Spontaneous preterm birth (PTB) (<37 weeks’ gestation) and preterm premature rupture of the membranes (pPROM) are 2 major pregnancy complications that contribute significantly to perinatal morbidity and mortality. Specifically, pPROM is directly antecedent to 35-40% of spontaneous PTB and its etiology includes several epidemiologic, behavioral, environmental, and genetic factors. The interactions of risk factors and overlapping biomolecular pathways make pPROM a complex syndrome. Lack of understanding of the risk factors and risk-induced pathophysiology for pPROM has hampered our ability to develop diagnostic markers for predicting high-risk pregnancies for pPROM or appropriate intervention strategies to reduce its risk.

Human fetal membranes are comprised of a single layer of amnion epithelial layer that lines the interior of amniotic cavity and multilayered chorion trophoblast layers connected to maternal decidua. The 2 layers are attached through an extracellular matrix (ECM) region, rich in various types of collagens providing the tensile strength and structural integrity to the membranes. Type IV collagen-rich basement membrane connects the 2 layers to the ECM. Proteolysis, specifically directed to dismantle the structural integrity of the ECM, weakens the membranes, and causes immunological, mechanical, and functional disruption of the layers resulting in pPROM. In essence, pPROM is a disease of the fetal membranes and currently there are no strategies to reduce or manage the risks to minimize its impact on maternal-fetal health.

Researchers are engaged in addressing 3 major questions related to pPROM: (1) what are the initiators of proteolytic activity resulting in pPROM? (2) what are the effectors (biochemicals) of proteolysis, membrane weakening, and rupture and can these biochemicals serve as markers to predict risk prior to rupture? and (3) if diagnosed early, how do we reduce the risk of pPROM prior to its occurrence and minimize the impact on maternal-fetal well-being?

Careful characterization of the pPROM phenotypes is required to better understand causality and pathophysiology. pPROM can be classified into 3 major groups: (1) pPROM in the absence of cervical change (longer latency); (2) pPROM associated with cervical changes (shorter latency), which is more similar to spontaneous preterm labor with intact membranes; and (3) pPROM involving bleeding disorders or coagulopathies that may be related to placental abruption. Regardless of the pPROM group, the majority of cases are associated with intraamniotic infection and inflammation, as well as clinical and histologic chorioamnionitis. Oxidative stress is an inseparable component of inflammation and pPROM risk factors are known inducers of oxidative stress. The fate of the fetal membranes is determined by the individual’s ability to regulate inflammation and oxidative stress.

The study by Kumar et al demonstrates the role of granulocyte-macrophage colony-stimulating factor (GM-CSF) in reducing fetal membrane structural strength and compromising fetal membrane integrity. GM-CSF activity is enhanced by tumor necrosis factor (TNF)-α, a known inducer of apoptotic cell death, and thrombin, a factor of hematologic dysfunctions in pPROM. Interaction between GM-CSF-TNF-α-thrombin and augmentation of inflammation creates a vicious cycle of events weakening the membranes. An important factor contributing to GM-CSF-mediated mechanical and structural disruption is fetal membrane cell senescence, a mechanism associated with aging of cells. Oxidative stress associated with fetal membrane senescence and senescence-associated secretory phenotype, a unique set of inflammatory markers, are documented factors in normal term parturition. Recent reports indicate that pPROM risk factors result in oxidative stress in fetal membranes, damage to cellular elements including DNA that can lead to telomere attrition, a marker of biologic aging. In pPROM, oxidative stress-induced cellular damage can lead to premature and pathologic activation of senescence. Oxidative stress-induced senescence and senescence-associated changes are likely the initiators in pPROM patients without cervical change and a subset of those with cervical change. GM-CSF, an inflammatory marker, is increased in the intrauterine cavity in response to oxidative stress-induced fetal cell senescence, a mechanism that can structurally and mechanically weaken the membranes. The report by Kumar et al not only explains the mechanistic insult on fetal membranes by GM-CSF leading to membrane weakening and rupture but also demonstrates progesterone’s usefulness in curtailing this pathologic mechanistic pathway of pPROM.

As the authors correctly note, exposure of the fetal membranes to both inflammation and thrombin is from the maternal compartment, closest to choriodecidua layers of the fetal membranes. The pathologic fetal membrane changes associated with pPROM described by Fortner et al include significant thinning of the fetal chorion layer of the membranes in pPROM even distant from the rupture site. This thinning in pPROM patients remained when subjects with histologic chorioamnionitis were excluded suggesting that chorion thinning was an important pathologic feature leading to pPROM. In addition, bacteria presence was characterized and found to be higher in pPROM and inversely correlated with chorion thickness. Using an in vitro system, the results by Kumar et al further demonstrate the importance of the choriodecidua layer in the maintenance of fetal membrane integrity.

While progesterone for PTB prevention has had some success, this success could be improved if we were able to accurately predict which patients would benefit most from therapy and which agents would provide maximum benefit. Critical to advancing this goal is a clear understanding of the mechanism of actions or progesterone therapy in pregnancy. Kumar et al note that the molecular target for progestins in fetal membranes may be the nuclear progesterone receptor expressed in cells of the maternal decidua layer. The fetally derived cells of the fetal membranes including the amnion and chorion lack the classic nuclear progesterone receptor but express nonnuclear progesterone receptors including progesterone receptor membrane component 1 (PGRMC1). PGRMC1 binds progesterone with high affinity, is localized in the endoplasmic reticulum, and is responsible for the antiapoptotic effect seen with progesterone treatment in serum-starved granulosa and luteal cells. PGRMC1 also completely attenuates the apoptotic action of paclitaxel and cisplatin on progesterone receptor–negative ovarian cancer cells. In pregnancy tissues, PGRMC1 expression is reduced in the chorion layer of pPROM fetal membranes compared to nonlabored, term, and preterm fetal membranes suggesting a regulated functional role for PGRMC1 in chorion cells. Further, TNF activation of matrix metalloprotease-9 is inhibited by progestins and this inhibition is mediated through PGRMC1.

The authors describe a common pathway for either inflammation- or bleeding-induced membrane weakening through GM-CSF that is diminished with progestin pretreatment suggesting progesterone’s protective role that is critical to the maintenance of healthy fetal membranes. They suggest that the mechanism for progesterone’s effect on membranes may be through the nuclear progesterone receptor in decidual cells resulting in decreased GM-CSF production or through membrane progesterone receptors, such as PGRMC1, and inhibition of GM-CSF production. It is likely that the actions of progesterone therapy varies by the progestogen and its targeted receptor. Although it is unlikely that inhibition of a single biochemical marker can reduce the risk of a complex phenotype like pPROM, the mechanistic data provided are promising and it is critical that we continue to work to understand the role of various progesterone agents and the mechanisms by which they provide protection from prematurity.