Science1 publisher2 min readPublished
Without a pregnancy-only set of brain neurons, mouse embryos fail to implant
Nirao Shah's Stanford team found that female mice need a neuron type making up under 0.001 per cent of brain cells to sustain a pregnancy. Any role in unexplained miscarriage or failed IVF in women is still a hypothesis, because the deletion experiment can only be done in animals.
The Scientist · Science desk
What happened
- The neurons sit in the preoptic hypothalamus, a region deep in the centre of the brain that is also involved in sexual and parental behaviour.
- During a first pregnancy, gene expression in the cells shifted more than 30-fold, turning them into a new neuron class that worked differently by mid-gestation.
- Females engineered to lack the neurons mated and conceived, but every pregnancy ended early because the embryos could not implant in the uterine lining.
- Searching an existing human brain gene-expression atlas, the team found the same set of neurons in the preoptic hypothalamus of women.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint Because the deletion experiment cannot be run in people, any human role for these cells will rest on extrapolation from mice until there is an indirect way to measure them in patients.
- capability If the neurons turn out to release a signalling molecule needed for implantation, O'Chan said that molecule might be turned into a therapy for people with fertility problems.
- precedent Recurrent early loss and implantation failure with healthy-looking embryos now have a defined brain population that researchers can examine as a possible contributor.
The cells turned up in a broad screen. Shah's group was surveying gene-expression changes across regions of the female mouse brain during pregnancy, and one small population stood out as completely transformed [1]. The deletion experiment that followed is informative because of where it failed. Knockout females still mated and still became pregnant [6], so a loss of sexual behaviour or of conception does not explain the result. The losses came at implantation [6]. Cells in the brain appear to influence an event in the uterus.
The effect is large for so few cells. The population is fewer than one brain cell in every 100,000 [1], and Shah and colleagues write that eliminating it completely blocks the ability to establish viable pregnancies [7]. The New Scientist report does not say how many mice were tested, and that number determines how firmly to read "all" [6].
Deletion shows the cells are necessary in mice. Their receptors for progesterone and other pregnancy hormones may explain how they switch on after fertilisation [5]. The thing it doesn't tell you is how a neuron in the hypothalamus signals to the uterus. "Certainly, the fact that when you get rid of these neurons, you lose these pregnancies, is really suggestive," said Jennifer O'Chan of the Icahn School of Medicine at Mount Sinai in New York [8][16]. She said the route is still open: "They've shown that these neurons change robustly during pregnancy, but then how are they actually communicating with the uterus? Is it hormonal? Are they secreting some kind of factor? If they could identify that, I would be very convinced." [9]
The human evidence is thinner. A gene-expression atlas can show that a cell type exists in women. It cannot show what the cell does during a pregnancy. "You can't do these types of experiments in humans, so we have to extrapolate here, but mouse and human brains do tend to be fairly well-conserved," O'Chan said [11].
The clinical interest is in the unexplained cases. Just over half of miscarriages are known to stem from genetic errors in the embryo, which leaves just under half without a known cause [12][2]. Some people having IVF have several genetically healthy-looking embryos transferred, and none implant [12]. Shah's group links its finding to such cases with care: "We speculate that a subset of such events reflects failure of particular neural circuits to recognize, monitor, or adapt to gestation," they write [13]. I think it is a reasonable hypothesis to pursue, on the condition that the uterine signal is found in mice first.
What to watch
- Whether Shah's group or others identify how the neurons communicate with the uterus, by hormone or a secreted factor, the result O'Chan said would convince her.
- Whether the neurons stay transformed after pregnancy or return to their original state, a question O'Chan raised given her own findings of long-lasting brain changes from pregnancy.
- Any human study that ties the state of these preoptic neurons to recurrent miscarriage or to implantation failure in IVF patients.