Among the many mysteries that have vexed scientists about the ongoing Zika epidemic is the question of how, in pregnant women, the virus manages to cross the maternal-fetal barrier.

A woman’s body is usually quite good at protecting her growing baby. There are biological blockades to prevent the transmission of viruses to a fetus through the bloodstream, by way of the placenta; the same path for the nutrients and oxygen that sustain a developing baby. The placental membrane is similarly adept at keeping harmful agents out.

Zika, however, is a strange and troubling exception.

Not only does the virus move through the placenta and infiltrate the amniotic sac; it works its way into the fetal brain and can cause irreparable, even fatal, damage.

Now, scientists are beginning to understand what happens, on a cellular level, once Zika crosses the placental barrier. A new study, published Thursday in the Journal of Clinical Investigation, identifies key cells that enable the virus to replicate by examining how Zika behaved in the tissue of placentas from three women not infected by the virus.

First, researchers found that Zika thrives in special placental cells known as Hofbauer cells—which is surprising, since the cells' typical role is to keep pathogens out. “These Hofbauer cells usually take a viral antigen and create an immune response,” says Michael Simoni, an obstetrician and one of the study’s authors. Instead, with Zika, Hofbauer cells are “physically housing” the virus. (This reinforces a similar finding from May.) Additionally, Hofbauer cells may actually help disseminate Zika to the fetal brain.

Second, the researchers found that Zika is also able to replicate in fibroblast cells, the building blocks that make up the very structure of the placenta. “The fibroblasts ended up being a little bit of a surprise,” Simoni told me. “When we found out they were replicating [Zika] just as well if not better as the Hofbauer cells, that was definitely interesting.”

Another intriguing finding was that three separate strains of Zika—one from Africa, one from the Americas, and one from a past outbreak in Cambodia—didn’t seem to differ from one another in how infectious they were. That’s despite the fact that congenital Zika, which causes babies to be born with abnormally small heads and other health problems, seems to have emerged only with more recent strains of the virus as it has evolved in recent years.

The findings raise a host of questions, the biggest of which is: How is Zika infiltrating the womb in the first place? “Next, we really want to get at the mechanism of how the virus is getting across the fetal-maternal barrier,” says Kellie Ann Jurado, a virologist and the lead author of the study.

Scientists have several hypotheses. It’s possible, for instance, that Zika is piggybacking its way across the placenta via otherwise helpful antibodies—basically taking advantage of a pathway meant to help keep a fetus healthy. Or, it may be that the virus is so tiny it can push its way through a membrane that keeps most other viruses out. “This super-tight layer that usually nothing expect for oxygen and nutrients can cross,” Simoni says. “But, potentially, if Zika is small enough, it can just squeeze in between the cells.”

Other researchers are focusing on how Zika affects the health of the placenta itself. A recent essay in the New England Journal of Medicine describes severe placental degradation among mice infected with Zika in the first and second trimesters of pregnancy. In humans, scientists still don’t fully understand how risky Zika is in different stages of pregnancy. A better understanding of what Zika does to the placenta could be crucial for figuring it out.

Complicating these areas of Zika research is how little is understood about the placenta more broadly. “Given its vital role, shockingly little is known about the placenta,” Denise Grady wrote for The New York Times in 2014. “Only recently, for instance, did scientists start to suspect that the placenta may not be sterile, as once thought, but may have a microbiome of its own—a population of micro-organisms—that may help shape the immune system of the fetus and affect its health much later in life.”

As a virologist, Jurado told me she’s “not always thinking about the placenta,” but that she was stunned when she began to review placenta-related medical literature for her research into Zika. “Seeing all the missing pieces, I was shocked,” she said. “I didn’t understand why there hasn’t been more basic research. It’s definitely needed, especially in cases like this where the outbreak seems to be infecting the placenta.”

Part of the difficulty in learning more about the placenta is the ethical sensitivity of potential research subjects—understandably, expectant mothers aren’t necessarily jumping at the chance to participate in experiments that could yield unexpected outcomes for their unborn babies. But there’s still plenty that can be gleaned from working with the organ shortly after delivery, when a baby no longer needs it, as this recent Zika study illustrates. Research initiatives like the Human Placenta Project, a $42 million NIH undertaking, are increasingly doing just that.

“The placenta is the least understood human organ but arguably one of the most important,” the project’s website reads. “It influences not just the health of a woman and her fetus during pregnancy, but also the lifelong health of both.”