Reproducibility in Cancer Biology: The who, where and how of fusobacteria and colon cancer
The association between the bacterium Fusobacterium nucleatum and human colon cancer is more complicated than it first appeared.
- Johns Hopkins University School of Medicine, United States
Trillions of microbes live on or in the body of a typical human, and this relationship is mostly harmonious. Our colons harbor the highest density of bacteria, with a mucus barrier protecting our gut from them. For 50 years, scientists have researched who amongst these bacteria might cause colon cancer, where these bacteria might act in the colon, and how they might cause colon cancer.
In 2012 researchers at the BC Cancer Agency, Simon Fraser University and the University of Guelph reported that they had used sequence-based technologies to show that Fusobacterium nucleatum, a species not previously linked to cancer, might be associated with colon cancer (Castellarin et al., 2012). They found that the levels of F. nucleatum in colorectal carcinoma were significantly higher than the levels in adjacent normal tissue. An independent group also reported a similar finding at the same time (Kostic et al., 2012). This was a surprise as F. nucleatum is usually found in the mouth.
In 2016, as part of the Reproducibility Project: Cancer Biology, Repass et al. published a Registered Report which explained in detail how they would seek to replicate the experiment (figure 2 in Castellarin et al.) in which quantitative polymerase chain reaction (qPCR) was used to detect F. nucleatum in colon tissues taken from colon cancer patients (Repass et al., 2016). The results of this experiment have now been published as a Replication Study (Repass and Reproducibility Project: Cancer Biology, 2018). In short, F. nucleatum was detected in just 25% of colorectal carcinomas in the Replication Study, and the difference in the level of this species in colorectal carcinomas and adjacent tissues (there was 10% more F. nucleatum in the carcinomas) was not significant.
So why did the Replication Study not find what was reported in the original study? One possible explanation is that Castellarin et al. studied 99 colon tissue pairs (tumor and tumor-adjacent tissues) from colon cancer patients. The Replication Study, on the other hand, included samples from 40 patients with cancer (both tumor and adjacent tissue), along with 40 non-diseased control tissue samples from age, sex and ethnicity matched individuals. A power calculation had suggested that a sample size of 40 would be big enough to see the effect reported in the original paper, assuming that the two populations were clinically similar. However, Castellarin et al. did not fully report the clinical metadata for their study (such as age, gender, ethnicity and risk factors for disease), so differences between the populations might explain why the original results were not replicated. This possibility is also supported by the fact that the relative abundance of F. nucleatum in tumors in the original experiment is much higher than in the Replication Study.
Technical issues might also have contributed to the lack of replication: in particular, in the Replication Study, qPCR suggested that F. nucleatum DNA was present in 26 samples, albeit at low levels, but qPCR amplicon sequencing detected the presence of specific F. nucleatum gene products in just 16 of these samples (10 in carcinomas; 6 in adjacent tissues). Moreover, all the non-specific amplicons (that is, the 10 that were not due to F. nucleatum) were detected very close to the detection limit of this technique.
Given that numerous studies have already shown that there is an association between F. nucleatum and the microbiota of human colon cancer (see Repass et al., 2016 and Gholizadeh et al., 2017 for reviews), the Replication Study is not a reason to change our view of this association. Rather, it provides a critical opportunity to reflect on our growing, yet incomplete, knowledge regarding fusobacteria and colon cancer. First, and most obvious, there is a need for prospective human studies in well-defined populations – using both microbiology and bioinformatics approaches – to carefully probe how risk factor and other clinical features might alter colon health in the presence of fusobacteria. For example, while F. nucleatum has certainly been the species most consistently identified in association with colon cancer to date, many studies have raised the possibility that other species of fusobacteria, or bacteria that co-aggregate in the presence F. nucleatum, could contribute to the pathogenesis of colon cancer (Castellarin et al., 2012; Kostic et al., 2012; Drewes et al., 2017; Bullman et al., 2017).
Further, while it is clear that fusobacteria are highly genetically diverse (Castellarin et al., 2012; Manson McGuire et al., 2014), we do not as yet understand whether the strains associated with colon cancer exhibit unique genetic features or are related to the fusobacteria that are common in the mouth.
There are also other gaps in our understanding. For example, some experiments that used daily inoculations of F. nucleatum in mouse models of colon cancer suggest that F. nucleatum acts early in colon cancer (Kostic et al., 2013; Yang et al., 2017), whereas germ-free mouse experiments – which are, arguably, a more definitive way to test the role of F. nucleatum on its own in tumor induction – refute this result (Tomkovich et al., 2017). In contrast, at least two studies (Castellarin et al., 2012; Bullman et al., 2017) suggest that F. nucleatum acts late in the tumor process, and research by the present author and co-workers could not discern an association with tumor stage (Drewes et al., 2017).
There is also uncertainty about where F. nucleatum is found: some researchers have reported that it is found with increased frequency in cancers on the right side of the colon (Mima et al., 2016), whereas this effect was not seen in other experiments (Drewes et al., 2017). The question of 'where?' is also complicated by the fact that right colon cancer contains prominent mucus-invasive bacterial biofilms, and about one third of these exhibit blooms of F. nucleatum.
Lastly, we lack a clear picture of how F. nucleatum contributes to the emergence of human colon cancer. Limited data suggest a wide range of putative mechanisms (including Wnt nuclear signaling, immune cell recruitment, checkpoint molecules and specific miRNA induction). While each mechanism is plausible, we lack a cohesive, step-by-step story for F. nucleatum carcinogenesis (Gholizadeh et al., 2017; Kostic et al., 2013; Yang et al., 2017).
While the Replication Study did not replicate the results of Castellarin et al., it provided useful information about the importance of population differences and the need for accurate F. nucleatum detection methods, and highlighted how much we need to learn about the links between F. nucleatum and colon cancer. A better understanding of the who (which Fusobacterium species and/or associates), the where (where in the colon, where in the world) and the how (which disease mechanisms) will help with the development of new prevention approaches, diagnostics and/or therapies for a cancer that is increasing in the young and also across the globe.
Note
Cynthia L Sears was the Reviewing Editor for the Registered Report (Repass et al., 2016) and the Replication Study (Repass and Reproducibility Project: Cancer Biology, 2018).
References
-
Carcinogenesis mechanisms of Fusobacterium nucleatumBiomedicine & Pharmacotherapy 89:918–925.https://doi.org/10.1016/j.biopha.2017.02.102
-
Fusobacterium nucleatum in colorectal carcinoma tissue according to tumor locationClinical and Translational Gastroenterology 7:e200.https://doi.org/10.1038/ctg.2016.53
Article and author information
Author details
Publication history
Copyright
© 2018, Sears
This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 1,993
- views
-
- 243
- downloads
-
- 23
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Reproducibility in Cancer Biology: The who, where and how of fusobacteria and colon cancer
eLife 7:e28434.
https://doi.org/10.7554/eLife.28434
Further reading
-
- Cancer Biology
- Immunology and Inflammation
The current understanding of humoral immune response in cancer patients suggests that tumors may be infiltrated with diffuse B cells of extra-tumoral origin or may develop organized lymphoid structures, where somatic hypermutation and antigen-driven selection occur locally. These processes are believed to be significantly influenced by the tumor microenvironment through secretory factors and biased cell-cell interactions. To explore the manifestation of this influence, we used deep unbiased immunoglobulin profiling and systematically characterized the relationships between B cells in circulation, draining lymph nodes (draining LNs), and tumors in 14 patients with three human cancers. We demonstrated that draining LNs are differentially involved in the interaction with the tumor site, and that significant heterogeneity exists even between different parts of a single lymph node (LN). Next, we confirmed and elaborated upon previous observations regarding intratumoral immunoglobulin heterogeneity. We identified B cell receptor (BCR) clonotypes that were expanded in tumors relative to draining LNs and blood and observed that these tumor-expanded clonotypes were less hypermutated than non-expanded (ubiquitous) clonotypes. Furthermore, we observed a shift in the properties of complementarity-determining region 3 of the BCR heavy chain (CDR-H3) towards less mature and less specific BCR repertoire in tumor-infiltrating B-cells compared to circulating B-cells, which may indicate less stringent control for antibody-producing B cell development in tumor microenvironment (TME). In addition, we found repertoire-level evidence that B-cells may be selected according to their CDR-H3 physicochemical properties before they activate somatic hypermutation (SHM). Altogether, our work outlines a broad picture of the differences in the tumor BCR repertoire relative to non-tumor tissues and points to the unexpected features of the SHM process.
-
- Cancer Biology
- Computational and Systems Biology
Effects from aging in single cells are heterogenous, whereas at the organ- and tissue-levels aging phenotypes tend to appear as stereotypical changes. The mammary epithelium is a bilayer of two major phenotypically and functionally distinct cell lineages: luminal epithelial and myoepithelial cells. Mammary luminal epithelia exhibit substantial stereotypical changes with age that merit attention because these cells are the putative cells-of-origin for breast cancers. We hypothesize that effects from aging that impinge upon maintenance of lineage fidelity increase susceptibility to cancer initiation. We generated and analyzed transcriptomes from primary luminal epithelial and myoepithelial cells from younger <30 (y)ears old and older >55 y women. In addition to age-dependent directional changes in gene expression, we observed increased transcriptional variance with age that contributed to genome-wide loss of lineage fidelity. Age-dependent variant responses were common to both lineages, whereas directional changes were almost exclusively detected in luminal epithelia and involved altered regulation of chromatin and genome organizers such as SATB1. Epithelial expression variance of gap junction protein GJB6 increased with age, and modulation of GJB6 expression in heterochronous co-cultures revealed that it provided a communication conduit from myoepithelial cells that drove directional change in luminal cells. Age-dependent luminal transcriptomes comprised a prominent signal that could be detected in bulk tissue during aging and transition into cancers. A machine learning classifier based on luminal-specific aging distinguished normal from cancer tissue and was highly predictive of breast cancer subtype. We speculate that luminal epithelia are the ultimate site of integration of the variant responses to aging in their surrounding tissue, and that their emergent phenotype both endows cells with the ability to become cancer-cells-of-origin and represents a biosensor that presages cancer susceptibility.
