A Method for Identification of Biotype-Specific Salivary Effector Candidates of Aphid

SIMPLE SUMMARY: Aphids are generally dietary specialists, colonizing a specific plant or a group of closely related plants, but a few species are generalists, colonizing hundreds of hosts across multiple plant families. In these generalist aphids, host-specialized lineages or host-specialized biotypes are often observed in nature. This is the case for the cotton-melon aphid, Aphis gossypii Glover. When introduced to alternative hosts, the host-specialized biotypes show poor fitness and may even die within a few days. The underlying mechanisms of aphid host specialization remain unknown until now. We hypothesized that host-specialized biotypes express biotype-specific salivary effectors or elicitors that determine the compatibility of aphid-plant interactions. In this research, we described three strategies to identify biotype-specific effectors in two host-specialized biotypes of A. gossypii, a biotype specialized in Malvaceae and another in Cucurbitaceae. The strategy of combining transcriptome and proteome has the highest efficiency, obtaining less than one dozen effector candidates, and we strongly recommend this strategy to identify biotype-specific effectors in aphids and other sap-sucking insects. ABSTRACT: Polyphagous aphids often consist of host-specialized biotypes that perform poorly in non-native hosts. The underlying mechanisms remain unknown. Host-specialized biotypes may express biotype-specific salivary effectors or elicitors that determine aphid hosts. Here, we tried three strategies to identify possible effectors in Malvaceae- (MA) and Cucurbitaceae-specialized (CU) biotypes of the cotton-melon aphid Aphis gossypii Glover. The whole-aphid RNA-seq identified 765 differentially expressed genes (DEGs), and 139 of them were possible effectors; aphid-head RNA-seq identified 523 DEGs were identified, and 98 of them were possible effectors. The homologous genes of published aphid effectors were not differentially expressed between CU and MA. Next, quantitative proteomic analyses of saliva identified 177 possible proteins, and 44 of them were different proteins. However, none of the genes of the 44 proteins were differentially expressed, reflecting the discrepancy between transcriptome and proteome data. Finally, we searched for DEGs of the 177 salivary proteins in the aphid-head transcriptomes, and the salivary proteins with expression differences were regarded as effector candidates. Through this strategy, 11 effector candidates were identified, and their expression differences were all confirmed by RT-qPCR. The combinatorial analysis has great potential to identify biotype-specific effector candidates in aphids and other sap-sucking insects.