Sunday, August 16, 2015

Labeling animal genes in plants

A while back, a student I spoke to asked me how GMOs would be labeled if an animal gene was added to a plant. She was a vegetarian and was concerned about this. I told her that I didn't know, but thinking about it now, I would have challenged the notion of labeling something as an "animal" gene. In this post, I'm going to explain the concept of common DNA between species, a topic that I touched on briefly in an earlier post. This post will also explain at a genetic level how agriculture is not natural, a topic that I also described in an earlier post, but will go in more detail here.

You may have read that "humans share ~50% of their DNA with bananas" or that we share around 99% of our DNA with chimpanzees. What does this mean? Before I delve deeply into the topic, I need to remind everyone that DNA is the blueprint for proteins. Proteins are the building blocks of our cells. As such, the same protein can be found in different organisms. For example: we have livers and mice have livers, so it would make sense that many of the proteins that are involved in developing the liver and are involved in the liver's metabolic functions would be similar between mice and humans.

As I've mentioned before, changes in DNA are normal. Each one of us has a few differences that we didn't inherit from our parents. These are known as mutations. We tend to think of mutations as negative things: they can cause cancer or genetic disorders. But mutations can also be beneficial. In our species, mutations have allowed for adaptation to high altitude in Tibetans or have protected individuals from heart disease. The same is true in nature: mutations allow for plants to develop resistance to pests, or in the case of weeds, to pesticides.

When a mutation takes place, there are several things that can happen: if the mutation is "bad", then it won't propagate to the next generation. If a mutation is "good", then it gives the organism a selective advantage and it propagates. As an example, if a mutation gives an organism a leg up against a predator, then odds are that the "mutant" will live longer and have more kids. Some of the kids will inherit the mutation, and they in turn will live longer and have more kids. Pretty soon, there will be more individuals in the population that have the awesome mutation, because they've all lived longer and had more kids. After many generations, that mutation might spread throughout an entire population so that everyone has it. Sometimes a mutation is just neutral: it doesn't hurt the organism nor does it improve matters.

Throughout evolution, each species will have developed unique mutations that make that species what it is. Different populations of the same species will have also developed unique mutations to help the population adjust to the region, such as mutations in certain human populations that have allowed them to develop resistance against regional diseases.

So, I thought I'd share an example and I'm taking the easy way out and will share an image from my thesis. In grad school, I was working on a gene and we thought that it might be involved in autism (ASD - or autism spectrum disorder). So I examined the DNA from that gene in a few hundred individuals with ASD. I found quite a few mutations and got really excited, so the next step was to do the same study in controls (i.e. individuals who didn't have ASD). I found that my controls had just as many variations. This graphic has the results from all that work:

Spouse, don't freak out!! I know that it's a really crowded graphic and you want to shut down the computer and never look at the image again, but I'll walk you through it. You're looking at a bunch of rows with really small A, C, T, and Gs. That's the DNA sequence for the gene I was working on in various ethnic populations around the world. What the DNA sequence is doesn't matter for the point I'm trying to make so quit trying to squint at the screen. Each row is a different human population: Japanese, Caucasian, etc. There are also a few other primates included: Chimp and Gorilla. This is commonly done when we study evolution in humans so that we can figure out how ancient a mutation might be. You'll also see a row with letters other than A,C,T,G: that's the protein (i.e. amino acid) that is encoded for by the DNA sequence. Then, you see a few boxes: those are the parts of the DNA/protein that are thought to be crucial for the protein's function. Next, you'll see that there are a few bases that are highlighted: those are the bases that are different between human populations. The shaded letters are what I want you to focus on.

Keep in mind that these shaded letters were in my controls. These individual don't have any diseases (that we know of), and it just goes to show that differences between populations is "normal". But you may also have noted that there aren't any variations in those boxes: the parts of the protein that we think are crucial for its function. There aren't even any differences in these boxes between humans and other primates. So by looking at this example, you can see how similar we are to chimps: much more than we'd like to believe.

As to difference between species (i.e. not just between different populations of the same species), this is what that exact same gene looks like at the protein level between many different species (below). In this graphic, you're not looking for shading, rather, you're looking for stars along the bottom, indicating that there's no difference in the protein in all the different species indicated at that spot.

Published with my own permission. Original is here.  
Again, you can see that the regions of the protein in the boxes have lots of stars along the bottom, indicating that through the millions of years of mammalian evolution between mice and human, not much has changed in the "important bits" of this particular protein.

Of course, there are genes unique to each species and there are scientists who dedicate their research to studying those differences in attempts to better understand, for example, what makes Homo sapiens so different from our other primate relatives.

Now, here's a key point: could you say that this particular gene was a dog gene? Could you say that it was a human gene? Not really. Which one of them is "right"? Which version of the human gene in all the different populations is "right"? The example I've provided is for a gene that is unique to mammals, but there are genes that are shared from bacteria all the way up to humans. There are genes that are shared in plants and humans. So it's very difficult to use the term "belong" when we talk about proteins and genes.

And now we move into the next topic: if you take the rat version of the gene and add it to a dog, is it unnatural?

The evolutionary process that I've described throughout this piece doesn't really apply to agriculture or domestic animals including cows, horses, or common household pets. Think about it this way: if you take a chihuahua and abandon it in the forest, do you think that it could fend for itself? Is the chihuahua best adapted to the environment and predators that a dog faces? Not really. It needs this $2600 Gucci dog carrier to survive in this cruel world. We've bred our pets and selected our crops to meet OUR needs, not theirs. We have bred and selected our dogs to be docile and to shed less, when they should have evolved to be better hunters and faster runners.

The same goes with our crops. We've bred our crops for taste, flavor, and size. Plants in the wild are poison filled, disgusting things that no predator would want to eat. The mutations that we've selected for aren't the ones that nature would have selected had we let her do her thing.

So when someone states that the Innate Potato is unnatural because we've taken a segment of a wild potato gene and have added it to another potato, how is it more unnatural than what we've already done in agriculture? Even transgenesis (taking one gene from a species and adding it to another), happens naturally, as was recently found in the case of the sweet potato where bacterial genes were "added" by nature to the tuber.

Genes are genes: they change, get copied, erased, and broken apart throughout evolution, but we can also choose how to change, copy, erase, and break them apart to meet our needs. These needs are getting more difficult to meet as our environment changes and our population grows, so we should have all forms of trait development available at our disposal, including transgenesis.

Feel free to ask email questions or to comment below.


  1. It's just that more than 99.97% of GM crop acres (ISAAA 2012) are crops that either resist herbicides via insertion of one or more synthetic bacterial genes, crops that produce their own genetically engineered pesticides via insertion of one or more synthetic, chimeric bacterial genes, or both. So when we talk GM crops we're really looking at the difference between the native gene/s in the bacteria as inspiration source, compared with the entirely synthetic, codon-altered often chimeric code that lands in a plant cell. So a simple chat about 'putting in a gene' isn't really a fair description of what actually goes into a GM crop transformation plasmid, let alone into the crop. Together with the random site of insertion, unevaluated genome-wide mutations from the transformation process, tissue culture, antibiotics/selection herbicides, you get a pretty random outcome - maybe 100's of transformants, most of them different.

    1. Could you please clarify what you mean by "synthetic" bacterial genes and "codon-altered often chimeric code"? And what's the difference between a crop that makes its own pesticides because we added the gene for the toxin or a crop that evolved to produce its own toxin? See my post on natural/synthetic toxins:

      The site of insertion is very well characterized and can be found here: As for unevaluated genome-wide mutations, please read this post, where I highlight that transgenesis has fewer unintended genetic consequences than traditional breeding as well as mutagenesis.

      Regarding "100's of transformants", that's simply wrong. I get the impression that you think that scientists transform hundreds of cells lines willy-nilly and just use them all to make GMOs. No: they make hundreds of cell-lines, but ultimately select the single one that has the desired level of expression and has the transgene inserted in a desirable site. All seeds are genetically identical to one another by backcrossing. See: