Comparative Genomics of Two Closely Related Wolbachia with Different Reproductive Effects on Hosts

GBE Comparative Genomics of Two Closely Related Wolbachia with Different Reproductive Effects on Hosts Irene L.G. Newton1,*, Michael E. Clark2, Bethany N. Kent3, Seth R. Bordenstein3,4, Jiaxin Qu5, Stephen Richards5, Yogeshwar D. Kelkar6, and John H. Werren2 1 Department of Biology, Indiana University, Bloomington 2 Department of Biology, University of Rochester 3 Department of Biological Sciences, Vanderbilt University 4 Department of Pathology, Microbiology and Immunology, Vanderbilt University 5 Human Genome Sequencing Center, Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas 6 Department of Biostatistics and Computational Biology, University of Rochester *Corresponding author: E-mail: . Abstract Wolbachia pipientis are obligate intracellular bacteria commonly found in many arthropods. They can induce various reproductive alterations in hosts, including cytoplasmic incompatibility, male-killing, feminization, and parthenogenetic development, and can provide host protection against some viruses and other pathogens. Wolbachia differ from many other primary endosymbionts in arthropods because they undergo frequent horizontal transmission between hosts and are well known for an abundance of mobile elements and relatively high recombination rates. Here, we compare the genomes of two closely related Wolbachia (with 0.57% genome-wide synonymous divergence) that differ in their reproductive effects on hosts. wVitA induces a sperm–egg incompatibility (also known as cytoplasmic incompatibility) in the parasitoid insect Nasonia vitripennis, whereas wUni causes parthenogenetic development in a different parasitoid, Muscidifurax uniraptor. Although these bacteria are closely related, the genomic comparison reveals rampant rearrangements, protein truncations (particularly in proteins predicted to be secreted), and elevated substitution rates. These changes occur predominantly in the wUni lineage, and may be due in part to adaptations by wUni to a new host environment, or its phenotypic shift to parthenogenesis induction. However, we conclude that the approximately 8-fold elevated synonymous substitution rate in wUni is due to a either an elevated mutation rate or a greater number of generations per year in wUni, which occurs in semitropical host species. We identify a set of genes whose loss or pseudogenization in the wUni lineage implicates them in the phenotypic shift from cytoplasmic incompatibility to parthenogenesis induction. Finally, comparison of these closely related strains allows us to determine the fine-scale mutation patterns in Wolbachia. Although Wolbachia are AT rich, mutation probabilities estimated from 4-fold degenerate sites are not AT biased, and predict an equilibrium AT content much less biased than observed (57–50% AT predicted vs. 76% current content at degenerate sites genome wide). The contrast suggests selection for increased AT content within Wolbachia genomes. Key words: endosymbiont, reproductive manipulation, mutational bias. Introduction Wolbachia pipientis is a ubiquitous alphaproteobacterial symbiont of arthropods and nematodes, distantly related to the Rickettsial pathogens Ehrlichia and Anaplasma (Werren et al. 2008). Within arthropods, these bacteria can induce a variety of reproductive alterations, including feminization of males, male-killing, sperm–egg incompatibility (known as cytoplasmic incompatibility or CI), and parthenogenetic development (Stouthamer et al. 1993; Werren et al. 2008). Other effects on arthropod hosts include protection against viruses (Hedges et al. 2008), suppression of sterile and lethal mutants in Drosophila (Starr and Cline 2002), supplementation of essential B-vitamins in bedbugs (Hosokawa et al. 2010), and ß The Author 2016. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact 1526 Genome Biol. Evol. 8(5):1526–1542. doi:10.1093/gbe/evw096 Accepted: April 24, 2016 GBE Different Reproductive Effects on Hosts Genome Biol. Evol. 8(5):1526–1542. doi:10.1093/gbe/evw096 where it induces CI, whereas wUni infects Muscidifurax uniraptor, where it causes parthenogenesis. We use the comparison between these closely related Wolbachia to address general questions about changes that occur during the initial stages of intracellular bacterial genome evolution, possible associations of genotypic changes with host and phenotypic shifts, and the mutational patterns in Wolbachia. Materials and Methods Wolbachia DNA Isolation The same extraction and amplification procedures were used for both Wolbachia strains. The Nasonia strain IntG12.1 (Chafee et al 2011) was reared under standard laboratory conditions under low density with one female provided with two hosts for 48 h. Upon pupation, wasps were removed from hosts, placed in a clean tube, allowed to eclose, and aged for 1–3 days. They were then pooled and put on ice. Before transfer of wasps to 5.0 mm filter columns, each column was rinsed with 70% ethanol, immediately followed by a 2000 RPM spin for 2 min before transfer to a new tube. Wasps were then rinsed with 500 ml of sterile distilled water, followed by a 2000 RPM spin for 2 min. The insects are not homogenized prior to centrifugation. Rather, centrifugation releases hemolymph and cells from the intact insects. This method appears to provide cleaner Wolbachia preparations with lower amounts contamination from host or other bacterial DNA. Columns with wasps were then transferred to a new sterile tube and spun at 13,500 RPM for 20 min at 4 C. Columns were then removed and discarded. Supernatant was discarded and pellets resulting from the spin were suspended in sterile PBS. Resuspended pellets were then transferred to a new column filter (5.0 mm) and spun at 13,500 RPM for 30 s. The resulting pellet was resuspended in the PBS and transferred to a new column filter (5.0 mm) for a total passage through four columns total. For the final column, the spin was 5 min at 13,500 K to pellet the bacteria. The supernatant was removed and discarded. The pellet was frozen at 80 until DNA extraction. wUni was similarly harvested from Muscidifurax uniraptor. DNA was extracted using the Qiagen DNA tissue extraction kit according to the manufacturers instructions for gram negative bacteria. Whole-genome amplification was performed using the Qiagen REPLI-g multiple displacement amplification kit as per the manufacturers instructions. Prior to sequencing, purity of Wolbachia preparation was evaluated in two ways. During bacterial purification, following passage through filter and resuspension, a 5 ml sampl (...truncated)