Agents associated with colorectal cancer 4

CHAPTER CONCLUSIONS Colorectal cancer, like gastric cancer, is one of the leading neoplastic diseases worldwide. However, unlike gastric cancer, its development occurs in a highly complex and poorly characterized bacterial environment. This is because the colon contains between 109 to 1013 bacteria per gram of luminal contents and the composition of colonic microbial flora is highly heterogeneous and unpredictably influenced by factors such as age, diet, and geography. By using four selected gene probes and PCR-ISH, this study is able to show that specific species of bacteria within the complex intestinal flora can indeed play an active part in cancer cell formation. Intestinal E.coli can be pathogenic when the right conditions prevail. The presence of E. coli has been linked to inflammatory bowel diseases [Hooper LV and Gordon JI, 2001] and colorectal cancer [Swidsinski A et al., 1998]. On the other hand, the E.coli bacterium has also been shown to give protection against colon cancer [Carrithers SL, 2003; Pitari GM et al., 2003]. It is precisely this commensal relationship between bacterium and host that made it difficult to know if some of them indeed played a role in the onset of cancer and if so how did they so in the insidious way that escapes early detection of cancer risk. The detection of genetic information by gene probe-related technology will be the first step to quickly determine if the E.coli bacterium is associated with colorectal 125 cancer. This thesis aimed to address issues related to the bacterium’s mode of action in its host, to obtain some insight as to whether it has an active role in colorectal cancer formation or it merely remains a passive symbiont in the colon and rectum. This study addressed two areas of interest, namely, cancer prevention and early diagnosis. The 81B sequence is a unique sequence that is specifically found in a vast majority of the E.coli strain, including those possessing haemolytic properties (as observed on blood agar, such as the O157:H7), toxigenic genes, invasive genes, adherent genes, cytonecrotizing genes and those that other methods failed to detect. For example, the enterohemorrhagic E.coli serotype 0157:H7 which can be detected by 81B sequence, lacks the B-D-glucuronide enzyme and therefore cannot be detected by the MUG (4-methylumbelliferyl-BD-glucuronide) method, as this MUG method requires the bacteria to be able to hydrolyse MUG to a fluorescent end product. Additionally, 81B sequence is found in abundance in localized and metastatic cancer cells. Therefore, to discover this unique 81B sequence is an asset, as it can be used in an assay to improve the standard of food safety and sanitation that is important to prevent food contamination and for cancer prevention. It can also be useful as a biomarker for early detection of tissue infection and for the detection of colorectal cancer cell. The data derived from the use of the 81B sequence has helped in the identification of the other three bacterial determinants (Tia, HRA1 and CNF1). All 126 this information will lead to a better understanding on how bacteria may have contributed to cancer cell formation. Bacteria isolated from fecal and tissue specimens of subjects who are normal volunteers or those with pre-malignant (adenomas) and malignant (carcinoma) tumors, were found to possess the adherent (HRA1) or invasive gene (Tia) (with or without the cytotoxic necrotising factor (CNF1) gene) but not the other eae genes (which are specific for both enteropathogenic E.coli and enterohemorrhagic E.coli ) or for enterotoxins (heat-labile enterotoxin, LT; heatstable enterotoxin-a ,STa; heat-stable enterotoxin-b, STb; Shiga-like toxin-I, SLT1; Shiga-like toxin-II, SLT II) of enteropathogenic, enterohemorrhagic or enterotoxigenic E.coli as shown in the preliminary study (see Chapter 3). The above findings suggest that one is dealing with cryptic but potentially harmful bacteria as they are capable of attaching, invading and transforming the mucosal cells without triggering the relevant immune/host response that can be evoked by the enterotoxigenic determinants. What is more interesting is that the test to look for the genes for Tia, HRA1 and CNF1 are not routinely carried out in monitoring food safety and sanitation. This is important knowledge because horizontal transmission of pathogenic and virulent genes between E coli bacterium will make it difficult to assign a particular strain of bacteria, such that their detection can be overlooked by some conventional assays. What this may imply is that we are subjected to food and water sanitation standards that are not protective 127 against cancer-causing pathogens. This further suggests that risk of cancer can be reduced or prevented by improving on the standard of food and water safety measures through implementation of relevant diagnostic assays with a focus on newly emerging or newly identified strains of pathogenic bacteria involved in enteric diseases. At this juncture, it is worthwhile to mention that, by knowing that bacteria with Tia, HRA1 and CNF1 can be found in cancer cells but not in the majority of normal cells is much more informative than just knowing that E.coli ribosomal genes are associated with cancer [Swidsinski A et al., 1998]. The identity of the bacterial determinants such as the Tia, HRA1 and CNF1, is important for they give a better insight into how bacteria may be involved. The clues to their active involvement come from studies carried out by the laboratories that identified the determinants. E.coli pathogenicity is complex as its virulence varies according to the combination of factors it expresses in response to its milieu and the clinical response from its host. Bacteria will need at least attachment factors, host cell surface modifying factors, invasive factors, factors that affect or disrupt host structural/cellular equilibrium and also other yet to identified factors that it can secret, export and pilot them to the target host cells. How bacteria may be involved in gastrointestinal disease can be deduced from the properties of the three determinants, Tia, HRA1 and CNF1. 128 It has been shown that the Tia locus of ETEC strain H10407 is capable of adhering to and invading cultured human intestine epithelial cells [Elsinghorst EA, 1992; Fleckenstein JM, 1996; Mammarappallil JG et al., 2000] while those bacteria with the HRA1 gene has been shown to mediate attachment to cells derived from intestinal epithelium (human colonic adenocarcinoma, COLO 201). Many bacteria that cause intestinal diseases need to adhere to the gut mucosa, and adhesion of pathogenic E.coli with HRA1 is resistant to D-mannose [Lutwyche P et al., 1994 and 1995, Srinivasan U et al., 2003]. On the other hand, CNF1 protein produced by uropathogenic and some other Escherichia coli strains has been shown to induce stress fiber formation in cell cultures by activating Rho [Fiorentini C et al., 1988] and is involved in various signaling processes, including secretion, phagocytosis, endocytosis, cell cycle progression, transcriptional activation and tranformation [Zheng Y et al., 1995; Olson MF, 1996; Sahai E et al., 2001]. CNF1 can also by having a direct action on host microfilamental network, cause normal cultured cells to become giant multinucleated cells. Treatment of Hep-2 cells with CNF1 has been shown to enable them to be able to internalize latex beads and non-invasive bacteria and in certain circumstances, to allow the bacteria to multiply intracellularly [Falzano L et al., 1993]. CNF1 is also shown to cause significant changes in transepithelial resistance and in the permeability of epithelial monolayers [Gerhard F et al., 1998]. Last but not least, CNF1 is also found to be able to hinder apoptosis in epithelial cells by modulating the expression of the Bcl gene family [Fiorentini C et al., 1998] and the Akt/Iқb kinase pathway [Miraglia AG et al., 2007]. 129 One way to know how these determinants can be related to cancer development will be to study its prevalence in fecal matter and to find out the depth and pattern of distribution of these molecular markers within the mucosa as they evolve from pre-malignant to malignant stages of colorectal cancer. To address this issue, a controlled prospective clinical study was carried out as described in Chapter 4. The culture results from the fecal and tissue specimens showed a high prevalence of HRA1-postivie E.coli among the cancer subjects and subjects with adenomas, when compared with normal subjects and those with hyperplasia. Similar prevalence patterns were also observed in the fecal specimens collected from cancer subjects and subjects with adenomas with CNF1-positive E.coli. Tia-positive E.coli alone, however, was not shown to have a significant prevalence, except, when it appeared together with HRA1-positive E.coli, suggesting a possible synergic relationship at the earlier stage of attachment and invasion into the mucosal cells of the colon tissue (see chapter 4). The culture results for the four time points fecal collection study also showed a persistent prevalence HRA1-positive E.coli on more than one time points of fecal specimens collected from subjects with adenomas (p[...]... bacteria load with adhesive/invasive fimbrial and the virulent bacterial proteins that can alter normal cell physiology/morphology resulting in cell transformation Cell turnover and cell selection is ongoing in the presence of mutagenic bacteria This will be the first event that underlies colorectal cancer risk A classic example will be the case of the Helicobacter pylori (HP) and gastric cancer as one... can be instituted without selection for resistant strains How useful and safe are antibiotics when the bacteria can be out of reach or can even be in form not treatable with antibiotics Is vaccination going to be the answer? At this stage, there are many questions for the management of patients at risk However, by identifying E.coli as one of the potential etiological agents of colon cancer or simply... if they are screen out or treated without concomitant decrease in incidence rate thereafter The important question now is can one afford to ignore the association and the in vitro studies? What determinants should be considered worthwhile for screening? Should those adherent/invasive bacteria with the potential to pick up other virulent traits (secreting proteins with in vitro proven transforming capabilities,... diarrhoea or there is frequent intake of probiotics in the form of yogurt drinks with healthy Lactobacillus species and Bifidobacterium species [Reddy BS, 1998; Hirayama K & Rafter J, 2000; Rafter J, 2002; Carrithers SL, 2003] There is the other matter of high association of meat consumption with increased risk of colon cancer Can it just be a matter of by-products of fat metabolism and bacteria by-products... Tia, HRA1, CNF1 that exist in great abundance) or their products inside the tumor cells, can be a more useful approach to the treatment of colorectal cancer while sparing normal tissue from toxicity 139 This study does not exclude the role of other bacteria in colorectal tumorigenesis In fact, one is highlighting that tumorigenesis is not exclusive to viruses and that bacteria too may have ‘oncogenic’... be persistently present in the gut as they are not eliminated by the flushing motion that is caused by diarrheagenic bacteria The PCR-ISH data in our study (Chapter 4) indicated progressively higher bacteria load and persistent presence with gradual change in cell morphology (normal to metastatic cells) Dynamic self propagating event can happen as more bacteria product enhanced permeability which allowed... simply associating the presence of the bacteria in great abundance inside the tumor cells as shown in our PCR-ISH results, one is a step nearer to being able to reduce the morbidity and mortality associated with it The information observed in this study can be useful in preventive measures which include improvement in food safety and water sanitation test, surveillance, production of vaccines, early...At the genetic level, cancer can be caused by genetic recombination between bacteria and eukaryotic host [Fernando de la Cruz, 2000] Although this method of cellular transformation across these two biological kingdoms (i.e the bacteria and the animal kingdom) has not been noted in vivo, it has happened between the bacteria and the plant kingdom One known cancer- causing bacterium is Agrobacterium... and deeper by looking at their effects when the bacteria are fed to animals and to see how the knowledge gained from this study can be applied to the early diagnosis, prevention and therapy of the colorectal cancer The marker gene sequences identified in this study, including the 81B, HRA1, CNF1 and Tia, can also be useful to look at other conditions where E.coli has been implicated There are receptors... to chance upon encountering a virulent strain, when one’s immune system is compromised, to experience the effect One will be wondering how can such incidence rate remained relatively low, if the silent cancer causing virulent E.coli, which are not particularly screen out during food safety check, are still present as public contaminant for consumption It is important to emphasize that gram-negative bacteria . bacteria with Tia, HRA1 and CNF1 can be found in cancer cells but not in the majority of normal cells is much more informative than just knowing that E.coli ribosomal genes are associated with cancer. technology will be the first step to quickly determine if the E.coli bacterium is associated with colorectal 126 cancer. This thesis aimed to address issues related to the bacterium’s mode of. 125 CHAPTER 5 CONCLUSIONS Colorectal cancer, like gastric cancer, is one of the leading neoplastic diseases worldwide. However, unlike gastric cancer, its development occurs in a highly