Spasmolytic polypeptide-expressing metaplasia (SPEM) cell lineages can be an origin of gastric cancer

Abstract

Intestinal type gastric cancer arises in a field of precancerous metaplastic lineages. Two types of metaplastic glands are found in the stomachs of humans with the characteristics of pyloric metaplasia or intestinal metaplasia. While spasmolytic polypeptide-expressing metaplasia (SPEM) cell lineages have been identified in both pyloric metaplasia and incomplete intestinal metaplasia, it has been unclear whether SPEM lineages or intestinal lineages can give rise to dysplasia and cancer. A recent manuscript published in this journal describes a patient with evidence of an activating Kras(G12D) in SPEM that is propagated into adenomatous and cancerous lesions which manifest further oncogenic mutations. This case therefore supports the concept that SPEM lineages can serve as a direct precursor for dysplasia and intestinal-type gastric cancer.

Gastric cancer arises in the human stomach in the setting of a field of pre-cancerous metaplastic cell lineages. Metaplasias are thought to represent a reparative process that can be localized to areas of significant damage. The first response to damage in the gastric corpus epithelium appears to be pyloric metaplasia, where oxyntic glands are replaced by mucous-secreting glands similar in structure to the antral glands. In pyloric metaplasia, Muc6-expressing spasmolytic polypeptide-expressing metaplasia (SPEM) cell lineages are found at the bases with Muc5AC-expressing foveolar lineages towards the lumens. While pyloric metaplasia may be programmed to resolve after cessation of injury, a second metaplastic gland phenotype, intestinal metaplasia, can evolve in the setting of chronic injury and inflammation. Two types of intestinal metaplasia are characterized by the presence of intestinal goblet cells in the gastric glands and found in the stomach. Glands with complete intestinal metaplasia show well-developed goblet cells and enterocyte-like cells and often have Paneth cells at their bases. These complete intestinal metaplasia glands are considered a successful protective lineage and do not have significant pre-neoplastic potential. In contrast, incomplete intestinal metaplasia glands have immature goblet cells and recent studies have demonstrated the presence of SPEM cell lineages at their bases [1]. Incomplete intestinal metaplasia is associated with the highest risk for cancer induction [2].

A number of recent publications have sought to elucidate the origin of pre-cancerous lesions in mouse models. These previous studies have suggested that intestinal type gastric cancers are associated with increased activation of Ras signaling [3, 4]. Thus multiple investigations have focused on the expression of the G12D activating mutation in Kras in gastric cells. Notably, expression of Kras(G12D) in chief cells with the Mist1-CreERT2 driver leads to development of SPEM, which over time can evolve into incomplete intestinal metaplasia and dysplasia [5]. These studies indicated that activation of Kras is a critical initial driver for development of SPEM and for its progression towards intestinal metaplasia. Nevertheless, it has remained unclear whether SPEM cell lineages or intestinal metaplastic cell lineages represent the precursors for development of intestinal-type gastric cancers.

In a recent issue of The Journal of Pathology, Kumagai et al. [6] described a patient with a well-differentiated adenocarcinoma in the absence of detectable H. pylori infection. The investigators examined SPEM lesions in proximity to regions with cancer and adenoma as well as a non-adjacent fundic polyp. While the fundic polyp did not have a Kras mutation, they found that SPEM, gastric adenoma and gastric adenocarcinoma lesions showed a Kras(G12D) mutation. These findings suggest that all of the lesions originated from a founder SPEM cell lineage with the Kras(G12D) mutation [6]. This patient therefore appears to confirm findings in mice that Kras(G12D) expression or Kras activation in SPEM cells can be an initiating event in gastric carcinogenesis. Also, SPEM cell lineages with Kras activating mutations can give rise to gastric cancer.

The authors also found that gastric adenocarcinoma gained further mutations in p53 and CDKN2A [6]. These results are consistent with a number of recent investigations that suggest that, while Kras activation or activating mutations may set the stage for emergence of metaplasia and dysplasia, development of frank adenocarcinoma likely requires additional mutations in driver oncogenes [7, 8]. Indeed, in mouse models, dysplastic stem cells isolated from dysplastic gastroids developed from Mist1-CreERT2;LSL-Kras(G12D) (Mist1-Kras) mice (Meta4) do not harbor significant mutations beyond the Kras mutation. Nevertheless, these cells can give rise to adenocarcinoma with multiple mutations following implantation subcutaneously in nude mice [9]. These studies support the concept that SPEM cell lineages may give rise directly to dysplasia and adenocarcinoma (Figure 1).

Figure 1:

A scheme for derivation of dysplastic lineages from SPEM cells harboring an activating Ras(G12D) mutation. Red cells depict the SPEM lineages in pyloric metaplasia that have acquired the Kras(G12D) activating mutation. These SPEM lineages can then give rise to incomplete intestinal metaplasia and then, with further oncogene mutations, to dysplasia and cancer.

If SPEM cell lineages can give rise to gastric cancer, how does this fit into the concepts developed by Professor Correa implicating a cascade from atrophy to intestinal metaplasia to cancer [10]? The answer may lie with a more detailed analysis of the lineages in metaplastic glands. Previous studies have demonstrated that incomplete intestinal metaplasia represents the highest risk for development of gastric cancer [2]. Using AQP5 as a marker of SPEM lineages, we have recently documented that, while SPEM lineages are distinctly present at the bases of pyloric metaplasia glands, they also are present at the bases of glands with incomplete intestinal metaplasia [1]. Incomplete intestinal metaplasia glands are clearly identified with staining for SPEM cell lineages with AQP5, and staining of incomplete intestinal metaplasia/dysplastic cell lineages with TROP2 (TACSTD2). In the present manuscript [6], the gastric cancer appears to have developed adjacent to a region with SPEM cell lineages harboring the activating Kras(G12D) mutation. However, it is not clear whether regions of incomplete intestinal metaplasia were also present adjacent to the cancer, which may have served as an intermediate for development of adenocarcinoma. In any case, the findings in this case report [6] would support the concept that SPEM cell lineages may serve as origins for gastric adenocarcinoma, likely through the accumulation of further oncogenic mutations.

While this investigation clearly demonstrates a connection of SPEM cell lineages to the development of cancer, these studies do not address whether other intestinalized cell lineages within incomplete intestinal metaplasia glands also can serve as de novo sources of dysplastic stem cells. Certainly, normal isthmal progenitor cells can give rise to diffuse type cancers in patients with cadherin mutations, so there may be multiple origins for initiation of dysplasia progression. Further studies, like those performed by Kumagai et al [6] addressing the discrete connection of mutations within precancerous and cancerous lesions from individual patients, are needed to identify the connection of mutations in early lesions with the evolution of cancer.