Gene Editing’s Extra DNA Problem: Déjà Vu All Over Again

UC Davis researcher Alison Van Eenennaam described the experience of learning that the “poster animals for the gene-editing revolution” do not, after all, comprise the “same outcome [that] could be achieved by breeding in the farmyard,” like she and her collaborators at Recombinetics had been claiming for years, to Antonio Regalado of the MIT Technology Review by saying simply that “We were surprised, but when you get new information, you proceed ahead, that is what science does.”

My reaction back in the early 1990’s to hearing similar news about extra, unintended DNA being inserted into Flavr Savr™ tomatoes, the poster GMOs for the first generation of genetic engineering technology, was somewhat more intense. My boss at Calgene, Inc. at the time, Bill Hiatt, gave me the news as we were traveling on the subterranean moving walkway system that connects the United Airlines terminals at Chicago’s O’Hare airport. I stopped in my tracks (although I kept moving). It just couldn’t be true, I said to myself; the flashing lights and recurrent United theme song intensifying the surrealism I felt at that moment. (Thinking about it now still gives me a queasy stomach.) I remember asking myself whether genetically engineered (GE; my definition covers gene-editing as well) products were really ready to bring to market when we, evidently, still had a lot to learn about how the technology worked and what unintended changes it could produce.

Like Van Eenennaam’s Recombinetics collaborators, we at Calgene had also initially informed US regulators at FDA and USDA that the changes we had engineered into our tomatoes were very precise, and that we knew exactly what DNA sequences we had inserted into them. And like Van Eenennaam and her animal scientist collaborators, we plant scientists only learned the truth about plasmid vector “backbone” DNA sequences getting unintentionally inserted into our GE products–in addition to the genes we intended to insert–through our interactions with scientists at FDA.

In our case, FDA had asked Calgene to provide experimental evidence to prove our claim that only our genes of interest would be inserted into our GE products, a claim which had been based on 10 years of published reports. It was the results of those experiments that Hiatt had conveyed to me at O’Hare airport. I, with a couple of Calgene colleagues, then published our findings in the Plant Cell in 1994. (Our paper might have been published in a more widely read, general science journal had it not been for the fact that one of its anonymous peer-reviewers seemed not to believe our results.)

In the case of Recombinetics’ GE cows, which had been engineered using one of the new gene-editing methods (TALENs) to prevent them from growing horns, FDA scientists were the ones who conducted the experiments that provided the incriminating evidence. FDA then informed Van Eenennaam and Recombinetics, as well as the Comissão Técnica Nacional de Biossegurança (CTNBio) in Brazil. (CTNBio had been planning–in conjunction with Recombinetics–to create a herd of gene-edited, hornless dairy cows; FDA’s results put an end to those plans.) FDA scientists are now in the process of publishing their results; a preprint describing them was posted online on July 28, 2019.

Will Publishing “Contradictory Results” Help?

It’s important that scientists publish findings like these so that developers of other GE products are aware of them and can, as mentioned by FDA scientists Norris et al. in their paper describing this latest GE “surprise,” employ “screening methods suited to reliably detect the unintended integration of plasmids and multiple template copies.” Because only then, after developers identify engineered/edited organisms that contain DNA they did not intend to insert, can informed decisions be made about whether or how to commercialize those products; for example, in cases like Recombinetics’ cows, in which the unintended DNA contains genes that convey resistance to antibiotics like ampicillin and neomycin/kanamycin, I believe the decision should be to not commercialize.

But publishing is apparently not enough. Some 20 years after we at Calgene published our paper about the “extra DNA problem” associated with the Agrobacterium method of inserting genes into plants, a former Calgene colleague of mine informed me that the scientists at the ag biotech company she was then working for had not been checking their potential products for vector “backbone” DNA that could have been unintentionally inserted during the genetic engineering process…until she brought them up to speed on the issue.

Similarly, papers documenting “unwanted” insertion of plasmid/vector DNA sequences when using zinc-finger nucleases to modify genomic DNA targets in recipient cells had been in the scientific literature for half a decade by the time scientists at Recombinetics published their results on gene-edited cows (see references cited in Norris et al.). Yet those published papers describing unintended insertion of DNA, as well as our paper describing the phenomenon in plants, were apparently not read or otherwise taken into account by the authors of the Recombinetics paper.

The preprint posted by FDA scientists Norris et al. cites additional, more recently published papers that also document unintended integrations of foreign DNA, as well as scrambled DNA at target sites and off-target effects, in various GE organisms that have occurred via use of various gene-editing/engineering techniques, including the CRISPR/Cas9 system.

And so I ask again the question I asked myself in the tunnel between United terminals at O’Hare airport back in 1993: are we really ready to bring gene-edited/engineered products to market when, as described by Norris et al., “our understanding of the unintended alterations” produced via these technologies is still evolving?

Regulation of all GE products–including those that have been gene-edited–should be based on the fact that they were produced using lab-based techniques that we still don’t understand very well

Norris et al. make recommendations for technical ways genetic engineers might rectify what these authors refer to as a “blind spot” in GE product development that was highlighted by their discovery of unintended insertion of vector/plasmid DNA into Recombinetics’ cows.

But FDA should go further. I hope that Recombinetics’ blunder will also convince US regulators that the agricultural biotechnology industry–including developers using the new gene-editing techniques–requires more regulating, not less (as opposed to claims made previously by Recombinetics). Developers of GE products should be required to adequately screen their potential products for these newly discovered failures of precision in the processes of genetic engineering/gene-editing.

For one thing, requiring that all GE products be regulated in this way will ensure that all developers of GE products become aware of this lack of precision in their methods.

For another, regulatory requirements for all GE products could save the ag biotech industry from itself since these kinds of mistakes reflect badly on the entire sector as noted in this August 26th Tweet by Chris Thompson @centrekicker82: “Between Theranos, and now this mess, the impending biotech revolution sure seems a bit… incompetent?”

This entry was posted in Biotechnology and tagged , , , , , , , , , , , , , , , , , , , , , . Bookmark the permalink.

2 Responses to Gene Editing’s Extra DNA Problem: Déjà Vu All Over Again

  1. Mary says:

    Very well explained why GE methods need to be regulated. Even though some of the science was over my head – the gist of the article was clear. Bravo.

  2. Pingback: In Light of Big Mistakes Made by Developers of “Poster Child” GMO Products like Hornless Cattle and Golden Rice, FDA is Justified in Requiring Regulation | Biotech Salon

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s