Exosomal miR-196a derived from cancer-associated fibroblasts confers cisplatin resistance in head and neck cancer through targeting CDKN1B and ING5

Abstract

Background: Cisplatin resistance is a major challenge for advanced head and neck cancer (HNC). Understanding the underlying mechanisms and developing effective strategies against cisplatin resistance are highly desired in the clinic. However, how tumor stroma modulates HNC growth and chemoresistance is unclear.

Results: We show that cancer-associated fibroblasts (CAFs) are intrinsically resistant to cisplatin and have an active role in regulating HNC cell survival and proliferation by delivering functional miR-196a from CAFs to tumor cells via exosomes. Exosomal miR-196a then binds novel targets, CDKN1B and ING5, to endow HNC cells with cisplatin resistance. Exosome or exosomal miR-196a depletion from CAFs functionally restored HNC cisplatin sensitivity. Importantly, we found that miR-196a packaging into CAF-derived exosomes might be mediated by heterogeneous nuclear ribonucleoprotein A1 (hnRNPA1). Moreover, we also found that high levels of plasma exosomal miR-196a are clinically correlated with poor overall survival and chemoresistance.

Conclusions: The present study finds that CAF-derived exosomal miR-196a confers cisplatin resistance in HNC by targeting CDKN1B and ING5, indicating miR-196a may serve as a promising predictor of and potential therapeutic target for cisplatin resistance in HNC.

Keywords: CDKN1B; Cancer-associated fibroblasts; Cisplatin resistance; Exosome; Head and neck cancer; ING5; miR-196a.

Fig. 1

CAFs derived from HNC are innately resistant to cisplatin. a Immunofluorescence staining for α-SMA, FAP, and FSP1 expression of NFs and CAFs (scale bar, 20 μm). b Western blot analysis of α-SMA, FAP, and FSP1 protein levels in six paired NFs and CAFs. c NFs, CAFs, and HNC cells were treated with or without 10 μM cisplatin for 8 days, and cell viability was measured to obtain percent cell survival and was normalized to that of control cells. d NFs, CAFs, and HNC cells were treated with or without 10 μM cisplatin for 24 h, and cell viability was detected to calculate the percentage of proliferating cells during cisplatin treatment. e The percentage of surviving CAFs (CAF3) and cisplatin-resistant HN4-res cells, which exhibit similar proliferation retention rates, upon 10 μM cisplatin treatment for 8 days. f, g Western blot analysis of MRP2, ATP7B, CTR1, XIAP, ERCC1, ERCC4, GSTK1, and Bcl-2 protein levels in NFs, CAFs, CAL 27, SCC-25, HN4, and HN4-res cells. The band intensity was assessed. (ns, no significant difference; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001)

Fig. 2

CAF-derived exosomes increase HNC cell proliferation and cisplatin resistance. a CAL 27 and HN4 cells were grown in CAF-CM or control CM for 6 days, and cell viability was examined. b CAL 27 and HN4 cells were grown in CAF-CM or control CM for 6 days, and then MTT assays were performed to detect the cisplatin response of these cells. c CAL 27 and HN4 cells were grown in control CM, CAF-CM, or exosome-depleted CAF-CM for 6 days, and cell viability was examined. MTT assays were performed to assess the tolerance of these cells to cisplatin. d HNC cells were co-cultured with DMSO-treated HNC cells, DMSO-treated CAFs, or GW4869-treated CAFs for 6 days, and then HNC cell viability was assessed. The survival percentage of co-cultured HNC cells was measured by MTT assay upon cisplatin treatment. e NanoSight particle-tracking analysis of size distribution and the number of exosomes from CAFs or cisplatin-treated CAFs (10 μM). f NanoSight particle-tracking analysis of the number of exosomes from NFs, CAFs, or HNC cells with or without cisplatin treatment (10 μM). g Exosomal protein concentration in CM from NFs, CAFs, or HNC cells with or without cisplatin treatment (10 μM). h CAL 27 and HN4 cells were incubated with DiO-labeled exosomes (25 μg/mL) from CAFs for 24 h, and the green exosome signal was detected by confocal microscopy (scale bar, 20 μm). i Flow cytometric analysis of DiO-positive CAL 27 or HN4 cells after incubating with DiO-labeled exosomes (25 μg/mL) from CAFs for the indicated time. j HNC cells were treated with exosomes (25 μg/mL) from HNC cells, CAFs, or cisplatin-treated (10 μM) CAFs for 6 days. Cell viability was measured, and MTT assays were performed to evaluate the cisplatin response of these cells. k HNC cells were treated with exosomes (25 μg/mL) from HNC cells, cisplatin-treated (10 μM) HN4-res cells, or cisplatin-treated (10 μM) CAFs for 6 days. The viability and survival percentage of these cells were measured by MTT assays. (NT, without cisplatin treatment; CT, cisplatin treatment; ns, no significant difference; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001)

Fig. 3

Exosomal transfer of miR-196a from CAFs to HNC cells. a miR-196a expression in NFs, CAFs, normal oral epithelial cells (titled normal), primary cancer cells, and HNC cell lines was analyzed using real-time PCR. b CAL 27 and HN4 cells were incubated with control CM, CAF-CM, or exosome-depleted CAF-CM for 24 h. The miR-196a expression level in these cells was determined using real-time PCR. c CAL 27 and HN4 cells were co-cultured with DMSO-treated HNC cells, DMSO-treated CAFs, or GW4869-treated CAFs for 24 h. The miR-196a expression level was then detected in HNC cells using real-time PCR. d CAFs transiently transfected with Cy3-tagged miR-196a (Cy3-miR-196a) were co-cultured with CAL 27 or HN4 cells for 48 h. Fluorescence microscopy was used to detect the red fluorescent signals in HNC cells (scale bar, 20 μm). e Real-time PCR analysis of miR-196a expression in CAF-CM treated with RNase A (2 mg/mL) alone or combined with Triton X-100 (0.1%) for 20 min. f Real-time PCR analysis of miR-196a expression in exosomes, exosome-depleted CM, and whole CM derived from CAFs. g miR-196a expression in CAL 27 and HN4 cells was detected by real-time PCR at 24 h after incubation with exosomes (25 μg/mL) from HNC cells (Ctrl Exos), CAFs, or CAFs transfected with or without miR-196a mimics. h miR-196a expression in CAL 27 and HN4 cells was detected by real-time PCR at 24 h after incubation with exosomes (25 μg/mL) from HNC cells (Ctrl Exos), CAFs, and CAFs transfected with or without anti-miR-196a. i, j CAL 27 and HN4 cells were treated with the indicated exosomes (25 μg/mL) for 6 days, and the cisplatin response in these cells was determined with MTT assays. (ns, no significant difference; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001)

Fig. 4

The hnRNPA1 protein mediates miR-196a packaging into CAF-derived exosomes. a A specific interaction between the miR-196a sequence and RBP motifs was predicted through RBPDB analysis (threshold 0.7). b Western blot and real-time PCR results showing ZRANB2, hnRNPA1, and ELAVL1 expression levels in CAFs at 48 h after transfection with specific siRNAs. c miR-196a expression in exosomes from CAFs transfected with specific siRNAs targeting ZRANB2, hnRNPA1, or ELAVL1 was measured using real-time PCR. d Real-time PCR analysis showing miR-196a expression in CAFs with hnRNPA1 silenced. e Western blot analysis of hnRNPA1 expression in samples derived by miRNA pulldowns performed with nuclear, cytoplasmic, or exosomal CAFs lysates and the indicated biotinylated miR-196a or mutated miR-196a; biotinylated poly(G) was used as a negative control. f RIP assays with anti-hnRNPA1 antibody (or IgG as control) were performed on the cell or exosomal lysates from CAFs. miR-196a levels in immunoprecipitated samples were determined by real-time PCR and were reported as percentages in respect to the input sample (% input). g RIP assay to determine hnRNPA1 enrichment on miR-196a relative to IgG in cytoplasmic or exosomal lysates of CAFs treated with or without cisplatin. h CAL 27 and HN4 cells were co-cultured with CAFs concurrently transfected with Cy3-miR-196a and specific siRNAs targeting hnRNPA1 for 48 h. Fluorescence microscopy was used to detect red fluorescent signals in HNC cells (scale bar, 10 μm). i Nude mice were subcutaneously xenografted with a mixture of CAL 27 cells plus CAFs transfected with hnRNPA1, sh-NC, or sh-hnRNPA1, and the tumor growth curve and tumor volumes are shown. j The distribution of miR-196a in xenograft tumors was detected using FISH assay, and miR-196a expression in tumor cells was assessed. k The hnRNPA1 mRNA expression level in 108 pairs of HNC samples and adjacent normal tissues. l Correlation analysis was performed between miR-196a expression and hnRNPA1 expression in HNC tissues (n = 108) (NT, without cisplatin treatment; CT, cisplatin treatment; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001)

Fig. 5

miR-196a regulates HNC cell proliferation and survival by promoting G1/S transition and apoptosis resistance. a Cell proliferation assays with CAL 27 cells at 48 h after transfection with miR-196a or anti-miR-196a. b Representative micrographs and quantification of EdU-incorporating cells at 48 h after transfection with miR-196a or anti-miR-196a. c MTT assay of CAL 27 cells transfected with miR-196a or anti-miR-196a for 48 h, followed by cisplatin treatment at the indicated concentration for 72 h. d Plate colony formation assay of CAL 27 cells transfected with miR-196a or anti-miR-196a with cisplatin treatment (3 μM). e The cell cycle distribution was analyzed by a flow cytometer in CAL 27 cells transfected with miR-196a or anti-miR-196a at 48 h after transfection. f, g Flow cytometric analysis of cisplatin-induced (10 μM) apoptosis in CAL 27 cells transfected with miR-196a or anti-miR-196a at 48 h after transfection. (*p < 0.05; **p < 0.01; ***p < 0.001)

Fig. 6

CDKN1B and ING5 are direct targets of exosomal miR-196a in HNC cells. a A diagram showing the predicted candidate target genes of miR-196a by Gene Ontology analysis. b CDKN1B and ING5 mRNA expression in CAL 27 and HN4 cells transfected with miR-196a or anti-miR-196a at 48 h after transfection. c CDKN1B and ING5 mRNA levels in CAL 27 and HN4 cells at 48 h after incubation with exosomes (25 μg/mL) from HNC cells (Ctrl Exos) and CAFs. d, e Correlation analysis was performed between miR-196a expression and CDKN1B or ING5 expression in HNC tissues (n = 108). f Predicted miR-196a target sequences in the 3′ UTRs of CDKN1B and ING5 genes. g Relative CDKN1B or ING5 reporter activities in 293T cells co-transfected with miR-196a and luciferase reporters. h The effects of anti-miR-196a on CDKN1B or ING5 reporter luciferase activity in 293T cells. i The effects of CAF-derived exosomes (25 μg/mL) on CDKN1B or ING5 reporter luciferase activity in 293T cells. j A diagram showing the program for the immunoprecipitation of miRNA targets. k Interaction of target transcripts with miR-196a. CAL 27 or HN4 cells were transfected with biotinylated miR-NC or miR-196a for 48 h. Levels of CDKN1B and ING5 mRNA in the materials pulled down by biotin-miR-196a were analyzed by real-time PCR and normalized to β-actin. l CAL 27 and HN4 cells were transfected with miR-196a or anti-miR-196a or were incubated with CAF-derived exosomes (25 μg/mL) for 48 h. p27 and ING5 expression in the indicated cells was detected by western blot analysis. m The results from the cell cycle and cisplatin-induced cell apoptosis analyses in HNC cells transfected with miR-NC, miR-196a plus control vector, miR-196a plus CDKN1B plasmid, or miR-196a plus ING5 plasmid at 48 h after transfection. (ns, no significant difference; *p < 0.05; **p < 0.01; ***p < 0.001)

Fig. 7

miR-196a accelerates cisplatin resistance in HNC cells via CDKN1B and ING5 downregulation. a Western blot showing p27, p21, CDK2, CDK4, Cyclin D1, and Cyclin E1 protein levels in CAL 27 and HN4 cells at 48 h after transfection with siRNA specific for CDKN1B. b Western blot showing ING5, Bcl-2, Bax, full-length Caspase 3, cleaved Caspase 3, full-length PARP, and cleaved PARP protein levels in CAL 27 and HN4 cells at 48 h after transfection with siRNA specific for ING5. c Left: western blot showing p27 and ING5 protein levels in CAL 27 and HN4 cells at 48 h after transfection with siRNAs specific for CDKN1B or ING5. Right: MTT assay showing the cisplatin response of CAL 27 and HN4 cells at 48 h after transfection, as indicated. d Left: western blot showing p27 and ING5 protein levels in CAL 27 and HN4 cells at 48 h after concurrent transfection with miR-196a and an exogenous expression vector (CDKN1B or ING5). Right: cisplatin response in CAL 27 and HN4 cells at 48 h after transfection, as indicated. e Left: western blot showing p27 and ING5 protein levels in CAL 27 and HN4 cells at 48 h after transfection with anti-miR-196a and siRNAs specific for CDKN1B or ING5. Right: cisplatin response of CAL 27 and HN4 cells at 48 h after transfection, as indicated. f Left: western blot showing p27 protein levels in CAL 27 and HN4 cells at 48 h after transfection with miR-196a and an exogenous expression vector (CDKN1B or CDKN1B 3′ UTR). Right: cisplatin resistance in CAL 27 and HN4 cells at 48 h after transfection, as indicated. g Left: western blot showing ING5 protein levels in CAL 27 and HN4 cells at 48 h after transfection with miR-196a and an exogenous expression vector (ING5 or ING5 3′ UTR). Right: cisplatin resistance in CAL 27 and HN4 cells at 48 h after transfection, as indicated. (ns, no significant difference; *p < 0.05; **p < 0.01; ***p < 0.001)

Fig. 8

CAF-derived exosomal miR-196a enhances cisplatin resistance in HNC cells in vivo. a CAL 27 cells with or without CAFs were used to establish a xenograft mouse model. The mice were intraperitoneally injected with cisplatin (4 mg/kg, every 4 days) with or without GW4869 (2 mg/kg, every 2 days). The tumor growth curve, tumor volumes, and tumor weights are shown. b CAL 27 cells were stably transfected with or without miR-196a, and CAFs were stably transfected with miR-196a or anti-miR-196a. Nude mice were subcutaneously xenografted with pre-transfected CAL 27 cells or CAFs as indicated. The mice were intraperitoneally injected 5 times with cisplatin (4 mg/kg, every 4 days). The tumor growth curve, tumor volumes, and tumor weights are shown. c Upregulated miR-196a levels were correlated with increased tumor size, lymph node metastasis, and advanced tumor stage in HNC tissues. d Real-time PCR analysis showing miR-196a expression in HNC tissues from chemosensitive patients (n = 20) and chemoresistant patients (n = 20). e Kaplan-Meier analysis of overall survival. Compared with patients with low miR-196a expression, patients with high miR-196a expression had a significantly lower overall survival rate. f Plasma exosomal miR-196a levels were detected using real-time PCR with plasma from HNC patients and healthy donors. g Real-time PCR analysis of exosomal miR-196a in matched plasma from HNC patients pre- and postoperation (n = 40). h Real-time PCR analysis of exosomal miR-196a expression in the plasma of chemosensitive patients (n = 20) and chemoresistant patients (n = 20). i Kaplan-Meier analysis of overall survival in 74 HNC patients from high and low miR-196a groups, according to the median exosomal miR-196a level in pre-therapy plasma. j ROC curve analysis of plasma exosomal miR-196a expression for discriminating the chemoresistant group (n = 20) from the chemosensitive group (n = 20). AUC, area under the curve. k A proposed model illustrating the modulatory role of CAF-derived exosomal miR-196a in regulating HNC cell proliferation and cisplatin resistance. (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001)