Renal TLR-7/TNF-α pathway as a potential female-specific mechanism in the pathogenesis of autoimmune-induced hypertension

Abstract

Hypertension is prevalent in patients with systemic lupus erythematosus (SLE). The goal of the current study is to track the pathogenesis of hypertension and renal injury in SLE, identify contributory mechanisms, and highlight differences in disease development among sexes. Mean arterial pressure was measured in conscious male and female SLE (NZBWF1) and control (NZW) mice at 34–35 wk of age using indwelling arterial catheters. Measures of renal injury, renal inflammation, and renal hemodynamics were used to monitor the potential contributors to latent sex differences. Both male and female SLE mice were hypertensive at 35 wk of age, and the hypertension was linked to renal injury in females, but not in males. A known contributor of renal pathology in SLE, Toll-like receptor (TLR)-7, and its downstream effector, the proinflammatory cytokine tumor necrosis factor (TNF)-α, were lower in male SLE mice than in females. Male SLE mice also had higher glomerular filtration rate (GFR) and lower renal vascular resistance (RVR) than females. Our data suggest that although hypertension in female SLE mice is associated with renal mechanisms, hypertension in male SLE mice may develop independent of renal changes. Future studies will continue to dissect sex-specific factors that should be considered when treating patients with hypertension with underlying chronic inflammation and/or autoimmunity.

NEW & NOTEWORTHY There is a high prevalence of hypertension in male and female SLE; however, male SLE mice are hypertensive without renal involvement. The development of hypertension in female SLE mice is renocentric and strongly associated with injurious renal mechanisms like the TLR-7→TNF-α pathway. This clear difference in the pathogenesis among the sexes could have a significant impact on how we treat patients with hypertension with underlying chronic autoimmune/inflammatory diseases.

INTRODUCTION

Hypertension remains a tremendous health care burden not only because of its immediate damaging effects on the kidneys and vasculature but also because of the cumulative detrimental effects on the cardiovascular system (1, 2). Despite the widespread availability of antihypertensive therapies, there is a subset of patients that are resistant to these therapies and hypertension persists (3–5). It is up to the scientific community to identify mechanisms of resistant hypertension and then search for new approaches to treat such patients.

Because of growing evidence linking autoantibodies, immune system overactivation, and inflammation to the causation of hypertension (6–8), many hypothesize that hypertension develops because of autoimmune processes. There is indeed a high prevalence of hypertension in autoimmune populations (9). Such autoimmune diseases tend to be female-dominated; however, men are also in danger and often discounted (10, 11). Despite understanding that sex is an important variable in hypertension, sex differences in the pathogenesis of hypertension in a setting of autoimmunity have never been studied.

Systemic lupus erythematosus (SLE) is a prototypical autoimmune disease that predominantly affects women of childbearing age (12). Women with SLE produce autoantibodies that accumulate and initiate a chronic renal inflammatory response (i.e., lupus nephritis) (13). This chronic renal inflammation eventually contributes to hypertension and renal injury in female patients with SLE (14, 15). In fact, the prevalence of hypertension in women with SLE can reach up to 74% depending on the cohort being studied (16). In addition, up to 60% female patients with SLE have some renal abnormality (17). We and others have consistently shown that spontaneous development of autoantibodies in an established mouse model of SLE, the female NZBWF1 mouse, promotes renal inflammation, hypertension, and renal injury (8, 18–25), just as in humans.

The pathogenesis of SLE is widely studied in female patients and female mouse models. There are few studies in male patients, however, the outcomes vary depending on the cohort studied (26). Some studies indicate that men that do develop SLE have an increased risk of hypertension and renal involvement (27–29), while some studies did not find an association of male sex with renal involvement in SLE (26, 30). The reliability of these outcomes is questioned by the inconsideration of confounding factors like disease duration, ethnicity, comorbidities, and even selection bias. Therefore, in this study, we used NZBWF1 mice to track autoimmune-induced hypertension to determine if similar contributing factors exist among the sexes.

Toll-like receptors (TLR) are immune receptors that recognize specific microbial elements and subsequently activate the innate immune response, as well as modulate the adaptive immune responses (31, 32). The gene encoding the protein TLR-7 (TLR7 gene in humans and Tlr7 gene in mouse) is found on the X chromosome, and incomplete inactivation of one of the X chromosomes in females is linked to increased prevalence of SLE in females (33). Gene duplication or overexpression of the region encoding TLR-7 promotes the development of SLE-like symptoms (31), while TLR-7-deficient mice are partially protected from lupus (34). It is also suggested that TLR-7 plays a role in the sex differences in immune responses in SLE leading to predominance in female SLE mice (35, 36). In many autoimmune diseases, activation of TLRs occur by endogenous DNA or RNA or ligands released from dead or dying cells or injured tissue (37). TLR-7 activation eventually causes the production and release of interferon-α (IFN-α) and tumor necrosis factor (TNF)-α (38, 39). The type 1 interferon (type I IFN) signature has been well studied and implicated in many autoimmune disorders (e.g., SLE, rheumatoid arthritis) as a driver of apoptotic factor and inflammatory cytokine production. However, serum TNF-α is also increased in patients with SLE compared with healthy subjects indicating a possible link between TLR-7 and increase in inflammation in this disease (40). In addition, TLRs are recognized as the contributors to sex differences in blood pressure regulation as well as cardiovascular diseases (41–43). Activation of TLR-7 with TLR-7 ligands led to increased IFN responses in healthy women compared with men (44). Interestingly, the expression of Tlr7 gene in renal cortex was higher in male versus female spontaneously hypertensive rats (45). However, the differences in the expression of renal TLR-7 protein in these mice are not known. Based on this, we hypothesized that differences in the development of hypertension and renal injury in SLE are linked to renal TLR-7 and renal inflammatory responses.

MATERIAL AND METHODS

Animals

Male and female SLE (NZBWF1; RRID: IMSR_JAX:100008) and control (NZW/LacJ; RRID: IMSR_JAX:001058) mice from Jackson Laboratories (Bar Harbor, ME) were used in this study. The NZBWF1 strain is a spontaneous mouse model of lupus that has been used extensively since the 1960s (46). The strain is a result of a cross between New Zealand White (NZW) males (used in this study as controls) and New Zealand Black (NZB) females. Symptoms such as autoantibody production, splenomegaly, and glomerulonephritis develop in a similar manner to human SLE in the NZBWF1 model (47). The NZW strain contains a few of the many susceptibility loci present in NZBWF1 strain, which contributes to some limited autoimmunity within this strain, albeit to a much lesser degree than NZBWF1(48). Mice in our studies were obtained at 4–6 wk of age and aged in a temperature-controlled facility with free access to food and water. As these mice develop albuminuria around 30 wk of age as indicated in our previous studies (23, 49, 50), blood samples were collected at 30 wk for dsDNA autoantibody detection and the urine was collected every week from 30 to 35 wk for dipstick analysis of albumin. Other than the blood sample collection at 30 wk and urine collection from 30 to 35 wk, all the measurements for this study were conducted at 35 wk of age as this is the age at which female SLE mice have been shown to be hypertensive with renal injury (23, 49, 50). All animal studies were approved by the University of North Texas Health Science Center’s Institutional Animal Care and Use Committee (IACUC) and were in accordance with National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals.

Plasma dsDNA Autoantibody Detection

Plasma was derived from the anticoagulated blood taken by retro-orbital capillary puncture at the age of 30 wk and again at 34–35 wk. The levels of anti-double-stranded DNA (dsDNA) autoantibodies were measured in the plasma via ELISA (Cat. No. 5120; Alpha Diagnostic International, San Antonio, TX) per the manufacturer’s instructions as previously described (19, 20, 50).

Blood Pressure Measurements

Mice were anesthetized with isoflurane and P-50 catheters (Braintree scientific, Braintree, MA) filled with sterile saline containing 1% heparin were implanted and secured in the left carotid artery. The opposite end of the catheter was exteriorized through the back of the neck between the scapula, secured with sutures, and sealed with a hot clamp. The mice were allowed to recover for 1 day. On the next day, the sealed part of the catheter was cut, the catheter flushed with heparinized saline, and then connected to a pressure transducer Deltran II (DPT-200, Utah Medical Products, Midvale, UT) coupled to a preamplifier (Bridge Amp, ML221; ADInstruments) and Powerlab computer data acquisition system (PowerLab 16/35, PL3516 ADInstruments). After a 1-h acclimation period, arterial pressure measurements were recorded for 30 min in these conscious, freely moving mice in their cages using LabChart software (49, 51). This procedure was repeated the following day. Mean arterial pressure (MAP), systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR) were collected, recorded, and averaged using the LabChart software. At the end of the study, all mice were euthanized and tissues were collected.

Assessment of Renal Injury, Renal Function, and Renal Inflammation

Mice were weighed and placed in metabolic cages for 12–24 h once every week from 30 to 35 wk of age to collect urine for albumin measurement using the dipsticks (Cat. No. 2191; Siemens Healthcare Diagnostics, Tarrytown, NY). The albumin-to-creatinine ratio (µg/mg) was calculated for 35 wk urine samples using albumin ELISA kits (Cat. No. 6300; Alpha Diagnostic Int, San Antonio, TX) and Creatinine Companion ELISA kit (Cat. No. 1012; Ethos Biosciences, Logan Township, NJ) (19, 50, 52, 53).

Western blot analysis.

Renal cortical and medullary TLR-7 and TNF-α expression were assessed via Western blot and normalized to total protein as previously described (23). Briefly, the homogenates of renal cortex and medulla were fractionated by Criterion TGX stain-free precast gels (Cat. No. 5678125, Bio-Rad, Hercules, CA), transferred to PVDF membranes (Cat. No. 1704157; Bio-Rad), and probed with primary and secondary antibodies. TLR7 was detected by probing homogenates with rat monoclonal anti-TLR-7 antibody (1:1,000, Cat. No. MAB7156; RRID: AB_10972762; R&D Systems, Minneapolis, MN) and TNF-α was detected in a similar manner using of a mouse monoclonal anti-TNF-α antibody (1:250, Cat. No. sc52746; RRID: AB_630341; Santa Cruz Biotech, Dallas, TX). Horseradish peroxidase-conjugated donkey anti-mouse (Cat. No. 610-703-002) or donkey anti-rat (Cat. No. 612–703-002) from Rockland Immunochemicals was used as the secondary antibody in 1:10,000 dilutions. Western blots were imaged using the ChemiDoc MP Imaging System (Bio-Rad; Hercules, CA). This imaging system is capable of detection of the total proteins in each lane in the stain-free blot image and detection of the target protein with chemiluminescent blot images following Super Signal West Femto Luminol/Enhancer Solution (Cat. No. 34095, Thermo Scientific, Rockford, IL). The results were analyzed using Image Lab software version 5.1 where the software allows normalization of the target protein in each lane to the total proteins in that corresponding lane. This method of using total proteins in the lane as loading control has been found to be more accurate as opposed to single housekeeping protein controls (54–56).

Immunohistochemistry and histological analyses.

Periodic Acid-Schiffs staining was performed on the formaldehyde-fixed and paraffin-embedded sections of the kidney cut at a thickness of 4 µm to evaluate glomerular histology and glomerulosclerosis. The sections were treated with 0.5% periodic acid solution followed by Schiff reagent (Cat. No. 26052-05) and Harris hematoxylin (Cat. No. 26041-06) from Electron Microscopy Solutions, Hatfield, PA. The renal changes were assessed histopathologically in glomeruli in a blinded manner. About 80–100 random glomeruli were analyzed for each mouse. The glomerulosclerosis index (GSI) was determined for each glomerulus as the percentage of PAS-positive staining with loss of capillaries in the glomeruli as follows: grade 0 = normal glomerulus, grade 1 = upto 25%, grade 2 = 25–50%, grade 3 = 50–75%, and grade 4 = 75–100% damage. For interstitial fibrosis, the kidney sections were stained with the Masson Trichrome staining kit (Cat. No. KTMTR2; Stat Lab, Mckinney, TX) according to the manufacturer’s protocol. The images were captured using a Nikon Eclipse Ni microscope equipped with a Nikon DS-Fi2 color camera (Nikon) and NIS-Elements BR 4.30.01 software. We analyzed the percentage of blue, which primarily indicates the collagen in the captured image. First, the threshold for the blue staining in Masson’s trichrome-stained sections was set using NIS-Elements BR 4.30.01 software and the same thresholding parameters were used for all the images to measure renal fibrosis.

Quantification of markers of renal injury.

Using the ELISA technique, the markers of the renal tubular injury, namely, kidney injury molecule-1 (KIM-1; Cat. No. MKM100; R&D Systems; Minneapolis, MN) and neutrophil gelatinase-associated lipocalin (NGAL; Cat. No. MLCN20; R&D Systems; Minneapolis, MN) were measured in the urine at 34–35 wk according to the manufacturer’s protocol. The concentrations of urinary KIM1-1 and NGAL were normalized to the urinary creatinine concentration (Creatinine Companion ELISA kit, Cat. No. 1012, Ethos Biosciences, Logan Township, NJ).

Sinistrin clearance.

Sinistrin, an inulin analog, was used for accurate assessment of glomerular filtration rate (GFR). Mice were anesthetized with isoflurane and the transdermal fluorescence detector (MB 0309 Mini, MediBeacon Inc., St. Louis, MO) was directly attached to the skin. FITC-conjugated sinistrin was then administered by retro-orbital injection at 0.03 mg/g body wt (0.03–0.05 mL). Mice were allowed to recover from anesthesia and then measurements of the excitation kinetics of the exogenous GFR tracer, FITC-sinistrin, were recorded in freely moving, conscious mice for 1.5 h. GFR was calculated based on the half-life (t_1/2) of plasma FITC-sinistrin decay using the formula 14616.8/(t_1/2) and reported as µL/min/100 g body wt (57, 58).

Renal blood flow and vascular resistance.

Following a posterior incision, the right renal artery was isolated from the corresponding vein in a subset of anesthetized mice and placed in a Transonics (Ithaca, NY) flow probe to measure renal blood flow (RBF) at 35 wk of age. RBF (mL/min) and anesthetized MAP were gathered simultaneously for 30 min following a 30-min stabilization period using Powerlab software. Renal vascular resistance (RVR; mmHg·min·kg·mL⁻¹) was calculated by dividing MAP by RBF normalized to body weight.

Statistical Analysis

Statistical analysis was conducted using Graphpad Prism 9 (GraphPad Software, San Diego, CA). Data were calculated as mean ± SE. Unpaired t test or two-way ANOVA, followed by a Holm–Sidak post hoc test was used to determine differences between groups, with a P < 0.05 indicating a significant difference.

RESULTS

All SLE Mice Produce Autoantibodies over Time, with Higher Levels in Females than in Males

Plasma levels of anti-dsDNA autoantibodies, a clinical marker of SLE, were present in male SLE mice (2.1e5 ± 4.2e4 activity U/mL) and female SLE (6.1e5 ± 9.7e4 activity U/mL) mice at 30 wk of age and male SLE mice (2.5e5 ± 4.6e4 activity U/mL) and female SLE (6.3e5 ± 1.1e5 activity U/mL) mice at 35 wk of age. (Fig. 1). Although male SLE mice had tendency of higher dsDNA autoantibodies than male controls (4.3e4 ± 4.9e3 activity U/mL at 30 wk and 5.9e4 ± 7.4e3 activity U/mL at 35 wk) and female controls (5.8e4 ± 1.0e4 activity U/mL at 30 wk and 4.9e4 ± 7.6e3 activity U/mL at 35 wk), the difference was not statistically significant. However, female SLE mice were significantly different from control mice as well as the male SLE (P < 0.001) at 30 wk and 35 wk. Close examination of the antibody activity index indicates 54% of males with positive antibody titer versus 92% in females at 30 wk, and 60% versus 97%, respectively, at 35 wk. In addition, as the parental strain has traces of autoimmunity (59), we compared plasma dsDNA autoantibodies in male and female SLE mice with a separate group of naïve, age-matched C57 mice (4.6e4 ± 1.1e4; data not shown) and this yielded a significantly higher level of dsDNA autoantibodies in both the male (P = 0.010) and female (P = 0.002) SLE mice.

Figure 1.

Both male and female mice develop characteristic autoantibodies of SLE. Plasma levels of double-stranded DNA (dsDNA) autoantibodies (activity units) in male and female control and SLE mice at 30 wk and 34/35 wk of age (n = 14–29/group). The results of the two-way ANOVA with repeated measures are indicated on the graph. P values were calculated using the Holm–Sidak post hoc analysis. P < 0.05 was considered statistically significant (n = 5 or 6 mice/group); ****P vs. SLE/male and control mice. SLE, systemic lupus erythematosus.

Male and Female SLE Mice Develop Similar Level of Hypertension

MAP is higher in SLE mice than in controls. Because of a lack of significant interaction following a two-way ANOVA, no significant group comparisons can be reported in MAP (Fig. 2). However, if you consider the MAP of a normal mouse [e.g., C57BL6/J mice, ∼100 mmHg (60)], blood pressure in both sexes is higher and therefore these mice are hypertensive (males: 167 ± 8, females: 154 ± 6). In addition, if comparing only control female mice, our historical control, to male and female SLE mice (via one-way ANOVA), MAP was significantly elevated in both (males: 167 ± 8, P = 0.0014; females: 154 ± 6, P = 0.0212) compared with control females (125 ± 4 mmHg). There were similar hypertensive trends in systolic and diastolic pressures that coincided with elevated HR in both male and female SLE mice (Table 1).

Figure 2.

Both male and female SLE mice are hypertensive. Mean arterial pressure in male and female control and SLE mice at 35 wk of age analyzed by two-way ANOVA (n = 5 or 6 mice/group). The results of the two-way ANOVA are indicated on the graph. P < 0.05 was considered statistically significant. Mean arterial pressure is higher in SLE mice, but since there is no significant interaction following a two-way ANOVA, no group comparisons are included. Mean arterial pressure in male and female SLE mice when compared with our historical control only (control female mice) analyzed by one-way ANOVA indicate both male and female SLE mice had significantly higher blood pressure (mmHg) than female controls (data not shown). Also, if compared with normal C57BL6/J mice with mean arterial pressure of 100 mmHg, the male and female SLE mice are hypertensive (data not shown). SLE, systemic lupus erythematosus.

Table 1.

SBP, DBP, and HR in male and female control and SLE mice at 35 wk of age

	Control		SLE		P Values
	Male	Female	Male	Female	Interaction	Sex effect	SLE effect
SBP, mmHg	147.43 ± 4.94	131.45 ± 4.19	170.42 ± 8.33	156.82 ± 7.68	0.86	0.03	0.0015
DBP, mmHg	133.25 ± 7.22	121.58 ± 4.94	165.86 ± 8.31	154.46 ± 7.53	0.99	0.13	0.0003
HR, beats/min	357.01 ± 91.34	497.24 ± 130.46	784.38 ± 310.05	586.93 ± 48.80	0.71	0.052	0.50

Values are means ± SE; n, 5 to 6 number of mice per group. Systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR) in male and female control and systemic lupus erythematosus (SLE) mice at 35 wk of age were analyzed by two-way ANOVA. P < 0.05 was considered statistically significant. As there is no significant interaction following a two-way ANOVA, no group comparisons are reported.

Male SLE Mice Do Not Develop Renal Injury like Age-Matched Females

The percentage of mice with positive albuminuria (i.e., urinary albumin concentration of >300 mg/dL) was increased in female SLE mice from week 30 to 35. About 50% (13 out of 26) of female SLE mice were positive for albuminuria as indicated by dipstick at 35 wk, but only 4% (1 of 25) of male SLE mice, 4% (1 of 27) of female controls, and none of the male control mice had positive albuminuria at 35 wk (Fig. 3A). Albumin creatinine ratio (µg/mg) was significantly higher in female SLE mice (13,857.03 ± 4,231.01) than in male SLE mice (2,289.63 ± 1,706.38; P = 0.0032), female controls (1,614.15 ± 1,539.78; P = 0.0006), and male controls (40.42 ± 5.38; P = 0.0005, Fig. 3B).

Figure 3.

Male SLE mice do not develop renal injury. The percentage of mice with positive albuminuria (A) and albumin-to-creatinine ratio (µg/mg; B) in male and female control and SLE mice (n = 15 mice/group). The results of the two-way ANOVA are indicated. P < 0.05 was considered statistically significant. **P vs. SLE/male and control mice. SLE, systemic lupus erythematosus.

Representative images of PAS staining and Masson trichrome staining indicate increased glomerular pathology and renal fibrosis in female SLE mice as seen in Fig. 4, A and C, respectively. The corresponding analysis of the images confirms significantly high glomerular sclerosis index (Fig. 4B) in female SLE mice (2.93 ± 0.2) as compared with both the male SLE mice (1.00 ± 0.04; P < 0.0001) and female control mice (1.07 ± 0.05; P < 0.0001). Similarly, as seen in Fig. 4D, renal fibrosis (% blue staining) was higher in female SLE mice (21.34 ± 3.17) than in both male SLE mice (4.50 ± 0.57; P < 0.0001) and female control mice (2.53 ± 0.51; P < 0.0001). The concentration of tubular injury marker, urinary KIM-1 normalized to urinary creatinine (ng/mg) (Fig. 4E), was significantly higher in female SLE mice (33.27 ± 8.37) than in male SLE mice (2.44 ± 0.17; P = 0.0017). Urinary NGAL normalized to urinary creatinine (ng/mg) was higher in female SLE mice (4,465.69 ± 2,110.31) than in male SLE mice (36.06 ± 2.71; P = 0.1093) and female controls (54.69 ± 8.57; P = 0.1304), however, the results were not statistically significant (Fig. 4F).

Figure 4.

Female SLE mice demonstrate higher glomerular and tubulointerstitial injury. Paraffin-embedded kidney sections were stained with PAS stain (A) and Masson trichrome stain (C) to assess the glomerular injury and to assess the tubule-interstitial fibrosis respectively. P < 0.05 was considered statistically significant. The glomerulosclerosis index (B) was higher in female SLE mice (n = 6–10 mice/group; ****P vs. SLE/male and control mice). The percentage of area of blue staining indicating renal fibrosis (D) was also higher in female SLE mice as compared with the other groups (n = 6–10 mice/group; ****P vs. SLE/male and control mice). Similarly, female SLE mice had higher concentration of the markers of tubular injury, urinary KIM-1 normalized to creatinine (E), as compared with the male SLE mice (n = 8–12 mice/group; **P vs. SLE/male) and higher but not significant concentration of urinary NGAL normalized to creatinine (F), as compared with the control and male SLE mice (n = 7–12/group). Statistical comparisons were made using a two-way ANOVA followed by the Holm–Sidak post hoc analysis. NGAL, neutrophil gelatinase-associated lipocalin; SLE, systemic lupus erythematosus.

Female SLE Mice Have Impaired Renal Function than Males

Female SLE mice had significantly higher RVR (Fig. 5A) than male SLE mice (10.07 ± 1.23 mmHg·min·kg·mL⁻¹ vs. 5.15 ± 0.60 mmHg·min·kg·mL⁻¹; P = 0.008). Anesthetized MAP (Fig. 5B; male SLE mice 77 ± 4 mmHg; female SLE mice 89 ± 6 mmHg) was divided by renal blood flow normalized to body weight (Fig. 5C; male SLE mice 15.7 ± 1.35 mL·min⁻¹·kg⁻¹; female SLE mice 10.6 ± 2.18 mL·min⁻¹·kg⁻¹) to yield RVR. GFR (Fig. 5D) was lower in female SLE mice than in males (805 ± 77 vs. 1,066 ± 60 µL/min/100 g body wt; P = 0.029).

Figure 5.

Female SLE mice had higher renal vascular resistance and lower GFR than male SLE mice. Conscious, anesthetized mice were implanted with a catheter in the carotid artery to obtain anesthetized blood pressure (mmHg; B). Renal blood flow (RBF, mL/min) was obtained using a Transonics probe (C). Renal vascular resistance (RVR) was calculated dividing anesthetized blood pressure by RBF. Data were normalized by dividing by body weight (g). P < 0.05 was considered statistically significant. Female SLE mice had significantly higher RVR adjusted for body weight (mmHg·min·kg·mL⁻¹) than male SLE mice (A) (n = 6–9 mice/group; *P vs. SLE/male). Glomerular filtration rate (µL/min/100 g body wt) was measured in conscious, freely moving mice and was higher for male SLE mice (D). Statistical comparisons for all were made using unpaired t test. (n = 4–9 mice/group; *P vs. SLE/male). GFR, glomerular filtration rate; SLE, systemic lupus erythematosus.

Female SLE Mice Express Higher Levels of Renal TLR-7 than Males

Renal cortical expression of TLR-7 was increased in female SLE mice at 35 wk of age (Fig. 6A) when compared with both male SLE mice and female controls. Female SLE mice expressed a significantly higher renal cortical expression of TLR-7 (3.9e5 ± 8.0e4 intensity units) than both male SLE (7.0e4 ± 1.5e4 intensity units; P < 0.0001) and female control mice (7.5e4 ± 2.6e3 intensity units; P < 0.0001). This trend was also true of renal medullary TLR-7 (Fig. 6B) with female SLE mice again expressing significantly higher levels of TLR-7 (5.3e5 ± 1.2e5 intensity units) than male SLE (1.3e5 ± 4.6e4 intensity units; P = 0.01) and female control mice (7.7e4 ± 1.7e4 intensity units; P = 0.0037).

Figure 6.

Female SLE mice express higher levels of renal TLR-7 expression. Protein expression of TLR-7 was assessed via Western blot in the renal cortices (A) and medullas (B) of male and female SLE and control mice. Female SLE mice displayed a significantly higher expression of TLR-7 (at 66 kDa, the cleaved version) than both male SLE and female control mice. Statistical comparisons were made using a two-way ANOVA. The results of the two-way ANOVA are indicated on both graphs. P values were calculated using the Holm–Sidak post hoc analysis. P < 0.05 was considered statistically significant (n = 5 or 6 mice/group; ****P and **P indicates significant difference vs. SLE/male and control mice for cortical and medullary renal TLR-7, respectively). SLE, systemic lupus erythematosus; TLR-7, Toll-like receptor-7.

Female SLE Mice Express Higher Levels of Renal TNF-α than Males

Renal cortical expression of TNF-α was increased in female SLE mice (1.3e6 ± 3.7e5) (Fig. 7A) when compared with both male SLE mice (5.1e5 ± 1.6e5 intensity units; P = 0.046) and female control (1.2e5 ± 3.4e4 intensity units; P = 0.0033) mice. This trend was also true of renal medullary TNF-α (Fig. 7B) with female SLE mice again expressing significantly higher levels of TNF-α (9.5e6 ± 1.8e6 intensity units) than male SLE (1.7e6 ± 6.0e5 intensity units; P = 0.0002) and female control mice (1.0e6 ± 3.1e5 intensity units; P < 0.0001).

Figure 7.

Female SLE mice express higher levels of renal TNF-α expression. Protein expression of was assessed via Western blot in the renal cortices (A) and medullas (B) of male and female SLE and control mice. Female SLE mice displayed a significantly higher expression of TNF-α (at 26 kDa, the transmembrane form) than male SLE and female control mice in both the renal cortex and medulla. Statistical comparisons were made using a two-way ANOVA. The results of the two-way ANOVA are indicated on both graphs. The P values were calculated using the Holm–Sidak post hoc analysis. P < 0.05 was considered statistically significant (n = 5 or 6 mice/group; *P and **P indicate significant difference vs. SLE/male and control mice for cortical and medullary renal TNF-α, respectively). SLE, systemic lupus erythematosus.

DISCUSSION

The major finding of this study is autoimmunity causes hypertension in both male and female mice, albeit through different mechanisms. Hypertension in female SLE mice was associated with increased renal inflammation, renal injury, and impaired renal function at 35 wk of age, whereas hypertension in age-matched male SLE mice appears to have no renal connection. Specifically, the female SLE mice had increased expression of TLR-7 and TNF-α in the kidneys. This intriguing finding will help us understand sex differences in the development of both SLE and the pathogenesis of hypertension in the setting of chronic inflammatory disease.

Sex differences exist in the prevalence of SLE (61). SLE is a female-dominant disease, with nearly 90% of all patients being young women of childbearing age (62, 63). Sex hormones were thought to play an immediate role in the female predisposition to SLE, with estrogen playing a permissive role in the development of the disease (64). However, recent studies using animal models view the contribution of estrogen to SLE as complex, with the hormone alternating between a protective and pathogenic role at different points in life (65). Although the notable female bias is still being characterized, sex differences in the development of the disease itself are even less defined. According to cohort studies, SLE in men is diagnosed at a later age and associated with higher prevalence of organ damage and cardiovascular manifestations (e.g., myocardial infarction, peripheral vascular disease) (66). However, it is not clear whether these outcomes are due to the progression of the disease or because male patients do not seek medical expertise as often or as soon as symptoms present, which itself could lead to delayed intervention and more severe health consequences (27, 67). Regardless, the underlying mechanisms behind these apparent sex differences in the pathogenesis of SLE are not clearly understood. In particular, the reason for the sex disparity in SLE hypertension has not been explored until now.

We have consistently found that as the disease progresses in female SLE mice, dsDNA autoantibodies develop, form complexes, and deposit into the kidney, causing lupus nephritis (20, 22, 23, 50). In our study, the dsDNA autoantibodies reached maximum values at earlier time points (30 wk) in female SLE mice, whereas in male SLE mice, the levels tend to increase slowly even after 30 wk (Fig. 1). Looking at earlier timepoints may reveal the time-course in the development of these autoantibodies in female SLE mice, whereas later timepoints may reveal their progression in male SLE mice. After the dsDNA autoantibodies developed, SLE female mice had albuminuria starting at 30 wk (Fig. 3A) indicating immune complex-mediated damage. Interestingly, the resultant renal inflammation is associated with hypertension in female SLE mice by the time the animals reach 34–35 wk of age (18–22). Our central hypothesis is that male SLE mice would have a similar disease progression as the females, but our data indicate male SLE mice do not develop renal injury as early as the female SLE mice, yet both male and female SLE mice are hypertensive by 35 wk of age (Table 1, Fig. 2). This implies that hypertension in male SLE mice may arise through an alternative mechanism. Furthermore, this may indicate that the development of autoimmune-associated hypertension, and potentially resistant hypertension, could occur through different pathways in males and females. One of the findings of this study was that the control male mice also displayed MAP on the higher side. This may have caused the lack of significant statistical differences in MAP between the control and SLE mice. The NZBWF1 mouse is the cross between a male NZW and female NZB and there is evidence of autoimmunity in both (68–70). From Table 1, it is clear that because of a lack of significant interaction following a two-way ANOVA, no significant group comparisons can be reported in the parameters SBP, DBP, and HR.

The higher expression of renal TLR-7 in female SLE mice was consistent with our hypothesis and supports findings of previous studies asserting that female predisposition to SLE could be linked to increased expression of TLR-7 because of incomplete X inactivation (33, 36). The proposition that an increased expression of the TLR-7 gene is linked to heightened susceptibility to autoimmune disorders is strengthened by other mouse models of SLE. The BXSB mouse model develops SLE only in males and this is mainly attributed to the translocation of the region of X chromosome containing the Tlr7 locus onto the Y chromosome and subsequent overexpression of TLR-7 in these mice (71, 72). Although the results of our study do not directly investigate and quantify the gene expression of TLR-7 in female SLE mice, the expression of renal TLR-7 in these animals was shown to be higher than both male SLE and female controls, indicating a correlation between increased TLR-7 expression and renal injury in the context of SLE. Our findings are complementary to those by Conti et al. (73) who demonstrated higher tubulointerstitial TLR-7 in the kidney sections of patients from lupus nephritis and it was correlated with the chronicity index of the disease. Although this project confirmed a higher level of renal TLR-7 expression in female SLE mice, these data do not quantify the sensitivity of the receptor. Also, determining the cellular location of the TLR-7 in the kidneys can be an interesting cellular pathway to explore in future. Previous studies indicate that TLR-7s are expressed by many immune cells like plasmacytoid DCs (pDCs), B cells, macrophages, and in kidney cells including proximal and distal tubules and Bowman’s capsule (36, 74–76). Sex differences exist in the expression of lupus-associated micro-RNAs (mi-RNAs) (77, 78). Dai et al. (79) demonstrated that miR-182-96-183 cluster, miR-155, miR-31, miR-148a, miR-127, and miR-379, were markedly higher at 30 wk of age in the splenocytes of female NZBWF1 mice when they exhibit moderate to severe lupus as compared with the male NZBWF1 mice. Administration of estrogen in orchidectomized male NZBWF1 mice in this study increased the expression of most of these lupus-associated miRNAs. Deng and team (80) provided evidence that miRNA-3148 may regulate the expression of TLR-7. As miRNAs are involved in the pathogenesis of cardiovascular and kidney diseases, it will be interesting to study the sex differences in the miRNAs and their association with renal TLR7 and development of hypertension in lupus.

The results of this study provide a possible mechanism for some of the renal injury and hypertension seen in female SLE, which we found to differ between males and females. The observational design of this study limits the ability to make a causative determination between activation of TLR-7 and hypertension. Administration of a TLR-7 antagonist (i.e., hydroxychloroquine) in female SLE mice alleviates renal inflammation in SLE and attenuates the subsequent development of hypertension and renal injury (81). It is not yet known whether the same would occur in male SLE mice. Interestingly, chloroquine was also antihypertensive and protected the endothelium of spontaneously hypertensive males (82). A recent study by Hayakawa et al. (83) used TLR-7 agonist, imiquimod (IMQ), in SLE-prone mice and found that IMQ aggravated lupus nephritis in female SLE mice, whereas in male SLE mice, there was minimal aggravation of lupus nephritis. It is likely that lack of effect of IMQ in male SLE mice as compared with female SLE mice in this study is due to the lower expression of renal TLR-7 itself. Although past drugs targeting TLR-7 have been used as a potential treatment for human SLE (82, 84), the use of these drugs has not been examined within a context of mitigating eventual cardiovascular outcomes in addition to the “normal” symptoms of SLE such as erythematic rash and pain (85). Investigating the effects of a TLR-7 antagonist and its potentially differential effects in males and females would afford the opportunity to further examine the link between autoimmunity and hypertension.

TLR-7 activation of type I IFN is well known. The TULIP (Treatment of Uncontrolled Lupus via the IFN Pathway) 2 clinical trial for the drug, anifrolumab completed phase III in August 2019, and the drug was approved by Food and Drug Administration in 2021 for the treatment of lupus (86). The target of anifrolumab is the type I IFN receptor; however, induction of this pathway is largely due to a preceding activation of TLR-7. The higher expression of renal TLR-7 in female SLE mice indicates elements of this pathway may be hyperactive in female SLE (33, 36, 44). The results of our study insinuate that the efficacy of this new therapy may differ between males and females because of the differential expression of TLR-7. Supporting our hypothesis, Davison et al. (87) showed that myeloid-derived suppressor cells (MDSCs) that are prominent in NZBWF1 male mice express protein calprotectin S100a9, which exerts the immunosuppressive effect by inhibiting the type I interferon signal in these male mice. Although both men and women were included in the TULIP trial, there is no evidence to indicate the efficacy of the drug was analyzed by sex, especially in the context of hypertension. Accordingly, the type I interferon receptor may not be as beneficial a target for treating the symptoms of SLE in male patients, creating a need to consider and develop more efficacious therapies for this population. In our study, we focused on TNF-α, as TNF-α antagonist etanercept, decreased kidney injury and blood pressure in female NZBWF1 mice (52). The results of our study indicate that the effect of etanercept may not be similar in male NZBWF1 mice. Further in-depth investigation of the renal TLR-7/TNF-α pathway and its role in the development of hypertension in male and female lupus mice is warranted to understand the pathogenic contribution of this pathway in autoimmune hypertension.

Suspecting impaired renal hemodynamics as an alternative contributor to the sex difference in renal injury and hypertension in SLE mice, we hypothesized that male SLE mice have increased RVR, which could explain hypertension without signs of renal injury. We found that female SLE mice had higher RVR than male SLE mice and that GFR was higher in males. Renal function appears to be better in male SLE mice. There was no difference between the control and the SLE groups but there were obvious differences in male and female mice for RVR (data not shown). Although in our study the progression of development of hypertension could not be studied over time, the results do indicate that in female SLE mice hypertension may be associated with renal mechanisms and that in male SLE mice at least the development of hypertension occurs through other mechanisms to begin with.

In conclusion, the results of this study confirm sex differences in the development of autoimmune-induced hypertension. Although hypertension was observed in both male and female SLE mice, only the females displayed signs of renal inflammation and renal injury. The current study focused on increased expression of renal TLR-7 in female SLE as an explanation for the increased prevalence of renal injury in these mice. Although the data supported our hypothesis, the cause of hypertension in male SLE mice is still unidentified. These data imply that some complications (specifically hypertension) in SLE result from dysfunctions in separate pathways that may not respond to a singular intervention, which may contribute to the increased prevalence of resistant hypertension in patients with SLE. Although traditional medications for hypertension may help reduce blood pressure in some SLE patients, they would not effectively target the contributions of autoimmunity such as TLR-7 activation and renal inflammation.

Furthermore, current therapeutics for SLE that target the TLR-7/type I IFN pathway may only effectively reduce complications associated with the disease in women, who make up a majority of patients with SLE. However, future efforts to study and treat SLE and other autoimmune disorders will need to closely consider the effect of sex on the development of symptoms, specifically hypertension, to effectively treat all patients.

GRANTS

This study was partially funded by National Institutes of Health Grants R01HL153703 and K01HL139859 (to K.W.M.) and R01DK079968-01 (to R.M.) and Lupus Research Alliance Grant 550778 (to K.W.M.).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

B.M.D. and K.W.M. conceived and designed research; S.C., B.M.D., J.Y.M., C.M.Y-S., C.G.S., R.M., and K.W.M. performed experiments; S.C., B.M.D., J.Y.M., C.M.Y-S., C.G.S., R.M., and K.W.M. analyzed data; S.C., B.M.D., C.G.S., and K.W.M. interpreted results of experiments; S.C., B.M.D., J.Y.M., C.G.S., R.M., and K.W.M. prepared figures; B.M.D. drafted manuscript; S.C. and K.W.M. edited and revised manuscript; S.C., C.M.Y-S., C.G.S., R.M., and K.W.M. approved final version of manuscript.