Caspase Mediated Synergistic Effect of Boswellia serrata Extract in Combination with Doxorubicin against Human Hepatocellular Carcinoma

The study investigated the growth-inhibiting and apoptosis mediating effects of B. serrata extract as monotherapy and combination therapy with DOX against hepatocellular carcinoma cell lines. Boswellic acid rich fraction of B. serrata extract was prepared. MTT assay on HepG2 and Hep3B cells was carried out using B. serrata alone and in combination with DOX. Further, caspase-3 activity, TNF-α level, and IL-6 level were estimated. Isobolographic analysis was carried out to evaluate the effect of combination therapy. Additionally, protective effect of B. serrata extract on DOX induced hepatic toxicity was also evaluated in Wistar rats. B. serrata extract inhibited growth of HepG2 (IC50 value of 21.21 ± 0.92 μg/mL) as well as HepG2 (IC50 value of 18.65 ± 0.71 μg/mL). DOX inhibited growth in HepG2 and Hep3B cells with an IC50 of 1.06 ± 0.04 μg/mL and 1.92 ± 0.09 μg/mL. Isobolographic analysis showed combination index (CI) of DOX and B. serrata extract of 0.53 ± 0.03 to 0.79 ± 0.02 suggesting synergistic behavior against the two cell lines. B. serrata extract also caused dose dependent increase in caspase-3 activity, TNF-α level, and IL-6 level which was higher (P < 0.001) with DOX (1 μM) and B. serrata extract (20 μg/mL) combination. B. serrata extract also protected Wistar rats against DOX induced hepatic toxicity.


Introduction
Hepatocellular carcinoma (HCC) is one of the most common cancers and the major leading cause of cancer related mortality globally [1]. The incidence of HCC is also increasing in Asia, Africa, Europe, and USA [2]. The lack of reliable biomarkers to detect HCC at early stage has caused the manifestation of advanced stages in most patients when surgical resection is not feasible [3]. Therefore, chemotherapy remains the viable strategy for treatment of inoperable cases of HCC. Doxorubicin (DOX) is one of the widely used anticancer drugs in the treatment of various malignancies including HCC [4]. However, the objective response rate with DOX as single agent is less than 20% whereas the median survival is only 4 months. Furthermore, clinical use of DOX is limited by severe adverse effects to nontumorous hepatic tissue [5,6]. Impaired liver function, severe pancytopenia, and mucositis with DOX therapy can prompt to reduce the dosage [7]. However, the mechanisms of DOX-mediated cytotoxicity are different in cancer and normal tissues [8][9][10]. This difference in DOX-mediated toxicity can be used as an opportunity to improve the antitumor therapy with DOX.
Boswellia extracts, derived from the plant Boswellia serrata (B. serrata), have been known for their anti-inflammatory effects in the treatment of arthritis, ulcerative colitis, and Crohn disease [11]. Boswellic acids from B. serrata have also been proposed to provide antineoploastic activity through their antiproliferative and proapoptotic properties in multiple human cancer cell lines including meningioma cells [12], leukemia cells [13], melanoma cells, fibrosarcoma cells [14], colon cancer cells [15], and prostate cancer cells [16]. Boswellic acids (BA) from Boswellia serrata can induce apoptosis in cancer cells through activation of proapoptotic bcl-2 family and caspase-3 and upregulation of cell death receptors DR4 and TNF-R1 levels, leading to caspase-8 activation [17].
Multidrug therapy is a common practice in oncology since it can achieve better therapeutic outcome than monotherapy and can also minimize the adverse effects [18]. Thus it is imperative to design new combinatorial approaches with DOX that allow dose reduction, enhance the drug effectiveness, and reduce the toxicity.
The present study aimed to investigate the growth-inhibiting and apoptosis mediating effects of B. serrata extract alone as well as in combination with DOX in order to provide a new adjuvant therapy for HCC.

Procurement of Drug.
Gum of B. serrata was procured from local market and samples were authenticated by Dr. H. B. Singh at NISCAIR (New Delhi), India.

Preparation of Methanolic Extract of B. serrata.
Weighed quantity of B. serrata gum (50 g) was size-reduced and powdered gum was mixed with 100 mL methanol in a round bottom flask of 250 mL capacity and refluxed for 2 hours. The solvent was filtered to obtain methanolic extract (Part A) and process was repeated twice (parts B and C). All the three extracts, that is, parts A, B, and C, were mixed to obtain total methanolic extract. Methanolic extract was further concentrated on water bath and evaporated overnight. Residue obtained was weighed to obtain percentage yield.

Isolation of Boswellic
Acids from B. serrata. Concentrated methanolic extract of B. serrata was prepared as described. To the concentrated 20% Aq. KOH was added with constant stirring to dissolve till pH is 9-10. After the desired pH was obtained, the alkaline extract was filtered to remove any precipitate and further washed with equal volume of hexane thrice. After washing it was acidified with conc. HCl and final pH was adjusted to 2 so that formation of precipitate occurred. Precipitate so obtained was filtered and washed with water till neutral, dried in the air, and weighed.  (c) Isolated Boswellic Acids. 50 mg of isolated boswellic acids were dissolved in 10 mL of HPLC grade methanol and solution was subjected to sonication for 20 minutes.

Quantification of Boswellic Acids in B. serrata Extracts.
The analysis was carried out on a Waters Alliance e2695 separating module (Waters Co., MA, USA) using photo diode array detector (waters 2998) with autosampler and column oven. The instrument was controlled by use of Empower software installed with equipment for data collection and acquisition. Standard and sample solutions were filtered through 0.22 m syringe filter before injection and the separation was achieved by using a LiChroCART C 18 column (25 × 4.6 mm, particle size 5 m). The mobile phase used was acetonitrile and 0.05% acetic acid in the ratio of 90 : 10 in gradient elution mode. Individual peaks were identified from retention time and concentrations were determined from the peak area for appropriate sample solutions using regression equation obtained from calibration plot. The detection of boswellic acids was done at the wavelength of 254 using PDA detector.

In Vitro Cytotoxicity Assay on HepG2 Cells
2.5.1. Cell Culture Treatment. HepG2 and Hep3B cells were cultured in respective media supplemented with 100 U/mL penicillin, 100 U/mL streptomycin, and 10% foetal bovine serum (FBS) and kept at 37 ∘ C in a humidified atmosphere of 5% CO 2 . All cells were plated in cell culture flasks at least 24 h before treatment.

Cell Viability Assay for B. serrata Extract Per Se.
Cells were plated at an initial density of 8 × 10 3 cells/well in a 96well plate and treated with 10, 20, 40, 80, and 160 g/mL of B. serrata extract for 72 h. MTT solution was added to each well to a final concentration of 0.5 mg/mL. After 4 h of exposure at 37 ∘ C, the absorbance was measured at 570 nm using ELISA reader.

Evaluation of B. serrata Extract with DOX.
Cells were plated at an initial density of 8 × 10 3 cells/well in a 96-well plate and treated with 1 M DOX per se and 1 M DOX in combination with 20 g/mL of B. serrata extract for 72 h. MTT solution was added to each well to a final concentration of 0.5 mg/mL. After 4 h of exposure at 37 ∘ C the absorbance was measured at 570 nm using ELISA reader.
2.6. Cytokine Assay. Cells were collected after treatment and lysed with 50 L cell lysis buffer. Then the cell lysate was centrifuged and the supernatants were retained. TNFquantification, caspase-3 activity, and IL-6 level estimation were carried out using commercially available ELISA kits. NF-B expression was evaluated using western blotting.

Isobolographic Analysis.
To study the interaction between DOX and B. serrata extract, Cells were treated with various dilutions of DOX in the presence or absence of B. serrata extract at 10, 20, and 30 g/mL and IC 50 values were determined by plotting the percentage of cell survival as a function of drug concentration [18,19]. The interactions between DOX and B. Serrata extract were evaluated by isobolographic analysis, a dose-oriented geometric method of assessing drug interactions [20]. For 50 percent toxicity, the combination index (CI) values were calculated based on the equation stated below: Combination Index (CI) where 1 is dose of drug 1 to produce 50 percent cell kill alone; 1 is dose of drug 1 to produce 50 percent cell kill in combination with 2; 2 is dose of drug 2 to produce 50 percent cell kill alone; 2 is dose of drug 2 to produce 50 percent cell kill in combination with 1; and = 1.

Animals and Treatment Protocols.
Adult albino (Wistar strain) rats weighing between 150 and 200 gm were used for the study. The animals were procured from Central Animal House Facility, Jamia Hamdard. Throughout the experimental period, the animals were housed in cages under room temperature (20 ± 2 ∘ C) and relative humidity (60-70%) and exposed to 12 : 12 h light : dark cycle. They were fed with standard laboratory diet supplied by M/S Ashirwad feed industry and water ad libitum.
The study was approved by Institutional Animal Ethics Committee of the University and animals received humane treatment as per Committee for the Purpose of Control and Supervision on Experiments on Animals (CPCSEA) guidelines. Animals were divided into five treatment groups. Group I was control and received normal saline 1 mL/kg p.o. for 30 days. Group II (toxic control) received normal saline 1 mL/kg p.o. for 30 days with DOX 30 mg/kg i.p. on 28th day. Group III (DOX + BS250) and IV (DOX + BS500) were given B. serrata extract at 250 and 500 mg/kg p.o., respectively, for 30 days with DOX 30 mg/kg i.p. single dose on 28th day. Group V (DOX + SY100) served as standard and was administered silymarin 100 mg/kg for p.o. for 30 days with DOX 30 mg/kg i.p. on 28th day.

Doxorubicin Hepatic Toxicity
Assay. 72 hours after DOX administration, blood was withdrawn from the tail vein for serum estimations. Followed by this animals were sacrificed under high dose of anaesthesia and liver tissue was excised out washed with ice cold saline and stored for biochemical and histopathological studies.
Biochemical Estimations in Serum and Tissue. Serum samples were used for liver function test and estimation of serum SGOT, SGPT, albumin, bilirubin, and alkaline phosphatase levels using commercially available kits. Extent of lipid peroxidation was measured as by thiobarbituric acid (TBA) reaction with malondialdehyde (MDA) according to the method described by [21].
2.10. Histopathology. The excised liver tissues were fixed in 10% neutral buffered formalin and sections of maximum 5 m thickness were cut. Further, sections were stained with hematoxylin and eosin (H&E) for histopathology. The histomorphological evaluation of the liver sections was done by a pathologist unacquainted with the treatment groups.
2.11. Statistical Analysis. The Data was represented as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was used to compare the means of all groups. The Tukey-Kramer post hoc test was used to test the significance among groups. The value of < 0.05 was considered to be statistically significant. GraphPad Prism was used for statistical analysis.

HPLC Analysis of Boswellic Acids.
The HPLC-PDA method was developed and successfully employed for the identification and quantification of acetyl-keto--boswellic acid (AKBA) and keto--boswellic acid (KBA) in prepared extracts of B. serrata. Identification of the peaks in the sample chromatograms were carried out by comparing retention and PDA spectra of each component (Figures 1(a)-1(c)). The contents AKBA and KBA were calculated from the corresponding calibration curve. The quantities of AKBA and KBA were 5% and 1.87% w/w, respectively, in prepared extract whereas the content of these two components were 0.917% and 0.71% w/w, respectively, in methanolic extracts.    Figure 3. B. serrata extract treatments at 5 g/mL showed insignificant difference compared to control treatment. Treatment at 10 g/mL increased caspase activity significantly ( < 0.01) to almost 1.5 fold of control level. However, treatment at 20-40 g/mL increased caspase-3 activity to even higher level ( < 0.001). Results showed that B. serrata alone and in combination with DOX showed significant ( < 0.001) increased expression of caspase-3 activity compared to control. The expression of caspase-3 in combination treatment was significantly ( < 0.001) different from single agent treatment.

TNF-Level. B
. serrata extract treatments significantly increased TNF-level compared to control (Figure 4) except at the dose of 5 g/mL ( > 0.05). TNF-level at the doses of 20-40 g/mL was significantly higher ( < 0.001 versus control) than 10 g/mL in both HepG2 cells ( < 0.01 versus control, Figure 4(a)) as well as in Hep3B cells ( < 0.05 versus control, Figure 4(b)). Further, results showed that treatment with DOX alone and in combination with B. serrata extract significantly ( < 0.001) enhanced TNF-level compared to control. TNF-level in combination treatment was significantly ( < 0.001) higher from treatment with B. serrata extract or DOX.

IL-6 Level
. IL-6 level in HepG2 treated with B. serrata extract showed dose dependent significant ( < 0.001) increase compared to control ( Figure 5(a)). In Hep3B cells treatment showed insignificant change at a dose of 5 g/mL and dose dependent increase in IL-6 level at doses above 5 g/mL. DOX and combination treatment of DOX and B. serrata extract also caused increased IL-6 levels ( < 0.001 versus control). The expression of IL-6 in combination treatment was significantly ( < 0.001) different from any single drug treatment.  single treatments. Similar results were obtained in Hep3B cells (Figure 6(b)).

Hepatic Toxicity Assay
3.4.1. Hepatic Enzyme Activity. The activity of serum markers of liver injury SGOT, SGPT, and ALP was significantly elevated ( < 0.001) in rats treated with DOX due to toxic effect (Table 2). Although the enzyme activities were significantly lower in the control group. Treatment with B. serrata extract both at 250 and 500 mg/kg/day significantly reduced ( < 0.001) activities of SGOT, SGPT, and ALP in comparison to DOX treated group. In the present study, enzyme levels in the standard group treated with sylimarin (100 mg/kg/day) showed comparable improvement ( > 0.05) to groups treated with B. serrata extract.

Serum Protein Levels.
The serum albumin levels in toxicant group were significantly reduced ( < 0.001) compared to control animals ( Table 2).Treatment with B. serrata extract restored albumin levels to significantly higher level compared to DOX treated animals ( < 0.001); this effect was also comparable to standard group ( > 0.05). Serum bilirubin level was significantly increased in DOX treated animals ( < 0.001). All the treatment significantly restored bilirubin level ( < 0.001 versus control). Within the liver tissues of DOX treated animals, there was significant alteration detected in the MDA levels compared to the control group, which was >6 fold of control level. A substantial decrease was observed in the MDA levels of the rats, which were administered B. serrata extract at 250 and 500 mg/kg/day ( < 0.05).  cells with prominent round nuclei and well preserved cytoplasm, prominent nucleus, nucleolus, and visible central vein (Figure 7(a)) (normal hepatic architecture). The liver sections of the animal intoxicated with DOX showed massive fatty changes, necrosis, ballooning degeneration and broad infiltration of lymphocytes, loose cellular boundaries, vacuolated hepatocytes, and mild periportal inflammatory cell infiltration (Figure 7(b)). The histological architecture of liver sections of animal treated with B. serrata extract at 250 mg/kg showed more or less normal lobular pattern with mild degree of fatty changes (Figure 7(c)). The animal treated with B. serrata extract at 500 mg/kg showed the centrizonal area with almost nil inflammatory cell infiltration (Figure 7(d)).
The results are comparable with standard (Figure 7(e)) and control.

Discussion
HCC is one of the most lethal forms of cancers. This is substantiated by the facts that even though there are significant advances in surgery and chemotherapy, the majority of patients with HCC die within one year of diagnosis [22]. Additionally, the current treatment approaches are associated with side effects leading to inadequate outcomes. Thus, there is desperate need for new treatment options of HCC. Cancer therapy with single drug confers quick development of drug resistance where combination therapy simultaneously with two or more drugs has ability to improve results and reduce the chances of resistance [23]. Although there have been advancement in drug targeting as well as combination therapy, no curative treatment options are yet achieved. This problem requires development of better therapeutic agents and combination strategies [24,25]. Recent studies on brain tumors, leukemic, and colon cancer cells indicate that boswellic acids from B. serrata may have antiproliferative effects [15]. DOX is one of the most commonly used anticancer agents against HCC; however, it is associated with severe toxicities to vital organs [19,20,26]. In the present study, we demonstrated that B. serrata extract containing boswellic acids in combination with DOX was effective in a synergistic manner in inhibition of tumor growth of HCC in vitro and also protected against toxicity in vivo. Our results indicate that the anticancer activity of B. serrata alone and in combination with DOX was mediated via activation of caspase cascade and induction of apoptosis. To our knowledge, this is the first study that demonstrates effectiveness of B. serrata in combination with DOX against HCC. In this study, we demonstrated that B. serrata inhibits HepG2 cells proliferation in vitro with IC50 value of 21.21 ± 0.92 g/mL (Figure 2(a)) and Hep3B cells with 18.65.21 ± 0.71 g/mL (Figure 2(b)). This was lower than the IC50 observed with AKBA treatment in other cancer cells [16,27,28].
The antiproliferative potential of different agents has been found to vary based on sensitivity and type of cancer. In our study, HepG2 and Hep3B cells were used to establish interaction between B. serrata extract and DOX using isobolographic analysis. Isobolographic analysis has been commonly used to assess the interaction between two antitumor agents. It also provides qualitative as well as quantitative measure of nature and extent of interaction [29]. In the present investigation, isobolographic analysis showed that B. serrata enhanced the cytotoxicity of DOX in both HepG2 cells (CI value 0.53-0.76) Hep3B cells (CI value 0.55-0.79) in a synergistic manner ( Table 1).
Inhibition of cell proliferation by induction of apoptotsis is one of the important mechanisms by which anticancer agents act [30]. Ottewel et al. [2008] have previously shown that combination of zoledronic acid and DOX caused significant ( < 0.05) increase in caspase-3 positive cells in MDA-G8 breast tumor xenografts compared to single treatment. To study the possible mechanism involved in the anticancer activity of B. serrata extract and combination, we evaluated induction of apoptosis of HepG2 cells by measuring caspase-3 activity. Induction of caspase-3 has been demonstrated following boswellic acid treatment in colon cancer [15], lukemic cells [31], and prostate cancer cells [32]. Induction of apoptosis and expression of cleaved caspase 3 was significantly ( < 0.001) induced in vitro by combination treatment compared to B. serrata or DOX alone (Figure 3). In in vitro increase in caspase-3 activity correlated very well with cytotoxicity thus confirming that caspase mediated apoptosis is an important pathway associated with the anticancer activity of these agents. Further, TNF-are cytokines which can stimulate the acute inflammatory response. TNF-is also able to induce apoptotis and inhibit tumorigenesis via caspase-3 [33][34][35][36]. In this study, the secretion of TNFincreased in a dose dependent manner in both HepG2 and Hep3B cells ( Figure 4). As observed with caspase-3, TNFlevel with combination treatment was significantly higher ( < 0.001) than single use of B. serrata or DOX. These results reinforced that apoptosis of HCC cells as induced by TNF-treatment is mediated through caspase-3. The tumor regression by DOX and combination was also mediated through increased expression of IL-6 levels ( Figure 5). DOX is reported to increase IL-6 expression through generation of oxidative stress. B. serrata is a proven anti-inflammatory agent in normal tissue, but our study showed that it can per se increase IL-6 level in carcinoma cells. This effect was synergistic in combination with DOX. To gain more insights on the mechanisms of increased IL-6 level with boswellic and combination therapy, other nonapoptotic signaling pathways need to be investigated. NF-B has been shown to promote cell survival signals leading to inhibition of apoptosis and cancer growth [37]. Previously boswellic acid has been shown to inhibit TNF-induced activation of NF-B [38] and in our experiment B. serrata extract and combination treatment inhibited NF-B activation ( Figure 6). This confirmed that combination of B. serrata extract and doxorubicin exhibit synergistic apoptosis mediated via NF-B.
After establishing the efficacy of B. serrata extract alone and combination with DOX on HepG2 cells in vitro, we designed in vivo experiments to test the efficacy of B. serrata extract against DOX-induced hepatic toxicity in Wistar rats. Previous studies have shown that prophylactic oral administration of B. serrata extracts prevented the development and progression of hepatic fibrosis mice [39]. Therefore, we evaluated B. serrata extract for in vivo hepatoprotective efficacy at two different dose levels of 250 to 500 mg/kg/day against DOX. Our results demonstrated that oral administration of B. serrata extract at 250 and 500 mg/kg/day showed significant improvement in liver function (Table 2). Animals were treated with DOX at 30 mg/kg (i.p.), a dose that causes acute toxicity [40]. The acute hepatotoxicity of DOX was clearly revealed by the increase in serum biochemical markers SGOT, SGPT, ALP, and bilirubin and decrease in serum albumin level. The most important cause of DOX toxicity is oxidative stress [41]. DOX generates free radical either by enzymatic formation of semiquinone by NADH dependent reductase or by a nonenzymatic reaction with iron. Further, DOX also decreases liver tissues ability to detoxify reactive oxygen species [42]. In the present study, generation of oxidative stress by one of these mechanisms should have contributed to hepatic dysfunction. However, the cotreatment of DOX and B. serrata extract resulted in a partial reversal of DOX-induced changes in these biomarkers ( < 0.001). Acute histopathological changes in DOX-treated livers were also reversed by B. serrata extract administration. B. serrata gum is a clinically proven anti-inflammatory agent. Boswellic acids from gum have been successfully used to prevent injury responses [32] and are effective agents in preventing oxidant-induced injury responses [43]. Boswellic acids can scavenge a broad spectrum of reactive oxygen and nitrogen species [44]. Recently the extract of B. serrata has been shown to possess active antioxidant substances which can exert protective effects against acute oxidative stress [45]. This scenario is suggestive of B. serrata extracts hepatoprotective nature through reduction of DOX-mediated oxidative stress. To further confirm this mechanism, Malondialdehyde (MDA), a product of lipid peroxidation, was measured in hepatic tissues of the treatment groups to evaluate the DOXinduced damage caused by oxidative stress [46]. MDA levels were significantly increased ( < 0.001) from control level showing DOX-induced lipid peroxidation. Results showed that treatment with B. serrata extract at 250 and 500 mg/kg significantly ( < 0.001) prevented DOX-induced lipid peroxidation and oxidative damage thus confirming our results.

Conclusion
In conclusion, our data provides convincing evidence that combination treatment is effective against HepG2 and Hep3B cells by induction of apoptosis. While the currently available chemotherapeutic options are associated with unavoidable adverse effects, oral B. serrata extract provides hope as a useful anticancer agent with significantly lower toxicity on normal liver tissue. Thus the use of synergistically acting B. serrata extracts and DOX combination therapy could be a novel strategy for the treatment for HCC and probably will have lesser toxicity compared to currently used regimens.