Introduction

In pregnancy, the development of hypercholesterolemia (HC) is a physiological process that is required to support the normal growth and development of the fetus1 but, at the same time, excessive (e)HC in pregnancy (total cholesterol > 280 mg/dL2–6) is a significant risk factor for the development of pregnancy complications, such as preeclampsia (PE, reviewed in7). This total cholesterol cut-off point was established based on the documented adverse effects of eHC on the placenta and fetoplacental vasculature2,3,4,5,6. And, it seems that it is not only pre-existing HC, but also eHC that develops specifically during pregnancy8,9,10, as a meta-analysis has reported that, compared to women who had uncomplicated pregnancies, women who develop PE have greater increases in cholesterol levels during pregnancy10. Further supporting this argument, we and others have demonstrated that pregnant rats fed a hypercholesterolemic diet during pregnancy (leading to pregnancy-specific eHC) present PE-like signs, including high blood pressure, proteinuria, and vascular dysfunction that persists after delivery11,12,13,14. Moreover, the fetal-to-placental weight ratio is reduced in eHC pregnancies, regardless of the fetal sex, suggesting reduced placental efficiency11,13. Placental insufficiency underlies the maternal syndrome of various pregnancy complications (by releasing placental-derived factors)15, while also having a negative impact on offspring health16. However, the mechanisms underlying impaired placental function in eHC pregnancies are not known.

In conditions of maternal obesity, which often are associated with eHC, there is accumulation of lipids in the placenta17, but whether such a process happens in pregnancies complicated by eHC that develops during pregnancy remains to be assessed. An increased lipid deposition in the placenta may lead to oxidative stress, endoplasmic reticulum (ER) stress (see below), and increase the trafficking of lipoproteins to the fetal circulation. The transfer of lipoproteins from the maternal to the fetal circulation requires their uptake by the syncytiotrophoblast via receptors, such as the scavenger receptor class B type I (SR-BI), low-density lipoprotein receptor (LDLR), very-low-density lipoprotein receptor (VLDLR), and lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1). Thereafter, free cholesterol particles are released from the basal side of the syncytiotrophoblast via ATP-binding cassette transporter A1 (ABCA1) and ATP-binding cassette sub-family G member 1 (ABCG1). Upon contact with fetal endothelial cells, these cholesterol particles are uptake via SR-BI and LDLR, and subsequently ABCA1 and ABCG1 release these cholesterol particles to acceptors in the fetal circulation18,19. Interestingly, ABCA1, but not ABCG1, is also expressed in the apical side of the syncytiotrophoblast, likely contributing to bidirectional cholesterol trafficking18,20. However, if eHC in pregnancy alters the levels of these receptors and/or transporters in the male and female placentas is also not known.

Previous studies have demonstrated that both maternal obesity and PE are associated with oxidative stress and the accumulation of misfolded/unfolded proteins, which induces ER stress and activates the unfolded protein response (UPR)17,21. Under normal conditions, the ER sensors of the UPR are bound to glucose-regulated protein 78 (GRP78), and consequently, they are held in an inactive state21. However, accumulation of misfolded/unfolded proteins, with subsequent displacement of GRP78, can activate the three highly conserved arms of the UPR; the activating transcription factor 6 (ATF6; to increase protein folding capacity), the inositol-requiring enzyme 1α (IRE1α; to regulate cell survival), and the protein kinase RNA-like ER kinase (PERK; which phosphorylates eIF2α [eukaryotic translation initiation factor 2α] to inhibit non-essential protein synthesis)21. Of note, phosphorylation of eIF2α also favors translation of stress-related mRNAs, including ATF4, which functions to induce the transcription of C/EBP homologous protein (CHOP), a multifaceted transcription factor that has the capacity to dephosphorylate eIF2α via GADD3422. In addition, dysregulated/persistent activation of the UPR can lead to the activation of the NOD-, LRR-, and pyrin domain-containing protein 3 (NLRP3) inflammasome23 by increasing thioredoxin-interacting protein (TXNIP) mRNA stability24. Interestingly, activation of the NLRP3 inflammasome requires two steps: the priming step (which induces the expression of the components of the NLRP3 inflammasome) and the activating step (which induces the assembly of the NLRP3 inflammasome)25,26. The priming step can happen in response to an inflammatory stimuli, such as activation of the innate immune receptor toll-like receptor 4 (TLR4), while the activating step can be triggered pathogen- or damage-associated molecular patterns, such as cholesterol crystals25,26. Nevertheless, activation of the NLRP3 inflammasome can lead to the release of pro-inflammatory mediators (IL-1β and IL-18) and alarmin signals (such as high mobility group box 1 [HMGB1]), as well as pyroptosis (a form of inflammatory cell death that is mediated via N-terminal gasdermin D)25,26. However, if eHC in pregnancy induces ER stress, activating the placental UPR and NLRP3 inflammasome in the male and female placentas remains to be determined.

In this study, we investigated the impact of eHC in pregnancy on the placentas of both the male and female offspring. We hypothesized that, regardless of the fetal sex, eHC leads to the accumulation of lipids in the placentas, which alters the levels of cholesterol receptors and transporters and induces ER stress, triggering the activation of the UPR and NLRP3 inflammasome.

Materials and methods

Ethics approval

All animal procedures comply with the Canadian Council on Animal Care Guidelines, and were approved by the Animal Care and Use Committee at the University of Alberta (AUP-3692). All animal data are reported in accordance with the ARRIVE reporting guidelines27.

Animal model and experimental design

The eHC model has been previously characterized by our group and others11,12,13,28,29. In brief, Sprague Dawley rats of both sexes (15 females and 6 male breeders) were purchased from Charles River Laboratories at twelve weeks of age. Same sex rats were housed in pairs in the animal facility at the University of Alberta under standard laboratory conditions with unrestricted access to water and food. All rats were allowed to acclimate for at least one week before experimentation. Female and male rats were mated overnight (1:1 ratio), and the presence of sperm in a vaginal smear the next morning indicated that the female rat was pregnant. This was considered gestational day 0, after which dams were single housed until endpoint. On gestational day 6, dams were arbitrarily allocated to receive a control diet (CD, standard rat chow – total digestible nutrients: 76% yielding 3.04 kcal/g) or a high cholesterol diet (to model eHC, Modified LabDiet® 5001 with 2% cholesterol and 0.5% sodium cholate – total digestible nutrients: 74.5% yielding 3.36 kcal/g) with all other macro- and micro-nutrients being comparable between the two diets until gestational day 20. Dams were then anesthetized with inhaled isoflurane (4% in 100% O2) and sacrificed by exsanguination via cardiac puncture. The humane endpoints set for this study were body weight loss > 10% or visible signs of illness in the dams, neither of which were observed in this cohort. Thereafter, each fetus/placenta was sexed (based on the fetus ano-genital distance), and the whole placentas (including the labyrinth and junctional zones) were either snap-frozen embedded in Tissue-Tek OCT Compound (Sakura Finetek Inc., Torrance, CA, USA) for cryosectioning or snap-frozen for protein extraction. Maternal phenotype and pregnancy outcomes data, including fetal and placental weights, for this cohort have been published elsewhere11.

Detection of lipids in the male and female CD and eHC placentas

Placenta sections (thickness: 9–12 μm) were fixed in 70% ethanol for 5 min, and stained with 1% Sudan IV (Sigma-Aldrich, St. Louis, MO, USA) diluted in 95% ethanol for 1 min. Then, slides were washed in 50% ethanol for 5 s, rinsed 2 times in Scott’s Bluing Reagent (RICCA Chemical Company, Arlington, TX, USA), and mounted with Fisher Chemical™ Permount™ Mounting Medium (ThermoFisher Scientific, Waltham, MA, USA) for visualization. Slides were imaged under 10x magnification (labyrinth zone) using a bright field microscope (EVOS XL Core Imaging System, ThermoFisher Scientific), and images were analyzed with FIJI software version 2.14.0/1.54f30. Color deconvolution tools were used to assess lipid levels, excluding non-specific staining and folded areas, and the reported data are presented as the % of positive area.

Total cholesterol levels in the male and female CD and eHC plasma

Male and female CD and eHC fetuses were sacrificed via decapitation, and pooled trunk blood was collected from the exposed neck vessels into EDTA-coated tubes (Microvette® CB 300 EDTA K2E, SARSTEDT AG & Co. KG, Germany). Within 30 min of collection, the blood was centrifuged at 2,000 rpm for 5 min at 4° C, and the plasma fraction was collected and snap-frozen until further analysis. Total cholesterol levels were assessed using a colorimetric assay kit (catalog number NBP3-25834, Bio-Techne, Toronto, ON, Canada) according to the manufacturer’s instructions.

Detection of reactive oxygen species (ROS) in the male and female CD and eHC placentas

Placenta sections (thickness: 9–12 μm) were assessed using dihydroethidium (Biotium Inc., Fremont, CA, USA) fluorescent staining, as previously described by our group11,31. Slides were imaged under 10x magnification (labyrinth zone; 3 images per slide) using an Olympus IX81 Microscope (Olympus Canada Inc., Toronto, ON, Canada), and the mean fluorescent intensity (arbitrary units) was calculated with ImageJ software (version 1.53), excluding non-specific staining and folded areas. The mean fluorescent intensity of the 3 images was averaged, and the data are reported as arbitrary units.

Western blotting for cholesterol receptors and transporters, ER stress markers, and NLRP3 inflammasome in the male and female CD and eHC placentas

The Pierce™ BCA Protein Assay kit (ThermoFisher Scientific) was used to determine the total protein concentration in placental homogenates. Equal amounts of protein (ranging from 25 to 100 µg, depending on the protein of interest) were loaded into SDS-polyacrylamide gels, separated by electrophoresis, and transferred to 0.2 μm nitrocellulose membranes (100 V for 1 h or 25 V overnight [used only for NLRP3]; Bio-Rad Laboratories Inc., Hercules, CA, USA). Li-Cor Revert™ 700 Total Protein Stain (Li-Cor Biosciences, Lincoln, NE, USA) was used following the manufacturer’s instructions to stain the membranes, which were subsequently visualized with either the ChemiDoc™ MP Imaging System (Bio-Rad Laboratories Inc.) or the Li-Cor Odyssey Imaging System (Li-Cor Biosciences). Membranes were then blocked for 1.5 h using a blocking solution with 50% Blocking Buffer for Fluorescent Western Blotting (Rockland, Limerick, PA, USA) diluted in 1x phosphate-buffered saline (PBS) and incubated overnight at 4º C or at room temperature for 1.5 h [used only for NLRP3] with primary antibodies (for details, please see Supplemental Table 1) diluted in 1x PBS with 0.1% Tween 20 (PBST). After the membranes were washed 3 times for 10 min with PBST, they were incubated with their corresponding secondary antibodies (IRDye® 800CW donkey anti-rabbit IgG or anti-mouse IgG or anti-goat IgG, or IRDye® 680RD donkey anti-mouse IgG or anti-goat IgG) at 1:10,000 dilution in 0.1% PBST with 25% Blocking Buffer for Fluorescent Western Blotting (Rockland) for 1 h, and washed 3 times for 10 min with PBST. Membranes were visualized either with the ChemiDoc™ MP Imaging System (Bio-Rad Laboratories Inc.) or the Li-Cor Odyssey Imaging System (Li-Cor Biosciences). Band densitometry quantified with Image Lab software version 6.1.0 or Li-Cor Image Studio (version 3.0; Li-Cor Biosciences). The reported values were normalized to the total protein staining, with the exception of phosphorylated IRE1αSer724 (normalized to IRE1α), phosphorylated eIF2αSer51 (normalized to eIF2α), cleaved ATF6 (normalized to full length ATF6), caspase 1 (normalized to pro-caspase 1), and caspase 8 (normalized to pro-caspase 8). Results are presented as % of the mean of the CD group in each membrane.

Detection of IL-1β and IL-18 levels in the male and female CD and eHC placentas

The levels of IL-1β and IL-18 were assessed in placental homogenates of equal protein concentration as determined with the Pierce™ BCA Protein Assay kit (ThermoFisher Scientific). All samples were analyzed with Luminex® xMAP® Technology by Eve Technologies (Calgary, AB, Canada), and the optimal protein concentration (8 µg/mL) was determined based on pilot studies.

Statistical analysis

Data were analyzed within each sex with a two-tailed unpaired t-test followed by an F-test or a two-tailed unpaired t-test with Welch’s correction (if no equal variance) using the GraphPad Prism software version 10.0.3 (GraphPad Software, San Diego, CA, USA). Outliers were identified with the Grubb’s test, and removed from the statistical analysis. Data are shown as means±SEM and n represents the number of dams from which male and female placentas were arbitrarily selected (significance = p≤0.05).

Results

Impact of eHC on placental lipid and ROS levels

Lipid levels were increased in the placentas of the male (Fig. 1Ai; p = 0. 0.0352) and female (Fig. 1Bi; p = 0.0144) eHC offspring compared to CD. However, regardless of the fetal sex, the levels of ROS were similar between the CD and eHC groups (Fig. 1Aii and 1Bii).

Impact of eHC on placental cholesterol receptors and transporters

In the male placentas, the protein expression levels of LDLR were decreased (Fig. 2Ai and 2Aiii; p = 0.0158), while the protein expression levels of ABCA1 were increased (Fig. 2Ai and 2Avi; p = 0.0035), in the eHC group compared to CD. The protein expression levels of SR-BI (Fig. 2Ai and 2Aii), VLDLR (Fig. 2Ai and 2Aiv), LOX-1 (Fig. 2Ai and 2Av), and ABCG1 (Fig. 2Ai and 2Avii) were similar between the CD and eHC groups.

In the female placentas, the protein expression levels of LDLR (Fig. 2Bi and 2Biii; p = 0.0388), VLDLR (Fig. 2Bi and 2Biv; p = 0.0068), and LOX-1 (Fig. 2Bi and 2Bv; p = 0.0043) were decreased in the eHC group compared to CD. No differences were observed in the protein expression levels of SR-BI (Fig. 2Bi and 2Bii), ABCA1 (Fig. 2Bi and 2Bvi), and ABCG1 (Fig. 2Bi and 2Bvii) between the CD and eHC groups.

Impact of eHC on fetal cholesterol levels

Total cholesterol levels were reduced in the plasma of the male eHC offspring compared to CD (Fig. 3A; p = 0.0131). On the other hand, total cholesterol levels were similar between the CD and eHC female offspring (Fig. 3B).

Fig. 1
figure 1

Lipid levels were increased in the male and female eHC placentas. Representative images and quantitative analysis of Sudan IV (lipids are stained in orange; i) and dihydroethidium (DHE, reactive oxygen species are stained in red; ii) staining in the placentas of the control diet (CD, closed symbols) and high cholesterol diet (eHC, open symbols) male (A) and female (B) offspring. Images were adjusted for brightness and contrast using the same parameters. Scale bar = 100 μm. Data are shown as means ± SEM; *p < 0.05 using a two-tailed unpaired t-test, n = 5/group. a.u.: Arbitrary units; MFI: Mean fluorescence intensity.

Full size image
Fig. 2
figure 2

LDLR levels were reduced in the male and female eHC placentas, VLDLR and LOX-1 levels were reduced in only the female eHC placentas, and ABCA1 levels were increased in only the male eHC placentas. Representative blots (i) and densitometry of scavenger receptor class B type I (SR-BI, normalized to total protein; ii), low-density lipoprotein receptor (LDLR, normalized to total protein; iii), very-low-density lipoprotein receptor (VLDLR, normalized to total protein; iv), lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1, normalized to total protein, v), ATP-binding cassette transporter A1 (ABCA1, normalized to total protein; vi), and ATP-binding cassette sub-family G member 1 (ABCG1, normalized to total protein; vii) in the placentas of the control diet (CD, closed symbols) and high cholesterol diet (eHC, open symbols) male (A) and female (B) offspring. Data are shown as means ± SEM; *p < 0.05 and **p < 0.01 vs. CD using a two-tailed unpaired t-test, n = 4–5/group. Full uncropped blots are presented in Figure S1-S6.

Full size image
Fig. 3
figure 3

Total cholesterol levels were reduced in only the male eHC offspring. Total cholesterol levels in the plasma of the control diet (CD, closed symbols) and high cholesterol diet (eHC, open symbols) male (A) and female (B) offspring. Data are shown as means ± SEM; *p < 0.05 vs. CD using a two-tailed unpaired t-test, n = 5–6/group.

Full size image
Fig. 4
figure 4

Phosphorylated IRE1α levels were reduced in only the male eHC placentas while phosphorylated eIF2α were reduced in only the female eHC placentas. Representative blots (i) and densitometry of glucose-regulated protein 78 (GRP78, normalized to total protein; ii), cleaved activating transcription factor 6 (c-ATF6, normalized to full length ATF6; iii), phosphorylated inositol requiring-enzyme 1αSer724 (p-IRE1αSer724, normalized to IRE1α; iv), phosphorylated eukaryotic translation initiation factor 2αSer51 (p-eIF2αSer51, normalized to eIF2α; v), activating transcription factor 4 (ATF4, normalized to total protein; vi), and C/EBP homologous protein (CHOP, normalized to total protein; vii) in the placentas of the control diet (CD, closed symbols) and high cholesterol diet (eHC, open symbols) male (A) and female (B) offspring. Data are shown as means ± SEM; *p < 0.05 vs. CD using a two-tailed unpaired t-test, n = 5–8/group. Full uncropped blots are presented in Figure S7-S12.

Full size image
Fig. 5
figure 5

NLRP3 levels were increased and TLR4 levels were reduced in only the male eHC placentas while TXNIP were reduced in only the female eHC placentas. Representative blots (i) and densitometry of thioredoxin-interacting protein (TXNIP, normalized to total protein; ii), NOD-, LRR-, and pyrin domain-containing protein 3 (NLRP3, normalized to total protein; iii), and toll-like receptor 4 (TLR4, normalized to total protein; iv) in the placentas of the control diet (CD, closed symbols) and high cholesterol diet (eHC, open symbols) male (A) and female (B) offspring. Data are shown as means ± SEM; *p < 0.05 vs. CD using a two-tailed unpaired t-test, n = 5–8/group. Full uncropped blots are presented in Figure S13-S15.

Full size image
Fig. 6
figure 6

IL-1β levels were reduced and caspase 8 levels were increased in only the male eHC placentas while HMGB1 levels were reduced in only the female eHC placentas. Interleukin (IL)−1β (i) and IL-18 (ii) levels, as well as representative blots (iii) and densitometry of N-terminal gasdermin D (NT-GSDMD, normalized to total protein; iv) and high mobility group box 1 (HMGB1, normalized to total protein; v) and representative blots (vi) and densitometry of caspase 1 (normalized to pro-caspase 1; vii) and caspase 8 (normalized to pro-caspase 8; viii), in the placentas of the control diet (CD, closed symbols) and high cholesterol diet (eHC, open symbols) male (A) and female (B) offspring. Data are shown as means ± SEM; *p < 0.05 vs. CD using a two-tailed unpaired t-test, n = 4–7/group. Full uncropped blots are presented in Figure S16-S19.

Full size image

Impact of eHC on the placental unfolded protein response

In the male placentas, the protein expression levels of phosphorylated IRE1αSer724 were decreased (Fig. 4Ai and 4Aiv; p = 0.0244) in the eHC group compared to CD. The protein expression levels of GRP78 (Fig. 4Ai and 4Aii), cleaved ATF6 (Fig. 4Ai and 4Aiii), phosphorylated eIF2αSer51 (Fig. 4Ai and 4Av), ATF4 (Fig. 4Ai and 4Avi), and CHOP (Fig. 4Ai and 4Avii) were similar between the CD and eHC groups.

In the female placentas, the protein expression levels of phosphorylated eIF2αSer51 were reduced in the eHC group compared to CD (Fig. 4Bi and 4Bv; p = 0.0060). No differences were observed in the protein expression levels of GRP78 (Fig. 4Bi and 4Bii), cleaved ATF6 (Fig. 4Bi and 4Biii), phosphorylated IRE1αSer724 (Fig. 4Bi and 4Biv), ATF4 (Fig. 4Bi and 4Bvi), and CHOP (Fig. 4Bi and 4Bvii) between the CD and eHC groups.

Impact of eHC on the placental NLRP3 inflammasome

In the male placentas, the protein expression levels of NLRP3 (Fig. 5Ai and 5Aiii; p = 0.0197) and caspase 8 (Fig. 6Avi and 6Aviii; p = 0.0340) were increased, while the protein expression levels of TLR4 (Fig. 5Ai and 5Aiv; p = 0.0525) tended to be reduced and the protein expression levels of IL-1β (Fig. 6Ai; p = 0.0372) were reduced, in the eHC group compared to CD. The protein expression levels of TXNIP (Fig. 5Ai and 5Aii), IL-18 (Fig. 6Aii), N-terminal gasdermin D (Fig. 6Aiii and 6Aiv), HMGB1 (Fig. 6Aiii and 6Av), and caspase 1 (Fig. 6Avi and 6Avii) were similar between the CD and eHC groups.

In the female placentas, the protein expression levels of TXNIP (Fig. 5Bi and 5Bii; p = 0.0255) and HMGB1 (Fig. 6Biii and 6Bv; p = 0.0327) were reduced in the eHC group compared to CD. No differences were observed in the protein expression levels of NLRP3 (Fig. 5Bi and 5Biii), TLR4 (Fig. 5Bi and 5Biv), IL-1β (Fig. 6Bi), IL-18 (Fig. 6Bii), N-terminal gasdermin D (Fig. 6Biii and 6Biv), and caspase 1 and 8 (Fig. 6Bvi-viii) between the CD and eHC groups.

Discussion

In this study, we show that eHC in pregnancy leads to lipid accumulation in the male and female placentas without altering ROS levels in either sex. However, eHC reduced placental cholesterol receptors (LDLR in males and females and VLDLR and LOX-1 only in females), increased placental cholesterol transporters (ABCA1 only in males), lowered fetal cholesterol levels (total cholesterol only in males), and altered the placental UPR (reduced phosphorylated IRE1αSer724 in males and phosphorylated eIF2αSer51 in females) in a sex-specific manner. Moreover, TLR4 levels tended to be reduced (indicating prolonged receptor activation), while NLRP3 levels were increased, in only the male eHC placentas. This was associated with reduced IL-1β levels in only the male eHC placentas (which is rapidly released by the cell), but not with pyroptosis. The levels of caspase 8, but not caspase 1, were increased in only the male eHC placentas, suggesting that IL-1β processing may have happened via a non-canonical pathway. In summary, eHC in pregnancy impacts the placentas of both the male and female offspring (Fig. 7), but the mechanisms are sex-specific.

Fig. 7
figure 7

Schematic summary of the effects of eHC on the male and female placentas. ABCA1: ATP-binding cassette transporter A1; ABCG1: ATP-binding cassette sub-family G member 1; ATF6: Activating transcription factor 6; eIF2α: Eukaryotic translation initiation factor 2α; ER: Endoplasmic reticulum; GRP78: Glucose-regulated protein 78; GSDMD: Gasdermin D; GSDMDNT: N-terminal gasdermin D; HMGB1: High mobility group box 1; IL: Interleukin; IRE1α: Inositol requiring-enzyme 1α; LDLR: Low-density lipoprotein receptor; LOX-1: Lectin-like oxidized low-density lipoprotein receptor-1; NLRP3: NOD-, LRR-, and pyrin domain-containing protein 3; ROS: Reactive oxygen species; SR-BI: Scavenger receptor class B type I; TLR4: Toll-like receptor 4; TXNIP: Thioredoxin-interacting protein; VLDLR: Very low-density lipoprotein receptor. Created in BioRender.com.

Full size image

Maternal circulating levels of cholesterol (including total, HDL, and LDL cholesterol) are increased in rat eHC pregnancies and associated with reduced placental efficiency (as measured by the fetoplacental weight ratio) in both fetal sexes11. Here, we found increased lipid, but not ROS levels in both the male and female eHC placentas in late-gestation. This increase in lipid deposition in the placentas of the eHC groups compared to controls aligns with previous studies showing high placental lipid contents in conditions of maternal HC associated with obesity32,33,34,35. And, since late-gestation placental lipid accumulation induces lipotoxicity (a process that causes cellular distress and dysfunction)36, our data support the notion that excessive pregnancy-specific HC detrimentally impacts the placenta.

We have also observed that the protein expression levels of two main cholesterol receptors, LDLR (in males and females) and VLDLR (only in females), are reduced in the placentas of the eHC groups compared to controls. Interestingly, no changes in the protein expression levels of SR-BI (the HDL receptor) were observed between the CD and eHC groups, regardless of the fetal sex. Trophoblasts isolated from term-placentas of women with excessively high cholesterol levels during pregnancy (total cholesterol > 280 mg/dL) have a reduced capacity to uptake cholesterol, including LDL and HDL37. Here, we speculate that downregulation of the cholesterol receptors in eHC pregnancies may be an adaptive response to lessen the impact of eHC on the placenta (and fetus). However, as mentioned, lipid contents are increased in eHC placentas compared to controls, suggesting an insufficient adaptative response. Moreover, we found that the protein expression levels of ABCA1, but not ABCG1, are increased in only the placentas of the male eHC offspring compared to controls. Our findings with ABCA1 in male eHC placentas, but not ABCG1, are in line with a previous study that showed increased placental levels of ABCA1 (primarily located in the apical membrane) in women with excessively high cholesterol levels in pregnancy coupled with reduced ABCG137, however the sex of the offspring was not reported. Our data suggest that male eHC placentas have a greater capacity to efflux lipids, but since we have performed Western blotting using whole tissue, further studies are warranted to assess whether ABCA1 colocalizes to the apical membrane, basal membrane, or both. Interestingly, plasma cholesterol levels were reduced only in male eHC fetuses, indicating that increased protein expression levels of ABCA1 did not result in fetal HC. Furthermore, the circulating levels of oxidized LDL are also increased in eHC pregnancies11. Thus, we assessed if eHC affects the protein expression levels of LOX-1, the main receptor for oxidized LDL38. Here, we found that LOX-1 protein expression levels are reduced in only the female eHC placentas, and continuous activation of LOX-1 can lead to its downregulation via shedding of its ectodomain39. Accordingly, reduced LOX-1 levels in female eHC placentas suggest prolonged LOX-1 activation, which, in turn, indicates that female eHC placentas may have internalized more oxidized LDL compared to controls.

As mentioned, HC can induce ER stress, including in conditions of maternal obesity17. Thus, we next assessed if excessive pregnancy-specific HC impacts this pathway. We found that eHC reduces the protein expression levels of phosphorylated IRE1αSer724 in only the placentas of the male offspring. IRE1α is as specialized ER resident protein that is critical for placental development and embryonic viability40, and the displacement of GRP78 (the master regulator of the UPR) from IRE1α allows for its phosphorylation and dimerization, which can initiate pro-survival or pro-apoptotic (in the presence of irremediable ER stress) pathways41. While we did not observe differences between the CD and eHC male placentas in the protein expression levels of GRP78, previous studies have shown that IRE1α activation may occur in response to its direct interaction with misfolded/unfolded proteins and, therefore, independently of GRP7842. Paradoxically, conditions of high, prolonged ER stress can lead to the attenuation of IRE1α expression and signaling (reviewed in41,43), which is associated with apoptotic cell death44. Likewise, diminished IRE1α phosphorylation at Ser724 has been shown to impair the IRE1α autophosphorylation (activation) process and to exacerbate ER stress-induced cell damage45. Thus, reduced phosphorylated IRE1αSer724 levels in male eHC placentas may indicate that these placentas were exposed to high, prolonged ER stress, which aligns with the timepoint (late-gestation) in which our analyses were performed. On the other hand, we found that, compared to controls, the protein expression levels of phosphorylated eIF2αSer51 are reduced in only the placentas of the female eHC offspring. eIF2α is the downstream target of PERK (another ER resident protein)41 and, interestingly, PERK signaling remains active even in conditions of high, prolonged ER stress exposure. Reduced phosphorylated eIF2αSer51 protein expression levels have been reported in aged tissues and were associated with a decline in the overall capacity of the ER to avoid the accumulation of misfolded/unfolded proteins46. Therefore, it may be suggested that eHC impacts the UPR in the placentas of both the male and female offspring, but via sex-specific mechanisms. Sex-specific differences in the activation of the placental UPR have also been reported in other models of pregnancy complications, including prenatal hypoxia47 and advanced maternal age48, which highlights the importance of sexing the offspring/placentas and assessing each sex separately.

The NLRP3 inflammasome is a known inducer of inflammatory cell death downstream of IRE1α24. Thus, we next assessed if eHC alters placental NLRP3 levels. We found that, compared to controls, the protein expression levels of NLRP3 are increased in only the placentas of the male eHC offspring. While information on NLRP3 inflammasome activation in placentas of women with eHC that developed specifically during pregnancy lack in the literature, it has been demonstrated that, in mice, exposure to palmitic acid (a saturated free fatty acid that is increased in obese pregnant women) can activate the placental NLRP3 inflammasome, leading to IL-1β secretion49. Increased NLRP3 protein expression levels have also been observed in placentas from women with PE50,51 and in placentas of the reduced uterine perfusion pressure model of PE52. However, to our knowledge, the present study is the first to report that offspring sex can impact placental NLPR3 levels. Of note, the link between high, prolonged ER stress via IRE1α and activation of the NLRP3 inflammasome has been established through TXNIP24. Here, we hypothesized that the placentas of the male eHC offspring would have increased TXNIP levels compared to controls. Contrary to our hypothesis, we did not observe changes in the protein expression levels of TXNIP in male eHC placentas. This suggests that the modulation of the NLRP3 inflammasome in male eHC placentas may occur independently of changes in the ER. However, we cannot rule out a potential contribution of the UPR via IRE1α in the mechanisms underlying increased NLRP3 levels in male eHC placentas, as this process may have occurred at an earlier timepoint. As well, since we only assessed TXNIP levels in late-gestation this may be why we did not observe changes in TXNIP. Noteworthy, we have also found signs of prolonged TLR4 activation in male eHC placentas in this study, and TLR4 is a known inducer of the NLRP3 inflammasome53. In addition, cholesterol crystals (which are likely increased in these placentas due to lipid accumulation) have been shown to modulate the NLRP3 inflammasome54. However, since the placentas of both the male and female eHC offspring have lipid accumulation, a potential mechanism by which only the male eHC placentas would be affected remains to be addressed. We speculate that it is the synergy of all of these mechanisms (i.e., irremediable ER stress, prolonged TLR4 activation, and lipid accumulation) in the male eHC placentas that culminates in the initiation of the NLRP3 inflammasome. Nevertheless, our findings of increased NLPR3 protein expression levels in only the placentas of the male eHC offspring suggest that the male offspring may be more susceptible to the impact of eHC.

Upon priming and activation of the NLRP3 inflammasome there can be release of inflammatory mediators (IL-1β and IL-18) and alarmin signals (such as HMGB1), as well as pyroptosis (via N-terminal gasdermin D). Thus, we next assessed whether eHC affects the expression levels of these proteins. We found that the protein expression levels of IL-1β, but not IL-18 or N-terminal gasdermin D (indicating that pyroptosis is not occurring), are reduced in only the male eHC placentas compared to controls. As pro-IL-1β is cleaved into IL-1β (which is rapidly released by the cell)55, we speculate that this reduction in placental IL-1β indicates that this cytokine is being cleaved and released into the extracellular milieu. The cleavage of pro-IL-1β into IL-1β is primarily mediated by caspase 1, but contrary to our expectations, the protein expression levels of caspase 1 were similar between the CD and eHC male placentas. In this case, we theorized that the NLRP3 inflammasome may be using a non-canonical pathway to process placental IL-1β. For example, the NLRP3 inflammasome can utilize caspase 8 as an IL-1β-converting protease56, and the protein expression levels of caspase 8 are increased in only the male eHC placentas; which provides a mechanism to explain the processing of pro-IL-1β into IL-1β. However, this mechanism remains to be assessed. Furthermore, it is important to note that although pyroptosis is not occurring in male eHC placentas, IL-1β can be secreted by other mechanisms such as microvesicle shedding55. It has been previously reported that pregnant women with excessively high cholesterol levels in pregnancy have more extracellular vesicles than controls57, but if IL-1β is part of the cargo of these vesicles requires further investigation.

We have also observed reduced protein expression levels of TXNIP and HMGB1 in only the female eHC placentas. Besides its role in activating the NLPR3 pathway, TXNIP can suppress the antioxidant activity of thioredoxins, acting as a redox regulator58. Thus, reduced protein expression levels of TXNIP in placentas of the female eHC offspring may be a compensatory mechanism to prevent alterations in ROS levels. On the other hand, it is important to mention that the alarmin signal HMGB1 is released in response to cellular stress, which can happen independently of the NLRP3 inflammasome. And, female eHC placentas had lipid accumulation, were likely exposed to more oxidized LDL, and presented a blunted UPR, all of which can cause cellular stress. HMGB1 is expressed in human term-placentas59and HMGB1 levels are increased in the circulation of women with PE60; hence, reduced HMGB1 protein expression levels in female eHC placentas may reflect the fact that these placentas are secreting HMGB1. Interestingly, HMGB1 can activate TLR461, and we have (1) observed signs of TLR4 activation in only the male eHC placentas and (2) previously reported that TLR4 activation impairs uterine artery endothelial function in eHC pregnancies11. Thus, it is possible that female eHC placental-derived HMGB1 modulates these responses.

In conclusion, our findings support the concept that eHC that develops specifically during pregnancy impacts the placentas of both the male and female offspring, but via sex-specific mechanisms. Noteworthy, as eHC activated the NLRP3 inflammasome in only the male placentas, the male offspring may be more susceptible to excessive increases in maternal cholesterol levels during pregnancy. It is well-accepted that the quality of the environment during fetal development is critical in determining health and disease in adult life62. However, even though it has been estimated that ≈25% of all pregnancies develop eHC2,3,4,5,6, the maternal lipid profile is not routinely assessed during pregnancy, and therefore, the true scope of the problem (i.e. the exact number of pregnancies that develop eHC and impacted offspring) is not known. Nevertheless, this study advances our understanding of the mechanisms underlying impaired placental efficiency in eHC pregnancies, and advocates for routine maternal lipid profiles during pregnancy.