Crosstalk between coagulation and complement activation promotes cardiac dysfunction in arrhythmogenic right ventricular cardiomyopathy

Aims: We previously found that complement components are upregulated in the myocardium of patients with arrhythmogenic right ventricular cardiomyopathy (ARVC), and inhibiting the complement receptor C5aR reduces disease severity in desmin knockout (Des-/-) mice, a model for ARVC. Here, we examined the mechanism underlying complement activation in ARVC, revealing a potential new therapeutic target. Methods: First, immunostaining, RT-PCR and western blot were used to detect the expression levels of complement and coagulation factors. Second, we knocked out the central complement component C3 in Des-/- mice (ARVC model) by crossing Des-/- mice with C3-/- mice to explore whether complement system activation occurs independently of the conventional pathway. Then, we evaluated whether a targeted intervention to coagulation system is effective to reduce myocardium injury. Finally, the plasma sC5b9 level was assessed to investigate the role in predicting adverse cardiac events in the ARVC cohort. Results: The complement system is activated in the myocardium in ARVC. Autoantibodies against myocardial proteins provided a possible mechanism underlying. Moreover, we found increased levels of myocardial C5 and the serum C5a in Des-/-C3-/- mice compared to wild-type mice, indicating that C5 is activated independently from the conventional pathway, presumably via the coagulation system. Crosstalk between the complement and coagulation systems exacerbated the myocardial injury in ARVC mice, and this injury was reduced by using the thrombin inhibitor lepirudin. In addition, we found significantly elevated plasma levels of sC5b9 and thrombin in patients, and this increase was correlated with all-cause mortality. Conclusions: These results suggest that crosstalk between the coagulation and complement systems plays a pathogenic role in cardiac dysfunction in ARVC. Thus, understanding this crosstalk may have important clinical implications with respect to diagnosing and treating ARVC.

vivo evaluation and no history of myocardial disease. Endomyocardial biopsy specimens (EMB) derived from ARVC patients were also used to perform the above experiments.
Secondly, 87 ARVC patients, 39 DCM patients (both from the outpatients at early or intermediate stage), 48 pulmonary arterial hypertension (PAH) patients, and 79 healthy volunteers were enrolled to measure the circulating complement factor levels. Their clinical data are presented in Table 1. All the ARVC patients enrolled were diagnosed according to 2010 revised Task Force Criteria [3] and the DCM patients according to the diagnostic criteria of Mestroni et al. [4] To rule out the effects of genetic mutations, we have specially filtered out those DCM patients without gene mutation and family history (DCM patients used as a positive control of heart failure only). The clinical characteristics of the DCM cohort used to measure sC5b9 levels (N=39) are presented in Table 1. The characteristics of transplanted DCM patients (n=16) used in WB/qPCR analysis are presented in Supplementary Table S3. Baseline demographics and medical details of patients, including symptoms, NYHA class, comorbidities, arrhythmia occurrence, noninvasive and invasive examination (ECG, Echocardiography, CMR, and biochemical tests) were obtained retrospectively from the chart review by persons blinded to the assay results. Age of onset was defined as the age when the first symptoms or signs most likely to be attributable to the disease occurred. Detailed family history was acquired by ARVC genetic counseling, as well as comprehensive mutation testing. The patient questionnaires or telephone interview regarding information about the adverse cardiovascular events, including the all-cause death, heart transplantation or in waiting list, ventricular fibrillation (VF) and ICD discharge, were updated regularly after patient enrollment.

Genetic screening
The whole-genome DNA was extracted from peripheral blood cells of ARVC patients. Targeted next-generation sequencing was performed based on the Hi-seq2000 platform (Illumina, USA).
GAPDH was used as the internal reference of the protein expression. The clinical baseline characteristics of HTx ARVC and DMC patients involved in WB analysis are presented in supplementary Tables 2 and 3.

RT-qPCR
The relative gene expression levels in human cardiac samples (RV) were detected by a SYBR green based PCR kit (Applied Biosystems, Foster City, CA, USA). The specific protocols for RNA extraction and reverse transcription were described in a previous study. The ribosomal protein L5 (RPL5) was used as the internal reference gene of the qRT-PCR. The ΔΔCt method was used for the calculation of relative gene expression levels. The primer pairs for qRT-PCR were; C5aR: forward 5'-TATCCACAGGGGTGTTGAGG -3' and reverse 5'-GCCCAGGAGACCAGAACAT -3'.

Autoantibody detection
For the detection of autoantibody in plasma, indirect immunofluorescence (IIF) was performed on the unfixed fresh-frozen sections of normal human ventricular tissue as previously reported [7], using a 1/10 dilution of the plasmas from healthy controls and patients with ARVC and DCM.
In order to evaluate the presence of autoantibodies in mice sera, immunofluorescent analysis of 12μm thick heart sections was performed eventually as described in the previous paragraph (Immunofluorescent/immunohistochemical staining). The antibodies included: Des-/-and wt mouse sera in a 1:100 dilution and anti-desmin (H76, 1:50, Santa Cruz Biotechnology, Dallas, Texas, USA) and anti-N-cadherin (Alomone labs, Israel 1:100).

Multiplex immunohistochemistry
In order to detect the co-expression of multiple proteins such as complement, coagulation and IgG in the myocardium, using a single section, a multiplex immunohistochemistry array was

Assessment of replacement index
Routine histologic procedures and staining with hematoxylin-eosin (H-E) and Masson's trichrome were performed in paraffin-fixed sections. To assess the "replacement index," which represents the areas of cardiac tissue replaced by fibrosis and/or calcification and/or infiltration of inflammatory cells, the hearts were sectioned across the longitudinal axis (at the midline), and eight sections per heart (including the right and left ventricular free walls and the septum as indicated in Figure 5B) were analyzed by two independent, blinded observers as previously described [8]. Sections were graded from 0 to 4 as follows: 0= no tissue injury, 1=one or two foci of limited replacement, usually in the right ventricle; 2=two to three foci in two different areas of the cardiac tissue; 3=multiple foci of extended replacement, usually in all ventricular compartments (right and left ventricular free walls and the septum); 4=multiple foci of extended and diffuse fibrosis occupying more than 50% of the myocardial surface in a given section. A Leica DMRA2 microscope was used for all bright field microscopy, and photographs were taken using a Leica DFC 500 camera and the Leica Application suite V3.6 program (Leica Microsystems, Wetzlar, Germany).

Treatment of young mice with the thrombin inhibitor lepirudin.
Des -/or Des -/-C3 -/mice were treated with injections into the peritoneal cavity of lepirudin (Refludan, Hoechst Marion Roussel, Athens Greece), a highly specific direct inhibitor of thrombin. At day 16 after birth, before the onset of the acute inflammatory reaction, Des -/animals were injected with 20mg (per kg body weight) of lepirudin, every 12 hrs for 5 consecutive days. Control Des -/animals were injected with PBS. After the treatment, the animals were sacrificed and cardiac tissue sections were analyzed for infiltration of inflammatory cells and/or fibrosis and/or calcification as described above (replacement index).

Echocardiography
Echocardiographic experiments were performed in male mice of all genetic permutations at the age of 4 and 12 months, using an ultrasound system (Vivid 7; GE Healthcare, USA) with a 13-MHz linear transducer, as previously described [8]. Mice were anesthetized with an intraperitoneal injection of 50 to 100 mg/kg ketamine.
Mouse serum was analyzed for C5a/C5adesArg concentrations, by the Biacore 2000 instrument (GE Healthcare, Munich, Germany) using the anti-C5a antibody (R&D Systems, clone 295108, MAB21501) as previously described [8].          Both are composed of serine proteases that are activated through partial cleavage by an upstream enzyme. The coagulation cascade is divided into TF pathway and contact activation.
The TF pathway could be activated by TCC, trauma, and some cytokines. Both pathways will merge at factor X level, which will generate thrombin. Thrombin could activate platelets and consequently induce platelet polyphosphate degranulation, thus consequently triggering contact activation. Activated platelets could additionally initiate the complement classical pathway by participating in C3 cleavage. Moreover, platelets contribute to the amplification of complement through the phosphorylation of C3b, which prolongs its life span. FXIIa can activate the classical complement pathway and kallikrein can activate both C3 and C5. Thrombin and plasmin can independently activate both C3 and C5. The final step of the coagulation process, catalyzed by thrombin, requires partial cleavage of soluble fibrinogen and polymerization to insoluble fibrin.
Complement is activated through the classical (DAMPs, immune complexes and PAMPs), lectin (PAMPs and apoptotic cells) or alternative (PAMPs) pathways all leading to C3 activation through C3 convertase which cleave C3 to C3a and C3b. C3b contributes to the formation of C5 convertase, which cleaves C5 to C5a and C5b. Subsequently, C5b will lead to TCC formation, which apart from cause lysis of microorganisms could also lyse host cells, releasing DAMPs.
TCC will induce TF pathway and platelet activation, and will enhance coagulation by negatively charged phospholipid surfaces. C3a and C5a anaphylatoxins generated by complement activation will recruit and activate leukocytes, as well as induce platelet activation and aggregation, inducing thrombosis and inflammation, which are known to further enhance coagulation.
Lepirudin: specific thrombin inhibitor. Blue: Components of complement system, Green: Components of coagulation/ fibrinolysis cascades, Orange: Initiatory components of pathways.
Abbreviations: DAMPs, damage-associated molecular patterns; MASP-2, mannose-binding protein-associated serine protease 2; MBL, mannan-binding lectin; PAMPs, pathogen-associated molecular patterns; TCC, terminal complement complex; TF, tissue factor. Figure S2      Cardiomyocytes degeneration is observed in the outer subepicardial myocardium of desmin-null mice which is extended towards the endocardium (A and C, red arrows), for comparison see WT, panels B and D (paraffin cardiac section, 24 days old animals, HE staining). This extended cardiomyocyte necrosis is followed by an inflammatory response and subsequently by injury repair and replacement of the necrotic fibers by dense collagenous scar (E, manson trichrome staining, 1.5 months old animals, collagen is blue).