Targeting cyclin-dependent kinase 6 partially reverses drug resistance in ripretinib-resistant gastrointestinal stromal tumor cells in vitro

Abstract

BACKGROUND

The development of drug resistance remains a major challenge in clinical practice for patients with gastrointestinal stromal tumors (GISTs).

AIM

To investigate whether a cyclin-dependent kinase 6 (CDK6) inhibitor can partially reverse ripretinib-resistant GIST cells and to explore the underlying mechanisms.

METHODS

Ripretinib-resistant GIST cell lines were established through continuous exposure to increased concentrations of ripretinib. The effects of a CDK6 inhibitor, alone or in combination with ripretinib, on cell viability were evaluated using the Cell Counting Kit-8. Wound healing and invasion indices were used to assess cell migration and invasion capabilities. Cell cycle distribution and apoptosis were analyzed by flow cytometry. The expression levels of genes and proteins related to the cell cycle, apoptosis, and drug resistance pathways were measured by real-time quantitative polymerase chain reaction and western blot, respectively.

RESULTS

CDK6 was significantly upregulated in ripretinib-resistant GIST cells at both the mRNA and protein levels. The CDK6 inhibitor palbociclib had no significant effect on parental GIST cells but induced G1 phase arrest, promoted apoptosis, and suppressed proliferation in resistant cells. Its combination with ripretinib produced a synergistic effect that reversed ripretinib resistance. Moreover, protein kinase B (AKT) expression and phosphorylation were elevated in resistant cells, and AKT inhibition significantly reduced CDK6 expression. The CDK6 inhibitor also modulated the expression of apoptosis-related proteins (B-cell lymphoma-2-associated X protein, B-cell lymphoma-2, cleaved caspase-9) and cell cycle-related proteins (Cyclin D, p21) in resistant cells.

CONCLUSION

CDK6 mediates ripretinib resistance in GIST by regulating the AKT signaling pathway. Targeting CDK6 can partially reverse ripretinib-resistant GIST cells. The combination of CDK6 inhibitors and ripretinib represents a promising novel therapeutic strategy for patients with ripretinib-resistant GIST.

Key Words: Combination therapy; Apoptosis; Drug resistance; Protein kinase B signaling pathway; Cyclin-dependent kinase 6; Ripretinib; Gastrointestinal stromal tumor

Core Tip: This study identifies cyclin-dependent kinase 6 (CDK6) as a key mediator of ripretinib resistance in gastrointestinal stromal tumors (GIST), with its overexpression driven by protein kinase B hyperactivation. CDK6 inhibitor induces G1-phase cell cycle arrest and apoptosis in ripretinib-resistant GIST cells, and combining CDK6 inhibitor palbociclib with ripretinib exerts a synergistic effect to reverse resistance, restoring the tumor cells’ sensitivity to ripretinib. This uncovers the protein kinase B/CDK6 axis as a novel target for treating ripretinib-resistant GIST, providing a new precision therapy strategy.

INTRODUCTION

Gastrointestinal stromal tumor (GIST) is the most prevalent mesenchymal neoplasm arising from the gastrointestinal tract, with an estimated annual global incidence of 10-15 cases per million individuals[1,2]. GIST is believed to originate from interstitial cells of Cajal within the gastrointestinal wall or from primitive mesenchymal stem cells. Its histological origin is closely associated with distinct molecular phenotypic features. Immunohistochemically, GIST consistently expresses CD117 and anoctamin 1 (ANO1/DOG-1), which are canonical biomarkers serving as the cornerstone for definitive diagnosis and therapeutic stratification. Before GIST was formally recognized as a distinct disease entity, the lack of specific molecular markers often led to misdiagnosis as leiomyoma, leiomyosarcoma, or other mesenchymal malignancies, resulting in suboptimal clinical decision-making and treatment determination. Currently, surgical resection remains the standard curative treatment for localized primary GIST. However, a significant proportion of patients present with metastatic disease at diagnosis, and many eventually experience postoperative tumor recurrence and distant metastasis. Notably, GIST is inherently resistant to conventional cytotoxic chemotherapy and radiotherapy, and patients with advanced, recurrent, or metastatic GIST generally have poor clinical outcomes[3-6]. The recognition and biological characterization of GIST have evolved over the past decades. It was not until 1983 that GIST was formally classified as an independent tumor entity[7]. A landmark study in 1998 first documented abundant KIT (CD117) protein expression in GIST specimens, followed by validation that oncogenic KIT mutations are present in the majority of GIST cases[8-10]. This pivotal breakthrough elucidated the core molecular mechanisms underlying GIST oncogenesis and progression, established actionable molecular targets for targeted therapy, and revolutionized the clinical management of GIST.

In 2001, imatinib received regulatory approval, marking the beginning of a new era in targeted oncological therapy. This milestone drug has significantly extended the median overall survival of patients with advanced GIST, increasing it from 20 months to 57 months[11-13]. Despite its transformative impact on GIST treatment, imatinib has limitations-approximately 50% of patients develop acquired drug resistance within 2 to 3 years of starting therapy[14-16]. This resistance poses a major clinical challenge, as it often leads to disease progression and limits the long-term efficacy of imatinib. So far, standardized treatment methods exist for stromal tumors from first-line to fourth-line therapy. Ripretinib has been approved as the standard fourth-line systemic therapy for unresectable or metastatic GIST. Its clinical approval is based on the INVICTUS trial (NCT03353753), a global, multicenter, randomized, double-blind, placebo-controlled phase III clinical study[17]. However, the clinical benefit of ripretinib is still limited, with a median progression-free survival of only 6.3 months and a median overall survival of 15.1 months in the INVICTUS trial[17]. This highlights the urgency to explore the underlying mechanisms of resistance to ripretinib and develop more effective therapeutic strategies for advanced GIST patients who progress on this agent.

CDK6 is a critical mediator of cellular transition into S phase, and is important for the initiation, growth, and survival of many cancer types[18-20]. In the context of GIST, emerging studies have begun to implicate CDK6 in the pathogenesis and progression of the disease; particularly in cases where tumors develop resistance to tyrosine kinase inhibitors (TKIs)[21]. Aberrant activation of CDK6 has been frequently observed in various malignancies, where it contributes to uncontrolled cell proliferation and tumor progression[22,23]. Emerging evidence suggests that CDK6 plays a role in modulating drug resistance in cancer cells by regulating cell cycle checkpoints, promoting DNA repair, or interacting with oncogenic signaling pathways[24,25]. For instance, overexpression of CDK6 has been associated with resistance to TKIs in several cancer models, such as chronic myeloid leukemia and non-small cell lung cancer[26,27]. However, the specific mechanisms by which CDK6 contributes to TKI resistance in GIST remain poorly understood. Recent studies have indicated that CDK6 may be upregulated in TKI-resistant GIST cell lines and patient samples, suggesting a potential link between CDK6 activity and therapeutic resistance in this disease setting[28]. CDK6 has been shown to interact with key signaling molecules involved in GIST pathogenesis, including the phosphatidyl-inositol-3-kinase (PI3K)/AKT pathways, which are known to contribute to TKI resistance when dysregulated[28]. Understanding the precise role of CDK6 in GIST, particularly its involvement in the acquisition and maintenance of TKI resistance, could therefore provide valuable insights for the development of novel therapeutic strategies to overcome this limitation.

This study aimed to investigate the potential role of CDK6 in the development of ripretinib resistance in GIST cell lines and to explore whether targeted inhibition of CDK6 could reverse this drug resistance in GIST cells. Additionally, we sought to elucidate the underlying molecular mechanisms by which CDK6 modulates ripretinib resistance, with a particular focus on its potential association with the PI3K/AKT signaling pathway. The ultimate objective of this research is to identify a novel therapeutic target and to provide a promising combinatorial treatment strategy for advanced GIST patients who experience treatment failure with ripretinib.

MATERIALS AND METHODS

Cell lines and cell culture

GIST cell lines GIST-T1 (T1), GIST-430 (430), and GIST-882 (882) were provided by Professor Wen-Bin Ou of Zhejiang Sci-Tech University. GIST-T1 harbors a KIT exon 11 mutation, while GIST-882 has a KIT exon 13 mutation. Both cell lines are sensitive to imatinib. GIST-430 harbors a primary mutation in KIT exon 11 and a secondary mutation in KIT exon 13. This cell line exhibits primary resistance to imatinib. The GIST-882 cell line was cultured in Roswell Park Memorial Institute 1640 (RPMI 1640) medium (Thermo Fisher Scientific, Waltham, MA, United States). GIST-T1 and GIST-430 cell lines were cultured in Dulbecco’s modified Eagle’s medium (Thermo Fisher Scientific, Waltham, MA, United States) supplemented with 10% fetal bovine serum, 100 μg/mL streptomycin, and 100 U/mL penicillin. All cells were maintained in a constant-temperature, constant-humidity incubator at 37 °C with 5% CO₂. Subculturing was performed when the cells reached 80%-90% confluence. Cell suspensions were seeded into new culture dishes at a 1:3 ratio. After replenishing with fresh medium, the cells were returned to the incubator at 37 °C with 5% CO₂ for continued culture. The culture medium was replaced promptly according to the cell growth rate, typically every 2-3 days.

Culture and identification of drug-resistant cell lines

GIST cell lines T1, 430, and 882 were used to establish and culture drug-resistant cell lines. Cells in the logarithmic growth phase were seeded into culture flasks for routine culture. When cell density reached approximately 90% confluence, cells were treated with 1 μM ripretinib (Sigma-Aldrich, St Louis, MO, United States) for 48 hours, followed by replacement with normal culture medium for continued growth. Once the cell density again reached approximately 90% confluence, cells were cultured starting with a low concentration of 5 nM ripretinib. Fresh medium containing the same drug concentration was replaced every 2-3 days to remove dead cells. Upon reaching full confluence, cells were digested with trypsin for subculture and continuously cultured at the same concentration for one or two passages until stable proliferation was observed. After cells adapted to the current drug concentration, the medium was replaced with a higher concentration, increasing by 1.5-2 times the previous concentration each time. This cycle of culture - medium change - subculture was repeated, with an adaptation period of 2-4 weeks for each round. When cells could grow normally in the presence of 1 μM ripretinib, were stably passaged for > 3 passages, and showed no obvious morphological abnormalities, preliminary drug-resistant cell lines were established and designated as GIST-T1 resistance index (RI; T1RI), GIST-430RI (430RI), and GIST-882RI (882RI). Ripretinib was continuously maintained at a concentration of 1 μM in the culture medium during the cultivation of resistant cells to prevent the loss of drug resistance after drug withdrawal. The RI, defined as the ratio of the half maximal inhibitory concentration of resistant cells to that of parental cells, was used to evaluate cellular drug resistance. Flow cytometry and genetic testing were performed to verify and characterize the established drug-resistant cell lines.

Cell Counting Kit-8 assay

Cell Counting Kit-8 (CCK-8; Sigma-Aldrich, St Louis, MO, United States) was used to assess cell viability. GIST cells were seeded into 96-well plates at 2000 per well, with 100 μL culture medium per well. After incubation for up to 5 days with daily measurements according to the treatment groups (palbociclib was 100 nM, and ripretinib was 40 nM), 10 μL CCK-8 solution was added to each well at the same time each day. The plates were gently shaken to ensure thorough mixing and incubated at 37 °C with 5% CO₂ for an additional 2 hours. Following incubation, the absorbance (optical density) of each well was measured at 450 nm using a microplate reader (BioTek Instruments, Winooski, VT, United States). Cells without drug treatment served as the blank control group. Each group included three replicate wells, and all experiments were independently repeated three times.

Wound healing assay

GIST cells were seeded into six-well plates. Once the cells reached full confluence, they were serum-starved for 12 hours to achieve synchronization. A straight, uniformly wide scratch was made vertically across the center of each cell monolayer using a sterile 200-μL pipette tip, ensuring the scratch extended across the entire well bottom with clear edges. Immediately after scratching, the cells were gently washed three times with prewarmed sterile phosphate-buffered saline (PBS) to completely remove detached cells and debris generated by the scratch. Serum-free medium containing the appropriate drug concentration (palbociclib was 100 nM, and ripretinib was 40 nM) was added according to the experimental treatment groups, and the plates were returned to the incubator at 37 °C with 5% CO₂ for continued incubation. Images were captured at two time points: 0 hour (immediately after scratching) and 24 hours post-scratch. At least three distinct fields of view were photographed for each sample at each time point to ensure reliable results. The width of the scratched area at each time point was measured using ImageJ software. Using the scratch width at 0 hour as the baseline, the wound healing rate at 24 hours was calculated. Differences in migratory capacity among the different cell lines were evaluated accordingly.

Invasion assay

GIST cells were seeded into 12-well plates and treated under different conditions for 48 hours (palbociclib was 100 nM, and ripretinib was 40 nM). Twenty-four hours before plating, cells were cultured in serum-free medium for 12 hours. Cells were then trypsinized, counted, resuspended in serum-free medium, and adjusted to the desired concentration. Matrigel was diluted with serum-free RPMI 1640 medium at a ratio of 1:8. A total of 50 μL diluted Matrigel was added to the upper chamber of each well, followed by incubation at 37 °C for 4-6 hours to allow gel solidification. Subsequently, 100 μL cell suspension (1 × 10⁵ cells) was added to the upper chamber, and 600 μL RPMI 1640 or Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum was added to the lower chamber. Cells were incubated for 24 hours at 37 °C in a 5% CO₂ incubator. Transwell inserts were removed, and the medium in the upper chambers was discarded. Cells were gently washed twice with PBS. Noninvaded cells and residual Matrigel on the upper surface of the membrane were carefully removed with a cotton swab. Cells were fixed in 500 μL 4% paraformaldehyde for 30 minutes. After fixation, cells were washed twice with PBS, stained with 0.1% crystal violet for 30 minutes, and washed three times with PBS. Five random, nonoverlapping fields were selected under a microscope for imaging. The number of cells that had invaded through the membrane was counted in each field, and the mean value was calculated for subsequent statistical analysis.

Cell cycle assay

Cells were seeded into 12-well plates and treated under different conditions for 48 hours (palbociclib was 100 nM, and ripretinib was 40 nM). Cells were harvested and centrifuged at 1200 rpm for 5 minutes, and the supernatant was discarded. The cells were washed twice with PBS (centrifuged at 1000 rpm for 5 minutes each time) to remove residual medium. Then, 1 mL ice-cold 70% ethanol was added, and the cells were gently pipetted and mixed, followed by fixation overnight at 4 °C. After fixation, cells were centrifuged at 1200 rpm for 5 minutes to discard the fixative and washed twice with PBS (centrifuged at 1000 rpm for 5 minutes each time). Subsequently, 500 μL of containing 50 μg/mL RNase A was added, and the cells were incubated in a 37 °C water bath for 30 minutes to eliminate RNA interference. Next, 500 μL staining solution containing 50 μg/mL propidium iodide (PI) was added and mixed gently, followed by staining in the dark at 4 °C for 30 minutes. After staining, the cell suspension was filtered through a 300-mesh nylon membrane to remove cell clumps. Flow cytometric analysis was performed using a Beckman Coulter flow cytometer (Beckman Coulter, Inc., CA, United States) with an excitation wavelength of 488 nm and emission wavelength of 610 nm. A total of 10⁴ cells were collected per sample. The proportions of cells in different phases of the cell cycle (G0/G1, S, and G2/M) were analyzed using ModFit LT software (Verity Software House, Inc., ME, United States). All experiments were independently repeated in triplicate.

Apoptosis assay

Cells in the logarithmic growth phase were seeded into six-well plates at specified densities per well. When the cells adhered and reached 70%-80% confluence, they were treated with drugs according and cultured for an additional 48 hours (palbociclib was 100 nM, and ripretinib was 40 nM). The cells were digested and collected into 1.5 mL Eppendorf tubes, followed by centrifugation at 1000 rpm for 5 minutes, and the supernatant was discarded. The collected cell pellets were washed twice with ice-cold PBS, with centrifugation at 1000 rpm for 5 minutes each time. The apoptosis detection kit (Beyotime, Shanghai, China) was equilibrated to room temperature. For the Annexin V-fluorescein isothiocyanate (FITC)/PI apoptosis detection kit, 100 μL binding buffer was added to resuspend the cell pellets, followed by 5 μL Annexin V-FITC and 5 μL PI staining solution. The mixture was gently mixed and incubated at room temperature in the dark for 15 minutes. After incubation, 400 μL binding buffer was added and mixed gently. Flow cytometric analysis was performed within 1 hour, using an excitation wavelength of 488 nm for FITC and 535 nm for PI, to determine the apoptosis rate.

Real-time quantitative polymerase chain reaction

GIST cell lines were seeded into 12-well plates and treated under various conditions for 48 h (palbociclib was 100 nM, and ripretinib was 40 nM). After treatment, the culture medium was discarded, and the cells were washed twice with PBS. Total RNA was extracted using the TRIzol method. TRIzol reagent (1 mL) was added to the cells, followed by thorough pipetting to ensure complete cell lysis, and the mixture was allowed to stand at room temperature for 5 min. Chloroform (200 μL) was added per 1 mL of TRIzol, and the mixture was vigorously shaken for 15-30 seconds, allowed to stand at room temperature for 3 minutes, and centrifuged at 12000 rpm for 15 minutes at 4 °C. The upper aqueous phase (approximately 500 μL) was carefully transferred to a new RNase-free centrifuge tube, mixed with an equal volume of isopropanol (500 μL) by gentle inversion, and incubated at -20 °C for 30 minutes. After centrifugation at 12000 rpm for 10 minutes at 4 °C, the supernatant was discarded. The RNA pellet was washed twice with 75% ethanol, each time followed by centrifugation at 7500 rpm for 5 minutes at 4 °C. The supernatant was removed, and the RNA pellet was air-dried at room temperature. Then, 50 μL of RNase-free water was added to dissolve the RNA. RNA concentration and purity were determined using a Nanodrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, United States), with the optical density 260/optical density 280 ratio ranging from 1.8 to 2.0. cDNA was synthesized from the extracted total RNA using a reverse transcription kit (Beyotime, Shanghai, China). After reverse transcription, quantitative polymerase chain reaction was performed using SYBR Green polymerase chain reaction master mix. The reaction mixture consisted of the cDNA template, forward and reverse primers, SYBR green mix, and nuclease-free water. Amplification was conducted on a real-time fluorescence quantitative polymerase chain reaction instrument under the following conditions: Pre-denaturation at 95 °C for 2 minutes, followed by 40 cycles of denaturation at 95 °C for 5 seconds and annealing/extension at 60 °C for 30 seconds. The expression levels of target genes were calculated using the 2^-ΔΔCt method, with beta-actin serving as the internal reference gene. The primer sequences were as follows: CDK6: Forward, 5’-CGTGGTCAGGTTGTTTGATG-3’; reverse, 5’-CCTCGGAGAAGCTGAAACAT-3’. AKT: Forward, 5’-AGCGACGTGGCTATTGTGAAG-3’; reverse, 5’-GCCATCATTCTTGAGGAGGAAGT-3’. Beta-actin: Forward, 5’-CATGTACGTTGCTATCCAGGC-3’; reverse, 5’-CTCCTTAATGTCACGCACGAT-3’.

Western blotting

GIST cells were seeded into 6-cm culture dishes and treated under different conditions for 48 hours (palbociclib was 100 nM, and ripretinib was 40 nM). After treatment, the culture medium was discarded, and the cells were washed twice with ice-cold PBS. Cells were lysed on ice for 30 minutes using 1 mL radioimmunoprecipitation assay lysis buffer (supplemented with protease and phosphatase inhibitors) per 10⁷ cells. The lysate was transferred to a 1.5-mL centrifuge tube and centrifuged at 12000 rpm for 15 minutes at 4 °C, and the supernatant was collected. Protein concentration was determined using a bicinchoninic acid protein quantification kit (Beyotime, Shanghai, China), and concentrations were normalized across samples. Equal amounts of protein were mixed with 5 × sodium dodecyl sulfate loading buffer at a 4:1 ratio and boiled at 100 °C for 10 minutes to denature the proteins. The denatured samples were loaded onto sodium dodecyl sulfate-polyacrylamide gels for electrophoresis (80 V for the stacking gel and 120 V for the separating gel). After electrophoresis, proteins were transferred to a polyvinylidene difluoride membrane using the wet transfer method (constant current of 300 mA for 40 minutes). The membrane was blocked with rapid blocking buffer (Saiweier, Hubei Province, China) for 5 minutes, followed by three washes with Tris-buffered saline Tween (TBST) for 10 minutes each. The membrane was incubated with diluted primary antibodies on a shaker at 4 °C overnight. After three additional washes with TBST (10 minutes each), the membrane was incubated with the corresponding horseradish-peroxidase-conjugated secondary antibody on a shaker at room temperature for 1.5 hours. The membrane was washed three more times with TBST for 10 minutes each. Visualization was performed using an enhanced chemiluminescence kit (Thermo Fisher Scientific, Waltham, MA, United States). Images were captured using a gel imaging system (Bio-Rad, CA, United States), and the gray values of target protein bands were analyzed with ImageJ software. Expression levels of target proteins were calculated using glyceraldehyde-3-phosphate dehydrogenase as the internal reference. Primary antibodies used in this study included: Glyceraldehyde-3-phosphate dehydrogenase (1:5000; Cell Signaling Technology, MA, United States), CDK6 (1:2000; Cell Signaling Technology, MA, United States), AKT (1:1000; Cell Signaling Technology, MA, United States), phosphorylated (p)-AKT (1:1000; Cell Signaling Technology, MA, United States), B-cell lymphoma-2 (Bcl-2; 1:1000; Cell Signaling Technology, MA, United States), Bcl-2-associated X protein (BAX; 1:1000; Cell Signaling Technology, MA, United States), P21 (1:1000; Cell Signaling Technology, MA, United States), cyclin D (1:1000; Cell Signaling Technology, MA, United States), and caspase 9/cleaved caspase 9 (1:1000; Cell Signaling Technology, MA, United States).

Statistical analysis

All experimental data are presented as mean ± SD. Data analysis was conducted using SPSS version 22.0 statistical software and GraphPad Prism 5. One-way analysis of variance was used for comparisons among multiple groups, followed by the least significant difference t-test for pairwise comparisons. P < 0.05 was considered statistically significant. Each experiment was independently repeated three times to ensure the reliability and reproducibility of the results.

RESULTS

CDK6 expression significantly upregulated in ripretinib-resistant cell lines

First, we conducted tests on the T1, 430, and 882 cell lines, along with their drug-resistant counterparts: T1RI, 430RI, and 882RI. Flow cytometry revealed that all six cell lines exhibited significant and high expression of CD117. Whole-genome sequencing indicated that all six cell lines harbored KIT gene mutations, while no mutations were detected in genes such as platelet-derived growth factor receptor alpha (PDGFRA), neurofibromin 1, or B-Raf proto-oncogene, serine/threonine kinase. Polymerase chain reaction showed that the mutation types in the T1, 430, and 882 cell lines were consistent with those in their corresponding ripretinib-resistant cell lines, with no secondary gene mutations observed in drug-resistant cells (Supplementary material 1, Supplementary Figure 1).

We performed transcriptomic and proteomic analyses on all six cell lines. Compared with nonresistant cells, 35 genes were highly expressed at the transcriptomic and proteomic levels across the three types of resistant cells (Supplementary material 2). CDK6 expression was significantly elevated (Figure 1A-C). Polymerase chain reaction and western blotting validation confirmed that CDK6 protein and mRNA levels in drug-resistant cells were significantly higher than those in nonresistant cells (P < 0.01) (Figure 1D and E). These findings indicate that high CDK6 expression is a hallmark of ripretinib-resistant cell lines and may be closely associated with acquisition of the drug-resistant phenotype.

Figure 1 In ripretinib-resistant cell lines, cyclin-dependent kinase 6 expression was significantly elevated at both the mRNA and protein levels. A: Detection of differentially expressed genes between drug-resistant cell lines and their parental drug-sensitive counterparts. T1 resistance index (RI) vs T1, 882RI vs 882 and 430RI vs 430; B: Detection of differentially expressed proteins between drug-resistant cell lines and their parental drug-sensitive counterparts. T1RI vs T1, 882RI vs 882 and 430RI vs 430; C: Co-expression analysis of transcriptomic and proteomic profiles. T1RI vs T1, 882RI vs 882 and 430RI vs 430; D: The quantitative polymerase chain reaction results demonstrated that the mRNA level of cyclin-dependent kinase 6 was significantly elevated in the ripretinib-resistant group. T1RI vs T1 (2.90 ± 0.311, P < 0.01), 882RI vs 882 (6.15 ± 0.56, P < 0.01) and 430RI vs 430 (2.67 ± 0.04, P < 0.01); E: The western blot results demonstrated that the protein level of cyclin-dependent kinase 6 was significantly elevated in the ripretinib-resistant group. T1RI vs T1 (1.55 ± 0.08, P < 0.01), 882RI vs 882 (1.37 ± 0.10, P < 0.01) and 430RI vs 430 (5.03 ± 0.48, P < 0.01). ^aP < 0.01. RI: Resistance index; DEP: Differentially expressed protein; NDEG: Non-differentially expressed gene; NDEP: Non-differentially expressed protein; DEG: Differentially expressed gene; CDK6: Cyclin-dependent kinase 6; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

CDK6 inhibitor suppresses proliferation of ripretinib-resistant cell lines

We determined the drug concentrations used in the experiment through CCK-8 assays. In the T1, 882, and 430 cell lines, 40 nM ripretinib was applied for 2 days and exhibited a significant inhibitory effect, whereas no significant inhibition was observed in the drug-resistant cell lines (Supplementary Figure 2). All RI values for the three resistant cell lines (T1RI, 882RI and 430RI) were greater than 5, meeting the widely accepted threshold for a stable resistant phenotype. Conversely, in the same cell lines, 0.5 nM-2 μmol/L palbociclib did not show significant inhibitory effects (Supplementary Figure 2). However, in drug-resistant cell lines, palbociclib at concentrations starting from 100 nM demonstrated significant inhibitory effects (Supplementary Figure 2). We selected 40 nM ripretinib and 100 nM palbociclib for subsequent experiments.

We assessed cell viability in each group over a period of up to 5 days. In non-drug-resistant cell lines, there was no significant difference in cell viability between the CDK6 inhibitor group and the control group (P > 0.05), nor between the ripretinib group and the ripretinib plus palbociclib group (P > 0.05; Figure 2). These findings suggest that in non-drug-resistant cell lines, palbociclib alone or in combination with ripretinib did not significantly affect cell proliferation. In contrast, in drug-resistant cells, cell viability in the CDK6 inhibitor group differed significantly from that in the control group (P < 0.05; Figure 2). Combination of ripretinib and palbociclib resulted in significantly greater inhibition of cell proliferation compared to ripretinib alone (P < 0.05; Figure 2). These results indicate that in ripretinib-resistant cell lines, palbociclib alone and in combination with ripretinib affect cell proliferation, and that palbociclib partially reverses resistance to ripretinib. High CDK6 expression may be a key factor enabling ripretinib-resistant cell lines to maintain their abnormal proliferative activity.

Figure 2 The effect of cyclin-dependent kinase 6 on the proliferation of gastrointestinal stromal tumor cells. The effect of cyclin-dependent kinase 6 inhibitors on the proliferation of T1, 882, and 430 cell lines: T1 vs T1 + palbociclib (P), T1 + ripretinib (Ri) vs T1 + Ri + P (P > 0.05), 882 vs 882 + P, 882 + Ri vs 882 + Ri + P (P > 0.05), 430 vs 430 + P, 430 + Ri vs 430 + Ri + P (P > 0.05). The effect of cyclin-dependent kinase 6 inhibitors on the proliferation of T1 resistance index (RI), 882RI, and 430RI cell lines: T1RI vs T1RI + P, T1RI + Ri vs T1RI + Ri + P (P < 0.05), 882RI vs 882RI + P, 882RI + Ri vs 882RI + Ri + P (P < 0.05) and 430RI vs 430RI + P, 430RI + Ri vs 430RI + Ri + P (P < 0.05). ^aP < 0.05. P: Palbociclib; Ri: Ripretinib; RI: Resistance index.

CDK6 inhibitors do not affect migration and invasion of GIST cells

Migration and invasion are key characteristics of malignant tumors. We examined the effects of CDK6 inhibitors on the migratory and invasive capabilities of stromal tumor cells. Scratch assay results showed that in the T1, 882, and 430 cell lines, ripretinib significantly inhibited cell migration and invasion, and CDK6 inhibitors had no significant impact on cell migration (P > 0.05; Figure 3). Similarly, in the invasion assay, whether the CDK6 inhibitor palbociclib was used alone to treat the drug-resistant cell line or combined with ripretinib, no significant difference was observed in the number of cells passing through the basement membrane compared to the control group (P > 0.05; Figure 4). These findings indicate that in both parental and resistant cells, CDK6 inhibition did not significantly alter migration or invasion under the conditions tested.

Figure 3 The effect of cyclin-dependent kinase 6 on the migration of gastrointestinal stromal tumor cells. T1 vs T1 + ripretinib (Ri), T1 + palbociclib (P) vs T1 + Ri + P (P < 0.001), 882 vs 882 + Ri, 882 + P vs 882 + Ri + P (P < 0.001), 430 vs 430 + Ri, 430 + P vs 430 + Ri + P (P < 0.001). T1 vs T1 + P, T1 + Ri vs T1 + Ri + P (P > 0.05), 882 vs 882 + P, 882 + Ri vs 882 + Ri + P (P > 0.05), 430 vs 430 + P, 430 + Ri vs 430 + Ri + P (P > 0.05). T1 resistance index (RI) vs T1RI + P, T1RI + Ri vs T1RI + Ri + P (P > 0.05), 882RI vs 882RI + P, 882RI + Ri vs 882RI + Ri + P (P > 0.05), 430RI vs 430RI + P, 430RI + Ri vs 430RI + Ri + P (P > 0.05). ^aP < 0.001. Scale bar = 100 μm. P: Palbociclib; Ri: Ripretinib; RI: Resistance index.

Figure 4 The effect of cyclin-dependent kinase 6 on the invasion of gastrointestinal stromal tumor cells. T1 vs T1 + ripretinib (Ri), T1 + palbociclib (P) vs T1 + Ri + P (P < 0.001), 882 vs 882 + Ri, 882 + P vs 882 + Ri + P (P < 0.001), 430 vs 430 + Ri, 430 + P vs 430 + Ri + P (P < 0.001). T1 vs T1 + P, T1 + Ri vs T1 + Ri + P (P > 0.05), 882 vs 882 + P, 882 + Ri vs 882 + Ri + P (P > 0.05), 430 vs 430 + P, 430 + Ri vs 430 + Ri + P (P > 0.05). T1 resistance index (RI) vs T1RI + P, T1RI + Ri vs T1RI + Ri + P (P > 0.05), 882RI vs 882RI + P, 882RI + Ri vs 882RI + Ri + P (P > 0.05), 430RI vs 430RI + P, 430RI + Ri vs 430RI + Ri + P (P > 0.05). ^aP < 0.05. Scale bar = 100 μm. P: Palbociclib; Ri: Ripretinib; RI: Resistance index.

In ripretinib-resistant cell lines, CDK6 inhibitor significantly increases the proportion of cells in G1 phase

To clarify the specific role of CDK6 in regulating the cell cycle of ripretinib-resistant cells, we used flow cytometry to analyze changes in the cell cycle distribution of drug-resistant cell lines after treatment with CDK6 inhibitor. In the T1, 882, and 430 cell lines, there were no significant differences in the proportions of cells in the G1, S, and G2/M phases between the untreated control and palbociclib-treated groups. Compared with the ripretinib monotherapy group, the combined use of a CDK6 inhibitor did not alter the distribution proportion of cells in the G1, S and G2/M phases (P > 0.05; Figure 5). However, in the drug-resistant cell lines T1RI, 882RI, and 430RI, treatment with palbociclib significantly increased the proportion of cells in G1 phase compared to the untreated controls (P < 0.05; Figure 5), while the proportions of cells in the S and G2/M phases correspondingly decreased (P < 0.05; Figure 5). Compared with the ripretinib monotherapy group, the percentage of cells in the G1 phase was also markedly elevated in the group receiving ripretinib plus a CDK6 inhibitor (P < 0.05; Figure 5). These results indicate that in ripretinib-resistant cell lines, CDK6 influences cell proliferation by promoting cell cycle arrest at the G1 phase.

Figure 5 The effect of cyclin-dependent kinase 6 on the cell cycle of gastrointestinal stromal tumor cells. In the T1, 882, and 430 cell lines, cyclin-dependent kinase 6 inhibitors had no significant effect on the cell cycle. However, in the ripretinib-resistant cell lines T1 resistance index (RI), 882RI, and 430RI, cyclin-dependent kinase 6 inhibitors significantly increased the proportion of cells in the G1 phase. P: Palbociclib; Ri: Ripretinib; RI: Resistance index.

CDK6 inhibitor combined with ripretinib enhance apoptosis in ripretinib-resistant cells

To further investigate the potential therapeutic value of CDK6 in ripretinib-resistant cells, we examined the effects of CDK6 inhibitor on apoptosis. In non-drug-resistant cell lines T1, 882, and 430, 40 nM ripretinib significantly induced tumor cell apoptosis. However, the proportion of apoptotic cells in the group treated with CDK6 inhibitor and ripretinib did not increase significantly compared to the ripretinib-only group (P > 0.05; Figure 6). In contrast, in drug-resistant cell lines T1RI, 882RI, and 430RI, 40 nM ripretinib did not significantly induce apoptosis (P > 0.05; Figure 6). CDK6 inhibitor alone promoted tumor cell apoptosis in T1RI, 882RI, and 430RI cells lines compared with the control group. The apoptosis rate in the combined treatment group (palbociclib plus ripretinib) was significantly higher than in groups treated with either agent alone (P < 0.05; Figure 6), indicating a synergistic effect in promoting apoptosis. These findings demonstrate that CDK6 inhibitors can induce apoptosis in ripretinib-resistant cells and partially reverse tumor cell resistance to ripretinib. This suggests that CDK6 inhibitors enhance ripretinib-induced apoptosis in resistant cells, and their combined use may overcome or mitigate ripretinib resistance through synergistic mechanisms. This combination therapy offers a promising new strategy for managing ripretinib resistance in GIST patients in clinical practice.

Figure 6 The effect of cyclin-dependent kinase 6 on the apoptosis of gastrointestinal stromal tumor cells. In the T1, 882, and 430 cell lines, cyclin-dependent kinase 6 inhibitors had no significant effect on cell apoptosis. However, in the ripretinib-resistant cell lines T1 resistance index (RI), 882RI, and 430RI, cyclin-dependent kinase 6 inhibitors significantly increased the proportion of apoptotic cells. T1 vs T1 + ripretinib (Ri), T1 + palbociclib (P) vs T1 + Ri + P (P < 0.001), 882 vs 882 + Ri, 882 + P vs 882 + Ri + P (P < 0.001), 430 vs 430 + Ri, 430 + P vs 430 + Ri + P (P < 0.001). T1 vs T1 + P, T1 + Ri vs T1 + Ri + P (P > 0.05), 882 vs 882 + P, 882 + Ri vs 882 + Ri + P (P > 0.05), 430 vs 430 + P, 430 + Ri vs 430 + Ri + P (P > 0.05). T1RI vs T1RI + P, T1RI + Ri vs T1RI + Ri + P (P > 0.05), 882RI vs 882RI + P, 882RI + Ri vs 882RI + Ri + P (P > 0.05), 430RI vs 430RI + P, 430RI + Ri vs 430RI + Ri + P (P > 0.05). ^aP < 0.001. P: Palbociclib; Ri: Ripretinib; RI: Resistance index.

Increased expression of CDK6 is associated with excessive expression and phosphorylation of AKT

Further analysis was performed to assess the significance of the observed association between CDK6 expression and AKT expression and phosphorylation. We examined the activation status of the AKT signaling pathway in drug-resistant cells and found that, compared with parental cells, total AKT levels increased significantly in protein and mRNA levels (P < 0.05; Figure 7A and B), and the phosphorylation level of AKT (p-AKT) in drug-resistant cells was significantly increased compared with nonresistant cells (P < 0.05; Figure 7A). We treated the resistant cells with an AKT-specific inhibitor (MK-2206 dihydrochloride, 100 nM). AKT inhibitor exerted no significant effect on the protein levels of AKT in cells, yet it markedly suppressed AKT phosphorylation, which is indicative of the inhibition of AKT activation. In the AKT inhibitor group, both the protein and mRNA levels of CDK6 were significantly decreased (P < 0.05; Figure 7C). These results strongly support the observed upregulation of AKT expression and phosphorylation in drug-resistant cells, reinforcing the association between CDK6 and AKT. This indicates that in ripretinib-resistant GIST cells, excessive expression and phosphorylation of AKT may be one of the key factors driving the increased expression of CDK6. There is a close regulatory relationship between the two, jointly contributing to the biological behavior changes of drug-resistant cells.

Figure 7 Cyclin-dependent kinase 6 expression correlates with protein kinase B expression and activation. A: Protein levels in gastrointestinal stromal tumor cell lines, including cyclin-dependent kinase 6 (CDK6), protein kinase B (AKT) and phosphorylated-AKT; B: MRNA levels in gastrointestinal stromal tumor cell lines, including CDK6, AKT; C: In the ripretinib-resistant cell lines, both the protein and mRNA levels of CDK6 were significantly decreased in the group treated with the AKT inhibitor; D: Effects of CDK6 inhibitors on the expression of apoptosis-related proteins in ripretinib-resistant cells. ^aP < 0.05, ^bP < 0.001. CDK6: Cyclin-dependent kinase 6; AKT: Protein kinase B; p: Phosphorylated; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; Ri: Ripretinib; P: Palbociclib; RI: Resistance index; BAX: B-cell lymphoma-2-associated X protein; Bcl-2: B-cell lymphoma-2.

Since CDK6 inhibitors promote tumor cell apoptosis, we further investigated the expression of apoptosis-related proteins in drug-resistant cells. The results showed that in the three drug-resistant cell lines, CDK6 inhibitor likely increased the expression of cyclin D protein, while the expression of p21 was also upregulated concomitantly. In the CDK6 inhibition group, expression of the proapoptotic protein BAX was significantly increased, whereas expression of Bcl-2 was markedly decreased (P < 0.05; Figure 7D). Expression of caspase-9/cleaved caspase-9 in the CDK6 inhibitor group increased significantly (P < 0.05; Figure 7D). These results indicate that CDK6 inhibition modulates expression of cell-cycle-associated and apoptosis-associated proteins in ripretinib-resistant GIST cells, thereby inducing apoptosis.

DISCUSSION

GISTs are the most common mesenchymal tumors of the gastrointestinal tract, originating from Cajal stromal cells. Their occurrence and development are closely associated with mutations in the KIT or PDGFRA genes. Although targeted therapeutic drugs, such as imatinib and ripretinib, have significantly improved the prognosis of patients, those with advanced-stage or drug-resistant conditions still encounter severe challenges. Ripretinib, as a new-generation broad-spectrum KIT/PDGFRA inhibitor, has been approved for fourth-line and higher-order treatments. However, the phenomenon of ripretinib resistance in clinical practice is becoming increasingly prominent, and there is an urgent need to explore new strategies to overcome this resistance. In the present study, it was discovered in the established ripretinib-resistant GIST cell lines that CDK6 inhibitors could significantly promote the apoptosis of drug-resistant cells and arrest the cell cycle at the G1 phase. Further in vitro experiments demonstrated that the CDK6 inhibitor alone had a weak inhibitory effect on the proliferation of drug-resistant cells. Notably, combination treatment with palbociclib and ripretinib produced prominent synergistic anti-tumor effects and partially reversed the ripretinib-resistant phenotype in ripretinib-resistant GIST cells. Preliminary exploration of the mechanism revealed that CDK6 was overexpressed in drug-resistant cells and positively correlated with excessive expression and phosphorylation of AKT, suggesting that high expression of CDK6 might be related to the AKT signaling pathway. This finding provides a new insight into the mechanism of ripretinib resistance in GISTs. The overexpression of CDK6 in drug-resistant cells and its positive correlation with the activation of the AKT signaling pathway imply that CDK6 plays a crucial role in maintaining the survival and proliferation of ripretinib-resistant GIST cells through the AKT pathway. Our study found that CDK6 inhibition only induced G1 phase cell cycle arrest and promoted apoptosis in ripretinib-resistant GIST cells, but had no significant regulatory effect on cell migration and invasion. This suggests that CDK6 specifically mediates the proliferation and survival of drug-resistant GIST cells rather than participating in the regulation of tumor cell metastasis-related biological behaviors. The absence of significant changes in migration and invasion of CDK6 inhibition is consistent with the primary role of CDK6 as a cell-cycle regulator[28,29]. This clarifies the functional specificity of CDK6 in ripretinib-resistant GIST and provides a theoretical basis for avoiding potential off-target effects of CDK6 inhibitors in clinical application.

Our functional experiments confirmed that the inhibitory effect of CDK6 inhibitors alone on the proliferation and apoptosis of ripretinib-resistant cells is limited. However, when CDK6 inhibitors are used in combination with ripretinib, they not only significantly enhance the proliferation inhibition of drug-resistant cells but also promote their apoptosis and arrest at the G1 phase. This synergistic effect indicates that inhibiting CDK6 can partially relieve the adaptive resistance of drug-resistant cells to ripretinib. The synergistic effect observed when combining CDK6 inhibitor with ripretinib is particularly noteworthy. It indicates that even though the CDK6 inhibitor alone has limited efficacy against the resistant cells, it can effectively sensitize these cells to ripretinib. This synergy might be attributed to the combined effect of blocking both the CDK6/AKT survival axis and the KIT/PDGFRA-driven oncogenic signaling. By inhibiting CDK6, we might be reversing some of the adaptive changes that the resistant cells have developed to bypass the inhibition of KIT/PDGFRA by ripretinib, thus restoring the sensitivity of the cells to the targeted drug. This not only provides a potential therapeutic strategy to overcome ripretinib resistance but also highlights the importance of combinatorial approaches in targeting complex and adaptive resistance mechanisms in GISTs. The drug doses selected in this study were determined based on pre-experimental CCK-8 results, where 40 nM ripretinib showed inhibitory activity in parental GIST cells and 100 nM palbociclib significantly suppressed the proliferation of drug-resistant cells without obvious cytotoxicity in nonresistant cells. This dose selection principle is consistent with the clinical medication logic of low-dose combination to reduce adverse reactions, and provides a reference for the subsequent exploration of the optimal drug combination ratio in vivo.

The pathogenesis of GISTs is primarily driven by activating mutations in receptor tyrosine kinases, which constitutively activate downstream signaling pathways such as PI3K/AKT/mammalian target of rapamycin and rat sarcoma viral oncogene homolog/mitogen-activated protein kinase[30-33]. This activation promotes aberrant tumor cell proliferation and inhibits apoptosis. Notably, the significant heterogeneity in GIST biological behavior presents considerable challenges in assessing malignant potential and managing drug resistance. Circulating tumor DNA studies have identified secondary KIT gene mutations as key drivers of targeted therapy resistance in GIST patients[34-36]. The PI3K/AKT pathway plays a critical role in the development of primary imatinib resistance in GISTs, and targeted PI3K inhibition effectively induces apoptosis in imatinib-resistant GIST cells[37,38]. Daniels et al[39] further demonstrated that B-Raf proto-oncogene, serine/threonine kinase mutations and PI3K pathway overactivation are closely associated with the drug-resistant phenotype in wild-type GISTs[39]. Aberrant PI3K phosphorylation and downstream signaling regulate GIST cell behaviors, including proliferation, apoptosis, and autophagy[40-43]. The AKT signaling pathway is a well-known regulator of cell survival, proliferation, and metabolism, and its aberrant activation has been implicated in various cancers, including GISTs, contributing to drug resistance[41,44]. Previous studies have shown that CDK6 not only regulates the cell cycle in various solid tumors but also participates in DNA damage repair, metabolic reprogramming, and activation of drug-resistance signaling pathways. In breast cancer, cyclin-dependent kinase 4/CDK6 inhibitors combined with endocrine therapy improve outcomes for estrogen receptor positive/human epidermal growth factor receptor 2-negative breast cancers[45,46]. Our results suggest that AKT might be an upstream activator or a key component in the sustained activation of the CDK6 in ripretinib-resistant cells. To further validate this hypothesis, we analyzed the expression levels of total AKT and p-AKT and CDK6 in both parental and ripretinib-resistant GIST cell lines. As expected, the resistant cells exhibited a significant upregulation of AKT and p-AKT compared to their parental counterparts, which was accompanied by a concurrent increase in CDK6 protein levels. This positive correlation between AKT and CDK6 expression strongly supports the notion that AKT signaling could be driving CDK6 activation in the resistant setting. When we treated the resistant cells with an AKT-specific inhibitor, we observed a marked reduction in CDK6 expression. These findings provide direct evidence that the AKT/CDK6 axis plays a crucial role in mediating ripretinib resistance in GIST cells. Previous studies have confirmed that the PI3K/AKT pathway is an important mediator of TKI resistance in GIST. Our study identified CDK6 as a key downstream effector of the AKT pathway in ripretinib resistance, which complements the regulatory network of the PI3K/AKT pathway in GIST drug resistance and provides a new specific target for the clinical intervention of this pathway. In addition, unlike the reported role of CDK4/6 in GIST proliferation, our study first revealed the specific role of CDK6 in ripretinib resistance, which enriches the research on the biological function of CDK family members in GIST.

From a clinical perspective, the upregulation of CDK6 in ripretinib-resistant GIST cells identified in this study also suggests that CDK6 may serve as a potential predictive biomarker for ripretinib resistance in GIST patients. Detecting the expression level of CDK6 in clinical tumor samples may help to screen patients who are prone to ripretinib resistance and are suitable for CDK6 inhibitor combined therapy, which is expected to realize personalized precision treatment for advanced GIST patients.

This study had some limitations. Firstly, this study was mainly based on cell lines, and the applicability of its results to clinical patients remains to be verified. Future research needs to collect more clinical drug-resistant GIST samples, detect the expression of CDK6 and AKT phosphorylation levels, and analyze their correlation with the clinicopathological characteristics and prognosis of patients. Secondly, the specific molecular interaction mechanism between CDK6 and AKT remains unclear. Whether it involves direct binding or is mediated through intermediate aptamer proteins remains to be clarified. Published evidence in other cancer types suggests that AKT signaling may regulate CDK6 expression through forkhead box O3a-mediated transcriptional activation[47,48]. Additionally, whether CDK6 can feedback-activate AKT, establishing a bidirectional regulatory loop, remains to be explored. Investigating this possibility in GIST cells could provide further insights into the complex regulatory network underlying ripretinib resistance. Finally, the impact of CDK6 inhibition on other drug resistance pathways (such as mitogen-activated protein kinase, Janus kinase/signal transducer and activator of transcription) was not systematically evaluated. Future research needs to construct patient-derived xenograft models or organoid platforms to verify the efficacy and safety of this combined strategy and explore the potential of CDK6 as a biomarker for GIST drug resistance.

CONCLUSION

This study is the first to identify CDK6 as a mediator of ripretinib resistance in GIST, and elucidates that the aberrant upregulation of CDK6, may be driven by the overexpression and hyperphosphorylation of AKT, modulates cell cycle progression and apoptotic signaling to sustain the drug-resistant phenotype of GIST cells. Mechanistically, CDK6 inhibition induces G1-phase cell cycle arrest and promotes apoptosis in ripretinib-resistant GIST cells by regulating the expression of cell cycle-related proteins (cyclin D and p21) and apoptosis-associated factors (BAX, Bcl-2, and caspase-9/cleaved caspase-9). We have demonstrated that combination of a CDK6 inhibitor with ripretinib partially restores the sensitivity of drug-resistant GIST cells to ripretinib. Our findings provide robust experimental evidence for the clinical application of CDK6 inhibitor-ripretinib combination therapy in ripretinib-resistant GIST. These findings point to an AKT/CDK6 regulatory axis implicated in ripretinib resistance; however, the precise molecular mechanism warrants further investigation. This work also identifies CDK6 as a potential therapeutic target and predictive biomarker for ripretinib-resistant GIST, offering a new direction for the development of precision treatment strategies for advanced GIST.