Evaluating the determinants of the spread of COVID-19 between and within rural communities in Mali, West Africa based on blood-fed mosquitoes

National Institute Of Allergy And Infectious Diseases

Investigators

Abstract

The spread of COVID-19 in rural sub-Saharan Africa is poorly understood and reliable data are unlikely to be obtained as a result of low capacity for disease surveillance, compounded by high proportion of young population that present mostly asymptomatic or low-symptomatic infections. To fill this gap and assess how the unique ecological and cultural settings affect it, we have initiated a novel epidemiological study, monitoring COVID-19 infections by sampling blood-fed mosquitoes collected monthly in houses of 20 Malian villages spread across 500km. By May 2022, a total of 48,805 females mosquitoes were collected over 27,900 house visits from 20 villages. We expect at least 30,000 females being blood-fed. Experiments using these and laboratory mosquitoes were carried out in Mali and in NIH to provide a proof of concept to the feasibility of this project as a prerequisite before expansion to the full scope of this work. Immunoglobulin-G antibodies were found to be readily detectable within the mosquito bloodmeals by a bead-based immunoassay through 10 (and up to 30) hours post-feeding, indicating that most mosquitoes collected via indoor aspiration catches (that likely fed the previous night and are now resting), are viable samples for analysis. Using 90 pre-pandemic bloodfed mosquitoes we established antigen specific cutoffs (mean 5*STD) for antibodies against the i) nucleocapsid, ii) the receptor binding domain (RBD), and iii-iv) the two spike domains, to identify mosquitoes that fed on Covid-seropositive and seronegative humans. By requiring at least 2 antibody titers to exceed their cutoffs the estimated sensitivity was 0.90 (+/=0.059) and specificity was 0.92 (+/-0.080), indicating that most blood-fed mosquitoes collected by early morning indoors, have likely fed the previous night, and are viable samples for analysis. Using this assay, we evaluated the change in seroprevalence of one urban and four rural communities in Mali. Consistent with, albeit a little lower than, a conventional serological study in Mali (Sagara et al. 2021), crude sero-positivity of blood sampled via mosquitoes was 6.3% in October/November 2020 over all sites, and increased to 25.1% overall, with the town closest to Bamako (Sotuba) reaching 46.7% in February of 2021. We have adjusted to mosquitoes feeding on non-human hosts, and to the possibility of multiple mosquitoes feeding on the same individual human. In finding this a viable technique amenable to conventional immunoassays, we feel sero-surveillance of human diseases, regardless if they are vector-borne, is attainable, non-invasive sampling option, especially suited to areas with high density of human-biting mosquitoes. A paper summarizing the results of 5 villages at these two time points is near completion (Krajacich et al. 2022: In Prep). Additionally, our team in Mali have carried out ELISA to detect antibodies against the Spike and RBD based on protocols optimized by the group of Patrick Duffy (LMIV) and using antigens they kindly provided. Currently over 1,700 mosquitoes were assayed for both antigens. Based on cutoffs established for these ELISA assays in Mali (Woodford et al. 2021), overall seroprevalence rates increased from 21.9% in October 2020 to 51.2% in January-February 2021 (21=55.4, P<0.001), in accord with sero-prevalence change measured in resident of 3 Malian communities (Sagara et al. 2022). Note that the assortment of villages used in this study was different from the subset used in the four-antigen bead assay (above). Our goal is to complete the assays on 12,000 blood-fed mosquitoes representing multiple time points from October 2020 to September 2022 from 20 rural communities, which will provide the dynamics of spread of COVID-19 in Mali during the first years of the pandemic.