Good Guys, Bad Guys

by Merry

The Good Guy. Aphidius wasps are commonly used for
for biocontrol of aphids. Here, a female wasp hits
her mark. Source.

Because it prefers to dine on some of our valued crop plants, the pea aphid (Acyrthosiphon pisum) is considered a major pest—thus a Bad Guy from our perspective. Pea aphids are not without their enemies. Enemy number one is a parasitoid wasp, Aphidius ervi. As parasitoid wasps are wont to do, females provide for their offspring by depositing their eggs in the haemocoel of another insect, in this case a pea aphid. The wasp larva feeds and develops there for about a week, eventually killing the host. When the host viscera have been consumed, the larva causes the aphid’s cuticle to harden and dry. Soon, an adult wasp emerges from the aphid "mummy." Agriculturally speaking, parasitoid wasps are Good Guys, important partners in our integrated pest management strategies. You can buy wasp eggs for your fields by the hundreds or thousands online.

Early on it was noticed that aphid (A. pisum) clones vary greatly in their resistance to the wasps. In some, the development of the wasp larva is arrested and the host survives. Because the pea aphid was known to harbor Buchnera aphidicola as a primary endosymbiont and also as many as five other bacteria as secondary endosymbionts, it was suspected that this variation in resistance might be due to which endosymbionts are on board. (What makes a symbiont a primary rather than a secondary symbiont? That depends on who you ask. Some distinguish based on the evolutionary age of the symbiosis, others mention the obligate nature of the relationship for both partners.)

The Bad Guy. SEM of an aphid, sure to have
Buchnera on board. Source.

The nutritional role of the Buchnera endosymbionts is well known. Aphids feed on phloem—a high sugar, low amino acid plant sap. The Buchnera genome encodes the genes required to make all the essential amino acids needed by the aphids. The Buchnera live within specialized aphid cells (called bacteriocytes) along with a variable number of secondary symbionts. The secondary symbionts characterized so far include three γ-proteobacteria, a Rickettsia, and a Spiroplasma. In order for these maternally-inherited symbionts to persist or spread in aphid populations, they must confer some benefit, at least under some conditions. Of the three γ-proteobacteria, the presence of one (Regiella insecticola) improves the survival rate of pea aphids infected by a major aphid fungal pathogen (the Entomophthorales fungus Erynia neoaphidis). For the other two (Hamiltonella defensa and Serratia symbiotica), the presence of either one confers some resistance to the parasitoid wasp—thus making them Bad Guys on the side of the aphid.

We now know quite a bit more about H. defensa, thanks to work carried out in Nancy Moran's lab. Some insights came from their recent sequencing of its complete genome. At 2.1 Mb, H. defensa has undergone significant genome reduction compared to its free-living close relatives such as Yersinia and Serratia with their 4.6 – 5.4 Mb genomes. On the other hand, it has suffered far less gene loss than Buchnera, whose genome is a mere 640 Kb. However, H. defensa lacks the genes needed to synthesize eight essential amino acids. For those, it relies—like its aphid host—on synthesis by Buchnera. What H. defensa does contribute is defense against the wasp. It earns its keep when parasitoids are present but otherwise carries a fitness cost. As the result, H. defensa occurs only sporadically in these aphids, its infection frequency fluctuating in concert with parasitoid pressure.

Accomplices. TEM of aphid bacteriocyte showing the
larger Buchnera primary symbionts (arrowheads) and
the smaller H. defensa secondary symbionts (arrows).
Bar = 10 μm. Source.

The story becomes even more complex because not all strains of H. defensa are defenders. Those effective in blocking development of the wasp larvae carry a λ-like phage known as APSE. (The source of this acronym? A. pisum secondary endosymbiont.) Even among the phage-carrying H. defensa, some are better defenders than others, likely due to differences between their phage. All of the ASPE phage studied so far have been found to have 95% or more sequence identity and to encode some nasty toxins. Some of their toxin genes are homologous to those found in well-known mammalian pathogens, e.g., genes for Shiga and Campylobacter jejuni toxins. The particular toxin-encoding genes present differ from one phage strain to another, likely accounting for the variable effectiveness of their H. defensa hosts.

That H. defensa's defense of its aphid host requires the ASPE phage does not mean that H. defensa doesn't play a part—only that it can't do it alone. H. defensa comes from a line of pathogens and has retained abundant weaponry, including type-III secretion systems, putative toxin genes, and related regulatory factors. Nevertheless, it seems clear that the phage is an essential partner. To quote the investigators, …the evolutionary interests of phages, bacterial symbionts, and aphids are all aligned against the parasitoid wasp that threatens them all.

But the wasp, too, is an essential partner, bringing the count of participants so far up to five: wasp, aphid, Buchnera, H. defensa, and ASPE phage. Without the wasp, H. defensa would disappear from the aphid population like so much excess baggage. Without H. defensa and its phage, the aphids would be more vulnerable to the wasps. Without Buchnera, the aphids would starve and the wasp would have to look for some other host to nourish its young. In this pentangle, there are neither Good Guys or Bad Guys, only an interwoven ménage à cinq.

Oliver KM, Degnan PH, Hunter MS, & Moran NA (2009). Bacteriophages encode factors required for protection in a symbiotic mutualism. Science (New York, N.Y.), 325 (5943), 992-4 PMID: 19696350

Moran NA, Degnan PH, Santos SR, Dunbar HE, & Ochman H (2005). The players in a mutualistic symbiosis: insects, bacteria, viruses, and virulence genes. Proceedings of the National Academy of Sciences of the United States of America, 102 (47), 16919-26 PMID: 16195380

Comments

[Tim Sampson]

A beautiful article on a beautiful system. Perhaps another example of how all the world is a phage!

[PWorthen]

"You can buy wasp eggs for your fields by the hundreds or thousands online."

... but you have to implant them in the aphids yourself.

[Nathan Myers]

It's in the wasp's interest, then, for H. defensa to be there, and to have some effectiveness -- otherwise, maybe, no more prey! -- but not to be absolute proof against parasitization -- or no more wasps! Ideally, it should be proof against every other wasp including her sisters, but defenseless against herself. Statistically defensive seems a good-enough approximation to that ideal.

How that unstable approximation might be maintained is curious. The wasp is always experimenting with ways to get past the defense, and perhaps succeeding occasionally, thereby wiping out the defenseless fraction of the population, and sometimes herself if that's too large. H. defensa, if it gets too good, wipes out the wasp population but also renders itself redundant and soon dispensed with. Probably this has happened frequently in isolated pockets, with repopulations from nearby and a tendency to reduced experimentalism in all participants?