Gastric Lgr5-positive Stem Cells Are The Cell Origins Of Gastric Cancer

Gastric cancer is the fifth most common malignancy and the third leading cause of cancer mortality worldwide in 2012. Several studies from mouse genetic models have demonstrated specific genetic alterations as the drivers of gastric cancer. However, most of gastric cancer mouse models result from germline mutation or somatic deletion occurred in all compartments of the epithelium. Therefore, it remains unclear which cell type sustains the cancer-initiating mutation. The cellular origin of gastric cancer remains elusive. Theoretically, gastric stem cells are the favored targets of transformation due to their inherent capacity for self-renewal and longevity, which permits the sequential acquisition of mutations and/or epigenetic changes required for tumorigenesis. Because the lack of gastric stem cell marker for a long time, there is a lack of in vivo evidence whether gastric stem cells could act as cancer-initiating cells to drive gastric cancer formation. The first identified Lgr5~+ stem cells reside at the base of each gland in adult pyloric antrum and gastro-esophageal junction. However, the role of Lgr5~+ stem cells in driving malignant gastric cancer is not fully validated. In vivo, gastric Lgr5~+-stem-cell-specific Apc deletion or Notch activation only gives rise to microadenoma or benign adenoma, and fails to give rise to malignant or invasive gastric cancer. To further complicate this, some clinical studies have revealed the increased LGR5-expressing cells or elevated LGR5 expression in intestinal gastric cancer(IGC) specimens, while some studies found LGR5+ cells were absent in the premalignant IGC lesions. The role of Lgr5~+ stem cells in gastric cancer formation and progression is not determined.So we raised the following scientific issues: 1) Does the gastric Lgr5~+ stem cell act as the cell-of-origin of malignant gastric cancer? 2) Are the gastric Lgr5~+ stem cells more prone to drive malignant transformation than the differentiated cell type? 3) How do LGR5 expression and some related signaling pathways alter in human gastric cancer?To study whether the gastric Lgr5~+ stem cell act as the cell-of-origin of malignant gastric cancer, we used the inducible Cre-LoxP system to modify the key mediators of TGF-β, Akt, Wnt and Kras pathways in mouse gastric Lgr5~+ stem cells, and meanwhile marked mutant Lgr5~+ cells and their progeny by Cre-reporter Rosa26 tdTomato. To demonstrate the role of Lgr5~+ stem cells in gastric carcinoma formation and progression, we next investigated the capability of Lgr5~+ stem cell self-renewal and differentiation and the alterations of signaling pathways responsible for the tumor formation and progression. To demonstrate the gastric Lgr5~+ stem cells were more prone to drive malignant transformation than differentiated cell, we used the Cre-LoxP system to modify the same mutants in three gastric differentiated cell types. Finally, using the large-scale human gastric cancer data from TCGA(The Cancer Genome Atlas) database, we detected the LGR5 expression and the alterations of the pathways.At day 90 post-induction, 65%(13/20) of Smad4 and PTEN double mutant mice developed polyps in the gastric antrum. Notably, 40%(8/20) of Smad4 and PTEN compound deficient mice developed intestinal-type gastric cancer(IGC) that invaded into the submucosa. Adenoma exhibited a marked increase in proliferative cells and loss of differentiated cells. Next, we investigated the downstream effectors of Smad4 and PTEN pathways responsible for tumor formation and progression. We found Smad4 deletion resulted in a dramatic increase in the expression of Smad4 targets, cyclin D1 and Spp1, two key effectors in cancer growth and metastasis. We also found that MAPK, Stat3 and EGFR pathways were activated in adenoma and invasive adenocarcinoma. Next, we assessed how Lgr5~+ stem cells initiated malignant gastric cancer. The Lgr5~+ cell number was similar between wild type and mutant mice. However, right above the Lgr5~+ stem cells domain, a marked increase in the Ki67+ cells were found in double mutant mice. Meanwhile, mutant-Lgr5~+-derived cells failed to differentiate into mature cells. These results suggested that Lgr5~+-derived progenies mainly contributed to hyperplasia by increasing proliferation while blocking differentiation. Furthermore, we examined the role of mutant Lgr5~+ stem cells in gastric caner progression. We found mutant Lgr5~+ cells still resided at the very base in benign adenoma or intramucosal lesion of IGC. Unexpectedly, in the invasive lesion of the antral IGC, the Lgr5~+ cells showed an extensive distribution throughout the invasive lesion. More importantly, mutant Lgr5~+ cells, regardless of their position, displayed proliferative capability and co-expressed CD44. These results suggested that these mutant Lgr5~+ cells contribute to gastric tumor growth and progression. Additionally, at the gastro-esophageal junction, mutant Lgr5~+ stem cells produced identical phenotypes with a relative low frequency. When KrasG12 D was expressed in the background of Smad4 and p53 double deletion, these triple mutant mice showed polyps in gastric antrum. These above results confirmed that gastric Lgr5~+ stem cells served as the cellular origin of invasive IGC.We next examined whether Lgr5~+ stem cells were more prone to drive malignant transformation than their differentiated progeny. A previous study has shown that Lgr5~+ stem cells give rise to pit cells in the antrum as well as parietal cells that are located at the transition zone between the antrum and corpus on the lesser curvature. We used the Cre-LoxP system to delete Smad4 and PTEN in these two gastric differentiated cell types, and we found no morphological aberration in the gastric antrum or transition zone of these two murine lines. Lgr5 also marks some chief cells at the gastric lesser curvature. In the same double mutant mouse, Lgr5~+ chief cells with Smad4 and PTEN deletion did not lead to morphological alteration at the gastric lesser curvature even after 90 days post-induction. These results suggested that, at least in the context of Smad4 and PTEN loss, the gastric Lgr5~+ stem cells are favored targets for gastric tumorigenesis.Finally, using the large-scale human gastric cancer data from TCGA(The Cancer Genome Atlas) database, we found elevated LGR5 expression in the intestinal subtype of gastric adenocarcinoma occurred at the gastric antrum and GEJ. Though analysis of the same data, we found that gene deletion in the SMAD4 or PTEN locus was frequent in the intestinal subtype. As expected, SMAD4 or PTEN genes deletion was tightly associated with low mRNA and protein expression in human intestinal subtype gastric cancer. Notably, concurrent deletions in the SMAD4 and PTEN loci were significant more than single deletion of SMAD4 and PTEN. Consistent with this, the mRNA levels of SMAD4 versus PTEN were positively correlated in the intestinal type samples. In line with our mouse modle, these results indicated that elevated LGR5 expression and aberrations in the SMAD4 and PTEN pathways in human IGC, and suggested the contribution of SMAD4 and PTEN downregulation in human gastric cancer growth.The conclusions and significances are included in this research as following: 1) We provided in vivo evidence that gastric Lgr5~+ stem cells could act as cancer-initiating cells to drive malignant IGC formation. We used the inducible Cre-LoxP system to concurrently delete Smad4 and PTEN genes in mouse gastric Lgr5~+ stem cells. We found mutant Lgr5~+ stem cells can give rise to the mutant clone, which consequently progressed into microadenoma, adenoma and invasive adenocarcinoma. 2) We provided evidence that gastric Lgr5~+ stem cells were more prone to drive malignant transformation than differentiated cell type. 3) We found that concurrent deletions in the SMAD4 and PTEN loci were significant more than single deletion of SMAD4 or PTEN, and the mRNA levels of SMAD4 versus PTEN were positively correlated in the intestinal type samples. By analyzing the TCGA data, we found the elevated LGR5 expression and aberrations in the SMAD4 and PTEN pathways in human IGC.In conclusion, we provided novel cellular and molecular mechanisms for the gastric tumorigenesis. Gastric Lgr5~+ stem cells as the cellular origin of invasive gastric cancer contribute to gastric tumor formation and progression. We provided the theoretical basis and models for screening of gastric cancer therapeutic targets.