|Home | About | Journals | Submit | Contact Us | Français|
We showed previously that 17β Estradiol (E2) led to improved survival in nephrotoxic serum induced nephritis (NTN) in male mice. In this study we determined whether E2 regulates vascular cell adhesion molecule (VCAM)-1, an adhesion molecule that is upregulated in kidney during autoimmune nephritis, in mesangial cells (MC). We show that E2 inhibited VCAM-1 up-regulation in kidneys in vivo during NTN, and in MCs upon TNFα stimulation. VCAM-1 up-regulation in MCs was controlled by the transcription factor NFκB. E2 inhibited RNA polymerase II recruitment to the VCAM-1 promoter, but not p65 recruitment. Interestingly E2 inhibited TNFα stimulated interaction between Poly (ADP-Ribose) Polymerase -1 (PARP-1) and p65. As PARP-1 is required for VCAM-1 upregulation in MCs, our data suggest that E2 may inhibit pre-initiation complex formation at VCAM-1 promoter by inhibiting PARP-1 recruitment to p65. We propose that E2 plays an important role in regulating renal inflammation locally.
The immuno-modulatory roles of estrogens, although studied extensively, are not completely understood. The higher prevalence of autoimmunity in females is often to linked to estrogens, on the other hand estrogens are proposed to protect from osteoporosis and kidney failure in diabetic females. The anti-inflammatory role of estrogens is well studied. 17β estradiol (E2) at periovulatory to pregnancy levels inhibited pro-inflammatory cytokines from human peripheral blood mononuclear cells. Similarly in mouse primary macrophages and macrophage cell lines E2 inhibited LPS stimulated TNFα and nitrite production [1; 2; 3] and decreased TNFα and IL-12 production in mature mouse dendritic cells (DC), with a shift toward cytokines such as IL-4 and IL-10 . These data support the anti-inflammatory and protective effects of E2. Contrary to these reports, several groups have suggested that signaling through estrogen receptor alpha (ERα) promotes an inflammatory pathway of DC development, in addition to regulating myeloid hematopoietic progenitor numbers . Evidence also suggests that E2 is able to stimulate antibody secretion [6; 7; 8], and therefore possibly contributes to the increased auto-antibody production in B cell driven autoimmune diseases such as systemic lupus erythematosus.
The role of estrogens in renal disease however, is not studied extensively. Pre-menopausal women have lower incidence of end stage renal disease (ESRD) in diabetic nephropathy, whereas post-menopausal women show an increase in ESRD. Male sex has been identified as a predictor of poor renal outcome [9; 10].
Estrogens reverse proteinuria and progressive renal injury in mice over-expressing TGFβ presumably by inhibiting TGFβ stimulated type IV collagen gene transcription [11; 12]. Estradiol also stimulates Matrix Metalloproteinase 2 (MMP2) secretion by mesangial cells , thereby preventing matrix accumulation and sclerosis in diabetic nephropathy.
A role for estrogens in modulating immune response during immune mediated renal disease has also been implied. Estrogens have been suggested to ameliorate hypertension-induced renal injury by reducing oxidative stress . Using an induced model of immune-mediated nephritis, we showed previously that supplementing male mice with E2 results in reduced severity of disease and improved survival . In the present study we further investigated the possible mechanism(s) by which E2 tames local inflammation and damage to renal tissue. To this extent we focused on the role of adhesion molecules on intrinsic renal cells.
Vascular cell adhesion molecule-1 (VCAM-1) is an adhesion molecule, which through its interaction with very late antigen-4 (VLA-4) present on leukocytes regulates recruitment of inflammatory cells. In a spontaneous mouse model of lupus, MRL/lpr, VCAM-1 expression was increased in the endothelium, glomeruli and tubules . Increased VCAM-1 gene expression was also observed in microarray analysis of glomeruli from nephritic MRL/lpr mice when compared to control MRL/MpJ . Soluble VCAM-1 (sVCAM-1) levels correlated with disease activity, and further more in patients with renal, hematological and vascular activity, significant differences in sVCAM-1 levels were reported [18; 19]. Urinary VCAM-1 levels were elevated in class IV LN . Finally blocking VLA-4/VCAM-1 interactions ameliorated crescentic nephritis in rats .
In this study we show that estrogens inhibit VCAM-1 upregulation in vivo during nephritis and in vitro in mesangial cells upon TNFα stimulation. This upregulation was mediated through activation of transcription factor NFκB. Moreover, E2 impaired NFκB activation by inhibiting the formation of pre-initiation complex (PIC) at the VCAM-1 promoter. Our data suggest that E2 may block PIC formation by inhibiting the interaction between p65 and co-factor Poly (ADP-Ribose) Polymerase-1 (PARP-1). We therefore propose that during renal inflammation estrogens are protective locally.
Mesangial cells  were maintained in DMEM with 10% fetal bovine serum (FBS). To determine the effect of estrogens, cells were cultured in phenol red free medium with charcoal/dextran treated FBS (Hyclone) for 72h. Cells were then treated with 17β-estradiol (E2 10−5 – 10−11 M) for 48h prior to TNFα stimulation. 10−7 and 10−5 M E2 were used to determine effects on TNFα stimulated VCAM-1 mRNA levels, with similar results. For PARP-1 inhibition, cells were pretreated with the inhibitors 3-aminobenzamide (3AB, Sigma, 0.01–1mM) or CEP 8983 (CEP, Cephalon/Teva, 1–10 μM) 30 min prior to TNFα stimulation. For NFκB inhibition, cells were pretreated with selective inhibitor of IκB kinase, IKK16 (Tocris Biosciences, 1μM) 30min prior to TNFα (1ng/ml) stimulation.
129sv mice were obtained from Jackson Laboratories. Breeding colonies were maintained at Temple University in accordance with the guidelines of the University Laboratory Animal Resource Office of Temple University. All experimental procedures were conducted according to the guidelines of the Institutional Animal Care and Use Committee. NTN was induced as described previously by a single injection of NTS . For estrogen treatment, 17β-estradiol (E2) pellets (Innovative Research of America) were implanted s.c. NTS (6mg/ml) was injected 5 days later. 30h following NTS injection, mice were euthanized, kidneys perfused, and frozen in OCT.
Mesangial cells were plated on chambered coverglass and stimulated with TNFα for various time points. Cells were fixed with 4% paraformaldehyde, permeabilized with saponin, blocked with 5% goat serum and incubated overnight with anti-p65 polyclonal antibody (Cell signaling). The cells were stained with Rhodamine conjugated goat anti-rabbit antibodies (Life Technologies) and DAPI and imaged using Zeiss META confocal microscope equipped with ZEN software. The data were analyzed by ImageJ software. Cells with nuclear p65 were counted with the help of ‘counter’ plug-in. A minimum 6 fields/condition were analyzed.
Frozen sections were fixed with acetone and stained with anti-VCAM-1 (BD biosciences, clone 429), followed by FITC conjugated anti-mouse IgG and DAPI for nuclear staining. The sections were mounted in Prolong gold (Life Technologies) and imaged using Zeiss META confocal microscope equipped with ZEN software. The data were analyzed by ImageJ software. The channels were separated (blue and green). The mean fluorescence intensity (MFI) was calculated for each field in the green channel. The MFI for the sections with secondary antibody only were used for background calculations and were subtracted from the MFI values obtained from the sections stained with primary antibody. A minimum of 5 fields for each section/mouse were analyzed.
For surface staining of VCAM-1, after stimulation, cells were incubated with biotin conjugated anti-VCAM-1 antibodies (eBioscience), followed by streptavidin conjugated APC (BD Biosciences), fixed in 1% paraformaldehyde and acquired on BD FACS Canto. 50,000 total events were acquired. For some experiments, cells were additionally stained with Annexin V FITC and 7AAD, and VCAM-1 up-regulation was measured in cells negative for both Annexin V and 7AAD (live cells). For intraceullular PARP-1 staining, the cells were first stained for surface VCAM-1 as above. After fixation, the cells were permeabilized with saponin, blocked with 5% goat serum and incubated with either anti-PARP-1 monoclonal antibody (BD Biosciences) or mouse IgG1 isotype antibody (Biolegend). Cells were washed, stained with FITC conjugated goat anti-mouse antibody (Jackson Immunoresearch) and acquired on FACS Canto. For intracellular Poly ADP-Ribose Polymer (PAR) staining we followed previously published method with some modifications . Briefly, cells were fixed and permeabilized first with 70% ethanol and then with 0.1% Triton X 100. After blocking with 5% goat serum, cells were incubated with anti-PAR monoclonal antibody (Trevigen) or isotype IgG3 antibody (Biolegend) overnight at 4°C. Cells were washed, stained with anti-mouse FITC conjugated secondary antibodies and acquired on BD FACS Canto. The data were analyzed using FlowJo software.
Following stimulation, chromatin was crosslinked by treating cell with 1% paraformaldehyde, and crosslinking was quenched with 0.125% glycine. Cells were then washed with PBS and lysed with 1% ChIP lysis buffer (1% SDS, 10mM EDTA, 50mM Tris) for 15 min on ice. The chromatin was sheared by sonication. Cleared lysates were diluted with ChIP dilution buffer (0.01% SDS, 1.2mM EDTA, 16.7mM Tris-HCl pH 8, 1.1% Triton X 100, 167mM NaCl) and incubated with either no antibody or rabbit anti-p65 (Millipore) or rabbit anti-Polymerase II (Millipore) antibodies overnight at 4°C. The lysates were then incubated with Protein A/DNA slurry (Millipore) for 2h. The beads were washed and immune-complexes were eluted using elution buffer (1% SDS, 0.1 M NaHCO3). Crosslinks were reversed by incubating overnight at 65°C in presence of 200mM NaCl. The eluates were treated with Proteinase K (Sigma) for 1h. DNA was extracted using Qiagen PCR clean up kit and stored at -20°C. A SYBR green-based assay was used for quantitative PCR (qPCR). Primers from a region of mouse VCAM-1 promoter upstream of NFκB binding site were used for qPCR. Primer sequences used were: Forward: 5′-AGTGTCGTGTTTCCCAGGAC-3′, Reverse: 5′-GCCAGGGAAAAAGTTTAACTGA-3′. The Ct values were normalized to input and the fold changes were calculated by ΔΔCt method.
For immunoprecipitations, nuclear proteins were enriched by first lysing cells in hypotonic lysis buffer (10mM Hepes, pH 7.9, 1mM EDTA, 10mM KCl, 0.5%NP40, protease inhibitor cocktail). The pellets were then lysed in high salt buffer (20mM Hepes, pH 7.9, 0.5M NaCl, 1mM EDTA, 1mM EGTA, 1mM DTT, 0.5%NP40) by rigorous vortexing and rotating for 20min at 4°C. The lysates were cleared and diluted with immunoprecipitation wash buffer (10mM Tris, 100mM NaCl, 0.5% NP40), and incubated overnight with Protein A/G agarose beads covalently linked to p65 antibodies. The covalent linking of antibodies to agarose beads was performed with dimethyl pimelimidate. The beads were washed with wash buffer and protein complexes eluted by boiling in presence of 2X Laemmli buffer. The proteins were separated on a 10% SDS PAGE and transferred to nitrocellulose membrane. Anti-PARP-1 monoclonal and rabbit anti-p65 (Cell Signaling technology) antibodies were used to detect proteins on membranes. To determine nuclear translocation of p65, equal amounts of protein from nuclear and cytoplasmic extracts were separated on 10% SDS PAGE and transferred to nitrocellulose membrane. Anti-LaminA/C (Cell Signaling technology) and anti-GAPDH (Thermofisher-pierce) were used as loading controls for nuclear and cytoplasmic extracts respectively.
Total RNA was isolated using Qiagen RNeasy kit. RNA (400ng) was reverse transcribed using Applied Biosystems high capacity reverse transcription kit. VCAM-1, normalized to GAPDH, was detected by real-time PCR using QuantiTect primer assays (Qiagen) and Applied Biosystems Step2 plus thermal cycler. Relative gene expression was calculated by ΔΔCt method.
ANOVA and t-tests were performed using GraphPad Prism 4.0c software for Mac (GraphPad). Differences among groups were determined using the Tukey-Kramer post hoc test. Statistical significance was defined as p < 0.05.
To determine whether estrogens protect mice from nephritis by down-regulating of adhesion molecules, we investigated the effect of E2 on VCAM-1 expression in vivo. Mice were treated with E2 by implanting E2 pellets s.c., as we described previously [15; 23]. Direct renal injury was induced by injecting nephrotoxic serum 5 days after E2 implantation [15; 23] and kidneys were harvested 30h later. We observed that E2 treatment inhibited VCAM-1 up-regulation in kidneys in vivo during nephrotoxic serum induced nephritis (NTN) in both male and female mice (Figure 1A, B). To determine whether the reduction in VCAM-1 expression in the kidneys was due to inhibition of VCAM-1 up-regulation in mesangial cells, we determined the direct effect of E2 on TNFα stimulated VCAM-1 up-regulation in mouse mesangial cells. Our data show that E2 pre-treatment of mesangial cells inhibited TNFα stimulated VCAM-1 at both protein and mRNA levels (Figure 1C, D).
NFκB regulates VCAM-1 transcription in endothelial cells and is widely accepted as a transcription factor that regulates VCAM-1 [24; 25; 26; 27]. However, the role of NFκB in regulating VCAM-1 expression in mesangial cells has not been previously reported. We therefore determined whether NFκB is responsible for VCAM-1 up-regulation in mesangial cells as well. To this end we used an inhibitor of IκB kinase activity to inhibit NFκB activation. Figure 2 shows that inhibition of IKK inhibited VCAM-1 upregulation and therefore our data show that NFκB regulates TNFα stimulated VCAM-1 up-regulation in mesangial cells.
We then determined the molecular mechanism of regulation of VCAM-1 by E2. Inhibition of NFκB activation and inhibition of transcription of genes regulated by NFκB can be inhibited at different stages: 1) IκB kinase (IKK) activation and/or IκB degradation, either of which would inhibit nuclear translocation of p65/p50 heterodimer, 2) p65 binding to the promoter region, or 3) recruitment of co-activators at the promoter site. We first determined whether E2 inhibits IKK activation or IκB degradation by measuring nuclear p65 upon TNFα stimulation. Figure 3 shows that there is peak of nuclear p65 within 30 minutes following TNFα stimulation. However, the cells that were pretreated with E2 did not show any impairment in p65 nuclear translocation. These data suggest that estrogens do not regulate NFκB activation by inhibiting either IKK activation or IκB degradation. To further determine whether estrogens inhibit p65 DNA binding, we performed ChIP assays. Figure 4A shows that, as expected, there was an increase in p65 binding to VCAM-1 promoter upon TNFα stimulation. E2 pretreatment, however, did not inhibit p65 binding to the VCAM-1 promoter. Surprisingly we saw that E2 inhibited RNA polymerase II recruitment to the VCAM-1 promoter (Figure 4B), suggesting that E2 may inhibit the recruitment of co-activators and the formation of pre-initiation complex (PIC) at the VCAM-1 promoter.
We showed previously that absence of Poly (ADP-Ribose) Polymerase-1 (PARP-1) inhibited TNFα stimulated VCAM-1 up-regulation in mouse mesangial cells . PARP-1 has been shown to interact with estrogen receptor  and has been also proposed as a co-factor for NFκB activation . We therefore determined whether E2 regulates VCAM-1 up-regulation through PARP-1. Figure 4C shows that PARP-1 interacts with p65 upon TNFα stimulation and this interaction is inhibited in the presence of E2. These data suggest that inhibition of PARP-1 recruitment to VCAM-1 promoter by E2 may be responsible for inhibition of recruitment of RNA Polymerase II to the promoter and therefore, inhibition of VCAM-1 transcription.
PARP-1 is known to interact with transcription factors through direct interaction or through Poly (ADP-Ribose) Polymers (PARs) generated as a result of PARP-1 activation. We showed previously that E2 inhibits PARP-1 activity in bone marrow derived macrophages . We therefore determined whether PARP-1 activity regulates VCAM-1 up-regulation stimulated by TNFα. Figure 5A–D show that pretreatment with PARP-1 inhibitors did not inhibit VCAM-1 up-regulation. We then determined whether TNFα stimulation activates PARP-1 in mesangial cells. PARP-1 activity was measured as PAR accumulation by flow cytometry. Figure 5E shows that although E2 inhibited basal PARP-1 activity, TNFα did not induce PARP activity, further supporting our data that PARP-1 activity is not essential for TNFα stimulated VCAM-1 upregulation in mesangial cells.
Another mechanism by which E2 can inhibit PARP-1/p65 interaction is by lowering the levels of PARP-1 protein through transcriptional regulation, thereby making PARP-1 unavailable to bind to p65. We therefore determined intracellular PARP-1 levels after stimulation with TNFα in cells pretreated with E2. Figure 6A and 6B show that although E2 inhibited VCAM-1 upregulation, E2 or TNFα treatment did not affect PARP-1 levels in mesangial cells.
There is increasing evidence that estrogens regulate the functional responses of immune cells and thereby influence the immune response. We therefore sought to determine whether estrogens may also regulate the immune response in the kidney during renal inflammation. In addition to the established role of estrogens in reducing renal sclerosis, our data also support the notion that estrogens regulate the inflammatory response in the kidney, and therefore they exert a protective effect on renal injury through amelioration of inflammation.
Our data show that estrogens inhibit up-regulation of the adhesion molecule VCAM-1 in nephritic kidneys. Using mesangial cells we further show that estrogens inhibit VCAM-1 up-regulation by inhibiting the formation of pre-initiation complex at the VCAM-1 promoter, thereby inhibiting the recruitment of RNA polymerase II and transcription at the VCAM-1 promoter. Our data suggest that the inhibition of RNA Polymerase II recruitment may be due the inhibition of PARP-1/p65 interaction by estrogens (Figure 7). We propose that estrogens play an important role in regulating renal inflammation locally.
Estrogens were shown to inhibit Matrix Metalloproteinases 2, MMP2, which possibly inhibits accumulation of extracellular matrix and reduce sclerosis . Estrogens also inhibit TGFβ-dependent collagen IV synthesis in mesangial cells by limiting the availability of the transcription factor Sp1. This inhibition by estrogens is mediated through down-regulation of Casein Kinase 2 (CK2) . However, the role of estrogens in regulating immune response in renal inflammation by influencing the activation of intrinsic renal cells was not studied previously. Mesangial cells that constitute the mesangium are embedded in extracellular matrix and have a unique location in the glomerulus where they can communicate with both the vasculature and the interstitium. Due to their presence in this unique environment, apart from the contractile function of mesangial cells, these cells can play a major role in facilitating the infiltration of cells from circulation. Indeed, mesangial cells have been proposed to regulate neutrophil and monocyte infiltration in the kidney by up-regulating adhesion molecules [30; 31]. The role of estrogens in regulating renal inflammatory cell infiltration has not been studied previously. We show that estrogens inhibit VCAM-1 upregualtion in the kidney, and therefore may ameliorate renal inflammation. These data are in line with our previous published report that showed protection of mice from nephritis by estrogen treatment .
The role of estrogens in regulating NFκB activation has been studied previously. Estrogens inhibit IKK and reduce IκBa phosphorylation in HeLa cells [32; 33]. In vitro studies show that estrogen-induced activation of ER inhibits expression of IL6 in osteoblasts and thus underscore the importance of estrogens in osteoporosis . On similar lines, ER was suggested to inhibit NFκB DNA binding in several cell types namely breast cancer cell line MCF7 , human umbilical vein endothelial cells  and vascular smooth muscle cells . However, these in vitro studies did not differentiate between ability of estrogens to inhibit NFκB nuclear translocation and to inhibit DNA binding. To overcome this problem, we employed a different strategy. We determined the nuclear translocation of p65 and also determined the ability of NFκB to bind VCAM-1 promoter in vivo in the presence of estrogens using chromatin immunoprecipitation. However, we found that in mesangial cells, estrogens did not inhibit p65 nuclear translocation or binding to VCAM-1 promoter. We showed previously that the absence of Poly (ADP-Ribose) Polymerase-1 (PARP-1) inhibited the up-regulation of VCAM-1 in mesangial cells. We now show that estrogens inhibit the interaction of PARP-1 with p65-subunit of NFκB, the transcription factor that regulates VCAM-1 up-regulation in mesangial cells. These data suggest that estrogens inhibit VCAM-1 up-regulation in mesangial cells by inhibiting p65 and PARP-1 interaction (Figure 7).
PARP-1 dependent NFκB activation was shown to be dependent on PARylation of p65-binding proteins in Trypanosoma cruzi infected cardiomyocytes, and inhibition of PARP-1 by phamacological inhibitors inhibited cytokine gene expression in infected cardiomyocytes. Our results show that in mesangial cells, PARP-1 was not activated upon TNFα stimulation. However, as we published previously in bone marrow derived macrophages , estrogens inhibited basal PARP-1 activity. Similarly, PARP-1 inhibitors did not inhibit VCAM-1 up-regulation. These data suggest that in mesangial cells, although PARP-1 interacts with p65 upon TNFα stimulation, PARP-1 activity is not required for NFκB activation. Our data, however, are in line with previous report that enzymatic activity and the DNA binding activity of PARP-1 are not required for NFκB activation in mouse fibroblasts . We also show that E2 does not affect the levels of PARP-1 protein in the cells therefore excluding the possibility that inhibition of PARP-1 interaction with p65 is due to unavailability of PARP-1. A possible reason for this inhibition of interaction may be due to sequestration of PARP-1 by E2/ER complex . It remains to be determined how inhibition of PARP-1/p65 interaction inhibits RNA polymerase II recruitment. Other post-translation modifications of PARP-1 that may facilitate its binding to p65 that are disrupted by estrogens may be involved and understanding those modifications require further studies.
Our data show that estrogens regulate VCAM-1 up-regulation in the kidney during renal inflammation by interfering with NFκB signaling. Our data also suggest that PARP-1 may be the co-activator that is responsible for NFκB activation required for VCAM-1 up-regulation in mesangial cells. Based on our data and published reports we propose that despite of the known autoimmune enhancing effects of estrogens in B cell driven autoimmune diseases, estrogens are protective locally in the kidney. The protection in the kidney is through 1) reducing the inflammatory response, and 2) reducing sclerosis thereby inhibiting and possibly delaying end stage renal disease. These data might explain why autoimmune renal disease is worse in males and could lead to a sex-based anti-inflammatory therapy.
This work was supported by NIH/NIAMS R01AR061569 to RC
Conflict of interest statement
The author(s) declare that there are no conflicts of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.