The current practice of freezing embryos - used to assist reproduction in humans or animals or to conserve endangered species - routinely causes ice to form within the cells, ripping through cell membranes, changing the way proteins behave and ultimately leading to fewer viable embryos.
Now a multidisciplinary team has found that speeding up the freezing process - by 30 times - prevents ice and the damage it causes.
"In developmental studies, we saw that fast-cooled bovine embryos behave much more like unfrozen, uncryopreserved embryos," said Robert Thorne, professor of physics and a Stephen H. Weiss Presidential Fellow in the College of Arts and Sciences. "This is going to have a major impact, especially for the propagation of animals where current cryopreservation methods don't yield good results."
The research published May 16 in Scientific Reports. Masters' students in animal science Abdallah Abdelhady, now a Ph.D. student at University of Missouri, and Jingzhi Zhang, now a Ph.D. student at Peking University, are co-first authors.
Standard cryopreservation involves soaking embryos in cryoprotectant chemicals to prevent ice and then plunging them into liquid nitrogen. While frozen embryos may show little ice, lots of ice forms during the thawing process. Previous work in Thorne's Cornell lab indicated that faster freezing might reduce ice during thawing. To freeze faster, the researchers used concepts developed in Thorne's lab and technology developed at Thorne's company, MiTeGen, to freeze biomolecular crystals for X-ray studies of their molecular structure.
The team worked with bovine embryos, which are difficult to freeze and prone to ice formation, in part because of their large size. After using the new method, the embryos remained ice-free, even when cryoprotectant concentrations were reduced by 30%.
The method could improve the cryopreservation process in human IVF (in vitro fertilization) - providing more consistency across clinics and better outcomes. Thorne said the bigger impact may be in breeding livestock and other domestic animals, which have lower success rates using current methods. The method could also be used for endangered species conservation or biomedical research.
"There are a lot of endangered species where we want to save the embryos to preserve biodiversity, genetic diversity," Thorne said. "In biomedical research, scientists make genetic variants of research animals and may want to preserve those stocks. More broadly, you could think about using this method for stem cells, thin tissue samples - there are a lot of applications of cryopreservation that could benefit from our methods."
After the bovine embryos thawed, the researchers, led by the late Soon Hon Cheong, professor of clinical sciences in the College of Veterinary Medicine (CVM), compared their development with embryos that had never been frozen, finding that the embryos that had undergone ultrafast cooling developed nearly as well as the unfrozen, "fresh" embryos. The fast-cooled embryos fared significantly better than those frozen at standard rates and led to successful pregnancies. Cheong died in December 2025.
"Dr. Cheong played a critical role in this," Thorne said. "The choice of the bovine model, procurement of ovaries, production of embryos, the preparation for freezing, the actual freezing at my company, thawing, post-thaw incubation and developmental characterization - those were all his piece, as well as coordinating the embryo transfers that resulted in pregnancies, evidence that the cryopreserved embryos were viable."
Cheong also brought in Jingyue Ellie Duan, assistant professor of animal science in the College of Agriculture and Life Sciences, to further verify and deepen the results using genomic analyses. Led by Duan, the genomics team looked at entire transcriptomes to reveal which and how much of more than 10,000 genes were expressed after the embryos were frozen and thawed. They discovered that genes associated with DNA damage repair pathways were more highly expressed in the embryos frozen at standard speeds and protocols.
"That indicates that there is some damage happening that the cells want to work really hard to repair," Duan said. "There's evidence that all the embryos underwent some kind of stress, but only the group cooled with standard protocols showed that actual DNA damage response."
By eliminating ice formation, Duan and Thorne hope they'll be better able to assess other impacts of the cryopreservation process - like the effect of cryoprotectant chemicals.
"It has never been possible to study what those non-ice-related impacts are in a definitive way," Thorne said. "Part of what Ellie's results are telling us is that we still have transcriptomic impacts not associated with ice, so now we can study what those are."
The team is continuing to experiment with the cryoprotectants, to find the lowest-possible concentrations that still prevent ice formation. The breadth of the genomic analyses provides biomarkers for more focused study.
"If we want to further improve the process, the technology, the media that's added during cryopreservation, these genes can be the markers," Duan said.
Thorne and Duan will carry on this work with Cheong's wife and co-author, Yoke Lee Lee, who manages Cheong's former lab, now led by Scott Bailey, associate professor of clinical sciences (CVM). Duan said the lab provides crucial training in embryology for students and faculty. Duan, who has always worked at the molecular level, described seeing time-lapse video of embryos' growth for the first time in Cheong's lab.
"They were expanding and contracting, like they were breathing, and it was the first time I was able to visualize it clearly - it was an amazing experience for me," Duan said. "Because of Cheong's expertise in embryology and his platform, we're able to use this facility to continue multiple projects, and it's all because he established it."
Additional co-authors include assistant professors of clinical sciences Mariana Diel de Amorim and Ian Porter (CVM); lab technician Chia-Kuan Wu M.S. '24, former lab technician Patrick Crane '18, postdoctoral researcher in animal science Meihong Shi and CVM clinician Qiudi Zheng DVM '20.
Funding for the study came from the USDA's National Institute of Food and Agriculture, the National Institutes of Health, the Cornell University Center for Advanced Technology and the Cornell University Center for Vertebrate Genomics.
