Effects of autoantibodies against β1-adrenoceptor in hepatitis virus myocarditis on action potential and L-type Ca2+ currents

Abstract

AIM: To investigate the effects of autoantibodies against β₁-adrenoceptor in hepatitis virus myocarditis on action potential and L-type Ca²⁺ currents.

METHODS: Fifteen samples of autoantibodies against β₁-adrenoceptor positive sera of patients with hepatitis virus myocarditis were obtained and IgGs were purified by octanoic acid extraction. Binding of autoantibodies against β₁-adrenoceptor to guinea pig cardiac myocytes was examined by immunofluorescence. Using the patch clamp technique, the effects on the action potential and I_Ca-L of guinea pig cardiac myocytes caused by autoantibodies against β₁-adrenoceptor in the absence and presence of metoprolol were investigated. Cell toxicity was examined by observing cell morphology and permeability of cardiac myocytes to trypan blue.

RESULTS: The specific binding of autoantibodies against β₁-adrenoceptor to guinea pig cardiomyocytes was observed. Autoantibodies against β₁-adrenoceptor diluted at 1:80 prolonged APD₂₀, APD₅₀ and APD₉₀ by 39.2%, 29.1% and 15.2% respectively, and only by 7.2%, 5.3% and 4.1% correspondingly in the presence of 1 μmol/L metoprolol. Autoantibodies against β₁-adrenoceptor diluted at 1:80, 1:100 and 1:120 significantly increased the I_Ca-L peak current amplitude at 0 mV by 55.87 ± 4.39%, 46.33 ± 5.01% and 29.29 ± 4.97% in a concentration-dependent manner. In contrast, after blocking of β₁-adrenoceptors (1 μmol/L metoprolol), autoantibodies against β₁-adrenoceptor diluted at 1:80 induced a slight increase of I_Ca-L peak amplitude only by 6.81 ± 1.61%. A large number of cardiac myocytes exposed to high concentrations of autoantibodies against β₁-adrenoceptor (1:80 and 1:100) were turned into rounded cells highly permeable to trypan blue.

CONCLUSION: Autoantibodies against β₁-adrenoceptor may result in arrhythmias and/or impairment of myocardiums in HVM, which would be mediated by the enhancement of I_Ca-L.

Key Words: $[Keywords]

INTRODUCTION

Viral hepatitis remains a worldwide public health problem[1-4] and is reported by the Chinese Ministry of Public Health as an infectious disease with the highest morbidity and mortality in China. Many complications are considered to be involved in viral hepatitis[5-17]. Hepatitis virus myocarditis (HVM) as a consequence of hepatitis virus infection, secondary to Coxackie virus myocarditis in morbidity in hepatitis virus predominated areas of China, displays severe cardiac manifestations in addition to liver impairment[18]. It was recently found that some cases of severe fulminant hepatitis virus myocarditis were accompanied by arrhythmias and/or cardiac injury. These cases were screened with a high prevalence of circulating autoantibodies against β₁-adrenoceptor (anti-β₁-receptor autoantibodies)[18], which have been recognized in idiopathic dilated cardiomyopathy (DCM) and Chagas’ cardiomyopathy[19,20]. Using the patch clamp technique, we examined the effects of anti-β₁-receptor autoantibodies from patients with HVM on the electrophysiological properties of cardiomyocytes so as to find the possible clinical significance of the autoantibodies.

MATERIALS AND METHODS

Samples collection

A peptide was synthesized according to the sequence of β₁-adrenoceptor[21]: H-W-W-R-A-E-S-D-E-A-R-R-C-Y-N-D-P-K-C-C-D-F-V-T-N-R and was used as an antigen in ELISA as described by our laboratory[22] for screening of anti-β₁-receptor autoantibodies in serum samples of patients diagnosed as HVM according to the Chinese diagnostic criteria for adult acute myocarditis approved in 1999[23]. Fifteen serum samples were selected.

Antibody purification

As described by Seddik et al.[24], IgGs in anti-β₁-receptor autoantibodies-positive sera of patients with HVM were purified by octanoic acid extraction. The purity of IgGs was over 90%.

Cell preparation

Guinea pig ventricular myocytes were isolated essentially as described by Heubach et al[25] with a modification. An adult guinea pig weighing 200-300 g was anesthetized with sodium pentobarbitone (30 mg/kg i.p.), and the trachea was cannulated for artificial respiration. The chest was opened, and the aorta was cannulated in situ. The heart was dissected and retrograde perfused on a Langendorff perfusion system at 37 °C. It was perfused first with Tyrode’s solution for about 1 min at a hydrostatic pressure of 70 cm, second with nominally Ca²⁺-free Tyrode’s solution for 3 min, and subsequently with 0.25 g/L collagenase type I, 0.20 g/L protease type XIV and 0.25 g/L bovine serum albumin for about 5 min. Finally the heart was washed with KB solution for 4 min and the ventricles were cut and teased into pieces in KB solution. All the solutions were bubbled with 1 000 mL/L oxygen. The dissociated cells were then kept in KB solution at room temperature for at least 1 h before the experiment.

Antibody binding

The binding of anti-β₁-receptor autoantibodies was examined by immunofluorescence. The isolated guinea pig cardiac myocytes were fixed by acetone at 4 °C. The fixed cells were washed three times with cold PBS, incubated with anti-β₁-receptors autoantibody-positive or negative sera which were absorbed by the above peptide sequence (1:10) overnight at 4 °C, and then washed 3 times in cold PBS, followed by incubation with goat anti-human IgG labeled with FITC (1:90) for 1 h at 37 °C. After rinsed with PBS, the slides were mounted with cover slips for fluoscopy.

Electrophysiology

Cardiac myocytes were placed in Tyrode’s solution and the Ca²⁺ resistant cells adhered to the coverslips of the recording chamber were selected for electrophsiological recording. Action potential and I _Ca-L were recorded using the whole-cell configuration of the patch-clamp technique. The cells were held in current-clamp mode or voltage-clamp mode using an Axopatch 200-A amplifier (Axon Instruments, USA). Action potential was elicited by constant current pulses of 1nA amplitude and a 6ms duration at a rate of 0.2 Hz.

In voltage-clamp experiments, the holding potential was set at –80 mV. Na⁺ and T-type Ca²⁺ channels were inactivated by applying a 100-ms prepulse to –40 mV immediately before each test pulse. The time course of changes in Ca²⁺ conductance was monitored by applying a 300-ms test pulse to 0 mV once every 5 seconds. For analysis of I-V relationship, after a 100-ms voltage step to –40 mV, 300-ms depolarizing voltage steps from –40 mV to +60 mV in 10 mV increments were used to elicit currents. Data were acquired and analyzed using the ISO2 (MFK, Germany) analysis software package.

Cell toxicity

Cell toxicity was examined by observing cell morphology and the permeability of cardiac myocytes to 20 g/L trypan blue. The live cardiac myocytes were rod-shaped and excluded trypan blue, whereas the dead cells were rounded and permeable to trypan blue.

Reagents

Tyrode’s solution contained (mmol/L) NaCl 135, KCl 5.4, MgCl₂ 1.0, NaH₂PO₄ 0.33, CaCl₂ 1.8 ,HEPES 10 and glucose 10 (pH adjusted to 7.4 with NaOH). KB solution contained (mmol/L) MgCl₂ 5, KCl 40, KH₂PO₄ 20, Taurine 20, Glutamicacid 50, EGTA 0.5, HEPES 10, glucose 10 (pH adjusted to 7.4 with KOH). All the chemicals for the electrophsiological experiment, enzymes and octanoic acid were purchased from Sigma Aldrich Company, USA. Metoprolol was a generous gift from AstraZeneca Company, USA. Goat anti-human IgG labeled with FITC was purchased from Wuhan Yafa Biotech Corp, China.

Statistical analysis

Data were expressed as mean ± SEM and analysed using Student’s t test. P < 0.05 was considered statistically significant.

RESULTS

Specific binding of anti-β₁-receptors autoantibodies to cardiac myocytes

Green fluorescence in rod-shaped cardiac myocytes only in slides incubated with anti-β₁-receptor autoantibodies-positive sera suggested the specific binding of anti-β₁-receptor autoantibodies to cardiac myocytes (Figure 1).

Figure 1 Specific binding of anti-β₁-receptor autoantibodies to guinea pig cardiac myocytes. Rod-shaped cardiac myocytes were stained by immunofluorescence (second IgG labeled with FITC), suggesting the specific binding of anti-β₁-receptor autoantibodies to cardiac myocytes.

Effects of anti-β₁-receptor autoantibodies on action potential in cardiac myocytes in the absence and presence of metoprolol

The effects of anti-β₁-receptor autoantibodies on action potential properties were tested in isolated cardiac myocytes in a current clamp mode of the patch clamp technique. After 5 min with stable action potentials, the antibodies diluted at 1:80 was superfused. The action potential duration (APD) was assessed as a duration to 20%, 50% and 90% repolarization (APD₂₀, APD₅₀ and APD₉₀, respectively). Anti-β₁-receptor autoantibodies diluted at 1:80 prolonged the APD in all phases of repolarization, and the increase averaged 39.2%, 29.1% and 15.2% for APD₂₀, APD₅₀ and APD₉₀vs control. Thus the plateau was markedly prolonged (Figure 2 and Table 1).

Table 1 Effects of anti-β₁-receptor autoantibodies diluted at 1:80 on action potential phases (n = 6).

	Control	Anti-β1-receptor autoantibodies
APD20(mV)	67.3 ± 1.8	93.7 ± 3.6b
APD50(mV)	148.8 ± 5.4	192.1 ± 6.7b
APD90(mV)	195.6 ± 8.2	225.3 ± 11.5b

^bP < 0.01 vs control.

Figure 2 AP recordings before and after superfusion with anti-β₁-receptor diluted at 1:80 in the absence and presence of 1 μmol/L metoprolol. A: effect of anti-β₁-receptor diluted an 1:80 on AP. ^arepresents AP before superfusion with 1:80 anti-β₁-receptor autoantibodies, ^brepresents AP after superfusion with 1:80 anti-β₁-receptor autoantibodies. B: no marked changes of AP after superfusion with 1:80 anti-β₁-receptor autoantibod-ies in the presence of 1 μmol/L metoprolol.

Whereas exposure of cells to anti-β₁-receptor autoantibodies diluted at 1:80 in the presence of 1 μmol/L metoprolol resulted in a slight prolongation of the APD. As in Table 2 shown, the increase averaged only 7.2%, 5.3% and 4.1% for APD₂₀, APD₅₀ and APD₉₀vs control (P < 0.01 vs anti-β₁-receptor autoantibodies diluted at 1:80).

Table 2 Effects of anti-β₁-receptor autoantibodies diluted at 1:80 on action potential phases in the presence of 1 μmol/L metoprolol (n = 6).

	Control	Anti-β1-receptor autoantibodies and metoprolol
APD20 (mV)	65.8 ± 1.1	70.5 ± 2.3
APD50 (mV)	147.2 ± 4.6	155.0 ± 6.3
APD90 (mV)	196.2 ± 5.1	204.3 ± 9.4

Effects of anti-β₁-receptor autoantibodies on I_Ca-L channels in the absence and presence of metoprolol

As described above, anti-β₁-receptor autoantibodies markedly prolonged the plateau, suggesting the antibodies might enhance I_Ca-L. We investigated the effect in the voltage clamp mode. Basal I _Ca-L was recorded 5-6 min after the cell membran rupture, when I_Ca-L was stable. Exposure of cells to anti-β₁-receptors diluted at 1:80 led to a significant increase of the amplitude of I_Ca-L, as the time course of changes (Figure 3A left) and the current traces of I_Ca-L (Figure 3A right) illustrated. The enhancement of I_Ca-L was rapid, and generally 20-30 s were sufficient for the full effect of the antibodies to take place. The rapid run–down of I_Ca-L followed after around 1 min. In light microscopy, the clamped cells were observed contracting and deteriorating. Beyond this dilution, the potentiating effect of the antibodies on I_Ca-L was too large to make effective voltage control. In 5 cells, anti-β₁-receptor autoantibodies diluted at 1:80 caused an increase of 55.87 ± 4.39% in the basal current (P < 0.01 vs control), and the enhancement of I_Ca-L was in a concentration-dependent manner (Figure 3C).

Figure 3 Effects of anti-β₁-receptor autoantibodies on peak current amplitude of I_Ca-L in the absence and presence of 1 μmol/L metoprolol. A and B (left): time-course changes of peak cur-rent amplitude of I_Ca-L before and after superfusion with anti-β₁-receptor autoantibodies diluted at 1:80 (A) or in the pres-ence of 1 μmol/L metoprolol (B). A and B (right): traces of I_Ca-L before and after superfusion with anti-β₁-receptor autoanti-bodies diluted at 1:80 (A) or before and after superfusion with anti-β₁-receptor autoantibodies diluted an 1:80 in the pres-ence of 1 μmol/L metoprolol (B). ^arepresents the maximal peak current amplitude of I _Ca-L before superfusion with anti-β₁-receptor autoantibodies diluted at 1:80, ^brepresents the maximal peak current amplitude of I_Ca-L after superfusion with anti-β₁-receptor autoantibodies diluted an 1:80. ^crepresents the maximal peak current amplitude of I_Ca-L after superfusion with anti-β₁-receptor autoantibodies diluted at 1:80 in the pres-ence of 1 μmol/L metoprolol. C: concentration dependence of anti-β₁-receptor autoantibodies. The numbers in brackets in-dicate the number of cells studied at each concentration.

I_Ca-L was recorded longer in cells exposed to anti-β₁-receptor autoantibodies diluted at 1:80 in the presence of 1 μmol/L metoprolol than that in the absence of metoprolol, with a slight increase of 6.81 ± 1.61% (P < 0.01 vs anti-β₁-receptor autoantibodies diluted at 1:80) (Figure 3B).

We also investigated the current-voltage (I-V) relationship by addition of anti-β₁-receptor autoantibodies in the absence and presence of metoprolol. Without shifting the I-V relationship, the antibodies caused a marked increase of current densities of I_Ca-L, whereas in the presence of 1 μmol/L metoprolol, a slight increase was found at positive and negative potentials (Figure 4).

Figure 4 Effects of anti-β₁-receptor autoantibodies diluted at 1: 100 on current-voltage (I-V) relationships normalized to cell membrane capacitance measured in 5 cells (mean ± SEM) dur-ing voltage steps from –40 to +60 mV in the absence (A) and presence of 1 μmol/L metoprolol (B).

Overall, these results indicated that anti-β₁-receptor autoantibodies could enhance I_Ca-L as agonists for β₁-adrenoceptors.

Cell toxicity effect

A large number of cardiac myocytes were exposed to high concentrations (1:80 and 1:100) of anti-β₁-receptor autoantibodies. Approximately over 90% rod –shaped cells with regular striation excluding trypan blue were turned into rounded cells highly permeable to the dye within 2 min (Figure 5).

Figure 5 Cell toxicity effect of cardiac myocytes exposed to high concentrations of anti-β₁-receptors autoantibodies. A: rod-shaped cells excluding trypan blue. B: rounded cells perme-able to trypan blue.

DISCUSSION

It was documented that there was a high homology between mouse hepatitis virus and β₁-adrenoceptors[26]. Therefore, autoimmune response directed against β₁- adrenoceptors can be initiated by cross reaction between these two molecules. In our clinical investigation, we found that the positive rate of hepatic virus antibodies and anti-β₁-receptor autoantibodies was highly consistent in patients with HVM[18]. Photomicrographs of guinea pig cardiac myocytes treated with anti-β₁-receptor autoantibodies before and after absorbed by the peptide sequence of β₁-adrenoceptor epitope and subsequently a fluorescently-tagged second antibody revealed the specific binding of autoantibodies in HVM to cardiac myocytes.

In heart diseases, interaction between catecholamine and β₁-receptors could result in various arrhythmias especially tachycardia and was the main cause of sudden death[27-29]. Based on our study, anti-β₁-receptor autoantibodies in HVM, as isoprenaline-like agonists for β₁-adrenoceptors were able to prolong the plateau of APD and enhance the Ca²⁺ permeability of cardiac myocytes via L-type Ca²⁺ channel, which triggers early after depolarization (EAD).

Ca²⁺ disorder was considered to be associated with myocarditis and cardiomyopathy[30]. It was reported anti-β₁-receptor autoantibodies in DCM could display a positive chronotropic effect on cultured neonatal rat ventricular cells without desensitization, thus increasing the metabolic pressure of myocardial cells[31], which was further demonstrated to be mediated by I_ca-L[32]. The time-course changes of I_ca-L in the presence of anti-β₁-receptor autoantibodies in HVM might reflect the process of the impairment of cardiac myocytes. About one minute after reaching the maximal peak value within 20-30s, I_Ca-L was observed running down and accompanied by the clamped cells contracting and deteriorating in the presence of anti-β₁-receptor autoantibodies diluted at 1:80. Whereas I_Ca-L was recorded longer in cells exposed to anti-β₁-receptor autoantibodies diluted at 1:80 in the presence of 1 μmol/L metoprolol. Furthermore, cell toxicity was observed by cell morphology and the permeability of cells to trypan blue. Rod-shaped cardiac myocytes exposed to high concentrations (1:80 and 1:100) of the antibodies were turned into rounded cells highly permeable to trypan blue. The overall influx of I_Ca-L caused by anti-β₁-receptor autoantibodies in HVM triggered electrophysiological and biochemical changes of cardiac myocytes and β₁- adrenoceptor blocker (metoprolol) could inhibit these effects. These results may in fact contribute to the proper explanation of the findings in our clinical investigation. The clinical investigation indicated that myocardial injury of patients with positive anti-β₁-receptor autoantibodies was more severe than that of patients with negative anti-β₁-receptor autoantibodies. Some arrhythmias such as sinus tachycardia or ventricular arrhythmias were associated with anti-β₁-receptor autoantibodies. These patients benefited from the use of β₁- adrenoceptor blocker[18].

In conclusion, autoantibodies against β₁-adrenoceptors can result in arrhythmias and/or the impairment of myocardiums in HVM, which would be mediated by the enhancement of I_Ca-L. As a basic research, our investigation would help to give early diagnosis by examining the presence of autoantibodies against β₁-adrenoceptor and beneficial treatment by properly utilizing selective β₁- adrenoceptor blocker for hepatitis virus myocarditis.