Preterm Birth and Corticotrophin-Releasing Hormone as a Placental Clock

Abstract

Preterm birth worldwide remains a significant cause of neonatal morbidity and mortality, yet the exact mechanisms of preterm parturition remain unclear. Preterm birth is not a single condition, but rather a syndrome with a multifactorial etiology. This multifactorial nature explains why individual predictive measures for preterm birth have had limited sensitivity and specificity. One proposed pathway for preterm birth is via placentally synthesized corticotrophin-releasing hormone (CRH). CRH is a peptide hormone that increases exponentially in pregnancy and has been implicated in preterm birth because of its endocrine, autocrine, and paracrine roles. CRH has actions that increase placental production of estriol and of the transcription factor nuclear factor-κB, that likely play a key role in activating the myometrium. CRH has been proposed as part of a placental clock, with early activation of placental production resulting in preterm birth. This article will review the current understanding of preterm birth, CRH as an initiator of human parturition, and the evidence regarding the use of CRH in the prediction of preterm birth.

Preterm birth is defined as birth occurring before 37 completed weeks of gestational age. Preterm birth affects families substantially both in the United States and worldwide, accounting for approximately 10.09% and an estimated 11% of births, respectively (1, 2). Rates of preterm birth in most countries have gradually risen over the past decade (3). Preterm birth is the leading cause of childhood mortality, accounting for approximately 1 million childhood deaths every year (2). Long-term studies have shown that preterm birth carries deleterious effects into adult life, including effects on cardiovascular and neurodevelopmental health (4). Enhanced understanding of the pathologic mechanisms by which preterm birth occurs is required to improve the prediction and prevention of preterm delivery. In this review, we will outline the current understanding of preterm birth mechanisms and specifically the role of corticotrophin-releasing hormone (CRH) as a potential pathologic mechanism for preterm birth.

 

Preterm Birth Principles

Clinically preterm birth is the result of 3 scenarios: (1) spontaneous preterm labor, (2) spontaneous preterm prelabor rupture of membranes, and (3) delivery due to a maternal or fetal indication, otherwise referred to as indicated. National statistics often do not separate these clinical scenarios, though they are important from an etiologic perspective. The relative proportions are often quoted as 70% spontaneous preterm births (preterm labor or premature rupture of membranes) and 30% indicated, though there are large variations among regions and countries (5). In higher income countries, preterm rates appear to be increasing, though methodological considerations must be accounted for given enhanced reporting, reductions in thresholds of viability, and increased multiple birth rates (5, 6).

Gestational age subdivisions of preterm birth have been defined by various organizations. The American College of Obstetricians and Gynecologists defines preterm birth occurring between 34 and 36 completed weeks as late preterm and before 34 weeks as early preterm (7). The World Health Organization defines birth before 28 completed weeks as extremely preterm, 28 to 32 completed weeks as very preterm, and from 32 to 37 weeks as moderate to late preterm (8). These definitions seek to convey the degree of prematurity, as morbidity is gestational-age dependent. Short-term morbidity of late preterm infants includes temperature instability, respiratory distress, apnea, hypoglycemia, seizures, jaundice, kernicterus, feeding difficulties, and periventricular leukomalacia (9). Longer-term effects of late preterm birth on motor skills, cognition, and behavior have been observed (10). Increasing morbidity is observed in extremely preterm infants, and includes considerable respiratory distress requiring ventilatory support, increased rates of interventricular hemorrhage with associated moderate to severe neurodevelopmental impairment, and significant behavioral sequalae (9). Infant mortality is increased across all categories of preterm birth. In the United States, 66% of infant deaths occur in those born preterm (11).

Preterm Birth as a Syndrome

The physiologic pathways that lead to parturition at term remain incompletely understood. At present, preterm birth appears to be not a single pathway or condition, but rather a syndrome (12). That is to say that there are multiple pathologic etiologies that may or may not share common pathways with parturition at term and result in preterm birth.

Events necessary for parturition at term include increasing myometrial contractility, cervical remodeling, and rupture of membranes. Myometrial quiescence is favored through the majority of gestation due to the actions of progesterone, which antagonizes the nuclear factor-κB (NF-κB) pathway and suppresses expression of proinflammatory cytokines (eg, interleukin-1 [IL-1], IL-6, IL-8) (13). At term progesterone action is reduced by several different mechanisms. Progesterone action during pregnancy is mediated by the progesterone receptor subtype B. Within the myometrium, subtype B inhibits expression of contraction-associated proteins and induces the transcription factor zinc finger E-box-binding homeobox 1 (ZEB1), which in turn acts to inhibit contraction-associated proteins and inhibit expression of progesterone-metabolizing enzymes (eg, 20 alpha-hydroxysteroid dehydrogenase) (14). At term inflammatory signal–regulated phosphorylation of progesterone receptor A antagonizes the action of the B subtype and creates a functional progesterone withdrawal (15, 16). While progesterone action is reducing at term, synthesis of active forms of the estrogen receptor 1 is increasing concurrently with increasing plasma estriol concentrations (17, 18). Estrogens in turn promote myometrial changes conducive for contractions as well as cervical ripening (19). Recent transcriptome analysis suggests that the myometrial changes in progesterone receptors, progesterone metabolism, and estrogen action are downstream from increases in the transcription factor NF-κB, which in turn may be driven by increasing CRH concentrations, as within the brain CRH has been shown to stimulate NF-κB activity (20). The exact cell types within myometrial tissue that mediate these changes are not yet clear, although work on single-cell transcriptomics will likely resolve this uncertainty (21). Recent work suggests that the term transformation of myometrium occurs by the reorganization of gene transcription within topologically associated domains, that is, groups of genes that are coregulated and coexpressed to generate a change in the tissue phenotype (22). This new knowledge regarding term labor provides a necessary framework to understand preterm labor in which the mechanisms of premature uterine activation, cervical remodeling, and membrane rupture may be different (23).

Cervical remodeling for dilation is mediated by changes in the structure and mechanical function of the extracellular matrix (24). Through pregnancy the replacement of fibrillar collagen containing strong cross-links with collagen containing weak cross-links is achieved in the progesterone-dominant phase termed cervical softening (25). Loss of progesterone function in late pregnancy during the cervical ripening phase promotes the influx of tissue monocytes into the cervix with activation of proinflammatory and tissue repair macrophages evident in the postpartum repair of the cervix (26). Similar to the myometrium, changes in progesterone receptor isoforms and increased metabolism of progesterone in late pregnancy contribute to cervical dilation necessary for parturition (27).

Membrane rupture results through the dynamic interactions with the decidua from either increased expression of inflammatory cytokines (eg, tumor necrosis factor-α and IL-1) or thrombin activation (decidual bleeding/abruption) associated with changes in activity of matrix metalloproteases and tissue inhibitors of matrix metalloproteases, increased granulocyte-macrophage colony-stimulating factor activity, dissolution of cellular cements such as fibronectin, and apoptosis (28–30). Again, these changes may be downstream from NF-κB activation and NF-κB has been emphasized as a target for novel tocolytics (31). Pathologic disruption of these 3 processes can occur at multiple levels—maternal, uterine (eg, overdistention from polyhydramnios, multifetal gestation), placental, decidual and/or cervical and via multiple pathways—stress, infection, vascular disorders, and/or altered immune tolerance (Fig. 1) (12, 32). Genetic and environmental factors also likely play a role.

Figure 1.

Potential compartments and pathologic pathway(s) involved in preterm birth. Preventive treatment(s) of preterm birth are also shown.

Intra-amniotic infection is one mechanism that has been causally linked to spontaneous preterm birth (33). Bacteria may gain access to the amniotic cavity via ascending infection, transplacental passage, retrograde seeding from the fallopian tubes, or introduction via invasive procedures (eg, amniocentesis). Ascending infection is the most common route of entry, wherein microorganisms that colonize the cervix, decidua, and membranes may then enter the amniotic sac (32). Most studies have defined intra-amniotic infection as those with a positive amniotic fluid culture for a pathologic microorganism, rather than “clinical chorioamnionitis,” or the clinical syndrome associated with bacterial invasion of the amniotic cavity (34). Based on this definition, colonization with an infectious agent has been detected in 25% to 40% of all preterm births (35), though this number may be limited by culture techniques (36). The mechanisms by which infection leads to preterm birth are related to the innate immune system—with activation of pattern recognition receptors that elicit inflammatory cascades that stimulate the production of prostaglandins and matrix-degrading enzymes (35, 37). Transcriptomic analysis of myometrium from individuals with clinical chorioamnionitis show marked upregulation of many inflammation-related genes, but even those without evidence of chorioamnionitis still show an immune signature suggesting that immunological activation is common to preterm labor of differing etiologies (38). Another mechanism linked to spontaneous preterm birth is decidual activation by thrombin (decidual bleeding/abruption) (30, 39).

Prediction of Preterm Birth

Numerous tests and monitoring modalities have been proposed as markers for preterm birth risk due to known association. These include cervical length assessment (40, 41); biomarkers such as fetal fibronectin (42), phosphorylated insulin-like growth factor–binding protein 1 (43), and placental alpha microglobulin 1 (44); and uterine monitoring (45). However, there is no single test that can accurately predict a women's risk for preterm birth. At present prediction of preterm birth is primarily based on a summative clinical picture determined by a patient's risk factors. Several risk factors have been identified. A prior history of preterm birth remains the most important. A prior history of preterm birth, however, predicts only approximately 10% of preterm births. In the United States persistent racial disparities have been observed in risk for preterm birth, with higher rates observed in non-Hispanic Black women compared to White and Hispanic women (1). Pregnancy factors such as a threatened abortion, fetal abnormalities, multifetal gestation, polyhydramnios, short cervical length, and short or long interpregnancy interval are known to increase the risk of preterm birth (32). Additional contributing factors for preterm birth risk include lifestyle (cigarette and illicit drug use, prepregnancy body mass index), genetic, infectious (mycoplasma, bacterial vaginosis), and periodontal disease (32). Given the multiple pathologic etiologies of preterm birth, it is worth noting that it is unlikely that a single test will predict all preterm birth, but rather a test may be designed to target the specific etiology of preterm birth, as we will review with CRH in later sections.

Prevention of Preterm Birth

Three treatments are recommended strategies by the Society for Maternal-Fetal Medicine and American College of Obstetricians and Gynecologists for the prevention of preterm birth in selected populations: vaginal progesterone, intramuscular progesterone, and cervical cerclage (7). Progesterone supplementation is given on the premise that withdrawal in mammals is a parturition-triggering event and that supplementation during pregnancy may prolong uterine quiescence. Data on the ability of progesterone to prevent preterm birth have been inconsistent in clinical trials (46–52), and there remains a dearth of information to support its mechanistic benefit in humans. Vaginal progesterone is recommended in women with a singleton gestation without a prior history of preterm birth but with a short cervix (< 25 mm); either vaginal or intramuscular progesterone can be considered in women with a singleton gestation and a prior history of preterm birth (7). Cervical cerclage is a treatment for the subset of women identified to have cervical dysfunction consistent with cervical insufficiency or a short cervix (< 25 mm) and a prior history of preterm birth (53). Serial cervical length screening is recommended in women with a singleton gestation and a prior history of preterm birth because of the potential for treatment with cerclage (7). In addition, modification of known risk factors (eg, cigarette and illicit drug use, low prepregnancy weight) can contribute to preterm birth risk reduction.

Present Preterm Birth Needs

There are many gaps in the knowledge of prediction and prevention of preterm birth. With progesterone there remains a need to better understand mechanistically its function to better identify those women who would benefit from supplementation across various indications. Moving beyond progesterone is the next step. There is a current need to identify alternative pathologic pathways, determine rationales for identifying populations at risk of preterm birth based on pathology, and then develop novel strategies for the prevention of preterm birth. One such pathway is CRH.

Corticotrophin-Releasing Hormone in Humans

CRH is a 41 amino acid peptide hormone synthesized in the hypothalamus that regulates the secretion of the anterior pituitary adrenocorticotropin (ACTH) secretion in response to stress (54). Outside pregnancy, CRH is barely detectable in the human circulation. However, in pregnancy, placental synthesis of CRH leads to exponential increases in the maternal circulation, the amniotic fluid, and the fetal circulation (55, 56). CRH has a known specific circulating binding protein (CRH-BP) (57). When CRH levels are high in late gestation, the saturated binding protein dimerizes and is cleared, leading to increased free, biologically active CRH, which is potentially able to contribute to triggering parturition (58–60).

Transcriptional and Epigenetic Regulation of the CRH Gene

Transcriptional and epigenetic regulation of the CRH gene varies by species and tissue type. Placental expression of CRH within the syncytiotrophoblast is specific to human and anthropoid primates. In New and Old World monkeys, placental CRH production peaks in mid-gestation (61, 62), and only in the apes, including humans, does placental CRH production increase exponentially across pregnancy, peaking at term (63). The single-copy CRH gene is also expressed in the hypothalamus, where it is under negative feedback control by glucocorticoids and is a key regulator of the stress response (64). In contrast, placental expression of CRH is under positive feedback by glucocorticoids and by cyclic adenosine monophosphate (cAMP), leading to the exponential increase in expression across gestation. The regulatory elements of the CRH gene in humans is shown in Fig. 2. The placenta-specific expression of CRH in humans and anthropoid primates is attributed to the presence of the retroviral long terminal repeats (LTRs) of the transposon-like human element 1B (THE1B) in the upstream region of the human CRH gene (65). The LTR element of THE1B acts as an enhancer for the CRH gene. Transgenic expression of the human CRH gene with the LTR of THE1B in mice delayed parturition by 15 hours and induced the expression of human CRH in mouse placenta (66). Deletion of the LTR of THE1B in transgenic mice returned the mice to normal parturition and stopped human CRH production in the mouse placenta (66). The human CRH gene promoter has 2 cAMP response elements (CRE), a consensus CRE, and a caudal-type homeobox response element (CDXRE) (67). The CRH promoter also has a negative glucocorticoid response element (nGRE), through which glucocorticoids suppress CRH production in the hypothalamus. However, King et al (67) identified a particular section of the CRH promoter (−213 to −99 base pairs) mediating glucocorticoid stimulation of CRH within the syncytiotrophoblast. The CRH promoter also has an NF-κB2 enhancer site, and chromatin immunoprecipitation showed that this enhancer is associated with RelB and P52 in trophoblast, which are part of the noncanonical pathway that can be upregulated in the placenta and myometrium in preterm birth (68). This enhancer interaction with RelB and P52 is increased by glucocorticoids and decreased by progesterone (68). Glucocorticoids have also been shown to enhance CRH expression by acetylating H3K9 histone in primary trophoblast culture (69), while cAMP stimulates the expression of CRH messenger RNA (mRNA) by increasing histone-3, lysine-4 trimethylation (H3K4me3), and histone-4 acetylation (acH4) (70). In normal pregnancies the mRNA for CRH is upregulated in the placenta after the fifth week of gestation (71).

Figure 2.

Regulatory elements in the CRH gene in humans. An endogenous retrovirus long-term repeat enhancer sequence (ERV-LTR), NF-κB enhancer sequence, negative glucocorticoid response element (nGRE), cAMP response element (CRE), caudal-type homeobox response element (CDXRE), and a TATA element are present in the promoter region of the human CRH gene.

CRH Downstream Effects in Pregnancy

The effect of CRH expression in pregnancy is tissue specific based on in vivo and in vitro studies. In the maternal pituitary, CRH has a stimulatory effect on ACTH, β-endorphin, and cortisol production that is subsequently tempered by desensitization of the pituitary receptor to CRH in late pregnancy (Fig. 3) (72–74). In the myometrium, CRH potentiates the action of uterotonics such as oxytocin and prostaglandin E2 (PGE₂) (75–78). Work by Grammatopoulos and Hillhouse elucidated that there are several isoforms of the CRH receptor in the myometrium (79). When nearing parturition, there appears to be a shift in isoforms that reduces cAMP signaling pathways promoting relaxation, with a consequent increase in contractility (80, 81). In the membranes and amniotic compartment, CRH increases synthesis of prostaglandins (PGE₂ and PGF_2α), promoting cervical ripening and uterine contractions (82). Additionally, prostaglandins stimulate further CRH secretion, creating a positive feedback loop in the placenta (82, 83). In the maternal circulation, CRH stimulates an increase in IL-6 from peripheral blood monocular cells (84). These cells migrate to the placenta and cervix and are notably increased in parturition and infection (84). IL-6 along with IL-1 also increase placental CRH production as another positive feedback loop (85).

Figure 3.

CRH pathways and feedback loops in the maternal, placental, and fetal compartments. Solid lines represent potential roles for CRH in parturition. Dashed lines represent positive feedback loops that could lead to progressive amplification of CRH. Adapted from Herrera 2021.

In the fetal circulation, by mid-trimester CRH stimulates ACTH in the fetal pituitary that may increase cortisol and dehydroepiandrosterone sulfate (DHEA-S) production (86). CRH also directly stimulates cortisol production from the fetal adrenals (87). Cortisol in turn has a direct effect on fetal lung maturity, as indicated by increased indices of fetal lung maturity at birth in women with higher CRH levels (88). Fetal cortisol positively feeds back to the placenta, unlike the hypothalamus, promoting increased placental CRH production (see Fig. 2) (89, 90). CRH directly stimulates DHEA-S production in the fetal zone of the fetal adrenal (91, 92). Fetal DHEA-S in turn is the major substrate for placental production of estriol, after 16 hydroxylation in the fetal liver. Estriol is then released into the maternal bloodstream, where it can promote labor via actions on the myometrium, cervix, and fetal membranes. Again, like cortisol, DHEA-S also positively feeds back to increase CRH secretion (93). Estrogen and progesterone have been shown to modulate placental CRH production through a separate CRE in the CRH promoter (see Fig. 2) (94–96). The multiple positive feedback loops of CRH allow progressive amplification of CRH until parturition occurs (see Fig. 3) (97).

CRH and Preterm Birth

Multiple cross-sectional studies reported that high concentrations of CRH are associated with preterm delivery (98–101). This led McLean et al to conduct a prospective study in 1995 that introduced the idea of CRH as a placental clock (102). CRH was found to increase exponentially in the maternal plasma throughout pregnancy. Controlling for gestational age compared to women who delivered at term, CRH levels were higher in those women who experienced preterm birth and lower in those with postterm delivery. These differences were apparent as early as 16 to 18 weeks of gestation. There was no difference in CRH-BP concentrations among women in the 3 populations. From these findings, CRH was proposed as a placental clock—earlier elevated levels of CRH sufficient to saturate CRH-BP in the maternal bloodstream may result in increased bioavailable CRH capable of initiating parturition. Within this framework, parturition is suggested to result from processes beginning before the mid-trimester and at a rate that could be inferred from observation of CRH levels, enabling preterm birth prediction. Following the work of McLean and colleagues, several studies (103–112) confirmed the association of higher second-trimester CRH with preterm birth.

However, a few studies failed to observe an association between maternal plasma CRH concentrations and preterm birth (113, 114). There are 2 explanations for these disparate data. One, technical considerations have led to difficulty in comparison of published results. Over the years differences have been reported in plasma processing, extraction technique, and the method of measurement of CRH (115). Extraction is necessary as CRH is bound by plasma proteins, including CRH-BP, and the degree of binding is highly variable between individuals (116). Failure to use an extraction technique can lead to erroneous results (97, 113). The second issue has to do with heterogeneity of populations. CRH is altered by race and ethnicity, with non-Hispanic Black (111) and Hispanic (117) women found to have lower CRH levels compared to non-Hispanic White women, though women delivering preterm still demonstrate a high CRH level for their racial and ethnic group (111).

Despite the association of increased CRH and preterm birth, in an unselected population the predictive capability of CRH alone is low. Promisingly, a recent study did find that CRH was increased in women who experienced recurrent preterm birth as compared to women who had a prior preterm birth and subsequent term birth (97). This suggests CRH does specify a population predisposed to preterm birth. That is to say that CRH plays a clear role in normal pregnancy physiology that can be altered to form a pathophysiologic process in a subset of women that results in preterm birth.

Pathologic Interactions of CRH

Placental CRH has been found to be elevated in the setting of preterm prelabor rupture of membranes with chorioamnionitis, suggesting that placental expression of CRH is activated by stress-related pathways that are active in infective processes (118). In twin gestations, CRH is elevated to more than 3 times a singleton gestation; however, preterm birth does not always occur. Variability in ratios of progesterone, estriol, and estradiol may explain this nuance (119). CRH is also elevated in the setting of many indicated preterm births, including fetal growth restriction (112, 120), pregnancy-induced hypertension (110, 121, 122), and gestational diabetes (122). In these cases, CRH elevation may be a compensatory mechanism as a result of stress-related pathways independent of infection status, as induction of these stress-related pathways can influence early brain development (123).

Future Potential of CRH

There is convincing evidence that early increases in plasma CRH correlate with an increased risk of preterm birth. However, the specific threshold at which substantial risk develops has not yet been defined. As a result of racial and ethnic variation, the establishment of large-scale data sets would be necessary for comparative analysis. In addition, any future data must be obtained in a standardized fashion for analysis. Given proven reproducibility of results, this would be via an extraction technique followed by radioimmunoassay (97). If the target population can be identified for diagnostic purposes, the next step would be the potential for therapeutic intervention. A CRH receptor antagonist is then an attractive option for the prevention of preterm birth. Both peptide and nonpeptide antagonists have been developed (124, 125). In sheep CRH antagonism was found to significantly delay delivery (126). Development of CRH antagonism for application in human pregnancy is promising provided diagnostic identification can be performed. Moving beyond CRH, patient-specific genomic profiling may also be another area of benefit, as recent research shows RNA profiling may indicate those at risk for preterm delivery (127).

Summary

In summary, comprehensive understanding of preterm birth must recognize the multifactorial etiology and contributing factors of this clinical syndrome. Current prediction and prevention of preterm birth are limited by our lack of pathophysiologic understanding. One potential pathologic pathway resulting in preterm birth is placentally produced CRH. Despite strong evidence for its biologic plausibility and preterm birth association, CRH measurement for clinical prediction remains problematic as a clear threshold for use has not been established.

Abbreviations

ACTH

adrenocorticotropin

cAMP

cyclic adenosine monophosphate

CRE

cAMP response element

CRH

corticotrophin-releasing hormone

CRH-BP

CRH binding protein

DHEA-S

dehydroepiandrosterone sulfate

IL

interleukin

LTR

long terminal repeat

mRNA

messenger RNA

NF-κB

nuclear factor κB

THE1B

transposon-like human element 1B

Disclosures

The authors have nothing to disclose.

Data Availability

Data sharing is not applicable to this article as no data sets were generated or analyzed during the present study.