A machine learning approach to integrating genetic and ecological data in tsetse flies (Glossina pallidipes) for spatially explicit vector control planning.

Vector control is an effective strategy for reducing vector-borne disease transmission, but requires knowledge of vector habitat use and dispersal patterns. Our goal was to improve this knowledge for the tsetse species Glossina pallidipes, a vector of human and animal African trypanosomiasis, which are diseases that pose serious health and socioeconomic burdens across sub-Saharan Africa. We used random forest regression to (i) build and integrate models of G. pallidipes habitat suitability and genetic connectivity across Kenya and northern Tanzania and (ii) provide novel vector control recommendations. Inputs for the models included field survey records from 349 trap locations, genetic data from 11 microsatellite loci from 659 flies and 29 sampling sites, and remotely sensed environmental data. The suitability and connectivity models explained approximately 80% and 67% of the variance in the occurrence and genetic data and exhibited high accuracy based on cross-validation. The bivariate map showed that suitability and connectivity vary independently across the landscape and was used to inform our vector control recommendations. Post hoc analyses show spatial variation in the correlations between the most important environmental predictors from our models and each response variable (e.g., suitability and connectivity) as well as heterogeneity in expected future climatic change of these predictors. The bivariate map suggests that vector control is most likely to be successful in the Lake Victoria Basin and supports the previous recommendation that G. pallidipes from most of eastern Kenya should be managed as a single unit. We further recommend that future monitoring efforts should focus on tracking potential changes in vector presence and dispersal around the Serengeti and the Lake Victoria Basin based on projected local climatic shifts. The strong performance of the spatial models suggests potential for our integrative methodology to be used to understand future impacts of climate change in this and other vector systems.

Developing and quality testing of microsatellite loci for four species of Glossina.

Microsatellite loci still represent valuable resources for the study of the population biology of non-model organisms. Discovering or adapting new suitable microsatellite markers in species of interest still represents a useful task, especially so for non-model organisms as tsetse flies (genus Glossina), which remain a serious threat to the health of humans and animals in sub-Saharan Africa. In this paper, we present the development of new microsatellite loci for four species of Glossina: two from the Morsitans group, G. morsitans morsitans (Gmm) from Zimbabwe, G. pallidipes (Gpalli) from Tanzania; and the other two from the Palpalis group, G. fuscipes fuscipes (Gff) from Chad, and G. palpalis gambiensis (Gpg) from Guinea. We found frequent short allele dominance and null alleles. Stuttering could also be found and amended when possible. Cryptic species seemed to occur frequently in all taxa but Gff. This explains why it may be difficult finding ecumenical primers, which thus need adaptation according to each taxonomic and geographic context. Amplification problems occurred more often in published old markers, and Gmm and Gpg were the most affected (stronger heterozygote deficits). Trinucleotide markers displayed selection signature in some instances (Gmm). Combining old and new loci, for Gmm, eight loci can be safely used (with correction for null alleles); and five seem particularly promising; for Gpalli, only five to three loci worked well, depending on the clade, which means that the use of loci from other species (four morsitans loci seemed to work well), or other new primers will need to be used; for Gff, 14 loci behaved well, but with null alleles, seven of which worked very well; and for G. palpalis sl, only four loci, needing null allele and stuttering corrections seem to work well, and other loci from the literature are thus needed, including X-linked markers, five of which seem to work rather well (in females only), but new markers will probably be needed. Finally, the high proportion of X-linked markers (around 30%) was explained by the non-Y DNA quantity and chromosome structure of tsetse flies studied so far.

Phylogeography and population structure of the tsetse fly Glossina pallidipes in Kenya and the Serengeti ecosystem.

Glossina pallidipes is the main vector of animal African trypanosomiasis and a potential vector of human African trypanosomiasis in eastern Africa where it poses a large economic burden and public health threat. Vector control efforts have succeeded in reducing infection rates, but recent resurgence in tsetse fly population density raises concerns that vector control programs require improved strategic planning over larger geographic and temporal scales. Detailed knowledge of population structure and dispersal patterns can provide the required information to improve planning. To this end, we investigated the phylogeography and population structure of G. pallidipes over a large spatial scale in Kenya and northern Tanzania using 11 microsatellite loci genotyped in 600 individuals. Our results indicate distinct genetic clusters east and west of the Great Rift Valley, and less distinct clustering of the northwest separate from the southwest (Serengeti ecosystem). Estimates of genetic differentiation and first-generation migration indicated high genetic connectivity within genetic clusters even across large geographic distances of more than 300 km in the east, but only occasional migration among clusters. Patterns of connectivity suggest isolation by distance across genetic breaks but not within genetic clusters, and imply a major role for river basins in facilitating gene flow in G. pallidipes. Effective population size (Ne) estimates and results from Approximate Bayesian Computation further support that there has been recent G. pallidipes population size fluctuations in the Serengeti ecosystem and the northwest during the last century, but also suggest that the full extent of differences in genetic diversity and population dynamics between the east and the west was established over evolutionary time periods (tentatively on the order of millions of years). Findings provide further support that the Serengeti ecosystem and northwestern Kenya represent independent tsetse populations. Additionally, we present evidence that three previously recognized populations (the Mbeere-Meru, Central Kenya and Coastal "fly belts") act as a single population and should be considered as a single unit in vector control.

Genetic Differentiation of Glossina pallidipes Tsetse Flies in Southern Kenya.

The tsetse fly Glossina pallidipes, the major vector of the parasite that causes animal African trypanosomiasis in Kenya, has been subject to intense control measures with only limited success. The G. pallidipes population dynamics and dispersal patterns that underlie limited success in vector control campaigns remain unresolved, and knowledge on genetic connectivity can provide insights, and thereby improve control and monitoring efforts. We therefore investigated the population structure and estimated migration and demographic parameters in G. pallidipes using genotypic data from 11 microsatellite loci scored in 250 tsetse flies collected from eight localities in Kenya. Clustering analysis identified two genetically distinct eastern and western clusters (mean between-cluster F ST = 0.202) separated by the Great Rift Valley. We also found evidence of admixture and migration between the eastern and western clusters, isolation by distance, and a widespread signal of inbreeding. We detected differences in population dynamics and dispersal patterns between the western and eastern clusters. These included lower genetic diversity (allelic richness; 7.48 versus 10.99), higher relatedness (percent related individuals; 21.4% versus 9.1%), and greater genetic differentiation (mean within-cluster F ST; 0.183 versus 0.018) in the western than the eastern cluster. Findings are consistent with the presence of smaller, less well-connected populations in Western relative to eastern Kenya. These data suggest that recent anthropogenic influences such as land use changes and vector control programs have influenced population dynamics in G. pallidipes in Kenya, and that vector control efforts should include some region-specific strategies to effectively control this disease vector.

Temporal genetic differentiation in Glossina pallidipes tsetse fly populations in Kenya.

BACKGROUND: Glossina pallidipes is a major vector of both Human and Animal African Trypanosomiasis (HAT and AAT) in Kenya. The disease imposes economic burden on endemic regions in Kenya, including south-western Kenya, which has undergone intense but unsuccessful tsetse fly control measures. We genotyped 387 G. pallidipes flies at 13 microsatellite markers to evaluate levels of temporal genetic variation in two regions that have been subjected to intensive eradication campaigns from the 1960s to the 1980s. One of the regions, Nguruman Escarpment, has been subject to habitat alteration due to human activities, while the other, Ruma National Park, has not. In addition, Nguruman Escarpment is impacted by the movement of grazing animals into the area from neighboring regions during the drought season. We collected our samples from three geographically close sampling sites for each of the two regions. Samples were collected between the years 2003 and 2015, spanning ~96 tsetse fly generations. RESULTS: We established that allelic richness averaged 3.49 and 3.63, and temporal Ne estimates averaged 594 in Nguruman Escarpment and 1120 in Ruma National Park. This suggests that genetic diversity is similar to what was found in previous studies of G. pallidipes in Uganda and Kenya, implying that we could not detect a reduction in genetic diversity following the extensive control efforts during the 1960s to the 1980s. However, we did find differences in temporal patterns of genetic variation between the two regions, indicated by clustering analysis, pairwise FST, and Fisher's exact tests for changes in allele and genotype frequencies. In Nguruman Escarpment, findings indicated differentiation among samples collected in different years, and evidence of local genetic bottlenecks in two locations previous to 2003, and between 2009 and 2015. In contrast, there was no consistent evidence of differentiation among samples collected in different years, and no evidence of local genetic bottlenecks in Ruma National Park. CONCLUSION: Our findings suggest that, despite extensive control measures especially between the 1960s and the 1980s, tsetse flies in these regions persist with levels of genetic diversity similar to that found in populations that did not experience extensive control measures. Our findings also indicate temporal genetic differentiation in Nguruman Escarpment detected at a scale of > 80 generations, and no similar temporal differentiation in Ruma National Park. The different level of temporal differentiation between the two regions indicates that genetic drift is stronger in Nugruman Escarpment, for as-yet unknown reasons, which may include differences in land management. This suggests land management may have an impact on G. pallidipes population genetics, and reinforces the importance of long term monitoring of vector populations in estimates of parameters needed to model and plan effective species-specific control measures.

Haplotype of non-synonymous mutations within IL-23R is associated with susceptibility to severe malaria anemia in a P. falciparum holoendemic transmission area of Kenya.

BACKGROUND: Improved understanding of the molecular mechanisms involved in pediatric severe malarial anemia (SMA) pathogenesis is a crucial step in the design of novel therapeutics. Identification of host genetic susceptibility factors in immune regulatory genes offers an important tool for deciphering malaria pathogenesis. The IL-23/IL-17 immune pathway is important for both immunity and erythropoiesis via its effects through IL-23 receptors (IL-23R). However, the impact of IL-23R variants on SMA has not been fully elucidated. METHODS: Since variation within the coding region of IL-23R may influence the pathogenesis of SMA, the association between IL-23R rs1884444 (G/T), rs7530511 (C/T), and SMA (Hb < 6.0 g/dL) was examined in children (n = 369, aged 6-36 months) with P. falciparum malaria in a holoendemic P. falciparum transmission area. RESULTS: Logistic regression analysis, controlling for confounding factor of anemia, revealed that individual genotypes of IL-23R rs1884444 (G/T) [GT; OR = 1.34, 95% CI = 0.78-2.31, P = 0.304 and TT; OR = 2.02, 95% CI = 0.53-7.74, P = 0.286] and IL-23R rs7530511 (C/T) [CT; OR = 2.6, 95% CI = 0.59-11.86, P = 0.202 and TT; OR = 1.66, 95% CI = 0.84-3.27, P = 0.142] were not associated with susceptibility to SMA. However, carriage of IL-23R rs1884444T/rs7530511T (TT) haplotype, consisting of both mutant alleles, was associated with increased susceptibility to SMA (OR = 1.12, 95% CI = 1.07-4.19, P = 0.030). CONCLUSION: Results presented here demonstrate that a haplotype of non-synonymous IL-23R variants increase susceptibility to SMA in children of a holoendemic P. falciparum transmission area.

Association between Fcγ receptor IIA, IIIA and IIIB genetic polymorphisms and susceptibility to severe malaria anemia in children in western Kenya.

BACKGROUND: Naturally-acquired immunity to Plasmodium falciparum malaria develops after several episodes of infection. Fc gamma receptors (FcγRs) bind to immunoglobulin G (IgG) antibodies and mediate phagocytosis of opsonized microbes, thereby, linking humoral and cellular immunity. FcγR polymorphisms influence binding affinity to IgGs and consequently, can influence clinical malaria outcomes. Specifically, variations in FcγRIIA -131Arg/His, FcγRIIIA-176F/V and FcγRIIIB-NA1/NA2 modulate immune responses through altered binding preferences to IgGs and immune complexes. Differential binding, in turn, changes ability of immune cells to respond to infection through production of inflammatory mediators during P. falciparum infection. METHODS: We determined the association between haplotypes of FcγRIIA-131Arg/His, FcγRIIIA-176F/V and FcγRIIIB-NA1/NA2 variants and severe malarial anemia (SMA; hemoglobin < 6.0 g/dL, any density parasitemia) in children (n = 274; aged 6-36 months) presenting for their first hospital visit with P. falciparum malaria in a holoendemic transmission region of western Kenya. FcγRIIA-131Arg/His and FcγRIIIA-176F/V genotypes were determined using TaqMan® SNP genotyping, while FcγRIIIBNA1/NA2 genotypes were determined using restriction fragment length polymorphism. Hematological and parasitological indices were measured in all study participants. RESULTS: Carriage of FcγRIIA-131Arg/FcγRIIIA-176F/FcγRIIIBNA2 haplotype was associated with susceptibility to SMA (OR = 1.70; 95% CI; 1.02-2.93; P = 0.036), while the FcγRIIA-131His/ FcγRIIIA-176F/ FcγRIIIB NA1 haplotype was marginally associated with enhanced susceptibility to SMA (OR: 1.80, 95% CI; 0.98-3.30, P = 0.057) and higher levels of parasitemia (P = 0.009). Individual genotypes of FcγRIIA-131Arg/His, FcγRIIIA-176F/V and FcγRIIIB-NA1/NA2 were not associated with susceptibility to SMA. CONCLUSION: The study revealed that haplotypes of FcγRs are important in conditioning susceptibility to SMA in immune-naive children from P. falciparum holoendemic region of western Kenya.

Interleukin (IL)-13 promoter polymorphisms (-7402 T/G and -4729G/A) condition susceptibility to pediatric severe malarial anemia but not circulating IL-13 levels.

In holoendemic Plasmodium falciparum transmission areas such as western Kenya, severe malarial anemia [SMA, hemoglobin (Hb) < 6.0 g/dL, with any density parasitemia] is the most common clinical manifestation of severe malaria resulting in high rates of pediatric morbidity and mortality in these regions. Previous studies associated interleukin (IL)-13 with pathogenesis of different infectious diseases, including P. falciparum malaria. However, the functional roles of polymorphic variants within the IL-13 promoter in conditioning susceptibility to SMA remain largely unexplored. As such, the association between the IL-13 variants -7402 T/G (rs7719175) and -4729G/A (rs3091307) and susceptibility to SMA was determined in children (n = 387) presenting with clinical symptoms of falciparum malaria and resident in a holoendemic transmission region in western Kenya. Our results indicated no difference in the proportions of individual genotypes among children presenting with non-SMA (n = 222) versus SMA (n = 165). Similarly, there was no associations between the individual genotypes (-7402 T/G and -4729G/A) and SMA. Additional analyses, however, revealed that proportions of individuals with -7402 T/-4729A (TA) haplotype was significantly higher in children presenting with SMA than non-SMA group (P = 0.043). A further multivariate logistic regression analyses, controlling for confounding factors, demonstrated that carriage of the TA haplotype was associated with increased susceptibility to SMA (OR; 1.564, 95% CI; 1.023-2.389, P = 0.039). In addition, circulating levels of IL-13 were comparable between the clinical groups as well as across genotypes and haplotypes. Collectively, findings presented here suggest that haplotypes within the IL-13 promoter at -7402 T/G and -4729G/A may modulate SMA pathogenesis, but do not affect circulating IL-13 levels.

Polymorphisms in the Fc gamma receptor IIIA and Toll-like receptor 9 are associated with protection against severe malarial anemia and changes in circulating gamma interferon levels.

An understanding of the immunogenetic basis of naturally acquired immunity to Plasmodium falciparum infection would aid in the designing of a rationally based malaria vaccine. Variants within the Fc gamma receptors (FcγRs) mediate immunity through engagement of immunoglobulin G and other immune mediators, such as gamma interferon (IFN-γ), resulting in erythrophagocytosis and production of inflammatory cytokines in severe malarial anemia (SMA). The Toll-like receptors (TLRs) trigger transcription of proinflammatory cytokines and induce adaptive immune responses. Therefore, these receptors may condition malaria disease pathogenesis through alteration in adaptive and innate immune responses. To further delineate the impacts of FcγRIIIA and TLR9 in SMA pathogenesis, the associations between FcγRIIIA -176F/V and TLR9 -1237T/C variants, SMA (hemoglobin [Hb] < 6.0 g/dl), and circulating IFN-γ levels were investigated in children (n = 301) from western Kenya with acute malaria. Multivariate logistic regression analysis (controlling for potential confounders) revealed that children with the FcγRIIIA -176V/TLR9 -1237C (VC) variant combination had 64% reduced odds of developing SMA (odds ratio [OR], 0.36; 95% confidence interval [CI], 0.20 to 0.64; P = 0.001), while carriers of the FcγRIIIA -176V/TLR9 -1237T (VT) variant combination were twice as susceptible to SMA (OR, 2.04; 95% CI, 1.19 to 3.50; P = 0.009). Children with SMA had higher circulating IFN-γ levels than non-SMA children (P = 0.008). Hemoglobin levels were negatively correlated with IFN-γ levels (r = -0.207, P = 0.022). Consistently, the FcγRIIIA -176V/TLR9 -1237T (VT) carriers had higher levels of circulating IFN-γ (P = 0.011) relative to noncarriers, supporting the observation that higher IFN-γ levels are associated with SMA. These results demonstrate that FcγRIIIA-176F/V and TLR9 -1237T/C variants condition susceptibility to SMA and functional changes in circulating IFN-γ levels.