The human epidermoid carcinoma cell line, KB-3-1, was used as the parental cell line. KB-C2, an MDR cell line that is resistant to colchicine due to the overexpression of P-gp, was created by exposing KB-3-1 cells to increasing concentrations of colchicine over a period of at least 2 months (Akiyama et al., 1985; Yoshimura et al., 1989). Both cell lines were kindly provided by Dr Shinichi Akiyama (Kagoshima University, Kagoshima, Japan). KB-3-1 and KB-C2 cells and their CDK4- and CDK6-deleted sublines were cultivated with DMEM supplemented with 10% FBS and 1% penicillin/streptomycin in a humidified incubator containing 5% CO₂ at 37°C. CRISPR/Cas9 all-in-one plasmids, encoding single guide RNA (SgRNA) and Cas9, were purchased from GeneCopoeia Inc (Rockville, MD). Because KB-C2 cells are widely used for studying P-gp-mediated MDR in cancers, KB-C2 cells were used to determine the reversal of P-gp-mediated MDR.

Ribociclib was kindly provided by ChemieTek (Indianapolis, IN). Paclitaxel, colchicine, and doxorubicin were purchased form Sigma Chemical Co. (St. Louis, MO). Mouse anti-P-gp, HRP ligated or fluorescent secondary rabbit or goat-anti mouse antibodies, were purchased from Invitrogen, Thermo Fisher (Carlsbad, CA). Mouse-anti-CDK4 and CDK6 antibodies were purchased from R&D Systems (Minneapolis, MN). All other reagents were purchased from VWR International (West Chester, PA).

Exponentially growing cells were seeded into 96 well plates at 5 × 10³ cells/well. These experiments were conducted in triplicate. After 72 h of incubation, 20 μl of MTT (5 mg/ml) was added to each well. After incubation for an additional 4 h, the medium containing MTT was discarded and replaced with 150 μl of DMSO. The plates were gently shaken until the dark blue-purple crystal were completely dissolved in DMSO. The absorbance was measured at a wavelength of 490 nm, using an ELx 800 Universal Microplate Reader (Bio-Tek, Inc. Winooski, VT). The relative survival rate (%) for the cells was analyzed using the SPSS 20 program (SPSS Inc., Chicago, IL) and the survival rate - drug concentration curves were generated using Origin 9.0 software (OriginLab corporation, Northampton, MA). The concentration of drug required to inhibit cell viability by 50% (IC₅₀ value) was determined using Origin 9.0 software.

Cells were seeded into 96-well plates (5 × 10³ cells per well) and cultured overnight. The cells were incubated with ribociclib (0, 0.3, 1 and 3 µM) for 1 or 2 h, followed by incubation with gradient concentrations of colchicine and paclitaxel, which are substrates for the P-gp transporter. The IC₅₀ values of the anti-cancer drugs were determined using the MTT assay as described above.

Parental KB-3-1 cells and MDR KB-C2 cells were incubated with 9 μM of ribociclib for 2 h and co-cultured with paclitaxel or colchicine for 24–72 h. Western blot and immune-fluorescence (IF) assays were conducted to determine the effect of ribociclib on the expression of the P-gp protein. For the Western blot assay, the cells were lysed with SDS lysate reagent and separated on a gradient polyacrylamide gel (4%–20%, containing 0.1% SDS). The proteins on the SDS-PAGE gel were transferred to a PVDF membrane. After blocking with 5% milk, the membrane was washed with TBST buffer (150 mM NaCl, 10 mM Tris pH 8.0, 0.1% v/v Tween20) three times, incubated with mouse anti-P-gp antibody at 4°C for 2 h, adequately washed with TBST, and incubated with goat anti-mouse IgG-HRP (horseradish peroxidase) at RT for 2 h. The membrane was then washed with TBST four times and exposed to the SignalFire™ ECL Reagent developing reagent (Cell Singling Technology, Danvers, MA), and the results were quantified using an AI600 RGB GEL Imaging System (GE, Fairfield, CT) set for the chemiluminescence mode.

RT-PCR of ABCB1 mRNA level was performed to investigate the influence of ribociclib on the ABCB1 expression on the transcription level. Ribociclib (9 µM) showing the effect of reversing ABCB1 mediated MDR in cancer cells was added to the cell culture medium. The cells cultured without ribociclib were set as control. After 48 h of cell culture, the cells were sampled for mRNA extraction and RT-q-PCR (QuantStudio™ 5 Real-Time PCR System, ThermoFisher Scientific, CA). ABCB1 and GAPDH were the objective and internal reference genes, respectively. Primers for amplification of ABCB1 gene were: GAAAGTGAAAAGGTTG (Forward), and CTGGCGCTTTGTTCCA (Reverse). Primers for amplification of GAPDH gene were ATTGACCTCAACTACA (Forward), and AGAGATGATGACCCTT (Reverse). Expression deviation was calculated according to the formular: Ratio _{Sample/Control} = 2^−ΔΔCT, where C_T value was automatically calculated by QuantStudio™ Design and Analysis Software v1.3.1 according to the ΔRn-Cycle curve.

The ATPase activity of P-gp was determined using the PREDEASY ATPase Kits (SOLVO Biotechnology, Szeged, Hungary), according to the manufacturer’s protocol (Ambudkar, 1998). Briefly, the membranes (20 μg) were incubated in assay buffer (50 mM of MES at pH 6.8, 50 mM of KCl, 5 mM of sodium azide, 2 mM of EGTA, 2 mM of DTT, 1 mM of ouabain and 10 mM of MgCl₂). The membrane vesicles from Sf9 cells were provided by the manufacturer, expressing high levels of human ABCB1, were incubated with ribociclib (0.05, 0.1, 0.2, 0.4, 0.6, 0.8, 1, 5, 10 and 20 µM) for 3 min. ATP hydrolysis was initiated by adding 5 mM of Mg-ATP and the reaction was terminated using a 5% SDS solution. Subsequently, the light absorption was measured at 800 nm using a Bio-Rad SmartSpec 30,000 UV/Vis spectrophotometer (Bio-Rad Laboratories, Hercules, CA).

Because doxorubicin is a fluorescent substrate that can be extruded from cancer cells by the P-gp transporter (Mi et al., 2010), its intracellular accumulation was used to determine if ribociclib can directly inhibit the efflux function of the P-gp transporter, thereby increasing the accumulation of doxorubicin. The parental and MDR cells were incubated with 9 µM of ribociclib for 1 h. Doxorubicin (0.2 µM) was added and co-incubated with the cells for 2 h. This incubation period and the relatively low concentration of doxorubicin were used to maintain cell viability. The cells were gently washed with PBS buffer three times and lysed using SDS lysate reagent (50 mM, pH 8.0 Tris, 1% SDS, 2 mM of sodium pyrophosphate, 25 mM of β-glycerophosphate, 1 mM of EDTA, 1 mM of Na₃VO₄ and 0.5 μg/ml of leupeptin). The fluorescence intensity, an indicator of doxorubicin accumulation, was determined using a SynergyTM 4 Multi-Mode Microplate Reader (Bio Tek Instruments, Inc. VT, Excitation/Emission: 450/550 nm).

For the long-term evaluation of the efflux of doxorubicin, cells were seeded at 3 × 10³ cells per well, cultured for 8 h, and incubated with 9 μM of ribociclib and doxorubicin (1 µM for the MDR KB-C2 cells and 0.1 µM for drug-sensitive KB-3-1 cells to maintain MDR and cell viability above 60%) for 72 h. Because the membrane of the dead or severely apoptotic cells was damaged, which could lead to the inaccurate quantification of the intercellular doxorubicin levels, the distribution of doxorubicin was determined in triplicate, using fluorescent microscopy.

We first determined the transcripts for the P-gp and CDK4/6 proteins in the KB-C2 cell line. According to the open reading frame (ORF) sequences detected in the major transcripts of P-gp in KB-C2 cell line, the structure was achieved by structure modelling using a Swiss model platform (https://swissmodel.expasy.org/). The interactions between ribociclib and P-gp, CDK4 or CDK6 were calculated using HEX 8.0 software (LORIA/Inria Nancy Research Centre in Nancy, France), based on the most stable calculated structures of the complexes by molecular docking (Kaczor et al., 2013). PyMOL (version 1.8. x) was used to analyze the data and determine the most stable complex structures, binding positions (such as residues or chemical groups) and interactions.

We conducted experiments to determine if ribociclib could increase the efficacy of colchicine (i.e., reversing resistance to colchicine) in MDR KB-C2 cancer cells. As previously reported, colchicine potently decreased (IC₅₀ = 13.15 nM) the viability of the colchicine-sensitive KB-3-1 cancer cells (Figure 1). In contrast, the IC₅₀ value for colchicine was 4.98 µM in the colchicine-resistant KB-C2 cancer cells, indicating these cells were almost three orders of magnitude more resistant to colchicine compared to the parental KB-3-1 cancer cells (Figure 1).

Ribociclib non-significantly decreased the IC₅₀ value of colchicine in the KB-3-1 cancer cells, whereas in the colchicine-resistant KB-C2 cancer cells, 9 µM of ribociclib significantly decreased the IC₅₀ of colchicine in KB-C2 cancer cells (Figure 1). These results suggested that the resistance of KB-C2 cancer cells to colchicine, which was mediated by the overexpression of the P-gp transporter³⁶, could be partially attenuated by ribociclib (Figure 1).

It is possible that ribociclib increases the efficacy of colchicine in KB-C2 cancer cells by affecting the expression of the P-gp protein. Therefore, we used Western blotting to determine the effect of ribociclib on P-gp expression. Our results indicated the incubation of KB-C2 cancer cells with 9 µM (a non-toxic concentration) of ribociclib for 72 h produced a significant decrease in the expression of P-gp protein levels compared to cells incubated with vehicle (Figure 2A). The remaining P-gp transporters expressed by the KB-C2 cells most likely mediated the lowered drug resistance produced by P-gp (Figure 2A). In contrast, P-gp protein expression was not significantly altered in KB-3-1 cancer cells incubated with 9 µM of ribociclib compared to cells incubated with vehicle (Figure 2A).

Treatment with ribociclib remarkably downregulated ABCB1 transcription in the KB-C2 cells (Figure 2B). Approximately 5.6% of the ABCB1 mRNA amounts was detected in the cells treated with ribociclib (9 µM), as compared with that detected in the cells without ribociclib treatment. This conclusion was coherent with the that supported by Western blot (Figure 2B). This phenomenom implied that the ribociclib down regulated P-gp at both the translational and transcriptional levels.

Although ribociclib showed decent inhibition activity when it binds CDK4 or CDK6 (Supplementary Figures S1–S4), we also performed docking analysis experiments to determine if ribociclib interacted with the P-gp transporter and if so, what chemical interactions were involved. Our results indicated that a strong interaction between P-gp and ribociclib existed. Docking studies indicated that ribociclib interacted with the drug-substrate binding site of P-gp, and had a Docking Score/Etotal (Eforce + Eshape, Kalaiselvi et al., 2015) of - 271.06. The molecular modeling indicated electrostatic interactions between N,N-dimethylamide cluster (positively charged with a proton at physiological conditions) in ribociclib and E273 and E1129 (with negative charges), in a trough-like structure between TMDs and NBDs in P-gp, which is adjacent to the interphase of the inner membrane (Figure 3, Figure 4). Since ribociclib was estimated to bind in a non-representative, drug-substrate pocket of P-gp, it was unknown as to whether this interaction results in a change in the efflux function of P-gp, which is frequently associated with a change in the ATPase activity of P-gp and the intracellular accumulation of antitumor drugs (Zhang et al., 2020). Therefore, we conducted experiments to determine if ribociclib affected ATPase activity.

Although ribociclib has been reported to be highly efficacious in inhibiting CDK4/6 (Kwapisz, 2017; Tripathy et al., 2017), it remains to be elucidated whether ribociclib interacts with P-gp. Therefore, we conducted experiments to determine if ribociclib 1) interacts directly with P-gp and alters the efflux activity and 2) alters ATPase activity of human P-gp in the membrane vehicles.

Studies have shown that P-gp transporter hydrolyzes ATP, which is involved in drug efflux (Kim and Chen, 2018) and ATPase activity can be stimulated or inhibited by various P-gp substrates (Feng et al., 2020; Wu et al., 2020). It has been postulated that the stimulation of the P-gp ATPase activity by an experimental compound suggests that it is interacting with the transporter at the drug-substrate binding site (Zhang et al., 2020). Therefore, we determined the effect of various concentrations of ribociclib on P-gp ATPase activity. The incubation of membrane vesicles from sf9 insect cells (which express high levels of P-gp) with ribociclib (0.05–20 µM) produced a maximal increase of 3.5-fold in the basal activity of the P-gp transporter ATPase and the EC₅₀ value was 0.04 μM (Figure 5A). The stimulation of P-gp transporter ATPase activity by ribociclib suggests that it may interact with the transporter at the drug-substrate binding site, although this remains to be definitively proven.

It is possible that ribociclib reverses the drug resistance of KB-C2 cancer cells by inhibiting the efflux function of P-gp. Therefore, we determined the effect of ribociclib (using a 2 h incubation period) on the intracellular accumulation on doxorubicin, a substrate for the P-gp transporter, in KB-C2 cancer cells. As previously reported (Zhang et al., 2020), the accumulation of doxorubicin was significantly greater in the parental cell line, KB-3-1, compared to MDR KB-C2 cells, which overexpress the P-gp transporter (Figure 5B) (Akiyama et al., 1985; Yoshimura et al., 1989). Doxorubicin accumulation was significantly increased in KB-C2 cancer cells incubated with 9 µM of ribociclib compared to cells incubated with vehicle (Figure 5B). In contrast, doxorubicin accumulation in the parental KB-3-1 cells, which do not overexpress the P-gp transporter, was not significantly altered by 9 µM of ribociclib. The relationship between the interaction of ribociclib with the P-gp transporter and its inhibition of drug efflux and increase in ATPase activity is summarized in Figure 5C.

These results indicated that ribociclib increases the ATPase activity of P-gp and inhibits the drug-efflux function of P-gp, suggesting that ribociclib can interact with P-gp directly, which may contribute to the reversal of P-gp-mediated MDR in KB-C-2 cells.

These studies were conducted to ascertain if the prolonged incubation (72 h) of KB-3-1 and KB-C2 cancer cells with ribociclib would inhibit the efflux of doxorubicin. The incubation of the parental KB-3-1 cells with 9 µM of ribociclib for 72 h did not significantly alter the intracellular accumulation of doxorubicin compared to cells incubated with vehicle (Figure 6). We concluded that the presence of ribociclib had no effect on the accumulation of DOX within KB-3-1 cells, which did not express P-gp and could not extrude DOX efficiently (Zhang et al., 2022). In contrast, doxorubicin accumulation was significantly increased in the KB-C2 cells incubated with 9 µM of ribociclib compared to cells incubated with vehicle (Figure 6). Thus, the accumulation of doxorubicin in the KB-C2 cells that overexpress the P-gp transporter is significantly increased after 2 or 72 h of incubation with 9 µM of ribociclib, suggesting that ribociclib increases the intracellular accumulation of doxorubicin by inhibiting the efflux function of P-gp in KB-C2 cells.