A Simple Lab Ingredient Derailed Science Experiments

Last year, in July, Reine Protacio’s experiments suddenly stopped working. Every scientist encounters baffling results from time to time; you chalk it up to error, repeat the experiment, and hope for the best. But in this case, the problem didn’t resolve and in fact spread to other members of the lab: Their yeast, which normally multiples with such intense fecundity that 500 colonies might bloom across a single laboratory dish, had become stunted. Now they were getting just two colonies, maybe three—lonely white dots in a sea of nothing. It was as if something was poisoning the yeast.

After two straight months of failed experiments, Protacio went looking for a culprit. Her lab once had a faulty water purifier, so she switched the water source. No difference. She systematically replaced the sugar and other nutrients for growing yeast. No difference. The mystery, she eventually learned, ran deeper and wider than she thought. And when she and her colleagues at the University of Arkansas for Medical Sciences started sharing her findings, several scientists around the world reported similar stories of ruined experiments. The cases all pointed to the same suspect: agar.

Agar is and has been a staple of microbiology labs for a century. “We buy it in bulk. We buy kilograms at a time,” Protacio told me. Mixed with water, the seaweed-derived white powder forms a sturdy, transparent gel perfect for growing microbes. In my own brief foray into the laboratory as an undergrad, I poured agar into probably hundreds of petri dishes, a tedious but necessary first step for many experiments. The lab where Protacio works uses agar to grow model organisms called fission yeast, whose chromosomes have striking similarities with ours. The bad agar derailed their experiments for two months. Although the lab could recoup the cost of the agar, she said, “they can’t reimburse us for the lost time and the lost productivity.” So the lab started raising the alarm.

In February, Wayne Wahls, who co-leads the lab where Protacio works, wrote to an email list of fission-yeast scientists asking if anyone else had encountered similar problems. One researcher replied yes, and then another. A biologist in Massachusetts even had this agar problem way back in 2006. The more that Wahls, Protacio, and a growing group of other scientists spoke publicly about the problem—in a preprint paper, then an article in Science—the more stories they started to hear. A few of the  scientists joined a study of the agar as collaborators, and the preprint has since been submitted to a journal.

The full pattern of agar failure that emerged is confusing, though. The problems in agar seem to have come and gone not just once but several times, sporadically, over the years—suggesting surprising variability in a standard lab product. They also seem to fade under certain conditions: when petri dishes are kept in the dark, according to one lab, or when yeast are fed a nutrient-rich diet, according to Protacio’s own work. Sunrise Science Products, the company that supplied the seemingly toxic batch to her lab, told me it’s been able to successfully grow fission yeast on the same batch of agar. “Please understand that we are NOT disputing their findings in their experimental situation,” the CEO, Liz Kylin, wrote in an email. Perhaps the problem shows up only in certain batches and under certain conditions, which Sunrise is still trying to understand. “Whatever this issue turns out to be, it is certainly elusive, probably extremely specific,” Kylin wrote.

Scientists have started to wonder if the potential toxicity originated in the seaweed used to make the agar. That could explain the variability from batch to batch: Perhaps certain factors—ecological, meteorological—alter the biochemical makeup of seaweed, the same way a wheat harvest differs from season to season and wine grapes vary from year to year.

Agar is also used in food, particularly in desserts in Asia. (Protacio is from the Philippines, and she originally knew agar as an ingredient in sago at gulaman, a cool, sweet drink that often contains bits of agar jelly.) And laboratory agar actually has its origins in food too: In the 1880s, Fanny Hesse suggested that her microbiologist husband use agar in his work, because she had used it to set fruit and vegetable jellies; her mother had heard about it from friends who had lived in Java. Today, however, culinary and laboratory agar are typically made from different types of seaweed. Agar in food is usually extracted from Gracilaria, which grows readily in large artificial ponds and tanks.

Laboratory agar is a more rarefied product. It comes from Gelidium, a slowly growing wild seaweed that yields a higher-quality agar whose lower gelling temperature is more suitable for lab work. These days, Gelidium is harvested primarily off the coast of Morocco, according to Dennis Seisun and Nesha Zalesny, who run the industry-analysis firm IMR International. The red, frilly seaweed can be collected when it washes ashore, but the finest-quality agar comes from Gelidium gathered from the seabed by professional divers in the summer. “If you can reproduce the waters of Morocco in a pond, the company would do it,” Zalesny told me, but Gelidium has so far resisted attempts at mass cultivation.

The reliance on wild seaweed has caused headaches for labs before. In 2015, a Gelidium shortage caused the wholesale price to nearly triple. But scientists have not, up to this point, been particularly keen to find a replacement for their agar. Seisun and Zalesny used to work for a company that makes gellan gum, an agar alternative that can be manufactured entirely in a factory—no divers needed, no finicky wild seaweed. Yet the product never took off. “Agar still is the king and queen and the gold standard,” Seisun told me. Protacio’s lab ended up switching to a different agar supplier—a cheaper one, actually—and since then everything has been just fine.