INTRODUCTION

Molecular mechanisms underlying carcinogenesis and pathogenesis of malignant neoplasms are very complex, but specific at the same time [1]. Multiple factors, such as genetic and epigenetic features, age, lifestyle, diet, bad habits, exposure to damaging environmental factors (ultraviolet and ionizing radiation, chemical substances, etc.), can initiate malignant cell transformation and promote progressive tumor growth [1]. A growing number of reports have demonstrated that understanding the role of the factors might be critical in choosing the targets for the anticancer therapy [2]. In last years, it has become evident that microbiota can impact malignant cell transformation at the metabolomic and genetic levels. An increasing number of studies have been dedicated to elucidating the effect of symbiotic microbiota on the host’s metabolism. By coevolving with the hosts, the microbiota has substantially shaped the phenotypes of our ancestors [3]. The connections between the metabolisms of a macroorganism and its microbiota, as well as sharing the same signaling molecules for communication and regulation of mutual metabolism in a holoorganism have resulted in that microorganisms, or rather their metabolites, have become profoundly involved in the pathogenesis of many human diseases, first of all, metabolic disorders, such as obesity, non-alcoholic fatty liver disease, dyslipidemia, insulin resistance, and type 2 diabetes mellitus [4]. However, development of some oncological diseases is also associated with microorganisms [5]. Despite that the International Agency for Research on Cancer (IACR) acknowledges only 11 infectious agents [Epstein-Barr virus, hepatitis B and C viruses, Kaposi’s sarcoma herpesvirus [human herpes virus type 8, human immunodeficiency virus (HIV), human papillomavirus type 16 virus, human T-cell lymphotropic virus type 1 (HTLV-1), Helicobacter pylori, Clonorchis sinensis, Opisthorchis viverrini, and Schistosoma haematobium] as group 1 carcinogens (i.e., agents triggering malignant transformation of host’s cells) [6], other microbes are no less important in carcinogenesis as a part of commensal microbiota ecosystem. The mechanisms by which the microbiota affects malignant cell transformation by either increasing or decreasing the risk of carcinogenesis can be divided into three major groups: (i) altering a balance between proliferation and death of host’ cells, (ii) modulating the functioning of host’s immune system, and (iii) affecting formation of metabolites synthesized by the host, supplied with food, or produced by the microbiota [6].

Here, we briefly discuss the role of commensal microbiota and microbiota-produced metabolites in the induction of carcinogenesis in the host organism, as well as their impact on the efficacy of anticancer therapy.

MICROBIOTIC FACTORS AS MUTAGENS AND REGULATORS OF HOST CELL PROLIFERATION

In the course of evolution, many microbes have developed various DNA-damaging mechanisms in order to destroy competitor species, and thereby, to survive in the microbial world. However, in host cells, such microbial protective factors [e.g., colibactin expressed by group B2 Escherichia coli cells and other enterobacteria [6] and enterotoxigenic Bacteroides fragilis toxin (Bft) and cytolethal distending toxin (CDT) produced by many Gram-negative bacteria including E. coli, Shigella dysentery, and H. pylori] can act as mutagens and promote carcinogenesis [7]. Both colibactin and CDT damage double-stranded DNA resulting in the chromosomal instability and promote senescence in eukaryotic cells [8, 9]. Bft acts indirectly by stimulating formation of reactive oxygen species (ROS), high levels of which damage the host’s DNA, overwhelming the activity of DNA repair systems [7].

Proteins expressed by commensal microorganisms [e.g., CagA protein encoded by the cytotoxin-associated gene A (CagA) and expressed by the oncogenic strains of H. pylori type 1; H. pylori virulence factor] can affect signaling pathways regulating proliferation, survival, and migration of eukaryotic cells [10]. After penetrating into the cytoplasm of the host’s cell, CagA aberrantly activates β-catenin, thus eliciting the Wnt signaling pathway resulting in the upregulated transcription of genes involved in the maintenance of cell stemness, proliferation, and migration, i.e., promotes events required to maintain malignant cell growth. Likewise, Fusobacterium nucleatum as a component of oral microbiota associated with the development of colorectal adenoma and human adenocarcinoma expresses FadA protein (adhesin A) that acts as an adhesion factor on the bacterial cell surface and binds host E-cadherin, leading to the β-catenin activation [11]. Salmonella typhi strains secrete AvrA, an effector protein accounting for the microbe interaction with the host tissues, that activates β-catenin signaling and is associated with the development of hepatobiliary cancer [12].

This activation of the Wnt/β-catenin axis reflects the convergent evolution of the microbiota and host cells and might be related to the formation of adaptive responses by the microbes in the process of creation of a new habitat niche. On the other hand, the presence of cancer-potentiating bacteria and their ability to interact with E-cadherin in developing tumors suggests that disturbance of the host’s tissue barrier is a critical step in the development of some types of malignant tumors. This hypothesis was corroborated by the finding that ammonia produced by H. pylori disrupts cellular tight junctions, thus affecting cell integrity and damaging the gastric epithelium. Recently, it was reported that urease plays a key role in progression of gastric cancer, as it promotes tumor growth and metastatic spread by inducing angiogenesis [10]. As much as 20% oncopathologies are related to the mutagenic effects of pathogenic bacteria and imbalance in the intestinal microbiota (dysbiosis) [13].

ONCOGENIC ROLE OF MICROBIOTIC METABOLITES

The microbiota can also affect malignant transformation of host’s cells by modulating the activity of immune cell. Co-evolution of the host and its microbiota system has been accompanied not only by the emergence of mutually beneficial relationships, but also by the development of barriers preventing pathological effects of the symbionts on the macroorganism. The disturbance in these barriers, i.e., when the microbes and host immune system meet the conditions fully dissimilar to those under which they have co-evolved, can lead to the oncogenic transformation and inflammatory diseases. Once the barriers are breached, microbial metabolites can affect immune responses in the developing tumor microenvironment by exerting the pro-inflammatory and/or immunosuppressive effects. On the other hand, long-term coexistence of symbiotic micro- and macroorganisms has resulted in the production by bacterial cells of protective (anti-inflammatory and antitumor) molecules that defend the host organism from the growth of malignant tumors, thereby ensuring mutual survival of the holoorganism.

Bile acids. Secondary bile acids (SBAs) are important metabolites produced by the fermentation of primary bile acids by intestinal microbiota. The most common SBAs are deoxycholic acid (DCA), lithocholic acid (LCA), and ursodeoxycholic acid (UDCA). SBAs can affect the composition of gut microbiota communities [14, 15]. Several studies have demonstrated that SBAs act as regulatory molecules that activate multiple signaling pathways [16, 17], in particular, induce proliferation of tumor cells while simultaneously inhibiting apoptosis, stimulate invasion and metastasis of tumor cells, and initiate transformation of malignant cells into cancer stem cells (CSCs) [16]. Moreover, SBAs promote carcinogenesis by regulating the functioning of immune cells [17]. Interestingly, a high-fat diet increases a relative abundance of sulfate-reducing bacteria (e.g., Desulfovibrio vulgaris) that generate pro-carcinogenic LCA and DCA [18].

SBAs exerts their effects via nuclear farnesoid X receptor (FXR), G protein-coupled bile acid receptor 1 (GPBAR1, also known as TGR5), vitamin D receptor (VDR), pregnane X receptor (PXR), and constitutive androstane receptor (CAR) [19, 20]. Primary bile acids mainly activate FXR, whereas SBAs activate TGR5 [21]. Activation of TGR5 stimulates proliferation of intestinal cells, promotes DNA damage, and induces senescence-associated secretory phenotype (SASP) [21]. Moreover, SBAs acting on TGR5 can inhibit the functioning of natural killer T cells (NK cells), B cells, dendritic cells (DCs), and macrophages. Thus, DCA and LCA inhibit activation of splenic and intestinal macrophages induced via Toll-like receptor 4 (TLR4) [22], downregulate secretion of interleukin-6 (IL-6), interferon-γ (IFN-γ), and tumor necrosis factor-α (TNF-α), and induce polarization of antitumor M1 macrophages into pro-carcinogenic M2 macrophages [23]. DCA and LCA downregulate secretion of TNF-α and IL-12, thereby suppressing the functioning of DCs [19]. SBAs inhibit the antitumor activity of B cells that act via antibody secretion, phagocytosis, and activation of the complement system [24]. DCA and LCA downregulate IL-6 secretion and suppress maturation of B cells, thereby lowering immunoglobulin (IgE and IgG) levels [22]. NK cells secrete IFN-γ and TNF-α to stimulate apoptosis of tumor cells [25]. However, DCA and LCA inhibit secretion of IFN-γ and TNF-α by suppressing the activity of NK cell [23]. Also, DCA and LCA stimulate IL-10 secretion by NK cells resulting in the downregulation of TNF-α secretion and T cell activity [26]. SBAs enhance the functioning of regulatory T cells (Tregs), which promotes formation of immunosuppressive microenvironment and tumor progression. Foxp3 is one of the key transcription factors of the FOX family controlling the development and functioning of Tregs [27]. IsoalloLCA (LCA derivative) upregulates Foxp3 expression in naive CD4+ T cells via activation of ROS production in the mitochondria [28]. SBAs can bind not only to TGR5, but also to the nuclear FXR expressed mainly in the intestine, liver, and immune cells [22]. Macrophages and DCs express both TGR5 and FXR, whereas NK cells express solely FXR. By acting on these receptors, bile acids increase production of IL-10 and decrease production of IL-6 and IFN-γ by macrophages, downregulate the synthesis of TNF-α and IL-12 by DCs, and suppress secretion of osteopontin by NK cells [22].

Therefore, the signaling functions of SBAs are associated with a higher risk of developing colorectal cancer, as well as some tumors targeting liver, pancreas, esophagus, lungs, and stomach [29-33].

However, it should be mentioned that a nation-wide cohort study in Taiwan revealed the oncoprotective properties of synthetic UDCA in colorectal cancer [34].

Lipopolysaccharide (LPS). LPS with its lipid A moiety is the most efficient protective toxin of bacterial cell wall in Gram-negative bacteria that elicits the pro-inflammatory effect in a host organism [35]. The lumen of the intestine, which harbors many trillions of commensal bacteria, is the main reservoir of LPS in the human body [36]. Interestingly, the binding of LPS to the apical, but not basolateral, receptors of the intestinal epithelial cells induces their apoptosis via caspase-3 activation and promotes destruction of tight junctions formed by ZO-1 (zonula occludens-1 scaffold protein), thereby increasing the permeability of the intestinal epithelium. In the plasma, LPS is transported bound to either LPS-binding protein (LBP) or plasma lipoproteins and triggers systemic inflammation [37]. LPS is involved in the oncogenic cell transformation via multiple mechanisms. Thus, it stimulates TNF-α production resulting in the recruitment of intracellular NF-κB (nuclear factor kappa-light-chain-enhancer of activated B cells) followed by the release of chemokines and inflammatory cytokines ( IL-1β, IL-6, IL-8, TNF-α, etc.) and upregulation of expression of inducible nitric oxide synthase (iNOS) and, accordingly, increase in the level of prooxidant NO [38]. TNF-α is responsible for the activation of various cellular signaling pathways leading to cell necrosis or apoptosis and, therefore, plays a key role in the organism resistance to infections and cancer [39]. Activation of the TLR4-dependent FAK/MyD88/IRAK4 axis controls LPS-induced intestinal inflammation and permeability of tight junctions [39].

Elevated serum levels of LPS increase the risk of developing prostate cancer and also promote tumor metastasis [40]. Thus, colorectal cancer tissues without metastases and with metastases to regional lymph nodes significantly differed in the LPS levels [36, 41]. LPS also markedly increases the motility of tumor cells and promotes lymphangiogenesis. LPS upregulates expression of angiogenic and lymphogenic vascular endothelial growth factor C (VEGF-C) in a dose-dependent manner. Moreover, LPS directly activates the TLR4/NF-κB/JNK axis leading to the upregulation of the VEGF-C expression [41]. By acting as a trigger of inflammatory response, LPS is associated with carcinogenesis, in particular, development of gastrointestinal mast cell tumors [42] (figure).

figure 1

Molecular ensembles of microbiotic metabolites in tumor initiation and progression

Some microbial metabolites, e.g., TLR4 agonists, promote the TLR4-associated signaling cascades, thus playing an essential role in controlling survival and progression of tumor cell growth in colon, pancreatic, liver, and breast cancers [43, 44]. For instance, TLR4 overexpression found in colorectal cancer promotes proliferation of malignant cells, their invasion, and metastasis, as well as generation of tumor-beneficial cellular microenvironment and suppression of apoptosis. In the last few years, there has been an increasing interest in the regulation of TLR4-mediated signaling in colitis-associated oncogenesis [44].

Tryptophan and its catabolites. Tryptophan (Trp) metabolism via the kynurenine pathway and its microbial conversion to indole compounds are of fundamental importance to the host’s health, especially, in colon cancer [45]. Microbiota is a key component of the tumor microenvironment; it affects initiation of malignant cell transformation, tumor growth, and tumor response to treatment. Changes in the Trp metabolism have been observed already at the early stages of tumor growth, when they serve as an adaptive mechanism allowing malignant cells to evade immunosurveillance and to metastasize [46].

One of the key enzymes limiting the rate of Trp catabolism is indoleamine 2,3-dioxygenase 1 (IDO1) that was found in Firmicutes [47]. IDO1 is strongly expressed by representatives of the Clostridium, Lachnoclostridium, Ruminoclostridium, and Roseburia genera [47]. Pro-inflammatory cytokines (IFN-γ, TNF-α, prostaglandins, and LPSs) activate IDO1 expression [48]. IDO1 converts Trp to N-formylkynurenine, which is rapidly transformed into kynurenine (the first stable catabolite in this pathway). Kynurenine is converted to metabolites that control the activity of immune cells [48].

Kynurenine regulates immune homeostasis of the host organism [49] by lowering the number of activated T cells, DCs, and NK cells, as well as by eliciting apoptosis of Th1 cell to control excessive inflammatory response [45]. Each metabolite downstream of kynurenine performs specific functions. For instance, kynurenic acid induces the anti-inflammatory response due to intrinsic antioxidant properties, whereas picolinic acid exhibits the antitumor activity by suppressing activation of T cells and transcription factor c-Myc (proto-oncogenic protein Myc). c-Myc accelerates Trp uptake by colon cancer cells by upregulating expression of Trp transporters (SLC7A5 and SLC1A5) [45].

3-Hydroxyanthralic and quinoline acids can act as toxins, in particular, in colon cancer [46]. The activities of IDO1 and arylformamidase, as well as the level of kynurenine, were found to be elevated during the tumor growth [50]. Higher serum kynurenine levels were observed in cancer patients vs. healthy individuals. At the same time, expression and activity of enzymes involved in further kynurenine conversion remained unaltered, suggesting that kynurenine represents a dominant metabolite in the Trp metabolism elevated in colorectal cancer [50].

Increased kynurenine levels promote tumorigenesis mainly via two pathways: (i) a fraction of produced kynurenine can directly induce inactivation and apoptosis of T cells resulting in immune evasion; (ii) the remaining kynurenine can constitutively activate aryl hydrocarbon receptor (AhR), which upregulates transcription of genes necessary for the tumor escape from immunosurveillance, as well as promotes proliferation and metastasis of malignant cells. Thus, the kynurenine/AhR axis represents one of the major factors contributing to the development of colon cancer [46, 50].

It should be mentioned that the rise in systemic kynurenine levels was found in the patients with tumors of various location. Many studies explain this by upregulation of IDO1 expression by pro-inflammatory cytokines, whose level increases during malignant tumor growth in vivo. Moreover, patients with colon cancer demonstrated reduced indole content and increased microbiota-derived production of kynurenine in the feces [45], while administration of kynureninase, an enzyme that degrades kynurenine, stopped tumor growth [51].

Recently, it was demonstrated that butyrate downregulated IDO1 expression, suggesting that Trp metabolism and, therefore, kynurenine generation via metabolism of short chain fatty acids (SCFAs), are controlled by the commensal bacteria [47].

Indole-3-acetamide, indole-3-acetaldehyde, indole-3-pyruvate, indole-3-aldehyde, indole-3-acetate, tryptamine, indole-3-propionic acid, and indole-3-acrylic acid are the major indole derivatives of Trp metabolism in the gut. Apart from being critical components of bacterial metabolism, indole and its derivatives play an important mediating and signaling role in the host–microbiota crosstalk. Indole production in the gut microbial communities affects spore formation, plasmid stability, biofilm formation, antibiotic resistance, cell division, and virulence of bacterial cells [52]. Indoles activate host’s signaling pathways that affect the barrier function of the intestinal epithelium and reduce its permeability, promote immune tolerance, cause pathogen displacement, reduce inflammation, and control mucin production [53]. Indole compounds can also act via the AhR axis and exert both pro- and anti-inflammatory effects [54]. Indoles downregulate expression of pro-inflammatory IL-8 and NF-κB, thus promoting expression of anti-inflammatory cytokines, in particular IL-10. Moreover, indole compounds regulate intestinal homeostasis by lowering secretion of IL-22, which improves the barrier function. However, at the late stages of cancer, IL-22 production can contribute to the tumor growth [55] (figure). Most indoles and their derivatives protect against inflammatory bowel diseases that often precede cancer development [52].

The studies of the antitumor effect of Lactobacillus gallinarum demonstrated that this bacterium produces large amounts of Trp and indole-3-lactate, thereby suppressing the growth of colorectal cancer cells [56]. Exposure to indole-3-lactate activated apoptosis in cultured tumor cells, whereas AhR agonists reduced the antitumor effect of indole-3-lactate [56].

Excessive intake of dietary cholesterol lowers the number of normal members of the gut microbiota – Bacteroides and Bifidobacterium, which is accompanied by a decline in the content of indole-3-propionate, as well as development of fatty liver disease, ultimately resulting in the emergence of hepatocellular carcinoma [57].

Short-chain fatty acids. SCFAs are important mediators in the metabolic interplay between gut microbiota and host organism. SCFAs not only affect the colon, but can also influence other organs and systems by acting via systemic circulation.

Recent epidemiological studies demonstrated that the development of gastric and breast cancers correlates with a low fecal SCFA level [58]. Clinical studies also revealed that the SCFA concentration in the stool of patients with colorectal cancer was lower than in healthy subjects, which might be accounted for by a decreased content of SCFA-synthesizing bacteria, such as Lachnospiraceae, Roseburia spp., and Bifidobacterium spp. [59].

SCFAs significantly reduce the risk of developing malignant cells via suppression of cell growth and migration, downregulation of histone deacetylase (HDAC) activity, and induction of apoptosis [60]. The major SCFAs are butyrate, acetate, and propionate [61]. It was established that the binding of intracellular butyrate to HDAC suppresses the enzyme activity, resulting in histone hyperacetylation and altered gene expression. Moreover, butyrate inhibits the growth of tumor cells by promoting cell cycle arrest and apoptosis. For instance, mice experimentally colonized with the wild-type butyrate-producing bacteria and fed with a fiber-containing diet exhibited suppressed tumor growth upon the treatment with azoxymethane and sodium dextran sulfate [62]. Moreover, it was further shown that the glycolytic metabolism characteristic of tumor cells suppressed butyrate metabolism and caused intranuclear accumulation of this compound, which resulted in the increased histone acetylation followed by activation of apoptosis and slowing of cell proliferation. On the other hand, microbiota and butyrate stimulated tumor growth in the mouse model of colon cancer due to mutations in the genes for Apc (regulator of Wnt signaling pathway) and Msh2 (protein of DNA mismatch repair system, tumor suppressor) [63]. The major effect of butyrate in this model was stimulation of hyperproliferation of Msh2-deficient epithelial cells. Hence, the effects of SCFAs were due to their ability to selectively activate certain signaling pathways and depended on the genetic characteristics of transformed cells, as well as the butyrate level. These studies once again emphasized the importance of correct interpretation of data of microbiome analysis and tumor cell genomics for the development of proper dietary recommendations for reducing a risk of malignant neoplasms.

Butyrate also affects other epigenetic mechanisms, including histone phosphorylation and methylation, DNA methylation, and hyperacetylation of nonhistone proteins [13]. It was found that a diet containing sodium butyrate alleviated diarrhea by reducing intestinal permeability via upregulation of expression of tight junction proteins claudin-3 and occludin [64].

Faecalibacterium prausnitzii and Eubacterium rectal are important butyrate-producing microbial species [65]. Butyrate is consumed by the mitochondria of colonocytes, which helps to maintain their energy balance and promotes cell proliferation [66]. Butyrate also induces expression of the P21 gene by inhibiting the transcription factor AP-1 axis and potentiating phosphorylation of the proto-oncogene c-Fos and mitogen-activated kinase ERK1/2 [67]. Butyrate-dependent activation of the Akt/mTOR pathway in hepatocellular carcinoma cell line upregulated expression of autophagic proteins (Beclin 1, ATG5, LC3-II) along with the ROS-induced autophagy [68, 69].

Moreover, butyrate downregulates microRNA-92a (miR-92a) expression via c-Myc, which slows down proliferation of colon cancer cells and stimulates apoptosis [7]. Overexpression of miR-92a in colon cancer promotes cell growth and invasion by targeting Krüppel-like factor 4 (KLF4) and downstream p21 protein, while downregulation of miR-92a elicits apoptosis of cancer cells [7].

SCFAs can affect immunoregulation via Treg cells and exhibit the anti-inflammatory and anticarcinogenic effects [21].

Acetate and propionate bind to the G protein-coupled receptors GPR41 and GPR43. Similar to butyrate, stimulation of cells by acetate and propionate cells via GPR41/43 triggers a signaling pathway that prevents inflammation and lowers a risk of malignant cell transformation, as shown in human renal epithelial cells. In particular, SCFAs suppressed TNF-α-stimulated production of monocyte chemoattractant protein-1 (MCP-1) through the inhibition of phosphorylation of p38 and JNK. It was also shown that inflammation in human cell lines (HeLa, HEK293) can be regulated via desensitization of GPR41/43 by β-arrestins. In the case of GPR43, β-arrestin abrogated degradation of NF-κB/IκB and nuclear translocation of NF-κB, resulting in the reduced expression of pro-inflammatory cytokines IL-6 and IL-1β [70].

Summarizing all the data above, SCFAs activate various cellular mechanisms that prevent carcinogenesis via regulating signaling pathways, transcription factors, and epigenome. SCFAs act not only trough binding to the transmembrane receptors, but can also enter the cell and directly interact with multiple intracellular targets. However, it should be taken into account that the effects of SCFAs can be reversed into the opposite (procarcinogenic) ones due to the unique genetic characteristics of some tumor cells, as well as depend on the SCFA concentration in the tumor microenvironment (figure).

THE IMPACT OF MICROBIOTIC METABOLITES ON THE SUCCESS OF ANTITUMOR THERAPY

Gut microbiota can significantly affect the efficacy of chemotherapy. For instance, the efficacy of the platinum-based drug oxaliplatin used to treat several malignant neoplasms of the gastrointestinal tract is influenced by the interaction between the gut microbiota and host’s immune system. Gut microbiota stimulates myeloid cells to produce high amounts of ROS, so that the emerging oxidative stress promotes oxaliplatin-associated DNA damage in tumor cells leading to their death [71]. Cyclophosphamide, an alkylating agent used for the treatment of hematological malignancies and solid tumors, is another example of a microbiota-dependent chemotherapeutic agent. Cyclophosphamide can damage the epithelium of the small intestine and impair its barrier function, which, in turn, leads to the translocation of gut commensal microbes to the secondary lymphoid organs, where they elicit an increase in the pool of tumor-associated antigen-specific Th17 lymphocytes [72]. Treatment with antibiotics prevents microbiota translocation, as well as associated polarization of T cells, thereby lowering the efficacy of antitumor chemotherapy [73].

Recent success of immunotherapy (cytokine therapy, targeted immunotherapy, and vaccine therapy) is one of the most important achievements of modern oncology [74]. Because of a close crosstalk between microbiota and immune system, it is expected that microbiotic metabolites would affect the response of host cells to immunotherapy. Thus, dysbiosis caused by the use of antibiotics reduced the efficacy of immunotherapy with CpG oligonucleotides in mice with subcutaneous tumors [71]. Taking into consideration that immunotherapy is effective in melanoma, bladder, renal and lung cancers, but not in colon cancer (i.e., organ most densely populated with bacteria) [1, 2, 75], it rises an essential question of how microbiota contributes to the success of immunotherapy.

Studies in mice revealed that the gut microbiota member Bifidobacterium spp. potentiated the efficacy of anti-PD1 immunotherapy [76], whereas Bacteroides thetaiotaomicron and B. fragilis were associated with a stronger effect of the checkpoint inhibitors targeting CTLA-4 (cytotoxic T-lymphocyte-associated protein 4, cellular receptor of the immunoglobulin superfamily) [76, 77]. The antitumor efficacy of anti-PD-1/L1 therapy was found to depend on several bacterial species, including Akkermansia, Faecalibacterium, Clostridia spp., and Bifidobacterium spp. [78], due in part to the exposure to microbial metabolites such as butyrate and propionate. However, despite that in some cases high fecal SCFA content was associated with longer progression-free survival or enhanced antitumor response, high systemic SCFA levels were accompanied with poor therapeutic response (treatment failure) [79]. Butyrate can also restrict the potential of DCs to elicit tumor-specific T cells and memory T cells, thereby limiting the efficacy of CTLA-4 checkpoint inhibitors [80]. The alternative pathways for the microbiome–host interactions in the context of cancer immunotherapy include direct stimulation of DCs in lymph nodes by Akkermansia muciniphila [81] or Bacteroides spp. via induced antitumor Th1 and CD8+ T cell response [77].

Moreover, gut microbiota also likely affects the immunotherapy-related toxicity, as some patients receiving targeted therapy [82] (e.g., anti-CTLA-4 and anti-PD-L1 antibodies) are known to develop severe colitis, the pathogenesis of which is strongly influenced the gut microbiota.

MOLECULAR ENSEMLES OF MICROBIOTIC METABOLITES IN CARCINOGENESIS

Metabolites produced by the gut microbiota exert multiple and often multidirectional effects depending on their concentration, as well as expression and abundance of relevant cognate receptors. However, we believe that “metabolomic stellar sky”, which differs clearly in healthy people and patients with malignant neoplasms, has its own specific “constellations” – unique molecular shifts, which, according to the “domino effect”, become involved in coordinated and irreversible pathological mechanism of carcinogenesis.

As a result of disturbances in the host’s physiology and in the composition of microbial community during induction of carcinogenesis, commensal microbes can exert a pathogenic effect by promoting intestinal inflammation and biofilm formation [83]. An increased content of Enterococcus faecalis and E. coli upregulates production in the intestinal cells of pro-inflammatory signaling molecules (IFN-γ and IL-4) that activate IDO1 expression and alter patient’s metabolism [84]. F. nucleatum causes a significant increase in the expression of the pro-inflammatory cytokine TNF-α resulting in the IDO1 hyperactivation [84]. F. nucleatum and Peptostreptococcus anaerobius can attach to the tumor cells via cell surface adhesion proteins leading to the activation of the PI3k/Akt-axis followed by the activation of cell proliferation [11]. It should be noted that both bacterial species can produce high amounts of indole molecules through Trp catabolism. Therefore, the microbiome acts as a key factor in inflammation precisely due to alterations in the Trp metabolism [45].

Another important oncometabolite is kynurenine (including that produced by the microbiota). As described above, kynurenine synthesis depends on the IDO1 expression level and activity, which in turn depend on the SCFA content [47]. It was experimentally confirmed that butyrate and, to a lesser extent, propionate, isobutyrate, isovalerate, and valerate inhibit the IDO1 activity in bacterial supernatants. A decrease in the fecal and blood plasma levels of SCFAs is a poor prognostic factor in tumor progression. Moreover, the content of Trp and microbial indole catabolites performing numerous protective functions also decreases [47].

Bacterial metabolites can produce multiple immunomodulatory effects by directly affecting tumor immune evasion or, conversely, ability of immune cells to target cancer cells. For instance, LPS-activated macrophages play a critical role in the stimulation of formation and activity of inflammasome, which produces large amounts of pro-inflammatory cytokines (CCL2, TNF-α, IL-12, and IL-6). Nastasi et al. [85] showed that butyrate and propionate played a pivotal role in modulating immune response in mature human DCs. Moreover, propionate and butyrate can significantly decreased IL-6 expression and secretion. Along with that, butyrate and propionate downregulated LPS-induced gene expression and production of IL-12B (IL-12p40), a common subunit necessary to form both IL-12p40 and IL-23 [86]. The authors hypothesized that both these SCFAs account for naïve T cell polarization by decreasing the pro-inflammatory potential of Th1 and Th17 cells, thereby shifting the balance towards the emergence of anti-inflammatory populations, such as Treg cells, at the expense of compromised IL-12 and IL-23 production. Moreover, the activity of butyrate and propionate turned out to be selective, as they affected expression of primary LPS response genes (in particular, TNF-α and CCL2 gene families), whereas expression of other genes (e.g., HLA-DR genes encoding major histocompatibility complex class II proteins, CD86, IL1A, IL1B) remained unaltered or even activated [85].

Intestinal alkaline phosphatase neutralizes LPS by catalyzing dephosphorylation of lipid A active (toxic) moiety, thereby preventing local inflammation and translocation of active LPS to the systemic circulation [37]. The activity of this enzyme strongly depends on the level of free L-amino acids, especially, L-phenylalanine (Phe) [87]. Dysbiosis and increased levels of Phe, tyrosine (Tyr), and Trp are typical of microbiota forming the tumor environment [88]. In turn, an increase in the Phe content results in the inhibition of intestinal alkaline phosphatase, expectedly leading to impaired LPS dephosphorylation and its active translocation to the systemic circulation. By acting via TLR4 and NF-κB-axis, LPS creates a pro-inflammatory and pro-carcinogenic status. Pro-inflammatory cytokines regulate Trp catabolism and formation of kynurenine. The latter suppresses immunosurveillance and allows malignant cell transformation. In turn, SCFAs protective against tumor formation by suppressing enzymes regulating kynurenine biosynthesis [47]. However, the emerging primary pro-oncogenic dysbiotic profile is characterized by a decrease in the content of microorganisms producing butyrate and other SCFAs [89]. The pro-oncogenic dysbiosis profile is also characterized by the elevated formation of SBAs (factors of aggression in tumor cell transformation). Thus, SBAs inhibit TLR4-dependent activation of spleen and intestinal macrophages [22] and, therefore, along with kynurenine, suppress the functioning of immunocompetent cells and create profiles of signaling molecules and related target pathways favorable for immunosuppression (figure).

CONCLUSION

Almost all human tissues and organs are influenced by the commensal microbiota. Molecular interactions between bacterial and eukaryotic cells have been established in the course of evolution, resulting in the emergence of a complex network of interactions with multiple intersections represented by signaling pathways, whose activity is determined by both microbiotic metabolites and host cells. Expectedly, the microbiota not only affects normal physiological functions, but is also involved in the pathogenesis of human diseases including cancer. It is not surprising that dysbiosis with its polymorphic microbiome is recognized as one of the hallmarks of carcinogenesis [90]. By acting as mutagens or signaling molecules aberrantly modulating the activity of host’s signaling pathways, microbiotic metabolites can initiate malignant cell transformation and promote tumor growth and metastasis. On the other hand, many metabolites show the oncoprotective effect by directing the host’s immune system toward combatting the tumor or by directly promoting apoptosis and senescence of malignant cells. The efficacy of antitumor therapy strongly depends on the microbiome composition, and microbiota transplantation is viewed as one of the promising approaches to increase the therapy success and survival of cancer patients [91]. However, despite an avalanche-like increase in the number of studies on the role of microbiota in carcinogenesis, the exact mechanisms by which microbial metabolites affect the host remain unclear. Some metabolites (e.g., butyrate) produce pleiotropic and multidirectional effects on different cells depending on the cell differentiation stage and genetic background, as well as on the concentration of metabolite itself. Moreover, the effects of these regulatory molecules are often influenced by other signaling metabolites produced by the microbiota and tumor microenvironment. Studying possible scenarios for the action of microbiotic metabolites will allow in the future to use their modulating potential for fighting cancer and increasing the efficacy of antitumor therapy.