How do you define a GMO?

What is the definition of a GMO? This seemingly simple question is difficult to answer, and it’s about to get a whole lot tougher.

It’s no mystery that GMO stands for genetically modified organism, but what exactly does that mean? You might have heard claims like: “we’ve been genetically modifying organisms for centuries”

Historical context for GMOs

That’s arguably true. Modern corn is about as similar to its ancestor “teosinte” as corgis are to wolves. Breeding has allowed us to select for useful traits such as large golden grains or small docile canines.  Modern techniques help to speed up the slow, laborious breeding process. “Marker-assisted breeding” involves taking a tiny chip out of a seed coat, checking to see whether a target gene is present, and growing up only the winners. “Mutagenesis” involves using chemicals or radiation to randomly introduce mutations into a population of seeds, and selecting plants with improved traits. Surprisingly, crops produced this way are not regulated or labeled as genetically modified and can even be certified organic.

Regulatory definition of GMO

So what crops are regulated as GMOs? All of the genetically modified crops currently on the market are “transgenic”. As the name implies, this means that a new/foreign gene has been inserted into these plants. This gene can come from bacteria as in herbicide-tolerant or pest-resistant corn, soybeans, cotton, alfalfa, canola, and sugarbeets. It can also come from a virus as in disease-resistant papayas and squash. Transgenic plants can even contain a gene from another plant of the same species. This is the case for the non-browning potatoes and apples coming to market soon.

Redefining genetic modification with CRISPR

Seems simple enough right? GMO’s have been directly engineered to contain novel genes. That definition has worked for a while, but now there’s a new trick in a breeder’s box of tools called CRISPR-CAS9. CRISPR-CAS9 is a bacterial defense system that  has all the features of Microsoft Word’s find/replace, cut, copy, and paste tools. It allows scientists to target extremely precise regions of the genome and make very specific changes. For example, if a single mutation is known to cause resistance to disease, scientists can now directly introduce that mutation into an existing gene. This change would be completely impossible to differentiate from a naturally occurring mutation. Nonetheless, many are calling to have plants altered in this way regulated as “GMO,” as they are technically genetically engineered.

So what should really qualify as “genetic modification”? Conventional or marker-assisted breeding, which involves shuffling and scrambling whole genomes? Mutagenesis, which causes unkown mutations at random sites throughout the genome?  Trans-genic technology, which involves adding a single gene that, depending on the gene, might have been possible to introduce by breeding on a much longer timescale?  Gene editing by CRISPR-CAS9 which causes one or a few precise changes to genes?

Focusing on the result, not the process

And these are only a handful of the many techniques used to improve crops. I haven’t even touched on grafting, protoplast fusion, polyploidy, or conventional breeding of non-compatible species with chemical assistance! You can see the issue is pretty complicated. Perhaps the real questions should revolve around risks not processes.

Why continue to heavily regulate any crop produced by transgenic technology when every major scientific organization in the world has agreed the technique is no more inherently risky than conventional breeding? And why regulate crops produced by gene editing which causes one or a few precise changes, but not those produced by mutagenesis which introduces many unknown mutations?

For more resources, there is a great infographic by Dr. Layla Katiraee of Integrated DNA Technologies comparing different crop improvement techniques, and a very thorough analysis of the GMO definition problem by Grist science journalist Nate Johnson titled “It’s practically impossible to define ‘GMOs’”