
èƵs have coaxed human cells to form a miniature replica of the cervix during pregnancy. This so-called cervix-on-a-chip reveals how inflammation and the vaginal microbiome can contribute to premature birth – and identifies a possible treatment to prevent it.
Premature birth – when a baby is born before 37 weeks of pregnancy – affects more than 13 million infants each year and is the second leading cause of childhood mortality and disability. Yet there are no effective therapies to slow or prevent it, largely because there are no animals that replicate human pregnancy and labour, making it difficult to study.
So, at Harvard University and his colleagues developed a cervix-on-a-chip using cervical cells collected from two women who had their uteruses surgically removed. The researchers arranged the cells on a 3D chip in a formation that mimicked the inner cervix’s lining and structural layers. They then coaxed the cells to grow and differentiate using a cocktail of hormones and nutrients. Finally, they bathed them in a mix of human hormones equivalent to the levels seen in the early third trimester of pregnancy, when preterm birth most often occurs.
Advertisement
The final model replicated many features of the cervix during pregnancy, including a mucus plug that develops in the cervical canal. This plug typically forms a barrier between the uterus and the vaginal canal, protecting the fetus from infection. During labour, the cervix shortens and the mucus plug dissolves, enabling the baby to pass through the birth canal.
The device has already yielded insights into preterm birth. Because previous research suggested the vaginal microbiome may contribute to this problem, the team applied bacteria resembling either a healthy vaginal microbiome or one commonly seen in bacterial vaginosis infections to six chips each.
After 96 hours, chips with a healthy microbiome saw a roughly 20 per cent increase in a protein essential for mucus formation. By contrast, those with pathogenic bacteria had a weakened mucus plug and a more than 60 per cent increase in a mucus-degrading enzyme, providing further evidence that the vaginal microbiome plays a role in premature birth.
Further experiments revealed that immune cells pump out inflammatory molecules in response to pathogenic bacteria. These molecules trigger structural cells in the cervix to undergo changes also seen during delivery, which could explain how the microbiome contributes to premature labour.
The researchers traced these effects to one key inflammatory molecule called interleukin-1 beta. A drug that blocks this inflammatory molecule is already approved to treat rheumatoid arthritis. Applying the drug to four cervix chips led to a roughly 77 per cent decrease in mucus-degrading enzymes, compared with untreated chips. This suggests the therapy could be used to prevent or slow preterm birth.
In addition to flagging a potential new treatment for premature birth, the model has also provided further insight into the factors that trigger labour, says at Stanford University in California. “Although it’s not the first effort to mimic the conditions of the cervix in the laboratory, it is, I would say, clearly the most savvy and effective model that has been created.”
Still, it lacks certain cell types and hormones relevant to pregnancy and labour, he says. “Humans are just amazingly complicated. Not only are there many other factors that wouldn’t have been incorporated into this model, but those factors are also changing all of the time.”
This ultimately makes it difficult to know whether these findings apply to premature birth in real people. “It’s not perfect, but you really have to compare it to other models, and there are really no other good models out there” says Ingber. “And the fact that it is human [cells] is, I think, what makes it so relevant.”
bioRxiv