Understanding the role of placental pathophysiology in the development of bronchopulmonary dysplasia

Abstract

The associations between bronchopulmonary dysplasia (BPD) and the gestational pathologies of chorioamnionitis (CA) and hypertensive disorders of pregnancy (HDP) have become increasingly well recognized. However, the mechanisms through which these antenatal conditions cause increased risk of BPD remain less well characterized. The objective of this review is to discuss the role of the placenta in BPD predisposition as a primary driver of intrauterine alterations adversely impacting fetal lung development. We hypothesize that due to similarities in structure and function, placental disorders during pregnancy can uniquely impact the developing fetal lung, creating a unique placental-pulmonary connection. In the current review, we explore this hypothesis through analysis of clinical literature and preclinical model systems evaluating BPD predisposition, discussion of BPD phenotypes, and an overview on strategies to incorporate placental investigation into research on fetal lung development. We also discuss important concepts learned from research on antenatal steroids as a modulator fetal lung development. Finally, we propose that the appropriate selection of animal models and establishment of in vitro lung developmental model systems incorporating primary human placental components are key in continuing to understand and address antenatal predisposition to BPD.

INTRODUCTION

Associations Highlighting Antenatal Risk Factors for Bronchopulmonary Dysplasia

Although bronchopulmonary dysplasia (BPD) has been studied for several decades, the acknowledgment of antenatal predisposition to this disease has only more recently come into stronger consideration. Among antenatal conditions influencing BPD risk, chorioamnionitis (CA), hypertensive disorders of pregnancy (HDP) and resultant intrauterine growth restriction (IUGR) have emerged as some of the strongest correlations in the clinical literature. The associations between these pregnancy disorders and BPD are challenging to understand because of the myriad confounders that cloud their relationships—for example, gestational age at birth or exposure to antenatal steroids. Because these disorders stem from unique intrauterine pathologies with primary involvement of the placenta, thus the role of placental pathology within the pathways directing increased risk for BPD warrants ongoing exploration. Because the placenta and developing fetal lung have many structural and functional similarities, we hypothesize the existence of a unique placental-pulmonary connection. Perturbations in this connection can alter fetal lung development, potentially putting infants at increased risk for developing BPD postnatally. In this review, we will present an overview of clinical data on CA, HDP, and IUGR associations with BPD, discuss the potential role of the placenta as a driver of these associations, and examine what placental lessons can be learned from studies on antenatal steroids for pulmonary benefit. Finally, we will discuss critical components of placental interface that are key in examining the role of this organ in altering fetal lung development.

Chorioamnionitis and BPD

CA is clinically defined as an acute intraamniotic infection leading to inflammation of the placenta and fetal membranes. CA is caused by several pathogens, most commonly genital mycoplasmas such as Ureaplasma species and Mycoplasma hominis (1). Several early observational studies were the first to flag CA as a potential antenatal risk factor for BPD (2, 3). Especially salient was a prospective cohort trial by Watterberg et al. (4) involving 53 preterm neonates that found exposure to CA was associated with lower risk of developing RDS but higher risk of developing BPD. From this work emerged the hypothesis that infection/inflammation of the intrauterine environment might promote lung maturation in the short-term and thus be protecting against RDS. However, it was further posited that this inflammatory environment could predispose patients to higher risk of longer-term abnormal lung development leading to BPD, especially in conjunction with other postnatal proinflammatory insults like mechanical ventilation or sepsis (5, 6).

Subsequent studies however found no independent association between exposure to CA and BPD risk, complicating the contemporary understanding of BPD pathogenesis (7, 8). Aiming to better elucidate this area, Hartling et al. (9) conducted a meta-analysis of 59 studies encompassing data from 15,295 preterm neonates that found a significant association between exposure to CA and BPD risk with an adjusted pooled odds ratio (OR) of 1.58 (95% CI 1.11–2.24). Villamore-Martinez et al. (10) also conducted a more recent meta-analysis of 158 studies encompassing data from 244,096 preterm neonates that also found a significant association between exposure to CA and risk of BPD at 36-wk postmenstrual age with an adjusted pooled OR of 1.25 (95% CI 1.01–1.54).

Given the heterogeneity of the findings discussed above, it remains unclear whether CA is truly correlated with an increased risk for BPD, indicating a need for ongoing investigation. Of note, all meta-analyses cited above were conducted with random-effects statistical models, which may impact their power. Ongoing studies, with larger cohorts and additional statistical analyses, such as prediction intervals (11), will be helpful to provide further clarity for these associations.

Hypertensive Disorders of Pregnancy and BPD

Maternal hypertensive disorders of pregnancy (HDP) include both gestational hypertension (GH) and the more severe systemic disease of preeclampsia (PE). GH and PE are clinical diagnoses made traditionally after 20 wk’ gestation. Although GH is the isolated occurrence of maternal hypertension, PE reflects a systemic vasculopathy with symptoms including maternal hypertension, proteinuria, liver dysfunction, and if left untreated, maternal seizures (12, 13). Although the pathogenesis of HDP is multifactorial, a strong consensus points to altered placentation and placental ischemia leading to systemic maternal vasculopathy (14, 15). In the case of PE, this vasculopathy has been attributed to the release of the antiangiogenic soluble fms-like tyrosine kinase 1 (sFLT-1) from the placenta into maternal circulation (13). Elevated sFLT-1 levels have also been identified within the fetal cord blood and amniotic fluid of preeclamptic pregnancies (16–18) strongly suggesting PE can also result in an anti-angiogenic intrauterine environment. Altered placentation and subsequent compromise of nutrient/oxygen delivery to the developing fetus is also why HDP, with PE in particular, are strongly associated with intrauterine growth restriction (IUGR) (19).

Several studies to date have reported significant associations of BPD with HDP, PE and IUGR supporting the hypothesis that placental vascular abnormalities, chronic hypoxia to the fetus, and/or imbalance of angiogenic signals can impact the fetal lung resulting in postnatal disease (20–26). More recent higher-powered studies have identified similar associations. Bi et al. (27) conducted a meta-analysis of 15 studies encompassing data from 20,779 preterm neonates that found a significant association between exposure to HDP and BPD risk with an adjusted pooled OR of 1.29 (95% CI 1.01–1.65). In some contrast, Razak et al. (28) conducted a meta-analysis of 9 studies that found no significant association overall, but subgroup analysis of patients born less than 29 wk yielded a significant association between exposure to HDP and BPD risk with an adjusted pooled OR of 1.15 (95% CI 1.06–1.26), suggesting potential importance of gestational age in mediating effects of HDP.

Finally, two studies published in 2022 highlight the importance of distinguishing an IUGR subgroup within analyses of HDP and BPD. Pierro et al. (29) conducted a meta-analysis of 211 studies encompassing data from 347,963 preterm neonates. Here authors identified a significant association between IUGR and risk of BPD at 36-wk postmenstrual age with an adjusted pooled OR of 1.56 (95% CI 1.37–1.79) (29). In addition, Hart et al. (30) showed an association with IUGR and long-term lung dysfunction in childhood (OR 1.783, 95% CI 1.06–3.00). Within the intrauterine compartment, dysfunctional placentation leading to IUGR might increase BPD risk more significantly as compared with placental alteration leading to HDP alone. Thus, HDP with IUGR may reflect a more significant disease severity leading to increased BPD risk. Earlier studies grouping HDP/IUGR together may have detected significant overall associations that could have been attributed to the IUGR subgroup within the broad HDP classification. Limitations to the current literature include nonstandardized definitions of HDP and BPD and IUGR being a mediator for further NICU interventions that may increase BPD risk. Nonetheless, the above studies lend depth to the dialogue of pregnancy disorders with altered placental function correlating with antenatal BPD predisposition.

ANTENATAL BPD RISK: PLACENTAL IMPACT ON THE PULMONARY DEVELOPMENTAL NICHE

Given the associations of CA and HDP/PE/IUGR with BPD, these diseases may be contributors to the continued prevalence of the “new BPD,” driven more by aberrant lung development rather than postnatal exposures (31). As such, research evaluating the intrauterine environment surrounding the developing lung should take equal importance to studies evaluating the fetal lung itself. The placenta, as the main physiological interface between mother and fetus, is the primary director of the intrauterine environment and thereby a significant contributor to the pulmonary developmental niche. Incorporating a perspective of placental study into BPD research is a key area of focus for studies moving forward in this field (32). The placenta and lung have significant structural and functional similarities, particularly during fetal lung development (Fig. 1). Both organs undergo parallel branching morphogenesis throughout gestation (34) resulting in functional subunits for gas exchange. In the case of the placenta, this unit is the placental villous, a blood-epithelial interface which enables gas exchange between the fetal and maternal circulations during pregnancy, similar to the air-epithelial interface of the lung alveolus. (Fig. 1). These similarities have led an expanding dialogue on the “placental- pulmonary” connection during fetal development. Although this connection is still largely speculative, it stems from the hypothesis that changes in the placenta may impact the fetal lung more prominently than other organ systems (33).

Figure 1.

Similarities of placental and lung development. A: schematic representations of human placental villous and fetal alveolar networks throughout gestation. B: histological cross sections of a term human placental villous and neonatal sheep lung alveolus. Boxes outline area of enlarged schematic. C: exchange interface of maternal blood space with placental villous; alveolar air space with pulmonary blood vessels. ALV, lung alveolus; BV, blood vessel; cTB, cytotrophobalst; EC, endothelial cell; MBS, maternal blood space; sTB, syncytiotrophoblast; T1, type I pneumocyte; T2, type II pneumocyte. [Image adapted from Taglauer et al. (33). Used with permission from Elsevier.]

Supporting the hypothesis of a placental-pulmonary connection, a key study by Mestan et al. (35), in 2014, identified a significant association between placental lesions called maternal vascular malperfusion (MVU) and neonatal BPD. Maternal vascular underperfusion (MVU) consists of a constellation of histopathologic alterations in the placental bed in response to aberrant maternal blood flow to the placental compartment and chronic fetal-hypoxic-ischemic state (36). Accordingly, MVU is a common placental finding among growth restricted infants. Mestan et al. (35) found that placental MVM was independently associated with BPD risk, even after adjusting for other neonatal complications commonly associated with BPD (adjusted OR 2.6, 95% CI 1.4–4.8). This study highlighted an important connection between placental aberrancy and risk of postnatal lung disease.

To further understand how the placenta connects to postnatal lung disease, other studies have examined specific alterations within the intrauterine environment known to result from placental alteration. Several preclinical models of intraamniotic exposures to proinflammatory lipopolysaccharide (LPS) (37–39) or anti-angiogenic sFLT-1 (41, 42) have demonstrated that these intrauterine exposures can alter postnatal lung alveolar structure and vascularization, correlating back to respective associations of CA and PE with BPD. More recent work by Taglauer et al. (40) used the heme-oxygenase-1 null (HO-1^−/−) mouse as an established model of PE with IUGR to further examine the influence of this disorder on the developing fetal lung. When compared with control pregnancies, the maternal preeclamptic (HO-1^−/−) environment resulted in altered fetal lung branching morphogenesis and differences in developmental gene expression at embryonic days 15–17. Postnatally on day 14, pups born to preeclamptic (HO-1^−/−) mothers had increased alveolar simplification and bronchial airway epithelial flattening. In vitro fetal lung explant and amniotic fluid culture assays in this study also identified altered branching morphogenesis and developmental gene expression changes following exposure to PE versus control amniotic fluid. Overall, Taglauer et al. findings suggested that the global preeclamptic intrauterine environment can impact early developmental changes in the fetal lung translating to postnatal morphological alterations.

Of note, the postnatal lung alterations (i.e., alveolar simplification, altered vascularization and bronchial airway epithelial flattening) found within the LPS, sFLT-1, and the HO-1^−/− antenatal models discussed above were all identified under normoxic conditions (37, 38, 40–42). Taken together, these findings strongly suggest the abnormal intrauterine environments of CA and PE can result developmental lung alterations that may predispose infants toward neonatal lung disease before postnatal exposures of mechanical ventilation and/or oxygen exposure.

PREDISPOSITION OF BPD SUBTYPES

When considering the placental impact of the pulmonary developmental niche, the in utero pathophysiology leading to differing BPD subtypes must also be considered. Here we speculate that distinct intrauterine pathologies resulting from CA and HDP (PE in particular) may result in clinical subtypes of the “new BPD” beyond the traditional categorization of mild/moderate/severe. Although still being defined, these subtypes may include the complication of pulmonary interstitial emphysema and the development of pulmonary hypertension (31).

Although both CA and HDP have been associated with BPD, the pathogenic mechanisms through which they negatively impact fetal lung development are likely very different (Table 1). CA is primarily driven by pathogen response pathways and/or idiopathic inflammation (1, 43). CA is also more common driver of spontaneous preterm birth, resulting in additional inflammatory factors released from the prematurely laboring uterus/placenta into the intraamniotic compartment (1). Among the HDP phenotypes, PE is strongly associated with the antiangiogenic influence of sFLT. Therefore, we speculate that the preeclamptic intrauterine compartment is likely more directly tied with vascular dysregulation rather global inflammation. In addition, PE is more commonly associated with IUGR. In line with this premise, both HDP and IUGR have been significantly associated with BPD-associated pulmonary hypertension (29). The precise mechanisms through which placental pathology in CA and PE in particular lead to altered fetal lung development and how these pathways can result in different BPD phenotypes requires ongoing exploration.

Table 1.

Comparative evaluation of antenatal conditions significantly correlated with BPD

	Chorioamnionitis (1)	Preeclampsia (12, 13)
Pathogenesis	Intrauterine bacterial infectionIdiopathic inflammation	Multifactorial causes leading to altered placentation and the release of placental sFLT-1 into maternal circulation and intrauterine compartment
Predominant intrauterine environment	Proinflammatory	Anti-angiogenic
Typical duration	Acute and chronic	Acute and chronic
Reason for preterm birth	Premature labor	Induction/cesarian section for maternal indications or intrauterine growth restriction

BPD, bronchopulmonary dysplasia.

ADDRESSING CA AND PE INTRAUTERINE PATHOPHYSIOLOGY: LESSONS LEARNED FROM ANTENATAL STEROIDS

One of the most significant clinical advances to impact the pulmonary developmental niche has been the introduction of maternal antenatal steroid (AS) administration. Placental-pulmonary investigations seeking to address antenatal BPD risk factors can build on the important groundwork laid by the clinical and preclinical studies pioneering the use of this therapy. The 1972 landmark study by Liggins and Howie (44) demonstrated that AS improve infant survival and reduce the incidence of acute RDS at birth. This study resulted in a many subsequent investigations leading to the current NIH consensus on the benefit of AS for hastening fetal lung maturation and the American College of Obstetricians and Gynecologists (ACOG) recommendation to use corticosteroids in pregnant women at risk for preterm birth within 7 days (45, 46). These recommendations have led to the widespread use of AS resulting in a large proportion of infants with antenatal risk of developing BPD being exposed to AS intrauterine effects.

Although the use of AS is strongly supported by these consensus statements, the mechanisms through which AS confer their benefit are still being defined. Preclinical animal studies have identified that AS can inhibit inflammation, induce the surfactant system, and enhance lung structural maturation (47, 48). However, alteration of an inflammatory intrauterine environment seems to require a delicate balance for the developing lung (39, 48). Gestational timing of administration and the order of exposure (i.e., before or after the onset of intrauterine inflammation) seem to influence how AS impacts the developing fetal lung (49, 50). In addition, there may be a synergistic benefit between inflammation and glucocorticoids. In a nonhuman primate model in which antenatal steroids were administered in the setting of LPS-induced intrauterine inflammation, Schmidt et al. (51) found that both exposures had a greater impact on improving neonatal lung compliance and surfactant production than either steroids or inflammation alone. However, these early benefits of were balanced by gene alterations in extracellular matrix development, so the long-term impact of synergistic inflammatory-steroid exposures remains to be determined.

Overall, conclusions on how antenatal steroids benefit the pulmonary developmental niche have largely been drawn from the setting of intrauterine inflammation and thus tie closely into pathways to mitigate BPD predisposition related to CA. But what of the risks associated with the predominantly anti-angiogenic environment of PE? In this case, important information can be drawn from work examining how antenatal steroids can impact neonatal pulmonary hypertension. In preclinical rat and sheep models, multiple studies have demonstrated that antenatal steroids (particularly betamethasone) can also induce the expression of vascular endothelial growth factor (52) and endothelial nitric oxide synthase (eNOS) (53–55). These proangiogenic aspects of antenatal steroids can also be of key consideration in ongoing studies seeking to optimize the BPD prone physiology of the preeclamptic intrauterine environment.

INTRAUTERINE IMPACT OF ANTENATAL STEROIDS- NEED FOR PLACENTAL EVALUATION

Despite widespread use, antenatal steroids have not resulted in a significant long term reduction of BPD among the preterm population. Reasons for this have remained unclear. Additional investigation is required to perhaps identify more targeted or additional pathways to address antenatal aspects of BPD prevention. To date, the majority of research on antenatal steroids has focused on the developing lung. However, because maternal antenatal steroids are administered systemically, it remains to be determined whether antenatal steroids primarily work by 1) direct impact on fetal lung tissue or 2) impact on the placenta to alter the pulmonary developmental niche.

Addressing this differentiation, Jobe et al. (56) showed that AS had a greater impact if administered systemically to pregnant ewes rather than direct intraamniotic treatment to the fetus. Although this effect may be secondary to differences in dosing/bioavailability, it also suggests that alteration of maternal intrauterine physiology may be more beneficial than direct exposure of AS to the fetal lung. Subsequently, a limited number of studies have investigated the effects of antenatal steroids on placental structure and function, finding notable differences between species. In sheep, antenatal steroids were found to impact placental epithelium, vasculature, and steroidogenic pathways, and were associated with a reduction in fetal weight (57–59). Studies in mice showed more beneficial effects, with a favorable impact on reducing feto-placental vascular resistance (60).

More recent work by Czamara et al. (61) evaluating DNA methylation in human placental tissue biopsies from mothers exposed to antenatal steroids found even more divergent effects, identifying epigenetic changes in genes particularly associated with inflammation and immune response. This study in particular highlights an important point: research evaluating the placental and pulmonary impact of antenatal steroids requires a stronger focus on the intrauterine immune environment. Within the maternal-fetal interface, specialized maternal immune cells located in modified uterine tissue called the decidua (Fig. 2B), surround the placenta and work in close relationship with placental parenchymal cells to dictate physiologic events of pregnancy initiation maintenance and parturition (62). The cytokines and chemokines secreted by these immune cells the maternal-fetal interface can contribute significantly to the intrauterine environment. Because steroids have such well characterized immunomodulatory effects, we speculate they are likely causing underrecognized effects on the maternal intrauterine immune environment (i.e., the pulmonary developmental niche). Evaluating the impact of antenatal steroids on maternal intrauterine leukocytes may be a key future area of investigation in understanding pathways in which to address antenatal BPD risk.

Figure 2.

Key considerations for studying placental influences in bronchopulmonary dysplasia (BPD). A: comparison of placental structure in models commonly used for the study of fetal lung development. B: accessible human intrauterine samples for investigation of the placental influence on fetal lung development. FBV, fetal blood vessel; MBS, maternal blood space; sTB, syncytiotrophoblast; TB, trophoblast; UEC, uterine epithelial cell. [Image created with BioRender.com and published with permission.]

CONCLUSIONS AND FINAL CONSIDERATIONS

Appropriate Selection of Animal Models

Mechanistic studies evaluating the placental influence on fetal lung development will require the appropriate use of animal models. Reviews published in 2022 by Raia-Barjat et al. (63) and Bakrania et al. (64) have excellent summaries on the respective benefits of key animal models to examine CA and PE. However preclinical models for these disorders much be approached with an important caveat: from a placental perspective, appropriate species selection is of utmost importance.

Placental structure, placentation, and the subsequent relationship at the maternal-fetal interface can differ significantly between species (65). Sheep models, commonly used for fetal lung research, have a significantly different placental structure than that of the rodent/human models. These placental types differ significantly in the relative proximity of placental epithelial cells called trophoblasts and the maternal/fetal circulations (Fig. 2A). The sheep placenta has discrete epithelial barriers between mother and fetal circulations during pregnancy (synepithelialchorial). In the rodent/human hemochorial placenta, maternal blood comes into direct contact with placental trophoblasts (Fig. 2A) creating a much more intimate blood-epithelial interface (hemochorial). Because of this more intimate relationship, alterations within the maternal-fetal interface of rodents and humans are more likely to result in downstream consequences for the developing fetus. The intrauterine immune physiology of humans and rodents are also similar, particularly the predominant role of decidual natural killer cells, macrophages and T lymphocytes (62), all of which can contribute significantly to the pulmonary developmental niche. Further, rodent models are amenable to genetic manipulation for mechanistic interrogations, a technology currently not achievable in sheep. Thus, the use of rodent models will be of key importance for parallel lung and placental studies related to BPD moving forward.

Development of In Vitro Humanized Placental-Lung Models

Although animal models can approximate human physiology, model systems incorporating primary human components also have significant translational value. The pioneering work led by Kotton and colleagues differentiating human induced pluripotent stem cells (iPSC) into lung epithelial cells is a particularly attractive option for the study of human placental-lung connections in vitro (66, 67). Incorporation of placental components into these models remains the next challenge. However, the human placenta is ideally suited for translational study. Because it is routinely discarded following delivery, the placenta one of the most widely available sources of primary human tissue for research. Human amniotic fluid (which is also discarded at delivery) can also be of scientific value, through as this fluid contains a mixture of maternal and fetal secreted components, it may be more valuable as a diagnostic “biopsy” to reflect intrauterine changes rather than a component to use within in vitro model systems (Fig. 2B).

In choosing analysis techniques for these in vitro model systems, single-cell RNA sequencing approaches are a particularly important consideration. Although single-cell transcriptomic profiling has been separately explored within the placenta (68) and fetal lung (69, 70), this technology can provide exquisitely sensitive tools for investigators to identify the comprehensive molecular impact of placental components on the developing fetal lung. Finally, appropriate selection of placental components that can shape the pulmonary developmental niche (Fig. 2B) and how they can be incorporated in vitro requires a detailed knowledge of human placental collection and manipulation. Thus, collaborations between human placental researchers and lung developmental biologists will be important partnerships to move these studies forward.

In conclusion, ongoing clinical investigation for drivers of BPD will continue to be a strong foundation to understand how these associations are shifting as our clinical practice evolves. However, these clinical studies will do well to partner with continued basic science investigations focused on pathogenesis and therapeutic development, particularly as the concept of BPD phenotypes continues to evolve. These BPD phenotypes may have origins in their associations of antenatal predisposition, and research on these phenotypes will continue to benefit from increased placental investigation, shifting the placental-pulmonary connection from the “bedside to the bench.”

GRANTS

This work was supported in part by Boston University Clinical and Translational Science Institute Grant 1UL1TR001430 (to E.S.T.) and National Heart, Lung, and Blood Institute Grant 2R01HL068702-13A1 (to S.H.A.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.P., A.N., and E.S.T. drafted manuscript; A.P., A.N., G.S., S.H.A., and E.S.T. edited and revised manuscript; A.P., A.N., G.S., S.H.A., and E.S.T. approved final version of manuscript.