Skip to main content
PMC home page
eLife logoLink to eLife
. 2021 Oct 22;10:e67686. doi: 10.7554/eLife.67686

A single synonymous nucleotide change impacts the male-killing phenotype of prophage WO gene wmk

Jessamyn I Perlmutter 1,2,3,✉, Jane E Meyers 1,3, Seth R Bordenstein 1,3,4,5,✉
Editors: Dieter Ebert6, Patricia J Wittkopp7
PMCID: PMC8555981  PMID: 34677126

Abstract

Wolbachia are the most widespread bacterial endosymbionts in animals. Within arthropods, these maternally transmitted bacteria can selfishly hijack host reproductive processes to increase the relative fitness of their transmitting females. One such form of reproductive parasitism called male killing, or the selective killing of infected males, is recapitulated to degrees by transgenic expression of the prophage WO-mediated killing (wmk) gene. Here, we characterize the genotype-phenotype landscape of wmk-induced male killing in D. melanogaster using transgenic expression. While phylogenetically distant wmk homologs induce no sex-ratio bias, closely-related homologs exhibit complex phenotypes spanning no death, male death, or death of all hosts. We demonstrate that alternative start codons, synonymous codons, and notably a single synonymous nucleotide in wmk can ablate killing. These findings reveal previously unrecognized features of transgenic wmk-induced killing and establish new hypotheses for the impacts of post-transcriptional processes in male killing variation. We conclude that synonymous sequence changes are not necessarily silent in nested endosymbiotic interactions with life-or-death consequences.

Research organism: D. melanogaster

Introduction

Wolbachia are maternally transmitted, obligate intracellular bacteria primarily residing in the cells of germline tissues in many arthropod species worldwide (Hurst and Frost, 2015; Taylor et al., 2018.) To facilitate their spread through the host matriline, these bacteria hijack sex ratios, sex determination systems, and embryonic viability to cause various reproductive parasitism phenotypes (Yen and Barr, 1971; Hurst et al., 1999; Bouchon et al., 1998; Hunter, 1999). One such phenotype is male killing, whereby sons of infected females are selectively killed (Hurst et al., 1999; Hurst and Jiggins, 2000; Charlat et al., 2005). Male killing selfishly drives the spread of the bacteria when, for instance, brothers and sisters compete for limited resources, and male lethality yields females (the transmitters) a relative fitness benefit due to reduced competition with fewer siblings (Jaenike et al., 2003; Unckless and Jaenike, 2012; Skinner, 1985; Hurst, 1997). This form of reproductive parasitism can have profound impacts on host behavior and evolution, including a sex role reversal (Jiggins et al., 2000), host nuclear genome changes to resist the phenotype (Majerus and Majerus, 2010; Hornett et al., 2006), and potentially host population extinction (Groenenboom and Hogeweg, 2002). Population modelling also suggests male killing may be deployed as a control method to crash the population size of arthropod pests and disease vectors (Berec et al., 2016).

Many lines of evidence suggest a complex relationship between male-killing genotype and phenotype. For example, male killing can be suppressed by host background (Majerus and Majerus, 2010; Hornett et al., 2006; Jaenike, 2007; Mitsuhashi et al., 2011), where male hosts exhibit resistance to the phenotype even in the presence of Wolbachia. In addition, Wolbachia that do not cause male killing in one host species can cause male killing in a second host species upon introgression or transfer to a naïve host (Jaenike, 2007; Fujii et al., 2001). Furthermore, some strains induce death at different host developmental stages (embryo vs. larvae) or across different sex determination systems (ZW lepidopterans, XY dipterans, XO arachnids), and the number of surviving males may vary widely (Riparbelli et al., 2012; Sasaki et al., 2002; Zeh et al., 2005). Such findings specify that the expression of male killing, and particularly the genotype-phenotype relationship in these symbioses, is more complex than simply the presence or absence of a male-killing gene.

We recently identified a male-killing candidate gene, wmk (WO-mediated killing), from prophage WO in the wMel Wolbachia strain of Drosophila melanogaster based on comparative genomic, transgenic, and cytological approaches (Bordenstein and Bordenstein, 2016; Perlmutter et al., 2019). It is a putative transcription factor with two predicted helix-turn-helix (HTH) DNA-binding domains, and transgenic expression in D. melanogaster embryos recapitulates many cytological and molecular aspects of male killing including accumulation of DNA damage in male embryos that overlaps with sites of dosage compensation activity (Riparbelli et al., 2012; Perlmutter et al., 2019; Harumoto et al., 2018). The wmk gene and two cytoplasmic incompatibility factor (cif) genes that underlie cytoplasmic incompatibility (a parasitism phenotype whereby offspring die in crosses between infected males and uninfected females) occur nearby specifically in the eukaryotic association module (EAM) of prophage WO (Bordenstein and Bordenstein, 2016; Perlmutter et al., 2019; LePage et al., 2017), which refers to the phage genome that is inserted into the bacterial chromosome. The EAM is common in WO phages across several Wolbachia strains (Masui et al., 2001; Wu et al., 2004) and is rich in genes that are homologous to eukaryotic genes or annotated with eukaryotic functions (Bordenstein and Bordenstein, 2016). As such, the expression of reproductive parasitism genes from the EAM and tripartite interactions between phage WO, Wolbachia, and eukaryotic hosts are central to Wolbachia’s ability to interact with and modify host reproduction.

The discovery of wmk has now enabled investigation of the impacts of genetic variation on the transgenic male-killing phenotype. Although the role of wmk in male killing has not yet been assessed in nature, evidence based on native wmk genotypes and transgenic wmk expression in flies suggest refined intricacy to the interactions between genotype and phenotype. For example, the native Wolbachia wmk homologs in three closely-related strains (wMel of D. melanogaster, wRec of D. recens, and wSuzi of D. suzukii) have nearly identical nucleotide sequences that vary by one to three SNPs (Perlmutter et al., 2019), and yet these strains induce different forms of reproductive parasitism. wMel and wRec induce cytoplasmic incompatibility, wRec also causes male killing in some strains of a sister species, and wSuzi is not known to induce any reproductive phenotype (Jaenike, 2007; Hamm et al., 2014; Hoffmann, 1988). Previous transgenic testing also demonstrated that addition of nine additional amino acids to the N-terminal region of the Wmk protein at the site of a putative alternative start codon ablates the phenotype (Perlmutter et al., 2020). Therefore, the Wmk N-terminus is particularly sensitive to genetic alterations, which is notable since small modifications such as protein tags in this location are often common additions to proteins that do not typically interfere with function. Furthermore, D. melanogaster does not harbor Wolbachia known to cause male killing, although transgenic wmk expression recapitulates the phenotype in this host (Perlmutter et al., 2019), and the wMel strain is most closely-related to the wRec male-killing strain (Metcalf et al., 2014). In addition, some D. melanogaster contain male-killing Spiroplasma symbionts whose plasmid gene Spaid can also transgenically recapitulate male killing (Harumoto and Lemaitre, 2018). In summary, the evidence for Wolbachia and wmk transgenic male killing suggests an intricate and multifaceted genotype-phenotype landscape.

Building on this background, several key questions emerge: Do closely-related or distantly-related homologs induce male killing? How sensitive is the wmk phenotype to small or large genetic changes? And how adapted are wmk homologs to their arthropod hosts? Here, we evaluate codon-optimized wmk homologs that span a spectrum of genetic divergence (Figure 1), including homologs (i) from distantly-related hosts, (ii) with putative upstream alternative start codons, (iii) with a single amino acid change, or (iv) with one or more synonymous codon changes, all in the D. melanogaster host. In this way, we aimed to investigate how a variety of genetic alterations to wmk, from a single synonymous nucleotide to the gene level, affect the phenotype. In particular, we hypothesized that divergent strains would not induce the phenotype due to co-adaptation with a distantly-related host. We also anticipated that the alternative start codons would inhibit function based on previous results (Perlmutter et al., 2020). In contrast, we hypothesized that neither a single amino acid change nor synonymous codon changes would alter the male-killing phenotype. We report that while distant homologs do not cause a male-killing phenotype in this host, single amino acid and synonymous nucleotide changes remarkably do alter the phenotype. Thus, wmk male killing is sensitive to the full spectrum of genetic alterations at a fine-scale level not previously recognized. Notably, synonymous sequence changes and post-transcriptional processes appear to play a role in controlling the genotype-phenotype relationships that underpin wmk male killing.

Figure 1. Overview of experimental design.

To investigate the genotype-phenotype landscape, we transgenically expressed wmk homologs with varying degrees of genetic changes. These sequences are codon-optimized based on different codon biases due to different tRNA abundances in the divergent bacterial source and eukaryotic destination species (Plotkin and Kudla, 2011; Gustafsson et al., 2004). Transgenic wmk in Drosophila melanogaster embryos results in three different phenotypes: no killing, male killing, and killing of males and females. Compared to wMel wmk, these transgenes were either divergent homologs from other Wolbachia strains, a homolog with a single amino acid change, homologs with an additional nine codons at the 5’ ends of the genes starting at an alternative upstream start codon, or variants with a single synonymous codon or nucleotide difference. These genotypes resulted in varying degrees of RNA sequence- and amino acid-level changes. Created with BioRender.com.

Figure 1.

Figure 1—figure supplement 1. Homologs of wmk tested in this study include variation in native gene and transgene sequence identity as well as host species.

Figure 1—figure supplement 1.

(Top) Bayesian nucleotide phylogeny of insect hosts based on 652 bp of the cytochrome oxidase subunit 1 (COI) gene from D. melanogaster, H. bolina, D. bifasciata, C. cautella, D. innubila, D. borealis, D. suzukii, and D. recens. Branch labels and scale bar indicate posterior probability. Species names are colored either by group within the Drosophila genus or by Order Lepidoptera. Underlines indicate hosts in which male-killing strains have been reported. Colored circles indicate host species group. (Middle) Bayesian nucleotide phylogenies of native (middle left, non-transgenic) or transgene (middle right) wmk sequences based on 690 or 686 bp, respectively. Label colors reflect groups from host phylogeny for comparison. Colored circles indicate host species group. Branch labels and scale bar indicate posterior probability. wBif wmk branches distantly due to a highly divergent sequence. wBor and wInn wmk share a branch because they share the same transgenic sequence. (Bottom) Nucleotide alignment of transgenes tested in this study as compared to the previously tested wMel wmk, with the regions encompassing the Helix-turn-helix (HTH) protein domains marked with black lines. Black ticks indicate sequence differences with the wMel wmk reference strain, light gray indicates sequence matches to the reference sequence, and white indicates an indel in at least one strain. In parentheses under strain names, percentages refer to nucleotide similarity compared to the wMel wmk transgene. MK indicates a male-killing strain in its native host, transgenic MK indicates the ability to induce a biased sex ratio transgenically only, and MK in sister species indicates the ability to kill males in a sister host species but not in the native host species. Alignment excludes the HA tag in the tagged strain, which is located between the two HTH domains.
Open in a new tab

Results

Closely-related homologs of wmk

We tested homologs from strains related to wMel that occur in several groups of species in the Drosophila genus, along with two from Order Lepidoptera (Figure 1—figure supplement 1). The phylogenetic clades of both the native Wolbachia genes and their transgenes are similar to those of the host species. Roughly divided, the transgenes group into two clusters that offer a range of genetic divergence for evaluating genotype-phenotype relationships (Figure 1—figure supplement 1) - those distantly related to the wMel wmk transgene (less than 90 % codon-optimized nucleotide sequence similarity to transgenic wMel wmk) and those closely related (greater than 90 % identity, from wRec, wSuzi, and an HA-tagged wMel homolog).

The closely-related wmk homologs to wMel include those from Wolbachia strains (i) wSuzi of the fruit pest species D. suzukii that has no confirmed reproductive phenotype but notably occurs in populations with female-biased sex ratios (Hamm et al., 2014; Drummond et al., 2019) and (ii) wRec of D. recens that kills males when introgressed into its sister species, D. subquinaria (Jaenike, 2007). The natural wSuzi wmk homolog has one synonymous single nucleotide polymorphism (SNP) compared to the wMel wmk reference, thus yielding the same amino acid sequence. The natural wRec wmk homolog has three SNPs, one of which is non-synonymous relative to wMel wmk, located in the first HTH DNA-binding domain (Figure 2A). These transgenes were codon optimized for expression in D. melanogaster. We hypothesized that both transgenes would induce a biased sex ratio comparable to wmk when expressed in D. melanogaster. However, transgenic expression of both homologs unexpectedly resulted in death of all expressing flies, both male and female (Figure 2B). In addition, we simultaneously tested a transgene of wMel wmk with an internal 3X HA tag epitope in the linker region between the two HTH DNA-binding domains. This transgene, HA-wmk, exhibited a sex ratio bias comparable to wMel wmk, as expected (Figure 2B). Therefore, two transgenes of wMel wmk (one with a tag) resulted in the biased sex ratio previously reported (Perlmutter et al., 2019), while two closely-related transgenes from strains that at least associate with female-biased host sex ratios yielded an all-killing phenotype.

Figure 2. Transgenic expression of closelyrelated wmk homologs causes male-killing and all-killing phenotypes in D. melanogaster.

(A) Schematic of wMel, wSuzi, and wRec wmk native nucleotide sequences. The blue tick mark indicates a non-synonymous nucleotide difference. Black tick marks indicate synonymous nucleotide changes. Numbers indicate nucleotide position across the entire 912 nucleotide sequence. (B) Sex ratios of adult flies are shown for expressing (Act5c-Gal4) and non-expressing (CyO) embryonic offspring. Each sample point represents the adult offspring (N = 50–157, mean 86) produced by a replicate family of ten mothers and two fathers, with expressing and non-expressing flies of a given genotype being siblings. Bars represent the mean sex ratio. Statistics are based on a Kruskal-Wallis, one-way ANOVA followed by Dunn’s correction across either expressing or non-expressing flies. wRec and wSuzi wmk have no points in the expressing category due to death of most or all males and females. HA-wmk contains a 3 X HA tag in the linker region between the two helix-turn-helix domains. This experiment was performed twice. Data and statistical outputs are available in Figure 2—source data 1 and Figure 2—source data 2, respectively. (C) Gene expression in embryos 4–5 h AED of each indicated wmk transgene from (B), relative to Drosophila housekeeping gene, rp49. There is no significant difference in expression based on a Kruskal-Wallis one-way ANOVA followed by Dunn’s correction. Data and statistical outputs are available in Figure 2—source data 3 and Figure 2—source data 4, respectively. (D) Gene expression in embryos 4–5 h AED of the host msl-2 dosage compensation gene relative to rp49 under simultaneous expression of the indicated transgene. There is no significant difference in expression based on a Kruskal-Wallis one-way ANOVA followed by Dunn’s correction. Data and statistical outputs are available in Figure 2—source data 5 and Figure 2—source data 6, respectively. (E) Predicted RNA secondary structures of native wMel wmk and several transgene strains. Black arrows point to the location of the start codon within each structure.

Figure 2—source data 1. Data for sex ratios of closely-related homologs corresponding to Figure 2B.
Figure 2—source data 2. Statistical output of Kruskal-Wallis test corresponding to sex ratios of closelyrelated homologs in Figure 2B.
Figure 2—source data 3. Data for qPCR of closely-related transgenes corresponding to Figure 2C.
Figure 2—source data 4. Statistical output of Kruskal-Wallis test corresponding to qPCR for transgene expression in Figure 2C.
Figure 2—source data 5. Data for qPCR of msl-2 expression with transgene expression corresponding to Figure 2D.
Figure 2—source data 6. Statistical output of Kruskal-Wallis test corresponding to qPCR for msl-2 expression in Figure 2D.

Figure 2.

Figure 2—figure supplement 1. wRec and wSuzi transgenes expressed with an alternative start codon lose their transgenic phenotypes.

Figure 2—figure supplement 1.

(Top) Sex ratios of adult flies are shown for expressing (Act5c-Gal4) and non-expressing (CyO) offspring. Each sample point represents the adult offspring (N = 50–120, mean 84) produced by a replicate family of 10 mothers and 2 fathers, with expressing and non-expressing flies of a given genotype being siblings. Bars represent the mean sex ratio. Statistics are based on a Kruskal-Wallis one-way ANOVA followed by Dunn’s correction across either expressing or non-expressing flies. This experiment was performed twice. Data and statistical outputs are available in and , respectively. (Bottom) Predicted RNA secondary structures of the wRec and wSuzi wmk transgenes with the additional 5’ sequence exhibit slight structural changes compared to their non-lengthened counterparts (included again from Figure 2 for ease of comparison). Black arrows point to the location of the start codon within each structure.
Figure 2—figure supplement 1—source data 1. Data for sex ratios of 5’ alternative start codon transgene expression corresponding to Figure 2—figure supplement 1.
Figure 2—figure supplement 1—source data 2. Statistical output of Kruskal-Wallis test corresponding to sex ratios of 5’ alternative start codon transgene expression in Figure 2—figure supplement 1.
Open in a new tab

Gene expression similarities

To assess if variation in gene expression underpins the phenotypic variation, we measured the transcript levels of the transgenes in embryos 4–5 h AED (after egg deposition) when wmk kills males (Perlmutter et al., 2019). Gene expression levels are not significantly different across wSuzi, wMel, wRec, or HA tag wMel wmk transgenes, indicating that transcript levels do not account for the phenotypic differences (Figure 2C). An alternative hypothesis is that transgenic expression of the wSuzi and wRec homologs impacts native expression of a host gene in D. melanogaster that causes male and female lethality. For instance, transgenic expression of the DNA-binding dosage compensation gene, male-specific lethal 2 (msl-2), can induce total lethality with male-killing Spiroplasma (Cheng et al., 2016). Based on this, we quantified msl-2 transcript levels in embryos expressing wMel wmk (sex ratio bias), wSuzi wmk (all expressing hosts die), and wBif wmk (no killing or sex ratio bias) and found that msl-2 levels were comparable across all genotypes and phenotypes (Figure 2D).

RNA structural model variation

We next considered mRNA secondary structural differences of the transgene transcripts as a factor explaining the observed phenotypic variation. Modeling RNA structures showed they were substantially different, even in the case of few to no amino acid level changes from one sequence to another. Notably, some RNA structural features grouped by phenotype, such as the location of the start codon (Figure 2E). Two transgenic transcripts, wMel wmk and HA-wmk, cause a sex ratio bias and have start codons in the middle of the structure, while the other transgenes, wSuzi wmk and wRec wmk, that kill all flies have start codons on outer loops or hairpins. Although caution is warranted with this predicted structural analysis, mRNA secondary structure could explain some phenotypic outcomes of closely-related wmk homologs. Native wMel wmk also has a start codon on an outer loop but does not induce an ‘all killing’ phenotype like the wSuzi and wRec wmk transgenes, which may be due to lower expression of native genes (Perlmutter et al., 2019) and/or other structural differences.

Distantly-related homologs of wmk

To determine if wmk homologs from distantly-related strains induce a biased sex ratio in D. melanogaster, we transgenically expressed four codon-optimized homologs from known male-killing strains of Wolbachia: the wBol1b strain from Hypolimnas bolina butterflies (Dyson et al., 2002), the wBif strain from Drosophila bifasciata flies (Riparbelli et al., 2012), the wCaub strain from Cadra cautella moths (Sasaki et al., 2002), and the wInn strain from D. innubila flies (Dyer and Jaenike, 2004) (same gene sequence as the wBor male-killing strain from D. borealis flies Carson, 1956; Figure 1—figure supplement 1). While transgene expression of wMel wmk induces a biased sex ratio (~one third of expressing males die), none of the more distantlyrelated wBol1b, wBif, wCaub, or wInn/wBor wmk homologs yield a biased sex ratio, demonstrating that they do not recapitulate male killing when transgenically expressed in D. melanogaster under the conditions tested (Figure 3).

Figure 3. Divergent homologs of wmk from male-killing strains do not induce a biased sex ratio in D. melanogaster.

Figure 3.

Open in a new tab

Sex ratios of adult flies are shown from either expressing (Act5c-Gal4) or non-expressing (CyO) offspring. WT refers to the background insertion line and Control gene refers to the WD0034 control transgene that induces no sex ratio bias. Each sample point represents the adult offspring (N = 50–132, mean 69) produced by a replicate family of ten mothers and two fathers, with expressing and non-expressing flies of a given genotype being siblings. Bars represent the mean sex ratio. Statistics are based on a Kruskal-Wallis one-way ANOVA followed by Dunn’s correction across either expressing or non-expressing flies. This experiment was performed twice. Data and statistical outputs are available in Figure 3—source data 1 and Figure 3—source data 2, respectively.

Figure 3—source data 1. Data for sex ratios of distantly-related homologs in Figure 3.
Figure 3—source data 2. Statistical output of Kruskal-Wallis test corresponding to sex ratios of divergent homologs in Figure 3.

Alternative start codon variation and male killing

Relevant to the studies here, we previously provided evidence that some strains contain alternative start codons upstream of the annotated start for wmk, and these upstream regions are expressed in the wMel strain (Perlmutter et al., 2020). When wMel wmk was transgenically expressed with the most likely upstream start codon, the phenotype was lost, and no biased sex ratio resulted. We also showed that some non-male-killing strains tended to have more of the alternative start codons (Perlmutter et al., 2020). To determine if wRec and wSuzi wmk transgene phenotypes are similarly sensitive to transcript changes, we expressed them with upstream codons that are native to each of their genetic sequences. As previously observed with other homologs, they lost their killing phenotype with only nine amino acids added to the 5’ end of the gene, despite being smaller than many commonly used protein tags (Figure 2—figure supplement 1). All expressing flies survived with no sex ratio bias. Returning to the RNA structure models, we find that simply adding the corresponding nucleotides at the 5’ end of each homolog resulted in several predicted differences in RNA secondary structure for each transgene compared to the structures without the additional 5’ nucleotides (Figure 2—figure supplement 1). This includes additional loops and different predicted placement of the start codon.

Silent site variation and male killing

Finally, to identify particular nucleotides that may account for phenotypic variation among the homologs, we aligned the sequences of the four transgenes in Figure 2 and investigated codons that clustered by phenotype (sex ratio bias for wMel wmk and HA-wmk, or all killing for wRec and wSuzi wmk). Across the length of the genes (and excluding the HA tag), there were only two codon differences: one at the sixteenth amino acid position and another near the end of the gene. As previous work demonstrated that changes at the 5’ end of this gene affect phenotype (Perlmutter et al., 2020) and since approximately the first 10 codons in model prokaryotic genes are known to substantially affect mRNA structure and resulting translation rate (Bentele et al., 2013), we focused on the earlier codon at site 16. This codon, which codes for Serine in all homologs, segregates among sequences by phenotype. The HA-wmk and wMel wmk transgenes, which recapitulate male killing, have a TCG codon, while the all-killing transgenes wSuzi and wRec wmk have TCC and AGC, respectively (Figure 4A).

Figure 4. Synonymous nucleotide changes in the 16th codon position of wmk alters resulting phenotype.

Figure 4.

Open in a new tab

(A) Sequence alignment of transgenic wmk homologs. The codon farthest on the left is the fourth codon in the sequence, and the highlighted codon is the 16th, with the farthest right representing the 23rd codon, and ellipses indicating codons continuing on either side. The red box outlines where the genotypes cluster by phenotype. The ‘HA-wmk’ and ‘wMel wmk’ genotypes share the same codon in this position, and both induce male-specific death. The ‘wRec wmk’ and ‘wSuzi wmk’ genotypes both exhibit different codons from the previous two and exhibit an all-killing phenotype. Colors correlate with amino acid identity. (B) Sequence alignment of transgenes with either the wMel wmk sequence made anew (wMel wmk new), or with the 16th codon (red box) replaced with the synonymous codons from the wRec and wSuzi wmk transgenes. The colors and symbols reflect those in (A). (C) Sex ratios of adult transgenic flies are shown for expressing (Act5c-Gal4) and non-expressing (CyO) offspring that include the original transgene wMel wmk strain used in previous figures, along with the newly created identical wMel wmk (new) transgene and the additional transgenes with the single codon swapped out for the indicated codons noted in (A). wSuzi codon and wRec codon refer to the strains that have the same sequence as the wMel wmk, but with one or three silent sites changed in the single codon at the 16th amino acid position. Each sample point represents the adult offspring (N = 50–161, mean 73) produced by a replicate family of 10 mothers and two fathers, with expressing and non-expressing flies of a given genotype being siblings. Bars represent the mean sex ratio. Statistics are based on a Kruskal-Wallis one-way ANOVA followed by Dunn’s correction across either expressing or non-expressing flies. This experiment was performed twice. Data and statistical outputs are available in Figure 4—source data 1 and Figure 4—source data 2, respectively. (D) Gene expression in embryos 4–5 h AED denotes expression of each transgene relative to that of rp49. There is no significant difference in expression based on a Kruskal-Wallis one-way ANOVA followed by Dunn’s correction. Data and statistical outputs are available in Figure 4—source data 3 and Figure 4—source data 4, respectively. (E) Predicted RNA secondary structures are shown for the wMel wmk transcript compared to both of the wRec or wSuzi codon transgenes exhibiting slight structural differences. The structure for transgene wMel wmk is included again from Figure 3 for ease of comparison. Black arrows point to the location of the start codon within each structure. Black circles highlight a key area of difference between the structures, with a stem absent in the wSuzi codon strain, and different base pair match probabilities calculated for each as indicated by color (blue to red, low to high probability). Within the black circle, the wSuzi codon transgene structure is missing a predicted stem that the others have. The stem in the wRec codon line, while present, has a weaker prediction as noted by the cooler colors, so there may be structural differences compared to the wMel wmk model.

Figure 4—source data 1. Data for sex ratios from expression of transgenes with single codon changes corresponding to Figure 4C.
Data for qPCR from expression of transgenes with single codon changes corresponding to Figure 4D.
Figure 4—source data 2. Data for qPCR from expression of transgenes with single codon changes corresponding to Figure 4D.
Figure 4—source data 3. Data for qPCR from expression of transgenes with single codon changes corresponding to Figure 4D.
Figure 4—source data 4. Statistical output of Kruskal-Wallis test corresponding to qPCR from expression of transgenes with single codon changes in Figure 4D.

To functionally test if the silent changes in this Serine codon accounted for the phenotype differences, we generated three new transgenes (Figure 4B): (i) wMel wmk control with no changes compared to the previously tested transgene (TCG codon, labeled ‘wMel wmk (new)’), (ii) wMel wmk with three nucleotide changes in the codon that reconstitutes the AGC present in the wRec transgene line (labeled ‘wRec codon’), and (iii) wMel wmk with one nucleotide change in the codon that reconstitutes the TCC present in the wSuzi transgene line (labeled ‘wSuzi codon’). When these three otherwise identical genes were transgenically expressed, the wMel wmk (new) transgene with no changes caused a biased sex ratio as expected; however, and remarkably, expression of transgenes with two different Serine codons ablated the phenotype and resulted in a non-biased sex ratio with normal numbers of expressing flies (Figure 4C). This ablation occurs even though transcript levels remain similar across all transgenes and despite sequencing confirmation of the single codon differences (Figure 4D). Thus, a minimum of one single synonymous site change in the 5’ region was sufficient to alter the sex ratio phenotype. However, while the nucleotide changes in the codon changed the phenotype, they did not recapitulate the all-killing phenotypes of their corresponding homologs. The predicted RNA secondary structures from the transgenes with the single codon changes are similar to the original wMel wmk transgene, but they differ in some aspects such as presence or absence of stems and loops and the probability score of the base pair match as indicated by color (scale of red to blue, warmer colors indicate high probability, cooler colors indicate low probability). (Figure 4E).

Discussion

Linking mutations to function is crucial in resolving the evolutionary dynamics of adaptations, especially when the functions emerge in a nested system of tripartite phage-bacteria-animal interactions. In a simple case, the direct impact of genetic divergence is functional divergence with increasing numbers of mutations leading to increased functional divergence. Here, we investigated evolutionary and molecular hypotheses related to the genotype-phenotype relationships between the prophage WO-mediated killing (wmk) gene (Perlmutter et al., 2019; Perlmutter et al., 2020) and male killing in D. melanogaster. Sequences among wmk homologs in nature can differ across divergent hosts, leading to the hypothesis that highly divergent alleles are functionally fine-tuned. Using a variety of sequences, we report three key results: (i) mutations even at the synonymous codon and single nucleotide levels can alter the male-killing phenotype, (ii) phenotype, genotype, and RNA structural variation exhibit some correlation with each other, and (iii) distantly-related homologs do not induce a male-killing phenotype in transgenic D. melanogaster, while some closely-related homologs from Drosophila species/populations with female-biased sex ratios induce all-killing phenotypes. We discuss how these findings expand and support mechanistic models of wmk male killing, emphasize important implications for transgenic assays in endosymbiont studies, and relate these findings to aspects of male killing such as phenotype switching and host resistance.

The most surprising and unanticipated result was that transgene expression of highly similar homologs and even single synonymous site changes alter phenotype, and they make the difference between life and death for some males. We tested wmk homologs with high native sequence identity compared to native wMel wmk: wRec wmk from the mushroom-feeding D. recens (native sequence has two synonymous and one non-synonymous nucleotide differences) and wSuzi wmk from the fruit crop pest D. suzukii (native sequence has one synonymous nucleotide difference). Although we anticipated similar results to wMel wmk expression, we found that transgenic expression of these genes killed all flies (Figure 2), even though the wSuzi wmk transgene produces an identical protein to transgenic wMel Wmk. Of particular note, transgenic wSuzi wmk (all killing) and the original transgenic wMel wmk sequence (male killing) produce proteins of the exact same amino acid sequence, but they shared only a 91% sequence identity in the codon-optimized transgene, largely due to an updated codon optimization algorithm. Codon optimization is the norm in transgenic symbiosis research under the common assumption that synonymous codons are functionally redundant. However, codon optimization algorithms often choose a codon based on factors including codon adaptation, mRNA folding, regulatory motifs, nucleotide bias, or codon correlations and biases (Plotkin and Kudla, 2011). With even a few different codons input into the algorithm as well as algorithm updates over the years, the tested transgenes had different nucleotide sequences.

Based on the aforementioned results, we next sought to assess the sensitivity of male killing to wmk transcriptional and post-transcriptional changes and asked if one synonymous codon or site change was sufficient to alter phenotype. We created three transgenic lines, each with different DNA sequences coding for the same Serine at position 16. With only these minor changes that encode an identical protein sequence to wMel Wmk, the transgenic male-killing phenotype was ablated (Figure 4). Thus, one single codon and even a single nucleotide, remarkably determined male viability. Importantly, this key result implies at the molecular level that endosymbiont phenotypes of reproductive parasites may not simply be governed by DNA sequence alone.

A key question related to our findings is how synonymous changes cause vastly different phenotypes. It is often assumed in endosymbiont transgenic experiments that synonymous changes do not affect function, and codons for the same amino acid are functionally redundant. However, several decades of research have uncovered mounting evidence that the functional redundancy hypothesis is not always accurate (Plotkin and Kudla, 2011). Indeed, codon bias varies across species (hence the need for codon optimization) (Sharp et al., 1997; Andersson and Kurland, 1990), and there are several mechanisms through which single or few codons can influence transcription or translation. When there is a rare codon, there are fewer tRNAs available, and the translation rate may be correspondingly lowered (Li et al., 2012). Notably, there is a bias for rare codons in N-terminal regions of bacterial genes, which is likely due to their influence on mRNA structure (Goodman et al., 2013). Lower 5’ RNA stability or weaker structure can also be correlated with ease of initiation and faster translation (Gu et al., 2010), and stem loop structures may affect translation rate in varying ways. Indeed, strong 5’ RNA secondary structure may inhibit ribosomal initiation or translational efficiency (de Smit and van Duin, 1990; Bettany et al., 1989), and there can be selection against SNPs that alter RNA secondary structure (Salari et al., 2013). There is also evidence for selection across the domains of life on codon usage in the first 30–60 nucleotides of a gene, likely due to their impact on mRNA structure near the site of initiation (Gu et al., 2010). Additionally, early codons with low frequencies can be beneficial by slowing elongation, thus preventing ribosome collisions, or potentially helping to recruit chaperones to the emerging peptide for proper folding (Tuller et al., 2010; Fredrick and Ibba, 2010). In addition, some specific codons across the sequence are favored for their lower likelihood of mistranslation (Qin et al., 2004). Indeed, synonymous codon differences in genes can result, for instance, in altered gene expression (Kudla et al., 2009), lower organism fitness (Agashe et al., 2013), or disease (Sauna and Kimchi-Sarfaty, 2011).

Beyond codon sequence, gene expression levels across representatives of all phenotypes are similar (Figure 2), supporting post-transcriptional differences as the source of functional variation and refuting the functional redundancy hypothesis in this context. Examining the three synonymous transgenes more closely, we find that codon usage indeed correlates with phenotype. According to the codon usage table for D. melanogaster on the Genscript website, the original wMel wmk codon at this position (TCG) has a frequency 16.7 per 1000 codons. In contrast, the wSuzi codon (TCC) has a frequency of 19.5/1000, and the wRec codon (AGC) has a frequency of 20.5/1000 (GenScript, 2020). Therefore, there is a pattern where the more common codons are present in the all-killing transgenes, and at similar frequencies, while the strain expressing partial male killing has a codon at a lower frequency. Thus, some of the phenomena described in studies on synonymous codons and function likely underlie the phenotypic variation observed here.

Given that transgenic wmk expression with silent site changes results in different phenotypes, we used software that estimated RNA secondary structures for the tested transgenes. Indeed, we found that the predicted structures correlate somewhat with phenotype, so synonymous codons may change transcript structure or other post-transcriptional features of the transcript or protein (Figure 2). Other small changes including adding a 9-codon N-terminal sequence ablated the phenotype and reversed it to no death, despite the additional sequence being smaller than many common gene tags that typically do not interfere with function. This small change also alters predicted RNA structure (Figure 2—figure supplement 1). We were unable to assess protein levels to determine translational differences due to lack of an antibody. However, the data demonstrates that the wmk transgene is functionally sensitive to some post-transcriptional changes, at least in the N-terminal region.

What, then, do these results illuminate about transgenic wmk function and the utility of transgenic research in light of the potential for marked influences of synonymous changes on phenotype? First, as discussed, the wmk transgenic phenotypes are likely sensitive to post-transcriptional processes. This has important implications for understanding wmk in a natural male-killing context as soon as feasible techniques are available, since heterologous expression and its reliance on codon optimization may obscure our understanding of the gene’s biology. Second, since three different transgenic phenotypes (sex ratio bias, all killing, no killing) have been found so far with only a few sequences analyzed, it is possible that further testing of new codons may increase the partial male killing to a full male-killing phenotype. Therefore, it would be fruitful and may be possible to continue to uncover a transgenic sequence that fully recapitulates the phenotype to refine this system as a study tool for Wolbachia male killing. Third and critically, although the results show that there is some relationship between synonymous codons and phenotype, several points remain for further testing. For example, we cannot conclude that the particular codon tested here is responsible for phenotype alterations in other host genetic backgrounds or species. It is possible that this codon plays a functional role only in a singular host genetic context. Here, we changed wmk sequences while holding the host genetic background fixed, but the reverse is required to conclude whether or not the particular codon plays a general role in other genotypes or natural contexts. Second, due to possible coevolution, various codons may or may not yield similar functional effects across different host backgrounds, and additional synonymous sites may contribute to the male-killing phenotype. Thus, the results here illuminate a previously unrecognized need for future research on the functional impacts of synonymous substitutions in endosymbionts. Future work may focus on determining if there is one specific synonymous codon that affects the male-killing function in all cases, if a more general feature exists where alteration of any or a subset of N-terminal or other wmk codons affects function, or if the effect of synonymous changes is specific to this background.

In addition, these findings are informative with regards to the more general study of phenotypes induced by endosymbiont transgenes. It is standard practice to codon optimize genes for maximizing host expression when testing endosymbiont gene function (Perlmutter et al., 2019; LePage et al., 2017; Harumoto and Lemaitre, 2018), with the assumption that synonymous codons will not alter function. However, if codon optimization potentially changes the interpretation of transgenic findings, then phenotypes should be corroborated in natural contexts once tools such as genetic editing are available in the relevant organism. Specifically, wmk should be knocked out in native contexts once it is more technically feasible. In addition, codon optimization algorithms are updated with new information periodically with the assumption that they yield improved results, although it is unclear in practice whether an algorithm is better optimized to produce results that reflect the true biology of a transgene. Future work is necessary to explore these concepts further. For example, comparisons of alleles may need to be performed with alleles identical in sequence except for any engineered differences, and the algorithm should remain constant across all transgenes that are compared to each other. Further, careful analysis and comparison of transgenic phenotypes produced by different algorithms may be necessary in some cases where the phenotype is known or expected. This approach could ensure the algorithm produces a transgenic phenotype that most closely resembles the natural phenotype. In addition, the reliability of certain molecular evolutionary signatures, such as dN/dS, may be compromised since synonymous mutations are assumed to be neutral in these analyses. These principles are particularly important for research on endosymbionts that increasingly relies on heterologous gene expression for functional studies. The smaller genomes of endosymbionts tend to have low GC content overall (McCutcheon et al., 2009), and GC content is the strongest driver of codon usage bias (Knight et al., 2001; Chen et al., 2004) and contributes to the strength of mRNA secondary structure (Plotkin and Kudla, 2011), so careful attention to the effects of synonymous changes may be of acute interest to the endosymbiont research community.

We also find interesting support for the hypothesis that there is co-adaptation between wmk homologs and their hosts since male-killing genes may be evolutionary matched for the host sex determination and molecular machinery that they manipulate. We first analyzed wmk homologs that are more distantly related to wMel wmk, including from male-killing strains wBol1b, wBif, wCaub, and wInn/wBor (Sasaki et al., 2002; Dyson et al., 2002; Dyer and Jaenike, 2004; Hurst et al., 2000; Sheeley and McAllister, 2009). They did not recapitulate male killing when transgenically expressed in D. melanogaster. The lack of a sex ratio bias is expected if either wmk does not underpin male killing in these systems or divergent homologs are required to be closely co-adapted to their hosts. This latter hypothesis is based on observations that resistance to male killing is common (evidence of a potential host-microbe arms race) (Charlat et al., 2005; Majerus and Majerus, 2010; Jaenike, 2007; Mitsuhashi et al., 2011; Hornett et al., 2008), and some strains can cause male killing anew when transferred to another, usually closely-related host (Jaenike, 2007; Fujii et al., 2001; Sasaki et al., 2002) while not causing male killing in all recipient host species (Veneti et al., 2012; Hughes and Rasgon, 2014). It is possible, then, that the wmk homologs tested induce a sex ratio bias in their natural hosts, but not in D. melanogaster.

If we make the assumption that wmk is involved in male killing in nature, which requires confirmation beyond transgenic recapitulation of the phenotype, then the results here give the basis for additional hypotheses that require further testing. First, the killing of both males and females by the closely-related wRec and wSuzi wmk homologs could indicate that the target is something shared in both males and females, but functions differently within each sex (assuming no off-target effects). Indeed, previous studies on wmk and Wolbachia male killing demonstrate a positive correlation with between dosage compensation complex (DCC) activity and DNA defects in male embryos (Riparbelli et al., 2012; Perlmutter et al., 2019; Harumoto et al., 2018). Notably, four of the five protein components of the DCC are expressed in both males and females, and it is only msl-2 that is male-specific and catalyzes formation of the complex (Lucchesi and Kuroda, 2015). Thus, there are several non-sex-specific wmk target candidates that the gene product may interact with to cause lethality in males and females. Second, protein divergence and resultant conformational changes may impact the specificity between host target and the Wmk toxin, and could underlie development of host resistance. Male-killing genes are expected to evolve rapidly within hosts in order to counteradapt host resistance mechanisms and keep evolutionary pace with the rapid evolution of sex-related genes that may be the target of a male-killing toxin (Marín and Baker, 1998; Rodriguez et al., 2007). As such, major or minor divergence in protein or transcript sequence of either the host target or the microbial toxin may underpin changes that lead to common host resistance, wmk-host coadaptation, and functional transgene differences in the foreign D. melanogaster host. Third, these findings leave open the possibility of a variety of functionally relevant wmk protein or transcript conformations in nature, which could contribute to the marked diversity of Wolbachia male killing in terms of host species and sex determination systems (Hurst et al., 1999; Charlat et al., 2005; Sasaki et al., 2002; Zeh et al., 2005; Dyer and Jaenike, 2004; Hurst et al., 2000; Fialho and Stevens, 2000; Van Borm et al., 2008).

Taken together, this work reports previously unrecognized relationships for wmk-induced killing and establishes new hypotheses for the impacts of RNA structure and post-transcriptional processes in wmk-induced male killing. It also highlights several critical features for the research community regarding assumptions about the broad use of transgenes and the role of synonymous mutations in gene function. If wmk is involved in natural male killing, then this work could indicate how silent sequence changes may relate to known male-killing phenomena such as frequent host resistance or male-killing function in diverse hosts.

Materials and methods

Key resources table.

Reagent type (species) or resource Designation Source or reference Identifiers Additional information
Gene (Wolbachia pipientis) WD0626 NCBI NCBI:WD_RS02815 Also known as wmk
(WO-mediated killing)
Genetic reagent (D. melanogaster) Act5c-Gal4/CyO Bloomington Drosophila Stock Center BDSC:3953; FlyBase FBti0012290 P{AyGAL4}25
Genetic reagent (D. melanogaster) WT (y1w67c23; P[CaryP]P2) Bloomington Drosophila Stock Center BDSC:8622; FlyBase FBti0040535 WT strain used in this
study; P{CaryP}attP2
Genetic reagent (D. melanogaster) wMel wmk This paper; Perlmutter et al., 2019; PMID:31504075 Expresses codon-
optimized transgene;
UAS promoter
Genetic reagent (D. melanogaster) wBol1b wmk This paper Expresses codon-
optimized transgene;
UAS promoter
Genetic reagent (D. melanogaster) wBif wmk This paper Expresses codon-
optimized transgene;
UAS promoter
Genetic reagent (D. melanogaster) wCaub wmk This paper Expresses codon-
optimized transgene;
UAS promoter
Genetic reagent (D. melanogaster) wInn/wBor wmk This paper Expresses codon-
optimized transgene;
UAS promoter; wInn
and wBor wmk have
same exact sequence
Genetic reagent (D. melanogaster) wSuzi wmk This paper Expresses codon-
optimized transgene;
UAS promoter
Genetic reagent (D. melanogaster) wRec wmk This paper Expresses codon-
optimized transgene;
UAS promoter
Genetic reagent (D. melanogaster) HA-wmk This paper Expresses codon-
optimized transgene;
UAS promoter; 3 X HA
tag epitope in linker
between HTH domains
of wMel wmk
Genetic reagent (D. melanogaster) 5’ wRec wmk This paper Expresses codon-optimized transgene; UAS promoter;
Sequence has additional nine amino acids
starting at upstream alternative start codon
Genetic reagent (D. melanogaster) 5’ wSuzi wmk This paper Expresses codon-optimized transgene; UAS promoter;
Sequence has
additional nine amino acids starting at
upstream alternative
start codon
Genetic reagent (D. melanogaster) wMel wmk (new) This paper Expresses codon-optimized transgene; UAS promoter;
Same exact sequence
as wMel wmk, newly transformed strain
Genetic reagent (D. melanogaster) wSuzi codon This paper Expresses codon-optimized transgene; UAS promoter; Same as wMel wmk, but with 16th amino acid position using TCC Serine codon from wSuzi wmk strain
Genetic reagent (D. melanogaster) wRec codon This paper Expresses codon-optimized transgene; UAS promoter; Same as wMel wmk, but with 16th amino acid position using AGC Serine codon from wRec wmk strain
Recombinant DNA reagent pTIGER (plasmid) Ferguson et al., 2012; PMID:22328499 Modified pUASp
plasmid for enhanced germline expression under Gal4/UAS control
Sequence-based reagent Rp49_F This paper PCR primers CGGTTACGGAT
CGAACAAGC
Sequence-based reagent Rp49_R This paper PCR primers CTTGCGCTTCT
TGGAGGAGA
Sequence-based reagent wmk_homologs_opt_F This paper PCR primers CTGTATGCCATTG
CCGAGACCCT
Sequence-based reagent wmk_homologs_opt_R This paper PCR primers TCACCAGATCCTTG
GCGATCTTCATC
Sequence-based reagent Msl-2_F This paper PCR primers GGATTAACGCGGT
CTAAGCATGTGTAACTG
Sequence-based reagent Msl-2_R This paper PCR primers GTATGCCGTCTG
GGCCATGATG
Commercial assay or kit Direct-zol RNA MiniPrep Kit Zymo R2051
Commercial assay or kit Superscript VILO cDNA Synthesis Kit ThermoFisher 11754050
Chemical compound, drug DNase, RNase-free Ambion, Life Technologies AM2222
Chemical compound, drug iTaq Universal SYBR Green Mix Bio-Rad 1725120
Software, algorithm GraphPad Prism 8 GraphPad Prism 8 RRID:SCR_002798
Software, algorithm Geneious Pro v.2019.2; Geneious Pro v.2020.2.4 Geneious RRID:SCR_010519
Software, algorithm jModelTest jModelTest RRID:SCR_015244
Software, algorithm RNAfold WebServer University of Vienna,
Gruber et al., 2008,
Lorenz et al., 2011 		PMID:18424795; PMID:22115189 http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi
Open in a new tab

**Reagents source from this paper may be obtained from Bordenstein lab.

Drosophila strains and maintenance

D. melanogaster strains used in this study include Act5c-Gal4/CyO (BDSC 3953, ubiquitously-expressing zygotic driver), the WT background line of genotype y1w67c23; P[CaryP]P2 (BDSC 8622), the WD0626 (wmk) and WD0034 (control gene) transgene constructs previously described (Perlmutter et al., 2019), and several new transgene constructs. Briefly, each Wolbachia gene of interest was codon-optimized for optimal D. melanogaster expression using algorithms developed by GenScript Biotech (Piscataway, NJ). These sequences were then synthesized as DNA nucleotides and cloned by GenScript into the pTIGER plasmid (Ferguson et al., 2012). The pTIGER vector is a UASp-based plasmid optimized for germline expression and uses PhiC31 integrase (Groth et al., 2004) for targeted integration into the D. melanogaster genome. It also includes UAS promoters for inducible expression with a Gal4 driver (Southall et al., 2008) as well as a w+ red eye marker for transformant screening. The plasmids were then sent to BestGene (Chino Hills, CA), which performed injections of the vectors into embryos of the BDSC 8622 background line with PhiC31 integrase to integrate the vector into the attP2 insertion site in the genome. Successful transformants were selected based on the red eye marker, and each transgene line is descended from the offspring of a single transformant (isofemale).

D. melanogaster were reared on standard cornmeal, molasses, and yeast (CMY) media (5% w/v cornmeal (Quaker, Chicago IL), 1.875% w/v yeast (Red Star Yeast, Milwaulkee WI), 7.8% v/v molasses (Sweet Harvest Foods, Rosemount MN), 0.5% w/v type II Drosophila agar (Genesee Scientific, San Diego CA), 0.056% w/v tegosept (Genesee Scientific, San Diego CA), and 0.39% v/v propionic acid (Sigma Aldrich)). Stocks were maintained at 25 °C with virgin flies stored at room temperature. During virgin collections, stocks were kept at 18 °C overnight and 25 °C during the day. All flies were kept on a 12 hr light/dark cycle.

Sex ratio assays

To assess the effect of transgene expression on adult sex ratios (measurement of male killing), sex ratio assays were performed as previously described (Perlmutter et al., 2019). Briefly, twenty biological replicates of 10 uninfected, 4- to 7-day-old virgin, female Act5c-Gal4/CyO driver flies and two uninfected, 1- to 2-day-old virgin, male transgene flies were each set up in vials with CMY media. Individuals were randomly allocated to each vial after all females or males of a given genotype were mixed together. They were left on the media to lay eggs for 4 days at 25 °C with a 12 hr light/dark cycle, at which point adults were discarded. The vials are then left at 25 °C until the offspring are counted. After 9 days of adult offspring emergence, they were scored for both sex and expression (red eye color from Act5c-Gal4 chromosome) or non-expression (curly wings from CyO balancer chromosome). The number of adult offspring per vial across all experiments ranges from 50 to 170, with a mean of 120 (ranges and means per experiment are included in figure captions). Any vials with fewer than 50 adult offspring were removed from the analysis, as this indicates either abnormally poor egg laying or hatching (typically 0–2 vials per group). In addition, vials with no adult emergence, while others of the same genotype had typical levels of offspring, were also excluded (typically 2–3 vials per group). Results were graphed in GraphPad Prism 8.4.0, which applies a ‘Standard Plot Appearance’ correction for visibility of data distribution where the width of distribution of points is proportional to the number of points at that y-value.

RNA secondary structures

RNA secondary structures were generated by uploading the nucleotide sequences of the indicated gene to the RNA fold web server (Gruber et al., 2008; Lorenz et al., 2011). The structures shown are the graphical outputs of the MFE (minimum free energy) secondary structures. Colors indicate base pair probabilities, from blue to red, with blue indicating a probability of 0 and red indicating a probability of 1.

Gene expression

Gene expression was measured in Drosophila embryos aged 4–5 hr AED. Each point represents a biological replicate with the RNA of 30 pooled embryos from crosses between a unique set of 60 uninfected, 4- to 7-day-old virgin, female Act5c-Gal4/CyO driver flies and 12 uninfected, 1- to 2-day-old virgin, male transgene flies of the indicated genotype. Each point represents a biological replicate from different bottles. Individuals were randomly allocated to each bottle after all females or males of a given genotype were mixed together. Each collection chamber consisted of a grape juice agar plate with yeast in an eight oz round bottom bottle, and flies. These were placed in a 25 °C incubator overnight (16 hr). Then, the plates were swapped with fresh ones. The flies were allowed to lay eggs for 1 hr. The plates were then left at 25 °C for an additional 4 hr to age them to be 4–5 hr old (the estimated time of male death in wmk crosses). Embryos were then gathered in groups of 30 (each group from a unique bottle/biological replicate) and flash frozen in liquid nitrogen. RNA was extracted using the Direct-zol RNA MiniPrep Kit (Zymo), RNase-free DNase (Ambion, Life Technologies), cDNA was generated with SuperScript VILO (Invitrogen), and RT-qPCR was run using iTaq Universal SYBR Green Mix (Bio-Rad). qPCR was performed on a Bio-Rad CFX-96 Real-Time System. Primers are listed in the Key Resources Table. Conditions were as follows: 50 °C 10 min, 95 °C 5 min, 40 x (95 °C 10 s, 55 °C 30 s), 95 °C 30 s. Differences in gene expression were done by calculating 2-Δct (difference in ct values of two genes of interest). Data points were excluded if a sample had low-quality cDNA that did not amplify in qPCR. Data points of each biological replicate are measured as the mean of two technical replicates from each sample.

Phylogenetic trees

The nucleotide phylogenetic trees of host COI genes and wmk native gene or transgene sequences were inferred based on a MUSCLE alignment in Geneious Prime 2020.2.4 followed by stripping all sites with gaps. The resulting 652, 690, 686 nucleotide base pair alignments (respectively) were then analyzed via jModelTest 2.1.10 v20160303 (Darriba et al., 2012,Guindon and Gascuel, 2003). The AICc-corrected best model, JC, was predicted for all three alignments and was used to build the trees using the MrBayes (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck, 2003) Geneious plugin with the JC69 model (Jukes and Cantor, 1969) and equal rate variation.

Transgene sequence alignments

The sequence alignments of different wmk transgenes in Figure 2 and Figure 4 were conducted in Geneious Pro v.2019.2 using a MUSCLE alignment. Black bars in Figure 2 indicate sequence mismatches compared to the wMel wmk transgene reference sequence with any gaps stripped. Codons in Figure 4 are colored by amino acid.

Statistical analyses

Sample sizes for experiments were based on previous publications, which demonstrated repeatability in relative differences between treatment groups. Each experiment was completed twice, and statistical tests were applied to both to confirm repetition in differences between treatment groups. In all plots, the first experiment is the representative one shown. For sex ratios, we tested different sample sizes for reliability in sex ratio measurements, and found that 20 biological samples per group, with 10 females and 2 males per sample, resulted in consistently replicable data, which is the standard we apply here. qPCR data was approached similarly, by previous work in the lab demonstrating that with embryos 4–5 h AED, we are able to get consistent, high-quality, replicable data with 30 embryos per sample, and at least eight samples per group.

All statistical analyses for sex ratios and were performed using GraphPad Prism eight software. For sex ratios, a non-parametric Kruskal-Wallis one-way ANOVA followed by Dunn’s multiple comparisons test was applied to all gene-expressing categories, followed by the same test but on all non-expressing categories. For gene expression, groups were compared using the same Kruskal-Wallis one-way ANOVA with Dunn’s multiple comparisons test. Significant results are indicated with * symbols in figures, with accompanying p values in captions. Any comparisons with no symbol are nonsignificant. Full statistical information and outputs for all sex ratio and qPCR data is available in the Source Data file.

Acknowledgements

This work was supported by National Institutes of Health (NIH) grant R21 AI133522 to SRB, the Vanderbilt Microbiome Innovation Center, and NIH grants F31 AI143152, K-INBRE P20 GM103418, and National Science Foundation (NSF) Postdoctoral Research Fellowship (PRFB) DBI 2109772 to JIP. We would also like to thank RL Unckless for helpful comments on phylogenies.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Contributor Information

Jessamyn I Perlmutter, Email: jessamyn.perlmutter@ku.edu.

Seth R Bordenstein, Email: s.bordenstein@vanderbilt.edu.

Dieter Ebert, University of Basel, Switzerland.

Patricia J Wittkopp, University of Michigan, United States.

Funding Information

This paper was supported by the following grants:

  • National Institutes of Health R21 AI133522 to Seth R Bordenstein.

  • National Institutes of Health F31 AI143152 to Jessamyn I Perlmutter.

  • Vanderbilt Microbiome Innovation Center General Funds to Seth R Bordenstein.

  • National Institutes of Health P20 GM103418 to Jessamyn I Perlmutter.

  • National Science Foundation DBI 2109772 to Jessamyn I Perlmutter.

Additional information

Competing interests

Is listed as an author on a patent related to the use of wmk in vector control. US Patent 20210000092 16/982708.

No competing interests declared.

Author contributions

Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Visualization, Writing – original draft, Writing – review and editing.

Investigation, Writing – original draft.

Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review and editing.

Additional files

Transparent reporting form
Source data 1. Source data for all graphical data sets and statistical tests performed for this study.
elife-67686-supp1.xlsx (67.1KB, xlsx)

Data availability

All data generated or analyzed during this study are included in the manuscript and supporting files. Source data files have been provided for Figures 3-6.

References

  1. Agashe D, Martinez-Gomez NC, Drummond DA, Marx CJ. Good codons, bad transcript: large reductions in gene expression and fitness arising from synonymous mutations in a key enzyme. Molecular Biology and Evolution. 2013;30:549–560. doi: 10.1093/molbev/mss273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Andersson SG, Kurland CG. Codon preferences in free-living microorganisms. Microbiological Reviews. 1990;54:198–210. doi: 10.1128/mr.54.2.198-210.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bentele K, Saffert P, Rauscher R, Ignatova Z, Blüthgen N. Efficient translation initiation dictates codon usage at gene start. Molecular Systems Biology. 2013;9:675. doi: 10.1038/msb.2013.32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Berec L, Maxin D, Bernhauerová V, Rohr J. Male-killing bacteria as agents of insect pest control. Journal of Applied Ecology. 2016;53:1270–1279. doi: 10.1111/1365-2664.12638. [DOI] [Google Scholar]
  5. Bettany AJ, Moore PA, Cafferkey R, Bell LD, Goodey AR, Carter BL, Brown AJ. 5’-secondary structure formation, in contrast to a short string of non-preferred codons, inhibits the translation of the pyruvate kinase mRNA in yeast. Yeast. 1989;5:187–198. doi: 10.1002/yea.320050308. [DOI] [PubMed] [Google Scholar]
  6. Bordenstein SR, Bordenstein SR. Eukaryotic association module in phage WO genomes from Wolbachia. Nature Communications. 2016;7:13155. doi: 10.1038/ncomms13155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bouchon D, Rigaud T, Juchault P. Evidence for widespread Wolbachia infection in isopod crustaceans: molecular identification and host feminization. Proceedings. Biological Sciences. 1998;265:1081–1090. doi: 10.1098/rspb.1998.0402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Carson H. A female producing strain of D. borealis Patterson. Drosoph Inf Serv. 1956;30:109–110. [Google Scholar]
  9. Charlat S, Hornett EA, Dyson EA, Ho PPY, Loc NT, Schilthuizen M, Davies N, Roderick GK, Hurst GDD. Prevalence and penetrance variation of male-killing Wolbachia across Indo-Pacific populations of the butterfly Hypolimnas bolina. Molecular Ecology. 2005;14:3525–3530. doi: 10.1111/j.1365-294X.2005.02678.x. [DOI] [PubMed] [Google Scholar]
  10. Chen SL, Lee W, Hottes AK, Shapiro L, McAdams HH. Codon usage between genomes is constrained by genome-wide mutational processes. PNAS. 2004;101:3480–3485. doi: 10.1073/pnas.0307827100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cheng B, Kuppanda N, Aldrich JC, Akbari OS, Ferree PM. Male-killing Spiroplasma alters behavior of the dosage compensation complex during Drosophila melanogaster embryogenesis. Current Biology. 2016;26:1339–1345. doi: 10.1016/j.cub.2016.03.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Darriba D, Taboada GL, Doallo R, Posada D. jModelTest 2: more models, new heuristics and parallel computing. Nature Methods. 2012;9:772. doi: 10.1038/nmeth.2109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. de Smit MH, van Duin J. Secondary structure of the ribosome binding site determines translational efficiency: a quantitative analysis. PNAS. 1990;87:7668–7672. doi: 10.1073/pnas.87.19.7668. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Drummond F, Ballman E, Collins J. Population Dynamics of Spotted Wing Drosophila (Drosophila suzukii (Matsumura)) in Maine Wild Blueberry (Vaccinium angustifolium Aiton) Sects. 2019;10:205. doi: 10.3390/insects10070205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dyer KA, Jaenike J. Evolutionarily stable infection by a male-killing endosymbiont in Drosophila innubila: molecular evidence from the host and parasite genomes. Genetics. 2004;168:1443–1455. doi: 10.1534/genetics.104.027854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dyson EA, Kamath MK, Hurst GDD. Wolbachia infection associated with all-female broods in Hypolimnas bolina (Lepidoptera: Nymphalidae): evidence for horizontal transmission of a butterfly male killer. Heredity. 2002;88:166–171. doi: 10.1038/sj.hdy.6800021. [DOI] [PubMed] [Google Scholar]
  17. Ferguson SB, Blundon MA, Klovstad MS, Schüpbach T. Modulation of gurken translation by insulin and TOR signaling in Drosophila. Journal of Cell Science. 2012;125:1407–1419. doi: 10.1242/jcs.090381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fialho RF, Stevens L. Male-killing Wolbachia in a flour beetle. Proceedings of the Royal Society of London. Series B. 2000;267:1469–1473. doi: 10.1098/rspb.2000.1166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fredrick K, Ibba M. How the sequence of a gene can tune its translation. Cell. 2010;141:227–229. doi: 10.1016/j.cell.2010.03.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Fujii Y, Kageyama D, Hoshizaki S, Ishikawa H, Sasaki T. Transfection of Wolbachia in Lepidoptera: the feminizer of the adzuki bean borer Ostrinia scapulalis causes male killing in the Mediterranean flour moth Ephestia kuehniella. Proceedings of the Royal Society of London. Series B. 2001;268:855–859. doi: 10.1098/rspb.2001.1593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. GenScript Codon Usage Frequency Table(chart) Tool. GenScript. 2020 https://www.genscript.com/tools/codon-frequency-table
  22. Goodman DB, Church GM, Kosuri S. Causes and effects of N-terminal codon bias in bacterial genes. Science. 2013;342:475–479. doi: 10.1126/science.1241934. [DOI] [PubMed] [Google Scholar]
  23. Groenenboom MAC, Hogeweg P. Space and the persistence of male-killing endosymbionts in insect populations. Proceedings. Biological Sciences. 2002;269:2509–2518. doi: 10.1098/rspb.2002.2197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Groth AC, Fish M, Nusse R, Calos MP. Construction of transgenic Drosophila by using the site-specific integrase from phage phiC31. Genetics. 2004;166:1775–1782. doi: 10.1534/genetics.166.4.1775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Gruber AR, Lorenz R, Bernhart SH, Neubock R, Hofacker IL. The vienna RNA websuite. Nucleic Acids Research. 2008;36:W70–W74. doi: 10.1093/nar/gkn188. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Gu W, Zhou T, Wilke CO, Snel B. A universal trend of reduced mRNA stability near the translation-initiation site in prokaryotes and eukaryotes. PLOS Computational Biology. 2010;6:e1000664. doi: 10.1371/journal.pcbi.1000664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Guindon S, Gascuel O. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology. 2003;52:696–704. doi: 10.1080/10635150390235520. [DOI] [PubMed] [Google Scholar]
  28. Gustafsson C, Govindarajan S, Minshull J. Codon bias and heterologous protein expression. Trends in Biotechnology. 2004;22:346–353. doi: 10.1016/j.tibtech.2004.04.006. [DOI] [PubMed] [Google Scholar]
  29. Hamm CA, Begun DJ, Vo A, Smith CCR, Saelao P, Shaver AO, Jaenike J, Turelli M. Wolbachia do not live by reproductive manipulation alone: infection polymorphism in Drosophila suzukii and D. subpulchrella. Molecular Ecology. 2014;23:4871–4885. doi: 10.1111/mec.12901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Harumoto T, Fukatsu T, Lemaitre B. Common and unique strategies of male killing evolved in two distinct Drosophila symbionts. Proceedings. Biological Sciences. 2018;285:20172167. doi: 10.1098/rspb.2017.2167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Harumoto T, Lemaitre B. Male-killing toxin in a bacterial symbiont of Drosophila. Nature. 2018;557:252–255. doi: 10.1038/s41586-018-0086-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Hoffmann AA. Partial cytoplasmic incompatibility between two Australian populations of Drosophila melanogaster. Entomologia Experimentalis et Applicata. 1988;48:61–67. doi: 10.1111/j.1570-7458.1988.tb02299.x. [DOI] [Google Scholar]
  33. Hornett E. A, Charlat S, Duplouy AMR, Davies N, Roderick GK, Wedell N, Hurst GDD. Evolution of male-killer suppression in a natural population. PLOS Biology. 2006;4:e283. doi: 10.1371/journal.pbio.0040283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Hornett EA, Duplouy AMR, Davies N, Roderick GK, Wedell N, Hurst GDD, Charlat S. You can’t keep a good parasite down: evolution of a male-killer suppressor uncovers cytoplasmic incompatibility. Evolution; International Journal of Organic Evolution. 2008;62:1258–1263. doi: 10.1111/j.1558-5646.2008.00353.x. [DOI] [PubMed] [Google Scholar]
  35. Huelsenbeck JP, Ronquist F. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics. 2001;17:754–755. doi: 10.1093/bioinformatics/17.8.754. [DOI] [PubMed] [Google Scholar]
  36. Hughes GL, Rasgon JL. Transinfection: a method to investigate Wolbachia-host interactions and control arthropod-borne disease. Sect Molecular Biology. 2014;23:141–151. doi: 10.1111/imb.12066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Hunter MS. The influence of parthenogenesis-inducing Wolbachia on the oviposition behaviour and sex-specific developmental requirements of autoparasitoid wasps. Journal of Evolutionary Biology. 1999;12:735–741. doi: 10.1046/j.1420-9101.1999.00082.x. [DOI] [Google Scholar]
  38. Hurst LD. The incidences and evolution of cytoplasmic male killers. Proceedings of the Royal Society of London. Series B. 1997;244:91–99. doi: 10.1098/rspb.1991.0056. [DOI] [Google Scholar]
  39. Hurst GDD, Jiggins FM, Hinrich Graf von der Schulenburg J, Bertrand D, West SA, Goriacheva II, Zakharov IA, Werren JH, Stouthamer R, Majerus MEN. Male–killing Wolbachia in two species of insect. Proceedings of the Royal Society of London. Series B. 1999;266:735–740. doi: 10.1098/rspb.1999.0698. [DOI] [Google Scholar]
  40. Hurst GD, Jiggins FM. Male-killing bacteria in insects: mechanisms, incidence, and implications. Emerging Infectious Diseases. 2000;6:329–336. doi: 10.3201/eid0604.000402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Hurst GD, Johnson AP, Schulenburg JH, Fuyama Y. Male-killing Wolbachia in Drosophila: a temperature-sensitive trait with a threshold bacterial density. Genetics. 2000;156:699–709. doi: 10.1093/genetics/156.2.699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Hurst GDD, Frost CL. Reproductive parasitism: maternally inherited symbionts in a biparental world. Cold Spring Harbor Perspectives in Biology. 2015;7:a017699. doi: 10.1101/cshperspect.a017699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Jaenike J, Dyer KA, Reed LK. Within-population structure of competition and the dynamics of male-killing Wolbachia. Evolutionary Ecology Research. 2003;5:1023–1036. [Google Scholar]
  44. Jaenike J. Spontaneous emergence of a new Wolbachia phenotype. Evolution; International Journal of Organic Evolution. 2007;61:2244–2252. doi: 10.1111/j.1558-5646.2007.00180.x. [DOI] [PubMed] [Google Scholar]
  45. Jiggins FM, Hurst GD, Majerus ME. Sex-ratio-distorting Wolbachia causes sex-role reversal in its butterfly host. Proceedings. Biological Sciences. 2000;267:69–73. doi: 10.1098/rspb.2000.0968. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Jukes TH, Cantor CR. Evolution of protein molecules. Mammal Prot Metabol. 1969;3:21–132. [Google Scholar]
  47. Knight RD, Freeland SJ, Landweber LF. A simple model based on mutation and selection explains trends in codon and amino-acid usage and GC composition within and across genomes. Genome Biology. 2001;2:research0010. doi: 10.1186/gb-2001-2-4-research0010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Kudla G, Murray AW, Tollervey D, Plotkin JB. Coding-sequence determinants of gene expression in Escherichia coli. Science. 2009;324:255–258. doi: 10.1126/science.1170160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. LePage DP, Metcalf JA, Bordenstein SR, On J, Perlmutter JI, Shropshire JD, Layton EM, Funkhouser-Jones LJ, Beckmann JF, Bordenstein SR. Prophage WO genes recapitulate and enhance Wolbachia-induced cytoplasmic incompatibility. Nature. 2017;543:243–247. doi: 10.1038/nature21391. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Li GW, Oh E, Weissman JS. The anti-Shine-Dalgarno sequence drives translational pausing and codon choice in bacteria. Nature. 2012;484:538–541. doi: 10.1038/nature10965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Lorenz R, Bernhart SH, Höner Zu Siederdissen C, Tafer H, Flamm C, Stadler PF, Hofacker IL. ViennaRNA Package 2.0. Algorithms for Molecular Biology. 2011;6:26. doi: 10.1186/1748-7188-6-26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Lucchesi JC, Kuroda MI. Dosage compensation in Drosophila. Cold Spring Harbor Perspectives in Biology. 2015;7:a019398. doi: 10.1101/cshperspect.a019398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Majerus TMO, Majerus MEN. Intergenomic arms races: detection of a nuclear rescue gene of male-killing in a ladybird. PLOS Pathogens. 2010;6:e1000987. doi: 10.1371/journal.ppat.1000987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Marín I, Baker BS. The evolutionary dynamics of sex determination. Science. 1998;281:1990–1994. doi: 10.1126/science.281.5385.1990. [DOI] [PubMed] [Google Scholar]
  55. Masui S, Kuroiwa H, Sasaki T, Inui M, Kuroiwa T, Ishikawa H. Bacteriophage WO and virus-like particles in Wolbachia, an endosymbiont of arthropods. Biochemical and Biophysical Research Communications. 2001;283:1099–1104. doi: 10.1006/bbrc.2001.4906. [DOI] [PubMed] [Google Scholar]
  56. McCutcheon JP, McDonald BR, Moran NA, Matic I. Origin of an alternative genetic code in the extremely small and GC–rich genome of a bacterial symbiont. PLOS Genetics. 2009;5:e1000565. doi: 10.1371/journal.pgen.1000565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Metcalf JA, Jo M, Bordenstein SR, Jaenike J, Bordenstein SR. Recent genome reduction of Wolbachia in Drosophila recens targets phage WO and narrows candidates for reproductive parasitism. PeerJ. 2014;2:e529. doi: 10.7717/peerj.529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Mitsuhashi W, Ikeda H, Muraji M. Fifty-year trend towards suppression of Wolbachia-induced male-killing by its butterfly host, Hypolimnas bolina. Journal of Insect Science. 2011;11:92. doi: 10.1673/031.011.9201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Perlmutter JI, Bordenstein SR, Unckless RL, LePage DP, Metcalf JA, Hill T, Martinez J, Jiggins FM, Bordenstein SR. The phage gene wmk is a candidate for male killing by a bacterial endosymbiont. PLOS Pathogens. 2019;15:e1007936. doi: 10.1371/journal.ppat.1007936. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Perlmutter JI, Meyers JE, Bordenstein SR. Transgenic testing does not support a role for additional candidate genes in Wolbachia male killing or cytoplasmic incompatibility. MSystems. 2020;5:e00658. doi: 10.1128/mSystems.00658-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Plotkin JB, Kudla G. Synonymous but not the same: the causes and consequences of codon bias. Nature Reviews. Genetics. 2011;12:32–42. doi: 10.1038/nrg2899. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Qin H, Wu WB, Comeron JM, Kreitman M, Li WH. Intragenic spatial patterns of codon usage bias in prokaryotic and eukaryotic genomes. Genetics. 2004;168:2245–2260. doi: 10.1534/genetics.104.030866. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Riparbelli MG, Giordano R, Ueyama M, Callaini G. Wolbachia-mediated male killing is associated with defective chromatin remodeling. PLOS ONE. 2012;7:e30045. doi: 10.1371/journal.pone.0030045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Rodriguez MA, Vermaak D, Bayes JJ, Malik HS. Species-specific positive selection of the male-specific lethal complex that participates in dosage compensation in Drosophila. PNAS. 2007;104:15412–15417. doi: 10.1073/pnas.0707445104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003;19:1572–1574. doi: 10.1093/bioinformatics/btg180. [DOI] [PubMed] [Google Scholar]
  66. Salari R, Kimchi-Sarfaty C, Gottesman MM, Przytycka TM. Sensitive measurement of single-nucleotide polymorphism-induced changes of RNA conformation: application to disease studies. Nucleic Acids Research. 2013;41:44–53. doi: 10.1093/nar/gks1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Sasaki T, Kubo T, Ishikawa H. Interspecific transfer of Wolbachia between two lepidopteran insects expressing cytoplasmic incompatibility: a Wolbachia variant naturally infecting Cadra cautella causes male killing in Ephestia kuehniella. Genetics. 2002;162:1313–1319. doi: 10.1093/genetics/162.3.1313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Sauna ZE, Kimchi-Sarfaty C. Understanding the contribution of synonymous mutations to human disease. Nature Reviews Genetics. 2011;12:683–691. doi: 10.1038/nrg3051. [DOI] [PubMed] [Google Scholar]
  69. Sharp PM, Averof M, Lloyd AT, Matassi G, Peden JF. DNA sequence evolution: the sounds of silence. Philosophical Transactions of the Royal Society of London. Series B. 1997;349:241–247. doi: 10.1098/rstb.1995.0108. [DOI] [PubMed] [Google Scholar]
  70. Sheeley SL, McAllister BF. Mobile male-killer: similar Wolbachia strains kill males of divergent Drosophila hosts. Heredity. 2009;102:286–292. doi: 10.1038/hdy.2008.126. [DOI] [PubMed] [Google Scholar]
  71. Skinner SW. Son-killer: a third extrachromosomal factor affecting the sex ratio in the parasitoid wasp, Nasonia (=Mormoniella) vitripennis. Genetics. 1985;109:745–759. doi: 10.1093/genetics/109.4.745. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Southall TD, Elliott DA, Brand AH. The GAL4 system: A versatile toolkit for gene expression in Drosophila. Cold Spring Harbor Protocols. 2008;2008:top49. doi: 10.1101/pdb.top49. [DOI] [PubMed] [Google Scholar]
  73. Taylor MJ, Bordenstein SR, Slatko B. Microbe Profile: Wolbachia: a sex selector, a viral protector and a target to treat filarial nematodes. Microbiology. 2018;164:1345–1347. doi: 10.1099/mic.0.000724. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Tuller T, Carmi A, Vestsigian K, Navon S, Dorfan Y, Zaborske J, Pan T, Dahan O, Furman I, Pilpel Y. An evolutionarily conserved mechanism for controlling the efficiency of protein translation. Cell. 2010;141:344–354. doi: 10.1016/j.cell.2010.03.031. [DOI] [PubMed] [Google Scholar]
  75. Unckless RL, Jaenike J. Maintenance of a male-killing Wolbachia in Drosophila innubila by male-killing dependent and male-killing independent mechanisms. Evolution; International Journal of Organic Evolution. 2012;66:678–689. doi: 10.1111/j.1558-5646.2011.01485.x. [DOI] [PubMed] [Google Scholar]
  76. Van Borm S, Wenseleers T, Billen J, Boomsma JJ. Wolbachia in leafcutter ants: a widespread symbiont that may induce male killing or incompatible matings. Journal of Evolutionary Biology. 2008;14:805–814. doi: 10.1046/j.1420-9101.2001.00321.x. [DOI] [Google Scholar]
  77. Veneti Z, Zabalou S, Papafotiou G, Paraskevopoulos C, Pattas S, Livadaras I, Markakis G, Herren JK, Jaenike J, Bourtzis K. Loss of reproductive parasitism following transfer of male-killing Wolbachia to Drosophila melanogaster and Drosophila simulans. Heredity. 2012;109:306–312. doi: 10.1038/hdy.2012.43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Wu M, Sun LV, Vamathevan J, Riegler M, Deboy R, Brownlie JC, McGraw EA, Martin W, Esser C, Ahmadinejad N, Wiegand C, Madupu R, Beanan MJ, Brinkac LM, Daugherty SC, Durkin AS, Kolonay JF, Nelson WC, Mohamoud Y, Lee P, Berry K, Young MB, Utterback T, Weidman J, Nierman WC, Paulsen IT, Nelson KE, Tettelin H, O’Neill SL, Eisen JA. Phylogenomics of the reproductive parasite Wolbachia pipientis wMel: a streamlined genome overrun by mobile genetic elements. PLOS Biology. 2004;2:E69. doi: 10.1371/journal.pbio.0020069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  79. Yen JH, Barr AR. New hypothesis of the cause of cytoplasmic incompatibility in Culex pipiens L. Nature. 1971;232:657–658. doi: 10.1038/232657a0. [DOI] [PubMed] [Google Scholar]
  80. Zeh DW, Zeh JA, Bonilla MM. Wolbachia, sex ratio bias and apparent male killing in the harlequin beetle riding pseudoscorpion. Heredity. 2005;95:41–49. doi: 10.1038/sj.hdy.6800666. [DOI] [PubMed] [Google Scholar]

Editor's evaluation

Dieter Ebert 1

This study identified the genetic mechanisms underlying sex-ratio distortion through male-killing in Drosophila melanogaster flies infected with the endosymbiont Wolbachia. The endosymbiont carries the prophage WO, which is the center of interest in this study. The key result of this study is that a synonymous mutation in a prophage gene can explain the differences between sex-ratio distorting and not distorting symbionts. The finding, that a synonymous SNP plays a key role is not entirely novel in biology, but there are only few examples known of this type of genotype–phenotype association.

Decision letter

Editor: Dieter Ebert1

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Decision letter after peer review:

Thank you for submitting your article "A single synonymous nucleotide change impacts the male-killing phenotype of prophage WO gene wmk" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, one of whom is member of our Board of Reviewing Editors, and the evaluation has been overseen by Patricia Wittkopp as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

1) The most important comment from the review is that it is plausible that the observed changes in the effect and strength of killing are due to an interaction between host and wmk genotype. This has implications for unravelling the underlying genetic basis to the male-killing phenotype more widely. Therefore, it will be critical to repeat some of the main findings in other D. melanogaster genotypes to determine the importance of the variation in the wmk homologs more generally.

2) The manuscript requires extensive streamlining. The text is hard to follow and has too many details. The main points are often hidden among details. A more focussed and shortened manuscript would be highly welcome.

Reviewer #1:

This study aims to find the genetic mechanisms underlying sex-ratio distortion through male-killing in Drosophila melanogaster flies infected with the endosymbiont Wolbachia. The endosymbiont carries the prophage WO, which is in the center of interested in this study. The key result of this study is that a synonymous mutation in a prophage gene can explain the differences between sex-ratio distorting and not distorting symbionts. The study uses transgene technology to modify phage genes and to investigate which changes in the gene is involved in the phenotype. The finding, that a synonymous SNP plays a key role is not entirely novel in biology, but there are only few examples known of this type of genotype–phenotype association. The study does not include experiments to show that the main finding is not limited to one particular background of the fly line used. An experiment including multiple genotypes would be needed to show this.

The study is mostly clear and easy to follow, but requires a lot of attention. The authors choose to build up the story as I guess it was carried out in the lab. Thus, the reader is guided through every step of the process. While I see that this is appealing from the way the study was carried out, it results in a very long manuscript with a lot of material that would be much better placed in a supplement.

The introduction seems unfocused. It meanders around, jumping from topic to topic and does not give the reader a sense of where things will go. Figure 1 gives an overview about the different aspects addressed here, but it is not used to guide the reader through the different lines of thought addressed in the introduction. If Figure 1 will stay (I actually think it is not needed) it should be introduced earlier and used as a road map for the paper. Alternatively, the introduction could stay more general and only in the last paragraph the different ways the system is studied will be summarized. Along these lines, it would be good to have a better reasoning for the combination of experiments conducted. It is left to the reader to understand why certain types of experiments have been done. On the other hand, the introduction misses a section on the biology of the phage and its interaction with the host(s). It is hard to understand the biology of the system without getting an understanding of the insect – Wolbachia – phage interactions. For non-specialist, understanding the role of the three players is essential for the system.

The result section could be easily shortened by focusing on the essential experiments. Experiments that do not contribute to the final result can go into the supplement.

Also the discussion is much too long. I suggest to reduce it to half and focus on the important points and the take-home messages. Currently the discussion follows the way the results are presented in the result section. However, this is not needed. The important finding should be discussed first. Findings that are important in the development of the project, may not be important for the biology of the system overall. And they may not be important for the reader.

The text contains many abbreviations. For less specialist readers this become rather difficult. I suggest to reduce abbreviations and to add a list with them, so one can find them fast. All abbreviations have to be explained when they are used for the first time. Furthermore, abbreviations are used in different forms making part of the text or figures messy. As an example, look at Figure 4: going plot by plot from A to E the names of the treatments changes from plot to plot. This is issues persists throughout text and figures of the entire manuscript.

On line 158-160 it is said that wSuzi differs only by one SNP from wMel. On lines 162-164 it is then said that the transgenes were codon optimized. Why is this necessary, if only 1 position differs? More generally: " codon-optimization " plays an important role in this study. Please explain why codon-optimization was done and what was exactly done.

For several experiment it is said "The experiment has been performed twice". How where the data treated? The stats described do not allow a complex design with repeated experiment. Did you pool the data of the two experiments? Did you test if the outcome of the experiments differed?

Line 241: "Thus, we conclude that small or large wmk transcript sequence changes can lead to alterations in predicted RNA structure or possibly protein structure that may relate to altered phenotypes in wmk homologs." This is a very vague conclusion. It basically says nothing.

For the plots of the individual data points a jitter function was used (please say so!). Was the function used to jitter in both dimensions or only in the X-axis dimension?

Animals were kept on a standard media. Please give references when you make such statements.

The methods section regarding the transgene constructs is rather short. I would like to see more information to understand what was done, why it was done and how it was done.

Reviewer #2:

This study aims to unravel the genomic basis to wmk-induced male killing by transgenically expressing homologs of varying relatedness, with synonymous nucleotide changes, and predicted alternative start codons in D. melanogaster flies. The study builds on previous work showing that expression of wmk in fly embryos recapitulates several aspects of male killing. While more distantly-related homologs did not induce male killing when expressed in D. melanogaster, more closely-related wmk homologs induce either killing of both sexes or male killing only. However, the male-killing phenotype was not due to amino acid differences, but associated with RNA structural differences of the different wmk homologs. In addition, only one synonymous nucleotide change was sufficient to ablate the killing phenotype. These findings suggests that minor and even silent nucleotide differences impact on the expression of male killing in D. melanogaster. It is concluded that a new model incorporating the impacts of RNA structure and post-transcriptional processes in wmk-induced male killing needs to be developed.

The strength of the study lies in the systematic and carefully controlled approach to quantify the phenotypic effects of both sequence and structural changes to various wmk homologs for inducing the male-killing phenotype. Detailed dissection of the phenotypic impact of minor changes to the wmk homologs including sequence variation, silent nucleotide changes, and RNA structural differences was quantified. This approach reveals a complex genotype-phenotype relationship, but highlights the importance of including post-translational processes. The data is novel in that previous work have largely ignored structural changes and assumed that synonymous differences in codons has no effect on protein function, whereas the current study based on updated codon optimization algorithms reveal that this assumption is incorrect. The finding highlights the importance of considering also structural genetic variation for phenotypic expression differences. This suggestion is further corroborated by the lack of difference in wmk homologue expression levels, indicating that the functional differences are due to post-translational effects.

There are limitations to the findings of this complex genotype-phenotype relationship. The current study only examined the phenotypic impact by expressing the different homologs in one D. melanogaster genetic background. Given the variability of the phenotypic pattern revealed based on minor changes to the wmk homologs, it will be critical to repeat some of the main findings in other D. melanogaster genotypes to determine the importance of the variation in the wmk homologs more generally. It is entirely plausible that the observed changes in the effect and strength of killing is due to an interaction between host and wmk genotype. This has implications for unravelling the underlying genetic basis to the male-killing phenotype more widely. It is as yet to be demonstrated whether wmk is involved in male killing in natural population, and to what extent there are shared patterns and mechanisms of male killing induced by other bacterial endosymbionts such as Spiroplasma.

I found the study of interest and enjoyed reading this manuscript. Overall, it represents a careful dissection of the functional genomic basis of the wmk candidate gene for the male-killing phenotype in D. melanogaster. The experimental design and choice of the putative male-killing wmk homologs of varying relatedness provides an opportunity to examine the importance of genetic similarity, in conjunction with wmk-variants with synonymous nucleotide changes, and predicted alternative start codons for the male killing expression. The complex pattern of genotype-phenotype relationship that is unravelled is perhaps unexpected, but possibly more important is the finding that RNA structural differences are associated with male killing. This finding has several implications such as the need to consider the importance of post-translational processes, but also that the assumption that synonymous differences in codons has no effect on protein function is erroneous. On the whole these complex patterns suggest that we need to re-visit the model for the genotype-phenotype relationship of male killing. As such, the study makes a valuable contribution to the field.

However, the study also has some shortcomings and limitations. Importantly, the impact of the various wmk homologs were only examined in one D. melanogaster genotype. Given the variability of the phenotypic pattern based on minor changes to the wmk homologs that was unravelled, it will be critical to repeat the main findings in other D. melanogaster genotypes to determine the importance of the variation in the wmk homologs for male killing irrespective of host genetic background. The suggestion that caution need to be taken when examining sequence variation and in particular synonymous nucleotide changes as their impact may be dependent on RNA structure is also based on few observations.

In general, little consideration is made to the evolutionary implications of the finding. It is currently unclear what patterns are to be expected in other host-gene combinations and as such the findings have perhaps low predictive power. It is entirely plausible that the observed changes in the effect and strength of killing is due to the interaction between host and wmk genotype shaped by co-evolution. This caveat has implications for understanding the underlying genetic basis to male-killing more widely, especially since one of the aims of the study was to address the genotype-phenotype relationship that underpins wmk function. Here the wmk genotype is varied while keeping the host genotype constant. This inevitably limit the generality of the findings, especially as small wmk differences were associated with big phenotypic differences. More work is therefore required to address the genotype-phenotype relationship by expressing some of the wmk homologs across different D. melanogaster genotypes.

Overall, while I found the discussion interesting, it is currently highly speculative and perhaps some of the suggestions that are based on limited amount of supporting data could be toned down. For example, the finding that only one synonymous nucleotide change was sufficient to remove the killing phenotype is intriguing. However, it is not clear if the suggestion made is that resistance to male killing is easily achieved by minor changes to wmk that are functionally dependent on codon biases or if minor changes in general can have large phenotypic effects. If correctly interpreted, then more D. melanogaster genotypes need to be assessed to verify this suggestion, and if resistance is achieved this way, then outlining some specific predictions about the expected variation in resistance strains (i.e. non-synonymous or structural) would be valuable.

It is unclear how robust the updated codon optimization algorithms are regarding how accurately they predict protein structure and function. It is noticeable that significant differences were found using the updated codon optimization algorithm compared to the older one from four years ago. How certain can we be that the new algorithm better reflect protein structure and function as it is rightly stated that the codon optimization may change the interpretation of results from transgenic experiments. The caution raised is valid, but perhaps it would be valuable to provide some suggestion regarding how to best mitigate this potential issue.

Finally, it is yet to be demonstrated whether wmk is involved in male killing also in natural population, and to what extent there are shared patterns and mechanisms of male killing induced by other bacterial endosymbionts such as Spiroplasma.

eLife. 2021 Oct 22;10:e67686. doi: 10.7554/eLife.67686.sa2

Author response

Essential revisions:

1) The most important comment from the review is that it is plausible that the observed changes in the effect and strength of killing are due to an interaction between host and wmk genotype. This has implications for unravelling the underlying genetic basis to the male-killing phenotype more widely. Therefore, it will be critical to repeat some of the main findings in other D. melanogaster genotypes to determine the importance of the variation in the wmk homologs more generally.

We understand the importance of this point and address it in more detail below in the response to the first comment by Reviewer 1. We would also like to thank the Editor and Reviewers for considering and acknowledging the request to not require this experiment for the reasons outlined below. Per the follow-up recommendation, we now acknowledge the limitations of examining one genetic background in the discussion and altered language to avoid perceptions of generalizability. In individual responses to reviewer comments below, we included some more detailed explanation and example text.

2) The manuscript requires extensive streamlining. The text is hard to follow and has too many details. The main points are often hidden among details. A more focussed and shortened manuscript would be highly welcome.

Thank you for this feedback. We agree and removed unnecessary elements in various sections, moved some parts of the results to the supplement, removed any part of the results that were redundant with other parts of the paper, and reordered and streamlined the discussion as well. The discussion is now a few pages shorter, as are the results. Specific changes are outlined where reviewers suggested them below.

Reviewer #1:

This study aims to find the genetic mechanisms underlying sex-ratio distortion through male-killing in Drosophila melanogaster flies infected with the endosymbiont Wolbachia. The endosymbiont carries the prophage WO, which is in the center of interested in this study. The key result of this study is that a synonymous mutation in a prophage gene can explain the differences between sex-ratio distorting and not distorting symbionts. The study uses transgene technology to modify phage genes and to investigate which changes in the gene is involved in the phenotype. The finding, that a synonymous SNP plays a key role is not entirely novel in biology, but there are only few examples known of this type of genotype –phenotype association. The study does not include experiments to show that the main finding is not limited to one particular background of the fly line used. An experiment including multiple genotypes would be needed to show this.

We agree that recapitulating the results in other backgrounds is intriguing and important for establishing a broader role of these findings. We thank the Reviewers and Editor for allowing us to pursue this line of investigation separately from this work, and we now discuss what experiments can be completed to answer these and other questions. We also edited the manuscript to tone down any conclusions that would imply generalizability of the findings at this point. For example:

“For example, we cannot conclude that the particular codon tested here is responsible for phenotype alterations in other host genetic backgrounds or species. It is possible that this codon plays a functional role only in a singular host genetic context. Here, we changed wmk sequences while holding the host genetic background fixed, but the reverse is required to conclude whether or not the particular codon plays a general role in other genotypes or natural contexts. Second, due to possible coevolution, various codons may or may not yield similar functional effects across different host backgrounds, and additional synonymous sites may contribute to the male-killing phenotype. Thus, the results here illuminate a previously unrecognized need for future research on the functional impacts of synonymous substitutions in endosymbionts. Future work may focus on determining if there is one specific synonymous codon that affects the male-killing function in all cases, if a more general feature exists where alteration of any or a subset of N-terminal or other wmk codons affects function, or if the effect of synonymous changes is specific to this background.”

Text summarizing the 06/21/2021 query to the Editor and Reviewers for further clarification: We believe there are several reasons why the results can stand on their own, while appropriately acknowledging caveats. First, we note the lack of genetic background testing on previous transgene experiments driving the major discoveries of Wolbachia genes involved in reproductive parasitism. This requirement would therefore hold the current work to a novel bar not previously applied by the field. In addition, the genetic background here is the same as used in previous work on these phenotypes, making it the most pertinent to test and inform previous and ongoing studies by many research groups. Second, the results shown here would still stand no matter the results of genetic background testing and would demonstrate that it is possible for synonymous changes to have functional relevance in the transgenic wmk phenotype. The major findings are still novel in the field, relevant to ongoing studies of reproductive parasitism, and informative regarding one of the most common genetic backgrounds. Finally, we note that two different lines with unique synonymous codon changes (the final experiment) independently created the same result that a synonymous codon change ablates phenotype, providing additional robustness to our findings. Doing additional experiments would be logistically difficult. Barriers include the relocation of the first author of the work to another lab for a postdoctoral position, completion of the funding for the project, remaining institutional COVID-19 restrictions, and lack of replacement personnel in the lab to continue the work. Notably, there is also the non-trivial requirement to create and test new transgene lines that would be costly and take nearly a year to complete (the experiments in the manuscript already took several years and the new fly lines would cost thousands to make).

The study is mostly clear and easy to follow, but requires a lot of attention. The authors choose to build up the story as I guess it was carried out in the lab. Thus, the reader is guided through every step of the process. While I see that this is appealing from the way the study was carried out, it results in a very long manuscript with a lot of material that would be much better placed in a supplement.

We thank the reviewer for pointing this out. We shortened the manuscript by removing redundant information and transferring some parts of the results to the supplement. We also removed about three pages of text from the discussion (before adding in new sections as requested by reviewers).

The introduction seems unfocused. It meanders around, jumping from topic to topic and does not give the reader a sense of where things will go.

We added a few topics into the Introduction as recommended in other comments, and we edited various portions of the Introduction to connect the ideas together more clearly. We hope the changes are now satisfactory, and we are of course happy to consider further feedback.

Figure 1 gives an overview about the different aspects addressed here, but it is not used to guide the reader through the different lines of thought addressed in the introduction. If Figure 1 will stay (I actually think it is not needed) it should be introduced earlier and used as a road map for the paper. Alternatively, the introduction could stay more general and only in the last paragraph the different ways the system is studied will be summarized.

We edited the final paragraph of the Introduction to more comprehensively cover the content of the figure and full direction of the paper. For readers not familiar with the biological system or questions, we believe this figure will serve as a gateway to the genetic alterations conducted in the experiments.

Along these lines, it would be good to have a better reasoning for the combination of experiments conducted. It is left to the reader to understand why certain types of experiments have been done.

It was not clear to us at the outset of these experiments what results would ultimately emerge and what follow-up experiments would be necessary as our initial hypotheses were proven wrong with many of the surprises from the work. So, there was no a priori reasoning for why experiments were done until we had the results of the previous experiments. We agree that this makes the reading a bit confusing. As such, we clarified the logic flow in the Results section as the narrative progresses from experiment to experiment, and we reorganized some of the introduction to improve transition statements and offer a roadmap to readers earlier on.

On the other hand, the introduction misses a section on the biology of the phage and its interaction with the host(s). It is hard to understand the biology of the system without getting an understanding of the insect – Wolbachia – phage interactions. For non-specialist, understanding the role of the three players is essential for the system.

Thank you for the suggestion. We now add a section introducing phage WO and its relevance to the phenotypes tested here.

“The wmk gene and two cytoplasmic incompatibility factor (cif) genes that underlie cytoplasmic incompatibility (a parasitism phenotype whereby offspring die in crosses between infected males and uninfected females) occur in the eukaryotic association module (EAM) of prophage WO, which refers to the phage WO genome that is inserted into the bacterial chromosome. The EAM is common in WO phages across several Wolbachia strains and is rich in genes that are homologous to eukaryotic genes or annotated with eukaryotic functions. As such, the expression of reproductive parasitism genes from the EAM and tripartite interactions between phage WO, Wolbachia, and eukaryotic hosts are central to Wolbachia’s ability to interact with and modify host reproduction.”

The result section could be easily shortened by focusing on the essential experiments. Experiments that do not contribute to the final result can go into the supplement.

We removed redundant sentences and made some figures supplemental.

Also the discussion is much too long. I suggest to reduce it to half and focus on the important points and the take-home messages. Currently the discussion follows the way the results are presented in the result section. However, this is not needed. The important finding should be discussed first. Findings that are important in the development of the project, may not be important for the biology of the system overall. And they may not be important for the reader.

We reordered the discussion to cover the biggest findings first, and removed about a third of the original writing in the discussion.

The text contains many abbreviations. For less specialist readers this become rather difficult. I suggest to reduce abbreviations and to add a list with them, so one can find them fast. All abbreviations have to be explained when they are used for the first time. Furthermore, abbreviations are used in different forms making part of the text or figures messy. As an example, look at Figure 4: going plot by plot from A to E the names of the treatments changes from plot to plot. This is issues persists throughout text and figures of the entire manuscript.

Thank you for pointing this out. In the manuscript, we used labels such as “wMel” or “wMel wmk” synonymously. Since that resulted in too many labels and acronyms to follow and sift through, we unified the terminology throughout by ensuring that the same label for each genotype remains constant in the text, and we edited the figures to ensure that the terminology is kept constant. For example, we now refer to the original homolog exclusively as “wMel wmk” as opposed to using this term in some cases and “wmk” in others.

On line 158-160 it is said that wSuzi differs only by one SNP from wMel. On lines 162-164 it is then said that the transgenes were codon optimized. Why is this necessary, if only 1 position differs? More generally: " codon-optimization " plays an important role in this study. Please explain why codon-optimization was done and what was exactly done.

We added a short explanation to the first figure to explain the codon optimization to readers, but here is an additional, expanded explanation for reference: Codon-optimization is standard for transgenic expression analyses since the DNA sequence comes from an organism (bacteria) that is divergent compared to its destination organism (flies). For synonymous codons, although all code for the same amino acid, each organism has a specific bias for which tRNAs are more or less abundant. Synonymous codon differences can result in variations in RNA folding (based on subtle molecular differences in the codons) as well as differences in translation such as translation rate (based on codon availability in terms of rarity). Expressing a bacterial gene in a fly host without optimizing the codons would mean that the gene would likely contain many codons that are not preferred/as common in the fly host, and this could result in altered translation rates or even early termination when the fly expresses the gene. For this reason, we and others utilizing transgenics take all the Wolbachia sequences and put them through an algorithm that gives the same protein sequence, but with the best codon combinations for optimal fly expression. This new “codon-optimized” sequence is then utilized to synthesize the gene anew for insertion into the fly host.

In the specific lines referenced here, we clarify that while the bacterial sequences differ by a single SNP, they must be codon optimized for fly expression for the reasons explained above, which then of course changes the DNA sequence. Since this is an important principle that we want readers to clearly understand before going into the results, we added a new section to the Introduction to introduce the concept of codon optimization and transgenes. We also expanded the Discussion of this topic as well. The Introduction change is noted below:

“These sequences are codon-optimized based on different codon biases due to different tRNA abundances in the divergent bacterial source and eukaryotic destination species.”

For several experiment it is said "The experiment has been performed twice". How where the data treated? The stats described do not allow a complex design with repeated experiment. Did you pool the data of the two experiments? Did you test if the outcome of the experiments differed?

In each case, experiments were done twice to confirm that the results were repeatable. We applied the same statistical tests to each individual experiment, and we determined that they both showed the same results. However, the data from each individual experiment were not combined. This is because exact y-values often differ from experiment to experiment due to small, uncontrollable environmental factors, even if relative differences between treatment groups are repeatable. This is typical of Drosophila egg-laying experiments, where the average number of eggs laid naturally varies within a wide range due to sensitivity to environmental factors and gives minor differences in results. The large sample size and sample size cutoff values were chosen because they largely correct for these small variations and give consistent results relative to groups within each experiment. However, there is still some variation in absolute values from experiment to experiment, so we did not combine data. For the data in the manuscript, all plots were from the first iteration of the experiment as a representative result of the two. Brief statements on this have been added to the methods section.

Line 241: "Thus, we conclude that small or large wmk transcript sequence changes can lead to alterations in predicted RNA structure or possibly protein structure that may relate to altered phenotypes in wmk homologs." This is a very vague conclusion. It basically says nothing.

The statement has been deleted.

For the plots of the individual data points a jitter function was used (please say so!). Was the function used to jitter in both dimensions or only in the X-axis dimension?

We graphed these plots using GraphPad Prism software, which does apply a function to horizontally distribute points at the same y-value evenly across for greater visibility of data distribution. We have now included a statement regarding this in the methods, including which particular function was used.

Animals were kept on a standard media. Please give references when you make such statements.

We used standard CMY media, but recipes may vary slightly lab to lab. For clarity, we added details of our specific recipe in the methods section.

The methods section regarding the transgene constructs is rather short. I would like to see more information to understand what was done, why it was done and how it was done.

We expanded the Methods section to include additional details on making Drosophila transformants so it can be repeated. The text is copied here below:

“D. melanogaster strains used in this study include Act5c-Gal4/CyO (BDSC 3953, ubiquitously-expressing zygotic driver), the WT background line of genotype y1w67c23; P[CaryP]P2 (BDSC 8622), the WD0626 (wmk) and WD0034 (control gene) transgene constructs previously described25, and several new transgene constructs. […] The plasmids were then sent to BestGene (Chino Hills, CA), which performed injections of the vectors into embryos of the BDSC 8622 background line with PhiC31 integrase to integrate the vector into the attP2 insertion site in the genome. Successful transformants were selected based on the red eye marker, and each transgene line is descended from the offspring of a single transformant (isofemale).”

Reviewer #2:

[…] There are limitations to the findings of this complex genotype-phenotype relationship. The current study only examined the phenotypic impact by expressing the different homologs in one D. melanogaster genetic background. Given the variability of the phenotypic pattern revealed based on minor changes to the wmk homologs, it will be critical to repeat some of the main findings in other D. melanogaster genotypes to determine the importance of the variation in the wmk homologs more generally. It is entirely plausible that the observed changes in the effect and strength of killing is due to an interaction between host and wmk genotype. This has implications for unravelling the underlying genetic basis to the male-killing phenotype more widely. It is as yet to be demonstrated whether wmk is involved in male killing in natural population, and to what extent there are shared patterns and mechanisms of male killing induced by other bacterial endosymbionts such as Spiroplasma.

We addressed this point in more detail above in the first response to the comments from Reviewer 1.

I found the study of interest and enjoyed reading this manuscript. Overall, it represents a careful dissection of the functional genomic basis of the wmk candidate gene for the male-killing phenotype in D. melanogaster. The experimental design and choice of the putative male-killing wmk homologs of varying relatedness provides an opportunity to examine the importance of genetic similarity, in conjunction with wmk-variants with synonymous nucleotide changes, and predicted alternative start codons for the male killing expression. The complex pattern of genotype-phenotype relationship that is unravelled is perhaps unexpected, but possibly more important is the finding that RNA structural differences are associated with male killing. This finding has several implications such as the need to consider the importance of post-translational processes, but also that the assumption that synonymous differences in codons has no effect on protein function is erroneous. On the whole these complex patterns suggest that we need to re-visit the model for the genotype-phenotype relationship of male killing. As such, the study makes a valuable contribution to the field.

However, the study also has some shortcomings and limitations. Importantly, the impact of the various wmk homologs were only examined in one D. melanogaster genotype. Given the variability of the phenotypic pattern based on minor changes to the wmk homologs that was unravelled, it will be critical to repeat the main findings in other D. melanogaster genotypes to determine the importance of the variation in the wmk homologs for male killing irrespective of host genetic background. The suggestion that caution need to be taken when examining sequence variation and in particular synonymous nucleotide changes as their impact may be dependent on RNA structure is also based on few observations.

We addressed this point in more detail above in the first response to the comments from Reviewer 1.

In general, little consideration is made to the evolutionary implications of the finding. It is currently unclear what patterns are to be expected in other host-gene combinations and as such the findings have perhaps low predictive power. It is entirely plausible that the observed changes in the effect and strength of killing is due to the interaction between host and wmk genotype shaped by co-evolution. This caveat has implications for understanding the underlying genetic basis to male-killing more widely, especially since one of the aims of the study was to address the genotype-phenotype relationship that underpins wmk function. Here the wmk genotype is varied while keeping the host genotype constant. This inevitably limit the generality of the findings, especially as small wmk differences were associated with big phenotypic differences. More work is therefore required to address the genotype-phenotype relationship by expressing some of the wmk homologs across different D. melanogaster genotypes.

We edited the Discussion to reflect the reviewer’s concern, such as the statement below:

“Third and critically, although the results show that there is some relationship between synonymous codons and phenotype, several points remain for further testing. […] Here, we changed wmk sequences while holding the host genetic background fixed, but the reverse is required to conclude whether or not the particular codon plays a general role in other genotypes or natural contexts."

Overall, while I found the discussion interesting, it is currently highly speculative and perhaps some of the suggestions that are based on limited amount of supporting data could be toned down. For example, the finding that only one synonymous nucleotide change was sufficient to remove the killing phenotype is intriguing. However, it is not clear if the suggestion made is that resistance to male killing is easily achieved by minor changes to wmk that are functionally dependent on codon biases or if minor changes in general can have large phenotypic effects. If correctly interpreted, then more D. melanogaster genotypes need to be assessed to verify this suggestion, and if resistance is achieved this way, then outlining some specific predictions about the expected variation in resistance strains (i.e. non-synonymous or structural) would be valuable.

We edited the Discussion to include clearer caveats and statements on the limitations of our study, as well as what remains to be determined in future work. An example is included below.

“Second, due to possible coevolution, various codons may or may not yield similar functional effects across different host backgrounds, and additional synonymous sites may contribute to the male-killing phenotype. […] Future work may focus on determining if there is one specific synonymous codon that affects the male-killing function in all cases, if a more general feature exists where alteration of any or a subset of N-terminal or other wmk codons affects function, or if the effect of synonymous changes is specific to this background.”

It is unclear how robust the updated codon optimization algorithms are regarding how accurately they predict protein structure and function. It is noticeable that significant differences were found using the updated codon optimization algorithm compared to the older one from four years ago. How certain can we be that the new algorithm better reflect protein structure and function as it is rightly stated that the codon optimization may change the interpretation of results from transgenic experiments. The caution raised is valid, but perhaps it would be valuable to provide some suggestion regarding how to best mitigate this potential issue.

We agree that it is not clear which algorithm is ideal or best reflective of the biology of the systems tested here. We expanded how this kind of work can be approached in the future in the Discussion.

“In addition, codon optimization algorithms are updated with new information periodically with the assumption that they yield improved results, although it is unclear in practice whether an algorithm is better optimized to produce results that reflect the true biology of a transgene. […] This approach could ensure the algorithm produces a transgenic phenotype that most closely resembles the natural phenotype.”

Finally, it is yet to be demonstrated whether wmk is involved in male killing also in natural population, and to what extent there are shared patterns and mechanisms of male killing induced by other bacterial endosymbionts such as Spiroplasma.

This is true and we note relevant text in the Discussion. We aimed to keep this emphasis in the new, streamlined Discussion by bringing it up several times.

“First, as discussed, the wmk transgenic phenotypes are likely sensitive to post-transcriptional processes. This has important implications for understanding wmk in a natural male-killing context as soon as feasible techniques are available, since heterologous expression and its reliance on codon optimization may obscure our understanding of the gene’s biology.”

“However, if codon optimization potentially changes the interpretation of transgenic findings, then phenotypes should be corroborated in natural contexts once tools such as genetic editing are available in the relevant organism. Specifically, wmk should be knocked out in native contexts once it is more technically feasible.”

“If we make the assumption that wmk is involved in male killing in nature, which requires confirmation beyond transgenic recapitulation of the phenotype, then the results here give the basis for additional hypotheses that require further testing.”

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    Figure 2—source data 1. Data for sex ratios of closely-related homologs corresponding to Figure 2B.
    elife-67686-fig2-data1.xlsx (18.8KB, xlsx)
    Figure 2—source data 2. Statistical output of Kruskal-Wallis test corresponding to sex ratios of closelyrelated homologs in Figure 2B.
    elife-67686-fig2-data2.xlsx (10.5KB, xlsx)
    Figure 2—source data 3. Data for qPCR of closely-related transgenes corresponding to Figure 2C.
    elife-67686-fig2-data3.xlsx (11KB, xlsx)
    Figure 2—source data 4. Statistical output of Kruskal-Wallis test corresponding to qPCR for transgene expression in Figure 2C.
    elife-67686-fig2-data4.xlsx (9.7KB, xlsx)
    Figure 2—source data 5. Data for qPCR of msl-2 expression with transgene expression corresponding to Figure 2D.
    elife-67686-fig2-data5.xlsx (11.4KB, xlsx)
    Figure 2—source data 6. Statistical output of Kruskal-Wallis test corresponding to qPCR for msl-2 expression in Figure 2D.
    elife-67686-fig2-data6.xlsx (9.4KB, xlsx)
    Figure 2—figure supplement 1—source data 1. Data for sex ratios of 5’ alternative start codon transgene expression corresponding to Figure 2—figure supplement 1.
    elife-67686-fig2-figsupp1-data1.xlsx (15KB, xlsx)
    Figure 2—figure supplement 1—source data 2. Statistical output of Kruskal-Wallis test corresponding to sex ratios of 5’ alternative start codon transgene expression in Figure 2—figure supplement 1.
    elife-67686-fig2-figsupp1-data2.xlsx (10.6KB, xlsx)
    Figure 3—source data 1. Data for sex ratios of distantly-related homologs in Figure 3.
    elife-67686-fig3-data1.xlsx (19.3KB, xlsx)
    Figure 3—source data 2. Statistical output of Kruskal-Wallis test corresponding to sex ratios of divergent homologs in Figure 3.
    elife-67686-fig3-data2.xlsx (11.1KB, xlsx)
    Figure 4—source data 1. Data for sex ratios from expression of transgenes with single codon changes corresponding to Figure 4C.

    Data for qPCR from expression of transgenes with single codon changes corresponding to Figure 4D.

    elife-67686-fig4-data1.xlsx (15.4KB, xlsx)
    Figure 4—source data 2. Data for qPCR from expression of transgenes with single codon changes corresponding to Figure 4D.
    elife-67686-fig4-data2.xlsx (10.2KB, xlsx)
    Figure 4—source data 3. Data for qPCR from expression of transgenes with single codon changes corresponding to Figure 4D.
    elife-67686-fig4-data3.xlsx (10.9KB, xlsx)
    Figure 4—source data 4. Statistical output of Kruskal-Wallis test corresponding to qPCR from expression of transgenes with single codon changes in Figure 4D.
    elife-67686-fig4-data4.xlsx (9.7KB, xlsx)
    Transparent reporting form
    elife-67686-transrepform1.docx (245.8KB, docx)
    Source data 1. Source data for all graphical data sets and statistical tests performed for this study.
    elife-67686-supp1.xlsx (67.1KB, xlsx)

    Data Availability Statement

    All data generated or analyzed during this study are included in the manuscript and supporting files. Source data files have been provided for Figures 3-6.

    Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

    RESOURCES