Wednesday, April 16, 2014

Wolbachia Variants Induce Differential Protection to Viruses in Drosophila melanogaster

A truly awesome paper on Wolbachia, variation in the pathogen-blocking phenotype, and genetics was published a while back by Luis Teixeira's group.  I've been eager to write a blog post on this particular paper, one of my favorite papers from 2013, so here goes!  
<Follow along with the paper here.>

You are reading this blog, so maybe you don't need convincing that Wolbachia are totally awesome, relevant, and interesting bacteria.  They infect about half of the insect species on the planet and do so by targeting the germ line: that's right folks, these babies come pre-loaded with their bacterial symbiont.  Recently, Wolbachia have become more medically relevant because folks (including Texieira himself with Michael Ashburner) found out that they protect their insect hosts from virus infection -- either by reducing the load that the host carries (resistance) or by preventing disease even if the virus replicates (tolerance).  However, different Wolbachia strains vary in their ability to pathogen block.

In Chrostek et al., (2013), the authors attempt to figure out what's behind this variability -- why do some Wolbachia strains protect their hosts better than others? Is this variability consistent across virus strains?  These are relevant questions to ask because 1) it may reveal the underlying biology of Wolbachia-host interaction and 2) it may lead us to different Wolbachia strains that could be utilized in pathogen blocking. 

The authors start by sequencing some variants of Wolbachia (they also use previously published whole genome data from Casey Bergman's group) and by figuring out if these variants differ in their ability to protect the host from Drosophila C virus.  Just because I'm a big fan of BIG trees - here's the phylogeny resulting from their genomic analysis.
Figure 1

Interestingly, the variants do differ. In Figure 2A (below), you can see that each of the strains provide differing degrees of protection, with the uninfected line (iso) dying quite shortly after infection.  They also repeated this experiment with Flock house virus with similar trends (see Figure 3).

Figure 2 from Chrostek et al., 2013

Next, they look at the density of Wolbachia variants within the same host background.  Interestingly, they found that wMelCS variants exist at MUCH higher titers than wMel variants.  In Figure 4 below, you can see the results of qPCR on Wolbachia genomic DNA from flies that are 3-4 or 6-7 days old.  Strikingly, you see a huge upward trend in wMelCS infection as the flies age (Figure 4D), but no so much for wMel.  Interestingly, some of these wMelCS variants reduce host lifespan!

Figure 4 from Chrostek et al., 2013

So...there must be some difference in these Wolbachia strains.  Chrostek et al were quite careful in their crosses -- they removed confounding variables such as the non-Wolbachia microbiome and  host genetic background.   So, they look at potential genomic differences in these strains -- remember, they sequenced the genomes to characterize their Wolbachia into the different variant clades.  They present a very large table of SNPs in the pairwise comparisons ... some interesting looking genes with ankyrin repeat domains...potentially cool stuff for future work from this group.  

BUT! The rub is that none of these indels are found in wMelPop, the variant that protects BEST against viruses and infects at highest titer (is the most pathogenic to the fly).  So...what else could be different? What is driving the titer differences in these variants?

Could be copy number variation be driving the difference between the strains?  Indeed so! The authors identify a region in wMelPop, containing 8 genes, that is elevated in copy number (between ~2-8x) compared to the other wMel variants. In a brilliant stroke of creativity (or as a result of a potential late night pub crawl), the authors name this region the "Octomom" region (see below):

Figure 7: So called "Octomom" region increase in coverage in the wMelPop genome (A) and qPCR based amplification in wMelPop vs other variants (B)

This region has some interesting genes in it, some of which are phage related, some of which have homologs to mosquitoes! Although we don't know what these genes are doing, these proteins could be of interest for those researchers interested in Wolbachia pathogenesis.  

< Oh, and also, this entire body of work is all in the context of the global replacement of certain Wolbachia strains in Drosophila melanogaster, as it turns out. >