Breast Cancer Genetic Characteristics in North Africa: A Genome Wide Haplotype Study in the General Tunisian Population

1. Breast Cancer Genetic Characteristics in North Africa: A Genome Wide
Haplotype Study in the General Tunisian Population
Yosr Hamdi1, Mariem ben Rekaya1, Najah Mighri1, Maroua Boujemaa1, Soumaya Labidi11,2, Jingxuan Shan3, Majdi Nagara1, Lotfi Chouchane3, Sonia Abdelhak1, Lilia Romdhane1,4, Hamouda
Boussen2, on the bahalf of the PEC Consortium2 .
1 Laboratoire de Génomique Biomédicale et Oncogénétique Institut Pasteur de Tunis, Université de Tunis El Manar
2 Service d’Oncologie Medicale Hopital Abdel Rahmen Mami Ariana
3 Department of Genetic Medicine, Weill Cornell Medical College-Qatar, Doha, Qatar
4 Department of Biology, Faculty of Science of Bizerte, Université Tunis Carthage, Tunisia
The vast majority of breast cancer genetic variations reported to date are from populations with European ancestry. Because of different linkage disequilibrium
patterns, important differences in genetic architecture have been observed between ethnicities. Moreover, the major challenge of breast cancer molecular genetics
is to identify causal variants. Thus, association studies and haplotype analysis in different populations can help to overcome this challenge. In this study, we
investigated the genetic architecture of several common variants in the Tunisian population and we compared their characteristics in different ethnic groups. We
characterized rs9911630-BRCA1 as a putative functional SNP that seems to affect the expression levels of BRCA1 and NBR2 genes in a cis-miR-eQTL manner.
We performed a genome-wide haplotype study by genotyping a set of
135 Tunisian individuals from the general population using the Affymetrix
6.0-Array in order to investigate the genetic characteristics of several
common variants known to be associated with breast cancer risk. We
phased haplotypes and constructed LD-blocks. We also performed Principle
Component Analysis based on 79 breast cancer associated variants. Finally,
We assessed the putative function of some selected variants using in silico
prediction tools and by conducting eQTL-assays .
Abstract
Methods
Haplotype analysis showed that Tunisians are more predisposed to
breast cancer variants on 2p24, 4q21, 6q25, 9q31, 10q26, 11q13 and
14q32 loci (Fig.1). We also identified 4 polymorphisms (rs2046210,
rs941764, rs3803662, and rs13329835) that showed significant frequency
differences between Tunisians and 11 populations from European, Asian
and African origins. Moreover, our PCA-plots and LD-blocks showed that
the genetic characteristics of breast cancer common variants in the
Tunisian population are similar to those of European populations (Fig.2).
Results
Population genetics:
Functional Analysis:
We explored the functional role of 79 breast cancer common variants.
rs9911630-BRCA1 get the highly ranked RegulomeDB score (1b).
Interestingly, rs9911630 is also the most strongly expression-associated
variant and the highly significant eQTL evidence was associated with NBR2
gene (p =1.2x10-23). Using the Genevar platform, we assessed eQTL
associations for rs9911630 in 8 populations. Data showed that rs9911630 is
significantly associated with the expression level of BRCA1 gene in Asian
and Caucasian populations but not in Africans (Fig.3). Consequently, we
compared the allelic frequencies of rs9911630 between different
populations. The frequency of rs9911630 is significantly different between
African populations on one side and Europeans, Asian and Tunisian
populations on another side (Table 1). Finally, We performed in silico
predictions of micro RNA binding sites using mirBase. Results showed that
rs9911630 causes a gain of mse-miR-2766 binding site and the loose of
bmo-miR3287 and ssa-miR-19d-5p binding sites.
Conclusion
This Population genetic analysis was usefull to compare the genetic
architecture of several common variants in several ethnic groups and to
identify a putative breast cancer causal variant on the BRCA1 gene in a cis-
miR-eQTL manner. The observed discordance in the genetic background
between populations highlights the necessity to establish a specific
genotype profile for each population. Therefore, caution should be
exercised when applying any genetic risk prediction model based on
tagSNPs outside of the ancestry group in which it was derived.
Figure 1. Distribution of breast cancer at risk
haplotypes in the general Tunisian population.
Figure 3. eQTL analysis of the rs9911630- BRCA1 variant in different populations.
Figure 2. Principle component analysis.
CEU TSI CHB JPT MEX CHD GIH TUN ASW MKK LWK YRI
MAF 0,336 0,392 0,327 0,273 0,25 0,435 0,477 0,426 0,623 0,535 0,728 0,795
Minor
allele
G G G G G G G G A A A A
Table 1. Comparison of rs9911630 frequencies between different populations.
Africans
Asians
Europeans