| Human Molecular Genetics | Pages |
Global sequence diversity of BRCA2: analysis of 71 breast cancer families and 95 control individuals of worldwide populations
Introduction
Results
General sequence variation
Disease-associated BRCA2 mutations
Sequence variants in HBC/HBOC families
Sequence variants detected in the control population
Frequency and global distribution of continent- and population-specific sequence variants
Discussion
Materials And Methods
Breast and breast-ovarian cancer families
Control individuals
DNA extraction and PCR conditions
DHPLC
Direct DNA sequencing
Cloning of BRCA2 mutations with the TA cloning kit
Acknowledgements
Abbreviations
References
Global sequence diversity of BRCA2: analysis of 71 breast cancer families and 95 control individuals of worldwide populations
INTRODUCTION
Since the identification of the BRCA2 gene in 1995 (1,2), >468 sequence variants have been detected (3). These sequence variants were identified predominantly in individuals from breast and breast-ovarian cancer (HBC/HBOC) families and primary breast cancer (BC)/ovarian cancer (OC) cases. Approximately 68% were protein truncating mutations, and 32% were polymorphisms or missense mutations of unknown biological relevance, i.e. unclassified variants. Although truncating mutations may be assumed to cause disease, pathogenicity involving unclassified variants is more equivocal because of insufficient information concerning both protein function and genetic variation. Knowledge with regard to potential functional domains as well as the overall function of BRCA2 itself remains limited (4-8). Comparative information regarding BRCA2 genetic diversity in various populations is also limited. The temptation to implicate a missense mutation that occurs at low frequency as a potential functional lesion must be tempered by the possibility that the variant may exist more frequently in another population and thus be considered a mere polymorphism. Schrijver et al. reported such a case for the fibrillin-1 gene (FBN1) (9). Initially, a missense mutation was detected predominantly in Caucasian probands with Marfans syndrome (MFS) and consequently classified as likely to be disease associated. Subsequent screening of 416 control individuals from various world populations permitted the classification of the allele as a polymorphism, rather than a mutation, prevalent in Asian and Latin American populations.
The aim of this study was to produce a comprehensive summary of BRCA2 sequence variants in representatives of diverse global populations. DNA from 40 populations was analyzed in conjunction with that of 71 Austrian HBC/HBOC families. Analysis was accomplished by denaturing high-performance liquid chromatography (DHPLC), a novel automated heteroduplex detection method with a proven sensitivity and specificity of >95% (10,11).
RESULTS
General sequence variation
The coding region and intron-exon boundaries (13 056 bp) of BRCA2 of 95 control individuals and 71 HBC/HBOC families were analyzed by means of DHPLC, with heteroduplexes subsequently being identified through direct sequence analysis. A total of 82 different simple sequence variants were detected (66% transitions, 29% transversions and 5% deletions). As of September 1998, only 31 of those variants had been reported previously (Table 1). In the coding regions, 53 out of 10 257 nucleotides (1 in 194) and in the non-coding regions 26 out of 2799 nucleotides (1 in 108) varied.
Table 1.
| Exon/intron | Nucleotidea | Codon | Nucleotide change | Amino acid changeb | Global heterozygosity | Geographic distributionc | BIC (as of 9/98) citations |
| 2 | 203 | - | G->A | non-coding | 0.260 | Eurasia/Oceania | BIC,2,12-16 |
| 217 | - | T->G | non-coding | 0.006 | Oceania | - | |
| 4 | 550 | 108 | A->C | Asn->His (nc) | 0.006 | Africa | - |
| 10 | 1093 | 289 | A->C | Asn->His (nc) | 0.120 | Global | BIC,15,16 |
| 1207 | 327 | A->G | Lys->Glu (nc) | 0.012 | Austria | - | |
| 1342 | 372 | A/C | Asn/His (nc) | 0.420 | Global | BIC,2,15-18 | |
| 1515 | 429 | A->G | silent | 0.006 | Pakistan | - | |
| 1593 | 455 | A->G | silent | 0.110 | Global | BIC,15,18 | |
| 1972 | 582 | A->C | Thr->Pro (nc) | 0.006 | Central Asia | - | |
| 11 | 2347 | 707 | G->T | Asp->Tyr (c) | 0.006 | Africa | - |
| 2457 | 743 | T->C | silent | 0.130 | Global | 2,15 | |
| 2578 | 784 | A->G | Met->Val (c) | 0.006 | Asia | BIC | |
| 2778 | 850 | A->G | silent | 0.006 | Asia | - | |
| 2885 | 886 | A->T | Asn->Ile (nc) | 0.006 | Africa | - | |
| 3031 | 935 | G->A | Asp->Asn (nc) | 0.006 | Austria | BIC | |
| 3111 | 961 | G->A | silent | 0.006 | Austria | - | |
| 3199 | 991 | A->G | Asn->Asp (nc) | 0.150 | Global | 2,15-17 | |
| 3624 | 1132 | A->G | silent | 0.340 | Global | - | |
| 4035 | 1269 | T->C | silent | 0.230 | Global | BIC,2,15 | |
| 4086 | 1286 | delAAA | del Leu | 0.006 | Africa | - | |
| 4097 | 1290 | G->A | Cys->Tyr (c) | 0.006 | Africa | - | |
| 4194 | 1322 | C->T | silent | 0.006 | Africa | - | |
| 4469 | 1414 | C->Td | Thr->Met (nc) | 0.012 | Africa | - | |
| 4486 | 1420 | G->T | Asp->Tyr (c) | 0.030 | Austria | BIC | |
| 4765 | 1513 | G->A | Asp->Asn (nc) | 0.006 | Australia | - | |
| 4791 | 1521 | G->A | silent | 0.018 | Africa | - | |
| 5007 | 1593 | A->C | Glu->Asp (c) | 0.006 | Pakistan | - | |
| 5427 | 1733 | C->Td | silent | 0.036 | Europe/Asia | - | |
| 5540 | 1771 | G->A | Gly->Asp (nc) | 0.006 | America | BIC | |
| 5642 | 1805 | A->G | Asn->Ser (c) | 0.006 | Africa | - | |
| 5868 | 1880 | T->G | Asn->Lys (nc) | 0.012 | Africa | BIC,15,17 | |
| 5932 | 1902 | G->A | Asp->Asn (nc) | 0.012 | Africa | BIC | |
| 5972 | 1915 | C->Td | Thr->Met (nc) | 0.048 | Austria/Oceania | BIC,13,15-18 | |
| 6328 | 2034 | C->Td | Arg->Cys (nc) | 0.006 | Austria | BIC,15-17 | |
| 6448 | 3074 | C->A | His->Asn (nc) | 0.006 | Africa | BIC | |
| 6640 | 2138 | G->T | Val->Phe (c) | 0.006 | Africa | 13 | |
| 6741 | 2171 | C->G | silent | 0.018 | Africa | - | |
| 7049 | 2274 | G->T | Gly->Val (nc) | 0.006 | Austria | 15,17,19 | |
| 14 | 7470 | 2414 | A->G | silent | 0.250 | Global except America | BIC,2,12,14-16,18 |
| 7625 | 2466 | C->T | Ala->Val (c) | 0.018 | Africa | BIC,17,18 | |
| 15 | 7666 | 2480 | T->G | Leu->Val (c) | 0.006 | Africa | - |
| 7697 | 2490 | T->C | Tyr->His (nc) | 0.018 | America | BIC | |
| 7772 | 3015 | C->T | Thr->Ile (nc) | 0.006 | Austria | BIC | |
| 18 | 8320 | 2697 | G->Ad | Arg->His (c) | 0.006 | Africa | - |
| 22 | 9058 | 2944 | A->T | Ile->Phe (c) | 0.006 | Africa | BIC |
| 9079 | 2951 | G->A | Ala->Thr (nc) | 0.006 | America | BIC | |
| 9133 | 2969 | G->Ad | Val->Met (c) | 0.006 | Asia | - | |
| 24 | 9465 | 3079 | T->C | silent | 0.006 | Africa | - |
| 25 | 9642 | 3138 | A->G | silent | 0.006 | Africa | - |
| 27 | 9998 | 3257 | A->G | Lys->Arg (c) | 0.006 | Africa | - |
| 10 056 | 3276 | A->T | Arg->Ser (nc) | 0.006 | Africa | - | |
| 10 152 | 3308 | C->Td | silent | 0.006 | Africa | BIC | |
| 10 204 | 3326 | A->T | 3326 STOP | 0.012 | Austria/Sardinia | BIC,15,17,20,21 | |
| 10 319 | 3364 | delCT | 3366 STOP | 0.006 | America | - | |
| 10 462 | 3412 | A->G | Ile->Val (c) | 0.024 | Africa/Austria/America | BIC,15,17,20 | |
| 2 | 295+62 | T->G | non-coding | 0.006 | Asia | 14 | |
| 3 | 545-22 | C->T | non-coding | 0.006 | Africa | - | |
| 4 | 654-47 | G->T | non-coding | 0.006 | Austria | - | |
| 6 | 744+18 | T->C | non-coding | 0.006 | Asia | - | |
| 745-4 | C->G | non-coding | 0.006 | Africa | - | ||
| 8 | 909+56 | C->T | non-coding | 0.024 | Africa/Asia/Oceania | - | |
| 11 | 7069+74 | T->A | non-coding | 0.006 | Africa | - | |
| 7069+80 | delTTAA | non-coding | 0.340 | Global except America | BIC | ||
| 13 | 7236-62 | A->G | non-coding | 0.006 | Austria | - | |
| 14 | 7663+53 | C->Td | non-coding | 0.130 | Global | BIC | |
| 15 | 7830-27 | T->A | non-coding | 0.006 | Asia | - | |
| 16 | 8034-14 | T->C | non-coding | 0.048 | Austria | BIC | |
| 17 | 8204+12 | G->A | non-coding | 0.012 | Africa | - | |
| 18 | 8560-66 | T->C | non-coding | 0.006 | Africa | - | |
| 8560-64 | A->G | non-coding | 0.018 | Africa | - | ||
| 19 | 8715+47 | C->T | non-coding | 0.018 | Africa | - | |
| 21 | 8983-79 | G->A | non-coding | 0.006 | Asia | - | |
| 8983-66 | T->C | non-coding | 0.220 | Global | - | ||
| 22 | 9181+16 | C->T | non-coding | 0.006 | Austria | - | |
| 9181+98 | T->C | non-coding | 0.012 | Austria | - | ||
| 25 | 9730-71 | G->T | non-coding | 0.006 | Africa | - | |
| 26 | 9876+54 | G->Ad | non-coding | 0.012 | Africa | - | |
| 9876+84 | G->A | non-coding | 0.006 | Asia | - | ||
| 9876+106 | delT | non-coding | 0.006 | Africa | - |
Disease-associated BRCA2 mutations
One frameshift mutation with a deletion of 5 bp in exon 11 and two splice site mutations in introns 5 and 21, respectively, were found in six out of 71 HBC/HBOC families (8%). All three are novel mutations and are likely to be disease associated. Pedigrees of the six HBC/HBOC families are shown in Figure






Figure 1. Pedigrees of the six HBC/HBOC families with apparently disease-associated BRCA2 mutations.
The frameshift mutation in exon 11 at nucleotide 6633 delCTTAA results in a stop codon at 2137. In the family with this mutation, three cancer cases could be identified: two BC cases at the age of 35 (carrier of the mutation) and 50 years (not available for testing), and one OC case at the age of 68 years (deceased, no DNA available for testing). The splice site mutation in intron 5 showed an A to G change in the third base pair of the conserved intronic splice donor site. This sequence variant was detected once in 332 chromosomes in a family with three cases of BC (at the ages of 42, 57 and 62 years, all carriers) and one case of BC and OC (at the age of 46 years, deceased, no DNA available for testing). Sequence analysis of the two non-diseased members of the third generation (aged 39 and 30 years) revealed that one inherited the sequence variant. The splice site mutation in intron 21 showed a G to A substitution in the first base of the conserved intronic splice acceptor site. The four families identified with this mutation were A3, G3, G4 and F47. A3 contained three cases of BC at the ages of 38, 48 and 62 years. Only the two younger cases were found to be carriers of this sequence variant. The older case is a sporadic BC. G3 presented three cancer cases, two of which could be tested for the presence of the sequence variant: one case of male BC at the age of 45 years (deceased, no DNA available for testing), one female with BC (at the age of 47 years), melanoma (at the age of 50 years) and OC (at the age of 60 years), and one male with stomach cancer at the age of 88 years. G4 contained three cases of male BC [at the ages of 54 (refused to be tested), 60 and 76 years (deceased, no DNA available for testing)], three cases of female BC [at the ages of 45 (carrier), 54 and 68 years (the latter two are deceased and no DNA has been available for testing)] and one case of prostate cancer (65 years) and stomach cancer (60 years), respectively (both patients are deceased and no DNA has been available for testing). In F47, three BC cases were identified at the ages of 36, 39 and 42 years, with the two younger patients definitely carrying the mutation, while no DNA had been available for testing from the third deceased case. In all four families, the mutation segregated from one generation to the other, and all identified carriers were affected with cancer. Unfortunately, due to the greater number of deceased family members with insufficient pathological specimens for DNA extraction and the refusal of others to be tested because of fear of discrimination, penetrance of the mutations could not be assessed. There was at least one case of sporadic BC in one of the six families (A3).
To resolve the changes caused by these splice site mutations, cDNA was cloned. In three different cloning experiments, a total of 15 plasmids were sequenced. For the intron 5 mutation, four clones showed a complete loss of exon 5. For the intron 21 mutation, three clones showed a complete loss of exon 22 and two clones a loss of exon 22 plus the first 51 bp of exon 23. On the basis of these findings, we deduce that the mutation in the intronic splice donor site of intron 5 causes a complete loss of exon 5, and the mutation in the intronic splice acceptor site of intron 21 can cause both a complete loss of exon 22 and a loss of exon 22 plus the first 51 bp of exon 23.
Sequence variants in HBC/HBOC families
Eight sequence variants were detected only in 11 HBC/HBOC families: five missense mutations, one silent mutation, one nonsense mutation and two sequence variations in introns 4 and 13 (Table 2). Four missense mutations, K327E, D935N, R2034C and T3015I, resulted in non-conservative amino acid changes (Table 2). One family, F54, with two BC cases at the ages of 48 and 49 years, presented with three sequence variants that were only detected in HBC/HBOC families: D935N, G961G and a non-coding sequence change in intron 13.
Table 2.
| Exon/intron | Nucleotide (codon) | Nucleotide change | Effect on coding sequence | Amino acid change | Family | No. of BC/OC (age of onset in years) |
| 10 | 1207 (327) | A->G | Lys->Glu | non-conservative | F40 | 6 BC (34, 37, 42, 42, 46, 47) |
| M16 | 2 BC (29, 49) | |||||
| 11 | 3031 (935) | G->A | Asp->Asn | non-conservative | F54 | 2 BC (48, 49) |
| 11 | 3111 (961) | G->A | silent, Gln | silent | F54 | 2 BC (48, 49) |
| 11 | 4486 (1420) | G->T | Asp->Tyr | conservative | F35 | 3 BC (38, 49, 72) |
| F50 | 3 BC (39, 52, 60) | |||||
| M26 | 3 BC (36, 50, 55 bilateral) | |||||
| M28 | 2 BC (39, 47 bilateral) | |||||
| M41 | 2 BC (33, 38) | |||||
| 11 | 6328 (2034) | C->T | Arg->Cys | non-conservative | M31 | 1 BC (29) |
| 15 | 7772 (3015) | C->T | Thr->Ile | non-conservative | M2 | 2 BC (32, 48) |
| Intron 4 | 654-47 | G->T | non-coding | F42 | 3 BC (43, 53, 59) | |
| Intron 13 | 7236-62 | A->G | non-coding | F54 | 2 BC (48, 49) |
Table 3.
| Continent (no. of individuals) |
Nucleotide | |||||||||||
| 203 | 1093 | 1342 | 1593 | 2457 | 3199 | 3624 | 4035 | 7470 | 7069+80 | 7663+53 | 8983-66 | |
| Africa (21) | 0.000 | 0.143 | 0.190 | 0.048 | 0.143 | 0.238 | 0.143 | 0.286 | 0.190 | 0.190 | 0.190 | 0.381 |
| Austrian controls (18) | 0.389 | 0.167 | 0.500 | 0.111 | 0.111 | 0.111 | 0.444 | 0.167 | 0.333 | 0.500 | 0.111 | 0.167 |
| Austrian HBOC (71) | 0.343 | 0.029* | 0.429 | 0.029 | 0.029 | 0.057 | 0.429 | 0.243 | 0.243 | 0.400 | 0.043 | 0.029* |
| Other Europeans (12) | 0.333 | 0.167 | 0.583 | 0.250 | 0.250 | 0.250 | 0.333 | 0.333 | 0.250 | 0.333 | 0.250 | 0.500 |
| Asia (31) | 0.258 | 0.194 | 0.484 | 0.194 | 0.258 | 0.226 | 0.290 | 0.161 | 0.323 | 0.323 | 0.258 | 0.452 |
| Oceania (7) | 0.143 | 0.286 | 0.286 | 0.286 | 0.286 | 0.286 | 0.143 | 0.143 | 0.143 | 0.143 | 0.429 | 0.286 |
| America (6) | 0.000 | 0.333 | 0.500 | 0.333 | 0.333 | 0.333 | 0.000 | 0.333 | 0.000 | 0.000 | 0.333 | 0.167 |
| Total (166) | 0.267 | 0.121 | 0.424 | 0.109 | 0.133 | 0.152 | 0.333 | 0.230 | 0.248 | 0.339 | 0.152 | 0.218 |
Sequence variants detected in the control population
In 95 control individuals, a total of 71 different sequence variants were detected (Table 1). Forty six were found in the coding region of BRCA2: 30 missense mutations, one nonsense mutation 10 204 A->T, resulting in a stop at codon 3326, one deletion of 2 bp (CT) with a stop at codon 3364, one deletion of a complete codon (AAA) that results in the loss of leucine, and 14 silent mutations were identified. Of the 30 missense mutations, 17 resulted in non-conservative amino acid changes (Table 1). The non-conservative missense mutation H2074N is located in the eigthth BRC repeat and was detected once in an African control individual. Twenty four changes were identified in the non-coding region, including one deletion of 4 bp, one deletion of 1 bp (T), and two in the 5[prime]-untranslated region (UTR) in exon 2.
Frequency and global distribution of continent- and population-specific sequence variants
Twelve sequence variants were detected on at least three continents. The frequency of heterozygotes for these sequence variants ranged from 10.9 to 42.4%. Table 3 shows a comparison of observed heterozygosities for Africa, Europe, Asia, Oceania, America and both affected and unaffected Austrians. A general deficit of heterozygosity, that attains statistical significance (P [ap] 0.025) in nucleotide positions 1093 and 8983-66, respectively, has been observed in HBC/HBOC cases. This deficit of heterozygosity is most probably due to large genomic deletions that are missed by PCR-based mutation screening methods such as DHPLC, using genomic DNA as template. Sixty three sequence variants (80%) were specific for a continent (55 in both the control population and HBC/HBOC families, eight in HBC/HBOC families only). Fifty one sequence variants (65%) were detected only in specific ethnic groups, of which 47 occurred only once. On a regional level, 33 (42%) were detected only in the African control individuals who contributed 13% of the 332 chromosomes. Twelve sequence variants were specific to the Austrian population (eight were detected only in HBC/HBOC families), nine to the Asian, four to the American Indian, two to the Indian/Pakistani, and one each to the Australian and Oceanian populations. One sequence variant was detected only in the Austrian, African and American Indian populations, one only in the Asian, European and Australian populations, one only in the African, Asian and Oceanian populations, one only in the Austrian and Sardinian populations, one only in the Austrian (n = 7) and Australian (n = 1) populations, and one only in the European (n = 5) and Asian (n = 1, Crimean Tartar) populations. Only the latter two (nucleotides 5427 and 5972) were associated with methylatable CpG/CpNpG motifs. Since such motifs are thought to be mutational hotspots, recurrence of those two polymorphisms cannot be excluded.
The fraction of segregating sites for the coding region (1/194) is roughly half that of the non-coding region (1/108). Given the number of segregating sites, heterozygosity, or genetic diversity, is relatively low. As noted elsewhere, 59% of the sequence variants are only found in one out of 332 chromosomes. Our estimates of nucleotide diversity ([thetas]), the number of differences per site between any two randomly chosen sequences, are 22 ± 6.5 × 10-5 for the coding region and 4.4 ± 1.8 × 10-4 for the non-coding region. These estimates are based on the assumption that the human groups under consideration constitute, as a whole, a random mating population. To the extent that the human species is subdivided, these values are likely to be underestimates.
DISCUSSION
A disease-causing BRCA2 mutation was identified in 8% of the 71 HBC/HBOC families analyzed. Similar results were determined in 100 Finnish HBC/HBOC families (8% BRCA2 mutations) and in 106 Scandinavian families (11% BRCA2 mutations) (12,20). Though Schubert et al. identified 27% BRCA2 mutations in US families, it is important to note that these were families with at least four cases of BC or OC (13).
All 71 Austrian families had been analyzed previously for BRCA1 mutations (22, unpublished data) and six different BRCA1 mutations were detected. Twelve (17%) out of those 71 families were found to carry a disease-associated mutation in either BRCA gene. The probability of detecting such a mutation was clearly related to the family history, with mutations having been identified in three out of four HBOC families, and two out of three families with male BC. Among the 61 HBC families, 37 and 30% of which presented with two and three cases of BC, respectively, only seven (12%) carried a mutation in either BRCA gene. The latter observation is in accordance with a study of 980 HBC families by T. Bishop (11th Breast Cancer Linkage Consortium Meeting, Pisa, 1997), who was able to show that among the group of families with two cases of BC prior to the age of 50 years, the percentage of both BRCA1 and BRCA2 mutations amounts to ~8% each. In addition, three of the six families with BRCA2 mutations showed one case of OC (one within and two outside the OC cluster region) (23).
Two sequence variants that lead to a stop codon were identified in exon 27. The nonsense mutation 10204 A->T was identified previously with a frequency of 0.02% in a US control population and was not associated with an increase of susceptibility to BC and OC (21). The other variant was a not previously reported CT deletion with a stop at codon 3364 detected once in an American-Indian individual. As this deletion is even closer to the C-terminal end than the 10204 A->T change, it is unlikely to be implicated in disease, but this has not been tested.
Eight sequence variants were detected only in HBC/HBOC families. None were located inside the eight BRC repeats (24,25). Seven could be of functional relevance: four resulted in non-conservative amino acid changes, one in a conservative change and two were silent mutations. The non-conservative missense mutation K327E (nucleotide 1207) is locatedin one of the RAD51-binding domains (residues 982-1066 and 1139-1266) and might therefore interfere with the BRCA2-RAD51 interaction (8). The silent mutation S1733S was detected in four HBC/HBOC families (M24, M28, M37 and M47). In M47, the protein truncating mutation 6633del5 was identified. It seems unlikely, therefore, that S1733S is of clinical relevance. However, recently two examples of silent mutations inducing exon skipping in the fibrillin-1 gene and the calpain gene have been reported (26,27). These findings imply that all sequence variants specific to HBC/HBOC might, at the very least, be of potential, functional relevance, but further tests regarding their effect on RNA stability or splicing are required. One non-conservative missense mutation H2074N, that was identified once in an African control individual and was located in the eighth BRC repeat in exon 11, might also be of functional relevance: the BRC 1, 2, 3, 4, 7 and 8 motifs are highly conserved between species (human, monkey, rat and mouse) (17,28), and Chen et al. (7) demonstrated that BRC repeats are critical for RAD51 binding. Therefore, changes in these motifs might interfere with the binding of RAD51 and lead to a loss of function.
Of particular interest is the observation of a deficit of heterozygosity in the HBC/HBOC cases compared with a group of controls from the same population with no history of BC/OC. This deficit is most probably caused by one or more large genomic deletion that is missed by PCR-based mutation screening such as direct sequencing, DHPLC and the protein truncation test, which was performed on exons 10 and 11 (data not shown), using genomic DNA as a template. No such deficit of heterozygosity had been observed in the same individuals on BRCA1 (unpublished data), and Southern blot analysis did not yield any evidence for the existence of large BRCA1 genomic deletions in Austrian BC/OC patients. However, BRCA1 genomic deletions have been reported to be major founder mutations in Dutch BC patients (29). Southern blot analysis to be carried out on the Austrian HBC/HBOC cases will eventually prove whether large genomic deletions account for the observed deficit of heterozygosity in BRCA2 and are major founder mutations.
The analysis of control individuals from a wide variety of populations, for the purpose of identifying polymorphisms, has particular significance: among the 95 control individuals included in this study, a total of 71 different sequence variants were detected, of which 63 (80%) were specific for a continent and 51 (65%) were detected only in specific populations. The data regarding the prevalence of certain sequence variants in specific populations have a direct bearing on the prevention of inaccurate classification. The control populations commonly consist of unaffected individuals of the same ethnic background as the families. Consequently, a sequence variation with an extremely rare occurrence among the Caucasian population, though more frequent in, for example, the African population, will not turn up in a Caucasian control population either. Vehmanen et al. detected the conservative missense mutation I3412V in exon 27 only in two families but not in the control population (20). In this study, this missense mutation was detected in one out of 89 Austrians, two out of 21 Africans and one out of six American Indians, and can therefore be classified as a polymorphism. Takahashi et al. detected the conservative missense mutation A2466V only once in a patient with OC out of 260 chromosomes (18). In our population sample, it was detected in three out of 21 African control individuals.
In evaluating the sequence variants found in our control populations, those variants occurring with a frequency of >1% in both global and specific populations may be classified reliably as polymorphisms. However, even where only a single occurrence of a sequence variant is detected in the control individuals (frequency 0.6%), this may be considered a strong indication of polymorphism. Additional research into control populations is needed to clarify this issue.
Finally, it is interesting to note that our estimate of nucleotide diversity ([thetas]) of 4.4 × 10-4 for the non-coding region of BRCA2 is very similar to that reported for 16 725 random human sequence-tagged sites (a total of 1 981 030 bp screened), which was estimated to be 4.58 × 10-4 (30). The latter estimate was based on 2748 candidate single nucleotide polymorphisms detected in 14 European chromosomes. Given the specificity of oligonucleotide array-based sequencing of 85% (30), their estimate can be considered an overestimate. On the other hand, this may be compensated by the fact that a more ethnically diverse sample is expected to yield a somewhat greater value of nucleotide diversity. A preliminary survey of 31 individuals representing all major racial groups yielded a value of [thetas] that was ~30% larger (30). In contrast, a recent survey of a 9.7 kb region of the human lipoprotein lipase (LPL) gene, that included 998 bp of coding sequence, in 71 individuals of African-American, Finnish and European-American descent reported an estimate of 5.0 ± 5.0 × 10-4 for the coding and 2.1 ± 1.0 × 10-3 for the non-coding region of LPL (31). Of the 88 variable sites found, 87 of which were diallelic, 78 had relative allele frequencies >1%, and more than half of the variant sites exceeded a heterozygosity of 10%. In the present study, only 12 out of 79 variable sites (15%) had a heterozygosity >10%, and 46 (58%) of the variable sites were observed only once. Although the nucleotide diversity estimates for LPL are significantly higher than those for BRCA2, the relative numbers of varying sites in coding (1 in every 142 bp) and non-coding (1 in every 108 bp) regions are very similar to those reported here, namely 1 in every 194 bp for the coding and 1 in every 108 bp for the non-coding regions of BRCA2.
MATERIALS AND METHODS
Breast and breast-ovarian cancer families
All 71 families with BC and/or OC were identified through the genetic counseling program of the Division of Senology, University of Vienna. Selection criteria were: (i) two relatives once removed affected with either BC diagnosed before the age of 50 years, or OC diagnosed at any age; (ii) three or more affected relatives, once or twice removed, at least two of whom hadBC diagnosed before the age of 60 years; (iii) one case of BC diagnosed before the age of 35 years; or (iv) male and femaleBC occurring in the same lineage at any age.
Four families were HBOC families, three families presented with two OC cases (HOC), 61 were HBC families (including four single cases of BC before the age of 30 years), two families with male and female BC, and one family presented with male BC, stomach cancer and one female with BC and OC. All patients were from Austria (Central European origin) and none of the families showed evidence of being related when traced back a minimum of three generations. From all available family members, blood samples and, where feasible, tumor tissue samples were collected. Genomic DNA isolated from lymphocytes was analyzed from 71 family members (43% of 166 screened individuals). One affected individual per family was chosen for screening. Preferably, this was the family member where diagnosis had occurred at the earliest age. All subjects gave their written informed consent regarding their participation in the study.
Control individuals
Ninety five control individuals were screened for sequence variants in the BRCA2 gene. The populations surveyed were as follows. Africa (n = 21): five Biaka (Central African Republic Pygmy), four Mbuti Zaire Pygmy, three Lisongo, two Omega San, four Ethiopian (one Birale, one Hamar, one Berta and one Surma), one Bozo, one Dogon Mali/West Africa and one Tuareg/North Africa; Asia (n = 22): four Japanese, five Han Chinese, four Cambodian, two Taiwanese (Atayal and Ami), two Arab, one Uzbek, one Crimean Tartar, one Turkmen, one Iranian and one Druze; India and Pakistan (n = 9): two Brushaski, one Brauhi, one Balochi, three Sindhi and two Pathan; Pacific (n = 7): three Melanesian, two New Guinean and two Australian; America (n = 6): two Karitiana, one Surui, two Mayan and one Colombian; Europe (n = 30): 18 Austrian, two Sardinian, two Italian, four Basque and four American of European descent.
Control individuals from Austria were women aged 60 years or older with no personal or familial history of cancer. All Austrian control individuals were informed about this research project and signed consent forms. East and West African samples were provided by M. Seielstad (Harvard University, MA) and Central Asian samples by R.S. Wells (Stanford University, CA). All other DNA samples of control individuals were from the Stanford DNA collection (L. Cavalli-Sforza). Samples representing the five continents were collected according to approved human subject protocols.
DNA extraction and PCR conditions
Genomic DNA was isolated from blood samples as described elsewhere (32). PCR was performed using 48 primer pairs in a 50 µl volume containing 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5-2.5 mM MgCl2, 50 µM dNTPs, 0.25 µM of each primer and 100 ng of genomic DNA. For all PCRs, AmpliTaq Gold (Perkin Elmer, Foster City, CA) was used. The intron sequences included upstream of the 3[prime] splice site a minimum of 14 nucleotides, average 74 nucleotides, and downstream of the 5[prime] splice site a minimum of two nucleotides, average 75 nucleotides. The PCR cycling regime comprised an initial denaturation step at 95°C for 10 min to activate AmpliTaq Gold. Subsequent denaturing steps were 94°C for 20 s and extension steps of 72°C for 45 s; annealing temperatures typically were decreased from 63°C by 0.5°C increments with each of the initial 14 cycles, followed by 20 cycles at 56°C for 20 s. Sometimes different annealing temperatures were used in spanning the 7°C touchdown temperature window (see Table 4 for detailed information on primer sequences, magnesium chloride concentrations and annealing temperatures).
Table 4.
| Exon | Forward (F) and reverse (R) primers (5[prime]->3[prime]) | MgCl2 (mM) | Annealing temperature (°C) |
Size (bp) | DHPLC temperature (°C) |
| 2 | F: CCA GGA GAT GGG ACT GAA TTA G | 1.5 | 69-62/62 | 311 | 56 |
| R: CTG TGA CGT ACT GGG TTT TTA GC | |||||
| 3 | F: TTC CTT ATG ATC TTT AAC TGT TCT G | 2.5 | 63-56/56 | 406 | 56 |
| R: GCT AAG ATT TTA ACA CAG GTT TGC | |||||
| 4 | F: AGA ATG CAA ATT TAT AAT CCA GAG TA | 2.5 | 63-56/56 | 249 | 51, 55 |
| R: AAT CAG ATT CAT CTT TAT AGA ACA AA | |||||
| 5+6 | F: TTC CAA CAA TTT ATA TGA ATG AGA ATC | 2.5 | 63-56/56 | 362 | 54 |
| R: CTC AGG GCA AAG GTA TAA CGC | |||||
| 7 | F: CCT TAA TGA TCA GGG CAT TTC | 2.5 | 63-56/56 | 214 | 57 |
| R: CAA CCT CAT CTG CTC TTT CTT G | |||||
| 8 | F: GTA GAT GTG CTT TTT GAT GTC TGA C | 2.5 | 63-56/56 | 315 | 52 |
| R: GAG AGA CAG CAG AGT TTC ACA GG | |||||
| 9 | F: CAG ATA ACT GAA ATC ACC AAA AGT G | 2.5 | 63-56/56 | 262 | 54 |
| R: ACA ACA ACA AAA AAA CCT GTA GTT C | |||||
| 10A | F: TAT AAA ATA TTA ATG TGC TTC TGT T | 2.5 | 63-56/56 | 374 | 54 |
| R: AAA GGG CTT CTG ATT TGC TAC | |||||
| 10B | F: ATC TGA AGT GGA ACC AAA TGA TAC | 2.5 | 63-56/56 | 280 | 56 |
| R: ACG TGG CAA AGA ATT CTC TGA AGT AA | |||||
| 10C | F: TTT CAG AAA AAG ACC TAT TAG ACA | 2.5 | 63-56/56 | 242 | 55 |
| R: CTT TTT GAT ACC CTG AAA TGA AGA AG | |||||
| 10D | F: TAA AGC AGG CAA TAT CTG GAA CTT CT | 2 | 65-58/58 | 295 | 56 |
| R: GTG GAT ATT AAA CCT GCA TTC TTC AA | |||||
| 10E | F: TTT AAT TGA TAA TGG AAG CTG G | 2.5 | 63-56/56 | 268 | 55 |
| R: TTA CAA AAA AAA AAA GAC AGA GGT | |||||
| 11A | F: TTG TCA GAT TTA ACT TTT TTG GAA G | 2.5 | 63-56/56 | 342 | 55 |
| R: CAA CTG GGA CAC TTT CTT TCA G | |||||
| 11B | F: GCT CAA GAA GCA TGT CAT GG | 2.5 | 63-56/56 | 394 | 55 |
| R: TAT GAA AAC CCA ACA GAG TAG GT | |||||
| 11C | F: GAA AGA AAG TGT CCC AGT TG | 2.5 | 63-56/56 | 360 | 58 |
| R: ACC ACA GTC TCA ATA GAA ACA AGG | |||||
| 11D | F: TGA GAC CAT TGA GAT CAC AGC | 2.5 | 63-56/56 | 709 | 53 |
| R: TAG TCA CAA GTT CCT CAA CGC A | |||||
| 11E | F: TGA TTG ATG GTA CTT TAA TTT TGT CAC | 2.5 | 63-56/56 | 338 | 54 |
| R: AGC CAA GAC CTC TTC TTT TAT ATC TG | |||||
| 11F | F: AAG CTG ATT CTC TGT CAT GCC TG | 2.5 | 63-56/56 | 456 | 55 |
| R: GAT TTG TGT TTT GGT TGA ATT GTA CC | |||||
| 11G | F: AAA ATA CAT GAG AGT AGC ATC ACC | 2.5 | 63-56/56 | 330 | 55 |
| R: AAA TCT TTT TTA ATT GAC ACT TGG | |||||
| 11H | F: CGA ACC CAT TTT CAA GAA CTC TAC CA | 2.5 | 63-56/56 | 215 | 54 |
| R: TGT AAT CAT TAT TTT TTT CTG G | |||||
| 11I | F: TTG GTT TAT GTT CTT GCA GAG GAG | 2.5 | 63-56/56 | 487 | 54 |
| R: CCT TTT GGC TAG GTG TTA AAT TAT GG | |||||
| 11J | F: TGG CAT TAG ATA ATC AAA AGA AAC TG | 2.5 | 63-56/56 | 500 | 55 |
| R: CCT AAA CCC CAC TTC ATT TTC ATC | |||||
| 11K | F: GAA ATT AAA CGG AAG TTT GCT GG | 2.5 | 63-56/56 | 465 | 54 |
| R: TGA ATC ACT GCC ATC AAA TTC TAA G | |||||
| 11L | F: AAT GAC TAC TGG CAC TTT TGT TG | 2.5 | 63-56/56 | 401 | 54 |
| R: CAC TTG CAG TCT GAA AAA ATG TAT C | |||||
| 11M | F: GCC AGT ATT GAA GAA TGT TGA AGA TC | 1.5 | 68-61/61 | 443 | 55 |
| R: AAA CCT TAT GTG AAT GCG TGC TAC | |||||
| 11N | F: AAC GAA AAT TAT GGC AGG TTG TTA C | 1.5 | 68-61/61 | 536 | 55 |
| R: GCT TTC CAC TTG CTG TAC TAA ATC C | |||||
| 11O | F: CCA GCT CAC AAG AGA AGA AAA TAC TG | 1.5 | 68-61/61 | 503 | 54 |
| R: TTA CGT TTT TAG GTG AAG CCT GTT C | |||||
| 11P | F: AAA CCC AGA GCA CTG TGT AAA CTC | 1.5 | 68-61/61 | 487 | 53 |
| R: TCT CCT CTT CTT TTT CCA ATT CTT G | |||||
| 11Q | F: TAC AGA TTC TAA ACT GCC AAG TCA TG | 1.5 | 68-61/61 | 265 | 54 |
| R: TAA CCA TAC TCC CCC AAA CTG AC | |||||
| 12 | F: AAT TGA CAT TGA AGA CTG ACT TTA CTC | 2.5 | 63-56/56 | 370 | 52 |
| R: AGC ACT TTG GAG AGG CAG G | |||||
| 13 | F: GCA TCC GTT ACA TTC ACT GAA A | 2 | 65-58/58 | 310 | 54 |
| R: ACG GGA AGT GTT AAC TTC TTA ACG | |||||
| 14A | F: ACC ATG TAG CAA ATG AGG GTC T | 2.5 | 63-56/56 | 391 | 55 |
| R: GCT TTT GTC TGT TTT CCT CCA A | |||||
| 14B | F: CAC AGA GTT GAA CAG TGT GTT AGG | 1.5 | 68-61/61 | 297 | 55 |
| R: GGG CTT TAA AAT TAC CAC CAC C | |||||
| 15 | F: GGC CAG GGG TTG TGC TTT TT | 2.5 | 63-56/56 | 369 | 52, 55, 59 |
| R: ATT TCA TTC ATC CAT TCC TGC | |||||
| 16 | F: TTT GGT AAA TTC AGT TTT GGT TTG | 2.5 | 63-56/56 | 396 | 55 |
| R: AGC CAA CTT TTT AGT TCG AGA G | |||||
| 17 | F: CAG AGA ATA GTT GTA GTT GTT GAA | 2.5 | 63-56/56 | 306 | 57 |
| R: AGA AAC CTT AAC CCA TAC TGC | |||||
| 18A | F: TCA GTT TTT ATT CTC AGT TAT TCA GTG | 2.5 | 63-56/56 | 298 | 55 |
| R: GCA TAC CAC CCA TCT GTA AGT TC | |||||
| 18B | F: TGT TTC TGA CAT AAT TTC ATT GAG C | 2.5 | 63-56/56 | 420 | 53, 57 |
| R: AAA CTT TAA CTG TCT GAA GAA TAT GC | |||||
| 19 | F: CTT ATT TAC TGT CTT ACT AAT CTT CCT | 2.5 | 63-56/56 | 389 | 52, 56 |
| R: GAC CGA AAC TCC ATC TCA AAC | |||||
| 20 | F: GGT GAT CCA CTA ATC TCA GCC TC | 2.5 | 63-56/56 | 451 | 54, 57 |
| R: GTC CCT TGT TGC TAT TCT TTG TCT | |||||
| 21 | F: GGG TGT TTT ATG CTT GGT TCT | 2.5 | 63-56/56 | 303 | 55, 59 |
| R: CAT TTC AAC ATA TTC CTT CCT G | |||||
| 22 | F: AAC CAC ACC CTT AAG ATG AGC | 2.5 | 63-56/56 | 455 | 53, 56 |
| R: GGG CAT TAG TAG TGG ATT TTG C | |||||
| 23 | F: ACT TCT TCC ATT GCA TCT TTC TCA | 2.5 | 66-59/59 | 290 | 53 |
| R: AAA ACA AAA CAA AAA TTC AAC ATA | |||||
| 24 | F: GCA GCG ACA AAA AAA ACT CA | 2.5 | 63-56/56 | 365 | 53, 56 |
| R: ATT TGC CAA CTG GTA GCT CC | |||||
| 25 | F: GCT TTC GCC AAA TTC AGC TA | 2 | 65-58/58 | 427 | 57 |
| R: TAC CAA AAT GTG TGG TGA TGC | |||||
| 26 | F: GTC CCA AAC TTT TCA TTT CTG C | 2.5 | 63-56/56 | 379 | 54, 58 |
| R: GGA GCC ACA TAA CAA CCA CA | |||||
| 27A | F: CTG TGT GTA ATA TTT GCG TGC T | 2.5 | 63-56/56 | 495 | 54, 58 |
| R: GCA AGT TCT TCG TCA GCT ATT G | |||||
| 27B | F: GAA TTC TCC TCA GAT GAC TCC A | 2.5 | 63-56/56 | 417 | 57 |
| R: TCT TTG CTC ATT GTG CAA CA |
DHPLC
DHPLC was carried out on automated HPLC instrumentation (Rainin Instrument, Woburn, MA) equipped with a DNA separation column (Transgenomic, San Jose, CA) (33-35). Crude PCR products-which were subjected to an additional 3 min, 95 denaturing step followed by gradual reannealing from 95 to 65 over a period of 30 min prior to analysis-were eluted with a linear acetonitrile (J.T. Baker, Phillipsburg, NJ) gradient at a flow rate of 0.9 ml/min. The start and endpoints of the gradient were adjusted according to the size of the PCR products. Generally, analysis took <6 min, including column regeneration and re-equilibration to the starting conditions. The temperature required for successful resolution of heteroduplex molecules was determined by use of the DHPLC melting algorithm that can be obtained at http://hardy-weinberg.stanford.edu/dhplc/melt.html . The temperatures at which PCR products were analyzed are given in the last column of Table 4. All 166 samples were analyzed for all BRCA2 exons and intron-exon boundaries, except for exon 1, which is not translated, by DHPLC.
Direct DNA sequencing
PCR products were purified by solid-phase extraction and bidirectionally sequenced with the Applied Biosystems Taq Dye Deoxy terminator cycle sequencing kit (Perkin Elmer) according to the manufacturers instructions. Samples were analyzed with an Applied Biosystems 373A sequencer.
Cloning of BRCA2 mutations with the TA cloning kit
For the splice site mutation in intron 5, a cDNA PCR product spanning exons 3-7 was generated according to standard protocols. For the splice site mutation in intron 21, nested PCR primers (first pair exons 20-24, second pair exons 21-23) were used. Both cDNA PCR products were ligated into pCRII and then transformed into One Shot (INVF[prime]) competent cells according to the manufacturers instructions (Invitrogen, San Diego, CA). The appropriate white colonies were selected after 24 h and analyzed with an Applied Biosystems 373A sequencer.
ACKNOWLEDGEMENTS
We thank the families that participated in this study. This work was supported by grants of the NIH (R01 HG01707), the European Committee DG-V, the Medizinischer Fond des Bürgermeisters der Stadt Wien, the Wiener Krebshilfe-Krebsgesellschaft, University of Vienna, and the Ludwig-Boltzmann Institut für Klinisch-Experimentelle Onkologie.
ABBREVIATIONS
BC, breast cancer; DHPLC, denaturing high-performance liquid chromatography; FBN1, fibrillin-1; HBC, human breast cancer; HBOC, human breast-ovarian cancer; LPL, lipoprotein lipase; MFS, Marfan syndrome; OC, ovarian cancer.
REFERENCES
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 6 Feb 1999
Copyright©Oxford University Press, 1999.
![]()
CiteULike
Connotea
Del.icio.us What's this?
![]()
![]()

![]()
![]()
![]()
J D Fackenthal, L Sveen, Q Gao, E K Kohlmeir, C Adebamowo, T O Ogundiran, A A Adenipekun, R Oyesegun, O Campbell, C Rotimi, et al.
Complete allelic analysis of BRCA1 and BRCA2 variants in young Nigerian breast cancer patients
J. Med. Genet.,
March 1, 2005;
42(3):
276 - 281.
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
N. M. Suter, R. M. Ray, Y. W. Hu, M. G. Lin, P. Porter, D. L. Gao, R. E. Zaucha, L. M. Iwasaki, L. P. Sabacan, M. C. Langlois, et al.
BRCA1 and BRCA2 Mutations in Women from Shanghai China
Cancer Epidemiol. Biomarkers Prev.,
February 1, 2004;
13(2):
181 - 189.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
G. Sasaki, T. Ogata, T. Ishii, K. Kosaki, S. Sato, K. Homma, T. Takahashi, T. Hasegawa, and N. Matsuo
Micropenis and the 5{alpha}-Reductase-2 (SRD5A2) Gene: Mutation and V89L Polymorphism Analysis in 81 Japanese Patients
J. Clin. Endocrinol. Metab.,
July 1, 2003;
88(7):
3431 - 3436.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
Y. R. Thorstenson, A. Roxas, R. Kroiss, M. A. Jenkins, K. M. Yu, T. Bachrich, D. Muhr, T. L. Wayne, G. Chu, R. W. Davis, et al.
Contributions of ATM Mutations to Familial Breast and Ovarian Cancer
Cancer Res.,
June 15, 2003;
63(12):
3325 - 3333.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
G. Palmieri, G. Palomba, A. Cossu, M. Pisano, M. F. Dedola, M. G. Sarobba, A. Farris, N. Olmeo, A. Contu, A. Pasca, et al.
BRCA1 and BRCA2 germline mutations in Sardinian breast cancer families and their implications for genetic counseling
Ann. Onc.,
December 1, 2002;
13(12):
1899 - 1907.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
D. Trikka, Z. Fang, A. Renwick, S. H. Jones, R. Chakraborty, M. Kimmel, and D. L. Nelson
Complex SNP-Based Haplotypes in Three Human Helicases: Implications for Cancer Association Studies
Genome Res.,
April 1, 2002;
12(4):
627 - 639.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
U Hamann, X Liu, S Lange, H U Ulmer, A Benner, and R J Scott
Contribution of BRCA2 germline mutations to hereditary breast/ovarian cancer in Germany
J. Med. Genet.,
March 1, 2002;
39(3):
e12 - 12.
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
S. Sale, R. Sung, P. Shen, K. Yu, Y. Wang, G. E. Duran, J.-H. Kim, T. Fojo, P. J. Oefner, and B. I. Sikic
Conservation of the Class I {beta}-Tubulin Gene in Human Populations and Lack of Mutations in Lung Cancers and Paclitaxel-resistant Ovarian Cancers
Mol. Cancer Ther.,
January 1, 2002;
1(3):
215 - 225.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
B. J. C. van den Bosch, R. F. M. de Coo, H. R. Scholte, J. G. Nijland, R. van den Bogaard, M. de Visser, C. E. M. de Die-Smulders, and H. J. M. Smeets
Mutation analysis of the entire mitochondrial genome using denaturing high performance liquid chromatography
Nucleic Acids Res.,
October 15, 2000;
28(20):
e89 - e89.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
J. PLASCHKE, T. COMMER, C. JACOBI, H. K SCHACKERT, and J. CHANG-CLAUDE
BRCA2 germline mutations among early onset breast cancer patients unselected for family history of the disease
J. Med. Genet.,
September 1, 2000;
37(9):
17e - 17.
[Full Text]
![]()
This Article ![]()
![]()
Abstract
![]()
FREE Full Text (PDF)
![]()
Alert me when this article is cited
![]()
Alert me if a correction is posted
![]()
Services ![]()
![]()
Email this article to a friend
![]()
Similar articles in this journal
![]()
Similar articles in ISI Web of Science
![]()
Similar articles in PubMed
![]()
Alert me to new issues of the journal
![]()
Add to My Personal Archive
![]()
Download to citation manager
![]()
Search for citing articles in:
ISI Web of Science (62)
![]()
Request Permissions ![]()
Google Scholar ![]()
![]()
Articles by Wagner, T. M.
![]()
Articles by Oefner, P.
![]()
Search for Related Content
![]()
PubMed ![]()
![]()
PubMed Citation
![]()
Articles by Wagner, T. M.
![]()
Articles by Oefner, P.
![]()
Social Bookmarking ![]()
![]()
What's this?