Medroxyprogesterone Opposes Estradiol-Induced Renal Damage in Midlife Ovariectomized Long Evans Rats

Abstract

Objective:

Our lab previously published that long-term administration of estradiol (E2) was detrimental to the kidneys of midlife ovariectomized Long Evans rats, contrasting clinical studies in showing that menopausal hormone therapy is associated with decreased albuminuria. However, it is unknown whether this renal benefit was due to estrogen and/or the combination with progestogen. Therefore, the objective of the current study was to determine the impact of medroxyprogesterone (MPA) on E2-mediated renal damage using a rodent model.

Methods:

Female Long Evans retired breeders underwent ovariectomy at 11 months of age and were treated for 40 days with subcutaneous E2, E2+MPA or vehicle at doses mimicking that of menopausal hormone therapy (N=5–7 per group). Systolic blood pressure was measured along with indices of renal damage and function to investigate the impact of MPA on E2-mediated renal outcomes. Renal estrogen receptor alpha and G protein-coupled estrogen receptor transcript copy numbers were measured in all treatment groups through droplet digital PCR.

Results:

Middle-aged female Long Evans rats displayed spontaneous hypertension with similar systolic blood pressures and heart weights between groups. Even though blood pressure was comparable, E2 reduced glomerular filtration rate and increased proteinuria indicating pressure-independent renal damage. Co-administration with MPA prevented E2-induced glomerular filtration rate impairment and proteinuria by promoting renal hypertrophy and preventing renal interstitial fibrosis. Both E2 and E2+MPA reduced renal ERα and increased renal GPER mRNA, but neither ERα nor ERß protein was different between groups.

Conclusion:

MPA was protective against E2-induced renal damage and dysfunction in middle-aged female Long Evans rats. Assessing the impact of hormone therapy on renal outcomes may be an important clinical factor when considering treatment options for postmenopausal women.

Introduction

Hormone therapy for healthy, symptomatic women is considered beneficial if they have no contraindications and are within 10 years of menopausal onset or younger than age 60 years.¹ Hormone therapy improves quality of life for postmenopausal women by alleviating many of the physical symptoms of menopause, such as mood changes, sleep disruption, vasomotor instability and vaginal atrophy.² However, the number of physicians prescribing hormone therapy for menopausal symptoms has drastically declined due to safety concerns raised from the Women’s Health Initiative study which showed that hormone therapy increased the risk for breast cancer, coronary heart disease, stroke, and pulmonary embolism.³ In response to the Women’s Health Initiative study, hormone therapies were shifted to a lower dose and shorter regimen, but the overall number of prescriptions never fully recovered as safety concerns still remain.³

Systematic review and meta-analysis identifying the role of hormone therapy on kidney function suggests a potential benefit for reduced albuminuria, but there is still controversy due to discrepancies in the reported outcomes and study designs.⁴ While hormone therapy may provide benefits against menopause-related symptoms, a subset of women experience increased microalbuminuria and a reduction in glomerular filtration rate (GFR) in response to estrogen therapy.^5,6 Similarly, we previously published that chronic estrogen therapy worsened renal health in midlife ovariectomized (OVX) Long Evans rats,⁷ and confirmed other reports showing increased proteinuria in castrated 14 month-old male Otsuka-Long-Evans-Tokushima Fatty rats.⁸ Therefore, understanding the impact of hormone therapy on renal outcomes will remain a critical issue as the prevalence of menopause increases among a growing aging population.

Hormone therapy for postmenopausal symptoms can include a combination of estrogen and progestogen, but women who have undergone a hysterectomy need only estrogen therapy. Medroxyprogesterone (MPA) is a progesterone derivative used to oppose uterine hyperplasia and reduce the risk of endometrial cancer. Our previously published results in the midlife female Long Evans rat mimics the response in postmenopausal women where estrogen therapy exhibits a negative effect on renal health.⁵ While other studies show that postmenopausal women using hormone therapy can experience a decrease in albuminuria,⁴ it is unknown whether this benefit was due to estrogen alone and/or due to the progestogen combination. Therefore, the goal of the current study was to determine if co-administration of MPA antagonizes estrogen-mediated renal damage. The use of a rodent “menopausal” model will address a fundamental question about the impact of menopausal hormone therapy on the renal system and provide a foundation for future clinical studies.

Methods

Animals.

Female retired breeder Long Evans rats aged approximately 11 months were received from Envigo (Indianapolis, IN) and housed in a temperature-controlled AAALAC-accredited vivarium under a 12h light-dark cycle. Food and water were provided ad libitum and animals were maintained on Teklad Global Soy Protein-Free Extruded Rodent Diet 2020X from Envigo. All animals were OVX at 11.5–12 months of age and implanted with subcutaneous dorsal capsules or minipumps to deliver treatments for 40 days. Animals were matched by baseline blood pressure and assigned to receive either vehicle (N=6), estradiol (E2, N=7) or E2 plus medroxyprogesterone (MPA, N=6). E2 and vehicle treatments were administered using 5 mm silastic capsule (0.058 in. inner diameter and 0.077 in. outer diameter; Dow Corning, Midland, MI) containing either 25% 17β-estradiol (Sigma-Aldrich, St. Louis, MO) diluted with cholesterol as previously described.⁷ Estradiol implants made with these exact dimensions maintain blood plasma estradiol levels in the physiological range and also match the estradiol levels achieved in postmenopausal women administered transdermal estradiol.^9–11 Control rats received cholesterol-only implants, which we previously showed increase neither uterine weight nor serum estradiol levels and maintain a predominance of leukocytes in vaginal smears.^11,12 MPA was administered using 2ML4 osmotic minipumps at 1 mg/kg/day (ALZET, Cupertino, CA) and dissolved in 1:1 polyethylene glycol and DMSO solution as described elsewhere.¹³ This dose is slightly higher than the daily oral MPA dose for endometrial protection in postmenopausal women of approximately 0.15 mg/kg/day.¹⁴ End of study wet weights of the uterus, kidney, and heart were obtained and normalized to body weight. Blood samples were collected via cardiac puncture and divided to collect plasma and serum samples. Serum and plasma samples were both centrifuged at 1000xg before removal of red blood cells. All experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved and monitored by the Tulane University Institutional Animal Care and Use Committee.

Blood Pressure.

Systolic blood pressure was recorded weekly using tail cuff plethysmography (CODA system, Kent Scientific, Torrington, CT). Animals were acclimated to the restraint for two days before baseline measurements. Animals were placed in restrainers and tails were warmed to 34°C. Blood pressures were taken over three days, taking the average of 15 readings per day excluding outliers (±2SD). Final weekly pressures were averaged over the three consecutive days.

Urinary markers.

Metabolic cages were used to collect the baseline and end of study urine. Samples were spun at 1000 × g for 10 min to remove particulates. Urinary protein concentration was measured via Bradford assay (Bio-Rad, Hercules, CA) using the standard bovine serum albumin and normalized against kilograms of body weight per day. Serum and urine creatinine were measured using the Jaffe Method and a 4:1 ratio of 1% Picric acid to 1N sodium hydroxide solution.¹⁵ Creatinine values were measure at 492 nm with known concentrations of creatinine as assay standards. GFR was calculated as previously described.⁷ All assays were performed in triplicate.

Histology.

Renal morphology was assessed in formalin-fixed and paraffin-embedded 4-μm kidney sections mounted onto slides (N=5–6 per group). Slides were heated at 55 °C for 30 min prior to being deparaffinized and hydrated. Trichrome Gomori One Step Aniline Blue Stain kit (Newcomer Supply, Middleton, WI) and NovaUltra Periodic Acid-Schiff Stain Kit (IHC World, Woodstock, MD) were used according to the manufacturer’s directions. All histology measurements were analyzed using ten randomly selected fields of view at 20X magnification. For glomerulosclerosis, an average number of scored glomeruli between 31–41 per slide. Glomerulosclerosis was assessed using a semiquantitative score from 0 to 4 (0 = no fibrosis, 1 = 0–25% of glomeruli fibrotic, 2 = 25–50% of glomeruli fibrotic, 3 = 50–75% of glomeruli fibrotic, and 4 = 75–100% of glomeruli fibrotic). Damage to the renal proximal tubule brush border was graded using a scale of 0 to 3 for Periodic Acid-Schiff staining (0 = none, 1 = poor, 2 = moderate, and 3 = excellent), and the average score was obtained from ten randomly selected fields of view. A scoring system using four grades of severity in kidney damage (0 = absence, 1 = mild, 2 = moderate, and 3 = severe) was used to evaluate whole kidney tubulointerstitial fibrosis, protein cast formation, vasa recta hyalinosis, peritubular congestion, and proximal tubule dilation. All histological analyses were assessed and quantified by one independent investigator who was blinded to the treatment groups. Intraobserver variability ranged from 10–18%.

Droplet Digital PCR.

Estrogen receptor alpha (ERα) and G protein-coupled estrogen receptor (GPER) mRNA was measured from five randomly selected animals per group. ERß data is not provided because its expression was extremely low (<1 copy/ng RNA). Kidneys were collected and stored in RNAlater solution (ThermoFisher Scientific, Waltham, MA). Purified RNA was collected from harvested kidney tissue (50mg) via the QIAGEN RNeasy Mini Kit (QIAGEN, Germantown, MD). Dual-labeled fluorescent probes for GPER (Assay ID: dRnoCPE5151056) and ERα (Assay ID: dRnoCPE5176827; Bio-Rad, Hercules, CA) were used with the Bio-Rad One-Step RT-ddPCR Advanced Kit, and droplets were analyzed in triplicate using the Bio-Rad QX200 system and QuantaSoft software, as previously described.¹⁶

Immunoblotting.

Kidney cortex and uterus was homogenized and immunoblotted as previously described.¹⁷ RIPA Lysis buffer containing protease and phosphatase inhibitors was added to 50 mg of tissue in a glass homogenizer. After obtaining a homogenous consistency, samples were centrifuged for 6 minutes at 12000 × g at 4°C and the supernatant was extracted. After protein determination using the BCA assay, 50 μg of kidney lysate and 20 μg uterine lysate was loaded onto a gel NuPage 10% Bis-Tris Gel (Invitrogen) and electrophoresis proceeded at 200 V. Gels were transferred to nitrocellulose membranes using the iBlot system (Invitrogen; Program 0, 7 min). After blocking with Odyssey Blocking Buffer (Li-Cor), membranes were incubated overnight at 4°C with monoclonal antibodies against ERα (Santa Cruz SC-8002, Lot G0717) and ERβ (Santa Cruz SC-373853, Lot G0117). GPER was not probed because antibodies failed validation in our hands. Odyssey IRDye 680RD Goat anti-Mouse IgG (Li-Cor 926–68070) was used as a secondary antibody and blots were imaged using the Odyssey system. Blots were reprobed with GAPDH (Cell Signaling 2118S; Lot 10) and secondary anti-rabbit as a loading control. Band densitometries were obtained using the ImageJ Gel Analyzer function.

Statistical analysis.

Blood pressure, tissue weights, RNA, and urinary and serum markers were analyzed using a parametric one-way ANOVA, followed by Tukey’s multiple-comparisons test. Outliers were identified using the ROUT method, and data are presented as means ± SEM. Renal glomerulosclerosis and brush border scores were analyzed using a nonparametric Kruskal-Wallis test, followed by a Dunn’s multiple-comparisons test when appropriate. Renal tubulointerstitial fibrosis, protein cast formation, vasa recta hyalinosis, peritubular congestion, and proximal tubule dilation were analyzed using Fisher-Freeman-Halton exact test, followed by Bonferroni correction when appropriate. Differences were considered statistically significant when P < 0.05. Analyses were performed using Prism version 6.0 (GraphPad Software, La Jolla, CA) and SPPS (IBM, v.22, Chicago, IL). All investigators were blinded to the treatment groups for the entirety of the study.

Results

Baseline systolic blood pressure did not differ between groups (Veh: 144 ±4; E2: 143 ±4 and E2+MPA: 147 ±4 mmHg; P = 0.24). There were no significant alterations from baseline blood pressure following 40 days of treatment or between groups (Veh: 135 ±5; E2: 143 ±6 and E2+MPA: 143 ±5 mmHg; Figure 1A; P = 0.55). Body weights were comparable between groups prior to OVX (Veh: 348 ±7; E2: 345 ±11 and E2+MPA: 335 ±10 grams; P = 0.44) and following 40 days of treatment (Veh: 373 ±13; E2: 365 ±19 and E2+MPA: 339 ±13 grams; P = 0.31). Uterine weights were used to confirm treatment efficacy following OVX. As expected, E2 significantly increased uterine weights compared with vehicle (P < 0.001). Unexpectedly, co-administration with MPA enhanced the impact of E2 on uterine weight (Figure 1B; P = 0.003). E2 therapy did not alter kidney weights from vehicle unless it was co-administered with MPA (Figure 1C; P = 0.002). There were no alterations in heart weights between treatment groups (Figure 1D).

Figure 1.

Animals received 40 days of vehicle (Veh), estradiol (E2) or estradiol + medroxyprogesterone (E2+MPA) immediately following OVX. (A) Final systolic blood pressure (BP), (B) uterine weight, (C) kidney weight, and (D) heart weight normalized to body weight. Values are means ± SEM, N=6–7 per group; One-way ANOVA, ^αP < 0.05 vs. Veh, ^βP < 0.05 vs E2.

E2 therapy for 40 days significantly increased proteinuria compared with vehicle (Figure 2A; P = 0.028). The E2-induced proteinuria was absent during MPA co-administration (P = 0.100). E2 therapy did not alter urinary creatinine (Figure 2B; P = 0.124) but significantly increased serum levels (Figure 2C; P = 0.003) compared with vehicle. In contrast, E2+MPA significantly increased urinary creatinine (P = 0.046) but reduced serum levels (P = 0.032) compared with E2 therapy. E2 therapy significantly reduced GFR (P < 0.001 vs. vehicle), but co-administration of MPA with E2 reversed the reduction (Figure 2D; P = 0.139 vs. vehicle).

Figure 2.

Post-treatment (A) 24 h proteinuria, (B) urinary creatinine, (C) serum creatinine, and (D) glomerular filtration rate (GFR). Values are means ± SEM, N=6–7 per group; One-way ANOVA, ^αP < 0.05 vs. Veh, ^βP < 0.05 vs E2.

Glomerulosclerosis was mild and not significantly different between treatment groups (Table 1; P=0.86). Renal brush border scores were not significantly different between groups (P = 0.151), with most sections displaying poor to moderate staining. Additional pathological changes in the kidneys due to E2 and E2+MPA were observed but not statistically significant. Renal protein cast formation was present in 1 out of 5 animals in the vehicle group compared with 5 out of 6 animals in the E2 group and 3 out of 6 in the E2+MPA group (P = 0.213). Vasa recta hyalinosis was present in 1 out of 5 animals in the vehicle group compared with 4 out of 6 in the E2 group and 1 out of 6 in the E2+MPA group (P = 0.382). Peritubular congestion and proximal tubule dilation were similar between groups (P = 0.127 and P = 1.000, respectively). The degree of renal fibrosis was significantly different between groups (Figure 3, P = 0.014). Specifically, fibrosis was undetectable in 100% of the vehicle group compared with only 16.7% of the E2 group and 66.7% in rats that were co-administered MPA.

Table 1.

Renal pathological changes assessed using index scores for glomerulosclerosis, brush border staining, protein cast formation, vasa recta hyalinosis, peritubular congestion, proximal tubule dilation, and interstitial fibrosis. Chi-square analysis results are reported in the text.

	Veh (n=5)	E2 (n=6)	E2+MPA (n=6)
Glomerulosclerosis
no fibrosis	0 (0%)	0 (0%)	0 (0%)
0–25% fibrotic	3 (60%)	1 (17%)	3 (50%)
25–50% fibrotic	2 (40%)	5 (83%)	2 (33%)
50–75% fibrotic	0 (0%)	0 (0%)	1 (17%)
75–100% fibrotic	0 (0%)	0 (0%)	0 (0%)
Brush border
None	0 (0%)	0 (0%)	0 (0%)
Poor	5 (100%)	3 (50%)	5 (83%)
Moderate	0 (0%)	1 (17%)	1 (17%)
Excellent	0 (0%)	2 (33%)	0 (0%)
Protein casts
Absence	4 (80%)	1 (16%)	3 (50%)
Mild	1 (20%)	3 (52%)	1 (16%)
Moderate	0 (0%)	1 (16%)	0 (0%)
Severe	0 (0%)	1 (16%)	2 (34%)
Vasa recta hyalinosis
Absence	4 (80%)	1 (17%)	5 (83%)
Mild	1 (20%)	4 (66%)	1 (17%)
Moderate	0 (0%)	0 (0%)	0 (0%)
Severe	0 (0%)	1 (17%)	0 (0%)
Peritubular congestion
Absence	4 (80%)	4 (66%)	2 (34%)
Mild	0 (0%)	2 (34%)	4 (66%)
Moderate	0 (0%)	0 (0%)	0 (0%)
Severe	1 (20%)	0 (0%)	0 (0%)
Proximal tubule dilation
Absence	4 (80%)	4 (66%)	5 (83%)
Mild	1 (20%)	2 (34%)	1 (17%)
Moderate	0 (0%)	0 (0%)	0 (0%)
Severe	0 (0%)	0 (0%)	0 (0%)
Interstitial fibrosis
Absence	5 (100%)	1 (17%)	4 (66%)
Mild	0 (0%)	2 (33%)	1 (17%)
Moderate	0 (0%)	3 (50%)	0 (0%)
Severe	0 (0%)	0 (0%)	1 (17%)

Figure 3.

Representative images of renal tubulointerstitial fibrosis assessed in trichrome stained kidney sections at 20X magnification (100 μm scale bar) following 40 days of (A) vehicle (Veh), (B) estradiol (E2), or (C) estradiol + medroxyprogesterone (E2+MPA). (D) Violin plot showing the data distribution using a scoring system to assess four grades of severity (0 = absence, 1 = mild, 2 = moderate, and 3 = severe). N=5–6 per group; Fisher exact test, ^αP=0.014 vs. Veh.

We recently published that ERα is the predominant estrogen receptor in the kidney followed by GPER.¹⁸ Moreover, ERα and GPER are suggested to play a renoprotective role in rodents.^17,19 Therefore, we evaluated whether alterations in renal estrogen receptor RNA or protein levels correlated with the adverse renal outcomes observed in our animal model. Results in the current study showed that renal ERα in both the E2 and E2+MPA treatment groups were comparable to each other but significantly lower than the vehicle group (Figure 4A; P = 0.018). In contrast, renal GPER levels were comparable between E2 and E2+MPA groups but significantly greater than the vehicle group (Figure 4B; P = 0.020). Renal ERß was below the level of detection, similar to our previous findings in Sprague-Dawley rats.¹⁸ Immunoblotting for ERα and ERß showed no differences between treatment groups (Figure 4C and 4D). We did not probe for GPER because no commercially available antibodies were able to pass validation studies using knockout tissue. ERα and ERß protein expression was highly correlated (Pearson r = 0.97, P < 0.0001, n = 12), and there was a negative but nonsignificant correlation between ERα protein and mRNA (Pearson r = −0.35, P = 0.39, n = 8). Full length blots are provided in Supplemental Digital Content 1.

Figure 4.

(A) Renal ERα and (B) renal GPER transcript copies per ng RNA after 40 days of vehicle (Veh), estradiol (E2) or estradiol + medroxyprogesterone (E2+MPA) treatment. Values are means ± SEM, N=5–6 per group; One-way ANOVA, ^αP < 0.05 vs. Veh. (C) Renal ERα and (D) renal ERß protein expression was not different between groups. Values are means ± SEM, N=3 per group; One-way ANOVA, P > 0.05.

Discussion

The novel finding from the current study was that co-administration of MPA blunted E2-induced renal damage and dysfunction in midlife OVX Long Evans rats. This study confirmed our previously published results that chronic E2 administration reduced GFR and increased proteinuria, while new data shows that these effects are prevented by MPA co-administration. In addition, our results support clinical data indicating that estrogen therapies may have worse renal outcomes when not combined with a progestogen.⁵

The physiological effects of estrogen are diverse and dependent on the target tissue. In the kidney, female sex hormones are assumed to be protective since renal disease progresses more slowly in women versus aged-match men prior to the onset of menopause.^20,21 Meta-analysis reveals mixed results for when evaluating the impact of estrogen therapy on albuminuria and proteinuria.⁴ In a small prospective study, estrogen plus norgestrel reduced proteinuria and improved creatinine clearance in postmenopausal women with hypertension and type 2 diabetes.²² The Rancho Bernardo Study cross-sectional study found better GFR in hormone users, especially with long-term estrogen use.²³ Other studies show negative renal outcomes, including a large case-controlled study using data from the Prevention of Renal and Vascular End Stage Disease cohort which found that menopausal hormone therapy is associated with increased microalbuminuria.²⁴ An observational study of premenopausal women with or without diabetes also found a positive association between oral contraceptive use and macroalbuminuria.²⁵ After correcting for age, diabetes, comorbidities, and baseline eGFR, the decline in eGFR was significantly greater with hormone therapy versus non-users.²⁶ This confusing mixture of positive, negative, and neutral results has lead to a lack of consensus on the effects of hormone therapy on renal outcomes.

Clinical studies assessing a potential renoprotective mechanism using progestogens in combination with estrogen add to the difficulty because not all progesterone derivatives are equal. For example, Norgestrel and gestodene stimulate the proliferation of breast cancer cells, while MPA does not impact cell growth.²⁷ Unexpected adverse cardiovascular effects in response to hormone therapy may also be amplified by natural aging. Estrogen adminsitration in aged female spontaneously hypertensive rats is less effective at reducing cardiac hypertropy when compared to young animals,²⁸ indicating that the process of aging disrupts the protective effects. Therefore, discrepancies in clinical studies assessing renal health outcomes may also occur due to differences in hormone formulation, route of administration, treatment duration, and timing of menopause onset versus hormone therapy initiation.

The current study showed that middle-aged female Long Evans rats display spontaneous hypertension, with the final systolic blood pressure and heart weights consistent between groups following the 40-day treatment. Therefore, E2 reduced GFR and increased proteinuria independent of blood pressure alterations. Co-administration with MPA prevented E2-induced renal dysfunction through compensatory renal hypertrophy, a mechanism that enhances the functional capacity of the residual nephrons to maintain normal renal function. Tubulointerstitial fibrosis promotes renal disease progression due to the excessive accumulation of extracellular matrix in the renal interstitium,²⁹ and our results suggest that MPA reduced E2-mediated renal damage by preventing kidney fibrosis. Progestogens are designed to stimulate progesterone receptors but can also influence additional steroid receptors, such as mineralocorticoid, glucocorticoid and androgen-receptors.^30–32 Therefore, modulation of non-progesterone receptors may play a role in the MPA response to protect the kidney against E2-mediated damage. Further studies will need to determine the molecular mechanism by which MPA influences the extracellular matrix, primarily its role in renal interstitial collagen formation.

Depletion of sex hormones with OVX resulted in reduced uterine weight, and treatment efficacy for E2 was confirmed by an increase in uterine weight compared with vehicle. Our adverse renal effects are not suspected to be the result of high E2 levels since we previously reported serum E2 levels to be approximately 15 pg/ml,⁷ which is lower than the plasma E2 levels of 30 pg/ml measured in intact female Long Evans rats.³³ Progestogens are commonly used to antagonize estrogen-induced growth in uteri tissue since it reduces and inhibits the replenishment of uterine estrogen receptors, and therefore, leads to reduced sensitivity to further estrogen stimuli and depressed uterine weight.³⁴ Therefore, we expected to see attenuation of uterine weight in response to MPA as reported elsewhere.^35–37 However, MPA-induced uterotrophic effects^38,39 or lack of antagonism of estrogen-induced uterotrophy^35,40–42 are also reported. These differences are independent of dosing, as the same dose of MPA produces opposite results in similar animal models.^36,39 Like most other steroids progesterone binds to a membrane receptor in addition to its nuclear targets.⁴³ Progesterone receptor membrane component (PGRMC) 1 and PGRMC2 are both expressed in rat ovary^44,45 as well as in mouse and human uterine tissue.⁴⁶ Some actions of MPA can be inhibited with PGRMC1 silencing,⁴⁷ and PGRMC knockout mice have abnormal uteri with increased glandular content.⁴⁸ A role for PGRMC in either uterine or kidney hypertrophy, however, is yet to be established.

Sex hormones regulate estrogen receptor expression,^49,50 but the function of these receptors in the kidney is not well defined. Female global ERα knockout mice display greater proteinuria and glomerular damage compared with the wild-type,¹⁹ while renal mesangial ERα expression is inversely correlated with glomerulosclerosis.⁵¹ These studies indicate that ERα plays a renoprotective role and may impact renal health outcomes. The current study showed that 40 days of E2 treatment in OVX Long Evans rats reduced renal ERα mRNA but not protein, confirming a similar effect following 120 days of E2 therapy in OVX Sprague Dawley rats,⁵⁰ however this transcriptional regulation did not correlate with ERα protein. Additional studies are needed to determine the feedback loop for ERα regulation in the kidney. Progesterone modulates ERα expression in uterine tissue during estrogen administration,⁵² but we did not find a difference in renal ERα levels between E2 and E2+MPA groups. ERα levels in the uterus are necessary for appropriate proliferation and thickening of the endometrium and are approximately two-fold higher than in the kidney of Sprague Dawley rats.¹⁸ Therefore, the ability of progesterone to regulate ERα expression may be absent in non-reproductive tissues such as the kidney.^53,54

The differing regulation of estrogen receptors in response to E2 may protect estrogen-sensitive tissues from over responding. E2 treatment in MCF-7 breast cancer cells causes a rapid reduction in both ERα and GPER, but GPER levels recover after 96 hours.⁵⁵ GPER was assessed in the current study since it is the second predominant renal estrogen receptor in rats and protective in cardiovascular and renal tissues.^18,56 We found an inverse relationship between renal ERα and GPER transcript levels, where ERα was downregulated and GPER was upregulated compared with controls. However, there was no difference between the E2 and E2+MPA groups despite GFR and proteinuria being worse in the E2 group. Therefore, our study indicates that the kidneys are estrogen-sensitive, but the adverse E2-mediated renal outcomes may be independent of alterations in expression.

Potential Clinical Value

There is a lack of consensus on the effects of hormone therapy on renal outcomes because clinical trials are limited and inconsistent. A survey of nephrologists shows that 43% are uncertain about the impact of menopausal hormone therapy in women with chronic kidney disease,⁵⁷ indicating that a standard of care for hormone therapy use in healthy women and in patients with reduced kidney function is needed. The current study used an aging rodent “menopausal” model to show that co-administration of MPA prevented the adverse renal outcomes associated with E2 therapy. These findings indicate that clinical administration of hormone therapies containing MPA may offer protection for patients that are susceptible to renal damage during estrogen use. Future studies will need to determine the molecular mechanism by which MPA protects renal health and whether the protective benefits extend to other progestogen formulas.

Conclusions

The present study demonstrated that chronic estrogen therapy had a different impact on renal health when not administered with a progestogen. Specifically, MPA attenuated the negative renal effects of E2 in midlife OVX Long Evans rats, independent of alterations in blood pressure. Furthermore, our results show that chronic exposure to E2 or E2+MPA therapy impacts renal estrogen receptor RNA levels, where ERα was downregulated and GPER was upregulated. Therefore, assessing renal markers of injury in postmenopausal women on hormone therapy may be clinically important and may help in decisions about dosage and discontinuation.

Supplementary Material

Supplemental Data File (.doc, .tif, pdf, etc.)

SDC 1: MENO-D-20-00130 SDC.docx