ACE2 downregulation contributes to adverse cardiovascular and reproductive outcomes in female mice on high fat diet

Abstract

Obesity is a growing concern among women of reproductive age, as it can significantly impact both maternal and fetal health. A potential mechanism linking obesity to cardiometabolic dysfunction is the imbalance between the classical (pressor) and the compensatory (protective) arms of the renin angiotensin system (RAS). The objective of this study was to investigate how high fat diet (HFD) impacts cardiovascular and reproductive health in female mice and its associated mechanisms with an emphasis on the protective angiotensin converting enzyme 2 (ACE2)/Angiotensin (Ang)-(1–7) axis. Female C57BL/6J mice at 8 weeks of age were assigned to either a HFD (60% kcal fat) or a regular diet (RD, 22 kcal% fat). Following 10–12 weeks’ dietary exposure, HFD-fed females exhibited significantly greater weight gain and glucose intolerance compared to RD-fed controls. Telemetry recordings revealed that HFD exposure led to significantly elevated blood pressure (BP), increased heart rate, reduced spontaneous baroreflex sensitivity, and dampened parasympathetic tone. The above cardiometabolic dysfunction persisted or further exacerbated during pregnancy. Although HFD exposure did not alter fertility (ability to conceive), delivery or litter size, offspring from HFD mothers suffered low birth weight and markedly increased postnatal mortality. Examination of the RAS revealed reduced ACE2 mRNA and protein expression in both the heart and the kidney of HFD-fed females. Importantly, Ang-(1–7) infusion in HFD-fed females starting two weeks before mating improved glucose tolerance, reduced BP, and improved offspring survival. In conclusion, downregulation of ACE2/Ang-(1–7) axis contributes to HFD-induced cardiometabolic and reproductive dysfunction in female mice.

NEW & NOTEWORTHY

This study identifies downregulation of ACE2 in the setting of high fat diet-induced obesity as a contributing factor to maternal cardiometabolic dysfunction and adverse pregnancy outcomes. Restoring ACE2/Ang-(1–7) function may have therapeutic value in improving cardiovascular and reproductive functions in females with obesity.

INTRODUCTION

Obesity is a global health crisis, affecting approximately one in eight adults worldwide and serving as a major risk factor for cardiovascular diseases (1, 2). The most important cause of obesity is an energy imbalance driven by overconsumption of a hypercaloric diet rich in fat, coupled with reduced physical activity (3). Obesity contributes to hypertension, type 2 diabetes, and heart failure in part through chronic activation of the renin-angiotensin system (RAS)(4–6). The burden of obesity is particularly concerning in pregnancy; in a recent analysis of 40 million pregnancies worldwide, an estimated 18.7% of pregnant women in North America were classified as obese and 47% as overweight/obese (7). Maternal obesity increases the risk of gestational hypertension, preeclampsia, gestational diabetes, and preterm delivery, with consequences extending to both mother and offspring (8, 9). Beyond these clinical associations, obesity disrupts the maternal cardiovascular adaptations required for a healthy pregnancy (10). Normal gestation is characterized by reduced systemic vascular resistance, increased cardiac output, and tightly regulated blood pressure (BP) (11, 12). These adaptations rely heavily on balanced autonomic regulation and RAS signaling (13–15). In obesity, however, sympathetic overactivation (16), impaired baroreflex sensitivity (17), and endothelial dysfunction (18, 19) contribute to an exaggerated hemodynamic burden during pregnancy. These alterations increase maternal cardiovascular risk and impair placental perfusion, ultimately compromising fetal growth and neonatal survival (20, 21).

A potential mechanism linking obesity to cardiometabolic and reproductive dysfunction is the imbalance between the classical (pressor) and compensatory (protective) arms of the RAS (22–24). Angiotensin II (Ang II), acting through angiotensin type I receptors (AT₁R), promotes vasoconstriction, oxidative stress, and inflammation, whereas angiotensin-converting enzyme 2 (ACE2) counteracts these effects by metabolizing Ang II into the vasodilatory and anti-inflammatory peptide Ang-(1–7). The role of ACE2 in obesity-associated maternal cardiometabolic dysregulation and its impact on pregnancy is incompletely understood. Obesity has been shown to regulate the expression of ACE2 in a sex- and tissue-specific manner. Lung ACE2 expression was upregulated in diet-induced obese male but not female mice (25), and lung epithelial cells isolated from obese human subjects (sex unknown) have significantly higher ACE2 expression compared to those isolated from non-obese subjects (26), which may contribute to the increased risk for severe complications in obese patients infected by SARS-CoV-2. In a diet-induced obesity mouse model, increased ACE2 expression in adipose tissue of obese female mice was associated with protection against obesity-induced hypertension (27). In human studies, however, obesity has been associated with decreased ACE2 expression in adipose tissue (28) as well as in the left ventricle (LV) of patients with end-stage heart failure (29).

Using a high-fat diet (HFD)-induced obesity model, the objective of the present study was to determine how HFD affects cardiovascular, metabolic, and reproductive functions in female mice and the role of the RAS, with an emphasis on the dysregulation of the ACE2/Ang-(1–7) axis.

MATERIALS AND METHODS

Animals

All animal procedures conformed to the National institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Louisiana State University Health Sciences Center Institutional Animal Care and Use Committee (Protocols #2553 and #8343).

Male and female C57BL/6J mice (6–7 weeks old) were obtained from Jackson Laboratories and housed in a temperature (~27°C) and humidity (30–60%) controlled facility on a 12-h light/dark cycle. Mice were provided with standard chow or regular diet (RD, 22 kcal% fat, Teklad Extruded Rodent Diet 2019S; Inotive, West Lafayette, IN) and water ad libitum and allowed to acclimate for at least one week prior to experimental procedures. At 8 weeks of age, female mice were randomly assigned to either RD or HFD groups. The RD group continued receiving the standard mouse chow ad libitum, whereas the HFD group was provided with a diet containing 60 kcal% fat (D12492; Research Diets, Inc., New Brunswick, NJ) ad libitum. Weight gain was monitored weekly. Two experimental cohorts (cohort 1 and 2) were used for cardiometabolic assessment following identical protocols (Figure 1A), and data were combined for final analysis (7–10 mice per group per cohort).

Figure 1. High-fat diet (HFD) induces weight gain, metabolic dysfunction and increased inflammation in female mice.

A: Experimental timeline. B: HFD females gained significantly more weight than regular diet (RD) controls beginning at week 1 and continuing through the timeline of the study. C: Insulin tolerance test. Main effects: Time, P < 0.0001; Group, P < 0.0001. Interaction between Time and Group, ns. D: Glucose tolerance test demonstrating elevated glucose levels and delayed clearance in HFD mice. Main effects: Time, P < 0.0001; Group, P < 0.0001. Interaction between Time and Group, P < 0.0001. AUC, area under the curve. E: Non-fasting glucose concentrations were significantly elevated in HFD females compared to RD controls. F: Plasma cytokine levels. A-E: n=10–17 per group, F: n=7–14 per group. Statistical analyses were performed using unpaired two-tailed Student’s t-test (B, E and F) and repeated-measures ANOVA followed by Bonferroni post-hoc analysis (C and D). *P < 0.05, **P <0.01, ***P < 0.001, ****P < 0.0001.

After 12 weeks on respective diet, RD and HFD females were mated with previously RD-fed C57BL/6J males at the onset of the dark cycle (6 PM). Copulatory plugs were checked each morning at approximately 8 AM. The presence of a vaginal plug was designated as embryonic day 0.5 (E0.5). Plug checks were performed by gently lifting the female by the base of the tail and visually inspecting the vaginal opening using a sterile 200-μL pipette tip or cotton swab. Because vaginal plugs do not always indicate successful fertilization, females were group-housed with males until pregnancy was confirmed by body weight gain at E7.5-E8.5. An increase in body weight that exceeds 1.75 g during the first week of gestation was used to confirm pregnancy (30). Confirmed pregnant females were then single housed.

Glucose and Insulin Tolerance Tests (GTT and ITT)

GTT and ITT were performed in mice following a 5-hour fast, with free access to water. All blood glucose measurements were obtained via tail vein puncture using a handheld glucometer. Baseline (time 0) blood glucose levels were recorded immediately prior to intraperitoneal (IP) injection. For GTT, mice received an IP injection of D-glucose at a dosage of 2 g/kg, prepared as a 25% (w/v) glucose solution in sterile saline. For ITT, mice received an IP injection of insulin at a dosage of 0.5 IU/kg. Following injection, blood glucose levels were measured at 15, 30, 60, 90 and 120 minutes. Mice were closely monitored throughout both procedures for signs of distress, including reduced movement, piloerection, hunched posture, or lethargy. During ITT, if blood glucose levels fell below 20 mg/dL or mice exhibited signs of distress, a glucose solution was immediately administered, and animals were excluded from further testing to ensure welfare.

Echocardiography

Cardiac structure and function were assessed at 10 weeks of dietary exposure using the Vevo F2 imaging system equipped with 30-MHz transducer (VisualSonics, Bothell, WA). Mice were anesthetized with 2–3% isoflurane for induction and maintained at 1–2% on a heated platform. M-mode images were acquired in the parasternal long-axis view to assess systolic function and LV wall thickness. Diastolic function was evaluated from the apical four-chamber view using pulsed-wave Doppler to measure transmitral flow (E and A waves) and tissue Doppler imaging to assess mitral annular velocities (e’). Heart rate (HR) was maintained between 450–520 bpm for systolic measurements and 320–400 bpm for diastolic measurements. Images were analyzed using Vevo LAB 5.9.0 (VisualSonics). For each animal, measurements were averaged from at least three consecutive cardiac cycles.

Radiotelemetry Recording

Subsets of mice were implanted with radiotelemetry transmitters (PA-C10 or HDX10; Data Sciences International, St. Paul, MN) at week 10 of dietary exposure. Following recovery (~10 days), radiotelemetry recordings (BP, HR and activity) were performed continuously for 24 hours once per week, as previously described (31, 32). Autonomic function testing was performed during the inactive phase (starting at 9 AM). For these tests, hemodynamic parameters were recorded for 30 min before and one hour following IP injection of atropine (muscarinic receptor blocker, 1 mg/kg) or chlorisondamine (ganglionic blocker, 2.5 mg/kg) with a minimum washout period of 4 hours between drug administrations. Changes in HR (ΔHR) and mean arterial BP (ΔMABP) were quantified and plotted to assess parasympathetic and sympathetic contributions to cardiovascular regulation (32). Telemetry data were analyzed using the Ponemah software (Data Sciences International).

Baroreceptor Reflex Gain Analysis

Spontaneous baroreflex sensitivity (SBRS) was quantified from 24-hour radiotelemetry recordings using HemoLab software (Harald Stauss Scientific, Iowa City, Iowa). Raw Ponemah files were imported directly into HemoLab, and the two-channel transformation module was used to generate beat-to-beat pulse interval and arterial pressure signals. SBRS was calculated using the Baroreflex Bertinieri sequence analysis tool, which identifies sequences of three or more consecutive beats in which arterial pressure and pulse interval change in the same direction. For each valid sequence, the slope of the regression between pressure and the subsequent pulse interval was computed, and sequences with correlation coefficients below the default threshold were excluded. Mean SBRS values were obtained by averaging all accepted sequence slopes across the full 24-hour recording. The reported SBRS metric corresponds to the “BRR_gain_all” output generated by HemoLab.

Ang-(1–7) Infusion

In a separate cohort (cohort 3), Ang-(1–7) (Bachem Americas, Inc., Torrance, CA) in sterile saline was delivered using 6-week Alzet osmotic minipumps (model 1006) at a constant rate of 600 ng/kg/min, starting at week 10 on HFD. Following the first two weeks of infusion, BP and metabolic assessments were performed. Females were then mated with RD-fed C57BL/6J males, and Ang-(1–7) infusion continued throughout pregnancy and early postpartum for the remaining 4 weeks.

Western Blotting

Western blotting was performed as previously described (32). LV and kidney tissues were homogenized in RIPA buffer containing protease and phosphatase inhibitors (Cell Signaling Technology, Danvers, MA). Protein concentrations were determined using the bicinchoninic acid assay (Pierce Biotechnology, Waltham, MA). Equal amounts of protein were separated on 4–15% Mini-PROTEAN TGX gels (Bio-Rad, Hercules, CA) and transferred to PVDF membranes. Membranes were blocked in Tris-buffered saline with 0.1% Tween-20 and 3% BSA. Primary antibodies included ACE2 (1:1,000 dilution; ab108252, Abcam, Waltham, MA) and GAPDH (1:5,000 dilution; 5174, Cell Signaling Technology). Antibody specificity has been previously validated (33, 34). HRP-conjugated secondary antibodies (Cell Signaling Technology) were used at 1:10,000–1:20,000. Bands were visualized using Amersham ECL Prime Western Blotting Detection Reagent and the Amersham Imager 680 (GE Healthcare Bio-Sciences, Marlborough, MA). Protein sizes were estimated using the EX-Run pre-stained ladder (Fisher Scientific, Hampton, NH). Densitometry was performed using NIH ImageJ.

Quantitative Reverse Transcription PCR (RT-qPCR)

Total RNA was extracted from LV and kidney tissues collected at endpoint using the RNeasy Plus Mini Kit (Qiagen, Ann Arbor, MI). One microgram of RNA was reverse-transcribed using the iScript cDNA Synthesis Kit (Bio-Rad). Real-time PCR was performed using the CFX Opus 96 Real-Time PCR Detection Systems (Bio-Rad). Relative gene expression was determined using the 2^−ΔΔCT method with data normalized to the mRNA level of the housekeeping gene Tbp (32). Pre-validated primers were purchased from Integrated DNA Technologies (IDT, Coralville, IA).

Plasma Cytokine Measurement

At endpoint, blood was collected in BD microtainer K₂EDTA blood collection tubes and subsequently centrifuged at 1,500 g for 10 min at 4°C to obtain plasma. Plasma cytokine levels were determined by Milliplex mouse cytokine assay (MilliporeSigma, Burlington, MA).

Ang-II and Ang-(1–7) Peptide ELISA Assays

For Ang II and Ang-(1–7) peptide ELISA assays, blood was collected using BD microtainer K₂EDTA blood collection tubes, and protease inhibitors (Cell Signaling Technologies) were added to the plasma immediately following the centrifugation step to prevent peptide degradation.

Plasma and kidney levels of Ang-II were quantified in RD and HFD female mice using a commercially available ELISA kit (#ADI-900–204, Enzo Life Sciences, Farmingdale, NY) without C₁₈ column extraction as previously described (35, 36). Plasma (50 μL) and kidney tissue extract in 50 μL of cell lysis buffer were used in the assay with final concentration normalized to total protein. To extract proteins from kidney, 50 mg kidney tissue was rinsed with 1X PBS, homogenized in 500 μL of 1 X cell lysis buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM Na₂EDTA, 1 mM EGTA, 1% Triton, 2.5 mM sodium pyrophosphate, 1 mM beta-glycerophosphate, 1 mM Na₃VO₄, 1 μg/ml leupeptin; 9803, Cell Signaling Technology), and centrifuged for 20 minutes at 16,000 × g at 4°C. The supernatant was collected and assayed immediately.

Plasma and kidney levels of Ang-(1–7) were quantified in RD and HFD female mice using a commercially available Ang-(1–7) ELISA kit (#CSB-E13763m, Cusabio, Houston, TX) according to the manufacturer's protocols. Plasma (100 μL) and kidney tissue extract in 100 μL of PBS with protease inhibitors (Cell Signaling Technology) were used in the assay. To extract proteins from kidney, 50 mg kidney tissue was rinsed with 1X PBS, homogenized in 0.5 mL of 1X PBS with protease inhibitors, and stored overnight at −20°C. After two freeze-thaw cycles were performed to break the cell membranes, the homogenates were centrifuged for 5 minutes at 5,000 × g at 4°C. The supernatant was removed and assayed immediately.

Based on the information provided by the manufacturer, the Ang-(1–7) ELISA exhibits high specificity without significant cross-reactivity with other Ang analogues. The Ang-II ELISA assay exhibits 0.053% cross reactivity with Ang-(1–7), minimal cross reactivity with Ang-(1–12), Ang I or Ang-(1–9), and 100% cross reactivities with Ang II metabolites Ang A, Ang III and Ang IV. The specificity of the Ang II and Ang-(1–7) ELISA kits was not confirmed or tested in our studies.

SFlt-1 Quantification

Maternal blood was collected via cheek vein puncture during gestational week 2 (cohort 2) and allowed to clot at room temperature for 1 hour. Samples were then centrifuged at 1,000 × g for 20 minutes, and the resulting supernatant (serum) was transferred to fresh tubes and stored at −80°C until analysis. Serum soluble fms-like tyrosine kinase-1 (sFlt-1) was measured using a commercial ELISA kit per manufacturer’s instructions (EK248749, AFG Bioscience, Northbrook, IL).

Statistical Analysis

Data are expressed as mean ± standard error of the mean (SEM). Statistical analyses were performed using unpaired two-tailed Student’s t-tests, one-way ANOVA or two-way repeated-measures ANOVA, followed by Bonferroni post-hoc test, as appropriate for each experimental design. Survival curves were analyzed using Kaplan–Meier survival analysis with the log-rank test. All analyses were conducted using GraphPad Prism 11 (GraphPad Software, San Diego, CA). P-value < 0.05 was considered statistically significant.

RESULTS

HFD exposure leads to increased body weight, metabolic dysfunction and inflammation in female mice

Female mice exposed to HFD exhibited a significant increase in body weight following 1 week of diet exposure compared to RD-fed controls, which persisted throughout the study (Figure 1B). This excess body weight gain was accompanied by metabolic impairment, as evidenced by glucose intolerance revealed by the GTT assay (Figure 1D). HFD females responded similarly to insulin in the ITT assay compared to RD females, although baseline glucose levels were elevated (Figure 1C). Non-fasting blood glucose levels measured during pregnancy were also elevated in HFD females (166.5±6.3 mg/dL in HFD group vs. 141.1±2.6 mg/dL in RD group, P < 0.001) (Figure 1E), further supporting the presence of metabolic dysfunction. Additionally, in plasma collected at endpoint (Figure 1F), HFD females exhibited significant increases in plasma levels of IL-6 (23.5±5.7 pg/mL in HFD group vs. 4.9±1.9 pg/mL in RD group, P < 0.05), keratinocyte-derived chemokine (KC/CXCL1; 281.2±55.1 pg/mL in HFD group vs. 115.9±22.1 pg/mL in RD group, P < 0.05), macrophage colony-stimulating factor (M-CSF; 184.8±48.3 pg/mL in HFD group vs. 8.3±3.2 pg/mL in RD group, P < 0.01), granulocyte colony-stimulating factor (G-CSF; 354.5±37.7 pg/mL in HFD group vs. 205.8±35.7 pg/mL in RD group, P < 0.05), and tumor necrosis factor-α (TNF-α; 14.6±5.3 pg/mL in HFD group vs. 0.8±0.8 pg/mL in RD group, P < 0.05). Of note, the increase in TNF-α levels occurred independently of any change in ADAM17 (aka TNF-α convertase) expression (Supplemental Figure S1). These findings indicate that HFD exposure promotes a pro-inflammatory circulating cytokine profile in female mice, consistent with diet-induced metabolic stress and low-grade chronic inflammation (37).

HFD exposure leads to elevated BP and diastolic dysfunction in female mice

Radiotelemetry recordings were employed to assess BP in conscious freely moving mice using continuous 24-h weekly recordings (Figure 2A). Compared to RD females, HFD females exhibited significantly higher MABP (103.7±1.9 mmHg in HFD group vs. 91.7±1.8 mmHg in RD group, P < 0.01), with increases in both systolic (116.9±1.9 mmHg in HFD group vs. 103.8±2.5 mmHg in RD group, P < 0.001) and diastolic BP (90.0±2.0 mmHg in HFD group vs. 79.7±1.7 mmHg in RD group, P < 0.001) across the 24-h cycle. When data were separated into active and resting phases, BP was significantly elevated in HFD mice in both phases (Supplemental Figure S2A). During gestation, MABP remained higher in HFD females compared with RD controls, although the difference did not reach statistical significance (Supplemental Figure S2B). HR was also significantly increased throughout the 24-h recording period, indicating relative tachycardia in HFD females (551.9±10.6 bpm in HFD group vs 469.9±12.7 bpm in RD group, P < 0.001; Figure 2B). Activity was also measured by radiotelemetry using the Ponemah software, which is based on the animal movement in the cage versus a reference point. HFD females exhibited reduced activity compared to the RD controls in both active (12.30±1.06 counts/min in HFD group vs 20.14±3.21 counts/min in RD group, P < 0.05) and resting (4.47±0.43 counts/min in HFD group vs 6.18±0.50 counts/min in RD group, P < 0.05) phases (Figure 2C), similar to the sedentary lifestyle often observed in obese human subjects (3).

Figure 2. High-fat diet (HFD) exposure leads to elevated blood pressure and heart rate (HR) and reduced activity in female mice.

A: HFD exposure leads to elevated mean arterial blood pressure (MABP), systolic blood pressure (SBP) and diastolic blood pressure (DBP) compared to regular diet (RD) controls throughout the 24-h period. B: HFD exposure leads to elevated HR in both the active and resting phases. C: HFD exposure leads to reduced activity in both the active and resting phases. n=5–6 per group. Unpaired two-tailed Student’s t-tests, *P < 0.05, **P < 0.01.

Cardiac structure and function were evaluated using transthoracic echocardiography. After 10 weeks of dietary exposure, long-axis M-mode imaging revealed increased LV posterior wall thickness during systole in HFD mice (1.17±0.01 mm in HFD group vs. 1.09±0.03 mm in RD group, P < 0.05; Supplemental Figure S3A), suggesting concentric remodeling in response to chronic pressure loading (38). LV anterior wall thickness at both systole and diastole (Supplemental Figure S3 C and D) and LV posterior wall thickness at diastole (Supplemental Figure S3B) were not significantly different between the two groups. Additionally, HFD did not significantly alter ejection fraction or fractional shortening (Supplemental Figure S3 E and F). Diastolic function was assessed using the E/A ratio, representing early (E) to late (A) diastolic transmittal flow velocity, and the E/e′ ratio, which reflects the relationship between early transmittal flow and early diastolic mitral annular tissue velocity. HFD females displayed a significantly elevated E/A ratio compared with RD controls (1.90±0.09 in HFD group vs. 1.52±0.04 in RD group, P < 0.05; Supplemental Figure S3G), which may reflect impaired ventricular relaxation and increased chamber stiffness associated with elevated left atrial pressure (39). The E/e′ ratio was also significantly increased in HFD females (23.35±0.85 in HFD group vs. 20.43±0.61 in RD group, P < 0.05; Supplemental Figure S3H), supporting the presence of higher LV filling pressure and early diastolic dysfunction (40, 41).

HFD exposure causes dysregulation of parasympathetic tone and baroreflex sensitivity

Atropine administration produced the expected tachycardic response in both RD and HFD females; however, the magnitude of this increase was significantly greater in RD females compared with HFD females (274.1±36.2 bpm in RD group vs. 143.4±15.8 bpm in HFD group, P < 0.01; Figure 3A), indicating reduced parasympathetic tone to the heart in the HFD group. Despite a trend, chlorisondamine did not reveal significant differences in vascular tone between the RD and the HFD groups, possibly due to high within-group variability (Figure 3B). HFD females demonstrated a significant reduction in SBRS compared to RD controls (2.72±0.15 msec/mmHg in HFD group vs. 3.61±0.15 msec/mmHg in RD group, P < 0.01; Figure 3C), consistent with impaired autonomic regulation. Together, these findings indicate that HFD disrupts both parasympathetic tone and baroreflex function, contributing to cardiovascular dysregulation.

Figure 3. High-fat diet (HFD) exposure impairs autonomic cardiovascular regulation and baroreflex sensitivity.

A: Changes in heart rate (ΔHR) following intraperitoneal atropine injection were significantly reduced in HFD females, indicating dampened parasympathetic tone. B: Changes in mean arterial pressure (ΔMABP) following chlorisondamine injection did not significantly differ between regular diet (RD) and HFD groups. C: Spontaneous baroreflex sensitivity (SBRS) was significantly lower in HFD females compared to RD females. n=5–6 per group. Data were analyzed by unpaired two-tailed Student’s t-test; **P < 0.01.

HFD leads to adverse pregnancy outcome

Maternal serum collected during mid-gestation revealed a significant elevation in sFlt-1 levels in HFD-fed females, with approximately a three-fold increase compared with RD controls (Figure 4A). This rise in circulating sFlt-1 suggests an anti-angiogenic shift and impaired placental vascularization in obese pregnancies. Although HFD didn’t cause a significant change in litter size between the groups (Figure 4B), offspring born to HFD mothers exhibited significantly lower birth weight (0.94±0.07 g in HFD-offspring group vs. 1.23±0.07 g in RD-offspring group, P < 0.01; Figure 4C) and dramatically reduced postnatal survival (Figure 4D). Approximately 71% of the offspring born to HFD mothers died within the first four postnatal days compared with 15% of the offspring born to RD mothers (P < 0.001).

Figure 4. High fat diet (HFD) elevates maternal circulating sFlt-1 and impairs offspring growth and survival.

A: Serum sFlt-1 levels were significantly increased in HFD females compared with regular-diet (RD) controls. n=10 per group. B: Litter size did not significantly differ between the RD and the HFD groups. RD, 18 litters; HFD, 20 litters. C: Offspring from HFD mothers had significantly lower birth weight. Average birth weight of the offspring from each litter is shown (12 litters from RD group and 10 litters from HFD group). D: Offspring survival was markedly reduced in the HFD group compared to the RD group. RD, 113 offspring from 18 litters; HFD, 101 offspring from 20 litters. Statistical analyses were performed using unpaired two-tailed Student’s t-tests for panels A–C and Kaplan–Meier survival analysis with log-rank test for panel D. **P < 0.01, ****P < 0.0001.

HFD reduces ACE2 expression and Ang-(1–7) levels

To investigate how HFD alters the RAS, LV and kidney tissues collected at endpoint were analyzed for mRNA expression of key components of the RAS, including ACE, ACE2, AT₁R and Mas receptor (MasR), the receptor for Ang-(1–7). As shown in Figure 5 A and B, HFD exposure led to decreased ACE2 transcripts in both the LV and the kidney. Reduced ACE and MasR transcripts were also observed in the kidney of HFD females compared to the RD controls (Figure 5B). Western blotting confirmed the reduction of ACE2 expression at protein levels in both the LV (Figure 5C) and the Kidney (Figure 5D) of HFD females. Furthermore, correlation analysis (Supplemental Figure S4) revealed significant inverse correlation between LV and kidney ACE2 mRNA expression and cardiometabolic indices, including MABP, HR and GTT values.

Figure 5. High fat diet (HFD) suppresses ACE2 expression in left ventricle (LV) and kidney tissues in female mice.

A: LV gene expression in regular diet (RD) and HFD females. B: Kidney gene expression in RD and HFD females. C: LV ACE2 protein expression was significantly reduced in HFD females compared with RD controls. D: Kidney ACE2 protein expression was significantly reduced in HFD females compared with RD controls. n=5–7 per group. Data were analyzed by unpaired two-tailed Student’s t-tests. *P < 0.05, **P < 0.01, ***P < 0.001.

Ang-II and Ang-(1–7) levels in plasma and kidney tissues were measured by ELISA. As shown in Figure 6A, a trend of decrease in plasma Ang-(1–7) levels was observed in HFD females compared to the RD controls (P = 0.0529), while a significant decrease in Ang-(1–7) levels was observed in the kidney of HFD females compared to the controls (P < 0.05, Figure 6B). Ang-II levels were significantly increased in the kidney of HFD females (Figure 6D); however, no significant differences were observed in plasma Ang-II levels between the two groups (Figure 6C). Because of the significant changes in Ang-II and Ang-(1–7) levels in the kidney, we examined kidney fibrosis and inflammatory cell infiltration on kidney tissue sections using Masson’s trichrome and anti-CD45 immunostaining, respectively (Supplemental Figure S5A). Our results show that HFD did not lead to increased kidney fibrosis, which is further confirmed by unchanged Col1a1 mRNA expression (Supplemental Figure S5B). However, CD45 positive inflammatory cells were increased in kidney sections from HFD females compared to RD controls (Supplemental Figure 5A), consistent with increased plasma cytokines in these mice (Figure 1F).

Figure 6. High fat diet (HFD) alters Ang-II and Ang-(1–7) levels in female mice.

A: Plasma Ang-(1–7) levels. B: Kidney Ang-(1–7) levels. C: Plasma Ang-II levels. D: Kidney Ang-II levels. n=8 per group. Data were analyzed by unpaired two-tailed Student’s t-tests. *P < 0.05, **P < 0.01.

Ang-(1–7) supplementation improves cardiometabolic function and offspring survival in HFD females

To further examine the protective role of the ACE2/Ang-(1–7) axis in maternal cardiometabolic and reproductive function, Ang-(1–7) supplementation was conducted starting at week 10 of HFD feeding via osmotic minipumps. No significant differences in body weight was oberved between HFD females with or without Ang-(1–7) infusion (Supplemental Figure S6). Following two weeks of Ang-(1–7) administration (weeks 10–12 of HFD feeding), HFD females exhibited marked improvements in hemodynamic function, with reduction of MABP to values observed in RD fed controls (Figure 7A), which was maintained during pregnancy with continued Ang-(1–7) infusion (Figure 7B). In parallel with these hemodynamic improvements, Ang-(1–7) treatment partially restored metabolic function in HFD females. Glucose tolerance (Figure 7D) was improved in the HFD group treated with Ang-(1–7) compared with untreated HFD group, although glucose levels did not fully normalize to RD levels. Offspring outcomes were also improved by Ang-(1–7) supplementation, with significantly improved offspring survival (Figure 7G).

Figure 7. Ang-(1–7) supplementation improves cardiometabolic function and offspring survival in high fat diet (HFD) exposed females.

A: Ang-(1–7) supplementation reduces mean arterial blood pressure (MABP) in HFD females following two weeks’ treatment to levels in regular diet (RD) controls. B: Reduction in MABP was maintained during pregnancy with continued Ang-(1–7) infusion in HFD females. C: Insulin tolerance test (ITT). Main effects: Time, P < 0.0001; Group, P < 0.001. Interaction between Time and Group, P < 0.05. D: Glucose tolerance test (GTT). Main effects: Time, P < 0.0001; Group, P < 0.0001. Interaction between Time and Group, P < 0.0001. AUC, area-under-the-curve. E: Litter size. F: Offspring weight at birth. G: Offspring survival was significantly reduced in the HFD group and improved with Ang-(1–7) supplementation. One-way ANOVA (A, B, E and F) or two-way repeated-measures ANOVA (C and D) followed by Bonferroni post-hoc test were used where appropriate; Kaplan–Meier survival analysis with log-rank test was used for panel G. A-D, n = 4–7 per group. Data shown in E and G included data from all three cohorts, 20 litters from RD group with a total of 128 offspring, 28 litters from HFD group with a total of 155 offspring, and 8 litters from HFD+Ang-(1–7) group with a total of 53 offspring. For Weight at Birth (F), data from cohorts 2 and 3 are included (we did not measure birth weight in cohort 1); average birth weight of the offspring from each litter is shown, which includes 13 litters from RD, 16 litters from HFD, and 7 litters from HFD+Ang-(1–7) group. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, HFD vs. RD; $P < 0.05, $ $P < 0.01, $ $ $P < 0.001, HFD+Ang-(1–7) vs RD; #P < 0.05, ##P < 0.01, ####P < 0.0001, HFD vs. HFD+Ang-(1–7).

Examination of plasma cytokine levels revealed significant reduction of M-CSF and TNF-α levels in Ang-(1–7) treated females on HFD (Supplemental Figure S5C), coinciding with reduced inflammatory cell infiltration in the kidney ((Supplemental Figure S5A).

DISCUSSION

Obesity is a major modifiable risk factor for cardiovascular and metabolic diseases, and its prevalence among women of reproductive age continues to rise (42). The impact of obesity on maternal cardiometabolic health and reproductive function is the focus of the present study. Using a murine HFD-induced obesity model, our study revealed several important findings: 1. Chronic HFD consumption leads to cardiometabolic dysfunction in female mice, including glucose intolerance, reduced diastolic function, elevated BP and HR, and autonomic dysregulation; 2. The above cardiometabolic dysfunction persists or exacerbates during pregnancy, which is associated with reduced birth weight and increased mortality of the offspring; and 3. HFD consumption is associated with downregulation of ACE2, and importantly, supplementation with Ang-(1–7) mitigates HFD-induced cardiometabolic dysfunction and improves pregnancy outcomes. Our study identifies ACE2 downregulation as a critical contributing factor to the adverse cardiovascular and reproductive outcomes induced by HFD.

In this study, chronic HFD-induced metabolic dysfunction is associated with marked increases in circulating IL-6, KC/CXCL1, M-CSF, G-CSF and TNF-α levels, indicating that maternal HFD consumption drives a systemic pro-inflammatory state characteristic of chronic metabolic stress. This cytokine profile is consistent with obesity associated immune activation and endothelial dysfunction that have been implicated in adverse pregnancy outcomes (43–45). During pregnancy, the HFD females exhibited a significant elevation of circulating sFlt-1, a potent anti-angiogenic factor strongly associated with placental dysfunction and preeclampsia (46, 47). This placental dysfunction is likely responsible for the impaired fetal growth (reduced birth weight) and the early postnatal mortality.

The hemodynamic changes observed in our HFD-fed females are consistent with the literature that obesity promotes sympathetic overactivation and blunts baroreflex function (17). In addition to a trend toward sympathetic overactivation, our study also showed that HFD feeding is associated with reduced parasympathetic tone, which often precedes sympathetic overactivation (48), with increased HR and reduced ΔHR in response to atropine. Echocardiographic assessment revealed early diastolic dysfunction in HFD females, including increased E/A and E/e′ ratios, along with increased LV posterior wall thickness. Although systolic function remained preserved, these findings are consistent with early concentric remodeling and elevated filling pressures, a condition strongly associated with obesity (39–41).

Sex differences have been reported in obesity associated hypertension by Gupte et al (27), where female mice were found to be protected from HFD-induced hypertension. The reason for the discrepancy between our study and this published study is currently unclear but could be due to different animal housing conditions. Most mouse studies are conducted at room temperature (20–22°C), which is far below their thermoneutral zone (29–33°C). Chronic housing of mice in thermally stressed conditions has profound effects on many physiological parameters including BP and HR and their intrinsic capacity to adapt to environmental challenges (49). Although our animal housing temperature (~27°C) is also below their thermoneutral point, the total energy expenditure at this temperature (~27°C) in mice equals about 1.7 times of their basal metabolic rate, which is the same for humans at ambient temperature of ~20–22°C (50). Another difference between these two studies is that our study utilized a 22 kcal% fat diet as our control diet or RD, whereas the Gupte study utilized a 10% kcal fat diet (Research Diets D12450B) as control. It is important to note, however, that HFD feeding resulted in downregulation of ACE2 in the kidney in both studies.

A central mechanistic finding of this study is that chronic HFD exposure suppresses the expression of ACE2 in organs important for cardiovascular function, including the LV and the kidney. In the kidney, reduction of MasR expression by HFD was also observed. In a previous study, HFD feeding starting at peripubertal stage was shown to reduce ACE2 protein expression in the LV but not in the kidney of female C57BL/6N mice (51), whereas HFD feeding was shown to increase ACE expression in the kidney of C57BL/6 (specific strain unknown) male mice (52). Many factors may contribute to the differences observed between our study and the previously published reports, including the mouse strain, age at the start of dietary exposure, sex, as well as animal housing conditions.

The downregulation of ACE2 in our study is associated with a significant decrease in Ang-(1–7) levels in the kidney and a trend of decrease in plasma Ang-(1–7). In addition, reduced ACE2 expression in the kidney is associated with elevated kidney Ang-II levels, which could contribute to the increased inflammatory cell infiltration in the kidney tissue.

The causal relationship between ACE2 downregulation and phenotypes observed in HFD female mice is supported by our Ang-(1–7) rescue experiment, which led to significant improvement of cardiometabolic and reproductive functions. Although the protective arm of the RAS has been implicated in cardiovascular and metabolic protection, our findings extend its relevance to neonatal survival, an outcome that, to our knowledge, has not been previously reported in maternal obesity models. It is important to note that plasma ACE2/Ang-(1–7) levels are elevated in healthy pregnancies but suppressed in preeclampsia and pre-term birth (53–55), and our study suggests that diet-induced obesity could be responsible for this suppression. Whether HFD exposure also alters ACE2/Ang-(1–7)/MasR in the placenta and its contribution to placental insufficiency requires further investigation.

Our study has several potential limitations. First, nutrient-matched control diet was not used for the HFD, and energy consumption between the RD and HFD groups were not measured. Second, chronic HFD may alter estrogen production, estrogen levels or estrogen receptor expression, which was not evaluated in this study. Third, we did not examine lactation or mammary tissues in HFD mothers nor their feeding or grooming behaviors, which might have contributed to reduced offspring survival. Fourth, in our osmotic minipump Ang-(1–7) infusion experiment, we did not include a saline control. Although osmotic minipump implantation did not significantly alter body weight, the potential effects of surgery and volume change were not accounted for in our study. And finally, we used ELISA to measure plasma and kidney Ang II and Ang-(1–7) peptide levels instead of HPLC-radioimmunoassay (RIA) or RIA methods. In contrast to previous studies which showed maintenance or even elevation of plasma Ang-(1–7) during chronic HFD administration ((27, 56)), our study found a decrease in Ang-(1–7) by HFD. In addition, plasma Ang II levels measured by ELISA in our study appear to be lower than the levels reported by the previous studies ((27, 56).

In summary, our study provides new insight into how diet-induced maternal obesity disrupts metabolic, cardiovascular and autonomic pathways before and during pregnancy and identifies downregulation of the ACE2/Ang-(1–7) axis as a central mechanistic contributor to these changes. By demonstrating that Ang-(1–7) supplementation not only improves maternal cardiometabolic function but also markedly enhances neonatal survival, this study reveals a previously unrecognized protective role for the ACE2/Ang-(1–7) pathway in obese pregnancies. This is particularly important due to the limited antihypertensive treatment options deemed safe during gestation, leaving a critical therapeutic gap for managing obesity-related cardiovascular risk. Given the rising prevalence of obesity during pregnancy, understanding how to restore protective RAS signaling has important implications for improving maternal cardiometabolic health, supporting placental function, and ultimately reducing the burden of adverse perinatal outcomes.

Supplementary Material

Supplemental Figure S1: https://doi.org/10.6084/m9.figshare.33201141

Supplemental Figure S2: https://doi.org/10.6084/m9.figshare.32086668

Supplemental Figure S3: https://doi.org/10.6084/m9.figshare.32086743

Supplemental Figure S4: https://doi.org/10.6084/m9.figshare.33179336

Supplemental Figure S5: https://doi.org/10.6084/m9.figshare.33179408

Supplemental Figure S6: https://doi.org/10.6084/m9.figshare.33201153

DATA AVAILABILITY

All data included in this study are available upon request. This study does not generate any unique reagents or original code. Requests for further information should be directed to Xinping Yue (xyue@lsuhsc.edu).