March 09, 2016 at 04:26AM

Today I Learned: 1) Walbachia are a very interesting clade of bacteria that chronically infect insects. Walbachia are *extremely* common, and come in many varieties with many kinds of relationships with their hosts ranging from parasitic to mutualistic. The most interesting ones, from my perspective, are the Walbachia that live either specifically in their hosts' eggs or their sperm. Because they only reproduce through female or male offspring of the host, they are incentivized to maniuplate their host's reproduction toward one sex or the other. And they do. There are a number of mechanisms, most of which involve somehow killing off eggs of the wrong sex, but the effect is that they skew the sex ratios of their hosts, which is really unusual in the animal kingdom. Another cool Walbachia effect -- some Walbachia will kill off any egg they find themselves in with a different strain of Walbachia. I'm a bit surprised they're so scorched-earth about it, but apparently Walbachia don't like cohabitating with other Walbachia. 2) Apparently a common genetic pattern in new endosymbionts like mitochondria, chloroplasts, and certain kinds of intracellular parasites is that the symbiont will very rapidly break a bunch of its no-longer-essential genes (by indels, loss of promoters, mutation of start codon, acquiring new stop codons, etc), then slowly delete those lost genes. This results in extremely rapid (evolutionarily-speaking -- this is still a multi-million-year or tens-of-millions-of-years long process) shrinkage of the symbiont genome, which is how you get things like mitochondria with itty bitty tiny little genomes. It's apparently also fairly common for those genomes to eventually start to balloon up in size again, usually from non-coding RNA. Plant mitochondria are a good example of this -- there are some VERY large plant mitochondrial genomes out there, even though they almost all have the same genes. 3) Speaking of bacteria that live in insects, and speaking of endosymbiont genetics, let's talk about cicada endosymbionts! It turns out that most or all cicadas are obligate symbionts with bacteria that they cultivate inside specialized cells, which are housed in specialized organs. Cicadas cannot manufacture all of the amino acids they need, and their diet is ridiculously poor in protein*, so they have bacterial symbionts that are specialized for amino acid manufacture that give them the amino acids they need to grow. There are lots of cicadas with an apparently wide variety of bacterial endosymbionts, but the talk I atteneded today (by John McCutcheon) was about cicadas of the genus Diceroprocta, which have two (sort of -- we'll get to that) species of bacterial endosymbiont, Sulca and Hodgkinia. Both are really weird-looking bacteria that form giant (multi-micron-long) tubes that seem to be *much* more permeable than the average bacteria (fun quote from the talk: "our theory is that these bacteria are permeable to everything except genomes"). Neither bacteria can synthesize a full set of amino acids; neither bacteria can live outside of the cicada's specialized cells; the cicada cannot survive without the full set of amino acids produced by the two symbionts. It's a fully-obligate three-way symbiosis, and it looks like Sulca and Hodgkinia are on their way to becoming cicada organelles. This makes Sulca and Hodgkinia potentially invaluable models of the early evolution of endosymbiotic organelles (mitochondria and chloroplasts, primarily). We can already see some of the expected genetic signatures -- Sulca and Hodgkinia have already lost key metabolic genes that are supplemented by the other species. Sulca is the relatively normal species of the pair. Hodgkinia has some wild stuff going on. For example, they're *missing key tRNA and tRNA synthetase genes*!!! That's absurd. Organisms basically can't produce any proteins without a full set of tRNA and tRNA synthetase... unless they're fed those molecules by a host. Hodgkinia apparently get by off of tRNAs produced either by the cicada or Sulca in the same cell (again, these cells are super permeable**). It goes a layer deeper, though. It turns out that Hodgkinia in many species of cicada have more than one form of genome, meaning that there are several (2 to possibly 80, depending on the species) different genome variants and each Hodgkinia cell has one of those variants (or, in some cases, two or three associated genome variants)***. Each genome variant has different deletions or pseudogenizations, but the population as a whole within each cicaca has complete coverage of the Hodgkinia genome. So in the two-genome case, there are some Hodgkinia that have deletions in one set of genes that are present in the other Hodgkinia, and vice versa. Somehow the cicada has to get at least some of every Hodgkinia variant (arguably, each Hodgkinia species, depending on how you want to define species here) from its mother, or it dies. We have no idea how the cicada ensures that it gets some of each. The point is that early endosymbiont evolution apparently isn't as simple as "they lose genes until they get to some sort of minimal set". Hodgkinia seems to be going through a weird *fragmenting* of its genome into overlapping and complementary populations. My suspicion is that the end product of this fragmentation will be consolidation into a single, minimal genome, which is what we see in many chloroplast and mitochondria... but even some mitochondria have multiple genomes! There's also a question of whether the genomic fragmentation of Hodgkinia is selectively advantageous, deleterious, or neutral. McCutcheon claims it's neutral, or possibly slightly deleterious, and I'm decently well convinced. Question to tuck away for later use -- could genomic fragmentation of this sort have been responsible for the development of eukaryotic chromosomes? * I also learned today that cicacas exclusively eat tree xylem sap, which is the sap that runs from the roots to the leaves. This is a terrible food source for several reasons. For one thing, it's really dilute for a nutrient-bearing liquid. For another thing, it mostly carries water, with just a bit of sugar. For another thing, it's under negative pressure, which means if you stick a straw into xylem, it will suck air into the tree rather than let sap out. To deal with this, cicadas need specialized mouthparts that actually suck liquid. In contrast, they *could* have fed on phloem sap, which runs from the leaves to the roots, and carries much more sugar and nitrogen, and is under positive pressure so that you just have to stick a mouthpart in it and it flows right out. I guess it just wasn't worth the extra difficulty of getting to the phloem sap... EXCEPT THAT PHLOEM IS CLOSER TO THE SURFACE THAN XYLEM IN MOST TREES. ** Maybe. That's the theory, at least. *** This took a bit of a heroic sequence assembly effort to figure out for the first time. One of McCutcheon's poor grad students tried sequencing the genome of a new cicada species, thinking rather reasonably that it would only have one. Nope. See the first two paragraphs of Results and Discussion here for details: http://ift.tt/1SyNjsz.