- Email: [email protected]
Hereditary Hemochromatosis: Generation of a Transcription Map within a Refined and Extended Map of the HLA Class I Region
GENOMICS
31, 319–326 (1996) 0054
ARTICLE NO.
Hereditary Hemochromatosis: Generation of a Transcription Map within a Refined and Extended Map of the HLA Class I Region ANGELA TOTARO, JOHANNA M. ROMMENS,*,† ANNA GRIFA, CLAUDIO LUNARDI,‡ MASSIMO CARELLA, JACK J. HUIZENGA,* ANTONELLA ROETTO,§ CLARA CAMASCHELLA,§ GIORGIO DE SANDRE,‡ AND PAOLO GASPARINI1 Servizio di Genetica Medica, IRCCS-Ospedale ‘‘CSS,’’ I-71013 San Giovanni Rotondo, §Dipartimento di Scienze Biomediche ed Oncologia Umana, Ospedale S. Luigi, Orbassano, 10100 Turin, and ‡Istituto di Clinica Medica, Universita’ di Verona, 35100 Verona, Italy; *Department of Genetics, Research Institute, The Hospital for Sick Children and †Department of Molecular and Medical Genetics, University of Toronto, Toronto, M5G 1X8 Canada Received August 21, 1995; accepted November 10, 1995
Hereditary hemochromatosis, a common severe inherited disease, maps to the short arm of chromosome 6 close to the HLA-A locus. Recently, linkage data on Italian and French populations confirmed this location, while a similar analysis on Australian and British populations located the gene closer to D6S105, a marker residing telomeric of HLA-A. To increase our knowledge on the region of highest linkage disequilibrium in our population and possibly to identify the disease gene, a 1.2-Mb detailed physical and transcription map was generated, spanning the HLA class I region. Thirty-eight unique cDNA fragments, retrieved following the hybridization of immobilized YACs to primary pools of cDNAs prepared from RNA of fetal brain, adult brain, liver, placenta, and the CaCo2 cell line, were characterized. All cDNA fragments were positioned in a refined and extended map of the human major histocompatibility complex spanning from HLAE to approximately 500 kb telomeric of HLA-F. The localization of known genes was refined, and a new gene from the RNA helicase superfamily was identified. Overall, 14 transcription units in addition to the HLA genes have been detected and integrated in the map. Thirteen cDNA fragments show no similarity with known sequences and could be candidates for the disease. Their characterization and assessment for involvement in hemochromatosis are still under investigation. Seven new polymorphisms, some tightly linked to the disease, were also identified and localized. q 1996 Academic Press, Inc.
INTRODUCTION
Hemochromatosis (HFE) is a common autosomal recessive disease characterized by chronic dietary iron 1 To whom correspondence should be addressed at Servizio di Genetica Medica, IRCCS-Ospedale ‘‘CSS,’’ I-71013 San Giovanni Rotondo, Italy. Telephone: /39 882 410825. Fax: /39 882 411616.
overload. The disease prevalence is estimated to be 2– 5/1000 in the Caucasian population, with a corresponding carrier frequency of 0.045–0.071 (McKusick, 1994). The underlying biochemical defect of the disease is unknown, but the gene has been mapped to the short arm of chromosome 6, containing the histocompatibility antigen (HLA) class I region (Simon et al., 1987). Linkage disequilibrium studies in French and Italian populations initially suggested a narrow candidate region for the HFE gene, including a 400-kb stretch of DNA flanking the HLA-A gene, spanning from I.82 to HLA-F (Boretto et al., 1992; Gasparini et al., 1993; Yaouanq et al., 1994). However, association studies, performed in the Australian population (Jazwinska et al., 1993), have extended this region to the D6S105 locus (Weber et al., 1991), which is estimated to be at least 2 cM telomeric of HLA (Stone et al., 1994; Volz et al., 1994). We recently reported new polymorphisms and sequence tagged sites from the HLA class I region and identified a polymorphic marker, named Y52, which showed the highest allelic association in our population (Totaro et al., 1995). Allele 1 of Y52, together with HLA-A serotype A3 and D6S265 allele 1, mark the ancestral HFE haplotype in our population. Y52 and D6S265 were positioned very close to the HLA-A locus (Totaro et al., 1995). These data support this area as being the most likely candidate region for the disease gene. Several novel coding sequences have already been obtained from this region (El Kahloun et al., 1993; Wei et al., 1993; Goei et al., 1994), but despite the extensive effort, there has as yet been no indication that these coding sequences are hemochromatosis candidates. We have extended the cloned region by isolation of an overlapping YAC clone, 169H11, and applied the cDNA selection technique to the entire region of 1.2 Mb containing the HLA-A locus, spanning from 90 kb proximal to HLA-E to about 550 kb distal to HLA-F. Moreover, to refine and extend the physical and transcription map beyond the distal part of the human HLA
319
/ m4990$3868
01-04-96 18:28:00
gnma
0888-7543/96 $12.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AP-Genomics
320
TOTARO ET AL.
class I region and to gain insight into its organization, we have performed long-range restriction pulse-field gel mapping by electrophoresis (PFGE) of yeast artificial chromosome clones using both known markers from this region and the new clones obtained by cDNA selection. MATERIALS AND METHODS Preparation and isolation of genomic DNAs. YAC 225B1 has been obtained from CEPH, while YAC 169H11 has been isolated from the Mega YAC CEPH library using a fragment located on the telomeric end of YAC 225B1 as a probe. YAC clones were isolated, grown on plates with medium lacking tryptophan and uracil, and used to prepare total chromosome DNA, as previously described (Rommens et al., 1994). After separation by PFGE, the artificial chromosomes were transferred to nylon membrane (Hybond-N, Amersham) and immobilized by UV-crosslinking for the selection experiments (Rommens et al., 1994). Genomic DNAs were isolated from human leukocytes or from rodent and rodent hybrid cell lines by established procedures (Miller et al., 1988). The mouse–human hybrid cell line GM10629 was obtained from the NIGMS Human Genetic Mutant Cell Repository (Camden, NJ). Restriction digestions were carried out according to the supplier’s recommendations. Electrophoresis, blotting, and hybridization were carried out according to standard procedures (Sambrook et al., 1989). After hybridization, the blots were washed (0.21 SSC with 0.1% SDS) at 607C and exposed with autoradiography for 2–72 h. For long-range physical mapping, high-molecular-weight DNA embedded in agarose of whole yeast chromosomes was digested with rare-cutting restriction enzymes. Products were separated by PFGE and transferred to nylon membrane for hybridizations as previously described (Church and Gilbert, 1984). Specific probes for the HLAE (Totaro et al., 1995), HLA-B30, HLA-A, HLA-F, MOG (Amadou et al., 1995), D6S131 (Amadou et al., 1995), and RFP (Amadou et al., 1995) loci were used to establish the orientation and framework for the physical map. The HLA probes were generated using gene-specific oligonucleotides and the polymerase chain reaction (PCR). cDNA preparation and selection. cDNA selection was carried out as previously described (Rommens et al., 1994). Briefly, randomly primed cDNA was prepared from poly(A)/ RNA of fetal liver, fetal brain, pancreas, and placenta tissues and from total RNA of adult liver and of the CaCo2 cell line (ATCC HTB 37). Selection was carried out for two consecutive rounds of hybridization to immobilized artificial chromosome DNA of 225B1 and 169H11. Hybridizing cDNA was collected, amplified, and cloned as described (Rommens et al., 1993, 1994). Characterization of retrieved cDNAs. Two hundred eighty-eight individual colonies were picked and gridded onto an agar plate containing 100 mg/ml ampicillin for ordering, prescreening, and storage. Initial analysis of cDNA clones included a prescreening for ribosomal sequences. Approximately 8% of the clones were eliminated because they hybridized with radiolabeled cDNA obtained from total RNA. Plasmids from 20 to 40 clones from each selection experiment that did not hybridize in the prescreenings were isolated for further analysis. Of clones tested, approximately 70% mapped appropriately to the starting YAC clones. The membranes of the grids were also repeatedly hybridized with individual cDNA clones to minimize the number of overlapping or redundant clones. RNA preparation and hybridization. Total RNA was isolated from frozen tissues (liver, duodenal mucosa, and placenta), cultured cells (endothelial cells and fibroblasts), and freshly isolated peripheral blood mononuclear cells by the guanidium isothiocyanate method (Chomczynski and Sacchi, 1987), size-fractionated by electrophoresis on 1% agarose–formaldehyde gels, and blotted onto nylon membranes. Hybridization with 32P-labeled probes was carried out at 427C for at least 12 h in a formamide-based solution. Membranes
/ m4990$3868
01-04-96 18:28:00
gnma
were then washed twice in 51 SSPE (11 SSPE: 360 mM NaCl, 20 mM Na2HPO4 , 2 mM EDTA) at 427C for 20 min, twice in 11 SSPE containing 0.1% SDS at 427C for 20 min, and exposed to HyperfilmMP (Amersham, UK). Hybridization with b-tubulin and/or -actin was performed to verify the integrity of RNA. DNA sequencing and sequence analysis. The sequences of both strands of the retrieved cDNAs were obtained with the dideoxy chain termination method (Sanger et al., 1977) using the Taq dye primer cycle sequencing kit (ABI) and the 373A DNA sequencer (ABI). The resulting sequences were then analyzed by comparison with the EMBL and GenBank Data Banks (PCGENE, Intelligenetics). RNA single-strand conformation polymorphism. Oligonucleotides for PCR primers corresponding to the first four clones isolated (GT253, GT256, GT257, and GT260) were designed. The T7 phage promoter sequence was incorporated at the end of one of each set as previously described (Sarkar et al., 1992). Annealing temperatures for amplification and length of each fragment are reported in Table 3. PCR was performed on 500 ng genomic DNA from 20 unrelated individuals according to standard protocols. Subsequent transcription was carried out as previously described (Bisceglia et al., 1994). An aliquot of 4.5 ml of the transcription reaction was then loaded onto a 6.5% nondenaturing polyacrylamide gel. Electrophoresis was performed at 30 W constant power for 13 h. After electrophoresis, the gel was dried and subjected to autoradiography for 12 h. Fragments with electrophoretically altered migration patterns were then sequenced with the dideoxy chain termination method (Sanger et al., 1977) as described above. When a polymorphism was detected, frequencies of alleles were assessed in a sample of 50 affected patients and their relatives attending our Centers. The statistical significance of linkage disequilibrium with the disease was tested by utilizing the x2 test of homogeneity for each marker and then standardized to obtain the d coefficient whose range is in the interval 0– 1, to measure the association. The d coefficient is defined as q
(ad 0 bc)/ (a / b)(c / d)(a / c)(b / d), where a and b are the frequencies of allele 1 in affected and unaffected chromosomes, respectively, and c and d are the frequencies of allele 2 in the same sets of chromosomes.
RESULTS
Thirty-eight unique cDNA clones were selected, sequenced, and verified to map to the YAC clones 225B1 or 169H11. To confirm the assignment to chromosome 6, each cDNA was hybridized to total human DNA and to a mouse hybrid cell line that contains chromosome 6 as its only human material. The pattern of hybridization (i.e., number of distinct EcoRI restriction fragments on chromosome 6 and on total human genomic DNA) is indicated in Table 1. For several clones, including GT256, GT511, and GT512, among others, only a single EcoRI restriction fragment was clearly detectable on chromosome 6 DNA, while a band of identical size and additional bands were present in genomic DNA. The DNA sequence of each clone was analyzed for coding potential and for alignment to the public databases and to each other. Thirteen corresponded to a new sequence, while the remaining 25 could be assigned to six subgroups that shared overlapping sequence or that were similar to previously known genes. One clone (GT554) showed homology with an EST (R85804 from the Washington University-Merck EST
AP-Genomics
321
TRANSCRIPTION MAP OF THE HLA CLASS I REGION
TABLE 1 cDNA Selection Results GT clone
Size
Chromosome 6 bands
Genomic bands
GT253 GT257 GT260 GT469 GT475 GT478 GT483 GT511 GT512 GT545 GT546 GT554 GT555 GT470 GT472 GT479 GT513 GT549 GT548 GT256
647 395 672 644 919 701 722 520 636 611 501 387 397 445 390 610 422 302 498 407
1 1 2 1 1 1 1 1 1 1 3 1 1 1 1 1 1 1 3 1
1 1 2 1 1 1 1 3 3 1 3 1 1 2 2 2 2 2 3 2
GT258
420
Several
Several
GT259 GT474
418 611
Several Several
Several Several
GT482
701
Several
Several
GT515 GT481
622 653
Several Several
Several Several
GT473
521
Several
Several
GT289
359
1
Several
GT477
409
1
4
GT486
510
1
4
GT510
711
1
4
GT254 GT291 GT509
610 612 529
2 2 2
Several Several Several
GT547
335
2
Several
GT480
572
1 faint
Several
GT514
364
1 faint
Several
GT550
393
1 faint
Several
Remarks New sequence EMBL Acc. No. X90532 New sequence EMBL Acc. No. X90534 New sequence EMBL Acc. No. X90535 New sequence EMBL Acc. No. X90536 New sequence EMBL Acc. No. X90537 New sequence EMBL Acc. No. X90538 New sequence EMBL Acc. No. X90539 New sequence EMBL Acc. No. X90540 New sequence EMBL Acc. No. X90541 New sequence EMBL Acc. No. X90542 New sequence EMBL Acc. No. X90543 New sequence EMBL Acc. No. X90544 EST (R85804) New sequence EMBL Acc. No. X90545 Overlap with GT256 Overlap with GT256 Overlap with GT256 Overlap with GT256 Overlap with GT256 Overlap with GT546 EMBL Acc. No. X90533 Identical to HSDNAHCGV gene from base 2856 to 3263 Identical to HLA-B clone PTMHCAA from base 1043 to 1461 Identical to HLA-E clone HSHLA1EA from base 600 to 994 Identical to MHC clone HSMHCFAN4 from base 1 to 135; chimeric clone Similar to HLA-B clone HSMHB17W from base 2577 to base 3269 Similar to HLA-A2 clone HSHLAA2 from base 163 to 788 Similar to MHC clone HSMHCFANA from base 3 to 120; chimeric clone Similar to MHC clone HSMHCFANB from base 1 to 120; chimeric clone with GT256 Identical to TB-3 gene clone HSTB31A from base 582 to 937 Identical to TB3-1 gene clone HSTB31A from base 1934 to 2252 Identical to TB3-1 gene clone HSTB31A; entirely contained in clone GT510 Identical to TB3-1 gene clone to HSTB31A from base 1087 to 1794 Identical to TC4 gene from base 323 to base 651 Identical to TC4 gene from base 323 to base 653 Identical to human S17 ribosomal protein gene clone HSRPS17 from base 3 to 477 Identical to human S17 ribosomal protein gene; entirely contained in clone GT509 Identical to RCK/RNA helicase clone HSRCK from base 1176 to 1756 Identical to RCK/RNA helicase clone HSRCK from base 1018 to 1407 Identical to RCK/RNA helicase clone HSRCK from base 1010 to 1425
Note. Summary of selected cDNA clones. The first 14 gene fragments correspond to new sequences, while the remaining ones shared overlapping sequence and/or were homologous to previously known genes.
Project). A complete summary of the results obtained is reported in Table 1. cDNAs of Known Human Genes Twenty-five clones shared overlapping sequence and/ or were homologous to already known genes, as de-
/ m4990$3868
01-04-96 18:28:00
gnma
scribed in Table 1. In particular, 7 clones (GT258, GT259, GT473, GT474, GT481, GT482, and GT515) showed identity or near identity with HLA class I sequences. Four clones (GT289, GT477, GT486, and GT510) showed complete sequence alignment with different portions of a TB3-1 gene, a member of a human
AP-Genomics
322
TOTARO ET AL.
FIG. 1. The physical and transcription map of the HLA class I region from HLA-E to D6S131. B, M, N, and R correspond to the BssHII, MluI, NotI, and NruI sites, respectively. The isolated cDNAs and the class I genes were positioned by PFGE. The relative position of each gene fragment within a given restriction fragment interval is not known. The numbering above the marked physical intervals indicates the 14 transcription units identified.
gene family with significant homology to yeast omnipotent suppressor 45 (Grenett et al., 1992). Two other clones (GT254 and GT291) were identical to the TC4 gene sequences (Drivas et al., 1990), a member of the family of human ras-like proteins. Two clones (GT509 and GT547) were identical to the S17 human ribosomal protein (Chen et al., 1986). This gene is strongly conserved through evolution and encodes a product that is fundamental to the translational apparatus. Three additional clones (GT480, GT514, and GT550) showed similarity with the RCK/RNA helicase gene (Lu and Yunis, 1992). The hybridization of these selected cDNA fragments detected only one faint band on chromosome 6 DNA, but several additional bands were observed in genomic DNA. Since the RCK/RNA helicase has been localized to band q23.3 of chromosome 11, our results are consistent with the identification of a new member of the expanding superfamily of RNA helicases. These genes are involved in specialized RNA functions and contain a series of 10 motifs that have been evolutionarily conserved. One clone (GT256) showed complete sequence alignment with the HCG-V gene, a gene recently submitted to the EMBL Data Bank (Gifon et al., Accession No. X89902). While all described clones are unique, five clones (GT470, GT472, GT479, GT513, and GT549) shared overlapping sequences with GT256, while one (GT548) overlapped with clone GT546. Physical Mapping A refined and extended physical map was obtained using both known markers from the HLA class I region and the new clones obtained by cDNA selection. It spans from about 90 kb centromeric of the HLA-E locus to approximately 550 kb telomeric of HLA-F (Fig. 1).
/ m4990$3868
01-04-96 18:28:00
gnma
The HLA-E locus resides in a 60-kb NotI–MluI fragment on YAC 225B1, while HLA-A is located in a NotI– BssHII fragment of 75 kb on both YAC225B1 and YAC169H11. PCR and PFGE results demonstrated that both HLA-F and MOG loci reside only on YAC 169H11. They are located in a 260-kb NotI fragment and must be contained in the region between the terminal right end of the YAC 225B1 and the following NotI site on YAC 169H11. The D6S131 probe hybridizes to a telomeric NruI fragment of 155 kb of 169H11. We attempted to integrate the map with other markers known to be localized telomeric of HLA, including RFP (Amadou et al., 1995) and D6S105 (Stone et al., 1994), using PCR analysis, but they were absent from YAC169H11, which extends most telomeric. The 13 clones with novel sequences have been localized by PFGE to their cognate genomic sequences on YAC 225B1 and YAC 169H11. They were found to be distributed across the YACs. The relative positions of clones that shared common restriction fragments could not be ordered. Among the clones localized on the centromeric side of HLA-A, two (GT260 and GT478) map centromeric to a BssHII site on the far left end of YAC 225B1, while those corresponding to the TC4 gene were within the same NotI–MluI fragment that contained the HLA-E locus. The others were located in the same 275-kb NotI fragment. The relative positions of the clones, from the centromere to the telomere, were GT257 in a NotI–BssHII fragment of 105 kb, GT469 in a MluI–BssHII fragment of 50 kb, and GT483 and GT554 in a BssHII–MluI fragment of 75 kb. Three clones (GT253, GT475, and GT555), plus those corresponding to the TB3-1 and HCG-V genes, are contained in the same NotI/BssHII fragment of 75 kb, in which
AP-Genomics
323
TRANSCRIPTION MAP OF THE HLA CLASS I REGION
TABLE 2 Northern Blotting Results Northern blotting CLONE GT253 GT260 GT469 GT483 GT511 GT512 GT545 GT546 GT554 GT257 GT475 GT478 GT555 HCG-V TC4 TB3-1
É É É É
Size of mRNA (kb)
Placenta
Duodenum
Liver
Lymphocites
Endothelium
Fibroblasts
5 4 2.8/3.9 3/4 2.6 3.8 3.8 2.8/4 2.8/4 4/4.5
— // // // — // — // // —
— // — — // // — // // —
— // — // // // — // // —
— // // // — — — // // —
// // // // — // — // // //
n.t. n.t. — — — // // — // —
// // //
// // —
// // —
— // —
— // —
n.t. n.t. n.t.
Below the sensitivity of our Northern assay 2 1.3/4.2 3
Note. RNA analysis of the gene fragments corresponding to new sequences. The tissues analyzed are indicated. Results obtained for the TB3-1 and TC4 genes are also included. n.t., not tested.
the HLA-A locus is also located. Finally, 6 clones reside telomeric to HLA-A. Clones GT545 and GT546 were not present on 225B1 but reside on a 260-kb NotI fragment on YAC 169H11. Thus, they must be restricted to the region of the NotI fragment that extends beyond the end of 225B1 and the following NotI site on YAC 169H11. The two clones (GT509 and GT547) corresponding to the human S17 ribosomal protein map to a NotI–NruI fragment of about 110 kb. Two clones (GT514 and GT550) that show homology with the RCK/ RNA helicase gene map within a 225-kb MluI–NruI fragment on YAC 169H11, while clones GT511 and GT512 are contained on the 150-kb end NruI fragment of YAC 169H11. Interestingly, Y52 (Totaro et al., 1994), the polymorphic marker that shows the highest linkage disequilibrium values in our population, maps a minimum distance of 90 kb centromeric of the HLA-A locus, in the same interval as GT469.
same gene. Expression data of the four gene fragments (GT257, GT469, GT483, and GT554) residing in the same 275-kb NotI fragment demonstrated four distinct patterns, corresponding most likely to four different genes. A clear example of expression pattern of one of these clones (GT469) is given in Fig. 2. Distinct expression patterns have also been obtained for at least three fragments (GT253, HCG-V, and TB3-1) residing in the same NotI–BssHII fragment of 75 kb. For the TB3-1 gene, levels of expression were detected only on placenta RNA, compared to its widespread distribution as previously described (Grenett et al., 1992). Northern results for GT511 and GT512 clones, residing in the same 150-kb NotI fragment, are also consistent with two different transcription units. Figure 3 gives an example of the pattern of expression of clone GT512. Only
RNA Hybridization Retrieved cDNA clones were also tested by hybridization to poly(A)/ or to total RNA of a series of tissues. The results are presented in Table 2. Five clones displayed transcripts in duodenum and liver, both tissues that are likely to express a gene involved in HFE. For four clones (GT257, GT475, GT478, and GT555), the expression levels appeared to be below the sensitivity of our Northern assay. In the case of clone GT257, it was possible to detect expression by PCR in the following tissues: brain, liver, pancreas, heart, and CaCo2 (data not shown). Gene fragments GT545 and GT546 showed identical expression patterns. As they are located in close proximity to each other, they most likely form a single transcription unit, corresponding to the
/ m4990$3868
01-04-96 18:28:00
gnma
FIG. 2. Northern blot results of the clone GT469. Two transcripts of 3.9 and 2.8 kb, respectively, were detected. A cross-hybridization with the 28S subunit is also present. 5, size of 28S; 2, size of 18S. D, duodenum; F, fibroblasts; P, placenta; L, lymphocites; E, endothelium; Li, liver.
AP-Genomics
324
TOTARO ET AL.
FIG. 3. Northern blot results of the clone GT512. A unique transcript of 3.8 was detected on fibroblasts RNA. 5, size of 28S; 2, size of 18S. D, duodenum; F, fibroblasts; P, placenta; L, lymphocites; E, endothelium; Li, liver.
fibroblasts are positive for the presence of a 3.8-kb transcript. Data obtained for the TC4 gene are in agreement with previous reports (Drivas et al., 1990). Combining the data from the physical map with expression results, a total number of 14 transcription units in addition to HLA genes was identified (Fig. 1). Identification of New Polymorphisms Specific primers were designed corresponding to the first four cDNA fragments isolated: GT253, GT256, GT257, and GT260. The RNA-SSCP technology allowed us to detect five new restriction enzyme polymorphisms and one insertion polymorphism. An additional EcoRI RFLP was also detected by Southern blotting using clone GT260. Details on each polymorphism are reported in Table 3. In each case, Mendelian inheritance was confirmed in 20 large kindreds. Allele frequencies on affected chromosomes were determined in a sample of 50 hemochromatosis patients, while the corresponding frequencies on the normal chromosomes were obtained from the unaffected parental chromosomes. Markers located more than 200 kb centromeric of HLA-A show absence of linkage disequilibrium with the disease, while alleles at the GT253 and GT256/ HCG-V loci, which are located in close vicinity to the HLA-A locus, show strong association with the disease. d Coefficients of 0.359 for the NsiI RFLP of GT253 and 0.269 for the MspI polymorphism of GT256/HCGV were observed. DISCUSSION
Linkage studies carried out on French and Italian families at risk for HFE have indicated a possible disease gene location in close vicinity to the HLA-A locus. This location has been recently confirmed by analysis with Y52, a polymorphic marker that showed the highest allelic association in our population, with a d coefficient of 0.419 (Totaro et al., 1995). The identification of
/ m4990$3868
01-04-96 18:28:00
gnma
new polymorphisms that have been described increases the number of polymorphic markers available, and the strong allelic association between alleles at the GT253 and GT256/HCG-V loci with the disease further support the candidacy of the interval surrounding Y52. A conservative centromeric limit of the HFE region, based on present data, should not extend more than 200 kb from HLA-A, as d coefficients obtained with markers residing more centromeric are very low or negative. Some ambiguity has arisen concerning the telomeric boundary, as recent reports on UK and Australian populations are consistent with extending the candidate region more telomeric, closer to D6S105 (Jazwinska et al., 1993; Stone et al., 1994; Raha-Chowdhury et al., 1995). D6S105 has not yet been placed on a physical map including HLA-A or HLA-F, but it should be at least 2 cM telomeric of HLA-A (Stone et al., 1994; Volz et al., 1994). Together, the association studies performed demonstrate the presence of a plateau of linkage disequilibrium spanning from HLA-A to D6S105, narrowed by two peaks, one higher in French and Italians (around HLA-A) and the other higher in British and Australians (around D6S105). Once D6S105 has been placed on a physical map linking this locus to HLA-F and an increased number of polymorphic markers are available between HLA-A and D6S105, an examination of these markers may make it possible to predict more precisely the location of the gene affected in HFE. Overlapping YACs covering the HLA-A region and the adjacent telomeric region were isolated and utilized as targets for cDNA selection experiments using cDNA pools from different tissues, including gut and liver. These tissues are anticipated to express the HFE gene. Thirteen new gene segments have been isolated. Two of them, GT260 and GT478, reside centromeric of the candidate area and thus are not likely to correspond to strong candidate genes. None of the remaining 11 show any striking similarity with known sequences or with characteristic features of genes involved in iron metabolism; however, they represent only fragments of genes. Their possible relevance to the hemochromatosis defect is currently being investigated. Both physical mapping and expression patterns were used to establish the number of transcription units identified by the 38 cDNA fragments analyzed. A number of the fragments were positioned at least 100 kb apart and thus are likely to correspond to different transcription units. For the fragments that map close together, different expression patterns or different RNA sizes were detected. These results are consistent with the identification, in addition to the HLA genes, of at least 14 transcription units, corresponding to nine new genes in the region. One clone, GT469, which detects two transcripts of 2.8 and 3.9 kb, is noteworthy because it maps very close to Y52, the polymorphic marker that shows the highest linkage disequilibrium values in our population. Additional genes, both known and novel, could be isolated either by increasing the
AP-Genomics
0.010 0.71 0.29
EcoRI MspI
AciI
ScrFI
GT260 GT256/HCG-V
GT257
GT257
GT260Ex3
ins A (Rna-SSCP) AciI GT260Ex1
/ m4990$3868
01-04-96 18:28:00
Note. Description of the polymorphisms identified in four gene fragments isolated by cDNA selection.
0.30 184 56
0.70
0.040 0.02 0.98 0.99 346 53
0.01
0.269 not tested 0.74 0.26 0.25 0.94 A1:21Kb, A2:14/7Kb 123 56
0.75 0.06
0.125 0.03 0.97 319 60
0 1
309 60
0.98
0.02
0.98
0.02
zero
7.44 (p:0.014)
6.95 (p:0.026) 0.359 0 1 0.13 0.87 197 56
F-AATGTTACCCAGCTATATGCA R-ACACCAGGAAAGCTACATGCT F-CTTAAGGAGGCTCCACAGAA R-CGGTGCATCTTTCATATGAG F-TCATGTTCCCTCAGGAGTGT R-CTGAACGAGTAAATGTGAATAAGA Southern blotting F-GACTCAGTGGTGGTTATGGATT R-CAGCCCAGATGTTCAGTTCTCT F-GCTTAGGAATAGATGTCCAA R-GACAAAGTCCTTACCTGTCT F-CTGAGCAGTAGCTAGAAGCCT R-GACAAAGTCCTTACCTGTCT NsiI GT253
2 1 2 Polymorphism
Oligonucleotide sequence
Annealing temp. (7C)
Amplification product size
1
Affected chromosome Normal chromosome
Clone
Polymorphism Results
TABLE 3
Allele frequencies
d Coefficient
x2
TRANSCRIPTION MAP OF THE HLA CLASS I REGION
gnma
325
number of unique clones to be characterized in detail or by additional cDNA selection experiments carried out with cDNAs from tissues not used in the present study. The novel fragments have already been useful as single-copy probes for screening larger-insert cDNA libraries and for refining and extending the physical map of the HLA class I region. Our mapping included a new mega YAC that contains a region extending telomeric of the HLA. The resulting map spans from about 90 kb centromeric of HLA-E to approximately 550 kb telomeric of HLA-F. Mapping data agree with an extended map of the region recently reported (Amadou et al., 1995). The positions of the HLA-F and MOG loci are slightly different, since we place them onto a distinct NotI fragment. The revised location does not alter the positions of the two genes significantly, as their positions shift only a few kilobases proximal to their previous ones. Moreover, our map confirms that D6S105, a marker linked to hereditary hemochromatosis, is positioned at least 1 Mb from HLA-A. Of the known genes that were identified, the position of the TC4 gene is in agreement with that recently reported (Wei et al., 1993). The localization of the TB31 and S17 genes has not been previously described. Moreover, the presence of a new member of the RNA helicase superfamily that maps telomeric of HLA-F has been identified and localized. In summary, we describe the isolation and mapping of 13 new gene fragments. Based on physical mapping of these fragments and of known genes, the HLA class I region is notably gene-rich, emphasizing the challenge of the identification of the HFE defect. ACKNOWLEDGMENTS This work was supported by grants from the Italian Ministry of Health, AIRC (Associazione Italiana Ricerca Cancro), and Telethon (Grant E.149). J.M.R. is a scholar of the Medical Research Council of Canada. The Authors thank Dr. Eve Roberts, Gastroenterology and Nutrition, H.S.C. (Toronto), for liver samples, L. Borgato (Verona) for technical assistance, and the YAC Screening Facility of the Canadian Genome Analysis and Technology Program.
REFERENCES Abderrahim, H., Sambucy, J., Iris, F., Ougen, P., Billault, A., Chumakov, I., Dausset, J., Cohen, D., and Le Paslier, D. (1994). Cloning the human major histocompatibility complex in YACs. Genomics 23: 520–527. Amadou, C., Ribouchon, M., Mattei, M., Jenkins, N., Gilbert, D., Copeland, N., Avoustin, P., and Pontarotti, P. (1995). Localization of new genes and markers to the distal part of the human major histocompatibility complex (MHC) region and comparison with the mouse: New insights into the evolution of mammalian genomes. Genomics 26: 9–20. Bisceglia, L., Grifa, A., Zelante, L., and Gasparini, P. (1994). Development of RNA SSCP protocols for the identification and screening of CFTR mutations: Identification of two new mutations. Hum. Mutat. 4: 136–140. Boretto, J., Jouanoulle, A., Yaouanq, J., El Kahloun, A., Mauvieux, V., Blayau, M., Perichon, M., Le Treut, A., Clayton, J., Borot, N.,
AP-Genomics
326
TOTARO ET AL.
Le Gall, J., Pontarotti, P., and David, V. (1992). Anonymous markers located on chromosome 6 in the HLA class I region: Allelic distribution in genetic hemochromatosis. Hum. Genet. 89: 33–36. Chen, I. T., Dixit, A., Rhoads, D. D., and Roufa, D. J. (1986). Homologous ribosomal protein in bacteria yeast and humans. Proc. Natl. Acad. Sci. USA 83: 6907–6911. Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162: 156–159. Church, G. H., and Gilbert, W. (1984). Genomic sequences. Proc. Natl. Acad. Sci. USA 81: 1991–1995. Drivas, G., Shih, A., Coutavas, E., Rush, M., and D’Eustachio, P. (1990). Characterization of four novel ras-like genes expressed in a human teratocarcinoma cell line. Mol. Cell. Biol. 10: 1793–1798. El Kahloun, A., Chauvel, B., Mauvieeux, V., Dorval, I., Jouanolle, A., Gicquel, I., Le Gall, J., and David, V. (1993). Localization of seven new genes around the HLA-A locus. Hum. Mol. Genet. 2: 55–60. Gasparini, P., Borgato, L., Piperno, A., Girelli, D., Olivieri, O., Gottardi, E., Roetto, A., Dianzani, I., Fargion, S., Schinaia, G., Cappellini, N., Gandini, G., Pignatti, P. F., Fiorelli, G., De Sandre, G., and Camaschella, C. (1993). Linkage analysis of 6p21 polymorphic markers and the hereditary hemochromatosis: Localization of the gene centromeric to HLA-F. Hum. Mol. Genet. 2: 571–576. Goei, V., Parimoo, S., Capossela, A., Chu, T., and Gruen, J. (1994). Isolation of novel -HLA gene fragments from the hemochromatosis region (6p21.3) by cDNA hybridization selection. Am. J. Hum. Genet. 54: 244–251. Grenett, H., Bounelis, P., and Fuller, G. (1992). Identification of a human cDNA with high homology to yeast omnipotent suppressor 45. Gene 110: 239–243. Jazwinska, E., Lee, S., Webb, J., Halliday, W., and Powell, W. (1993). Localization of the hemochromatosis gene close to D6S105. Am. J. Hum. Genet. 53: 347–352. Lu, D., and Yunis, J. (1992). Cloning expression and localization of an RNA helicase gene from a human lymphoid cell line with chromosomal breakpoint 11q23.3. Nucleic Acids Res. 20: 1967– 1972. McKusick, V. A. (1994). ‘‘Mendelian Inheritance in Man,’’ 11th ed, The John Hopkins Univ. Press, Baltimore. Miller, S., Dykes, D., and Polesky, H. (1988). A simple salting out procedure for extracting DNA from human cells. Nucleic Acids Res. 16: 1215. Raha-Chowdhury, R., Bowen, D., Burnett, A., and Worwood, M. (1995). Allelic association and homozygosity at loci from HLA-B to D6S299 in genetic haemochromatosis. J. Med. Genet. 32: 446–452. Rommens, J., Lin, B., Hutchinson, G., Andrew, S., Goldberg, Y.,
/ m4990$3868
01-04-96 18:28:00
gnma
Glaves, M., Graham, R., Lai, V., McArthur, J., Nasir, J., Theilmann, J., McDonald, H., Kalchman, M., Clarke, L., Schappert, K., and Hayden, M. (1993). A transcription map of the region containing the Huntington disease gene. Hum. Mol. Genet. 2: 901– 907. Rommens, J., Mar, L., McArthur, J., Tsui, L-C., and Scherer, S. (1994). Towards a transcriptional map of the q21–q22 region of chromosome 7. In ‘‘Identification of Transcribed Sequences’’ (U. Hochgeshwender and K. Gardiner, Eds.), pp. 65–79, Plenum, New York. Sanger, F., Nicklen, S., and Coulson, A. (1977). DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74: 5463– 5467. Sarkar, G., Yoon, H-S., and Sommer, S. (1992). Screening for mutations by RNA single-strand conformation polymorphisms (rSSCP): Comparison with DNA-SSCP. Nucleic Acids Res. 20: 871–878. Simon, M., Le Mignon, L., Fauchet, R., Yaouanq, J., David, V., Edan, G., and Bourel, M. (1987). A study of 609 HLA haplotypes marking for the hemochromatosis gene: Mapping the gene near the HLA-A locus and characters required to define a heterozygous population; hypothesis concerning hemochromatosis-HLA association. Am. J. Hum. Genet. 41: 89–105. Stone, C., Pointon, J., Jazwinska, E., Halliday, J., Powell, L., Robson, K., Monaco, A., and Weatherall, D. (1994). Isolation of CA dinucleotide repeats close to D6S105; linkage disequilibrium with hemochromatosis. Hum. Mol. Genet. 3: 2043–2046. Totaro, A., Grifa, A., Roetto, A., Zelante, L., Camaschella, C., and Gasparini, P. (1994). A new complex polymorphic repeat close to HLA-A and HLA-E loci. Hum. Genet. 94: 578. Totaro, A., Grifa, A., Roetto, A., Lunardi, C., D’Agruma, L., Sbaiz, L., Zelante, L., De Sandre, G., Camaschella, C., and Gasparini, P. (1995). New polymorphisms and markers in the HLA class I region: Relevance to hereditary hemochromatosis (HFE). Hum. Genet. 95: 429–434. Volz, A., Boyle, J., Cann, H., Cottingham, R., Orr, H., and Ziegler, A. (1994). Report of the second international workshop on human chromosome 6. Genomics 21: 464–472. Weber, J., Kwitek, A., May, P., and Zoghbi, H. (1991). Dinucleotide repeat polymorphism at the D6S105 locus. Nucleic Acids Res. 19: 968. Wei, H., Fan, W-F., Xu, H., Parimoo, S., Shukla, H., Chaplin, D., and Weissman (1993). Genes in one megabase of the HLA class I region. Proc. Natl. Acad. Sci. USA 90: 11870–11874. Yaouanq, J., Perichon, M., Chorney, M., Pontarotti, P., Le Treut, A., El Kahloun, A., Mauviex, V., Blayau, M., Jouanolle, A., Chauvel, B., Moirand, R., Nouel, O., Le Gall, J., Feingold, J., and David, V. (1994). Anonymous marker loci within 400 kb of HLA-A generate haplotypes in linkage disequilibrium with the hemochromatosis gene (HFE). Am. J. Hum. Genet. 54: 252–263.
AP-Genomics
31, 319–326 (1996) 0054
ARTICLE NO.
Hereditary Hemochromatosis: Generation of a Transcription Map within a Refined and Extended Map of the HLA Class I Region ANGELA TOTARO, JOHANNA M. ROMMENS,*,† ANNA GRIFA, CLAUDIO LUNARDI,‡ MASSIMO CARELLA, JACK J. HUIZENGA,* ANTONELLA ROETTO,§ CLARA CAMASCHELLA,§ GIORGIO DE SANDRE,‡ AND PAOLO GASPARINI1 Servizio di Genetica Medica, IRCCS-Ospedale ‘‘CSS,’’ I-71013 San Giovanni Rotondo, §Dipartimento di Scienze Biomediche ed Oncologia Umana, Ospedale S. Luigi, Orbassano, 10100 Turin, and ‡Istituto di Clinica Medica, Universita’ di Verona, 35100 Verona, Italy; *Department of Genetics, Research Institute, The Hospital for Sick Children and †Department of Molecular and Medical Genetics, University of Toronto, Toronto, M5G 1X8 Canada Received August 21, 1995; accepted November 10, 1995
Hereditary hemochromatosis, a common severe inherited disease, maps to the short arm of chromosome 6 close to the HLA-A locus. Recently, linkage data on Italian and French populations confirmed this location, while a similar analysis on Australian and British populations located the gene closer to D6S105, a marker residing telomeric of HLA-A. To increase our knowledge on the region of highest linkage disequilibrium in our population and possibly to identify the disease gene, a 1.2-Mb detailed physical and transcription map was generated, spanning the HLA class I region. Thirty-eight unique cDNA fragments, retrieved following the hybridization of immobilized YACs to primary pools of cDNAs prepared from RNA of fetal brain, adult brain, liver, placenta, and the CaCo2 cell line, were characterized. All cDNA fragments were positioned in a refined and extended map of the human major histocompatibility complex spanning from HLAE to approximately 500 kb telomeric of HLA-F. The localization of known genes was refined, and a new gene from the RNA helicase superfamily was identified. Overall, 14 transcription units in addition to the HLA genes have been detected and integrated in the map. Thirteen cDNA fragments show no similarity with known sequences and could be candidates for the disease. Their characterization and assessment for involvement in hemochromatosis are still under investigation. Seven new polymorphisms, some tightly linked to the disease, were also identified and localized. q 1996 Academic Press, Inc.
INTRODUCTION
Hemochromatosis (HFE) is a common autosomal recessive disease characterized by chronic dietary iron 1 To whom correspondence should be addressed at Servizio di Genetica Medica, IRCCS-Ospedale ‘‘CSS,’’ I-71013 San Giovanni Rotondo, Italy. Telephone: /39 882 410825. Fax: /39 882 411616.
overload. The disease prevalence is estimated to be 2– 5/1000 in the Caucasian population, with a corresponding carrier frequency of 0.045–0.071 (McKusick, 1994). The underlying biochemical defect of the disease is unknown, but the gene has been mapped to the short arm of chromosome 6, containing the histocompatibility antigen (HLA) class I region (Simon et al., 1987). Linkage disequilibrium studies in French and Italian populations initially suggested a narrow candidate region for the HFE gene, including a 400-kb stretch of DNA flanking the HLA-A gene, spanning from I.82 to HLA-F (Boretto et al., 1992; Gasparini et al., 1993; Yaouanq et al., 1994). However, association studies, performed in the Australian population (Jazwinska et al., 1993), have extended this region to the D6S105 locus (Weber et al., 1991), which is estimated to be at least 2 cM telomeric of HLA (Stone et al., 1994; Volz et al., 1994). We recently reported new polymorphisms and sequence tagged sites from the HLA class I region and identified a polymorphic marker, named Y52, which showed the highest allelic association in our population (Totaro et al., 1995). Allele 1 of Y52, together with HLA-A serotype A3 and D6S265 allele 1, mark the ancestral HFE haplotype in our population. Y52 and D6S265 were positioned very close to the HLA-A locus (Totaro et al., 1995). These data support this area as being the most likely candidate region for the disease gene. Several novel coding sequences have already been obtained from this region (El Kahloun et al., 1993; Wei et al., 1993; Goei et al., 1994), but despite the extensive effort, there has as yet been no indication that these coding sequences are hemochromatosis candidates. We have extended the cloned region by isolation of an overlapping YAC clone, 169H11, and applied the cDNA selection technique to the entire region of 1.2 Mb containing the HLA-A locus, spanning from 90 kb proximal to HLA-E to about 550 kb distal to HLA-F. Moreover, to refine and extend the physical and transcription map beyond the distal part of the human HLA
319
/ m4990$3868
01-04-96 18:28:00
gnma
0888-7543/96 $12.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AP-Genomics
320
TOTARO ET AL.
class I region and to gain insight into its organization, we have performed long-range restriction pulse-field gel mapping by electrophoresis (PFGE) of yeast artificial chromosome clones using both known markers from this region and the new clones obtained by cDNA selection. MATERIALS AND METHODS Preparation and isolation of genomic DNAs. YAC 225B1 has been obtained from CEPH, while YAC 169H11 has been isolated from the Mega YAC CEPH library using a fragment located on the telomeric end of YAC 225B1 as a probe. YAC clones were isolated, grown on plates with medium lacking tryptophan and uracil, and used to prepare total chromosome DNA, as previously described (Rommens et al., 1994). After separation by PFGE, the artificial chromosomes were transferred to nylon membrane (Hybond-N, Amersham) and immobilized by UV-crosslinking for the selection experiments (Rommens et al., 1994). Genomic DNAs were isolated from human leukocytes or from rodent and rodent hybrid cell lines by established procedures (Miller et al., 1988). The mouse–human hybrid cell line GM10629 was obtained from the NIGMS Human Genetic Mutant Cell Repository (Camden, NJ). Restriction digestions were carried out according to the supplier’s recommendations. Electrophoresis, blotting, and hybridization were carried out according to standard procedures (Sambrook et al., 1989). After hybridization, the blots were washed (0.21 SSC with 0.1% SDS) at 607C and exposed with autoradiography for 2–72 h. For long-range physical mapping, high-molecular-weight DNA embedded in agarose of whole yeast chromosomes was digested with rare-cutting restriction enzymes. Products were separated by PFGE and transferred to nylon membrane for hybridizations as previously described (Church and Gilbert, 1984). Specific probes for the HLAE (Totaro et al., 1995), HLA-B30, HLA-A, HLA-F, MOG (Amadou et al., 1995), D6S131 (Amadou et al., 1995), and RFP (Amadou et al., 1995) loci were used to establish the orientation and framework for the physical map. The HLA probes were generated using gene-specific oligonucleotides and the polymerase chain reaction (PCR). cDNA preparation and selection. cDNA selection was carried out as previously described (Rommens et al., 1994). Briefly, randomly primed cDNA was prepared from poly(A)/ RNA of fetal liver, fetal brain, pancreas, and placenta tissues and from total RNA of adult liver and of the CaCo2 cell line (ATCC HTB 37). Selection was carried out for two consecutive rounds of hybridization to immobilized artificial chromosome DNA of 225B1 and 169H11. Hybridizing cDNA was collected, amplified, and cloned as described (Rommens et al., 1993, 1994). Characterization of retrieved cDNAs. Two hundred eighty-eight individual colonies were picked and gridded onto an agar plate containing 100 mg/ml ampicillin for ordering, prescreening, and storage. Initial analysis of cDNA clones included a prescreening for ribosomal sequences. Approximately 8% of the clones were eliminated because they hybridized with radiolabeled cDNA obtained from total RNA. Plasmids from 20 to 40 clones from each selection experiment that did not hybridize in the prescreenings were isolated for further analysis. Of clones tested, approximately 70% mapped appropriately to the starting YAC clones. The membranes of the grids were also repeatedly hybridized with individual cDNA clones to minimize the number of overlapping or redundant clones. RNA preparation and hybridization. Total RNA was isolated from frozen tissues (liver, duodenal mucosa, and placenta), cultured cells (endothelial cells and fibroblasts), and freshly isolated peripheral blood mononuclear cells by the guanidium isothiocyanate method (Chomczynski and Sacchi, 1987), size-fractionated by electrophoresis on 1% agarose–formaldehyde gels, and blotted onto nylon membranes. Hybridization with 32P-labeled probes was carried out at 427C for at least 12 h in a formamide-based solution. Membranes
/ m4990$3868
01-04-96 18:28:00
gnma
were then washed twice in 51 SSPE (11 SSPE: 360 mM NaCl, 20 mM Na2HPO4 , 2 mM EDTA) at 427C for 20 min, twice in 11 SSPE containing 0.1% SDS at 427C for 20 min, and exposed to HyperfilmMP (Amersham, UK). Hybridization with b-tubulin and/or -actin was performed to verify the integrity of RNA. DNA sequencing and sequence analysis. The sequences of both strands of the retrieved cDNAs were obtained with the dideoxy chain termination method (Sanger et al., 1977) using the Taq dye primer cycle sequencing kit (ABI) and the 373A DNA sequencer (ABI). The resulting sequences were then analyzed by comparison with the EMBL and GenBank Data Banks (PCGENE, Intelligenetics). RNA single-strand conformation polymorphism. Oligonucleotides for PCR primers corresponding to the first four clones isolated (GT253, GT256, GT257, and GT260) were designed. The T7 phage promoter sequence was incorporated at the end of one of each set as previously described (Sarkar et al., 1992). Annealing temperatures for amplification and length of each fragment are reported in Table 3. PCR was performed on 500 ng genomic DNA from 20 unrelated individuals according to standard protocols. Subsequent transcription was carried out as previously described (Bisceglia et al., 1994). An aliquot of 4.5 ml of the transcription reaction was then loaded onto a 6.5% nondenaturing polyacrylamide gel. Electrophoresis was performed at 30 W constant power for 13 h. After electrophoresis, the gel was dried and subjected to autoradiography for 12 h. Fragments with electrophoretically altered migration patterns were then sequenced with the dideoxy chain termination method (Sanger et al., 1977) as described above. When a polymorphism was detected, frequencies of alleles were assessed in a sample of 50 affected patients and their relatives attending our Centers. The statistical significance of linkage disequilibrium with the disease was tested by utilizing the x2 test of homogeneity for each marker and then standardized to obtain the d coefficient whose range is in the interval 0– 1, to measure the association. The d coefficient is defined as q
(ad 0 bc)/ (a / b)(c / d)(a / c)(b / d), where a and b are the frequencies of allele 1 in affected and unaffected chromosomes, respectively, and c and d are the frequencies of allele 2 in the same sets of chromosomes.
RESULTS
Thirty-eight unique cDNA clones were selected, sequenced, and verified to map to the YAC clones 225B1 or 169H11. To confirm the assignment to chromosome 6, each cDNA was hybridized to total human DNA and to a mouse hybrid cell line that contains chromosome 6 as its only human material. The pattern of hybridization (i.e., number of distinct EcoRI restriction fragments on chromosome 6 and on total human genomic DNA) is indicated in Table 1. For several clones, including GT256, GT511, and GT512, among others, only a single EcoRI restriction fragment was clearly detectable on chromosome 6 DNA, while a band of identical size and additional bands were present in genomic DNA. The DNA sequence of each clone was analyzed for coding potential and for alignment to the public databases and to each other. Thirteen corresponded to a new sequence, while the remaining 25 could be assigned to six subgroups that shared overlapping sequence or that were similar to previously known genes. One clone (GT554) showed homology with an EST (R85804 from the Washington University-Merck EST
AP-Genomics
321
TRANSCRIPTION MAP OF THE HLA CLASS I REGION
TABLE 1 cDNA Selection Results GT clone
Size
Chromosome 6 bands
Genomic bands
GT253 GT257 GT260 GT469 GT475 GT478 GT483 GT511 GT512 GT545 GT546 GT554 GT555 GT470 GT472 GT479 GT513 GT549 GT548 GT256
647 395 672 644 919 701 722 520 636 611 501 387 397 445 390 610 422 302 498 407
1 1 2 1 1 1 1 1 1 1 3 1 1 1 1 1 1 1 3 1
1 1 2 1 1 1 1 3 3 1 3 1 1 2 2 2 2 2 3 2
GT258
420
Several
Several
GT259 GT474
418 611
Several Several
Several Several
GT482
701
Several
Several
GT515 GT481
622 653
Several Several
Several Several
GT473
521
Several
Several
GT289
359
1
Several
GT477
409
1
4
GT486
510
1
4
GT510
711
1
4
GT254 GT291 GT509
610 612 529
2 2 2
Several Several Several
GT547
335
2
Several
GT480
572
1 faint
Several
GT514
364
1 faint
Several
GT550
393
1 faint
Several
Remarks New sequence EMBL Acc. No. X90532 New sequence EMBL Acc. No. X90534 New sequence EMBL Acc. No. X90535 New sequence EMBL Acc. No. X90536 New sequence EMBL Acc. No. X90537 New sequence EMBL Acc. No. X90538 New sequence EMBL Acc. No. X90539 New sequence EMBL Acc. No. X90540 New sequence EMBL Acc. No. X90541 New sequence EMBL Acc. No. X90542 New sequence EMBL Acc. No. X90543 New sequence EMBL Acc. No. X90544 EST (R85804) New sequence EMBL Acc. No. X90545 Overlap with GT256 Overlap with GT256 Overlap with GT256 Overlap with GT256 Overlap with GT256 Overlap with GT546 EMBL Acc. No. X90533 Identical to HSDNAHCGV gene from base 2856 to 3263 Identical to HLA-B clone PTMHCAA from base 1043 to 1461 Identical to HLA-E clone HSHLA1EA from base 600 to 994 Identical to MHC clone HSMHCFAN4 from base 1 to 135; chimeric clone Similar to HLA-B clone HSMHB17W from base 2577 to base 3269 Similar to HLA-A2 clone HSHLAA2 from base 163 to 788 Similar to MHC clone HSMHCFANA from base 3 to 120; chimeric clone Similar to MHC clone HSMHCFANB from base 1 to 120; chimeric clone with GT256 Identical to TB-3 gene clone HSTB31A from base 582 to 937 Identical to TB3-1 gene clone HSTB31A from base 1934 to 2252 Identical to TB3-1 gene clone HSTB31A; entirely contained in clone GT510 Identical to TB3-1 gene clone to HSTB31A from base 1087 to 1794 Identical to TC4 gene from base 323 to base 651 Identical to TC4 gene from base 323 to base 653 Identical to human S17 ribosomal protein gene clone HSRPS17 from base 3 to 477 Identical to human S17 ribosomal protein gene; entirely contained in clone GT509 Identical to RCK/RNA helicase clone HSRCK from base 1176 to 1756 Identical to RCK/RNA helicase clone HSRCK from base 1018 to 1407 Identical to RCK/RNA helicase clone HSRCK from base 1010 to 1425
Note. Summary of selected cDNA clones. The first 14 gene fragments correspond to new sequences, while the remaining ones shared overlapping sequence and/or were homologous to previously known genes.
Project). A complete summary of the results obtained is reported in Table 1. cDNAs of Known Human Genes Twenty-five clones shared overlapping sequence and/ or were homologous to already known genes, as de-
/ m4990$3868
01-04-96 18:28:00
gnma
scribed in Table 1. In particular, 7 clones (GT258, GT259, GT473, GT474, GT481, GT482, and GT515) showed identity or near identity with HLA class I sequences. Four clones (GT289, GT477, GT486, and GT510) showed complete sequence alignment with different portions of a TB3-1 gene, a member of a human
AP-Genomics
322
TOTARO ET AL.
FIG. 1. The physical and transcription map of the HLA class I region from HLA-E to D6S131. B, M, N, and R correspond to the BssHII, MluI, NotI, and NruI sites, respectively. The isolated cDNAs and the class I genes were positioned by PFGE. The relative position of each gene fragment within a given restriction fragment interval is not known. The numbering above the marked physical intervals indicates the 14 transcription units identified.
gene family with significant homology to yeast omnipotent suppressor 45 (Grenett et al., 1992). Two other clones (GT254 and GT291) were identical to the TC4 gene sequences (Drivas et al., 1990), a member of the family of human ras-like proteins. Two clones (GT509 and GT547) were identical to the S17 human ribosomal protein (Chen et al., 1986). This gene is strongly conserved through evolution and encodes a product that is fundamental to the translational apparatus. Three additional clones (GT480, GT514, and GT550) showed similarity with the RCK/RNA helicase gene (Lu and Yunis, 1992). The hybridization of these selected cDNA fragments detected only one faint band on chromosome 6 DNA, but several additional bands were observed in genomic DNA. Since the RCK/RNA helicase has been localized to band q23.3 of chromosome 11, our results are consistent with the identification of a new member of the expanding superfamily of RNA helicases. These genes are involved in specialized RNA functions and contain a series of 10 motifs that have been evolutionarily conserved. One clone (GT256) showed complete sequence alignment with the HCG-V gene, a gene recently submitted to the EMBL Data Bank (Gifon et al., Accession No. X89902). While all described clones are unique, five clones (GT470, GT472, GT479, GT513, and GT549) shared overlapping sequences with GT256, while one (GT548) overlapped with clone GT546. Physical Mapping A refined and extended physical map was obtained using both known markers from the HLA class I region and the new clones obtained by cDNA selection. It spans from about 90 kb centromeric of the HLA-E locus to approximately 550 kb telomeric of HLA-F (Fig. 1).
/ m4990$3868
01-04-96 18:28:00
gnma
The HLA-E locus resides in a 60-kb NotI–MluI fragment on YAC 225B1, while HLA-A is located in a NotI– BssHII fragment of 75 kb on both YAC225B1 and YAC169H11. PCR and PFGE results demonstrated that both HLA-F and MOG loci reside only on YAC 169H11. They are located in a 260-kb NotI fragment and must be contained in the region between the terminal right end of the YAC 225B1 and the following NotI site on YAC 169H11. The D6S131 probe hybridizes to a telomeric NruI fragment of 155 kb of 169H11. We attempted to integrate the map with other markers known to be localized telomeric of HLA, including RFP (Amadou et al., 1995) and D6S105 (Stone et al., 1994), using PCR analysis, but they were absent from YAC169H11, which extends most telomeric. The 13 clones with novel sequences have been localized by PFGE to their cognate genomic sequences on YAC 225B1 and YAC 169H11. They were found to be distributed across the YACs. The relative positions of clones that shared common restriction fragments could not be ordered. Among the clones localized on the centromeric side of HLA-A, two (GT260 and GT478) map centromeric to a BssHII site on the far left end of YAC 225B1, while those corresponding to the TC4 gene were within the same NotI–MluI fragment that contained the HLA-E locus. The others were located in the same 275-kb NotI fragment. The relative positions of the clones, from the centromere to the telomere, were GT257 in a NotI–BssHII fragment of 105 kb, GT469 in a MluI–BssHII fragment of 50 kb, and GT483 and GT554 in a BssHII–MluI fragment of 75 kb. Three clones (GT253, GT475, and GT555), plus those corresponding to the TB3-1 and HCG-V genes, are contained in the same NotI/BssHII fragment of 75 kb, in which
AP-Genomics
323
TRANSCRIPTION MAP OF THE HLA CLASS I REGION
TABLE 2 Northern Blotting Results Northern blotting CLONE GT253 GT260 GT469 GT483 GT511 GT512 GT545 GT546 GT554 GT257 GT475 GT478 GT555 HCG-V TC4 TB3-1
É É É É
Size of mRNA (kb)
Placenta
Duodenum
Liver
Lymphocites
Endothelium
Fibroblasts
5 4 2.8/3.9 3/4 2.6 3.8 3.8 2.8/4 2.8/4 4/4.5
— // // // — // — // // —
— // — — // // — // // —
— // — // // // — // // —
— // // // — — — // // —
// // // // — // — // // //
n.t. n.t. — — — // // — // —
// // //
// // —
// // —
— // —
— // —
n.t. n.t. n.t.
Below the sensitivity of our Northern assay 2 1.3/4.2 3
Note. RNA analysis of the gene fragments corresponding to new sequences. The tissues analyzed are indicated. Results obtained for the TB3-1 and TC4 genes are also included. n.t., not tested.
the HLA-A locus is also located. Finally, 6 clones reside telomeric to HLA-A. Clones GT545 and GT546 were not present on 225B1 but reside on a 260-kb NotI fragment on YAC 169H11. Thus, they must be restricted to the region of the NotI fragment that extends beyond the end of 225B1 and the following NotI site on YAC 169H11. The two clones (GT509 and GT547) corresponding to the human S17 ribosomal protein map to a NotI–NruI fragment of about 110 kb. Two clones (GT514 and GT550) that show homology with the RCK/ RNA helicase gene map within a 225-kb MluI–NruI fragment on YAC 169H11, while clones GT511 and GT512 are contained on the 150-kb end NruI fragment of YAC 169H11. Interestingly, Y52 (Totaro et al., 1994), the polymorphic marker that shows the highest linkage disequilibrium values in our population, maps a minimum distance of 90 kb centromeric of the HLA-A locus, in the same interval as GT469.
same gene. Expression data of the four gene fragments (GT257, GT469, GT483, and GT554) residing in the same 275-kb NotI fragment demonstrated four distinct patterns, corresponding most likely to four different genes. A clear example of expression pattern of one of these clones (GT469) is given in Fig. 2. Distinct expression patterns have also been obtained for at least three fragments (GT253, HCG-V, and TB3-1) residing in the same NotI–BssHII fragment of 75 kb. For the TB3-1 gene, levels of expression were detected only on placenta RNA, compared to its widespread distribution as previously described (Grenett et al., 1992). Northern results for GT511 and GT512 clones, residing in the same 150-kb NotI fragment, are also consistent with two different transcription units. Figure 3 gives an example of the pattern of expression of clone GT512. Only
RNA Hybridization Retrieved cDNA clones were also tested by hybridization to poly(A)/ or to total RNA of a series of tissues. The results are presented in Table 2. Five clones displayed transcripts in duodenum and liver, both tissues that are likely to express a gene involved in HFE. For four clones (GT257, GT475, GT478, and GT555), the expression levels appeared to be below the sensitivity of our Northern assay. In the case of clone GT257, it was possible to detect expression by PCR in the following tissues: brain, liver, pancreas, heart, and CaCo2 (data not shown). Gene fragments GT545 and GT546 showed identical expression patterns. As they are located in close proximity to each other, they most likely form a single transcription unit, corresponding to the
/ m4990$3868
01-04-96 18:28:00
gnma
FIG. 2. Northern blot results of the clone GT469. Two transcripts of 3.9 and 2.8 kb, respectively, were detected. A cross-hybridization with the 28S subunit is also present. 5, size of 28S; 2, size of 18S. D, duodenum; F, fibroblasts; P, placenta; L, lymphocites; E, endothelium; Li, liver.
AP-Genomics
324
TOTARO ET AL.
FIG. 3. Northern blot results of the clone GT512. A unique transcript of 3.8 was detected on fibroblasts RNA. 5, size of 28S; 2, size of 18S. D, duodenum; F, fibroblasts; P, placenta; L, lymphocites; E, endothelium; Li, liver.
fibroblasts are positive for the presence of a 3.8-kb transcript. Data obtained for the TC4 gene are in agreement with previous reports (Drivas et al., 1990). Combining the data from the physical map with expression results, a total number of 14 transcription units in addition to HLA genes was identified (Fig. 1). Identification of New Polymorphisms Specific primers were designed corresponding to the first four cDNA fragments isolated: GT253, GT256, GT257, and GT260. The RNA-SSCP technology allowed us to detect five new restriction enzyme polymorphisms and one insertion polymorphism. An additional EcoRI RFLP was also detected by Southern blotting using clone GT260. Details on each polymorphism are reported in Table 3. In each case, Mendelian inheritance was confirmed in 20 large kindreds. Allele frequencies on affected chromosomes were determined in a sample of 50 hemochromatosis patients, while the corresponding frequencies on the normal chromosomes were obtained from the unaffected parental chromosomes. Markers located more than 200 kb centromeric of HLA-A show absence of linkage disequilibrium with the disease, while alleles at the GT253 and GT256/ HCG-V loci, which are located in close vicinity to the HLA-A locus, show strong association with the disease. d Coefficients of 0.359 for the NsiI RFLP of GT253 and 0.269 for the MspI polymorphism of GT256/HCGV were observed. DISCUSSION
Linkage studies carried out on French and Italian families at risk for HFE have indicated a possible disease gene location in close vicinity to the HLA-A locus. This location has been recently confirmed by analysis with Y52, a polymorphic marker that showed the highest allelic association in our population, with a d coefficient of 0.419 (Totaro et al., 1995). The identification of
/ m4990$3868
01-04-96 18:28:00
gnma
new polymorphisms that have been described increases the number of polymorphic markers available, and the strong allelic association between alleles at the GT253 and GT256/HCG-V loci with the disease further support the candidacy of the interval surrounding Y52. A conservative centromeric limit of the HFE region, based on present data, should not extend more than 200 kb from HLA-A, as d coefficients obtained with markers residing more centromeric are very low or negative. Some ambiguity has arisen concerning the telomeric boundary, as recent reports on UK and Australian populations are consistent with extending the candidate region more telomeric, closer to D6S105 (Jazwinska et al., 1993; Stone et al., 1994; Raha-Chowdhury et al., 1995). D6S105 has not yet been placed on a physical map including HLA-A or HLA-F, but it should be at least 2 cM telomeric of HLA-A (Stone et al., 1994; Volz et al., 1994). Together, the association studies performed demonstrate the presence of a plateau of linkage disequilibrium spanning from HLA-A to D6S105, narrowed by two peaks, one higher in French and Italians (around HLA-A) and the other higher in British and Australians (around D6S105). Once D6S105 has been placed on a physical map linking this locus to HLA-F and an increased number of polymorphic markers are available between HLA-A and D6S105, an examination of these markers may make it possible to predict more precisely the location of the gene affected in HFE. Overlapping YACs covering the HLA-A region and the adjacent telomeric region were isolated and utilized as targets for cDNA selection experiments using cDNA pools from different tissues, including gut and liver. These tissues are anticipated to express the HFE gene. Thirteen new gene segments have been isolated. Two of them, GT260 and GT478, reside centromeric of the candidate area and thus are not likely to correspond to strong candidate genes. None of the remaining 11 show any striking similarity with known sequences or with characteristic features of genes involved in iron metabolism; however, they represent only fragments of genes. Their possible relevance to the hemochromatosis defect is currently being investigated. Both physical mapping and expression patterns were used to establish the number of transcription units identified by the 38 cDNA fragments analyzed. A number of the fragments were positioned at least 100 kb apart and thus are likely to correspond to different transcription units. For the fragments that map close together, different expression patterns or different RNA sizes were detected. These results are consistent with the identification, in addition to the HLA genes, of at least 14 transcription units, corresponding to nine new genes in the region. One clone, GT469, which detects two transcripts of 2.8 and 3.9 kb, is noteworthy because it maps very close to Y52, the polymorphic marker that shows the highest linkage disequilibrium values in our population. Additional genes, both known and novel, could be isolated either by increasing the
AP-Genomics
0.010 0.71 0.29
EcoRI MspI
AciI
ScrFI
GT260 GT256/HCG-V
GT257
GT257
GT260Ex3
ins A (Rna-SSCP) AciI GT260Ex1
/ m4990$3868
01-04-96 18:28:00
Note. Description of the polymorphisms identified in four gene fragments isolated by cDNA selection.
0.30 184 56
0.70
0.040 0.02 0.98 0.99 346 53
0.01
0.269 not tested 0.74 0.26 0.25 0.94 A1:21Kb, A2:14/7Kb 123 56
0.75 0.06
0.125 0.03 0.97 319 60
0 1
309 60
0.98
0.02
0.98
0.02
zero
7.44 (p:0.014)
6.95 (p:0.026) 0.359 0 1 0.13 0.87 197 56
F-AATGTTACCCAGCTATATGCA R-ACACCAGGAAAGCTACATGCT F-CTTAAGGAGGCTCCACAGAA R-CGGTGCATCTTTCATATGAG F-TCATGTTCCCTCAGGAGTGT R-CTGAACGAGTAAATGTGAATAAGA Southern blotting F-GACTCAGTGGTGGTTATGGATT R-CAGCCCAGATGTTCAGTTCTCT F-GCTTAGGAATAGATGTCCAA R-GACAAAGTCCTTACCTGTCT F-CTGAGCAGTAGCTAGAAGCCT R-GACAAAGTCCTTACCTGTCT NsiI GT253
2 1 2 Polymorphism
Oligonucleotide sequence
Annealing temp. (7C)
Amplification product size
1
Affected chromosome Normal chromosome
Clone
Polymorphism Results
TABLE 3
Allele frequencies
d Coefficient
x2
TRANSCRIPTION MAP OF THE HLA CLASS I REGION
gnma
325
number of unique clones to be characterized in detail or by additional cDNA selection experiments carried out with cDNAs from tissues not used in the present study. The novel fragments have already been useful as single-copy probes for screening larger-insert cDNA libraries and for refining and extending the physical map of the HLA class I region. Our mapping included a new mega YAC that contains a region extending telomeric of the HLA. The resulting map spans from about 90 kb centromeric of HLA-E to approximately 550 kb telomeric of HLA-F. Mapping data agree with an extended map of the region recently reported (Amadou et al., 1995). The positions of the HLA-F and MOG loci are slightly different, since we place them onto a distinct NotI fragment. The revised location does not alter the positions of the two genes significantly, as their positions shift only a few kilobases proximal to their previous ones. Moreover, our map confirms that D6S105, a marker linked to hereditary hemochromatosis, is positioned at least 1 Mb from HLA-A. Of the known genes that were identified, the position of the TC4 gene is in agreement with that recently reported (Wei et al., 1993). The localization of the TB31 and S17 genes has not been previously described. Moreover, the presence of a new member of the RNA helicase superfamily that maps telomeric of HLA-F has been identified and localized. In summary, we describe the isolation and mapping of 13 new gene fragments. Based on physical mapping of these fragments and of known genes, the HLA class I region is notably gene-rich, emphasizing the challenge of the identification of the HFE defect. ACKNOWLEDGMENTS This work was supported by grants from the Italian Ministry of Health, AIRC (Associazione Italiana Ricerca Cancro), and Telethon (Grant E.149). J.M.R. is a scholar of the Medical Research Council of Canada. The Authors thank Dr. Eve Roberts, Gastroenterology and Nutrition, H.S.C. (Toronto), for liver samples, L. Borgato (Verona) for technical assistance, and the YAC Screening Facility of the Canadian Genome Analysis and Technology Program.
REFERENCES Abderrahim, H., Sambucy, J., Iris, F., Ougen, P., Billault, A., Chumakov, I., Dausset, J., Cohen, D., and Le Paslier, D. (1994). Cloning the human major histocompatibility complex in YACs. Genomics 23: 520–527. Amadou, C., Ribouchon, M., Mattei, M., Jenkins, N., Gilbert, D., Copeland, N., Avoustin, P., and Pontarotti, P. (1995). Localization of new genes and markers to the distal part of the human major histocompatibility complex (MHC) region and comparison with the mouse: New insights into the evolution of mammalian genomes. Genomics 26: 9–20. Bisceglia, L., Grifa, A., Zelante, L., and Gasparini, P. (1994). Development of RNA SSCP protocols for the identification and screening of CFTR mutations: Identification of two new mutations. Hum. Mutat. 4: 136–140. Boretto, J., Jouanoulle, A., Yaouanq, J., El Kahloun, A., Mauvieux, V., Blayau, M., Perichon, M., Le Treut, A., Clayton, J., Borot, N.,
AP-Genomics
326
TOTARO ET AL.
Le Gall, J., Pontarotti, P., and David, V. (1992). Anonymous markers located on chromosome 6 in the HLA class I region: Allelic distribution in genetic hemochromatosis. Hum. Genet. 89: 33–36. Chen, I. T., Dixit, A., Rhoads, D. D., and Roufa, D. J. (1986). Homologous ribosomal protein in bacteria yeast and humans. Proc. Natl. Acad. Sci. USA 83: 6907–6911. Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162: 156–159. Church, G. H., and Gilbert, W. (1984). Genomic sequences. Proc. Natl. Acad. Sci. USA 81: 1991–1995. Drivas, G., Shih, A., Coutavas, E., Rush, M., and D’Eustachio, P. (1990). Characterization of four novel ras-like genes expressed in a human teratocarcinoma cell line. Mol. Cell. Biol. 10: 1793–1798. El Kahloun, A., Chauvel, B., Mauvieeux, V., Dorval, I., Jouanolle, A., Gicquel, I., Le Gall, J., and David, V. (1993). Localization of seven new genes around the HLA-A locus. Hum. Mol. Genet. 2: 55–60. Gasparini, P., Borgato, L., Piperno, A., Girelli, D., Olivieri, O., Gottardi, E., Roetto, A., Dianzani, I., Fargion, S., Schinaia, G., Cappellini, N., Gandini, G., Pignatti, P. F., Fiorelli, G., De Sandre, G., and Camaschella, C. (1993). Linkage analysis of 6p21 polymorphic markers and the hereditary hemochromatosis: Localization of the gene centromeric to HLA-F. Hum. Mol. Genet. 2: 571–576. Goei, V., Parimoo, S., Capossela, A., Chu, T., and Gruen, J. (1994). Isolation of novel -HLA gene fragments from the hemochromatosis region (6p21.3) by cDNA hybridization selection. Am. J. Hum. Genet. 54: 244–251. Grenett, H., Bounelis, P., and Fuller, G. (1992). Identification of a human cDNA with high homology to yeast omnipotent suppressor 45. Gene 110: 239–243. Jazwinska, E., Lee, S., Webb, J., Halliday, W., and Powell, W. (1993). Localization of the hemochromatosis gene close to D6S105. Am. J. Hum. Genet. 53: 347–352. Lu, D., and Yunis, J. (1992). Cloning expression and localization of an RNA helicase gene from a human lymphoid cell line with chromosomal breakpoint 11q23.3. Nucleic Acids Res. 20: 1967– 1972. McKusick, V. A. (1994). ‘‘Mendelian Inheritance in Man,’’ 11th ed, The John Hopkins Univ. Press, Baltimore. Miller, S., Dykes, D., and Polesky, H. (1988). A simple salting out procedure for extracting DNA from human cells. Nucleic Acids Res. 16: 1215. Raha-Chowdhury, R., Bowen, D., Burnett, A., and Worwood, M. (1995). Allelic association and homozygosity at loci from HLA-B to D6S299 in genetic haemochromatosis. J. Med. Genet. 32: 446–452. Rommens, J., Lin, B., Hutchinson, G., Andrew, S., Goldberg, Y.,
/ m4990$3868
01-04-96 18:28:00
gnma
Glaves, M., Graham, R., Lai, V., McArthur, J., Nasir, J., Theilmann, J., McDonald, H., Kalchman, M., Clarke, L., Schappert, K., and Hayden, M. (1993). A transcription map of the region containing the Huntington disease gene. Hum. Mol. Genet. 2: 901– 907. Rommens, J., Mar, L., McArthur, J., Tsui, L-C., and Scherer, S. (1994). Towards a transcriptional map of the q21–q22 region of chromosome 7. In ‘‘Identification of Transcribed Sequences’’ (U. Hochgeshwender and K. Gardiner, Eds.), pp. 65–79, Plenum, New York. Sanger, F., Nicklen, S., and Coulson, A. (1977). DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74: 5463– 5467. Sarkar, G., Yoon, H-S., and Sommer, S. (1992). Screening for mutations by RNA single-strand conformation polymorphisms (rSSCP): Comparison with DNA-SSCP. Nucleic Acids Res. 20: 871–878. Simon, M., Le Mignon, L., Fauchet, R., Yaouanq, J., David, V., Edan, G., and Bourel, M. (1987). A study of 609 HLA haplotypes marking for the hemochromatosis gene: Mapping the gene near the HLA-A locus and characters required to define a heterozygous population; hypothesis concerning hemochromatosis-HLA association. Am. J. Hum. Genet. 41: 89–105. Stone, C., Pointon, J., Jazwinska, E., Halliday, J., Powell, L., Robson, K., Monaco, A., and Weatherall, D. (1994). Isolation of CA dinucleotide repeats close to D6S105; linkage disequilibrium with hemochromatosis. Hum. Mol. Genet. 3: 2043–2046. Totaro, A., Grifa, A., Roetto, A., Zelante, L., Camaschella, C., and Gasparini, P. (1994). A new complex polymorphic repeat close to HLA-A and HLA-E loci. Hum. Genet. 94: 578. Totaro, A., Grifa, A., Roetto, A., Lunardi, C., D’Agruma, L., Sbaiz, L., Zelante, L., De Sandre, G., Camaschella, C., and Gasparini, P. (1995). New polymorphisms and markers in the HLA class I region: Relevance to hereditary hemochromatosis (HFE). Hum. Genet. 95: 429–434. Volz, A., Boyle, J., Cann, H., Cottingham, R., Orr, H., and Ziegler, A. (1994). Report of the second international workshop on human chromosome 6. Genomics 21: 464–472. Weber, J., Kwitek, A., May, P., and Zoghbi, H. (1991). Dinucleotide repeat polymorphism at the D6S105 locus. Nucleic Acids Res. 19: 968. Wei, H., Fan, W-F., Xu, H., Parimoo, S., Shukla, H., Chaplin, D., and Weissman (1993). Genes in one megabase of the HLA class I region. Proc. Natl. Acad. Sci. USA 90: 11870–11874. Yaouanq, J., Perichon, M., Chorney, M., Pontarotti, P., Le Treut, A., El Kahloun, A., Mauviex, V., Blayau, M., Jouanolle, A., Chauvel, B., Moirand, R., Nouel, O., Le Gall, J., Feingold, J., and David, V. (1994). Anonymous marker loci within 400 kb of HLA-A generate haplotypes in linkage disequilibrium with the hemochromatosis gene (HFE). Am. J. Hum. Genet. 54: 252–263.
AP-Genomics