Role of Differentially Expressed Proteins in Acquired Resistance to Cdk4/6 Inhibitor in Breast Cancer

CDK4/6 inhibitors (Abemaciclib, Ab and Palbociclib, Pb) stop the G1-phase in cell-cycle being used to cure advanced stage of breast cancer (BC). Acquired resistance is a major challenge in BC therapy. The molecular signature of the therapy resistance for Ab and Pb drugs in BC should be explored. Here, we developed Ab/Pb-resistant cell-models and explored the molecular changes. Drug’s resistance cells were developed in MCF-7 cells by continuous drug treatment and it was conrmed by MTT-assay, PI-staining-microscopy, and real-time-qPCR. Global proteome proling done by Labelled-free-Proteome-Orbitrap-Fusion-MS-MS technique. Bioinformatics tools used to analyse the proteome data. Ab-resistant and Pb-resistant MCF-7 cells showed increased tolerance for the respective drug. The BCL-2 and MCL-1 survival genes were up-regulated, while the apoptosis genes BAD, BAX, CASP-3 and PARP-1were down-regulated in the resistant cells. Expression of the MDR-1, ABCG2, ESR-1, CDK4, CDK6, and Cyclin-D1 genes were increased in both resistance cells. For proteomics, 237 and 239 proteins were expressed differently in the resistant Ab and Pb cells, respectively. The NUDT5, PEPD, ABAT, ATP1B1, GGCT, and SELENBP1 proteins were down-regulated and the SBSN, HSD17B10, CD9, PDIA3, PSMB4, SLC2A1, and VTN proteins were upregulated in Ab-resistant cells. The NUDT5, PEPD, and GGCT proteins were down-regulated, while CD47, HIST1H2BN, LMNA, VTN, PSMB5, HBB, PSMA7, FLNB, PRDX4, VDAC1, GOT2, HSPA5, SERPINH1, EIF4A2, FTH, and VIM proteins were up-regulated in Pb-resistant cells. These proteins are a new set of prognostic markers and drug targets for overcoming the respective drug resistance. However, it is necessary to perform an in vivo or clinical assessment.


Introduction
Breast cancer is the second most commonly diagnosed cancer in the world, after lung cancer; however, it is the most prevalent in the Indian population [1][2][3]. Breast cancers are classi ed by the presence or absence of a surface receptor, that is, the estrogen receptor (ER), the progesterone receptor (RP), and the human epidermal growth factor receptor 2 (HER-2). Radiotherapy, chemotherapy, hormone therapy, and targeted therapy are all common treatment methods [4]. To treat primary and metastatic breast cancer, chemotherapy drugs such as 5-uorouracil, methotrexate, cyclophosphamide, doxorubicin, and epirubicin are used. [5,6].ER inhibitors (tamoxifen, letrozole, and fulvestrant) are used to treat estrogen-positive breast cancer in hormone therapy. Tumor cells develop tolerance to drug cytotoxicity, resulting in the therapeutic resistance phenotype, which contributes to disease progression and recurrence. [5,6]. Tumor cells triggered other pathways essential for their survival in the process of acquiring the resistance trait [5,6].The investigation of the molecular mechanisms underlying therapeutic resistance in cancer care has resulted in a transition in drug development toward targeted therapy.By inhibiting certain proteins necessary for tumor cell proliferation and survival, targeted therapy prevents cancer progression. Proteins such as receptor tyrosine kinase, cytokine receptors, and intracellular serine-threonine kinases can be targeted.
Abemaciclib (Ab) and Palbociclib (Pb) are two synthetic (small-molecule) inhibitors of cyclin-dependent kinase 4 and 6 (CDK4 and CDK6) [7]. CDK 4 and 6 are intracellular, serine-threonine protein kinase enzyme associated with cell-cycle progression from G1 to S-phase. Inhibition of CDK4 and CDK6 will halt the G1 phase of the cells. The U.S. FDA approved Palbociclib in 2016, while Abemaciclib approved in 2017 to treat advanced breast cancer with ER, PR positive and HER2-negative receptors [8,9].
Deregulation of CDK4/6 causes uncontrolled cell proliferation, which may occur due to ampli cation of cyclin D1, a gain of CDK4/6, loss of p18, elevated Rb1, and loss of p16, etc. [10][11][12][13]. CDK4/6 inhibitors have emerged as breakthrough for breast cancer treatment. Recently, resistance to CDK4/6 inhibitors has been identi ed as a concern in the management of breast cancer [14][15][16]. PI3K-AKT-mTOR signal mediated resistance to CDK4/6 inhibitors is reported for a similar medicine ribociclib. [17]. However, resistance mechanisms for Ab and Pb drugs in breast cancer therapy have not been reported.
This study aims to develop drug-resistant cell models for the drugs Abemaciclib and Palbociclib in the breast cancer cell lineage and further assess the molecular changes involved in resistance.

Materials And Methods
Abemaciclib and Palbociclib drugs were purchased from Selleck Chem, USA. MCF-7 was bought from NCCS Pune in India. Cell culture media (Lonza), FBS (Gibco life technology), Plastic-wares for cell-culture (Corning), 96-well plate for real-time-qPCR of Thermo Scienti c, RNA isolation kit (Qiagen), cDNA synthesis kit (Takara), 2x SYBR Green-Master mix (Promega) and Primers synthesized from Imperial Life Sciences Pvt. Ltd. Both, control and resistant cells were maintained in high glucose DMEM medium supplemented with 10% FBS and 1% Penicillin and streptomycin antibiotics in 5% CO2, humidi ed air and at 37 0C temperature. To develop resistance models, MCF-7 cells were treated with Abemaciclib and Palbociclib drugs in separate T-25 asks at initial 5 nm concentrations. The drug doses were gradually increased up to 200 nm concentrations and keep maintaining cells at the same.
Cell-viability Assay MCF-7 and continued drug-treated cells called Ab-R-MCF-7 and Pb-R-MCF-7 with Ab and Pb drug, respectively, are seeded in a 96-well plate with 2. 5x10 4 cells count in each well. After 24 h of incubation, cells were treated with the respective drug from 100 nm to 6 μM concentrations in drug-sensitive MCF-7 cells and up to 8μM for drug-resistant MCF-7 cells for 48 and 72 h. Add 100 µl of 0.5 μg/μl MTT to each well and incubated for 2.5 h than lysed the cells with detergent (DMSO). OD-value was measured at 595 nm wavelength into micro-plate reader and percent cell viability was calculated.
PI-Staining-microscopy MCF-7, Ab-R-MCF-7, and Pb-R-MCF-7 cells were seeded (4x10 4 cells/well) in a 12-well plate and incubated into a CO2 incubator for 24 hrs. Cells were treated with Ab and Pb of 1.5 and 2.0 μM concentrations for 48 and 72 h, respectively. Cells were stained with 25 μg/ml propidium iodide (PI) for 15 min. PI-stained cells exposed to green mono-chromatic light and red-emission light were captured along with white light.
Images were taken on 20x magni cation into a Nikon inverted uorescence microscope.
Real-Time-qPCR MCF-7, Ab-R-MCF-7, and Pb-R-MCF-7 cells were seeded (7.5x104 cells/well) into 6-well plates and incubated into CO2 incubator for 24 hrs. Cells were treated with Ab (0.25 μM) and Pb (2.0 μM) drugs for 48 h. Cells detached and washed with cold PBS. Total RNA is extracted from cells palette using total RNA isolation kit followed the kit protocols. The quantity and quality of total RNA was measured by Nano-Drop. 0.5 μg total RNA is used for cDNA synthesis following the protocols mentioned in cDNA synthesis kit. Real-time-qPCR was performed in triplicate using Bio-Red connect real-time-qPCR system using the 2X Master mix of SYBR Green (from Promega) and gene-speci c primers.18S gene was selected as an internal control for normalization. Relative quantity (RQ) -value for each gene were analyzed for each sample from their Ct-value. The list of genes and primer sequences is listed in Supplemental Table 1.
Cells were seeded (7.5 x 10 4 cells/well) in T-25 asks, incubated in a CO 2 incubator for 48 h, after those cells were harvested and lysed in a cell lysis buffer (6 M urea, 2M thiourea, 2% CHAPS and 0.5% SDS) containing 1% protease inhibitor cocktail. Cell samples were sonicated on ice for 30 sec pulse with 10 sec gaps for 5 times, centrifuged the cell-lysate at 12000 rpm for 15 min at 4 0 C than supernatant was collected. Proteins were estimated using Bradford reagent (Sigma) and measured the OD value in the microplate reader at 595 nm wavelength light exposed.

Bioinformatics
Bioinformatics tools were used to analyse proteomics data. Venny tool used to select common proteins in replicating data. The reactome pathway database was used to nd the role of proteins in different reaction pathways. Gene Ontology (GO) analysis was performed using the STRING database. A network analyst tool used to nd the proteins hub. Generic PPI was analysed without any lter, but the subnetwork having maximum seed. miR-Net tool used for miRNA-Gene interaction analysis to explore the posttranscriptional regulation of gene expressions. Here, identi ed protein IDs were searched against the background of breast cancer tissues. The minimum network option was selected for the analysis.

Results
Increased tolerance to Abemaciclib and Palbociclib in Ab-R-MCF-7 and Pb-R-MCF-7 cells MCF-7 cells were treated with Abemaciclib and Palbociclib and the inhibitory concentration values were calculated. Cells treated with Ab exhibited an IC50 value of 1.5 µM at 48 h and cells treated with Pb showed an IC50 3.0 µM after 72 h. These sensitive MCF-7 cells were treated continuously with an increasing concentration ranging from 0.1 to 8 µM. As shown in Figure-1 cytotoxicity was assessed in the continuously drug-exposed cells and it was found that Ab treated cells could tolerate the drug up to a concentration of 7.0 µM after 48hrs and the Pb treated cells could tolerate up to 7.5 µM after 72 h. When increased IC50 values were compared to the sensitive cells could survive higher drug concentration, indicating that these cells develop resistance to the drugs.

Up-regulation Of Pro-survival And Down-regulation Of Apoptotic Genes
To check the cell death at the mRNA level the expression of pro-survival and apoptosis-associated genes were checked. qPCR results as shown in Fig. 2B, the expression of pro-survival genes BCL-2 and MCL-1 were found to be up-regulated in resistant cells. BCL-2 expression was enhanced by 3.13 and 1.60-fold in Ab-RMCF-7 and Pb-R-MCF-7 cells, respectively, whereas, MCL-1 were 4.90 and 5.80-fold higher expression in Ab-R-MCF-7 and Pb-R-MCF-7 respectively as compare to the control. Pro-apoptotic genes BAX, BAD, Casp-3 and PARP1 expressions in Ab-R-MCF-7 were similar to the control cells. They exhibited a change of 1.04, 0.80, 1.14, and 1.28 folds, respectively. However, in Ab-treated drug-sensitive cells, these gene expressions were found to be 2.25, 2.17, 2.14, and 5.17-folds higher, respectively ( Fig. 2. C). BAX, BAD, Casp-3 and PARP1 expression in Pb-RMCF-7 was 1.15, 0.95, 1.0, and 1.20-fold changed, respectively, however, in Pb-treated, drug-sensitive cells, these were 2.25, 2.15, 2.51, and 3.86-fold higher, respectively, compared to the control ( Fig. 2. D).
MDR-1 and ABCG-2 are two e ux proteins that are involved in drug expulsion and have been linked to cancer cell survival and resistance evolution. As depicted in Fig. 2A. MDR1 expression was found to be identical in control and drug-treated responsive cells. It was shown to be increased 3.0-fold and 8.2-folds in Ab-R-MCF-7 and Pb-R-MCF-7 cells, respectively. The expression of the ABCG2 gene was comparable in control and drug-treated responsive cells, but it was 4.0 and 5.9 times higher in Ab-RMCF-7 and Pb-R-MCF-7 cells, respectively. ESR1 genes are overexpressed in cancer cells, especially in breast cancer, and both Ab and Pb are CDK4/6 inhibitors. In this analysis, the expression of ESR-1, CDK-4, and CDK-6 in control and responsive cells was found to be identical, whereas the level of cyclin D1 decreased 0.55-fold.However, ESR-1, CDK-4, CDK-6, and Cyclin D1 genes increased 11.9, 3.6, 15.5, and 4.1-fold in Ab-resistant cells, respectively. In Pbresponsive cells ESR-1, CDK-4, and CDK-6 gene expression was unchanged compared to the controls, whereas, Cyclin-D1 was 0.43-fold down-regulated. However, in Pb-resistant cells, ESR-1, CDK-4, CDK-6, and Cyclin-D1 gene expression were increased 20.3, 2.8, 6.4, and 6.0-fold, respectively (  Table 2 and Table 1). The upregulation of SBSN, HSD17B10, CD9, PDIA3, PSMB4, SLC2A1, and VTN proteins in Ab-R-MCF-7 cells has been identi ed as a poor prognostic indicator or involved in acquired drug resistance, and has been proposed as a novel drug target for BC treatment (Supplementary Table 3 and Table 1). NUDT5, PEPD, and GGCT proteins, which have been identi ed as prognosis or drug-sensitive markers, were downregulated in Pb-R-MCF-7 cells (Supplementary Table 4 and Table 1). CD47, HIST1H2BN, LMNA, VTN, PSMB5, HBB, PSMA7, FLNB, PRDX4, VDAC1, GOT2, HSPA5, SERPINH1, EIF4A2, FTH1, and VIM proteins have been found to be upregulated in Pb-R-MCF-7 cells, which are associated with poor prognosis or drug resistance markers (Supplementary Table 5 and Table 1).  cellular components and further KEGG pathway analysis identi ed 30 signaling pathways. (Fig. 5. A,  B).We analysed both drug resistance samples using the Reactome pathway database to investigate the reactome pathway associated with these two conditions. DEPs are implicated in metabolism, immunity, signal transduction, programmed cell death, cellular response to external stimuli, DNA replication, cell cycle, and vesicle transport in this study. (Fig. 6A, B, C, and D).
Hub molecules in DEP's dataset were identi ed by the Network Analysis tool. FLNA, SSBP, SLC3A2, ATP5B, DLD, and SOD1 were found to be major hubs among down-regulated proteins, whereas, S100A7, VTN, and RAD are up-regulated proteins were identi ed as major hubs in Ab-resistance cells. YWHAZ, DDB1, HSPB1, S100A9, and ANXA2 were found to be major hubs among down-regulated proteins, while SERPINH1, AHCY, FBP1, TXNRD1, PSMA6, HSP90B1, and PKM were found to be major hubs among upregulated proteins in Pb-resistance cells. (Fig. 7. A, B, C, and D).

Discussion
Drug-resistant MCF-7 cells showed to tolerate the drug at higher concentrations compared to control cells. Microscopic images of PI-stained cells con rmed the ndings.MDR-1 and ABCG-2 are xenobiotic transporters that help cells eliminate toxins. These proteins have been linked to the phenotype of drug resistance in various cancers [18,19].Increased MDR-1 and ABCG-2 expression in MCF-7 cells exposed with Ab and Pb for a long-time result in drug e ux, which may be one of the reasons for drug resistance in these cells.
As shown in Fig. 2B, C, and D, the drug-resistance models were re-con rmed by increased expression of survival (Bcl-2 and MCL-1) genes and suppressed expression of pro-apoptotic (BAX, BAD, Caspase-3, and PARP1) genes.Bcl-2 blocks the intrinsic apoptosis signal by inhibiting Bax and Bak interactions on the mitochondrial outer membrane. Drug-resistant, cells have elevated levels of Bcl-2 and MCL-1 [20,21]. Estrogen activates ER, which leads to DNA replication and cell division, as encoded by the ESR1 gene [22].Increased ESR1 expression suggests that it can contribute to therapy resistance. CDK4 and CDK6 are targeted by Ab and Pb, which stimulates cyclin-D1 and promotes cell cycle progression. Increased CDK4/6 and cyclin D1 expression in our cells promotes successful cell division, which is a hallmark of drug resistance.
The roles of proteins in therapy resistance in Ab-R-MCF-7 and Pb-R-MCF-7 cells are described in Supplementary table 2-5. DEPs are manually curated from the PubMed literature search engine to determine their molecular functions, roles in cancer progression, therapy resistance, and prognostic importance (see Table 1). NUDT5 is linked to a poor prognosis and low overall survival in clear cell renal cell carcinoma [23]. PEPD binds to Her-2 and inhibits EGFR signaling, resulting in cancer cell growth inhibition [24]. Lower expression of GGCT, a crucial component of the GSH-pathway, has been linked to chemo-resistance [25]. In ammatory BC has an inverse association between ABAT expression and therapy resistance [26]. A high level of ATP1B1 is linked to metastasis and it is an essential energy transfer system for cells [27]. Low levels of SELENBP1 in ER + ve BC result in poor survival and selenium tolerance [28] (Supplementary table 2 and 4) and Table 1.
PSMB is a group β-subunit of the 20S proteasome (PSMB4 and PSMB5), and α-subunit PSMA7 are involved in the proteolytic degradation of intracellular proteins. PSMB4 overexpression promotes cell cycle progression from G1 to S phase and cell viability through NF-B signaling [29]. A high PSMB5 level suggests a poor prognosis [30,31]. PSMA7 levels in gastric cancer have been related to invasion, metastasis, and poor prognosis [32]. VTN is a part of the extracellular matrix that promotes integrin signaling. VTN, which is downstream of VEGF/VEGFR and PI3K/AKT signaling, promotes cell migration and metastasis in breast cancer [33,34,35]. SBSN is an oncoprotein that promotes tumorigenicity by increasing Wnt/β-catenin signaling [36,37]. The mitochondrial enzyme HSB17B10 is responsible for the oxidation of steroids, alcohol, and fatty acids. In osteosarcoma, a high level of HSD17B10 indicated a poor response to chemotherapy [38]. Chemo-resistance is caused by the tumour microenvironment facilitating CD9-mediated crosstalk between mesenchymal stem cells and BC cells via CCL5, CCR5, and CXCR12 [39]. PDIA3 expression in the tumor microenvironment is high, which favors invasion and metastasis [40,41].GLUT1 proteins encoded by SLC2A1 aid glucose transport. Hypoxia inducing factor-1(HIF-1) is activated as breast cancer grows aggressively. Due to hypoxic environments, HIF1 triggers GLUT1 expression [42,43].Under hypoxic conditions, HIF-1 also activates CD47 transcription. In BC, CD47 retains stemness, induces EMT and leads to a poor prognosis [44,45]. HIST1H2BN is a part of the H2B protein family whose unregulated expression causes cancer and is a predictor of poor prognosis in ovarian cancer [46]. Hemoglobin beta (HBB) is an oxygen transporter that promotes BC cell aggression and a poor prognosis [48]. By releasing FOXC1 transcription factor, FLNB exon 30 skipping (gene splicing) induces EMT, and expression of EMT gene signature induces tumorigenicity [49]. Cancer stem cell survival and proliferation are in uenced by redox control and oxidative stress. Peroxiredoxin 4 (PRDX4) catalyses hydrogen peroxide and regulate hydrogen peroxide signaling leading to tumor recurrence and therapy resistance [50,51]. BRD4 is a downstream target of voltage-dependent anion channels (VDAC1), which are found on the outer mitochondrial membrane. VDAC1 overexpression causes breast cancer proliferation, is associated with a poor prognosis, and is linked to therapy resistance to BRD inhibitors in BC [50]. The GOT2 promoter is regulated by the ZBRK1 and BRCA1 complex, which binds to it and regulates its expression. Impaired complex binding contributes to uncontrollable expression, which promotes cell proliferation [52]. In lapatinib-resistant BC, a high level of HSPA5 was found [53,54]. SERPINH1 is a chaperone protein, and its high expression has been linked to a more aggressive

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.