Host feeding preferences of malaria vectors in an area of low malaria transmission

Abstract

Studying the behaviour and trophic preferences of mosquitoes is an important step in understanding the exposure of vertebrate hosts to vector-borne diseases. In the case of human malaria, transmission increases when mosquitoes feed more on humans than on other animals. Therefore, understanding the spatio-temporal dynamics of vectors and their feeding preferences is essential for improving vector control measures. In this study, we investigated the feeding behaviour of Anopheles mosquitoes at two sites in the Sudanian areas of Senegal where transmission is low following the implementation of vector control measures. Blood-fed mosquitoes were collected monthly from July to November 2022 by pyrethrum spray catches in sleeping rooms of almost all houses in Dielmo and Ndiop villages, and blood meals were identified as from human, bovine, ovine, equine and chicken by ELISA. Species from the An. gambiae complex were identified by PCR. The types and numbers of potential domestic animal hosts were recorded in each village. The Human Blood Index (HBI) and the Manly Selection Ratio (MSR) were calculated to determine whether hosts were selected in proportion to their abundance. Spatio-temporal variation in HBI was examined using the Moran's index. A total of 1251 endophilic Anopheles females were collected in 115 bedrooms, including 864 blood fed females of 6 species. An. arabiensis and An. funestus were predominant in Dielmo and Ndiop, respectively. Of the 864 blood meals tested, 853 gave a single host positive result mainly on bovine, equine, human, ovine and chicken in decreasing order in both villages. Overall, these hosts were not selected in proportion to their abundance. The human host was under-selected, highlighting a marked zoophily for the vectors. Over time and space, the HBI were low with no obvious trend, with higher and lower values observed in each of the five months at different points in each village. These results highlight the zoophilic and exophagic behaviour of malaria vectors. This behaviour is likely to be a consequence of the distribution and use of LLINs in both villages and may increase risk of residual outdoor transmission. This underlines the need to study the feeding host profile of outdoor resting populations and how domestic animals may influence malaria epidemiology in order to tailor effective malaria vector control strategies in the two villages.

Figure 1

Geographical representation of the map of Senegal (A), the region of Fatick (B) and the two study villages (C). The Capital city of Senegal and The Gambia are shown.

Figure 2

Variations in the relative abundance of the main hosts in Dielmo and Ndiop. For the five hosts, means with different letters are significantly different.

Figure 3

Spatial and temporal distribution of HBI of malaria vectors in Dielmo. Significance levels are analysed using Getis-Ord Gi* statistics. Z scores < 0 indicate a clustering tendency of low HBI and Z scores > 0 indicate a clustering tendency of high HBI. Significant Z scores (p < 0.05) are shown in red.

Figure 4

Spatial and temporal distribution of HBI of malaria vectors in Ndiop. Significance levels are analysed using Getis-Ord Gi* statistics. Z scores < 0 indicate a clustering tendency of low HBI and Z scores > 0 indicate a clustering tendency of high HBI. Significant Z scores (p < 0.05) are shown in red.

Figure 5

Graphs showing the Manly host selection ratio with 95% confidence interval bars for human, bovine, ovine and equine in Ndiop from July to November 2022. The dashed horizontal line marks the Manly ratio value of 1. The red asterisks indicate Manly ratios significantly greater than 1 (host species over-selection) or less than 1 (under-selection).

Figure 6

Graphs showing the Manly host selection ratio with 95% confidence interval bars for human, bovine, ovine and equine in Dielmo from July to November 2022. The dashed horizontal line marks the Manly ratio value of 1. The red asterisks Manly ratios significantly greater than 1 (host species over-selection) or less than 1 (under-selection).