A variety of 'debris' is shed from the syncytial surface of the human placenta ranging from large deported multinuclear fragments to sub-cellular components. It is increasingly clear that at least some of this material has signalling functions...
So yeah, the "syncytial surface" is the projections into the maternal bloodstream, near enough. Hypothesis says the signals are being sent into the maternal bloodstream to tell our systems to do things that benefit the pregnancy, like to eat what it thinks we should eat, or to sleep when it thinks we should sleep, or to send MOAR BLOOD. The maternal conflict comes about when the placenta starts to try to send more resources to the kiddo than we think it should be getting, because we also want it to fit out.
(of course neither placentas nor we have active intent here - this is entirely genetic.)
I got to talk to the lead author about this work for a while at Saving Grace one year, and the tricky bit is that these micro particles are so small that conventional assays were not picking them up, so they had to design a whole novel method to even detect them, and then had to have independent researchers confirm that this was really happening and wasn't just an artifact of the test design. It looks like they're getting somewhere!