Microbiome in drug resistance to colon cancer

Abstract

Metagenomic analyses have revealed microbial dysbiosis in the gut of patients with colorectal cancer (CRC). The gut microbiota influences CRC via a variety of mechanisms, including microbial-derived factors such as metabolites or genotoxins. Pathogenic drivers and opportunistic passenger bacteria may underlie direct effect of the gut microbiota on carcinogenesis. We posit that metabolites generated by gut microbiota can influence CRC through a multitude of epigenetic or genetic effects on malignant transformation. A closer look at the cross talks between the commensals, epithelial cells, immune regulators etc., needs to be established with more substantiated studies. The recurrence of chemoresistant disease following therapy undoubtedly provides the impetus for morbidity and mortality; yet, the role of gut microbiome in drug resistance remains to be fully investigated. We review the current literature on microbial dysbiosis during CRC and discuss the mechanistic basis of CRC-associated bacteria in tumor initiation, progression and drug resistance.

Colorectal Cancer and Gut Microbiome

One of the leading causes of mortality and morbidity due to cancer, across the globe is colorectal cancer (CRC). It is the third most prevalent form of cancer and fourth most common cause of cancer deaths worldwide. The vast majority of CRC cases are sporadic (~70%) and non-inherited. Inherited predisposition syndromes including Lynch syndrome or familial adenomatous polyposis (FAP) account for about 5% of the new CRC diagnosis. 20–30% of the cases have a familial disposition with no associated or known germline mutation [1]. The disease originates from the epithelial cells that accumulate mutations and acquire malignant properties over time. In addition to genetic factors, factors such as diet, smoking, obesity, diabetes, alcohol consumption, and exposure to carcinogens increase the risk for developing sporadic CRC.

Human microbiome are complex ecosystems involving bacteria, viruses, archaea, or eukaryotes that are co-evolving in an environment subject to various selective pressures, such as antibiotic administration, diet and/or lifestyle. An optimal “balance” in the microbial population (termed “eubiosis”), is essential for the sustenance as well as maintenance of good health in humans. In the gut, such a complex bacterial community obtains energy from indigestible dietary components, provides accessory growth factors, protects against colonization by pathogens, provides post-natal maturation of mucosal structures and educates the immune system. The colon and ileocecal valve that exhibit the highest bacterial density along the gastrointestinal tract point towards an important role for the microbiota in CRC. While efforts are underway to specifically link microbial dysbiosis to consensus molecular subtypes (CMS), several bacterial species with causal and/or complicit roles in CRC have been identified [2*]. Studies have also revealed that during malignant transformation, loss of gut vascular barrier can facilitate dissemination of intestinal bacteria to the liver [3**] thereby suggesting that targeting tumor microbiome can be a viable strategy to target cancer progression and metastasis. It can therefore be gleaned from recent explosion of information on tumor microbiome that components of CRC including progression, treatment and even resistance to chemotherapy [4] and clinical outcomes can all be impacted by bacterial dysbiosis [5**].

Association vs. Causation – possible mechanisms implicating gut microbiome in CRC

Despite a plethora of studies linking changes in gut microbiome to CRC, the scientific community is still debating as to whether gut microbes should be considered causal, co-varying, or a necessary but not sufficient agent in CRC development. In 2011, Sears and Pardoll [6] proposed an ‘alpha-bug’ hypothesis wherein, species such as Bacteroides fragilis exert a central pro-oncogenic, enterotoxigenic role, thereby contributing to the onset of CRC. Later, Tjalsma H et al. [7] provided further credence to this concept by proposing the driver-passenger model in 2012 wherein, driver bacteria (e.g., B. fragilis) lead to a multistep development of colorectal tumorigenesis including inflammation, increased cellular proliferation, and/or the production of genotoxins. CRC with hypermutable microsatellite instability (MSI)-high and CpG island methylator phenotype (CIMP)-high phenotypes, as well as CRC with BRAF mutation, may alter the potential pathogenic influence of intestinal microbiota along the proximal-distal axis. To support this, it was shown by Dejea et al. that proximal but not distal colons consistently exhibit biofilm [8*] and interestingly, these biofilms invade the colonic crypts [9] consisting mainly of known CRC-associated bacteria [10]. Finally, Purcell et al. [11], showed that these bacteria are associated with the consensus molecular subtype (CMS) 1, which is called MSI immune subtype, marked by MSI, CIMP-high, BRAF mutations, and immune cell infiltration [11]. Thus, to evaluate their distinct features, it is important to consider microbiomes from proximal and distal colon cancer separately. Studies have also suggested different set of risk factors for these consensus molecular subtypes. Along these lines, smoking has been linked to proximal and rectal cancer but not distal colon cancer, while physical activity was inversely correlated to proximal and distal colon cancer, but not rectal [12–14]. Height and gender are also reported to impact the type of cancer, for example females are more prone to the proximal colon than distal cancer, compared to men, while greater height is correlated with higher risk for distal colon and rectal cancer, respectively [15]. Chronic inflammation resulting in polyps at the bowel lining has been linked with most of the CRC cases [16].

Propagation of chronic inflammation by genotoxins [17*] and altered signalling pathways [18] have been suggested as probable mechanisms of carcinogenesis in liu of gut dysbiosis. Other mechanisms include release of metabolites with context-dependent tumorigenic effects (Figure 1). For example, the production of secondary bile acids may alter immune function and influence tumour growth in the context of hepatobiliary cancer [19,20]; or levels of circulating oestrogens may be altered in the context of breast cancer [21, 22]. Studies have also shown that gut microbiota can promote colon cancer either through 8-oxoguanine-induced lesions in mismatch repair-deficient colons [23] or colibactin-induced DNA adduct formation [24]. In an elegant study, Kadosh et al. [25**] recently demonstrated that the gut microbiome switches mutant p53 from tumour-suppressive into an oncogenic protein by promoting the Tcf4-chromatin interaction and the hyperactivation of Wnt, thus conferring a malignant phenotype to the organoids and throughout the gut (Figure 1). Beyond the resident microbiota of the gut, existence of microbiota within the tumour microenvironment (TME) has been observed in tumours proximal and distal from the resident microbiota. The explanation for the existence of intratumoral bacteria however, is unclear although some studies suggest the role of altered mucosal integrity of microbiota-resident systems such as the GI tract and preferential homing of bacteria for the TME due to the rich vasculature and nutrients [26, 27]. Nevertheless, tumour microbiota may have an important role in the context of cancer treatment, including resistance to chemotherapy [28] and altered clinical outcomes [29].

Figure 1. Influence of gut microbiome on cellular transformation and metastasis.

In a prototypic example, certain components of a Western diet can influence gut microbiota and generation of metabolites through altered metabolism and an exaggerated immune response. A combination of these changes may lead to cellular transformation and neoplasia. The onset of a malignant disease can be further influenced by disruption of gut vascular barrier leading to metastatic spread. The gut microbiome has also been recently shown to promote switching of mutant p53 (mp53) from tumor-suppressive into an oncogenic protein to confer a malignant phenotype throughout the gut.

Evidence towards a functional role of the gut microbiota in tumour development

In CRC, alterations in both tumor microbiome and those of the adjacent mucosa have been suggested as important determinants of disease progression [3**]. A combination of intrinsic as well as environmental factors including medication, diet, smoking etc., have been shown to affect the gut microbiome wherein, changes in species such as Fusobacterium nucleatum, Escherichia coli, and Bacteroides fragilis are directly implicated in disease occurrence in preclinical studies [3**]. Here, we briefly review the potential mechanisms through which these bacteria can influence CRC development and progression.

A. Fusobacterium nucleatum:

F. nucleatum has been linked to human diseases including periodontal diseases, pregnancy disorders, rheumatoid arthritis, respiratory tract infections and cancer. Fusobacterium appears to be an important component of the TME in CRC. A study in Chinese population with CRC revealed the enrichment of F. nucleatum in colon tissues [30, 31]. F. nucleatum mediates carcinogenesis through virulence factors that upregulate the expression of inflammatory genes and downregulate acquired immunity mediated by T cells [32]. FadA antigen, from F. nucleatum binds to the extracellular domain of E-cadherin to stimulate tumor proliferation through the Wnt signaling pathway [33]. Rubinstein and team [33] used FadA knock-out mutants to prove the effect of adhesion, and the β-catenin signaling cascades. Another mechanism involves suppression of anti-tumor immune response by expansion of FOXP3 non-T_reg cells [34]. F. nucleatum was also correlated to the enhanced expression of another cell-adhesion protein, Fap2, a lactose-binding inhibitor that defends tumors from host immune attack, in the CRC patients [35].

B. Enterotoxigenic Bacteroides fragilis (ETBF):

ETBF produces toxin B. fragilis toxin (BFT), an established risk factor for CRC, and is more closely related to advanced stage of CRC [36]. BFT induces cancer proliferation by triggering pathways like nuclear Wnt/β-catenin and NF-κB signaling during CRC induced colitis and colonic tumors in Apc^Min/+ mice [37**]. BFT-mediated pathogenesis also increases gut permeability, which enhances translocation of microbial products [37**]. Research team led by Purcell et al. [11], advocates the use of ETBF as a potential marker for early diagnosis of CRC.

C. Escherichia coli:

Compared to healthy individuals, E. coli was present in abundance in CRC patients. Prevalence of this strain is involved in the initiation and development of CRC [38]. The colon cancer associated E. coli takes over the regular p38 MAPK signaling pathway, which regulates autophagy [38]. Another E coli (pks+) strain with the polyketide synthase gene complex, mediates carcinogenesis by production of colibactin that promotes activation of senescence-associated secretory phenotype in malignant or premalignant epithelial cells that is associated with secretion of various growth factors, cytokines, chemokines, and enzymes [39, 40]. Some of these factors could be protective by contributing to immune-mediated tumor control, but prolonged senescence-associated secretory phenotype could cause immunosuppression [41]. In addition, enzymes secreted by senescent tumor cells, such as matrix metalloproteinases, could promote tumor invasion and metastasis [42*]. Factors like COX-2/PGE2 from E. coli secretome create a conducive TME for self-propagation. E. coli invades the intestinal epithelial cells through adhesin Afa and Eae, to activate similar pathways which further enhance proliferative activity [43**].

D. Streptococcus gallolyticus:

S. gallolyticus triggers carcinogenesis through invasion in premalignant lesions and uncontrolled proliferation [44]. Increased cell proliferation was observed on introduction of S. gallolyticus to cell line [45], which was attributed to increase in nuclear β-catenin. The increased cell proliferation and elevated β-catenin signaling was observed in in vitro cultured cells, in xenografts, and in colonic crypt cells in the mouse model. Another study posited over expression of COX2 and inflammation coupled with prevention of apoptosis [43]. Recruitment of tumor-infiltrating immune cells was reported, on exposure to S. gallolyticus, leading to suppression of immune system and inducing neoplasia [46].

E. Campylobacter species.

Some invasive Campylobacter species have been suggested to contribute to tumor progression by inducing a pro-inflammatory response driven by IL-18 [47].

Microbial Effects on Drug Metabolism and Chemoresistance

The growing relevance of the gut microbiota to various human disease may also directly impinge on the efficacy of chemotherapeutics. Array of anti-cancer drugs and combinations, primarily targeting the cellular signaling pathways involved in tumor survival and growth, have demonstrated increased survival in CRC patients. 5-fluorouracil (5-FU)- remains the mainstay of therapy for patients with CRC. 5-FU is a pyrimidine analog used as anti-neoplastic agent to induce cell death [48]. Platinum, leucovorin, oxaliplatin, irinotecan or monoclonal antibodies (as monotherapy or combinatorial therapy) have received wide application for CRC patients [49, 50]. In 1996, irinotecan (CPT-11), a semi-synthetic camptothecin derivative that selectively inhibits topoisomerase I (Topo I), was approved by the FDA for CRC treatment. CPT-11 forms a topoisomerase-inhibitor-DNA complex hampering the DNA function. Cyclophosphamide mediates its antitumor activity by stimulating the antitumor immune response, and controls tumor growth by inducing immunogenic cell death, destroying immunosuppressive T cells, or promoting Th1 and Th17 cells [51**]. Diaminocyclohexane (DACH) together with its platinum compound causes DNA to be less prone to the repair mechanisms, thus detrimental to the tumor cells. Capecitabine is the oral anti-metabolite first-line therapy for patients with metastatic CRC. Bevacizumab is a humanized mAb against vascular endothelial growth factor receptor (VEGFR-A) [52, 53]. In combination with FOLFOX, FOLFIRI, XELOX, and XELXIRI, bevacizumab has been found to increase the tumor control rate [53]. With the increased usage however, patients develop resistance with nearly half of them resistant to 5-FU-based chemotherapies [54]. TP converts capecitabine to 5-FU and is the driver of its chemoresistance. Oxaliplatin chemoresistance is caused by multiple mechanisms including nucleotide excision repair and WBSCR22 protein [55].

The gut microbiome, through its ability to metabolize drugs can have implications on drug activity/efficacy based on increased or decreased toxicity. Gut bacteria harbor enzymes and pump out other molecules that can influence how medications are activated or broken down. A growing body of evidence supports the impact of gut microbiota on development of chemoresistance to some of the most frequently used anticancer drugs like oxaliplatin, cyclophosphamide, irinotecan, cyclophosphamide anthracyclines etc., through mechanisms such as microbial translocation, immunomodulation, metabolism, enzymatic degradation and reduced ecological diversity [54] (Figure 1). It is suggested that antibiotic treatment might induce dysbiosis which further leads to prevalence of Staphylococcus and Clostridium coupled with depletion of Bacteroides and Lactobacillus, as observed in CRC patients displaying chemoresistance [55]. A recent study discovered that subcutaneous tumors fail to respond to immunotherapy and platinum chemotherapy after antibiotic treatment, whereas another study reported that the effect of Cyclophosphamide on the anti-tumor immune response relies on the presence of a healthy gut microbiota [51**]. Thus, both studies clearly suggest that microbial disturbance through the use of antibiotics severely compromises efficacy of both immunostimulatory (CpG, Cyclophosphamide) and platinum-based chemotherapeutics. F. nucleatum adopts the strategy of increased autophagy to develop chemoresistance against 5-FU and oxaliplatin, by targeting the immune receptors TLR-4 & MYD88 and specific mRNAs [4**]. Patients harboring F. nucleatum in abundance are more susceptible to disease relapse due to chemotherapy failure. E. coli and Clostridium species are known to metabolize Irinotecan increasing its toxicity and thus the drug efficacy is reduced, eventually leading to chemoresistance. In an elegant study, Geller et al. [26] reported that gemcitabine resistance could be caused by intra-tumoral bacteria thereby suggesting that gemcitabine-ciprofloxacin therapy is a viable option to enhance chemotherapeutic efficacy.

Microbe-based approaches to counter drug resistance

A new frontier targeting bacterial dysbiosis has emerged in our efforts to prevent or treat CRC. The aim is to achieve the “optimally balanced flora” to alter the community which further influences the metabolite production and modulatory functions. Fecal microbiota transplantation (FMT) for recurrent C. difficile infection for example, has been utilized to improve bacterial, viral, fungal or archaeal diversity into the host receiving the transplant. Additionally, several clinical trials based on FMT are underway for a myriad of diseases ranging from inflammatory bowel diseases to cancer to psychiatric conditions. FMT however, is not a one-size-fits-all strategy and is subject to safety concerns particularly in the setting of current pandemic. Beyond undefined fecal microbiota transplants, targeted formulations such as probiotics with live microorganisms are being developed as therapeutic agents with defined clinical benefit. These next generation probiotics include microorganisms that have been modified genetically but are still within the regulatory guidelines. Probiotics not only produce anti-inflammatory factors to extend the immune stimulating function, but also secrete antioxidant, anti-cancer compounds, and short chain fatty acids to improve intestinal barrier function. In the context of CRC, Lactobacillus casei variety rhamnosus (Lcr35) prevented FOLFOX-induced intestinal mucositis in CRC-bearing mice [57].

More recently, scientists have turned to compounds produced by microorganisms, released from food components or microbial constituents, including non-viable cells - also called postbiotics that have a potential to promote health and well-being when administered in adequate amounts. Along these lines, heat-inactivated Akkermansia muciniphila was observed to alleviate features of metabolic syndrome in overweight and obese subjects [58]. Similarly, diet, pre- and postbiotics are equally relevant in either improving the gut microbiota or reversing the microbial dysbiosis. Indeed, diet in addition to medication, can shift the the microbial profile within 24 hours and can have long term effects as well. Prebiotics or dietary fibers when combined with live beneficial microbes have the ability to alter the patient’s microbiota into favorable microbiome akin to FMT [59]. In near future, targeted removal of early-stage carcinogenic members of the gut microbial community, by employing bacteriophages are proposed to reduce the risk factors for CRC [59]. Moreover, profiling the oral microbiome may offer an attractive screening method to detect CRC [60, 61].

Conclusion.

Based on elegant studies focusing on gut microbiome, it can be inferred that gut microbiome can impact CRC and other cancers at distant sites through immune system modulation. Microbial species have been identified in many tissue types with the advent of bacterial RNA sequencing including breast, lung, ovary, pancreas, melanoma, bone, and brain tumors etc., [62, 63**]. The hypoxic tumor microenvironment can facilitate the growth of anaerobic and facultative anaerobic bacteria such as Clostridia [64] while necrotic areas of the tumor can release chemotactic compounds to promote bacterial invasion [64]. The leaky vasculature of cancerous tissues with either absence or low abundance of immune cells, may allow bacteria to enter and grow [64] (Figure 1). Along this line, intra-tumoral administration of non-pathogenic bacteria has been suggested as a mechanism of direct drug or therapeutic delivery [65, 66]. At the same however, intra-tumoral bacteria can also drive resistance to anti-cancer drugs [28]. Thus, a holistic approach is needed to develop mechanistic insights into how the microbiome can be manipulated to augment microbe-based therapies in cancer.