Not long ago I was talking to a fellow environmentalist about fracking. It was her latest big cause, she said — really bad for the environment and on the rise across the United States.
“I don’t know a lot about fracking,” I told her. “I’ve heard a lot of people don’t like it, but yeah…So like what are they actually doing?”
It turned out she didn’t really know. She didn’t know how fracking worked; just knew it was bad because it involved lots of chemicals, oil companies are doing it and she’d heard it was bad for the environment.
We all do this sometimes. We decide whether we like something based on assumptions: what we want to believe, who we trust, what we think is natural, or what the Democrats or Republicans like (because we’re Democrats or Republicans and whatever our Party likes must be good). I tend to do this myself when it comes to political issues because I don’t follow politics very closely, so when I have to make up my mind about some kind of political issue, I tend to jump to knee-jerk conclusions. It’s not because I’m dumb or you’re dumb or we’re dumb, it’s because of time constraints. We’ve all got plenty of demands on our time — work, family, friends, credit card debt, significant other, hobbies, favorite TV shows, writing a science blog (!), playing beach volleyball, watching football, etc., and at the same time we’re expected to make up our minds about issues ranging from Obamacare to Proposition XYZ. We just don’t have time to investigate every single issue out there with the kind of thoroughness it deserves. And so we rely on assumptions, some of which are dumb assumptions if we took the time to stop and think about them. Natural is better than synthetic. Chemicals with long names are bad for you. If big companies like it, it must be bad for the environment. The Republicans care about middle-class values. I want to believe this, so it must be true.
Nowhere is this more evident than in the debate over genetically modified crops (GMOs). If you talk to people who don’t like GMOs, you’ll find some of them have wildly bizarre reasons (they’re destroying our gut flora, enabling corporate operations to take over farming, inserting foreign DNA into your body, etc.). More often, however, you’ll find people who don’t like GMOs don’t have any clear idea what they are. They know GMOS are unnatural, Monsanto makes them and Monsanto is a big company with an ugly history, and from this they deduce GMOs are bad for either us or the environment. Which makes about as much sense as assuming that since UnitedHealthcare has healthcare in their name and pictures of happy patients on their homepage, they must really care about making patients happy, right?
Like I said, we all make up our mind based on assumptions sometimes. But this is not the scientific way to make up your mind. In science you make up your mind based on the evidence. Is the evidence reliable? how was it collected? what does it suggest? and is your conclusion from that evidence consistent with what we know about the laws of nature (e.g. the laws of thermodynamics)? That’s it. We don’t care who disagrees or what their credentials are. I don’t care where you went to school or how many letters you have after your name. The ultimate arbiter in science is the data and what we know about the laws of Nature*. Period.
So let’s start over again and see if we can make up our mind about GMOs based on the science instead of the political crap. Let’s see why and how you make a genetically modified plant and think about what kinds of things could go wrong based on what we know.
*Or at least that’s how it should be. Unfortunately there’s politics in science as in everything else. But this is the way it should be.
So you’ve probably heard of proteins at one time or another. Proteins are all polymers made up of chemical units called amino acids. There are twenty common amino acids most organisms use. They’re all like this:
H2N-carbon with a side chain attached-COOH.
The only difference between these amino acids is the type of side chain they have. You can see what these look like in the diagram below.
Now if you think back to those posts from a while ago on illegal drugs, you may remember that as you go towards the upper right-hand corner of the periodic table, elements get more electronegative, meaning they get more selfish about how they share electrons. Oxygen and nitrogen are much more selfish than carbon, whereas carbon and hydrogen tend to share electrons pretty evenly. A bond where electrons are shared evenly is “nonpolar”, whereas uneven sharing is “polar”, and the more unequal that sharing becomes, the more polar it is. You may also remember that “like dissolves like”; water is very polar, so it likes to dissolve polar compounds.
If you look at these side chains, some of them like the ones in phenylalanine and leucine are made of ONLY hydrogen and carbon. These are not going to get along very well with water at all; they’re pretty nonpolar. Others like lysine or serine have OH or NH2 groups. These get along just great with water.
A protein is a chain of amino acids. When you put it in water, it’s generally going to fold up in such a way that the greasy amino acids like leucine are buried in the interior where they stay out of contact with the water surrounding the protein, while the amino acids that DO get along great with water are all on the outside. That’s kind of the basic principle. It does get a little more complicated than that because these amino acids can interact with each other and so can the backbone of the protein (the part that is NOT side chains), but at the end of the day that’s the basic idea. The protein wants to fold up into a 3-D shape that’s stable because it maximizes favorable interactions like a polar sidechain group exposed to water and minimizes unfavorable interactions like a nonpolar sidechain exposed to water.
When you’re dealing with a protein that has, say, three or four hundred amino acids, those 3-D shapes can get pretty complicated — a little like molecular origami. But at the end of the day the key take-home point, the thing you need to know is this. The sequence of amino acids in a protein determines what kind of shape or structure it’s going to become. And its shape in turn determines what it will do — how that protein is going to behave, how soluble it’s going to be, what it will react with and stick to.
For most of the past century, biologists have known there is a soil bacterium called Bt which produces proteins that kill certain insects. Don’t ask me why Bt wants to kill them; it’s just like Dexter. I don’t really know why he wants to kill people, but the show says he does. And as long as he’s just killing serial killers I guess that’s OK by me. The same applies here. Many microorganisms are embroiled in all kinds of crazy wars with both each other and much larger species. Regardless of the reason, Bt has developed a series of powerful tools for murdering insects: the Cry proteins.
The Cry proteins are very insoluble in water at acidic or neutral pH. Instead of dissolving, they form crystals, just like salt in the bottom of a glass. They only dissolve at high pH, up above 9.5 — the exact conditions found in the midgut of insects in the order Lepidoptera. Under these conditions the Cry protein dissolves — and that’s when it swings into action.
Enzymes that the insect uses to digest its food chop the Cry protein in two. One of the two parts, now that it’s on its own, can fold up into a new 3-D structure that will make it into a deadly weapon. This subunit is called delta-endotoxin. Thanks to its shape it can bind to proteins found on the surface of cells in the lepidopteran midgut. Once several delta-endotoxins come together, they form a pore in the cell membrane; the contents of the cell gush out through the pore and the cell dies. As thousands of cells fall apart and the lining of the insect’s gut disintegrates, the insect loses the ability to feed and slowly starves to death.
Cruel? No doubt. But Nature usually is. The important part from our point of view is that some species of insects in the order Lepidoptera want to eat our crops. The cry protein is highly toxic to them, and about as harmless to us as you can get. The important thing about it and the thing that makes it dangerous is a) the high pH of the Lepidopteran midgut and b) the fact that the shape of delta-endotoxin is a good fit for the shape of of proteins found on the surface of Lepidopteran midgut cells. From this point of view it makes an ideal pesticide, which is why some organic farmers have for many years sprayed Bt bacteria in water on their crops. The problem with spraying, however, is that insects have to eat the bacteria or the protein crystals and that limits its effectiveness more than you might think. Insects that tunnel into the plant, for example, will never eat the Bt, no matter how often you spray the bacteria on your plants.
What we really need to turn cry proteins into the ideal pesticides, then, is a way to put them into the plant. And this is where genetic engineering comes in.
Like any protein, the Cry protein is encoded by a stretch of DNA called a gene. The order of the As, Cs, Ts and Gs in that gene specifies which amino acids the protein will contain and in which order. The gene also contains a segment called a promoter that tells the cell when the gene should be “switched on” or activated and start coding for cry protein.
We can make lots of copies of the Cry gene using a technique called PCR, which makes a gazillion copies of a particular DNA sequence. (PCR is also used in DNA fingerprinting.) Once we have the Cry gene, we can use enzymes biologists have borrowed from bacteria called restriction enzymes that “cut-and-paste” DNA to insert the Cry gene into a small loop of DNA called a plasmid, a little like a tiny fruit loop. Now comes the hard part — getting the plasmid and the Cry gene it carries into a plant cell. (Fortunately, this is way easier than getting DNA into animal cells, which is a whole nother story altogether.)
Getting a plasmid into a bacterium is easy. One of the most common techniques is called calcium chloride transformation. You take some E. coli bacteria and combine them with a solution containing a) the plasmid and b) some calcium chloride (the same salt used for melting ice on driveways) then put it on ice. Next, you take the vial with your bacterial solution and warm it back up to 42 degrees Celsius for a minute or so, then transfer it back to the ice. Some of the bacteria now contain your plasmid. If your plasmid includes an antibiotic resistance gene, you can grow the bacteria on media that includes an antibiotic; only the ones that contain your plasmid will survive and grow. This is easy enough you could do it at home, if you know how to culture bacteria and have some basic reagents/lab equipment*.
So we can put the plasmid into some bacteria; but how do we get it into a plant? Here’s where another bacterium called Agrobacterium comes into play. (It’s kind of amazing, but pretty much all the tools that biologists use to manipulate DNA are tools we borrowed from nature. That’s why the history of molecular biology is so different from the history of chemistry and physics. The history of molecular biology is basically the history of biologists going, “oh wow, there’s this one cool bacterium that makes a protein that does X. Gee, why don’t we take that protein and use it to do X too?”) This bacterium infects plants inserting a plasmid into plant cells. The plasmid contains genes that code for proteins that will force plant cells to make plant hormones and food for the bacterium so the bacterium can form a colony inside plant tissue. That part is bad. But the cool part is that Agrobacterium knows how to take one of its own plasmids and put it into a plant cell.
So what we’ve done is this. We take the Agrobacterium plasmid and use restriction enzymes to cut out the bad genes, the ones for plant hormones and Agrobacterium food. We replace them with the Cry protein gene we copied earlier, then we put the plasmid back into Agrobacterium. Now Agrobacterium, like any bacterium, is not very smart. It’s just a little robot. It doesn’t know that we totally jacked up its little plasmid. It has no idea. It’s going to go out there and insert that plasmid into plant cells the same as it ordinarily would. In so doing it’s no longer accomplishing its own evil mission (because we took out all the genes it was trying to insert). Instead, it’s now inserting a plasmid containing the Cry protein gene, which is exactly what we wanted to do. If we put our Agrobacterium together with plant tissue in tissue culture, we can insert our plasmid and use the plant tissue to grow a plant. Every cell in that plant now contains the Cry protein gene, and when we breed that plant all its descendants now contain the Cry protein gene too.
This is kind of the bare-bones keeping-it-simple version of how genetic modification works. There are other ways to get DNA into plant cells now too, but I believe Agrobacterium remains one of the best and oldest. (Animal cells, again, are a whole nother story.)
The beauty of this is again the Cry protein itself. Remember what I said in Part I about the ideal pesticide? It’s highly toxic to the pest, but not to other species; it breaks down rapidly in the environment, but persists long enough to kill the pest; and finally it has very low toxicity to humans. Bt meets all of these criteria. It’s a protein, so if it gets into the environment somehow it will just get munched up by other bacteria (because there are hungry bacteria everywhere, and bacteria already know exactly how to digest proteins — no problem there). It has very low toxicity to humans. Because it’s inside the plant the only species it can kill are species of Lepidopteran insects that eat the plant! and if you are an insect and you are eating the plant, you are by definition a pest. Finally and perhaps most importantly, there are multiple Cry proteins. We can always genetically engineer a plant with several of them using the same technique and thereby make it very very tough for Lepidoptera to evolve resistance (they’d have to develop resistance to several different varieties of Cry at the same time).
The conclusion is simple: for now, genetically modified plants containing Cry proteins are probably about as close to the ideal pesticide as we can get. But we should as always be careful here and think about what can go wrong.
What could go wrong? Really there are two main possible concerns I can think of. One would be what happens when our crops cross-pollinate with wild relatives (which happens). Now you have wild plants that contain Cry proteins and can more effectively fight off pests, so they can out-compete other plants, which could alter an ecosystem in ways that are difficult to predict. That’s definitely a potential problem worth thinking about. But to my mind the potential for environmental impact there is dwarfed by the environmental impact of conventional pesticides. Imidacloprid slaughters bees; organophosphates and carbamates kill millions of birds a year. I feel like GMOs with Cry proteins are way more benign than most pesticides by any standard.
The other concern is toxicity to humans. Now the Cry proteins themselves are not toxic, although it’s always possible that a handful of people will be allergic to them. This in and of itself isn’t very serious, however, because there are people who are allergic to all kinds of foods. I know folks who say they are allergic to avocados and tomatoes, for example. Nobody would seriously suggest banning avocados and tomatoes because somebody somewhere is allergic to them. Same goes for plants with Cry proteins in them.
But what if there’s something more subtle going on? What if after Agrobacterium inserts the plasmid the DNA gets introduced into a region in the plant genome next to a promoter for some other gene, where it affects the activity of some other gene(s) in a way that somehow made the plant less nutritious (or in another way more undesirable)? This is an interesting question, and yeah, it’s something that the folks in Regulatory already thought of. EPA and USDA require a variety of field tests and animal studies before a GMO can be approved, together with test data to show it is nutritionally equivalent to the “normal” crop. The data strongly suggests this kind of “secondary effect” has not happened and is not happening.
It’s worth remembering how we got all our traditional crops — by breeding, often with wild relatives. Do we know what genes those wild relatives contain? No, not really. Could some of those genes and the proteins they produce be bad from a human health and nutrition standpoint? Sure. By this measure, genetic engineering is actually a more targeted approach.
What do critics have to say in response to the evidence that GMOs with Cry proteins are safe and reduce insecticide use? They complain it’s not natural. But neither are cars or condoms or caffeinated drinks, and I don’t see people giving those up just because you don’t find them in Nature.
So let me put it simply: just saying something “isn’t natural” is not a good argument in my book. There are legitimate reasons not to like GMOs with Cry proteins. “Not natural” isn’t one of them.
You’ll notice I’ve spent a lot of time talking about the Cry proteins and relatively little time talking about other GMOs. Clearly there are companies that want to make all sorts of GMOs. Just because I think GMOs with Cry proteins are a brilliant idea doesn’t mean I support all GMOs. In fact, there are cases where I do not support them.
My favorite example is salmon genetically modified to contain a gene for a hormone that makes them grow faster. Now with Cry proteins the benefit is clear: it helps us win the war with pests in a reasonably low-impact way. But with the salmon, there is every chance that some of these salmon will escape and cross-breed with wild populations, so that now some wild salmon will end up with the gene — and I have no idea what kind of effect that will have on the future trajectory of salmon evolution. I’d be prepared to live with that if I thought this would make salmon farming more environmentally friendly and increase the amount of salmon you could produce per kilogram of food. But to the best of my knowledge it doesn’t. So to my mind it just makes salmon farmers more economically competitive — at the cost of additional impact on the environment. No, thanks.
It’s important to realize that genetic modification of crops is a technique. There will be scenarios where application of this technique is a great idea, and scenarios where it’s not. And there are other scenarios where it will be useless because transferring a couple of genes from some bacterium doesn’t address the problem you need to address. To say that “I support GMOs” or “I don’t support GMOs” is about as specific as saying that you do or do not like a whole genre of music. I like Daft Punk and Kaskade, for example, but does that mean I like all EDM? If I say I love Thrift Shop, does that in turn mean I like all rap music? So people who claim that GMOs are The Way To Save The World from Hunger are just as off-base as people who call them Frankenfoods. This is a technique. It has limitations like any technique, but there are cases (as with the Cry proteins) where it’s going to be incredibly useful.
With that said, however, the people who refer to GMOs as “Frankenfoods” tend to be much more off-base than the supporters. Critics of GMOs have come up with all kinds of nonsensical objections. My favorite is the claim that DNA from GMOs might get taken up by bacteria in your intestines and thereby cause all kinds of problems. Let’s stop and think a minute here, folks. Any time you eat meat or fruits or vegetables, you’re eating cells that contain DNA. If you’re worried that bacteria in your digestive tract could take up “foreign DNA” and use it in ways that compromise your health, then you should really stop eating meat, fruits and vegetables and confine yourself to a steady diet of processed foods. To my knowledge, none of the Nature-Worshippers who object to GMOs have actually done so.
The strangest thing about this whole debate to me, however, is this. I support GMOs with Cry proteins at least in part because I am an environmentalist — because I care about the environment. I think that if Rachel Carson were alive today, she would support them too. And yet it is the same environmental movement she helped found that has done more than anyone else to fight not just GMOs that don’t make sense (like GMO salmon) but all GMOs, in a nonselective and willfully ignorant way. What this suggests to me is that they are opposed to GMOs at least in part because they have no idea what GMOs actually are.
In response to the inevitable objections: no, I do not work for Monsanto and I do not work in agricultural biotech. This is my personal blog, which means it represents my personal opinion and mine only. So no, I am not a “paid shill” for Monsanto or whatever.
*Some people do. There are a handful of hobbyists out there now called “biohackers” who are into this DIY garage-experiments-with-genetically-engineered-E. coli type of stuff. I think that’s pretty awesome. Obviously you could raise concerns about the potential for bioterrorism and the FBI has apparently begun to show some interest in this, but there’s no point getting paranoid; I think the potential for misuse here is far outweighed by the potential for good. Let’s not forget that Henry Ford started out as just this guy tinkering with stuff in his spare time, you know…