Human retinoblastoma (RB) cell lines Y-79 and SO-Rb50 were obtained from the Type Culture Collection of the Chinese Academy of Sciences and the Zhongshan Ophthalmic Center of Sun Yat-sen University, respectively. Both cell lines were cultured in RPMI 1640 (Sigma-Aldrich) supplemented with 10% FBS (HyClone, Cytiva) and penicillin (50 U/ml; Sigma-Aldrich) streptomycin (50 μg/ml; Sigma-Aldrich) in a 37°C, 5% CO₂ incubator.

Drug-resistant RB cells were established as previously described with modifications (3,13). In brief, Y-79 and SO-Rb50 cells were treated with slowly increasing doses of carboplatin until cells exhibited exponential growth at high drug concentration. Cells were treated with the initial carboplatin concentration of 0.05 µM, and the treatment dose was doubled every 10 days, for 100 days. The cells surviving carboplatin treatment were then cultured in drug-free medium (complete culture medium without carboplatin) for 2 weeks before downstream tests. The carboplatin-resistant Y-79 and SO-Rb50 cells were designated as Y-79/CR and SO-Rb50/CR, respectively.

The relative RNA level of RNF6 in RB cells was determined by RT-PCR, as previously described with modifications (7,14). In brief, total RNA was first extracted from Y-79, Y-79/CR, SO-Rb50 and SO-Rb50/CR cells using RNeasy Mini Kit (Qiagen), according to the manufacturer’s instructions. Then, extracted RNA was reverse transcribed into cDNA with TransScript® First-Strand cDNA Synthesis SuperMix (TransGen) according to the manufacturer’s instructions. The RNF6 mRNA level was then semi-quantified using a SYBR Green-based RT-PCR method on a Bio-Rad CFX Connect platform. The following primer pairs were used: RNF6, forward 5′-AGA​AGA​TGG​CAG​CAA​GAG​CG-3′ and reverse 5′-TCA​AGT​CAG​GCT​GAG​ATG​CTA​GT-3′; β-actin, forward 5′-CAC​CAA​CTG​GGA​CGA​CAT-3′ and reverse 5′-ACA​GCC​TGG​ATA​GCA​ACG-3′. The following thermal cycling protocol was used: 95°C, 1 min; 40 cycles of 95°C, 10 s and 60°C, 30 s. The relative RNF6 mRNA level was calculated with the 2^−ΔΔCt using β-actin was an internal control.

The RNF6 expression plasmid (p-RNF6) was constructed as previously described with modifications (15). In brief, total RNA was prepared from Y-79 cells using RNeasy Mini Kit (Qiagen) and then reverse-transcribed into cDNA with TransScript® First-Strand cDNA Synthesis SuperMix (TransGen) according to the manufacturer’s instructions. The full-length RNF6 gene was then applified by PCR using the cDNA as template. The primers used for PCR were: forward 5′-CCC​GGA​ATT​CAT​GAA​TCA​GTC​TAG​ATC​GAG​ATC​AG-3′ and reverse 5′-AAA​TAT​GCG​GCC​GCT​TAC​CCA​TTG​TTT​GCT​ATG​TTA​GAC​CC-3′. The applified RNF6 gene was inserted into expression vector pcDNA3.1 between EcoRI and NotI restriction sites. The gene sequence was verfied by Sanger sequencing.

The knockdown of RNF6 expression in Y-79/CR and SO-Rb50/CR cells was done with siRNA transfection, while the overexpression of RNF6 in Y-79 and SO-Rb50 cells was done with p-RNF6 plasmid transfection (100 ng/well was used for 96-well plates and 500 ng/well was used for 24-well plates, unless otherwise indicated). For all the transfections, cells were seeded at 1 × 10⁴/well for 96-well plates and 5 × 10⁴ for 24-well plates, 1 day before transfection. Both siRNA and plasmid were transfected using Lipofectamine 3,000 (Thermo Fisher Scientific), according to the manufacturer’s instructions. The following siRNAs (1 pmol/well was used for 96-well plates and 5 pmol/well was used for 24-well plates, unless otherwise indicated) were used in the current study: human RNF6 siRNA (HSS165268, Thermo Fisher Scientific), human p65 siRNA (sc-29410, Santa Cruz) and negative control siRNA (4404021, Thermo Fisher Scientific). The transfection complexes were removed and replaced with fresh culture medium after 6 h.

Y-79/CR and SO-Rb50/CR cells, and their parental cells (Y-79 and SO-Rb50) were treated with either ascendent doses of carboplatin or a fixed dose (50 µM) of carboplatin, vincristine, and etoposide for 24 h, and then cell viability was assessed. If cells were transfected with siRNA or plasmid and also treated with drugs, drugs were introduced at 24–26 h after transfection for another 24 h.

Signaling pathway inhibition was done using specific pathway inhibitors, as previously described with modifications (16). In brief, cells were first seeded in 24-well plates, and then specific signaling pathway inhibitors were added into the culture for 24 h. The following signaling pathway inhibitors were used in the current study: BAY-11–7,082 (10 μM; NF-κB inhibitor), CCT251545 (10 μM; Wnt inhibitor), FLLL32 (5 μM; JAK2/STAT3) inhibitor, PD98059 (20 μM; MEK inhibitor), SB203580 (10 μM; p38 inhibitor) and SP600125 (10 μM; JNK inhibitor). All the inhibitors were purchased from Selleck and used according to the manufacturer’s instructions.

Cell viability was assessed with MTT assay (R&D systems), as previously described with modifications (17,18). In brief, after 24 h of drug treatment, cells were first treated with 10-fold diluted MTT reagent for 4 h, and then Detergent Reagent was added, and the plate was further incubated for another 4 h in the dark. After incubation, the plate was read on a microplate reader, with the testing wavelength of 570 nm and the reference wavelength of 650 nm. The cell viability was calculated using cells without treatment being considered as 100% viable. Drug resistance index (RI) was defined as the ratio of IC50 of the drug-resistant cells divided by the IC50 of its parental cells.

Western blot was performed as previously described with modifications (18). In brief, cells were lysed with RIPA cell lysis buffer supplemented with protease inhibitors and phosphatase inhibitors (Santa Cruz). Lysed cell samples were centrifuged at 10,000 g for 10 min at 4°C, and cleared cell lysate was then mixed with loading buffer (Beyotime) and loaded onto a 10% SDS-PAGE gel. After electrophoresis, the resolved proteins were transferred from the gel onto a PDVF membrane. The membrane was subsequently blocked with 5% non-fat milk for 1 h at room temperature. After blocking, the membrane was sequentially incubated with primary antibodies and HRP-conjugated secondary antibodies over night at 4°C and 1 h at room temperature, respectively. Following incubations, the membrane was extensively washed with PBST and the immuno-bands were visualized with ECL substrate (Beyotime). Band density was analyzed using ImageJ 1.8.0(19). The following primary antibodies were used: rabbit anti-human RNF6 antibody (ab204506; Abcam), rabbit anti-human p65 antibody (10745-1-AP, Proteintech), mouse anti-human STAT3 antibody (ab119352; Abcam), rabbit anti-human p-STAT3 antibody (9,145; Cell Signalling Technology) and mouse anti-human β-actin antibody (66009-1-Ig, Proteintech). The following secondary antibodies were used: HRP-conjugated goat anti-rabbit IgG (H + L) (SA00001-2; Proteintech) and HRP-conjugated goat anti-mouse IgG (H + L) (SA00001-1; Proteintech). All primary antibodies were used at 1:1,000 dilution and all secondary antibodies were used at 1:20,000 dilution.

All data were expressed as mean ± standard deviation (SD) and all statistical analyses were performed with GraphPad Prism 9.2 (GraphPad). Mann-Whitney test was used for comparisons between two groups and Kruskal-Wallis test followed by Dunn’s multiple comparisons were used for comparing three or more groups. For all the tests, a p value less than 0.05 was considered statistically significant.

The sensitivity of the carboplatin-resistant cells and parental RB cells to carboplatin was assessed by cell viability assay. Our data showed that carboplatin-resistant RB cells could survive much higher dose of carboplatin treatment comparing to their parental cells (Figures 1A,B). The calculation of the IC50s of carboplatin on these cells further revealed that the IC50s of carboplatin on Y-79 and Y-79/CR were 7.63 µM and 48.39 µM, respectively. This represented a resistance index (RI) of 6.34 on Y-79/CR cells (Table 1). Similarly, the IC50s of the same drug on SO-Rb50 and SO-Rb50/CR were 13.92 µM and 71.11 µM, representing a RI of 5.11 on SO-Rb50/CR (Table 1). These data here indicate that the carboplatin-resistant Y-79/CR and SO-Rb50/CR cells were successfully established.

To test the role of RNF6 in RB drug resistance, the expression of RNF6 was then measured in Y-79/CR and SO-Rb50/CR and their parental cells. RT-PCR analysis revealed that RNF6 mRNA level was significantly increased in both Y-79/CR and SO-Rb50/CR, comparing to the parental cell line Y-79 and SO-Rb50, respectively (Figures 1C,D). In accordance, our western blot data showed that the RNF6 protein level was also considerably increased in the carboplatin-resistant RB cells comparing to the parental cells (Figures 1E,F). Together, our data here indicate that RNF6 is upregulated in carboplatin-resistant RB cells.

To further investigate the importance of RNF6 in RB drug resistance, RNF6 was knocked down with specific siRNA in carboplatin-resistant Y-79/CR and SO-Rb50/CR cells, and then the cell sensitivity to carboplatin was assessed. As shown in Figures 2A,B, the knock-down of RNF6 in both Y-79/CR and SO-Rb50/CR cells rendered the cells more sensitive to carboplatin, and the cell sensitivity to the drug appeared to be RNF6 siRNA dose dependent. The data implies that RNF6 expression might enhance carboplatin resistance in RB cells.

To further confirm that RNF6 expression was associated with RB drug resistance, RNF6 overexpression was performed in drug-sensitive Y-79 and SO-Rb50 cells, and then cell sensitivity to carboplatin was assessed. Our data showed that RNF6 overexpression significantly enhanced cell tolerance to carboplatin in both Y-79 and SO-Rb50 cells and such enhanced drug tolerance increased with p-RNF6 plasmid dose (Figures 3A,B). Taken together, these data further confirmed that RNF6 promotes carboplatin resistance in RB cells, implying that RNF6 may serve as a biomarker for RB carboplatin resistance.

We next further explored the possible mechanism underlying RNF6-associated carboplatin resistance in RB cells. In this experiment, Y-79/CR was used a carboplatin-resistant RB cell model. To check whether RNF6 induced carboplatin resistance in RB cells through certain signaling pathways, Y-79/CR cells were first treated with a panel of specific inhibitors targeting common drug resistance pathways and then the cell sensitivity to carboplatin was assessed. As shown in Figure 4A, of the tested 6 pathway inhibitors, only FLLL32, a specific inhibitor for JAK2/STAT3 pathway, significantly decreased the tolerance of Y-79/CR cells to carboplatin treatment, while the other 5 inhibitors showed no apparent impact on cell survival. To further confirm the involvement of JAK2/STAT3 pathway in carboplatin resistance in RB cells, Y-79/CR cells were first transfected with or without STAT3 siRNA or NF-κB siRNA, and then cell sensitivity to carboplatin was determined. As shown in Figure 4B, knock-down of STAT3 with siRNA dramatically decreased the cell viability upon carboplatin treatment, while NF-κB knock-down or control siRNA transfection did not induce apparent change in cell viability. These data together indicate that JAK2/STAT3 pathway is involved in carboplatin resistance in RB cells.

The involvement of JAK2/STAT3 pathway in RNF6-induced carboplatin resistance in RB cells was also investigated. Y-79/CR cells were first mock-treated, or treated with control siRNA, RNF6 or p65 siRNA, and the expression of p65, STAT3 and p-STAT3 was determined by western blot. Our data showed that RNF6 knock-down with RNF6 siRNA did not change the expression of STAT3, but inhibited the phosphorylation of this protein, as evidenced by the decrease of p-STAT3 (Figure 4C). By contrast, p65 knock-down did not show impact on either STAT3 or p-STAT3. These data indicate that RNF6 promotes carboplatin resistance in RB cells through JAK2/STAT3 pathway.

To test if RNF6 upregulation could render RB cells resistant to anti-cancer drugs besides carboplatin, Y-79 and SO-Rb50 cells were first overexpressed with RNF6 and then their sensitivities to carboplatin, vincristine and etoposide were assessed. Similar to the data obtained from carboplatin, these two RB cell lines with RNF6 overexpression demonstrated significantly increased tolerance to these drugs (Figures 5A,B). These data indicate that RNF6 can promote RB cells resistant to multiple anti-cancer drugs and may serve as a biomarker and treatment target for drug resistance in RB cells.