Introduction
Ovarian cancer remains the most lethal gynecological cancer among women [
1,
2]. Projection estimates by GLOBOCAN 2020 indicated that by 2040, the number of women worldwide diagnosed with ovarian cancer will increase about 37% to 428,966 [
3]. Further, the number of deaths from the disease is projected to surge over 50% to 313,617 from 2020. Current standard treatment for the most common ovarian cancer (i.e. epithelial ovarian cancer) includes surgery followed by platinum-based chemotherapy and radiation therapy [
4]. The five-year survival rate of ovarian cancer is around 47%, mainly due to high risk of relapse and resistance to chemotherapy. Moreover, early-stage detection of the disease is difficult due to lack of promising screening tools, and most patients are typically diagnosed at advanced stage of the cancer. New therapeutic methods have emerged from various biomarker-driven initiatives such as poly ADP-ribose polymerase inhibitors and antiangiogenic therapy [
5,
6]. Cancer immunotherapy, which is the modulation of the body’s innate immune system to treat cancer, has gained widespread interest as any immune-related adverse effects are relatively better tolerated than traditional chemotherapeutic agents [
7]. Current immunotherapies for ovarian cancer fall into five broad categories: monoclonal antibodies, checkpoint inhibitors and immune modulators, therapeutic vaccines, adoptive T cell transfer and oncolytic viruses [
4]. In particular, natural killer (NK)-cell based immunotherapy holds great promise for cancer treatment because NK cells can be easily isolated and expanded ex vivo for adoptive cell transfer therapy [
8‐
10]. NK cells recognize a broad panel of several dozen ligands which can each induce a cytolytic response [
10]. The advantage of NK cell-based therapy over T cells is that there is virtually no complication from graft-versus-host disease [
8,
9]. There is no good treatment for late stage ovarian cancer after relapse from the treatment of bevacizumab and olaparib [
5,
6]. NK cell therapy of relapsed cancer could potentially provide an alternative option [
8‐
10].
Majority of the chemotherapeutic agents exert their cytotoxic effects by apoptosis which is typically deemed to be non-inflammatory and non-immunogenic [
11]. However, it is now clear that certain agents such as anthracyclines and oxaliplatin, in addition to having cytotoxic properties, can also elicit immunogenic cell death [
12]. Immunogenic cell death is mediated largely by damage-associated molecular patterns (DAMPs), most of which are recognized by pattern recognition receptors on immune cells. Some DAMPs are actively induced by cells undergoing immunogenic cell death, such as calreticulin, and adenosine triphosphate (ATP), whereas others are induced passively, such as high-mobility group box 1 (HMBG1). Some like members of the tumor necrosis factor (TNF)-family like FAS ligand (FASL), TNF and TNF-related apoptosis inducing ligand (TRAIL) can induce tumor-cell apoptosis upon the formation of immune synapses. These DAMPs play a beneficial role in anti-cancer therapy by interacting with the immune system [
12,
13]. Chronic inflammation is typically associated with ovarian cancers, with high levels reactive oxygen species, cytokines, growth factors and inflammatory mediators [
14]. An important member of cytokines is the interleukin-1 (IL-1) superfamily which has critical functions in proper maintenance of the innate and adaptive immune system [
15]. Various genomic studies have shown that single nucleotide polymorphisms in the IL-1 superfamily can lead to higher susceptibility for immunological pathologies and disease presentation [
15].
Medicinal plants have been traditionally used to treat numerous human health conditions and offer a vast resource as drug leads or novel therapeutic agents [
16]. Despite the extensive biodiversity of medicinal plants around the world including Southeast Asia, there is scant documentation on the usage of fresh medicinal plants. Rapid urbanization poses a real threat to their natural habitat. Further, there is inadequate research on their pharmacological activities and scientific basis for their medicinal use.
Leea indica (Burm. f.) Merrill, which belongs to the genus
Leea and family Vitaceae, can be found in tropical and subtropical forests of Southeast Asia, China, India, and north Australia [
17,
18]. In Singapore, the plant is distributed in the coastal areas, mangroves, secondary forests and the undergrowth of primary forests [
19].
L. indica is also known as Bandicoot berry in English, or Yan Tuo 岩陀 in Chinese, Memali in Malay [
17,
18]. The leaves, roots and fruits of
L. indica have been traditionally used to treat a wide variety of ailments including cardiovascular diseases, cancer, diabetes, diarrhea, dysentery, eczema, fever, headache, and pain [
18,
20,
21]. In vitro studies showed that the leaves of
L. indica have various biological activities, including antihyperglycemic [
22], antimicrobial [
23], antioxidant [
23], anticancer [
24], anxiolytic [
25], thrombolytic [
26] and phosphodiesterase inhibitory effects [
27]. Essential oils from the flowers may have antimicrobial activity [
28], and the entire plant may have antioxidant and nitric oxide inhibitory activities [
29]. In view that medicinal plants are good sources of novel therapeutics while treatment options for refractory ovarian cancer are limited and NK cell therapy looks promising, we wish to explore the effects of NK cell killing of ovarian cancer cells triggered by a phytoconstituent identified in a medicinal plant. We have previously shown that the maceration methanolic leaf extract of
L. indica had good anti-proliferative activity against various human cancer cells, including ovarian cancer cells [
30]. However, the effect of
L. indica or its phytoconstituent on ovarian cancer cells and with chemotherapeutic drug oxaliplatin or NK cells are not known. Hence the objective of this study is to investigate the effects of
L. indica leaves and its selected phytoconstituents on human ovarian cancer cells and in combination with oxaliplatin and NK cells.
Discussion
L. indica is traditionally used to treat intestinal cancer and uterus cancer [
43] and the leaf extracts of
L. indica showed anticancer activity against Ehrlich Ascites Carcinoma (EAC) cells in Swiss albino mice, cervical epidermoid (Ca Ski) and most other human cancer cell lines [
24,
25,
30]. Mollic acid arabinoside and mollic acid xyloside identified from the ethanol leaf extract were reported to be responsible for the cytotoxic effects against human cervical Ca Ski cancer cells [
44], possibly via the stimulation of mitochondria-mediated apoptosis [
45]. However, it is unclear if there are also other phytoconstituent(s) responsible and other mechanisms involved. Herein we report for the first time that leaf extracts of
L. indica increased the susceptibility of human ovarian cancer cells to NK cell-mediated cytotoxicity (Fig.
1). The leaf extracts also suppressed the levels of TNF-α and IL-1β cytokines in human U937 macrophages (Fig.
6). We demonstrate for the first time that methyl gallate isolated and identified in
L. indica can enhance the sensitivity of human ovarian cancer cells to NK cell-mediated cytotoxicity (Figs.
2 and
4), and this increased cytolysis was likely due to the associated elevated expression of various stress ligands for NK cell receptors (i.e. DNAM-1 and NKG2D) on these cancer cells (Fig.
3B, Supplementary Fig.
2B). Cancer cells pretreated with methyl gallate and subsequently cultured in the presence of NK cells were unable to initiate their growth potential (Fig.
5). Combined methyl gallate with low concentration of chemotherapeutic drug oxaliplatin enhanced the levels of stress ligand expression on tumor cells (Fig.
3C, Supplementary Fig.
2C), and this was accompanied by higher susceptibility to NK cell-mediated cytolysis (Fig.
4).
It is well documented that although NK cells are innate immune cells that play crucial roles in immunosurveillance and eliminating tumors, NK cell function is often impaired during tumor development and progression due to the presence of multiple immunosuppressive factors in the tumor microenvironment [
46,
47]. Tumour variants can evade NK cell attack by mechanisms such as defective expression of activating ligands. Tumours may also upregulate ligands for inhibitory receptors and/or lose ligands for activating receptors, causing cells to be resistant against NK cell-mediated killing. For instance, high levels of circulating soluble MIC-A/B were associated with poor prognosis in a number of cancer types including colorectal, ovarian, liver, lung and prostate cancers [
48,
49]. Compounds that enhance NK cell-mediated lysis of cancer cells are therefore highly beneficial and valuable, and they include bortezomib [
50], doxorubicin [
51] and oxaliplatin [
52]. For example, the chemotherapeutic drug doxorubicin or the proteasome inhibitor bortezomib can trigger the upregulation of activating ligands for NKG2D receptor and DNAM-1 on multiple myeloma cells, thereby sensitizing them to NK cell-mediated lysis [
53]. Bortezobmib inhibited proliferation of liver cancer cells and increased MIC-A/B expression (
50). Natural products that are able to induce immunogenic cell death could represent novel lead compounds for cancer therapy. Some examples of reported natural products that induce immunogenic cell death include digoxin from
Digitalis species and capsaicin from
Capsicum species, as well as those derived from marine organisms such as
Spirulina maxima [
54], resveratrol [
55], daphnetin [
56], and stemphol [
57]. Resveratrol, a naturally occurring plant polyphenol, sensitized human leukemia KG-1a cells to NK cell killing through NKG2D ligands and TRAIL receptors [
58]. Lee et al.[
59] showed in mouse models that resveratrol upregulated NKG2D, NKp30 and CD107a expression, and effectively inhibited tumor growth and metastasis. Stemphol, a natural dialkyl resorcinol extracted from
Stemphylium globuliferum, induced caspase-independent cell death and released high-mobility group box 1 (HMGB1) in leukemia cells [
57]. Daphnetin, a dihydroxylated derivative of coumarin, is a potent stimulator of NK cells in that daphnetin enhances IFN-γ production and direct cytotoxicity in the presence of IL-12 [
56]. Daphnetin also suppresses inflammatory cytokine production in experimental autoimmune encephalomyelitis mice [
60]. In our study, we showed in ovarian cancer cells that methyl gallate significantly enhanced the expression of stress ligands for DNAM-1 and NKG2D NK cell receptors, i.e. CD112, CD155, MIC-A/B, ULBP-1/2/3, TRAIL-1 (DR4) and TRAIL-2 (DR5) (Fig.
3B, Supplementary Fig. 2B). Pre-treatment of ovarian cancer cells with methyl gallate rendered these cancer cells more susceptible to NK cell killing, compared to ovarian cancer cells that have not been previously exposed to methyl gallate (Figs.
2 and
4). However, gallic acid showed no significant effect on the expression of stress ligands in these cancer cells at the concentration tested (Supplementary Fig.
3), and therefore the relatively low level of NK cell-mediated cytolysis (Fig.
2). In contrast, Dedoussis et al.[
61] demonstrated in human leukemia K562 cell line that pre-treatment with 200 µg/ml gallic acid rendered the cells significantly susceptible to NK cell-mediated necrosis. It is likely that the difference in findings could be due to differences in cell type and concentration of gallic acid used. Nevertheless, it is possible that the stress ligands investigated in this study may not account for the whole picture of sensitizing the ovarian cancer cells to NK cell-mediated killing. There may be other ligands and factors not studied here that could potentially contribute to the methyl gallate-associated NK cell lysis or combined methyl gallate and oxaliplatin-associated NK cell lysis, such as B7-H6, calrecticulin, HMGB1, cytokines, and chemokines [
46,
47]. It is also unclear whether methyl gallate treatment of ovarian cancer cells inhibited specific signaling pathway, or dampened DNA methyltransferase or histone acetylases. Given the importance of dysregulation of epigenetic signaling pathways and cancer [
62], future studies exploring these possibilities are warranted.
As far as we are aware, our group is the first to report the identification and isolation of methyl gallate from leaves of
L. indica (reference [
32] and this study). Also known as methyl-3,4,5-trihydroxybenzoic acid, methyl gallate is a polyphenolic compound reported in plants such as maple leaf [
63], root bark of
Paenonia suffruticosa [
64],
Schinus terebinthifolius [
65],
Rosa rugosa [
66], and
Galla rhois [
67]. Methyl gallate has been reported to possess various biological properties including anti-oxidant [
64,
68] and anti-microbial properties [
69]. In human hepatocellular carcinoma, methyl gallate is reported to suppress cell proliferation via increasing the production of reactive oxygen species and apoptosis [
70]. Methyl gallate is also shown to have anti-inflammatory activities in zymosan-induced experimental arthritis animal model, wherein methyl gallate impaired zymosan-stimulated macrophages by inhibiting IL-6 and nitric oxide production, cylooxegenase-2 (COX-2) and inducible nitric oxide synthase (iNOS) expression (
71). Administration of methyl gallate in lipopolysaccharide-treated mice protected the mice against acute renal injury, increased anti-oxidant activity and decreased NF-kB activity [
64]. In mouse RAW 264.7 cells, methyl gallate blocked inflammation induced by Toll-like receptor ligands through attenuating NF-kB signaling and mitogen-activated protein kinase (MAPK) pathway [
72]. Methyl gallate also inhibited lipopolysaccharide-induced nitric oxide and IL-6 production in mouse-derived RAW 264.7 cells, most likely via the down regulation of extracellular-signal-regulated kinase 1/2 (ERK1/2) pathway [
73].
Combination of cisplatin-paclitaxel, which is a widely adopted “standard” treatment for advanced ovarian cancer, is frequently interrupted by the emergence of drug resistance cancer cells [
74,
75]. Oxaliplatin but not cisplatin was shown to trigger immunogenic cell death of colorectal cancer cells, activated dendritic cells by expressing danger signals such as heat shock proteins, calreticulin, HMGB1, and efficiently generated a pool of tumor antigen-specific T cells [
39]. Oxaliplatin is a third-generation platinum compound that is less studied and rarely used but promising in the treatment of ovarian cancer [
76]. Oxaliplatin has been the backbone of treatment of colorectal cancer. Its cytotoxic effect is mediated mainly through DNA damage. Like other chemotherapeutic agents when used at high doses, oxaliplatin is reported to have clinically adverse side effects such as peripheral neuropathy and nausea [
40‐
42]. We therefore studied oxaliplatin at low concentration and chose two ovarian cancer cell lines OVCAR-5 and SK-OV-3 with different spectrum of drug resistance. OVCAR-5 cells are known to show resistance to clinically relevant concentrations of adriamycin, melphalan and cisplatin, while SK-OV-3 cells are resistant to tumour necrosis factor and several cytotoxic drugs including diphtheria toxin, cis-platinum and adriamycin. Combination of low level oxaliplatin and methyl gallate in the presence of NK cells was capable of effecting cancer cell lysis despite the tumor resistance (Fig.
4). An E:T ratio of 1:4 was chosen to demonstrate that even at this low concentration of NK cells, the proliferation of these ovarian cancer cells which are resistant to drugs, can still be suppressed (Fig.
5). More importantly, our data suggests that a single treatment regime alone (either NK cell co-culture alone, or methyl gallate treatment alone) was insufficient to completely abrogate cancer cell growth. Likewise, oxaliplatin treatment alone was insufficient [
36]. The re-proliferation of minimal residual cancer cells in in vitro cultures was subsequently detectable over time. These “residual” cancer cells can potentially form the next resistant colony in the long term and eventually render resistance to the existing treatment regime. Our work suggests combination of methyl gallate and oxaliplatin, which can trigger antitumor immunogenicity, in conjunction with activated NK cells, warrants further investigation.
The devastating effect of immune dysregulation is well recognized in cancers. Cytokine storm and cytokine release syndrome are life-threatening systemic inflammatory syndromes which involve high levels of circulating cytokines and immune-cell hyperactivation [
77]. Multiple studies have demonstrated that ovarian cancer has immunosuppressive tumor micro-environment that poses serious challenge to existing treatment modalities. For instance, myeloid-derived suppressor cells were increased by vascular endothelial growth factor (VEGF) expression in human ovarian cancer, resulting in suppressed immunity [
78]. The number of intraepithelial CD8
+ tumor infiltrating lymphocytes and a high ratio of CD8
+/Treg are associated with a positive prognosis in epithelial ovarian cancer [
79]. Our study showed that methyl gallate inhibited TNF-α and IL-1β production in human U937 macrophages (Fig.
6). Our results are consistent with those findings observed using mouse macrophages [
72,
73] in that methyl gallate exert anti-inflammatory effects. In our study, we observed that at the same concentration examined, methyl gallate significantly suppressed TNF-α production (
p < 0.01), whereas gallic acid showed no appreciable effect on TNF-α production (Fig.
6B). Further, at the same concentration studied, methyl gallate significantly suppressed IL-1β production (
p < 0.05), while gallic acid had no significant effect on IL-1β production (Fig.
6D). Gallic acid, also known as 3,4,5-trihydroxybenzoic acid, is a natural secondary metabolite and widely present in various plants [
80]. Gallic acid is reported to suppress TNF-α and IL-1β levels in gouty arthritis mice model by inhibiting NLR family pyrin domain containing 3 (NLRP3) inflammasome activation [
81]. Interestingly, studies on the levels of TNF-α using either RNA in-situ hybridization of tissue arrays or semi-quantitative reverse polymerase chain reaction of mRNA in ovarian cancer have shown that TNF-α expression was present at higher levels in ovarian carcinoma compared to normal tissues [
82]. Future studies on understanding the pharmacological mechanism of methyl gallate and its effects on refractory ovarian cancer cells are warranted.
In conclusion, the leaf extracts of L. indica and its selected phytoconstituents were successfully investigated for their effects on human ovarian cancer cells and in combination with oxaliplatin and NK cells. The crude leaf extract of L. indica was found to enhance the susceptibility of ovarian cancer cells to NK cell cytolysis. Its phytoconstituent methyl gallate was found to upregulate the expression of stress ligands for NK cell receptors and elevate the sensitivity of drug-resistant human ovarian cancer cells to NK cell cytolysis. Our findings suggest that the combined effect of methyl gallate, oxaliplatin and NK cells in ovarian cancer cells warrants further investigation. Our work is a step towards better scientific understanding of the traditional anticancer use of L. indica.