A Blog by Ed Yong

One Species Becomes Two, Inside an Insect

It’s easy to imagine how a physical barrier, like a river or a mountain range, could create new species. If two populations of the same creature get stuck on either side of the divide, they won’t be able to breed. They’ll evolve along their own separate paths, until the differences between them become so big that breeding would be impossible, even if they were to meet. One species becomes two.

There are also many examples where new species arise without any barrier—where one population splits in two even though all of its members share the same space. This is called sympatric speciation (from the Greek for “same fatherland”). It was proposed more than a century ago, and has been controversial for much of that time. But scientists have found more and more examples that support the concept, where new wasps, flies, fish and trees evolve side by side. Just last week, news broke about a deadbeat ant that branched off from its parent species, while living in the same colony.

The latest intriguing example comes from James Van Leuven and John McCutcheon at the University of Montana. It involves a bacterium called Hodgkinia that split into two distinct species, while living in the cells of an insect. There is no barrier. Sardines in a can have nothing on Hodgkinia. These bacteria are crammed into the same tightly packed microscopic structures, but somehow, they’ve managed to become two distinct species.

The two daughter species are like two halves of their ancestor. They’ve each lost genes that the original Hodgkinia had, but they’ve jettisoned different genes. Each compensates for the losses of its sister species. They complement each other perfectly—put them together, and you’d (almost) the complete genome of the ancestor.

Van Leuven and McCutcheon made their discovery by studying cicadas—insects known for their ear-splitting songs. About five years ago, they showed that one species of cicada has two bacteria living inside its cells—Sulcia and Hodgkinia. This is pretty normal. Many insects have helpful internal bacteria or “endosymbionts”. In sap-sucking groups like cicadas, these microbes act like dietary supplements, making nutrients that are missing from their diet.

Things got strange when Van Leuven and McCutcheon analysed DNA from a South American cicada called Tettigades undata. They found many fragments of Hodgkinia DNA but, try as they might, they couldn’t unite those pieces into a single genome. They always assembled into two separate ones. For simplicity, I’m going to call these H1 and H2.

The two genomes belong to different bacteria; they’re never found in the same cell The team confirmed this by using fluorescent molecules designed to label each genome—a yellow one for H1 and a blue one for H2. You can see the results in the image below. Each little dot is a separate bacterium, and each contains either H1 or H2, but never both. (The green dots are Sulcia, and the magenta ones belong to the cicada itself.)

Two Hodgkinia species (blue and yellow) in a cicada. Credit: Van Leuven et al, 2014.
Two Hodgkinia species (blue and yellow) in a cicada. Credit: Van Leuven et al, 2014.

These two bacteria diverged from a common ancestor, which I’ll call H0, around 5 million years ago. The duo got a good idea of what H0 looked like by studying a closely related species of cicada, which only has one Hodgkinia genome with 137 genes. Out of these, 20 are there in H1 but not H2, and 44 are in H2 but not H1. All of them (except one) are found in one or both of the daughter species.

This pattern looks a lot like what happens when a species duplicates its entire genome, as has happened many times in the evolution of flowers, fish, and more. Suddenly, the species carries two copies of each gene. Since it only needs one, the second is free to pick up mutations that disable and destroy it, which is often what happens. The result is a genome with almost the same number of genes, but packed into twice as much material.

That’s exactly what Van Leuven and McCutcheon saw in their cicadas. The original Hodgkinia doubled up into two distinct genomes that add back up to the original. But in cases of whole-genome duplication, the doubled-up DNA is still part of the same genome. Not so here; in this case, H1 and H2 are separate entities. The process that created them is a bit like splitting a coin along its edge, so you get a heads-only coin and a tails-only one. There’s no novelty. H1 and H2 don’t do anything that H0 couldn’t already do.

That’s abundantly clear if you look at their genes. H0 makes nutrients like vitamin B12 and methionine to feed its cicada host. It devotes many genes to the task, one for each step in the chain of chemical reactions that eventually produce the nutrients. Between them, H1 and H2 can do the same, but neither of them has the complete set of genes for any chain. Neither one alone can give its cicada the nutrients it needs. They have to work together, passing chemicals between them like a production line that snakes between two adjacent factories.

How did this complicated set-up evolve? How did H1 and H2 arise from a population of H0 cells that were all living next to one another?

It’s hard to say for sure, but McCutcheon has an idea, illustrated in the diagram below. Each Hodgkinia bacterium contains thousands of copies of its genome (the green circles) in the same cell (the black outlines). At some point, one bacterium gets a mutation in just one of its genomes, which breaks one gene (marked in yellow in B). A second bacterium gets a mutation, again in just one genome, which breaks a different gene (marked in blue).

How one Hodgkinia species became two. Credit: Van Leuven et al, 2014.
How one Hodgkinia species became two. Credit: Van Leuven et al, 2014.

With so many genomes in each cell, these mutations don’t matter. They’re invisible to natural selection, and free to spread. As these bacteria divide, future generations of daughters might have two genomes with the mutation, then four, then eight (C).

Eventually, there’s some kind of bottleneck (D)—some event that produces lineages of cells where every genome has the yellow mutation or the blue one. Neither lineage is any use to the cicada on its own, because neither can make the full suite of nutrients that the insect demands. They have to work together, so they both stick around. They each start losing more and more genes, but always in a complementary way. They become H1 and H2.

It probably helps that cicadas can live for up to 17 years—long for an insect and an eternity for a bacterium. For most of those years, they live underground as immature nymphs that barely grow. In those years, the symbionts aren’t that important. They can change, evolve, and even build up detrimental mutations without affecting their host. The same symbiont shenanigans would be a much bigger problem for a short-lived insect like an aphid, and both the host and its deficient microbes would be quickly weeded out by natural selection. It’s probably no coincidence that, as McCutcheon puts it, “In the longest-lived cicadas, Hodgkinia genomes are off the deep end.”

“We think this is a very clear case of non-adaptive evolution,” he says. Hodgkinia didn’t gain anything from its split into two species, and neither did the cicada. It just happened through random events. It’s a great reminder that evolution isn’t a climb towards superior and more efficient forms. Sometimes, it leads to complexity for the sake of it. “This new symbiosis is not any better off than the simpler version,” says McCutcheon. “It’s just more complicated and it’s stuck.”

Reference: Van Leuven, Meister, Simon & McCutcheon. 2014. Sympatric Speciation in a Bacterial Endosymbiont Results in Two Genomes with the Functionality of One. Cell. http://dx.doi.org/10.1016/j.cell.2014.07.047

For another example of weird insect symbionts: Snug as a Bug in a Bug in a Bug

7 thoughts on “One Species Becomes Two, Inside an Insect

  1. Hi Ed,
    This sounds like a fascinating bit of biology. I need to read the study carefully, but it seems like it is somewhat outside the sympatric speciation debate (as usually framed). If the endosymbiont is asexual then sex and recombination– the usual factors that make sympatric speciation difficult– are absent. Thus while sympatry is a key part of this story, it’s not really a good case of sympatric speciation as we would usually talk about it in sexual organisms (but that might be my biased view on this).

    There’s actually relatively little discussion of sympatric speciation in the paper, which suggests (to me) that the authors dont see it as a strong case-study for how sympatric speciation could occur in sexual organisms.


  2. Thanks for the comment, Graham. Indeed, this paper is not meant to contribute to the sympatric speciation debate as it relates to sexual organisms. These endosymbionts are asexual, as far as we can tell. But we needed to call this something, and it seemed that, at least from the perspective of the endosymbiont lineage, sympatric speciation summed it up well. We thought about “sympatric lineage-splitting,” but that just seemed like a cop-out, and is awkward.


  3. Thanks for the reply John. I guess I was reacting somewhat to Ed’s description of it possibly being the best case of sympatric speciation to date. I’m definitely looking forward to working through the paper in more depth.


    [I’ve actually changed that line based on this discussion. Thanks, everyone. – Ed]

  4. I’m curious–humans are long-lived; is it possible we have some bacteria in our digestive tracts which have split up into two distinct groups, dividing their tasks in a way similar to these bacteria in cicadas? I wonder if any disorders or problems in assimilating foods may be caused by a lack of one of these complementary lines of bacteria.

  5. I wonder how the host responds to this. Its original symbiont H0 is replaced by two species H1 and H2. If it happened over the life time of an individual, the insect is not given the chance to co-evolve with its symbionts and to recognize them as such… There must be strong selection pressure on H1 and H2 to conserve the genes that are involved in the host-symbiont communication. It would be a very cool model to study host-symbiont interactions at the molecular level!

  6. I am a bit confused as I thought a single bacterium had a single genome, a single circular DNA molecule, which replicates and results in two bacterial cells. You talk about a bacterium having thousands of copies of its genome in a single cell. Is this a special feature of the Hodgkinia bacteria?

    [Yes, some bacteria have many copies of their genomes in a cell. Here’s an extreme example – Ed]

  7. Good article and a great example of non-adaptive evolution. Not exactly rocket-science is one of the few sites I return to on a daily basis. As this is an old article I hadn’t expected to find a typo, what with many thousands of people reading your blog (or what I think is a typo. I have English as a second language, so perhaps I’m mistaken. It happens :p)
    Where you wrote “put them together, and you’d (almost) the complete genome of the ancestor.”, I think you omitted ‘have’ after or before the ‘(almost)’.

Leave a Reply

Your email address will not be published. Required fields are marked *