Introduction

Doxorubicin (DOX) is a highly active antineoplastic agent, used in the treatment of several malignancies, including breast, ovarian, lung and uterine cancers as well as soft-tissue sarcomas [1]. However, its clinical use is limited due to its cumulative multi-organ toxicities including cardiotoxicity [2,3,4]. Several hazardous effects, including lipid peroxidation, oxidative stress, DNA/RNA damage, inhibition of autophagy, endoplasmic reticulum-mediated apoptosis and disturbance of calcium homeostasis have been associated with DOX-induced cardiotoxicity [5,6,7]. The metabolism of DOX through NADPH-cytochrome P-450 enzyme leads to the formation of superoxide anions and hydroxyl radicals, which in turn cause injury to cellular membranes [8]. Moreover, cardiac inflammatory changes were reported due to DOX overdosage [9].

Due to the high efficacy of DOX as a chemotherapeutic drug, research efforts are directed to prevent and treat its associated cardiotoxicity. Phytochemicals are widely available, plant-derived, small molecules with proven antioxidant, anti-inflammatory, and anti-apoptotic activities [10]. The beneficial effects of garlic against cardiovascular disorders, such as stroke, coronary artery disease and hypertension are well known [11, 12]. Among the bioactive substances in garlic, allicin is a highly active organosulphur compound [13].

Allicin (allyl 2-propenethiosulfinate or diallyl thiosulfinate) is naturally produced by the action of the enzyme alliinase on the stable precursor S-allyl cysteine-S-oxide (alliin), when garlic cloves are macerated [13]. Recent studies have shown that oral administration of allicin reduces systemic blood pressure and protects against coronary endothelial dysfunction and cardiac hypertrophy in rats [13, 14]. Other animal studies reported that allicin offers multi-organ protection against the toxicity of several drugs and chemicals [15,16,17,18,19,20]. Based on these findings, allicin may be effective in mitigating the side-effects of DOX. In view of these facts, we performed this study to investigate the beneficial effects of allicin against DOX-induced cardiotoxicity in mice.

Materials and methods

Chemicals

Doxorubicin (ADRIAMYCIN®: a vial contains 2 mg/ml of Doxorubicin HCl) was produced by Pfizer, Egypt), while allicin (25% powder) in a clinical formulation was purchased from AnHui Ruisen Biological Technology Co., Ltd., China (CasNo: 539-86-6). The ELISA kits, used for analyzing cytokines [Tumor necrosis factor (TNF-α) and Interleukin-1β (IL-1β)] and 8-Oxo-2′-deoxyguanosine (8-OhdG: a marker of DNA oxidative damage) were obtained from R&D (Mannheim, Germany) and Cayman Chemical (MI, USA) companies, respectively. The kits for creatine kinase (CK) and creatine kinase-MB (CK-MB) were purchased from Stanbio TM (Texas, USA), while lactate dehydrogenase (LDH) Kits were obtained from Randox Laboratories. All other kits were purchased from Biodiagnostics Co. (Cairo, Egypt). Other chemicals, used in this study, were of analytical grades.

Animals and experimental design

Forty male Swiss albino mice (10 weeks age, 25 ± 2 g) were obtained from the animal breeding unit, Faculty of Veterinary Medicine, Zagazig University. Mice were housed in stainless steel cages in a clean well-ventilated room at an adjusted temperature (24 ± 2 °C) and relative humidity (50 ± 5%). They were provided rodent diet and water ad libitum. Animal care and housing, as well as the experimental protocol were approved by Animal Care and Ethics Review Committee at the Faculty of Veterinary Medicine, Suez Canal University, Ismailia, Egypt (The approval no. 201609).

After acclimatization for 1 week, mice were randomly divided into five groups (each of 8). The first group (control) orally received physiological saline. The second group (Allicin20) received allicin at a dose of 20 mg kg−1 BW for 2 weeks. The third group (DOX) received an oral saline solution for 2 weeks with intraperitoneal DOX injection at a dose of 10 mg kg−1 BW at 7, 9, and 11th day [21]. The fourth (DOX-Allicin10) and fifth groups (DOX-Allicin20) received intraperitoneal DOX injection (same dose of third group) plus allicin at a dose of 10 and 20 mg kg−1 BW [17], respectively, for 2 weeks.

Serum collection and tissue preparation

At the fourteenth day of the experiment, blood samples were collected in non-heparinized tubes and left for 30 min at room temperature to clot. The samples were then centrifuged at 3000 rpm for 20 min to obtain clear sera, which were preserved at −20 °C until used for biochemical assays. The mice were later killed by cervical decapitation under isoflurane anesthesia and the cardiac tissues were immediately removed from each mouse and washed in physiological saline. A half gram from each tissue was homogenized in 5 ml of phosphate buffer (pH 7.4) on ice, using an electric homogenizer. Homogenates were then centrifuged at 3000 rpm for 15 min at 4 °C and the resulting supernatants were kept at −20 °C until later use. The remaining cardiac specimens were immediately fixed in 10% neutral buffered formalin for histopathological and immunohistochemistry examination.

Assessment of serum biochemical parameters

The stored sera were used for evaluation of cardiac injury biomarkers. Serum aspartate transferase (AST) and LDH levels were assessed according to the methods of Reitman and Frankel [22] and Babson et al. [23], respectively. The CK activity was determined according to the method of Szasz et al. [24], while serum CK-MB activity was assessed according to the method illustrated by Wurzburg et al. [25]. Serum IL-1β/TNF-α and 8-OHdG concentrations were measured using commercially available ELISA kits from R&D (Mannheim, Germany) and Cayman Chemical Co. (MI, USA), respectively, according to the manufacturer’s instructions.

Assessment of oxidative/antioxidant status

Cardiac tissue homogenates were used for assessment of antioxidant enzymes’ activity, catalase [(CAT) according to the method of Aebi [26]], superoxide dismutase [(SOD) according to the method illustrated by Nishikimi et al. [27]], and glutathione peroxidase [(GPx) according to Paglia et al. [28]]. Reduced glutathione (GSH), a non-enzymatic antioxidant biomarker, was assessed according to the method of Beutler et al. [29]. Lipid peroxidation was evaluated by measuring the malondialdehyde (MDA) concentration according to the method developed by Mihara et al. [30], while nitric oxide (NO) concentration was determined according to Green et al. [31].

Histopathological and immunohistochemical examination

The formalin-fixed cardiac specimens were dehydrated in an ascending grade of ethyl alcohol, cleared in xylol, embedded in paraffin wax, and cut at 4–6 µm thick sections. They were later subjected to staining with Harri’s Hematoxylin & Eosin (H&E) for light microscopic examination. The paraffin sections were then stained using primary monoclonal antibodies against cyclooxygenase-2 (COX2) and caspase-3. The binding of antibodies was visualized by avidin–biotin complex (ABC kit, Vector laboratories) and the immunostaining reaction was labeled with diaminobenzidine (DAB) as a chromogen and counterstained with Mayer’s Hematoxylin.

Statistical analysis

All values were expressed as the mean ± standard error of mean (SEM). The data were analyzed, using the one way analysis of variance (ANOVA), followed by post hoc Duncan’s test. A p value <0.05 was considered statistically significant.

Results

Serum biochemical analysis

Intraperitoneal administration of DOX alone in mice caused a significant (p < 0.05) elevation in serum levels of AST, LDH, CK and CK-MB, compared to control mice. However, mice pretreated with both doses of allicin (10 and 20 mg kg−1 BW) for 2 weeks showed a dose-dependent significant (p < 0.05) reduction in the levels of these parameters, when compared to DOX-overdosed mice (Table 1).

Table 1 Serum enzymes’ activities and biochemical parameters’ concentrations in control and treated groups
Full size table

Moreover, DOX administration resulted in significant increases (p < 0.05) in serum levels of IL-1β, TNF-α and 8-OHdG (89.14 ± 2.37 pg/ml; 346.93 ± 14.45 pg/ml and 25.65 ± 0.92 ng/ml, respectively), when compared to control mice (38.65 ± 1.88 pg/ml; 185.05 ± 7.74 pg/ml and 13.83 ± 0.6 ng/ml, respectively). Pretreatment of DOX-intoxicated mice with both doses of allicin at 10 and 20 mg kg−1 BW resulted in dose-dependent significant reductions (p < 0.05) in serum levels of IL-1β (57.88 ± 2.61 and 39.79 ± 1.28 pg/ml, respectively), TNF-α (238.25 ± 6.47 and 189.54 ± 8.36 pg/ml, respectively) and 8-OHdG level (20.28 ± 1.11 and 13.92 ± 0.75 ng/ml, respectively), when compared to DOX-treated mice (Fig. 1).

Fig. 1
figure 1

Serum levels of proinflammatory cytokines and DNA marker in control and different treated groups. IL-1β; interleukin-1beta, TNFα; tumor necrosis factor alpha, 8-OHdG; 8-Oxo-2′-deoxyguanosine. Statistics; One-way ANOVA-test followed by Duncan’s Multiple Range Test; n = 8 (P ≤ 0.05), different letters indicate statistical significance (P ≤ 0.05)

Full size image

Effect on cardiac lipid peroxidation and antioxidant biomarkers

Doxorubicin-intoxicated mice showed significantly (p < 0.05) higher levels of NO and MDA, compared to control mice. These elevations were significantly (p < 0.05) ameliorated by treatment with both doses of allicin (10 and 20 mg kg−1 BW). Similarly, intraperitoneal DOX adminstration induced significant reductions in GSH concentration and the activity of cardiac CAT, SOD, and GPx enzymes, compared to untreated mice. However, oral allicin administration at (10 and 20 mg kg−1 BW) doses induced a significant (p < 0.05) dose-dependent increase in all of these parameters (Table 2).

Table 2 Cardiac tissue antioxidants and oxidative markers in control and treated groups
Full size table

Histopathological examination results

Examined H&E sections of the control group revealed normal myocardial architecture (a network of branching and anastomosing cardiac myofibers (each cardiomyocyte contains oval, central and euchromatic nuclei), separated by a loose connective tissue of endomysium with flat-nucleated fibroblasts and numerous blood capillaries (Fig. 2a).

Fig. 2
figure 2

Photomicrographs of examined groups stained with H&E. a Control group: Normal architecture of cardiac wall with branching and anastomosing myofibers bounded with endomysium, cardiomyocytes have central oval, euchromatic nuclei (arrow). Note that they are surrounded with numerous blood capillaries and flat nuclei of fibroblasts (arrowhead). b DOX-treated group: Focal myocytolysis (curved arrow) with either pyknotic or fade nuclei (arrow). Note the congested blood vessels (asterisk). Inset showing vacuolar degeneration of cardiomyocytes (arrowhead). c Allicin treated group (III) at a dose 10 mg/kg: Reorganization of cardiac myofibers with myocytolysis (arrow) and small vacuoles (arrowhead) of some cardiomyocytes. d Allicin treated group (IV) at a dose of 20 mg/kg: well-organized cardiac myofibers and oval and euchromatic cardiomyocytic nuclei (arrow). ×20 scale bar 50 µm

Full size image

Sections of DOX-treated animals showed fragmentation and degeneration of the myocardial tissue, manifested by focal myofibrillar loss, detached areas of cardiomyocytes, dispersed vacuoles, and eosinophilia of some myofibers. The cardiomyocytic nuclei had more than one appearance: either pyknotic and condensed or faint like a ghost. Moreover, sections displayed abundant congestion of blood vessels, extravasation of blood, and a widespread infiltration by fibroblasts and polymorphonuclear cells (Fig. 2b).

The histological sections of DOX-allicin10-treated group showed reorganization of cardiac myofibers with less cytoplasmic vacuoles and myofibrillar loss, but congestion of blood vessels was still evident (Fig. 2c). Further attenuation of DOX-induced histological changes was observed in DOX-allicin20-treated group with seemingly normal myocardial architecture, well-organized cardiac myofibers and a significant improvement in the myocardial nuclei profile. Congested blood vessels and vacuoles were occasionally observed (Fig. 2d).

Immunohistochemical examination results

The expression of COX2 in the myocardial wall was limited to the cardiomyocyte cytoplasm and fibroblastic nuclei. The control group demonstrated negative immunoreaction (Fig. 3a), while the most intense immunoreactivity was observed in the myocardial sections of DOX-treated mice (Fig. 3b). Treatment with allicin at doses of 10 and 20 mg kg−1 reduced COX2 expression in cardiomyocyte cytoplasm, but induced no change in fibroblastic nuclei expression (Fig. 3c, d).

Fig. 3
figure 3

Photomicrographs showing the expression of COX2 in the myocardium of the examined groups: (a) control, (b) DOX-treated group showing strong intensity of cardiomyocyte cytoplasm (asterisk) and fibroblast nuclei (arrow). Reduced COX-2 expression in allicin treated groups at a dose 10 (c) and 20 mg/kg B. wt. (d) (×20 scale bar 50 µm)

Full size image

The expression of caspase-3 in the myocardial wall was localized in the cardiomyocyte cytoplasm. The sections from control mice showed no immunoreaction (Fig. 4a), while an intense immunoreactivity was observed in the sections from DOX-overdosed group (Fig. 4b). This immunoreactivity was significantly ameliorated in sections, obtained from mice treated with DOX plus 10 or 20 mg kg−1 of allicin (Fig. 4c/d).

Fig. 4
figure 4

Photomicrographs showing the expression of caspase-3 in the myocardium of the examined groups: (a) control, (b) DOX-treated group, allicin treated at a dose of 10 (c) and 20 mg/kg B. wt (d) (×20 scale bar 50 µm)

Full size image

Discussion

The well-known cardiotoxic effects are primarily attributed to induction of inflammation [9] and overproduction of free radicals at high DOX doses [2, 32]. The heart is considered the main target organ for DOX-induced oxidative stress due to the lower levels of antioxidant enzymes (SOD, CAT, and GSH) in the cardiac muscle, compared to other organs [33, 34]. Moreover, the mitochondria in the cardiac muscle contains cardiolipin, which has a high affinity to DOX, leading to DOX accumulation within the cardiac mitochondria, impairment of the respiratory chain, and induction of apoptotic death [35]. All these events (oxidative damage, apoptosis, and inflammation) were confirmed in our biochemical, histopathological, and immunohistochemical analyses.

Reduced glutathione (GSH) is a non-enzymatic antioxidant, which scavenges reactive oxygen species (peroxynitrite, hydroxyl radical, lipid peroxyl radical, and H2O2) and other free radicals [36]. It acts as a substrate for GPx; therefore, unavailability of GSH results in reduced GPx activity [37]. Other antioxidant enzymes include SOD and CAT, which catalyze the breakdown of superoxide anions and hydrogen peroxide, respectively. Echoing other studies [8, 38, 39], we detected a significant reduction in the activities of cardiac SOD, CAT and GPx following DOX adminstration. Moreover, we observed a significant increase in cardiac tissue levels of NO, MDA, and 8-OHdG in DOX-intoxicated mice. Other studies explained this finding by DOX-induced upregulation of inducible nitric oxide synthase (iNOS), increasing NO production, lipid peroxidation, and DNA damage [40,41,42,43].

Interestingly, pretreatment of DOX-overdosed mice with allicin (10 and 20 mg kg−1 BW) significantly restored antioxidant enzymes’ (GPx, SOD, CAT) activity and GSH concentration and reduced NO, MDA, 8-OHdG levels in the cardiac tissue. Similar effects were described for allicin in the protection against acrylamide [17], cyclophosphamide [20], and gentamicin [18] cytotoxicity. Former studies have suggested that the antioxidant effect of allicin can be direct (by interaction with hydroxylated molecules or transferring its allylic hydrogen to the oxidized peroxyl radicals) or indirect through upregulation of phase II detoxifying enzymes (hemeoxygenase-1, glutathione-S-transferases, NAD(P)H-quinine oxidoreductase, and γ-glutamyl-cysteine synthetase) in a nuclear factor related-2 factor (Nrf2)-dependent pathway [44,45,46].

Confirming former studies [47, 48], our biochemical analysis showed that DOX induced an acute inflammatory reaction, manifested in elevation of serum levels of TNF-α and IL-1β. Moreover, we detected increased levels of serum cardiac injury markers (LDH, CK and CK-MB) in DOX-overdosed mice. Interestingly, allicin pretreatment lowered serum levels of inflammatory and cardiac injury biomarkers, which further confirms its anti-inflammatory and membrane-stabilizing effects [14, 49]. Liu and colleagues showed that allicin can offer cardioprotection through other mechanisms, e.g., allicin ameliorated cardiac hypertrophy and fibrosis via attenuating reactive oxygen species-dependent Jun N-terminal kinase (JNK1/2) signaling pathway. They also showed that allicin can prevent apoptosis by activating extracellular signal-regulated kinase (ERK) and PI3 K/Akt/GSK3β survival pathways [50].

Our histopathological analysis demonstrated cardiomyopathic changes following DOX adminstration, such as extensive sarcoplasmic vacuolization particularly around the blood vessels, myofibrillar loss, and distortion of normal architecture. Moreover, it demonstrated nuclear manifestation of degeneration, such as pyknotic or fading nuclei. These changes have been recorded earlier by several investigators who suggested oxidative stress, lipid peroxidation, and apoptosis as the underlying mechanisms for these observations [51,52,53]. Other studies have demonstrated that cellular vacuolization reflects dilatation of the smooth endoplasmic reticulum within cardiomyocytes [54, 55], while myofibrillar loss may be due to reduced mRNA expression of cardiac-specific proteins, such as troponin I and myosin light chain 22 [56].

Immunohistochemical analysis showed that DOX-induced apoptosis can be mediated by elevated cellular levels of caspase-3 enzyme. This was previously demonstrated by Davitashvili et al. who showed that DOX causes a marked release of cytochrome C into the cytoplasm, as well as an increased expression of caspase-3 and caspase-9 in cardiomyocytes [57]. Another apoptotic mechanism of DOX that was not investigated in this study is disrupting calcium homeostasis. Doxorubicin increases the opening likelihood of sarcoplasmic reticulum Ca channels and inhibits Na+–Ca2+ exchanger membrane proteins, inducing cellular and mitochondrial Ca overload, which may disrupt cellular metabolism, increase free radicals generation, and induce apoptosis [58].

In the current study, allicin-pretreated groups exhibited an improvement in the histological profile of the myocardium. The cardioprotective effects of allicin were dose-dependent, i.e., cardiac tissues subjected to the higher allicin dose (20 mg kg−1) displayed nearly normal histological architecture. These improvements are probably related to the previously mentioned antioxidant, anti-inflammatory, and antiapoptotic properties of allicin. Based on these findings, future clinical trials should consider allicin as a promising protective agent against DOX cardiotoxicity. Moreover, future animal studies should test the cytoprotective effect of allicin against other chemotherapeutic agents, especially in animal models with malignant tumors.

Conclusion

In experimental mice, DOX-induced cardiac oxidative damage following the exhaustion of cardiac antioxidant defense, as well as apoptotic and inflammatory reactions. Pretreatment with allicin significantly ameliorated the biochemical and histological changes, induced by DOX. Therefore, allicin may be a promising cytoprotective agent against DOX cardiotoxicity in human cancer patients.