Genetic variants in SEC16B are associated with body composition in black South Africans

• Published in: Nutrition & diabetes Vol. 8; no. 1; pp. 43 - 10
• Main Authors: Sahibdeen, Venesa, Crowther, Nigel J., Soodyall, Himla, Hendry, Liesl M., Munthali, Richard J., Hazelhurst, Scott, Choudhury, Ananyo, Norris, Shane A., Ramsay, Michèle, Lombard, Zané
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 19-07-2018; Nature Publishing Group
• Subjects: 45/43; 631/208; 692/699/2743/393; Adolescent; Adult; African Continental Ancestry Group - genetics; Body composition; Body Composition - genetics; Body Mass Index; Clinical Nutrition; Diabetes; DNA-Binding Proteins - genetics; Epidemiology; Female; Gene loci; Humans; Internal Medicine; Male; Medicine; Medicine & Public Health; Metabolic Diseases; Middle Aged; Polymorphism, Single Nucleotide; South Africa; South Africa
• ISSN: 2044-4052; 2044-4052
• DOI: 10.1038/s41387-018-0050-0

Objective
The latest genome-wide association studies of obesity-related traits have identified several genetic loci contributing to body composition (BC). These findings have not been robustly replicated in African populations, therefore, this study aimed to assess whether European BC-associated gene loci played a similar role in a South African black population.
Methods
A replication and fine-mapping study was performed in participants from the Birth to Twenty cohort (
N
 = 1,926) using the Metabochip. Measurements included body mass index (BMI), waist and hip circumference, waist-to-hip ratio (WHR), total fat mass, total lean mass and percentage fat mass (PFM).
Results
SNPs in several gene loci, including
SEC16B (P
adj
 <  9.48 × 10
−7
),
NEGR1 (P
adj
 < 1.64 × 10
−6
),
FTO (P
adj
 < 2.91 × 10
−5
),
TMEM18 (P
adj
 < 2.27 × 10
−5
), and
WARS2
(P
adj
 < 3.25 × 10
−5
) were similarly associated (albeit not at array-wide signficance (
P
 ≤ 6.7 × 10
−7
) with various phenotypes including fat mass, PFM, WHR linked to BC in this African cohort, however the associations were driven by different sentinel SNPs. More importantly, DXA-derived BC measures revealed stronger genetic associations than simple anthropometric measures. Association signals generated in this study were shared by European and African populations, as well as unique to this African cohort. Moreover, sophisticated estimates like DXA measures enabled an enhanced characterisation of genetic associations for BC traits.
Conclusion
Results from this study suggest that in-depth genomic studies in larger African cohorts may reveal novel SNPs for body composition and adiposity, which will provide greater insight into the aetiology of obesity.