A Minnesota gene-editing company that designed a hornless dairy bull in a lab five years ago found out the hard way that something went wrong in its editing process.

Scientists at the Food and Drug Administration’s (FDA’s) Center for Veterinary Medicine accidentally stumbled onto the editing glitch recently when analyzing the animal’s DNA, highlighting the inherent challenges for companies pioneering the complex world of livestock gene-editing.

Both the FDA and St. Paul-based Recombinetics, which had designed the animal’s DNA in 2014, don’t believe the error is harmful to cattle. But they also don’t want it to happen again.

The federal researchers who made the discovery published an early version of their findings in July that have not yet been peer reviewed.

“We don’t want to say this is important because it sets off alarm bells. We do not draw that conclusion at all. It’s quite possible it could be safe,” said Laura Epstein, senior policy adviser for the Center for Veterinary Medicine. “We really want to encourage the safe use of gene-editing, and we just believe it’s important to look at these things and get this information out to researchers in this field.”

The donor mechanism — called a plasmid and used to repair the DNA break caused by the insertion of the horn gene — was inadvertently copied into the DNA, alongside the newly inserted hornless gene.

At first, Alexis Norris, the FDA bioinformatician who discovered the anomaly, didn’t believe it.

“Like most times, when you find something you weren’t expecting, you don’t trust the result,” she said.

After double- and triple-checking their work, the FDA researchers notified Recombinetics of the findings.

“If something isn’t exactly the way it is intended, we would want to know the potential consequences of that,” said Heather Lombardi, director of the agency’s Division of Animal Bioengineering and Cellular Therapies at the center.

In 2014, when Recombinetics initially checked the results of its alterations made to the bull’s skin cells, the company wasn’t looking for the plasmid because it was supposed to disappear on its own. Those lab-edited skin cells were later successfully cloned twice, creating two hornless bulls, Spotigy and Buri.

Recombinetics wasn’t seeking FDA approval, and the company does not plan to for the first-generation proof-of-concept animal. But Buri, the bull, was an experiment whose DNA information was in the public domain.

This spring, Buri graced the cover of Wired magazine, which reported that his sperm was scheduled to be shipped to Brazil for experimental purposes.

That shipment was canceled after the FDA’s discovery. The presence of the plasmid would officially change Buri’s DNA from non-genetically modified to genetically modified, or GMO.

“We want to go through the regulatory process without the GMO categorization,” said Mark Platt, chief executive of Recombinetics. “It was very unfortunate we didn’t see the plasmid in 2014, but our technology would not allow that to happen today. We haven’t given up on Brazil by any stretch — once we have the animal we want.”

Recombinetics has a new gene-editing method to achieve the same hornless trait that doesn’t rely on the plasmid to deliver the change. Farmers commonly remove horns for worker and animal safety, but some decry the process as painful and inhumane.

The company maintains that setbacks and constant improvements are a natural part of scientific discovery.

“This technology continues to promise great improvement in animal welfare, productivity and disease resistance in a world where those things matter a lot,” the company said.