GMOs used as pesticides in a brand new way!

posted in: Science | 35

I just returned from Reno, Nevada attending the Entomological Society of America’s annual meeting. I went to a bunch of really neat talks, saw some old friends and met some new friends. It turned out to be a great networking opportunity, and I met some folks doing amazing research I would really like to work with in the future. Unfortunately, I wasn’t presenting data because I missed the submission deadline but I was still fortunate enough to be on the debate team.

The topic of the debate was “can organic agriculture solve food scarcity problems?”. The subjects were randomly chosen and don’t necessarily support the views of those engaged in the debate, so I will not speak for anybody but myself. I was on the con team, and we were charged with arguing that organic agriculture is an inferior method of food production. We were up against a very good team and all day folks were coming up to us and telling us how much they enjoyed our debate. Ultimately, we won the best overall debate team and took home an engraved trophy and left the meeting $125 richer after splitting a $500 prize between the four of us.

My role on the team was to look into the pesticides used in organic agriculture and their treatment regimes. To my surprise, I found that organic operations actually increase the amount of inputs put into the environment by requiring higher concentrations and more frequent applications of pesticides. The insecticides used in organic ag are often less effective, less selective, and can have greater non-target effects than synthetic insecticides. Some organic pesticides, like the biopesticide Beauveria bassiana, are assumed to have a very low environmental impact quotient (EIQ), but haven’t been tested for potential ecological side effects. My position (and position on the debate team) is that GMOs like Bt corn are better for the environment because they decrease the amount of pesticides that we must put on crops and that organic pesticides are worse for the environment because they must be constantly reapplied in very high concentrations.

This, however, wasn’t the idea that earned me my stripes during the debate. During the Q&A session, somebody asked us to clarify why we thought organic ag was able to innovate to a lesser extent than sustainable or conventional agriculture. My response was that we can modify pesticides to become less toxic, more easily degradeable and more difficult for insects to detoxify by producing insecticides synthetically and making it more or less difficult for the insecticides to degrade. While organic ag could certainly benefit from new chemistries, they reject them as soon as modifications such as these take place because the new pesticide is synthetic. In short, organic producers are unable to take advantage of novel chemistries. I used the example of adding carbon atoms or benzene rings in a specific place to keep beta-lactam antibiotics medically relevant during the debate, but there was a much better example I could have used but unfortunately neglected to discuss. But, hey… that’s what the blogosphere’s for isn’t it?

Very recently, the lab of Reddy Palli has figured out a way to genetically modify an organism to become a spray-on pesticide. To fully understand and appreciate what’s going on, there are a lot of things I need to explain. Fortunately, I’ve got about 12 hours of travel time ahead of me. Awesome, right?

A Colorado potato beetle. USDA photograph by Scott Bauer via Wikipedia.

First, let’s talk about the animal discussed in the paper. The Colorado Potato Beetle is what’s referred to as a ‘superpest’. It’s highly prolific, and essentially bulletproof. This insect specializes on solanaceous crops like potatoes and tomatoes, the crops most closely related to nightshade plants. These plants are famous for defending themselves by producing deadly secondary metabolites. By specializing on these plants, the Colorado Potato Beetle has evolved with some incredible detoxification mechanisms which shields it from our pesticides. As an unfortunate (for us) side effect, it manages to become resistant to every pesticide we throw at it very quickly. It can defoliate entire potato fields, and we can’t stop it very easily. We’re desperate for new chemistry to counteract this pest.

Next, let’s talk about a very basic part of insect physiology. Insects, like humans, are made from proteins encoded by DNA. When a protein needs to be made, an RNA polymerase translates DNA to RNA, and a ribosome transcribes the RNA molecule to protein. This is pretty constant throughout the kingdom of life plants, humans and insects all use a similar system and there is RNA in everything you consume. It can get a bit more complicated than this (see below), but there’s one thing I need to point out – mRNA is always single stranded in eukaryotic organisms. Some viruses use a double-stranded RNA (dsRNA) molecule. This is kind of like DNA, but it’s made out of slightly different stuff. Insect immune systems are good at picking up stuff that looks like it shouldn’t be there and dsRNA sticks out like a sore thumb.

The beetle has an immune system just like us. Ever get sick? Did you get better? That’s your immune system working. Beetles are exposed to pathogens just like we are every day. A good example of this is a cypovirus, which is kind of like an insect rotavirus. When the beetle gets a cypovirus, a series of enzymes pick the dsRNA it makes from the crowd of mRNA and selectively degrades it by using that dsRNA as a template to scan all the RNA in the insect and then degrade it. This is called RNA interference, or RNAi.

How can we use this to our advantage?

A special shoutout to Jafsica and Bug G Membracid, to whom I nerded out about this at ESA
The experimental setup Palli's team used. Everything's labeled pretty well, and very self explanatory. The larvae eat the leaf, eat the dsRNA which causes their own body to shut down vital systems.

Unlike our antibody production system the RNAi system is kind of stupid and won’t distinguish self from nonself mRNA. The reason for this is that RNAi is also used to make sure the beetle doesn’t produce too much of a particular protein. If it wants to shut down certain specific proteins, it can make small interfering RNA (siRNA) and allow the RNAi system to destroy the RNA. It’s physiologically important for the beetle to be able to do this, but there’s no doublecheck system. The beetle can’t tell if it produced the RNA or if the dsRNA came from another source.

Reddy Palli’s lab did something ingenious with bacteria. They inserted several sequences into a bacterium that made double stranded RNA to a variety of important proteins. These included the muscle protein actin, sec23 which is a protein involved in the transport of newly produced proteins, and a couple ATPases which are responsible for producing the ATP energy currency of the cell. After killing the bacteria but preserving the RNA, they sprayed the bacteria onto potato plants which contained Colorado Potato Beetles. They also did this with just straight dsRNA. The beetles eat the plants, they eat the bacteria and a whole load of dsRNA.

What happened?

Here’s the cool part: it actually worked. To me this is mind blowing because RNA is incredibly unstable, thanks to an oxygen attached in a rather unfortunate place which allows it to break the backbone of the molecule. There are also nucleases which degrade RNA so the bacteria had to be modified so they wouldn’t produce these enzymes. Keeping the molecule double stranded helps by making it more difficult for either of these reactions to occur, so dsRNA is more stable than regular mRNA. But it’s still an incredible thing to me that this even worked.

If you want to show this works, you need to first show that the mRNA levels drop in response to the treatment. Turns out that they do for all the genes involved. Actin is the muscle protein, the ATPases produce energy and sec23 and CopB are involved in protein transport. The control was something which has relatively constant transcription that wasn't target by RNAi.

The beetles ate the killed bacteria, digested the outer wall and released the dsRNA. The cells take up the RNA, and the RNAi process occurs just as described above. The RNA coding for actin gets degraded, so that the beetles don’t make new actin or repair their existing actin polymers. In short, their muscles fall apart, their cells don’t divide. Even their sperm wouldn’t move…all these processes are dependent on actin.  As a direct result, the beetles stop eating, stop moving and die. Similar things happened with the other genes. When sec23 and COPB are silenced, their proteins don’t properly get transported and modified, resulting in a buildup of nonfunctional machinery. When the ATPases are silenced, ATP is no longer produced and the beetle can’t produce enough energy to maintain vital life functions. From this research, it would appear there are a great diversity of genes we could target which opens up a lot more avenues of attack when making pesticides.

Now, there are some neat implications to this research. This was a ‘proof of concept’ paper, which means that this works on a particular organism with a particular set of proteins under ideal lab conditions but doesn’t directly deal with the economics, field conditions or range of pests that could be targeted. It’s exciting and this technique has a lot of potential, but a lot more research needs to be done before we could use this in the field. That doesn’t mean there aren’t good reasons to be excited to see this further developed, though. Even though this may be a somewhat limited technique (see below), I could still see this being used to create very highly specific insecticides that quickly degrade in the environment.

In general, this would be the ideal pesticide for an environmentalist because RNA is all around you, as are nucleases. The Colorado Potato Beetle produces RNA and siRNA. We produce RNA and siRNA. Bacteria produce RNA, but I’m not sure if they produce siRNA. This is essentially all-natural, with the only difference being that we’re telling the beetle to degrade proteins at the wrong time and at a much higher rate than it normally would. RNA degrades by itself pretty easily and RNA degrading nucleases can be found almost anywhere you look. The bacteria can degrade in the environment and have no components which aren’t found in soil bacteria except foreign RNA sequences. There’s no reason to think there would be any issues with the bacteria staying around in the soil for an extended period such as we’d see with DDT.

Despite my enthusiasm for this clever technique, I also don’t want to give anybody the impression this is a ‘magic bullet’ for pest control. Some critters take up RNA better than others. RNAi was discovered in nematodes using this technique, so we could potentially use this on nematodes as well as beetles. Honeybees are able to ingest RNA and acheive silencing, so we might even be able to target sawflies. We could not use this on moth pests because lepidopterans are notoriously difficult to perform RNAi in, which has led to caterpillars being more of a biochemistry rather than genetic model organism. Since a lot of pests like aphids pierce the plant and suck the juices out, this would be useless against them because they’re not actually ingesting anything on the outside of the plant. There also may be better ways to introduce the dsRNA and for all we know using viral machinery may be a better way to introduce and replicate the dsRNA. There’s a lot more basic research which needs to be done on this before I’d be willing to say ‘we could use this’. With this paper, there are good reasons to think this would work.

The results from the experiment. This is a survival curve, with the percentage of the larvae surviving plotted on the Y axis and the time of survival plotted on the X axis. As you can see, the survival curves dropped far below the controls. A is the bacteria encapsulated dsRNA, while B is the dsRNA without bacteria. Both work, despite the limitations I explained earlier.

In addition to needing to pay attention to the pests this could work on, we need to pay attention to the kinds of beneficial insects and other animals this would potentially harm just as we would any other pesticide. Actin tends to be pretty similar in all organisms. The other genes are really important, and are probably very conserved in genetic sequence. I would think this could have some potential nontarget effects on other beetles, flies or wasps that I’d be pretty concerned about the potential for syrphid flies to eat aphids coated in dsRNA filled bacteria, for example. I think it’s unlikely that RNAi would be able to be done for humans in this manner because we’re coated in nucleases and to perform RNAi we must actually envelope dsRNA viral components in artificial cell walls to prevent degredation in the bloodstream if we inject RNA into the body as we would with medication used to treat ebola. I’ll go into more detail about this in the next paragraph but even if we found that we could potentially perform RNAi in humans by doing this I wouldn’t expect any big nontarget effects because we could choose the systems interfered with in the insects and avoid using systems humans and insects have in common. We aren’t able to do this with conventional insecticides as well as we could with dsRNA because they often target systems humans and insects have in common like sodium channels and acetylcholinesterase. We do OK by making pesticides less toxic to humans (synthetic pyrethroids have LD50s 10x less than natural pyrethrum for a quick example), but we could always do better.

I’m not sure how big of a problem resistance would be, but I can kind of sort of speculate on this. RNA is difficult for some organisms to take up, so I don’t think it’s impossible for the organism to change its ability to uptake RNA. As far as easily imaginable forms of resistance go, I think this would be the most problematic form of resistance. Increased nuclease activity in the digestive tract would be an issue from a resistance management standpoint, as well. The beauty of this technique is that we can put any sequence of RNA into the bacteria to perform this technique. If we were to target insect specific insulin-like peptides, we could kill the beetles by causing growth deformities or by putting the insect in a diabetic coma. If we found that we could silence some of the metabolic machinery in a species specific manner we could target this. We could target single genes, or groups of genes and thus custom-tailor our pesticides to the pest itself. If the sequence of the RNA changed in response to the management, we could just determine if a different RNA sequence would work. It’s very exciting stuff, and it uses chemistry that’s already existing all around (and even inside) you.

It’s a good example of how technology can be applied in novel ways. In this particular example, we are doing something very simple-genetically modifying bacteria-to accomplish the relatively simple goal of killing crop pests. If we were to develop this further and get it ready for field use, organic agriculture proponents would be sadly unable to take advantage of this technique because they ban both synthetic insecticides and genetically modified organisms. Organic agriculture rejects many tools which could help them further goals which are certainly admirable. Unfortunately organic agriculture proponents attempt to maintain a false dichotomy between synthetic insecticides, genetically modified organisms and environmental issues. A lot of this stems from simple chemophobia, the idea that synthetic things are inherently bad. This causes the field to reject many good tools like this based on little more than fear and misunderstanding. Unfortunately, as a result of this I reject organic agriculture and refuse to buy anything organically produced despite the fact I agree with their goals wholeheartedly. I sincerely hope the field moves in a direction which places an emphasis on environmentally friendly solutions rather than perceived naturalness of interventions. Unfortunately, from what I’ve seen I don’t expect that to happen anytime soon.

ResearchBlogging.orgZhu, F., Xu, J., Palli, R., Ferguson, J., & Palli, S. (2011). Ingested RNA interference for managing the populations of the Colorado potato beetle, Leptinotarsa decemlineata Pest Management Science, 67 (2), 175-182 DOI: 10.1002/ps.2048
Zehnder, G., Gurr, G., Kühne, S., Wade, M., Wratten, S., & Wyss, E. (2007). Arthropod Pest Management in Organic Crops Annual Review of Entomology, 52 (1), 57-80 DOI: 10.1146/annurev.ento.52.110405.091337

Bahlai, C., Xue, Y., McCreary, C., Schaafsma, A., & Hallett, R. (2010). Choosing Organic Pesticides over Synthetic Pesticides May Not Effectively Mitigate Environmental Risk in Soybeans PLoS ONE, 5 (6) DOI: 10.1371/journal.pone.0011250

Kovach, J., Petzoldt, C., Degni, J., & Tette, J. (1992). A Method to Measure the Environmental Impact of Pesticides New York’s Food and Life Sciences Bulletin

Follow Joe Ballenger:
Joe Ballenger is an entomologist who works in the biotech industry as a contractor. In his spare time, he helps answer questions about bugs at Ask an Entomologist.