Characterization of Mutants of the Vitamin-D-Binding Protein/Group Specific Component: GC Aborigine (1A1) from Australian Aborigines and South African Blacks, and 2A9 from South Germany
Post on 30-Sep-2016
Vox Sang 1995;68:50-54
Andrea Kofer a Andreas Braun a* Trefor Jenkins Sue V! Serjeantson Hartwig Cleve a
a Institute of Anthropology and Human Genetics, and Dr. von Haunersches Kinderspital, University of Munich, Germany;
' MRC Human Ecogenetics Research Unit. The South African Institute for Medical Research and University of the Witwatersrand, Johannesburg, South Africa, and The John Curtin School of Medical Research, Canberra, Australia
Characterization of Mutants of the Vitamin-D-Binding Protein/Group Specific Component: GC Aborigine ( I A I ) from Australian Aborigines and South African Blacks, and 2A9 from South Germany
................................................................................................. Abstract The structure and organization of the human vitamin-D-binding protein gene (DBP, group-specific component, GC) have recently been determined. Each ex- on may now be amplified by the PCR method using oligonucleotide primers deduced from the intron sequences near their 5' ends and 3' ends. In this study we examined the anodal GC variants IAl and 2A9. Genomic DNA of the variant 1Al was obtained from Australian Aborigines and from South African Bantu-speak- ing Blacks. Amplification and sequencing of exon 11 of 1Al revealed a point mutation in codon 429 at the second position. It is remarkable that this mutation was found in the Australian 1Al variant and in the African IAl variant, and raises the question whether the mutation in these two ethnic groups has a common ori- gin. Genomic DNA of the 2A variant called 2A9 was obtained from South Ger- many and a point mutation also concerning position 429 in exon 11 was found. The nucleotide exchange in this case, however, was at the first position of the codon. The widely distributed genetic polymorphism of DBPiGC is located in exon 1 1 and is characterized by substitution at amino acid positions 416 and 420. Variant IAl is due to a second site mutation of the allele GC*lF; variant 2A9 is due to a mutation in the GC*2 allele. .....................
Since the discovery of the group-specific component (GC) [l], the polymorphism of this protein has been shown to be very complex. There are three common alleles (GC*2, GC*lS and GC*IF) and more than 120 variant alleles in the
human population . The molecular differences of the three common alleles reside in exon 11 at codons 416 and 420 [3,4]. At position 416, the codon for aspartic acid (GAT) was found for the alleles GC*lF and GC*2, the codon for glutamic acid (GAG) for GC*IS. At position 420, the codon for threonine (ACG) was found for GC*IF and GC*IS, the
Rccciwd: .January I X . 1494 10 IY95 S. Karger AG. Basel Reviscd manuscript Institute of Anthropology and I luman Genetics (1042- Y007.'Y5 068 I 0050
Acccptcd..luly I I . I Y W
received: Juiie29. lY94 University of Munich $ X.00~0 I7icIi;ird-Wagiicr-Str~iss~ I0 1)-XO333 Miinchen ((icrmany)
codon for lysine (AAG) for GC*2. Position 416 includes a Hue111 restriction site for the allele GC*IS and the position 420 includes a Sty I restriction site for allele GC"2. Thus, the three common genetic GC types can also be classified by restriction fragment analysis .
The australian variant 1Al was first reported by Cleve et al.  and called GC Aborigine (GC-Ab). The African var- iant 1Al was described by Hirschfeld  and Parker et al.  and was originally called GC-Y
The variants present in these two populations are indis- tinguishable by immunoelectrophoresis, electrophoresis on starch, agarose or polyacrylamide gels, antigen-antibody- crossed electrophoresis , and by isoelectric focusing . It was, therefore, suggested that they may represent the same mutation.
In this study, we examined the Australian and the African 1Al variants at the molecular level. Another anodal variant, 2A9, was also analyzed.
Material and Methods
Genomic DNA Genomic DNA of Australian Aborigines from North Western Aus-
tralia was obtained from the John Curtin School of Medical Research, Canberra, Australia. It was DNA of a GC*IAI homozygote and of a GC*IAI/IF heterozygote. Genomic DNA of a South African Bantu- speaking Black individual was obtained from the South African In- stitute of Medical Research, Johannesburg. It was DNA of a GC*lAI/IF heterozygous person.
The genomic DNA of the variant 2A9 was prepared in our lab- oratory as described in Braun et al. . I t was DNA of a GC*2A9/1S heterozygous individual as classified by isoclectric focusing.
Ampl&ution of Exon I 1 bv Polvmeruse Chuin Reuction For amplifying exon 1 I , we used the intron-specific oligonucleo-
tides EllF (5'ACATGTAGTAAGACCTTAC 3') and EllR (5'GATTG- GAGTGCATACGTTC 3') (MWG, Biotech, Germany) which flank the exon near the 5' end and the 3' end, respectively. The total reaction volume of 50 pl included about 15 ng to 500 ng of genomic DNA, 50 ng of each primer, 200 p M of each dNTP (Pharmacia, Freiburg, Germany), 1.25 mM MgCI2, and 1.25 U Taq DNA polymerase (Prome- ga, Heidelberg, Germany). The DNA polymerase buffer contained 50 mM KCI, 10 mMTris-HCI (pH 9.0), 0.1% Triton X-100 (Promega). Thc PCR was carried out in a Hybaid Thermal Reactor by the follow- ing program: 5 min 94C (IQ; 35 amplification cycles with 1 min 94"C, 1 min 50"C, and 30 s 72C; 5 min 72C (llE3).
Booster-PCR PCR ofthe Australian samples which contained only a small quan-
tity of DNA was not sufficient. Therefore, a booster-PCR using I pl of the cleaned (Magic PCR purification kit, Promega) first PCR product was carried out. PCR conditions were the same as described above, with the exception of only 20 amplification cycles.
Ligution, Siibcloning und Sequencing qfthe PCR Products The PCR products were blunt-ended with T4 DNA polymerase
(Boehringer, Mannheim. Gcrmany) and ligated in the plasmid vector pUC19 (BRL, Eggenstein, Germany) which was cut with Hindll. Li- gation was done with 1U T4 DNA ligase (Promega) at 15C for 14 h. The ligation products were subcloned in the Escherichiu coli strain MC 1061. The transformed cells were plated on LB agar containing ampicillin and grown overnight. After sweeping all colonies (about 300) from the agar with LB medium, the plasmids were purified with the Quiagen mini-kit (Diagen, Diisseldorf, Germany) according to the manufacturer's instructions. This procedure avoids the manifestation of a possible Taq DNA polymerase error and guarantees sequencing of both allelic products of each sample in a batch. To confirm the ob- tained results, single clones were sequenced. Sequencing was per- formed with the Sequenase version 2.0 sequencing kit (USB, Cleve- land, Ohio, USA) using the specific primers EllF and EllR ( 5 ngipl), and [u-~'P]~ATP (Amersham, Braunschweig, Germany) for labelling. Electrophoresis was done with a 6% acrylamidc, 6 M urea gel. The running conditions were 2,300 V at 50C for 2 h.
Restriction Anulwis ofthe PCR P r ~ d i ~ c t s with C fol The PCR products were ethanol precipitated and dissolved in double
distilled water. Digestion was done with 0.5 pl (6 U) of C'I (Boehr- inger) at 37C for 1 h. The fragments were separated in a 4% agarose gel (NuSieve 3:l agarose; Biozym Diagnostic, Hameln, Germany) and stained with ethidium bromide. The electrophoresis buffer contained 85 mMTris, 90 mM boric acid, and 2 mMEDTA. The I-kb ladder sup- plied by BRL (Eggenstein, FRG) was used as the size standard.
Single-Strund Confortnution Polvmorphisni Anulysis ofthe PCR Products To confirm that the IAl variant from South Africa and from Aus-
tralia are identical, all 12 coding exons ofthe hDBP gene, including the exon I I, were also analysed by single-strand conformation polymor- phism (SSCP). Each exon of the African and Australian variant IAl and of a GC*IF/IF homozygous control was amplified by the PCR method. Two microliters of each PCR product were mixed with forma- mide and loaded to a 15% polyacrylamide gel after denaturation at 95C. Exon I I of a homozygous GC*IS/IS, GC*2/2, a heterozygous GC*IS/2, and GC*IS/2A9 was analyzed as well. The method is de- scribed in detail elsewhere [lo].
Using the intron-specific primers El IF and ElIR, we ob- tained a 233-bp long PCR product including the 133-bp- long exon 1 I, 37 bp of the 3'end of intron 10 and 63 bp of the 5' end of intron 11. After sequencing the exon 11 of 1Al from Australia as well as from South Africa, we found the nucle- otide pattern of homozygous GC*lF in both, when evaluat- ing positions 416 and 420 (fig. la). At position 429, we dis- covered a nucleotide substitution from G to A in the second position of the codon (fig. la, b). This transition causes an amino acid exchange from arginine to histidine. This change could explain the different position of the 1Al var- iant in IEF in contrast to the normal 1F allele.
iants 1Al and 2A9 in exon 11 concerning the positions 416,420, and 429. Shown is the se- quence pattern of a heterozygous Bantu- speaking Black (IFAAI), a heterozygous Ab- origines (IFAAI), a homozygous Aborigines (IAMAI), and a heterozygous Caucasian with the allele combination ISi2A9. b Se- quence analysis of single clones at the posi- tion 429. From the left to the right: GC*IF, variant IAl, variant 2A9. b
Sequencing the exon 11 of the variant 2A9 confirmed the result of IEF. The sequence in codons 416 and 420, respec- tively, revealed the GC type 2 4 s . A point mutation was found at codon 429. C was replaced by T at the first position of the codon causing an amino acid change from arginine to cysteine (fig. la, b).
The unmutated positions 428 and 429 include a restric- tion site for the enzyme CfoI (GCG/C). Thus, the point mu- tations found in 1Al and 2A9 should change the restriction pattern of this enzyme. The result of the analysis is shown in figure 2' As expected' the l1 Of GC*lS and GC*lF ho- mozygoteswascutintofragmentsof108and 125 bp. In con- trast, exon 11 of the variants was not digested. The nucleo-
Fig. 2. C f i l restriction analysis of the PCR products, containing exon 11. (-) = without enzyme; (+) = with enzyme. Lane 1: size stan- dard; lane 2: GC*2/2 (-); lane 3: GC212 (+); lane 4: GC*IS/2A9 (+):
tide sequence differences in exon 11 are summarized in fig- ure 3.
The SSCP pattern of the exon 11 of the South African and Australian 1Al variant, respectively, was identical and it was clearly different from the pattern of the exon 11 of a homozygous lF/lF type. Analysis of the remaining 11 exons failed to reveal any differences in the SSCP pattern from the
lane 5 : GC*IFIIF (-); lane 6: GC*IF/IF (+); lane 7: GC*IF/IAI (+); lane 8: GC*IAI/IAI (+); lane 9: size standard.
52 Kofler/Braun/Jenkins/Serjeantson/Cleve DBP/GC Mutants: Analysis of GC 1 A 1 and 2AY
POS . Pos. Pos . 4 1 6 4 2 0 4 2 9
G C * l F 5 ' TTGCCTMGCCACACCCASGAACTGGCAAAGCTGGTTAACAAtmTCA 3 '
G C * l A l 5 ' TTGCCTWGCCACACCCAQ3GAACTGGCAAAGCTGGTTAACAAGWTCA 3 '
G C * l S 5 ' TTGCCTMGCCACACCCASGAACTGGCAAAGCTGGTTAACAAtSTCA 3 '
GC*2 5' TTGCCTWGCCACACCCAGAACTGGCAAAGCTGGTTAACAA$mTCA 3 '
GC*2A9 5 ' TTGCCTWGCCACACCCAAGGAACTGGCAACTGGCAAAGCTGGTTAACAAGmTCA 3 '
Fig.3. Sequence of the three comnion alleles (GC*IF, GC*IS, GC*2) and the variants 1.41 and 2A9 in the region coding for the amino acid 416,420, and 429. The recognition sitc for the restriction enzyme C\bl is marked by asterisks.
GC*lF/IF homozygous control. The pattern of the exon 11 of 2A9 was also clearly distinguishable from the controls (fig. 4).
Evaluation of the codons 416 and 420 showed variant 2A9 to be second-site mutation of allele 2 and variant IAl as a second site mutation of allele IF.
It is remarkable that we found this mutation in the Aus- tralian 1Al variant as well as in the African 1Al variant. This raises the question of the origin of this mutation in both eth- nic groups: Is it an ancient mutation, distributed in the two ethnic groups? or Have the mutations arisen independently? If so, is position 429 a mutation hot spot?
To provide further evidence for the identity, the recently discovered polymorphic variable poly (dA) Alu repeat (AluVpA) [ 1 I ] was analyzed. We investigated the 3' end of the Alu repeat which contains a tetranucleotide repeat (TAAA),,. The polymorphic repeat resides at the 3' end of intron 8 of the DBP gene and is located about 2,500 bp up- stream of codon 429. The Australian 1Al variant contained 9 TAAA repeats and the African variant had 6 repeats. The AluVpA polymorphism is differently distributed in human populations: in South Germany, there is an association be- tween alleles GC*IF and GC*I8-IO (intron 8) with linkage disequilibrium [ I l l . This association was not found in Afri- can Blacks, amongst whom a higher degree of variability exists at the I8 locus than in the German population. The distribution of 18 alleles is markedly different, with the GC*18-10 being very rare in African Blacks [Bichlmaier et al., submitted]. Thus, alleles of the serum GC polymor-
1 2 3 4 5 6 I 1 . . .
Fig.4. SSCP analysis of the exon 11 amplificate (GC*lF, *IS, *2 and the variants IAl, 2A9) ofthe human DBP gene on a 15% polyacry- lamidc gcl. Lane 1: GC*IF/IF; lane 2: GC*IFIIAI, lane 3 : GC*IS/IS; lane4: GC*2/2; lane 5 : GC*IS/2A9; lane 6: GC*IS/2.
phism and of the intron 8 polymorphism are not associated in African Blacks. I t is, therefore, not possible to distinguish between the two possibilities mentioned above.
We also examined the sequence polymorphism in the un- translated region of exon 1 which we have recently de- scribed [ 121. We found the Australian heterozygote lF/IAl to be homozygous for T-23 and the African heterozygote IF/IAl to be homozygous for C-23. Thus, no association with the alleles of the serum GC polymorphism can be found.
In summary, although the GC* 1Al variant in Australians and Africans would appear to be due to the same mutational event, the other differences between the mutant alleles in the two populations suggest that they are not identical by descent. Both the AluVpA polymorphic site in intron 8 and the polymorphic site at nucleotide 23 in the untranslated re-
gion of exon 1 differ between the two. However, if these two polymorphic sites are hypemutable, it is possible that the
mon origin and were further changed in subsequent linea- ges, leading to present-day populations. The identification of further variation between the DBP alleles will clarify the origin of the genetic variants within the indigenous peoples of Australasia and Africa.
Australian and the African GC*]A1 variants share a This study w a s supported b y Deu t sche Forschungsgemeinschaft . Bad Godesbe rg , by a g ran t t o us (H.C., Project CI 27114-2 and CI 27 / 15-1) for which we are
I Hirschfeld J: Immune-electrophoretic demon- stration of qualitative differences in human sera and their relation to the haptoglobins. Acta Pathol Microbiol Scand 1959;47: 160-1 68.
2 Cleve H, Constans J: The mutants of the vitamin D-binding protein: More than I20 variants of the GC/DBP system. Vox Sang 1988;54:215- 225.
3 Reynolds RL, Sensabough GF: Use ofthe poly- merase chain reaction for typing GC variants; in Polesky HF, Mayr WR (eds): Advances in Fo- rensic Haemogenetics. Springer. Berlin, 1990, vol. 3, pp 158-161.
4 Braun A, Bichlmaier R. Cleve H: Molecular analysis o f the gene for the human vitamin D- binding protein (group specific component): Allelic differences of the common genetic GC types. Hum Genet 1992;89:401406.
5 Cleve H, Kirk RL. Parker WC, Bearn AG. Schacht LE, Kleinnian H, Horsfall WR: Two genetic variants of the group-specific compo- nent of human serum: Gc Chippewa and Gc Ab- origine. Am J Hum Genet 1963;15:368-379.
6 Hirschfeld J: The Gc-system. Progr Allergy I 962;6: 155-1 86.
7 Parker WC, Clcve H. Beam AG: Determination of phenotypes in the human group-specific component (Gc) system by starch gel electro- phoresis. Am J Hum Genet 1963;18:201-214.
8 McDermid EM, Cleve H: A comparison of the fast migrating Gc-variant of Australian Aborig- ines. New Guinean Indigenes, South African Bantu. and Black Americans. Hun1 Hcred 1972; 22:249-253.
9 Constans J , Cleve H: Group-specific compo- nent. Report of the First International Work- shop. Hun1 Genet 1979;48: 143-149.
10 Burgemeister R, Rotzer E, Gutensohn W Iden- tification of a new missense mutation in exon 2 of the human hypoxanthine phosphoribosyl- transferase gene (HPRT,,,,). A further example of clinical heterogeneity in HPRT-deficiences. Hum Mutat, in press.
1 1 Braun A, Bichlmaier R, Miiller B,Cleve H: Mo- lecular evaluation o f an Alu repeat including a polymorphic variable poly(dA) (AluVpA) in the vitamin D binding protein (DBP) gene. Hum Genet 1993;90:526-532.
I2 Braun A, Kofler A, Bichlmaier R, Kammerer S. Cleve H: A novel sequence polymorphism in cxon 1 ofthe human vitamin D-binding protein (hDBP) gene. Hum Mol Genet 1993;2: 1750.
Kofler lBrauniJen kinslSerjeantsoniCleve D B P i G C Mutants : Analysis of GC 1 A 1 and 2A9