Mitochondrial Ultrastructural Alterations and Declined M2 Receptor Density Were Involved in Cardiac Dysfunction in Rats after Long Term Treatment with Autoantibodies against M2 Muscarinic Receptor

Background Previous studies showed that autoantibodies (M2-AA) against the second extracellular loop of M2 muscarinic receptor (M2AChR-el2) from dilated cardiomyopathy (DCM) serum could induce DCM-like morphological changes in mice hearts. However, the effects of M2-AA on the cardiac function during the process of DCM and the potential mechanisms are not fully known. The present study was designed to dynamically observe the cardiac function, mitochondrial changes, and M2 receptor binding characteristics in rats long-term stimulated with M2-AA in vivo. Methods M2-AA-positive model was established by actively immunizing healthy male Wistar rats with synthetic M2AChR-el2 peptide for 18 months. Meanwhile, vehicle group rats were administrated with physiological saline. The change of mitochondrial membrane potential (ΔΨm) was detected by radionuclide imaging. The ultrastructure of mitochondria was observed under electron microscopy. The M2 receptor binding characteristics were determined by radioactive ligand binding assay. Results After immunization for 12 months, compared with vehicle group, M2AChR-el2-immunized rats showed decreased myocardial contractility and cardiac diastolic function evidenced by declined maximal rate of rise of ventricular pressure and increased left ventricular end-diastolic pressure, respectively. Additionally, mitochondrial swelling and vacuolation were observed. At 18 months, M2AChR-el2-immunized rats manifested significant decreased cardiac systolic and diastolic function and pathological changes such as enlargement of right ventricular cavity and wall thinning; and the mitochondrial damage was aggravated. Furthermore, the M2 receptor maximum binding capacity (Bmax) of the M2AChR-el2-immunized rats significantly decreased, while the M2 receptor dissociation constant (Kd) was increased. Conclusions Our study suggested that long-term stimulation with M2-AA leaded to the ventricular dilatation and gradual deterioration of cardiac dysfunction. Mitochondrial damage and the down-regulation of M2 receptor density and affinity may be involved in the process.


Introduction
Idiopathic dilated cardiomyopathy (DCM), a refractory and primary myocardial disease, is a major cause of death in patients with heart failure. However, the pathogenesis of DCM is not fully understood. Autoimmunity, viral infection, and genetic predisposition have long been considered as major causes for idiopathic DCM, among which autoantibodies have become one of the cornerstones in mediating the pathophysiology in these patients [1].
It has long been known that there are high titers of autoantibodies against the second extracellular loop of the M 2 muscarinic receptor (M 2 -AA) in sera of patients with DCM [2]. Our previous study showed that both M 2 -AA positive rate and titer in the sera of 53 DCM patients were higher than those in 408 healthy subjects [3]. A previous study demonstrated that M 2 -AA could recognize the second extracellular loop of the M 2 muscarinic receptor (M 2 AChR-el2) and stimulate the muscarinic receptor on rats' heart membrane [4]. Recent studies also indicated that M 2 -AA was responsible for DCM-like morphological changes in mice that were immunized with M 2 receptor for 8 weeks [5]. However, whether long term exposure to M 2 -AA can worsen cardiac function during progression of DCM remains largely unknown.
Metabolic disorder is one of the key mechanisms of cardiac dysfunction in DCM, which is mainly caused by mitochondrial injury. Existing evidence suggests that mitochondrial injury in a rat model of DCM is responsible for cardiac dysfunction [6]. A clinical study has also shown that mitochondrial dysfunction could aggravate the pathogenicity of DCM [7]. These reports suggest that mitochondrial injury plays a detrimental role in the pathogenesis of DCM. However, whether long-term stimulation by M 2 -AA could contribute to mitochondrial injury in myocardium of DCM is still undetermined. M 2 receptor is an essential molecule in cardiac function regulation (i.e. negative inotropy and chronotropy), which displays contrarian functions to those of β 1 -adrenoceptors. M 2 receptor activation also plays important roles in regulation of mitochondrial function [8]. Studies have shown that like other G protein-coupled receptors, the expression and affinity of M 2 receptor can be up-or down-regulated by a multitude of factors, such as agonist, antagonist [9], and stress [10]. Additionally, abnormal receptor function may cause weakened cardio-protective effects of M 2 receptor. A previous study reported that M 2 -AA isolated from patients with DCM can directly activate M 2 receptor and exhibit agonist-like effects [11]. However, whether changes of binding characteristics of M 2 receptor are responsible for the M 2 -AA-induced cardiac dysfunction are still unknown.
In the present study, healthy rats were immunized with the synthetic peptide of M 2 AChR-el2 to clarify the pathological roles of M 2 -AA in DCM. We aimed to determine the progress of cardiac dysfunction, changes of mitochondrial membrane potential and ultrastructure, and M 2 receptor binding characteristics in the M 2 -AA positive rat model.

Peptide Synthesis
The peptide-corresponding to the sequence of the second extracellular loop of the human muscarinic receptor-2 (residues 168-193, [4] was synthesized by Shanghai Biochemical Institute of Chinese Medical Science Academy using an automated peptide synthesizer. The peptides were purified by reverse phase HPLC on a C-18 column (Vydac, Hesporia, CA) and assessed on an automated amino acid analyzer (Beckman Instruments, CA).

Immunization of animals
Ten-week old healthy male Wistar rats, lacking congenital M 2 -AA, were divided into two groups. The M 2 -AA group (n = 30) was immunized with a mixture of synthetic peptides corresponding to M 2 AChR-el2 and immunoadjuvants once a month for 18 months, and the vehicle group (n = 30) was immunized with a mixture of physiological saline and immunoadjuvants. The immunizing antigen was dissolved in 0.1 mol/L Na 2 CO 3 solution (pH = 11.0). In the M 2 -AA group, rats were immunized based on body weight with 0.4 mg/kg of M 2 AChR-el2 peptide, which was emulsified in an equal volume of Complete Freund's adjuvant (CFA) and injected subcutaneously. After a month, a booster injection of a mixture of the same dose of M 2 AChR-el2 antigen peptide with equal volume of Incomplete Freund's adjuvant (IFA) was administered subcutaneously. CFA contains paraffin oil, mannide monooleate and killed mycobacterium tuberculosis, which can attract macrophages to the injection site and enhance immune response. On the other hand, IFA contains paraffin oil, mannide monooleate but lacks bacteria, therefore can minimize the side effects of adjuvant, and was thus used for the immunization boosts. The rats of the vehicle group received the mixture of physiological saline and adjuvant according to the procedure as described above. All rats were bled one day before each booster injection and sera collected. At the end of immunization, rats were executed to collect the hearts and sera samples. Carbon dioxide (CO 2 ) inhalation was used for euthanasia of rats. The study conformed to AVMA Guidelines on Euthanasia and the Guide for the Care and Use of Laboratory Animals protocol published by the Ministry of the People's Republic of China and were approved by Shanxi Medical University Committee on Animal Care. During the study period, the rats were fed with rat chow and water ad libitum.

Streptavidin-enzyme-linked immunosorbent assay (ELISA)
Blood samples were collected from the caudal veins pre-immunization every month. The sera was separated by centrifugation at 3000 g for 15 min, aliquoted and frozen at -20°C until analysis. The titers of M 2 -AA were dynamically measured by ELISA as reported previously [3], and the results were expressed as optical density (OD) values. The OD value was measured at 405 nm using an ELISA reader (Spectra Max Plus, Molecular Devices, Sunnyvale, California, USA). We also calculated the positive/negative (P/N) ratio [(the OD of sample-the OD of blank control) / (the OD of negative control-the OD of blank control)] of each sample. Negative control samples were prepared as follows: 100 sera samples from healthy rats with an OD value of less than 2.5 times the background OD were pooled and centrifuged at 1500 rpm for 10 min, and the supernatants were then divided into small aliquots and stored until use. Samples with a P/ N value of 2.1 were considered as M 2 -AA-positive, and samples with a P/N value of 1.5 were considered as M 2 -AA-negative [12].

Measurement of cardiac function in vivo
The rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and the left ventricle was catheterized with a Statham gauge (PE50) through the right common carotid artery. Then left ventricular function was measured and the data were recorded by the PowerLab data acquisition system (AD Instruments). The following primary and derived variables were continuously recorded on a beat-to-beat basis: left ventricular systolic pressure (LVSP), left ventricular end diastolic pressure (LVEDP), and maximal rate of rise and decline of ventricular pressure (±dp/ dt max ).

Cardiac morphological and histopathological analysis
Histopathologic examination of the hearts was performed after routine fixation and paraffin embedding. The wall thickness, cavity dimension, and cavity area were measured at the midsections of each heart by a computerized image analysis system (MIAS-300, Sichuan, China).
Heart tissues were fixed with glutaraldehyde for transmission electron microscopy analysis. Briefly, small pieces of heart tissue were fixed with 2.5% glutaraldehyde in 0.1 mol/L phosphate buffer (pH 7.4) for 2 h at 4°C. After being washed with phosphate buffer, the specimens were post-fixed with 1% osmium tetroxide for 2 h at 4°C. Following dehydration with a graded series of acetone, the specimens were infiltrated with epoxies 618 ultra-thin sections (50 nm thicknesses) and cut using a LKB ultra-microtome IV after polymerization at 60°C, stained with uranyl acetate and lead citrate, and finally observed using a 100-CX transmission electron microscope (Japan).

Myocardial radionuclide imaging
Myocardial radionuclide imaging was used to detect the mitochondrial membrane potential (ΔCm) [13]. Myocardial uptake of 99m Tcmethoxyisobutylisonitrile ( 99m Tc-MIBI) was measured as heart and upper mediastinum (H/M) count ratio. The operation was performed as previously described [14]. Briefly, an 18.5-MBq dosage of 99m Tc-MIBI was slowly injected via rat tail vein and planar and single photon emission computed tomography views were obtained. Conventional gamma scintillation camera was utilized for image processing (Mobile Radioisotope Camera, Model BHP6602, Hamamatsu, Japan).

Radio-ligand receptor binding assay (RBA)
Rats were sacrificed at 12 and 18 months after the initial immunization. Tissue preparation was performed using methods previously reported by Vandermolen et al. [15]. Briefly, left ventricles from each group were collected and rapidly frozen in liquid nitrogen, and then stored at -70°C until sectioning. Ventricular tissue was trimmed to a dimension of 5×5 mm and embedded in OCT media (Miles, USA). The tissue sections were consecutively cut 40 μM in thickness in a -18°C cryostat microtome (Ryocut 1800) with their serial order being noted. One of five consecutive sections, which were made into a set, was used to measure tissue protein via Bradford method to ensure that consecutive serial sections did not vary in size [16]. [ 3 H]quinuclidinyl benzylate ([ 3 H]-QNB, 51 Ci/mmol, Amersham International, UK) was used as radioligand for M 2 -muscarinic receptor. Saturation-binding isotherms were obtained by incubating sections for 1h with varying concentrations of [ 3 H]-QNB (0.125 nM-8 nM) at 26°C, and experiments performed in duplicate. The reaction was terminated by dipping slides into 0°C Tris buffer for 22 min. The slides were rapidly dried and each set of four consecutive sections was scraped from the slides with a blade into tubes respectively, and individual scintillation vials with 5 ml scintillation fluid (PPO 4 g, Popop 100 g, dissolved in 1000 ml xylol) was added to each vial and stabilized overnight and counted in a liquid scintillation counter (Beckman LS-3801) for muscarinic binding. Non-specific binding was determined in the presence of 10 -4 mol/L atropine and amounted to less than 20% of total binding. Specific binding was obtained by subtracting non-specific binding from total binding. All binding data are given as specific binding. The saturation binding parameters B max and K d were determined using the Prism 2.01 Programs.

Statistical analysis
All measured values are expressed as mean ± SEM. Data was analyzed by unpaired Student's t test or ANOVA where appropriate. The analyses were carried out using SPSS 13.0 software. Statistical significance was set at P < 0.05.

M 2 -AA-positive rat models were successfully established and resulted in dilated cardiomyopathy-like morphological features
To determine the effect of long-term existence of M 2 -AA on cardiac structure and function in vivo, a M 2 -AA-positive rat model was established by active immunization with synthetic M 2 AChR-el2 antigen peptides for 18 months. Serum levels of M 2 -AA were detected by ELISA. In the immunization group, three rats (No. 3, 18 and 29) were euthanized after the 2 nd month for failing to generate M 2 -AA (P/N1.5). One rat (No.15) was euthanized due to inflammation at inoculation site after the 3 rd month. In the vehicle group, two rats (No.4, 16) were euthanized because of malocclusion induced-excessive weight loss and inflammation at inoculation site after the 2 nd and 3 rd month, respectively. One rat (No.30) died unexpectedly at the 4 th month. Therefore, there were 26 rats in the M 2 AChR-el2-immunized group and 27 rats in the vehicle group. During the immunization period, the M 2 -AA were positive (P/N2.1) in the M 2 AChR-el2-immunized group and maintained at a high level (Fig 1A), which peaked at the 11 th month (OD value, 1.76±0.14 vs. 0.39±0.07, P<0.01) and then declined gradually. The average titer of the antibodies during the 18 months was 1:1168.3±29.6. The Mg 2 -AA remained negative (P/ N1.5) in the vehicle group throughout the observation period. The level and variance trends of M 2 -AA in the M 2 AChR-el2-immunized group were matched with the detected result of clinical patients, which indicates that this autoimmunity animal model was able to imitate clinical pathological progress.
The morphology of the hearts excised from the rats was observed post-immunization. As shown in Fig 1B, hearts from the M 2 -AA-positive immunization group showed moderate dilatation in the ventricles, including wall thinning. The dilatation was more apparent in the right ventricle (RV) than the left ventricle (LV), the RV/LV ratios of hearts from the M 2 -AA group increased significantly in cavity dimension and cavity area, while the wall thickness decreased significantly (Table 1) compared with the hearts from rats in vehicle group.

Long-term active immunization with M 2 AChR-el2 peptide induced cardiac dysfunction in rats
To observe whether long-term active immunization with the M 2 AChR-el2 peptide influenced cardiac function in rats, hemodynamics parameters to evaluate left ventricular systolic and diastolic function were measured at 6, 12, and 18 months post-immunization. As shown in Fig 2, compared with the vehicle group, the left ventricular systolic pressure (LVSP) in the M 2 AChR-el2-immunized group had no significant change at the 12 th month, but markedly decreased at the 18 th month (Fig 2A). The left ventricular end-diastolic pressure (LVEDP) in the M 2 group increased at the 12 th month (1.5±1.3 KPa vs.  KPa/s, P<0.01) (Fig 2D). These results indicated that the cardiac function underwent no change in the first 6 months after M 2 AChR-el2 immunization; however, diastolic function and myocardial contractility of LV worsened at the 12 th month, and both the systolic and diastolic function of LV resulted in serious deterioration at the 18 th month post-immunization.
Active immunization rats with M 2 AChR-el2 peptide caused a decrease in myocardial ΔΨm and structural changes of mitochondria Mitochondrial membrane potential (ΔCm) is an important parameter of mitochondrial function [17]. Given that approximately 90% of myocardial 99m Tc-MIBI is localized within mitochondria and myocardial uptake of 99m Tc-MIBI is dependent on ΔCm, the current study used radionuclide 99m Tc-MIBI myocardial perfusion imaging method to determine the alteration in ΔCm as reported previously [14]. The regions of interest positioned over the heart (H) and upper mediastinum (M) represent the 99m Tc-MIBI uptake in the heart and mediastinum, respectively, and H/M uptake ratio can reflect the myocardial ΔCm changes. As shown in Fig  3A, at the 12 th month post initial immunization, 99m Tc-MIBI uptake ratio of heart to mediastinum in M 2 -AA group rats decreased compared to vehicle group rats (1.63±0.05 vs. 2.07±0.09, P<0.05), which indicated that long-term exposure of M 2 -AA in rats could reduce nuclide uptake of heart and declined myocardial ΔCm. Along with the extension of immunity time, mitochondrial ultrastructural damage occurred in the M 2 AChR-el2-immunized group rats at 12 th month after initial immunization. From electron micrographs, myocardial specimens from the M 2 -AA group showed irregular distribution of mitochondria, varying degrees of mitochondrial swelling, cristae disappearance, and vacuolar degeneration ( Fig 3B); moreover, the pathologic changes became more apparent at the 18 th month of immunization. Deposition of dense granules in sarcoplasm and focal-dissolved myofibrils were also observed at the end of the experiment (Fig 3C). On the other hand,

The influence of long-term stimulation of M 2 -AA on myocardial M 2 receptor binding characteristics in rats
To study the influence of myocardial M 2 receptor binding characteristics after long-term active immunization with M 2 AChR-el2 antigen peptide in rats, the largest combination capacity (B max ) and dissociation constant (K d ) of M 2 muscarinic receptor were determined by RBA at the 12 th and the 18 th month post initial immunization. As shown in Table 2, in the M 2 -AA group, the B max value reduced significantly at the 18 th month post M 2 AChR-el2-immunization compared to the vehicle group, and the K d value was up-regulated at the 18 th month timepoint compared to the vehicle group. These results indicated that the number of receptor sites Tc-MIBI uptake ratio of regions of interest positioned over the heart to mediastinum. Data are presented as mean ± SEM. n = 6/group. Alterations in the myocardial mitochondrial ultrastructure in two groups of rats after 12months (B) and 18 months (C) of immunization, respectively. Representative photomicrographs of the mid-papillary section of the heart from the vehicle group (panels B-I, C-II) and the M 2 group (panels B-II~III, C-II~IV) were shown. Panels B-I~III, C-II, C-IV were photographed at ×10,000 magnification and panels C-I and C-III were photographed at ×20,000 magnification. The micrographs in M 2 -AA group displayed the irregular distribution of mitochondria, swollen mitochondria, vacuolar degeneration (as the red arrow directed), focal dissolved myofibrils (as the blue arrow directed), and deposition of dense granules in the sarcoplasm (as the black arrow directed). were down-regulated by prolonged exposure to M 2 -AA; moreover, the remaining receptors were functionally declined.

Discussion
Dilated cardiomyopathy (DCM) is a cardiac disease characterized by ventricular cavity expansion and decline in contraction function. At present, etiology researches for DCM mainly focus on viral infection, cellular immunity, and autoimmunity; among which autoimmune mechanisms have received growing attention over the years [18]. The impact of autoimmune factors on heart disease has mainly focused on autoantibodies such as β 1 -AA and M 2 -AA [1]. In 1993, Fu et al. found that there was certain kind of autoantibody against M 2 AChR-el2 (M 2 -AA) in DCM patients, which suggested that autoimmune factors may participate in the development of DCM [2]. In our current study, a M 2 -AA positive rat model was established by method of active immunization with M 2 AChR-el2 and we found that: (1) long-term stimulation of M 2 -AA resulted in reduced myocardial contractility and diastolic function at the 12 th month, and both the systolic and diastolic dysfunction occurred at the 18 th month time-point; meanwhile, M 2 AChR-el2-immunized rats developed DCM-like pathological features, such as ventricular cavity expansion and wall thinning; (2) decreased mitochondrial membrane potential and mitochondrial ultrastructural damage were detected in M 2 -AA positive rats; and (3) the largest combination capacity (B max ) and affinity (1/K d ) of M 2 receptors decreased in myocardial tissue of active immunization rat. In summary, this study indicates that the elevated M 2 -AA in the body could lead to myocardial mitochondrial damage, which may be one of potential causes for the decline in cardiac function and the compensatory expansion of ventricle which is similar to clinical features of DCM. Furthermore, down-regulation of M 2 receptor density and affinity induced by M 2 -AA may also be involved in the process.
It has been shown that M 2 -AA, a subtype of IgG2a autoantibody [19], can bind to M 2 receptors on cell surface and play a role similar to receptor agonist [2]. Wallukat G et al. observed a negative chronotropic effect of M 2 -AA on myocytes via M 2 muscarinic receptor [20]. However, the pathophysiological roles of M 2 -AA in the development of DCM need further exploration. In the present model, long-term presence of M 2 -AA can lead to DCM-like morphological changes especially the right ventricular dilation, which is consistent with previous reports [5]. In addition, M 2 -AA can gradually deteriorate cardiac systolic and diastolic function. Cardiac catheterization is a classical method for detecting cardiac hemodynamics. The dp/dt refers to first-order differential of ventricular systolic pressure, which will generate the curve of pressure variation rate. The maximum of dp/dt (+dp/dt max ) appeared in the first half of the isovolumic contraction period when preload and afterload are almost constant. Therefore, +dp/dt max can be used as an important indicator to evaluate myocardial contractility under different functional statuses. Our study found that +dp/dt max significantly declined at the 12 th month after initial immunization, which means that M 2 -AA depleted the myocardial contraction force. However, LVSP which reflects the peak ventricular systolic pressure was still within normal range due to compensatory mechanisms. Until the end of the immunization, LVSP in M 2 -AA group was significantly decreased, indicating ventricular systolic dysfunction (Fig 2A). Ventricular diastole can be divided into two phases: active relaxation and passive stiffness. Ventricular active relaxation, mainly occurring in isovolumetric diastolic and rapid filling phases, needs to consume energy and can be represented by the change in diastolic pressure in the heart chamber per unit time (dp/dt). Ventricular passive stiffness mainly includes slow filling and atrial systolic phases during which no energy is needed. Either poor active relaxation or passive stiffness can induce elevated left ventricular end diastolic pressure (LVEDP). Additionally, LVEDP can be influenced by other factors, such as cardiac systolic force, heart rate, and intra-pericardial pressure, etc. Our study found that LVEDP increased at the 12 th month, and was further elevated at the 18 th month after initial immunization with M 2 AChR-el2 ( Fig  2B), indicating the gradual deterioration of cardiac diastolic function. The reasons for the diastolic dysfunction are complex, and may include a longer filling time, weaken myocardial contractility, delayed cross-bridge detachment and Ca 2+ dissociation from troponin C caused by energy shortage (evidenced by decreased-dp/d max in Fig 2D), myocardial remodeling, and limited myocardial compliance.
It is well known that there are abundant myofibrils and mitochondria in myocardial cells. Mitochondrial damage can lead to abnormal energy metabolism and cardiomyocyte deaths, both of which can lead to weakened cardiac contractility and relaxation. In this study, to detect whether mitochondria was involved in cardiac disorder caused by M 2 -AA, myocardial mitochondrial membrane potential (ΔCm) and ultrastructure changes were detected via radionuclide imaging and electron microscopy, respectively. Because of noninvasive examination, 99m Tc-MIBI perfusion imaging is widely used in clinical applications to diagnose and assess various cardiac diseases. As a free cationic complex, approximately 90% of 99m Tc-MIBI activity is associated with mitochondria in an energy-dependent manner in vivo. Previously, it has been shown that when mitochondrial membrane undergoes depolarization, uptake and retention of 99m Tc-MIBI by myocardium is inhibited. Hence the mitochondrial membrane condition can be reflected by the uptake rate of 99m Tc-MIBI [21]. Herein, we found that the ΔCm evidenced by 99m Tc-MIBI uptake decreased in M 2 -AA-positive rats at 12 months post initial immunization, which is indicative of impaired mitochondrial function [17]. Moreover, electron micrographs showed that myocardial mitochondria displayed irregular distribution, swelling, and vacuolation. These pathological mitochondrial perturbations became more severe at the end of the process of immunization. Numerous studies have demonstrated that swollen mitochondrion [22], vacuolar degeneration [23], and altered myocardial ΔCm [24] are widely used morphological and functional markers of abnormal mitochondria. These changes in mitochondria may lead to increased ROS formation, inhibition of mitochondrial respiratory chain, and eventually, cell death. While it is possible that attenuated cardiac systolic/diastolic function is due to abnormal energy metabolism from aberrant cellular activity as a result of mitochondrial damage, the exact mechanisms of M 2 -AA-induced mitochondrial damage remain to be studied. Given that cardiomyocytes entirely depend on aerobic oxidation for energy supply, identification of contributory factors that cause mitochondrial damage will undoubtedly be of great significance for myocardial salvage.
Previously, we showed that an autoantibody against the second extracellular loop of β 1adrenoceptor (β 1 -AA), contributed to the development of DCM [25]. Although both β 1 -AA and M 2 -AA have been shown to be involved in the development of DCM, they can lead to significantly different lesion characteristics. For example, (1) β 1 -AA-positive rat models established by active immunization exhibited elevated cardiac systolic and diastolic function at the 12 th month, which decreased significantly at the 18 th month after the initial immunization. On the other hand, cardiac dysfunction caused by M 2 -AA became more and more serious with the extension of immunization time. (2) Long-term existence of β 1 -AA caused mitochondrial condensation, while vacuolar degeneration represented the main feature of M 2 -AA-induced mitochondrial aberration. The data presented in the current study therefore suggests that M 2 -AA may trigger unique mechanism in cardiac injury. Moreover, there may be crosstalk between M 2 -AA and β 1 -AA, which could provide synergistic effects in causing declined cardiac function in DCM patients. It is also possible that the pathogenic role of M 2 -AA is prior to that of β 1 -AA. Notably, Stavrakis et al. recently demonstrated that M 2 -AA could neutralize the contractive effect of β 1 -AA on isolated Purkinje fibers [26], which further complicates the relationship between M 2 -AA and β 1 -AA. Therefore, the detailed interplay of β 1 -AA and M 2 -AA in DCM is worthy of further investigation.
M muscarinic receptors are cholinergic receptors that are widely present in humans. They are divided into three types, namely M 1 , M 2 , M 3 muscarinic receptor depending on their different expression sites and function. Among them, M 2 muscarinic receptor is mainly distributed in the heart and negatively regulates heart rate and contractile forces of cardiomyocytes, which balances with the roles of the β-adrenoceptor. In this study, specific ligands were bound directly to frozen heart tissue embedded on slides in order to measure maximal binding sites and affinity of specific receptors on cardiomyocytes in a manner that closely mimics in vivo condition. At the end of immunization, we found that M 2 receptor maximum binding capacity (B max ) decreased in M 2 -AA active immunization model of DCM, suggesting a decrease in the expression of M 2 receptors; meanwhile, the dissociation constant (K d ) standing for M 2 receptor-ligand binding affinity was increased, implying a declined function of the available receptors. Receptor endocytosis is a significant mechanism of receptor degradation which can lead to decreased density of membrane receptor [27]. Delaney K.A, et al. reported that upon agonist stimulation, M 2 receptor was internalized via clathrin-independent pathway and resulted in decreased receptor density [28]. In our study, long term stimulation of M 2 receptor by M 2 -AA may be the cause of decreased M 2 receptor density. The down-regulated affinity may be due to the changes of receptor conformation [29]. However, whether receptor endocytosis and conformational changes are critical mechanisms underlying reduced receptor binding ability need to be examined further. Additionally, associated signal pathways remain to be explored. However, studies have found that the inactivation of M muscarinic receptor plays an important role in cell apoptosis induced by mitochondrial injury or dysfunction [9]. This implies that decreased M 2 receptor binding ability may be involved in mitochondrial damage, which could further contribute to cardiomyocyte damage. Specifically, future studies need to examine whether M 2 -AA can directly induce mitochondrial injury through interfering with M 2 receptor.
Unexpectedly, a disturbance in the ratio of CD4 + and CD8 + T cells was also observed during the process of M 2 -AA-induced DCM in rats (S1 Fig). CD4 + and CD8 + are two subtypes of T lymphocytes which usually maintain a stable ratio under normal conditions, and often, increases or decreases in this ratio is indicative of autoimmune disorders. In this study, the CD4 + /CD8 + ratio increased during the three months post-immunization, and decreased gradually after that. The increased CD4 + /CD8 + ratio implied autoimmune hyperactivity during which more autoantibodies could be produced. This speculation was verified by the direct effect of M 2 -AA on the proliferation of T lymphocytes in vitro (S2 Fig). However, the decreased CD4 + /CD8 + ratio indicated the immune suppression. Taken together, long-term stimulation of M 2 -AA can result in more severe immune imbalance; a likely vicious circle which should be further explored in the future studies.

Study limitations
In this study, we have observed that long-term exposure to M 2 -AA results in DCM-like cardiac structural and functional changes, mitochondrial damage, and decreased M 2 -AA receptor density and affinity; however, more in vivo and in vitro experiments are needed in our further studies to understand the causal relationship between them and the potential mechanisms.