I spent part of my weekend reading a very nice, detailed paper by Soderlund and Knipple (thanks for the recommendation, Nick!) on the distribution and mechanisms of mutations that confer pesticide resistance on insects. The main message is that we have been using potent pyrethroid poisons that have a common mechanism, targeting the highly conserved sodium channel of the nervous system, and that similar mutations that reduce the affinity of the channel for the pyrethroid are popping up in many insect species.

First, a little elementary neurobiology. Signals are shuttled around the nervous system by pulses of ion flux that are primarily mediated by two ions, potassium (K⁺) and sodium (Na⁺). Neurons are typically sitting at a negative voltage at rest; a signal, or action potential, is a result of a transient opening of Na⁺ channels so that positive sodium ions rush in, depolarizing (or reducing the voltage difference towards zero) the cell. So, the resting state of a neuron is negative, and a change in state is signalled by short, sharp flicks towards 0.

Normally, these signals are very brief. There are two general reasons for that: one is that the sodium channel is kept on a tight leash, and after opening for that inward rush of positive sodium ions, it tends to automatically inactivate after a short period. The other is that there is an opposing current, the K⁺ current, that consists of positive K⁺ ions rushing outwards, balancing the sodium influx. You can see the two currents diagrammed here, the outward potassium current (I_K) in black, and the inward sodium current (I_Na) in purple. You can also see how fast these events occur, on the order of a few milliseconds.

What pyrethroid poisons do is mess up that purple line; they bind to the open sodium channel and prevent it from closing, so that the inward sodium flux continues for a longer period of time. This is not good for a finely-tuned, functional nervous system—it means the signals flick towards zero, and then stay near there longer than they should. It's a subtle difference, but since depolarizing a cell a little bit makes it more likely to send a signal, it means the whole nervous system is made hypersensitive, and storms of activity can go thundering through the whole thing. The poor bugs hit with this poison are taken out fast as the nervous system locks up, a process called knockdown.

You can knockdown lots of different kinds of insects with one poison because the sodium channel, the target of the pyrethroid, is highly conserved. The channel protein has a characteristic structure consisting of four domains that cluster in the membrane, forming a pore between them. The channel must have a characteristic shape and arrangement of charged amino acids in its interior to confer specificity: only Na⁺ is allowed through, not Cl^- or Ca⁺⁺ or K⁺ or any of the other ions wafting through the fluid-filled space around the cells. The molecule also has specific bits that mediate gating, and make it sensitive to the transmembrane voltage. It's not surprising that the sodium channel is highly conserved—it's an intricate apparatus with a lot of critical bits and pieces. There is variation, and the differences increase with increasing phylogenetic distance, but the overall structure of four domains forming a pore is found everywhere in animals, not just in insects, but in us mammals as well.

This diagram therefore applies everywhere in the animal kingdom. The sodium channel is made up of four domains (I-IV), each with 6 membrane spanning strands (S1-S6), that assemble to form the pore.

One of the useful things the Soderlund and Knipple paper does is put in one place a list of all the insect mutations that confer knockdown resistance. The idea is that what these mutations do is interfere with the binding of the pyrethroid to the channel, so mutations that block that binding are likely to be changes near the binding site. It's a way to map the location of the pyrethroid binding site, and here's the summary diagram:

Diagram of the extended transmembrane structure of voltage-sensitive sodium channel a subunits showing the four internally homologous domains (labeled I-IV), each having six transmembrane helices (labeled S1-S6 in each homology domain), and the identities and locations of mutations associated with knockdown resistance. The symbols used to identify mutations indicate their functional impact as determined in expression assays with X. laevis oocytes.

They may look a bit scattered, but keep in mind that in the 3-dimensional structure of the protein, the four domains are folded together, and it's S5 and S6 that all line the interior of the pore, so these are less widely dispersed than the initial impression might give. The site labeled L1014 (S, H, and F are 3 different variants at the same site) is the most common target of mutations that confer knockdown resistance.

The M918 site is also interesting. All insect species happen to have a methionine residue at that particular spot in the loop between S4 and S5 in domain II; animals that mutate that particular amino acid acquire a profound resistance to pyrethroids. As it turns out, all known mammalian sequences contain an isoleucine at that point, and all mammals are resistant to pyrethroid toxicity (which is why it's valuable as a relatively selective poison). Converting that isoleucine to methionine in a rat makes the rat 10 times as sensitive to pyrethroid poisons.

To me, the virtue of the paper is in the elucidation of mechanisms and the discussion of variations in the sodium channel, but hey, you know that one of the goals has to be better, more effective ways of killing bugs (I used to work in an insect lab, and know a number of insect neurobiologists, and one of the unfortunate facts of life in that field is that it helps get grant money to mention something about finding new ways to enhance arthropod neurotoxins.) The authors tout one of the techniques used here, the heterologous expression assay, as a great way to characterize resistant mutants and identify poisons that circumvent the mutations. The heterologous expression assay involves extracting the sodium channel RNA from the insect, expressing it in frog oocytes, and measuring its properties and responses to toxins there. Frog oocytes are a great tool for measuring channel properties, because they are huge, with wide expanses of membrane, making it relatively easy to measure the ion fluxes of a particular, artificially expressed channel type. The idea is that we could refine a suite of toxins that blast different common insect knockdown resistance mutations, and then design a rational pesticide rotation strategy that kills the typical variants as they emerge from each cycle of selection.

That's one of the dilemmas of insect research, I guess, that you may love this nifty little critter you study, but one of your jobs is to find more thorough ways to eradicate them.

Soderlund DM, Knipple DC (2003) The molecular biology of knockdown resistance to pyrethroid insecticides. Insect Biochemistry and Molecular Biology 33: 563-577.