In RFX-B-Deficient Cells Through Exogenous Class II Transactivator Partial Rescue of MHC and Related Genes Novel Mutations Withi

Novel Mutations Within the RFX-B Gene and Partial Rescue of MHC and Related Genes Through Exogenous Class II Transactivator in RFX-B-Deficient Cells This information is current as of October 1, 2021. Uma M. Nagarajan, Ad Peijnenburg, Sam J. P. Gobin, Jeremy M. Boss and Peter J. van den Elsen J Immunol 2000; 164:3666-3674; ; doi: 10.4049/jimmunol.164.7.3666

http://www.jimmunol.org/content/164/7/3666 Downloaded from

References This article cites 58 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/164/7/3666.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on October 1, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Novel Mutations Within the RFX-B Gene and Partial Rescue of MHC and Related Genes Through Exogenous Class II Transactivator in RFX-B-Deﬁcient Cells1

Uma M. Nagarajan,2* Ad Peijnenburg,2† Sam J. P. Gobin,† Jeremy M. Boss,3,4* and Peter J. van den Elsen3†

MHC class II deficiency or bare lymphocyte syndrome is a severe combined immunodeficiency caused by defects in MHC-specific regulatory factors. Fibroblasts derived from two recently identified bare lymphocyte syndrome patients, EBA and FZA, were found to contain novel mutations in the RFX-B gene. RFX-B encodes a component of the RFX transcription factor that functions in the assembly of multiple transcription factors on MHC class II promoters. Unlike RFX5- and RFXAP-deficient cells, transfection of exogenous class II transactivator (CIITA) into these RFX-B-deficient fibroblasts resulted in the induction of HLA-DR ␤ Downloaded from and HLA-DP and, to a lesser extent, HLA-DQ. Similarly, CIITA-mediated induction of MHC class I, 2-microglobulin, and invariant chain genes was also found in these RFX-B-deficient fibroblasts. Expression of wild-type RFX-B completely reverted the noted deficiencies in these cells. Transfection of CIITA into Ramia cells, a B cell line that does not produce a stable RFX-B mRNA, resulted in induction of an MHC class II reporter, suggesting that CIITA overexpression may partially override the RFX-B defect. The Journal of Immunology, 2000, 164: 3666–3674.

ajor histocompatibility complex class II genes encode indirectly to conserved regulatory elements in the proximal pro- http://www.jimmunol.org/ cell-surface glycoproteins that fulfill a crucial function moter of these genes (reviewed in Refs. 6–10). M in the immune response and in T cell development The conserved S, X (comprising X1 and X2), and Y box DNA (1, 2). The importance of these molecules is illustrated in patients sequence regulatory elements that have been identified in the prox- suffering from MHC class II deficiency or bare lymphocyte syn- imal promoter of MHC class II, class I, and their accessory genes drome (BLS),5 a severe combined immunodeficiency. This dis- are bound by several DNA-binding protein complexes (reviewed ease, which is characterized by an absence of MHC class II ex- in Refs. 3, 6, 7, and 11–14). These include RFX, X2BP (cAMP pression and frequently also by a reduced level of MHC class I response element binding protein; CREB), and NF-Y, which bind expression (3), results in the inability to present antigenic peptides cooperatively to the X1 box (8, 15), X2 box (8, 9, 16, 17), and Y by guest on October 1, 2021 to CD4ϩ T cells for initiation of the immune response. BLS pa- box (18, 19), respectively. Together these proteins are engaged in tients manifest severely impaired humoral and cellular immune the formation of a highly stable multiprotein DNA complex (20– responses, as well as impaired development of CD4ϩ T cells (4, 5) 24). Although individual binding of RFX or X2BP to some of the and therefore are extremely susceptible to many opportunistic in- isotypic MHC class II-conserved elements has been shown to be fections (5). BLS is inherited in an autosomal recessive fashion weak or absent (8, 22), the RFX/X2BP/NF-Y complex is formed and results from mutations in the genes encoding transacting reg- on X-Y DNA of all MHC class II isotypic promoters (22). In a ulatory proteins. These transcription factors are critical to the ex- similar fashion, multiprotein complex formation also occurs on ␤ ␤ pression of MHC class II genes and play an ancillary role in the X-Y DNA of the MHC class I and 2-microglobulin ( 2m) pro- transcriptional control of MHC class I genes by binding directly or moters.6 This complex formation allows the MHC class II isotypes to be regulated in a coordinate fashion. The lack of MHC class II expression and the reduction of MHC class I expression in BLS *Department of Microbiology and Immunology, Emory University School of patient-derived cell lines with defects in one of the RFX sub- Medicine, Atlanta, GA 30322; †Division of Molecular Biology, Department of Im- munohematology and Blood Bank, Leiden University Medical Center, Leiden, The units is attributed to the failure to assemble this complex in vivo Netherlands (25, 26).6 Received for publication October 5, 1999. Accepted for publication January 19, 2000. Formation of this multiprotein complex on X-Y DNA is not The costs of publication of this article were defrayed in part by the payment of page sufficient for activation of MHC class II genes. Activation requires charges. This article must therefore be hereby marked advertisement in accordance the activity of the MHC class II transactivator (CIITA) (27). with 18 U.S.C. Section 1734 solely to indicate this fact. CIITA is believed to act as a transcriptional coactivator through 1 This research was supported by grants from the National Institutes of Health to J.M.B. (GM47310 and AI34000) and The Netherlands Organization for Research to protein/protein interactions with the RFX complex and as such P.J.v.d.E. (Grant 901-09-243). S.J.P.G. is a Fellow of the Royal Netherlands Acad- connects the MHC-specific transcription complex with the basal emy for Arts and Sciences. transcription initiation apparatus (28–30). CIITA has strong trans- 2 U.M.N. and A.P. contributed equally to the content of this manuscript. activation properties that are mediated through its acidic NH2-ter- 3 J.M.B. ans P.vD.E. contributed equally to the supervision of this manuscript. minal domain (31, 32). Whereas CIITA is critical for expression of 4 Address correspondence and reprint requests to Dr. Jeremy M. Boss, Department of Microbiology and Immunology, Emory University School of Medicine, 1510 Clifton Road, Atlanta, GA 30322. E-mail address: [email protected] 5 Abbreviations used in this paper: BLS, bare lymphocyte syndrome; CREB, cAMP 6 Gobin, S. J. P., M. Van Zutphen, S. D. Westerheide, J. M. Boss, and P. J. van den ␤ ␤ response element binding protein; 2m, 2-microglobulin; CIITA, class II transacti- Elsen. MHC gene transactivation through the SXY regulatory module is controlled by vator; Ii, invariant chain; CAT, chloramphenicol acetyltransferase. a specific enhanceosome. Submitted for publication.

MHC class II genes, CIITA augments IFN-␥ induction of MHC GAPDH-speciﬁc RT-PCR and hybridization was performed as described ␤ previously (44). class I and 2m genes (8–10). Four distinct genetic complementation groups have been de- RNA and DNA hybridization analysis scribed in BLS (A–D) (33), and the genes affected in these complementation groups have been identiﬁed. BLS patients of comple- For Figs. 1 and 2, total cellular RNA was prepared using RNAzolB (Cinna/ Biotecx Laboratories, Houston, TX) following the manufacturer’s instruc- mentation group A are defective in CIITA (27, 34). The genes tions. Twenty micrograms of total RNA was separated on an 1.2% agarose affected in groups B, C, and D all encode subunits of the hetero- gel containing 2.2 M formaldehyde, transferred to a Hybond N membrane trimeric phosphoprotein RFX (35). BLS groups B, C, and D are (Amersham), and hybridized using probes that were labeled with 32 defective in RFX-B (also referred to as RFXANK) (36, 37), RFX5 [ P]dCTP by random priming (DuPont-NEN, Brussels, Belgium). Trans- fer and hybridization were performed according to the instructions of the (38, 39), and RFXAP (40, 41), respectively. RFX-B is the smallest manufacturer of the membranes. The human cDNA probes for MHC class ␤ (33 kDa) of the three subunits that constitute the RFX complex I, MHC class II, invariant chain (Ii), 2m, and GAPDH were described (36, 37). Despite evidence demonstrating that RFX-B interacts before (43–46). For Fig. 7, RNA was extracted from cells using the with DNA (42), the protein has no known DNA binding domain RNeasy kit (Qiagen, Valencia, CA). Total RNA (10 ␮g) was used for (36, 37). RFX-B contains three ankyrin repeats that may be im- Northern blot analysis. Random primed RFX-B cDNA (Rediprime II; Amersham) spanning exons 3–10 was used as a probe, and hybridization portant in the formation of the RFX complex and the assembly of was conducted in Ultrahyb (Ambion, Austin, TX). The blots were reprobed the multiprotein complex involving RFX, X2BP, and NF-Y on for GAPDH to verify equal loading of RNA. Southern blot analysis of X-Y DNA. Alternatively, these ankyrin repeats may mediate in- RT-PCR products was conducted as described (49) using an end labeled Ј teractions of the multiprotein/DNA complex with the MHC primer as probe. Primers corresponding to exons 4 and 5 were 5 -CCAC CACTCTCACCAACCGG and 5Ј-CTGTCCATCCACCAGCTCGCAG CIITA.

CACAG, respectively. Downloaded from Here we describe the molecular characterization of BLS patients that were found to belong to BLS complementation group B. The Plasmids and DNA transfections defects in RFX-B resulted in absence of MHC class II expression The following plasmids were used in this study: pREP4-CIITA (9, 31), and reduced levels of MHC class I expression. Expression of ex- pSVK-FLAG-CIITA (50), pRFX-B (37), pRFX-B⌬5 (37), and pRFX5 ogenous CIITA, in the absence of wild-type RFX-B, resulted in the (created by inserting the RFX5 cDNA into pcDNA3.1). The base vectors for the above plasmids, pREP4 (Invitrogen, Carlsbad, CA), pcDNA3.1 induction of appreciable amounts of HLA-DR and also of (Invitrogen), and pSVK (Pharmacia, Uppsala, Sweden) were used as con- http://www.jimmunol.org/ HLA-DP in RFX-B-deficient fibroblasts. Exogenous CIITA alone trols. The promoter reporter constructs pGL3-B250, pGL3-␤ m , pGL3-Ii, ␤ 2 1 was able to transactivate the MHC class I and the 2m promoter in pDRCAT, and pDRCATwt have been described previously (8, 9, 31, 46). RFX-B-deficient fibroblasts as well. These results reveal that cer- Stable cell lines were generated in fibroblasts by the calcium phosphate precipitation method (51) or electroporation (31) with the indicated plas- tain mutations in RFX-B can function in CIITA-mediated trans- ␮ ␤ mid and were selected in either 50 g/ml G418 (Life Technologies) or 50 activation of MHC class II, class I, and 2m promoters and that ␮g/ml hygromycin (Boehringer Mannheim, Mannheim, Germany) for an CIITA overexpression may override the deficiency of BLS group average of 2 wk. Resistant colonies were pooled and used for analysis of B patients. MHC surface expression. Promoter activity assays

Materials and Methods For luciferase containing reporter plasmids, in each of four wells of a by guest on October 1, 2021 Cell lines six-well plate, 1.5 ϫ 105 fibroblast cells were transfected by the calcium phosphate method, with a DNA mix containing 1 ␮g firefly luciferase BLS patient EBA was 8 mo of age at the time he was referred to the Leiden ␤ pGL3 reporter plasmid (pGL3-DRA, pGL3-B250, pGL3- 2m, or pGL3- University Medical Center for allogenic bone marrow transplantation (43). Ii), 1 ␮g Renilla luciferase pRL-TK control plasmid (Promega, Madison, Patient FZA was 14 years of age at the time of cell sampling. FACS WI), and 1 ␮g of pREP4 or 0.5 ␮g of pREP4-CIITA. Cells were harvested Ͻ analysis revealed that 1% of his lymphocytes stained weakly positive for 3 days after transfection. Luciferase activity was determined using the du- HLA-DR. Patient EBA was characterized by oligotyping as DR7/DR11, al-luciferase reporter assay system (Promega) and a luminometer (Tropix, and patient FZA was characterized by oligotyping as DR4/CDR10. Pri- Bedford, MA). mary fibroblasts from a skin biopsy of the two patients were first trans- For chloramphenicol acetyltransferase (CAT)-based reporter construc- formed with an SV40 ori plasmid and subsequently stably cotransfected tions, cells were cotransfected with 10 ␮g pRFX-B, 10 ␮g pSVK-FLAG- with pCMVEBNA and pRSVneo (44). SV40-transformed fibroblasts de- CIITA; 10 ␮g pDRCATwt, a reporter that contains the SXY box of class rived from BLS group C patients OSE (DR17/DR4) and SSI (DR7/DR10) II promoter (52); and 0.5 ␮g of pGL3-promoter vector (Promega), which and BLS group D patient ABI (DR16/DR7) were described previously encodes the luciferase gene driven by a constitutive promoter (SV40). Con- (44–47). SV40-transformed JVH (DR17/DR11) and ABL fibroblasts trol transfections were conducted similarly with the pDRCAT reporter that (DR1/DR15) were derived from BLS-group A and B patients, respectively does not contain the SXY box of class II promoter (52). Cell lysates were (48). The BLS group B patient-derived B cell line Ramia was described prepared 72 h posttransfection, and 5% of the lysate was analyzed for previously (48). Fibroblasts and B lymphoblastoid cell lines were grown in expression of luciferase product using the Luciferase Assay System (Pro- Iscove’s modified DMEM (Life Technologies, Paisley, U.K.) supple- mega). The remaining sample was analyzed for CAT protein using an mented with 10% (v/v) FCS (Life Technologies) penicillin (100 IU/ml), ELISA (Boehringer Mannheim) according to the manufacturer’s instruc- ␮ and streptomycin (100 g/ml). tions. The data were normalized to the expression of luciferase.

Transient heterokaryon analysis Flow cytometric analysis The generation and analysis of transient fibroblast homo- and heterokary- To measure MHC class II and class I surface expression, cells were stained ons were essentially as described before (44). Upon treatment of fibroblasts with mAbs against the HLA-DR backbone (B8.11.2) (53), -DQ backbone with polyethylene glycol 4000, cells were cultured in the absence or pres- (SPV-L3) (54), -DP backbone (B7/21) (55), or MHC class I (WT6/32) and ence of 500 U/ml of IFN-␥ for 48 h. RNA was isolated from the cells and FITC-conjugated goat anti-mouse IgG (Becton Dickinson, Mountain View, subjected to RT-PCR using HLA-DRB haplotype-specific oligonucleotides CA). Cells stained with the corresponding conjugated or unconjugated Ig (located in exon 2) as 5Ј primers and a generic HLA-DRB oligonucleotide isotype were used as controls. In some experiments, the Abs described in (located in exon 3) as the 3Ј primer. The RT-PCR products were size- Nagarajan et al. (37) were used. Approximately 5000 cells were analyzed fractionated on an 1% agarose gel, transferred to Hybond Nϩ membranes on a FACScan flow cytometer (Becton Dickinson) in each assay. (Amersham, Little Chalfont, U.K.), and hybridized with biotin-labeled RT-PCR HLA-DRB haplotype-specific probes. The generic 3Ј primer as well as the 5Ј primers and biotin-labeled hybridization oligonucleotides and the criti- PolyA RNA was made using PolyA Tract mRNA isolation system (Pro- cal washing temperatures were described before (44–46). To assure mega). Reverse transcriptase reactions were conducted using Superscript II that the quality and the amount of the various cDNAs were similar, (Life Technologies), and PCR was conducted using native PfuI polymerase 3668 RFX-B-DEPENDENT MHC EXPRESSION

(Stratagene, La Jolla, CA). The RFX-B cDNA from exons 2–10 was amplified using primers with restriction site overhangs: 5Ј-CTAGTCTA GACAGATCGCTGAGGGTCCG and 5Ј-CCGGAATTCCGGCAGGCG GCCTTCACTC. Thirty cycles of 30 s at 95°C, 30 s at 55°C, 1.0 min at 72°C, and a final extension at 72°C for 7 min, were used to amplify the RFX-B cDNA for all RT-PCR samples. GAPDH was amplified from the same reverse transcription reaction as a control, using the primers 5Ј-CCATGGGGAAGGTGAGGTAGGATC and 5Ј-GAGGAGTGGGTG TCGCTGTTGATC. RFX-B cDNAs obtained by RT-PCR from patient RNAs were cloned into the pCRBlunt vector (Invitrogen) for sequencing and also cloned into the mammalian expression vector pcDNA3.1Ϫ at XbaI/EcoRI sites.

Genomic DNA PCR and analysis of homozygosity Genomic DNA from the cell lines was made as described in Ausubel et al. (56). PCR of the RFX-B gene consisted of 30 cycles of 10 s at 92°C, 30 s at 60°C, and 2 min at 68°C. The last cycle was extended for 7 min. DNA fragments encoding exons 4–8 were amplified using the primers, 5Ј- CCACCACTCTCACCAACCGG and 5Ј-CGCATTTCACGTGGTTCC CGCG-3Ј. To analyze the mutations in FZA and family members, exon 8 was amplified using primers corresponding to introns 7 and 8: 5Ј-GAACT GCGCTGCGATGGCAGATG and 5Ј-GCAGGCAGGGACTACCTGC Downloaded from TCC, respectively. The PCR products were purified and sequenced. To detect homozygosity in EBA, PCR was conducted using Taq polymerase and primer sets specific for the mutated or wild-type sequence in exon 5. Ј A wild-type forward primer located in exon 4, 5 -CCACCACTCTCAC FIGURE 1. Northern blot analysis of mRNA isolated from SV40-trans- Ј CAACCGG, was used with either the wild type 5 -GCTCCTTCAGCTG formed fibroblasts of the two type III BLS patients, FZA and EBA, and of GTCCAGCTC or mutated 5Ј-GCTCCTTCAGCTGGTCCAGCTA reverse WSI, a normal control individual (51). Fibroblasts were grown in the ab- primers. sence (Ϫ) or presence (ϩ) of IFN-␥. http://www.jimmunol.org/

Results CIITA-mediated transactivation of MHC genes in the absence of Lack of MHC class II and reduced class I expression in BLS wild-type RFX-B patients IFN-␥ induces MHC class II genes through the induction and Patients FZA and EBA presented themselves with an apparent loss expression of the transactivator CIITA. Expression of exogenous of MHC class II surface expression on lymphocytes, although in CIITA via plasmid transfection results in the induction of MHC the case of FZA Ͻ1% of the lymphocytes stained weakly for class II expression of all three isotypes in many cell types (47, 57). by guest on October 1, 2021 HLA-DR. Immortalized fibroblast cell lines were established from To determine whether overexpression of CIITA could result in skin biopsies as previously described (44). To examine the expres- MHC class II expression, EBA fibroblasts were stably transfected sion characteristics of MHC class II (HLA-DRA), MHC class I, Ii, with pREP4-CIITA. Following hygromycin selection, these cells ␤ and 2m genes in fibroblasts derived from BLS patients FZA and were found to express HLA-DR and HLA-DP, and to a lesser EBA, RNA was isolated from untreated and IFN-␥-treated cells extent HLA-DQ (Fig. 3). Expression of all three MHC class II and subjected to Northern blot analysis (Fig. 1). Compared with a isotypes in the absence of functional RFX has not been observed normal control fibroblast cell line WSI, no HLA-DRA mRNA was in fibroblast cells with defects in RFX5 or RFXAP (46, 47). detected upon induction of FZA and EBA fibroblasts with IFN-␥. CIITA-transfected RFX5 or RFXAP-deficient cells were found to Furthermore, Ii expression both in FZA and EBA could not be express HLA-DR (results not shown; see Refs. 46 and 47). In detected upon treatment with IFN-␥. The amount of constitutively addition, these CIITA-transfected EBA cells also displayed a mod- ␤ expressed MHC class I and 2m transcripts were greatly reduced. erate increase in MHC class I cell-surface expression, which was The expression characteristics of MHC and accessory genes in also not noted in CIITA-transfected RFX5 and RFXAP-deficient FZA and EBA is similar to our previous observations in other BLS fibroblasts. patients that contain defects in the RFX5 and RFXAP subunits of RFX (3, 8, 44–46). RFX-B regulates constitutive and CIITA-mediated transactivation of MHC and accessory genes To further investigate the role of RFX-B in constitutive and FZA and EBA belong to BLS complementation group B CIITA-mediated transactivation of MHC class II, MHC class I, Somatic cell fusion experiments were performed to determine the and accessory genes, transient transfection experiments were per- complementation group assignment of FZA and EBA. FZA fibro- formed using FZA and EBA fibroblasts with either HLA-DRA ␤ blasts were fused with fibroblasts derived from BLS patients JVH, promoter-driven CAT reporter constructions or HLA-B7, 2m, ABL, SSI, and ABI, which were representative for complementa- and Ii promoter-driven luciferase reporter constructs. For compar- tion groups A, B, C, and D, respectively. The transient heterokary- ison to FZA and EBA, another BLS group B cell line, Ramia, was ons were analyzed for restoration of MHC class II expression by also analyzed. Ramia cells are EBV-transformed B lymphocytes RT-PCR and Southern blotting for HLA-DRB. As shown in Fig. 2, and like FZA and EBA are negative for class II expression (48). A and B, both FZA and EBA were complemented by JVH, SSI, Transient cotransfection of Ramia cells with the HLA-DRA re- and ABI, but not by ABL and EBA. These results demonstrate that porter or a control reporter, an RFX-B expression plasmid or a both FZA and EBA belong to complementation group B and sug- plasmid expressing a splice variant of RFX-B, RFXB⌬5, showed gest that they contain defects in the RFX-B subunit of the RFX that RFX-B but not the control vectors could complement the de- complex (36, 37). fect in Ramia cells (Fig. 4A). RFX-B⌬5 expression showed a slight The Journal of Immunology 3669 Downloaded from

FIGURE 3. CIITA partially restores MHC class II expression of all isotypes in EBA ﬁbroblasts. Vector (pREP-4)- and CIITA (pREP4-CIITA)- transfected cells were stained with Abs directed against HLA-DR, -DP, and http://www.jimmunol.org/ -DQ, and against HLA class I and analyzed by FACS.

Complementation of FZA and EBA fibroblasts with RFX-B alone has a significant impact on the constitutive levels of expres- ␤ sion of MHC class I and 2m promoters. This is in contrast to the HLA-DRA and Ii promoter, where RFX-B had no effect on the constitutive level of expression. In combination with CIITA, all of the above mentioned promoters displayed a substantial increase in by guest on October 1, 2021 the level of expression in these RFX-B-deficient fibroblasts. Inter- ␤ estingly, both the MHC class I and 2m promoters were transac- tivated by CIITA in the absence of wild-type RFX-B. This is in contrast to RFX5-and RFXAP-deficient cells that lacked CIITA- mediated transactivation in the absence of these wild-type proteins. Together, these results suggest that RFX-B plays a critical role in the constitutive and CIITA-mediated transactivation of MHC class ␤ I and 2m genes. FIGURE 2. Characterization of transient heterokaryons. FZA (A)or Expression of RFX-B leads to surface expression of MHC class EBA (B) fibroblasts were cocultured with fibroblast belonging to BLS II in both Ramia B cells and FZA fibroblasts complementation groups A, B, C, and D. The monolayers were exposed (P) or not exposed to polyethylene glycol 4000 and, subsequently, grown in the Stable cell lines expressing RFX-B were created to determine presence (ϩ) or absence (Ϫ) of IFN-␥. The transient heterokaryons were whether RFX-B could fully complement the defect of the endog- analyzed for reexpression of HLA-DRB by RT-PCR and Southern blotting enous MHC class II genes in both Ramia and FZA cells. Flow using oligonucleotide primers and probes specific for a particular HLA- cytometric analysis of vector-transfected or RFX-B-transfected DRB haplotype (indicated). As control, cultures of each of the fibroblasts cell lines showed that RFX-B could complement all three MHC were treated in a similar fashion. class II isotypes (Fig. 5). FZA cells coexpressing both CIITA and RFX-B showed enhanced expression of MHC class II compared with CIITA-expressing cells. Thus, mutations in the RFX-B gene rescue of class II expression. Because Ramia cells are B cells, are likely responsible for the lack of class II expression in these exogenous CIITA expression was not required. FZA and EBA cells. fibroblasts were similarly cotransfected; however, in these experiments a CIITA expression vector was required for expression. Ramia, FZA, and EBA have different RFX-B mutations Similar to the flow cytometry experiments in Fig. 3, transfection of The above results could suggest that the mutant RFX-B alleles the CIITA expression vector alone results in a moderate level of produce proteins that could provide some function in the presence expression (Fig. 4A). As with Ramia cells, transfection of RFX-B of excess CIITA. Alternatively, excess CIITA may be sufficient to rescues the class II defect. Moreover, transfection of the RFX5 drive class II expression in the absence of functional RFX-B pro- subunit in FZA fibroblasts did not result in rescue, demonstrating tein. To begin to distinguish between these hypotheses, the nature the specificity of the complementation. of the defect and the genotypes of these cell lines were determined. 3670 RFX-B-DEPENDENT MHC EXPRESSION Downloaded from

␤ FIGURE 4. RFX-B complements deficiencies in MHC class II, MHC class I, and 2m expression. A, Transient transfections were conducted in Ramia, FZA, and EBA cells by electroporation. Samples contained the plasmids expressing the indicated proteins and either the HLA-DRA SXY region-dependent reporter (pDRCATwt) or the control plasmid pDRCAT. CAT values were normalized to the expression levels of a constitutively expressed luciferase reporter and are expressed with the SE from three independent assays. B, FZA and EBA fibroblasts were transfected by calcium phosphate with the indicated ␤ http://www.jimmunol.org/ expression vectors and either the HLA-B or 2m promoter-dependent firefly luciferase reporter. Samples were normalized to the expression of a constitutively expressing renilla luciferase reporter and are expressed with the SE from three independent experiments.

Previous analysis of RFX-B-deficient cells identified two mutant products were obtained from FZA only. Shorter transcripts were alleles, both of which affected the splicing of exon 6. One allele obtained from EBA and Ramia. The RT-PCR products were se- contained a 26-bp deletion in a region encompassing the splice quenced after cloning into PCR blunt vector. Both FZA and EBA acceptor site of exon 6 (36, 37). The second allele identified was DNA sequences each contained a single base pair substitution (Fig. a 56-bp deletion of the DNA encompassing the splice donor site 5Ј 6A). FZA contained a T 3 C transition in exon 8, resulting in a to exon 6 (36) Both of these mutations cause a frame shift and a Leu195 3 Pro substitution. This mutation alters a conserved by guest on October 1, 2021 premature stop codon. To determine the mutations in each of these leucine residue in the third ankyrin repeat of the RFX-B protein. cell lines, RFX-B cDNA from each of the cell lines was generated Sequence analysis of EBA’s RT-PCR product showed that it cor- by RT-PCR; however, as discussed below, full-length RT-PCR responded to the sequence of RFX-B⌬5 as observed in wild-type Raji cells. This suggested that EBA could either contain a splice acceptor/donor mutation flanking exon 5 or a point mutation in exon 5 that could result in an unstable full-length transcript. To delineate the defect in EBA, genomic DNA was isolated from EBA cells and exons 4–8 of RFX-B gene were amplified by PCR. A transversion, which changed GAG (Glu103) 3 TAG (amb) was found in exon 5, thereby encoding a truncated protein. The mutation in Ramia was found to be the same 26-bp deletion described above (Fig. 6A). A smaller RT-PCR product, lacking both exons 5 and 6, was also found in Ramia cells. Experiments were performed to determine whether the mutations were homozygous within these patients. The mutation in FZA generates a novel BsiWI restriction site. DNA from FZA and cell lines from the parents of FZA, JZA, and MZA was isolated and amplified across exon 8 by PCR. The resulting DNA was subjected to BsiWI digestion and analyzed by gel electrophoresis (Fig. 6B). Wild-type DNA (Raji) and EBA DNA were analyzed in parallel. The results show that each of FZA’s parents carries one wild-type and one mutant allele and that FZA is homozygous for the point mutation. To determine whether EBA was homozygous for the nonsense codon, wild-type and mutation-specific PCR primers were generated and used to amplify genomic DNA by FIGURE 5. RFX-B induces class II surface expression. Stable transfor- mants were generated in Ramia and FZA with the indicated expression PCR (Fig. 6C). The results showed that EBA DNA was amplified plasmids or vector control. Following selection, cells were stained with the by the mutation-specific primer only. Control reactions included indicated Abs and analyzed by flow cytometry (dark histograms). The light wild-type DNA from Raji cells and DNA from FZA, which is wild histogram indicates isotype-matched control antiserum staining. type at this position. Thus, both FZA and EBA are homozygous for The Journal of Immunology 3671 Downloaded from

FIGURE 7. A, FZA fibroblasts express wild-type levels of RFX-B. Northern blot analysis of total RNA from Raji, EBA, FZA, and Ramia using an RFX-B cDNA probe were performed. GAPDH RNA was assayed for loading comparison between the cell types. B, EBA expresses the RFX- B⌬5 splice variant. Total RNA from Raji, FZA, and EBA amplified by RT-PCR using primers encompassing exons 2–10 was analyzed by gel http://www.jimmunol.org/ electrophoresis (top panel). Full-length and splice variant transcripts are represented by the upper and lower bands in the wild-type Raji cell lanes. Southern blots conducted on this gel using exon 5- or exon 4-specific probes are shown. Exon 5 is absent in the splice variant, whereas exon 4 is present in both the wild-type and splice variant. The lower panel shows an agarose gel of parallel RT-PCR for the GAPDH, which was used to com- pare sample and RNA loading. by guest on October 1, 2021 ducted to determine whether RNA could be detected in EBA cells FIGURE 6. A, Ramia, FZA and EBA have three distinct mutations. The (Fig. 7B). RT-PCR of RFX-B from the wild-type Raji cells shows RFX-B gene from Ramia, EBA, and FZA were amplified from RNA or both the full-length RFX-B band (upper) and the minor splice vari- genomic DNA by PCR as described in Materials and Methods and se- ant, RFX-B⌬5 (lower band). FZA produces an identical pattern to quenced. Mutations and exon numbers are indicated. B, FZA is homozy- the wild type. Surprisingly, EBA displays only the lower band. To gous for RFX-B. A BsiW1 site is generated due to the mutation. PCR amplification of DNA isolated from FZA and MZA and JZA (FZA’s par- demonstrate that this lower band was in fact the splice variant, a ents) was conducted, digested with BsiWI, and analyzed by gel electro- Southern blot was performed on this gel using probes specific for phoresis. PCR amplification of wild-type Raji DNA and EBA DNA are exon 5, which is missing in the splice variant, and exon 4, which shown as positive controls. C, EBA is homozygous for RFX-B. PCR am- is present in the splice variant. The result showed that EBA ex- plification of genomic DNA from the indicated cell types was conducted presses the splice variant, but not the full-length mRNA that would using a wild-type reverse primer or primer corresponding to EBA mutation contain the nonsense codon. As described above, this result was as the last nucleotide. The PCR products were analyzed by gel electro- verified by sequencing the RT-PCR product. Thus, by expressing phoresis. No product was found in EBA by using wild-type primer. the splice variant, EBA cells can bypass the nonsense codon generated by the mutation. However, this transcript is not stable in cells and is expressed at very low levels, suggesting that if any their mutations. Ramia was also found to be homozygous by protein is synthesized it would be low in abundance. Southern blot analysis (data not shown), as described in Nagarajan FZA et al. (37). RFX-B can provide a low level of expression Because FZA produces normal RFX-B mRNA levels, it is possible RNA analysis of the group B cell lines shows distinct expression that the RFX-BFZA protein can stimulate a low level of MHC class patterns II expression. Similarly, the RFX-B⌬5 splice variant, which is The 26-bp deletion mutation that is also found in Ramia cells was found in EBA, may also be able to provide some level of MHC shown previously to result in destabilization and loss of the RFX-B class II expression. To assay these variants, the RFX-BFZA mRNA mRNA. To determine whether the same were true for the FZA and and the splice variant were cloned into the mammalian expression EBA mutations, RNA analyses by Northern blotting and RT-PCR vector pcDNA3.1. Ramia cells, which are deficient for both the were performed (Fig. 7). Northern blot analysis of Ramia, FZA, wild-type and splice variant, were cotransfected with the above and EBA showed two patterns. Ramia and EBA were found to expression vectors and a DRA promoter-dependent reporter con- express undetectable and very low levels of RFX-B mRNA by struct (pDRCATwt) or its control vector (pDRCAT). The results Northern blot, respectively, whereas, FZA expressed wild-type showed that the splice variant could provide a small increase in RFX-B mRNA levels (Fig. 7A). RT-PCR of exons 2–10 was con- expression of the MHC class II promoter-dependent reporter over 3672 RFX-B-DEPENDENT MHC EXPRESSION

experiments eliminate the possibility that CIITA or IFN-␥ treatment might contribute to class II expression through an increase in RFX-B levels. Thus, these data demonstrate that CIITA overexpression can partially override the RFX-B defect and provide a mechanism by which EBA cells can express class I, class II, and ␤ ␥ 2m genes following CIITA transfection but not following IFN- treatment.

Discussion The cloning of the RFX-B/ANK gene (36, 37) completed the iden- tification of the genes responsible for the four major BLS comple- FIGURE 8. RFX-BFZA is partially functional when transfected into mentation groups. BLS group B, the largest of the complementa- Ramia cells. RFX-B mRNA from FZA and splice variant were cloned into tion groups, was associated initially with two deletion mutations in the pcDNA3.1 expression vector and transfected into Ramia cells by elec- the RFX-B gene, both of which involved the splicing of exon 6 troporation along with a HLA-DRA SXY-dependent reporter or its control (36, 37). Disease progression in this group is variable, and thus plasmid. CAT activity was assayed and normalized to the luciferase levels investigating the genetic basis of patients in this group may shed from a cotransfected luciferase expression plasmid. The average of three light on the regions of RFX-B required for expression of MHC and independent experiments is plotted with their SD. their related genes. The analysis of cells derived from two new patients, FZA and EBA, showed the ability of these cells to induce Downloaded from MHC class I and all MHC class II genes in response to overex- the background associated with this vector (Fig. 8). However, FZA pression of CIITA, a property not observed with other BLS RFX-B overexpression resulted in almost one-third of the wild- complementation groups. A third patient cell line, Ramia, belong- type activity. Thus, the ability to transactivate MHC class II ex- ing to the same complementation group was also analyzed. Ramia pression by the mutant proteins may be partially achieved when cells showed no class II expression despite normal constitutive the levels of CIITA reach a threshold level, such as that in normal levels of CIITA. As discussed below, the relative phenotype http://www.jimmunol.org/ B lymphocytes, or through the overexpression of CIITA from an caused by the mutations in FZA and EBA are distinct from that of exogenous promoter. Ramia and may explain the variability of disease seen in BLS RFX-B-independent activation of MHC class II expression group B patients. This may correlate with residual levels of MHC class II expression as noted on lymphocytes of patient FZA and As stated above, Ramia cells produce an unstable RFX-B mRNA EBV B lymphoblastoid cell lines of EBA (results not shown), that is not detectable by Northern analysis. Thus, these cells are which is corroborated by the in vitro expression studies presented most likely devoid of RFX-B protein. However, Ramia cells do in Figs. 4 and 8. express normal B cell levels of CIITA. Thus, this cell line could The partial restoration of MHC class II, class I, Ii, and ␤ m serve as a model for overexpression of CIITA in the absence of 2 by guest on October 1, 2021 expression following overexpression of CIITA in the FZA and RFX-B. Therefore, to determine whether CIITA could transacti- EBA fibroblasts suggests that the mutant RFX-B proteins are ei- vate in the absence of RFX-B, a CIITA expression vector was ther partially active or that the CIITA overexpression can com- transfected into Ramia cells (Fig. 9). This resulted in a 4- to 5-fold pensate in some manner for a mutated RFX-B protein. The data increase in the activity of the cotransfected DRA reporter. As argue for a combination of these two view points. Unlike RFX-B above, transfection of the RFX-B expression vector yielded a 12- mRNA in EBA and Ramia, RFX-BFZA mRNA is expressed at fold increase in expression. Interesting, expression of both RFX-B normal levels and is not rapidly degraded. The fact that one-third and CIITA vectors resulted in a 26-fold increase, suggesting that of wild-type activity is observed when the RFX-BFZA mRNA in CIITA expression was limiting in these cells. Moreover, mRNA Ramia cells suggests that the mutant protein is produced. The de- from wild-type RFX-B or the mutant alleles expressed in FZA or velopment of an antiserum to RFX-B will allow confirmation of EBA were not found to be induced by treatment of the cells with this prediction. Thus, with regard to RFX-BFZA, it is likely that the IFN-␥ or in cells transfected with CIITA (data not shown). These protein is partially functional. The RFX-BFZA mutation changes Leu195 3 Pro in the third ankyrin repeat of the RFX-B protein. These ankyrin repeats are conserved domains that are involved in protein-protein interactions and have ␣ helical structures. The pro- line substitution in RFX-BFZA may affect the ␣ helix within the third repeat. We have recently found that the third ankyrin repeat is required for association of RFX-B with the other two RFX subunits, RFX-5 and RFXAP (DeSandro, Nagarajan, and Boss, un- published data). It is not known at the current time whether FZA mutation specifically affects the interactions between the three subunits, although this is a likely prediction from our current data. The EBA mutation provides an opportunity to examine the role of the RFX-B splice variant. The stop codon caused by the single base pair mutation results in a frame shift and an unstable mRNA FIGURE 9. CIITA overexpression stimulates expression from an HLA- DRA reporter. Transient cotransfections were conducted in Ramia cells for the full-length transcript. However, because the splice variant with the pDRCATwt reporter plasmid and vectors expressing CIITA fortuitously removes the exon encoding the mutation, it is unaf- and/or wild-type RFX-B as indicated. pUC19 DNA was cotransfected with fected by the mutation. A low level of the splice variant mRNA the reporter gene as a control. The average of three experiments are shown can be detected by overexposure of a Northern blot. RT-PCR with the SD from the mean. readily detects this transcript as shown in Fig. 7. Expression of the The Journal of Immunology 3673 splice variant mRNA does not normally rescue RFX function or References MHC class II expression. This could be due to instability of the 1. Germain, R. N. 1994. MHC-dependent antigen processing and peptide presenta- RFX-B RNA or to the protein that it encodes. However, in the tion providing ligands for T lymphocyte activation. Cell 76:287. presence of excess CIITA, a low but significant level of MHC class 2. Jameson, S. C., and M. J. Bevan. 1998. T cell selection. Curr. Opin. Immunol. 10:214. II and an increase in MHC class I expression can be detected in 3. van den Elsen, P. J., M. C. Peijnenburg, J. A. Van Eggermond, and S. J. P. Gobin. EBA cells. Additionally, overexpression of the splice variant in 1998. Shared regulatory elements in the promoters of MHC class I and class II several of the experiments presented showed a low level of MHC genes. Immunol. Today 19:308. 4. Klein, C., B. Lisowska-Grospierre, F. LeDeist, A. Fischer, and C. Griscelli. 1993. class II expression. One interpretation of this result is that the low Major histocompatibility complex class II deficiency: clinical manifestations, im- level of splice variant-generated protein may function in these as- munologic features and outcome. J. Pediatr. 123:921. 5. Lambert, M., M. C Van Eggermond, M. Andrien, F. Mascart, E. Vamos, says. In contrast, the mutation in Ramia, which results in a frame E. Dupont, and P. van den Elsen. 1991. Analysis of the peripheral T-cell com- shift in exon 6 and unstable mRNA, is completely devoid of RFX partment in the MHC class Ii deficiency syndrome. Res. Immunol. 142:789. activity. The splice variant would also be affected by this mutation 6. Mach, B., V. Steimle, E. Martinez-Soria, and W. Reith. 1996. Regulation of MHC class II genes: lessons from a disease. Annu. Rev. Immunol. 14:301. as well, as it includes exon 6. Thus, these three mutations may be 7. Boss, J. M. 1997. Regulation of transcription of MHC class II genes. Curr. Opin. grouped into three categories with regard to the transcriptional Immunol. 9:107. 8. Gobin, S. J. P., A. Peijneburg, M. van Eggermond, M. van Zutphen, potential of RFX-B: partial activity in FZA, low activity in EBA, R. van den Berg, and P. van den Elsen. 1998. The RFX complex is crucial for the ␤ and no activity in Ramia. constitutive and CIITA-mediated transactivation of MHC class I and 2-micro- The finding that IFN-␥ treatment of FZA and EBA cells did not globulin genes. Immunity 9:531. 9. Gobin, S. J. P., A. Peijnenburg, V. Keijsers, and P. J. van den Elsen. 1997. Site induce MHC class I or II expression but that CIITA expression in ␣ is crucial for two routes of IFN␥-induced MHC class I transactivation: the these cell lines did was intriguing and suggested that CIITA may ISRE-mediate route and a novel pathway involving CIITA. Immunity 6:601. Downloaded from 10. Martin, B. K., K.-C. Chin, J. C. Olsen, C. A. Skinner, A. Dey, K. Ozato, and partially override the RFX-B defect. This hypothesis was tested by J. P.-Y. Ting. 1997. Induction of MHC class I expression by the MHC class II transfecting CIITA into Ramia cells, which lack RFX-B. The re- transactivator CIITA. Immunity 6:581. 11. Glimcher, L. H., and C. J. Kara. 1992. Sequences and factors: a guide to MHC sults showed a 4- to 5-fold increase in expression of the MHC class class II transcription. In Annual Reviews of Immunology, 10th ed. W. E. Paul, II reporter gene and demonstrated that CIITA could compensate C. Garrison Fathman, and H. Metzger, eds. Annual Reviews Inc., Palo Alto, CA, for the RFX-B defect. We have found in other experiments that p. 13. 12. Ting, J. P.-Y., and A. S. Baldwin. 1993. Regulation of MHC gene expression. CIITA expression from the pcDNA3.1 vector provides at least a Curr. Opin. Immunol. 5:8. http://www.jimmunol.org/ 3-fold increase in the level of CIITA mRNA in transfected cells as 13. van den Elsen, P. J., S. J. P. Gobin, M. C. J. A. van Eggermond, and ␥ A. Peijnenburg. 1998. Regulation of MHC class I and II gene transcription: dif- compared with IFN- -treated cells (data not shown). Additionally, ferences and similarities. Immunogenetics 48:208. in stable cell lines, CIITA is constitutively expressed and always 14. Rohn, W. M., J. Lee, and E. N. Benveniste. 1996. Regulation of MHC class II present, allowing accumulation of CIITA protein. Thus, the effect expression. CRC Crit. Rev. Immunol. 16:311. 15. Reith, W., S. Satola, C. H. Sanchez, I. Amaldi, B. Lisowska-Grospierre, is likely to be greater in the EBA and FZA cell lines stably trans- C. Griscelli, M. R. Hadam, and B. Mach. 1988. Congenital immunodeficency fected with CIITA. with a regulatory defect in MHC class II gene expression lacks a specific HLA-DR promoter binding protein, RF-X. Cell 53:897. What is CIITA’s role in these experiments? It is possible that 16. Moreno, C. S., P. Emery, J. E. West, B. Durand, W. Reith, B. Mach, and CIITA stabilizes weak interactions at the class I or even class II J. M. Boss. 1995. Purified X2BP cooperatively binds the class II MHC X box by guest on October 1, 2021 promoters. There is evidence to suggest that CIITA expression region in the presence of purified RFX, the X box factor deficient in the bare lymphocyte syndrome. J. Immunol. 155:4313. could enhance the in vivo footprint in cells that are normally class 17. Moreno, C. S., G. Beresford, P. Louis-Plence, A. C. Morris, and J. M. Boss. 1999. II negative (58–60). Thus, in the case of the EBA and Ramia CREB regulates MHC class II expression in a CIITA-dependent manner. Immu- nity 10:143. RFX-B alleles, CIITA may stabilize the promoter complex in the 18. Zeleznik-Le, N. J., J. C. Azizkhan, and J. P.-Y Ting. 1991. Affinity-purified absence of a functional RFX-B protein. In doing so, CIITA would CCAAT-box-binding protein (YEBP) functionally regulates expression of a hu- then be able to activate transcription. The FZA allele is expressed man class II major histocompatibility complex gene and the herpes simplex virus thymidine kinase gene. Proc. Natl. Acad. Sci. USA 88:1873. and may provide partial activity to the RFX complex, as indicated 19. Schoneich, J., J. L. Lee, P. Mansky, M. Sheffery, and S. Y Yang. 1997. The by partial rescue of Ramia cells. Because MHC class I genes use pentanucleotide ATTGG, the “inverted CCAAT”, is an essential element for ␥ HLA class I gene transcription. J. Immunol. 158:4788. other elements for their expression and induction by IFN- , which 20. Wright, K. L., B. J. Vilen, Y. Itoh-Lindstrom, T. L. Moore, G. Li, M. Criscitello, are located upstream to the SXY motifs, it is possible that these P. Cogswell, J. B. Clarke, and J. P.-Y. Ting. 1994. CCAAT box-binding protein, elements provide additional stability or aid in the assembly of the NF-Y, facilitates in vivo recruitment of upstream DNA-binding transcription factors. EMBO J. 13:4042. SXY-specific factors when CIITA is expressed. Thus, deficiencies 21. Reith, W., C. A. Siegrist, B. Durand, E. Barras, and B. Mach. 1994. Function of in RFX-B may not be sufficient to prevent transcription of MHC major histocompatibility complex class II promoters requires cooperative binding between factors RFX and NF-Y. Proc. Natl. Acad. Sci. USA 91:554. class II genes under all conditions. Therefore, it is not surprising 22. Louis-Plence, P., C. S. Moreno, and J. M. Boss. 1997. Formation of a regulatory that some of the RFX-B-deficient patients have lived into adult- factor X/X2 box-binding protein/nuclear factor-Y multiprotein complex on the hood. The analysis of the three patient cell lines in this report conserved regulatory regions of HLA class II genes. J. Immunol. 159:3899. 23. Fontes, J. D., N. Jabrane-Ferrat, and B. M. Peterlin. 1997. Assembly of functional provide evidence for single base pair substitution within an im- regulatory complexes on MHC class II promoters in vivo. J. Mol. Biol. 270:336. portant regulatory element. Thus, this analysis questions whether 24. Linhoff, M. W., K. L. Wright, and J. P. Y. Ting. 1997. CCAAT-binding factor NF-Y and RFX are required for in vivo assembly of a nucleoprotein complex that there are other alleles for complementation group B. It is possible spans 250 base pairs: the invariant chain promoter as a model. Mol. Cell. Biol. that such mutations may be plentiful in humans and that there is 17:4589. allowance at this locus and protein structure for expression and a 25. Kara, C. J., and L. H. Glimcher. 1991. In vivo footprinting of MHC class II genes: bare promoters in the bare lymphocyte syndrome. Science 252:709. viable immune system. 26. Kara, C. J., and L. H. Glimcher. 1993. Three in vivo promoter phenotypes in MHC class II deficient combined immunodeficiency. Immunogenetics 37:227. 27. Steimle, V., L. A. Otten, M. Zufferey, and B. Mach. 1993. Complementation cloning of an MHC class II transactivator mutated in hereditary MHC class II Acknowledgments deficiency (or bare lymphocyte syndrome). Cell 75:135. 28. Scholl, T., S. K. Mahanta, and J. L. Strominger. 1997. Specific complex forma- We thank Drs. T. Ottenhof and B. Roep for critically reading the manu- tion between the type II bare lymphocyte syndrome-associated transactivators script. We are grateful to Drs. B. Gerritsen, S. van Lierde, and J. M. Vossen CIITA and RFX5. Proc. Natl. Acad. Sci. USA 94:6330. 29. Fontes, J. D., S. Kanazawa, D. Jean, and B. M. Peterlin. 1999. Interactions be- for providing patient material and to M. van Eggermond and R. van den tween the class II transactivator and CREB binding protein increase transcription Berg for expert technical assistance. of major histocompatibility complex class II genes. Mol. Cell. Biol. 19:941. 3674 RFX-B-DEPENDENT MHC EXPRESSION

30. Mahanta, S. K., T. Scholl, F.-C. Yang, and J. L. Strominger. 1997. Transactiva- class II deficiency patient reveals a novel mutation in the RFX5 gene. Immuno- tion by CIITA, the type II bare lymphocyte syndrome-associated factor, requires genetics 49:338. participation of multiple regions of the TATA box binding protein. Proc. Natl. 46. Peijnenburg, A., M. C. J. A. van Eggermond, S. J. P. Gobin, R. van den Berg, Acad. Sci. USA 94:6324. B. C. Godthelp, J. M. J. J. Vossen, and P. J. van den Elsen. 1999. Discoordinate 31. Riley, J. L., S. D. Westerheide, J. A. Price, J. A. Brown, and J. M. Boss. 1995. expression of invariant chain and MHC class II genes in class II tranactivator- Activation of class II MHC genes requires both the X box region and the class II transfected fibroblasts defective for RFX5. J. Immunol. 163:794. transactivator (CIITA). Immunity 2:533. 47. Peijnenburg, A., S. J. P. Gobin, M. C. J. A. van Eggermond, B. C. Bodthelp, 32. Zhou, H., and L. H. Glimcher. 1995. Human MHC class II gene transcription N. van Graafeiland, and P. J. van den Elsen. 1997. Introduction of exogenous directed by the carboxyl terminus of CIITA, one of the defective genes in type II class II trans-activator in MHC class II-deficient ABI fibroblasts results in in- MHC combined immune deficiency. Immunity 2:545. complete rescue of MHC class II antigen expression. J. Immunol. 159:2720. 33. Kohonen-Corish, M. R. J., R. V. Blanden, and N. J. C. King. 1989. Induction of cell surface expression of HLA antigens by human IFN-␥ encoded by recombi- 48. Lisowska-Grospierre, B., M.-C. Fondaneche, M.-P. Rols, C. Griscelli, and nant vaccinia virus. J. Immunol. 143:623. A. Fischer. 1994. Two complementation groups account for most cases of inher- 34. Bontron, S., V. Steimle, C. Ucla, and B. Mach. 1997. Two novel mutations in the ited MHC class II deficiency. Hum. Mol. Gen. 3:953. MHC class II transactivator CIITA in a second patient from MHC class II defi- 49. Ausubel, K., and B. Gitler. 1987. Swallow syncope in an otherwise healthy young ciency complementation group A. Hum. Genet. 99:541. man. Am. Heart J. 113:831. 35. Moreno, C. S., E. M. Rogers, J. A. Brown, and J. M. Boss. 1997. Regulatory 50. Brown, J. A., E. M. Rogers, and J. M. Boss. 1988. The MHC class II transacti- factor X, a bare lymphocyte syndrome transcription factor, is a multimeric phos- vator (CIITA) requires conserved leucine charged domains for interactions with phoproteinn complex. J. Immunol. 158:5841. the conserved W box promoter element. Nucleic Acids Res. 26:4128. 36. Masternak, K., E. Barras, M. Zufferey, B. Conrad, G. Corthals, R. Aebersold, 51. Chen, C., and H. Okayama. 1987. High-efficiency transformation of mammalian J.-C. Sanchez, D. F. Hochstrasser, B. Mach, and W. Reith. 1998. A gene encod- cells by plasmid DNA. Mol. Cell Biol. 7:2745. ing a novel RFX-associated transactivator is mutated in the majority of MHC 52. Riley, J. L., and J. M. Boss. 1993. Class II MHC transcriptional mutants are class II deficiency patients. Nat. Gen. 20:273. defective in higher order complex formation. J. Immunol. 151:6942. 37. Nagarajan, U. M., P. Louis-Plence, A. DeSandro, R. Nilsen, A. Bushey, and J. M. Boss. 1999. RFX-B is the gene responsible for the most common cause of 53. Rebai, N., B. Malissen, M. Pierres, R. S. Accolla, G. Corte, and C. Mawas. 1983. the bare lymphocyte syndrome, a MHC class II immunodeficiency. Immunity Distinct HLA-DR epitopes and distinct families of HLA-DR molecules defined 10:153. by 15 monoclonal antibodies either DR or allo-anti-Ia crossreacting with human Downloaded from 38. Steimle, V., B. Durand, B. Emmanuele, M. Zufferey, M. R. Hadam, B. Mach, and DR. Eur. J. Immunol. 13:106. W. Reith. 1995. A novel DNA-binding regulatory factor is mutated in primary 54. Spits, H., J. Borst, M. Giphart, J. Coligan, C. Terhorst, and J. E. de Vries. 1984. MHC class II deficiency (bare lymphocyte syndrome). Genes Dev. 9:1021. HLA-DC antigens can serve as recognition elements of human cytotoxic T lym- 39. Villard, J., W. Reith, E. Barras, A. Gos, M. A. Morris, S. E. Antonarakis, phocytes. Eur. J. Immunol. 14:299. P. J. van den Elsen, and B. Mach. 1997. Analysis of mutations and chromosomal 55. Vandekerckhove, B. A. E., G. Datema, F. Koning, E. Goulmy, G. G. Persijn, localisation of the gene encoding RFX5, a novel transcription factor affected in J. J. van Rood, F. H. J. Claas, and J. E. de Vries. 1990. Analysis of the donor- major histocompatibility complex class II deficiency. Hum. Mutat. 10:430. specific cytotoxic T lymphocyte repertoire in a patient with a long term surviving

40. Durand, B., P. Sperisen, P. Emery, E. Barras, M. Zufferey, B. Mach, and allograft. J. Immunol. 144:1288. http://www.jimmunol.org/ W. Reith. 1997. RFXAP, a novel subunit of the RFX DNA binding complex, is 56. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, mutated in MHC class II deficiency. EMBO J. 16:1045. J. A. Smith, and K. Struhl. 1987. Current Protocols in Molecular Biology. Wiley, 41. Villard, J., B. Lisowska-Grospierre, P. van den Elsen, A. Fischer, W. Reith, and Boston. B. Mach. 1997. Mutation of RFXAP, a regulator of MHC class II genes, in 57. Steimle, V., C.-A. Siegrist, A. Mottet, B. Lisowska-Grospierre, and B. Mach. primary MHC class II deficiency. N. Engl. J. Med. 337:748. 1994. Regulation of MHC class II expression by interferon-␥ mediated by the 42. Westerheide, S. D., and J. M. Boss. 1999. Orientation and positional mapping of transactivator gene CIITA. Science 265:106. the subunits of the multicomponent transcription factors RFX and X2BP to the major histocompatibility complex class II transcriptional enhancer. Nucleic Acids 58. Rigaud, G., F. Paiola, and R. S. Accolla. 1994. In vivo modification of major Res. 27:1635. histocompatibility complex class II DRA promoter occupancy mediated by the 43. Godthelp, B. C., M. C. J. A. Eggermond, A. Peijnenburg, I. Tezcan, AIR-1 trans-activator. Eur. J. Biochem. 24:2415. S. van Lierde, M. J. van Tol, J. M. Vossen, and P. J. van den Elsen. 1999. 59. Rigaud, G., A. De Lerma Barbaro, M. Nicolis, T. Cestari, D. Ramarli,

Incomplete T-cell immune reconstitution in two major histocompatibility com- A. P. Riviera, and R. S. Accolla. 1996. Induction of CIITA and modification of by guest on October 1, 2021 plex class II-deficiency/bare lymphocyte syndrome patients after HLA-identical in vivo HLA-DR promoter occupancy in normal thymic epithelial cells treated sibling bone marrow transplantation. Blood 94:348. with IFN-␥: similarities and distinctions with respect to HLA-DR constitutive B 44. Peijneburg, A., B. Godthelp, A. van Boxel-Dezaire, and P. J. van den Elsen. cells. J. Immunol. 156:4254. 1995. Definition of a novel complementation group in MHC class II deficiency. 60. Wright, K. L., K.-C. Chin, M. Linhoff, C. Skinner, G. Li, J. M. Boss, G. R. Stark, Immunogenetics 41:287. and J. P.-Y. Ting. 1998. CIITA stimulation of transcription factor binding to 45. Peijnenburg, A., M. C. J. A. van Eggermond, R. van den Berg, O. Sanal, major histocompatibility complex class II and associated promoters in vivo. Proc. J. M. J. J. Vossen, and P. J. van den Elsen. 1999. Molecular analysis of an MHC Natl. Acad. Sci. USA 95:6267.