Oxidative stress promotes oral carcinogenesis via Thbs1-mediated M1-like tumor-associated macrophages polarization

Abstract

Although oxidative stress is closely associated with tumor invasion and metastasis, its' exact role and mechanism in the initial stage of oral cancer remain ambiguous. Glutamine uptake mediated by alanine-serine-cysteine transporter 2 (ASCT2) participates in glutathione synthesis to resolve oxidative stress. Currently, we firstly found that ASCT2 deletion caused oxidative stress in oral mucosa and promoted oral carcinogenesis induced by 4-Nitroquinoline-1-oxide (4-NQO) using transgenic mice of ASCT2 knockout in oral epithelium. Subsequently, we identified an upregulated gene Thbs1 linked to macrophage infiltration by mRNA sequencing and immunohistochemistry. Importantly, multiplex immunohistochemistry showed M1-like tumor-associated macrophages (TAMs) were enriched in cancerous area. Mechanically, targeted ASCT2 effectively curbed glutamine uptake and caused intracellular reactive oxygen species (ROS) accumulation, which upregulated Thbs1 in oral keratinocytes and then activated p38, Akt and SAPK/JNK signaling to polarize M1-like TAMs via exosome-transferred pathway. Moreover, we demonstrated M1-like TAMs promoted malignant progression of oral squamous cell carcinoma (OSCC) both in vitro and in vivo by a DOK transformed cell line induced by 4-NQO. All these results establish that oxidative stress triggered by ASCT2 deletion promotes oral carcinogenesis through Thbs1-mediated M1 polarization, and indicate that restore redox homeostasis is a new approach to prevent malignant progression of oral potentially malignant disorders.

Keywords: ASCT2; Macrophages; Oral carcinogenesis; ROS; Thbs1.

Graphical abstract

Fig. 1

The identification of oxidative stress in oral mucosa epithelium of ASCT2 knockout mice Western blot (a) and IHC (b) were used to detect the efficiency of ASCT2 knockout in day 1,3,7,14,21 after tamoxifen injection (120 mg/kg). (Fig. 1–b, magnification × 100 for the left, magnification × 100 for the right, scale bars, 400 μm for 100* magnification and 100 μm for 400* magnification. the red box showed the area of interest). (c) The GSH/GSSG ratio was measured in the 3 groups. (d) H₂O₂ levels were evaluated in oral epithelium using the H₂O₂ Content Assay Kit. (e) ROS levels in oral epithelium of the same gene background (Slc1a5^f/f Cre^+/−) were assessed by DHE staining in cryosections. The markers were ASCT2 (green), ROS (red) and nucleus (blue) respectively. Left: Representative captures from 3 mice for each group. (Scale bar 100 μm). Right: Quantification of ASCT2 and DHE staining using image J. (f) Staining of γ-H2AX of the 3 groups in paraffin sections. The markers were ASCT2 (green), γ-H2AX (red) and nucleus (blue) respectively. Left: Representative captures for each group (Scale bar 100 μm). Right: Quantification of ASCT2 and γ-H2AX staining using image J. (g) Staining of 8-OHdG of the 3 groups in paraffin sections. The markers were ASCT2 (green), 8-OHdG (red) and nucleus (blue) respectively. The above mice used to detect oxidative stress were all male. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

ASCT2 knockout promoted OSCC occurrence in 4-NQO-treatment mice (a) Schematic diagram of the experimental strategies in vivo. (b) Body weight was recorded every week in different groups. (c) Tongue images of different groups. (d) Representative HE images of normal, dysplasia and carcinoma in tongue. Magnification, 20* and 100*; scale bars, 2000 μm for 20* magnification, 400 μm for 100* magnification. (e) Scores of the histopathologic diagnosis in the four groups. (f) Quantification of the histological ratio of high grade lesion (severe dysplasia and carcinoma), low grade lesion (mild and moderate dysplasia) and normal tissues in the four group. (g) Counts of low grade lesions in the four group. (h) Counts of high grade lesions in the four group. (i) The levels of proliferation cell nuclear antigen (PCNA) were assessed by IHC assay in four groups. Left: Representative images for each group, magnification, 100* and 200*; scale bars, 400 μm for 100* magnification and 200 μm for 200* magnification. The red box showed the area of interest. Right: Quantification of PCNA in the four groups. The above mice were all male and 8 mice for each group. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

ASCT2 knockout promoted Thbs1 expression (a) Venn diagram of differentially expressed genes (DEGs) among the four groups. (b) 33 overlapping genes were clustered to generate a heatmap between control 3 group and test group. (c) GO enrichment plot for the 33 overlapping genes. IHC assay was used to measure the levels of ASCT2 (d) and Thbs1 (e) in the four groups. Left: Representative images for each group, magnification, 100* and 400*; scale bars, 400 μm for 100* magnification and 100 μm for 400* magnification. The red box showed the area of interest. Right: Quantification of ASCT2 and Thbs1 in the four groups. The above mice were all male and 8 mice for each group. (g, h) ASCT2 expression was measured by Western blot and PCR respectively after transduction with lentivirus targeting ASCT2. (h) Western blot was used to evaluated the expression of ASCT2 and Thbs1 in DOK cells after ASCT2 knockdown. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4

ASCT2 knockout induced M1-like TAMs infiltration in cancerous sites of mice tongue treated by 4-NQO The proportions of CD4⁺ (a), CD8⁺ (b) and F4/80⁺ (c) were detected by IHC assay in four groups respectively. Left: Representative images for each group, magnification, 100* and 400*; scale bars, 400 μm for 100* magnification and 100 μm for 400* magnification. The red box showed the area of interest. Right: Quantification of CD4, CD8 and F4/80 in the four groups. (d) mIHC assay was used to evaluate the colocalization of F4/80 with CD86 or CD206 respectively in control 3 and test group. Left: Representative images for the two groups, magnification, 200* and 1000*, scale bars, 200 μm for 200* magnification and 40 μm for 1000* magnification. The red box showed the area of interest. Right: Quantification of M1 and M2 in the two groups. The above mice were all male and 8 mice for each group. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5

Thbs1 induced the polarization of M1-like phenotype through extracellular vesicle transport pathway THP-1 cells were treated with gradient-diluted recombinant Thbs1 protein (0, 1, 2.5, 5, 10, 20 μg/mL). (a, b) Flow cytometry was used to assess the proportion of M1-like or M2-like macrophages respectively. (c, d) Cell lysates were harvested and analyzed by Western blot for detecting M1 markers (iNOS and CD80) or M2 markers (Arg-1 and CD206). (e) Uptake of exosomes by THP-1-derived macrophages. The markers were β-Tubulin (green), exosomes (red) and nucleus (blue) respectively. Scale bar, 15 μm. (f) Thbs1 expression of exosomes from ASCT2 knockdown cells was assessed by Western blot. (g, h) Flow cytometry was used to calculate the proportion of M1-like or M2-like macrophages after treated with exosomes respectively. The above experiments were repeated 3 times. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 6

Targeted Thbs1 inhibited the polarization of M1-like TAMs through p38, Akt and SAPK/JNK signaling (a, c) Thbs1 expression was measured by Western blot and PCR assay in DOK cells transfected with siRNAs targeting Thbs1 (#Si1, #Si2) based on ASCT2 knockdown. (b) Western blot was used to detect the Thbs1 level in exosomes. (d, e) The percentage of M1-like or M2-like macrophages was analyzed by flowcytometry after treated with exosomes from Thbs1 knockdown DOK cells. (f) THP-1-derived macrophages were treated with exosomes isolated from Thbs1 knockdown cells. Cell lysates were acquired an analyzed by Western blot for detecting iNOS, CD80 and p38, Akt, SAPK/JNK signals. (g) The markers of M2-like macrophages (Arg-1 and CD206) were assessed by Western blot. The above experiments were repeated 3 times.

Fig. 7

ASCT2 knockdown upregulated Thbs1 expression by inducing ROS accumulation and thereby promoted M1-like TAMs polarization (a) Schematic diagram of glutamine metabolism. (b–e) Steady-state metabolite levels of Gln, glutamate, GSH and ROS were assessed. (f, h) Thbs1 expression was detected in whole cell lysis and exosomes respectively. (g) TCA-cycle metabolites (α-KG, citrate, malate, fumarate, succinate and ATP levels) were measured using kits. (i, j) Flowcytometry was conducted to examine the M1-like and M2-like macrophages polarization respectively. (k) Western blot was used to detect iNOS, CD80 and p38, Akt, SAPK/JNK signals. (l) The markers expression of M2-like macrophages (Arg-1 and CD206) was assessed by Western blot. The above experiments were repeated 3 times. NAC, N-acetylcysteine.

Fig. 8

4-NQO treatment significantly increased carcinogenecitty of DOK cells (a) Cell growth status was acquired by inverted contrasting microscope, magnification, 40* and 100*, scale bars, 1000 μm for 40* magnification and 400 μm for 100* magnification. The red box showed the area of interest. HE (b) and Papanicolaou staining (c) were used to observe cell morphology, magnification, 100* and 200*, scale bars, 400 μm for 100* magnification and 200 μm for 200* magnification. The red box showed the area of interest. (d) Cytoskeleton staining was performed to assess cytoskeleton between DOK and DOK-4-NQO cells, scale bars, 80 μm. Cell migration (e) and cell invasion assay (f) were conducted to detect the abilities of cell migration and invasion, magnification, 40* and 100*, scale bars, 1000 μm for 40* magnification and 400 μm for 100* magnification. The red box showed the area of interest. CCK-8 assay (g), cell cycle assay (h), and colony formation assay (i) were used to evaluate the proliferation ability between DOK and DOK-4-NQO cells. (j) Tumor sphere formation assay was conducted to assess the stemness, magnification, 50* and 100*, scale bars, 800 μm for 50* magnification and 400 μm for 100* magnification. The red box showed the area of interest. (k) Western blot was performed to examine the expression of p53, Bim1, Nanog, Claudin 4, Occluding, MMP2, Ki 67, PCNA between DOK and DOK-4-NQO cells. (l) Xenograft experiment (male nude mice) was used to compare the tumorgenicity. Left: Representative images of xenograft tumors. Right: Tumor Volume and weight of xenografts in the ending point respectively, Tumor volume=(length × width²)/2. (m) Representative HE images of xenograft tumors. magnification, 50* and 200*; scale bars, 800 μm for 50* magnification and 200 μm for 200* magnification. The red box showed the area of interest. The above experiments were repeated 3 times. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 9

DOK-4-NQO cells showed increased tumorgenicity treated with CM from M1-like TAMs Cell cycle assay (a) and CCK-8 assay(b) were used to detect proliferation ability of CM on DOK-4-NQO cells. The effect of CM on the colony formation was evaluated by colony formation assay(c). The abilities of migration (d) invasion (e) were determined by, cell migration and cell invasion assay. magnification, 100* and 200*, scale bars, 400 μm for 100* magnification and 200 μm for 200* magnification. The red box showed the area of interest. (f) Tumor sphere formation assay was conducted to assess the stemness, magnification, scale bars, 800 μm for 50* magnification and 400 μm for 100* magnification. The red box showed the area of interest. (g) Western blot was conducted to further assess stemness (Bim1 and Nanog), cell junction (Claudin 4 and Occludin), invasion (MMP2) and proliferation (PCNA) of CM on DOK-4-NQO cells. (h) Xenograft experiment (male nude mice) was used to evaluate the tumorgenicity of CM on DOK-4-NQO cells. Upper: Representative images of xenograft tumors. Lower: Tumor Volume and weight of xenografts in the ending point respectively, Tumor volume=(length × width²)/2. (i) The levels of PCNA, Bim 1 and Occludin in xenograft tumors were assessed by IHC, magnification, 200* and 400*, scale bars, 200 μm for 200* magnification and 100 μm for 400* magnification. The red box showed the area of interest. The above experiments were repeated 3 times. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 10

Model diagram for the role of ROS in M1-like TAMs-mediated oral carcinogenesis by exosome-transferring Thbs1 In oral keratocytes, targeted interference with ASCT2-mediated Gln uptake inhibits the synthesis of GSH and thereby induces the accumulation of ROS, which in turn upregulates Thbs1 expression. Thbs1 then activates p38, Akt and SAPK/JNK signaling to polarize M1-like TAMs by exosome-transferred pathway, which finally promotes oral carcinogenesis.