Skip Navigation


Human Molecular Genetics Advance Access originally published online on March 17, 2004
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
13/9/967    most recent
ddh113v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (14)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Price, P.
Right arrow Articles by Christiansen, F. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Price, P.
Right arrow Articles by Christiansen, F. T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2004, Vol. 13, No. 9 967-974
DOI: 10.1093/hmg/ddh113
Human Molecular Genetics, Vol. 13, No. 9 © Oxford University Press 2004; all rights reserved

Polymorphisms at positions -22 and -348 in the promoter of the BAT1 gene affect transcription and the binding of nuclear factors

Patricia Price1,2,*, Agnes M.-L. Wong1,2, David Williamson1,2, Dominic Voon3, Svetlana Baltic3, Richard J.N. Allcock1,2, Alvin Boodhoo2,4 and Frank T. Christiansen1,2

1School of Surgery and Pathology, University of Western Australia, Nedlands 6009, Australia, 2Department of Clinical Immunology and Biochemical Genetics, Royal Perth Hospital, Perth 6001, Australia, 3GeneStream Pty Ltd, Western Australia and 4University of Mauritius, Reduit, Mauritius

Received January 19, 2004; Accepted March 9, 2004


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
BAT1 (D6S81E, UAP56) lies in the central MHC between TNF and HLA-B, a region containing genes that affect susceptibility to immunopathologic disorders. BAT1 protein may be directly responsible for the genetic association, as antisense studies show it can down-regulate inflammatory cytokines. Here we investigate polymorphisms at positions -22 and -348 relative to the BAT1 transcription start site. DNA samples from healthy donors were used to confirm haplotypic associations with the type 1 diabetes-susceptible 8.1 ancestral haplotype (AH; HLA-A1,B8,BAT1-22*C,BAT1-348*C,DR3 ) and the diabetes-resistant 7.1 AH (HLA-A3,B7,BAT1-22*G,BAT1-348*T,DR15). Alleles carried at BAT1-22 and -348 were in linkage disequilibrium. Electrophoretic mobility shift assays using nuclear proteins from T-cells (Jurkat and HT2), monocytes (THP1, U937) and epithelial cells (HeLa and MDA468) demonstrated DNA : protein complexes binding oligonucleotides spanning positions -22 and -348 on the 7.1 AH only. Competition assays, supershifts and molecular weight determinations suggest the complexes include the transcription factors YY1 (at -348) and Oct1 (at -22). Promoter activity was demonstrated using 520 bp and 336 bp fragments cloned from immediately upstream of the transcription start site and carrying all combinations of -22 and -348 alleles, suggesting an unidentified non-polymorphic sequence within 336 bp of the start site drives transcription. The 520 bp fragment of the BAT1 promoter cloned from the 8.1 AH was slightly less efficient than the equivalent from the 7.1 AH, whilst the reverse was observed with 336 bp fragments. This suggests BAT1 transcription on the 7.1 AH is modified by interactions involving DNA flanking positions -22 and -348.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The human BAT1 gene (D6S81E, UAP56) is a member of the DEAD-box family of ATP-dependent RNA helicases (1). It is located in the central major histocompatibility complex (MHC) on chromosome 6, in a region affecting several immunopathological disorders (2–4). Fine mapping of the gene(s) affecting disease is limited by linkage disequilibrium within the MHC. We have defined a group of six genes (LTA, LTB, TNF, IKBL, AT6PG and BAT1) which is conserved through modern human evolution (5) and are evaluating each constituent gene as candidate for the observed genetic associations. Here we describe our investigation of BAT1.

Our studies suggest that BAT1 affects a process relevant in disease. Monocyte and T-cell lines infected with retroviral constructs containing BAT1 antisense sequence produced higher levels of the acute phase cytokines TNF, interleukin-1 (IL-1) and IL-6 than cells infected with the vector alone. These results suggested that BAT1 protein is a negative regulator of inflammation (6). We also described 10 single nucleotide polymorphisms in the 1500 bp proximal promoter region of BAT1 in EBV-transformed B-cell lines homozygous for conserved MHC haplotypes (7). The coding region was monomorphic, except for a synonymous substitution at position +348.

As our aim is to identify polymorphisms that mediate susceptibility to type 1 diabetes, we use the 7.1 ancestral haplotype (AH; HLA-A3,B7,DR15), which is associated with dominant resistance, and the 8.1 AH (HLA-A1,B8,DR3), which confers susceptibility. We established that the first 520 bp of the BAT1 promoter allele 1 (7.1 AH) activates transcription more efficiently in Jurkat cells compared to allele 2 (8.1 AH haplotype). These alleles differ only at positions -22 and -348 in the 520 bp upstream of BAT1. Electrophoretic mobility shift assays (EMSA) demonstrated that both polymorphisms affect binding of nuclear proteins in Jurkat cells (7). Here we show that both loci contribute to the difference in transcriptional activity between the alleles carried on the 7.1 AH or 8.1 AH, and identify transcription factors binding differentially to the two alleles in cells of different lineages.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Haplotypic associations between BAT-22 and -348 in healthy donors
DNA from 296 donors from the West Australian Bone Marrow registry was genotyped for alleles of BAT1-22 and BAT1-348. Alleles carried at BAT1-22 and -348 were in significant linkage disequilibrium, with no donors homozygous for BAT1-22*CC/-348*TT (Table 1). Using a panel of workshop cell lines, we have shown previously that BAT1-22*C/-348*C is carried on the 8.1 AH and BAT1-22*G/-348*T is carried on the 7.1 AH. To confirm this association in a normal population, we examined the HLA-B and HLA-DR alleles carried by individuals who were homozygous for BAT1-22*C/BAT1-238*C or homozygous for BAT1-22*G/BAT1-348*T (Table 2). Of the 32 individuals homozygous for BAT1-22*C/BAT1-348*C, 19 carried HLA-B8 and HLA-DR3 including seven individuals homozygous for these alleles. Of the remaining 13, none carried HLA-B7 or DR15. Thus BAT1-22G occurs on the 8.1 AH and not on the 7.1 AH. Similarly, of the 16 individuals homozygous for BAT1-22*G/BAT1-348*T, 10 also carried HLA-B7 and HLA-DR15, including one individual who was homozygous. The only trace of an 8.1 AH was one donor carrying HLA-B8. Hence, BAT1-22*G/BAT1-348*T occurs on the 7.1 AH and not on the 8.1 AH.


View this table:
[in this window]
[in a new window]
  		Table 1. BAT1-22 and BAT1-348 are in linkage disequilibrium West Australian donorsa
 

View this table:
[in this window]
[in a new window]
  		Table 2. Alleles of BAT1-22 and BAT1-348 are associated with the 8.1 AH and 7.1 AH, respectively
 
Nuclear factors bind to sequences flanking BAT1-22*G and -348*T in multiple cell lines
Nuclear extracts from cells of six lineages were incubated with 32P-labelled probes spanning BAT1-22 and -348 (Fig. 1). Bands A and D were generated by the BAT1-22 probe using all cell lineages [T-cells (Jurkat and HT2), monocytes (THP1, U937) and epithelial cells (HeLa and MDA468)], as previously in Jurkat cells (7). Band A reflects specific binding to BAT1-22*G, since a 100-fold molar excess of unlabelled oligonucleotide inhibited binding of the labeled probe. Unlabelled probe carrying the alternative allele did not inhibit this band (data not shown). Band D was not allele-specific, so further analyses focused on Band A. Two intermediate bands (b and c) displayed variable intensity and were not uniformly inhibited by excess unlabelled probe in replicate studies (data not shown), so they were not investigated further.



View larger version (53K):
[in this window]
[in a new window]
  		Figure 1. Nuclear protein extracts from six cell lines of different lineage produce similar bands on EMSA. 32P-labeled BAT1 oligonucleotide probes were incubated with nuclear extracts from each cell line. Specificity was assessed by competition with 100-molar excess of unlabelled oligonucleotide carrying each allele (lanes labeled +). Specific bands are labeled in upper case (A, D and G).

 
Probes spanning BAT1-348 generated several bands (e,f,G,h). Band G was observed with extracts from all cell lines reacted with the BAT1-348*T probe, but not with BAT1-348*C. Incubation with excess unlabelled homologous oligonucleotide substantially reduced the intensity of Band G. Bands e, f and h were weak, variable and not reproducibly reduced by competition with excess unlabelled homologous oligonucleotide.

Nuclear factors binding to BAT1-22*G (Band A) and BAT1-348*T (Band G) can be inhibited by transcription factor consensus oligonucleotides
Jurkat nuclear extracts were probed using oligonucleotides containing the consensus-binding motif sequence for a series of transcription factors as unlabelled competitors (Fig. 2). GATA, NFAT and Oct1 consensus oligonucleotides inhibited Band A (seen with the BAT1-22*G probe). Incomplete competition occurred with the consensus {delta}EF1 and CREB oligonucleotides. Band G (seen with the BAT1-348*T probe) was inhibited by consensus NFAT and YY1 oligonucleotides.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 2. Transcription factor consensus oligonucleotides were used as a guide to the identity of Bands A and G. 32P-labeled BAT1-22*G or BAT1-348*T probes were incubated with Jurkat nuclear extracts. Identity was assessed by competition with 100-molar excess of unlabelled consensus oligonucleotide.

 
Molecular mass of proteins binding selectively to BAT1-22*G and -348*T
Jurkat nuclear extracts were incubated with the labelled oligonucleotides and then UV irradiated to cross-link proteins to the DNA. The samples were then denatured and separated by electrophoresis on 10% SDS–PAGE gels. Competition with excess unlabelled oligonucleotides was used to assess specificity.

The BAT1-22*G probe produced strong bands with approximate molecular weights of 66–70 and 95–100 kDa. These were not seen with the -22*C probe and were inhibited by excess unlabeled oligonucleotides (Fig. 3), so both may reflect components of Band A on the original EMSA (Fig. 1). Other bands were weaker, not allele-specific and/or not inhibited by unlabelled oligonucleotides. The BAT1-348*T probe also generated multiple bands (Fig. 3), but only the 68 kDa band was allele-specific and inhibited by unlabelled oligonucleotide. We suggest this protein represents Band G on the original EMSA (Fig. 1).



View larger version (62K):
[in this window]
[in a new window]
  		Figure 3. Molecular mass of Jurkat nuclear proteins binding DNA flanking BAT1-22 and BAT1-348 was assessed using 32P-labeled oligonucleotides incubated with nuclear extracts, UV-crosslinked and separated on a denaturing gel.

 
Oct1 and YY1 factors form sequence-specific complexes with the polymorphic promoter elements of the BAT1 gene
Using Jurkat cell nuclear extract and labeled BAT1-22*G probe, anti-Oct1 antibody substantially inhibited Band A and produced a low mobility complex (Fig. 4A, lane 4). Anti-Oct2 antibody had no effect (lane 5). Labeled consensus Oct1 oligonucleotide generated a complex equivalent to Band A (lane 6), which was inhibited by unlabeled BAT1-22*G oligonucleotide but not by BAT1-348*T. This band could be supershifted with anti-Oct1, suggesting Oct1 is an integral part of Band A—binding to the DNA if the G allele is present. Its molecular weight is consistent with the larger specific band demonstrated by UV-crosslinking (95–100 kDa; Fig. 3).



View larger version (69K):
[in this window]
[in a new window]
  		Figure 4. Identity of nuclear binding proteins was established using 32P-labeled BAT1-22*G, BAT1-348*T or specific transcription factor consensus oligonucleotides as probes, with Jurkat nuclear extracts. Binding profiles of Oct1 and BAT1-22 oligonucleotides were similar (A), as were profiles of YY1 and BAT1-348 oligonucleotides (C). YY1 and BAT1-22 probes produced distinct bands, but polyclonal anti-YY1 blocked Band A, so YY1 may form part of a large complex at this site (B). Bands produced by NFAT oligonucleotide were marginally inhibited by unlabelled BAT1-348, but not by any available antibody (see text; D). Other consensus oligonucleotides did not produce bands similar to BAT1-22 or -348 (E).

 
When the consensus YY1 oligonucleotide was used as a probe, the major complex was smaller than Band A (Fig. 4B, lane 6) and binding of labeled BAT1-22*G oligonucleotide was not inhibited by YY1 consensus sequence (Figs 2 and 4B, lane 3). However anti-YY1 antibody inhibited Band A (lane 4). Together these findings suggest YY1 (MW=68 kDa) is an integral part of Band A, which does not bind directly to the DNA. Its molecular weight is consistent with the smaller specific band demonstrated by UV-crosslinking (66–68 kDa; Fig. 3).

The complex generated with labelled YY1 consensus oligonucleotide resembled Band G generated with BAT1-348*T, and was partially blocked by competition with unlabelled BAT1-348*T oligonucleotide but not BAT1-22*G (Fig. 4C, lanes 7 and 8). Moreover anti-YY1 antibody inhibited Band G, producing a low mobility complex (Fig. 4C, lane 4). An unrelated antibody (anti-Sp1) lacked this activity (lane 5). This suggests YY1 is an integral part of Band G. The molecular weight of YY1 is consistent with the specific band demonstrated by UV-crosslinking (66–68 kDa; Fig. 3).

NFAT consensus sequence blocked Bands A and G (Fig. 2). When used as a probe this produced a band which corresponded to Band G, suggesting binding around -348*T. However the NFAT band was only marginally inhibited by excess BAT1-348*T and not affected by BAT1-22*G (Fig. 4D), so the inhibition of Band A is considered non-specific. An extensive series of antibodies was tested but did not cause a supershift using -348*T (data not shown) or NFAT consensus sequence (Fig. 4D) as a probe, so we were unable to confirm that NFAT is a component of Band G. Moreover our UV-crosslinking study (Fig. 3) suggests a single protein binds to -348*T, which appears to be YY1.

When consensus sequences for the other transcription factors able to inhibit binding to the BAT1 promoter (GATA, {delta}EF1 and CREB; Fig. 2) were used as probes, the bands did not correspond with those obtained with the BAT1-22*G probe (Band A) or BAT1-348*T probe (Band G) (Fig. 4E).

BAT1-22 and -348 affect transcription of reporter constructs
Four luciferase reporter constructs containing 520 bp of DNA upstream of the transcription start site were generated to assess all combinations of alleles at -22 and -348 (Table 1). Consistent with our previous studies (7), the relative luciferase activity of promoter with alleles (-22*C/-348*C) (i.e. from the 8.1 AH) was lower than that of a promoter carrying (-22*G/-348*T). However, all promoter fragments induced luciferase activity, suggesting non-polymorphic sequences within the promoter drive basal transcription. Moreover, luciferase activities induced by promoters carrying (-22*C/-348*T) or (-22*G/-348*C) were similar to that of promoters carrying (-22*G/-348*T) (Table 3). The simplest explanation for this is that binding of nuclear proteins to either site in the 7.1 AH promoter is sufficient to increase transcription above that seen with the 8.1 AH.


View this table:
[in this window]
[in a new window]
  		Table 3. Dual-Luciferase reporter activity generated by a 520 bp promoter fragment encompassing BAT1-22*C and -348*C is low in Jurkat T-cells
 
Lower reporter gene expression for the 520 bp BAT1 promoter fragment from the 8.1 AH was confirmed over an extensive time course by flow cytometry using HeLa cells transfected with fluorescent reporter constructs (Fig. 5, Exp. 1). To further address the contributions of the -22 and -348 sequences, 520 bp fragments of the 8.1 AH and 7.1 AH were compared with 336 bp fragments lacking BAT1-348. The 336 bp promoter from the 8.1 AH showed higher reporter activity than that from the 7.1 AH (Fig. 5, Exp. 2B). This finding was not expected as a follow-on from the data in Table 3 and may reflect the binding of unidentified proteins to non-polymorphic sequences affected by the truncation. Cell lineage may affect interactions between the -22 and -348 sequences, though the EMSA data (Fig. 1) and transfected 520 bp promoter fragments (Table 3 and Fig. 5, Exp. 1) would not support this.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 5. Activity of the BAT1 proximal promoter from the 7.1 and 8.1 AH was compared using HeLa cells transiently transfect with EGFP vectors. Activity is expressed relative to a control vector containing HcRed. Exp. 1: 520 bp promoter fragments assessed over 72 h after transfection; Exp. 2: 520 bp (A) and 336 bp (B) promoter fragments assessed over 48 h after transfection.

 
BAT1 mRNA is expressed in the cell types used to analyse transcription
BAT1 mRNA expression was assessed using real-time RT–PCR in Jurkat and HeLa cells and PBMC from single adult male donors homozygous for the 7.1 AH and 8.1 AH. Levels of expression were 146, 154, 103 and 128, respectively (mean of duplicate determinations expressed relative to ß-actin mRNAx10–3). Thus BAT1 is widely expressed, but the in vivo effects of promoter genotype require further analysis.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
This study demonstrates that combinations of alleles carried at -22 and -348 relative to the transcription start site of the BAT1 gene are in linkage disequilibrium (Table 1) and distinguish the 7.1 AH and 8.1 AH in a normal population (Table 2), although they are not unique to either AH. The alleles have moderate effects on transcription (Table 3 and Fig. 5). Since promoters from both alleles drive transcription efficiently, we conclude that the polymorphisms affect regulatory regions rather than sequences essential for transcription initiation. This conclusion is consistent with the high species conservation of BAT1 and the absence of coding polymorphisms, which suggest it is essential to life (1,8,9).

Binding of nuclear proteins to DNA flanking -22 and -348 was restricted to the 7.1 AH promoter, independent of cell lineage (Fig. 1). The proteins are tentatively identified as YY1 (binding directly at -348 and indirectly at -22) and Oct1 (binding directly at -22), based on competition with oligonucleotides carrying the consensus binding sequences, supershift assays and molecular weight determinations from protein : DNA UV-crosslinking (Figs 2–4). As the YY1 consensus oligonucleotide did not inhibit Band A, the supershift detected with YY1 antibody may be an artefact or YY1 (or an antigenically-related protein of similar molecular weight) may bind indirectly around -22. The polymorphic site at position -22 lies within the core consensus sequence for Oct1 (GCAAAT), but both alleles are discordant (GCACAT and GCAGAT). BAT1-348*T carries the core binding sequence for YY1 (CCAT ), whilst the equivalent sequence for BAT1-348*G is CCAC. This change would be expected to affect binding of YY1. Further analyses of YY1 binding using transcription factor databases produced widely divergent outcomes depending on the length of sequence and the database selected. For example, whilst the YY1 consensus sequence used here matched that used by Mordvinov et al. (10), multiple different consensus sequences are described by Zabel et al. (11). It is notable that several studies show interactions with YY1 at several sites in the same promoter (11,12), consistent with our finding that alleles at -22 and -348 affect BAT1 expression via YY1 (Table 3 and Fig. 5).

YY1 and Oct1 have complex roles in regulation of gene expression. YY1 (Yin Yang1) is best known for its ability to suppress or activate transcription depending on other features of the cell and/or promoter. For example; YY1 mediates transcriptional repression of the gene encoding interferon-{gamma} by inhibiting AP1 binding in Jurkat human T-cells (12), and can function as a transcriptional activator for interferon-{gamma} by interacting with NFAT in primary murine splenocytes (13). Interactions between Oct1 and YY1 have been implicated in the down regulation of IL-5 in T-cells (10). The possibility that YY1 and NFAT may interact at position -348*T to produce Band G (Figs 2 and 4D) requires further study.

In conclusion, our programme of research has defined a series of criteria required to evaluate polymorphisms that may be responsible for associations between conserved MHC haplotypes and complex disease traits (2). Through recombinant mapping we established that BAT1 lies in a region of the MHC which contains genes that influence the development or severity of several immunopathological conditions. We then showed BAT1 had a function relevant to disease, specifically as a negative regulator of inflammatory cytokines (6). Finally, we showed that transcription driven by the BAT1 promoter transfected into Jurkat cells was lower on a disease-susceptible haplotype (7). Here we use healthy homozygous donors to show that carriage of BAT1-22*C/-348*C or -22*G/-348*T marks haplotypes with known disease associations. We established that the loci have a concerted effect on transcription in Jurkat and HeLa cells, and show these cells and PBMC express BAT1 mRNA. Transcription factors likely to mediate differences in transcription were tentatively identified in Jurkat cells as YY1 and Oct1, with similar EMSA profiles suggesting this is common to other cell lineages. As YY1 and Oct1 are involved in immunological processes, our findings support and extend a model in which BAT1 affects immunopathological disease.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Genotyping
Healthy donors from the Western Australian Bone Marrow Registry were used to establish haplotypic associations. HLA typing and DNA extractions were performed as described previously (3). Alleles carried at -22 and -348 were determined by Amplifluor methodology at the Centre Nationale de Genotypage (Paris, France) (R.J.N. Allcock et al., submitted for publication).

Electrophoretic mobility shift assays (EMSA)
EMSA were performed using single-stranded oligonucleotides (Table 4; GeneWorks, South Australia and Santa Cruz, USA), annealed and gel purified according to standard protocols (7). T4 polynucleotide kinase was used to label oligonucleotides with [{gamma}-32P]-ATP. Cell extracts were prepared as described by Schreiber et al. (14) with minor modifications: Protease inhibitor cocktail (Boehringer Mannheim, Germany), 1 mM Na3VO4 (Sigma, USA) and 0.5 mM DTT (Sigma, USA) were added to reaction buffers prior to lysis. Protein concentrations were determined using a DC Protein Assay (Bio-Rad, USA).


View this table:
[in this window]
[in a new window]
  		Table 4. Oligonucleotides used as EMSA probes or competitors
 
Standard EMSA binding reactions contained 3 µg nuclear extract, 1 µg poly(dI-dC) (Pharmacia, USA), 12 mM HEPES (pH 7.9), 60 mM KCl, 0.1 mM EDTA, 12% glycerol, 1 mM DTT, 1 mM Na3VO4, Protease Inhibitor Cocktail and 25 fmol 32P-labelled double–stranded oligonucleotide probe. Nuclear extracts were incubated in the reaction buffer, followed by 32P-labelled DNA probe (each 30 min on ice). DNA–protein complexes were resolved by polyacrylamide gel electrophoresis (PAGE). X-ray films were exposed with intensifying screens at –70°C. Competition assays utilized 100-fold molar excess of unlabeled double-stranded oligonucleotides, added to the binding reaction 30 min before the labelled oligonucleotide. Super-shift assays utilized 1–2 µl antibody (100 µg/ml; Santa Cruz, USA), added to the binding reaction 60 min before the labelled DNA probe.

To determine the molecular masses of nuclear factors binding to the BAT1 promoter, labelled oligonucleotides and nuclear extracts were mixed and UV-irradiated (Strata-linker; USA) at 999 mV for 8 min. Standard SDS–PAGE loading dye was added for denaturation at 95°C for 5 min. UV-crosslinked DNA : protein complexes were separated on 10% SDS–PAGE gels, alongside 14C-RainbowTM protein molecular weight markers (Amersham, UK).

Reporter constructs
Inserts carrying -22*G with -348*T (7.1 AH), -22*C with -348*C (8.1 AH) and -22*G with -348*C (18.1 AH; HLA-B18,DR15) were generated by PCR amplification of genomic DNA from homozygous EBV-transformed B-cell lines carrying these haplotypes (7). Amplicons were purified and directionally ligated into the multiple cloning site of a promoter-less luciferase-encoding plasmid (pGL3-enhancer, Promega, USA). Inserts carrying -22*C with -348*T were generated by BglII/BglI digestion and re-ligation of DNA fragments derived from the 8.1 AH and 7.1 AH, for expansion in the same vector. Reporter constructs were sequenced using dye-terminator chemistry to confirm their identity. DNA concentration and purity of maxiprep DNA samples were measured spectrophotometrically and confirmed by gel analysis.

The 520 bp BAT1 promoter fragments were removed from the pGL3-luciferase vectors by digestion with SacI (upstream) and BglII (downstream) and cloned into the BTG1N4 vector, which contains a destabilized EGFP reporter (Gene Stream Pty Ltd, Australia) (D. Voon et al., manuscript in preparation). Truncation of the BAT1 promoter fragments to 336 bp required introduction of a further SacI restriction site by site-directed mutagenesis. Complementary oligonucleotide primers with the sequences: 5'-GGAGGGGTGCCTgagctcACAGAGAAGACCTGCG-3' and 5'-CGCAGGTCTTCTCTGTgagctcAGGCACCCCTCC-3' (Proligo, Australia) were modified using the QuickChangeTM (STRATAGENE, USA) site-directed mutagenesis kit. Briefly, 0.8 µM primer, 20 mM Tris–HCl (pH 8.0), 2 mM MgSO4, 2 mM dNTP, 1.25 U Platinum Pfx DNA polymerase (Invitrogen, USA) and 200 ng parental template DNA (BAT7.1G1N4 or BAT8.1G1N4) were reacted in a final volume of 25 µl. The reaction mixture is heated to 95°C for 10 min, followed by 18 cycles of 95°C for 30 s, 52°C for 1 min and 68°C for 15 min, and extension at 68°C for 15 min. Following amplification, the methylated parental templates were removed by DpnI (20 U, New England Biolabs, USA) digestion at 37°C for 2 h. The resulting DNA (2.5 µl) was introduced into Escherichia coli by standard heat-shock transformation. To generate a plasmid lacking BAT1-348, the -520 to -336 region of the BAT1 promoter was removed by digestion and re-ligation in the presence of SacI.

Culture and transfection of mammalian cell lines
Jurkat, THP1, U937, HeLa and MDA468 cells from the American Type Culture Collection (USA) were cultured in RMPI-1640 medium with 10% foetal calf serum. Peripheral blood mononuclear cells (PBMC) were isolated using Ficoll-Hypaque density gradient separation. BAT1 reporter plasmids were co-transfected with a control reporter marker into Jurkat or Hela cells using Fugene (Boehringer Mannheim, Germany) or Lipofectamine (Invitrogen, USA), according to manufacturers' instructions

For luciferase reporter vectors, cells were harvested and resuspended in 100 µl passive lysis buffer. Cell lysates were cleared by centrifugation and 40 µl aliquots assayed using the Dual Luciferase Assay Kit (Promega, USA). Light emission (wavelength 300–650 nm) was measured in a Victor 1420 multilabel reader for 10 s (Wallac, USA). Values were normalized for transfection efficiency by dividing test reporter activity by values obtained for Renilla luciferase activity. The promoter less control vector (pGL3-enhancer; Promega, USA) and a vector containing an SV40 promoter (pGL3-control; Promega, USA) were included as controls.

For the vectors containing EGFP, HeLa cells (50–70% of confluency in 75 cm2 flask) were co-transfected with 5 µg DNA using Lipofectamine 2000 (Invitrogen, USA). As an internal control of transfection efficiency, 1.5 µg BTR1N4 vector (D. Voon et al., manuscript in preparation) expressing HcRed was co-transfected with 3.5 µg of each BAT1 promoter construct. Six hours after transfection, cells were split into 6 cm petri dishes, one per time point. Cells were harvested at the required times by trypsinization. Expression of EGFP and HcRed was measured on a flow cytometer (Coulter, USA) and analysed using FlowJo software (Tree Star, USA).

Quantification of mRNA by real-time PCR
RNA was extracted using the RNeasy total RNA kit for cDNA synthesis using Omniscript (Qiagen, USA). PCR was performed on a LightCyclerTM (Roche) in 20 µl volumes containing 1.25 mM dNTP, 20 pmol each primer, 0.25 mg/ml bovine serum albumin and 1.5 units Platinum Taq DNA polymerase. Primer sequences were 5'-GATGACCCAGATCATGTTTGA-3' and 5'-GACTCCATGCCCAGGAAGGAA-3' for ß-actin and 5'-AGAGGCTTTCTCGGTATCA-3' and 5'-CATCAGGCAGCTCACTAA-3' for BAT1. The PCR protocol comprised 95°C for 5 min, and 35 cycles of 95°C for 2 s, 62–64°C for 10 s and 72°C for 20 s. Product formation was monitored using SYBR Green fluorochrome (Sigma, USA), generation of a single product was confirmed from the melting curve and quantitation utilized standard curves generated from serial 10-fold dilutions of purified PCR products. Ratios of BAT1 to ß-actin amplicons were multiplied by 1000 to yield whole numbers.


   ACKNOWLEDGEMENTS
 
The authors thank Dr Gretchen Schwenger, Professor Colin Sanderson and Dr Deirdre Coombe (Molecular Immunology, Curtin University, Australia) for training and reagents for EMSA; Professor Ivo Gut (Centre Nationale de Genotypage, Paris) and Ms Lydia Windsor for genotyping; Ms Meredith Glasson for technical assistance and Dr John Daly (GeneStream Pty Ltd) for the fluorescent expression vectors. M.G., D.W. and A.W. were supported by the National Health and Medical Research Foundation of Australia. This is publication 2003-31 of the Dept of Clinical Immunology and Biochemical Genetics, Royal Perth Hospital.


   FOOTNOTES
 
* To whom correspondence should be addressed at: Department of Clinical Immunology and Biochemical Genetics, Royal Perth Hospital, Wellington Street, Perth, WA 6001, Australia. Tel: +61 892240378; Fax: +61 892240204; Email: pprice{at}cyllene.uwa.edu.au


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Peelman, L.J., Chardon, P., Nunes, M., Renard, C., Geffrotin, C., Vaiman, M., Van Zeveren, A., Coppieters, W., van de Weghe, A., Bouquet, Y. et al., (1995) The BAT1 gene in the MHC encodes an evolutionary conserved putative nuclear RNA helicase of the DEAD family. Genomics, 26, 210–218.[CrossRef][Web of Science][Medline]

  2. Price, P., Witt, C., Allcock, R.J.N., Sayer, D., Garlepp, M., Kok, C.C., French, M.A.H., Mallal, S. and Christiansen, F.T. (1999) The genetic basis for the association of the 8.1 ancestral haplotype (A1, B8, DR3) with multiple immunopathological diseases. Immunogenetics, 167, 257–274.

  3. Cheong, K.Y., Allcock, R.J.N., Eerligh, P., Witt, C.S., Christiansen, F.T., McCann, V. and Price, P. (2001) Localization of central MHC genes influencing type 1 diabetes. Hum. Immunol., 62, 1363–1370.[CrossRef][Web of Science][Medline]

  4. Ota, M., Katsuyama, Y., Kimura, A., Tsuchiya, K., Kondo, M., Naruse, T., Mizuki, N., Itoh, K., Sasazuki, T. and Inoko, H. (2001) A second susceptibility gene for developing rheumatoid arthritis in the human MHC is localized within a 70-kb interval telomeric of the TNF genes in the HLA class III region. Genomics, 71, 263–270.[CrossRef][Web of Science][Medline]

  5. Price, P., Bolitho, P., Jaye, A., Glasson, M., Yindom, L.M., Sirugo, G., Chase, D., McDermid, J. and Whittle, H. (2003) A Gambian TNF haplotype matches the European HLA-A1,B8,DR3 and Chinese HLA-A33,B58,DR3 haplotypes. Tissue Antigens, 62, 72–75.[CrossRef][Web of Science][Medline]

  6. Allcock, R.J.N., Williams, J.H. and Price, P. (2001) The central MHC gene, BAT1, may encode a protein that down-regulates cytokine production. Genes Cells, 6, 487–494.[Abstract]

  7. Wong, A.M.L., Allcock, R.J.N., Cheong, K.Y.M., Christiansen, F.T. and Price, P. (2003) Alleles of the proximal promoter of BAT1, a putative anti-inflammatory gene adjacent to the TNF cluster, reduce transcription on a disease-associated MHC haplotype. Genes Cells, 8, 403–412.[Abstract]

  8. Allcock, R.J.N., Price, P., Gaudieri, S., Leelayuwat, C., Witt, C.S. and Dawkins, R.L. (1999) Characterisation of the human central MHC gene, BAT1: genomic structure and expression. Exp. Clin. Immunogenet., 16:, 98–106.[CrossRef][Web of Science][Medline]

  9. MacMorris, M., Brocker, C. and Blumenthal, T. (2003) UAP56 levels affect viability and mRNA export in Caenorhabditis elegans. RNA, 9, 847–857.[Abstract/Free Full Text]

  10. Mordvinov, V.A., Schwenger, G.T., Fournier, R., De Boer, M.L., Peroni, S.E., Singh, A.D., Karlen, S., Holland, J.W. and Sanderson, C.J. (1999) Binding of YY1 and Oct1 to a novel element that downregulates expression of IL-5 in human T cells. J. Allergy Clin. Immunol., 103, 1125–1135.[CrossRef][Web of Science][Medline]

  11. Zabel, M.D., Wheeler, W., Weis, J.J. and Weis, J.H. (2002) Yin Yang 1, Oct1, and NFAT-4 form repeating, cyclosporin-sensitive regulatory modules within the murine CD21 intronic control region. J. Immunol., 168, 3341–3350.[Abstract/Free Full Text]

  12. Ye, J., Cippitelli, M., Dorman, L., Ortaldo, J.R. and Young, H.A. (1996) The nuclear factor YY1 suppresses the human gamma interferon promoter through two mechanisms: inhibition of AP1 binding and activation of a silencer element. Mol. Cell. Biol., 16, 4744–4753.[Abstract]

  13. Sweetser, M.T., Hoey, T., Sun, Y.L., Weaver, W.M., Price, G.A. and Wilson, C.B. (1998) The roles of nuclear factor of activated T cells and yin-yang 1 in activation-induced expression of the interferon-gamma promoter in T cells. J. Biol. Chem., 273, 34775–34783.[Abstract/Free Full Text]

  14. Schreiber, E., Matthais, P., Muller, M.M. and Schaffner, W. (1989) Rapid detection of octamer binding proteins with ‘mini-extracts’ prepared from a small number of cells. Nucl. Acids Res., 17, 6419.[Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Hum Mol GenetHome page
W. Koch, P. Hoppmann, E. Michou, V. Jung, A. Pfeufer, J. C. Mueller, C. Gieger, H.-E. Wichmann, T. Meitinger, A. Schomig, et al.
Association of variants in the BAT1-NFKBIL1-LTA genomic region with protection against myocardial infarction in Europeans
Hum. Mol. Genet., August 1, 2007; 16(15): 1821 - 1827.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
13/9/967    most recent
ddh113v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (14)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Price, P.
Right arrow Articles by Christiansen, F. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Price, P.
Right arrow Articles by Christiansen, F. T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?