I'll admit it: I've never used Gizmodo's blog composition feature. I don't even know if it's available on all of Gawker or just on Giz. And honestly, I have no clue if it's been around since Kinja arrived or if it's brand-spanking-my-bottom-new. That said...I think I'll try to use it today.
So as the headline implies in the most obvious, least implicating way, this will be about GMOs, and it will be in favor of them. Why? Because one of my Facebook friends is completely, militantly against all GMOs and constantly spreads anti-GMO fear to prove it, and that frustrates me. And she fears GMOs because of many logical fallacies that, as someone who's studied the genetic modification process over several years of university classes and my own research, I just can't handle. I've tried talking science to her, but she doesn't want to hear it. But I hope that around here, my fellow Gizmodo readers actually embrace science in a way that my rant today will inspire rather than annoy.
What are GMOs?
So let's start by answering the most basic question: "What is a GMO?" GMO stands for "genetically modified organism", and it refers to any living thing which has had its genetic code altered artificially—that is, in a way other than spontaneous mutation and sexual recombination. The first thing to note here is that it is a very broad category of things. Not all GMOs are identical, and in fact, many are completely different. For example, the modifications in golden rice are extremely different from the modifications in Bt corn, and those are both completely different from the modifications in the freezer-burn-resistant Flavr Savr tomato. I mention this because given the wide range of modifications, claiming that ALL GMOs are harmful would require testing on ALL GMOs. If you test Bt corn and find it's dangerous (I'll talk about that below), it still would not mean that ANY other type of GMO is dangerous. And yet, many anti-GMO bandwagoners will try to claim that all GMOs are dangerous based on studies of specific types. This isn't science, it's unwarranted extrapolation.
Where do they come from?
Okay, so if GMOs have had their DNA changed...how did they do it? How are GMOs made? Here is where many anti-GMO people have a huge gap in their understanding. Many people have this idea in their head, this image of a group of mad scientists in lab coats just tossing a bunch of poisonous chemicals together, somehow ending up with GMOs, and putting them on store shelves with no regard for safety, no testing, and no ethics, often because companies like Monsanto are apparently bribing them all. I'll discuss the conspiratorial angle in the last section, but for now, just know this is not how genetic science is performed.
So how is it performed? How do we get, for example, the beta-carotene gene into a species of rice which doesn't normally have it? The answer is: a long process that's not very complex to understand, but requires some effort to parse the jargon involved. I want to explain this in detail, but I'm going to try my best to do so in way even the most novice science enthusiast can understand. Let's hope I succeed! And always remember: if you'd like anything clarified, comment below and I'll be glad to elaborate.
How it's done, part 1: Isolation
So the entire point of making a GMO is to take one gene from somewhere and move it into a new place. For example, the beta-carotene in golden rice is produced from two genes, called psy and crtl (abbreviated names, these). Together, they take the lycopene that's naturally in rice and convert it to beta carotene, which is found in carrots (and which your body turns into health-promoting vitamin A). But psy and crtl aren't normally in rice...so where do we get them?
Psy is a protein found in daffodils. Crtl is found in a harmless bacterium called E. uredovora, which is found in soil pretty much everywhere. They normally are dangerous to plants, but are completely harmless to humans. However, a key thing to note is that we're not using the entire bacterium; we're taking just the one Ctrl gene from it and leaving everything else, including its plant-damaging genes and all non-genetic material. It's like removing the trigger from a gun: you have the one part you need, but the rest of the gun, including its dangerous powder and bullets, are stripped away.
Cool. So now we know what genes we need. How do we get them into rice? Well, slow down there, reader! You can't just take a daffodil and a bacterium and put them on top of a rice seed. You need to isolate the genes first: that "stripping away" thing I mentioned. So how do we isolate a gene? The short answer is: we use nature.
Many organisms in nature have these things called "restriction enzymes". In fact, we know of over 800! These evolved as cellular defense mechanisms, protecting their creators from foreign DNA, and they're good at their jobs. Here's how they work, in summary: the restriction enzyme has a piece that bonds with a specific sequence of DNA. When this bonding occurs, the enzyme will literally tear the DNA apart at specific point in that sequence. Some enzymes make a clean cut through both strands, others will cut it in a slightly different place at each strand, creating a Z-shape. This is important: the ends of the DNA strands that stick out of this Z-cut are called "sticky" ends, as they are open and free to bond to a complimentary sequence of single-stranded DNA. Remember your biology lessons: in DNA, a G will (almost) always bond to a C, and an A to a T. That's what "complimentary" means.
So what happens if we take the DNA of a daffodil (isolated through sheer force in a centrifuge) and mix it with several different types of restriction enzymes with several different recognition sequences? The answer: that DNA will be cut up into many tiny, individual pieces of DNA. Some of these pieces will contain that psy gene we want, and the rest will not. So now that psy gene is not connected to anything else...how do we get it away from all the other little pieces of DNA that we don't want?
How it's done, part 2: Finding It
The way we get only the one gene by itself is by using bacteria, specifically E. coli. Now, wait a minute, relax! Despite what you've heard, only very specific strains of E. coli are dangerous. Most are harmless, and in fact, there's probably at least one colony of E. coli living and growing in your gut right now. And you'll never feel them. So don't get hung up on the name—they won't hurt you.
But what does E. coli have to do with anything? Well, bacteria have this very useful thing called a plasmid. These are tiny rings—yes, I said rings—of DNA that exist inside the bacterium but are completely separate from its main genome. But that's not why they're useful. Bacteria have a nice little trick they do to help them adapt: they can take DNA from their environment, mix it into their plasmids, and start expressing those new genes. If you've ever seen the TV show Heroes, it's like what Peter can do, only with DNA instead of superpowers.
That said, what if we took our many pieces of DNA and spread them out onto a colony of E. coli? Those bacteria would absorb the DNA, that's what! But they will only absorb the DNA that happens to find them, meaning the gene you're looking for will only end up in some of the bacteria and not all of them. This is exactly what we want. If we do this and give those bacteria a day or two to grow, we'll end up with multiple bacteria colonies and some of those colonies will contain our gene. We'll know which ones by testing for the protein that gene makes—in our example, psy.
So now we have some bacterial colonies, and we can tell which ones have our gene. While they may also have picked up some other genes, there will be far fewer than we started with, so if we repeat steps 1 and 2 with the plasmids of our friendly bacterial colonists, each time we'll get closer and closer to pinpointing just the one gene we need. Once we do, one more round with the restriction enzymes gives us a clean, uncontaminated sample of our target gene.
How it's done, part 3: MOAR.
Okay, we have our gene all alone. But how much do we need? Is one copy going to be enough? Probably not, and even though we actually have more than one copy by this point, it's still not much. Luckily, we can fix that. There's a process called PCR (polymerase chain reaction, if you like your mouth full of words). It uses a cycle of heat to split the DNA apart (the bonds between strands are fairly weak compared to the bonds holding each strand together) and some enzymes that almost every living thing has to create a complimentary strand for each half. Rinse and repeat, and in an automated PCR machine, you can have many thousands of identical copies of your gene in a few hours. I loved playing with the PCR machine in my biology lab, since I found it fascinating that I could have millions of copies of my own DNA if I waited long enough. But I'm a biomed nerd, and I digress.
Anyway, great, we have tons of copies of just the one gene that we wanted. What's next? Actually...the last step is next.
How it's done, part 3: Splicing
The word "splicing" may seem sci-fi and scary, but it's a real, pre-biomed word that simply means "joining together". And of course, we're trying to join our genes together with the host genome here. So how do we do it?
There are two ways. One way is with what's actually called a "gene gun". Essentially, this does what it sounds like: it uses pressurized gas to literally shoot the DNA into a seed or egg, with the hopes that enough will enter the cell for it to be used by the host. There are two possibilities here: either it will work and do just what we wanted, or it won't work and we'll have the host completely unaffected. Either way, no dangers, unless a failed project is a danger to you. This relatively low chance of success, however, is why the second method was developed.
If gene guns aren't cutting it for you, you can use a retroviral vector to splice your DNA. You hear "virus" and think "DANGER, WILL ROBINSON" (I'm too young to make that reference...), but again, think it through before you panic. What makes a virus dangerous? It has to do with how it reproduces. A virus will reach a host cell, inject its own DNA or RNA into it, and when the host cell starts reading that DNA/RNA, it will make more viruses. These "babies" then break out of the cell and run free, killing the host in the process. It's this reproduction and breaking out that makes viruses dangerous. Without it, a virus would be no more harmful than a speck of pollen.
The viral vectors used in gene splicing are modified so that they can't reproduce. You could quite literally inhale, drink, and inject a gallon of properly-prepared viral vectors and you wouldn't have a single problem. The viruses would inject their DNA/RNA into your cell, but that genetic material wouldn't code for any functional virus. The proteins it makes would just sit there in your cell, doing nothing, until the cell reuses them, no harm done.
So how does this help us with gene splicing? Retroviruses are a class of viruses that are a little "smarter" than most. A retrovirus not only injects its RNA into the cell, but it also includes some enzymes called "reverse transcriptases". These read the RNA, but instead of making proteins, they'll convert the RNA back into DNA...and then stick that DNA into the host cell's genome. Cold sores are an example of a retrovirus: they add their DNA to your own, and whenever your body starts reading that section of your DNA (i.e. when you're under stress), it makes baby viruses that start killing cells in your mouth, hence the sores.
But remember: our viral vectors can't reproduce. Instead, the DNA they're shoving into their hosts' genome is simply our isolated gene. Meaning not only are they harmless, but they can be (and often are) used to insert isolated genes from one organism into the DNA of another. Which is, of course, our final intended result.
How precise is this process? For example, can we really get a single gene with no other DNA around it? Well...not with the most common methods above, no. While we have over 800 restriction enzymes at our disposal, we can't control where any of them cut, so depending on the sequence of our target gene, we may or may not be able to cut directly at the ends. This usually isn't much of a problem, though, because of genetic regulation. Just because DNA exists in a genome doesn't mean it will be translated. In fact, 20% or more (depending on who you ask) of the human genome doesn't actually do anything. Organisms have evolved ways of only transcribing actual genes while generally leaving "junk DNA" (its real name!) alone. So those excess bits of non-gene around our isolated gene are likely not going to be touched and are probably not a problem—at least, no experiment has shown them to cause issues.
But still, it would be nice if we could be more accurate in our slice-and-dicing of DNA. And so we can, with a relatively new development. There's a class of artificial restriction enzymes called zinc-finger nucleases, or ZFNs. These can literally be custom-designed not only to target an exact genetic sequence that we want, but also to cut right where want it to inside that sequence. They've been around since the early 1990s, but only recently have they been developed enough to start showing some real promise. While most natural restriction enzymes can recognize sequences of about 4 or 5 base pairs, ZFNs have been developed that recognize sequences of 9 or more pairs. This greatly reduces the amount of cutting that occurs and greatly increases the accuracy of the cuts. For example, in 2011, a pair of 5-base-pair ZFNs were used on the genome of a C. elegans worm, and the result was almost perfect: only the desired sequences were cut, except for a single cut far away. Compare that to a natural restriction enzyme, which would have made many cuts at various places and none at the exact spot the researchers wanted.
ZFNs are the future of creating more accurate (and thus, more powerful and safe) genetically modified organisms. But even now, a time when ZFNs aren't quite common yet, our GMO processes are much more accurate and much less of a gamble than artificial selection through crossbreeding, which is what humans used to use for similar selective purposes. That process literally shuffled entire genomes around and continued this until it made something we wanted. Now, we can pinpoint only what we want and not have to worry about what ELSE might be getting mixed up in the process.
Is it safe?
Short answer: yes. Long answer: many studies have been performed to test the safety of GMOs. As I mentioned at the beginning of this article, GMOs are such a broad category that you can't claim anything about the safety of one based upon a study of another. That said, the most common study cited to say that GMOs are all bad is one by Gilles-Éric Séralini, a French molecular biologist. Published in 2012, his experiment studied a specific GMO, RoundUp-resistant corn. He used Sprague-Dawley rats as his subjects, feeding his experimental group the GM corn with RoundUp. He went on to show that the experimental groups all ended up with cancers and some had stomach ulcers, and concluded that the GM corn was the cause.
Seems pretty damning for GMOs, right? Well...not really. Once it was published, Séralini's study was quickly peer-reviewed; you know, like all good science. Only his didn't hold up to the testing. In the study, Séralini had neglected to mention that many of the control group rats, who weren't fed the GMOs, also ended up with cancer. In fact, the Sprague-Dawley breed of lab rat is known to have high rates of spontaneous cancer generation—over 80% of males and 70% of females get cancer under normal, non-experimental conditions. The study did not account for this. Statistically speaking, a valid experiment of this sort would need around 65 rats in each group to drown out the noise caused by spontaneous cancer; Séralini only used 10 per group. At a minimum, standards require 20 rats per group for toxicity studies and 50 per group for cancer studies.
The scientific flaws aside, the study contained multiple photos of experimental-group rats with tumors, but not a single photo of a control-group rat. Visual comparisons were therefore left out, and along with the book and movie Séralini released about GMOs around the time of publishing, many biological scientists are skeptical that the study seems to have been engineered as a political tool rather than a valid study. Even the French Society of Toxicologic Pathology, a nonprofit and nongovernmental conglomeration of researchers specializing in toxicology and pathology, criticized the study and denounced its claims.
Okay, so that's the most common study out of the way. Now, do we have studies that prove GMOs are safe? Well...yes and no. You see, "proving it's safe" is the same as "proving there does not exist a connection between GMOs and detrimental health." Science can never disprove the existence of anything; to paraphrase Bertrand Russel, the burden of proof is on the person making a claim. If you can't show that something does exist, it doesn't.
This is basically what's known in science as the "null hypothesis", and it applies to everything. The null hypothesis is the default position on any topic you can ever imagine testing, and its basic form is this: "The thing in question doesn't exist/happen unless there's evidence that it does." When an experiment is run, it's looking for that evidence to reject the null hypothesis. If it can't find any, the null hypothesis remains and we say that the thing doesn't exist or happen.
That said, don't think this is a cop-out. Many studies have been performed on the safety of GMOs, and the majority of valid ones (that is, the majority of those which have stood up to peer review) have found nothing. No link at all. They're what's called "inconclusive"—they found no evidence to conclude anything. Which doesn't mean we don't know if they're safe or not, it means that the null hypothesis stands and we say there is no danger because there's no evidence of danger.
Another problem with making a broad claim that all GMOs are harmful is that in science, and biology especially, there needs to be some mechanism of action. For X to cause Y, there must be some action by which it does. To claim that all GMOs are dangerous therefore assumes there's a mechanism by which a gene from one organism, which is natural and harmless on its own, can suddenly cause health issues when added to the genome of a different, also harmless organism. While it's true that two harmless things can come together to produce something harmful or vice-versa—sodium chloride, anyone?—without a mechanism of action or enough evidence, such a claim can't be made.
Rant's End: Appeal to Nature
So that's it. While GMOs are creating more amazing possibilities each day (i.e. the golden rice I mentioned helping to greatly reduce vitamin A deficiency in much of the poverty-stricken world), people continue to fear them and rally against them without actually knowing what they are, how they're made, or what they do. While this is often because of bandwagoning—that is, they fear it because other people make them think they should—a root cause is the all-to-common appeal to nature. This is a logical fallacy by which people say "if it's natural, it's healthy, but if it's artificial, it's bad for you". Nevermind the fact that "natural" is a murky concept—humans are natural, so why aren't the things we make? And what about, as in GMOs, when we combine two natural things together? Why isn't that natural?—but a bigger problem is that the concept just holds no water. There's no reason why something made by a human would intrinsically be more harmful than something that evolved in nature. Evolution is a random process, but it's had 3.5 billion years to work. Man-made foods and medicines have only had a few centuries or so to work (depending on what you consider man-made), but they're much more targeted and precise. The two even out. There are good and bad things under both categories, and if you don't believe me, then I'll rub some natural poison ivy on you and you can't have any artificial hydrocortisone cream for it.
The only reason artificial things would be more harmful by default than natural things is if you believe that humans have no idea what we're doing. If you think researchers don't test their work, peer review, go through years of safety testing by multiple groups of people, and try to learn more about their processes to improve them over time. If you think we randomly toss chemicals together and hope for the best. Of course, if you're afraid of man-made things for that reason, then you're essentially scared of the Boogeyman, since he's just as real as that kind of fantasy-horror world you've imagined.