Fusobacterium and Colorectal Cancer: Important Association or Random Coincidence?

Abstract

The role of the microbiome in human cancer has become an area of intensive research and controversy. Many reports have highlighted the physical association of Fusobacterium with colorectal cancer (CRC). This association has provided diagnostic and therapeutic promise but has also given rise to several controversies regarding the influence of Fusobacterium species on human CRC. Here we discuss two areas of controversy surrounding this emerging pathogen: the influence of Fusobacterium on CRC proliferation and the effect of Fusobacterium on the immune microenvironment of CRC.

Introduction

The bacterial genus Fusobacterium includes diverse species and subspecies characterized by nonmotility, a spindle morphology, and an inability to grow in the presence of oxygen. Originally studied as an organism of the mouth, fusobacteria are of increasing medical interest due to recent findings that they are associated with colon cancer.

Two independent genomic analyses originally identified fusobacteria as enriched in colon cancer [1,2]. Kostic and colleagues applied a computational subtraction pipeline, PathSeq, to whole genome DNA sequences from nine colorectal cancers and matched normal colon tissue and used linear discriminant analysis to identify enriched species in the colon cancer DNA; F. nucleatum, F. mortiferum, and F. necrophorum enrichment in the colon cancers was observed and then validated in a larger cohort of 95 paired specimens by 16S rDNA sequencing [2]. Castellarin and colleagues identified Fusobacterium RNA by sequencing total RNA from eleven colorectal cancer and matched normal patient samples and assessing for transcript enrichment in the cancer samples [1]. They also cultured one isolate directly from the tumor samples and identified it as F. nucleatum [1].

Subsequently, multiple studies have confirmed the statistically significant enrichment of Fusobacterium spp., primarily F. nucleatum, in colorectal cancers compared with matched non-cancerous tissues from patients of several countries. Given the weight of evidence from multiple studies, there appears to be little doubt that some colorectal cancers harbor Fusobacterium, and that Fusobacterium species are enriched in colorectal cancer tissue, on the population level. This association raises the question: Does Fusobacterium promote tumorigenesis or cancer progression, does it interfere with the cancer, or does it just find a growth benefit to dwelling in the tumor microenvironment? Here we focus on two areas of controversy: the impact of Fusobacterium species on cancer cell growth and the impact of fusobacteria on the immune microenvironment of the cancer.

Does Fusobacterium promote primary tumor formation and CRC cell line proliferation?

With the link between Fusobacterium and colorectal cancer confirmed by several groups using metagenomic methods, apparent questions arose regarding the relevance of this observation to the involvement and progression of tumorigenesis. Daily gavage of an invasive human F. nucleatum strain, isolated from a Crohn’s disease patient, into Apc^Min/+ mice resulted in a significantly higher number of colonic tumors as compared to mice fed with colon commensal Streptococcus spp [3]. Interestingly, Apc^Min/+ and Apc^Min/+; Il10^−/− mice colonized with F. nucleatum isolates from CRC patients did not promote intestinal tumorigenesis in the presence or absence of complex biota [4]. A limitation to utilizing in vivo mouse models is the variability in Fusobacterium colonization. Fusobacterium is a heterogenous species and strain diversity should be taken into consideration, as not all F. nucleatum strains will be able to colonize mice. For example, recently Queen et al. sought to develop a stable colonization mouse model which closely simulates the condition in humans. However, when inoculated weekly, CRC-derived F. nucleatum did not consistently colonize the gut of specific-pathogen-free (SPF) C57BL/6J wild type mice [5]. As suggested by the authors, the failure to establish a stable colonization for colon tumorigenesis could be the fact that mice are not a natural host for F. nucleatum.

Additional in vivo studies have sought to establish a tumorigenic role for F. nucleatum. Yang et al. found that infection of CRC cells with F. nucleatum strain 25586 increased formation of xenograft tumors in BALB/c nude mice [6]. Yet, when wild-type F. nucleatum strain 12230 was injected into HCT116 xenografts, no tumor growth was reported, only abscess formation was observed within 3–5 days [7]. Bullman et al. have provided antibiotic evidence for a role for Fusobacterium in growth of CRC, where metronidazole treatment of a xenograft model, derived from a CRC patient, reduced tumor growth [8]. Most recently, aspirin treatment was shown to reduce colonic adenoma growth in an Apc^Min/+ model infected with F. nucleatum [9].

Along with in vivo models, investigations have utilized CRC-derived cell lines cocultured with F. nucleatum to determine the potential role in cell proliferation. Studies by Rubinstein and colleagues initially reported that F. nucleatum strain 12230 stimulated proliferation of human CRC cell lines HCT116, DLD1, SW480, HT29, but weakly impacted RKO cell lines [7]. Another independent group has reported that F. nucleatum strain 25586 promoted cell growth of HCT116 and LoVo CRC cell lines [6]. These intriguing results will benefit from a more comprehensive approach utilizing multiple Fusobacterium strains, a large collection of CRC cell lines and additional cell proliferation assays to provide a clearer assessment of Fusobacterium impact on CRC cell line growth.

Thus, in our evaluation, it is still not clear whether Fusobacterium promotes the growth of either CRC cell lines or CRC tumors. Depending on the strain, F. nucleatum does not always induce tumor formation in vivo, and in vitro studies have only tested a limited number of CRC cell lines. Further studies with a variety of Fusobacterium strains and a range of cell and animal models are needed to assess the generality of reports that Fusobacterium can directly promote CRC growth in vivo or in vitro.

Does Fusobacterium affect the immune contexture of the tumor microenvironment?

In addition to a potential direct impact on CRC proliferation, various studies have proposed that Fusobacterium is associated with alterations in tumor immunity in colorectal cancer. Introduction of F. nucleatum to Apc^Min/+ mice was found to be associated with accelerated colonic tumorigenesis, infiltration of specific myeloid cell subsets into tumors, and an NF- κB proinflammatory signature [3]. Furthermore, this inflammatory gene signature was shared with human CRC tissue with a high Fusobacterium abundance [3]. Several studies have proposed that Fusobacterium may promote colorectal carcinogenesis through its suppressive effect on antitumor immunity, including both lymphocytes and myeloid cells. However, there is some inconsistency in that Fusobacterium is abundant in MSI-high CRC and the presence of F. nucleatum was negatively associated with levels of tumor-infiltrating lymphocytes (TILs) in MSI-high tumors, but positively associated with TIL density in non–MSI-high tumors [10]. This suggests there may be some biological variation in the immune contexture correlation to Fusobacterium, and/or the current sample numbers may not yet be sufficient for definitive conclusions.

Functional studies of the impact of Fusobacterium on immunomodulatory cells in CRC also reveal a mixed picture. In one intriguing result, Gur et al reported that the cytotoxicity of TILs is inhibited by F. nucleatum and that natural killer (NK) cell killing of tumor cells is also inhibited in the presence of various F. nucleatum strains [11]. Other studies have reported that F. nucleatum increases M2 polarization of macrophages in vivo and in vitro, and induces high expression of TLR4, IL-6, STAT3, p-STAT3, and c-MYC in cultured macrophages [12].

The studies detailed above have suggested that F. nucleatum may be associated with suppression of antitumor immunity, but the results vary in what immune cells are altered. More analysis of both human tissues and model systems will be required to determine whether Fusobacterium exerts a direct or indirect effect on immune cells in the tumor microenvironment.

Summary

There is convincing evidence to support the association of Fusobacterium and colorectal cancer. We have evaluated two controversial questions in the literature on the function of Fusobacterium in colorectal cancer.

1. Does Fusobacterium promote tumor formation and CRC cell proliferation? Our conclusion: there are intriguing hints but more analysis is required.

2. Does Fusobacterium affect the immune contexture of the tumor microenvironment? Our conclusion: as above, there are intriguing hints but more analysis is required.