Autoimmune activation and hypersensitization of the AT₁ and ET_A receptors contributes to vascular injury in scleroderma renal crisis

Abstract

Objectives

Scleroderma renal crisis (SRC) is a rare vascular complication of systemic sclerosis with substantial risks for end-stage renal disease and premature death. Activating autoantibodies (Abs) targeting the angiotensin II type 1 (AT₁R) and the endothelin-1 type A receptor (ET_AR) have been identified as predictors for SRC. Here, we sought to determine their pathogenic significance for acute renal vascular injury potentially triggering kidney failure and malignant hypertension.

Methods

IgG from patients with SRC was studied for AT₁R and ET_AR dependent biologic effects on isolated rat renal interlobar arteries and vascular cells including contraction, signalling and mechanisms of receptor activation.

Results

In myography experiments, patient IgG exerted vasoconstriction sensitive to inhibition of AT₁R and ET_AR. This relied on MEK-ERK signalling indicating functional relevance of anti-AT₁R and anti-ET_AR Abs. The contractile response to angiotensin II and endothelin-1 was amplified by patient IgG containing anti-AT₁R and anti-ET_AR Abs with substantial crosstalk between both receptors implicating autoimmune receptor hypersensitization. Co-immunoprecipitation experiments indicated heterodimerization between both receptor types which may enable the observed functional interrelation by direct structural interactions.

Conclusion

We provide experimental evidence that agonistic Abs may contribute to SRC. This effect is presumably related to direct receptor stimulation and additional allosteric effects, at least in heterodimeric receptor constellations. Novel therapies targeted at autoimmune hyperactivation of AT₁R and ET_AR might improve outcomes in severe cases of SRC.

Open in new tabDownload slide

Introduction

SSc is a rheumatic disease of yet unknown pathogenesis characterized by autoantibodies directed against various cellular targets, fibrosis of the skin and internal organs, and severe obliterative vasculopathy [1, 2].

A potentially life-threatening vascular complication is scleroderma renal crisis (SRC) occurring in 4.9% of patients [3]. An acute decline of renal function is the hallmark of SRC. Hypertension occurs in about 90% of cases [4], often with blood pressure-related end-organ damage such as hypertensive retinopathy, encephalopathy with seizures and thrombotic microangiopathy (TMA) with haemolytic anaemia (MAHA) and thrombocytopenia [5]. In kidney biopsies from patients with SRC, acute and chronic vascular lesions are predominantly found in small vessels rather than in glomeruli [6]. Early findings are endothelial swelling, thrombosis and fibrinoid necrosis while chronic changes include onion-skin lesions, fibrointimal sclerosis and glomerulosclerosis [6]. Recurring or continuous endothelial injury in the arcuate and interlobular arteries results in intimal thickening and progressive luminal narrowing leading to decreased cortical blood flow [4]. This triggers renin production and activation of the renin-angiotensin-aldosterone system (RAAS). Expression of ET-1 [7] and its receptors [8] is enhanced and systemic ET-1 levels are elevated [9], whereas endothelial synthesis of prostacyclin is decreased [4]. This imbalance of vasoconstrictive and vasodilative factors culminates in endothelial damage and SRC.

Introduction of angiotensin-converting enzyme inhibitors (ACEI) in the late 1970s reduced the very high 1-year mortality rate from 76% to <15% and restored kidney function in >50% of patients initially requiring dialysis treatment in a US single-centre cohort [10]. However, despite routine use of ACEI in SRC, patient survival is still limited to 50–60% after 5 years [11] and a history of SRC is an independent predictor of 3-year mortality in patients with SSc [12]. Furthermore, 25–50% of SRC patients progress to end-stage renal disease [12].

Our group has described agonistic autoantibodies (Abs) directed against AT₁R and ET_AR in patients with SSc. High Abs levels were associated with an increased risk for vascular complications including SRC, pulmonary arterial hypertension and SSc-related death [13, 14]. Anti-AT₁R and anti-ET_AR Abs can activate microvascular endothelial cells [14], stimulate neutrophil migration [15], diminish endothelial wound repair [15] and have been attributed a pathogenic role in SSc-related pulmonary vasculopathy [13]. Recently, we demonstrated that these antibodies activate an Extracellular signal-Regulated Kinases 1/2 (ERK1/2)-Erythroblastosis virus E26 homologue transcription factor-1 (Ets-1)-Tissue Factor (TF) signalling cascade leading to endothelial cell proliferation [16].

Here, we hypothesized that agonistic anti-AT1R and anti-ETAR Abs found in patients with SSc contribute to the development of SRC not only by activation of endothelial cells but also by increasing contractility of renal interlobar arteries. Enhanced responsiveness to the natural ligands AngII and ET-1 due to Abs effects as well as synergistic responses were explored.

Patients and methods

Extended methods are available in the supplementary material available at Rheumatology online.

Isolation of human IgG

Endotoxin-free IgG was isolated from SSc patients with SRC and healthy controls as described previously [17]. The institutional review board of Charité Universitätsmedizin Berlin approved the study and all participants provided written informed consent.

Measurement of AT₁R and ET_AR abs levels

Levels of anti-AT₁R and anti-ET_AR Abs in IgG preparations were measured with a solid-phase sandwich ELISA using membrane extracts from Chinese Hamster Ovary cells overexpressing human AT₁R or ET_AR in their native conformation (CellTrend GmbH, Luckenwalde, Germany). Conformational epitopes were maintained by addition of 1 mM calcium.

Small resistance artery myography

All animal studies were conducted in accordance with local animal care guidelines. Arteries were harvested as previously described [18]. Ring segments from third to fourth-generation interlobar renal arteries were isolated from male Lewis rats and were mounted on a small vessel wire myograph (Multimyograph 610, Danish Myo Technology, Aarhus, Denmark). Contraction was normalized to the maximal response induced by 60 mM KCl set to 100%. Numbers of individual artery rings tested in each experiment are indicated in the figure legends.

Cell culture and Western blot analysis

Vascular smooth muscle cells (VSMC) from rat aorta, human microvascular endothelial cells (HMEC-1) and human umbilical vein endothelial cells (HUVEC) were cultured as described previously [19]. Cells were stimulated with IgG from healthy controls (Control IgG) or SRC patients (SRC IgG) for 15 min with or without pre-treatment with the AT₁R blocker (AT₁RB) valsartan or the ET_A/BR blocker (ET_A/BRB) bosentan at 10 µM for 60 min. Protein extraction, SDS–PAGE, transfer and membrane treatment were performed following standard procedures [20].

Luciferase reporter assay

HMEC-1 were transfected with human wild type AT₁R and wild type ET_AR cloned into pcDNA3 and a serum response element luciferase reporter plasmid (Promega, Madison, WI, USA). Cells were incubated with indicated combinations of Control IgG or SRC IgG (1.0 mg/ml), AngII (1000 nM) or ET-1 (100 nM), and bosentan or valsartan (10 µM). Luciferase production was measured with the Luciferase Assay System (Promega).

Immunoprecipitation

Three different clonal lines of transfected human embryonic kidney 293 cells (HEK293) stably overexpressing Flag-tagged ET_AR and His-tagged AT₁R (clones 7, 22 and 32) were generated. Cell lysates were incubated with either an anti-AT₁R antibody (Santa Cruz Biotechnology, Dallas, TX, USA) or isotype control (Sigma-Aldrich, Saint Louis, Missouri, USA) coupled to agarose beads overnight at 4°C. Pelleted beads were then analysed by western blotting using anti-ET_AR (BD Biosciences, Franklin Lakes, NJ, USA) or anti-FLAG antibodies (Sigma-Aldrich).

Statistical analysis

Results are expressed as means (s.e.m.). Multiple groups were compared using one-way or two-way analysis of variance (ANOVA). Post-testing was performed with Bonferroni’s multiple comparisons test. Correlations were calculated with the Pearson correlation coefficient. A two-sided P-value <0.05 was considered statistically significant.

Results

IgG from patients with SRC induces contraction in renal interlobar arteries via AT₁R and ET_AR

To test our hypothesis that agonistic Abs targeting the G-protein coupled receptors (GPCR) with vasoconstrictive activity AT₁R and ET_AR are pathophysiologically relevant in SRC, we studied contractile responses of small renal resistance arteries ex vivo. mRNA transcripts for AT₁R and ET_AR were more abundant in rat renal interlobar arteries than in total kidney extracts (Supplementary Fig. S1, available at Rheumatology online). Expression levels in interlobar arteries were comparable to those found in rat VSMC (Supplementary Fig. S1, available at Rheumatology online).

IgG purified from SRC patients who were ELISA-positive for anti-AT₁R and anti-ET_AR Abs (SRC IgG) was compared with IgG from healthy controls (Control IgG). SRC IgG dose-dependently induced contraction of renal interlobar arteries whereas Control IgG did not (Fig. 1A).

Pre-treatment of the artery segments with the AT₁RB valsartan partly abolished the contractile response to SRC IgG (Fig. 1B). Both the ET_AR blocker (ET_ARB) sitaxsentan and the dual ET_A/BRB bosentan completely prevented SRC IgG induced vasoconstriction (Fig. 1B). Contractile responses of the artery rings were positively correlated to the concentrations of anti-AT₁R and anti-ET_AR Abs in the individual IgG preparations from SRC patients (r = 0.596, P < 0.001 for anti-AT₁R Abs, r = 0.790, P = 0.002 for anti-ET_AR Abs; Fig. 1C and D). These results indicate that contraction of renal resistance arteries induced by anti-AT₁R and anti-ET_AR Abs-positive IgG from SRC patients is mediated by AT₁R and ET_AR.

Autoantibody-induced vasoconstriction involves AT₁R and ET_A/BR-dependent activation of the MEK-ERK pathway

Since activation of ERK1/2 by agonistic anti-AT₁R and anti-ET_AR Abs from SSc patients has been described in HMEC-1 [14, 16] and this pathway has been implicated in AngII and ET-1 induced arterial vasoconstriction [21], we analysed its relevance for autoimmune modulation of renal vasoconstriction in SRC. In rat VSMC, we observed a 2–3-fold increase in ERK1/2 phosphorylation in response to SRC IgG compared with Control IgG (Fig. 2A and B). Inhibition of AT₁R or ET_A/BR prevented IgG-induced ERK1/2 phosphorylation (Fig. 2A and B). Similar effects were also found in HUVEC (Supplementary Fig. S2, available at Rheumatology online).

To test ERK1/2 activation mediated by anti-AT₁R and anti-ET_AR Abs containing patient IgG translates into contraction of renal interlobar arteries, we pre-treated the artery segments with an inhibitor of the MEK-ERK pathway. The contractile response to SRC IgG was diminished by 50% (Fig. 2C). Thus, the MEK-ERK pathway constitutes a functional link between autoimmune AT₁R and ET_AR stimulation and renal vasoconstriction in SRC.

Anti-AT₁R and anti-ET_AR abs hypersensitize renal interlobar arteries to AngII and ET-1

AngII and ET-1 have been ascribed crucial roles in the pathophysiology of SRC [4, 11]. To explore putative functional interactions with the agonistic Abs, we tested the response of renal interlobar artery rings to AngII and ET-1 alone and in combination with SRC IgG. Both AngII and ET-1 elicited a dose-dependent contractile response that was unaltered when the vessels were pre-incubated with Control IgG (Fig. 3A and B). However, exposure to agonistic Abs containing SRC IgG enhanced the response to AngII by 48–85% (Fig. 3A) and to ET-1 by 28–83% (Fig. 3B). Because the responses to the natural ligands in combination with the autoantibodies were supra-additive in comparison to separate stimulation, this indicated hypersensitization of both receptors to their natural ligands by binding of anti-AT₁R and anti-ET_AR Abs.

Inhibition of AT₁R abolished AngII induced vasoconstriction (Fig. 3A). Similarly, blockade of ET_AR alone or ET_A/BR completely precluded contraction upon exposure to ET-1 (Fig. 3B).

Crosstalk between AT₁R and ET_AR in response to anti-AT₁R and anti-ET_AR abs and natural ligands

In order to decipher the synergy between autoimmunity (anti-AT₁R and anti-ET_AR Abs) and the natural vasoconstrictors as a novel concept for SRC pathogenesis, we evaluated the response to SRC IgG in crossover blocking experiments where one receptor was stimulated with the natural agonist while the other receptor was pharmacologically inhibited.

When vessels were pre-treated with ET_A/BRB prior to stimulation with SRC IgG in combination with AngII, the total contractile response was reduced by almost 60% compared with treatment with SRC IgG and AngII without receptor blockade (Fig. 4A). Vice versa, pre-treatment with AT₁RB reduced the contraction induced by SRC IgG in combination with ET-1 to a similar amount (Fig. 4B). Remarkably, the observed crossover blocking effects exceeded the extent that could be attributed to stimulation of ET_AR (Fig. 4A) or AT₁R (Fig. 4B) by SRC IgG containing agonistic Abs. This observation was most prominent in the experiment with ET-1 and AT₁RB: AT₁RB reduced the maximal contraction elicited by SRC IgG together with ET-1 by 113.9 percentage points, whereas SRC IgG induced an additional contraction of only 5.9 percentage points over control IgG (Fig. 4B). With control IgG, crossover receptor blockade had no effect on vessel contraction (Supplementary Fig. S3, available at Rheumatology online). This indicates that inhibition of one of the receptors that is activated only by the agonistic Abs (e.g. AT₁R) also impairs activation of the other receptor type (e.g. ET_AR) that is stimulated by the agonistic Abs in combination with the natural ligand (e.g. ET-1). This finding clearly indicates a crosstalk between both receptors in activity regulation by ligand interaction.

To identify the molecular pathway operative in the observed interplay of both receptors, we studied ERK1/2 activity with a luciferase assay under one-sided crossover receptor inhibition and simultaneous stimulation with SRC IgG in combination with AngII or ET-1. These experiments support that ERK1/2 activation elicited by SRC IgG in synergy with one of the natural agonists can be reduced by inhibition of the other receptor that is not occupied by a natural ligand (Supplementary Fig. S4A and B, available at Rheumatology online). On the other hand, combined stimulation with both natural ligands was less than additive (Supplementary Fig. S4C, available at Rheumatology online). Additional blockade of one of the two receptors resulted in ERK1/2 activity of the same magnitude as stimulation with the unblocked natural ligand alone (Supplementary Fig. S4C, available at Rheumatology online). Thus, functional crosstalk between AT₁R and ET_AR was evident with SRC IgG containing anti-AT₁R and anti-ET_AR Abs in combination with AngII and ET-1 but not with the natural ligands alone.

This interdependency of AT₁R and ET_AR activation in the presence of agonistic Abs and natural ligands implies a direct contact between the two receptors, known as heterodimeric arrangement [18]. We conducted co-immunoprecipitation studies to detect putative heterodimers formed by AT₁R and ET_AR. Total lysates from HEK293 cells transfected with His-tagged AT₁R and FLAG-tagged ET_AR were precipitated with anti-AT₁R-antibodies or isotype control. Results from three independently generated clones selected for robust expression of His-tagged AT₁R and FLAG-tagged ET_AR are shown to exclude clone dependent effects. Staining of immunoblots from the precipitates with anti-ET_AR or anti-FLAG antibodies revealed a distinct band at 52 kDa consistent with ET_AR that was co-precipitated with anti-AT₁R-antibodies but not isotype control (Fig. 5A and B). This strongly supports a close spatial relation of both receptors and suggests the formation of heterodimers in line with the observed functional reciprocal dependency between AT₁R and ET_AR.

Discussion

Based on previous reports implicating agonistic autoantibodies targeting AT₁R and ET_AR in the pathogenesis of SSc [13–16], we aimed to determine their role in SRC. In ex vivo experiments with rat renal interlobar arteries, we found that IgG containing agonistic anti-AT₁R and anti-ET_AR Abs isolated from patients with SRC induced vasoconstriction via AT₁R and ET_AR. Anti-AT₁R and anti-ET_AR Abs positive IgG also hypersensitized the arteries to their natural ligands AngII and ET-1 priming them for an excessive contractile response. Furthermore, we found evidence for a crosstalk between AT₁R and ET_AR in the context of autoimmune receptor stimulation that can be explained by allosteric effects of agonistic Abs on the ligand-receptor interplay.

Our findings suggest that anti-AT₁R and anti-ET_AR Abs contribute to the development of SRC not only due to direct vasoconstrictive properties but also by amplifying the response of target organs such as small kidney arteries to AngII and ET-1. This mechanism can be considered particularly relevant, as both neurohumoral systems are activated in SRC [4, 11] and seem to be major drivers of detrimental pathophysiologic changes resulting in loss of kidney function and malignant hypertension. Anti-AT₁R and anti-ET_AR Abs from SSc patients have also been shown to augment vasoconstrictive responses to AngII and ET-1 in resistance arteries from rat lungs [13]. On the contrary, antibodies isolated from kidney transplant recipients with acute vascular rejection who were negative for donor-specific anti-HLA antibodies but positive for anti-AT₁R Abs only increased the contractile response of rat renal arteries to AngII by trend [18]. These variances might be related to functional heterogeneity of agonistic autoantibodies arising in differing pathological environments such as preeclampsia [22, 23], kidney transplantation [18, 20] or SSc [14] that is probably caused by antibody specificity directed against different epitopes of the receptors [18]. In line with this, a bioassay using AT₁R overexpressing Huh-7-cells revealed that anti-AT₁R Abs can feature either stimulatory or inhibitory effects [24].

Furthermore, we found evidence for a functional crosstalk of AT₁R and ET_AR in the presence of IgG from SRC patients positive for anti-AT₁R and anti-ET_AR Abs. This indicates that both potent vasoconstrictive systems are highly interrelated in this condition and therapeutic targeting of just one of them might be suboptimal. Although anti-AT₁R and anti-ET_AR Abs are not found exclusively in SSc patients with SRC, they might predispose for extreme intrarenal vasoconstriction and endothelial damage when a second hit strikes such as local RAAS and endothelin activation.

Interaction of AT₁R and ET_AR function might have resulted in some apparent inconsistencies in our experiments. For example, while blockade of ET_A/BR but not of AT₁R resulted in complete abolishment of SRC IgG-induced vasoconstriction as shown in Fig. 1B, phosphorylation of ERK1/2 was totally prevented by the AT₁RB and not by the ET_A/BRB (Fig. 2A and B). Moreover, as recently shown, direct agonistic effects of anti-AT₁R Abs or allosteric modulation of the receptors by the anti-AT1R Abs may depend on cell type and cellular function [25]. Accordingly, the Abs may induce different effects in primary ring segments vs isolated VSMC. Involved mechanisms might include varying receptor expression, receptor co-expression or the presence of different allosteric modulators [26].

Potential alterations of a GPCR adapting the response to binding partners that induce different conformational changes such as natural ligands, multiple types of agonistic autoantibodies, or different blocking molecules can be complex. They include differential recruitment of GPCR kinases, G-proteins or β-arrestins (biased signalling), surmountable and insurmountable antagonism, and short- or long-term desensitization [27, 28]. In addition, formation of homo- and/or heteromultimers of GPCRs has been associated with altered functionality in many cases [29] including enhanced signalling of AT₁R in response to AngII [30, 31].

How to explain the mechanisms of interaction between both receptors and hypersensitization? Structural data from X-ray crystallography of the AT_1/2 receptors and ET_BR revealing details of ligand binding, antibody binding and putative dimeric arrangement [32] may help to understand the complex interplay of the different components.

First, AngII and ET-1 bind at their receptors deeply into a crevice between the extracellular loops 1–3 (EL1-3) as known from previously determined receptor-ligand complexes (Fig. 6A and reviewed by Speck et al. [32]). The transmembrane helices (H) and the N-terminus (Nt) cover the bound ligand from the extracellular site (Fig. 6A). In a homodimeric arrangement between two AT₁R protomers, their contact surface is localized between H1-H2 and EL1 (Fig. 6B), one of the most observed dimeric interfaces in class A GPCR [33]. Additionally, for several GPCRs such as the melanocortin-4 receptor [34], it is known that dimer-arrangements can have allosteric effects on binding and signalling mutually between the protomers. This principal structural-functional relation in dimeric receptor units is also existent for heterodimeric GPCRs [35], as shown for the AT₁R/prostaglandin F2α-receptor dimer with signalling variations for arrestin-recruitment [36]. Thus, a heterodimeric receptor arrangement can generally contribute to specific signalling effects in cells expressing both AT₁R and ET_AR.

Second, an AT₂R-AngII analogue complex co-crystallized with an antibody-Fab has revealed a positive allosteric effect on signalling by the Fab [37]. Binding epitopes of this Fab fragment are located in the central part of EL2, the Nt and EL1 (Fig. 6C). Thus, the binding sites for Abs and endogenous ligand are not overlapping, but they are at the same structural entities like EL2 (ligand at the C-terminus of EL2, Abs at the central part of EL2). Based on these previous perspectives and insights, we suppose that the observed positive allosteric effects of agonistic Abs containing patient IgG can be explained by (i) Abs increasing the predisposition of the receptor to bind AngII by directly structurally altering the ligand binding site (e.g. at EL2), or (ii) increased signalling activity by bound Abs lowering the energetic barrier for the endogenous ligand to further stimulate the receptor thereby causing hyperstimulation ability. Such positive allosteric effect between the natural ligand and activating Abs [37] was also observed in this study for AT₁R (Fig. 3A). We postulate that such mechanisms of positive allosterism may also exist between Abs and AT₁R examined here because of high similarities in the sequences and structures of ligands and receptors involved. Moreover, heterodimerization of angiotensin and endothelin receptors is a widely known phenomenon [32].

Combining these insights on positive allosteric Abs effects with heterodimerization between AT₁R and ET_BR, we suggest the following scenario to explain hyperstimulating effects (Fig. 6A–D): agonistic ligands bind to their receptors, while activating Abs allosterically (laterally) support the activation at receptor protomers by structural modifications, such as close to or directly at EL2. In a heterodimeric receptor constellation, this effect should be accompanied by structural modifications at the dimeric interface (e.g. EL1) with potentially further allosteric (horizontal) positive effects on signalling. Interestingly, a heterodimeric complex (AT₁R/ET_AR)-ligand-Abs model (Fig. 6D) supposes a potentially extended scenario, where the Abs (Fab) bound at AT₁R also contacts the second ET_AR protomer. In this case, Abs trans-effects may occur at epitopes in EL1 and EL2, respectively. This idea is supported by findings of antibodies directly entering interleukin-4 receptor heterodimers [38] and by GPCR heterodimer selective Abs [39].

Conclusions

In conclusion, we report evidence for a pathophysiologic involvement of agonistic Abs directed against AT₁R and ET_AR in SRC. Pathogenic mechanisms include a direct contractile effect on renal resistance arteries as well as hypersensitization to the natural ligands AngII and ET-1 and mediation of a crosstalk between the two receptors. Further research is needed to elucidate the molecular mechanisms linking binding of Abs to AT₁R and ET_AR to the observed biologic effects in order to develop precise novel therapeutic strategies.

Supplementary data

Supplementary data are available at Rheumatology online.

Data availability statement

The data underlying this article will be shared on reasonable request to the corresponding author.

Funding

This work was supported by the Bundesministerium für Wirtschaft und Energie (BMWi, Federal Ministry for Economic Affairs and Energy; ZIM project KF2257305AJ1) and the Deutsche Stiftung Sklerodermie (DSS, German Scleroderma Foundation). G.K., P.S., A.P. and R.C. are funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) Project-ID 394046635—SFB 1365 subproject A03 and Project-ID 421152132—SFB 1423, subproject A01 (to P.S.).

Disclosure statement: The authors have declared no conflicts of interest.

Acknowledgements

This paper is dedicated to the memory of Duska Dragun, whose devotion to science and to understanding the biology of angiotensin II type 1 receptor activating autoantibodies was unsurpassed.

References

Author notes