Rickettsia induces strong cytoplasmic incompatibility in a predatory insect

Rickettsia, a group of intracellular bacteria found in eukaryotes, exhibits diverse lifestyles, with some acting as vertebrate pathogens transmitted by arthropod vectors and others serving as maternally transmitted arthropod endosymbionts, some of which manipulate host reproduction for their own benefit. Two phenotypes, namely male killing and parthenogenesis induction are known as Rickettsia-induced host reproductive manipulations, but it remains unknown whether Rickettsia can induce other types of host manipulation. In this study, we discovered that Rickettsia induced strong cytoplasmic incompatibility (CI), in which uninfected females produce no offspring when mated with infected males, in the predatory insect Nesidiocoris tenuis (Hemiptera: Miridae). Molecular phylogenetic analysis revealed that the Rickettsia strain was related to Rickettsia bellii, a common insect endosymbiont. Notably, this strain carried plasmid-encoded homologues of the CI-inducing factors (namely cifA-like and cifB-like genes), typically found in Wolbachia, which are well-known CI-inducing endosymbionts. Protein domain prediction revealed that the cifB-like gene encodes PD-(D/E)XK nuclease and deubiquitinase domains, which are responsible for Wolbachia-induced CI, as well as OTU-like cysteine protease and ankyrin repeat domains. These findings suggest that Rickettsia and Wolbachia endosymbionts share underlying mechanisms of CI and that CI-inducing ability was acquired by microbes through horizontal plasmid transfer.

Methods
Crossing experiment

Treatment

Females and males of the Nesidiocoris tenuis OW1 (Rickettsia infected) and OW1tet (cured) lines were crossed in all possible combinations to compare the number of eggs laid and the hatch rates. Males and females were collected separately from newly emerged adults. A 3–6day old female was allowed to mate with a male for 7 days and then transferred to a container with a leaf of Crassula ovata for oviposition. Leaves were replaced every 2 days, and oviposition continued for 6 days. Ten days after the last day of oviposition, hatched nymphs were counted and leaves were dissected to count unhatched eggs. The food source was renewed every 2 days during the crossing experiments.

Statistical analysis

The total number of eggs laid was analysed using the Kruskal–Wallis test. Egg hatch rates were analysed using a generalised linear mixed model (GLMM) with binomial error and a logit-link function. Each mating pair was assigned a random effect. Based on the results of the GLMM, ANOVA was performed for each treatment to estimate the P-value using the chi-squared test with Bonferroni correction. Mating pairs with no oviposition were excluded from the analysis of egg hatch rates. The analyses were performed using R version 4.2.2.

Phylogenetic analysis

16S rRNA

The 16S rRNA sequences (1,494 bp) obtained through next-generation sequencing of the Rickettsia rNten-OW1 isolate were used for phylogenetic analyses. Phylogenetic trees based on the nucleotide sequences were constructed by the maximum likelihood method using MEGA11. Kimura’s two-parameter model, evaluated using the best fit method, was applied for the calculation.

cif gene

To infer the phylogeny of the cif gene, we obtained the nucleotide sequences of previously identified cif homologues. After manual reannotation as described by Martinez et al. (2020) (doi:10.1093/molbev/msaa209), cifA and cifB nucleotide sequences were aligned according to their amino acid translations using MUSCLE implemented in MEGA11. The aligned data were cleaned using the GBlocks tool in NGPhylogeny.fr to remove weakly conserved regions. Phylogenetic trees based on the cleaned nucleotide sequences were constructed by the maximum likelihood method using MEGA11. A general time-reversible model, evaluated by the best fit method, was applied for the calculation.