Xanthorrhizol: a review of its pharmacological activities and anticancer properties

XNT is considered active against a variety of pathogenic microorganisms. Antimicrobial effects of XNT included antibacterial [15, 16, 22], anticandidal [19, 23] and antifungal activities [24, 25]. There have been evaluated by in vitro susceptibility tests such as minimum inhibitory concentration (MIC), minimum bactericidal concentration (MBC), minimum fungicidal concentration (MFC), NCCLS (M38-A) standard method and biofilm quantification.

Earlier study by Hwang and colleagues reported that XNT showed the highest antibacterial activity against dental caries causing bacteria (Streptococcus species) followed by periodontitis causing bacteria (Actinomyces viscosus and Porphyromona gingialis) [16]. XNT also strongly inhibited Gram-positive bacteria Staphylococcus aureus, methicillin-resistant Staphylococcus aureus (MRSA), Gram-negative bacteria Escherichia coli [34] and acne-causing bacteria Propionibacterium acnes [35].

Moreover, the ability of XNT in preventing dental plaque and removing oral bacterial biofilms has been demonstrated on the oral Streptococcus mutans biofilms in vitro [22]. Biofilms removal activities were affected by XNT concentration, exposure time and the biofilm phase growth. For example, XNT (5 µM) completely inhibited the formation of S. mutans biofilms at adherent growth phase, whilst XNT (50 µM) removed 76 % of biofilm at plateau accumulated phase after 60 min exposure. XNT killed S. mutans at planktonic growth due to its direct contact with biofilm outer layer cells [15, 22, 36]. The antimicrobial activities were induced by the capability of the hydrophobic chains of XNT to penetrate and reduce the viability of dental plaque biofilm [37].

For anticandidal activity, XNT inhibited planktonic cells of Candida albicans at MICs range of 1–15 µg/mL [19]. This finding is controversial with the previous result [16], where Candida albicans were found to be resistant to XNT. Since there is lack of information on the XNT’s condition used in the previous work [16], we infer that XNT dissolved in dimethyl sulfoxide (DMSO) [19] may enhance anticandidal activity towards C. albicans. The ability of XNT to prevent and kill C. albicans was further supported by Rukayadi and Hwang, where XNT at 8 µg/mL completely reduced C. albicans biofilms at adherent phase, whilst 32 µg/mL reduced 88 and 67.5 % of biofilm at intermediate and mature phase, respectively [23]. It was also active against pathogenic non-Candida albicans species such as C. glabrata, C. guilliermondii and C. parapsilosis biofilms in vitro [19, 38]. These results indicated that XNT might be used to cure biofilm-related candidal infections and treat candidiasis.

On the other hand, XNT performed antifungal activity against planktonic fungal cells such as Malassezia species [24] and opportunistic filamentous fungi [25]. Anti-Malassezia activity of XNT was reported in M. furfur and M. pachydermatis [24]. XNT also inhibited the conidial germination of all six filamentous fungi species such as Aspergillus flavus, Aspergillus fumigatus, Aspergillus niger, Fusarium oxysporum, Rhizopus oryzae and Trichophyton mentagrophtes based on NCCLS (M38-A) standard method. Its effect was comparable to amphotericin B [25].

Although antimicrobial mechanisms of XNT are not well understood, we believe that XNT may suppress nuclear factor kappaB (NF-kB) and mitogen-activated protein kinase (MAPK) induced by microbial infection. XNT has been demonstrated to inactivate both of them in skin cancer [26]. According to Wilken et al., infectious antigens could induce the activation of NF-kB [39]. For example, exposure of epithelial cells to C. albicans hyphae stimulates pro-inflammatory immune responses via NF-kB and MAPK pathways [40], which are also involved in the carcinogenesis [41, 42].

Based on epidemiologic studies, it has been estimated about 15 % of the worldwide cancer incidence is considerably related with microbial infection [43]. Chronic infection of human papilloma virus in immunocompetent hosts causes cervical carcinoma, whilst hepatitis B and C virus infection leads to hepatocellular carcinoma. Mirobes may also induce cancer incidence through opportunistic infection such as human herpes virus (HHV)-8 infection leading to Kaposi’s sarcoma [44‐46]. In addition, gastric cancer secondary to Helicobacter pylori colonization or colon cancer may occur in certain people due to abnormal immune responses to microbes contributed by chronic inflammatory bowel disease precipitated by the intestinal microflora [44‐46]. Since XNT has anticancer and antimicrobial properties, we suggest that its antimicrobial mechanism studies should be conducted not only to develop XNT as a potent antimicrobial agent, but also provides new insight on the suppression of microbes-induced cancer in the future.

First in vitro anti-inflammatory report of XNT has been shown in lipopolysaccharide-activated mouse leukaemic monocyte macrophage cell RAW 264.7 [27]. XNT reduced cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS) activity by inhibiting the production of prostaglandin E2 (PGE2) and nitric oxide (NO) respectively in lipopolysaccharide-activated mouse macrophage cell RAW 264.7. These results indicated XNT may be a potent COX-2 and iNOS inhibitors [27], which is suggested by another anti-inflammatory assay of XNT performed in activated primary cultured microglial cells induced by lipopolysaccharide [17]. It was found to inhibit COX-2, iNOS, proinflammatory cytokine interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) in activated microglial cells. It is clear that XNT is capable to inhibit COX-2 and iNOS as consistent with several findings [26, 27, 30], whilst IL-6 and TNF-α as consistent with recent report [6].

Further in vivo anti-inflammatory studies of XNT have been conducted in 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced mouse acute inflammation model [26]. XNT has been reported to counteract the effect of TPA-induced ornithine decarboxylase (ODC), COX-2 and iNOS activation in mouse skin. Since pro-inflammatory proteins COX-2 and iNOS are highly associated with cutaneous inflammation, cell proliferation and skin tumor promotion [26, 47], their suppression are important to alleviate inflammation and prevent cancer [26, 41, 42, 48]. The expression of COX-2 and iNOS might be regulated by transcription factor, NF-kB, as reported in cultured cell lines and TPA-induced cutaneous inflammation in mouse skin [17, 26]. When nuclear translocation and DNA binding of NF-kB increase in response to external stimuli, NF-kB stimulates COX-2 and iNOS transcription [26, 48]. Thus, NF-kB plays a pivotal role in inflammation and tumorigenesis.

Another study postulated that XNT may exert anti-inflammatory activity by blocking the neurogenic and inflammatory pain response in the formalin induced pain test in rats [10]. It may partly contribute to the analgesic effects or antinociceptive activity. However, the detailed mechanisms have not been worked out. From the integration of findings [10, 17, 26, 27], we summarize that anti-inflammatory mechanism of XNT involved inhibition of IL-6 and TNF-α, and suppression of COX-2 and iNOS expression via NF-kB pathway resulting PGE2 and NO reduction.

Antioxidant properties of XNT contribute to its neuroprotective [17] and LDL oxidation inhibitory effects. XNT has been known to possess in vitro antioxidant activity against murine hippocampal neuronal HT22 cell line [17] and copper-mediated isolated human low-density lipoprotein (LDL) oxidation [5]. In murine hippocampal neuronal HT22 cell line, XNT reduced the free radical-mediated oxidative damage [17]. Its antioxidant properties exerted potent neuroprotective effects by suppressing hydrogen peroxide (H₂O₂)-induced lipid peroxidation in rat brain homogenates, glutamate-induced neurotoxicity and reactive oxygen species (ROS) production in HT22 cells. These results indicated that XNT could be a potent agent to treat Alzheimer’s disease and ROS associated neurological disease [17].

On the other hand, the inhibition of copper-catalysed LDL oxidation was evaluated employing thiobarbituric acid reactive substances (TBARSs) assay with human LDL as the oxidation substrate [5]. XNT strongly inhibited human LDL peroxidation in a dose-dependent manner. The presence of phenolic hydroxyl group (sesquiterpene phenol) on the bisabolene skeleton of XNT, has most probably contributed to its strong antioxidant properties by chelating Cu²⁺. This in turns may suppress the initiation of LDL oxidation and generation of free radicals at the lipoprotein [5]. We suggest that XNT might be subjected to further investigation in cardiovascular disorders because high LDL antioxidant activity could reduce the risk of heart attack. In vivo antioxidant assay could also be conducted in the future.

In vivo antihyperglycemic effects of XNT have been demonstrated in the high-fat diet (HFD)-induced obese mice [6]. XNT and C. xanthorrhiza extract with standardized XNT reduced the levels of insulin, glucose, free fatty acid (FFA), and triglyceride (TG) in their serum. XNT also reduced the size of epididymal fat pad and adipocyte and decreased the production of inflammatory cytokines such as TNF-α, IL-6, interleukin-1ß (IL-1ß), and C-reactive protein (CRP) in adipose tissue, liver and muscle of HFD-induced obese mice. Thus, XNT may prevent fatty liver disease (accumulation of liver fat) and chronic inflammation [6].

These results showed that XNT’s antihyperglycemic and anti-inflammatory activities may restrict and treat type 2 diabetes, which is mainly caused by obesity-induced insulin resistance [6]. Insulin resistance is related to chronic low-grade inflammation states such as increased proinflmmatory cytokine levels. The inflammation process is initiated by the activation of TNF-α, IL-6, IL-1ß and CRP, which are known to disrupt the transduction of insulin signalling causing insulin resistance [6]. Based on this study, we reveal that XNT could suppress HFD-induced metabolic disorders including hyperglycemia, inflammation and hepatic injury by inhibiting fatty acid release from adipose tissue. We suggest that anti-obesity effects of XNT and its related mechanisms of action could be studied in the future.

XNT extracted from Iostephane heterophylla has shown potential antihypertensive activities [28]. A preliminary study demonstrated that XNT effectively inhibited precontractions induced by calcium chloride, potassium chloride and noradrenaline in rat thoracic aorta rings. The vasorelaxation effect of XNT indicated that it may act as a calcium antagonist by reducing calcium influx into vascular smooth muscle cells in rat aorta. In fact, its calcium antagonistic activity has been illustrated earlier in isolated rat uterine smooth muscle [21]. XNT attenuated the effect of rat uterus’ tonic contraction stimulated by calcium chloride, potassium and calcium channel agonist in a dose-dependent manner. This might be due to the ability of XNT to block the voltage operated calcium influx in myometrial cells. According to Grossman and Messerli, calcium antagonists reduce blood pressure via vasodilation and decreased peripheral resistance [49]. Since calcium antagonists have been well established as basic antihypertensive drugs [50], we believe that XNT may have blood pressure-lowering effect. However, detailed antihypertensive activities and mechanisms of XNT are yet to be elucidated.

In vitro antiplatelet activity of XNT (100 µg/mL) showed a strong inhibition towards platelet aggregation stimulated by arachidonic acid (100 %), collagen (81.3 %) and adenosine diphosphate (ADP) (78.6 %) in human whole blood [29]. Although previous studies reported that the antiplatelet activity of curcumin was higher than XNT [29], the potential of XNT as an antiplatelet compound should not be neglected. We suggest that its antiplatelet mechanism requires further investigation.

Nephroprotective and hepatoprotective effects of XNT have been performed in male ICR mice treated with cisplatin [30‐32]. Cisplatin is a potent chemotherapeutic drug [31, 51], but the occurrence of nephrotoxicity has become the main limitation of using cisplatin-based chemotherapy [32, 52]. XNT exhibited nephroprotective effect by attenuating the increased specific gravity of kidney induced by cisplatin [32]. Cisplatin-induced kidney injury was reported as increased kidney weight, enhanced lipid peroxidation in kidney tissues, weakened filtration and excretion process of kidney, and subsequently increased blood urea nitrogen and serum creatinine levels. Pretreatment of XNT obviously restored the kidney weight to the base level and attenuated the elevated levels of blood urea nitrogen and serum creatinine. Although DNA-binding activity of NF-kB and activator protein 1 (AP-1) did not contribute to the nephroprotective effect [32], the exact mechanism has not yet been identified.

High dose of cisplatin also induces hepatotoxicity [30, 31]. Cisplatin increased DNA-binding activity of NF-kB but suppressed DNA-binding activity of AP-1. The function of NF-kB is to stimulate COX-2 and iNOS, which are associated with inflammation and toxicity. XNT pretreatment has been shown to abrogate these effects. XNT elicited hepatoprotective effects by reducing blood glutamate-pyruvate transaminase (GPT) and glutamate–oxaloacetate transaminase (GOT) levels caused by cisplatin [30]. The mechanism involved XNT’s dose-dependent attenuation of c-Jun N-terminal kinases (JNKs) phosphorylation in MAPK signaling, especially JNK1 [31]. This action may inhibit the transcription of COX-2, iNOS and transcription factor subunits (c-fos and p50). When XNT suppressed cisplatin-induced c-Fos protein expression, it may modulate the DNA-binding activity of NF-kB and AP-1, which in turns regulate COX-2 and iNOS expression. Mitochondrial apoptosis was excluded since the expression of both cytochrome c and caspase-9 was not changed [31]. Thus, it has been concluded that XNT minimized side effects of cisplatin-induced hepatotoxicity by regulating the DNA-binding activities of transcription factors NF-kB and AP-1 [30] via blocking the phosphorylation of JNK(s) [31].

It was believed that XNT exerted better suppressing effect towards cisplatin-induced nephrotoxicity [32] and hepatotoxicity than curcumin [30, 31]. At the same dose, curcumin was less effective in attenuating the elevated levels of blood urea nitrogen and serum creatinine [32]. XNT downregulated COX-2 and iNOS gene expression, but curcumin suppressed only COX-2 gene [30]. Moreover, XNT abrogated the expression of NF-kB subunit, p50 and AP-1 subunit, c-fos, but not curcumin [31]. Combined with the findings of both nephroprotective and hepatoprotective effects, we assume that XNT could be clinically applied as a suppressant of toxicity for patients administrated with high dose cisplatin to prevent kidney and liver damage.

XNT has been known to possess estrogenic activity in estrogen receptor (ER)-positive MCF-7 cells during the state of hormone starvation [20, 33]. It has been reported that XNT treatment upregulated ER target gene expression, trefoil factor 1 (pS2) and promoted the interaction of ER-estrogen response elements (EREs) in MCF-7 cells. Since XNT has been proven to possess estrogenic activity in negligible estrogen level [20], we suggest that XNT could be further explored in the treatment of estrogen deficiency-induced menopausal symptoms, cardiovascular disease and osteoporosis.

In contrast, XNT was revealed as partial estrogen antagonist in T47D breast cancer cells [12]. In molecular docking simulation, the binding interaction between XNT and human estrogen receptor-α (hERα) indicated that XNT might be able to compete with estradiol. Both XNT and estradiol showed almost similar binding free-energy. Also, a strong hydrophobic interaction found between XNT and hERα may be due to the presence of hydroxyl group (1-OH) and alkyl chains, leading to its potential as partial antagonist hERα. The postulation was confirmed by pharmacophore modeling, which identified that 1-OH and alkyl chain were two important chemical features of XNT as partial antagonist hERα to strongly inhibit T47D cells. This molecular interaction with hERα also involved aromatic ring of XNT [12].

Based on the estrogenic [20, 33] and anti-estrogenic activities [12] reported, we suggest that XNT may act as a potent phytoestrogen with beneficial therapeutic potential. According to Tham et al., the partial estrogenic/anti-estrogenic behaviour is a common characteristic of phytoestrogens [53]. The estrogenic activity of phytoestrogens is 100 to 1000-fold weaker than 17β-estradiol, but its concentrations may be 100-fold higher than endogenous estrogens in the body [53]. Hence, we believe that abundant XNT molecules might act as competitive inhibitors of endogenous 17β-estradiol. XNT may block the actions of estradiol from binding to ERs of breast cancer cells, thus inhibiting tumor growth. Seeing that tumorigenesis of ER-positive luminal A cell lines (MCF-7 and T47D) can be suppressed by anti-estrogen therapy [54], XNT could be developed as a potential anti-estrogen agent.

To further study the effects of XNT as phytoestrogens in vitro, estrogen should not be excluded in experimental condition because circulating estradiol exists at all stages of the life cycle [53]. XNT may exert both estrogenic and anti-estrogenic effects on human metabolism, depending on XNT and endogenous estrogens concentration, gender and menopausal status.

According to Yamazaki et al., a single oral administration of 500 mg/kg XNT showed no mortality in mice [56]. Since 1 mg of C. xanthorrhiza ethanolic extract contained 0.1238 mg of XNT [10], we estimate that up to 619 mg/kg XNT in 5 g/kg of this extract was safe to be administrated in mice. However, efficacy and safety dosage of XNT in targeted therapeutic areas are yet to be conducted in the future. Moreover, there are still lacking of studies on genotoxicity, carcinogenicity and reproductive toxicity of XNT.

To date, XNT data is unavailable in clinical pharmacology and clinical efficacy studies [57]. Not every new candidate compound discovered is fully developed and marketed [58]. We reveal that some issues restraining the availability of clinical trials are funding limitations and difficulty getting approvals from Food and Drug Administration (FDA). Frohlich supported that one of the complexities surrounding clinical research is insufficient funding to conduct modern research [59]. Aside from funding, strict ethical and regulatory compliance enforced by our social structure [59] might prolong the new drug developmental process. The major concern of clinical research is the safety and efficacy of the candidate compound [58]. It normally takes 5–6 years for a candidate drug to be submitted in a new drug application (NDA) to the FDA. Then, it required 6–10 months to complete the review of all the safety and efficacy data [58]. To overcome the delay of widespread access to new therapies, questionable data and ethical issues must be resolved. We suggest that necessary safety data, efficacy data and overall risk/benefit analysis are important key elements in successful sponsorship of XNT for future clinical studies.

XNT’s potent antioxidant and free radical-scavenging properties may exert chemopreventive effects on carcinogenesis. XNT was first reported to inhibit hepatic cytochrome p450 enzyme system by Yamazaki et al. [56]. Cytochrome p450 plays an important role in the oxidation and detoxification of toxic compounds [39]. When it is exposed to toxin, oxidation occurs and subsequently produces carcinogenic metabolites forming DNA adducts, which could initiate carcinogenesis [39]. Therefore, we deduce that XNT’s potent inhibitory effect on cytochrome p450 activity might suppress the initial stages of carcinogenesis by reducing ROS production in liver cells. Menon and Sudheer also presumed that membrane lipids peroxidation mediated by free radical and oxidative damage of DNA and proteins might most likely link to cancer and neurodegenerative diseases [69]. It has been shown that XNT inhibited H₂O₂-induced lipid peroxidation in rat brain homogenates, glutamate-induced neurotoxicity, LDL peroxidation and ROS production [5, 17]. In addition, ODC-induced reactive oxygen intermediates (ROIs) may also involve in the tumor promotion [69]. It is supported by a previous study where superoxide dismutase (SOD) and catalase suppressed ODC induction in murine mammary tumor cells [70]. Since XNT was able to suppress ODC in mouse skin cancer [26], we suggest that it may exert antipromotional activity via radical scavenging effect.

Inflammation has long been correlated with the progression of cancer [44‐46, 69]. We believe that anti-inflammatory activities of XNT may inhibit tumor promotion via the modulation of transcription factor NF-kB. Increased NF-kB activation is common in many cancers because it involves the cellular pathways leading to inflammation, angiogenesis, tumorigenesis and metastasis [39]. As NF-kB is stimulated by stressful stimuli from cytokines (TNF-α and IL-1), multiple NF-kB regulated gene products (COX-2, iNOS, IL-6) might be upregulated [39].

As mentioned in Jantan et al., XNT was capable to reduce TNF-α, IL-6, IL-1β production [5]. We infer that it can switch off NF-kB activation by cytokines, thus giving anti-inflammatory effects. This is evident by several studies [17, 26, 27, 30] that showed downregulation and reduction of COX-2 and iNOS. Indeed, both enzymes are important to mediate inflammatory processes [69]. Improper upregulation can cause certain human cancers and inflammatory disorders [69].

In mouse skin with TPA-induced acute inflammation (mouse ear edema) and 19 weeks TPA-induced tumor promotion in 7,12-dimethylbenz[a]anthracene (DMBA)-initiated mouse skin, XNT suppressed cancer and inflammatory biomarkers including ODC, COX-2 and iNOS expression through MAPKs, NF-kB and or protein kinase B (Akt) [26]. XNT delayed or inhibited tumor formation, and reversed the carcinogenic process at premalignant stages [26]. Therefore, XNT’s anticancer activities may be associated with its anti-inflammatory property by inhibiting the NF-kB activity and subsequently the induction of COX-2 and iNOS.

In addition to apoptosis induction, growth inhibitory pathways of XNT have been demonstrated in colon cancer, esophageal cancer and tongue cancer cells. Considering cyclin D1 proto-oncogene is ultimately important to modulate the transition of G1 to S phase in various cell types [85], XNT reduced cyclin D1 expression in HCT116 colon cancer [65], TE-1 and TE-4 esophageal squamous cancer cell lines [68].

In HCT116 colon cancer cells, XNT was found to inhibit the cell cycle in the G1 or G2/M phase by triggering cyclin-dependent kinase inhibitors (CDKIs) such as p21 and p27 [65]. Cell cycle arrest in the G0/G1 and G2/M phase and increased sub-G1 peaks were closely associated with under-expression of cell cycle regulatory proteins such as cyclin A, B1, D1, cyclin-dependent kinase 1 (CDK1), CDK2, CDK4 and proliferating cell nuclear antigen (PCNA), which is a biomarker for the cell proliferation. Also, XNT stimulated G0/G1 and S + G2/M cell cycle arrest in human Tca8113 tongue cancer cell line [66]. To understand the molecular mechanisms by which XNT regulates cell cycle, extensive studies of its downstream signalling pathways will be necessary.

In TE-1 and TE-4 esophageal squamous cancer cell lines, double combination of XNT and astaxanthine or triple combination of XNT, astaxanthine and α-tocopherol exerted synergistic apoptotic effects [68]. They reduced not only the expression of p-Akt and cyclin D1, but also exhibited a higher caspase-3 expression than XNT’s treatment alone in both esophageal cancer cells. On the other hand, previous studies indicated that combined astaxanthine and α-tocopherol did not exhibit any cooperative apoptosis, where they increased the expression of p-Akt and maintained caspase-3 levels compared to control [91]. In fact, astaxanthine or α-tocopherol treatment alone was effective against TE-1 and TE-4 esophageal cancer cells, but not their combination [91]. Thus, we postulate that the addition of XNT to astaxanthine and α-tocopherol may stimulate their antiproliferative properties giving synergistic effects.

Cytoselective toxicity of a bioactive compound is hardly defined. A bioactive compound may be cytotoxic in certain cells while inactive towards others. XNT has been considered as cytoselective when it was tested on HeLa cervical cancer cells as compared to non-malignant Chang’s Liver and MDBK cells [61]. EC₅₀ value of XNT towards both cell lines was 4.7-fold and 2.8-fold higher than HeLa cells, respectively. Also, XNT was found to be more sensitive towards MCF-7 breast cancer cells than African green monkey kidney cells (COS-7) [20].

In another study [7], XNT was moderately cytoselective against Chang’s Liver cells but lowly cytoselective towards Vero kidney cells compared to HepG2 liver cancer cells. IC₅₀ value of XNT in both normal cell lines was 2.1-fold and 1.6-fold higher than HepG2 cells, respectively. Also, XNT exhibited less than twofold lower growth inhibition and cytotoxicity in both MDBK and Vero cells as compared to MDA-MB-231 breast cancer cells [63]. Surprisingly, non-cytoselective activity of XNT has been reported in normal fibroblast cell line CCD1114sk as compared to malignant melanoma HM3KO cells [62]. These results indicated that cytoselective and non-cytoselective effects of XNT depend on cell types and biological variation.