Clinical And Biological Implications Of Gastric Cancer Stem Cell In Omental Milky Spots And Lymph Nodes Micrometastasis

BackgroundIn the light of cancer stem cell theory, tumor growth and metastasis are driven by a small population of cancer stem cells(CSCs),and there is a hierarchical organization of cell populations in cancer cells which hierarchy includes CSCs that produce committed progenitor cells that, finally resulting in the generation of fully differentiated Cells .Current chemotherapeutic agents and radiation therapy largely target proliferating and differentiated cells that form the bulk of the tumor but not the CSCs and progenitor cells. Thus, this quiescence and resistance may account for many treatment failures. If this is the case, then the only effective way to treat cancer is to target the CSCs and progenitor cells. Until now, there are three strategies have been employed to identify CSCs as follows: (1)expression of cell surface markers,(2)ability to exclude fluorescent dye Hoescht33342 in a cell population termed“side population”(SP), (3)formation of tumor spheres in suspension culture. But these methods lead us to focus cancer stem cells on a minor subset of tumor cells, instead of initial CSCs and progenitor cells. How to target and isolate initial CSCs and progenitor cells is the key for further study.Peritoneal metastasis can occur to any organ in the peritoneal cavity, but it is well known that there are preferred locations in the early stages of peritoneal cancer dissemination . Many studys have reported that the omentum is one of the initial sites for peritoneal metastasis. The accepted reason is that omentum possess milky spots, submesothelial lymphoid structures composed of numerous macrophages and lymphocytes , which provided a microenvironment (physical and chemical factors)where cancer cells can survive and grow into micrometastasis easily.Milky spots are primitive lymphoid tissues in the peritoneal cavity of humans and animals, and exist mainly in the greater omentum, the mesenterium, and the pelvic floor. In particular, omental milky spots are composed of numerous macrophage and lymphocyte aggregation and involved in the clearance of particles, bacteria and tumor cells from the peritoneal cavity. Omental milky spot is a well-known site of metastases of carcinomas of the ovaries, stomach and colon. Many studies have indicated that gastric cancer cells selectively infiltrate into the milky spots in the early stages of peritoneal cancer dissemination, and then forming micrometastasis. Macrophages in omental milky spots not only have cytotoxic properties against tumor cells but also produce growth stimulatory factors.The tumor cells are attached to the milky spots in the greater omentum and find a microenvironment within milky spots in which they are capable of survival and growth into solid metastases. Combining the theory of cancer stem cell, we propose that a gastric cancer stem cell infiltrate into the milky spot, then proliferate and differentiate into some gastric cancer progenitor cells, which is the basic of forming metastasis.Omental milky spot provide a chance which let us target a single gastric cancer stem cell. S.J.Oosterling and colleagues have identified that tumor cells mainly adhered on omental milky spots 4 h after tumor cell injection . This preferred attachment can be explained by the fact that mesothelial cells lining milky spots showed higher levels of cellular adhesion molecules such as intercellular adhesion molecule-1 than other omental regions, which may contribute to enhanced adhesion as well . However, a sharp decline in numbers of tumor cells was seen in specific milky spot areas, then few proliferating tumor cell clusters were found in milky spots and the infiltrated cancer cell forming micrometastasis.On the basic of former studies, we propose that part of CSCs and mature cancer cells in some milky spot have been eliminated by the killing function of macrophages, but only survival CSCs in some milky spot proliferate and differentiate into gastric cancer progenitor cells and forming micrometastasis. So, in this situation, omental milky spots become a high efficiency“natural filter”for screening gastric cancer stem cells.In most solid tumors, cancer stem cells are rare. Many cancers contain a small but significant proportion ( < 1%) of CSCs. Catherine and colleagues calculated by limiting dilution analysis that there was only one CSC in 5.7×10~5 unfactionated human colon cancer cells . So it is very difficult to find the unique marker to isolated and purified CSCs and progenitor cells which were covered by thousands of mature cancer cells. In many cases, a combination of multiple markers is used to identify a particular cancer stem cell type. So now, researchers often identify stem cells in shorthand by a combination of marker names reflecting the presence (+) or absence (–) of them . But omental milky spots help us to resolve this problem. Cancer stem cell and progenitor cells consist of total cancer cells at the early stage of milky spot micrometastasis, the proportion of the cancer stem cell and progenitor cells markers may increase accordingly. Through comparing the change of the expression of cell surface molecules on cancer cells, we may primarily screen the cancer stem cell markers. So if a marker express higher on the surface of cancer cells which were isolated from milky spot micrometastasis, but lower express on surface of the bulk tumor cells or even negative. This marker will may be the unique maker used for isolating and purifying CSCs and progenitor cells.In the current study, we therefore used a milky spot model to investigate the biological significance of the cells in micrometastasis and its tumorigenic ability, and whether the omental milky spots can be a high efficiency“natural filter”for screening gastric cancer stem cells or not. And study several stem cell-like surface markers expression in gastric cancer specimens, and their relationship to clinicopathological features, including survival.1. Role of omental milky spot micrometastasis in peritoneal disseminationObjective:The omentum is one of the initial sites for peritoneal metastasis since it possesses milky spots which provide a microenvironment in which cancer cells are able to readily survive and grow into micrometastases. This study investigated the nature of gastric cancer cell dissemination in the omentum.Methods:Gastric cancer MFC cells (1x10~4) labeled with DiI were injected intraperitoneally in strain 615 mice. The mice were sacrificed at different times, and omenta were excised for immunohistochemistry. Results: Electron microscopy revealed the detailed cell composition of the milky spots which were largely composed of abundant macrophages with some lymphocytes, neutrophils and various stromal cells . No basement menbrane was noted beneath the cells covering the milky spot. The surface layer cells were in direct contact with the connective tissue matrix. The prominent intercellular gaps and stomata caused the submesothelial connective tissue and cells to be exposed to the peritoneal surface.In vitro, no differences in adhesion abilities and proliferation rates were shown between DiI-MFC and non-labeled MFC cells. After 10 days of culture, DiI-MFC cells displayed a high signal intensity which then began to decrease.The MFC cells began to adhere to the omental milky spots in the omental milky spot areas at 4 h post-injection. At 12 h post-injection, MFC cells were particularly concentrated in the milky spots, while no tumor cells were found in the non-milky spot areas of the omenta. A reduction in the number of DiI-MFC cells was noted in the milky spot areas at 48 h post-injection due to the killing function of the macrophages. After 72 h, the number of DiI-MFC cells further decreased in the milky spot areas, while proliferating tumor cells and the formation of micrometastases were noted. In contrast, sporadic DiI-MFC cells in the non-milky spot areas were observed at 48 h after intraperitoneal injection. After 10 days, some tumor cell clustersincreased in size and cell number, and the formation of metastasis was noted.At 48 h post-injection, sporadic tumor cells were found in the non-milky spot areas in omental non-milky spot areas, while no cell clusters were detected. After 72 h, the number of DiI-MFC cells in the milky spots increased. After 1 week, the number of DiI-MFC further increased in the non-milky spot areas, while tumor cells did not proliferate into cell clusters and only single cell-type micrometastasis was observed. After 10 days, the increased number of tumor cells only contributed to single cell-type micrometastasis in the non-milky spots.At 2 weeks after intraperitoneal injection, the milky spot areas were completely occupied by the proliferating gastric cancer cells, and cell cluster-type metastasis was noted. In contrast, proliferating cancer cell clusters were not observed in the non-milky spot areas, and cancer cells formed single cell-type metastasis.Conclusions:Cancer cells selectively infiltrate into milky spots in the early stages of peritoneal cancer dissemination, and that cell cluster-type metastasis in milky spots is the main cell source of further peritoneal metastasis.2. Tumorigenic ability of cells“cluster-type”micrometastasis in omental milky spotsObjective: To investigate the tumorigenic ability of gastric cancer cells of cluster-type micrometastasis in omental milky spots. Whether the gastric cancer cells of cluster-type micrometastasis have the property of stem cell , and wether the omental milky spots can be a highly efficient“natural filter”for screening gastric gastric cancer stem cells and progenitor cells or not.Methods:After 72h of intraperitoneal injection, animals were sacrificed and omenta were excised. The surface of omental milky spots was rinsed with RPMI 1640 and collected the gastric cancer cells of micrometastasis. The cells were suspended in a 1:1 mixture of media and injected subcutaneously into the right and left flank of NOD/SCID mice with anesthesia. After 16 weeks all mice were sacrificed by cervical dislocation, and tumors were removed , fixed in formalin and analyzed by histology(H&E、CK-20 and CEA). All of the gastric cancer cells of cluster-type micrometastasis in omental milky spots were originally implanted into NOD/SCID mouse subcutaneously and never cultivated or expanded in vitro. To study the tumorigenecity of cells of cluster-type micrometastasis, a different number of cells cluster-type micrometastasis was injected subcutaneously into NOD/SCID mice. Tumorigenecity was evaluated at 14 weeks after NOD/SCID transplantation. To determine whether this xenotransplant system was quantitative and able to detect the doses unable to initiate tumor growth, we performed limiting dilution experiments. Groups of NOD/SCID mice were transplanted with doses of cells cluster-type micrometastasis over a range from doses unable to initiate tumor growth to doses that always initiated tumor formation and normal MFCs as control group.Results: To determine whether omental milky spots enrich the gastric cancer stem cells, we collected the cells of cluster-type micrometastasis and injected into NOD/SCID mice. Of 40 mice (dose range: 1×101 to 1×103) injected with cells of cluster-type micrometastasis, one mouse transplanted with 5×102 cells generated a tumor. Two mice transplanted with 7.5×102 cells generated a tumor and tumors were consistently generated after injection of 1×103 cells of cluster-type micrometastasis. In contrast, one tumor was generated after injection of 5×104 control group of gastric cancer cells, one mouse transplanted with 1×105 cells generated a tumor and tumors were consistently generated after injection of 5×106 gastric cancer cells. The histology of the tumors, as expressed by H&E (haematoxylin and eosin) staining, shows same poor differentiated adenocarcinomas. The immunohistochemical markers CK-20 and CEA reveal similar staining patterns in both the bulk and cells of cluster-type micrometastasis xenografts.Conclusions:There are gastric cancer stem cells in“cluster-type”micrometastasis of omental milky spots3. The biological characters of the gastric cancer cells in omental milky spots“cluster-type”micrometastasisObjective: To investigate the biological characters of the gastric cancer cells in“cluster-type”micrometastasis of omental milky spotsMethods:After 72h of intraperitoneal injection, animals were sacrificed and omenta were excised. The surface of omental milky spots was rinsed with RPMI 1640 and collected the cells in“cluster-type”micrometastasis of omental milky spots. To detect the bilolgical characters of cells in“cluster-type”micrometastasis of omental milky spots, RT-PCR and immunocytochemistry analyse stem cell-like surface markers, such as CD44、CD24、CD133、CD324 and CD34 expression.Results:Ten random fields were respectively taken for counting the CD133 positive cells and total cells, CD133 and CD44 positive cells quantity in cells in“cluster-type”micrometastasis of omental milky spots was more than it was in control group of gastric cancer cells (P<0.01). Counting the CD324 negative cells and total cells in ten random fields, CD324 negative cells quantity in OMSS-GCCswas more than it was in unscreened gastric cancer cells (P<0.01). RT-PCR showed that CD44 mRNA increased and CD324 mRNA decreased.Conclusions:CD133~+、CD44~+and CD324~- expression maybe the gastric cancer stem cell surface markers.4.CD44+CD324- expression and prognosis in patients with gastric cancerObjective : To evaluate the frequency of the cancer stem cell marker CD44 combined CD324 expression in gastric cancer specimens , and the association of the expression with the survival of the gastric cancer patients .Methods : A total of 203 patients with primary gastric cancer who underwent curative gastric resection with lymphadenectomy . The expression of CD44 and CD324 was analysed by immunohistochemistry and the association of CD44 and CD324 expression with clinicopathological characteristics , and survival of the patients was evaluated.Results: CD44~+CD324~- tumors showed strong correlation with positive lymph node metastasis. Patients with CD44-CD324+ had a significantly better prognosis(54.8%) than that with CD44~+CD324~- (24.5%). A multivariate Cox proportional hazards model using those variables associated with survival in our study(depth of invasion,tumor size, lymph node metastasis, gastrectomy, CD44 expression, CD324 expression, CD44 and CD324 expression) revealed that CD44~+CD324~- expression(P =0.005) was one of significant independent prognostic indicator, whereas CD44 expression(P =0.081), CD324 expression (P =0.068)was not.Conclusion : CD44~+CD324~- expression was one of significant independent prognostic survival factor for patients with gastric cancer. 5. Adverse prognosis of“clustered-cell”versus single-cell micrometastases in pN0 early gastric cancerObjective:The clinical significance of lymph node micrometastasis (MM) for pN0 early gastric cancer is not well documented. The aim of this study is to clarify the risk factors of lymph node MM and the prognostic significance of the type of lymph node MM in patients with pN0 early gastric cancer.Methods: We investigated the lymph node MM with using an anticytokeratin immunohistochemical stain in 160 patients with pN0 early gastric cancer who underwent curative resection between 2000 and 2005.Results: By immunohistochemical staining for CK ,lymph node MM was identifed in 34 of 160 patients (21.3%), in 84 of 1656 lymph nodes (5.1%). MM was significantly associated with tumor size(P=0.041) and lymphatic invasion (P=0.008). However, gender, tumor location, depth of invasion , venous invasion and histological type did not significantly differ. Lymph node MM was present in the form of a single-cell metastasis (10/34) or cluster of tumour cells(24/34). There was no significant relationship between the single cells or cell clusters of lymph node MM and clinicopathological variables including patient gender, tumor location, depth of invasion, tumor size, lymphatic or venous invasion and histological type.By univariate survival analyses, the statistically significant prognostic factors affecting 5-year survival rates were found to be depth of tumor invasion(P=0.020), tumor size(P=0.035), lymphatic invasion(P=0.018), MM (P<0.001) and cluster type of MM(P<0.001).The 5-year survival rate of MM negative patients was significantly higher than MM positive patients. The 5-year survival rate for the group of patients with cluster-type MM was significantly lower than that for the group with negative or single-cell MMMultivariate Cox regression survival analysis showed that MM (P<0.001) and cluster type MM (P < 0.001)proved major independent prognostic factors for histologically node negative gastric cancer patients.Conclusions:The incidence of lymph node MM in patients with node negative early gastric cancer was 21.3%, and cancer cell“cluster type”of MM proved a primary independent prognostic factor for pN0 early gastric cancer patients.