Behind the paper: A bacteria strain that kills males and its co-evolution with its host

In this ‘behind the paper’ post, Ary Hoffmann explains how they found a Wolbachia strain that caused all-female offspring in infected flies and the finding that host genes are able to suppress this phenotype.

I’m Ary Hoffmann, a professor at the University of Melbourne in the School of BioSciences and the Bio21 Institute where I lead the Pest and Environmental Adaptation Research Group (PEARG). I’ve been interested in endosymbiotic bacteria since the mid-1980s when I discovered Wolbachia bacteria living in Drosophila cells by chance while I was a postdoc working with Michael Turelli at the University of California. I’ve seen this area expand rapidly over the last 40 years to the extent that endosymbiotic bacteria living in insect cells are now seen as a practical tool to suppress insect disease vectors and to prevent the transmission of diseases such as dengue viruses.

Our new paper came out of an observation made by Kelly Richardson from PEARG. She noted that some female Drosophila pseudotakahashii flies produced only female offspring whereas others produced a 50:50 ratio of male and female offspring. We had previously established that all individuals of this Australian endemic species are infected by a Wolbachia strain which we had already shown to cause cytoplasmic incompatibility (CI), the process where males infected by Wolbachia cause embryo death when they mate with uninfected females (but there is no death in the reciprocal cross). Because infected females can produce viable offspring regardless of whether they mate with infected or uninfected males, and pass on the bacteria to all their offspring, infected females can have a huge advantage over uninfected females, helping to spread the CI infection throughout a population.

However, the absence of male offspring in some crosses indicated that something else was afoot. We already knew from other work in a variety of insects including flies, butterflies, and beetles that some endosymbiotic bacteria including Wolbachia and Spiroplasma could cause male killing (MK); MK infected females still mate with males, but their bacteria cause the male offspring to die. This intriguing phenomenon can increase the fitness of MK infected females in several situations. One of these involves “sib competition” – the intense competition among sibling larvae developing on limited resource; when male siblings are killed, there is an opportunity for the remaining female larvae to successfully complete development.

Kelly, Perran Ross and others from PEARG completed several experiments to understand male killing in D. pseudotakahashii further. They showed that the MK females were all infected by Wolbachia and still compatible with males that carried the CI-causing Wolbachia — the MK females were protected from being sterilised by these males. They also showed that the male killing phenomenon was transmitted through the mother.

Image credit

Ary Hoffmann, CC BY 4.0

The next part in this story emerged when we started to characterize the Wolbachia from CI and MK strains at the molecular level. When sequencing sections of Wolbachia genes we noticed that sequences from the MK strain had two nucleotides whereas CI Wolbachia only had one of these nucleotides. This led us to suspect that there may have been two different Wolbachia infections in the male killer strain with (mostly but not entirely) similar sequences. We then started a collaboration with Brandon Cooper’s group at the University of Montana who were sequencing complete Wolbachia strains from different Drosophila species. By sequencing Wolbachia from both the CI and MK strains of D. pseudotakahashii, they found support for the double infection hypothesis. Wolbachia from the MK strain showed reduced coverage in two sections of the Wolbachia genome identified from the CI strain, which we interpreted as the genomic regions missing in the MK strain which was otherwise identical to the CI strain. We had therefore discovered a new male killer whose presence was normally masked by the CI Wolbachia to which it was closely related!

A new layer of complexity was added to this story when Kelly and Perran noticed that males started to appear again when some lines showing male killing were held in the laboratory for several generations. In fact, some MK lines eventually went back to producing a 50:50 sex ratio, even though they remained infected by Wolbachia. This unexpected result led us to propose that nuclear host genes may have rapidly evolved to suppress the male killing phenomenon. Such “suppressors” have previously been found in a butterfly from the Pacific which had almost gone extinct because the loss of males meant that females no longer had access to the males required for mating and offspring production.

Perran and others next used a series of backcrosses to show that the suppression of male killing was indeed inherited through the host genome. And by sequencing SNPs across the genome from lines where the suppressor had swept through the population. Tom Schmidt from PEARG was able to locate the suppressor genes involved in this sweep to one specific region of the host’s genome.

Image credit

Eddie Tsyrlin, CC BY 4.0

The work highlights the complex interactions between bacterial strains and their hosts in driving phenotypes associated with endosymbionts. An evolutionary interplay between them can lead to their appearance and loss. Perhaps male killing phenotypes due to Wolbachia and other endosymbionts are common in natural populations but their expression is blocked by past evolutionary processes that have led to the evolution of suppressor genes.

Clearly there is a lot to learn about evolutionary trajectories in insect–endosymbiont systems. Given their growing use in controlling pest and disease vectors and in suppressing disease transmission, it is important to understand these trajectories. The D. pseudotakahashii system and others like it also provide a unique opportunity to understand the molecular basis of sex ratio distortion and its suppression.  Since the 1980s, we have learnt an enormous amount about Wolbachia and other endosymbionts but there is still a long way to go.

About the author

Ary Hoffmann is a Melbourne Laureate Professor and heads the Pest and Environmental Research group at the University of Melbourne. 0000-0001-9497-7645 @PEARG_LAB