Systematic review of p38 mitogen-activated kinase and its functional role in reproductive tissues

This systematic review was written following the standards set by the MOOSE group and PRISMA statements and was registered in PROSPERO: registration number 42016033380. Articles for this systematic review were published between January 1980 and August 2017 in English and came from 4 databases: Web of Science, Pubmed, Medline, and CoCHRANE. Articles were found with the help of a library liaison. Our research strategy was designed to fill the knowledge gap of p38 MAPK’s role in the intrauterine compartments (placenta, chorioamniotic membranes, decidua, myometrium, and cervix) after implantation and through delivery (See Appendices 1 – 3 for full details of the search). This review is restricted to studies focusing on p38 MAPK in humans and all animal models of pregnancy. Original research studies selected investigated p38 MAPK in various intrauterine compartments (placenta, chorioamniotic membranes, decidua, myometrium, and cervix) after implantation and during pregnancy and parturition. Studies were excluded if they were not related to p38 MAPK, not related to reproductive tissue, if they were review articles, if they had a below average quality score, or if no full text was available online (published as an abstract only) or the full text was not available from authors upon request. Studies that collected samples from female patients at full term gestation (≥ 37 weeks) who were either in labor or not in labor at the time of specimen collection for assay and analysis. Studies were also included if results were solely related to patients with preterm labor and delivery (both spontaneous and induced), chorioamnionitis, or studies that utilized cell lines from gestational tissues. Any age or patient ethnicity was included. Animal studies, including samples collected during and at the end of term or preterm birth studies, were also included. This study measured p38 MAPK’s function and mechanistic role during pregnancy and parturition in human and animal models. Included studies assessed 1) changes in p38 MAPK RNA or protein expression, 2) its activators and repressors, 3) the mechanisms of activation or inactivation, and 4) biological pathways impacted in response to changes in p38 MAPK’s function ( Table 1 ). All articles identified through the search strategy mentioned above were downloaded into a common folder, and duplicate articles were removed. Two authors (L.R., S. S-M.) independently screened the titles and abstracts, removing reviews and nonrelevant articles, and selecting those that were potentially appropriate for full text analysis. A data extraction form was created and information from the articles was collected including primary author, year and journal of publication, tissue, type of study, methods, inhibitor and isoform information, relevance to parturition, and key findings about p38 MAPK. Four authors (L.R., S. S-M., L.M, J.J.) independently collected data from the articles, discussed their findings, and created a single document containing all article information ( Table 1 ). No quality assessment tool for systematic reviews has been well defined for basic science research. Hence, we followed the guidelines outlined by Hadley et al, 19 which specified criteria for quality assessment of such works and ranks them for quality. Areas, such as hypothesis, strength of study design, description of samples and reagents used, methods replication and reproducibility, experimental rigor as defined use of proper control, adequate power, assay and analytical approaches, appropriate use of statistical tests, sufficient explanation of results, and if the results support the conclusions ( Appendix 3 ), were analyzed per article, and a score was given ranking them as poor, acceptable, or good quality. Based on our data extraction from all included articles, we present a better comprehensive understanding of p38 MAPK’s role throughout pregnancy in the intrauterine cavity (placenta, chorioamniotic membranes, decidua, myometrium, and cervix) of human and animal models. We also present the overlapping and unique functions of p38 MAPK in each of these tissues.

p38 MAPK was first reported in reproductive tissue in 2000 when Peter et al and Scherle et al studied p38 MAPK’s role in trophoblast and endothelial cell differentiation. 20 , 21 Since then, more than 72 articles have been published highlighting p38 MAPK as the main regulator of many processes that occur throughout gestation and during preterm or term parturition. Specifically, the rate of publication increased 3.5-fold from 2000 to 2016 ( Figure 2 ), including documentation of p38 MAPK in more than 31 different reproductive cells or tissues, supporting p38 MAPK’s diverse and important role during pregnancy ( Figure 3 ). The main characteristics and findings of the included studies are summarized in Table 1 . Studies included in this systematic review discussed p38 MAPK function or mechanism of action in either the placenta, chorioamniotic membranes, decidua, myometrium, or cervix and various combinations of the gestational tissues ( Figure 4A ). Over the past 37 years, only 1 report (Wang et al) studied p38 MAPK expression, but not its function, in cervical tissues. 10 Functional and mechanistic analysis of p38 MAPK in human and animal models ( Figure 4B ) were analyzed by a variety of techniques (often performing multiple approaches in the same study for confirmation purposes) including Western blot (71), immunostaining—histochemistry (5) or cytochemistry (1), flow cytometry (1), ELISA (1), and bioplex (2) ( Figure 4C ). Mechanistic participation of p38 MAPK was analyzed by studying what proteins or pathways lead to its activation, whereas functional changes associated with p38 MAPK activation were reported by measuring signaling pathways or the endpoint attained (eg, cell death, inflammation, and prostaglandin synthesis). Total forms of p38 MAPK were measured by Western blot, and immunohistochemistry, while phosphorylated active forms were measured by Western blot immunohistochemistry, flow cytometry, ELISA/bioplex, and immunocytochemistry. Though there are 4 different isoforms of p38 MAPK (alpha, beta, gamma, delta), 65 of the included studies did not document the specific isoform studied ( Figure 4D ). Additionally, 13 studies did not use total p38 MAPK to normalize phospho-p38 MAPK data when Western blot analysis was performed, making it difficult to properly interpret data. It is highly recommended to perform normalization of active p38/total p38 over other housekeeping genes for appropriate biological assessment of changes associated with p38 MAPK activation. To add scientific rigor to experiments, most investigators have used inhibitors to p38 MAPK alpha or beta, such as PD169316 (3), SB203580 (35), SB202190 (13), SR141716 (1) or SB239063 (1), ( Figure 4E ). Surprisingly, no studies used p38 MAPK siRNA, leaving 19 of the studies in this systematic review to be mostly observational. The reports evaluated for this review are basic science or research laboratory-based studies, and no reference criteria exist to assess quality. However, based on the reported standards of quality assessment, we used a quality assessment tool established by Hadley, et al, 19 which ranks articles by quality for each of the included articles ( Appendix 3 ). Based on the criteria described in methods section, 55 articles were graded as above average quality, 42 as average quality, and 11 as below average quality. During the quality assessment, additional articles were identified that did not meet our inclusion criteria outlined in the methods (Not related to pregnancy or pregnancy timeline (9), Not related to p38 MAPK (11), Review articles (1), No full text available (2)) and were removed prior to the final analysis. Mechanisms and functions of p38 MAPK in various uterine tissues/cells were determined based on data extracted from 74 studies ( Figure 4 ). The majority of included studies reported on a single cell or tissue type and used more than one assay type for analysis. As shown in Table 1 , various gestational tissue types were studied: placenta (27), chorioamniotic membranes (14), decidua (8), myometrium (13), and cervix (1). Upon review, these reports suggested that p38 MAPK plays a role in gestation, specifically decidualization and trophoblast differentiation and invasion (28); parturition (21); adverse pregnancy outcomes like preterm birth, related and not related to infection (10); and myometrial quiescence or activation (8) ( Figure 4F ). All of these studies were carried out in vitro (cell culture or tissue explants) using human derived samples (65); however, some studies were carried out in vivo using rodent models (9). p38 MAPK’s activation and function are stimulant, tissue, and time-dependent. The following findings determined the diversity and importance of p38 MAPK in the intrauterine cavity throughout pregnancy and during parturition. The most studied upstream activator of p38 MAPK was different for each tissue type, namely, oxidative stress in the placenta and chorioamniotic membranes, proinflammatory cytokines in the decidua, and oxytocin signaling in the myometrium. The most studied downstream effects of p38 MAPK were cell fate in the placenta and chorioamniotic membranes, prosenescence and inflammatory cytokine production in the decidua, transition from quiescent to active state of the myometrium, and cervical ripening in the cervix ( Table 2 ). The cervix, 10 ovaries, 22 placental arteries, 23 and umbilical cord stromal cells 24 were understudied based on our review. Additionally, we did not examine vaginal expression of p38 MAPK as a part of this study. Tissue type breakdown and findings are discussed below. In studies related to placental cells, p38 MAPK was activated by hypoxia, LPS, and hydrogen peroxide 25 – 32 leading to the stimulation of cell fate pathways like cell survival 21 , 25 ,31−39 ( Table 2 ). The activation and function of p38 MAPK in the placenta was documented in villous explants and primary and immortalized cytotrophoblasts cells. Other p38 MAPK functions reported in placenta include cytotrophoblast differentiation and invasion during early gestation. 21 , 35 , 39 – 43 During differentiation, growth factors, such as epithelial and fibroblast growth factor (EGF & FGF), 21 , 44 can activate p38 MAPK, 45 potentially via protein kinase A (PKA), 35 increased placental lactogen-I, 21 human chorionic gonadotropin (hCG), and fusion genes 35 to establish pregnancy. Conversely, Costa et al showed endocannabinoid 2-arachidonoylglycerol can block p38 MAPK activation and prevent protein expression in syncytiotrophoblast. 43 Maternal serum factors and immune cell derivatives activate p38 MAPK, causing cytotrophoblast cell invasion in vitro, 42 , 46 whereas decorin 41 secreted by the decidua prevents cytotrophoblast migration by inhibiting p38 MAPK. Throughout gestation, p38 MAPK function is regulated to promote cell survival. Cindrova-Davies et al and Jessmon et al reported that hypoxia directly, or via heparin-binding EGF-like growth factor, activates p38 MAPK. This activation of p38 MAPK promoted anti-apoptotic factors 29 , 31 or promoted cytotrophoblast proliferation. 36 , 37 Besides hypoxia, p38 MAPK activation is also associated with hCG and leptin production in cytotrophoblast. 34 , 35 , 38 p38 MAPK is also reported to enhance placental immune response in vitro in inflammation models, 27 causing cytokine 30 and uterotonin production 28 , 47 at term and preterm parturition. Jiang et al found that prostaglandins activated a p38 MAPK-mediated increase in progesterone receptor A (PR-A) in the placenta, 48 a mechanism associated with functional progesterone withdrawal in the myometrium. However, this novel role of p38 MAPK has not yet been determined in the myometrium. Studies on p38 MAPK’s role in chorioamniotic membranes has mostly been studied in association with parturition ( Table 2 ), partly due to the impracticality of obtaining chorioamniotic membranes (chorioamniotic membranes) during gestation until after delivery. Placental studies have overcome this limitation by using placenta from aborted fetuses; however, membrane researchers have not attempted such early studies, thus creating a black box regarding p38 MAPK’s function during formation, development, remodeling, and expansion of the chorioamniotic membraness. Though chorioamniotic membranes researchers only study p38 MAPK activation and function in a short window of time, theirs are the only intrauterine tissue studies to document p38 MAPK using Western blots, RT-PCR, ELISAs, immunohistochemistry, and immunocytochemistry. This documentation should set the standard for researchers in other uterine tissues, which normally use only Western blots to determine p38 MAPK expression and function. Though chorioamniotic membranes comprise multiple layers, the prominent cell type studied is amnion epithelial cells (AECs). Higher natural levels of p38 MAPK are documented in cases of PTB 49 and in the region of the chorioamniotic membranes that overlap the cervix. 7 Recently, multiple laboratories have reported OS induced pathways of parturition in chorioamniotic membranes directed by p38 MAPK. Regardless of species (mice 50 or humans 14 , 21 , 51 , 52 ) or pregnancy conditions (PTB 28 , 53 or term labor 14 , 21 , 51 , 52 ), all these conditions support a role for p38 MAPK. Studies showed OS causes p38 MAPK-mediated senescence of chorioamniotic membranes, sterile inflammation, and prostaglandin production 28 to facilitate parturition. In addition to the OS induced pathways, oxytocin, 50 environmental toxins, 54 and lipid mediators 55 were all reported to activate p38 MAPK. In chorion cells, OS induces p38 MAPK activation, leading to production of PGE2, 56 Heme oxygenase 1 downregulation, 57 pro-inflammatory cytokine increase, 56 and apoptosis. 57 In summary, multiple stimulants at term activate p38 MAPK in chorioamniotic membranes cells to affect labor-inducing pathways. p38 MAPK’s role in the maternal decidua during gestation and at parturition was documented in endometrial stromal cells that were undergoing or had undergone decidualization. The role of p38 MAPK in decidualization was reported by Yoshino et al, Lockwood et al, Scherle et al and many others ( Table 1 ). In decidua cells, pro-inflammatory cytokines and chemokines induced p38 MAPK causing apoptosis, 33 cell survival, 32 angiogenesis, 58 inflammation, 59 collagenolysis, 58 and senescence ( Table 2 ). Therefore, p38 MAPK is reported to support cell survival and cell death, likely mediated by distinct stimuli. Only 2 reports documented decidual cells role at term and preterm parturition. In mice, Bonney et al showed association between p38 MAPK expression and decidual senescence at term, 17 while Li et al reported lack of a role for decidual p38 MAPK in immune cell infiltration during preeclampsia. 60 The role of p38 MAPK in maintaining myometrial quiescence and inducing contractions was reported using primary and immortalized myocytes, myometrial tissue, and uterine artery cells. Oxytocin, a hormone released by the pituitary gland during labor, was a reported activator of p38 MAPK through the oxytocin receptor in the myometrium ( Table 2 ). Kim et al reported that the oxytocin-oxytocin receptor complex activates p38 MAPK and induces COX-2 and IL-8 in myocytes. Xu et al validated these findings, showing PGF2α production mediated by p38 MAPK 61 through the same mechanism. A new oxytocin receptor antagonist, nolasiban, was shown to inhibit NF-κB and p38 MAPK signaling, downregulating cytokine production in myometrial cells. 62 Along with oxytocin, reports documented p38 MAPK activation by pro-inflammatory mediators that stimulates myometrial activation. 38 , 50 , 61 – 65 Specifically, IL-1β and interferons 66 have been documented as activators of p38 MAPK in myometrial cells, which induces prostaglandin synthesis 67 , 68 , 69 and inflammation. 66 Activation of these signaling pathways creates a feed-forward loop, perpetuating pro-inflammatory cytokines and COX-2 69 production associated with parturition. 61 , 64 , 70 Arrestins are a small family of proteins important for regulating signal transduction of G protein-coupled receptors (GPCRs) upstream of MAPK activators. Brighton et al reported arrestin 2 and 3 negatively regulate oxytocin-induced p38 MAPK activation by downregulating oxytocin receptors in myometrial cells, thus maintaining a quiescent phenotype. 71 Willets et al validated these findings using bradykinin, a potent pro-inflammatory peptide involved in infection-induced inflammation (to mimic preterm labor), to induce p38 MAPK signaling. 65 Though cyclic adenosine monophosphate (cAMP) is documented as a muscle relaxant 64 it has been shown to perform dual roles in myometrial activation and quiescence through p38 MAPK. Two reports have identified cAMP-mediated activation of p38 MAPK, 64 , 72 which in turn induces corticotropin-releasing hormone secretion and COX-2 expression; both are key components of labor initiation pathways. 64 , 72 These reports suggest p38 MAPK dependency of cAMP’s dual role in myometrial quiescence and activation. In the myometrium, p38 MAPK activation induces a pro-inflammatory environment associated with parturition, and inhibition of p38 MAPK maintains myometrial quiescence. However, there are no current reports on potential interventions for preterm labor that inhibit p38 MAPK activation or function in the myometrium.

p38 MAPK is a multifunctional signaling kinase. It is a responder to stress stimuli performing various pleiotropic functions ranging from maintenance of cellular homeostasis to contributing to cellular dysfunction, depending on the tissue environment. p38 MAPK helps to maintain pregnancy and promote parturition, and its untimely activation or overactivation may cause adverse pregnancy outcomes. A better knowledge of p38 MAPK’s function in various uterine tissues during normal and abnormal conditions may increase its potential as a target for intervention in adverse pregnancy outcomes. While the number of p38 MAPK publications has increased recently, current literature is still insufficient to provide a comprehensive description of p38 MAPK’s role in gestation and parturition in various uterine tissues. More basic research is essential to generate a comprehensive functional role of p38 MAPK, especially in under-reported tissues, such as the cervix, ovaries, placental arteries, and umbilical cord stromal cells, where its expression, activation, and function are relatively unknown.

Our systematic review documents p38 MAPK as a key intermediate kinase responsible for regulating signaling cascades during implantation, pregnancy, and parturition. These processes included decidualization, myometrial activation during parturition, placental growth, cervical ripening, and parturition at term and preterm. Our systematic review emphasizes challenges in determining the precise function of p38 MAPK due to redundancies in the activators and effectors in specific tissues, leading to contradictory reports on p38 MAPK function. Some of the challenges include 1) finding a specific upstream activator and downstream signaling molecule that correlate to the biological function of interest, 2) identifying the temporal activation of p38 MAPK in vitro and in vivo, 3) determining the functional isoform under a specific cellular environment, and 4) choosing the right inhibitor and model to study p38 MAPK’s functional role in different tissues and species. Knockout of p38 MAPK is embryonically lethal; therefore, functional studies using p38 MAPK knockout models are not attempted. Dependency on in vitro models alone to understand p38 MAPK function is challenging due to obvious limitations of such models in replicating the in vivo tissue microenvironment. Though our reports document p38 MAPK’s role in many intrauterine tissues during pregnancy and parturition, we also identified several knowledge gaps that need to be filled through future research. The most commonly reported downstream function of p38 MAPK, regardless of tissue, was activation of proinflammatory cytokines; however, the pathways by which p38 MAPK induces inflammatory activation are still unclear. Interestingly, NF-κB activation by p38 MAPK is well reported in various other systems, 73 – 75 but this mechanism was not reported in any of the uterine tissues in this review. It has been shown that p38 MAPK causes placental PRA-PRB ratio switch to favor progesterone-PRA mediated action, a well-known proinflammatory mechanism. The role of p38 MAPK as a mediator of functional progesterone withdrawal in the myometrium, where this ratio switch is considered a prerequisite for labor initiation, 76 – 78 has not yet been reported. p38 MAPK exists in 4 isoforms—alpha, beta, gamma, and delta—with tissue-specific functions. Isoform-specific function of p38 MAPK in reproductive tissues has yet to be fully elucidated, making it difficult to determine the exact functional form of p38 MAPK in a given cell. The gamma isoform may have a role in myometrial function as it is known to be expressed in muscle tissues 79 while the delta isoform might play a role in brain development. The knowledge of isoform function is essential in designing molecules that can inhibit p38 MAPK function to avoid any deleterious effects caused by its premature activation. Knowledge gaps still exist in p38 MAPK’s role in placental growth and development and fetal membrane collagen and cellular remodeling during gestation. Additionally, cervix, ovaries, placental arteries, and umbilical cord stromal cells were understudied based on our review. Immediate research attention is needed in understanding the functional and mechanistic role of p38 MAPK in these tissues during pregnancy and parturition.

Reactive oxygen species (ROS) arising from intrauterine redox reactions play a dual role in reproductive tissues, specifically, uterine tissue remodeling during pregnancy when ROS are balanced and promoting parturition when there is a redox imbalance. The intrauterine oxidative status changes during various stages of gestation; it is mostly hypoxic during the first and third trimester, but it becomes hyperoxic during the second trimester and at term. 1 , 2 Reactive oxygen molecules generated from oxido-reductive reactions are essential for fetal and placental tissue growth and pregnancy maintenance. Recent studies have shown that increased oxidative stress is essential to trigger labor and delivery at term. 3 Premature activation of oxidative stress during gestational periods can lead to adverse pregnancy events, like stillbirth, intrauterine growth restriction, preeclampsia, premature rupture of chorioamniotic membranes, and spontaneous preterm delivery. Excessive oxidative stress, if not balanced or controlled by uterine tissue antioxidant mechanisms, can damage cells and cellular organelles. Oxidative-stress-induced damage generate mechanistic pathways that trigger parturition either at term or prematurely at preterm. Understanding these mechanistic mediators and pathways are critical in preventing adverse pregnancy events. One of the responders to oxidative stress damage is p38 mitogen-activated protein kinase (MAPK), which can trigger a variety of events in various tissues. Recent findings suggest that regulating p38 MAPK activation and its downstream pathologic effects may be beneficial in reducing the incidence of adverse pregnancies. 4 However, a consensus has not emerged, and experimental evidence is still lacking to point to p38 MAPK as the molecule to regulate during pregnancy to avoid adverse events, partly because p38 MAPK can also function as a tissue remodeler through various pathways during pregnancy. Several reports have provided various functional roles for p38 MAPK in reproductive tissues at various stages of reproductive events 5 – 7 ; however, its origin, its mode of expression, signals that cause its induction and activation, and its specific role in various reproductive tissues are yet to emerge. The family of MAP kinases has highly conserved regulation and function. There are 3 well-characterized subfamilies of MAPKs in mammals: the extracellular signal-regulated kinases (ERKs), the c-Jun NH2-terminal kinases (JNKs), and the p38 MAPKs. It has been demonstrated that targeted disruption of the p38 MAPK gene is embryonically lethal, 8 indicating the function of p38 during embryogenesis is nonredundant, which is significantly different than JNK or ERK genes in which disruption is not lethal. 9 The p38 MAPK pathway has been implicated in various cellular responses controlling cell proliferation, differentiation, apoptosis, embryogenesis, inflammatory responses, and stress responses in human reproductive tissues. 7 , 10 p38 MAPK is activated by most environmental stresses, such as oxidative stress, and pro-inflammatory cytokines, including interleukin 1 (IL-1) and tumor necrosis factor α (TNF­α); yet the functional role of p38 MAPK is cell-, tissue-, and stimulant-dependent. 11 – 13 Recently, we reported that activation of the p38 MAPK pathway by oxidative stress in human fetal membrane tissues causes cellular senescence and sterile inflammation, which are postulated to trigger parturition. 14 – 16 Animal model studies have also confirmed its potential role in the initiation of labor. 17 , 18 However, it is premature to state that p38 MAPK is the key mediator of parturition based on its activation in chorioamniotic membranes to various stimuli. To examine the role of p38 MAPK, we conducted a systematic review of the literature. Our objectives are to 1) identify reports on p38 MAPK-associated function changes in human and animal reproductive tissues, 2) determine the mechanistic roles of p38 MAPK in human and animal reproductive tissues, and 3) determine the knowledge gaps in the functional role of p38 MAPK in human and animal reproductive tissues. Understanding the role of p38 MAPK in normal term pregnancies can provide insights into pathologic activation of its pathways that can cause adverse pregnancies. We expect that this systematic review will further characterize the role p38 MAPK plays in the initiation of labor, both at term and preterm, and lead to a better design and conduct of future experiments.