Journal of Applied Physiology

Diallyl trisulfide and diallyl disulfide ameliorate cardiac dysfunction by suppressing apoptotic and enhancing survival pathways in experimental diabetic rats

Yao-Te Huang, Chun-Hsu Yao, Chia-Li Way, Kung-Wei Lee, Cheng-Yen Tsai, Hsiu-Chung Ou, Wei-Wen Kuo


Cardiovascular disease is one of the major causes of mortality in diabetic patients. Mounting studies have shown that garlic exhibits, possibly through its antioxidant potential, diverse biological activities. In this study, we investigated the alleviating effects of garlic oil (GO) and its two major components, diallyl disulfide (DADS) and diallyl trisulfide (DATS), on diabetic cardiomyopathy in rats. Physiological cardiac parameters were obtained using echocardiography. Apoptotic cells were evaluated using TUNEL and DAPI staining. Protein expression levels were determined using Western blotting analysis. Our findings indicated that in diabetic rat hearts significantly decreased fractional shortening percentage, increased levels of nitrotyrosine, an elevated number of TUNEL-positive cells, enhanced levels of caspase 3 expression, and decreased PI3K-Akt signaling pathway activities were observed. Furthermore, all of these alterations were reversed following both GO and DATS (or DADS) administrations through increasing PI3K-Akt signaling pathway activities and inhibiting both the death receptor-dependent and the mitochondria-dependent apoptotic pathways. In conclusion, this study shows that DATS and DADS, with the efficacy order DATS > DADS, have the therapeutic potential for ameliorating diabetic cardiomyopathy. Furthermore, the therapeutic effects of GO on diabetic cardiomyopathy should be mainly from DATS and DADS.

  • diabetic cardiomyopathy
  • garlic
  • PI3k-Akt
  • apoptosis
  • echocardiography

the ubiquitous prevalence of diabetes mellitus (DM) has been noticeably increasing globally, and the number of affected patients was estimated to be 171 million (or 2.8% of the worldwide population) in 2000, and will be projected to rise to 366 million (or 4.4%) in 2030 (35). Besides, 12% of the global healthcare expenses were expected to be spent on diabetes in 2010 (40). Also, diabetes is a well-known serious factor for the development of cardiovascular disease (CVD). Indeed, diabetic patients have a two- to eightfold higher chance of developing CVD in comparison with normal subjects (15). Furthermore, CVD is the major cause of mortality and morbidity in patients with diabetes. CVD can explain up to 80% of premature excess mortality in diabetic patients (36). These highlight the gravity of the burden which the epidemic diabetes and its associated CVD put on the global economy and healthcare (16), and prevention or alleviation of CVD in diabetes will become a key issue.

Diabetic cardiomyopathy, first coined by Rubler (31), is a diabetes-induced cardiac disease independent of coronary artery disease, and, clinically speaking, manifests itself at first as asymptomatic diastolic dysfunction and cardiac hypertrophy and then progresses to symptomatic heart failure (1). The molecular mechanism underlying its pathogenesis is not well understood, but a rough picture of the process has started to emerge (19). Briefly, diabetes can induce hyperglycemia, hyperlipidemia (excessive free fatty acid oxidation), and mitochondrial uncoupling, resulting in the induction of oxidative stress [i.e., excessive production, insufficient removal, or mismanagement of highly reactive species including reactive oxygen (ROS) and reactive nitrogen species (RNS)] (27). The oxidative stress, in turn, activates activities of several pathways including production of advanced glycation end products (AGEs) and expression of the receptors for AGEs (RAGEs) and their upregulation of transcription factors (e.g., NF-κB) (14), polyol and hexosamine pathways, and tyrosine kinase/phosphatase pathways, leading to endothelial dysfunction, cardiovascular inflammation, and cardiovascular apoptosis and remodeling, ending up with diabetic cardiomyopathy (19). Thus the oxidative stress is located at the pivotal point during the progression of diabetic cardiomyopathy, and the antioxidant therapies or therapies based on suppression of ROS production should have a potential and beneficial role in their treatments (34).

Garlic (Allium sativum) and its components have a wide assortment of beneficial biological activities including cholesterol- and lipid-lowering effects, hypotensive effects, antithrombotic and antiplatelet aggregatory activities, enhancing immune functions, and antitumor actions (17, 28). Because of these healthy benefits, garlic has a long history as a natural folk remedy and as a dietary supplement (8). For many years, various kinds of garlic preparations have been studied, with somewhat inconsistent outcome, for their efficacy in preventing or treating cardiovascular disease in vitro and in vivo (2, 28), necessitating the further investigation. Hypothetically, these activities of garlic may be mediated through its antioxidant potential on alleviating the oxidative stress of many diseases including diabetic cardiomyopathy (2). This hypothesis was compellingly supported by the important finding that hydrogen sulfide (H2S) production comes from garlic-derived organic polysulfides and confers the vasoactivity and other cardioprotective actions on garlic, suggesting that the capacity to produce hydrogen sulfide is a good benchmark to standardize garlic dietary preparation (4, 9). Our lab has used garlic oil (GO) as a garlic preparation (26). In a typical preparation, an average of 2.5 g of GO was extracted from 1 kg of garlic cloves. Based on the gas chromatogram of the GO, its four major organosulfur components and their relative levels are diallyl disulfide (DADS) at 40.83%, diallyl trisulfide (DATS) at 38.93%, methyl allyl trisulfide at 7.17%, and diallyl sulfide (DAS) at 3.77% (26), indicating that DADS and DATS are the two most abundant garlic constituents. Our previous two reports (11, 26) used streptozotocin (STZ) -induced diabetic rats as our animal model system because STZ is a widely used chemical to induce diabetes in several kinds of animals by affecting their degranulation and necrosis of pancreatic β-cells. Besides, our previous two reports documented that the oral administration of GO, used as a whole, was able to ameliorate cardiac contractile dysfunction, cardiac hypertrophy, and apoptosis in STZ-induced diabetic rats (11, 26). Yet, the effects of individual garlic components on alleviating diabetic cardiomyopathy in diabetic rats were not investigated yet.

The purpose of this present study is to understand, at the molecular level, the cardioprotective roles of two major components of GO, DADS and DATS. Thus we fed the diabetic rats with DADS or DATS at a dose of 40 mg/kg body wt, which is calculated from the proportion of the content in GO at a dose of 100 mg/kg body wt (26). We employed echocardiography, TUNEL and DAPI staining, and Western blotting analysis to address this question. Here, we show that, like GO, both DATS and DADS, with the effective order DATS > DADS, possess therapeutic potential for protecting hearts from diabetic cardiomyopathy.



Garlic oil was prepared from fresh garlic (Allium sativum) that was purchased from the local market as previously described (26). Briefly, a steam distillation technique was used, and the essential components of garlic oil were analyzed and identified as 40% DADS, 40% DATS, 10% diallyl sulfide, and minor amounts of other compounds. For the animal study, DADS and DATS were purchased from Tokyo Kasei Chemical (Tokyo, Japan) and KLT Laboratories (St. Paul, MN). The monoclonal antibodies against caspase 3, PI3K, phospho-PI3K, Akt, and phospho-Akt were purchased from Cell Signaling Technology (Beverly, MA), and polyclonal antibodies against cytochrome c, SOD-1, α-tubulin, caspase 8, caspase 9, nitrotyrosine, COX-2, Fas, FasL, Bax, Bak, IGF-I, IGF receptor, and t-Bid were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Animal model and treatments.

Weanling male Wistar rats (4 wk old) were purchased from the National Animal Breeding and Research Center (Taipei, Taiwan). The animals were kept under a 12:12-h light-dark cycle and room temperature was maintained at 22 ± 1°C. All animals were given free access to water and standard laboratory chow (Lab Diet 5001; PMI Nutrition International, Brentwood, MO). Housing conditions and experimental procedures were performed according to the NIH Guide for the Care and Use of Laboratory Animals, and all protocols were approved by the Institutional Animal Care and Use Committee of China Medical University, Taichung, Taiwan. After 10-day acclimatization, diabetes was induced by injection of streptozotocin (STZ, 65 mg/kg body wt in citrate buffer, pH 4.5) into a lateral tail vein of animals. After three days of injection, blood glucose was measured with the Accu-Check Compack kit (Roche Diagnostics, Mannheim, Germany). Only animals in which hyperglycemia had been successfully induced were allowed to assigned to four groups (n = 8) (DM, DADS-40, DATS-40, and GO-100 groups), and received by gavage DADS (40 mg/kg body wt), DATS (40 mg/kg body wt), garlic oil (GO) (100 mg/kg body wt) or the vehicle (corn oil, 2 ml/kg body wt) every other day for 16 days. The dose of garlic oil in the present study was in accordance with the results of our previous study (26). The study showed dose-dependent antioxidative effects in rats fed with GO (0–100 mg/kg). The doses of DADS and DATS used were based on the analysis results of our garlic oil preparation, these two compositions representing 40% and 40% of the constituents, respectively; thus these two doses are similar to that in 100 mg/kg garlic oil. The other normal control animals (n = 8) were fed with corn oil (2 ml/kg body wt). After 16 days of treatment, all animals were anesthetized and echocardiographic analysis was performed. Then, they were killed and their hearts were collected for further analysis.

In vivo cardiac function.

Echocardiograms were recorded using a Hewlett-Packard Sonos 2500 (Andover, MA) sector scanner equipped with a 7.5-MHz phased-array transducer. From the parasternal long-axis view in two-dimensional mode (12, 18), an M-mode cursor was positioned perpendicularly to the interventricular septum and posterior wall of the left ventricle (LV) at the level of the papillary muscles and M-mode images were obtained. LV end diastolic diameter (LVEDD), LV end systolic diameter (LVESD), interventricular septal wall thickness at end diastole (IVSd), interventricular septal wall thickness at end systole (IVSs), left ventricular posterior wall thickness at end diastole (LVPWd), and left ventricular posterior wall thickness at end systole (LVPWs) were recorded with M-mode. During diastole, LV diameter and wall thickness were measured from the maximum chamber cavity. During systole, parameters were measured when maximum anterior motion of the posterior wall occurred. Fractional shortening percentage (FS%) was calculated from the M-mode LV diameters using the equation: [(LVEDD-LVESD)/LVEDD] × 100%.

TUNEL assay.

All of the procedures were performed as described previously (21). The 3-μm thick paraffin sections were deparaffinized. The number of left ventricular TUNEL-positive cardiac cells was determined. All morphometric analysis was performed by at least three independent individuals in a blinded manner.

Hematoxylin-eosin staining.

The left ventricles of rat hearts of the five groups (control, DM, DADS, DATS, and GO groups) were taken, fixed in 10% formalin, and embedded in paraffin. Paraffin blocks were sliced into 5-μm sections, and stained with hematoxylin-eosin solution. Photomicrographs were taken under the Zeiss Axiophot microscope.

Protein extraction from tissue.

The left ventricular tissues were homogenized, and protein therein was extracted in a PBS buffer (0.14 M NaCl, 3 mM KCl, 1.4 mM KH2PO4, 14 mM K2HPO4) at a concentration of 1 mg tissue/10 μl PBS for 5 min. The homogenates were centrifuged at 12,000 rpm for 30 min. Then supernatant fractions were collected for protein analysis.

SDS-PAGE and Western blotting analysis.

The protein level of cardiac tissue extract was determined using the Bradford protein assay (7). Extracted protein samples were then separated on 12% SDS-PAGE gels and transferred to nitrocellulose membranes. Nonspecific protein binding of antibody was stopped in blocking buffer (5% milk, 20 mM Tris-HCl, pH 7.6, 150 mM NaCl, and 0.1% Tween 20), and then the specific antibody was added in the blocking buffer for blotting at 4°C overnight. For repeated blotting, nitrocellulose membranes were stripped with Restore Western blot stripping buffer (Pierce Biotechnology, Rockford, IL) at room temperature for 30 min. Protein signal intensity was evaluated by using a Fuji LAS-3000 Imaging System (Fuji Photo Film). Alpha-tubulin was used as a loading control.

Statistical analysis.

Statistical differences were performed by one way-ANOVA followed by Tukey's post hoc test. The data were expressed as means ± SD. P < 0.05 was considered statistically significant.


Oral administration of DATS, DADS, or GO was able to alleviate cardiac malfunction in diabetic rats.

In order to assess the effect of the oral gavage of DATS, DADS, or GO on cardiac function and growth status of diabetic rats, we employed echocardiography to measure cardiac physiological parameters, and we, in addition, measured these rats' body weights and blood glucose levels. The results are tabulated in Table 1, and shown representatively in Fig. 1A. Rats suffering from diabetes gained significantly less weight, had much higher blood glucose levels, and had poorer cardiac physiological parameters (HR, SV, and FS%) than the control group (Table 1). These adverse conditions in diabetic rats were improved significantly in response to their oral administration of garlic oil (GO) or either of its two major components (DATS and DADS). Interestingly enough, in terms of FS%, the administration of GO, DATS, or DADS could restore cardiac contractile function to the normal level. Furthermore, in this aspect, DATS appeared to be the most effective, compared with DADS or GO (Table 1).

View this table:
Table 1.

Physiological parameters in rat hearts

Fig. 1.

Physiological and histological assessment of the inhibitory effect of diallyl disulfide (DADS), diallyl trisulfide (DATS), and garlic oil (GO) supplementation, respectively, on apoptosis in diabetic rat hearts. A: I–V are representative echocardiograms from control, diabetes mellitus (DM), DADS-40, DATS-40, and GO-100 groups, respectively. B: I–V are representative photos for the DAPI staining and TUNEL assay results from control, DM, DADS-40, DATS-40, and GO-100 groups. VI is the quantified representation of the TUNEL assay results. The y-axis values were shown as a percentage relative to the control group. Results are shown in means ± SD for each group (DM, DADS-40, DATS-40, and GO-100). C: I–V are representative hematoxylin and eosin (H and E) staining photos from control, DM, DADS-40, DATS-40, and GO-100 groups, respectively. DM, DADS-40, DATS-40, and GO-100 represent oral gavage of 0 mg of GO (garlic oil), 40 mg of DADS (diallyl disulfide), 40 mg of DATS (diallyl trisulfide), and 100 mg of GO per kilogram of body weight, respectively. **P < 0.01 compared with control group; #P < 0.05 compared with the DM group.

The administration of garlic oil or either of its two major components was able to prevent diabetic rats from undergoing apoptosis.

To evaluate histologically the status of apoptosis of diabetic cardiocytes, we employed DAPI staining and TUNEL assay. In the TUNEL assay, the apoptotic nuclei were identified by the characteristic fragmented nuclei. In the DAPI staining, the nuclear location could be seen. The results are shown in Fig. 1B. As shown in Fig. 1, B-VI, the DM group has more positive TUNNEL-positive cells than the control group. Interestingly enough, the DADS group, the DATS group, and the GO group have roughly the same number of TUNEL-positive cells as the control group. Additionally, the results of hematoxylin-rosin staining showed that the diameter of the left ventricular lumen increased in the DM groups, and decreased to the roughly normal value upon the treatment of DADS, DATS or GO (Fig. 1C). Taken together, garlic and its constituent feeding were able to inhibit apoptosis in diabetic rats.

The levels of RNS and ROS were elevated, and the level of SOD-1 was increased in diabetic rats, and these alterations were improved significantly by administration of garlic oil or either of its two major components.

We, next, studied the effects on the RNS and ROS levels in diabetic rats. We used the intensity of nitotyrosine (NY)-modified protein bands (detected by antibodies recognizing NY in Western blots) as a quantitative measure of the in vivo RNS level, used the intensity of COX-2 protein bands as a monitoring ROS marker, and used the intensity of SOD-1 band for accessing the ROS stress. The results are shown in Fig. 2. Diabetes caused the high expression levels of NY and COX-2, and low expression level of SOD-1. Moreover, feeding DATS, DADS, or GO could alleviate the RNS and ROS stress significantly. In terms of decreasing the level of NY-modified proteins, DADS appeared to be the most effective. Regarding increasing the level of SOD-1 in vivo, both DATS and GO were better than DADS.

Fig. 2.

Administrations of DADS, DATS, and GO dampen the oxidative stress in diabetic rats. A: the rat's cardiac nitrotyrosine (NY) and SOD-1 protein expression levels analyzed by Western blotting. Alpha-tubulin was used as a loading control. B: intensity of protein bands shown was quantified. The y-axis values were shown as a percentage relative to the control group. Results are shown in means ± SD for each group (DM, DADS-40, DATS-40, and GO-100). *P < 0.05, ***P < 0.001 compared with control group; ##P < 0.01, ###P < 0.001 compared with the DM group.

Garlic oil and its two major components suppressed both death receptor-dependent and mitochondria-dependent apoptotic pathways.

To further elucidate the molecular mechanism underlying the protective roles of garlic and its two components in apoptosis, we employed the Western blotting analysis to quantitatively determine expression levels of many related apoptotic proteins in the control, DM, DADS, DATS, and GO groups, respectively. The results are shown in Fig. 3, A–E. First, we turned our attention to the death receptor-dependent apoptotic pathway. Compared with the control group, Fas protein expression has a 1.48 (±0.12) -fold increase in the DM group, a 1.16 (±0.36) -fold increase in the DADS-40 group, a 0.90 (±0.24) -fold increase in the DATS-40 group, and a 0.80 (±0.25)-fold increase in the GO-100 group. Thus a rough trend emerges: In terms of the ability to inhibit the expression of Fas, GO is more effective than DATS, and DATS is more effective than DADS. Likewise, in comparison with the control group, a similar trend also holds for the inhibitory role in the expression of FasL: GO > DATS > DADS (0.92 ± 0.18 vs. 0.98 ± 0.23 vs. 1.25 ± 0.21 -fold increase) (Fig. 3A).

Fig. 3.

The rat's cardiac expression levels of proteins in apoptotic pathways analyzed by Western blotting. A: Western blots of FAS and FasL, and their quantitative representations. B: Western blots of active-caspase-8, and its quantitative representation. C: Western blots of t-Bid, Bax, Bak, and Cyto c and their quantitative representations. D: Western blots of active caspase-9, p-Bad, Bcl-2, and Bcl-xL and their quantitative representations. E: Western blot of active caspase-3 and its quantitative representation. The y-axis values were shown as a percentage relative to the control group. Results are shown in means ± SD for each group (DM, DADS-40, DATS-40, and GO-100). *P < 0.05, **P < 0.01, ***P < 0.001 compared with control rats; #P < 0.05, ##P < 0.01, ###P < 0.001 in comparison with the DM group.

Subsequently, we examined the inhibitory effect on the level of active form of caspase 8, which is converted from pro-caspase 8 catalyzed DISC (death-inducing signaling complex) when death receptor-dependent apoptotic signaling is activated. As shown in Fig. 3B, compared with the control group, the active caspase 8 level has a 4.66-fold increase in the DM group, a 1.53-fold increase in the DADS-40 group, a 1.25-fold increase in the DATS-40 group, and a 0.81-fold increase in the GO-100 fold increase. Thus, once again, the inhibitory order trend, GO > DATS > DADS, also held for the active caspase 8.

Third, we investigated the role of DADS, DATS, and GO, respectively, in affecting the expression levels of many members of the Bcl-2 protein family (possessing either pro-apoptotic activities or anti-apoptotic activities) and other proteins (cyto c and active caspase 9), involved in mitochondria-dependent apoptotic pathway. These results are shown in Fig. 3, C and D. Relative to the control group, pro-apoptotic proteins (including t-Bid, Bax, and Bak) were upregulated in the DM group, but showed significantly less increased expression levels in the DADS-40, DATS-40, and GO-100 groups (Fig. 3C). Furthermore, the effective order trend is GO > DATS > DADS. By the same token, with respective to the control group, anti-apoptotic proteins (including p-Bad, Bcl-2, and Bcl-xL) were downregulated in the DM group, but showed more increased expression levels in the groups fed with DADS, DATS, and GO, respectively (Fig. 3D). Moreover, the effective order trend is GO > DATS > DADS. Two molecules under study did not appear to follow the assumed effective order. One is active caspase 9 (DATS > GO > DADS, due to the noisy data of DATS); the other is cyto c (DADS > DATS > GO, due to the noisy data of DATS and GO).

Finally, we examined the role of DADS, DATS, and GO, respectively, in affecting the level of caspase 3, whose activation is very critical for the execution of apoptosis, and is mediated either caspase 8 activation or caspase 9 activation. The result is shown in Fig. 3E. With respect to the control group, we observed the 2.32-fold increase in the DM group, but the 1.08-fold increase in the DADS-40 group, the 0.76-fold increase in the DATS-40 group, and the 0.74-fold increase in the GO-100 group. The familiar effective order, GO > DATS > DADS, appeared again.

Garlic oil and garlic components were able to enhance the IGF1-PI3K-Akt pathway in diabetic rats.

Besides inhibiting two kinds of apoptotic pathways mentioned above, another way to gain resistance to apoptosis is enhancing cellular survival pathway(s). One obvious target was IGF1-PI3K-Akt pathway in diabetic rats. We used the Western blotting analysis to quantify the extent to which DADS, DATS, and GO, respectively, affected the expression levels of proteins in the IGF1-PI3K-Akt pathway in diabetic rats. The results were shown in Fig. 4. Relative to the control group, IGF1 was downregulated in the DM group, but was upregulated in the DADS-40, DATS-40, and GO-100 groups. Similarly, the level of phosphorylated Akt (pAkt) was decreased significantly (almost by 5-fold) in the DM group, but had a 0.8-fold increase in the DADS-40 group, a 1.13-fold increase in the DATS-40 group, and 1.33-fold increase in the GO-100 group. Taken together, the effective order, GO > DATS > DADS, held for IGF1 and pAkt. Also, the effective order held for p-IGF1R with two exceptions. Compared with DATS and DADS, GO was the third effective in improving the low level of p-IGF1R in diabetic rats.

Fig. 4.

The rat's cardiac expression levels of proteins in survival pathways analyzed by Western blotting. The upper part shows the Western blots of proteins (IGF1, p-IGF1R, IGF1R, p-Akt, and Akt), and the lower part displays their quantitative representations. The y-axis values were shown as a percentage relative to the control group. Results are shown in means ± SD for each group (DM, DADS-40, DATS-40, and GO-100). ***P < 0.001 compared with the control group; #P < 0.05, ##P < 0.01, ###P < 0.001 compared with the DM group.


In our previous reports, we utilized the STZ-induced diabetic rat model to show that diabetes developed cardiac hypertrophy (structurally characteristic of diabetic cardiomyopathy) via MAPKs (e.g., p38, JNK and ERK1/2) and IL-6/MEK5/ERK5 signaling pathways (11), and developed cardiac contractile dysfunction (pathophysiologically characteristic of diabetic cardiomyopathy) via p38-NFκB and apoptotic pathways (26). Furthermore, garlic oil supplementation mitigated these cardiac abnormalities and malfunctions in diabetic rats. In this present work, we provide strong evidence to show that in these diabetic rats, the intracellular ROS and RNS levels were elevated and cardiac contractile dysfunction developed by means of downregulating IGF1-PI3K-Akt signaling and upregulating death receptor-dependent and mitochondria-dependent apoptotic pathways. Moreover, here we establish that oral administration of garlic oil and its two major components (DATS and DADS) could ameliorate these cardiac alterations in diabetic rats. Taken together, a series of our work has paved the way for elucidating the molecular mechanism underlying diabetic cardiomyopathy, and may suggest strongly that DATS and DADS as well as garlic oil should have therapeutic potential for having cardioprotective roles and that the therapeutic effects of GO on diabetic cardiomyopathy should be from DATS and DADS (Fig. 5).

Fig. 5.

The schematic diagram summarizes our previous work (11) and current work on dissecting the molecular mechanism underlying diabetic cardiomyopathy. The oxidative stress (caused by hyperglycemia) downregulates the IGF1-PI3K-Akt pathway, and upregulates death-receptor-dependent apoptotic pathway and mitochondria-dependent apoptotic pathway, and upregulates the MAPKs (p38, JNK, and ERK1/2) survival pathways in diabetic rats, leading to the development of diabetic cardiomyopathy. However, in diabetic rats, the supplementation of garlic oil and its two major components (DADS and DATS) can reverse all of these abnormalities and improve the heart function presumably by acting as the donors of H2S which neutralizes the oxidative stress, and signals possibly mainly by S-sulfhydrating the Cys residues of key proteins (NF-κB, for example) in signaling cascades underlying cardioprotection.

Here, we also document that in terms of exerting cardioprotective effects, the order of efficacy is GO (100 mg/kg body wt) > (≥, in some cases) DATS (40 mg/kg body wt) > DADS (40 mg/kg body wt) with few exceptions. This observation that DATS is more effective than DADS is correlated with the number of sulfur atoms within the molecule (37), and with the total yield of H2S (hydrogen sulfide) which the organosulfur compound can produce (4). The very similar structure-activity relationship of garlic-derived polysulfides has been reported for their modulatory effects on the hepatic drug-metabolizing enzymes (37), and for their cancer-preventive effects in rats (24). Thus DATS and DADS, the major constituents of garlic oil, function as donors of H2S. Recently, hydrogen sulfide has been increasingly recognized as playing an important role in cardioprotection (9, 23) due to its powerful reducing potential (i.e., antioxidant properties) (which may dampen the oxidative stress, the major cause to trigger diabetic cardiomyopathy), and due to its ability to modulate cellular signaling pathways possibly by causing S-sulfhydration of the Cys residues of some key protein molecules (25). This provides the compelling and unifying basis for interpreting our data presented here as well as beneficial effects of garlic-related study reported by other research groups worldwide.

Antioxidant effects.

Recently, hydrogen sulfide has been reported to suppress the oxidative stress by increasing intracellular reduced GSH concentrations (20), and by increasing the local concentration of nuclear factor-E-2-related factor (Nrf2) in the nucleus and thereby upregulating the gene expression of many antioxidant enzymes (10). Interestingly, induced expression of these upregulated genes bearing the antioxidant responsive element (ARE) includes that of SOD (which works together with catalase in detoxifying superoxide and H2O2)(41), a finding consistent with our result shown in Fig. 2. Moreover, the finding that garlic and its major constituents, presumably inducing H2S production (4), could activate Nrf2-mediated NQO1 [NAD(P)H quinone oxidoreductase 1] gene expression strengthens the role of H2S in Nrf2-mediated antioxidant gene expression (13).

Antiapoptotic effects.

A large body of evidence indicates that hydrogen sulfide possesses antiapoptotic effects (9). Mostly notably, a very recent report indicated that hydrogen sulfide may sulfhydrate posttranslationally the Cys38 residue of RelA, the p65 subunit of NF-κB, which, in turn, is activated to bind to its coactivator RPS3, thereby mediating the transcription of several antiapoptotic genes (32). This interesting finding supports the emerging concept that H2S signals possibly mainly by S-sulfhydrating the Cys residue(s) of its target proteins (23, 25). In addition, using an in vivo model of H2S-mediated pharmacological preconditioning, during the late stage of the preconditioning period, H2S increased the expression of Bcl-2 and Bcl-xL (antiapoptotic proteins), and inactivated Bad (proapoptotic protein) and increased the expression of HSP70 and HSP90 heat shock proteins [which can suppress apoptosis in the intrinsic, extrinsic, and caspase-independent manners (22, 29, 38)], consistent with our current result (Fig. 3). Besides, denitrosylation of some nitrosylated caspases including procaspase 9 and procaspase 3 leads to caspase activation and promotes apoptosis and vice versa (5, 6). This line of evidence also supports our observation seen in Fig. 3 given that S-nitrosylation (mediated by NO) and S-sulfhydation (mediated by H2S) appear to act reciprocally to regulate the activities of key proteins in cellular signal transduction pathways (23).

Prosurvival effects.

The prosurvival properties of hydrogen sulfide have been documented in literature (9). Yet, H2S can either activate or inhibit prosurvival kinase activities in different cellular context (3). In our previous investigation, we showed that GO supplementation could downregulate, rather than upregulate, MAPKs (p38, JNK, and ERK1/2) in diabetic rats (11), in line with the observation that hydrogen sulfide inhibits p38 MAPK phosphorylation (30). In this present study, however, we document that the administration of GO, DADS, and DATS can upregulate the IGF1-PI3k-Akt survival pathway in diabetic rats, consistent with the report that H2S could stimulate Akt and protein kinase C activities in isolated hearts (39). Moreover, the observation that diabetic myocardium is associated with lower level of IGF-1R protein (which is due to the increased ubiquitination of IGF-1R) and with lower level of HSP 60 (which can suppress ubiquitination of IGF-1R) highlights the importance of the functional role of the IGF1-PI3k-Akt survival pathway in controlling the development of diabetic cardiomyopathy (33). This is in tune with our finding (Fig. 4) that the administration of GO, DADS, and DATS can enhance this survival pathway activity, thereby mitigating diabetic cardiomyopathy in diabetic rats.

Taken together with all of our findings augmented with literature survey, our current study suggests that the administration of GO along with its two major constituents, DADS and DATS, alleviates diabetic myocardiopathy in diabetic rats by inducing the production of H2S which dampens the oxidative stress, and modulates cellular signaling possibly mainly by S-sulfhydrating the Cys residues of key proteins (NF-κB, for example) in signaling cascades, leading to inhibiting apoptotic pathways and promoting the survival pathway. As to future perspectives, to explore the list of potential targets for H2S-mediated S-sulfhydration using mass spectrometry-based proteomic approach or to ascertain whether H2S-mediated S-sulfhydration is the only one posttranslational modification means for H2S signaling will await interested investigators.


This study was supported by Taiwan Department of Health Clinical Trial and Research Center of Excellence (DOH101-TD-B-111–004), by China Medical University (CMU100-S-08), and by the National Science Council, Taiwan (NSC 96-2320-B-039-035 MY3)


No conflicts of interest, financial or otherwise, are declared by the author(s).


Author contributions: Y.-T.H., C.-H.Y., C.-L.W., K.-W.L., and C.-Y.T. analyzed data; Y.-T.H., C.-H.Y., C.-L.W., and H.-C.O. interpreted results of experiments; Y.-T.H. and C.-L.W. prepared figures; Y.-T.H. drafted manuscript; Y.-T.H. and W.-W.K. edited and revised manuscript; Y.-T.H. and W.-W.K. approved final version of manuscript; C.-H.Y., C.-L.W., and W.-W.K. conception and design of research; C.-H.Y., C.-L.W., K.-W.L., and C.-Y.T. performed experiments.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
View Abstract