*2.1.3. CD44*

[28]. Through *in vivo* lineage tracing experiments, these authors found that LGR5-positive cells were self-renewing and multipotent and were responsible for the renewal of the gastric epithelium. Interestingly, they observed that the transformation of this population of stem cells could drive gastric tumorigenesis *in vivo* [28], fact that has been strongly demonstrated more recently by Li and collaborators [29]. Consistently with the observation in mice, in human gastric tissue, LGR5 expression has also been found in the bottom of gastric glands [30]. Supporting the putative role of LGR5 as a gastric CSC regulator in humans, independent studies have reported LGR5 overexpression in human gastric cancer samples respect to normal gastric mucosa in a progressively increasing manner from well-differentiated to poorly differentiated gastric carcinomas [20, 31]. Furthermore, LGR5 expression has been strongly linked to a high degree of tumor infiltration, high TNM stage, recurrence and dismal prognosis of gastric cancer patients [18–20, 31]. More recently, Wang and collaborators have shown that sphere cells derived from a gastric cancer cell line presented increased expression of some canonical stem regulators, being LGR5 particularly elevated [32]. They also showed that ectopic LGR5 overexpression potentiated the sphere cell growth and cell migration capabilities of gastric cancer cells and also their tolerance to oxaliplatin, associating LGR5 expression with the characteristic features of CSCs [32]. These findings are in concordance with previous observations showing the impairment of the invasiveness and the reduction in the expression of metalloproteinase 2 (MMP2) and β-catenin in gastric cancer cells in response to LGR5 silencing *in vitro*; and revealing a positive correlation between the expression of LGR5 and MMP2 in gastric cancer tissue samples [20]. Regarding the implications of the carcinogenic agent *H. pylori* in the LGR5 positive cells in the stomach, it has been found that this population of cells is expanded in gastric cancer tissues affected by the bacteria, indicating that LGR5 likely represents a marker

of stem cells susceptible of oncogenic transformation driven by *H. pylori* [13, 33].

CD133 (also Prominin 1) is a pentaspan transmembrane glycoprotein present in embryonic epithelial structures, thought to function as an organizer of plasma membrane topology, and regulating the maintenance of the appropriate lipid composition within the plasma membrane [34]. CD133 has been presented as a marker of cancer stem cells in colon, pancreas, brain or lung cancer [35], yet its role in gastric CSCs is controversial. Several findings related to different aspects of gastric CSCs have been published in support of its role as a gastric CSC marker and regulator. In gastric cancer cell lines, CD133 silencing abrogates sphere formation capacity [36] and, consistently, CD133 has been found overexpressed in gastric sphere cultures [37, 38]. Noteworthy, a large number of publications show increased CD133 expression in human gastric cancer tissue respect to non-neoplastic gastric mucosa and highlight the prognostic significance of CD133, associating its overexpression with a big plethora of adverse clinic-pathological features, such as elevated cellular proliferation rates, high T stage, venous invasion, lymph node and distant metastasis, chemoresistance, recurrence, poor 5-year disease-free and overall survival and so on [37, 39–42]. According to this, studies performed in gastric cancer cell lines demonstrate that CD133-positive gastric cancer cells present a CSC phenotype, since they are more tumorigenic, more chemoresistant and exhibit higher migration or invasion capacities than CD133-negative cells [37, 38, 43]. However, some controversial findings have been published indicating that CD133 expression is not a *sine qua non* condition for gastric cancer cells

*2.1.2. CD133*

46 Gastric Cancer

CD44 is a transmembrane glycoprotein expressed on leukocytes, endothelial cells, hepatocytes or gastric epithelial cells, which acts as a receptor for hyaluronic acid (HA) [47] and can also interact with other ligands, such as osteopontin, collagens and MMPs. CD44 is a fetal and adult hematopoietic stem cell regulator that is involved in cell-cell interactions, cell adhesion and migration and participates in a wide variety of cellular functions, including hematopoiesis and lymphocyte activation, recirculation and homing [48]. CD44 gene contains 20 exons. Ten of these exons (exons l–5 and 16–20) are expressed together on many cell types and the product is referred to as the "standard" form of CD44. Additionally, complex alternative splicing of the transcripts affecting exons from 6 to 15 (variant exons) results in many functionally distinct isoforms or variants (CD44v) [49]. The role of CD44 as a CSC marker has been broadly studied in myeloid leukemia and also in several solid tumors such as lung, brain, liver, head and neck or gastric cancer [50]. In gastric cancer, the first temptative characterization of CSCs in terms of markers was performed by Takaishi and collaborators, who found that CD44+ cells isolated from different gastric cancer cell lines presented sphere formation ability *in vitro* and tumorigenic potential when inoculated into stomach and skin of immunodeficient mice, abilities that were abrogated by CD44 silencing. Moreover, these CD44+ gastric cancer cells showed the stem cell characteristics of self-renewal and the ability to give rise to differentiated progeny [44]. In concordance, other authors have documented CD44 enrichment in spheres derived from gastric cancer cell lines [51] or have identified that CD44-positive cells derived from gastric cancer cell lines are resistant to 5-fluorouracil and cisplatin chemotherapy and also exhibit significantly more migration, invasion and anchorage-independent growth capabilities [52]. Regarding gastric cancer clinical samples, CD44 is expressed in 80% of gastric tumor resected specimens [40] and its high expression has been associated with tumor size, lymphatic vessel and intravenous invasion, moderate grade of differentiation and low response to chemotherapy [52–54]. Furthermore, the presence of CD44+ cancer cells at the invasive front of gastric tumors entails poor survival and constitutes a prognostic factor for this malignancy [55]. In relation to this aspect, Watanabe and collaborators have found that in gastric cancer patients, the frequency of circulating CD44-positive tumor cells correlates with disease stage, depth of tumors and venous invasion [56]. Moreover, it has been suggested that the emergence of gastric CSCs induced by *H. pylori* infection of gastric mucosa may rely on CD44 upregulation [57]. Nevertheless, in contraposition to these evidences, some works have not found CSC characteristics in the subpopulations of CD44-positive cells isolated from gastric tumors [45, 58]. Besides, it is being sustained the notion that some CD44 variants are more relevant for gastric CSCs than standard CD44. An example of it is the work of Lau and collaborators, who show that CD44v8-10 is the predominant CD44 variant expressed in gastric cancer cells, whose expression levels, unlike those of standard CD44, are increased in gastric tumors respect to adjacent normal tissue. The authors also showed that ectopic expression of CD44v8-10, but not standard CD44, in gastric cancer cells potentiates their ability to initiate tumors in mice at limiting cell concentrations and that total CD44 silencing impairs tumor-initiating potential of cells, which could be rescued by restoration of CD44v8-10, but not standard CD44, expression [46].
