Class II major histocompatibility complex (MHC-II) proteins play a central role in the control of normal immune homeostasis, while aberrant expression of MHC-II is frequently associated with abnormalities in immune responses. MHC-II proteins elicit immune activation through presentation of exogenously derived antigens to CD4 ϩ T cells and represent the seminal control of both peripheral T-cell activation and thymic selection (23,28,47). The level of MHC-II expression is exquisitely regulated. Constitutive MHC-II expression is restricted to B cells, monocytes, macrophages, and dendritic cells, whereas inducible expression is observed on a selected number of cell types in response to cytokines such as gamma interferon (IFN-␥) and tumor necrosis factor alpha (TNF-␣) (37, 47). The regulation of MHC-II expression resides predominantly at the transcriptional level and is globally controlled by the master regulator, class II transactivator (CIITA) (12, 47).CIITA was initially isolated by complementation cloning, using an Epstein-Barr virus-based library to rescue MHC-II expression in MHC-II-negative cells (45). CIITA is encoded by the MHC2TA gene, deletions in which represent the genetic defect in immunodeficient type II group A bare lymphocyte syndrome patients. Expression of CIITA is controlled by four distinct promoters, allowing for a complex pattern of constitutive and inducible MHC-II expression (31, 39). CIITA does not bind DNA but controls MHC-II and related genes by interacting with the requisite MHC-II transcription factors (RFX5, CREB, and NF-Y), which associate with conserved promoter motifs, termed X1, X2, and Y, respectively (9,26,29,42,58). These interactions are critical for the formation of a stable enhanceosome. CIITA also interacts with components of the basal transcription machinery (TFIIB, TATA binding protein, and TATA binding protein-associated factors) (6,25,27). Most relevant to this work, CIITA associates with several chromatin remodeling enzymes, including histone acetyltransferases (HATs) CBP/p300, and pCAF (16,43,44,59), and ATPdependent remodeling factors, such as BRG-1 (30, 38). These enzymes have all been demonstrated to modulate MHC-II promoter activation.Structure-function analysis of CIITA protein indicates that it can be divided into three important segments. The N terminus contains an acidic transactivation domain as well as target lysines for both acetylases and a HAT-like domain (16,40,44). The mid-section contains a nucleotide-binding domain (NBD) that is critical for nuclear import and contributes to self-association (10,17,21). The C terminus contains a stretch of leucine-rich repeats (LRRs) that are also involved in proteinprotein association (11,21). This unique combination of the NBD and LRR domains is a conserved feature among a new family of known and novel genes, which we have recently called the CATERPILLER family (11). The NBD domain is also shared by a more loosely related family of known genes, called the NACHT family. Members of this family range from plant

Although increasing evidence indicates that there is a direct link between ubiquitination and mono-ubiquitination and transcription in yeast, this link has not been demonstrated in higher eukaryotes. Here we show that the major histocompatibility complex (MHC) class II transactivator (CIITA), which is required for expression of genes encoding MHC class II molecules, is ubiquitinated. This ubiquitination enhanced the association of CIITA with both MHC class II transcription factors and the MHC class II promoter, resulting in an increase in transactivation function and in the expression of MHC class II mRNA. The degree of CIITA ubiquitination was controlled by histone acetylases (HATs) and deacetylases (HDACs), indicating that the crucial cellular processes mediated by these enzymes are linked to regulate transcription. Thus, ubiquitin positively regulates a mammalian coactivator by enhancing its assembly at the promoter.

The presence of the class II transactivator (CIITA) activates the transcription of all MHC class II genes. Previously, we reported that deletion of a carboxyl-terminal nuclear localization signal (NLS) results in the cytoplasmic localization of CIITA and one form of the type II bare lymphocyte syndrome. However, further sequential carboxyl-terminal deletions of CIITA resulted in mutant forms of the protein that localized predominantly to the nucleus, suggesting the presence of one or more additional NLS in the remaining sequence. We identified a 10-aa motif at residues 405–414 of CIITA that contains strong residue similarity to the classical SV40 NLS. Deletion of this region results in cytoplasmic localization of CIITA and loss of transactivation activity, both of which can be rescued by replacement with the SV40 NLS. Fusion of this sequence to a heterologous protein results in its nuclear translocation, confirming the identification of a NLS. In addition to nuclear localization sequences, CIITA is also controlled by nuclear export. Leptomycin B, an inhibitor of export, blocked the nuclear to cytoplasmic translocation of CIITA; however, leptomycin did not alter the localization of the NLS mutant, indicating that this region mediates only the rate of import and does not affect CIITA export. Several candidate nuclear export sequences were also found in CIITA and one affected the export of a heterologous protein. In summary, we have demonstrated that CIITA localization is balanced between the cytoplasm and nucleus due to the presence of NLS and nuclear export signal sequences in the CIITA protein.

RGS10 is an important regulator of cell survival and chemoresistance in ovarian cancer. We recently showed that RGS10 transcript expression is suppressed during acquired chemoresistance in ovarian cancer. The suppression of RGS10 is due to DNA hypermethylation and histone deacetylation, two important mechanisms that contribute to silencing of tumor suppressor genes during cancer progression. Here, we fully investigate the molecular mechanisms of epigenetic silencing of RGS10 expression in chemoresistant A2780-AD ovarian cancer cells. We identify two important epigenetic regulators, HDAC1 and DNMT1, that exhibit aberrant association with RGS10 promoters in chemoresistant ovarian cancer cells. Knockdown of HDAC1 or DNMT1 expression, and pharmacological inhibition of DNMT or HDAC enzymatic activity, significantly increases RGS10 expression and cisplatin-mediated cell death. Finally, DNMT1 knock down also decreases HDAC1 binding to the RGS10 promoter in chemoresistant cells, suggesting HDAC1 recruitment to RGS10 promoters requires DNMT1 activity. Our results suggest that HDAC1 and DNMT1 contribute to the suppression of RGS10 during acquired chemoresistance and support inhibition of HDAC1 and DNMT1 as an adjuvant therapeutic approach to overcome ovarian cancer chemoresistance.

No abstract