Strepsipteran genome brings us a step closer to solving an entomological enigma

Parasitoids play a huge role in regulating insect populations. They have a lot of tools to do this, including using genetic modification to disable the immune systems of their hosts, as I described in Polydnaviruses: Nature’s GMOs. In agriculture, if you remove parasitoids, you end up applying far more insecticides… and nobody wants that. In apiculture, removing parasitoids such as Apocephalus borealis could help honey bee populations. Parasitoids are important.
The Strepsiptera are one group of Parasitoids that I haven’t yet discussed. These are particularly bizarre insects. The males are relatively unremarkable… they’ve got six legs, antennae and wings. Relatively bug-like. The females, however, don’t even resemble insects. Out of all the Strepsiptera, the genus Xenos has been best described (shown here). The bottom picture is a male. The top picture, however, is a female. Yeah… they’re the same bug.
This is where things get freaky. In more ways than one.
Strepsiptera biology is completely unique amongst insects. While the males are readily identifiable as an insect, the females resemble worms. They’ve lost all the features that identify them as insects. No legs, wings, or antennae. The head and thorax are merged into one segment. The internal anatomy is even more bizarre. They don’t have reproductive organs, their eggs simply circulate in their bloodstream. Fertilization occurs when the male stabs her in the head with his penis and sperm meets egg in the bloodstream. The eggs develop inside the female, then turn into legged larvae which burrow out of the female – killing her in the process. The larvae then seek a host and burrow into it.
Their host patterns are even more unusual. It’s common for Strepsiptera males and females to seek out different hosts, something very rare in parasitoid wasps and something I’ve never heard of for parasitoid flies. They also rarely kill their hosts outright, instead they hijack the host’s brain and extend their lives. As I said… they’re weird bugs.
The males are weird, too… but in ways that are a bit more subtle. Stabbing penises have evolved multiple times in insects, and many species of insects have their own style of S&M. However, the order’s wing characteristics are weird. They have wings and balancing organs called ‘halteres’, similar to flies. However, on flies it’s the rear wings that are modified into halteres instead of the front. This can be explained by screwed up hox genes… but those mutations don’t appear to exist in Strepsiptera. Strepsiptera very much resemble beetles because they use their back wings to fly, whereas flies (Diptera) use their front wings. In some beetles, the front wing cases have been reduced in a superficially similar manner. In short, they’re intriguing insects.
The question of what the Strepsiptera are related to has intrigued entomologists since they were discovered. They were originally placed within the Hymenoptera. Some early phylogenetic analysis of their ribosomal RNA genes placed Strepsiptera with flies, despite the fact that they have morphology that’s more similar to beetles. Some entomologists have even argued that they’re not even related to insects that have complete metamorphosis.
Historically there have been five hypothesis about their relationships, with three taken seriously. The problem is that a lot of our attempts to choose between them have been foiled because of molecular convergence. There’s only so many ways a genome can change, so sometimes you end up with similar looking genomes only because of a statistical anomaly. Geneticists call this long branch attraction and morphologists call the phenomena convergent evolution.
The five hypothesis about Strepsiptera (as described in Genomic and Morphological Evidence Converge to Resolve the Enigma of Strepsiptera) are:

1. Strepsiptera are related to beetles. Based on morphology.
2. They are really highly derived beetles. There are some parasitoid beetles, and there are morphological similarities between Strepsiptera and some groups of beetles.
3. They are flies. Based on ribosomal RNA evidence, and some handwaving about hox mutations.
4. They’re related to groups of insects that go through a process similar to, but not identical to complete metamorphosis. (To be honest, the first time I’d heard of this theory was while reading this paper.)
5. They are parasitoid wasps. This was pretty obviously not the case to early researchers, and like #4 only a select few have heard this idea. They bear little morphological resemblance to wasps, and you pretty much have to go back to the time of Linnaeus to see this mentioned in the literature. But… technically it’s in the books.

So… how do we solve this mystery?
Use ALL the genes!
The Strepsipterans shown in this post belong to the families Stylopidae and Mengenillidae, which represent the two main groups of extant Strepsipterans. The latter are the more primitive groups, with females retaining their legs. They also leave the host before their larvae emerge, so they retain some characteristics of their ancestors. If you’re looking to minimize any sort of long branch attraction, this is the best way to do it.
The genome of a primitive Mengenillid Strepsipteran has been published. If you want to figure out the evolutionary history of the order, this is a good place to start. From this genome a group of entomologists attempted to recreate the evolutionary history by comparing it to the genomes of a bunch of other insects that have had their genomes sequenced. They also looked at the position of the genes.
By looking at the genomes of the insects, they were able to discern that the Strepsiptera were different enough genetically to separate them from the beetles, but still very closely related. They’re their own distinct order, but very closely related to beetles.

There’s still a lot more work that needs to be done. This analysis, while using a lot of genes, did not incorporate morphological data and relied solely on genetics. The taxa they used were also biased by those which have had their genomes sequenced… in short, they didn’t use a lot of different groups of insects. But they used a lot of genes.
There are some ways to fix this. Genbank houses a lot of expressed sequence tags (ESTs) from a lot of different insects. They could use ESTs to extend the number of insects they used in this analysis. This, of course, brings it’s own set of problems because then there’s going to be a lot of gaps… you might have gene A for insect 1 but not Gene B. For insect 2, you might have gene B but not gene A… that kind of thing.
I’d also have liked to see them incorporate other Strepsipterans. It’s possible that primitive insects have genes that are highly derived… and parasites tend to evolve pretty fast on a molecular level.
I’s also have liked to see inclusion of morphological data. The morphological data they refer to in this paper is things like gene order on the chromosome and similar phenemona, and not the features I discussed earlier, such as wings. The  morphological data comes out pretty strongly in favor of Strepsiptera being most like beetles… but I’d still like to see it incorporated along with the genetic data.
Genetic data isn’t the whole story, but neither is morphology. Thanks to this paper, however, it looks like the genetic and morphological data may no longer be in conflict. Still, there’s a lot more work to be done. Now… if only they could do something about the mess that is the lower Neoptera.
.
Citation: Niehuis O., Hartig G., Grath S., Pohl H., Lehmann J., Tafer H., Donath A., Krauss V., Eisenhardt C. & Hertel J. & (2012). Genomic and Morphological Evidence Converge to Resolve the Enigma of Strepsiptera, Current Biology, 22 (14) 1309-1313. DOI: 10.1016/j.cub.2012.05.018
 
Edited by Joe Ballenger at 1:07 PM on 3/21/2014 to correct spelling, and clarify some of the classification.