The use and misuse of animal analog models of human pregnancy disorders

It has been suggested that the differences between placentation in humans and rodents, such as mice, are sufficient to render human pregnancy unique and to justify ignoring data generated using mice. Detailed examination of the placenta-decidua interaction and decidual NK cell composition in humans, and mice, show that the principles are the same. Indeed, the rat placenta is useful in showing an intermediary arrangement between humans and mice. This is consistent with the thesis of Darwin that structures of older species evolve with development of new species to provide a survival advantage. Molecular details may differ between species, but also between individuals given gene polymorphisms. Human data on interaction of HLA-C2 with NK cell KIR receptors has been used to suggest that human pregnancy problems such as recurrent miscarriage, fetal growth retardation, and pre-eclampsia are due to lack of activation of true uterine NK cell (TuNK) functions that promote trophoblast cell growth and invasion which prevents such problems. But when TuNKs bear certain KIR phenotypes, pathology results. It is shown that such mechanisms could only be pertinent in less than one-third of recurrent miscarriage patients. Activated blood-type NK cells that enter the uterus (BuNKs) remain the major effector of pregnancy loss in humans, and this is consistent with data from the mouse. The importance of activated BuNKs in pre-eclampsia and fetal growth retardation merits further investigation as pre-eclampsia and fetal growth restriction are also manifest in the CBAxDBA/2 mouse model where activated NK cells are the initiator of abortions.