Spatio-temporal changes in endosymbiont diversity and composition in the African cassava whitefly, Bemisia tabaci SSA1

Abstract

Sap-sucking insects, including whiteflies, are amongst the most devastating and widely distributed organisms on the planet. They are often highly invasive and endosymbiont communities within these insects help them adapt to new or changing environments. Bemisia tabaci (Gennadius; Hemiptera: Aleyrodidae) whitefly species are vectors of more than 500 known plant-viruses and harbour highly diverse endosymbionts communities. To date, however, whitefly-endosymbiont interactions, community structure and their spatio-temporal changes are still poorly understood. In this study, we investigated the spatio-temporal changes in the composition and diversity of bacterial endosymbionts in the agricultural crop pest whitefly species, Bemisia tabaci sub-Saharan Africa 1-subgroup 1 and 2 (SSA1-SG1 and SSA1-SG2). 16S rRNA amplicon sequencing analysis was carried out to characterise endosymbiont compositionsin field-collected SSA1 (SSA1-SG1 and SSA1-SG2) populations infesting cassava in Uganda in 1997 and 2017. We detected Portiera, Arsenophonus, Wolbachia, Hamiltonella and Hemipteriphilus, with Arsenophonus and Wolbachia infections being predominant. Hemipteriphilus and Hamiltonella frequencies were very low and were detected in seven and two samples, respectively. Bacterial diversity based on three independent parameters including Simpson index, number of haplotypes and Bray-Curtis dissimilarity matrix was significantly higher in 1997 than in 2017. This period also coincided with the advent of super-abundant cassava-whitefly populations on cassava crops in Uganda. We discuss how endosymbionts may influence the biology and behaviour of whiteflies leading to population explosions.

Keywords: Illumina HiSeq; cassava; endosymbionts; temporal change; whitefly.

Figure 1

Map showing the distribution of whitefly sampling sites used in this study.

Figure 2

One-way ANOVA of diversity analysis including Shannon index and total haplotype in relation to date (A,B) and site (C,D). Non-metric multi-dimensional scaling analysis (NMDS) test on Bray–Curtis matrix (E) of S-endosymbionts only. Portiera was excluded from the statistical tests.

Figure 3

A stacked barplot of the three bacteria types that were detected in this study in each single whitefly showing their relative abundance in relation to date in 1997 (A) and 2017 (B).

Figure 4

A stacked barplot of S-endosymbiont haplotypes detected in this study in each single whitefly which exhibit significantly different relative abundance in relation to date, as indicated by ANCOM-BC (Supplementary Table S5), in 1997 (A) and 2017 (B).

Figure 5

Phylogenetic analysis of the endosymbionts haplotypes based on the region V4–V5 in 16S rRNA sequences using the neighbour-joining method with 1,000,000 bootstraps in Geneious. The coloured names showed the reference sequences with their accession numbers all the black names were generated from this study.

Figure 6

Heat map of the average of the relative abundance of S-endosymbiont haplotypes detected according to the date of the investigations. The dendrogram on the left was constructed based on the overall relative frequency of each haplotype, meaning that the haplotypes which are clustered together have similar overall relative prevalence.