLE, increased CREMα expression is driven by transcriptional upregulation, and CREMα mRNA expression can be suppressed by an antisense CREM plasmid [86].Elevated CREMα expression involves trans-activation of the CREM promoter by the transcription factor specificity protein-1 (SP-1) [81].An alternative intronic promotor located upstream of the second exon (Figure 3) is under regulatory control by activating protein (AP)-1 [87] which displayed enhanced activity following T cell activation via stimulation with anti-CD3/CD28 antibodies or phorbol 12-myristate 13-acetate (PMA)/ionomycin treatment.However, activation of this secondary promoter as a result of T cell activation appears to be defective in SLE T cells.In contrast to healthy donor T cells, transfection of T cells from SLE patients with promoter constructs did not result in increased promoter activity [87].T cells from SLE patients display elevated basal levels of CREMα, which is a negative regulator of AP-1 family member c-fos [88], which suggests an autoregulatory feedback mechanism between CREMα and AP-1.Lastly, CD4 + T cell subsetspecific CREMα expression profiles exists, with elevated expression observed in effector memory (CD45RA À CCR7 À ) CD4 + T cells when compared to central memory (CD45RA À CCR7 + ) and naïve (CD45RA + CCR7 À ) CD4 + T cells [68].In this context, epigenetic contributors to CREM expression has been suggested, with 32 CpG sites within the proximal 500 bp of its promoter region, where methylation results in distinct downregulation of CREMα expression [68], and out-of-context expression of DNMT3a potently reduced CREMα levels in primary human T cells.In CD4 + T cells from SLE patients, CREMα overexpression is partially regulated via H3K4me3 at the promoter region, which is controlled by increased SET domain containing 1 (Set1) enrichment [89].Notably, two studies showed that SLE disease activity correlates with CREM promoter activity [68,81], highlighting its potential role as a direct disease biomarker.Single nucleotide polymorphisms (SNPs) in the CREM gene have also been shown to associate with SLE, and following stratification by clinical features one such SNP (rs2295415) also associated with anti-Smith antibody positivity, but showed a potential protective effect against renal disease [90].Abovementioned protein kinase CaMK4 is required during Th17 cell differentiation.Naïve T cells display elevated levels of CaMK4 during stimulation with Th17-inducing cytokines, and CaMK4-deficient mice fail to produce IL-17A [23].Reduced activation of CREMα in CaMK4-deficient cells coincided with increased CpGmethylation of the IL17A gene in humans and mice [23].

CREM-Α AND ICER PROMOTE EFFECTOR T CELL PHENOTYPES
The role of CREMα in inflammation is complex.While it suppresses transcription of certain genes, it promotes the expression of others (Figure 4).The most in-depth studied cytokines regulated by CREMα are immuneregulatory IL-2 and pro-inflammatory IL-17A.A tightly controlled balance between IL-2 and IL-17A is crucial for effective pathogen clearance without overshooting tissue damage [91].Distinctly imbalanced cytokine expression is a hallmark of T cell-mediated autoimmune/inflammatory diseases, including SLE [92] and psoriasis [12].In both, cytokine imbalance has been (in part) linked to CREMα.

Suppression of immune regulatory IL-2 expression
As mentioned above, CD4 + T cells from patients with SLE and psoriasis exhibit increased expression of CREMα.In SLE, CREMα expression is specifically enhanced in effector memory (CD45RA À CCR7 À ) CD4 T + cells, which fail to produce IL-2 but abundantly express IL-17A [68].
Indeed, elevated CREMα expression in T cells coincides with its increased recruitment to the IL2 promoter [82,83,93].Whereas antigenic stimulation of T cells derived from healthy individuals result in CREB binding to the IL2 promoter, trans-activation and IL-2 expression, CREM binds to the IL2 promoter in SLE T cells even without stimulation T cell receptor stimulation resulting in trans-repression [83].Upon T cell activation in SLE, CREM appears to competitively replace CREB at the IL2 promoter [84].Because both CREM and CREB are activated by cAMP while having opposing effects on IL2 transcription, a chronic disruption of the balance between them will hinder the ability to maintain T cell homeostasis.Treatment of T cells isolated from healthy individuals with SLE-derived serum-results in increased CREMα and subsequently reduced IL-2 expression, which was linked to anti-TCR/CD3 IgG autoantibodies and increased receptor signalling [93].
Notably, CREB expression in T cells does not differ between SLE and healthy controls [82], suggesting that blockade of CREMα may ameliorate altered IL-2 repression.Notably, available treatments have the potential to modulate CREMα expression, for example, high doses of corticosteroids [82].
Given the link between increased expression of CREMα in T cells from patients with autoimmune disease, significant efforts have been made to study the molecule in vivo.In mice, selective overexpression of CREMα in T cells has been achieved, which confirmed enhanced binding of CREMα to the IL2 promoter, decreased IL-2 production and T cell proliferation [94].Crucially, this did not affect the expression of CD4 + CD25 + Foxp3 + regulatory T cells as well as other T cell subpopulations.To analyse functional effects of CREMα in lupus-prone mice, a Fas mutation was introduced in the transgenic strain.In this system, CREMα overexpression accelerated lymphadenopathy and splenomegaly and the numerical expansion of DN T cells [95].Notably, an increase in IFN-γ production and decrease in Treg proportion was observed which could both be reversed through treatment with IL-2 [95].
In addition to trans-repression of the IL2 promoter, CREMα inhibits IL-2 expression in SLE T cells through the induction of two epigenetic events, histone deacetylation and CpG DNA methylation [22].In healthy individuals, activation of naïve CD4 + T cells through CD3/28 stimulation results in histone H3K18 hyperacetylation and H3K27 hypomethylation that result in an increased IL-2 expression.However, CD4 + T cells from SLE patients display reduced H3K18 acetylation and increased H3K27 methylation, which is mediated by co-recruitment of histone deacetylase (HDAC)1 with CREMα [22].This coincides with increased CpG methylation near the IL2 gene in SLE patients' CD4 + T cells that is mediated by corecruitment of DNMT3a.
Lastly, CREMα also indirectly reduces IL-2 expression as it recruits to the proximal promoter region reducing the expression of the c-fos proto-oncogene.Reduced c-fos protein in turn results in altered AP-1 activity, thereby affecting IL2 transcription that is regulated by multiple upstream AP-1 binding sites [96].

Enhanced expression of pro-inflammatory IL-17
Dysregulation of IL-17 producing Th17 T cells has been described in several autoimmune disorders [97].
Thus, IL-17A or its receptor are targets of recently licenced biologic anti-inflammatory treatments, for example, secukinumab, ixekizumab, bimekizumab and brodalumab [32].In T cells from SLE patients, CREMα binds to CREs within the IL17A proximal promoter resulting in increased IL-17A expression through trans-activation and the induction of epigenetic remodelling [25].Overexpression of CREMα in T cells in a transgenic mouse model also resulted in increased IL-17A production in response to stimulation with acnti-CD3 and anti-CD28 antibodies ex vivo [94].CREMα instructs epigenetic opening of the IL17 gene cluster, which includes both IL17A and IL17F isoforms.While CREMα recruits to the promoters of these genes, it is not able to recruit DNMT3a and HDAC1 to induce DNA methylation and instead drives demethylation and histone acylation [24,25,68].The exact mechanisms involved are elusive, but a plausible explanation involves the recruitment of histone acetyltransferase p300 as proximity ligation assay indicates a physical interaction between p300 and CREMα in T cells from both healthy individuals and SLE patients [98].Conversely to its effects on the IL17A, CREMα reduces transcription of IL17F through trans-repression of its promoter, independent of activating epigenetic marks [21].At first, these observations may appear counterintuitive as both isoforms IL-17A and IL-17F are pro-inflammatory in nature and capable of forming both homo-and hetero-dimers [99].
While the IL-17F isoform is typically more abundant, the IL-17A isoform is more potent in promoting inflammation.Thus, a reduction of IL-17F may therefore steer cells towards an increased IL-17A to IL-17F ratio with an increased inflammatory phenotype [21].
In addition to direct effects on the IL2 and IL17 genes, CREMα also indirectly affects cytokine expression.CREMα induces the dual specificity protein phosphatase (DUSP)4 in effector CD4 + T cells through co-recruitment of p300 [100].Histone acylation at DUSP4 is mediated by co-recruitment of p300.Notably, DUSP4 regulates the balance between phosphorylated STAT3 and STAT5, thereby impacting on the balance between IL-17A (depending on STAT3) and IL-2 (depending on STAT5) [101].This likely contributes to the altered balance between IL-17A expressing effector and IL-2 expressing regulatory T cells in SLE and likely other autoimmune/inflammatory diseases [102][103][104].
An alternative CREM isoform utilizing an alternative promoter is inducible cAMP early repressor (ICER) [105].ICER is required for Th17 differentiation of CD4 + T cells.In vitro Th17 differentiation is impaired in ICERdeficient cells, but can be restored after forced expression of ICER [80].ICER recruits to the IL17A promoter where it drives the accumulation of canonical enhancer ROR-γt resulting in trans-activation.Notably, ICER is expressed at increased levels in CD4 + T cells from SLE patients where it promotes IL-17A expression [80].
ICER is furthermore involved in immune metabolic processes.In vitro experiments have shown that ICER deficient T cells display reduced activity of oxidative phosphorylation rates in the presence of glutamine and reduced glutaminase 1 expression, both of which could be restored by ICER overexpression, and that direct binding occurs between ICER and the glutaminase 1 promoter which increases its activity [106].Because Th17 (but not Th1, Th2 or Treg) differentiation was dependent on glutaminolysis and expression of glutaminase 1, this highlights a molecular link between ICER and immune metabolism.Further control of ICER-mediated regulation of Th17 differentiation occurs via its modulation of pyruvate dehydrogenase phosphatase catalytic subunit 2 (PDP2) [107], an enzyme that activates pyruvate dehydrogenase, which plays a role in the glycolytic pathway.ICER-deficient mice display increased pyruvate dehydrogenase activity in their Th17 cells, and ICER was found to bind the PDP2 promoter and suppress expression.
The generation of effector 'double negative'(DN) CD3 + TCR + CD4 À CD8À T cells In both SLE and autoimmune lymphoproliferative syndrome, numerical expansion of DN T cells has been described, contributing to increased expression of proinflammatory cytokines and tissue damage [51,108,109].CREMα contributes to transcriptional silencing of both CD8A and CD8B genes in CD8 + T cells, mediating downregulation of surface CD8 expression and the generation of DN T cells [20].Mechanistically, CREMα instructs epigenetic remodelling of the CD8 gene cluster through corecruitment of DNMT 3a and histone methyltransferase G9a [19].Thus, in T cell-mediated autoimmune/ inflammatory diseases characterized by increased numbers of pathogenic DN T cells, CREMα may be a potential therapeutic target.

Dysregulation of additional immuneregulatory molecules
IL-10 is an immune regulatory cytokine.In addition to several anti-inflammatory effects, it has been linked with B cell activation, immunoglobulin class switch and increased immunoglobulin production [110].In SLE, IL-10 expression is increased in CD4 + T cells, and therapeutic blockade of IL-10 showed promising results in a small cohort of otherwise treatment-refractory SLE patients [111].At the IL10 gene, recruitment of STAT3 and STAT5 mediates trans-activation and epigenetic remodelling through interactions with the transcriptional co-activator p300 [112].Competitive replacement of STAT5 with STAT3, which is promoted by the abovementioned CREMα-mediated expression of DUSP4, results in increased IL-10 expression in SLE [101].In CREMα transgenic mice, CREMα binds to a CRE half-site within the IL21 promoter resulting in enhanced promoter activity and gene expression [113].In these mice, IL-21 was crucial for IL-17 expression, and blockade of the IL-21 receptor resulted in reduced IL-17 transcription [113].
A central player involved in T cell-mediated immune homeostasis [114,115] is the inhibitory surface coreceptor Programmed Death (PD)-1 (also called CD279).PD-1 is expressed in activated T cells, as well as DN thymocytes [116][117][118][119][120], and plays a critical role in the limitation of inflammatory responses, including effector cytokine expression [121].Increasing focus has been laid on the role of PD-1 in T cell-mediated autoimmune diseases, such as SLE, rheumatoid arthritis (RA), psoriasis and psoriatic arthritis [122][123][124][125][126][127].Recently, a functional link between reduced PD-1 and increased CREMα expression was demonstrated in genetically modified CD4 + T cells and primary human CD4 + T cells isolated from patients with psoriasis.CREMα recruits to the PDCD1 gene (encoding for PD-1) where it induces DNA methylation through co-recruitment of DNMT3a [18].

FUTURE DIRECTIONS
While CREMα and the short ICER isoform have been established as a key regulator of T cells homeostasis, there are still several unanswered questions.
While the transcription factor microenvironment at specific loci appears to be involved, it remains lastly unclear why CREMα has trans-activating effects at some and trans-repressing effects at other promoters.The same applies for its interaction with epigenetic activators and repressors.Comprehensive unbiased molecular dissection of the CREMα interactome in general and at specific genomic sites will be invaluable.
Although abnormal CREMα expression and associated molecular events have been described in SLE and psoriasis, larger prospective cohort studies are needed to validate findings and test its ability to measure disease activity and/or predict flares.
Currently available animal models for the investigation of in vivo effects of CREM isoforms are limited.No CREMα specific knock-out animal model exists, and CREM-deficient mice exhibit breeding issues [128].CREMα-deficient and CREMα overexpressing Jurkat T cells are available [100] but are limited by their origin (immortalized T cell lymphoma line).Knowing limitations of current and future cell and animal models, they will provide insights into potential therapeutic targeting of CREMα (or other isoforms), for example, with small molecules.

CONCLUSIONS
The transcription factor CREMα is a key regulator of tissue-specific immunity.Several effector T cell-mediated diseases display altered CREMα expression that affects multiple cellular functions.Mechanistically, CREMα directly and indirectly modulates the expression of pro-and anti-inflammatory cytokines through transregulation of genes and the induction of epigenetic remodelling.Although some questions regarding its function currently remain unanswered, CREMα likely presents a promising biomarker candidate and/or therapeutic target in effector T cell mediated autoimmune/inflammatory diseases.
F I G U R E 2 DNA methylation and nucleosome structure.(a) Structure of cytosine, 5-methylcytosine, and 5-hydroxymethylcytosine, and their dynamic conversion via DNA methyltransferases (DNMTs) and ten-eleven translocation (TET) family enzymes.(b) Nucleosome arrangement of densely packed heterochromatin, characterized by a high degree of DNA methylation and repressive histone modifications.Transcriptionally active euchromatin is characterized by relaxation of the nucleosome structures with reduced DNA methylation and permissive histone modification, allowing access of regulatory transcription factors and gene expression.stimulates the enzymatic capabilities of protein kinases including protein kinase A (PKA), protein kinase C (PKC) and casein kinases I and II, which can phosphorylate and thereby activate CREB/CREM proteins.The cAMP-responsive element binding protein (CREB) is ubiquitously expressed in T lymphocytes and other immune cells.It is activated through a variety of immune-related factors, after which it induces the expression of several pro-and anti-inflammatory cytokines (including IL-2, IL-4, IL-10 and IFN-γ) directly or indirectly through the regulation of other transcription factors (induction of FoxP3, inhibition of NFkB) [77].

3
The CREM gene and transcription factor isoforms.(a) Schematic diagram of the CREM gene with exons, including the highly conserved DNA binding domains (DBDs).(b) Multiple isoforms of CREM, including CREMα and ICER which are preferentially expressed by T cells, are generated via the use of alternative promoters and splice mechanisms.