Obesity is a widespread condition affecting one in three adults in the United States (Yang and Barouch, 2007). Pre-existing obesity in women increases the risk of developing gestational comorbidities, including diabetes mellites, hypertension, and preeclampsia (PE) (Catalano and deMouzon, 2015; Denison et al., 2010; Howell and Powell, 2017). PE is a pregnancy-specific syndrome that occurs in the second half of gestation clinically recognized by hypertension, proteinuria, and/or other signs/symptoms (American College of Obstetricians and Gynecologists, 2013). This can lead to adverse maternal and fetal outcomes. While the exact pathophysiology of PE is unknown, inadequate placentation with poor remodeling of spiral arteries is often seen in PE and results in decreased placental perfusion. This negatively impacts fetal development, as seen in mouse studies that have shown smaller litter sizes and fetal growth restriction (FGR) (Sones et al., 2016). This highlights the necessity of appropriate blood flow to prevent placental ischemia and allow for normal fetal growth (Gilbert et al., 2009). Vascular endothelial growth factor (VEGF) is a key factor during placentation that aids in angiogenesis at the fetal-maternal interface for appropriate oxygen and nutrient supply to the placenta. This has been shown to be lower in PE pregnancies along with higher circulating soluble VEGF receptor (R)1, sFlt-1, produced with pathologic placental hypoxia and scavenges free VEGF preventing placental angiogenesis (Nevo et al., 2006). VEGF overexpression within the endometrium of mice has been shown to parallel upregulation of the anti-angiogenic factor, sFlt1 (Fan et al., 2014). This has also been observed in rodent models of PE (Woods et al., 2011; Reijnders et al., 2018).

To study the basic mechanisms of PE, we utilize a mouse model, borderline blood pressure high (BPH)/5 (Davisson et al., 2002). Female BPH/5 mice are prehypertensive and obese due to hyperphagia, and spontaneously develop a preeclamptic-like phenotype in pregnancy (Davisson et al., 2002; Sutton et al., 2017). Additional studies revealed overexpression of decidual VEGF in ad libitum fed BPH/5 female mice implantation sites at e7.5 prior to higher circulating sFlt-1 and lower placental VEGF at mid-gestation compared to control mice (Olson et al., 2020; Reijnders et al., 2018; Woods et al., 2011). BPH/5 mice exhibit FGR with lower pup and placenta weight with placental oxidative stress, and decreased litter size during pregnancy compared to control C57 mice (Dokras et al., 2006; Reijnders et al., 2019; Davisson et al., 2002). When BPH/5 mice are pair-fed to lean normotensive control mice beginning at conception they have normalization of decidual VEGF (Olson et al., 2020).

During early gestation hypoxia is necessary to promote trophoblast cell invasion and the formation of spiral arteries to increase oxygen delivery. Hypoxia stimulates the induction of different pathways, including the transcription of hypoxia inducible factor (HIF) which includes a constitutively expressed ß subunit and an oxygen-dependent α subunit (Soares et al., 2017). Activation of HIF leads to the expression of genes that encode proteins which control adaptations to hypoxia. Other pathways that serve as markers of hypoxia include stem cell factor (SCF) and heme oxygenase-1 (Ho-1), which we investigated in this study. Levels of HIF α correspondingly stabilize during early pregnancy with the occurrence of placentation. While maternal hypoxia is physiologic at specific time points in gestation, severe, prolonged, or transiently sensitive exposure to maternal hypoxia can have detrimental consequences to the fetoplacental unit (Soares et al., 2017). Such effects include shallow trophoblast invasion, narrowing and higher resistance placental arteries, which can lead to conditions such FGR and PE (Soares et al., 2017). Finally, previous studies have demonstrated increased leptin in pregnant females with PE, and that leptin gene expression is elevated during hypoxic conditions (Grosfeld et al., 2001). Elevated circulating leptin in women with PE may serve as physiological signaling for more nutrients due to decreased placental perfusion (de Knegt et al., 2021). This is an area of important research in the field. This was tested in vivo, which demonstrated that hyperleptinemic BPH/5 pregnant mice that are pair-fed have lower leptin along with reduced visceral adipose tissue (Reijnders et al., 2019). Furthermore, maternal weight loss and lower circulating leptin has been shown to be associated with attenuation of late-gestational hypertension and FGR (Beckers et al., 2023). However, the precise mechanism is under investigation. We hypothesized that leptin signaling contributes to excess hypoxia in BPH/5 decidua in early gestation resulting in poor placental blood flow and downstream adverse maternal and fetal outcomes.

To test our hypothesis, we measured hypoxia-related factors in early gestation implantation sites from BPH/5 ad libitum and pair-fed pregnant mice and compared to controls. Placental leptin expression and vasculature was assessed by histology after pair-feeding BPH/5 was continued into late-gestation to complement attenuation of FGR as previously described in this model.

Obesity is thought to contribute to PE. It is hypothesized that an obesogenic maternal environment leads to inadequate trophoblast invasion of the maternal uterine arteries. Impaired placentation with diminished placental vasculature would reduce delivery of oxygen and nutrients to the fetus and promote FGR in PE pregnancies. The role of maternal obesity and specifically leptin in early gestation in PE pregnancies has not been fully described. To better understand this, we utilized pre-hypertensive and obese BPH/5 mice that spontaneously develop a PE-like syndrome. The aim of this study was to provide evidence of pathologic hypoxia and leptin expression in early gestation that precedes impaired fetoplacental development associated with PE in this model. We hypothesized that ad libitum fed BPH/5 mice would have elevated hypoxia markers and leptin in implantation sites at e7.5 that is attenuated with maternal weight loss and reduced leptin via pair-feeding.

In this study, the tested markers all respond to hypoxia through different pathways. In the HIF pathway during normoxic conditions, HIFα subunit binds to von-Hippel-Lindau protein, allowing for it to become ubiquitinated and then degraded (Soares et al., 2017). During hypoxic conditions, the HIFα subunit instead stabilizes and interacts with the HIFß subunit and binds coactivators to regulate genes encoding proteins that control adaptions to hypoxia, including VEGF (Soares et al., 2017). Expression of SCF can also be induced by hypoxia and will bind to its receptor, c-kit, to aid in the regulation of placental hemopoietic cell clusters (Broudy, 1997; Sasaki et al., 2010). Ho-1 is a membrane protein that cleaves b-type heme molecules to produce biliverdin, carbon monoxide, and iron as the first rate-limiting step in heme catabolism (Lee et al., 1997). The expression of Ho-1 is induced by stressful conditions, including hypoxia, and the consequent production bilirubin (converted from biliverdin), allows for potent scavenging of free radicals (Foresti et al., 2001). Our results demonstrated a pathologic overexpression of hypoxia markers in ad libitum fed BPH/5 e7.5 implantation sites in comparison to physiological expression levels observed in control C57 mice. These markers were reduced with pair-feeding BPH/5 mice. Hypoxia in the decidua in early pregnancy may contribute to VEGF overexpression in ad libitum BPH/5 e7.5 implantation sites, which similarly normalizes with pair feeding (Olson et al., 2020).

Additionally, we investigated leptin expression during early BPH/5 gestation. Previous studies demonstrated elevated circulating leptin levels in pregnant BPH/5 mice serum (Reijnders et al., 2019), which could be due to adipose tissue and/or placental production of leptin. Therefore, we measured leptin and leptin receptor mRNA expression in e7.5 implantation sites and placental leptin at e18.5. Ad libitum BPH/5 e7.5 implantation sites had a greater than 3-fold expression of Lep compared to C57 controls, which normalized with pair-feeding. Contrary to Lep, LepR expression was not significantly elevated in ad libitum fed BPH/5 e7.5 implantation sites compared to control mice but did decrease with pair-feeding. These results could be due to the influence of systemic leptin exposure and priming of the maternal decidua early in BPH/5 pregnancy. This is an area of ongoing investigation in the laboratory.

After demonstrating both leptin and hypoxia marker expression was higher in ad libitum BPH/5 e7.5 implantation sites compared to ad libitum C57 e7.5, we investigated if there was a causal relationship between leptin exposure and hypoxia marker expression in the decidua. Cultured e7.5 decidua explants from only BPH/5 exhibited a statistically significant elevation in hypoxia markers, Hif1a and Ho-1, when cultured with leptin compared to decidua cultures without leptin. The discrepancy whereby BPH/5 but not C57 responded to exogenous leptin in vitro may be due to higher leptin sensitivity in BPH/5 tissues from exposure in vivo. Ho-1 has been shown to have potential protective capabilities in cells through possible antioxidative, anti-inflammatory, anti-apoptotic effects (Zeng et al., 2013). We speculate that the higher expression of Hif1a with leptin addition could potentiate the expression of Ho-1, which is a downstream target of Hif1a, and have a beneficial effect in BPH/5 mice due to a greater need for cytoprotective properties. It is possible that high levels of Ho-1 is a physiological response that promotes maintenance of pregnancy as BPH/5 mice do have term deliveries. Further investigation is needed to better understand the extent of the effect leptin has on hypoxia marker expression and the impact of this on pregnancy. We briefly tested if transient exposure to hypoxia would induce higher expression of leptin in C57 and BPH/5 decidua compared to normoxic controls. We did not show an elevation of leptin expression in hypoxic decidua suggesting leptin contributes more to expression of hypoxia markers, but these results warrant further experimentation. One hypothesized explanation for the detrimental impact of abnormally upregulated leptin on adverse pregnancy outcomes and PE development is inducible nitric oxide synthase (iNOS) expression. While iNOS is normally expressed in the first half of pregnancy to aid vasodilation and placental development, previous studies on BPH/5 mice have demonstrated an upregulation of iNOS expression (Heyward et al., 2017). Upregulated iNOS has been linked to poor pregnancy outcomes, such as embryo loss. While the intermediate steps are not completely understood, previous rodent studies demonstrated an elevation in leptin leads to increased NO synthesis through elevated iNOS expression (Yang and Barouch, 2007). Other research demonstrated iNOS to be a downstream target of STAT3, which is activated by the Ob-Rb/JAK2 complex that forms when leptin binds to its receptor, though this was seen in astrocytes during transformation (Osuka and Van Meir, 2017). This pathway of activation, from leptin, leptin receptor, Ob-Rb/JAK2, STAT3, leading to iNOS, may explain and contribute to adverse pregnancy outcome seen in the PE-like BPH/5 mice.

To investigate further downstream impacts hypoxia and leptin can have on placental angiogenesis, placental vasculature area was measured. While previous studies have demonstrated ad libitum BPH/5 mice exhibit inadequate spiral artery formation and smaller placental vessel area with oxidative stress, in this study pair-fed BPH/5 e18.5 placenta have a greater luminal area of placental vessels in comparison to ad libitum fed BPH/5 e18.5 placenta. These results could be associated with aberrant hypoxia and leptin levels in early BPH/5 gestation. Further investigation is warranted to better understand the effect of hypoxia and leptin on the development of PE.

Limitations of this study include the method of weight loss. Restricting the diet of the mice has the potential to place them in a negative energy balance which can allow for the development of adverse fetal outcomes including decreased nephron numbers, hypertension, proteinuria, or renal dysfunction after birth (Luyckx, 2020). While excessive dietary restriction can be harmful, the method of food restriction utilized, pair feeding, which is matched to lean, gestation-matched control mice, acts to normalize obese BPH/5 mice weight and adiposity, and serum leptin by e7.5 (Reijnders et al., 2019), and has demonstrated improvement in fetal and maternal outcomes (Beckers et al., 2023). Pair-feeding BPH/5 showed a similar reduction in serum leptin by e7.5 (83%) as by e18.5 (95%). Although the placenta contributes to circulating leptin more by late gestation than in early, the reduction at e18.5 in BPH/5 may be due to decreased adipose tissue as well as placental expression. This warrants additional exploration to determine the contribution of each organ. Further investigation into hypoxia at the maternal-fetal interface and its effects in the development of PE is necessary. For example, looking into the expression of other markers of hypoxia or proteins in hypoxia-induced pathways, such as hypoxia inducible factor 2a, insulin-like growth factor-II, E-cadherin, or Lysine Demethylase 3A. Another potential area of study includes further localization of the hypoxia at the fetal-maternal interface to determine if it is localized in the fetus, placenta, and/or endometrium. There is also a need to better understand the role of leptin in PE and the exact mechanism behind its elevated expression and its influence at the maternal-fetal interface. Other murine preeclamptic models could also be used to further verify and expand our current findings; such models to consider in future studies include Hif1a knock-out mice or the reduced uterine perfusion pressure (RUPP) model with similar placental ischemia. In conclusion, maternal obesity may contribute to pathologic hypoxia and leptin expression at the fetal-maternal interface during gestation. A reduction in maternal weight and adiposity through pair-feeding normalizes these abnormalities in a mouse model of PE, BPH/5, and the associated angiogenic imbalance. Further studies are required to better understand the impact of abnormal leptin and hypoxia on placental ischemia in PE.