Kruppel-associated box (KRAB) proteins in the adaptive immune system

Abstract

The ability of adaptive immune system to protect higher vertebrates from pathogens resides in the ability of B and T cells to express different antigen specific receptors and to respond to different threats by activating distinct differentiation and/or activation pathways. In the past 10 years, the major role of epigenetics in controlling molecular mechanisms responsible for these peculiar features and, more in general, for lymphocyte development has become evident. KRAB-ZFPs is the widest family of mammalian transcriptional repressors, which function through the recruitment of the co-factor KRAB-Associated Protein 1 (KAP1) that in turn engages histone modifiers inducing heterochromatin formation. Although most of the studies on KRAB proteins have been performed in embryonic cells, more recent reports highlighted a relevant role for these proteins also in adult tissues. This article will review the role of KRAB-ZFP and KAP1 in the epigenetic control of mouse and human adaptive immune cells.

Keywords: :

• epigenetic
• KRAB-ZFP
• KAP1
• T cells
• B cells
• mouse
• human

Epigenetic Control of the Adaptive Immune System

Two pathways of immune responses against foreign invaders characterize higher vertebrates: the innate and the adaptive immune systems. Innate immunity is the first line defense; it is mainly, but not only, mediated by the myeloid lineage of the hematopoietic compartment and relies on the direct recognition of pathogen associated molecules by specific receptors expressed on all innate immune cells. Upon engagement of their receptors, innate immune cells are able to directly and quickly neutralize the pathogen by different means. Adaptive immunity is the second line of defense; it is mainly mediated by the lymphoid (T and B) lineage of the hematopoietic compartment and relies on the recognition of pathogen-associated antigens by specific receptors (T Cell Receptor, TCR, and B cell Receptor, BCR) expressed on adaptive immune cells (each one having a different antigen specificity). Upon engagement of these receptors, the few antigen-specific cells can expand and amplify the signal by activating a cascade of subsequent events, which eventually eliminate the pathogen.

Adaptive immune cells are highly specialized cells characterized by (1) the ability to distinguish a danger (i.e., pathogen antigens) from a false alarm (i.e., self or food antigens) and react accordingly (response or anergy and/or tolerance), (2) a very high level of response/differentiation plasticity (i.e., different cell differentiation in response to different threats), and (3) the ability to persist after clearance of infection and respond faster in case of re-encountering of the same antigen (immunological memory) (see Fig. 1 for a schematic of adaptive cell differentiation and function). As a further peculiarity, together with the strict and temporal control of gene expression needed by all differentiation pathways, adaptive immune cell differentiation relies on the unique event of genomic DNA rearrangement at the antigen receptor loci in order to ensure maturation of clones with different antigen specificity. These features of adaptive immune cells are ensured at the molecular level by the cross talk between external stimuli and intrinsic cues during development and activation. In the recent years, epigenetics has been proposed as main hub for the integration of these signals.

Figure 1. Epigenetic control of adaptive immune cell differentiation and function. Adaptive immune system consists in the cellular (T cell mediated) and humoral (B cell mediated) arm. Differentiation of adaptive immune cells occurs initially in primary lymphoid organs (thymus for T cells and bone marrow, BM, for B cells), where each progenitor cell (Pre-T and Pre-B) rearranges its antigen receptor loci at the DNA level in order to express a different receptor. In these organs only cells expressing a functional receptor not recognizing self-antigens receive anti-apoptotic stimuli (positive selection) and survive from negative selection (depletion of autoreactive clones). Upon rearrangement and differentiation effector and regulatory mature cells (matT helper, matTh, or matT cytoxic, matTc, if expressing CD4 or CD8 as TCR co-receptor, respectively, matB and thymic derived CD4+ T regulatory, tTreg) migrate to secondary lymphoid organs (mainly spleen and lymphnodes), where naïve (nTh, nTc, nB, and tTreg) cells can encounter an antigen (Ag). Ag-TCR or Ag-BCR interaction induces pattern of expressions leading to the generation of activated (and thus functional) immune cells. TCR recognizes Ag only if processed and presented by antigen presenting cells (APC) through Major Histocompatibility Class (MHC) molecules, while BCR recognizes free Ag. Depending on the type of Ag and/or the MHC presenting it, naïve cells differentiate in different subsets with different functions: (1) nTh differentiate mainly in Th1, Th2, Th17 cells that produce different cytokines inducing different B cell- mediated or myeloid response to extracellular pathogens; (2) in case of non-harmful Ag, nTh differentiate into T regulatory cells (peripheral, pTreg) that, together with tTreg cells, induce tolerance and dampen immune response through different mechanisms; (3) nTc differentiate into Cytotoxic T Lymphocyte (CTL) that mediate killing of the infected target cells; (4) nB cells differentiate into activated B (actB) cells and this process requires antibody maturation and differentiation through class switch recombination (CSR) and somatic hyper mutation (SHM), respectively; actB cells further differentiate in plasma cells (PC) that are factories for the secretion of antibodies (Ab). Most of the expanded activated cells undergo apoptosis after pathogen clearance while a small fraction persists in the body as memory cells (mTh, mTc, and mB) and mediate faster secondary response in case of re-encountering of the Ag. The adaptive immune cell processes controlled at the epigenetic level are highlighted in stars with relative references.

Epigenetics collectively defines inheritable post-translational modifications of the chromatin components (DNA and histones) that are not directly dictated by the underlying DNA sequence. These modifications result in changes in the compaction status and nuclear localization of chromatin, and ultimately govern gene expression patterns. Histone acetylation is usually associated with transcription permissive chromatin, with acetylated lysine 9 and lysine 27 on histone 3 (H3K9Ac and H3K27Ac, respectively) among the best characterized permissive marks. With some exceptions, methylated histones are instead associated with non-permissive chromatin, such as H3K9me and H3K27me.^1-3

Control of T and B cell lineage specification has been originally ascribed to the activity of the so-called lineage-determining factors (i.e., transcription factors whose expression is sufficient and necessary for the determination of a given cell fate). Recently, the activity of these factors has been coupled to chromatin modifiers, and epigenetic modifications have been shown to play key roles in the control of hematopoietic cell differentiation and function.⁴ In particular, T cell differentiation in the thymus has been linked to progressive chromatin condensation and shown to be controlled by the differential expression of master transcription factors recruiting histone deacetylases, acetyl transferases, methyltransferases, and components of the Polycomb and NuRD complexes at specific genomic loci.⁵ In a very similar manner, B cell fate is determined by the interaction of key transcription factors with epigenetic modifiers that alters the epigenetic landscape and regulates the gene expression pattern of the differentiating cells.^6,7 Furthermore, the specific abilities of adaptive immune cells of rearranging DNA at the antigen receptor loci and “remembering” a stimulus (memory) are strictly controlled at the epigenetic level. Antigen receptor rearrangement (DNA recombination of DNA stretches encoding common and variable domains of the receptor) relies on both chromatin modifiers for the recruitment and activation of the recombinase enzymes, and chromatin conformation and nuclear structure to put genetic elements distributed along large genomic distances in proximity.^8,9 These data strongly supported the idea that lymphoid cell differentiation is governed by the ability of master transcription factors and chromatin modifying enzymes to induce activation of lineage specific genes and repression of alternative lineage-related ones. As occurring in several differentiation pathways, this might lead to an irreversible epigenetic landscape with mutually exclusive epigenetic modifications at the lineage-determining transcription factor loci (i.e., open chromatin conformation at the specific lineage factor and closed conformation at the alternative-lineage factors).¹⁰ Nevertheless, the development of high-throughput sequencing techniques has, at least partially, disproved this scenario. By assessing genome-wide chromatin modifications and gene expression, chromatin landscape in mature T and B cell subsets has been found far more dynamic than previously expected, showing “bivalent” states (i.e., loci marked by open and closed chromatin marks together) and fast changes in epigenetic modifications in response to external stimuli both in human and mouse cells.^11-13

As expected from the described role of epigenetics in regulating virtually all the aspects of the adaptive immune system, defects in chromatin modifiers and aberrant epigenetic control have been linked to human immune-related diseases. In particular, alteration in epigenetic regulation seems to concur with genetic predisposition in the pathogenesis of autoimmune disorders, such as systemic lupus erythematosus, rheumatoid arthritis, systemic sclerosis, and type 1 diabetes.¹⁴ Studies performed in monozygotic twins non-concordant for the development of the disease showed global DNA and/or histone hypomethylation together with locus specific DNA hypomethylation of genes associated to activation and/or response of adaptive immune cells in affected patients.¹⁴ Similarly, altered epigenetic regulation at key immune genes during development has been proposed as functional link between environmental exposures and chronic inflammation leading to allergies.¹⁵ Although the molecular mechanisms underlying this observation are still unclear, several mutations in chromatin modifying molecule expressing genes have been also associated to leukemia development.¹⁶

Overall, all of these studies assessed the pivotal role of epigenetics in controlling most of the molecular mechanisms that regulate adaptive immune cell features, including their peculiar high level of plasticity.

KRAB-ZFP and KAP1

Krüppel-associated box zinc finger proteins (KRAB-ZFPs or ZNFs) constitute the widest family of tetrapod-specific transcription repressors, which underwent a marked expansion by gene and segment duplication during evolution.^17-20 KRAB-ZFPs are characterized by tandem repeats of C2H2 zinc fingers at the C-terminus, which confer them with the ability to bind specific polynucleotidic sequences, and one or two KRAB domains at the N-terminus, responsible for recruiting KRAB-associated protein 1 (KAP1) (Fig. 2).^21-23 KAP1 is the so far only described co-factor of KRAB-ZFPs, is a ubiquitously expressed member of the tripartite motif-containing (TRIM) family, and is also known as TRIM28, TIF1β, or KRIP1. It encodes a TRIM/RBCC motif (RING finger, B box, coiled coil), plant homeodomain finger and bromodomain and functions as a strong transcriptional repressor when bound to DNA.^24-26 It acts as scaffold that recruits chromatin modifiers including the SETDB1 histone methyltransferase, the CHD3/Mi2 component of the NuRD complex and Heterochromatin Protein 1 (HP1). These KAP1-mediated complex leads to heterochromatin formation by histone 3 tri-methylation on lysine 9 (H3K9me3) and histone deacetylation (Fig. 2).^27-29 The rather advanced characterization of the biochemical mechanism of KRAB-ZFP/KAP1 action contrasts with our large ignorance of the physiological roles of this system, in particular in adult tissues. KRAB-ZFP genes are evolutively recent and their expansion in the tetrapod genome strongly suggests that new functions for the encoded proteins have been generated under the selective pressure of newly acquired biological pathways. It is tempting to speculate that one of these pathways may well be the adaptive immune system, as typical instance of acquisition of new specialized functions in higher vertebrates.

Figure 2. Function(s) of the KRAB proteins. Schematic of the main domains of KAP1 (top) and a prototypic KRAB-ZFP (bottom), and of the main molecules interacting with each of the two (direct interaction is depicted as dashed line). KAP1 domains (N to C-terminal). RING, Really Interesting New Gene zinc finger; B1 and B2, B box zinc fingers; CC, coiled coil. RING, B boxes, and CC form together the RBCC domain, needed by KAP1 to interact with the KRAB A domain of the KRAB-ZFP. PvXvL, hydrophobic pentapeptide needed for interaction with HP1 (Heterochromatin Protein 1); PHD, plant homeo-domain; BR, bromodomain. PHD and BR cooperate in binding the Mi2a/NuRD deacetylase complex and the SETDB1 histone 3 lysine 9 methyl transferase. Once recruited by KAP1 these two histone-modifying factors are able to induce heterochromatin formation by modifying histone 3 (H3) tail. KRAB-ZFP domains (N to C-terminal). KRAB A, Krüppel-associated box A. It is a transcriptional repressor module present in all KRAB-ZFP and mediates KAP1 recruitment. The second KRAB box is facultative and can be B or b depending on the primary structure of the sequence. Zinc fingers are present in tandem repeats and can vary from 2 to 36 in the family; they are able to bind specific DNA sequences. KAP1 has not a DNA-binding domain; KRAB-ZFP is thus the linkage between specific DNA stretches and KAP1-mediated complexes. Listed in the pop-ups the roles of the depicted protein or histone modification as assessed in the mouse (in blue) and in the human (in orange) adaptive immune system; in brackets the relative references.

Several studies on the KRAB-ZFP/KAP1 system have been focused on the mouse embryo and embryonic stem cell (ESC) biology. They have shown an essential role for KAP1 in embryonic differentiation and morphogenesis, establishment and/or maintenance of genomic imprinting and pluripotency and/or self-renewal maintenance in embryonic stem cells, where it also represses endogenous and exogenous retroviruses.^30-37 The KRAB-ZFPs mediating KAP1 action in these processes have been identified as zfp809, zfp568 (chato), and zfp57 for the control of exogenous retroviruses, extra-embryonic tissue development, and genomic imprinting, respectively.^38-41 Zfp819 has been proposed very recently as player in the control of endogenous retroelement in ESC.⁴² Also gametogenesis seems to be controlled by the KRAB-ZFP/KAP1 system, with KAP1 needed for spermatogenesis and KRAB ZNF899 (Prdm9) playing a key role in meiotic recombination and hybrid sterility.^43-46

In human adult tissues, consistent studies have demonstrated the role of the KRAB-ZFP/KAP1 system in vital cellular pathways.⁴⁷ They have found that during DNA damage response, Ataxia Telangiectasia Mutated- (ATM-) mediated KAP1 phosphorilation is required for chromatin relaxation and recruitment of the DNA repairing complex at the damage site.^48-50 Also, KRAB-ZFP/KAP1 system plays a double role in neoplastic transformation. On one side, the KRAB ZNF350 (ZBRK1) partakes in human cancer development by regulating the expression of oncogene and/or oncosuppressor genes and mediating the DNA damage response controlled by the tumor-suppressor molecule retinoblastoma.^51-54 On the other, KAP1 controls p21 and p53 pathways by different means and its overexpression is associated with several human malignancies.^55-58 Moreover, p53 activity seems to be directly regulated by the KRAB-ZFP ZNF420 (APAK).^59,60 A very recent study has also proposed KRAB ZFP1 as negative regulator of polymerase II-mediated transcription.⁶¹

Regarding the physiology of specific tissue, KAP1 and ZNF746 (PARIS) have been implicated in central nervous system disorders in mice and humans, respectively,^62,63 while KAP1 and the KRAB-ZFP rsl1-2 seem to control mouse liver metabolism and sexual dimorphism, respectively.^64,65 KRAB zfp157 has been proposed as player in controlling proper mammary gland development in the mouse.⁶⁶ Very recently, a prominent role of KAP1 in controlling microRNA expression regulating mouse erythroid differentiation has also been described.⁶⁷

KAP1/KRAB-ZFP in Mouse Lymphoid Cell Development

The role of KAP1 and KRAB-ZFPs in the mouse and human adaptive immune systems is far from being clearly assessed. Nevertheless, the study of mouse models conditionally knocked out for the KAP1 gene or genes encoding for its interacting partners has put forth a relevant role for this system in several aspects of the adaptive immune response.

B cell lineage specific Kap1 KO mice have been developed by crossing the CD19^Cre/+, in which the recombinase is expressed at the pro-B stage, with the kap1^fl/fl strain.^30,68 A first study by Reina-San Martin group has found an interaction between kap1 and Activation-Induced cytidine Deaminase (AID) in mouse B cells and used cells from KO mice to study the molecular relevance of this interaction.⁶⁹ AID is the enzyme required for initiating class switch recombination (CSR) and somatic hypermutation (SHM) in B cells.^70,71 SHM and CSR are two processes needed for antibody maturation and diversification in activated B cells, and are based on the introduction of point mutations and recombination in the immunoglobulin gene, respectively.^72,73 The epigenetic machinery strictly regulates these pathways by controlling accessibility of the loci to AID, three-dimensional localization of the recombining regions and recruitment of the DNA damage response machinery.⁷⁴ In the above-mentioned study, the authors showed a 50% reduction in CSR in kap1 KO when compared with control B cells in vitro and no defect in SHM. They proposed a role for kap1 in facilitating recruitment of AID to the switch region through the formation of a complex containing HP1 and recognizing H3K9me3.⁶⁹ By analyzing the same mouse model, we have found normal rates of CSR in vivo, probably because of compensating mechanisms masking in vivo the mild defect observed in vitro.⁷⁵ Instead, we have observed a significant defect in mature B cell development, in particular in the non-conventional subset, decreased levels of steady-state antibody production, and faster rates of antibody decays after viral immunization. This indicated a relevant role for KAP1 in B cell differentiation and maturation. By performing gene expression and chromatin precipitation studies, we found the PI3K antagonist PTEN to be directly regulated by kap1 and proposed altered regulation of this gene—known to be a main player in B cell biology⁷⁶—as molecular mechanism underlying (at least part of) the phenotypes observed in kap1 KO mice. Genome wide, we found that kap1 binding sites correlated positively with regions marked by the repressive histone modification H3K9me3 and negatively with B cell-specific regulatory elements bearing active marks (H3K4me1 and H3K4me3) or bound by the PU.1 transcription factor.^75,77,78 Kap1 binding sites seemed also to be associated to facultative rather than constitutive heterochromatin.⁷⁵

Three studies have investigated the function of kap1 in mouse T cells. By using conditional KO mice based on the expression of the cre recombinase under the T cell co-receptor CD4 or the T cell specific kinase lck promoter, we and others have reported kap1 involvement in controlling T cell-mediated response and tolerance.^79,80 The two studies performed gene expression analyses and proposed different molecular mechanisms as origin of the observed defects. Tasuku Honjo’s group proposed that kap1 regulated transforming growth factor pathway in mature T cells and that alteration of this axis led to autoimmune phenotype.⁷⁹ We found profound defects in immature T cell differentiation and proposed kap1-mediated direct control of the FoxO1 gene—encoding for the major regulator of thymocyte transcriptional network—as the main pathway underlying this phenotype.⁸⁰ The same year, another group confirmed the role of kap1 in T cell differentiation and extended it to invariant natural killer T cells, but attributed the phenotype to a role for this molecule in TCR rearrangement.⁸¹ We observed association between kap1, NuRD, and Ikaros binding sites, as expected.^28,82 Nevertheless, our chromatin studies showed an unpredicted landscape for kap1 binding sites in immature T cells. As for B cells, kap1 binding sites were highly enriched in repressive (H3K9me3 and H3K27me3) and depleted of active promoters and/or enhancer associated (H3K4me3, H3K4me1, and H3K9ac) histone marks. Differently from what observed in mature B cells and unexpected for a predicted transcriptional repressor, markers of open chromatin and/or active transcription, such as the histone acetyl-transferase CBP, ETS1 (a critical transcriptional regulator of T cells), TFIIB, and formaldehyde-assisted isolation of regulatory elements (FAIRE), were significantly enriched at kap1 binding sites. Although we could not rule out the possibility of retrieving different complexes containing kap1 in different cells, these data suggested that kap1 binds plastic cis-regulatory regions and may coexist at these sequences with partially assembled and/or weakly bound complexes of transcription and chromatin-remodelling factors in immature T cells. In order to assess if the different features of kap1 binding sites in T and B cells are linked to the different lineages or to the diverse differentiation stages (immature and mature, respectively), further investigation is needed. In line with the hypothesis of the formation of different KRAB complexes in different cellular contests, in non-lymphoid cells several studies have suggested that KRAB proteins may play different functions and even acquire activating transcriptional ability depending on the interacting partners.^83-86

Functional studies about KRAB-ZFPs in the mouse lymphoid system are not available. The only existing data are related to their level of expression in different hematopoietic lineages. Data from ChiP-seq studies also indicated that kap1 binds in proximity of KRAB-ZFP genes in mouse adaptive immune cells,^75,80 suggesting that it could be involved in the regulation of expression or maintenance of genomic stability of these highly repeated sequences as proposed for human cells.^87-91

Further hints about the functional role of kap1 in adaptive immune cell development might be extrapolated from studies performed on kap1-interacting molecules or associated chromatin modifications. In vitro and in vivo data have demonstrated kap1 association with the histone deacetylases NuRD complex, SETDB1 histone methyltransferase, HP1, and H3K9me3.^{27-29,75,80,87}

The role of H3K9me in lymphoid lineage development has been studied by taking advantage of mouse models knockout for the H3K9 methyltransferases (SUV39H1, G9a, and SETDB1). In the first report showing G9a KO mice, no major phenotype in the T cell compartment was observed (T helper specific functions were not assessed). Interestingly, B cell phenotype seemed fully consistent with the one observed in kap1 KO mice, as B cells showed a lower proliferation capacity and defects in rearrangement of the Ig locus and plasma cell maturation.⁹² Two studies showed the importance of H3K9me in T helper differentiation and function. By using G9a KO CD4⁺ cells, Zaph’s group proposed a role for H3K9me2 in controlling interleukin 17 gene locus and T helper function.⁹³ A few years later, Amigorena’s group used Suv39h1 and HP1 KO CD4⁺ cells and showed SUV39H1-HP1-H3K9me3 controlling interferon γ and interleukin 4 gene loci and thus T helper differentiation and stability.⁹⁴ All of these findings suggested that kap1 and H3K9me affect a quite restricted set of processes during lymphocyte development and activation, and this was rather unexpected for molecules with such a broad distribution in the genome.^75,80 Nevertheless, this behavior has been attributed to several histone-modifying factors, such as the MLL/SET H3K4 and the Polycomb H3K27 methyltransferase complexes and the H3K27 demethylase Jmjd3.^95-97 Thus, in analogy with other transcriptional co-regulators, kap1 and H3K9me3 may fine-tune expression rather than acting like an on-off switch of gene expression, at least in the lymphoid system.

A more pronounced role in lymphoid cell physiology seems to be played by the NuRD histone deacetylase complex, whose activity in this compartment seems to be mainly dictated by its interaction with the zinc finger Ikaros, a key regulator of lymphopoiesis.^82,98,99 Different reports have shown that NuRD and/or the Ikaros-NuRD complex regulates key pathways of lymphoid cell biology, including (1) CD8 and CD4 gene expression and T cell development, (2) cytokine loci and T cell immune response, tolerance and anergy, and (3) B cell specific genes and B cell differentiation.^100-106 Direct evidence of binding of kap1 to Ikaros/NuRD in lymphoid cells is still missing. Nevertheless, in vitro interaction with NuRD and in vivo overlapping of kap1 and Ikaros-NuRD binding sites might lead to contemplate kap1 involvement in the control of, at least some, Ikaros-NuRD target genes. Moreover, these data might suggest that kap1 can be recruited to target genes through interaction with other DNA binding factors and independently by KRAB-ZFP mechanisms, as also proposed in other contexts (see also following paragraph).⁸⁸

KAP1/KRAB-ZFP in the Human Immune System

Direct and indirect evidences link KRAB proteins to different aspects of human adaptive immune cell differentiation and function.

The best-characterized function of KRAB-ZFP/KAP1 system in human adaptive immune cells is the control of viral replication. Apparently contrasting results have been published about the role of this system in the control of B cell tropic herpes viruses. These viruses are characterized by two different stages of life cycle (the latent/silent and the lytic/active) that are strictly regulated by activation and/or repression of specific viral gene expression patterns. On one side, ZNF251 (KZLP) has been shown to activate K1 promoter of Kaposi’s sarcoma-associated herpes virus (KSHV) and contrast the LANA KSHV, which is the essential transcription factor to establish latency.¹⁰⁷ Nevertheless, this early report, which was mostly based on biochemical data, did not show binding of the ZFP to viral DNA and did not show a role for this ZFP in the replication of KSHV DNA. Thus, the relevance of this study is not obvious. More consistently, KRAB-ZFP and KAP1 have been linked to repression of KSHV activation and induction of latency. Although there is no evidence for a direct interaction between the two KRAB proteins in this context, ZNF426 (K-RBP) has been shown to directly repress transcription of the KSHV transactivator RTA gene and KAP1 to bind and repress several KSHV lytic genes.^108,109 KAP1 and the KRAB-ZFP ZBRK1 have also been shown to interact with Epstein-Barr herpes virus (EBV), although the role of this complex on EBV life cycle has not been clearly defined.¹¹⁰ Moreover, KRAB-ZFP/KAP1 are apparently involved in the control of Human Immunodeficiency Virus (HIV) replication at different stages. HIV targets CD4⁺ (T and monocyte) cells and is characterized by the ability of integrate into the human genome and establish in some, not fully elucidated, circumstances latent infection. Two KRAB-ZFPs have been proposed to regulate HIV transcription. Early studies have identified ZNF175 (OTK18) as gene upregulated upon HIV infection in monocyte-derived macrophages and shown its role in controlling several cellular genes as well as HIV transcription.^111,112 A more recent study has reported ZBRK1 to bind and control HIV promoter expression by recruiting KAP1.¹¹³ In this work, the authors have also shown that knocking down of ZBRK1 reduced KAP1 binding to HIV promoter by only 25%, supporting the idea that KAP1 binding to HIV sequences might be mediated by more than one (KRAB) factor. Another group has recently put forward a role for KAP1 in controlling HIV integration into the cell host genome.¹¹⁴ This study reported increased integrated viral copies upon KAP1 downregulation and proposed a role in post-translational modification of the HIV integrase protein, rather than epigenetic for KAP1. All of these observations suggest that KRAB proteins conserved the antiviral activity (mainly mediated by the control of expression of viral genes) from mouse to human, although by using different mechanisms in the two systems. Further investigating this issue in human adaptive immune cells might lead to the discovery of new pathways underlying the still obscure process of epigenetic control of viral latency and ameliorate both basic scientific knowledge and antiviral therapies.

As aforementioned, in order for the DNA damage response (DDR) machinery to be recruited at the DNA break point, KAP1 needs to be phosphorylated by ATM to detach and allow chromatin relaxation.^48-50 The DDR pathway is required for proper DNA rearrangement at the TCR and BCR loci and for CSR/SHM at the immunoglobulin locus in B cells. Indeed, although the recombinase proteins acting at the receptor loci are specific for the immature T and B cells, the machinery recruited to the DNA break induced by the recombinases is the canonical DDR. DDR is mediated by the non-homologous end-joining apparatus, which includes the DNA-dependent protein kinase catalytic subunit (DNA-PKcs) and ATM in this context.¹¹⁵ Although neither ATM nor DNA-PK are absolutely required, their combined deficiency results in a block in DNA recombination suggesting that these two kinases have overlapping activities.^116,117 The important role of this machinery in the assembly of antigen receptor genes is highlighted by the observation that mutations in DNA-PK leads to severe combined immune deficiency (SCID), while ATM deficiency causes ataxia-telangiectasia (A-T) disease in humans.^118-120 A-T patients show severe neurodegenerative disorder usually associated with immunological dysfunctions, such as decreased specific immunoglobulin production and CD4⁺ T cell number, increased radiation susceptibility and predisposition to lymphoid cancer.¹²¹ A-T immunological abnormalities have been associated to defective receptor locus recombination, and T and B cell progenitor maturation in mouse models.^122-127 Further investigation will clarify the function of KAP1 in this context confirming or not its relevance put forth by the recent data reporting its prominent role in ATM mediated response.

The activity of transcription factors playing pivotal roles in human lymphocyte biology has also been linked to KRAB proteins. c-Rel/NFkB is a family of transcription factors regulating immune response and inflammation through their control of immune genes, such as interleukins and lineage specific transcription factors.¹²⁸ Upon nuclear translocation NFkB transcriptional activity is controlled by post-translational modifications that regulate its nuclear retention and interaction with co-activators or repressors.¹²⁹ KAP1 has been proposed as negative regulator of NFkB transcriptional activity by interacting with the acetyltransferase p300/CBP and inhibiting acetylation of NFkB at the interleukin-6 inflammatory gene.¹³⁰ This KAP1 activity seemed to be mediated by its interaction with the signal transducers and activators of transcription 3 (STAT3).¹³¹ KAP1 seems also to play a role in the interferon-mediated inflammatory response by interacting with STAT1.¹³² These data are particularly interesting in light of the importance of STAT proteins in shaping immune response. STATs are indeed main activators of cytokine genes involved in T helper (Th), T cytotoxic, and B cell differentiation and/or activation and exert their positive function by recruiting the histone acetyltranferase p300/CBP.^133-135 In light of this, if confirmed, KAP1 interaction and/or regulation of STAT proteins might imply a major role for this protein in the regulation of gene expression patterns of human adaptive immune cells. KAP1 interaction with STAT3 seems particularly interesting in light of the role of the latter in controlling autoimmune manifestations. It has been clearly demonstrated, indeed, the main role of STAT3 in both human and mouse in inducing the expression of cytokine genes leading to Th17 phenotype, which is the main Th subset involved in inflammation during autoimmune manifestations.^136-138 On the other hand, there seems to be discordance, at least in the mouse, about the role of STAT3 in controlling the pattern of expression in Treg cells, which is the T subset counteracting autoimmune manifestations by dampening immune response through different mechanisms.^139,140 The master transcription factor regulating Treg cell specification is FOXP3, whose gene mutation induces a life-threatening autoimmune disorder called IPEX.^141,142 This forkhead-domain factor acts as either activator or repressor of gene expression thanks to its ability to bind gene promoters and cooperate with different component of the epigenetic machinery.¹⁴³ Among several other interactors, recent data indicated that FOXP3 associates with STAT3¹⁴⁴ and KAP1¹⁴⁵ in human Treg cells. FOXP3-KAP1 interaction seemed to be mediated by a human specific KRAB-containing protein derived from the alternative splicing of the ZFP90 and called FIK, and to be necessary for the repressive activity of FOXP3 at target gene loci. Thus, according to these reports and mirroring the mouse system, KAP1 seems to be a player in the maintenance of human immune tolerance through different means. This role of KAP1 in controlling autoimmunity seems highly consistent with the altered H3K9me pattern and HP1 recruitment at key immune genes found in the lymphocytes of patients suffering from several autoimmune diseases.^146-149

Concluding Remarks

Although not many reports have studied the role of KRAB proteins in controlling gene expression patterns in immune cells, several lines of evidence both in the mouse and in humans put forth a relevant role for these proteins in key aspects of the biology of these peculiar cells. Interestingly, KRAB proteins seem to form different complexes depending on the cellular contest (i.e., level of expression of common and cell type-specific transcription factors, post-translational modification of transcription, and chromatin regulators) and this seems to significantly affect the function of these proteins and the resulting chromatin modifications. More focused studies, including high throughput gene expression and chromatin approaches, will clarify the molecular patterns mediated by KRAB proteins and improve the understanding of the complicate epigenetic landscape controlling adaptive immune cell differentiation and function.

Abbreviations:
AID	=	Activation-Induced cytidine Deaminase
A-T	=	Ataxia-Telangectasia
ATM	=	Ataxia-Telangectasia Mutated
BCR	=	B Cell Receptor
CSR	=	Class Switch Recombination
DDR	=	DNA Damage Response
EBV	=	Epstein-Barr Virus
ESC	=	Embryonic Stem Cell
H3K9me3	=	Histone 3 Lysine 9 three-methylation
HIV	=	Human Immunodeficiency Virus
HP1	=	Heterochromatin Protein 1
KAP1	=	KRAB-Associated Protein 1
KRAB	=	Kruppel Associated Box
KSHV	=	Kaposi Sarcoma associated Herpes Virus
NuRD	=	Nucleosome Remodeling Deacetylase
SCID	=	Severe Combined Immune Deficiency
SETDB1	=	SET Domain, Bifurcated 1
SHM	=	Somatic HyperMutation
TCR	=	T Cell Receptor
Th	=	T helper
ZNF	=	Zinc Finger
ZFP	=	Zinc Finger Protein

Disclosure of Potential Conflicts of Interest

No potential conflict of interest was disclosed.

Acknowledgments

Thanks to Didier Trono, Monica Ievolella, Laura Passerini, and Rosa Bacchetta for fruitful discussions. F.R.S.d.S is supported by the Italian Ministry of Health (Grant GR-2010-2317550).

10.4161/nucl.28738