The G protein-coupled receptor T-cell death-associated gene 8 (TDAG8) facilitates tumor development by serving as an extracellular pH sensor

Abstract

Tumors often are associated with a low extracellular pH, which induces a variety of cellular events. However, the mechanisms by which tumor cells recognize and react to the acidic environment have not been fully elucidated. T-cell death-associated gene 8 (TDAG8) is an extracellular pH-sensing G protein-coupled receptor that is overexpressed in various tumors and tumor cell lines. In this report, we show that TDAG8 on the surface of tumor cells facilitates tumor development by sensing the acidic environment. Overexpression of TDAG8 in mouse Lewis lung carcinoma (LLC) cells enhanced tumor development in animal models and rendered LLC cells resistant to acidic culture conditions by increasing activation of protein kinase A and extracellular signal-regulated kinase in vitro. Moreover, shRNA-mediated knockdown of endogenous TDAG8 in NCI-H460 human non-small cell lung cancer cells reduced cell survival in an acidic environment in vitro as well as tumor development in vivo. Microarray analyses of tumor-containing lung tissues of mice injected with TDAG8-expressing LLC cells revealed up-regulation of genes related to cell growth and glycolysis. These results support the hypothesis that TDAG8 enhances tumor development by promoting adaptation to the acidic environment to enhance cell survival/proliferation. TDAG8 may represent a therapeutic target for arresting tumor growth.

Tumor cells frequently exhibit elevated glucose uptake and glycolysis (1, 2). Surplus production and insufficient clearance of glycolytic metabolites cause a significant decrease in the local extracellular pH in tumor tissues, which can reach as low as 6.4 (1, 3, 4). This acidity is believed to promote a variety of cellular events in tumors such as metastasis (5, 6), angiogenesis (7, 8), and cell death (9, 10). However, the mechanisms through which tumor cells recognize the extracellular pH and transmit signals intracellularly have not been fully elucidated.

T-cell death-associated gene 8 (TDAG8, also known as GPR65) originally was identified as an orphan G protein-coupled receptor (GPCR) in apoptotic thymocytes (11). We and other groups have found that TDAG8 functions as an extracellular pH sensor which enhances cAMP production in response to an acidified extracellular environment (12–14). Im et al. (15) previously proposed that TDAG8 is activated by psychosine (1-β-d-galactosylsphingosine). However, they did not provide data showing a specific interaction of psychosine with TDAG8, and subsequent reports have not supported this finding (12–14). Consistently, Radu et al. (16) have demonstrated that TDAG8 is dispensable for psychosine-induced formation of multinucleated cells. Together, these reports consistently indicate that psychosine is not a physiological ligand for TDAG8 and demonstrate that TDAG8 functions as an extracellular pH sensor.

In an earlier study, Sin et al. (17) demonstrated that TDAG8 shows slight oncogenic activity, although the involvement of its pH-sensing ability in this oncogenicity has not been clarified. Radu et al. (16) have generated Tdag8-deficient mice, which show normal immune development and glucocorticoid-induced thymocyte apoptosis. More recently, we have shown that TDAG8 partly mediates the extracellular acidification-induced inhibition of proinflammatory cytokine production in mouse macrophages (18).

Human TDAG8 mRNA has a relatively limited gene expression profile but is highly expressed in the immune system (11, 19). However, strong expression of TDAG8 mRNA, along with other pH-sensing GPCRs, is observed in a variety of human tumors (17, 20). Further evidence that TDAG8 mRNA overexpression is observed in certain tumors and tumor cell lines is provided by the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/; e.g., accession nos. GDS1962 and GDS2736) and the BioGPS database [formerly Genomics Institute of the Novartis Research Foundation (GNF) SymAtlas, http://biogps.gnf.org/]. Therefore, we hypothesized that TDAG8 expressed on the surface of tumor cells may be involved in tumor malignancy by sensing the acidic environment. Here, we show that overexpressed TDAG8 protects tumor cells from cell death under acidic conditions in vitro and enhances tumor development in vivo. Furthermore, shRNA-mediated knockdown of endogenous TDAG8 attenuated cell survival under acidic conditions in vitro and tumor development in vivo. Therefore, we propose a mechanism by which tumor cells adapt to an acidic environment to survive and grow.

Results

TDAG8 Enhances Tumor Development in Vivo.

To analyze the role of TDAG8 in tumor cells, we used mouse Lewis lung carcinoma (LLC) cells (21), which are derived from C57BL/6 mice and commonly used as a model of tumor growth. To date, four GPCRs, namely ovarian cancer G protein-coupled receptor 1 (OGR1) (22), G protein-coupled receptor 4 (GPR4) (22), G2 accumulation (G2A) (23), and TDAG8, have been identified as extracellular pH sensors. LLC cells show endogenous expression of Ogr1 mRNA, whereas transcripts of the other pH-sensing GPCRs are barely detectable (Fig. S1A). We established LLC cells stably expressing human TDAG8 (TDAG8-LLC cells) by purifying a polyclonal TDAG8-expressing cell population from drug-resistant cells by FACS sorting, which avoids clonal deviation (Fig. 1A). Consistent with previous reports (12, 13), TDAG8-LLC cells exhibited increased cAMP levels in response to low-pH stimulation, whereas LLC cells transfected with the empty vector (control cells) showed no change in cAMP production under acidic conditions (Fig. 1B).

Fig. 1.

TDAG8-mediated tumor development in mice. (A) The expression level of exogenous TDAG8 protein in TDAG8-LLC cells was examined by FACS. (B) The pH dependency of TDAG8-mediated cAMP production in LLC cells is shown. Results in a second experiment were similar. Substantially the same experiment was performed in Fig. 4C. (C) TDAG8-LLC or control cells were injected s.c. into C57BL/6 mice. The time course of tumor development is shown (n = 10; *P < 0.05; two-way repeated measures ANOVA). The results of a second experiment were similar. (D–H) TDAG8-LLC or control cells were injected i.v. into C57BL/6 mice. (D) TDAG8-mediated tumor malignancy was evaluated by survival analysis (n = 17; *P = 0.001; log-rank test). The results shown are from two independent experiments. (E) Lung sections at day 15 after injection were stained with H&E. (Scale bar: 1 mm.) (F) Lungs at day 19 after injection are shown. (G) Tumors on the lung surface were counted 15 d after injection. Each point represents the tumor count from one mouse. Horizontal bars denote the mean (*P < 0.005; Mann–Whitney U test). (H) Lung wet weights were measured 19 d postinjection. Data from age-matched naïve mice are shown for comparison. Each point represents one mouse. Horizontal bars denote the mean (*P < 0.0005; Mann–Whitney u test). The results of a second experiment were similar.

We first examined the impact of TDAG8 overexpression on tumor development. TDAG8-LLC or control cells were injected s.c. into syngeneic C57BL/6 mice. Mice injected with TDAG8-LLC cells (hereafter designated “TDAG8-LLC mice”) showed significantly enhanced tumor development (Fig. 1C). In addition, tumor formation was observed in all the TDAG8-LLC mice but in only 10 of 16 mice injected with control LLC cells (hereafter designated “control mice”). The effect of TDAG8 on tumor formation was examined further using A549 human lung carcinoma cells, which do not endogenously express mRNA for any pH-sensing GPCRs (Fig. S1B). A polyclonal population of A549 cells that stably expressed TDAG8 (TDAG8-A549 cells) formed significantly larger tumors than control cells when injected s.c. into athymic nude mice (Fig. S2).

To evaluate further the role of TDAG8, TDAG8-LLC or control cells were injected i.v. into the tail vein of C57BL/6 mice. TDAG8-LLC mice had a significantly shorter lifespan than control mice (Fig. 1D). Microscopic and macroscopic observations revealed that TDAG8-LLC cells form more and larger tumors both inside and on the surface of the lungs, compared with control cells (Fig. 1E at day 15 and Fig. 1F at day 19). Indeed, a significantly greater number of tumors were detected on the lung surface of TDAG8-LLC mice at day 15 postinjection than in control mice (Fig. 1G). Consistently, injection of TDAG8-LLC cells resulted in significant increases in both wet and dry lung weights at day 19 (Fig. 1H and Fig. S3A), whereas the lung weights of control mice were nearly equal to those of naïve mice (Fig. 1H and Fig. S3A). Together, these results indicate that TDAG8-LLC cells form tumors across the entire lung and that control mice probably have very low tumor burdens in the lung at this stage. By immunostaining, HA-tagged TDAG8 was detected in the tumor foci in the lungs of TDAG8-LLC mice at day 25 (Fig. S3B), suggesting that the expression of TDAG8 in LLC cells is stable in vivo. Many groups have succeeded in measuring or imaging the low extracellular pH values in tumors in vivo by means of magnetic resonance spectroscopy, MRI, or PET, demonstrating that the tumor microenvironment is acidic in various types of tumors in humans, mice, and rats (1, 3, 4, 24, 25). Therefore, these results imply that TDAG8 enhances tumor development in vivo, likely through its extracellular pH sensitivity.

Microarray Analyses Reveal Increased Expression of Genes Regulating Cell Growth and Glycolysis in the Lung Tissues of TDAG8-LLC Mice.

To understand better the global effects of TDAG8 overexpression on the transcriptome in vivo, microarray analysis using Affymetrix GeneChip Mouse Genome 430 2.0 Arrays was performed with total RNA prepared from whole lungs of TDAG8-LLC mice or control mice 19 d after i.v. injection. This analysis identified a list of 1,425 up-regulated (>1.5-fold) and 793 down-regulated (<0.5-fold) genes in the lungs of TDAG8-LLC mice compared with those of the control mice (Fig. 2A). We queried this list against the Reactome database (26), a curated and peer-reviewed resource of human biological processes. SkyPainter, an annotation tool available in this database, identified and visualized up-regulated genes that are involved in cellular processes including “cell cycle,” “DNA repair,” “metabolism of nucleotides,” “metabolism of carbohydrates,” and other processes (Fig. 2B). Consistent results were obtained by analyzing the genes that were up-regulated in tumor-containing TDAG8-LLC lung tissues using a functional annotation clustering tool provided by the Database for Annotation, Visualization and Integrated Discovery (DAVID) (27, 28) (Table S1). These results suggest that cell growth is facilitated by overexpression of TDAG8 in the tumors.

Fig. 2.

Up-regulation of genes related to cell growth and glycolysis in lung tissues of TDAG8-LLC mice indicated by microarray analyses. (A) The schema for microarray analyses is shown. The dysregulated genes were analyzed subsequently as shown in B and C. (B) The Reactome reaction map illustrating all the events (reactions and/or pathways) that involve the analyzed genes, using the SkyPainter tool. Each arrow represents one event. The color of each arrow represents the ratio of the level of expression of the analyzed genes (TDAG8-LLC/control) involved in the event, as indicated by the color bar on the right. Gray arrows correspond to reactions that do not involve the analyzed genes. The areas highlighted by solid black lines indicate events that are up-regulated in the tumor-containing lung tissues of TDAG8-LLC mice. (Further details of Reactome are available at http://www.reactome.org/.) (C) GenMAPP pathway-based representation of altered glycolysis/gluconeogenesis. Metabolites are indicated by black letters. Boxed names are genes involved in the metabolic pathways that are connected by black arrows. Genes shown in a red box were up-regulated (>1.5-fold) in TDAG8-LLC cells compared with control-LLC cells. Genes shown in a blue box were down-regulated (<0.5-fold). Fold change is shown also. Genes in gray boxes were not found in the selected gene list. All abbreviations are defined in Table S2.

Enhanced glycolysis, resulting in an acidic environment, is one of the hallmarks of primary and metastatic cancers (1, 2, 29, 30). Consistently, we also noted an up-regulation of genes related to carbohydrate metabolism. A detailed profile of the glycolysis/gluconeogenesis pathway was constructed by analyzing the list of genes dysregulated in the lungs of TDAG8-LLC mice using the GenMAPP program (31, 32). The expression levels of many glycolysis-related genes such as hexokinases and phosphofructokinases were increased in the lung tissues of TDAG8-LLC mice, whereas gluconeogenesis-related genes such as glucose-6-phosphatase, fructose bisphosphatases, and phosphoenolpyruvate carboxykinase 1 remained unchanged (Fig. 2C). In addition, genes encoding a glucose transporter [solute carrier family 2 (facilitated glucose transporter), member 1 (SLC2A1)/glucose transporter 1] and monocarboxylic acid transporters [SLC16A1/monocarboxylate transporter 1 (MCT1) and SLC16A3/MCT4] also were up-regulated. These results, which are consistent with previous reports (1, 2, 29, 30), suggest that during tumor development glycolysis, glucose uptake, and lactate release are enhanced, creating an acidic microenvironment.

TDAG8 Facilitates Cell Survival and Growth Under Acidic Environment.

We then undertook a series of in vitro experiments to elucidate the mechanisms by which TDAG8 enhances tumor development. TDAG8-LLC cells displayed a higher survival rate than control cells when stimulated with a serum-free medium at pH 6.4 (Fig. 3 A and B), suggesting that expression of TDAG8 has a positive effect on cell survival under acidic conditions. Cell proliferation assays using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) also revealed that TDAG8-LLC cells show facilitated cell proliferation at pH 6.4 compared with control cells, although they grow more slowly at pH 6.4 than at pH 7.4 (Fig. 3C), possibly because the acidic culture conditions inhibit cell growth and cause cell death through a variety of signaling pathways (10). We speculate that TDAG8 counteracts those signals and enhances cell health but not strongly enough to maintain cell growth fully. The augmented proliferative ability of TDAG8-LLC cells under acidic conditions was observed consistently in cell-counting assays (Fig. S4A). Similar results also were obtained with another polyclonal TDAG8-LLC cell population in the MTT assay (Fig. S4B).

Fig. 3.

Contribution of TDAG8 to LLC cell survival and growth under acidic conditions in vitro. (A and B) After culture in a serum-free medium at pH 6.4, viable and dead LLC cells were counted by flow cytometry. (A) Cell viability was determined by propidium iodide staining followed by FACS analysis. Representative FACS plots at 48 h are shown. (B) The time course of cell survival is shown (n = 3; *P < 0.0001; two-way repeated-measures ANOVA). The results of a second experiment were similar. (C) The proliferative effect of TDAG8 on LLC cells was assessed by MTT assay (n = 3; *P < 0.0005; unpaired t test). The experiments were repeated three times with similar results. (D) TDAG8-mediated ERK phosphorylation (P-ERK) was detected by Western blotting. Cells were treated with a serum-free medium at pH 6.4 or 7.4 for 24 h. The effect of 10 μM H89 on ERK phosphorylation also was observed. The results of a second experiment were similar. (E) The effects of ERK and PKA inhibition on LLC cell growth were measured by MTT assay at pH 6.4 (n = 3; *P < 0.001; one-way ANOVA followed by Tukey's multiple comparison test). The concentration of the inhibitors was 10 μM. The experiments were repeated three times with similar results.

Western blotting analysis then was conducted to clarify the molecular mechanisms underlying the TDAG8-mediated resistance to acidic conditions. Extracellular signal-regulated kinase (ERK) phosphorylation in TDAG8-LLC cells was increased under acidic conditions compared with neutral pH conditions, whereas ERK phosphorylation in control cells was attenuated under acidic conditions (Fig. 3D). Furthermore, the protein kinase A (PKA) inhibitor H89 dramatically abrogated ERK phosphorylation in TDAG8-LLC cells at pH 6.4, as well as at pH 7.4, implying that PKA mediates ERK phosphorylation.

MTT assays at pH 6.4 in the presence of kinase inhibitors consistently demonstrated that the MAP kinase/ERK kinase (MEK) 1/2 inhibitor U0126 greatly attenuates the proliferative ability of TDAG8-LLC cells compared with control cells (Fig. 3E), whereas the inactive analog U0124 has almost no effect (Fig. S5). The PKA inhibitor H89 also inhibited TDAG8-LLC cell proliferation at pH 6.4, an observation that is consistent with our data showing that TDAG8 promotes cAMP production in an acidic pH-dependent manner (Fig. 1B). These results suggest that TDAG8 improves cell proliferation and prevents cell death of LLC cells under acidic conditions by activating the PKA and ERK signaling pathways.

pH-Sensing Ability of TDAG8 Is Essential for TDAG8-Mediated Tumor Development.

Ludwig et al. (22) reported that several histidine residues are important for the pH-sensing ability of OGR1. An amino acid sequence homology search among three known pH-sensing GPCRs (TDAG8, OGR1, and GPR4) identified three conserved histidine residues at positions 10, 14, and 243 of TDAG8. According to the predicted 3D model of OGR1 (22), His-10 and His-14 are located in the extracellular N-terminal domain, whereas His-243 is located near the extracellular side of helix VI (Fig. 4A). To evaluate the importance of the extracellular pH-sensing ability in TDAG8-mediated tumor development, we constructed two TDAG8 mutants that contain substitutions at the conserved histidine residues (H10/14F or H243N). We prepared LLC cell populations stably expressing these mutants at levels comparable to wild-type TDAG8 using a cell sorter (Fig. 4B). Both mutants displayed impaired pH sensitivity in the cAMP accumulation assay (Fig. 4C). We then evaluated tumor development following i.v. injection of these cell populations. Compared with wild-type TDAG8, cells expressing these two mutants consistently showed a diminished ability to promote tumor development in the lung (Fig. 4D and Fig. S3C), leading to improved animal survival (Fig. S3D). In accord with these in vivo data, the proliferative ability of these mutants under acidic culture conditions was diminished significantly in vitro (Fig. 4E). These data further support the idea that the pH-sensing ability is essential for TDAG8-mediated tumor progression as well as for cell proliferation under acidic conditions.

Fig. 4.

Importance of TDAG8 pH-sensing ability for tumor development. (A) Histidine residues of TDAG8 at positions 10, 14, and 243 were mutated. The approximate location of each mutation relative to the plasma membrane (PM) is indicated. Ex, extracellular space; In, intracellular space. (B) Wild-type and mutant (H243N and H10/14F) TDAG8-LLC cells were sorted by FACS so that cell populations with similar receptor expression levels were obtained. (C) The pH-sensing ability of TDAG8 mutants was assessed by cAMP accumulation (n = 4). Both mutants show attenuated cAMP production compared with the cells expressing wild-type TDAG8. The experiments were repeated three times with similar results. (D) Wild-type or mutant (H243N or H10/14F) TDAG8-LLC or control cells were injected i.v. into C57BL/6 mice. Lung wet weight was measured 19 d after injection. (Left) LLC cells (5 × 10⁵) were injected. The results are from two independent experiments (n = 10–11). (Right) Consistent results were obtained in another independent experiment using 1 × 10⁶ LLC cells (n = 7). *P < 0.05; **P < 0.01; ***P < 0.001; Kruskal–Wallis test followed by Dunn's multiple comparison test. (E) The proliferative effect of TDAG8 mutants on LLC cells was assessed by MTT assay (n = 5). **P < 0.01; one-way ANOVA followed by Dunn's multiple comparison test. The results of a second experiment were similar.

Endogenous TDAG8 Is Important for Cell Survival Under Acidic Conditions in Vitro and Tumor Development in Vivo.

Analysis of available gene expression data from the BioGPS database and our own RT-PCR data (Fig. 5A) consistently showed that NCI-H460 human non-small cell lung cancer cells endogenously express TDAG8 mRNA at a high level. Interestingly, NCI-H460 cells showed enhanced survival at pH 6.6 compared with the normal physiological pH of 7.5 (Fig. 5B). Treatment with the PKA inhibitor H89 reduced the survival of NCI-H460 cells at pH 6.6 (Fig. 5C). To examine if endogenous TDAG8 is responsible for this phenotype, we stably knocked down TDAG8 mRNA in two independent NCI-H460 cell lines with different shRNAs (shRNA1 or shRNA3). Consistent with the results obtained using LLC cells, NCI-H460 clones stably expressing either shRNA (clone 1–11 expressing shRNA1 or clone 3–8 expressing shRNA3; Fig. 5D) showed significantly reduced cell survival at pH 6.6 compared with vector-transfected control cells (Fig. 5E). Furthermore, shRNA-mediated knockdown of TDAG8 attenuated s.c. tumor development in athymic nude mice (Fig. 5F). These data strongly support a distinct role for TDAG8 in tumor development.

Fig. 5.

Attenuated cell survival under acidic conditions in vitro and tumor development in vivo following shRNA-mediated knockdown of endogenous TDAG8 in NCI-H460 cells. (A) TDAG8 mRNA was detected in several human cancer cell lines by RT-PCR, with a rank order of LOX-IMVI melanoma cells ≥ NCI-H460 non-small cell lung cancer cells > H111 gastric carcinoma cells > HT1080 fibrosarcoma cells. GAPDH was used as a control for RNA integrity. (B) After culture in a serum-free medium at pH 6.4 or 7.5, the survival of NCI-H460 cells was determined (n = 3; *P < 0.0001; unpaired t test). (C) The effect of 10 μM H89 on the survival of NCI-H460 cells was measured (n = 3; *P < 0.0005; unpaired t test). (D) Endogenously expressed TDAG8 mRNA was knocked down using shRNA in NCI-H460 cells. The expression levels of TDAG8 mRNA in two independent stable clones (clone 1–11 and clone 3–8, transfected with shRNA expression vectors shRNA1 and shRNA3, respectively) are presented. The values were normalized to GAPDH mRNA levels (n = 3; *P < 0.01; one-way ANOVA followed by Dunnett's multiple comparison test). (E) After culture in a serum-free medium at pH 6.4 or 7.5, the survival of TDAG8-shRNA stable clone cells (clone 1–11 and clone 3–8) was determined (n = 3; *P < 0.01, one-way ANOVA followed by Dunnett's multiple comparison test). (F) TDAG8-shRNA stable clone (clone 1–11) or control cells were injected s.c. into athymic nude mice. The time course of tumor development is shown (n = 10; *P < 0.05; two-way repeated-measures ANOVA followed by Bonferroni's test). The data in A–E are representative of two independent experiments with similar results; the results in F were obtained from two independent experiments.

Discussion

A rapidly growing tumor quickly exceeds its vasculature supply and thus lacks oxygen and nutrients. Glycolysis is up-regulated to obtain growth and survival advantages both hypoxic (1, 2, 7, 33) and normoxic conditions (the Warburg effect) (29, 30). This increase in glycolysis, which produces lactic acid, causes microenvironmental acidosis that requires tumor cell adaptation to prevent acidosis-induced toxicity and to promote continued survival and further proliferation. Normal cells undergo apoptosis under acidic conditions, whereas certain tumor cells are able to survive because of mutations that provide survival advantages (e.g., the well-characterized mutations in components of the apoptotic pathway such as p53) (34, 35). In the present study, we identified TDAG8 as a candidate molecule that contributes to acid resistance and tumor proliferation. Acid-sensing ion channels and transient receptor potential vanilloid 1 are expressed in sensory neurons and are known to be involved in acid sensitivity (36). However, whether these molecules also play a role in tumor development remains uncertain.

Here we demonstrate that overexpression of TDAG8 in tumor cells promotes cell survival under acidic conditions in vitro and enhances tumor development in vivo following both i.v. (LLC cells) and s.c. (LLC and A549 cells) injections in mice. In addition, we show that TDAG8 mutants with minimal pH-sensing ability display decreased tumor development and cell proliferation under acidic conditions. Consistent phenotypes were observed in vitro and in vivo when endogenous TDAG8 was knocked down in NCI-H460 cells using a targeted shRNA. Given these results, along with reports that TDAG8 is overexpressed in a variety of tumor tissues and tumor cell lines, we propose that TDAG8 can function as an extracellular pH sensor on the surface of tumor cells, inducing intracellular adaptations in response to an acidic environment to enhance tumor cell survival/proliferation. No metastatic foci were observed in other organs, such as liver or brain, in either animal model. However, conclusions about the role of TDAG8 in tumor metastasis would be premature, because metastatic foci typically require months to grow big enough to be visualized (37). To address this issue, a better metastasis model will be needed (38).

Microarray analyses of lung tissues of TDAG8-LLC mice revealed up-regulation of both growth-related genes (cell cycle, DNA repair, and other functions) and glycolysis-related genes. These results suggest that the surplus production of acidic glycolytic metabolites from TDAG8-LLC cells, together with the facilitated cell growth, leads to further acidification of the tumor microenvironment, which in turn activates TDAG8 (Fig. 6). Activation of the extracellular pH-sensing machinery may trigger intracellular adaptations that allow the cells to survive in a more acidic environment. Likewise, in some human cancer cells that overexpress TDAG8, such as NCI-H460 cells, this positive feedback may confer a growth advantage under a lower extracellular pH, an intrinsic feature of tumors (1–3). We note that acidification of the tumor microenvironment also may be caused by other factors derived from the surrounding cells, such as cancer-associated inflammation.

Fig. 6.

Proposed mechanism for TDAG8-mediated tumor development. The excessive production of acidic glycolytic metabolites in TDAG8-expressing tumor cells leads to the accumulation of protons in the extracellular space and subsequent acidification in the tumors, which in turn activates TDAG8, forming a positive feedback loop. Therefore, TDAG8 is likely to trigger adaptations that allow the tumor cells to survive in an acidic environment and promote further tumor development.

We also demonstrate that TDAG8-mediated LLC cell growth under acidic conditions is dependent on PKA and ERK activation (Fig. 6). The importance of PKA is consistent with increased cAMP production upon activation of the Gs-coupled TDAG8 (12, 13). The ERK signaling pathway also is reported to play a pivotal role in cancer progression (39). It is known that Gs-coupled GPCRs can use the PKA/B-Raf pathway to activate ERK in a cell-specific manner (40, 41), although PKA also may contribute to TDAG8-mediated cell growth independently of ERK activation. Because there is functional cross-talk between the GPCR and the Wnt canonical signaling pathways (42), and because ERK is involved in transcriptional regulation of cancer-related genes (43), future analyses of signaling networks downstream of TDAG8 will be important for further elucidating the mechanisms underlying TDAG8-mediated tumor development.

Aberrant overexpression and activation of GPCRs are common mechanisms by which malignant cells undergo autonomous proliferation, invade surrounding tissues, evade the immune system, and disseminate to other organs (20, 44). The GPCRs protease-activated receptor 1, endothelin receptors, and chemokine receptors are known to be overexpressed in tumor cells, leading to increased cell motility and proliferation (44). It is not yet clear how TDAG8 expression is up-regulated in some cancer cells. However, signal transducer and activator of transcription 3α (Stat3α) was shown to promote the expression of Tdag8 mRNA in mouse 32D myeloid progenitor cells (45). Because Stat3 is overexpressed and/or constitutively activated in a large number of different cancers (46), this transcription factor is a strong candidate for this role.

Although a previous report has suggested that TDAG8 is activated by psychosine (15), we and other groups have demonstrated that TDAG8 functions as an extracellular pH sensor and have not observed such agonistic effects of psychosine (12–14). In this study, we show that TDAG8 promotes cell survival and growth in response to an acidic environment. Taking these findings together, we propose that TDAG8 enhances tumor development through its pH-sensing ability. Singh et al. (47) recently reported that when OGR1, another pH-sensing GPCR, is overexpressed in PC3 human prostate cancer cells it suppresses tumor metastasis by abrogating cell migration. However, this phenotype is independent of the extracellular pH-sensing ability of OGR1. Thus, our report demonstrates tumor development facilitated by a pH-sensing GPCR through its extracellular pH-sensing ability, indicating a connection between tumor acidity and tumor cell survival/proliferation. GPCRs are predominantly expressed on the cell surface and represent a major therapeutic target (48). Based on our findings, we propose that antagonists of TDAG8 or anti-TDAG8 antibody drugs may be promising therapeutic candidates for the treatment of TDAG8-expressing tumors.

Materials and Methods

Detailed methods, including reagents, cell culture, cell stimulation, stable expression of TDAG8, mutagenesis, cAMP measurement, cell survival, cell counting, cell proliferation assay, ERK phosphorylation, stable knockdown of TDAG8 in NCI-H460 cells, animal experiments, histology, immunohistochemistry, microarray analysis, mRNA expression profile of extracellular pH-sensing GPCRs, and statistical analysis, appear in SI Materials and Methods.The information of PCR primers for pH-sensing GPCRs is shown in Table S3. The care and the use of animals were in accordance with the University of Tokyo's Ethics Committee for Animal Experiments.