AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 1 of 34
Vascular
receptor
autoantibodies
in
pulmonary
arterial
hypertension
associated with systemic sclerosis Mike O. Becker, MD1*, Angela Kill1*, Marissa Kutsche1, Jeannine Guenther1, Angelika Rose1, Christoph Tabeling, MD2, Martin Witzenrath, MD2, Anja A. Kühl, PhD3, Harald Heidecke, PhD4, Hossein A. Ghofrani, MD5, Henning Tiede, MD5, Ralph T. Schermuly, PhD5, Nils Nickel, MD6, Marius M. Hoeper, MD6, Ivo Lukitsch, MD7, Maik Gollasch, MD7, Wolfgang M. Kuebler, MD8, Sebastian Bock1, Gerd R. Burmester, MD1, Duska Dragun, MD7*, Gabriela Riemekasten, MD1*
* these authors equally contributed to the manuscript
Contributions of all of those named in the author line, and details of their substantial contribution to the study: Conception and design: MOB, MW, HH, IL, MG, WMK, GRB, DD, GR; data acquisition: MOB, AK, MK, JG, AR, CT, MW, AAK, HH, HAG, HT, NN, MH, IL, MG, WMK, SB, DD, GR; data analysis and interpretation: MOB, AK, MK, JG, AR, CT, MW, AAK, HH, HAG, HT, RS, NN, MH, IL, MG, WMK, SB, GRB, DD, GR; drafting or critically revising of the manuscript for important intellectual content: MOB, AK, MK, JG, CT, MW, AAK, RS, NN, MH, WMK, SB, GRB, DD, GR.
Running title: Anti-AT1R/ETAR Abs in SSc-PAH
1
Departments of Rheumatology and Clinical Immunology†, and German Rheumatism
Research Centre, a Leibniz institute,
2
Department of Infectious Diseases and
Respiratory Medicine†, 3Dept. of Medicine I for Gastroenterology, Infectious Diseases
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 2 of 34
and Rheumatology / Research Center ImmunoSciences, CBF†, 4CellTrend GmbH Luckenwalde, Germany, 5University of Giessen and Marburg Lung Center (UGMLC), member of the German center for lung research (DLZ), Center for Pulmonary Hypertension,
University Hospital
Giessen
and
Marburg,
Germany
(HAG),
6
Department of Respiratory Medicine, Hannover Medical School, Hannover,
Germany, 7Nephrology and Intensive Care Medicine, CVK†, 8Keenan Research Centre, St. Michael´s Hospital, Toronto and Dept. of Physiology† †
Charité University Hospital Berlin
Address for Correspondence:
Prof. Dr. G. Riemekasten Charité University Hospital T170, Rheumatology and Clinical Immunology Charitéplatz 1 10117 Berlin Germany Tel: +49 30 450513017 Fax: +49 30 450513928 E-mail:[email protected]
Prof. Dr. D. Dragun Charité University Hospital Nephrology and Intensive Care Medicine Augustenburger Platz 1 13353 Berlin Tel: +49 30 450653485 2
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 3 of 34
Fax: +49 30 450565925 E-mail: [email protected]
Descriptor
number:
9.35
(Pulmonary
Hypertension:
Clinical-
Diagnosis/Pathogenesis/Outcome)
Abstract: Objective: Systemic sclerosis (SSc)-associated pulmonary arterial hypertension (PAH) portends worse outcome than other forms of PAH. Vasoconstrictive and vascular remodeling actions of endothelin-1 (ET-1) and angiotensin II (Ang II) via endothelin receptor type A (ETAR) and angiotensin receptor type-1 (AT1R) activation are
implicated
in
PAH
pathogenesis.
We
hypothesized
that
stimulating
autoantibodies (Abs) targeting and activating AT1R and ETAR may contribute to SScPAH pathogenesis and tested their functional and biomarker relevance. Methods and Results: Anti-AT1R and -ETAR Abs detected by ELISA were significantly higher and more prevalent in patients with SSc-PAH (n = 81) and connective tissue disease (CTD)-associated PAH (n=110) compared to other forms of PAH/pulmonary hypertension (n=106). High anti-AT1R and anti-ETAR Abs predicted development of SSc-PAH and SSc-PAH-related mortality in a prospective analysis. Both Abs increased endothelial cytosolic Ca2+ concentrations in isolated perfused rat lungs which could be blocked by respective specific receptor antagonists. Ab-mediated stimulation of third to fourth-generation intralobar pulmonary rat artery ring segments in a myograph increased vasoconstrictive responses to Ang II and ET-1 and 3
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 4 of 34
implicated cross-talk between both pathways demonstrated by reciprocal blockade with respective antagonists. Transfer of SSc-IgG containing both autoantibodies into healthy C57Bl/6J mice led to more abundant vascular and airway alpha-smooth muscle actin expression and inflammatory pulmonary vasculopathy. Conclusions: Anti-AT1R and -ETAR Abs are more frequent in SSc-PAH/CTD-PAH compared to other forms of PH and serve as predictive and prognostic biomarkers in SSc-PAH. Both antibodies may contribute to SSc-PAH via increased vascular endothelial reactivity and induction of pulmonary vasculopathy.
Key words: systemic sclerosis, PAH, AT1R, ETAR, PAH model, autoantibodies
Introduction Pulmonary arterial hypertension (PAH) is a rare yet frequently fatal clinical condition (1). Patients with connective tissue diseases (CTD) account for 20-30% of PAH patients in clinical trials. The majority suffers from systemic sclerosis (SSc, e.g. 2). Despite substantial therapeutic improvements, CTD-PAH and SSc-PAH patients have a reduced survival compared to patients with idiopathic PAH (IPAH) (3). The complex pathogenesis of PAH includes sustained vasoconstriction and vascular remodeling of small pulmonary arteries (4). Endothelin-1 (ET-1) or angiotensin II (Ang II) and their receptors have been implicated in disease pathogenesis. Pharmacologic targeting of ET-1 actions, mainly via activation of ET-1 receptor type A (ETAR), has become a mainstay of PAH therapy (5,6). Increased angiotensin II type-1 receptor (AT1R) activation was linked to disease progression and mortality in patients with
4
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 5 of 34
IPAH (7,8). Pharmacologic targeting of AT1R with losartan alleviated experimental PAH (9). In addition, dysregulated immunity and inflammation are also common in CTD-PAH and other PAH entities as suggested by inflammatory infiltration, growth factor expression in the remodeled pulmonary vasculature and increased cytokine and chemokine levels in the circulation (10). However, heterogeneity in therapeutic responses asks for a more precise definition of pathophysiological differences among PAH entities. Therefore new diagnostic and prognostic biomarkers will be critical to identify patient subsets that are likely to benefit from targeted therapies. Recently, we have identified simultaneous presence of functional autoantibodies (Abs) against AT1R and ETAR in patients with SSc (11) and linked them to increased prevalence of SSc-related vascular and fibrotic complications and higher risk for SScrelated mortality. Both antibodies induce phosphorylation of Erk1/2 and TGF-β expression in endothelial cells and hence activate endothelial cells to secrete cytokines (11,12). In addition, the antibodies mediate decreased wound repair by endothelial cells and act on immune cells (12,13). Based on these data and inducible renal obliterative lesions after transfer of anti-AT1RAbs into animals (14), we hypothesized an involvement of anti-AT1RAbs and anti-ETAR Abs in the pathogenesis of CTD-, especially SSc-related PAH. After investigating their presence in different PAH entities, we studied their potential relevance as predictive and prognostic biomarkers for SSc-PAH, and their contribution to its pathogenesis. Some of the results of these studies have been previously reported in the form of an abstracts (15,16)
Material and Methods Patients. 5
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 6 of 34
Anti-AT1R/ETAR Ab levels were first measured in serum samples from 62 consecutive SSc-PAH patients admitted to the Rheumatology department of the Charité. Additional serum samples were obtained from 182 patients with different pulmonary hypertension (PH)/PAH entities followed at the Department of Respiratory Medicine (n=115) at the University Hannover, Germany, and at the Justus-Liebig University Giessen, Germany (n=67). 62 of these 182 patients had idiopathic PAH (IPAH), 31 were classified as chronic thromboembolic PH (CTEPH), 14 patients had PH due to congenital heart disease (CHD), 19 SSc-PAH and 56 patients were PAH patients associated with CTD, which was not further specified. Combining all cases from the three centers, the cohort included 110 CTD-PAH patients, of which 81 had SSc-PAH. SSc patients fulfilled the new EULAR/ACR criteria for systemic sclerosis (17). In addition, 253 consecutive SSc patients entering our outpatient clinic between 2006-2011, without right heart catheter (RHC)-proven PAH and naive to the use of ETAR or AT1R blockers, were tested for anti-AT1R and anti-ETAR antibody levels and were prospectively observed. The epidemiologic data of the patients are shown in Table 1. Under clinical routine conditions, patients were screened for PAH at least in one-year intervals by assessment of WHO functional class, echocardiography, lung function with detection of predicted diffusion capacity for carbon monoxide (DLCOSB), and, during the last years, also by detection of N-terminal pro-brain natriuretic peptide (NT-proBNP) levels. Only if PAH was suspected, RHC was performed, using a mean pulmonary artery pressure ≥25 mm Hg and a pulmonary capillary wedge pressure ≤ 15mmHg at rest for PAH diagnosis (18). All patients were examined in experienced PAH centers. Pulmonary vascular resistance (PVR) and cardiac index (CaI) were determined according to the local guidelines, yet not reported for all patients. The study was approved by the local ethics committee of each contributing center and included written consent of each patient. 6
Copyright © 2014 by the American Thoracic Society
Page 7 of 34
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Detection of anti-AT1R and anti-ETAR Abs. Serum antibody levels were measured by an ELISA (CellTrend GmbH, Germany, and since 2012 One Lambda, Inc., U.S.A.) as previously described (11). Briefly, polystyrene plates were coated with extracts from cells overexpressing the human AT1R or the human ETAR in native conformation, respectively. Detection of the Ab levels was performed by staff unaware of patient data.
Isolation of human anti-AT1R and anti-ETAR Abs containing IgG and control IgG All ex-vivo and in-vivo experiments were conducted with endotoxin-free IgG isolated from serum samples as previously described (11,12). For in vitro experiments, final concentration was 1 mg/mL if not stated otherwise. Purified pooled IgG from 14 SSc patients with high levels of anti-AT1R Abs (mean concentration 21.8 U) and of high anti-ETAR Abs (mean concentration 17.91 U) was used for animal experiments. IgG from 15 healthy donors (HD) served as control (mean concentration 3.85 U for antiAT1R Abs and 2.5 U for anti-ETAR Abs, respectively). Real time fluorescence imaging of endothelial calcium. Isolated lungs from male Sprague-Dawley rats (350 to 450 g body weight) were prepared (n=9 per group), perfused and ventilated as described (19, 20). For imaging of endothelial cytosolic Ca2+ concentration ([Ca2+]i), lungs were positioned under an upright fluorescence microscope and superfused with normal saline at 37°C. Fura-2acetoxymethylester (Fura-2 AM, 5 µmol/L), which is de-esterified intracellularly to the Ca2+ sensitive dye fura-2, was loaded to lung endothelial cells for 20 min via a microcatheter. Fura-2 fluorescence in endothelial cells of subpleural lung single venular capillaries of 15-30 µm in diameter was excited as described (21). 7
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 8 of 34
Endothelial [Ca2+]i was determined from the 340/380 ratio based on a Kd of 224 nmol/L and appropriate calibration parameters (21). Following baseline recordings, purified IgG from either SSc patient or from HD sera was directly infused in HEPES dextran buffer into the imaged microvessels via a microcatheter over 30 min. The AT1R blocker valsartan (gift by D. N. Müller, Max Delbrück Center for Molecular Medicine Berlin, Germany) and ETAR blocker sitaxsentan (provided by Pfizer Deutschland GmbH) were each administered 10 min prior to IgG infusion in respective experiments (1 µmol/L each). All animal studies were approved by the local government authorities. Small vessel myography Third to fourth-generation intralobar pulmonary rat artery ring segments from Sprague-Dawley rats (external diameter 100-300 µm, length 2 mm) were freed from connective tissue and mounted in a small vessel myograph (DMTDanish Myotechnology, Aarhus, Denmark). Contractile responses were measured as described (22). As ET-1 induced contraction in rat small pulmonary arteries is insensitive to either BQ-123, a selective ETAR antagonist, or the selective ETBR antagonist BQ-788 (22), we used the combined ETA/BR-antagonist bosentan (gift from Actelion, Basel, Switzerland). The vessels were incubated with 10 µM bosentan for 20 min after which IgG (0.1 mg/mL) was added for 30 min. Thereafter, vessels were exposed to cumulatively increasing concentrations of Ang II. After repeated washing, the vessels were treated following the same protocol substituting valsartan 10 µM for bosentan and ET-1 for Ang II. Contractile responses to Ang II and ET-1 were measured as a percentage of the response to 60 mM K+, the effects were reversible by the respective receptor blockers. Antibody transfer experiments 8
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 9 of 34
For short-term experiments, female 7 week-old C57Bl/6J mice received a single intravenous injection of 800 µg IgG dissolved in 100 µL NaCl 0.9%. Lungs were harvested for histological and immunological analyses at day 7. For long-term experiments, mice received repeated IgG injections of 200 µg IgG dissolved in 100 µL NaCl 0.9% at days 1, 17, 30 and 93. Organs were harvested at day 100. Body weight (day 30, 58, 78 and at read out day) and rectal temperature control (at read out day; BAT-12 Microprobe; Physitemp Instruments, Clifton, NJ) were monitored.
Histology and immunohistochemistry For immunohistochemistry, lungs were removed and immediately snap frozen in O.C.T.TM medium (Tissue-Tek®, Sakura, Europe). Sections of 3 µm were prepared and stained with a polyclonal anti-alpha smooth muscle actin (α-SMA) antibody (Abcam, U.K.) and with an anti-human IgG F(ab')2-fragment (mouse absorbed, Dianova, Germany) to detect human IgG deposition. DAPI (Sigma Aldrich, Germany) was used for nuclear DNA staining. For histological assessment, lungs were fixed in 4% paraformaldehyde at room temperature for 24 hours and embedded in paraffin using standard procedures. Anti-CD31 antibody was used (clone SZ31, Dianova, Germany) to stain endothelial cells. Hematoxylin and eosin (H&E) staining was performed to visualize cellular organization. Signals were detected by light microscopy (Axioplan, Carl Zeiss MicroImaging GmbH, Germany) as well as by immunofluorescence (LSM710, Carl Zeiss Microscopy GmbH, Germany) and analyzed by ZEN 2009 Light Edition software (Carl Zeiss Microscopy GmbH, Germany).
9
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 10 of 34
Statistics Comparisons between groups were analyzed by Mann-Whitney U test for continuous variables. Kaplan-Meier analyses were performed to investigate the onset of PAH development and mortality after antibody detection. Significance was determined by Log-rank test. Receiver-operating-characteristic (ROC) curves were plotted to determine area under the curve (AUC), sensitivity, and specificity using calculated distance to 0.1. Small vessel myography data were analyzed by analysis of variance (ANOVA). Results are expressed as means ± SEM, in myograph experiments n stands for the number of the studied pulmonary artery ring segments. Differences were considered as statistically significant at a p value ≤0.05. Graphpad Prism 5.0 as well as IBM SPSS 21.0 were used for statistical analyses.
Results Anti-AT1R and anti-ETAR antibodies are higher in SSc- and CTD-PAH than in other forms of PH/PAH Epidemiologic and hemodynamic data of the cohorts studied are shown in Table 1. PH/PAH groups did not differ significantly among centers. Higher percentage of female patients and shorter duration of PAH were observed in CTD-PAH and SScPAH. Mean pulmonary arterial pressure (mPAP) as detected by RHC in these two groups was significantly lower as compared to IPAH (p
Page 1 of 34
Vascular
receptor
autoantibodies
in
pulmonary
arterial
hypertension
associated with systemic sclerosis Mike O. Becker, MD1*, Angela Kill1*, Marissa Kutsche1, Jeannine Guenther1, Angelika Rose1, Christoph Tabeling, MD2, Martin Witzenrath, MD2, Anja A. Kühl, PhD3, Harald Heidecke, PhD4, Hossein A. Ghofrani, MD5, Henning Tiede, MD5, Ralph T. Schermuly, PhD5, Nils Nickel, MD6, Marius M. Hoeper, MD6, Ivo Lukitsch, MD7, Maik Gollasch, MD7, Wolfgang M. Kuebler, MD8, Sebastian Bock1, Gerd R. Burmester, MD1, Duska Dragun, MD7*, Gabriela Riemekasten, MD1*
* these authors equally contributed to the manuscript
Contributions of all of those named in the author line, and details of their substantial contribution to the study: Conception and design: MOB, MW, HH, IL, MG, WMK, GRB, DD, GR; data acquisition: MOB, AK, MK, JG, AR, CT, MW, AAK, HH, HAG, HT, NN, MH, IL, MG, WMK, SB, DD, GR; data analysis and interpretation: MOB, AK, MK, JG, AR, CT, MW, AAK, HH, HAG, HT, RS, NN, MH, IL, MG, WMK, SB, GRB, DD, GR; drafting or critically revising of the manuscript for important intellectual content: MOB, AK, MK, JG, CT, MW, AAK, RS, NN, MH, WMK, SB, GRB, DD, GR.
Running title: Anti-AT1R/ETAR Abs in SSc-PAH
1
Departments of Rheumatology and Clinical Immunology†, and German Rheumatism
Research Centre, a Leibniz institute,
2
Department of Infectious Diseases and
Respiratory Medicine†, 3Dept. of Medicine I for Gastroenterology, Infectious Diseases
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 2 of 34
and Rheumatology / Research Center ImmunoSciences, CBF†, 4CellTrend GmbH Luckenwalde, Germany, 5University of Giessen and Marburg Lung Center (UGMLC), member of the German center for lung research (DLZ), Center for Pulmonary Hypertension,
University Hospital
Giessen
and
Marburg,
Germany
(HAG),
6
Department of Respiratory Medicine, Hannover Medical School, Hannover,
Germany, 7Nephrology and Intensive Care Medicine, CVK†, 8Keenan Research Centre, St. Michael´s Hospital, Toronto and Dept. of Physiology† †
Charité University Hospital Berlin
Address for Correspondence:
Prof. Dr. G. Riemekasten Charité University Hospital T170, Rheumatology and Clinical Immunology Charitéplatz 1 10117 Berlin Germany Tel: +49 30 450513017 Fax: +49 30 450513928 E-mail:[email protected]
Prof. Dr. D. Dragun Charité University Hospital Nephrology and Intensive Care Medicine Augustenburger Platz 1 13353 Berlin Tel: +49 30 450653485 2
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 3 of 34
Fax: +49 30 450565925 E-mail: [email protected]
Descriptor
number:
9.35
(Pulmonary
Hypertension:
Clinical-
Diagnosis/Pathogenesis/Outcome)
Abstract: Objective: Systemic sclerosis (SSc)-associated pulmonary arterial hypertension (PAH) portends worse outcome than other forms of PAH. Vasoconstrictive and vascular remodeling actions of endothelin-1 (ET-1) and angiotensin II (Ang II) via endothelin receptor type A (ETAR) and angiotensin receptor type-1 (AT1R) activation are
implicated
in
PAH
pathogenesis.
We
hypothesized
that
stimulating
autoantibodies (Abs) targeting and activating AT1R and ETAR may contribute to SScPAH pathogenesis and tested their functional and biomarker relevance. Methods and Results: Anti-AT1R and -ETAR Abs detected by ELISA were significantly higher and more prevalent in patients with SSc-PAH (n = 81) and connective tissue disease (CTD)-associated PAH (n=110) compared to other forms of PAH/pulmonary hypertension (n=106). High anti-AT1R and anti-ETAR Abs predicted development of SSc-PAH and SSc-PAH-related mortality in a prospective analysis. Both Abs increased endothelial cytosolic Ca2+ concentrations in isolated perfused rat lungs which could be blocked by respective specific receptor antagonists. Ab-mediated stimulation of third to fourth-generation intralobar pulmonary rat artery ring segments in a myograph increased vasoconstrictive responses to Ang II and ET-1 and 3
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 4 of 34
implicated cross-talk between both pathways demonstrated by reciprocal blockade with respective antagonists. Transfer of SSc-IgG containing both autoantibodies into healthy C57Bl/6J mice led to more abundant vascular and airway alpha-smooth muscle actin expression and inflammatory pulmonary vasculopathy. Conclusions: Anti-AT1R and -ETAR Abs are more frequent in SSc-PAH/CTD-PAH compared to other forms of PH and serve as predictive and prognostic biomarkers in SSc-PAH. Both antibodies may contribute to SSc-PAH via increased vascular endothelial reactivity and induction of pulmonary vasculopathy.
Key words: systemic sclerosis, PAH, AT1R, ETAR, PAH model, autoantibodies
Introduction Pulmonary arterial hypertension (PAH) is a rare yet frequently fatal clinical condition (1). Patients with connective tissue diseases (CTD) account for 20-30% of PAH patients in clinical trials. The majority suffers from systemic sclerosis (SSc, e.g. 2). Despite substantial therapeutic improvements, CTD-PAH and SSc-PAH patients have a reduced survival compared to patients with idiopathic PAH (IPAH) (3). The complex pathogenesis of PAH includes sustained vasoconstriction and vascular remodeling of small pulmonary arteries (4). Endothelin-1 (ET-1) or angiotensin II (Ang II) and their receptors have been implicated in disease pathogenesis. Pharmacologic targeting of ET-1 actions, mainly via activation of ET-1 receptor type A (ETAR), has become a mainstay of PAH therapy (5,6). Increased angiotensin II type-1 receptor (AT1R) activation was linked to disease progression and mortality in patients with
4
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 5 of 34
IPAH (7,8). Pharmacologic targeting of AT1R with losartan alleviated experimental PAH (9). In addition, dysregulated immunity and inflammation are also common in CTD-PAH and other PAH entities as suggested by inflammatory infiltration, growth factor expression in the remodeled pulmonary vasculature and increased cytokine and chemokine levels in the circulation (10). However, heterogeneity in therapeutic responses asks for a more precise definition of pathophysiological differences among PAH entities. Therefore new diagnostic and prognostic biomarkers will be critical to identify patient subsets that are likely to benefit from targeted therapies. Recently, we have identified simultaneous presence of functional autoantibodies (Abs) against AT1R and ETAR in patients with SSc (11) and linked them to increased prevalence of SSc-related vascular and fibrotic complications and higher risk for SScrelated mortality. Both antibodies induce phosphorylation of Erk1/2 and TGF-β expression in endothelial cells and hence activate endothelial cells to secrete cytokines (11,12). In addition, the antibodies mediate decreased wound repair by endothelial cells and act on immune cells (12,13). Based on these data and inducible renal obliterative lesions after transfer of anti-AT1RAbs into animals (14), we hypothesized an involvement of anti-AT1RAbs and anti-ETAR Abs in the pathogenesis of CTD-, especially SSc-related PAH. After investigating their presence in different PAH entities, we studied their potential relevance as predictive and prognostic biomarkers for SSc-PAH, and their contribution to its pathogenesis. Some of the results of these studies have been previously reported in the form of an abstracts (15,16)
Material and Methods Patients. 5
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 6 of 34
Anti-AT1R/ETAR Ab levels were first measured in serum samples from 62 consecutive SSc-PAH patients admitted to the Rheumatology department of the Charité. Additional serum samples were obtained from 182 patients with different pulmonary hypertension (PH)/PAH entities followed at the Department of Respiratory Medicine (n=115) at the University Hannover, Germany, and at the Justus-Liebig University Giessen, Germany (n=67). 62 of these 182 patients had idiopathic PAH (IPAH), 31 were classified as chronic thromboembolic PH (CTEPH), 14 patients had PH due to congenital heart disease (CHD), 19 SSc-PAH and 56 patients were PAH patients associated with CTD, which was not further specified. Combining all cases from the three centers, the cohort included 110 CTD-PAH patients, of which 81 had SSc-PAH. SSc patients fulfilled the new EULAR/ACR criteria for systemic sclerosis (17). In addition, 253 consecutive SSc patients entering our outpatient clinic between 2006-2011, without right heart catheter (RHC)-proven PAH and naive to the use of ETAR or AT1R blockers, were tested for anti-AT1R and anti-ETAR antibody levels and were prospectively observed. The epidemiologic data of the patients are shown in Table 1. Under clinical routine conditions, patients were screened for PAH at least in one-year intervals by assessment of WHO functional class, echocardiography, lung function with detection of predicted diffusion capacity for carbon monoxide (DLCOSB), and, during the last years, also by detection of N-terminal pro-brain natriuretic peptide (NT-proBNP) levels. Only if PAH was suspected, RHC was performed, using a mean pulmonary artery pressure ≥25 mm Hg and a pulmonary capillary wedge pressure ≤ 15mmHg at rest for PAH diagnosis (18). All patients were examined in experienced PAH centers. Pulmonary vascular resistance (PVR) and cardiac index (CaI) were determined according to the local guidelines, yet not reported for all patients. The study was approved by the local ethics committee of each contributing center and included written consent of each patient. 6
Copyright © 2014 by the American Thoracic Society
Page 7 of 34
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Detection of anti-AT1R and anti-ETAR Abs. Serum antibody levels were measured by an ELISA (CellTrend GmbH, Germany, and since 2012 One Lambda, Inc., U.S.A.) as previously described (11). Briefly, polystyrene plates were coated with extracts from cells overexpressing the human AT1R or the human ETAR in native conformation, respectively. Detection of the Ab levels was performed by staff unaware of patient data.
Isolation of human anti-AT1R and anti-ETAR Abs containing IgG and control IgG All ex-vivo and in-vivo experiments were conducted with endotoxin-free IgG isolated from serum samples as previously described (11,12). For in vitro experiments, final concentration was 1 mg/mL if not stated otherwise. Purified pooled IgG from 14 SSc patients with high levels of anti-AT1R Abs (mean concentration 21.8 U) and of high anti-ETAR Abs (mean concentration 17.91 U) was used for animal experiments. IgG from 15 healthy donors (HD) served as control (mean concentration 3.85 U for antiAT1R Abs and 2.5 U for anti-ETAR Abs, respectively). Real time fluorescence imaging of endothelial calcium. Isolated lungs from male Sprague-Dawley rats (350 to 450 g body weight) were prepared (n=9 per group), perfused and ventilated as described (19, 20). For imaging of endothelial cytosolic Ca2+ concentration ([Ca2+]i), lungs were positioned under an upright fluorescence microscope and superfused with normal saline at 37°C. Fura-2acetoxymethylester (Fura-2 AM, 5 µmol/L), which is de-esterified intracellularly to the Ca2+ sensitive dye fura-2, was loaded to lung endothelial cells for 20 min via a microcatheter. Fura-2 fluorescence in endothelial cells of subpleural lung single venular capillaries of 15-30 µm in diameter was excited as described (21). 7
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 8 of 34
Endothelial [Ca2+]i was determined from the 340/380 ratio based on a Kd of 224 nmol/L and appropriate calibration parameters (21). Following baseline recordings, purified IgG from either SSc patient or from HD sera was directly infused in HEPES dextran buffer into the imaged microvessels via a microcatheter over 30 min. The AT1R blocker valsartan (gift by D. N. Müller, Max Delbrück Center for Molecular Medicine Berlin, Germany) and ETAR blocker sitaxsentan (provided by Pfizer Deutschland GmbH) were each administered 10 min prior to IgG infusion in respective experiments (1 µmol/L each). All animal studies were approved by the local government authorities. Small vessel myography Third to fourth-generation intralobar pulmonary rat artery ring segments from Sprague-Dawley rats (external diameter 100-300 µm, length 2 mm) were freed from connective tissue and mounted in a small vessel myograph (DMTDanish Myotechnology, Aarhus, Denmark). Contractile responses were measured as described (22). As ET-1 induced contraction in rat small pulmonary arteries is insensitive to either BQ-123, a selective ETAR antagonist, or the selective ETBR antagonist BQ-788 (22), we used the combined ETA/BR-antagonist bosentan (gift from Actelion, Basel, Switzerland). The vessels were incubated with 10 µM bosentan for 20 min after which IgG (0.1 mg/mL) was added for 30 min. Thereafter, vessels were exposed to cumulatively increasing concentrations of Ang II. After repeated washing, the vessels were treated following the same protocol substituting valsartan 10 µM for bosentan and ET-1 for Ang II. Contractile responses to Ang II and ET-1 were measured as a percentage of the response to 60 mM K+, the effects were reversible by the respective receptor blockers. Antibody transfer experiments 8
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 9 of 34
For short-term experiments, female 7 week-old C57Bl/6J mice received a single intravenous injection of 800 µg IgG dissolved in 100 µL NaCl 0.9%. Lungs were harvested for histological and immunological analyses at day 7. For long-term experiments, mice received repeated IgG injections of 200 µg IgG dissolved in 100 µL NaCl 0.9% at days 1, 17, 30 and 93. Organs were harvested at day 100. Body weight (day 30, 58, 78 and at read out day) and rectal temperature control (at read out day; BAT-12 Microprobe; Physitemp Instruments, Clifton, NJ) were monitored.
Histology and immunohistochemistry For immunohistochemistry, lungs were removed and immediately snap frozen in O.C.T.TM medium (Tissue-Tek®, Sakura, Europe). Sections of 3 µm were prepared and stained with a polyclonal anti-alpha smooth muscle actin (α-SMA) antibody (Abcam, U.K.) and with an anti-human IgG F(ab')2-fragment (mouse absorbed, Dianova, Germany) to detect human IgG deposition. DAPI (Sigma Aldrich, Germany) was used for nuclear DNA staining. For histological assessment, lungs were fixed in 4% paraformaldehyde at room temperature for 24 hours and embedded in paraffin using standard procedures. Anti-CD31 antibody was used (clone SZ31, Dianova, Germany) to stain endothelial cells. Hematoxylin and eosin (H&E) staining was performed to visualize cellular organization. Signals were detected by light microscopy (Axioplan, Carl Zeiss MicroImaging GmbH, Germany) as well as by immunofluorescence (LSM710, Carl Zeiss Microscopy GmbH, Germany) and analyzed by ZEN 2009 Light Edition software (Carl Zeiss Microscopy GmbH, Germany).
9
Copyright © 2014 by the American Thoracic Society
AJRCCM Articles in Press. Published on 02-September-2014 as 10.1164/rccm.201403-0442OC
Page 10 of 34
Statistics Comparisons between groups were analyzed by Mann-Whitney U test for continuous variables. Kaplan-Meier analyses were performed to investigate the onset of PAH development and mortality after antibody detection. Significance was determined by Log-rank test. Receiver-operating-characteristic (ROC) curves were plotted to determine area under the curve (AUC), sensitivity, and specificity using calculated distance to 0.1. Small vessel myography data were analyzed by analysis of variance (ANOVA). Results are expressed as means ± SEM, in myograph experiments n stands for the number of the studied pulmonary artery ring segments. Differences were considered as statistically significant at a p value ≤0.05. Graphpad Prism 5.0 as well as IBM SPSS 21.0 were used for statistical analyses.
Results Anti-AT1R and anti-ETAR antibodies are higher in SSc- and CTD-PAH than in other forms of PH/PAH Epidemiologic and hemodynamic data of the cohorts studied are shown in Table 1. PH/PAH groups did not differ significantly among centers. Higher percentage of female patients and shorter duration of PAH were observed in CTD-PAH and SScPAH. Mean pulmonary arterial pressure (mPAP) as detected by RHC in these two groups was significantly lower as compared to IPAH (p
