• DOI:10.1002/ijc.22314
• Corpus ID: 116669938

Streptococcus bovis: Causal or incidental involvement in cancer of the colon?

@article{Hausen2006StreptococcusBC,
  title={Streptococcus bovis: Causal or incidental involvement in cancer of the colon?},
  author={Harald Hausen},
  journal={International Journal of Cancer},
  year={2006},
  volume={119}
}

• H. Hausen
• Published 1 November 2006
• Medicine, Biology
• International Journal of Cancer

In this issue, Tjalsma and colleagues from the University of Nijmegen Medical Center report on the detection of immune reactions against Streptococcus bovis antigens in sera of 11 out of 12 colon cancer patients and in 3 out of 4 patients with colon polyps. No positive reaction was observed in 8 control subjects. They used immunocapture tandem mass spectrometry to identify an immune reaction against bacterial antigens and found that one of the diagnostic antigens represents a surface-exposed heparin-binding protein that, according to their speculation, might be involved in the attachment of S. bovis to tumor cells. They claim that profiling of the humoral immune response against S. bovis infections may represent a promising diagnostic tool in early detection of human colon cancer. In spite of the still relatively low number of sera analyzed, these data are of substantial interest since they stress the possibility that this immune reaction may serve as a sensitive marker for early detection of colon neoplasias. A large number of previous reports point to a close linkage between S. bovis infections and tumors of the human colon. Initially, a paper by McCoy and Mason appeared in 1951, suggesting a relationship between cancer of the colon and enterococcal endocarditis. The coincidence of S. bovis septicemia and S. bovis endocarditis was noted in several subsequent publications, first reported in 1974 and confirmed shortly thereafter. S. bovis endocarditis is considered to be a clinical indication for colonoscopy since the majority of these patients present with either colon carcinomas or colon polyps. The same holds for bacteremias caused by S. bovis. In the presence of a carcinoma of the colon, the relative risk for developing S. bovis-linked endocarditis is in the range of 3–6%. Conversely, 65–70% of patients with S. bovis endocarditis simultaneously revealed not previously recognized malignant gastrointestinal tumors. The high percentage of colon cancer found in patients with S. bovis endocarditis or septicemia has been explained by the fact that these bacteria can easily penetrate the blood stream in ulcerating carcinomas of the colon. This interpretation, however, finds little support by data demonstrating a high risk of S. bovis endocarditis and/or septicemic patients for nonulcerating adenomas. In addition, patients with a history of S. bovis endocarditis with no colonoscopic abnormalities at that time developed colon tumors 2–4 years later. (This again does not seem to support such an interpretation.) As shown by Tjalsma et al., it is unlikely that the acquisition of serum antibodies to S. bovis antigens originates from antigen presentation due to bleeding from microlesions since antibody titers against lipopolysaccaride antigens of Gram-negative bacteria prevalent in the colon did not differ between colon cancer patients and controls, as also previously reported by Darjee and Gibb. Some preliminary findings suggest an active role of S. bovis in the promotion of intestinal carcinogenesis: adult rats treated with azoxymethane for 2 weeks and which subsequently received either injections with S. bovis bacteria or wall-extracted antigens twice weekly. The authors observed progression of preneoplastic lesions, enhanced expression of proliferation markers and increased production of interleukin-8 in the colonic mucosa in these rats. The same group used a partially purified S. bovis S300 fraction representing 12 different proteins and triggered the synthesis of proinflammatory proteins (human interleukin-8 and prostaglandin E2, correlated with the in vitro overexpression of cyclooxygenase2 [Cox-2]) in human colon carcinoma cells (Caco-2) and in rat colonic mucosa. These data could point to a role of oxygen radicals in colon carcinogenesis induced by a chronic infection with S. bovis. The mechanism could be similar to the one suspected for the development of gastric carcinomas after persisting Helicobacter pylori infections. Presently it would still be important to know whether the increased presence of S. bovis in colonic cancers and polyps results from the preferential bacterial colonization of these cancers and their precursors or whether S. bovis represents a carcinogen that is causally involved in colon cancer. Obviously, a number of questions are still open, as presented clearly in the discussion of Tjalsma and his colleagues. The available data do not prove there is a causal relationship of this streptococcal infection for colon cancer. They should, however, encourage further research in this area. The molecular events initiating sporadic colon cancer development are still poorly understood, although a number of well-defined molecular characteristics in the course of tumor progression have been elucidated. A high level of red meat consumption has been defined as a risk factor for colon carcinogenesis. A role of S. bovis in this malignancy would provide an exciting explanation for the reported risk increase: S. bovis is one of the contaminants frequently detected in commercially available meat. Similarly, the overexpression of Cox-2 in colon cancer could be triggered by S. bovis infections. Thus, further studies of a possible etiologic relationship between S. bovis and related bacterial infections and colon cancer are highly desirable. Irrespective of a possible role of S. bovis in cancer development, however, the seroresponse to these bacterial antigens seems to represent a sensitive marker for the early diagnosis of colon neoplasias. This in itself should have important clinical implications.