A minisatellite-based MLVA for deciphering the global epidemiology of the bacterial cassava pathogen Xanthomonas phaseoli pv. manihotis

Abstract

Cassava Bacterial Blight (CBB) is a destructive disease widely distributed in the different areas where this crop is grown. Populations studies have been performed at local and national scales revealing a geographical genetic structure with temporal variations. A global epidemiology analysis of its causal agent Xanthomonas phaseoli pv. manihotis (Xpm) is needed to better understand the expansion of the disease for improving the monitoring of CBB. We targeted new tandem repeat (TR) loci with large repeat units, i.e. minisatellites, that we multiplexed in a scheme of Multi-Locus Variable number of TR Analysis (MLVA-8). This genotyping scheme separated 31 multilocus haplotypes in three clusters of single-locus variants and a singleton within a worldwide collection of 93 Xpm strains isolated over a period of fifty years. The major MLVA-8 cluster 1 grouped strains originating from all countries, except the unique Chinese strain. On the contrary, all the Xpm strains genotyped using the previously developed MLVA-14 microsatellite scheme were separated as unique haplotypes. We further propose an MLVA-12 scheme which takes advantage of combining TR loci with different mutation rates: the eight minisatellites and four faster evolving microsatellite markers, for global epidemiological surveillance. This MLVA-12 scheme identified 78 haplotypes and separated most of the strains in groups of double-locus variants (DLV) supporting some phylogenetic relationships. DLV groups were subdivided into closely related clusters of strains most often sharing the same geographical origin and isolated over a short period, supporting epidemiological relationships. The main MLVA-12 DLV group#1 was composed by strains from South America and all the African strains. The MLVA-12 scheme combining both minisatellite and microsatellite loci with different discriminatory power is expected to increase the accuracy of the phylogenetic signal and to minimize the homoplasy effects. Further investigation of the global epidemiology of Xpm will be helpful for a better control of CBB worldwide.

Fig 1. Visualization of multiplex PCR amplicons on agarose gels.

MW: Molecular marker. A. Pool I, loci Xpm 2–22, Xpm 2–18 and Xpm 2–35. B. Pool II, loci Xpm 2–33, Xpm 2–5 and Xpm 2–29. C. Pool III loci Xpm 2–3, Xpm 2–20 and Xpm 2–23.

Fig 2. Minimum spanning tree displaying relationships between haplotypes using the MLVA-8 minisatellite scheme.

Colors indicate the geographical origin of the haplotype, and the circle size indicates the number of strains of each haplotype. Numbers indicate the number of locus variants between haplotypes.

Fig 3. Genetic structure of a worldwide collection of Xanthomonas phaseoli pv. manihotis based on the discriminant analysis of principal components (DAPC) of minisatellite and microsatellite data (MLVA-12).

Numbers and colors represent the seven genetic clusters retained from Bayesian information criterion (BIC) values. (A) Scatterplot representing axes 1 and 2 of the DAPC. (B) Scatterplot representing axes 1 and 3 of the DAPC.