Molecular Effects of Auto-Antibodies on Angiotensin II Type 1 Receptor Signaling and Cell Proliferation

The angiotensin II (Ang II) type 1 receptor (AT1R) is involved in the regulation of blood pressure (through vasoconstriction) and water and ion homeostasis (mediated by interaction with the endogenous agonist). AT1R can also be activated by auto-antibodies (AT1R-Abs), which are associated with manifold diseases, such as obliterative vasculopathy, preeclampsia and systemic sclerosis. Knowledge of the molecular mechanisms related to AT1R-Abs binding and associated signaling cascade (dys-)regulation remains fragmentary. The goal of this study was, therefore, to investigate details of the effects of AT1R-Abs on G-protein signaling and subsequent cell proliferation, as well as the putative contribution of the three extracellular receptor loops (ELs) to Abs-AT1R signaling. AT1R-Abs induced nuclear factor of activated T-cells (NFAT) signaling, which reflects Gq/11 and Gi activation. The impact on cell proliferation was tested in different cell systems, as well as activation-triggered receptor internalization. Blockwise alanine substitutions were designed to potentially investigate the role of ELs in AT1R-Abs-mediated effects. First, we demonstrate that Ang II-mediated internalization of AT1R is impeded by binding of AT1R-Abs. Secondly, exclusive AT1R-Abs-induced Gq/11 activation is most significant for NFAT stimulation and mediates cell proliferation. Interestingly, our studies also reveal that ligand-independent, baseline AT1R activation of Gi signaling has, in turn, a negative effect on cell proliferation. Indeed, inhibition of Gi basal activity potentiates proliferation triggered by AT1R-Abs. Finally, although AT1R containing EL1 and EL3 blockwise alanine mutations were not expressed on the human embryonic kidney293T (HEK293T) cell surface, we at least confirmed that parts of EL2 are involved in interactions between AT1R and Abs. This current study thus provides extended insights into the molecular action of AT1R-Abs and associated mechanisms of interrelated pathogenesis.


Introduction
The superfamily of G protein-coupled receptors (GPCRs) constitutes the largest membrane-spanning protein family in the human genome. More than 800 different human GPCRs transmit a huge variety of extracellular signals into the cytoplasm. Activating stimuli include peptides, neurotransmitters, chemokines, ions, metabolites, fatty acids, and even physical stimuli, such as light, mechanical forces, and pH shifts [1,2].
Angiotensin II type 1 receptor (AT 1 R) is a class A GPCR endogenously activated by angiotensin II (Ang II) and is evolutionarily related to angiotensin II type 2 receptor (AT 2 R). AT 1 R is expressed in the kidneys, adrenal gland, nervous system, heart and blood vessels [3]. Stimulation of AT 1 R results in the intracellular recruitment of various effectors, but especially in activation of G q/11 [4][5][6][7], which triggers, e.g., the Extracellular signal-Regulated Kinases 1/2 (ERK1/2) signaling pathway, thus driving blood pressure regulation via vasoconstriction, or water and ion homeostasis [8,9].
Binding of AT 1 R-Abs promotes downstream signaling through the activation of AT 1 R [14,25,26]. While Ang II receptor binding has been explored in detail [27][28][29][30][31], and a receptor-ligand complex structure (PDB ID: 6oS0 [32]) is known, the detailed molecular mechanisms of the antibody-receptor interaction and associated effects are still unknown, hampering a comprehensive understanding of AT 1 R-Abs-related pathogenesis at the molecular level.
Of particular note is a newly recognized link between COVID-19 infection and Ang II-AT 1 R signaling, which is involved in inflammatory processes, collateral tissue damage, and systemic failure [33]. Such viral infection dysregulates renin-angiotensin-aldosterone system (RAAS) homeostasis by elevating Ang II levels. AT 1 R blockers or biased AT 1 R agonists are considered to potentially complement COVID-19 treatment strategies, including severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) [34][35][36].
Angiotensin Converting Enzyme inhibitors (ACEis) are the drugs currently recommended to treat scleroderma renal crisis (SRC) [37] and have been found to result in a substantial improvement in the prognosis of this pathogenic condition [38]. However, recent studies have questioned the use of this drug because it was shown that ACEis are associated with a shorter time to occurrence of vascular lesions in systemic sclerosis (SSc) patients [39]. In a retrospective study, it was found that development of SRC was linked to the prescription of ACEis [40]. In addition, an analysis of EULAR registers unraveled that ACEi treatment was an independent risk factor for SRC [41]. On the contrary, Angiotensin Receptor Blockers (ARB) were not associated with any risks in these studies. Hence, better understanding of the activation of AT 1 R after stimulation with its natural ligand Ang II or patient antibodies could hold the key to developing innovative therapeutic strategies to prevent the occurrence of and treat SRC.
We elucidated here the impact of AT 1 R-Abs in SSc patients on the AT 1 R-mediated signaling in endothelial cells. In addition, we aimed to investigate the influence of the three AT 1 R extracellular loops (ELs) on autoantibody-induced receptor activation by substituting the sequence of AT 1 R ELs with alanine amino acid blocks and thereby delineate and further describe the determinants and regions mediating the effects of AT 1 R Abs.
The goal of this study was to provide further information about potential targets for pharmacological intervention strategies in AT 1 R-related signaling.

Endothelial Cell Proliferation Is Induced by AT 1 R-Abs-Mediated G q/11 Activation and Is Inhibited by Basal G i Signaling
Ang II is known to activate G-proteins G i and G q/11 intracellularly [42]. The latter induces phospholipase C activation and increases cytosolic Ca 2+ concentration, which, in turn, triggers cellular responses, such as protein kinase C stimulation and nuclear factor of activated T-cells (NFAT) signaling [43], which we measured here as a signaling readout.
Moreover, G q/11 and G i activation have also been reported to be potentially involved in the proliferation of specific cell types [42,[44][45][46]]-a key mechanism in obliterative vasculopathy. Therefore, we assessed NFAT signaling pathway stimulation by AT 1 R-Abs and the impact of such signaling on endothelial cell proliferation. For this purpose, we isolated AT 1 R-Abs from SSc patients with renal crisis. Of note, these specific SSc patients carried AT 1 R-Abs as well as antibodies targeting another class A GPCR, the endothelin-1 type A receptor (ET A R). To focus on AT 1 R-mediated effects, we used Human Microvascular Endothelial Cells (HMEC-1s), which endogenously express angiotensin receptors (ATRs) and ET A R. In the non-transfected state, HMEC-1s already show a higher expression of AGTR1 compared with EDNRA (Supplementary Materials Figure S1A), but we additionally transiently transfected these cells to over-express the human wild type receptor (WT AT 1 R), which finally leads to a 10,000-fold increase in AGTR1 mRNA relative to EDNRA mRNA (Supplementary Materials Figure S1B). This over-expression ensured the relative significance of induced effects at AT 1 R above potential effects of ET A R in interplay with Abs.
After six hours of stimulation with 1 mg/mL AT 1 R-Abs, NFAT activation levels were increased (~30%) compared with non-stimulated cells ( Figure 1A). Moreover, cell proliferation was moderately increased by the addition of AT 1 R-Abs, while Ang II effects remained insignificant ( Figure 1B).
To potentially distinguish between the effects of either G q/11 or G i in AT 1 R-Absinduced NFAT stimulation, cells were pre-treated with pertussis toxin (PTx) one hour prior to stimulation. PTx specifically inhibits G i signaling [47] and therefore decreases G i -related effects on NFAT stimulation by endogenous ligands. Of note, pre-incubation with PTx significantly reduced even basal level NFAT activation (non-stimulated cells, Figure 1C), which indicates G i -induced NFAT activation independently of AT 1 R-Abs or Ang II stimulation.
Compared with cells stimulated with AT 1 R-Abs without PTx, NFAT activation was significantly lower in PTx pre-treated cells stimulated with AT 1 R-Abs ( Figure 1C). However, the relative decrease in the signaling extending between AT 1 R with or without PTx and between AT 1 R-Abs with or without PTx was similar. Thus, we conclude that G i stimulated by AT 1 R-Abs has no direct effect on the regulation of NFAT signaling. This finding strongly suggests that G q/11 -mediated signaling through AT 1 R-Abs is a key pathway for NFAT stimulation, while basal G i signaling displays only a minor contribution to the effects on NFAT. (C) One hour pre-incubation with pertussis toxin (PTx) did not change NFAT activation or (D) cell proliferation induced by AT1R-Abs. Cells transfected to over-express wild-type (WT) AT1R were stimulated with 1mg/mL AT1R-Abs or 1µM Ang II, as indicated, with or without a pre-incubation with 5ng/mL PTx for one hour. SEMs derived from five to eight experiments are shown along with p-values for Wilcoxon or Mann-Whitney tests. (* p < 0.05 as compared with AT1R WT without stimulation; # p < 0.05 as compared with AT1R stimulated with Ang II; § p < 0.05, § § p < 0.01 as compared with AT1R WT pre-incubated with PTx.) To potentially distinguish between the effects of either Gq/11 or Gi in AT1R-Abs-induced NFAT stimulation, cells were pre-treated with pertussis toxin (PTx) one hour prior to stimulation. PTx specifically inhibits Gi signaling [47] and therefore decreases Gi-related effects on NFAT stimulation by endogenous ligands. Of note, pre-incubation with PTx significantly reduced even basal level NFAT activation (non-stimulated cells, Figure 1C), which indicates Gi-induced NFAT activation independently of AT1R-Abs or Ang II stimulation.
Compared with cells stimulated with AT1R-Abs without PTx, NFAT activation was significantly lower in PTx pre-treated cells stimulated with AT1R-Abs ( Figure 1C). However, the relative decrease in the signaling extending between AT1R with or without PTx and between AT1R-Abs with or without PTx was similar. Thus, we conclude that Gi stimulated by AT1R-Abs has no direct effect on the regulation of NFAT signaling. This finding strongly suggests that Gq/11-mediated signaling through AT1R-Abs is a key pathway for (C) One hour pre-incubation with pertussis toxin (PTx) did not change NFAT activation or (D) cell proliferation induced by AT 1 R-Abs. Cells transfected to over-express wild-type (WT) AT 1 R were stimulated with 1 mg/mL AT 1 R-Abs or 1 µM Ang II, as indicated, with or without a pre-incubation with 5 ng/mL PTx for one hour. SEMs derived from five to eight experiments are shown along with p-values for Wilcoxon or Mann-Whitney tests. (* p < 0.05 as compared with AT 1 R WT without stimulation; # p < 0.05 as compared with AT 1 R stimulated with Ang II; § p < 0.05, § § p < 0.01 as compared with AT 1 R WT pre-incubated with PTx).
While we showed that AT 1 R-Abs-mediated NFAT signaling increases independently from G i , we tested a potential G i involvement in HMEC-1 proliferation by inhibiting G i via PTx in proliferation assays. Surprisingly, we detected a significant increase in cell proliferation when only blocking G i basal activity (non-stimulated cells, Figure 1D), which suggests that basal G i activity reduces cell proliferation. The proliferation-supporting effect triggered upon G i inhibition is further increased in cells pre-treated with PTx and stimulated with AT 1 R-Abs ( Figure 1D). Cell proliferation is then doubled due to the combination of G i inhibition and the presence of AT 1 R-Abs, a result in stark contrast to Ang II's lack of effect on cell proliferation.
In summary, G i inhibition strengthens AT 1 R-Abs-mediated HMEC-1 proliferation (likely mediated by G q/11 ) and points to a "protective" effect of G i towards proliferation mechanisms, an effect which also differs between Ang II and AT 1 R-Abs-induced signaling.

Auto-Antibodies Attenuate Ang II-Mediated Internalization of AT 1 R
These identified AT 1 R-Abs effects on AT 1 R signaling are distinct from Ang II-induced signaling. We next examined the number of receptors expressed at the cell surface; this factor can have major effects on signaling properties [48] and all the more since it is already known that Ang II-mediated AT 1 R signaling capacity depends on internalization [49], whereby, in contrast, a recent study showed that AT 1 R-Abs isolated from preeclamptic women do not induce AT 1 R internalization [50].
Short-time incubation of cells with the endogenous AT 1 R ligand Ang II indeed resulted in a decreased number of receptors on the cell surface (Figure 2A), which is in line with the ligand-induced internalization already described [49]. In contrast, ten minutes stimulation with AT 1 R-Abs did not significantly alter AT 1 R numbers at the surface ( Figure 2A). This remained true after three hours of stimulation (Figure 2A), while we observed a significant decrease in AT 1 R numbers at the membrane after six hours of stimulation ( Figure 2A). In addition, although effects of AT1R-Abs on AT1R internalization in the presence of Ang II have not been investigated so far, this condition resembles the endogenous complex situation. We thus investigated whether auto-antibodies could modify Ang II-mediated internalization of the receptor. To address this, Human Embryonic Kidney293T cells (HEK293T) expressing HiBiT-tagged AT1Rs were pre-incubated (or not) with AT1R-Abs and then stimulated with increasing doses of Ang II ( Figure 2B-E). Cells pre-incubated with AT1R-Abs and then stimulated with 100 nM or 1 µM Ang II showed significantly transfected to express a HiBiT-tagged AT 1 R were incubated for 10 min, three or six hours with 1 µM Ang II or 1 mg/mL AT 1 R-Abs. (B-D) Ang II stimulation in HEK293T dose-dependently triggered a decrease in AT 1 R expression at the plasma membrane. Cells transfected with the HiBiT-tagged receptor were pre-incubated with or without 1 mg/mL AT 1 R-Abs for six hours before stimulation with increasing concentrations of Ang II for 10 min. Results are gathered in (E). SEMs derived from 7 to 14 experiments (A-D) are shown along with p-values for Wilcoxon or Mann-Whitney tests. (* p < 0.05 as compared with HiBiT-AT 1 R WT without stimulation; # p<0.05, ## p < 0.01, ### p < 0.001 as compared with HiBiT-AT 1 R WT stimulated with Ang II; § § p < 0.01, § § § p < 0.001 as compared with AT 1 R WT pre-incubated with AT 1 R-Abs.) In addition, although effects of AT 1 R-Abs on AT 1 R internalization in the presence of Ang II have not been investigated so far, this condition resembles the endogenous complex situation. We thus investigated whether auto-antibodies could modify Ang II-mediated internalization of the receptor. To address this, Human Embryonic Kidney293T cells (HEK293T) expressing HiBiT-tagged AT 1 Rs were pre-incubated (or not) with AT 1 R-Abs and then stimulated with increasing doses of Ang II ( Figure 2B-E). Cells pre-incubated with AT 1 R-Abs and then stimulated with 100 nM or 1 µM Ang II showed significantly higher numbers of AT 1 Rs on the plasma membrane than those that received Ang II stimulation alone ( Figure 2C,D). Therefore, it can be concluded that AT 1 R-Abs attenuate Ang II-mediated internalization of the receptors.

The Second Extracellular Loop Is Involved in AT 1 R-Mediated Signaling Induced by AT 1 R Abs
To evaluate the potential impact of the ELs on the interaction of AT 1 R with AT 1 R-Abs in SRC, we designed and generated poly-alanine mutations in each of the three ELs and tested these mutants first on their capacity to express at the cell surface. Of note, two amino acid stretches in EL2 at positions 173-178 and 182-187 have been previously supposed to be epitopes involved in antibody-binding to AT 1 R [10,14]. Already determined AT 1 R structures revealed the spatial localization of these EL2 fragments, both being exposed to the extracellular solvent region accessible for ligand contacts ( Figure 3A). Indeed, bound Ang II is in direct contact with the amino acid residues in the C-terminal EL2 fragment 182-187 ( Figure 3B). higher numbers of AT1Rs on the plasma membrane than those that received Ang II stimulation alone ( Figure 2C,D). Therefore, it can be concluded that AT1R-Abs attenuate Ang II-mediated internalization of the receptors.

The Second Extracellular Loop Is Involved in AT1R-Mediated Signaling Induced by AT1R Abs
To evaluate the potential impact of the ELs on the interaction of AT1R with AT1R-Abs in SRC, we designed and generated poly-alanine mutations in each of the three ELs and tested these mutants first on their capacity to express at the cell surface. Of note, two amino acid stretches in EL2 at positions 173-178 and 182-187 have been previously supposed to be epitopes involved in antibody-binding to AT1R [10,14]. Already determined AT1R structures revealed the spatial localization of these EL2 fragments, both being exposed to the extracellular solvent region accessible for ligand contacts ( Figure 3A). Indeed, bound Ang II is in direct contact with the amino acid residues in the C-terminal EL2 fragment 182-187 ( Figure 3B). . Active Angiotensin II type 1 receptor (AT1R) conformation bound with angiotensin II (Ang II). AT1R crystal structures either in an inactive or active state conformation provide detailed structural insights into this receptor (e.g., PDB IDs: 4zud [51], 6do1 [52], 6os0 [32]). To visualize our findings regarding loop mutants in the context of structural receptor properties, we used the (A) active state-like AT1R structure in complex with Ang II [32]. This active state AT1R structure (amino acids D9-Q315, PDB ID: 6os0) provided details of Ang II binding and extracellular loop (EL) conformations (different colours for EL1-3) and the N-terminal tail (Ntt). (B) Of note, beside the highly conserved G-protein coupled receptor (GPCR) disulfide bridge between a cysteine (yellow sticks) in transmembrane helix 3 (H3, C101) and a cysteine in the EL2 (C180), a second disulfide bridge fixes the N-terminus (C18) to the EL3/transmembrane helix 7 (H7) transition (C274). Amino acids studied here by blockwise alanine mutants are highlighted by line representation. Specific residues of the loops 1-3 are exposed to the extracellular solvent phase and may function as contact points for extracellularly interacting molecules.
All mutants were tagged with the HiBiT tag to evaluate how the alanine mutations affect the expression of AT1R at the cell surface. In HEK293T, we observed that most mu- . Active Angiotensin II type 1 receptor (AT 1 R) conformation bound with angiotensin II (Ang II). AT 1 R crystal structures either in an inactive or active state conformation provide detailed structural insights into this receptor (e.g., PDB IDs: 4zud [51], 6do1 [52], 6os0 [32]). To visualize our findings regarding loop mutants in the context of structural receptor properties, we used the (A) active state-like AT 1 R structure in complex with Ang II [32]. This active state AT 1 R structure (amino acids D9-Q315, PDB ID: 6os0) provided details of Ang II binding and extracellular loop (EL) conformations (different colours for EL1-3) and the N-terminal tail (Ntt). (B) Of note, beside the highly conserved G-protein coupled receptor (GPCR) disulfide bridge between a cysteine (yellow sticks) in transmembrane helix 3 (H3, C101) and a cysteine in the EL2 (C180), a second disulfide bridge fixes the N-terminus (C18) to the EL3/transmembrane helix 7 (H7) transition (C274). Amino acids studied here by blockwise alanine mutants are highlighted by line representation. Specific residues of the loops 1-3 are exposed to the extracellular solvent phase and may function as contact points for extracellularly interacting molecules.
All mutants were tagged with the HiBiT tag to evaluate how the alanine mutations affect the expression of AT 1 R at the cell surface. In HEK293T, we observed that most mutants, with the exception of AT 1 R polyA182-187, were significantly less expressed than AT 1 R WT (Figure 4). Variant AT 1 R polyA182-187 showed an expression level comparable to AT 1 R WT ( Figure 4). Thus, only AT 1 R polyA182-187 was selected for further experiments. to AT1R WT ( Figure 4). Thus, only AT1R polyA182-187 was selected for further experiments. As mentioned above, the patients whose material was investigated presented both AT1R-Abs and auto-antibodies targeting ETARs. To avoid cross-reactivity and obtain a direct read-out of the effects of structural changes on the receptor's function, we opted to develop a MMY14-AT1R yeast system (described in Material and Methods). To first test the reliability of the yeast growth assay, we used human WT AT1R expressed in MMY14 yeast cells. Stimulation with Ang II revealed that human WT AT1R indeed responds dosedependently to its endogenous peptide agonist, as indicated by increased cell growth (Figure 5A). In support, pre-incubation of yeast cells with the AT1R inhibitor (antagonist) Valsartan inhibited Ang II ligand-induced yeast growth, even at the highest concentration of supplemented Ang II ( Figure 5A). Secondly, MMY14-AT1R stimulated with AT1R-Abs (isolated from patients with SRC) also induced positive yeast growth levels. AT1R-Abs dose-dependently activates AT1R, as shown by a growth level comparable to the level induced by Ang II (Figure 5B). Pre-incubation with Valsartan resulted in a statistically significant decrease in yeast growth compared with AT1R-Abs/AT1R-mediated cell growth. Altogether, these findings confirmed the reliability of the assay for testing AT1Rmediated signaling induced by ligands. Transformation of yeasts with alanine-mutant receptors bearing AT1R polyA182-187A revealed that Ang II and AT1R-Abs-mediated activation of the receptors was significantly impaired compared with yeasts transformed with the WT receptor ( Figure 5C,D). This supposes an essential role of the EL2 in Abs-mediated signaling effects. As mentioned above, the patients whose material was investigated presented both AT 1 R-Abs and auto-antibodies targeting ET A Rs. To avoid cross-reactivity and obtain a direct read-out of the effects of structural changes on the receptor's function, we opted to develop a MMY14-AT 1 R yeast system (described in Material and Methods). To first test the reliability of the yeast growth assay, we used human WT AT 1 R expressed in MMY14 yeast cells. Stimulation with Ang II revealed that human WT AT 1 R indeed responds dose-dependently to its endogenous peptide agonist, as indicated by increased cell growth ( Figure 5A). In support, pre-incubation of yeast cells with the AT 1 R inhibitor (antagonist) Valsartan inhibited Ang II ligand-induced yeast growth, even at the highest concentration of supplemented Ang II ( Figure 5A). Secondly, MMY14-AT 1 R stimulated with AT 1 R-Abs (isolated from patients with SRC) also induced positive yeast growth levels. AT 1 R-Abs dose-dependently activates AT 1 R, as shown by a growth level comparable to the level induced by Ang II (Figure 5B). Pre-incubation with Valsartan resulted in a statistically significant decrease in yeast growth compared with AT 1 R-Abs/AT 1 R-mediated cell growth. Altogether, these findings confirmed the reliability of the assay for testing AT 1 R-mediated signaling induced by ligands. Transformation of yeasts with alanine-mutant receptors bearing AT 1 R polyA182-187A revealed that Ang II and AT 1 R-Abs-mediated activation of the receptors was significantly impaired compared with yeasts transformed with the WT receptor ( Figure 5C,D). This supposes an essential role of the EL2 in Abs-mediated signaling effects.

AT1R Auto-Antibodies Inhibit Ang II-Mediated Receptor Internalization
We show here that AT1R-Abs action not only lacks subsequent receptor internalization but also attenuates Ang II-induced AT1R internalization or leads to a prolonged surface expression of the receptor (Figure 2). This effect is potentially associated with differences regarding β-arrestin recruitment and receptor phosphorylation, as supposed by a previous study [53].
By impairing β-arrestin recruitment, AT1R-Abs, in contrast to Ang II, are categorised as GPCR-biased ligands that only alter a specific signaling pathway compared with endogenous signaling [54]. This contrasts with most recent findings on AT1R-Abs in transplantation patients showing that these antibodies induced β2-arrestin recruitment [55]. However, previous studies have shown that β-arrestin signaling is privileged and has (A,B) * p < 0.05 as compared with AT 1 R WT without stimulation and without inhibitorscompared with; & p < 0.05, && p < 0.01 as compared with AT 1 R WT pre-incubated with AT 1 R inhibitors (Valsartan). (C,D) ** p < 0.01, *** p < 0.001 as compared with AT 1 R WT without stimulation; $ p < 0.05, $$ p < 0.01 as compared with AT 1 R polyA182-187.)

AT 1 R Auto-Antibodies Inhibit Ang II-Mediated Receptor Internalization
We show here that AT 1 R-Abs action not only lacks subsequent receptor internalization but also attenuates Ang II-induced AT 1 R internalization or leads to a prolonged surface expression of the receptor (Figure 2). This effect is potentially associated with differences regarding β-arrestin recruitment and receptor phosphorylation, as supposed by a previous study [53].
By impairing β-arrestin recruitment, AT 1 R-Abs, in contrast to Ang II, are categorised as GPCR-biased ligands that only alter a specific signaling pathway compared with endogenous signaling [54]. This contrasts with most recent findings on AT 1 R-Abs in transplantation patients showing that these antibodies induced β2-arrestin recruitment [55]. However, previous studies have shown that β-arrestin signaling is privileged and has beneficial effects; for example, TRV120027, a β-arrestin biased ligand of AT 1 R, increases cardiomyocyte contraction and reduces Ang II effects in rats [56]. Similarly, TRV12023, another β-arrestin biased ligand of AT 1 R, increases cardiac contractility and induced pro-survival signaling after cardiac injury in mice [57]. Of note, TRV12023 also exhibited protective effects in a rat model infused with Ang II [58] by preventing cardiac hypertrophy of dilated cardiomyopathy [59]. Finally, β-arrestin-biased AT 1 R agonists have been suggested to treat COVID-19 in terms of inducing cardioprotective signaling pathways [36].
AT 1 R-Abs have been associated with increased intracellular signaling, particularly the ERK1/2 pathway, in SSc patients [26]. In vitro, AT 1 R-Abs induced higher IL-8 secretion by endothelial cells, resulting in decreased endothelial repair as well as increased monocyte migration and ROS production [60]. Moreover, these auto-antibodies triggered an increase in intracellular calcium in endothelial cells and higher vasoconstriction of rat pulmonary arteries ex vivo [61]. According to our findings, this stronger signaling might be due to a prolonged stay of the receptor at the cell surface and impaired signaling regulation.
In addition, a recently published study showed that heterodimerization of ET A R with endothelin-1 type B receptor (ET B R) could also delay β-arrestin recruitment, phosphorylation, and related receptor internalization [62]. Unfortunately, little is known yet regarding the heterodimerization capacity of endothelin with Ang II receptors. Therefore, we cannot exclude that AT 1 R heterodimerizes with ET A R in vivo, which may also have an impact on β-arrestin recruitment by receptors. Of note, antibodies targeting both receptors appear concomitantly in various diseases, such as preeclampsia, graft rejection, and systemic sclerosis [26,[63][64][65].
Of note, we cannot exclude completely that the HiBiT tag affects AT 1 R internalization. However, recent data published by the Promega Company showed that the β2-adrenergic receptor tagged with HiBiT internalized in the same way and to the same extent as reported in previously published studies [66].

Cell Proliferation Is Mediated by G q/11 Activation and Inhibited by Basal G i Signaling
G q/11 is activated by Ang II at the AT 1 R [67], also shown in tissues where Ang II triggers physiological effects [68]. Hence, G q/11 mediates the cardiac hypertrophic effects of Ang II involving vascular smooth muscle cells but not cardiomyocytes in transgenic mice [69], which was also confirmed ex vivo [7]. It has been demonstrated that neural progenitor proliferation was impaired by G q/11 inhibition [44]. Moreover, AT 1 R activation via Ang II results in the proliferation of Human Umbilical Vascular Endothelial Cells (HU-VEC) and is associated with increased expression of Vascular Endothelial Growth Factor (VEGF) [70]. Recently, it was reported that G q/11 regulates VEGF-induced HUVEC proliferation, and inactivation of G q/11 resulted in vivo in neo-angiogenesis impairment [46]. These effects could be blocked with Candesartan, a specific AT 1 R inhibitor.
It should also be considered that AT 1 R heterodimerizes with different receptors, such as Bradykinin 2 receptors [71][72][73], these receptors constituting a potential secondary mechanism for G q/11 activation. Complementary experiments are required to prove such a mechanism.
Under our experimental in vitro conditions, the proliferating effect of Ang II on HMEC-1s was very limited and more pronounced for AT 1 R-Abs ( Figure 1B). Although it is possible that the previously observed differences between the two stimuli resulted from the effects of AT 2 R with native HMEC-1s (which showed no proliferation with Ang II [25]), AT 2 R anti-proliferative effects (as described in vascular smooth muscle cells (VSMCs) [74]) are here very unlikely due to AGTR1 mRNA being 100,000-fold more expressed than AGTR2 mRNA (Supplementary Materials Figure S1). Furthermore, we recently showed that AT 2 R inhibitors fail to block SSc antibodies-induced endothelial cell proliferation [25]. In any case, further complementary studies are still required to determine the exact difference between Ang II-and antibodies-induced endothelial cell proliferation.
Our data also suggest that immune activation of AT 1 R by Abs stimulates G q/11mediated signaling (Figure 1C) but not G i signaling, which in turn increases endothelial cell proliferation. Notably, blocking basal Gi signaling activity strongly enhances cell proliferation induced by AT 1 R-Abs but not by Ang II (Figure 1D). This implies that, on the one hand, a basal G i signaling activity may have a "protective" effect towards cell proliferation induced by Abs, but, on the other hand, it again indicates that Ang II and AT 1 R-Abs differ in their signaling effects. Interestingly, in 2016, it was reported that G α/ i proteins are involved in the hyperproliferation of VSMCs in hypertensive rats [75], a finding confirmed in two other studies [76,77]. Similarly, another recent publication has shown that Gi-mediated signaling activity influences cancer cell proliferation [78]. However, why AT 1 R-Abs induces proliferation of HMEC-1 more efficiently than the natural ligand Ang II and how reduced basal G i signaling enhances the proliferative effect of Abs needs to be addressed ( Figure 1C).
The reason why AT 1 R-Abs induces a stronger proliferation of HMEC-1s than Ang II and how reduced basal G i signaling contributes to an enhanced proliferative effect of Abs needs to be clarified in more detail in future work.

EL2 Contributes to Antibody-Mediated Signaling in Systemic Sclerosis
In addition to the auto-antibodies involved in preeclampsia, it has been suggested that AT 1 R-Abs found in transplant rejection recognize not just one epitope located in the AT 1 R EL2 but interact with at least two distinguishable epitopes in this loop [10,14]. In general, the EL2 among many GPCRs is known as one of the key elements in ligand binding and signaling regulation [79,80]. Our AT 1 Rs using blockwise alanine mutants also indicate that this receptor has an essential role in EL2 ( Figure 5). In our experimental MMY model, an alanine mutation in the amino acid stretch 182 to 187 led to a complete inhibition of the receptor capability for signaling ( Figure 5C,D), yet without specificity for immunogenic activation. We conclude that the amino acid composition and probably the associated spatial arrangement of the EL2 relative to other receptor components is important for antibody and simultaneously for ligand recognition (Figure 3) or signaling regulation. This is in accordance with structural information about the AT 1 R-Ang II complex that shows that Ang II directly interacts with EL2 amino acids, further confirmed by binding studies of Ang II at the AT 1 R [30]. An additional contribution of Abs binding at particular EL2 parts must be assumed. Altogether, modifications at this loop, like the alanine substitutions tested here, disturb the ligand-peptide binding pocket and/or modify antibody-recognition motifs (Figure 3).
To date, only limited information is available on the potential contribution of AT 1 R EL1 and EL3 for ligand and Abs binding. Both loops expose various amino acids to the extracellular site, which may function as antibody epitopes (Figure 3). Between 1995 and 2003, studies were published about peptides derived from the first EL of rat AT 1 R which aimed to determine the structure of this loop [81]. However, no functional characterization accompanied these approaches. Our alanine substitutions of EL1 and EL3 led to impaired cell-surface expression (Figure 4), at least in HEK293T cells, which makes it unfortunately impossible to exclude or infer their contribution to AT 1 R-Abs binding.

Conclusions
Altogether, our investigations revealed that Ang II-induced internalization of AT 1 R is attenuated by a simultaneous action of AT 1 R-Abs. Importantly, G q/11 but not G i activation induced by AT 1 R-Abs mediates cell proliferation. Furthermore, G i basal signaling activity of AT 1 R has a decreasing effect on cell proliferation; inhibition of G i basal activity potentiates the proliferation-enhancing effect of AT 1 R-Abs but not Ang II, which requires further clarification of the molecular causes. In summary, our data show that Ang II and AT 1 R-Abs elicit different effects in terms of cell proliferation and G q/11 activation in patients with SRC, at least under our experimental conditions. This current study, therefore, provides insights into the molecular action of AT 1 R-Abs and the associated mechanisms of related pathogenesis.

Clinical Samples of Patients with Systemic Sclerosis
Plasma was obtained from three patients treated for Angiotensin Converting Enzyme I (ACEI) refractory SRC in our clinic (Charité-Universitätsmedizin, Berlin, Germany) between January 2006 and October 2010. SRC was defined in terms of an otherwise unexplained rapid decline in renal function (increase in serum creatinine ≥ 50%) in a patient with systemic sclerosis. Diagnosis was confirmed by renal biopsy that assessed obliterative vasculopathy of arteries and arterioles in all cases. Moreover, all SRC individuals were tested for the presence of AT 1 R antibodies, as previously described (CellTrend GmbH, Luckenwalde, Germany) [25,26,61]. Patients were considered positive if their antibody level reached at least 17 U/mL [26]. The local Medical Ethics Committee approved the study protocol (EA1/013/705). All experiments were conducted with immunoglobulin G (IgG) extracted from patient plasma samples (AT 1 R-Abs), as previously described [14,25,26]. Based on the results obtained with the MMY yeast model in which AT 1 R was the sole GPCR expressed (see the description of the experimental model below), where AT 1 R-Abs dose-dependently activated AT 1 Rs ( Figure 5B), we decided to perform the experiments with the smallest dose of 1 mg/mL activating AT 1 R.

Expression of Constructs and Site-Directed Mutagenesis
Complementary DNA of human full-length AT 1 R (kindly given by the group of Pierre Lavigne, Université de Sherbrook, Sherbrook, QC, Canada) was cloned into p426GPD (kindly given by Simon Dowell, GSK, Stevenage, United Kingdom) for expression into S. cerevisiae. Sub-cloning into pcDNA3 was also performed to allow expression in Human Microvascular Endothelial-1 cells (HMEC-1s) using the InFusion kit from Takara/Clontech (Mountain View, CA, USA). Site-directed mutagenesis was performed with the Q5 Sitedirected mutagenesis kit from New England Biolabs (Frankfurt am Main, Germany) using specific primers (Table 1).
For stimulation, HMEC-1 and HEK293T cells were cultured in starvation medium.

GPCR-Induced Yeast-Growth Assay
A specific yeast growth assay enabled a controlled activation and characterization of a particular GPCR. The methods and protocols used in this assay are well-established [84,85]. In brief, modified yeast strains grow under dependency of activation by a single inserted human GPCR. Furthermore, these yeasts express a chimeric yeast G-protein with five amino acids of a particular human G-protein C-terminus, conferring to each strain a forced human G-protein specificity. Upon activation of the transformed human GPCR, the chimeric G-protein is activated and triggers the activation of the Mitogen Activated Protein Kinase (MAPK) signaling pathway, resulting in the pairing and growing of the yeasts. This system has in principle already been demonstrated by studies with human adenosine 1 and 2B receptor subtypes [85,86]. Here, in analogy, MMY14 S. cerevisiae yeast were transformed with human AT 1 R (MMY14-AT 1 R) and, upon activation of the receptor, G q/11 activation could induce yeast growth. This assay, finally, allows detection of constitutively activated receptors and/or receptors with increased response to external stimuli through increased growth rate levels.
The transformed yeasts grew for three days on WHAUL plates supplemented in histidine (WHAUL-His). Four clones were picked-up and transferred individually onto a new WHAUL-His plate. On the day after, each clone was transferred into liquid WHAUL-His medium and allowed to grow overnight at 30 • C under shaking. To evaluate the activation of the receptor in response to stimulation, each pre-cultured clone was added to liquid WHAUL Medium without histidine and supplemented with BU salts, fluorescein di-β-D-galactopyranoside (10 nM final) to monitor yeast growth, and 3-amino-1,2,4 triazole (2 mM final) to suppress basal growth in His-deficient medium. First, this system was stimulated with increasing concentrations of Ang II ranging from 0 to 1 µM to prove the reliability of the system for experimental studies, then we investigated the ligand-mediated action of WT AT 1 Rs or AT 1 R variants on yeast growth. AT 1 R-Abs were isolated as described above and dialyzed against WHAUL medium. Yeast growth was measured on a FLUOStar OPTIMA reader after 20 h incubation.

Luciferase Reporter Assay
HMEC-1s were plated on 24-well plates with 50,000 cells per well and allowed to grow in complete medium until they reached 70% confluence. Cells were then transfected with 250 ng WT or mutated human AT 1 R cloned into pcDNA3 and 250 ng NFAT luciferase reporter plasmids. Medium was changed 24 h after to complete medium. On the next day, the medium was changed to starvation medium with or without 5 ng/mL pertussis toxin. After one hour, cells were treated with medium with or without AT 1 R-Abs (1 mg/mL). AT 1 R-Abs were previously dialyzed against Dulbecco's Modified Eagle's Medium low Glucose for six hours. The cells were subsequently washed in PBS and lysed in 1x passive lysis buffer (Promega, Madison, WI, USA) according to the manufacturer's instructions. Luciferase production was measured (luciferase assay system, Promega Corporation, Madison, WI, USA) according to the manufacturer's instructions.

EdU Proliferation Assay
HMEC-1s were plated and cultivated as described above for the luciferase reporter assays. Transient transfection was performed with 250 ng of wild type or mutated human AT 1 Rs cloned into pcDNA3. The next day, the medium was changed to complete medium. After 24 h, the medium was changed to starvation medium with or without 5 ng/mL pertussis toxin. After one hour, cells were stimulated with medium with or without 1 mg/mL AT 1 R-Abs (dialyzed against DMEM Low Glucose for 24 h). Two hours prior to the end of stimulation, half the medium was replaced with starvation medium containing 20 µM edU. Immunofluorescence was performed according to the manufacturer's instructions (iclick EdU Andy Fluor 488 Imaging kit, ABP Biosciences, Beltsville, MD, USA) and nuclei were counter-labelled with a PBS/DAPI 5 µg/mL solution.

Receptor Expression at the Cell Surface
HEK293T or HMEC-1 cells were plated on white 96-well plates (Corning, NY, USA) with 25,000 or 50,000 cells per well, respectively, and allowed to grow in complete medium. Cells were then transfected with 400 ng human AT 1 Rs tagged with HiBiT cloned into pcDNA3. Four hours after transfection, the medium was changed to complete medium. The cells were then incubated in starvation medium for a further day. On the next day, the cells were treated with the Nano-Glo HiBiT extracellular detection system (Promega Corporation, Madison, WI, USA) according to the manufacturer's instructions.

Receptor Internalization after Ligand Exposure
HEK293T cells were plated on white 96-well plates (Corning, NY, USA) with 25,000 cells per well and allowed to grow in complete medium. Cells were then transfected with 400 ng human AT 1 Rs tagged with HiBiT cloned into pcDNA3. Four hours after transfection, the medium was changed to complete medium. The cells were then incubated in starvation medium for a further day. On the day of stimulation, HEK293T cells were incubated for different durations (10 min, 3 and 6 h) and stimulated with 1 µM Ang II or 1 mg/mL AT 1 R-Abs (dialyzed against DMEM Low Glucose for 24 h).
Alternatively, HEK293T cells were pre-incubated with AT 1 R-Abs (1 mg/mL) or DMEM low glucose for six hours, then stimulated with different concentrations of Ang II (10 nM, 100 nM, or 1 µM) for 10 min.
In every case, cells were treated following stimulation with the Nano-Glo HiBiT extracellular detection system (Promega Corporation, Madison, WI, USA) according to the manufacturer's instructions.

Institutional Review Board Statement:
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Charité-Universitätsmedizin Berlin (EA1/013/705).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors. The data are not publicly available as they may be used for further studies by the group.

Acknowledgments:
The authors would like to dedicate this manuscript to Duska Dragun who was the first to understand the potential of this study. The authors are grateful to Heike Biebermann (Charité-Universitätsmedizin Berlin) for providing the pGL4.34 plasmids used in the luciferase reporter assays. This research was supported by the German Research Foundation (DFG, Deutsche Forschungsgemeinschaft) project-ID 394046635-SFB1365 (subproject A03 to D.D. and P.S.), project-ID 421152132-SFB1423 (subprojects A01/A05/Z03 to P.S. and A03/C01/Z04 to P.W.H.), and project-ID 25065445-SFB740 (subproject B6 to P.W.H. and P.S.). This research was also funded by the European Union's Horizon 2020 MSCA Program under grant agreement 956314 (ALLODD) (G.K. and P.S.). The work was further supported through Stiftung Charité and the Einstein Center Digital Future (to P.W.H.).

Conflicts of Interest: H. Heidecke is employed by CellTrend GmbH and S. Dowell is employed by
GlaxoSmithKline. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.