BET inhibitor and CDK4/6 inhibitor synergistically inhibit breast cancer by suppressing BRD4 stability and DNA damage repair

Highlights

• •Abemaciclib and OTX-015 exhibit a synergistic anti-tumor effect in ER⁺ and triple-negative breast cancer in vitro and in vivo.
• •The synergistic effect of combining CDK4/6 inhibitors and BET inhibitors is mediated by promoting DNA damage and cell apoptosis.
• •CDK4 inhibition can promote the degradation of BRD4 through the proteasome pathway, leading to the downregulation of RAD51 expression and inhibition of DNA damage repair.

Abstract

CDK4/6 inhibitors have shown clinical benefits in hormone receptor positive breast cancer. However, monotonous indications and unclear resistance mechanisms greatly limit the clinical application of these inhibitors. We attempt to improve the therapeutic effect of CDK4/6 inhibitors against breast cancer by combination with BET inhibitors. Although this combination therapy has begun to be studied in recent clinical trials, the mechanism of action is not clear. We provide the evidence that CDK4/6 inhibitor LY2835219 plus BRD4 inhibitor OTX-015 synergistically inhibits both ER positive and triple-negative breast cancer cells growth in vitro and in vivo. Mechanistically, LY2835219 accelerates the degradation of BRD4 through the proteasome pathway via inhibition of CDK4 activity. This instability of BRD4 protein in turn enhances the anti-tumor effect of CDK4/6 inhibitor by suppressing transcription of DNA damage repair gene RAD51, and synergistically promotes γ-H2AX accumulation and DNA double-strand breaks. Overall, we demonstrated the potential combined therapeutic value of CDK4/6 and BRD4 inhibitors and elucidated the mechanisms, which may provide a new rational approach for breast cancer patients.

Graphical abstract

Introduction

Although endocrine therapy, chemotherapy, and early diagnosis have improved the prognosis [[1], [2], [3]], breast cancer is still the most common and fatal malignancy in the world. And the poor prognosis and limited treatment options of triple-negative breast cancer (TNBC) exacerbate this situation. All these highlights the desire for new therapies [[4], [5], [6]]. CDKs as targets of anti-tumor inhibitors have been widely concerned for many years. There are more than 20 CDKs inhibitors in clinical trials, most of them are pan-CDKs inhibitors with high toxicity and slow progress [7]. It was not until the advent of paradigms FDA-approved CDK4/6 specific inhibitors that the situation was changed. CDK4 and CDK6, together with d-type cyclins (cyclin D1, D2, and D3), are responsible for the cell cycle transition from G1 to S phase in most cells through phosphorylating and inhibiting retinoblastoma protein (RB) and its related protein family members [8]. The progression of some breast cancer is highly dependent on estrogen receptor (ER)-CDK4/6-RB signal pathway [9,10], which causes the significant efficacy of CDK4/6 inhibitors on these patients. Currently, three ATP competitive small molecule CDK4/6 specific inhibitors, including Palbociclib (PD033291, PD for short) [11], Abemaciclib (LY2835219, LY for short) [12] and Ribociclib (LEE011) [13], have been approved by FDA, as the first-line treatment for hormone receptor-positive advanced breast cancer patients with monotherapy or combination with endocrine therapy [[11], [12], [13]]. In triple-negative breast cancer (TNBC), loss of ER and RB [14], and the elevated expression of cyclin E [15], all contribute to the insensitivity to CDK4/6 inhibitors [16]. However, even for ER positive breast cancer, only some patients benefit from these inhibitors and they may also eventually become insensitive to CDK4/6 inhibitors due to upregulation of cyclin D, CDK6, activation of CDK2, and retention of the inhibitor in the lysosome [17]. Despite numerous preclinical studies, there is still no consensus on the sensitive markers of CDK4/6 inhibitors to indicate patients who may benefit, except for ER [18,19]. Fortunately, CDK4/6 inhibitors are usually well tolerated in monotherapy in clinic. So now a lot of efforts have been devoted to find appropriate combination therapies to improve the efficacy and expand the clinical indications of CDK4/6 inhibitors, hoping to benefit more patients.

As a member of the bromodomains and extra-terminal (BET) family, bromodomain-containing protein 4 (BRD4) is a transcriptional and epigenetic regulator characterized by two tandem bromodomains (BD1 and BD2) at its N-terminal, and acts as a reader of hyper-acetylated chromatin and is mainly involved in regulating the initiation and elongation of gene transcription during G1/S phase [[20], [21], [22]]. Studies have also shown that BRD4 has a certain preference for cancer-associated genes [21,23,24], such as c-Myc [23]. Therefore, inhibition of BRD4 activity with BET inhibitors is considered to be a promising anti-tumor strategy for MYC-addicted cancers, including breast cancer. Currently, most BET inhibitors in clinical trials are acetyl moiety competitive small molecule inhibitors, among which Birabresib (OTX-015, OTX for short) has completed phase I and phase Ib trials [25,26] and has entered phase II clinical trials, for the treatment of acute myeloid leukemia and glioblastoma multiforme [[26], [27], [28]]. Previous studies have shown that BRD4 inhibitors are potential therapeutic drugs for breast cancer [29], but they are also prone to develop drug resistance [30], which has not yet been successful in clinical.

Recently, we have found that the combination therapy of CDK4/6 inhibitors and BRD4 inhibitors has begun to be studied in clinical trials (NCT05372640), and some published studies show that the synergistic effect of the combination therapy may be due to cell cycle arrest [31], cellular senescence [32], or ferroptosis [32], however, the detailed molecular mechanisms still need to be further clarified. In this study, we attempt to find the efficacy of the combination therapy, and understand how they exert the synergistic anti-tumor effects, which may improve the outcomes of these two drugs, and provide new therapeutics for breast cancer patients, especially TNBC patients.

Results

CDK4/6 and BRD4 inhibitors synergistically inhibit the proliferation of various breast cancer cells

The inhibitory activities of LY and OTX on the proliferation of ER⁺ breast cancer cells and TNBC cells were evaluated. We found that ER⁺ breast cancer cells MCF-7 and BT474 were visibly more sensitive to LY and OTX than those TNBC cells (Fig. 1A). Preliminary combined application in MCF-7 and BT549 cells showed that LY together with OTX had a more notable anti-tumor effect than each agent alone (Fig. 1B). To further confirm the synergistic effect of the two types of inhibitors, we expanded to TNBC cell lines. As displayed in the dose-effect curve, compared with single-drug groups, the curve of the LY+OTX was visibly shifted to the left, indicating that the combination can more effectively inhibit the proliferation of tested breast cancer cells (Fig. 1C). In addition, CalcuSyn software was applied to calculate combination index (CI) values to evaluate the combined effect. It is generally considered that CI<0.8 has a synergistic effect, and a CI value of 0.8<CI<1.0 has an additive effect [33]. Except that the CI values of CAL-51 and BT549 cells were close to 0.8, the CI values of the other cell lines were all between 0.04 and 0.47, with an average value of 0.27, much less than 0.8, suggesting that the two compounds had a significant synergistic effect in these cells (Fig. 1D). Given that MCF7 and MDA-MB-231 cell lines are widely recognized as representative models for ER⁺ and TNBC cells [[34], [35], [36], [37]], respectively, we selected these two cell lines as primary models for subsequent experiments, with additional cell lines serving as supplementary validation tools. Furthermore, the clone formation assay was also applied to assess the synergistic effect of CDK4/6 inhibitor plus BRD4 inhibitor. Compared with the single-compound groups, whether in ER⁺ breast cancer or TNBC, the number of clones in the LY+OTX treatment group was reduced more conspicuously (Fig. 1E). All above indicates that CDK4/6 inhibitor and BRD4 inhibitor have synergistic effects on inhibiting ER⁺ and TNBC cells proliferation in vitro.

Fig. 1.

LY and OTX synergistically inhibit the proliferation of breast cancer cells.

A The average IC₅₀ values of breast cancer cell lines after being treated with LY or OTX for 72 h B Heatmap of inhibition ratio after MCF-7 and BT549 cells treated with a series of doses of LY and OTX in combination for 72 h C Inhibition rate curve of breast cancer cells treated with different concentration gradients of LY, OTX or LY+OTX for 72 h, (from the highest concentration of the label, carry out a 2-fold dilution and 8 dilutions). D The average combined index (CI) values after 72 h of combined treatment with different doses of LY and OTX. E Colony formation assay of MCF-7, MDA-MB-231 and Hs578T cells subjected to LY, OTX or LY+OTX treatment for 12 days. All error bars represent mean ± SD. Differences between indicated groups were measured by one-way ANOVA with Dunnett's multiple group comparison test, *p < 0.05, **p < 0.01, ***p < 0.001.

Combination therapy synergistically induces apoptosis

Due to the important roles of CDK4/6 and BRD4 in G1/S transition [38,39], we firstly investigated the effect of LY combined with OTX on cell cycle. Consistent with the previous reported results, both LY and OTX alone could slightly increase the proportion of G1 cells in MCF-7, MDA-MB-231 and HBL-100 cells. However, combination treatment did not enhance the G1 phase arrest in MCF-7 and TNBC cell lines MDA-MB-231 and HBL-100 (Fig. 2A).

Fig. 2.

The effect of LY combined with OTX on cell cycle and apoptosis.

A The effect of LY and OTX alone or in combination on the cell cycle of MCF-7, MDA-MB-231 and HBL-100 cells. B, C The effect of LY and OTX alone or in combination on apoptosis of breast cancer cells. All error bars represent mean ± SD. Differences between indicated groups were measured by one-way ANOVA with Dunnett's multiple group comparison test, ***p < 0.001.

Since apoptosis is another important mode of cell death induced by anticancer drugs, we also detected apoptosis ratio in breast cancer cells after treatment with the combination therapy. The representative flow graphs of MDA-MB-231 and Hs578T were shown in Fig. 2B. And the quantitative data of all tested cells was summarized in Fig. 2C. We found that in ER⁺(MCF-7) and TNBC (MDA-MB-231, Hs578T and HCC1143) cells, LY or OTX alone resulted in a slight increase in apoptosis ratio, and the increase became more visible when combined two compounds. For example, in Hs578T cells, the apoptotic rate was 12.56% and 9.47% in two monotherapy group, and it was increased to 36.65% in the combined group.

Increased the instability of BRD4 by CDK4 inhibition improves the efficacy of CDK4/6 inhibitors

Further, the molecular mechanism was explored. As expected, LY inhibited phosphorylation of RB on S780 (pRb1 S780), a major downstream substrate of CDK4/6 and the proliferation-related protein c-Myc in a dose-dependent manner in MCF-7, MDA-MB-231 and CAL-51 cells (Fig. 3A). LY and PD also time-dependently decreased pRb1 S780 and c-Myc in MCF-7 or CAL-51 (Fig. 3B). We also found that OTX alone noteworthily downregulated c-Myc (Fig. 3C). And the combination of LY and OTX induced a dramatic synergistic decrease of c-Myc in MCF-7 and CAL-51 cells (Fig. 3D). To our surprise, the level of BRD4 dose- and time- dependently declined after treatment with CDK4/6 inhibitors LY or PD (Fig. 3A, B), which attracted our attention. According to the proliferation inhibition data of 40 breast cancer cells treated with CDK4/6 inhibitor palbociclib in the public database GDSC (https://www.cancerrxgene.org/) and CCLE (https://depmap.org/portal/), we found that the level of BRD4 was in direct proportion to the IC₅₀ of the cells against palbociclib (Fig. 3E). The left shifted dose-response curves of LY (Fig. 3F) and reduced IC₅₀ (Fig. 3G) were observed in these BRD4 depleted breast cancer cells, which was similar to the effect of LY+OTX in Fig. 1. Moreover, the CI values of BRD4 siRNA combined with LY were also less than 0.8 in all tested cells (Fig. 3H). All indicated that BRD4 is a key factor in the synergistic effect of LY combined with OTX.

Fig. 3.

Knockdown of BRD4 sensitizes breast cancer to CDK4/6i.

A, B Western blot analysis of BRD4 and related target protein after treated with CDK4/6 inhibitors alone. C Western blot analysis of BRD4 and related target protein after treated with OTX alone. D Changes of BRD4 and related target proteins after being treated by LY, OTX or LY+OTX for 12 h E Violin plot showing IC₅₀ of palbociclib (from GDSC) in breast cancer cell lines with different BRD4 mRNA levels (from CCLE). F Western blot analysis of BRD4 in MCF-7, MDA-MB-231, BT549 and Hs578T knockdown cells and inhibition rate curve of BRD4 knockdown cells treated with different concentrations of LY for 72 h. G The average IC₅₀ values of BRD4 knockdown breast cancer cells after being treated with LY for 72 h. H The average combined index (CI) values are calculated as (IC₅₀ of siBRD4)/ (IC₅₀ of siNC). All error bars represent mean ± SD. Differences between indicated groups were measured by one-way ANOVA with Dunnett's multiple group comparison test, *p < 0.05, **p < 0.01, ***p < 0.001.

Then, siRNA was used to interfere with CDK4 or CDK6 to explore the cause of BRD4 down-regulation. Due to the lack of CDK6 in MCF-7 cells, we knocked down CDK4 and CDK6 in BT-549 and MA-MB-231 cells. Only loss of CDK4 obviously reduced the protein level of BRD4, but did not affect its mRNA level (Fig. 4A–C), while knockdown of CDK6 had almost no effect, indicating that inhibition of CDK4 might affect the stability of BRD4 protein. The mRNA of CDK4 were almost unchanged with the reduction and pharmacological inhibition of BRD4 in MDA-MB-231 cells (Fig. 4B, E). And we also proved that loss of BRD4 did not affect the protein level of CDK4 in MDA-MB-231 and CAL-51 cells (Fig. 4D). Further, MCF-7 cells were used to investigate the effect of CDK4 inhibition on the stability of BRD4. Protein synthesis inhibitor Cycloheximide (CHX), proteasome inhibitor MG132, and lysosomal inhibitor Bafilomycin A1 (Baf-A1) were used combined with CDK4 inhibitors to observe the stability of BRD4. As expected, with the extension of CHX treatment time, the protein level of BRD4 gradually decreased. Moreover, as the simultaneous treatment time of LY and CHX increased, the protein level of BRD4 was further reduced, which led to a decrease in the half-life of BRD4 degradation (Fig. 4F). This data again indicated that LY accelerated the degradation of BRD4. In MG132 treated MCF-7 cells, it was noted that the BRD4 protein in DMSO treatment group was not accumulated with the prolongation of MG132 treatment time, but gradually accumulated in LY treatment group (Fig. 4G). Conversely, in Baf-A1 treated MCF-7 cells, BRD4 protein was accumulated with the prolongation of Baf-A1 treatment time only in DMSO treatment group (Fig. 4H). The above results demonstrated that BRD4 in breast cancer cells was mainly degraded through the lysosomal pathway in the natural state, and the treatment of LY might promote the degradation of BRD4 through the proteasome pathway, which further increased the sensitivity of breast cancer cells to CDK4/6 inhibitor.

Fig. 4.

Inhibition of CDK4 downregulates BRD4 in breast cancer cells.

A Western blot analysis of BRD4 after knockdown of CDK4 (siCDK4–1 and siCDK4–2) or CDK6 (siCDK6–1 and siCDK6–2). B, D RT-qPCR and Western blot analysis detects the changes of mRNA and protein after knockdown of CDK4 (siCDK4–1 and siCDK4–2) and BRD4 (siBRD4–1 and siBRD4–2). C, E The mRNA levels of related genes after LY or OTX treatment alone were detected by RT-qPCR. F Western blot analysis of BRD4 changes with the prolonged treatment time of Cycloheximide (CHX) and LY, and the grayscale quantification using Image J. G, H Western blot analysis of BRD4 changes with the prolonged treatment time of MG132 or Bafilomycin A1 (Baf-A1) after the presence of LY or DMSO, and the grayscale quantification using Image J. All error bars represent mean ± SD.

Combination therapy decreases transcription of DNA homologous recombination repair genes

Since BRD4 plays an important role in gene transcription, we performed RNA-seq analysis to detect probable transcriptome changes in MDA-MB-231 cells after treatment with LY, OTX, and both. Three clusters of differentially expressed genes were obtained by pairwise comparisons (Fig. 5A). Subsequently, we performed a Venn analysis on the differential gene sets obtained from the above three comparisons, and found 997 notably changed genes (568 upregulated and 429 downregulated) in the intersection (Fig. 5B). The clustering heat map of the intersection genes indicated that the LY+OTX group was obviously different from the other three groups. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis revealed that these genes were mainly enriched in three signaling pathways, cell cycle (26 genes), homologous recombination (HR) (11 genes) and p53 signaling pathway (10 genes) (Fig. 5C, Supplementary Table S1). Then, the interaction analysis of KEGG enriched genes by STRING (https://cn.string-db.org/) revealed that the core genes were BRCA1, RAD51, CCNA2 and CHEK1 (Fig. 5D). And the level of these altered genes after drugs treatment was also verified by RT-qPCR in MDA-MB-231, HBL-100 and MCF-7 cells (Supplementary Fig. 1). Based on the function and expression levels in RNA-seq, two key genes in HR [[40], [41], [42], [43]], BRCA1 and RAD51, were finally selected, which may mediate the synergistic effect of the combination therapy. Our results illustrated that LY combined with OTX did synergistically reduce the mRNA and protein level of RAD51 and BRCA1 in MDA-MB-231, HBL-100 and MCF-7 cells (Fig. 5E, 5F), strongly indicating that HR repair pathway was crucial to the synergistic effect of LY plus OTX.

Fig. 5.

The effect of LY combined OTX on the transcription of the homologous recombination pathway of breast cancer cells.

A The differential genes between the LY+OTX group and DMSO group (left), between the LY+OTX group and LY group (middle), between the LY+OTX group and OTX group (right). p-adjust<0.05, |log2FoldChange|≥1.5. B Venn analysis model diagram of the differential gene sets in the LY+OTX group compared to the DMSO, LY and OTX groups (left). Heatmap of cluster analysis performed on the overlapped genes (right). C Bubble chart of KEGG pathway enrichment analysis of the overlapped genes. Heatmap of enriched terms is colored by p-adjust values. D Protein-protein interaction network diagram of the KEGG enriched genes. E RT-qPCR detects the mRNA level of RAD51 and BRCA1 in MCF-7, MDA-MB-231 and HBL-100 cells treated with LY, OTX, or LY+OTX. F Western blot analysis of RAD51 and BRCA1 after treated with LY, OTX, or LY+OTX in MCF-7, MAD-MB-231, and HBL-100 cells. All error bars represent mean ± SD.

CDK4/6 and BRD4 inhibitors exhibit synergistic activity in DNA damage response both in vitro and in vivo

With the extension of LY+OTX treatment time, RAD51 gradually decreased, and the accumulation of γH2AX significantly increased at 9 h and 12 h in MCF-7, MDA-MB-231 and BT549 cells (Fig. 6A). And compared with the monotherapy, LY + OTX led to a further decrease in RAD51 and a more significant accumulation of γH2AX in MCF-7 and MDA-MB-231 cells (Fig. 6B). Then, we measured DNA damage induced by LY alone or in combination with OTX using the neutral comet assay in MDA-MB-231 cells. The combination led to a further increase in moments and DNA contents in tails of comet (Fig. 6C, 6D). Similarly, LY could also result in an effective reduction of RAD51 and an increase in γH2AX in BRD4 silencing cells, although loss of BRD4 had already suppressed RAD51 transcription to boost γH2AX (Fig. 6E, 6F, 6G). Compared with the siNC group, the inhibition rate curve of the combination therapy shifted to the left in MDA-MB-231 cells depleted of RAD51 or BRCA1, indicating that RAD51 and BRCA1 were indeed key factors in the synergistic effect (Supplementary Fig. 2).

Fig. 6.

The effect of Inhibiting BRD4 and CDK4/6 inhibitors on DNA damage repair in vitro and in vivo.

A Western blot analysis showing dynamic expression of γH2AX/RAD51 after different treatment times of LY+OTX. B Western blot analysis of γH2AX/RAD51 after treated with LY, OTX or LY+OTX in MCF-7 and MDA-MB-231 cells. C DNA damage in MDA-MB-231 cells treated with LY, OTX or LY+OTX, measured by the neutral comet assay. Scar bar=10 μm. D Quantification of DNA damage by tail moments and DNA% in tail. E Western blot analysis of γH2AX/RAD51 after knockdown of BRD4 (siBRD4–1 and siBRD4–2). F RT-qPCR detects the changes of mRNA in MCF-7, MDA-MB-231 and CAL-51 cells after knockdown of BRD4 (siBRD4–1 and siBRD4–2). G After siRNA (siBRD4–1 and siBRD4–2) treatment for 48 h, then co-treated with LY for 24 h to detect the changes of RAD51 and γH2AX in the MCF-7, MDA-MB-231 and CAL-51 cells. H Nude mice bearing MDA-MB-231 cells were treated with drugs for 21 consecutive days. The tumor volume of mice is calculated as [length × (width)2]/2. A Student's t-test is used for statistical analysis. I Western blot results of mouse tumor tissue protein samples. J Schematic diagram of LY combined with OTX to inhibit breast cancer synergistically. All error bars represent mean ± SD. Differences between indicated groups were measured by one-way ANOVA with Dunnett's multiple group comparison test, *p < 0.05, **p < 0.01, ***p < 0.001.

Furthermore, in the subcutaneous nude mice xenograft model, combination treatment also showed a synergistic inhibitory effect on tumor growth. Daily dosed 75 mg/kg LY or 20 mg/kg OTX alone only slightly delayed the growth of MDA-MB-231 xenografts, and the percentage of T/C obtained on the 21st day were 64.10% and 56.55%, respectively (Fig. 6H, Supplementary Table S2). The inhibitory effect of LY+OTX on tumor growth was significantly enhanced with the T/C% reaching 32.52% at the end of the experiment, which was notably better than each monotherapy group (Fig. 6H, Supplementary Table S2). And there was no obvious weight loss in the combination therapy compared with the single therapy (Supplementary Fig. 3). Western Blot analysis of tumor tissues demonstrated that, LY also reduced the level of BRD4 protein in tumor tissues, and the combination with OTX synergistically downregulated RAD51 and c-Myc (Fig. 6I), consistent with the in vitro results.

Discussion

Although endocrine therapy and early diagnosis have significantly improved the prognosis of breast cancer, it is still the most common malignant tumor and the main cause of cancer death throughout the world [44]. GLOBOCAN 2020 data produced by the International Agency for Research on Cancer from 185 countries reported an estimated 2.3 million new cases of breast cancer and a mortality rate of 6.9% [44]. In particular, TNBC has a worse prognosis because of its heterogeneity, aggressive tumor phenotypes, higher recurrence risk, and fewer treatment options [45]. Therefore, more new potent therapeutic applications need to be approached [1,2,46]. Recently, the emergence of CDK4/6 inhibitors has brought new strategy for the treatment of breast cancer, but the beneficiaries are still very limited. Here, we found that CDK4/6 inhibitor LY combined with BRD4 inhibitor OTX could synergistically induce DNA damage in ER⁺ and TNBC breast cancer cells, and then inhibit tumor growth in vitro and in vivo. These results highlight the feasibility of this combination therapy for ER⁺ and TNBC treatment in clinical.

We also sought to underline the potential molecular mechanisms of the combination therapy. BRD4 is a transcription and epigenetic regulator that plays an important role in embryogenesis and tumorigenesis [47]. It was originally identified as a cell cycle controlling protein that associates with chromosomes during mitosis, marking genes in G1 phase that are ready to be transcribed to ensure cell cycle progression [20,48,49]. BRD4 also functions as a scaffolding platform for a variety of transcription factors [49], such as p53²¹, C/EBP²², positive transcription elongation factor b⁴⁹ and so on. Previous studies indicated that BRD4 is important to the occurrence and development of breast cancer, and BET inhibitors are potential therapeutic agents for breast cancer [30]. It was surprising to find that CDK4 inhibition promoted the degradation of BRD4 protein, and knockdown of BRD4 enhanced the efficacy of CDK4/6 inhibitor. We thought that decreased BRD4 by CDK4 inhibition contributes to the synergistic effect of the combination therapy against breast cancer. Our study also indicated that high level of BRD4 in breast cancer cells correlates with reduced sensitivity to CDK4/6 inhibitors, suggesting that BRD4 may serve as a potential biomarker for predicting the efficacy of CDK4/6 inhibitors. Although it still needs to be further validated in clinical.

Recently, it has been illustrated that CDK4/6 inhibitors repress homologous recombination proteins, prevent DNA repair and recovery and then synergize with several antimitotic and DNA-damaging agents after sequential administration [50]. In addition, BRD4 also has been reported to function in the DNA repair system as a scaffold protein to stabilize the structure of histones and DNA repair machines, dependent on or independent of its classical transcriptional activity [51,52]. In this study, gene analyses gave us the similar results that combined CDK4/6 inhibitors with OTX dramatically downregulated genes related in DNA damage and repair signaling pathway. We found that BRD4 inhibition, either by pharmacological inhibitor or deletion by specific silencing, resulted in directly decrease in HR-related genes transcription and double-strand break (DSB) damage. Further, the combination therapy inhibited the transcription of RAD51 and BRCA. And the marker of DSB damage, γH2AX was accumulated in breast cancer cells and xenografts after treatment with CDK4/6 inhibitors plus OTX. Although existing literature mentions that the combination of these two types of inhibitors can induce DNA damage in RB-deficient lung cancer cells, it has not thoroughly explored the genes mediating this process [32]. Through RNA sequencing and protein-protein interaction (PPI) analyses, we identified RAD51 and BRCA1 as key executors of this effect. Subsequent validation using siRNA confirmed that both genes mediate this combinatorial effect. This might be another important mechanism for the synergetic antitumor efficacy of the combination therapy.

Taken all together, our study strongly suggests that CDK4/6 and BRD4 inhibitors exhibit synergistic effects on proliferation and DNA damage in both ER⁺ and TNBC in vitro and in vivo. Mechanistically, inhibition of CDK4 promotes the degradation of BRD4 via the proteasomal degradation pathway. Then, BRD4 inhibitor OTX and decreased BRD4 in turn enhances the efficacy of CDK4/6 inhibitor through synergistically suppresses DNA damage repair by downregulation of RAD51, which is transcriptionally regulated by BRD4 (Fig. 6J). Most importantly, we found a possible mechanism of action independent of cell cycle regulation even in CDK4/6 inhibitor insensitive TNBC cells. These data provide a basis for the combination of CDK4/6 and BRD4 inhibitors, and are expected to broaden the clinical indications for these inhibitors. Despite progress, significant gaps remain in understanding CDK4/6 inhibitor responses and resistance mechanisms. Future research should focus on identifying additional biomarkers [53], exploring resistance strategies, and developing new treatment combinations. Advances in predictive tools and personalized medicine are crucial for improving treatment outcomes and overcoming resistance in breast cancer.

Materials and methods

Cells and compounds

The selection of these cell lines aims to cover the two main subtypes of breast cancer, ER⁺ and TNBC, hoping to provide a more comprehensive perspective on how the combined treatment affects different subtypes of breast cancer (Table 1). Among them, MCF-7 is a well-established ER⁺breast cancer cell line that has been widely used in numerous studies. MDA-MB-231 is a frequently used TNBC cell line known for its aggressive nature and is commonly used to develop xenograft models in mice, making it suitable for assessing anti-tumor effects of drugs in vivo. Given the insensitivity of TNBC to CDK4/6 inhibitors, we placed more emphasis on investigating the effects and mechanisms of combination therapy within TNBC contexts. Then, we included more TNBC cell lines, such as MDA-MB-231, HBL-100, CAL-51 and Hs578T in our assays.

Table 1.

Pathological features of breast cancer cell lines used in this study [[54], [55]].

Cell line	ER	PR	HER2	Classification
MCF-7	+	+	–	ER+
BT474	+	+	+	ER+
CAL-51	–	–	–	TNBC
BT549	–	–	–	TNBC
MDA-MB-231	–	–	–	TNBC
MDA-MB-468	–	–	–	TNBC
Hs578T	–	–	–	TNBC
HCC1143	–	–	–	TNBC
HCC1806	–	–	–	TNBC

Details for compounds used in this work are available in the supplemental methods.

MCF-7, BT-474, CAL-51, BT-549, MDA-MB-231, Hs578T, MDA-MB-468, HCC1143, and HCC1806 were purchased from the American Type Culture Collection (Manassas, USA). The human breast epithelial cell line HBL-100 was purchased from the Cell bank of typical culture preservation Committee of Chinese Academy of Sciences (Shanghai, China). All these cells were maintained according to the providers’ instructions.

Western Blot analysis

Details for Western Blot analysis are available in the supplemental methods.

Cell proliferation assay

The Sulforhodamine B method (Sigma-Aldrich, St. Louis, MO, USA) was utilized to detect cell viability and proliferation. The inhibition rate (IR) and half-maximal inhibition concentration (IC₅₀) were calculated using softmax (Pro v5.4.1, molecular device) software. More details for Cell Proliferation assay are available in the supplemental methods.

Determination of synergism

The combination index (CI) value calculated by CalcuSyn software is often used to characterize the synergistic effect of two drugs. A significant synergistic effect is seen when the CI value is smaller than 0.8. When the CI value is between 0.8 and 1.2, it means there is an additive effect. When the CI value is more than 1.2, an antagonistic effect is present.

Colony formation assay

See the supplemental methods for Colony formation assay.

Flow cytometry

Cell cycle analysis and cell apoptosis analysis were analyzed on a FACS Calibur cytometer (BD, San Jose, USA). Data analysis was done with FlowJo V10. More details for Flow cytometry are available in the supplemental methods.

Comet assay

Comet slides (R&D Systems, USA), 1 × TBE electrophoresis solution (Meilunbio, Dalian, China) and DAPI (Beyotime, Shanghai, China) were used for the Comet assay. More details for Comet assay are available in the supplemental methods.

Real-time quantitative PCR (RT-qPCR)

See the supplemental methods for details about RT-qPCR.

Small interference RNA-mediated protein knockdown

Details for Small interference RNA-mediated protein knockdown are available in the supplemental methods.

RNA-sequencing analysis

MDA-MB-231 cells were used for RNA-sequencing (RNA-seq). Samples were sent and entrusted to RiboBio (Guangzhou, China) for sequencing. After Ribobio's sequencing was completed, we used RStudio 4.0.2 software, referring to the process of using DESeq2 to process RNA-seq data in Bioconductor. The RNA-seq data (PRJNA1060487) is accessible via the Sequence Read Archive (https://dataview.ncbi.nlm.nih.gov/?archive=bioproject). More details for RNA-sequencing analysis are available in the supplemental methods.

In vivo studies

The animal experiment operation was carried out after approval by the Institutional Animal Care and Use Committee (IACUC) of the Shanghai Institute of Materia Medica (No. 2020–04-DJ-56). MDA-MB-231 tumor tissue during the vigorous growth period, was cut it into about 1.5 mm³, and inoculated under aseptic conditions into the right axillary subcutaneously of female BALB/c nude mice aged 4 weeks. More details for In vivo studies are available in the supplemental methods.

Statistical analysis

The data are the results of at least 3 independent experiments displayed in the form of mean±SD, except for special cases. GraphPad Prism7.0 software (GraphPad Prism, California, USA) was used to compare the data differences between indicated groups were measured by one-way ANOVA with Dunnett's multiple group comparison test. When the p value<0.05, it indicates that the data has a statistically significant difference, *** means p < 0.001; ** means p < 0.01; * means p < 0.05.

CRediT authorship contribution statement

Shuaishuai Chi: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Fan Wei: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Yangsha Li: Software, Formal analysis, Data curation. Lei Yu: Methodology, Investigation, Data curation. Chuyao Ma: Methodology, Data curation. Yanfen Fang: Formal analysis, Data curation. Biyu Yang: Methodology, Data curation. Yi Chen: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Jian Ding: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.