Renal tubular estrogen ß receptors are expressed at high levels in small vessel vasculitis and are primarily localized in the distal tubule

This study investigated the possible roles of renal estrogen receptors (ER) in glomerulonephritis associated with small vessel vasculitis. The relationships of ERs were investigated in antineutrophilic cytoplasmic antibody (ANCA)‐associated glomerulonephritis and immunoglobulin A (IgA) nephropathy groups, which are small vessel vasculitis subtypes with two different glomerulonephritis development pathophysiologies. The design of this study was prepared as a retrospective cohort study. The study included 42 patients with ANCA‐associated vasculitis and 18 with IgA nephropathy in the small vessel vasculitis group. For the control group, intact renal tissues of 28 patients who underwent nephrectomy due to renal cell carcinoma were used. Renal biopsy samples of the groups were stained with ER beta (ß) and ER alpha (α). Tubular ER ß expression score (TERßES) median values were found to be significantly higher in ANCA‐ associated vasculitis (B = 0.724, OR [95%CI]: 2.064 [1.141–3.731], p = .016) and IgA nephropathy (B = 0.898, OR [95%CI]: 2.454 [1.307–4.609], p = .005) than in intact kidney tissue. It was determined that tubular ERß was most frequently localized in the distal tubule at 57.9% and the second most common in the proximal tubule at 20.4%. The expression of tubular ERß is increased in glomerulonephritis due to small vessel vasculitis. Tubular ERßs are most commonly localized in the distal tubule. Further studies are needed to understand the physiological and pathophysiological effects of altered renal ER levels in small vessel vasculitis.


| INTRODUCTION
2][3][4][5] Three estrogen receptors (ER), G protein-coupled estrogen receptor (GPER), ERα, and ERß mediate the biological and pathophysiological effects of estrogen. 6lthough ERs are primarily found in the nucleus, they are also located in the mitochondria and cytoplasm. 7The classical ER subfamily is considered to consist mainly of ERα and ERß. 8 ERs bind to intracellular DNA motifs and participate in various processes such as nuclear translocation, transcription, and post-translational modification. 9strogen and ERs have essential roles in various physiological processes such as endothelin-1 (ET-1) system regulation 10 and maintenance of mitochondrial system homeostasis in the care of renal functions. 11In rat kidneys, GPER1 activation stimulates natriuresis via endothelin receptors A/B, and activation of this receptor prevents the increase in mean arterial pressure due to ovariectomy. 12,13strogen activates NaPi-IIa, a phosphate transporter, by binding to ERα/ERß localized in renal proximal tubules and thus plays physiological roles in regulating phosphorus homeostasis. 14In addition, estrogen attaching to ERs creates beneficial therapeutic effects for improving coagulopathic conditions by controlling the impaired larginine/nitric oxide (NO) pathway in uremic rats. 9,157][18][19][20][21][22][23] Estrogen acts on the principle of phosphorus metabolism through its receptors in the proximal tubule. 14In addition, there is a close relationship between the susceptibility and progression of various renal diseases and ERα polymorphisms. 24Recent publications show the contribution of pathological changes in estrogen and ER signaling pathways in different clinical conditions such as acute kidney injury (AKI) models, lupus nephritis (LN), diabetic nephropathy, immunoglobulin A (IgA) nephropathy, and complications of chronic renal failure.Experimental and clinical studies investigating the relationship of estrogen and ERs with various renal diseases have shown that approaches targeting estrogen and ER signaling pathways can have protective effects against renal diseases. 9o our knowledge, only three studies have previously evaluated renal ERs in humans.While one of these studies compared renal ERs between IgA nephropathy and LN, 25 the other one investigated renal ERs expressed in diabetic glomerular mesangial cells in cell culture. 26The last study was conducted by us, evaluating renal ERs in LN, familial Mediterranean fever (FMF) amyloidosis, antineutrophilic cytoplasmic antibody (ANCA)-associated vasculitis, and intact kidney tissue. 27Renal ERs have not been compared between ANCA-associated glomerulonephritis, IgA nephropathy, and intact kidney tissue, and the relationship between renal histopathological patterns.This study aimed to investigate the possible roles of ERs in glomerulonephritis due to small vessel vasculitis.The relationships of ERs were investigated in ANCA-associated glomerulonephritis and IgA nephropathy groups, which are small vessel vasculitis subtypes with two different glomerulonephritis development pathophysiologies.In this study, we showed the relationship between the clinical and histopathological features of ANCA-associated glomerulonephritis and IgA nephropathy and renal ER expression scores for the first time.We also revealed information about the renal tubular localization of ERs.

| Study design
This retrospective cohort study evaluated renal biopsies of ANCA-associated vasculitis patients with pauciimmune glomerulonephritis and patients with IgA nephropathy with immunocomplex deposition between 1/6/2019 and 1/6/2022.While the renal histopathological parameters were obtained retrospectively from the pathology reports, the data of the ERs were obtained from the immunohistochemical analysis of the renal tissue samples in the pathology inventory.Forty-two patients who were followed up in the rheumatology department of Ankara Bilkent City Hospital, who were diagnosed with pause-immune glomerulonephritis in renal biopsy, and whose clinical and laboratory findings were compatible with ANCA-associated glomerulonephritis, were included in the ANCA group.Eighteen patients with consistent clinical and laboratory findings and diagnosed with IgA nephropathy in renal biopsy were included in the IgA nephropathy group.In the control group, only normal renal tissues without tumor tissue (taken from a distance of at least >2 cm from the tumor tissue) from 30 patients who underwent nephrectomy due to renal cell neoplasm were included.These kidney tissues were included in the study after checking that they did not contain neoplastic cells.All ANCA-associated glomerulonephritis patients and patients with IgA nephropathy met the criteria of the International Chapel Hill Consensus Conference on the Nomenclature of Systemic Vasculitides. 28Patients were considered positive for the presence of any comorbid diseases such as hypertension (HT), chronic heart disease, chronic obstructive pulmonary disease (COPD), or diabetes mellitus (DM).
Patients in the IgA nephropathy group were drug-naive individuals who underwent renal biopsy due to persistent macroscopic or microscopic hematuria or proteinuria and were diagnosed histopathologically as IgA nephropathy.Patients in the ANCA-associated glomerulonephritis group consisted of drug-naive individuals who were admitted to the hospital with clinical symptoms such as acute renal failure, pulmonary symptoms, hearing loss, or pulmonorenal syndrome and were diagnosed with ANCA-associated glomerulonephritis by renal biopsy, imaging studies, and serological tests.The radiological and laboratory data used in the study were obtained from the examinations performed at the same time with the renal biopsy.Detection of more than five glomerular erythrocytes in each region at 10 × 40 magnification under the microscope in three consecutive centrifuge samples obtained at least seven days apart was considered positive for hematuria.Glomerular filtration rate (GFR) was determined using the Cockcroft-Gault formula.The data on systemic organ involvement of ANCA-associated glomerulonephritis and IgA nephropathy groups were obtained from the online hospital system or patient files.Dates are given in DD/MM/YYYY format.

| Analysis of tubule types by immunohistochemical method
Differentiation of tubule types was made under a light microscope with Hematoxylin Eosin (H&E) and periodic acid schiff (PAS) dyes.Histopathologically, the tubules observed in the cortex with columnar shape, with nuclei located basally or at the base, with brush-like borders and dark eosinophilic cytoplasm were considered as proximal tubules, while the others were considered as distal tubules.In the medulla, the tubules that do not have brushlike edges, have narrow cytoplasm, and nuclei protruding into the lumen, are partially lined with endothelium and resemble capillaries, are considered as loops of Henle, and the tubules outside this are considered as distal tubules. 30

| Statistical analysis
Statistical Packages for the Social Sciences (SPSS) version 22.0 package program was used to evaluate the statistical analyses in the study.Kolmogorov Smirnov test and Q-Q Plot, Box Plot, and histogram graphs were used to assess the normal distribution for continuous variables.Descriptive statistics were presented using the mean and standard deviation (mean ± SD) for normally distributed quantitative variables and the median (interquartile range [IQR], [25%-75%]) for non-normally distributed quantitative variables.Fisher's exact or chi-square tests were used to compare categorical data.Independent samples t-test was used for normally distributed variables.Mann-Whitney U test was used for non-normally distributed variables to determine statistically significant differences in pairwise comparisons between groups.In comparisons between multiple groups after Bonferroni correction, the One-way ANOVA post hoc Tukey test was used for normally distributed quantitative variables, and the Independent Samples Kruskal-Wallis test was used for non-normally distributed quantitative variables.Spearman correlation analysis was used to determine the correlation between continuous variables.Multinomial logistic regression analysis was used to evaluate the effect of independent variables on categorical variables.The impact of independent variables on continuous dependent variables was tested using binary or multivariate linear regression analysis.The accepted significance level for the p-value in pairwise comparisons was <.05, while in multiple comparisons, it was determined after Bonferroni correction.

| RESULTS
A comparison of demographic characteristics and laboratory data between groups is shown in Table 1.A total of 42 people, 23 women and 19 men, were included in the ANCA group, a total of 18 people, four women, and 14 men, were included in the IgA group, and a total of 28 people, 12 women and 16 men, were included in the control group.The median serum creatinine values were found to be considerably higher in the ANCA group (3.38 [1.73-4.67])than in the IgA nephropathy (1.42 [0.93-2.13])and control (0.84 [0.71-1.05])groups (p < .001).
A comparison of renal histopathological patterns and renal ER levels between the groups is shown in Table 2. Expression of ERα was not detected in both glomeruli and tubules in all study groups.While there was no significant difference between the groups in terms of glomerular ERß expression score (GERßES) median values (p = .498),tubular ERß expression score (TERßES) median values were found to be significantly higher in ANCA (1 [0-2.25]) and IgA nephropathy (2 [1-3.25])groups than in the control group (1 [0-1]) (p = .002).
In the multinomial logistic regression model created with renal histopathological patterns and the TERßES variable, the median TERßES values were found to  Immunohistochemical images of ERα and ERß expressions of breast carcinoma used as positive control are shown in Figures 1 and 2. Immunohistochemical images of ERα and ERß expressions of renal biopsy samples belonging to the study groups are shown in Figures 3-5.
The localization of tubular ERß expression in the renal tubular system is shown in Table 3.The proportion of subjects with no staining for TERß was 34%, and these were not included in the multiple comparisons based on the localization of TERß in the tubules.It was determined that tubular ERß was the most frequently localized in the distal tubule with 57.9% (40.9% + 17%), and the second most common localization was in the proximal tubule with 20.4% (3.4% + 17%).In comparison between tubules where tubular ERß was localized, the distal tubule was found to have significantly higher ERß expression than other localizations (p = .003).
The comparison of renal ERs according to the presence of some demographic and clinical conditions of the study groups is shown in Table 4. Renal ER expression levels were unaffected by gender, menopause, C-ANCA or P-ANCA positivity, and systemic organ involvement.
Spearman correlation analysis results between renal ERs and some variables are shown in Table 5.No significant correlation relationship was detected between renal ERs and other variables.
Spearman correlation analysis results between renal ERs and other renal histopathological findings belonging to ANCA and IgA nephropathy groups are shown in Table 6.No significant correlation relationship was detected between renal ERs and other variables.

| DISCUSSION
In this study, we aimed to investigate the roles of renal ERs in ANCA-associated glomerulonephritis and IgA nephropathy and to emphasize the possible pathophysiological effects of ERs in glomerulonephritis due to small vessel vasculitis.Our results showed that TERßES levels were higher in ANCA-associated glomerulonephritis and IgA nephropathy than in healthy kidney tissue, and there was no renal ER-α expression in all study groups.Additionally, this study found that tubular ERß was most frequently localized in the distal tubule (57.9%).
Epidemiologically, it is accepted that the typical onset for ANCA-associated vasculitides is between the ages of 45-75 and that the gender distribution is equal. 31For IgA nephropathy, it has been shown that the age of onset is between 31 and 52 years, with a 71% male predominance in adults. 32In our study, consistent with these epidemiological data, the average age in the ANCAassociated vasculitis group was found to be 61.2 years, and the female/male ratio was 1.21.The average age in the

T A B L E 6 Correlation analysis results
between GERßES and TERßES and other renal histopathological findings.
IgA nephropathy group was 37.7%, and 77.7% male predominance was detected.In addition, our study found no significant difference in renal GERßES and TERßES levels between men and women.Similar to these results, we showed in a recent study that there was no difference between genders regarding renal ER levels. 27Although these current data suggest that gender has no or minimal effect on the possible roles of renal ERs in the pathogenesis of small vessel vasculitis, further studies will contribute to a better understanding of this field.
In our study, no ERα expression was detected in all groups of glomeruli and tubules.However, we found the expression of ERß more intense in tubules and lower in glomeruli.Similar to our results, a human study reported that ERα expression was not detected in renal biopsy samples of patients with LN and IgA nephropathy.However, ERβ was expressed in the tubules but not the glomeruli. 25n experimental animal study showed that ERß was described but not ERα in epithelial cells of the bladder, renal pelvis, and ureter. 33Our previous study showed ERß expression was more intense in tubules and lower in glomeruli in LN, ANCA-associated vasculitis, FMF amyloidosis, and healthy controls.In this study, ERα expression was not detected in both the tubules and glomeruli of all groups. 27Unlike these results, studies investigating the expression of renal ERs in human diabetic glomerular mesangial cell culture 26 and mouse glomerular podocytes 34 showed the face of both ERα and ERβ.Still, these studies did not evaluate the presence of tubular renal ERs.In summary, studies provide conflicting results about renal ERα expression in humans but support the existence of renal ERβ expression.
Although studies show the association of ERs with some renal diseases, most of the data on the role of ERs in renal disease comes from experimental animal studies, and there are only a few human studies on the part of ERs in renal diseases.One human study showed that ERα and ERß were expressed in diabetic glomerular mesangial cell culture. 26Another study reported that tubular ERß expression is higher in IgA nephropathy than in LN. 25 In a previous study, we compared renal ERs between groups with LN, FMF amyloidosis, ANCAassociated vasculitis, and healthy renal tissue.We evaluated the relationship between renal ER expression levels and clinical conditions seen in these diseases.In this study, we demonstrated significantly higher TERßES levels in FMF amyloidosis than in other groups.Our results showed that renal TERßES and GERßES median values were higher in ANCA-associated vasculitis than in the group with healthy renal tissue.However, we did not detect a significant difference in the logistic regression analysis. 27Although the low sample number (n = 10) in the ANCA-associated vasculitis group in our previous study may have affected the accuracy of the results in the regression analysis, these results led us to evaluate renal ERs in ANCA-associated vasculitis with more extensive sample participation.No comparison has been made before in terms of renal ERs between intact kidney tissue and ANCA-associated glomerulonephritis and IgA nephropathy, which represent two different subtypes of small vessel vasculitis in humans, and the clinical significance of renal ERs in these groups is not well known.In this study, we did not detect a significant difference in renal ERs between ANCA-associated vasculitis and IgA nephropathy, two subtypes of small vessel vasculitis.However, our results showed that these two small vessel vasculitis subtypes had higher levels of TERßES than intact kidney tissue.There is no literature information other than our previous study evaluating the relationship between ANCA-associated vasculitis and renal ERs, 27 and the clinical importance of high TERßES in ANCA-associated vasculitis, which we showed in this study, needs to be clarified with further studies.
This study also showed that TERßES was significantly higher in IgA nephropathy than intact kidney tissue.Although no previous research compares IgA nephropathy with intact kidney biopsy specimens regarding renal ERs, a human study reported that TERßES is higher in IgA nephropathy than in LN. 25 Data on the roles of estrogen and ERs in IgA nephropathy mostly come from experimental animal studies.One study showed that mice with trichothecene vomitoxin (VT) induced IgA nephropathy had higher disease activity in the male sex. 35In addition, VT exacerbated IgA nephropathy in castrated female mice, but estrogen supplementation increased disease severity. 36Yamamoto et al. determined that polymorphisms in the ERα gene may contribute to the pathogenesis of IgA nephropathy. 37Mirfazeli et al. reported that ERα functions as a central protein by affecting some transcription factors and proteins that regulate mesangial cell proliferation, a histopathological pattern of IgA nephropathy. 38ome recent publications have reported that glomerular ERα expression decreases as the severity of IgA nephropathy increases.ERα is a critical factor that may affect the prognosis of IgA nephropathy, 39 and some responsible genes involved in IgA nephropathy are upregulated in the estrogen signaling pathway. 40Our study showed that IgA nephropathy has higher TERßES than healthy kidney tissue.Although this result supports the existing literature data, further studies are needed to elucidate the roles of estrogen and its receptors in the pathogenesis of IgA nephropathy.
We have previously shown that renal ERs are most frequently expressed in the distal tubule in the tubular system, with 84.1%, and that TERßES correlates with serum creatinine and the presence of crescent. 27Similar to our previous study, we found that TERßES was most frequently localized in the distal tubule with 57.9%.On the other hand, unlike our previous study, we did not detect a significant correlation between serum creatinine levels, the presence of renal crescent, and TERßES.The dissimilarity between the results of these two studies may come from the differences between the groups included in the study and the sample sizes of these groups or the small sample size (n = 10) in the ANCA group in our previous study.However, the physiological or pathological effects of renal ERs being most frequently expressed in the distal tubule need to be elucidated with further molecular studies.
Although it is known that pathological changes of vascular origin are primarily responsible for the development of ANCA-associated glomerulonephritis, 41 in our study, we detected significantly higher renal tubular ERß expressions, especially in the distal tubule, in the ANCA-associated glomerulonephritis group compared to the controls.Although the role of tubular renal ERs in the pathogenesis of renal diseases is not well known, some experimental studies have shown that estrogen may have pathogenic effects in LN through renal ERs. 9 These studies showed that estrogen administration to wildtype mice induced the development of LN by increasing the levels of immune system cells (CD4+/CD8+ T lymphocytes, macrophages, and dendritic cells) expressing ERα. 42Additionally, it has been reported that estrogen administration to lupus-prone mice increases the levels of anti-dsDNA antibodies, BAFF, anti-C1q, and nonimmunotolerant T and B lymphocytes, thus leading to the progression of LN. 43,44 Additionally, studies have found that environmental agents such as bisphenol-A and diethylstilbestrol, which have estrogen agonist activity, increase renal immune complex deposits by promoting autoantibody production in experimental animals prone to lupus. 45Some pathophysiological processes, such as abnormalities in B cell tolerance, complement-mediated kidney damage due to alternative pathway activation, T cell abnormalities and high B-cell activating factor (BAFF), which play a role in the development of LN, 46 also play a role in the development of ANCA-associated glomerulonephritis 31 suggests that renal ERs may have a possible contribution to the pathogenesis of ANCA-associated glomerulonephritis.These literature data support possible associations between estrogen and renal ERs and ANCAassociated glomerulonephritis, which has autoimmune processes similar to LN.
Recent electron microscopic studies showing that tubule-interstitial area involvement is not uncommon in ANCA-associated glomerulonephritis [47][48][49] may provide a possible pathophysiological explanation for the increased tubular ERß expressions in ANCA-associated vasculitis shown in our results.Although ANCAassociated vasculitis is a disease of vascular origin, 41 data obtained from electron microskope studies have shown that breaks occur in the capillary basement membrane due to the development of destruction in the type IV collagen of peritubular capillaries.Additionally, these studies report infiltration of mononuclear cells and neutrophils in the peritubular capillary lumen and surrounding interstitial tissues, and tubulitis develops in most cases.][49] Another pathophysiological feature of ANCAassociated glomerulonephritis is that, although it is well known that there is a strong relationship between increased tubular epidermal growth factor (EGF) levels and the progression of chronic renal failure and the development of interstitial fibrosis, 50 there are lower EGF levels in the tubules of ANCA-associated glomerulonephritis than in controls. 51These results suggest that non-EGF factors are responsible for the pathogenesis of tubulointerstitial damage in ANCA-associated vasculitis.In addition, in the data indexed in the nephr oseq.org data set, the authors, who investigated human chronic renal failure models in terms of tubular EGF levels, 51 also interestingly found lower ESR2 (the gene name of ERβ) expressions in the tubulointerstitial area of ANCArelated vasculitis compared to healthy controls.This data contradicts the high TERβES result we found in our study, especially in the distal tubule of ANCA-related vasculitis.However, in this data set, ESR2 expression was evaluated only in the tubulointerstitial area, and information about the tubular system localization of ESR2 was not shared.Further molecular studies that evaluate tubular ERβ levels and ESR2 gene expressions together in terms of renal tubular localizations will contribute to a better understanding of this issue.Although tubular or peritubular capillary localizations of ERs could not be evaluated by electron microscopy in our study, our results showed the presence of increased tubular ER ß expression levels in small vessel vasculitis compared to controls.This result indicates that renal tubular ERs should be investigated in further studies as potential candidates responsible for the pathogenesis of tubulitis and peritubular capillary damage seen in ANCAassociated vasculitis.
The limitations of this study include the failure to investigate how renal ERs change according to disease activity through repeated renal biopsies in small vessel vasculitis, the inability to evaluate tubular ER gene expressions and renal GPER levels, or the failure perform electron microscopic studies.Although we used neoplastic cell-free kidney tissues taken from tumor-free areas of kidneys with renal cell carcinoma as the control group in this study, recent publications have shown that there is a relationship between the ER receptor and the progression of renal cell carcinoma and may even increase the number of cancer stem cells. 23Another limitation of this study is that we used renal tissues from patients with renal cell carcinoma as a control group due to the difficulties in obtaining renal tissues from healthy people.However, this study showed that patients with ANCA-associated glomerulonephritis and IgA nephropathy, which are small vessel vasculitis, had higher TERßES than healthy controls, and tubular ERßs are most commonly localized in the distal tubule.

| CONCLUSION
Expression of tubular ERß is increased in small vessel vasculitis compared to healthy controls.Tubular ERßs are most commonly localized in the distal tubule.Further studies are needed to understand the physiological and pathophysiological effects of altered renal ER levels in small vessel vasculitis.

F I G U R E 3
In IgA nephropathy, weak positive staining for ERß in the glomeruli (short black arrow) and strong positive staining in the tubules (long red arrow) (×200).F I G U R E 4In ANCA-associated glomerulonephritis, weak positive staining for ERß in the glomeruli (short black arrow) and strong positive staining in the tubules (long red arrow) (×200).F I G U R E 5In the control group, weak positive staining (×200) for ERß in glomeruli (short black arrow) and tubules (long red arrow).

T A B L E 3 F I G U R E 2
Distribution of TERß expression in the renal tubule system.Nuclear staining (arrow) with estrogen receptor estrogen alpha (ERα) is observed in breast ductal adenocarcinoma cells used as a positive control in the sections (×200).
Comparison of demographic characteristics and laboratory data between groups.
Comparison of renal estrogen receptor expression scores according to the presence of some clinical conditions.Correlation analysis results between renal estrogen receptor expression scores and some study parameters.