Mice pups conceived with in vitro fertilization (IVF) in the lab have slightly increased rates of DNA errors, or mutations, compared to pups conceived naturally, a new study on artificial reproductive technologies suggests.
While the results do not directly apply to humans, they highlight the importance of understanding how fertility treatments affect an offspring's DNA. The research is newly published in the journal Genome Research .
"What we are seeing is a true biological signal, but we cannot make an apples-to-apples comparison relative to what happens in a clinic. Still, the fact that we see this trend in mice does argue for additional research investigating this possibility in humans," said Beth Dumont , a geneticist at The Jackson Laboratory who led the study.
The researchers compared genome sequences of lab mice conceived naturally and mice conceived through assisted reproductive technologies, including hormone treatments, IVF, and embryo transfer. They discovered pups born through these fertility treatments had about 30% more new single-nucleotide variants, or tiny changes in DNA sequences.
Nucleotides are DNA's building blocks or "letters." Arranged in specific sequences, these letters compose the instructions cells use to grow and function. Single-nucleotide variants are simply genetic differences (or mutations) involving a change in just one DNA letter. They can occur when cells replicate their DNA.
The mutations observed in the study are unlikely to be harmful. Scientists estimate that fewer than 2% of new mutations arising in a genome are deleterious or have an impact on an individual's phenotype or disease susceptibility, Dumont said.
The mutations appeared spread across the genome, rather than clustered in particular genes. The timing of when these new mutations appeared in early embryos also looked similar between fertility-treated and natural groups, implying that fertility treatment increases the overall chance of new DNA changes but does not impact when they occur during development.
"Even though we see an increase in mutation rate, most of these mutations are peppered across the genome, and they are neutral mutations that have no impact overall on the organism's phenotype," Dumont said. "Because it's not a huge leap in mutation rate, the risk that any one of those new mutations would be deleterious is really, really small—almost negligible."
Even with a 30% increase in new mutations, the absolute number of harmful new mutations per mouse remains low. For about every 50 mice conceived with IVF, scientists expect roughly one additional harmful DNA change compared to natural conception. That is one problematic change out of many possible ones, since the mouse genome is about 2.7 billion DNA letters long. A similar effect is expected if the male parent's age increased by about 30 weeks, Dumont explained, since paternal age is a major driver of mutation rates in mammals.
The biological mechanisms underlying these genetic changes are not clear. Further research is needed to study whether the new mutations come from a specific step in the IVF process or from the combined effects of several steps. One possible factor is the use of hormone treatments that stimulate the ovaries, since these hormones push eggs to restart meiosis, a stage of cell division known to be prone to mistakes. Other aspects of the fertility treatment protocol could also play a role, such as physical handling of embryos or the chemical conditions of the lab culture environment, Dumont said.
The study does not show whether the same effect happens in humans. Fertility procedures vary between mice and humans, and both have different reproductive biology. For example, mice do not menstruate. Also, people seeking IVF will likely encounter environmental factors that may already have affected their genetics.
"Nothing in our study directly speaks to the potential of mutations in human IVF, but there have been associations in the literature suggesting some steps in IVF might induce genetic changes. That is clearly worth studying to ensure patients are maximally informed," Dumont said.
Other authors include Laura Blanco-Berdugo and Alexis Garretson of The Jackson Laboratory.
This work was supported by start-up funds from The Jackson Laboratory and a Maximizing Investigators' Research Award from The National Institute of General Medical Sciences (R35 GM133415).
About The Jackson Laboratory
