Viral myocarditis (VMC), an inflammatory disease of heart muscle secondary to viral infection (Cooper, 2009), is an important cause of dilated cardiomyopathy worldwide (Schultz et al., 2009). A recent study showed that myocarditis was the cause of sudden cardiac death in 8.6% of cases (Fabre and Sheppard, 2006). Enteroviruses, particularly coxsackievirus B (CVB), as the most common viruses resulting in viral myocarditis, are reportedly contribute to at least 50% cases of infection-caused heart diseases (Seong et al., 2001).

Treatment of VMC is dependant on the clinical presentation and severity of disease (Swedberg et al., 2005; Jessup et al., 2009; Schultz et al., 2009). Seeing that the most common inducement of myocarditis is virus infection, it would be credible that antiviral vaccines or antiviral medications could be valuable in the treatment of viral myocarditis (Blauwet and Cooper, 2010). The effects of some antiviral substances in the treatment of acute myocarditis have been evaluated in animal models and a few clinical cases. The role of antiviral therapy for more chronic myocarditis associated with persistent viral genomes has still being investigated (Feldman and McNamara, 2000; Schultz et al., 2009). Although interferon alpha and ribavirin can improve the survival rate of mice with acute myocarditis (Matsumori et al., 1987, 1988), optimal virus-specific preventive procedures against myocarditis have not been clinically evaluated and no standard exists for their use in clinical practice. Looking for potent antiviral agents without affecting host cells’ normal physiological function is the primary task in designing new effective therapeutic drugs.

Previous studies have shown that cinnamaldehyde (3-phenyl-2-propenal), a major component of the essential oil of cinnamon bark isolated from Cinnamomum trees, possesses multiple biological activities including anti-platelet aggregation, anti-bacterial, anti-viral, and anti-inflammation properties (Beige et al., 1996; Youn et al., 2008). Our recent study showed that cinnamaldehyde reduced the viral titer and improved the survival rate of VMC mice. Furthermore, cinnamaldehyde reduced plasma nitric oxide (NO) content, NF-κB expression and decreased inflammatory cell infiltrate in the myocardium of VMC mice (Ding et al., 2010). Despite these potential advantages, cinnamaldehyde, as an aldehyde with reducing property, can be oxidized to cinnamic acid rapidly in blood. The therapeutic effect of cinnamic acid on viral myocarditis was not observed in vivo. In order to overcome the instability of cinnamaldehyde in vivo, we synthesized a series of derivatives using cinnamaldehyde as the leading compound (structure shown in Fig. 1). The evaluation of the antiviral and cytotoxic effects of these derivatives on HeLa cells infected by CVB3 and neonatal rat cardiomyocytes are described in this report. We present a series of derivatives to determine if added substituents raise antiviral activity.

Synthesis of (E)-3-p-tolylacrylaldehyde (1), (E)-3-(4-chlorophenyl) acrylaldehyde (2), and (E)-3-(4-methoxyphenyl)acrylaldehyde (3)

0.25 mol substituted-benzaldehyde (4-methylbenzaldehyde, 30.0375 g; 4-chlorobenzaldehyde, 35.1425 g; 4-methoxybenzaldehyde, 34.0370 g) and KOH (1.60 g) in ethanol (50 ml) were stirred at 5–10°C and protected by nitrogen. Acetaldehyde (8.81 g, 0.20 mol) was added in a dropwise manner within 15 min. The mixture was stirred at 5°C for 10 h, adjust pH to 5 with acetic acid. After pouring this whole mixture into water, the oily material was extracted with diethyl ether and then the organic layer was washed with saturated aqueous NaCl solution and water consecutively. The solvent was removed in vacuo, recrystallized with 50% ethanol, and the products were yellow crystals 1 (4.38 g, 12%), 2 (6.25 g, 15%) and 3 (9.32 g, 23%). 1: Mp: 40–42°C; IR (solid): 2822, 2744, 1684, 1626, 1604, 1508, 1448, 1324, 1210, 1129, 1109, 1008, 808 cm⁻¹; NMR (400 MHz; CDCl₃): δ 9.68 (1H, d, J=8.0), 7.46 (2H, d, J=8.0), 7.45 (1H, d, J=16.0), 7.23 (2H, d, J=8.0), 6.68 (1H, dd, J=8.0 16.0), 2.38 (3H, s). 2: Mp: 59–61°C; IR (solid): 2992, 2853, 1687, 1626, 1590, 1490, 1411, 1299, 1247, 1121, 1085, 1012, 977, 807 cm⁻¹; NMR (400 MHz; CDCl₃): δ 9.69 (1H, d, J=8.0), 7.80 (1H, d, J=16.0), 7.54 (2H, d, J=8.0), 7.48 (2H, d, J=8.0), 6.89 (1H, dd, J=8.0 16.0). 3: Mp: 56–57°C; IR (solid): 2939, 2843, 2765, 1666, 1626, 1602, 1570, 1511, 1264, 1249, 1177, 1131, 1008, 978, 826, 808 cm⁻¹; NMR (400 MHz; CDCl₃): δ 9.65 (1H, d, J=8.0), 7.52 (2H, dd, J=4.8 6.4), 7.42 (1H, d, J=16), 6.94 (2H, dd, 4.8 6.8), 6.61 (1H, dd, 8.0 15.6), 3.86 (3H, s).

Synthesis of (Z)-2-bromo-3-p-tolylacrylaldehyde (4)

Compound 1 (3.67 g, 0.025 mol) and acetic acid (10 ml) were added and the solution was stirred in an ice bank to drop the temperature below 5°C. Dropwise liquid bromine (4.15 g) was slowly added in a dropwise fashion and this mixture was stirred in the ice bank for 30 min. Then, K₂CO₃ (2.76 g, 0.02 mol) was added to the mixture until no bubbles were present. This solution was stirred at 80°C for 1.5 h. After cooling the mixture to 30°C, 25 ml water was added into it to precipitate. The yellow crystals were recrystallized with 80% ethanol, and to create a yellow compound 4 (6.95 g, 81.26%). Mp: 64–66°C; IR (solid): 2851, 1933, 1693, 1595, 1560, 1284, 1106, 1082, 895, 817 cm⁻¹; NMR (400 MHz; CDCl₃): δ 9.31 (1H, s), 7.92 (2H, d, J=8.0), 7.85 (1H, s), 7.29 (2H, d, J=8.0), 2.41 (3H, s).

Synthesis of (Z)-2-bromo-3-(4-chlorophenyl) acrylaldehyde (5)

Compound 2 (4.17 g, 0.025 mol) and acetic acid (10 ml) were added and the solution was stirred in an ice bank to drop the temperature below 5°C. Liquid bromine (4.10 g) was slowly added in a dropwise manner and stirred in the ice bank for 30 min. Then, K₂CO₃ (2.76 g, 0.02 mol) was added to the mixture until no bubbles were present and the solution was stirred at 80°C for 1.5 h. After cooling the mixture to 30°C, 25 ml water was added into it to precipitate. The yellow crystals were recrystallized with ethanol to create a yellow compound 4 (4.45 g, 72.48%). Mp: 91–92°C; IR (solid): 2832, 2359, 1687, 1601, 1586, 1489, 1408, 1282, 1116, 1091, 1016, 895, 822 cm⁻¹; NMR (400 MHz; CDCl₃): δ 9.34 (1H, s), 7.95 (2H, d, J=8.0), 7.85 (1H, s), 7.46 (2H, d, J=8.0).

Cardiomyocytes culture and cytotoxicity assay

Cardiomyocytes culture and cytotoxicity assay were performed according to a previously described procedure (Zhang et al., 2012). Briefly, rat cardiomyocytes were isolated from the ventricles of one to three-day-old Sprague-Dawley rats. The hearts were excised from neonatal rats under ethylether anesthesia, minced with scissors and dispersed with PBS. The cells were digested with 0.05% collagenase type II and 0.05% trypsin in PBS for 10 min at 37°C. The supernatants were transferred to DMEM containing 10% FBS, and the digestion was repeated four times. The resulting cell suspension was centrifuged, and the pellet was re-suspended in DMEM. The cells were plated in culture flasks for 1.5 h to remove nonmyocytes. Cardiomyocytes were cultured on 96-well plates maintaining in DMEM (containing 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin) for 3 days. The cells were incubated with 0.01–1000 μM compounds 1–5 or cinnamaldehyde for 72 h. Dimethyl sulfoxide (DMSO) was used to dissolve compounds 1–5 and cinnamicaldehyde. The solutions were diluted with the medium before use to ensure all assays’ final DMSO concentration didn’t exceed 0.05%. Colorimetric MTT assay was used to analyze cell proliferation. 20 μl of MTT (5 mg/ml) was added into each well of the plate which was then maintained at 37°C for 4 h. After aspirating the supernatant, 150 μl of DMSO was added into each well. A 96-well microplate reader (Bio-Rad, Tokyo, Japan) was used to measure the absorbance at 570 nm. The following formula was used to determine the cytotoxicity: percent of inhibition (%)=100−(absorbance value of test compound-absorbance value of blank)/(absorbance value of control-absorbance value of blank)×100. The 50% toxic concentration (TC₅₀) was determined by the compound’s concentration inhibiting 50% of the viability of untreated cell cultures. At least three tests of the mean dose-response curve was used to calculate the TC₅₀.

Virus and anti-viral activity assay

The CVB3 (Nancy strain), purchased from the Microbiology Department of Fourth Military Medical University (Xi’an, China), was maintained by passage through human cervical carcinoma HeLa cells (ATCC CCL-2). Viral titers were evaluated by cytopathic effect in HeLa cells. Virus stocks were titrated on HeLa monolayer cells in 96-well plates by making tenfold dilutions (eight wells per dilution). Plates were incubated at 37°C in a CO₂ incubator for 5 d and the results were read under the light microscope. Titres were expressed as 50% tissue culture infective dose (TCID₅₀) values, calculated according to Reed and Muench (1938).

The antiviral activity of test compounds against CVB1 was evaluated by the XTT (2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-5-[(phenylamino) carbonyl-2H-tetrazolium hydroxide]) method as previously described (Chiang et al., 2002). Briefly, cardiomyocytes were cultured to confluency on 96-well plates (final cell density: ≈1×10⁴ cell /well) and then infected with CVB3 (100TCID50, 20 μl per well) for 2 h in six replicates. After that, the cells were washed with phosphate-buffered saline (PBS) for three times, followed by incubation with serial dilutions of compound 1–5 and cinnamaldehyde. After 72 h, 50 μl of mixed solution of 0.1 ml phenazine methosulfate and XTT (5 mg/5 ml) were added to each well and then incubated for 2 h to producing XTT formazan. A 96-well microplate reader (Bio-Rad, Tokyo, Japan) was utilized to detect the optical densities with the test wavelength being 450 nm and the reference wavelength being 690 nm. The effect of viral inhibition was determined by the following formula: viral inhibition rate=(ODtv−ODcv)/(ODcd−ODcv)×100%. ODtv, ODcv and ODcd denote the absorbance of virus infected cells treated with test compound, the virus control and the cell control, respectively. The half-maximal inhibitory concentrations (IC₅₀) and the therapeutic index (TI, TC₅₀/IC₅₀) of compound 1–5 and cinnamicaldehyde were calculated.

Treatment of CVB3 infected mice

Four-week-old male BALB/c mice were purchased from the Center of Experimental Animals of the Fourth Military Medical University and housed under pathogen-free condition. These animals were randomly divided into fourteen groups (n=10) as follows: normal group, vehicle group (0.5% carboxymethyl-cellulose (CMC)-saline), model group, cinnamaldehyde group (40 mg /kg), interferon alpha (10 mg /kg) and compound 2, 4, 5 groups (20, 40 and 60 mg/kg). To establish VMC models, animals were inoculated intraperitoneally (i.p.) with 0.1 ml RPMI 1640 medium that contains 100×TCID₅₀ of CVB3. The vehicle group was treated with only 0.1 ml RPMI 1640 medium. From the 2^nd day the interferon alpha group was intraperitoneally administered interferon once per day. The animals in the compound 2, 4, 5, vehicle and model groups were administered compound 2, 4, 5 or 0.5% CMC-saline interferon by oral gavage at the same time. At day 7, animals were euthanatized by an overdose of sodium pentobarbital. To assess the severity of acute CVB3-induced myocarditis, ventricular tissue samples of isolated hearts were prefixed in buffered formalin, embedded in paraffin and finally sectioned into 4-μm slices. Sections were subsequently stained with H&E and examined by light microscopy. The impairment degree of CVB3-induced myocarditis was evaluated by the ratio of the heart section with inflammation to the entire heart section using a microscope eyepiece grid with magnification ×200. The score system was as follow: no lesion (grade 0); <25% of the heart section involved (grade 1); 25–50% involved (grade 2); 50–75% involved (grade 3); >75% involved (grade 4) (Nishio et al., 1999). The animal experiments were performed in accordance with the ‘Guidelines for Animal Experimentation’ of the Fourth Military Medical University.

Virus titration from heart homogenate

Part of the heart was weighed, homogenized in 10% (w/v) PBS and then centrifuged at 5000 rpm for 10 min. The supernatant was collected for TCID₅₀ assay to determine virus titer. Briefly, 96-well plates was seeded with HeLa cell monolayers and then serial 10-fold dilutions (100 μl) of supernatant added into the wells. After 5 days after infection, virus titers were calculated according to the last dilution leading to cells in 50% of wells showing a cytopathic effect. Results were presented as Log TCID₅₀/mg of tissue.

TNF-α, IL-1β and IL-6 expression

After infected with CVB3, cardiomyocytes were incubated with cinnamaldehyde (10 μM) and compound 2, 4 and 5 (10 μM). After 12 h, the TNF-α, IL-1β and IL-6 mRNA expression in cardiomyocytes was assayed using real-time PCR (Chen et al., 2005). The supernatants were collected after 72 h, and the levels of TNF-α, IL-1β and IL-6 in the culture supernatant were measured with ELISA, using commercially available ELISA kits (Rapidbio, West Hills, CA, USA).

Western blot analysis

The cardiomyocytes infected with CVB3 were incubated with cinnamaldehyde (10 μM) and compound 2, 4 and 5 (10 μM). After 24 h, whole-cell lysates were prepared according to the procedure previously described (Cao et al., 2011). Samples containing equal amounts of protein were separated by 15% SDS-PAGE and transferred onto nitrocellulose membranes. After washing, the membranes were blocked with 5% skimmed milk at room temperature for 2 h, and then incubated overnight at 4°C with the TNF-α, IL-1β, IL-6 and GAPDH antibodies (R&D Systems, Minneapolis, MN, USA). The membranes were washed and incubated with appropriate horse-radish peroxidase-conjugated secondary antibodies for 1 h at room temperature. Blots were visualized using Supersignal chemiluminescence substrate (Pierce, Rockford, IL, USA). Band densities were determined using Quantity One Software (BioRad, Hercules, CA, USA). The intensity of the resulting bands was expressed as the ratio between proteins of interest and GAPDH.

Statistics

All data were expressed as mean ± SD. Significant differences between two groups were determined by Student’s t-test and one-way ANOVA was used for multiple comparisons. p<0.05 was accepted as statistically significant for all tests

Preparation of compounds

There are several published synthetic routes for cinnamaldehyde derivatives (Battistuzzi et al., 2003; Nordqvist et al., 2011). We modified and built synthetic schemes to yield compound 1–5 (Fig. 2). The substituted-benzaldehydes were combined with acetaldehyde to yield substituted- cinnamaldehydes (compound 1–3). Compound 1 and 2 were combined with liquid bromine to yield 2,3-dibromo-3-p-tolylpropana or 2,3-dibromo-3-(4-chlorophenyl) propanal, respectively. After dehydrobromination, yellow crystals (compound 4, yield was 81.26%; compound 5, yield was 72.48%) were obtained.

Cytotoxic and anti-viral activities of cinnamaldehyde derivatives on cardiomyocytes

Cardiomyocytes were infected with CVB3 and treated with compound 1–5 and cinnamaldehyde at concentrations in the range of 0.01–1000 μM for 72 h and then evaluated by the MTT assay. IC₅₀ of compound 1–5 were calculated (De Logu et al., 2000). In addition, the cytotoxic effects of compound 1–5 were evaluated in the neonatal rat cardiomyocytes using MTT assay. In our experiments, compound 2, 4, 5 and cinnamaldehyde inhibited the replication of CBV3 in the cardiomyocytes in concentration-dependent manners (Fig. 3A). Compound 4 and 5 showed strong inhibitory effects with IC₅₀ values of 11.38 ± 2.22 μM and 2.12 ± 0.37 μM, respectively, compound 2 showed a weak inhibitory effect with IC₅₀ value of 2109.08 ± 157.69 μM, while the other compounds exhibited no activities (Table 1). Fig. 3B showed the cytotoxic effects of compound 1–5 and cinnamaldehyde on cardiomyocytes. The CC₅₀ values of compound 1, 3 and cinnamaldehyde were 3.96 ± 0.29 μM, 7.88 ± 0.82 μM and 0.98 ± 0.12 μM, respectively (Table 1). Compound 4 and 5 showed weak cytotoxic effects with CC₅₀ values of 38.61 ± 4.89 μM and 85.69 ± 7.63 μM, while compound 2 exhibited no obvious cytotoxicity. The TI values of compound 2, 4, 5 and cinnamaldehyde were 3.31, 3.39, 40.42 and 0.04, respectively. So in the next study, we chose compound 2, 4, 5 for further study and compared their therapeutic effects on VMC in vivo.

Therapeutic effects of cinnamaldehyde derivatives on mouse model of VMC

The morphometrical analysis was used to assess the severity of myocarditis (Fig. 4, Table 2). In the infected control group, histopathological investigations showed large infiltrates of lymphocytes associated with severe myocardial necrosis. The cardiac pathological score was 2.95 ± 0.60. Compared with the infected control group, the myocardial damage was significantly reduced by compound 4 and 5 in dose-dependent manners. However, there was no significant difference between the compound 2 group and model group.

Furthermore, viral titers in the infected mice were significantly decreased by the treatment of compound 4, 5 and interferon. The cardiac CVB3 titer in the model group was 3.40 ± 0.70 Log TCID₅₀/mg. Compound 4 and 5 inhibited viral titers in dose-dependent manners. The level of CBV3 titers in compound 4 (60 mg/kg) group and compound 5 (40 and 60 mg/kg) groups were significantly decreased, compared with those in the model group (p<0.05). Results showed that compound 5 had highest efficiency on 11 VMC than other cinnamaldehyde derivatives, being consistent with in vitro results.

Anti-inflammatory activities of cinnamaldehyde derivatives

Neonatal rat cardiomyocytes were infected with CVB3 and then incubated with cinnamaldehyde (10 μM) and compound 2, 4 and 5 (10 μM) to examine their effects on TNF-α, IL-1β and IL-6 levels. After 12 h, the TNF-α, IL-1β and IL-6 mRNA expressions in cardiomyocytes were assayed using real-time PCR (Chen et al., 2005). The culture supernatants were collected after 72 h, and their TNF-α, IL-1β and IL-6 levels were measured with ELISA. According to our data, the levels of TNF-α, IL-1β and IL-6 in viral-infected cardiomyocytes were significantly inhibited and reached 43.60 ± 4.77, 112.72 ± 13.60 pg/ml and 232.80 ± 19.62 pg/ml, respectively, after application of compound 5. These values are lower than those of the viral control group (80.12 ± 8.40, 163.64 ± 19.75 and 337.46 ± 29.22 pg/ml, respectively) (Fig. 5A). Compound 4 and 5 also inhibited the expressions of TNF-α, IL-1β and IL-6 mRNA (Fig. 5B). We further evaluated the effect of cinnamaldehyde, compound 2, 4 and 5 on the expression of TNF-α, IL-1β and IL-6 by Western blot analysis (Fig. 5C). The TNF-α, IL-1β and IL-6 expression in viral-infected cardiomyocytes was markedly increased (3.52 ± 0.31, 2.53 ± 0.28 and 2.27 ± 0.24, respectively) compared with normal cells. Notably, compound 4 and 5 markedly diminished CVB3-mediated TNF-α, IL-1β and IL-6 expression. These results indicate that the two compounds could suppress CVB3-induced TNF-α, IL-1β and IL-6 expressions at the transcription and translation levels.

One of the major causes of acute myocarditis is CVB3 which can lead to tissue remodeling and even result in the disease itself (Kashimura et al., 2004; Dennert et al., 2008). In this study, we compared the anti-CVB3 activities of compound 1–5 and cinnamaldehyde on CVB3-induced viral myocarditis. Among the derivatives, compound 2, 4 and 5 inhibited the CVB3 in HeLa cells with IC₅₀ values of 2109.08 ± 157.69, 11.38 ± 2.22 and 2.12 ± 0.37 μM, respectively. CC₅₀ in compound 2, 4 and 5-treated cardiomyocytes were 6972.47 ± 734.08, 38.61 ± 4.89 and 85.69 ± 7.63 μM, which were 7114-fold, 39-fold and 87-fold higher than in the cinnamaldehyde group, suggesting that the toxicities of compound 2, 4 and 5 were much lower than that of cinnamaldehyde. In addition, the present studies demonstrated that compound 4 and 5 had effectively protective effect on CVB3-infected mice. They could reduce the viral titers and cardiac pathological changes in dose-dependent manners. Furthermore, compound 4 and 5 significantly inhibited the secretion and mRNA expressions of inflammatory cytokines TNF-α and IL-β in CVB3-infected cardiomyocytes, indicating that brominated cinnamaldehyde not only improved the anti-vital activity, but also had potent anti-inflammatory effects in cardiomyocytes induced by CVB3.

At present, the treatment of viral myocarditis is conventional supportive therapy. Specific therapies are lacking and in great demand (Swedberg et al., 2005; Schultz et al., 2009; Blauwet and Cooper, 2010). It is a primary task to look for potent anti-viral agents without affecting host cells’ normal physiological function in designing new effective therapeutic drugs. Cinnamaldehyde, α, β-unsaturated carbonyl derivative with a mono-substituted benzene ring, has demonstrated to suppress the growth of influenza A/PR/8 virus and CVB3 (Hayashi et al., 2007). It has been reported that cinnamaldehyde is unstable in vivo and can be oxidized to cinnamic acid rapidly in circulating blood. Resulted from β-oxidation of cinnamic acid, glycine or glucuronic acid conjugates of benzoic acid can be formed as the primary urinary metabolites of cinnamaldehyde (Yuan et al., 1992). Our previous studies showed that 10 to 1000 μM cinnamic acid displayed significant anti-CVB3 activity in primary cultured myocardial cells. Cinnamaldehyde possessed similar antiviral activity to cinnamic acid when being administered intraperitoneally (Ding et al., 2010). However, cinnamaldehyde could reduce plasma nitric oxide (NO) content, NF-κB, inducible nitric oxide synthase and TLR4 expression in myocardium from VMC mice 7 days after viral inoculation, while not observed in the mice treated with cinnamic acid. These results suggested that cinnamaldehyde might be a new lead compound for treating VMC.

Cinnamaldehydes are commonly synthesized in one or more steps by the Wittig reaction, Horner-Wadsworth-Emmons reaction, Peterson reaction, oxidation of primary allylic alcohols or crossed aldol condensation (van Staden et al., 2002; Nordqvist et al., 2011; Gu and Tian, 2012). We used substituted-benzaldehyde and acetaldehyde in one step to obtain cinnamaldehyde derivatives. This synthesis was readily available, low toxic, and easy to handle. Then, we compared their VMC therapeutic effects and cytotoxicity with cinnamaldehyde at a concentration range of 0.01–1000 μM. Among them, compound 4 and 5 had promising therapeutic effects and less toxicity than cinnamaldehyde in vitro.

Acute VMC has three phases including an original viral infection, autoimmune response and reshaping of cardiac structure and function (Dennert et al., 2008). In the incipient stage of CVB3 infection, the direct attack of the virus on the cardiac muscle cells is the main pathogenic process (Liu and Mason, 2001; Wang et al., 2007). We used the CVB3-inoculated male BABL/C mice for duplicating the VMC model. On the 3rd day, the mice exhibited signs of illness, including lethargy, progressive weight loss, and death. The structure of normal myocardium was clear, and the cell nucleus was clear. We found severe inflammation infiltration, necrosis, and fibrosis in infected murine hearts. With the activation of host immunologic function, the viral titer gradually dropped in the target organs. But in cardiomyocytes, non-renewable cells, viral titer might be long-standing, even being achieved at 30 days (Mall et al., 1991). So to detect VMC infected with CVB3, the activities of biochemical markers, virus titers of the organs and the histology changes of the heart are the most particular characteristics (Wang et al., 2009). After the mice were given compound 4 and 5 for 6 days, the damage of cardiomyocytes was reduced and necrosis and infiltration were significantly decreased, compared with the infected control group. Furthermore, the cardiac viral titers in the mice administrated with compound 4 and 5 were significantly decreased compared with the CVB3 control. These results suggested that compound 4 and 5 might improve the cardiac injury in the VMC animal models.

Although virus in the evolution of acute VMC is very important, it can attack on myocardial cells directly. The main cause of myocyte injury is an excessively activated immune response, which is triggered by the virus (Dennert et al., 2008). Previous studies showed that the expression of many pro-inflammatory cytokines in the heart of mice with viral myocarditis were too much. Among these cytokines, TNF-α, IL-1β and IL-6 conduce to the CBV3 pathological changes in myocardial disease (Shen et al., 2009; Gui et al., 2012). So, in this study, we discussed the effects of compound 2, 4 and 5 on the TNF-α, IL-1β and IL-6 production and mRNA and protein expression in CVB3 infected neonatal rat cardiomyocytes. Our experiments showed that compound 4 and 5 could inhibit not only the production, but also the mRNA and protein expressions of TNF-α, IL-1β and IL-6 in CVB3 infected cardiomyocytes. These results suggested that compound 4 and 5 had potent anti-inflammatory activities. In addition, we observed that compound 5 inhibited CVB3-induced NF-κB activation, IκB-α degradation and phosphorylation by dose-dependent method, and decreased hearts’ Toll like receptor (TLR) 4 protein level (Zhang et al., 2012). However, further studies involving structure-activity relationship are needed.

Furthermore, cinnamaldehyde is unstable in rat blood with a half-life of 4 min (Yuan et al., 1992). Our previous study showed that cinnamaldehyde maintained a very low concentration in plasma by the oral route (Zhao et al., 2014). In this study, we showed the antivital activities were markedly increased when cinnamaldehyde di-substituted with bromine and chlorine/methyl. We assumed that these substituents might raise the stability of cinnamaldehyde and further increases anti-viral activity. To evaluate the pharmacokinetic behavior of compound 2, 4 and 5 in rats, we recently developed a high performance liquid chromatographic method with UV detection for the determination of them in rats after a single oral administration at a dose of 20 mg/kg (unpublished data). A period of 24 h stability was tested by analyzing compounds in blood at room temperature and indicated that compound 2, 4 and 5 were stable for 24 h. In addition, the preliminary data showed that the peak concentrations of compound 2, 4 and 5 were achieved about 1.2 h, 3.3 h and 2.1 h after oral administration. From above, it seems likely that the substituted cinnamaldehyde exhibited more satisfactory biological effects and pharmacokinetic properties. However, the precise details of pharmacokinetics and the tissue distribution of these compounds need more elucidation so that the molecular mechanisms involved could be better understood.

In conclusion, our results demonstrated that chlorinated and brominated cinnamaldehyde can inhibit CVB3 replication and CVB3-induced cardiac injury potently both in vitro and in vivo. Compound 4 and 5, particularly compound 5, showed improved anti-vital and anti-inflammatory activity, low toxicity relative to cinnamaldehyde.

This work was supported by the grant from the National Natural Science Foundation of China (No. 20872180 to W.-S. Wang, No.81370241 and No.81170107 to X.-Q. Li.).