Sex and death in the cornfields: What is a refuge?

Written by Stephanie Gorski

A lot of people have sent me news articles about the spread of Bt-resistant corn rootworm because they know I am interested in transgenic Bt.  These articles have alarming titles like Voracious worm evolves to eat biotech corn engineered to kill it and Evolution one-ups genetic modificationYes,  unfotunately, corn rootworm that are resistant to transgenic Bt corn are real (Cullen 2013, EPA 2014a). But the story is more complex than you’d think from these headlines. To explain, I’d like to talk about IRM (insect resistance management), specifically, the IRM for Bt corn targeting pest caterpillars (Lepidoptera).  Please note that corn rootworms are not Lepidopterans, but Lepidopterans are a good model for understanding how IRM works.

Spodoptera frugiperda
Fall armyworm, a pest targeted by Bt. Image by Donald Hobern, Flickr Creative Commons.

Insect resistance is a problem with every possible pest management system.  So when you hear doomsaying articles about how transgenic crops cause insect resistance, well, they’re not necessarily wrong, but they’re really missing the point.  All pesticides can result in insect resistance.  Yes, even organic pesticides (Feng and Isman 1995).  Insects can even develop resistance to crop rotation!

Farmers are able to control some pests with crop rotation (such as rotating corn and soybeans). But one species, the northern corn rootworm, has managed to overcome crop rotation in some areas by remaining in the egg stage for 2 winters instead of one, meaning that the fields will likely be rotated back to corn by the time the larvae hatch (Krysan 1984).  Some populations of another species, the western corn rootworm, now like to lay eggs in fields that aren’t corn (Levine 2002), increasing the likelihood that the babies will be born in corn fields.  Clever ladies!
Unlike in the bad old days, when resistance to pest management strategies wasn’t well understood, we are now actively trying to combat this problem of resistance, particularly with Bt crops.  So it’s more newsworthy when insects become resistant to transgenics because we do have such a good IRM program for them.
Around the time transgenic technology started to be adopted, some really smart, cool people invented the refuge system to slow the development of insect resistance.  Let’s look at caterpillar-resistant strains of Bt crops, because that’s the easiest system and the one for which refuges were developed.
Every Bt crop planted in the US must be planted next to a block or strip of non-Bt plants of the same species (EPA 2014b).  The specific percent refuge differs depending on the system, but we’ll say for our example that 10% of a field must be planted with non-Bt plants.  The non-Bt block is called the refuge.
Refuge diagram
Simplified diagram of a refuge. Image by S Gorski.

Say you’re a caterpillar with resistance to Bt, and you grow up on a Bt plant.  When you emerge from your pupa as a beautiful moth, you’ll go looking for love.  You might start your search in the transgenic field, but you’ll get pretty lonely because most of the other caterpillars were killed by Bt.  So you’ll quickly find yourself in the refuge area, where there is no Bt.  Relatively speaking, it’s a busy singles bar over there.
Since there’s no selection pressure in the refuge, your new boy/girlfriend is pretty unlikely to have a gene for resistance.  (This assumption only works if resistance genes are rare in the population to begin with, which is usually the case. If it isn’t, then we have a whole other set of problems besides refuges.)  So your kids might inherit genes for resistance from you, but they’ll inherit genes that aren’t resistant from your boy/girlfriend (Gould 1998).
If you ever had to do a Punnett square in science class, you might remember this means your kids are heterozygous- they’ll have one resistant gene and one nonresistant gene. A gene is “dominant” if you see its effects in a heterozygous individual and “recessive” if you don’t see its effects in a heterozygous individual.  If resistance genes act dominant, then Farmer Joe’s Bt won’t work on your kids.  If resistant genes act recessive, the Bt will work.  The real-life answer is (as always) a bit more complicated (see for example Kacser and Burns 1980).  Most genes aren’t 100% dominant or recessive.  They’re partially dominant or recessive.  So they might have a partial effect, or they might have an effect only under certain environmental conditions.
Resistance Punnett
Simplified Punnett square for dominant and recessive resistance genes. Image by S Gorski.

So Farmer Joe wants to increase his chances that your resistance genes will act recessive, so that his Bt will work.  Can he do that?  Yes, if his Bt is strong enough that it would kill a susceptible insect many times over.  Deciding how much Bt this requires is a numbers game, but it’s generally accepted that if a Bt plant contains 25x the amount of Bt needed to kill nonresistant insects, it will be strong enough to kill an insect that has inherited resistance from only one parent (EPA 1998).
So now Farmer Joe’s Bt is killing nonresistant insects AND insects that have inherited resistance genes from only one parent.  It still can’t kill insects that have inherited resistance genes from both parents (because then by definition they wouldn’t be resistant, right?).  But they’re rare. And because they’re rare, they will almost surely mate with nonresistant insects, so their larvae won’t have a full complement of resistance genes.  And that’s a dead end, because of the high dose of Bt.
Is that a good system?  Yes, it must be.  Because for all the transgenic Bt crops used globally since 1996, there have only been a few cases of resistance- somewhere around four, if you count the rootworms.  If the system didn’t work, we would expect much more resistance than that (Tabashnik 2003)!
All of the cases on record can be attributed to some sort of problem with the high dose/refuge system, as it is called.  In at least one case (Tabashnik 2008), in the southern U.S., lack of high dose Bt expression appears to be a major factor in resistance development.  The other recorded times that caterpillars have been found to be resistant to Bt crops, in Puerto Rico and in South Africa, are suspected to involve suboptimal refugia (Matten 2008, van Rensburg 2007).
Works Cited

  • Cullen, E., M. Gray, A. Gassmann, and B. Hibbard.  2013.  Resistance to Bt corn by western corn rootworm (Coleoptera: Chrysomelidae) in the U.S. Corn Belt.  Journal of Integrated Pest Management 4(3):D1-D6.
  • [EPA] Environmental Protection Agency.  1998.  Transmittal of the final report of the FIFRA Scientific Advisory Panel Subpanel on Bacillus thuringiensis (Bt) plant-pesticides and resistance management, meeting held on February 9 an 10, 1998.
  • [EPA] Environmental Protection Agency.  2014.  White paper on corn rootworm resistance monitoring for Bt plant-incorporated protectants.
  • [EPA] Environmental Protection Agency.  2014.  Insect resistance management fact sheet for Bacillus thuringiensis (Bt) corn products.
  • Feng, R. and M. Isman.  Selection for resistance to azadirachtin in the green peach aphid, Myzus persicae.  Experientia 51(8): 831-833.
  • Gould, F.  1998.  Sustainability of transgenic insecticidal cultivars: Integrating pest genetics and ecology.  Annual Review of Entomology 43: 701-726.
  • Kacser, H., and J. Burns.  1980.  The molecular basis of dominance.  Genetics 97: 639-666.
  • Krysan, J., J. Jackson, and A. Lew.  1984.  Field termination of egg diapause in Diabrotica with new evidence of extended diapauses in D. barberi (Coleoptera: Chrysomelidae).  Environmental Entomology 13(5): 1237-1240.
  • Levine, E., J. L. Spencer, S. A. Isard, D. W. Onstad, and M. E. Gray.  2002.  Adaptation of the western corn rootworm to crop rotation: evolution of a new strain in response to a management practice.  American Entomologist 48(2):  94-107.
  • Matten, S., G. Head, and H. Quemada.  2008.  in J. Romeis, A. Shelton, and G. Kennedy, [eds.],  Integration of Insect-Resistant Genetically Modified Crops within IPM Programs, Springer, New York, 27-39.
  • Tabashnik, B., Y. Carriére, T. Dennehy, S. Morin, M. Sisterson, R. Roush, A. Shelton, and J. Zhao.  2003.  Insect resistance to transgenic Bt crops: lessons from the laboratory and field.  Journal of Economic Entomology 96(4): 1031-1038.
  • Tabashnik, B., A. Gassmann, D. Crowder, and Y. Carriére.  2008.  Insect resistance to Bt crops: evidence versus theory.  Nature Biotechnology 26(2):  199-202.
  • Van Rensburg, J.  First report of field resistance by the stem borer, Busseola fusca (Fuller) to Bt-transgenic maize.  South African Journal of Plant and Soil 24(3): 147-151.

Written by Guest Expert

Stephanie Gorski has a PhD in entomology from North Carolina State University. She has a strong interest in environmental sustainability and conducted research on Western corn rootworm and transgenic Bt crops. She is currently an Assistant Professor at Wiley College.


  1. I have often wondered if bt was the correct way to go after a soil dwelling organism. If bt is naturally occurring is soils. Is it possible that rootworms have long had exposure and were more likely to quickly develop resistance than the earworm?

  2. Hi Eric!
    Excellent point! I hope if anyone besides me has something to add, they feel free to step in.
    You may be right about Bt being a bad choice for control of soil-dwelling insects. However, I’ll point out two things:
    1) Bacillus thuringiensis is also often found to colonize leaves. I can’t find the original paper I read, but here’s a quick reference for you on Bt ecology:
    Maduell et al. 2002. Distribution and Characterization of Bacillus thuringiensis on the Phylloplane of Species of Piper (Piperaceae) in Three Altitudinal Levels. Microbial Ecology 44(2): 144-153.
    2) The fact that rootworms developed resistance to Bt rather fast probably has a lot to do with how rootworm distribution and ecology differs from moth ecology. I’m writing a sequel to this article on that exact topic, and I’ll have it out to you shortly.
    If I think of anything else to add to this puzzle, I’ll let you know. Thanks for reading!

  3. Hi Steph, I thought the problem with the rootworm technology is that the Bt dose is not high enough on rootworms for the high dose/refuge strategy to work long term. This may be confounded by rootworm biology.

  4. You are correct about the lack of high dose, and I will be discussing that as well. There are lots of things that make rootworms a good candidate for rapid development of resistance. I just wanted to give a quick answer, so I said the first few things that came to my head. 🙂

  5. Wow Steph 3 exclamation points for my question. I may never delete this e-mail. Used to work at an isolated nursery near Quincy Fl. and we always had Bt in our pond water. Several have told me it probably got there from mosquito control spraying. This was in mid 80s and mosquitoes were clearly not controlled near that pond. Your opinion? Thanks for posting the abstract. I had no idea they could colonize leaves.

  6. Hi Eric,
    It’s my first post so I’m flattered someone is reading!
    Interesting question. I’m going to give you my answer for tonight and hope I don’t regret it tomorrow.
    I’m no expert on mosquito control, but I’m pretty sure there was no known resistance to Bti in the 80s, so you’re correct to be suspicious if the pond had a mosquito problem. This was the only reference I was able to churn up:
    Paul et al. Journal of the American Mosquito Control Association 21(3):305-309. 2005
    Also, sprayed Bti is not known to overwinter so I don’t believe it would have persisted from a previous mosquito treatment.
    I did manage to turn up a few articles alluding to naturally occurring Bt isolated from water, but they’re few and far between. This one looks promising, but I tried clicking through the references and couldn’t find exactly what aquatic environments the Bt was isolated from. Probably because it’s time for bed.
    Martínez C, Caballero P. Contents of cry genes and insecticidal toxicity of Bacillus thuringiensis strains from terrestrial and aquatic habitats. J Appl Microbiol. 2002; 92(4):745-52.
    So I suspect that yes, you had a naturally occurring Bt population. How exactly was the Bt identified?

  7. Hey Steph, Routine water sampling to make sure we weren’t didn’t have disease inoculant in out irrigation water is all I remember.

  8. Were you able to identify the Bt in the pond as israelensis? That is the serovar active against Diptera larvae.
    Except for things like the briquet (Mosquito Dunk) formulations, most vector control Bti products are effective for only a couple of days before breaking down.
    Bti is not effective against all taxa of mosquitoes. The [i]Culex[/i] mosquitoes are a good example. They are better controlled with Bacillus sphaericus products.

  9. Sorry, Bug Guy, I do not remember what type. Back then I may not have even been aware that there different types. The area is about 25 miles west of Tallahssee and just S.W. of Quincy. So, I was skeptical when folks recently told me that the source was probably from mosquito spraying.

  10. Considering that Gadsden County had a fairly small mosquito control program back when I was in the business in the mid-2000s and even now do larviciding by briquet (effective, but for small habitats and labor intensive), I have my doubts, too.

  11. OK, How in the Sam Hill would you possibly be aware of Podunk,…I mean Gadsden county’s mosquito program or lack there of?

  12. Steph!
    Your article is amazing!
    Only few people can explain easily a difficult concept like refuge.

  13. I think this is part I for part II, right (I went backwards, heh)? I’m on science but from different fields – but the way you have written both articles was pretty straightforward, easy to understand the concepts. Keep it up!

Comments are closed.