GPR68 limits the severity of chemical-induced oral epithelial dysplasia

Abstract

Head and neck cancer is the sixth most common malignancy, and there is an urgent need to identify physiological processes contributing to tumorigenesis. Extracellular acidification caused by aerobic glycolysis within tumor microenvironments can stimulate proton-sensing receptors. GPR68, or ovarian cancer G protein-coupled receptor 1, responds to extracellular acidity and is highly expressed in head and neck squamous cell carcinoma (HNSCC) as well as normal esophageal tissue. To study the role of GPR68 in oral dysplasia, wild-type and GPR68^-/- mice were treated with 4-Nitroquinoline N-oxide (4NQO) in drinking water for 11-13 weeks, followed by normal water for 11-12 weeks. 4NQO treatment resulted in 45 percent of GPR68^-/- mice developing severe dysplasia or squamous cell carcinoma compared to only 10.5 percent of GPR68^+/+ mice. This correlated with increased frequencies of regulatory T cells in the spleens of male GPR68^-/- mice. Dysplastic regions of the tongue had increased CD31 staining compared to normal regions in both GPR68^-/- and GPR68^+/+ mice, suggesting that angiogenesis was GPR68-independent. RNA knockdown studies using HNSCC cell lines demonstrated no direct effect of GPR68 on survival or growth. Overall, we demonstrate that GPR68-deficiency worsens the severity of chemical-induced oral dysplasia, suggesting a protective role for this gene in tumorigenesis.

Figure 1

Endogenous GPR68 expression does not significantly affect body weight, water consumption or survival following oral 4-Nitroquinoline N-oxide (4NQO) treatment. WT (closed symbols) and KO (open symbols) mice were separated into male (M) and female (F) groups and placed on 4NQO drinking water (0.05 mg/mL) for 11–13 weeks, followed by normal water for 11–12 weeks. (A) Body weight reported as a percent of starting weight (week 0 = 100%) at the indicated timepoints. Area under the curve analysis revealed no differences between the groups. (B) Body weight at week 22. (C) Average amount of 4NQO water consumed per mouse each day. Water was replaced each week, with the volume consumed divided by the number of mice in each cage. This number was then divided by the days since the last water change. (D) Time course showing the percent of WT (solid line) or KO (dotted line) mice surviving at each weekly timepoint. Data are combined from three independent experiments and shown as the mean +/− SEM (WT M n = 9; WT F n = 10; KO M n = 11, KO F n = 9). Asterisks indicate statistically significant differences between the groups (**P < 0.01).

Figure 2

GPR68 deficiency increases the severity of 4NQO-induced tongue dysplasia. (A) Representative H&E-stained squamous mucosa with varying degrees of dysplasia on week 22 or 24. Mild dysplasia is characterized by changes confined to the lower one-third of the mucosa. Moderate and severe dysplasia have progressive involvement of the lower two-thirds to the upper one-third of the mucosa. Invasive squamous cell carcinoma has full thickness severe dysplasia with stromal invasion and permeation of malignant cells beyond the basement membrane. E = epithelium, S = stroma, M = muscle. The magnification for each panel is indicated. (B) Stacked chart showing the number of WT (left) or KO (right) mice with each grade of dysplasia. Statistical analysis was performed using a Chi-square test. (C) Stacked chart showing the mean (+/−SEM) number of large (red) or small (green) lesions per tongue on week 22 or 24. Statistical analysis was performed using Student’s t-tests for large and small lesions. Data are from the same experiments described in Fig. 1.

Figure 3

Dysplastic regions have increased CD31 expression independently of GPR68. Tongue sections from week 22–24 were stained for CD31 by immunohistochemistry (IHC). (A) Representative stain delineating the dysplastic region (red outline) from normal region (yellow). (B) Percent of cells staining positive for CD31 within the dysplastic (closed bar) or normal (open bar) regions. Data are from the same experiments described in Fig. 1 (*P < 0.05; **P < 0.01; ***P < 0.001).

Figure 4

Role of GPR68 on Treg frequencies following 4NQO treatment. Single cell suspensions from cervical lymph nodes and spleen were stained for CD4 and Foxp3, and analyzed by flow cytometry. (A) Representative dot plots showing SSC vs. CD4 (top) and Foxp3 versus CD4 (bottom) from cervical lymph nodes of untreated and 4NQO-treated mice on week 24. (B) Mean +/− SEM of the percent of CD4 T cells expressing Foxp3 in cervical lymph nodes of WT (closed symbols) and KO mice (open symbols), as indicated. The dotted line represents Foxp3 frequencies in age-matched untreated mice. (C) Mean + /- SEM of the percent of CD4 T cells expressing Foxp3 in spleen of WT and KO mice, as indicated. Data are from the same experiments described in Fig. 1.

Figure 5

GPR68 does not intrinsically regulate Treg differentiation. Naïve CD4 T cells from GPR68^fl/fl (Cre-negative) and GPR68^fl/fl x CD4-Cre (CD4-Cre +) mice were cultured at pH 6.5, 7.0 and 7.5 with anti-CD3/CD28 beads (1:3 bead:cell ratio), IL-2 (10 ng/mL) and TGF-b (5 ng/mL). On day 4, cells were stained with CD4 and Foxp3, and analyzed by flow cytometry. The percent of CD4 T cells expressing Foxp3 is shown as the mean + /- SEM. Data are combined from four experiments with n = 13–17.

Figure 6

GPR68 knockdown had no significant effect on cell line survival. Cal-27 and UD-SCC-2 cells were cultured at pH 6.6, 7.0 and 7.4 with siRNA specific for GPR68 (open bars) or a negative control (closed bars), as indicated. At 48 h, cells were scraped, stained for Annexin V and 7-AAD, and analyzed by flow cytometry. (A) Representative flow cytometry plots showing Annexin V vs. 7-AAD staining. (B) Charts showing the percent of live double-negative (Annexin V^- 7-AAD^-) cells as mean +/− SEM. Data are combined from 6 experiments with n = 16–18 (Cal-27), or from three experiments with n = 9 (UD-SCC-2).

Figure 7

GPR68 knockdown had no significant effect on cell line growth. Cal-27 (A), MDA-1483 (B) and UD-SCC-2 (C) cells were cultured for 72 h with siRNA specific for GPR68 (open bars) or a negative control (closed bars), as indicated. An SRB assay was used to measure cell growth, with data reported as the mean +/− SEM of the optical density at 565 nm. Data are combined from 4 experiments with n = 16 (A), 3 experiments with n = 14 (B) or 3 experiments with n = 12 (C).