Molecular mechanisms of CRC
Several mechanisms, including those summarized below, have been related to the onset and metastasis of CRC (Figure 1).
Figure 1 The general molecular mechanisms of colorectal cancer.
CRC: Colorectal cancer; MPE: Molecular pathological epidemiology; PA: Physical activity; CIN: Chromosomal instability; 67LR: 67 kDa laminin receptor; MSI: Microsatellite instability; EMT: Epithelial-to-mesenchymal transition; WE: Warburg effect.
Diet and lifestyle: CRC is generally reported as diet- and lifestyle-related pathology and is associated with several main factors: Diet, physical activity, consumption of alcohol, cigarettes and aspirin.
Diet: Findings from a systematic review have demonstrated that various foods are associated with CRC, positively or negatively. In general, the higher or lower risk of CRC is related to the proinflammatory or anti-inflammatory property of the food, respectively. Moreover, different foods can exert the function via different mechanisms. We will take some of these foods as examples to explain the mechanism that they act on CRC, briefly.
(1) Red and processed meats: There are several systematic reviews and epidemiological studies indicating that intake of red and processed meats will increase the risk of CRC[10-12]. Besides, a study demonstrated that the consumption of red and processed meats was associated more strongly with an increased risk of CRC with KRAS-wildtype, indicating that the potential mechanism should be studied. Intake of red and processed meats at high temperatures results in the formation of heterocyclic amines and polycyclic aromatic hydrocarbons, and then allows the formation of DNA adducts that subsequently cause DNA damage to promote tumorigenesis[13,14]. In red meat, heme is present in high concentrations in the form of myoglobin and a large amount of heme iron intake has been associated with a higher risk of CRC. Heme iron from red and processed meats can stimulate the metabolism of nitrate/nitrite and the formation of N-nitroso compounds, and induce oxidative stress and lipid peroxidation (LPO) to trigger inflammatory response, and thereby promote the development of CRC[14,15].
(2) Dietary fats: Dietary fats are also associated with CRC. A high intake of ω-6 polyunsaturated fatty acid (PUFA) and saturated fat has tumor-enhancing effects. Rapid metabolism of arachidonic acid (AA), increased activities of phospholipases, and elevated levels of cyclooxygenase (COX) and lipoxygenase (LPO) may suggest the potential mechanism of fatty acid promoting the incident of CRC. However, ω-3 PUFA intake can reduce the risk of CRC, particularly with microsatellite instability (MSI)-high cancer subtype or high FOXP3+ regulatory T cell (Treg cell) counts[18,19]. ω-3 PUFA exerts the effect of anticancer through several potential molecular mechanisms, including suppression of AA-derived eicosanoid biosynthesis, impact on transcription factor activity, gene expression, and signal transduction pathways, increased or decreased production of free radicals and reactive oxygen species (ROS), and so on. In addition, marine ω-3 PUFA also protects against CRC through inhibition of the T cell-suppressive activity of Treg cells. In addition, oleic acid, the main monounsaturated fatty acid in olive oil also exerts a protective effect on CRC[21,22]. A competitive inhibition by oleic acid of the ∆6-desaturase will suppress the eicosanoid biosynthesis of AA to disrupt the tumor growth progress.
(3) Vitamin D: Vitamin D can inhibit the development of CRC, particularly of some specific subtypes of CRC. The beneficial survival association of high vitamin D level is stronger for CRC with lower-level peritumoral lymphocytic reaction than for carcinoma with higher-level reaction. Vitamin D is hydroxylated in the liver to produce 25(OH)D that serves as a standard indicator of vitamin D activity. And, then, 25(OH)D is hydroxylated further in the kidneys to produce a hormonally active metabolite, 1,25-dihydroxyvitamin D. Vitamin D and its metabolites exert their antineoplastic effect by binding with the transcription factor vitamin D receptor. Vitamin D may suppress signaling pathways and cytokines and modulate adaptive immune cells, such as B cells, helper T cells (Th cells) and Treg cells. Moreover, vitamin D diet can also cause significant changes in the fecal microbial community structure. During the development of CRC, vitamin D deficiency can not only cause a sharp decrease in Akkermansia muciniphila but also induce changes in the expression of mucus and goblet-cell associated genes, so that the gut barrier integrity is destroyed.
(4) Dietary fiber: A high intake of dietary fiber, particularly derived from vegetables and fruit, was inversely associated with CRC risk[26,27]. This association was driven mainly by the position of the tumor, which was stronger for the risk of rectal cancer. However, a new study also indicated that the relationship of fiber and risk of CRC is independent of tumor subsite or molecular marker. In addition, higher intake of dietary fiber was more strongly associated with lower risk for Fusobacterium nucleatum (F. nucleatum)-positive CRC, but not F. nucleatum-negative CRC. The findings suggest a potential role for intestinal microbiota in mediating the association between fiber and CRC. Fiber can be fermented by the gut bacteria into short-chain fatty acids, such as butyrate, acetate, and propionate, that possess a diversity of tumorsuppressive effects. High level of short-chain fatty acids produced from fiber might alter pH, increase transit time of gut contents, and lead to differences in local immune surveillance, and thereby reduce the growth of harmful species, such as F. nucleatum. In addition, the potential mechanism underlying fiber inhibiting the development of CRC also contains other following aspects: increasing the stool bulk; shortening the bowel transit time; diluting the effect of potential carcinogens; and altering bile acid metabolism[26,32].
(5) Selenium (Se): Epidemiological investigation has demonstrated that higher Se levels were inversely associated with the risk to develop CRC in Europeans. Usually, dietary Se intake is essential for synthesizing selenoproteins that are important in inhibiting oxidative and inflammatory processes linked to colorectal carcinogenesis. Se supply might play an important role in regulating expression of some selenoproteins, such as glutathione peroxidases (i.e., GPX1), selenoprotein F (SELENOF), selenoprotein P (SELENOP), selenoprotein K (SELENOK), and components of the thioredoxin reductase system (TXNRD1-3), to reduce the oxidative stress and inflammatory response[34,35]. However, some studies also suggested that the selenoprotein expression may affect CRC development independent of the Se status, even leading to the development of CRC with suboptimal Se status[34,35]. In addition, a study also indicated that intake of Se nanoparticles can activate autophagy to promote cancer cell death, through up-regulation of beclin 1-related signaling pathways.
(6) Folic acid: Accumulating evidence displays that folic acid can also be an effective chemopreventive agent for CRC[37,38]. Supplemental folic acid has been shown to prevent the loss of heterozygosity of the tumor suppressor gene that is deleted in CRC and to stabilize its protein in normal appearing rectal mucosa of patients with colorectal adenomas. In addition, deficiency folic acid may lead to inadequate purine and pyrimidine synthesis and changes in methylation, with a concomitant impact on DNA replication and cell division due to the disruption of folate cycle. Thus, deficiency of folic acid can promote epidermal growth factor receptor (EGFR) expression through reducing methylation of CpG sequences within its promoter.
And (7) Others: Other foods are also associated with CRC positively or negatively. High sugar foods and spicy foods might have a positive association with CRC risk; however, vegetables, soy bean/soy products, seafood, and vitamins C, E and B12 play a protective role against CRC risk[6-8]. These foods exert the promotive or protective effect on CRC through modulating the inflammatory response, insulin resistance, and the composition of gut microbiota mainly[40,41].
Physical activity: Physical inactivity has also been well demonstrated as a lifestyle risk factor for CRC. There are many epidemiological studies indicating that physical activity (PA) is associated with a statistically significant reduction in CRC risk[42-45]. Moreover, the association may depend on the location of tumor and gender[42-45]. Some meta-analysis and systematic reviews indicated that PA is associated with reduced risk of both proximal colon and distal colon cancer, but the difference was observed between the colon and the rectum[42,43]. There is even no association observed between PA and rectal cancer. And, gender is another factor to impact the relationship between the PA and CRC. A systematic review has observed an apparent interaction between sex and PA in relation to CRC risk, that being a statistically significant reduced risk among men but statistically nonsignificant reduced risk among women. In addition, there is a potential interactive effect of PA and sedentary time on CRC risk. The benefits of moderate to strenuous-vigorous PA on CRC risk are observed most clearly among those with more sedentary time because these individuals have lower total activity. Several plausible biological mechanisms have been proposed, including changes in endogenous sexual and metabolic hormone levels and growth factors, decreased obesity and central adiposity, and possibly changes in immune function and so on. Peroxisome proliferator-activated receptor gamma coactivator 1α (PGC-1α) is a mitochondrial regulator in a wide variety of biological processes, such as thermogenesis, circadian rhythm, fatty acid oxidation, glucose metabolism, mitochondrial organization, and biogenesis. PA, as a stressor that demands energy, stimulates PGC-1α expression, increasing biological processes of CRC and suppressing the development of CRC.
Consumption of alcohol: High consumption of alcoholic beverages may lead to an increasing risk of CRC[48,49]. Consumption of alcohol is also relative to molecular subtypes of CRC. Alcohol intake was positively related to risk of BRAF-tumors but not to risk of BRAF-positive tumors, irrespective of their KRAS status. Similarly, a study has demonstrated that higher alcohol consumption was associated with risk of CRC with insulin-like growth factor 2 (IGF2) differentially methylated region-0 (DMR0) hypomethylation but not risk of cancer with high-level IGF2 DMR0 methylation. IGF2 up-regulation by DMR0 hypomethylation caused by alcohol may promote tumorigenesis in colorectal tissue. Alcohol can also interfere with one-carbon metabolism, a complex network of interrelated biochemical reactions that involve the transfer of one-carbon (methyl) groups from one compound to another. Excess alcohol can antagonize methyl donors, including vitamin B6, vitamin B12, methionine, and folate, leading to a lower concentration of S-adenosylmethionine in the liver, and thereby cause abnormal DNA methylation. Thus, alcohol can impair the bioavailability of dietary folate as well as folate-dependent intermediary metabolisms to cause carcinogenesis. Besides, monocyte chemoattractant protein-1 (MCP-1) is a chemokine that plays an important role in regulating tumor microenvironment and metastasis. Alcohol can increase the expression of MCP-1 and its receptor CCR2 at both protein and mRNA levels. The study demonstrated that alcohol may promote the metastasis of CRC through modulating the glycogen synthase kinase 3β (GSK3β)/β-catenin/MCP-1 pathway.
Cigarette: Cigarette smoke is considered as a risk factor for CRC. A study found that individuals with heavy, long-term cigarette smoke exposure were significantly younger at the time of CRC diagnosis compared to lifelong never-smokers. And, smoking is also correlative to some specific subtypes of CRC, such as MSI-high, CpG island methylator phenotype (CIMP)-positive, and BRAF mutation-positive subtypes. This finding from the study also indicated that epigenetic modification may be involved in smoking-related carcinogenesis. In general, heterocyclic aromatic amines and polycyclic aromatic hydrocarbons may play an important role in CRC associated with smoking[56,57]. N-Acetyltransferases 1 and 2 (NAT1 and NAT2) are also considered to participate in the metabolism of aromatic and heterocyclic aromatic amines. Glutathione S-transferases (GSTs), particularly GSTM1, GSTT1 and GSTP1, are detoxification enzymes that have been known to metabolize a wide range of carcinogens from cigarette smoke, such as heterocyclic aromatic amines and polycyclic aromatic hydrocarbons. Thus, NAT1 and NAT2, and GSTs gene polymorphisms may be involved in cigarette smoking-CRC risk[56,57]. A study demonstrated that individuals with fast acetylation capacity achieved by NAT1 and NAT2, may more efficiently activate heterocyclic aromatic amines, thereby increasing the induction of DNA damage and, consequently, increasing susceptibility to CRC. Besides, GST gene polymorphisms influence interindividual susceptibility to smoking-associated CRC, which can play an important role in the detoxification of colorectal carcinogenesis during smoking. A novel opinion is that smoking may increase cancer cell survival and induce some events associated to epithelial-to-mesenchymal transition (EMT) process. Smoking may reduce cell necrosis, deregulate Claudin-1 and E-cadherin expression and enhance the expression of miR-21 to induce EMT.
Aspirin: Abundant evidence indicates that regular use of aspirin is associated with a significant reduction in the incidence of CRC[59-62]. Not only that, the beneficial function of aspirin may be emphasized in some specific molecular subtypes of CRC. Several studies have indicated that regular use of aspirin is associated with better prognosis and clinical outcome in COX-2-positive and PIK3CA-mutated CRC[61,62]. Aspirin might inhibit the expression of COX-2 to reduce the prostaglandin (PG)E2 synthesis, and thereby to reduce the inflammatory response and suppress cancer cell proliferation and survival[59,61]. As to the status of PIK3CA mutation, pho-sphatidylinositol-3 kinase (PI3K) and the downstream Akt pathway can be activated to enhance COX-2 activity and PGE2 synthesis, resulting in inhibition of apoptosis in CRC cells. Therefore, aspirin can attenuate PI3K activity through inhibiting PGE2 signaling[59,61,63]. Meanwhile, aspirin might inhibit mTOR, a downstream effector of the PI3K pathway, by activation of adenosine monophosphate-activated protein kinase (AMPK) in CRC. In addition, aspirin may also inhibit Wnt signaling either directly or through down-regulation of PGE2 to suppress the onset of CRC.
Genomic level: Genomic instability is an essential feature that underlies CRC. There are three aspects to achieving genomic instability that can contribute to CRC: Chromosomal instability, MSI, and CpG island methylation. First, chromosomal instability, a common and efficient mechanism, can lead to the physical loss of tumor suppressor genes, such as adenomatous polyposis coli (APC), P53, and SMAD family member 4, and the activation of oncogenes, such as KRAS and PI3KCA[64,65]. These changes can transform the normal phenotype into a malignant phenotype[1,64,65]. Second, MSI can silence mismatch repair genes, such as MLH1, in patients with hereditary nonpolyposis colon cancer, who then have an even higher risk of developing CRC. Finally, the aberrant methylation of CpG islands has been demonstrated to result in the CIMP or CIMP-high, which accounts for 15% of CRC cases and exists in nearly all CRC tumors with aberrant methylation of MLH1.
In addition to genomic mutation, microRNAs (miRNAs) and long noncoding RNAs (commonly referred to as lncRNAs) are also expressed abnormally in CRC. Existing evidence indicates that miRNAs, such as miR-93 and miR-328, are aberrantly expressed in CRC and regulate the proliferation and metastasis of cancer stem cells[66,67]. MiR-200c can promote the EMT to induce proliferation and metastasis. Moreover, the up-regulation of a novel lncRNA, colorectal neoplasia differentially expressed (CRNDE), has been observed in the early stages of colorectal neoplasia (> 90%), except for CRNDE-d.
Modification of the signaling pathways: It has been demonstrated that tumor cells in CRC are maintained by the deregulation of specific signaling pathways. Genetic events are also part of a larger network that alters signal pathways, resulting in an increase in tumor cell proliferation and a decrease in tumor cell death. The onset and migration of CRC involves several signaling pathways, including the mitogen-activated protein kinase (MAPK) pathway, PI3K pathway, Wnt/β-catenin pathway, Janus-activated kinase/signal transducers and activators of transcription (JAK/STAT) pathway, 67 kDa laminin receptor (67LR) pathway, nuclear factor-kappa B (NF-κB) pathway, and nuclear factor-erythroid 2-related factor (Nrf2) pathway. Moreover, the crosstalk between pathways can promote the development and invasion of CRC and increase its resistance to drugs. The author details these specific pathways in the following section. Another novel signaling pathway, the Hippo pathway, is responsible for cell proliferation, differentiation, apoptosis, and tumorigenesis and exists in many malignant tumors, including CRC[72,73]. The Hippo pathway was initially defined as a tumor suppressor pathway, but its major effector, Yes-associated protein (YAP1), is viewed as an oncogene; therefore, the down-regulation of the Hippo pathway is connected to CRC initiation and progression[72,74]. It has been emphasized that the interaction between the Hippo pathway and the Wnt/β-catenin pathway is crucial in the development of CRC.
Cytokines: Chronic inflammation can promote the development of CRC. In chronic inflammation, immune cells such as lymphocytes, plasma cells, macrophages, and neutrophils infiltrate the colon and enrich ROS and reactive nitrogen species (RNS). In addition to exogenous mutagens, ROS and RNS can also cause DNA damage, which facilitates the initiation of cancer, as observed in a mouse model. Infiltrating inflammatory cells can also produce high levels of protumorigenic cytokines that drive tumor progression.
Cytokines are low-molecular-weight proteins that can mediate cell-to-cell communication and induce cell transformation and malignancy. Many proinflammatory cytokines, such as tumor necrosis factor (TNF)-α and interleukin (IL)-6, are involved in the creation of the tumor microenvironment. Furthermore, there are immunosuppressive cells in the tumor microenvironment that secrete vascular endothelial growth factor (VEGF), IL-6, IL-10, transforming growth factor-beta (TGF-β), soluble FasL, and indolamine-2,3-dioxygenase to promote the growth and metastasis of CRC by generating ROS and RNS, potentiating the EMT, and inducing angiogenesis[77,78]. Following the activation of these cytokines, some signaling pathways that can induce inflammation and stress are activated to induce the proliferation of cancer cells.
Some enzymes, such as Rab GTPase, can also have an essential function in the growth and metastasis of CRC, and they are correlated with cytokines. Rab GTPases, a large family of Ras small GTPases, play a crucial role in normal human physiology by controlling membrane identity and vesicle traffic. A study using a metastatic mouse model suggested that high Rab3C expression in patients might increase the migration and invasion ability of colon cancer as a result of IL-6 secretion and JAK2/STAT3 signaling pathway activation. Besides the Rab3C-IL-6-STAT3 axis, another Rab GTPase, Rab25, is also linked to CRC. The loss of Rab25 is associated with a poor prognosis in CRC because Rab25 can influence the trafficking and recycling of numerous key regulators of polarity and signaling that are involved in transformation.
Oxidative stress: Another factor that promotes the development of CRC is oxidative damage, which is characterized by elevated ROS levels and accumulated mutations that cause oxidative DNA damage. ROS, which includes superoxide (O2−), the hydroxyl radical (•OH), and hydrogen peroxide (H2O2), act as crucially important mediators in multiple cell signaling pathways. Sustained and excessive ROS are not only strongly correlated with the tumorigenic potential of cancer cells but also render cancer cells resistant to anticancer drugs. Studies have demonstrated that both gut microbiota and inflammation can cause oxidative stress. Gut microbiota can generate reactive metabolites and induce chronic mucosal inflammation. Inflammatory cells can mediate immediate cellular stress responses through the activation of NF-κB, signal transducer and STAT3, hypoxia-inducible factor-1α, activator protein-1 (AP-1), and Nrf2. Meanwhile, LPO, protein oxidation, nitric oxide (NO) production, enzymatic activity alteration and DNA damage can be mediated by oxidative stress, to injure cells, induce gene mutation, and influence signaling pathways and transcription factors.
Gut microbiota: Gut microbiota is viewed as a forgotten organ that participates as an essential contributing factor in the initiation and development of CRC. The balance of gut microbiota is conducive to the metabolism of nutrients, maintenance of the intestinal barrier, modulation of the immune system, and protection from pathogens. However, some bacterial species have been identified and suspected to play a role in colorectal carcinogenesis; these include Helicobacter pylori, Bacteroides fragilis, F. nucleatum, and so on. Dysbiosis is characterized by reduced Firmicutes to Bacteroidetes ratio (known as FIR/BAC), depletion of short-chain fatty acid-producing members of Lachnospiraceae and Ruminococcaceae, and the presence of putative pathobionts of oral origin. Dysbiosis contributes to increased mucosal permeability, bacterial translocation, and increased activation of components of the innate and adaptive immune system. These changes promote chronic inflammation and further downstream changes that promote colon carcinogenesis.
Gut microbiota can regulate some immune cells of the immune system to impact the development of CRC.
(1) T lymphocytes: Gut microbiota can exert an important effect on T lymphocytes to modulate the progression of CRC. On the one hand, gut microbiota plays an important role in triggering chemokines production, such as that of CCL3, CCL4, CCL5, CCL20 and CXCL10, ultimately leading to T lymphocyte recruitment in tumor tissues and improved prognosis of CRC. Bacteria-induced chemokine gene expression may also be initiated by Toll-like receptor (TLR) triggering on CRC cells. On the other hand, gut microbiota can regulate the differentiation of T lymphocytes. Different T lymphocytes can exert different effects on CRC. Th1 cytokine interferon gamma (IFNγ) plays an antitumorigenic role, whereas the Th2/Treg cytokines IL-4, IL-5, and IL-10 mediate a protumorigenic role. Besides, Th17 cells are known to be protumorigenic in CRC, and IL-17A is also linked to the gut microbiota. Gut microbiota depletion can increase numbers of antitumor IFNγ-secreting T cells and decrease numbers of protumor IL-17A and IL-10 secreting immune populations to reduce the development of CRC. Another study also demonstrated that a remodel of the gut microbiota can enhance anti-inflammatory capacity through promoting the induction of Tregs. In addition, F. nucleatum, a proinflammatory bacterial species in tumor tissue but rarely found in normal intestinal microbiota, is associated with increased lymph node metastases and a worse outcome in CRC patients[92,93]. F. nucleatum is likely to possess immunosuppressive activities through its inhibition of human T cell responses[94,95]. F. nucleatum has been shown to expand myeloid-derived immune cells, which inhibit T cell proliferation and induce T cell apoptosis in CRC[95,96]. F. nucleatum also expresses the virulence factor FadA on their bacterial cell surface, which has been shown to activate the Wnt signaling pathway and down-regulate the T cell-mediated antitumor immune response. Similarly, F. nucleatum can recruit proinflammatory cytokines, such as IL-17A, TNF, and CCL20, which induce inflammation and suppress immunity. Meanwhile, rats with depletion of gut microbiota also show an increase in cytotoxic T lymphocyte cells. Finally, a study demonstrated that fecal bacteria from CRC patients can up-regulate degranulation and cytotoxicity of CD8+T cells.
(2) B lymphocytes: The human gut homeostasis requires microbiota coated by both secretory immunoglobulin M (SIgM) and secretory immunoglobulin A (SIgA) emerging from B lymphocytes. SIgA deficiency will cause dysbiosis, which may promote the development of CRC. The study indicated that SIgM may emerge from pre-existing memory B cells and could help SIgA anchor highly diverse commensal communities to intestinal mucus. Meanwhile, IL-33 might participate in modulating the IgA-microbiota axis to prevent IL-1α-dependent colitis and tumorigenesis. IL-33 can promote IgA production to maintain gut microbial homoeostasis and inhibit IL-1α-mediated inflammation to prevent the onset of CRC. Similarly, bacteria in CRC can also induce the production of IL-17, which promotes influx of intratumor B cells that promote tumor growth and progression.
(3) Natural killer (NK) cells: Some certain bacteria may favor recruitment of immune cells such as NK cells other than T cells, to achieve a favorable prognosis. NK cells and CD8+T cell crosstalk in the tumor microenvironment may benefit patient outcome. Nlrp3 inflammasome components exacerbate liver CRC metastatic growth by impairing IL-18 signaling and further impacting maturation of hepatic NK cells. In addition, Nlrp3 activation might be mediated by a microbial ligand derived from the remaining intestinal microbiota.
(4) Neutrophils: Neutrophils are also believed to modulate growth of colon tumors, and correlate with outcomes of patients with colon cancer. It has been indicated that neutrophil depletion is correlated with increased numbers of bacteria in tumors and proliferation of tumor cells, and an inflammatory response mediated by IL-17, thereby inducing the development of CRC.
(5) Eosinophils: Eosinophils in CRC patients are strongly linked with a decreased disease risk, better prognosis, and extended patient survival. Dysbiosis might impair eosinophil-driven responses to promote the development of CRC. However, the specific mechanism is not clear.
(6) Macrophages: Macrophages are also involved in the development of CRC. Monocytes/macrophages may polarize as M1 or M2 cells. Overall, M1 macrophages display a pro-inflammatory potential mediating antitumor activities, while M2 macrophage display an anti-inflammatory promoting cancer cell growth[106,107]. In the tumor microenvironment, tumor-associated macrophages undergo polarization into M1 and M2 phenotypes. The specific interaction of gut microbiota and macrophages on CRC is still required to investigate further. Some studies have provided insights into the relative mechanism. A metastasis-related secretory protein, cathepsin K, activated by the imbalance of intestinal microbiota, stimulates CRC progression through accelerating M2 polarization of tumor-associated macrophages through a TLR4-mTOR-dependent pathway. Besides, another study also indicated that defects in the subepithelial band of lamina propria-indigenous macrophages barrier in inflammatory bowel disease encourage the trespassing of the gut microflora into the host, thereby destabilizing host immunity and promoting the development of CRC. High amounts of F. nucleatum intratumorally are correlated with increased macrophage infiltration and CDKN2A promoter methylation in MSI-H CRC. Although it can be hypothesized that the repression of CDKN2A via promoter methylation may be connected with the increased M2 macrophages in F. nucleatum-high CRC, the M2 macrophage density was not significantly associated with F. nucleatum status in MSI-H CRCs, as displayed by the study. Besides, a strong association between lower frequency of macrophages, increased Firmicutes, and decreased tumorigenesis was also observed in CRC.
(7) Dendritic cells: Dendritic cells play critical roles in maintaining tolerance and immune homeostasis in the gut. And, some species in the gut can also induce dendritic cell maturation and the induction of Tregs and IL-10 production to regulate tumorigenesis. Overall, the specific immune-microbiota mechanism needs to be investigated by more animal studies and epidemiological studies before it is proven.
Molecular pathological epidemiology: Molecular pathological epidemiology (MPE) has emerged as an integration of molecular pathology and epidemiology, to address the need to investigate the inherent heterogeneity of pathogenic processes even for a single disease entity[113,114]. Overall, MPE discusses the interrelationship between exogenous and endogenous factors, tumoral molecular signatures, and tumor progression. On the one hand, MPE can uncover potential risk factors that are not detectable in conventional epidemiological research without using molecular pathology methods. On the other hand, MPE can help us refine the association between exogenous or endogenous factors and validate specific etiological hypotheses, thereby augmenting causal inference[113-116]. Meanwhile, MPE study can provide novel etiologic and pathogenic insights, potentially contributing to precision medicine for personalized prevention and treatment[116,117]. In addition, MPE can also integrate several disciplines to evolve subfields of MPE, including pharmaco-MPE, immuno-MPE and microbial MPE, to provide novel opinions into underlying etiologic mechanisms.
Some progression has been made in CRC. The MPE research has determined the strength of the association for between the exposures and the specific subtypes of CRC, which can help to establish causality and speculation on the relative mechanism of exposure acting on CRC. A MPE study has demonstrated that both obesity and physical inactivity are associated with a higher risk of CTNNB1 (β-catenin)-negative CRC but not with CTNNB1-positive cancer risk. Hence, the study implied that energy balance and metabolism status might exert impact on the development of CRC independent of WNT/β-catenin activation. Then, pharmaco-MPE, integrating MPE into pharmacoepidemiology, will play a vital role in identifying target individuals who will most likely benefit from use of a particular drug, clinically. MPE studies have demonstrated that regular use of aspirin can reduce the risk of CRC with overexpression of COX-2 but not of that with weak or absent expression of COX-2[119,120]. Many pharmaco-MPE studies have shown that regular aspirin use was associated with lower risk of BRAF-wildtype and PIK3CA-mutated CRC but not with BRAF-mutated and PIK3CA-wildtype CRC[121,122].
Immuno-MPE, the integration of immunology and MPE, can mainly discuss exposures impacting CRC through regulating the immune system and disease-immune interactions. An MPE research project has revealed that the association of aspirin use with CRC survival is stronger in patients with the programmed cell death ligand 1 (PD-L1)-low tumors than the PD-L1-high CRC. It indicated that PD-L1 expression might serve as a biomarker that predicts resistance to aspirin use. In addition, microbial MPE is also studied in CRC. Typically, a high level of F. nucleatum might be associated with molecular features of CRC, including MSI-high and CIMP-high[124,125]. Meanwhile, another MPE study demonstrated that a greater amount of F. nucleatum was associated with a lower density of CD3+T cells in CRC, indicating that the interaction of target microbiota and immune system should be discussed further for CRC prevention and precision treatment. In addition to F. nucleatum, other components of gut microbiota need to be investigated in the future.
Although the MPE has many strengths, the pitfalls and challenges should be considered. Challenges in MPE mainly include sample size selection, need for rigorous validation of molecular assays and study findings, and paucities of interdisciplinary experts, education programs, international forums, and standardized guidelines. In addition, MPE research needs to face the issue of multiple hypothesis testing, so it is necessary to form a priori hypotheses based on earlier exploratory findings or on potential biological mechanisms. Similarly, MPE also may create a higher chance of yielding spurious findings.
Other mechanisms: The EMT and the Warburg effect (WE) are considered to be involved in tumor metastasis. EMT occurs when polarized epithelial cells lose their adhesion property and obtain mesenchymal cell phenotypes as a result of the loss of membrane E-cadherin expression. Although the exact mechanism in CRC is not clear, EMT-related molecular mechanisms have been described, including the activation of many signaling pathways, such as the TGF-β/Wnt pathway and PTEN/Akt/HIF-1α pathway, as well as many activated genes, such as APC and Akt. The WE is the result of pyruvate being directed away from the tricarboxylic acid cycle and metabolized to lactate, resulting in a buildup of glycolytic intermediates. Briefly, the WE is the process of aerobic glycolysis. Many mechanisms can inhibit the WE in CRC to affect the metastasis of cancer; these include some miRNAs[130,131], such as miRNA-98 and Pim1. Gas1, a tumor suppressor, can inhibit both EMT and the WE in CRC through AMPK activation and the mTOR pathway.
Because of the drug resistance and side effect that can arise in targeted therapy of CRC, studies have investigated treatments that involve natural bioactivate materials found in various foods, such as tea polyphenols (TPs)[134,135]. We have learned that TPs may become a novel medicine to prevent and treat disease with fewer side effects than traditional medicines. Also, the combination of chemotherapeutic drugs and TPs could synergistically enhance treatment efficacy and reduce the adverse side effects of anticancer drugs. Over the past several decades, we have learned that TPs can be utilized effectively as chemopreventive and chemotherapeutic agents for some diseases, including obesity, diabetes mellitus, Alzheimer’s disease, Parkinson’s disease, cardiovascular disease, and cancers. TPs can play an essential role in the treatment of most cancers by causing G0/G1 phase cell cycle arrest and inhibiting angiogenesis[143,144]. One study demonstrated that green TPs could suppress the pathological formation of new blood vessels by inhibiting members of the VEGF family.
Chemical structure of TPs: Tea, which originates from the plant species Camellia sinensis, has become the second most commonly consumed beverage following water, with teas such as green tea, black tea, and oolong tea among those frequently consumed. The main difference between these three kinds of tea is the fermentation level, which leads to the presence of different TPs. Green tea is made from dry tea leaves, which do not undergo the process of fermentation. Thus, green TPs contain more oligomeric polyphenols, with the main content comprising flavan-3-ols or tea catechins (approximately 59%), including (-)-epigallocatechin (EGC), (-)-epigallocatechin-3-gallate (EGCG), (-)-epicatechin (EC), and (-)-epicatechin-3-gallate (ECG) (Figure 2), among which EGCG is the most abundant polyphenol in green tea[145-147]. Although gallic acid (GA) can also be found in green tea, it is mostly contained in the fully fermented Pu-erh tea, usually. Black tea must undergo high or full fermentation, and the level of fermentation of oolong tea falls in the middle of this range[136,149]. Therefore, black tea contains lower monomeric polyphenol content (3%-10% of solids) and higher concentrations of polymeric polyphenols (23-25% of solids), such as theaflavin (TF), theaflavin-3-gallate (TF2a), theaflavin-3’-gallate (TF2b), theaflavin-3,3’-digallate (TF3 or TFdiG), and thearubigin (Figure 3)[147,150]. Oolong tea polyphenol content includes epitheflagallin (ETG) and EGCG, among others.
Figure 2 The structures of green tea polyphenols, including (−)-epigallocatechin-3-gallate, (−)-epicatechin-3-gallate, (−)-epigallocatechin, catechin, (−)-epicatechin.
Figure 3 The structures of black tea polyphenols, including, theaflavin, theaflavin-3-gallate, theaflavin-3'-gallate and theaflavin- 3,3'-digallate.
Oligomeric and polymeric polyphenols undergo mutual transformation. The formation of black TPs involves two steps: Oxidation and polymerization, which are regarded as the fermentation of green tea. In the first step, catechins are partially oxidized to quinones as a result of the enzymatic catalysis of polyphenol oxidase or peroxidase, which exist in nature. Subsequently, polymerization produces gallocatechin quinones, and further oxidation and rearrangement lead to the synthesis of the core of black TPs, namely, benzotropolone. For instance, EC and EGC form TF1, ECG and EGC form TF2a, EC and EGCG form TF2b, and ECG and EGCG form TF3. Therefore, the chemical structure of black TPs and green TPs share some similarities. All oligomeric polyphenols have the same basic chemical structure of two aromatic rings (A and B) linked by three carbons that usually form an oxygenated heterocycle (C ring), which consists of a C6–C3–C6 skeleton[152,153]. In the B ring, OH or OCH3 groups usually occupy up to three positions. In flavan-3-ols, the C ring, as the activated center, is a saturated heterocycle with a hydroxyl group that provides different arrangements of hydroxy, methoxy, and glycosidic groups and bonds with other monomers. Moreover, the chemical structures of TPs are not simple linear oligomers because they contain gallate groups.
Bioavailability of TPs: After tea is consumed, TPs can be decomposed into different fractions and absorbed in the gut, which is considered to be a complex physiological process. Bioavailability is used to describe the extent of absorption that an ingested compound is released from food, and its fate in the organism. Furthermore, many studies have investigated the kinetics and extent of polyphenol absorption by measuring plasma concentrations and/or urinary excretion after the ingestion of TPs. A number of studies have demonstrated that TPs have poor bioavailability from in vivo and in vitro gastrointestinal digestion[154,156]. In nature, most flavan-3-ols undergo epimerization and exist as stereoisomers in a cis or trans configuration [(-)-epicatechin or (+)-catechin, respectively]. Different stereoisomers have different bioavailability. The bioavailability of the stereoisomers has been ranked as (-)-epicatechin > (+)-epicatechin = (+)-catechin > (-)-catechin. Also, the in vivo effects of flavan-3-ols, major components of green TPs, rely on their absorption and metabolism in the gastrointestinal tract. Thus, we should discuss the health effects of not only TPs but also their metabolites.
The different metabolites can be found in the small intestine and large intestine. The absorption and metabolism of TPs, including the processes of methylation, glucuronidation, and sulphation, mainly transpire in the small intestine. Among the TPs, EGCG is the only known polyphenol present in plasma with a large proportion (77%-90%) in the free form. Others are highly conjugated with glucuronic acid and/or sulfate groups, such as epicatechin-3’-glucuronide and 4’-O-methylepicatechin-3’-glucuronide, among others, after being metabolized. One study observed that the conjugated forms of two major phenolic catabolites, (-)-5-(3’,4’,5’-trihydroxyphenyl)-gamma-valerolactone (M4) and (-)-5-(3’,4’-dihydroxyphenyl)-gamma-valerolactone (M6), which accounted for up to 40% of the amount of ingested pure EGC and EC, could be detected in plasma, urine, and feces. 4’,4’’-di-methyl-EGCG was also detected in human plasma and urine following green tea ingestion. Although the metabolism of black TPs has been researched less, the metabolites might contain 3-methylgallic acid, 4-methylgallic acid, and 3,4-di-methylgallic acid, which also exist in green TPs.
In the large intestine, the metabolic fate of TPs after in vitro gastrointestinal digestion was studied, and bioaccessibility activity of TPs was shown to be higher in the colon than in the duodenum, suggesting that, in vivo, the gut microbiota might be able to metabolize dietary polyphenols, resulting in an increase in their beneficial effects in the large intestine[136,161]. The study demonstrated that green tea catechins were more bioavailable when colonic ring fission metabolites were taken into consideration. A possible ring-fission metabolite, (-)-5-(3’,5’-dihydroxyphenyl)-γ-valerolactone (M6’), was detected in human urine after green tea ingestion. The fraction of flavan-3-ols that is not absorbed in the small intestine reaches the large intestine, where it can undergo several microbial processes that finally lead to smaller molecules that can be absorbed and reach the liver and, subsequently, the systemic circulation. Studies have indicated that catechin and epicatechin, which can enter the portal vein at a relatively high concentration as a result of ileal transfer, can be further metabolized to methylated and glucuronidated forms by phase I and II metabolism in the liver. From the action of microbiota, ingested flavan-3-ols can be converted to C6-C5 phenylvalerolactones and phenylvaleric acids, which undergo side-chain shortening to produce C6-C1 phenolic and aromatic acids that enter the bloodstream and are excreted in urine.