H3Africa is developing capacity for health-related genomics research in Africa

BackgroundIdentification of pathogens associated with bovine mastitis is helpful in treatment and management decisions. However, such data from sub-Saharan Africa is scarce. Here we describe the distribution and antimicrobial susceptibility patterns of bacteria from cows with clinical mastitis in Kampala, Uganda. Due to high concern of zoonotic infections, isolates from milkmen are also described.Methodology/Principal FindingsNinety seven milk samples from cows with clinical mastitis and 31 nasal swabs from milkmen were collected (one sample per cow/human). Fifty eight (60%) Gram-positive isolates namely Staphylococci (21), Enterococci (16), Streptococci (13), Lactococci (5), Micrococci (2) and Arcanobacteria (1) were detected in cows; only one grew Staphylococcus aureus. Furthermore, 24 (25%) coliforms namely Escherichia coli (12), Klebsiella oxytoca (5), Proteus vulgaris (2), Serratia (2), Citrobacter (1), Cedecea (1) and Leclercia (1) were identified. From humans, 24 Gram-positive bacteria grew, of which 11 were Staphylococci (35%) including four Staphylococcus aureus. Upon susceptibility testing, methicillin-resistant coagulase-negative staphylococci (CoNS) were prevalent; 57%, 12/21 in cows and 64%, 7/11 in humans. However, methicillin-resistant Staphylococcus aureus was not detected. Furthermore, methicillin and vancomycin resistant CoNS were detected in cows (Staphylococcus hominis, Staphylococcus lugdunensis) and humans (Staphylococcus scuiri). Also, vancomycin and daptomycin resistant Enterococci (Enterococcus faecalis and Enterococcus faecium, respectively) were detected in cows. Coliforms were less resistant with three pan-susceptible isolates. However, multidrug resistant Klebsiella, Proteus, Serratia, Cedecea, and Citrobacter were detected. Lastly, similar species grew from human and bovine samples but on genotyping, the isolates were found to be different. Interestingly, human and bovine Staphylococcus aureus were genetically similar (spa-CC435, spa-type t645 corresponding to ST121) but with different susceptibility patterns.Conclusions/SignificanceCoNS, Enterococci, Streptococci, and Escherichia coli are the predominant pathogens associated with clinical bovine-mastitis in Kampala, Uganda. Multidrug resistant bacteria are also prevalent. While similar species occurred in humans and cows, transmission was not detected.

BackgroundAccurate diagnosis of tuberculosis, especially by using rapid molecular assays, can reduce transmission of drug resistant tuberculosis in communities. However, the frequency of resistance conferring mutations varies with geographic location of Mycobacterium tuberculosis, and this affects the efficiency of rapid molecular assays in detecting resistance. This has created need for characterizing drug resistant isolates from different settings to investigate frequencies of resistance conferring mutations. Here, we describe the prevalence and patterns of rifampicin- and isoniazid- resistance conferring mutations in isolates from Uganda, which could be useful in the management of MDR-TB patients in Uganda and other countries in sub-Saharan Africa.ResultsNinety seven M. tuberculosis isolates were characterized, of which 38 were MDR, seven rifampicin-resistant, 12 isoniazid-mono-resistant, and 40 susceptible to rifampicin and isoniazid. Sequence analysis of the rpoB rifampicin-resistance determining region (rpoB/RRDR) revealed mutations in six codons: 588, 531, 526, 516, 513, and 511, of which Ser531Leu was the most frequent (40%, 18/45). Overall, the three mutations (Ser531Leu, His526Tyr, Asp516Tyr) frequently associated with rifampicin-resistance occurred in 76% of the rifampicin resistant isolates while 18% (8/45) of the rifampicin-resistant isolates lacked mutations in rpoB/RRDR. Furthermore, sequence analysis of katG and inhA gene promoter revealed mainly the Ser315Thr (76%, 38/50) and C(-15)T (8%, 4/50) mutations, respectively. These two mutations combined, which are frequently associated with isoniazid-resistance, occurred in 88% of the isoniazid resistant isolates. However, 20% (10/50) of the isoniazid-resistant isolates lacked mutations both in katG and inhA gene promoter. The sensitivity of sequence analysis of rpoB/RRDR for rifampicin-resistance via detection of high confidence mutations (Ser531Leu, His526Tyr, Asp516Tyr) was 81%, while it was 77% for analysis of katG and inhA gene promoter to detect isoniazid-resistance via detection of high confidence mutations (Ser315Thr, C(-15)T, T(-8)C). Furthermore, considering the circulating TB genotypes in Uganda, the isoniazid-resistance conferring mutations were more frequent in M. tuberculosis lineage 4/sub-lineage Uganda, perhaps explaining why this genotype is weakly associated with MDR-TB.ConclusionSequence analysis of rpoB/RRDR, katG and inhA gene promoter is useful in detecting rifampicin/isoniazid resistant M. tuberculosis isolates in Uganda however, about ≤20% of the resistant isolates lack known resistance-conferring mutations hence rapid molecular assays may not detect them as resistant.

Large-scale, population-based genomic studies have provided a context for modern medical genetics. Among such studies, however, African populations have remained relatively underrepresented. The breadth of genetic diversity across the African continent argues for an exploration of local genomic context to facilitate burgeoning disease mapping studies in Africa. We sought to characterize genetic variation and to assess population substructure within a cohort of HIV-positive children from Botswana-a Southern African country that is regionally underrepresented in genomic databases. Using whole-exome sequencing data from 164 Batswana and comparisons with 150 similarly sequenced HIV-positive Ugandan children, we found that 13%-25% of variation observed among Batswana was not captured by public databases. Uncaptured variants were significantly enriched (p = 2.2 × 10) for coding variants with minor allele frequencies between 1% and 5% and included predicted-damaging non-synonymous variants. Among variants found in public databases, corresponding allele frequencies varied widely, with Botswana having significantly higher allele frequencies among rare (<1%) pathogenic and damaging variants. Batswana clustered with other Southern African populations, but distinctly from 1000 Genomes African populations, and had limited evidence for admixture with extra-continental ancestries. We also observed a surprising lack of genetic substructure in Botswana, despite multiple tribal ethnicities and language groups, alongside a higher degree of relatedness than purported founder populations from the 1000 Genomes project. Our observations reveal a complex, but distinct, ancestral history and genomic architecture among Batswana and suggest that disease mapping within similar Southern African populations will require a deeper repository of genetic variation and allelic dependencies than presently exists.

Purpose The Collaborative African Genomics Network (CAfGEN) aims to establish sustainable genomics research programs in Botswana and Uganda through long-term training of Ph.D. students from these countries at Baylor College of Medicine. Here we present an overview of the CAfGEN Ph.D. training program alongside trainees’ perspectives on their involvement. Background Historically, collaborations between High Income Countries (HICs) and Low and Middle Income Countries (LMICs), or North-South collaborations, have been criticized for the lack of a mutually beneficial distribution of resources and research findings, often undermining LMICs. CAfGEN purposes to address this imbalance in the genomics field through a program of technology and expertise transfer to the participating LMICs. Methods An overview of the training program is presented. Trainees from the CAfGEN project summarized their experiences, looking specifically at the training model, benefits of the program, challenges encountered relating to the cultural transition, and program outcomes after the first two years. Conclusion Collaborative training programs like CAfGEN will not only help establish sustainable long-term research initiatives in LMICs, but foster stronger North-South and South-South networks. The CAfGEN model offers a framework for the development of training programs aimed at genomics education for those for whom genomics is not their “first language.”

BackgroundIn the absence of an effective vaccine, malaria treatment and eradication is still a challenge in most endemic areas globally. This is especially the case with the current reported emergence of resistance to artemisinin agents in Southeast Asia. This study therefore explored the prevalence of K13-propeller gene polymorphisms among Plasmodium falciparum parasites in northern Uganda.MethodsAdult patients (≥18 years) presenting to out-patients department of Lira and Gulu regional referral hospitals in northern Uganda were randomly recruited. Laboratory investigation for presence of plasmodium infection among patients was done using Plasmodium falciparum exclusive rapid diagnostic test, histidine rich protein-2 (HRP2) (Pf). Finger prick capillary blood from patients with a positive malaria test was spotted on a filter paper Whatman no. 903. The parasite DNA was extracted using chelex resin method and sequenced for mutations in K13-propeller gene using Sanger sequencing. PCR DNA sequence products were analyzed using in DNAsp 5.10.01software, data was further processed in Excel spreadsheet 2007.ResultsA total of 60 parasite DNA samples were sequenced. Polymorphisms in the K13-propeller gene were detected in four (4) of the 60 parasite DNA samples sequenced. A non-synonymous polymorphism at codon 533 previously detected in Cambodia was found in the parasite DNA samples analyzed. Polymorphisms at codon 522 (non-synonymous) and codon 509 (synonymous) were also found in the samples analyzed. The study found evidence of positive selection in the Plasmodium falciparum population in northern Uganda (Tajima’s D = −1.83205; Fu and Li’s D = −1.82458).ConclusionsPolymorphism in the K13-propeller gene previously reported in Cambodia has been found in the Ugandan Plasmodium falciparum parasites. There is need for continuous surveillance for artemisinin resistance gene markers in the country.Electronic supplementary materialThe online version of this article (doi:10.1186/s12879-016-1777-7) contains supplementary material, which is available to authorized users.

BackgroundRhomboids are ubiquitous proteins with unknown roles in mycobacteria. However, bioinformatics suggested putative roles in DNA replication pathways and metabolite transport. Here, mycobacterial rhomboid-encoding genes were characterized; first, using the Providencia stuartii null-rhomboid mutant and then deleted from Mycobacterium smegmatis for additional insight in mycobacteria.Methodology/Principal FindingsUsing in silico analysis we identified in M. tuberculosis genome the genes encoding two putative rhomboid proteins; Rv0110 (referred to as “rhomboid protease 1”) and Rv1337 (“rhomboid protease 2”). Genes encoding orthologs of these proteins are widely represented in all mycobacterial species. When transformed into P. stuartii null-rhomboid mutant (ΔaarA), genes encoding mycobacterial orthologs of “rhomboid protease 2” fully restored AarA activity (AarA is the rhomboid protein of P. stuartii). However, most genes encoding mycobacterial “rhomboid protease 1” orthologs did not. Furthermore, upon gene deletion in M. smegmatis, the ΔMSMEG_4904 single mutant (which lost the gene encoding MSMEG_4904, orthologous to Rv1337, “rhomboid protease 2”) formed the least biofilms and was also more susceptible to ciprofloxacin and novobiocin, antimicrobials that inhibit DNA gyrase. However, the ΔMSMEG_5036 single mutant (which lost the gene encoding MSMEG_5036, orthologous to Rv0110, “rhomboid protease 1”) was not as susceptible. Surprisingly, the double rhomboid mutant ΔMSMEG_4904–ΔMSMEG_5036 (which lost genes encoding both homologs) was also not as susceptible suggesting compensatory effects following deletion of both rhomboid-encoding genes. Indeed, transforming the double mutant with a plasmid encoding MSMEG_5036 produced phenotypes of the ΔMSMEG_4904 single mutant (i.e. susceptibility to ciprofloxacin and novobiocin).Conclusions/SignificanceMycobacterial rhomboid-encoding genes exhibit differences in complementing aarA whereby it's only genes encoding “rhomboid protease 2” orthologs that fully restore AarA activity. Additionally, gene deletion data suggests inhibition of DNA gyrase by MSMEG_4904; however, the ameliorated effect in the double mutant suggests occurrence of compensatory mechanisms following deletion of genes encoding both rhomboids.

BackgroundWe determined prevalence of pertussis infection and its associated host and environmental factors to generate information that would guide strategies for disease control.MethodsIn a cross-sectional study, 449 children aged 3 months to 12 years with persistent cough lasting ≥14 days were enrolled and evaluated for pertussis using DNA polymerase chain reaction (PCR) and ELISA serology tests.ResultsPertussis prevalence was 67 (15% (95% Confidence Interval (CI): 12–18)) and 81 (20% (95% CI: 16–24)) by PCR and ELISA, respectively among 449 participating children. The prevalence was highest in children with >59 months of age despite high vaccination coverage of 94% in this age group. Study demographic and clinical characteristics were similar between pertussis and non-pertussis cases. Of the 449 children, 133 (30%) had a coughing household member and 316 (70%) did not. Among 133 children that had a coughing household member, sex of child, sharing bed with a coughing household member and having a coughing individual in the neighborhood were factors associated with pertussis. Children that had shared a bed with a coughing household individual had seven-fold likelihood of having pertussis compared to children that did not (odds ratio (OR) 7.16 (95% CI: 1.24–41.44)). Among the 316 children that did not have a coughing household member, age <23 months, having or contact with a coughing individual in neighborhood, a residence with one room, and having a caretaker with >40 years of age were the factors associated with pertussis. Age <23months was three times more likely to be associated with pertussis compared to age 24–59 months (OR 2.97 (95% CI: 1.07–8.28)).ConclusionFindings suggest high prevalence of pertussis among children with persistent cough at a health facility and it was marked in children >59 months of age, suggesting the possibility of waning immunity. The factors associated with pertussis varied by presence or absence of a coughing household member.