Gastric-to-intestinal transdifferentiation and cancer

In The Curious Case of Benjamin Button, an original short story authored by F. Scott Fitzgerald, the main character ages in reverse, beginning life as a mature adult (i.e., well-differentiated state) and then progressing inexorably toward a childlike existence (i.e., progenitor state). A molecular counterpart to this process, which frequently occurs during oncogenesis, is transdifferentiation, in which cells transition from a well-differentiated state to a progenitor state, which is then followed by the reemergence of a different cell lineage (i.e., metaplasia). Intestinal-type gastric cancer epitomizes this process during its progression from normal mucosa, to chronic gastritis and atrophy, to intestinal metaplasia, and finally to dysplasia and adenocarcinoma. The annual incidence of gastric cancer is estimated to be 0.1% for patients with atrophy but increases 2.5 fold for patients with intestinal metaplasia, underscoring the premalignant potential of this lesion (1). In PNAS, Fujii et al. provide fresh insights into the molecular underpinnings that regulate the development of intestinal metaplasia within the stomach (2).

The strongest known risk factor for intestinal-type gastric adenocarcinoma is chronic colonization by the bacterial pathogen Helicobacter pylori (3). However, only a percentage of colonized cases show the development of neoplasia, and enhanced risk is related to H. pylori strain differences, host responses governed by genetic diversity, and/or specific interactions among host, microbial, and environmental determinants (4). One H. pylori strain-specific virulence locus that augments cancer risk is the cag pathogenicity island, which encodes a type IV secretion system (TFSS) that functions as a molecular syringe to inject microbial proteins into host cells. The product of the cagA gene (CagA) is translocated by the TFSS into epithelial cells and undergoes targeted tyrosine phosphorylation by Src and Abl kinases at motifs (termed A, B, C, or D) containing the amino acid sequence EPIYA (5–7). Phospho-CagA activates a cellular phosphatase (SHP-2) and ERK MAPK, leading to morphological aberrations that mirror changes induced by growth factor stimulation (8, 9). However, nonphosphorylated CagA also exerts effects with carcinogenic potential, including activation of β-catenin (10), which can occur via PI3-kinase–dependent inhibition of GSK-3β (11). Previously, Murata-Kamiya et al. identified another mechanism through which CagA can activate β-catenin in gastric epithelial cells: physical interaction with E-cadherin and disruption of the E-cadherin–β-catenin complex, leading to β-catenin translocation from the membrane into the nucleus (12) (Fig. 1). Of interest in that study, the authors noted that CagA-deregulated β-catenin transactivated the intestinal specific transcription factor CDX1, which was followed by up-regulation of the intestinal differentiation marker Muc2 (12), findings that provided the framework for the work of Fujii et al. (2).

Fig. 1.

Effect of H. pylori CagA on CDX1-mediated intestinal transdifferentiation. The H. pylori cag TFSS translocates CagA into gastric epithelial cells. CagA can activate β-catenin through disruption of the E-cadherin–β-catenin complex, leading to β-catenin translocation from the membrane to the nucleus. β-Catenin subsequently transactivates the intestinal-specific transcription factor CDX1, which then up-regulates the reprogramming transcription factors SALL4 and KLF5. Stemness factors regulated by SALL4 and KLF5 facilitate the transdifferentiation from well-differentiated gastric epithelial cells to well-differentiated intestinal epithelial cells, a key step in the development of intestinal metaplasia of the stomach.

Fujii et al. (2) begin with a manipulatable gastric epithelial cell system, Tet-Off

Fujii et al. provide fresh insights into the molecular underpinnings that regulate the development of intestinal metaplasia within the stomach.

MKN28 cells, in which overexpression of CDX1 could be precisely regulated. A comparison of results from RNA expression arrays and ChIP-chip assays identified 166 genes that were potential targets of CDX1. Based on previous data implicating SALL4 and KLF5 as reprogramming factors, the authors focus on these two constituents and demonstrate that they were indeed up-regulated by CDX1, which bound to cognate sites within each of the respective promoter regions. Importantly, the authors extend these findings into the gastric niche by demonstrating that levels of SALL4 and KLF5 were significantly increased within gastric mucosa of CDX1-overexpressing vs. WT mice, and within human gastric tissue harvested from patients with intestinal metaplasia, compared with patients without this lesion. Returning to mechanistic studies, Fujii et al. (2) subsequently determine that other markers of intestinal cell stemness, including GATA binding protein 6 (GATA6), follistatin (FST), leucine-rich repeat containing G protein-coupled receptor 5 (LGR5), and BMI1 polycomb ring finger oncogene (BMI1), were up-regulated in CDX1-overexpressing cells. CDX1-dependent up-regulation of stemness factors was followed by an increase in expression of a subset of intestinal differentiation markers, including sucrase-isomaltase (SI) and membrane metallo-endopeptidase (MME), which characterize absorptive enterocytes, but no increase in markers of Paneth cells or enteroendocrine cells. To more definitively implicate SALL4 and KLF5 in this pathway, stable suppression studies were performed, which demonstrated the requirement for these reprogramming factors in CDX1-mediated intestinal transdifferentiation. Finally, the role of H. pylori and the cag TFSS was investigated by infecting GES-1 gastric epithelial cells with a WT cag⁺ strain or its isogenic cagA⁻ mutant; levels of induction of CDX1, SALL4, KLF5, and LGR5 were each significantly attenuated in cells cocultured with the cagA⁻ isogenic mutant compared with the parental WT H. pylori strain (Fig. 1).

This study provides information regarding how a chronic bacterial infection may affect carcinogenesis; more importantly, it establishes a framework for future directions. Most of this work uses a reductionist system consisting of modifiable gastric epithelial cells. Progression from normal gastric mucosa to intestinal metaplasia in vivo occurs within the context of H. pylori-induced inflammation. Subsequent studies should focus on the role of the host immune response in this process. For example, DNA damage resulting from inflammation-associated reactive oxygen and nitrogen species plays a key role in the development of premalignant lesions within H. pylori-infected gastric mucosa (13), and H. pylori can directly induce DNA damage in gastric epithelial cells via translocation of CagA (14). In addition to its ascribed role as a reprogramming factor, KLF5 can exert other functions depending on the cellular context. In cultured cells, KLF5 functions as a molecular chaperone for β-catenin, promoting its nuclear localization and modifying its transcriptional activity (15). McConnell et al. previously demonstrated that intestinal cell-specific deletion of klf5 in mice leads to impaired barrier function, inflammation, and a regenerative phenotype (16, 17). Tissue-specific depletion of klf5 in the intestine also resulted in disruption of β-catenin signaling, as evidenced by reductions in the levels of β-catenin target genes in klf5-deficient compared with WT mice. Thus, additional pathways mediated by KLF5 may contribute to transdifferentiation. As the authors acknowledge (2), the role of CDX2 in this process remains an unexplored area of investigation. Which motifs within CagA lead to activation of CDX1-dependent intestinal metaplasia? Because not all persons infected with H. pylori cag⁺ strains develop gastric cancer, CagA proteins from East Asian H. pylori strains, which are derived from patients with higher cancer risk and are more potent in inducing cellular morphologic aberrations, could be compared with CagA proteins derived from Western H. pylori strains. Regardless, the findings from this study (2) are important not only for understanding the development of gastric cancer, but also because such results may be extended to other malignancies that arise from inflammatory foci.