The role of key gut microbial metabolites in the development and treatment of cancer

ABSTRACT In recent years, the role of gut microbial metabolites on the inhibition and progression of cancer has gained significant interest in anticancer research. It has been established that the gut microbiome plays a pivotal role in the development, treatment and prognosis of different cancer types which is often mediated through the gut microbial metabolites. For instance, gut microbial metabolites including bacteriocins, short-chain fatty acids and phenylpropanoid-derived metabolites have displayed direct and indirect anticancer activities through different molecular mechanisms. Despite the reported anticancer activity, some gut microbial metabolites including secondary bile acids have exhibited pro-carcinogenic properties. This review draws a critical summary and assessment of the current studies demonstrating the carcinogenic and anticancer activity of gut microbial metabolites and emphasises the need to further investigate the interactions of these metabolites with the immune system as well as the tumour microenvironment in molecular mechanistic and clinical studies.


Introduction
Over the past decade, the gut microbiome has been extensively investigated in the context of the maintenance of human health. Gut microbiota has been observed to maintain a mutually beneficial relationship with the host through modulation of gut homeostasis and the preservation of the epithelial barrier which are crucial for gut immunity 1 . These microorganisms are important in the normal physiological function and structure of the host innate immune system, which has a number of implications on gut health. 2 Emerging evidence has indicated that in addition to gut microbiota and their structural components, the myriad of metabolites produced by gut microbial communities also influence the host physiology and health by acting as signalling molecules and substrates for metabolic reactions. 3 Our recent review has underlined the pivotal role of gut microbiota in the prevention, therapy and clinical outcome of the five most prevalent cancers while emphasising the direct and indirect impacts of gut microbial metabolites on tumours. 4 Several studies in the literature have also demonstrated the potential effect of gut microbial metabolites in different diseases including cancer. This review provides a critical summary and assessment of the current studies performed to understand the role of the key gut microbial metabolites in cancer development, prevention, and treatment. This review also proposes several potential future directions in this emerging area of cancer research.
Gut metabolites have been shown to exhibit a variety of pro-and anti-carcinogenic effects on cancer. Recent reviews demonstrated the dual role of gut metabolites in promoting as well as preventing colorectal cancer (CRC) 5,6 ( Figure 1). It has also been suggested that the cumulative effects of the microbial metabolites should be considered to predict and prevent the progression of CRC. 5 Furthermore, it has been postulated that the increase in CRC risk is due to an imbalance between health-promoting metabolites such as butyrate and potentially carcinogenic metabolites including secondary bile acids (BAs). 7 In addition to the pro-carcinogenic activity of secondary BAs, microbial metabolism may also highlight the role of dietary fat (which increases the synthesis of BAs in the liver) on colon cancer progression as observed in rural African and African American populations. 7 The influence of diet on colonic health was further observed in a study that assessed the impact of a high-protein and low-carbohydrate diet on the metabolic profile in the colon. 8 The author reported that the high-protein and reducedcarbohydrate diet had a detrimental effect on colon health by causing an observable decrease in faecal anticancer metabolites and an increase in concentrations of carcinogenic metabolites, constituting an increased risk of CRC in individuals who adhere to this diet long-term. 8 Further evidence supporting this trend has led to the understanding that protein fermentation in the distal colon has detrimental effects on host health by producing toxic ammonia, amines, phenols and sulfides, in comparison to carbohydrate fermentation which generates health benefitting short-chain fatty acids (SCFAs). 9 The review by Windey et al. in 2012 9 also indicated that a diet rich in meat increases fermentation of proteins in addition to increased intake of fat, heme and heterocyclic amines, which may collectively contribute to the higher prevalence of CRC in Figure 1. A schematic representation of (a) the signalling effect of gut microbial metabolites on the colon epithelium, (b) initiating an invasion of the bacterial species into the colorectal cancer tissue. 6 In addition, (c) gut microbial metabolites induce apoptosis in cancer cells and activate immune cells to inhibit pro-inflammatory cytokines. 5 Western society. However, more in-depth mechanistic studies are necessary to understand the relationship between protein fermentation and gut health in relation to CRC.
Prebiotic and probiotic-based strategies targeted at improving systemic health has led to an increased interest in the role of SCFAs including butyrate, acetate, and propionate which are the byproducts (secondary metabolites) of carbohydrate fermentation by the gut microbial communities. 10 Whilst SCFAs exhibit cancer-protective properties during dietary fibre fermentation, secondary BAs, at high physiological levels in the colon, elicit an opposite effect through induction of colonic inflammation. 10,11 A high abundance of secondary BAs correlates with a high-fat diet, in which exposure to BAs can generate reactive oxygen species and disrupt the cell membrane and mitochondria. 12 Other bacterial metabolites such as bacteriocins have been used safely in the food industry as a food preservative and are emerging as potential therapeutic agents against colon, head and neck, breast, brain, skin, and liver cancers. [13][14][15][16][17][18][19][20] Nisin is a well-researched bacteriocin that has demonstrated cytotoxic effects on CRC and head and neck squamous cell carcinoma both in vitro and in vivo mediated via induction of apoptosis. 13,16 Despite the reported potential pro-and anticancer activities, further animal and clinical studies are required to develop a better understanding of the role of gut metabolites in the progression as well as prevention of cancer and to develop precision anticancer therapies.
Phenylpropanoids are a diverse family of plant secondary metabolites synthesized from the amino acids-phenylalanine and tyrosine. Plantbased diets consisting of phenylpropanoids have been suggested to improve human health. However, specific microbial species have the capacity to ferment the three aromatic amino acids (AAAs)-phenylalanine, tyrosine and tryptophan to phenylacetic acid (PAA) and 4-hydroxylphenylacetic acid (4-hydroxyPAA) indicating that protein fermentation is the probable source of phenylpropanoid-derived by-products within the colon. 21 Although the study by Russell et al. in 2013 21 indicated that gut microbiota can ferment proteins to produce major phenylpropanoid-derived metabolites, the same research group earlier in 2011 suggested that a highprotein and low total carbohydrates and fibre diet can significantly decrease faecal cancerprotective metabolites and increase the concentrations of hazardous metabolites. 8

Bacteriocins
Bacteriocins are cationic peptides produced by certain probiotic bacteria in the gut through ribosomal activity and structurally classified as bacterial antimicrobial peptides. [22][23][24] Bacteriocins have demonstrated significant inhibition of other bacteria such as antibiotic-resistant strains with narrow to broadspectrum activity. 22 These antimicrobial peptides can also inhibit pathogenic bacteria in the gut 25 and therefore, are important in maintaining gut homeostasis. Lactic acid bacteria is one of the most significant sources of bacteriocins, especially the genus Enterococcus belonging to the phylum Firmicutes. 26,27 A clinical study profiled the prevalence of bacteriocin production by the pathogenic Escherichia coli strains in CRC patients. 28 The clinical trial included 30 patients with colorectal cancer, 30 patients with colorectal adenoma, and 20 healthy controls, and evaluated the bacteriocinscolicin Ia, colicin M, microcin mH47, microcin mV, and microcin mM. 28 This study observed that advanced stage CRC patients presented with more virulent strains of E. coli, and this correlated with increased production of bacteriocins in comparison to less advanced stage diagnoses. 28 Bacteriocins have exhibited significant cytotoxicity against cancer cells in vitro (Table 1) and low cytotoxicity towards normal intestinal epithelial cells. 18 In addition, the anticancer activity of bacteriocins is also attributed to their capacity to inhibit the colonisation of competing pathogenic bacterial strains in a phenomenon known as 'colonisation resistance' (Figure 3), as well as the immunomodulation of the gut microbial composition. 66 The cellular membrane is the primary target of bacteriocins in eukaryotic cells, in which bacteriocins increase the expression of negatively charged cell-surface molecules on cancer cells and encourage cytotoxicity. 67 The proposed mechanisms of action of this activity are the induction of apoptotic cell death and the depolarization of the cell membrane leading to changes in cell membrane permeability, indicating a non-receptor-modulated process. 67 In addition to selective cytotoxicity against cancer cells, the nonimmunogenic and biodegradable nature of bacteriocins make them a promising candidate for novel anticancer therapy. 67 Therefore, the possibilities to modulate the production of bacteriocins by the probiotic bacteria in the gut as well as their bioengineering have also been proposed for their clinical applications. 22 As bacteriocins have been shown to have a direct impact on gut microbiota in addition to modulating the host immune system, these metabolites may play a key role in the processes of inhibiting carcinogenesis in the intestine as well in determining the efficacy of anticancer treatments and the clinical outcome of cancer. However, there are very limited in vivo studies on the anticancer effects of bacteriocins with most studies performed in vitro. Additionally, several limitations exist related to the survival and function of bacteriocins in vivo as they are mostly dependent on different factors including a) the survival of bacterial strain in the gut, b) specificity of the bacteriocins, and, c) the type of animal model used. 66 This was further validated in a study that suggested that bacteriocins may not be synthesised or available in high quantities within the GIT, however, there is uncertainty regarding the efficacy of bacteriocin production in existing studies. 68 Given the inconsistencies with bacteriocin production and availability, the analysis of microbial ecology should be an integral stage of the novel drug discovery process to increase the efficacy of bacteriocin treatment. 68 The probiotic effects of bacteriocins exemplify the possibility of incorporating these metabolites as novel alternatives to existing antibiotic treatment, as well as pharmabiotics. 69 The specificity of bacteriocin species ensures that they can target certain pathogens, which is a potential replacement for traditional antibiotics, especially for drug-resistant pathogenic strains. 69 To investigate this, an in vitro study assessed the antibacterial activity of bacteriocins produced by lactic acid bacteria against various strains of the Helicobacter pylori species, 70 which is responsible for a significant number of gastrointestinal cancers. The authors identified that the bacteriocins lacticin A164 and BH5, produced by Lactococcus lactis, exhibited the most substantial antimicrobial activity against the H. pylori strains, of which the ATCC 43504 strain was the most tolerant of the bacteriocins. 70 These findings were further supported by another study, which acknowledged that the inhibitory activities of bacteriocins support the use of probiotics in control of H. pylori infection and related diseases. 71 Whilst the Induced the apoptotic pathway, which was also supported by observed morphological changes in cancer cells. 30 In vitro MTT and annexin V-FITC /propidium iodide staining.
Induced apoptotic processes and cell death, down-regulated cell viability dosedependently. 20 In vitro MTT and flow cytometry.
Induced cell cycle arrest and apoptotic cell death. This correlated with an increase in reactive oxygen species (ROS) and mitochondrial membrane potential modulatory action. 37 In vitro Induced apoptotic cell death and led to significant changes in cell morphology after treatment. 38 In vitro MTT and flow cytometry.
Higher concentrations increased the level of apoptosis. Cell cycle arrest in the G 1 growth phase was also observed. 40 In vitro MTT and flow cytometry.
Inhibited cell growth and cell proliferation in a dose-dependent manner and caused a blockage in stage G 1 of the cell cycle. Induced apoptotic cell death dose-dependently. 40

In vitro
Lung cancer H1299 and H1703 non-small cell lung carcinoma cells.
Inhibited cell proliferation, however, no IC 50 values were reported.
Western blot and PCR.
At 10 mM, sodium propionate suppressed cell growth and proliferation. Induced cell cycle arrest in the G 2 /M phase leading to apoptotic cell death. Upregulated the expressions of p21 and survivin leading to suppression of cell proliferation. 43 (Continued) In vitro Induced apoptotic cell death and inhibited cell migration of the HN4 and HN6 cells.

In vivo
Demonstrated biocompatibility without adverse effects, with increased apoptotic cell death and decrease in cancer cell survival. Decreased nuclear factor (NF)-κB activation leading to suppression of mRNA and protein expression of matrix metalloproteinase-9 thereby inhibiting tumour cell metastasis. Downregulated anti-apoptotic Bcl-2/Bcl-XL expression and upregulated apoptotic Bax expression. 45 In

In vitro
The flow cytometry method observed early and late-stage apoptosis, as well as inhibition of cell proliferation. HIPK2 regulated the phosphorylation of p53, as well as the concentration of Bcl-2 and Bax in these cancer cells.
In vivo Enhanced the expression of pro-apoptotic HIPK2, p53, and Bax proteins in tumours, but decreased expression of anti-apoptotic protein Bcl-2, in a dose-dependent manner. Human tissue In human colorectal cancer tissues, the expression of HIPK2 was significantly lower compared to normal tissues. The expression of HIPK2 in human colorectal cancer significantly correlated with the degree of differentiation. 46 In vitro Inhibited cell proliferation in a concentration-and time-dependent manner. Induced cell cycle arrest at the S phase (Caco-2 and HT-29 cells) with no effect on the cell cycle of SW480 cells. Induced oxidative phosphorylation and the TCA cycle, inhibited epithelial-mesenchymal transition, Vascular Endothelial Growth Factor A expression and boosted antitumor immunity. The molecular mechanism of action of LCA was found to be TGR5 receptor-mediated. Bacterial LCA production was reduced in early-stage breast cancer patients. 62

MTT, Western blot and PCR.
Induced ER stress and TGF-β expression. 59

e2038865-12
structural and functional capacities, as well as immunomodulatory activities of bacteriocin have been well-researched, further studies are required to develop a more comprehensive understanding of the factors that modulate bacteriocin production in the intestinal system. Collectively, based on the available reports, bacteriocins might be promising for the development of novel therapies, especially to target microorganisms that are responsible for carcinogenesis including H. pylori. Nisin is one of the most explored bacteriocins (also known as lantibiotic) with anticancer potential as evident in several studies on cancer cells. Thus far, four variants of nisin-nisin A, Z, Q and U have been discovered from L. lactis and Streptococcus uberis. 72 Nisin is a polycyclic peptide produced by the process of bacterial fermentation known for its antibacterial activity against a broad range of Gram-positive bacteria including Staphylococcus aureus, and Listeria monocytogenes 73 and has been studied recently as a potential anticancer peptide against colorectal cancer cells. 13,16,74 The normal physiological function of nisin is to compete with other Grampositive bacteria for colonisation of cell surfaces and other cellular resources, which contributes to its prevalent use as a safe food preservative in dairy products. 75 Due to its safety profile, nisin has been approved in more than 50 countries for use as a food preservative and is generally regarded as safe for humans by the World Health Organisation. 67 Recent studies have explored its use in the inhibition of tumorigenesis in head and neck squamous carcinoma cells both in vivo and in vitro. 16 Nisin has been shown to increase the apoptotic index in cancer cell lines via the intrinsic apoptotic pathway. 13 This was further supported by another study that observed that increasing concentrations of a specific nisin variant correlated with increasing levels of apoptotic cancer cell death and a decrease in cell proliferation of head and neck cancer cells. 16 It was earlier proposed that nisin exerted these cytotoxic effects on cancer cells through CHAC1 (ChaC Glutathione Specific Gamma-Glutamylcyclotransferase 1), which is a proapoptotic cation transport regulator and is considered an apoptotic mediator in relation to tumorigenesis. 15 Nisin-induced CHAC1 expression increased calcium influxes and induced cell cycle arrest in the G 2 phase which led to apoptosis and a decrease in tumour cell proliferation ( Figure 4). 15 However, the authors also acknowledged that the optimal therapeutic dose must be determined for the potential use of nisin in cancer therapy which is plausible based on the history of safe human consumption of nisin. 15 Furthermore, an in vivo study on the association of nisin ZP against head and neck squamous cancer cells (HNSCC) demonstrated that nisin ZP reduced tumorigenesis in mice models, and long-term treatment with nisin ZP extended the survival of the mice with normal organ histology 16 (Table 1). This was further supported by another in vivo study, in which nisin decreased tumorigenesis of HNSCC through the induction of apoptosis via upregulating the CHAC1 gene expression. 15 The same study showed that nisin at 80 μg/mL inhibited the proliferation of the UM-SCC-17B HNSCC cells via cell cycle arrest in the G 2 phase after 24 h. 15 Recent studies have also cloned and expressed the fusion protein derived from the three bacteriocin-nisin, enterocin, and epidermicin in E. coli to explore the possibility of utilising the fusion protein for gastric cancer. 76 Additionally, nisin has been shown to inhibit the proliferation of blood, breast, brain, colon, gastrointestinal, liver and skin cancer cells pre-clinically mostly through induction of apoptosis as shown in Table 1. Despite its promising anticancer activity and safety profile as depicted in these studies, nisin has not been evaluated alone or in combination with standard anticancer therapy clinically. Due to its selective toxicity towards cancer cells as compared to normal cells, 67 further clinical studies should be performed to explore the therapeutic potential of nisin. Current literature has demonstrated the potential use of bacteriocins in conjunction with standard chemotherapeutic drugs as an alternative approach to cancer treatment. A recent study identified that the application of nisin and 5-FU as a combination therapy exhibited synergistic activity against 7,12-dimethylbenz(a)anthracene-induced skin cancer in vivo and lowered their IC 50 values by an eight-fold against the A431 epidermoid carcinoma cells with a combination index value of 0.188. 77 The anticancer activity of the combination was mediated by modulation of apoptotic, angiogenic and cell proliferative pathways with significant reduction of tumour size and number (mean tumour volume and mean tumour burden) compared to the mono treatments in that study. 77 Furthermore, within the food and pharmaceutical industries, bacteriocins have been observed to be capable of replacing antibiotics which could assist in combatting multi-drug resistant pathogens. 24 This approach might also be useful in eliminating carcinogenic pathogens from the gut. However, further research is necessary to understand the efficacy of bacteriocins in anticancer therapy both as mono and combination regimens with standard chemotherapy.

Short-chain fatty acids (SCFAs)
SCFAs have been well-researched in recent years for their inhibitory effects on, especially colon and breast cancer cells (Table 1). It has been established that dietary nutritional factors play a substantial role in the progression of CRC, and the absence of certain factors can disturb metabolic and homeostatic pathways within the intestinal system that in turn promote tumourigenesis. [78][79][80] SCFAs are derived from the breakdown of dietary fibre, in which fermentation of the non-digestible carbohydrates occurs in the lower gastrointestinal system, the cecum and large intestine, by anaerobic cecal and colonic microorganisms. 81 This fermentation process results in a group of metabolites with SCFAs as the primary metabolites. 81 The predominant bacterial species responsible for producing SCFAs are the Faecalibacterium prausnitzii, Clostridium leptum, Eubacterium rectale, and Roseburia species, as well as lactate-utilising species that synthesise SCFAs from lactate and acetate, including Anaerostipes species and Eubacterium hallii. 82 A profiling study on CRC patients observed that the dominant group of Bifidobacterium species disappeared, and different spectrums of Bifidobacterium was present in the CRC patients (n = 14) compared to the non-CRC participants (n = 14). This observation correlated with significantly lower SCFA levels in the CRC patients compared to the non-CRC group. 83 The same research group earlier demonstrated that CRC patients (n = 14, Indonesian citizens, 18 years of age or older) were presented with lower levels of acetate, propionate, and butyrate, than the non-CRC participants (n = 14, Indonesian citizens, 18 years of age or older), indicating the indirect contribution of SCFAs in the prevention of CRC development. 84 Several key SCFAs have exhibited protective action against colon carcinogenesis, including butyrate, acetate, and propionate, which are synthesised via anaerobic bacterial and carbohydrate fermentation 21,79,85,86 ( Figure 5). In particular, the major health benefits of fibre consumption are attributed to the production of SCFAs through the fermentation processes occurring in the colon. 79 At the molecular level, butyrate has been observed to inhibit cell proliferation and induce apoptosis and cell differentiation through the initiation of histone hyperacetylation in cancer cells. 5,78,86,87 Furthermore, SCFAs can recognise G proteincoupled receptors-GPR41, GPR43 and GPR109A on the surface of colonocytes and immune cells including macrophages and T cells which in addition to histone hyperacetylation leads to an enhancement in total colonic regulatory T cell numbers and the levels of the anti-inflammatory cytokines interleukin-10 (IL-10) and transforming growth factor-β (TGFβ) (Figure 1). 5 A study investigated the effects of three SCFAsbutyrate, acetate, and propionate on the growth of the HT29 human colorectal adenocarcinoma cells and showed that butyrate and propionate were more effective in inhibiting the growth of HT29 cells, in comparison to acetate which had no observable effect. 85 Similar observations were made in another study 86 where butyrate, propionate and valerate (a type of SCFA) inhibited the human colon carcinoma cells with no activity reported for acetate and caproate (a type of SCFA). Butyrate in this study also significantly increased apoptosis in the cancer cells. 86 The inhibitory effect of butyrate and propionate on the proliferation of cancer cells was associated with the activation of ornithine decarboxylase, which is an important enzyme of polyamine metabolism, however, it was also noted that these SCFAs did not act solely on the polyamine pathway. 85 Data from that study revealed that butyrate and propionate caused an increase in alkaline phosphatase activity indicating that they may play an important role in the normal physiology of the colon and could also be a contributing factor in the protective influence of dietary fibres on colon carcinogenesis. 85 The tumour suppressive functions of SCFAs, most particularly butyrate, are believed to be caused by the histone hyperacetylation-mediated pathway which results in the conversion of inactive procaspase-3 to catalytically active protease (apoptotic) (Figure 6). 32,86,87 This antitumour effect of SCFAs has also been supported by epidemiological studies that highlight a fibrerich dietary lifestyle correlates with the reduction of CRC risks. 32,88 Other health benefits of butyrate dietary supplementation include the prevention of insulin resistance and obesity induced by a high-fat diet, which is achieved through a decrease in adiposity and an increase in insulin sensitivity in peripheral tissues. 79 This was further supported in a mice model study that observed that the administration of butyrate as a dietary supplement prevented and treated diet-induced insulin resistance. 89 The authors also acknowledged that the mechanism of action of butyrate was directly associated with the induction of mitochondrial Figure 6. A diagrammatic depiction of the biological activities of butyrate on cancer cells via G protein-coupled receptor 43 (GPR43) and modulation of immune cells. 32,86,87 activity and the promotion of energy expenditure, which was a significant development in understanding the anticancer mechanisms of butyrate. 89 As aforementioned, the introduction of butyrate into cells leads to histone hyperacetylation via the inhibition of histone deacetylase activity. 5,78,86,87 Therefore, butyrate may serve as a key factor in determining the role of histone acetylation in the structure and function of chromatin, demonstrating a strong association between butyrate and histone deacetylase inhibitors in preventing and managing cancer cell proliferation. 90 The mechanism of action of this inhibitory activity has also been studied and showed that there is a close association between the SCFA-mediated activation of the GPR41/GPR43 receptor signalling pathways and the inhibition of histone deacetylases. 91 GPR43 is recognised by SCFAs, and expression of this receptor is found predominantly in the large intestine and haematopoietic tissues and is frequently lost in colon cancer cell lines. 32 It has been identified that restoration of GPR43 receptor expression in the HCT8 human colonic adenocarcinoma cells led to an increased apoptotic cancer cell death following G 0 /G 1 cell cycle arrest. 32 In particular, the treatment of the HCT8 cancer cells with butyrate and propionate led to an increase in the GPR43 receptor expression and apoptotic cell death ( Figure 6). 32 Therefore, it has been speculated that the GPR43 receptor serves as a functional tumour suppressor to mediate the apoptotic effects of SCFAs in CRC. 32,85,92,93 This was further supported by another study that observed agonist activity for both butyrate and propionate on the GPR43 receptor, in which the highest concentrations of GPR43 were identified in immune cells under certain pathophysiological conditions. 92 Another report also suggested that SCFA-mediated growth arrest in colon carcinoma cells requires the p21 gene as SCFAs were ineffective against the p21-deleted HCT-116 colon cancer cells. In normal cells, p21 functions as a cell cycle inhibitor and antiproliferative effector, whereas in some cancers it is dysregulated. 94 The role of p21 in cancer has been established in several reports as a tumoursuppressor protein under the p53 transcription factor activity. 94 The complex interplay among gut microbiota, the immune system and dietary factors have been researched in recent years. 4 Whilst more investigations are required to adequately determine the association between SCFAs and the immunometabolism of T cells, including specific metabolic targets, it has been shown that diets rich in SCFAs have displayed suppressive action on T cellmediated autoimmune responses, which may be achieved via the regulation of cytokine expression and T cell function by these secondary metabolites. 91 Despite the promising antitumoural action of butyrate identified in existing studies, research has also observed its pro-tumoural effects in the development of CRC. A study assessing the impacts of pathogenic bacteria on CRC development acknowledged that the pro-or antitumoural effects of butyrate are dependent on multiple factors, including duration and amount of exposure to the treatment, and the studied cell type. 95 That study coined the term 'butyrate paradox', where the effects of butyrate are determined by its concentration, with low levels promoting tumorigenesis and high levels inhibiting tumour. 95 Another study validated the 'butyrate paradox' by utilising a mice model and observed that, despite the evident anti-tumoural activity of butyrate, low concentrations of butyrate promoted the development of CRC by increasing the proliferation of colonic epithelial cells. 5 Many studies have observed similar carcinogenic effects of butyrate in animals and humans, however, further investigation is required to understand the interactions between the host's genetics, microbial composition, and presence of other gut metabolites to constitute this paradoxical effect. 96 The low levels of butyrate have also been shown to initiate a proinflammatory environment within the host that disrupt the gut microbial composition by suppressing potential pathogens and encouraging colonisation by butyrate-producing species. 5 The complexity of butyrate activity dependent upon concentration is a vital consideration in its use as a potential anticancer therapeutic. In relation to standard chemotherapeutic drugs, one study identified that butyrate significantly improved the efficacy of 5-fluorouracil (5-FU) against colon cancer cells and increased impairment of DNA synthesis caused by 5-FU. 97 Parallel observations were made in a previous study that reported that forming a conjugate targeted delivery system with a standard chemotherapeutic drug such as doxorubicin, and a SCFA could improve the efficacy of the standard drug, limit the occurrence of drug resistance, and more efficiently target the tumour microenvironment. 98 Through the inhibition of histone deacetylase (HDAC), butyrate might be beneficial in improving the clinical efficacy and reducing the toxicity of standard chemotherapy. 99 Overall, SCFAs are promising specifically in the context of colon cancer. Future studies should evaluate the effects of SCFAs on other cancer types including pancreatic and gastric cancer to understand their molecular mechanisms of action. Studies should also explore the impact of SCFAs on the efficacy and safety of standard chemotherapy and the prognosis of cancer.

Phenylpropanoid-derived metabolites
Phenylpropanoid-derived metabolites, such as phenolic acids, are a significant component of plant secondary metabolism and have been demonstrated to inhibit the growth of different cancer cell types in several studies through a number of molecular pathways. [100][101][102][103] As a derivative of plant secondary metabolism and a constituent of diets rich in plant foods, phenylpropanoids have exhibited chemopreventive, antioxidant, anti-inflammatory, and antimitotic activities in the host. [104][105][106] The biosynthesis of phenylpropanoids has been extensively researched in the past decade to develop a greater scientific understanding of the upstream and downstream enzymes responsible for the development of these secondary metabolites. 104 It has been speculated that specialised phenylpropanoid products can be developed from the recognised mechanistic foundations of phenylpropanoid metabolising enzymes, which can include diverse novel compounds with both dietary and medicinal properties in human health. 104 One study on human faecal samples found that phenylpropanoidderived compounds, including phenylacetic acid and 4-hydroxylphenylacetic (the two most abundant metabolites detected) were synthesised from both plant-rich diet and the microbial fermentation of AAAs in the colon. 21 In particular, phenylpropanoids-derived compounds such as phenylacetic acid (PAA) and 4-hydroxylphenylacetic acid (4-hydroxyPAA) are produced through microbial fermentation of AAAs-phenylalanine, tyrosine and tryptophan in the colon by Bacteroidetes (Bacteroides thetaiotaomicron, Bacteroides eggerthii, Bacteroides ovatus, Bacteroides fragilis, Parabacteroides distasonis), and Firmicutes (Eubacterium hallii and Clostridium bartlettii). 21 A profiling study of colorectal cancer patients detected increased levels of Bacteroides fragilis in patients with advanced diagnoses (stage III and IV CRC), and these levels were greater in the colon than in the rectum. 107 This study involved patient participation from two cohorts (average age 59 years with an equal division of gender); the first cohort consisted of 55 paired CRC patient samples with no pre-selected condition in addition to CRC, and the second cohort consisted of 18 patients that also had been diagnosed with sporadic microsatellite instability. 107 This increased presence of bacteria responsible for producing phenylpropanoidderived metabolites was indicative of the potential causative association between advanced progression of the tumour and increase in anti-tumoural bacteria and metabolite production. 107 An in vitro study on probiotic metabolites of Lactobacillus rhamnosus observed that 4-hydroxyPAA activated mitochondrial-regulated apoptosis and induced cell proliferation against the HepG2 liver cancer cell line. 108 Similarly, an in vitro study examined the activity of a novel Zn(II) complex combining phenylacetic acid and the 4,4'-bipyridine ligand, in which the complex was identified to inhibit the HeLa cervical cancer cell line through induction of apoptosis. 109 Phenylpropanoids are well-known for their microbial diversity and biosynthetic origins, however, further studies are required to better understand the mechanisms of action and antimicrobial activities of these compounds, 110 especially against carcinogenic bacteria and viruses. Dietary verbascoside, a phenylpropanoid compound, has been shown to influence gut morphology due to its antimicrobial and antioxidant properties. 111 The authors demonstrated that verbascoside protected the gastrointestinal tract from oxidative stress with potential appetitestimulatory effect via modulation of the protein expression of the gastrointestinal taste receptors. 111 Verbascoside has also been found to arrest the MKN45 gastric epithelial cancer cells at the sub-G 1 and G 2 /M phases of the cell cycle ( Figure 7). 47 A study showed that human and rat gut microbiota can break down acteoside, a type of verbascoside into 14 metabolites including 8 degradation metabolites, 2 isomers in intestinal bacteria and intestinal enzyme samples and 4 parent metabolites. 112 Acteoside also exhibited a significant anti-inflammatory effect by inhibiting LPS-induced PGE 2 , nitric oxide and TNF-α in mouse peritoneal macrophages in a concentration-dependent manner. 113 Whilst further research is required to understand the anti-inflammatory effects of these compounds, it has been suggested that polyphenols target multiple inflammatory components and modulate immune processes via the synthesis of proinflammatory cytokines, immune cell regulation, and gene expression. 114 These biologically active compounds promote extended health benefits for several chronic inflammatory diseases, including cancers, by supporting the immune system and preventing the onset of chronic disease as observed in preclinical experimental models and clinical studies. 114 In recent years, research has prioritised focusing on the use of natural substances that are most cost-effective and present with fewer adverse effects. 115 Existing studies have also assessed the combined implementation of phenylpropanoids and standard chemotherapeutic agents in the treatment of cancer. An in vitro study aimed to assess the combined therapeutic potential of eight distinct phenylpropanoids in conjunction with 5-FU against the HeLa cervical cancer cells and identified that eugenol, ferulic acids, and caffeic acids demonstrated synergy when combined with 5-FU. 116 Additionally, phenylpropanoids exhibited minimal haemolytic activity on human erythrocytes supporting the use of these compounds as pharmaceutical drugs without causing toxicity within the host. 116 Despite their broad spectrum of biological properties against cancer as well as in the modulation of gut microbiota, further studies are required to understand the metabolic and catabolic pathways of phenylpropanoid by the gut microbiota and the mechanisms by which these compounds modulate inflammatory and microbial processes. 117

Prenylflavonoids
Xanthohumol (XN) is a prenylated flavonoid found in hops and it has shown promising anticancer activity in recent years. 52 The gut microbiota can metabolites XN to produce 8-prenylnaringenin (8-PN), a very potent phytoestrogen. 52 Primarily, 8-PN has demonstrated strong Figure 7. A schematic representation of the molecular mechanisms of action of verbascoside, a gut microbial metabolite, on the HN4 and HN6 human oral squamous cell carcinoma and the MKN45 gastric epithelial cancer cells. 45,47 anticancer action across multiple cancer cell lines (Table 1) and has been acknowledged for other health benefits. 49,118 The bioactivity of 8-PN might be attributed to its greater oral bioavailability in healthy individuals than its isomer 6-PN, despite a clinical trial demonstrating that both compounds had a similar effect on increasing cell viability of peripheral blood mononuclear cells. 118 A study assessed the anticancer activity of 8-PN in silico and in vitro against the SK-MEL-28 and BLM human metastatic melanoma cells and observed that it mediated anticancer action via the inhibition of HDAC. 49 An earlier study also showed that 8-PN inhibited the proliferation of the MCF7 human breast cancer cells through induction of apoptosis. 51 The proposed mechanism of action of this activity was the increased proliferation of estrogen-responsive cells by the 8-PN metabolite, through the interference with the estrogen receptor-associated PI3K pathway. 51 The strong estrogenic action of 8-PN was also observed in other studies where it showed greater activity than that of other established phytoestrogens, including genistein, daidzein, and coumestrol. 119 Another study compared the anticancer and apoptotic potential of 8-PN with other side-chain variants of prenylflavanones and found that 8-PN could target multi-drug resistant leukaemia cells and induced mitochondria-dependent apoptosis. 120 The in vitro action of 8-PN on various stages of colorectal tumourigenesis was also investigated. 48 This metabolite inhibited cell proliferation of the HT115 cells in a dose-dependent manner, with a growth reduction of up to 46%, in comparison to the untreated control. 48 The authors demonstrated that 8-PN exerted the anticancer activity at various key stages of colorectal tumourigenesis, which could be beneficial to improve the poor prognoses of CRC. 48 A more recent study specifically assessed the anticancer activities of 8-PN against HCT-116 colon cancer cells and determined that it conferred anti-proliferative activity via the induction of extrinsic and intrinsic pathway-mediated apoptosis. 50 These in vitro findings of the anticancer activity of 8-PN warrant further in vivo and clinical studies to examine its mechanisms of action and potential use as a natural anticancer agent.

Natural purine nucleosides
Natural purine nucleosides have demonstrated anticancer potential via different molecular mechanisms. The efficacy of nucleoside-based anticancer drugs is determined by the cellular transporters that modulate the movement of drugs into and out of the cell. 121 Analyses of different parasitic species have identified two prominent purine nucleoside transporters-an adenosine transporter and an inosine transporter. 122 Inosine is a metabolite synthesised by the catabolism of the adenosine compound, which exhibited diverse anti-inflammatory and immunomodulatory effects in vivo by acting directly on adenosine receptors. 123 The proposed modulatory mechanism of action of these effects is through the adenosine A 2A receptor (A 2A R), in which inosine-regulated activation of A 2A R initiates cAMP production and extracellular signal-modulated phosphorylation of kinase-1 and −2. 123 The findings of an in vivo study identified that inosine initiates ERK1/ 2-biased signalling as an agonist, in which it can amplify and extend A 2A R activation, and this has significant pharmacological implications. 123 Another in vivo study on the inosine metabolite acknowledged that it enhanced T cell antitumour activity in colorectal, bladder, and melanoma cancer types while amplifying the effects of checkpoint blockade immunomodulation. 58 The authors also confirmed that the bacteria Akkermansia muciniphila, associated with responsiveness to immune checkpoint blockade (ICB) therapy in humans, used inosine-A 2A R signalling for its ICBpromoting effect. 58 The proposed mechanism of action of inosine on bladder cancer was the enhancement of the function of anti-CTLA-4 to increase infiltration of IFN-ɣ + CD4 + and IFN-ɣ + CD8 + T-cells into the tumour, as well as reducing overall tumour weight when administered with CpG (DNA oligonucleotides containing unmethylated deoxycytidylyl-deoxyguanosine dinucleotides 124 ) as a combination therapy. 58 The study also reported that inosine produced by Bifidobacterium pseudolongum increased the activation of a cDC-dependent T H 1 cell circuit, which enhanced the overall effect of ICB therapies in mouse models of intestinal and epithelial tumours. 58 Whilst inosine has presented with promising anticancer action in combination e2038865-20 with other immunotherapies, further investigation is required into its stand-alone anticancer activity and mechanisms of action within the host.

Carcinogenic activity
Secondary BAs are metabolised from the dehydroxylation of primary bile acids by anaerobic bacteria in the large intestine, in which primary BAs are originally synthesised from cholesterol in the liver hepatocytes prior to being released into the intestinal system. 125,126 The bacterial species primarily involved in the production of secondary bile acids are members of the Clostridium genus including C. scindens, C. hiranonis, C. hylemonae, and C. sordellii. 127 Unlike SCFAs, BAs have been shown to exhibit procarcinogenic activity. Studies have demonstrated that the exposure of the gastrointestinal (GIT) tract cells to high levels of secondary BAs are a major contributing risk factor towards the development of GIT cancers and a high level of BAs is most commonly seen in individuals with a high dietary fat intake. 12,125,128,129 High-fat diets lead to an increase in the levels of secondary BAs in the enterohepatic circulation, including deoxycholic acid (DCA) and lithocholic acid (LCA), both of which can be risk factors for the induction of inflammation and cancer in the colon. 11 This accumulation of secondary BAs, particularly DCA, is due to the incapability of the human liver in returning 7α-hydroxylating secondary BAs via the portal vein, causing the high-level accumulation in humans consuming a 'Western diet'. 127 It has been well-established that DCA is capable of initiating cellsignalling pathways involved in the onset of various diseases. 127 To support this, a clinical study profiling metabolite concentrations in CRC patients observed an increased DCA level in the faeces, blood serum, and bile of participants. 130 Similarly, it has also been reported that mice fed with a high-fat diet presented with higher levels of the C. sordellii compared to other microbial species suggesting its role in increasing the DCA levels. 127,131 Excessive exposure to BAs can further lead to the generation of ROS with subsequent disruption of the mitochondrial and cell membrane as well as DNA damage. 12 The oxidative and DNA damage-related stress caused by prolonged exposure of cells to BAs initiates genomic instability within the cells, leading to the development of apoptotic resistance and the eventual onset of cancer. 12,129 It has been reported that nuclear receptors are directly associated with the modulation of BA metabolism and detoxification as they operate as transcription factors in the protection from the tumour promoting action of secondary BAs. 132 This is a significant factor of consideration in circumstances of human genetic mutational predispositions, in which secondary BAs can accelerate the intestinal adenoma-adenocarcinoma sequence through the initiation of the Wnt/β-catenin signalling pathway. 133 The genetic and environmental factors associated with the pro-carcinogenic activity of secondary BAs provide a foundation for further investigation into the role these metabolites play in the origin and prognosis of cancer.
As a common secondary BA present in individuals with a fat-rich diet, DCA functions as a significant environmental trigger in the onset of CRC. 134 Whilst the exact mechanism of action of DCA on intestinal tumorigenesis requires further investigation, it has been observed that it disrupts the intestinal mucosal barrier and increases proinflammatory cytokine production in the intestine, which is a key precursor to the development of intestinal cancer. 134 Gut microbial communities are inherently responsible for the modulation of intestinal homeostasis, in which dysbiosis to the microbiome is directly associated with intestinal tumorigenesis that is induced by high concentrations of DCA in the intestine. 133 In an in vivo study, the DCA-treated mice presented with alterations to the intestinal microbiome composition, which was coupled with an impaired intestinal barrier, inflammatory processes, and tumorigenesis. 133 The findings of that study provided evidence that the introduction of DCA into the intestine led to disturbances in the gut microbial composition and promoted carcinogenesis in the intestine. These findings, in correspondence with other available studies, emphasised that the interactions between secondary BAs and gut microbiota were responsible for the initiation of intestinal carcinogenesis, which is important for developing novel therapeutic strategies especially for GIT cancers. 133,135 The hyperproliferation of the colon mucosa is a preliminary stage in the progression of colorectal carcinogenesis and in animal models, DCA was observed to stimulate the proliferation of colorectal epithelial cells with a tumour promoting activity. 136 Another study assessed the effects of DCA on the migration of the Caco-2 human colon cancer cells and found that the migration of the cancer cells was likely associated with protein kinase C. 137 The tumour promoter activity of DCA was further validated in another study showing that DCA could activate protein kinase C and phospholipase C via increased Ca 2+ entry at the plasma membrane of the BHK-21 fibroblast cells. 138 A biopsy study on 19 patients with and without colon cancer in 1999 demonstrated that a significant increase in colorectal proliferation was correlated with the serum DCA levels but not with the serum levels of other BAs such as lithocholic, cholic, chenodeoxycholic, and ursodeoxycholic acid. 136 More human interventional studies are prudent to examine whether a decrease in DCA levels lowers the risk of carcinogenesis. It has been demonstrated that a higher risk of CRC in the American population was likely in part due to their high-fat and high-protein diet, which leads to the promotion of microbial species that can produce potentially carcinogenic secondary BAs. 80 Similarly, a more recent investigation identified that there are multiple mechanisms of action and a diverse range of signals involved in the promotion of CRC development by BAs and their derivatives, which highlights the potential of targeting primary and secondary BAs in the prevention of CRC. 139 However, to date, there are limited studies exploring these mechanisms and signalling pathways, which emphasises the need for future preclinical and clinical research on BAs to assist in the prevention and treatment of cancers.

Anticancer activity
Despite the predominantly carcinogenic nature of secondary BAs, lithocholic acid (LCA), a derivative of cholic acid, demonstrated anti-proliferative action on different cancer cell lines (Table 1). [61][62][63][64]140 In vitro studies on the human breast adenocarcinoma (MCF-7 and MDA-MB-231), human prostate cancer (PC-3, LNCaP and DU-145), hepatic cancer (HepG2) and neuroblastoma cancer (BE(2)-m17, SK-n-SH, SKn-MCIXC and Lan-1) cells have found that LCA inhibited the cancer cell growth by different molecular mechanisms of action (Table 1). [59][60][61]63,64 Notably, LCA inhibited the growth of the human prostate cancer (LNCaP and PC-3) cells through caspase-3, 8 and 9 mediated apoptosis. 60 LCA was also able to induce endoplasmic reticulum (ER) stress and transforming growth factor-β (a potent profibrogenic factor that induces apoptosis of hepatocytes and liver fibrosis) in the HepG2 liver cancer cells. 59 The cytotoxicity and ER stress in the HepG2 liver cancer cells were largely dependent on the hydrophobicity of the secondary BAs, with chenodeoxycholic (a hydrophobic secondary BA) exhibiting the greatest activity among the tested secondary BAs. 59 Another study 62 demonstrated that LCA is anti-proliferative against breast cancer cells in vitro (against the MCF7, 4T1 and SKBR3 cells without affecting primary fibroblast cells) and in vivo (4T1 xenograft female BALB/c mice) and induced oxidative phosphorylation and the TCA cycle, inhibited epithelial-mesenchymal transition, vascular endothelial growth factor A expression and boosted antitumor immunity. The molecular mechanism of action of LCA was found to be TGR5 receptor-mediated. 62 The authors also reported that in early-stage breast cancer patients, bacterial LCA production was reduced. 62 Despite these new findings on the anticancer activity of some secondary BAs, its in-depth mechanisms of mitochondrial dysfunction and cytotoxicity in cancer cells are yet to be confirmed and requires further investigation.

Conclusion and future directions
The vital role of the gut microbiota in the maintenance of gut homeostasis and immune health has been welldocumented in the literature. This acknowledgement led to the investigation into the influence of gut microbial metabolites on human health, most specifically in relation to cancer. The complex cross-talk between specific gut microbial metabolites and the progression or inhibition of cancer cell growth is an emerging area of anticancer research. Studies conducted in vitro and in vivo have been crucial in understanding the antiand pro-cancer activity of these gut metabolites in the host. For instance, gut metabolites including SCFAs and bacteriocins have been increasingly reported to have cytotoxic activity on various cancer cell types. SCFAs are commonly known for their antioxidant and anti-inflammatory action on host health, which help in inhibiting cancer cell proliferation. Furthermore, SCFAs have been shown to be a key player in the inhibition of obesity-induced insulin resistance, which is an important consideration in the development of colorectal cancer. Similarly, bacteriocins displayed anti-tumour potential via direct (by induction of the apoptotic pathway in cancer cells) and indirect (by conferring colonisation resistance on epithelial surfaces to inhibit pathogenic microbes) actions. In addition, phenylpropanoid-derived metabolites are observed in high concentrations in individuals that consume a plant-rich diet and have been extensively researched over the past decade for their antioxidant, anti-inflammatory, and inhibitory activities on cancer cell proliferation. Comparatively, secondary BAs have been found to express procarcinogenic activity in the host. These toxic byproducts can initiate genetic instability within the cell and encourage apoptotic resistance increasing cancer cell growth.
Despite the current evidence on the anticancer potential of gut metabolites such as SCFA, bacteriocins and phenylpropanoid-derived compounds, most of these studies were conducted in vitro with limited in vivo testing and no reported clinical studies. The mechanisms of action of these gut microbial metabolites are yet to be comprehensively understood in the context of carcinogenesis and anticancer activity. Further investigation will be crucial to determine the appropriate therapeutic dose of the gut metabolites for their safe clinical use in anticancer therapy. Therefore, future research should prioritise studies-a) to further understand the molecular mechanisms of action and b) to evaluate the potential toxicity of these metabolites, and c) to determine the therapeutic dose needed. Furthermore, it would be exciting to evaluate if these gut metabolites interact favourably with standard chemotherapies to increase their efficacy and safety in clinical settings.