(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



A male-killing Wolbachia endosymbiont is concealed by another endosymbiont and a nuclear suppressor [1]

['Kelly M. Richardson', 'School Of Biosciences', 'Institute', 'University Of Melbourne', 'Parkville', 'Victoria', 'Perran A. Ross', 'Department Of Chemistry', 'Bioscience', 'Aalborg University']

Date: 2023-03

Bacteria that live inside the cells of insect hosts (endosymbionts) can alter the reproduction of their hosts, including the killing of male offspring (male killing, MK). MK has only been described in a few insects, but this may reflect challenges in detecting MK rather than its rarity. Here, we identify MK Wolbachia at a low frequency (around 4%) in natural populations of Drosophila pseudotakahashii. MK Wolbachia had a stable density and maternal transmission during laboratory culture, but the MK phenotype which manifested mainly at the larval stage was lost rapidly. MK Wolbachia occurred alongside a second Wolbachia strain expressing a different reproductive manipulation, cytoplasmic incompatibility (CI). A genomic analysis highlighted Wolbachia regions diverged between the 2 strains involving 17 genes, and homologs of the wmk and cif genes implicated in MK and CI were identified in the Wolbachia assembly. Doubly infected males induced CI with uninfected females but not females singly infected with CI-causing Wolbachia. A rapidly spreading dominant nuclear suppressor genetic element affecting MK was identified through backcrossing and subsequent analysis with ddRAD SNPs of the D. pseudotakahashii genome. These findings highlight the complexity of nuclear and microbial components affecting MK endosymbiont detection and dynamics in populations and the challenges of making connections between endosymbionts and the host phenotypes affected by them.

Funding: This research was supported by an Australian Research Council ( https://www.arc.gov.au ) Discovery grant DP120100916 to AAH, as well as a National Institutes of Health ( https://nigms.nih.gov ) MIRA grant R35GM124701 and a National Science Foundation ( https://beta.nsf.gov ) CAREER grant 2145195 to BSC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability: All non-molecular data are available in the paper and from https://doi.org/10.26188/21862119.v1 . Molecular data are available from https://melbourne.figshare.com/articles/dataset/D_pseudotakahashii_ddRADseq/21310478 and Genbank (accession number NZ_JAPJVH010000000). Numbers in figures can be found at https://doi.org/10.26188/21892974.v1 , https://doi.org/10.26188/21863961.v1 , and https://doi.org/10.26188/21862119.v1 .

Copyright: © 2023 Richardson et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

In Drosophila pseudotakahashii, Richardson and colleagues [ 23 ] described a CI Wolbachia infection (wPse) present at a high incidence in natural populations and causing strong CI; however, the CI was weaker in older males from which the infection could be absent despite its high density in all females. Our analyses here demonstrate that CI-causing wPse is a Group-A Wolbachia, that is outgroup to a clade containing both wMel-like and wRi-like variants ( S1 Fig , [ 24 , 25 ]). Here, we describe a second Wolbachia strain in D. pseudotakahashii that is present at a low frequency and occurs alongside the CI strain where it causes MK. Unusually, male death occurred only after embryo development and was modifiable through a nuclear gene that segregated in some laboratory lines where it increased in frequency to the extent that sex ratio reverted. We use molecular approaches to characterize the MK strain that differs for some genomic regions to the coinhabiting CI strain but is identical in other regions. We explore the presence and phylogenetic relationships of several functionally relevant regions of the Wolbachia genome: the wmk gene linked to MK [ 26 ], loci known to cause CI (cifs) [ 27 , 28 ], and the broader WO prophage regions that house these loci [ 29 ]. We also identify the genomic region associated with nuclear suppression through segregating crosses using the newly sequenced D. takahashii genome [ 30 ]. Our findings raise the issue of whether male-killers have been much more common in natural populations than previously assumed given that they are unlikely to be detected in the presence of common CI infections and may be affected by nuclear suppressors.

A well-documented example of MK suppression is in the butterfly Hypolimnas bolina, where nuclear suppression of MK revealed a CI phenotype [ 15 , 16 ]. In this case, a high frequency of MK in a population that persisted for many years [ 17 ] was expected to produce strong selection for a nuclear suppressor because of the fitness advantage of rare males required for offspring production [ 18 ]. Rapid recovery of male production for male-killers associated with endosymbionts has also been documented in other systems including lacewings [ 19 ] and planthoppers [ 20 ]. The genetic basis of nuclear suppression is still unclear although in Hypolimnas bolina it involves a single chromosomal region [ 21 ] and in suppression generated following lab-based hybridization between 2 Drosophila species it is polygenic [ 22 ].

Although male-killers can result in all-female broods, there is variability in sex-ratio effects in some MK systems. In Drosophila innubila, Wolbachia density can vary among females which in turn correlates with female-biased offspring ratios, an effect that also has an epigenetic component and could contribute to stability of this infected system [ 5 , 6 ]. Moreover, while some male-killer phenotypes can be stable across long time periods with little resistance to them evolving over thousands of years [ 14 ], MK phenotypes associated with endosymbionts can also be suppressed by nuclear genes.

Male-killers in Drosophila typically result in embryo death; this includes MK associated with both Spiroplasma [ 8 ] as well as Wolbachia [ 6 , 7 , 9 ] endosymbionts. Typically, such male-killers are detected by a reduction in hatch rate coupled with changes in sex ratio; this can be one reason for their underappreciation in natural populations given that male-killers are not maintained in stocks when males are required to produce offspring [ 10 ]. However, while reproductive effects of Wolbachia involving CI and MK are typically mediated through effects on embryonic development that results in a loss of egg hatch, they may also be affected by sex-specific mortality later in development. For example, in mites and thrips, CI associated with Wolbachia has been reported as involving postembryonic mortality [ 11 , 12 ] and in planthoppers, mortality can occur late in development [ 13 ].

Male-killing (MK) phenotypes associated with endosymbionts were first investigated in ladybugs and butterflies [ 1 ]. While MK endosymbionts often occur at low frequencies in populations, they can persist and spread through horizontal transmission or by providing a fitness advantage, such as through resource allocation or the avoidance of sib mating [ 2 , 3 ]. They can also invade populations with endosymbiont strains that cause cytoplasmic incompatibility (CI) as long as they are compatible with the CI strain [ 4 ]. In Drosophila, several male-killers associated with Wolbachia and Spiroplasma endosymbionts have been described [ 5 , 6 ]. However, their incidence in this genus is likely to be underestimated, partly because they can be uncommon in populations compared to CI strains that are often at a high frequency [ 7 ].

Results

A rare Wolbachia strain causes female-biased sex ratios We established 188 D. pseudotakahashii isofemale lines from collections in Nowra, south-eastern Queensland, and northern Queensland, Australia. Of these, 3.72% (N = 7) were found to have only female F1 offspring (Table 1), and there was no significant difference in the incidence of female-biased lines across the collection sites (G = 3.87, df = 6, P = 0.769). No female-biased lines were found in the collections in northern Queensland or in a few individuals from Moorland. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 1. Percentage of female-biased lines in the isofemale lines set up from field populations of D. pseudotakahashii. https://doi.org/10.1371/journal.pbio.3001879.t001 Sequences of individuals from female-biased and non-female–biased lines using the Pgi and CO1 nuclear and mitochondrial markers showed almost no variation while sequences of the Ddc nuclear marker showed only a small amount of variation, and for all genes there was no separation of female-biased and non-female–biased lines (sequences in Genbank (accession number NZ_JAPJVH010000000)). These results, in addition to morphological examination of occasional males emerging from the female-biased lines, supported the conclusion that the female-biased lines are indeed D. pseudotakahashii. Nucleotide sequences from multiple female-biased lines were obtained for the 5 Wolbachia MLST loci and wsp [31]. Sequences presented a series of double peaks interspersed with sections without double peaks. These patterns were identical for the forward and reverse primers and present for all primer types and samples. Upon investigation, the “background” sequence was the same as the wPse CI [23] strain while the double peaks presented evidence for a second strain sharing many bases in common with wPse. We designed strain-specific primers and screened a subset of the isofemale lines (N = 111) using standard PCR. Only the female-biased lines amplified with the MK primers, suggesting that the lines with female-only offspring were indeed the only lines with the double infection. Further genomic analyses (outlined below) confirmed the presence of 2 Wolbachia strains. Treating copies of the female-biased lines with tetracycline resulted in emergence of male progeny and sex ratios that were closer to 50:50 compared to copies of the lines that were not treated with tetracycline (Table 2). RT-PCR with wsp_validation primers confirmed their uninfected status, and these lines became self-sustaining and no longer required the introduction of males from other lines, suggesting that the female-bias is indeed related to Wolbachia infection. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 2. Sex ratios of female-biased lines over the first 10 generations of laboratory maintenance. https://doi.org/10.1371/journal.pbio.3001879.t002

A double Wolbachia infection causes late MK We assessed sex-ratio distortion by the Wolbachia double infection and its maternal transmission through crosses. We first crossed females from female-biased lines N101MK and B302MK with males from the non-female–biased B116CI and N51CI lines (N = 15). Three lines originating from B302MK females produced both males and females (% female offspring of 16.67%, 20.00%, and 42.86%); however, the remainder of crosses had all female progeny. Female offspring from female-only lines (denoted by “MK1”) and a line derived from B302MK that produced male and female offspring (“MK2”) were chosen for a second set of crosses. Note that we use italics to designate lines expressing MK and CI phenotypes associated with Wolbachia. When MK1 females were crossed with males from the CI lines (B116CI and N51CI), uninfected line (TPH35-) or the mixed sex line (MK2), egg hatch proportions were somewhat lower than those from crosses involving females carrying CI-causing Wolbachia (Table 3) although differences between the 5 crosses were marginally nonsignificant (Kruskal–Wallis test, H = 8.01, df = 4, P = 0.072). However, the mean percent egg-to-adult viability was half that of the control crosses and this variable differed among the treatments (Kruskal–Wallis test, H = 16.86, df = 4, P = 0.002). Sex ratio in the progeny also differed significantly (Kruskal–Wallis test, H = 59.09, df = 4, P < 0.001) and progeny from MK females were almost all female (Table 3). This suggests that MK is occurring, and mostly at a later time point than expected based on studies in other species (e.g., D. pandora [7]). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 3. Crosses between individuals from MK, CI, and uninfected (-) lines. https://doi.org/10.1371/journal.pbio.3001879.t003 When females from the mixed sex MK2 line were crossed with CI males, males emerged from 5 of the 11 replicates (% female progeny of the 5 replicates ± SD = 53.63% ± 11.91), while the other 6 replicates had 100% female progeny, suggesting that the leakiness in MK in the female parent was not passed on to all her daughters. Mean % egg to adult viability (± SD) for the 5 crosses with males emerging was 68.31% ± 7.83 compared to 46.26 ± 13.00 for the replicates with only female progeny, indicative of MK rather than feminization. Screening indicated that all but 2 females from the N101MK and B302MK lines used in the crosses had both the MK and CI Wolbachia strains (N = 73), suggesting incomplete maternal transmission, which is common in Drosophila (e.g., [32,33]). Males from the B116CI and N51CI lines used in the crosses were 80% and 60% infected with the CI strain, respectively (N = 45 and 25, respectively), consistent with findings elsewhere [23]. All MK2 males had the CI strain (N = 9), while all but one had the MK strain.

Rapid loss of MK, but long-term stability of MK Wolbachia during laboratory culture Over time, males emerged in all the MK lines that were not tetracycline-treated (Table 2). By F10, the lines had only a slightly female-biased sex ratio. At F11, we tested 2 males and 2 females from a range of MK and non-female–biased lines and found that all but one female and all males from the female-biased lines carried the MK strain (S2 Table). We tested a further 34 males from the B289MK, B280MK, and B302MK lines (N = 7, 8, and 19, respectively) and all but 2 carried the MK strain, suggesting that despite the appearance of males, the MK infection was still present in the female-biased lines. After 68 generations, we again screened a subset of the lines for infection status. Despite the MK lines no longer having strong sex-ratio biases, all individuals tested from 5 MK lines were infected with the MK strain (females: N = 47, males: N = 47). Densities of the MK Wolbachia strain were similar between the sexes (S2 Fig, GLM: F 1,84 = 0.281, P = 0.597). Additionally, all females across the CI and MK lines were infected with the CI strain (N = 76). Consistent with previous research [23], presence of the CI strain was variable in males, with 90% of males from the CI lines (N = 30) and 89% of males from the MK lines (N = 47) carrying the CI strain, and Wolbachia density also being much lower in males than females (S2 Fig, F 1,129 = 145.531, P < 0.001). Sanger sequencing of the MLST genes from males singly infected with MK confirmed the presence of only the MK strain with alternate bases to the CI strain in locations where double peaks were present in double-infected individuals. Sequences published in Genbank (accession number NZ_JAPJVH010000000). Several factors may contribute to the reappearance of males in these lines, including the potential effects of laboratory rearing on the expression of sex ratio distortion. In 2 lines that we examined in detail (see below), we show that one of the main reasons was likely to have been the emergence of nuclear-based suppression of the MK phenotype.

Molecular analysis of segregating lines highlights a region with a selective sweep We performed ddRADseq on lines derived from N101MK and B302MK that produced mixed-sex offspring (MKS) or female-only offspring (MK) to identify genomic regions associated with MK suppression (Fig 2A). We identified 3 contigs of >1 Mbp length where SNPs were structured in line with MK-suppression phenotype (Fig 2B). Of these, a specific region on contig NW_025323476.1 (positions: 3,321,074 to 4,677,392) showed strongly reduced variation in the MK lines but normal patterns of variation in the MKS lines, which matched expectations of a selective sweep on the MK lines. This region was estimated to contain 131 unique genes and 153 gene products (see S4 Table for a list). Genome-wide heterozygosity was lower in MK lines than in MKS lines (S3 Table), though analysis of each contig in isolation showed that this difference was wholly due to heterozygosity differences in the 3 contigs from Fig 2B, where H O was 23% smaller than other contigs in MK lines but 49% larger than average in MKS lines. Backcrossing was expected to produce negative genome-wide F IS , and this was more negative in MKS lines than MK lines. In the 3 contigs from Fig 2B, F IS was 84% less negative than other contigs in MK lines but 60% more negative in MKS lines. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Molecular analysis of segregating lines; (A) is a cartoon detailing the experimental crosses and (B) shows frequency plots of non-reference alleles on the 3 contigs where structure followed MK suppression phenotype. A selective sweep pattern denoted by arrows is apparent on the NW_02532476.1 contig. The data underlying this figure can be found in https://doi.org/10.26188/21863961.v1. https://doi.org/10.1371/journal.pbio.3001879.g002

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3001879

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.