Molecular Medicine Reports Molecular Medicine Reports 1791-2997 1791-3004 D.A. Spandidos 10.3892/mmr.2014.1980 mmr-09-04-1209 Articles BMI-1 is important in bufalin-induced apoptosis of K562 cells LIANXIAOYUN1 WANGHAO2 WEIXUCANG3 WANGYI3 WANGQISHAN3 GUOLIANG2 ZHAOYUAN3 CHENXIEQUN1 Department of Hematology, Xijing Hospital, Fourth Military Medical University, Xi’an, Shaanxi 710032, P.R. China Department of Hematology, Xi’an Central Hospital, Xi’an, Shaanxi 710003, P.R. China Department of Hematology, Shaanxi Provincial People’s Hospital, Xi’an, Shaanxi 710068, P.R. China Correspondence to: Dr Xiequn Chen, Department of Hematology, Xijing Hospital, Fourth Military Medical University, 17 Changle West Road, Xi’an, Shaanxi 710003, P.R. China, E-mail: xiequnbchen@163.com 4 2014 21 02 2014 9 4 1209 1217 21 05 2013 23 01 2014 Copyright © 2014, Spandidos Publications 2014 This is an open-access article licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License. The article may be redistributed, reproduced, and reused for non-commercial purposes, provided the original source is properly cited.

The purpose of this study was to analyze the effects of bufalin on the gene expression of K562 cells and on the expression of BMI-1 pathway constituents in K562 cell apoptosis. K562 cells were treated with bufalin, and the inhibition rate and apoptosis were detected by an MTT assay, flow cytometry and a microarray assay. BMI-1, p16INK4a and p14ARF were examined by quantitative polymerase chain reaction (qPCR). Bufalin induced significant changes in the gene expression of the K562 cells; 4296 genes were differentially expressed, 2185 were upregulated and 2111 were downregulated. The most upregulated genes were associated with transcription regulation, while the most downregulated genes were associated with the non-coding RNA metabolic processes and DNA repair. qPCR analysis demonstrated that BMI-1 was overexpressed in the K562 cells. Bufalin is able to downregulate BMI-1 expression levels in K562 cells prematurely and cause an increase in the expression levels of p16INK4a and p14ARF. Moreover, bufalin downregulated BCR/ABL expression levels in a time-dependent manner, and the expression of BCR/ABL was not associated with the upregulation or downregulation of BMI-1 expression. Bufalin may induce K562 cell apoptosis by downregulating BMI-1 expression levels and accordingly upregulating the expression levels of p16INK4a and p14ARF. Bufalin may also induce K562 cell apoptosis via downregulating BCR/ABL expression levels, and this pathway may be independent of the BMI-1 pathway.

 bufalin-induced apoptosis BMI-1 K562 cells Introduction

Bufalin is a traditional Chinese medicine and it is the major digoxin-like immunoreactive component of Chan Su, which is obtained from the skin and parotid venom glands of toads (1). It is a cardioactive C-24 steroid, with the molecular formula C24H34O4 and a relative molecular weight of 386.5. Bufalin exhibits a variety of biological activities, such as cardiotonic, anesthetic, blood pressure stimulatory, respiratory and antineoplastic effects (2). In terms of its antitumor activity, bufalin has been demonstrated to inhibit the growth of gastric, colon, breast, prostate, gynecological, hepatocellular and bladder tumors (3). It is also able to induce strong differentiation and apoptosis of myeloid leukemia cells, such as HL60, U937 and K562 cells. The mechanisms by which bufalin induces leukemia cell apoptosis are complex and involve a diverse range of cell signals (4–7). It is capable of inhibiting sodium-potassium-ATPase, and then activating apoptotic pathways (8,9). It is also able to activate the mitogen-activated protein kinase (MAPK) signaling pathway (9); downregulate the expression of the proto-oncogene c-myc and the anti-apoptotic protein Bcl-2 (10); and reduce the quantity, activity and mRNA levels of topoisomerase II (11). However, the precise pathways by which bufalin induces apoptosis remain unclear, and an improved understanding of such pathways is an essential goal of further study. To determine the global molecular mechanisms underlying bufalin-induced leukemia cell apoptosis, the present study analyzed the effects of bufalin on the gene expression of K562 cells using a Microarray assay. Based on this experiment, the expression of the BMI-1 pathway components in K562 cell apoptosis was also analyzed.

BMI-1 is a member of the polycomb-group gene family, that is known to be essential in supporting the self-renewal of stem cells through their epigenetic transcriptional regulation (12–15). BMI-1 is located on chromosome 10p13, a region involved in chromosomal translocations in infant leukemia (16). BMI-1 is a transcriptional repressor, the most widely known of the INK4a/ARF locus, that encodes the cell cycle regulators and tumor suppressors p16INK4a and p14ARF (17,18). Downregulation of p16INK4a and p14ARF inhibits cell apoptosis that is induced by p53 and retinoblastoma protein (pRB). It has become clear that BMI-1 also functions in protecting against oxidative stress. In the absence of BMI-1, reactive oxygen species (ROS) accumulate, which are associated with the activation of DNA damage response pathways and increased apoptosis (19). Therefore, overexpression of BMI-1 results in cell immortalization and tumorigenesis, whereas downregulation in tumor (stem) cells results in impaired self-renewal and proliferation (20,21). It has been demonstrated that BMI-1 is often overexpressed in various types of leukemia and its depletion in those leukemia cells leads to cell proliferation arrest, differentiation and apoptosis (22). Lessard and Sauvageau (20) demonstrated that the proliferative potential of acute myeloid leukemia (AML) stem and progenitor cells that lacked BMI-1 was compromised, as the cells eventually underwent proliferation arrest and exhibited signs of differentiation and apoptosis. In addition, the absence of BMI-1 led to transplant failure of the leukemia, and these proliferative defects were completely rescued by BMI-1. These findings suggest that the BMI-1 gene is a potential target for therapeutic intervention in leukemia. Zhu et al (22) demonstrated that small interfering RNA (siRNA)-mediated silencing of BMI-1 in U937 cells led to reduced cell growth and proliferation, and increased cell apoptosis. Meng et al (23) also demonstrated that K562 cells transfected with an antisense BMI-1 plasmid grew significantly slower than controls. Their colony-forming ability decreased significantly, and their p16INK4a expression levels were upregulated more than that of the controls.

In the present study, the effect of bufalin on BMI-1 expression in K562 cells was investigated, and the expression of p16INK4a and p14ARF during the bufalin-induced apoptosis was also determined.

 Materials and methods Materials

Bufalin (10.0 mg/vial) was purchased from Sigma-Aldrich (St. Louis, MO, USA) and was dissolved in anhydrous alcohol to make 0.01 mol/l stock solution. The solution was stored at −20°C and diluted in RPMI-1640 when used. K562 cells were obtained from the central laboratory of The First Affiliated Hospital of Xi’an Jiaotong University (Xi’an, China). Bone marrow mononuclear cells were acquired from two female healthy iron-deficiency anemia volunteers (ages, 38 and 29 years) as normal controls. The study was approved by the Ethics Committee of Shaanxi Provincial People’s Hospital and written informed consent was obtained from the volunteers.

 Cell culture

The K562 human erythrocyte leukemia cell line was grown in suspension culture in RPMI-1640 supplemented with 10% (v/v) fetal bovine serum (HyClone, Logan, UT, USA), at 37°C in a humidified atmosphere with 5% CO2. K562 cells at an exponential growth stage were employed in all of the experiments.

 MTT assay

Cells were seeded in 96-well plates at a density of 105 cells/ml in 100 μl of medium and cultured overnight. Each experiment was performed in triplicate. Subsequently, the cells in the wells were treated with fresh medium containing different concentrations of bufalin diluted from the stock solution. The resulting concentrations of bufalin were 0.025, 0.05, 0.1, 0.5, 1.0 and 2.0 μmol/l. The culture solution was applied to the wells as a blank control. The cells that were not treated with bufalin served as negative controls. Cells were cultured for 48 or 72 h. At the indicated times, MTT solution [5 mg/ml in 20 μl phosphate-buffered saline (PBS)] was added to each well and incubated for 4 h. Following removal of the medium, 200 μl dimethylsulfoxide was added to each well to dissolve the formazan crystals. The absorbance at 490 nm was determined using a microplate reader (Beijing Pulangxin Technology Co., Ltd., Beijing, China). The inhibition rate of cell proliferation was calculated as follows: Inhibition rate (%) = A490 (control) − A490 (test) / A490 (control) − A490 (blank) × 100.

 Flow cytometric analysis

To evaluate the effect of bufalin on K562 cell early apoptosis and to determine a suitable concentration of bufalin for microarray experiments, an Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) apoptosis detection kit (BD Immunocytometry Systems, Franklin Lakes, NJ, USA) for flow cytometry was used to examine the apoptosis. K562 cells were treated with 0.025, 0.05, 0.5 or 1.0 μmol/l bufalin for 48 h. The cells were washed twice with cold PBS and then stained with Annexin V-FITC and PI according to the manufacturer’s instructions. Samples were analyzed using a FACSCalibur flow cytometer (Beckman Coulter, Inc., Brea, CA, USA) following staining.

 RNA isolation and microarray analysis RNA extraction and purification

Total RNA was extracted using TRIzol reagent (cat. no. 15596-018; Life technologies, Carlsbad, CA, USA) following the manufacturer’s instructions, and checked for an RNA integrity number to inspect RNA integration by an Agilent Bioanalyzer 2100 (Agilent technologies, Santa Clara, CA, USA). Qualified total RNA was further purified using an RNeasy mini kit (cat. no. 74106; Qiagen GmBH, Hilden, Germany) and RNase-Free DNase set (cat. no. 79254; Qiagen GmBH).

 RNA amplification and labeling

Total RNA was amplified and labeled using a Low Input Quick Amp Labeling kit, one-color (cat. no. 5190-2305; Agilent Technologies), according to the manufacturer’s instructions. Labeled cRNA were purified using the RNeasy mini kit (cat. no. 74106; Qiagen GmBH).

 Hybridization

Each slide was hybridized with 1.65 μg Cy3-labeled cRNA using a Gene Expression Hybridization kit (cat. no. 5188–5242; Agilent Technologies) in a hybridization oven (cat. no. G2545A; Agilent Technologies), according to the manufacturer’s instructions. Following 17 h of hybridization, slides were washed in staining dishes (cat. no. 121; Thermo Shandon, Waltham, MA, USA) with the Gene Expression Wash Buffer kit (cat. no. 5188–5327; Agilent Technologies), following the manufacturer’s instructions.

 Data acquisition

Slides were scanned with an Agilent Microarray Scanner (cat. no. G2565CA; Agilent Technologies) with the following default settings: Dye channel, green; scan resolution, 5 μm; PMT, 100%, 10%. Additionally, 16-bit Feature Extraction software 10.7 (Agilent Technologies) was used, and raw data were normalized by a quantile algorithm (GeneSpring software 11.0; Agilent Technologies).

 qPCR

Total RNA was isolated and purified as previously described. Reverse transcription of the total RNA was conducted using the Cloned AMV First-Strand cDNA Synthesis kit (Invitrogen Life Technologies, Carlsbad, CA, USA). qPCR was performed using SYBR-Green chemistry in an ABI StepOnePlus system (Applied Biosystems, New York, NY, USA). GAPDH served as a housekeeping gene. PCR primer sequences were as follows: Forward: 5′-TGGGTGTGAACCATGAGAAGT-3′ and reverse: 5′-TGAGTCCTTCCACGATACCAA-3′ for human GAPDH; forward: 5′-CTTGGCTCGCATTCATTTTCT-3′ and reverse: 5′-CTCAGTGATCTTGATTCTCGTTGT-3′ for BMI-1; forward: 5′-GACAAAGAAAACGCCACAAATC-3′ and reverse: 5′-CTGAAACTGAATCCTGATCCAAC-3′ for MDM2; forward: 5′-GTCAGGACCTTCGTAGCATTG-3′ and reverse: 5′-CTCAGGGCACAGGAAAACATC-3′ for E2F; forward: 5′-GAGGGCTTCCTGGACACG-3′ and reverse: 5′-TCTTTCAATCGGGGATGTCTG- 3′ for p16INK4a; forward: 5′-TGTGGAGTTGGACTGAATGCT-3′ and reverse: 5′-TGACAAGGTCTTCTCTTCATAGGTT-3′ for p14ARF; and forward: 5′-GGAAGAAATACAGCCTGACGG-3′ and reverse: 5′-AGGAGGTTCCCGTAGGTCAT-3′ for BCR/ABL.

Triplicates were made for each sample. For each sample, an amplification plot and corresponding dissociation curves were examined. Relative quantification analysis was performed using the comparative CT method (2−ΔΔCT). The formula was as follows: 2−ΔΔCT = 2-[(CT gene of interest − CT internal control) sample A − (CT gene of interest − CT internal control) sample B)]

 Statistical analysis

The data are expressed as the mean ± standard deviation from experiments performed in triplicate. Student’s t-test was used to identify statistically significant differences between the experimental and control groups. The statistical analyses were performed using SPSS software, version 13.0 (SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

 Results Effect of bufalin on the growth of K562 cells

To investigate the effects of bufalin on K562 cells and to determine a suitable concentration for microarray experiments, the effect of various doses of bufalin on the viability of K562 cells was tested using an MTT assay and flow cytometry.

MTT results

The cells were treated for 48 and 72 h as shown in Fig. 1A, and cell growth was inhibited by bufalin in a dose- and time-dependent manner (Fig. 1A–C). The calculated IC50 values (50% inhibitory concentration) were 0.153 and 0.028 μmol/l for cells treated for 48 and 72 h, respectively. The difference in the apoptosis rate between bufalin treatment at a concentration of 0.025 and 0.05 μmol/l at 48 h was significant (P<0.05).

 Flow cytometric analysis

K562 cells were treated with 0.025, 0.05, 0.5 or 1.0 μmol/l bufalin for 48 h. Flow cytometric analysis showed that bufalin induced the apoptosis of cells in a dose-dependent manner (Fig. 2A–E).

 Microarray analysis of bufalin-induced K562 cell apoptosis

As there is no standard concentration for the use of bufalin in cell culture for microarray experiments, several different concentrations were used to determine the effect of bufalin on the growth of K562 cells at various time points. Chen et al (5) used IC50 values as their standard concentrations for microarray analysis of bufalin-induced HL-60 apoptosis. In the present study, based on the results of the MTT and flow cytometry assays, a concentration of 0.5 μmol/l for 48 h was selected for the microarray experiments. Cells that were not treated with bufalin served as the control. Triplicates were made for the samples treated with bufalin. Gene functional annotation analysis and gene network analysis were performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov/). All data underwent log transformation. A ratio of ≥1 was regarded as upregulated and a ratio of ≤1 was considered as downregulated. The results showed that bufalin induced significant changes in the gene expression of K562 cells; 4296 genes were differentially expressed, 2185 were upregulated and 2111 were downregulated. As the focus of the study was on enriched functional categories rather than on individual genes, DAVID Annotation Cluster was employed to analyze the molecular mechanisms of bufalin-induced apoptosis. Among the genes with high fold-change values, the most upregulated genes were associated with transcription regulation (Table I), while the most downregulated genes were associated with the non-coding RNA (ncRNA) metabolic process and DNA repair (Table II). Moreover, a significant signaling pathway, the MAPK pathway, was identified, which had been verified as a pathway in bufalin-induced apoptosis (9). Of the genes overexpressed by bufalin in K562 cells, BMI-1, E2F and MDM2 were selected for qPCR analysis and the results were in agreement with the microarray data (Table III).

 Effect of bufalin on the BMI-1, p16<sup>INK4a</sup>, E2F, p14<sup>ARF</sup>, MDM2 and Brc/Abl expression

K562 cells were treated with 0.5 or 0.05 μmol/l bufalin for 3, 6, 18, 24 or 48 h, respectively, and then qPCR was performed. The results indicated that BMI-1, E2F and MDM2 expression levels were upregulated in the K562 cells treated with 0.5 μmol/l bufalin for 48 h, which was in agreement with the results of the microarray experiment (Table III).

The expression of BMI-1 was decreased relative to the baseline level at 18 and 24 h following treatment with a concentration of 0.5 μmol/l bufalin, and then increased 1.63-fold 48 h after treatment. The minimum expression of the gene was at 24 h after bufalin treatment. In addition, the expression of BMI-1 target genes p16INK4a and p14ARF, and the effect of bufalin on BCR/ABL expression was investigated. The results indicated that the expression of p16INK4a began to increase 18 h following bufalin treatment, and increased 107.60-fold 48 h after bufalin treatment. The rate of increase of p14ARF expression was slower than that of p16INK4a, and p14ARF expression initially decreased and returned to the orignial level at 48 h after treatment (Fig. 3A). The expression of BCR/ABL decreased gradually in a time-dependent manner, and had decreased 6.73-fold 48 h after bufalin treatment (Fig. 3B).

The effect of bufalin treatment at a concentration of 0.05 μmol/l on K562 cells was the same as that of 0.5 μmol/l, except that the minimum expression of BMI-1 was at 18 h following treatment. p16INK4a expression levels began to increase 6 h following bufalin treatment and increased 30.18-fold 48 h after treatment. p14ARF expression began to increase 24 h following bufalin treatment, and increased 1.26-fold 48 h after the treatment.

In conclusion, p16INK4a expression began to increase as BMI-1 expression decreased, and the increase in p14ARF expression lagged behind that of p16INK4a. The expression of BCR/ABL does not appear to be correlated with the upregulation or downregulation of BMI-1. Bufalin-induced apoptosis may be mediated by downregulating the expression of BMI-1, and accordingly upregulating the expression of p16INK4a and p14ARF, and downregulating the expression of BCR/ABL in K562 cells.

In addition, BMI-1 expression was observed in K562 cells treated with bufalin for 24 h. Bone marrow mononuclear cells from healthy iron-deficiency anemia volunteers served as the normal control. BMI-1 expression levels in the untreated control cells were higher than those in the normal control cells, and they were downregulated in the drug-treated cells (Fig. 4).

 Discussion

The potential antitumor activity of natural products used in traditional Chinese medicines has been noted (24). Cinobufacini (Huachansu), a Chinese medicine prepared from dried toad skin, has been widely used for the treatment of various types of cancer in China (25). Bufalin, which is the main active component of cinobufacini, has attracted much attention from researchers since the 1990’s (2). Experiments have demonstrated that it is capable of inducing apoptosis in a variety of types of tumor cell, including leukemia cells such as U937, HL-60, ML1 and THP1 (5,6,9–11,26). The mechanisms by which bufalin induces apoptosis are complex and involve a diverse range of cell signals (2,3). Data suggest that activation of the MAPK signaling pathway is one of the most important signal transduction pathways for the induction of apoptosis by bufalin (2). Bufalin is also able to induce apoptosis of prostate cancer cells via p53- and Fas-mediated apoptotic pathways (27). In addition, bufalin induces lung cancer A549 cell apoptosis via inhibition of the PI3K/Akt signaling pathway (28). Recently, it has been indicated that bufalin-induced apoptosis is associated with the disruption of the mitochondrial membrane potential, cytochrome c release, ROS generation and caspase activation (29).

In the present study, it was demonstrated that bufalin is able to induce K562 cell apoptosis in a time- and dose-dependent manner, and the microarray analysis results suggest that bufalin is able to induce significant changes in the gene expression of K562 cells. The most upregulated functional annotation cluster of bufalin-induced apoptosis of K562 cells was associated with transcription regulation. For example, BMI-1, AFF4, DDIT3, ING3, E2F1, E2F8, DENND4A, ATF, CCNL1, CCNL2, CCTI, CCT2, CDK9, JUN, CDKN1B and certain zinc finger proteins were upregulated in K562 cells treated with bufalin. DDIT3, also known as CCAAT/enhancer binding protein ζ, has been demonstrated to be involved in the regulation of cellular growth and differentiation (30). The levels of DDIT3 transcription have been demonstrated to be downregulated in myeloid malignancies (31), and overexpression of DDIT3 transcripts has been shown to induce increased apoptosis of myeloid cells and block cell progression from the G1 to S phase (32,33). ING3, a member of the ING family, has been reported to be involved in p53-mediated transcription modulation, cell cycle control and the induction of apoptosis. It was demonstrated that the expression levels of ING3 are decreased in melanoma and head and neck squamous cell carcinoma, and predict a poor prognosis (34). DENND4A, also known as C-myc promoter-binding protein MBP-1, acts as a general transcriptional repressor (35). It binds to c-myc promoter sequences and transcriptionally represses the c-myc gene (36–38). It has been demonstrated that overexpression of the MBP-1 gene induces apoptosis in several types of cancer cell line (39). The family of zinc finger proteins is composed of regulatory proteins that participate in a number of molecular and cellular pathways, and is considered to be one of the most profuse regulatory protein families in eukaryotic cells. Their functions are extremely diverse and include DNA recognition, RNA packaging, transcriptional activation, regulation of apoptosis, protein folding and assembly, and lipid binding (40–41).

The present study demonstrated that bufalin downregulated certain genes participating in ncRNA metabolic processes, such as DGCR8, FTSJ3, MINA, NSUN2, U2AF1, CPSF3, GEMIN4, LCMT2 and PRMT5. DGCR8 is an RNA-binding protein that assists the RNase III enzyme Drosha in the processing of microRNAs (miRNAs); it is required for the processing of pri-miRNAs to pre-miRNAs. Using RNA from wild-type embryonic stem cells as a reference, Wang et al (42) observed a global loss of miRNAs in Dgcr8 knockout cells but identified normal expression in heterozygous Dgcr8 cells, by using miRNA microarray analysis. As a result, bufalin is able to interfere with the processing of miRNAs via downregulating certain genes, such as DGCR8.

FTSJ3 is a human NIP7-interacting protein. NIP7 is one of the numerous trans-acting factors required for eukaryotic ribosome biogenesis, which interacts with nascent pre-ribosomal particles and dissociates as they complete maturation and are exported to the cytoplasm. Conditional knockdown revealed that depletion of FTSJ3 affects cell proliferation and causes pre-rRNA processing defects (43). Mina53 is involved in cell division (44) and is directly involved in ribosome biogenesis, most likely during the assembly process of pre-ribosomal particles (45). It may be important in cell growth and survival. The inhibition of Mina53 expression results in the suppression of cell proliferation in certain cell lines (46). PRMT5 is a type II PRMT that catalyzes the symmetrical dimethylation of arginine residues within target proteins (47). PRMT5 is highly conserved among yeast, animals and higher plants, and has been implicated in diverse cellular and biological processes including transcriptional regulation, RNA metabolism, ribosome biogenesis, Golgi apparatus structural maintenance and cell cycle progression (48). Overexpression of PRMT5 promotes tumor cell growth and is associated with poor disease prognosis (49).

Bufalin also downregulated certain genes associated with DNA repair, such as RUVBL1, Cdc7, ERCC5, Grx2, MSH6, MUTYH, NEIL1, NTHL1, PRMT6, tyrosyl-DNA phosphodiesterase I (Tdp1) and MYC. RUVBL1 is a member of the AAA+ (ATPase associated with diverse cellular activities) family of proteins that is associated with several chromatin-remodeling complexes and is important in transcriptional regulation, the DNA damage response, telomerase activity, snoRNP assembly, cellular transformation and cancer metastasis (50–52). RUVBL1 is overexpressed in a number of different types of cancer and interacts with major oncogenic factors, such as by regulating the function of β-catenin and c-Myc (50–51). RUVBL1 blocks p53-mediated apoptosis by repressing the expression of p53 and its target genes, and siRUVBL1 is also able to induce the apoptotic death of HCT116 (p53−/−) cells, suggesting that RUVBL1 is able to suppress apoptosis that is mediated by p53 and other molecules (53). Cdc7 is a conserved serine-threonine kinase, as well as an essential replication regulator (54). Knockdown of Cdc7 has been demonstrated to cause the death of cancer cells, but not normal cells, via p53-dependent pathways which are able to arrest the cell cycle, presumably in the G1 phase (55,56). It has also been reported that Cdc7 knockdown induced p38-dependent cell death in HeLa cells (57). Cdc7 depletion is able to induce DNA damage in cancer cells (58), and activated Chk2 is capable of stabilizing the FoxM1 transcription factor, which induces the expression of cyclin B1 (59). Thus, anticancer drugs, such as bufalin, that facilitate Cdc7 depletion are able to induce cancer cell death via numerous pathways.

Grx2 belongs to the oxidoreductase family and it has been reported that Grx2 protects cells against a variety of oxidative insults (60–62). Conversely, knockdown of Grx2 sensitizes HeLa cells to anticancer drugs (63). A study indicated that Grx2 also has an anti-apoptotic function. Grx2 was demonstrated to protect cells against H2O2-induced apoptosis via its peroxidase and dethiolase activities; in particular, Grx2 prevents complex I inactivation and preserves mitochondrial function (64).

The DNA mismatch repair (MMR) system is one of a number of DNA repair mechanisms. MSH6 is an MMR-related protein that belongs to the MutS family. MSH proteins recognize errors in the genome sequence during replication, and prevent the duplication of the damaged strand and repair single strand breaks (65). Bufalin also downregulates certain genes involved in the base excision repair and nucleotide excision repair pathways of oxidization-induced DNA damage, such as MUTYH, NEIL1 and NTHL1. Bufalin is capable of inducing apoptosis via a ROS-dependent mitochondrial death pathway (29) and as a result, bufalin induces apoptosis by inducing ROS production and inhibiting repair of DNA damage induced by oxidization.

Tdp1 is a cellular enzyme that repairs the irreversible topoisomerase I (Top1)-DNA complexes and confers chemotherapeutic resistance to Top1 inhibitors. Inhibiting Tdp1 provides an attractive approach to potentiating clinically used Top1 inhibitors (66,67). Bufalin is able to reduce the quantity, activity and mRNA levels of topoisomerase II (11). Thus, bufalin is a Top II inhibitor and a sensitizer that increases the cytotoxic effects of Top I inhibitors.

The qPCR analysis of the present study demonstrated that bufalin is able to downregulate BMI-1 expression levels in K562 cells after 3 h. The expression levels in the K562 cells were higher than in the normal controls and decreased following treatment with bufalin. BMI-1 is required for maintenance and self-renewal of normal and leukemia stem and progenitor cells. Leukemia stem and progenitor cells lacking BMI-1 eventually undergo proliferation arrest and exhibit signs of differentiation and apoptosis, leading to transplant failure of the leukemia (20). Rizo et al (68) suggested that repression of BMI-1 in cord blood CD34+ cells impaired their long-term expansion and progenitor-forming capacity, in cytokine-driven liquid cultures and in bone marrow stromal co-cultures. In addition, the long-term culture-initiating cell frequencies were markedly decreased upon knockdown of BMI-1, indicating an impaired maintenance of stem and progenitor cells. The reduced progenitor and stem cell frequencies were associated with increased expression levels of p14ARF and p16INK4A and enhanced apoptosis (68). It has been verified that BMI-1 is overexpressed in hematological malignancies and is correlated with a poor prognosis (22,69). The findings of the present study also demonstrated that BMI-1 is overexpressed in K562 cells, and that the increase in p16INK4a and p14ARF expression levels is associated with the downregulation of BMI-1 expression levels. We propose that bufalin may induce K562 cell apoptosis via downregulating BMI-1 expression and accordingly upregulating the expression of p16INK4a and p14ARF.

To date, it has not been demonstrated that expression of BMI-1 alone is sufficient to induce leukemia. However, various studies suggest that BMI-1 may be an important collaborating factor in the transformation process. For example, BMI-1 is able to cooperate with H-RAS to induce aggressive breast cancer with brain metastases (70,71). Using qPCR, Merkerova et al (72) demonstrated significantly increased levels of BMI-1 transcripts in chronic myelogenous leukemia (CML) cells. Using siRNA silencing, the impact of BMI-1 inhibition on the growth and cell cycle of BCR/ABL-positive CML cell lines was measured, and it was demonstrated that BMI-1 inhibition did not affect the proliferation rate or the cell cycle of these cells. As it is well-known that the growth deregulation in CML cells is caused by the BCR/ABL oncogene, BMI-1 may have a secondary role in the proliferation of these cells (73). Merkerova et al (72) also reduced BCR/ABL expression by siRNA in K562 cells to determine if BMI-1 upregulation was dependent on BCR/ABL expression, and demonstrated that the BMI-1 transcript levels did not change after BCR/ABL inhibition, which suggests that the increased expression levels of BMI-1 are not directly associated with BCR/ABL. In the present study, the expression of BCR/ABL did not appear to be correlated with the upregulation or downregulation of BMI-1 expression levels. Bufalin may induce K562 cell apoptosis via downregulating BCR/ABL expression levels, and this pathway may occur independently of the BMI-1 pathway.

It has become clear that BMI-1 also functions in protecting against oxidative stress. In the absence of BMI-1, ROS accumulate and are associated with the activation of DNA damage response pathways and increased apoptosis. BMI-1-mediated control over ROS levels may occur via impaired mitochondrial functions, independent of the INK4a/ARF pathway (19). BMI-1 may be required to protect hematopoietic stem/progenitor cells from apoptosis induced by oxidative stress conditions (68). As mentioned previously, bufalin is able to induce apoptosis via a ROS-dependent mitochondrial death pathway and inhibit repair of DNA damage induced by oxidization (29). The association between bufalin, BMI-1 and DNA damage repair genes requires further investigation.

The signaling pathways that regulate BMI-1 have not been extensively studied. SALL4, an oncogene that is expressed in AML, is capable of upregulating BMI-1 expression by directly binding to its promoter (74). Furthermore, BMI-1 expression appears to be under the control of miRNAs, including miRNA-128. miRNA-128 causes a marked decrease in the expression levels of BMI-1 by direct regulation of the BMI-1 mRNA 3-untranslated region, through a single miRNA-128 binding site (75,76). miRNA-128a increases intracellular ROS levels by targeting BMI-1 and inhibits medulloblastoma cancer cell growth by promoting senescence (77). However, BMI-1 is not expressed in a cell cycle-regulated manner in various types of primary and transformed cells and is not affected by overexpression of p16INK4a (19). In the present study, BMI-1 expression levels increased 48 h following bufalin treatment. Although the mechanism requires in-depth research, we propose certain potential possibilities: i) Bufalin may upregulate genes upstream of BMI-1, such as SALL4; ii) bufalin may downregulate mRNAs and eliminate their suppression of BMI-1, for example, bufalin is capable of interfering with the processing of miRNA via downregulating DGCR8; and iii) transcription factor E2F-1 may directly regulate BMI-1, which represses the INK4A gene, thereby promoting phosphorylation of the pRb and activation of E2F, as has been demonstrated previously (78). The present study showed that bufalin is able to upregulate E2F-1 expression and this may be a reason why BMI-1 increased 48 h after drug treatment. The phenomena of BMI-1 expression levels increasing at later time also suggests that BMI-1 alone cannot induce apoptosis of K562 cells.

 References KrennLKoppBBufadienolides from animal and plant sourcesPhytochemistry481291998 QiFLiAInagakiYKokudoNTamuraSNakataMTangWAntitumor activity of extracts and compounds from the skin of the toad Bufo bufo gargarizans CantorInt Immunopharmacol113423492011 TakaiNKiraNIshiiTYoshidaTNishidaMNishidaYNasuKNaraharaHBufalin, a traditional oriental medicine, induces apoptosis in human cancer cellsAsian Pac J Cancer Prev133994022012 ZhangLNakayaKYoshidaTKuroiwaYInduction by bufalin of differentiation of human leukemia cells HL60, U937, and ML1 toward macrophage/monocyte-like cells and its potent synergistic effect on the differentiation of human leukemia cells in combination with other inducersCancer Res52463446411992 ChenAYuJZhangLSunYZhangYGuoHZhouYMitchelsonKChengJMicroarray and biochemical analysis of bufalin-induced apoptosis of HL-60 cellsBiotechnol Lett314874942009 HuangCChenAGuoMYuJMembrane dielectric responses of bufalin-induced apoptosis in HL-60 cells detected by an electrorotation chipBiotechnol Lett29130713132007 NumazawaSShinokiMAItoHYoshidaTKuroiwaYInvolvement of Na+, K(+)-ATPase inhibition in K562 cell differentiation induced by bufalinJ Cell Physiol1601132101994 KawazoeNAiuchiTMasudaYNakajoSNakayaKInduction of apoptosis by bufalin in human tumor cells is associated with a change of intracellular concentration of Na+ ionsJ Biochem1262782861999 WatabeMMasudaYNakajoSYoshidaTKuroiwaYNakayaKThe cooperative interaction of two different signaling pathways in response to bufalin induces apoptosis in human leukemia U937 cellsJ Biol Chem27114067140721996 MasudaYKawazoeNNakajoSYoshidaTKuroiwaYNakayaKBufalin induces apoptosis and influences the expression of apoptosis-related genes in human leukemia cellsLeuk Res195495561995 HashimotoSJingYKawazoeNMasudaYNakajoSYoshidazTBufalin reduces the level of topoisomerase II in human leukemia cells and affects the cytotoxicity of anticancer drugsLeuk Res218758831997 ParkIKQianDKielMBmi-1 is required for maintenance of adult self-renewing haematopoietic stem cellsNature4233023052003 DickJEStem cells: self-renewal writ in bloodNature4232312332003 LessardJSauvageauGBmi-1 determines the proliferative capacity of normal and leukaemic stem cellsNature4232552602003 IwamaAOguroHNegishiMEnhanced self-renewal of hematopoietic stem cells mediated by the polycomb gene product Bmi-1Immunity218438512004 AlkemaMJWiegantJRaapAKBernsAvan LohuizenMCharacterization and chromosomal localization of the human proto-oncogene BMI-1Hum Mol Genet2159716031993 JacobsJJKieboomKMarinoSDePinhoRAvan LohuizenMThe oncogene and polycomb-group gene bmi-1 regulates cell proliferation and senescence through the Ink4a locusNature3971641681999 DhawanSTschenSIBhushanABmi-1 regulates the Ink4a/Arf locus to control pancreatic beta-cell proliferationGenes Dev239069112009 SchuringaJJVellengaERole of the polycomb group gene BMI1 in normal and leukemic hematopoietic stem and progenitor cellsCurr Opin Hematol172942992010 LessardJSauvageauGBmi-1 determines the proliferative capacity of normal and leukaemic stem cellsNature4232552602003 Valk-LingbeekMEBruggemanSWvan LohuizenMStem cells and cancer; the polycomb connectionCell1184094182004 ZhuWHuangLXuXQianHXuWAnti-proliferation effect of BMI-1 in U937 cells with siRNAInt J Mol Med258898952010 MengXXLiuWHLiuDDZhaoXYSuBLConstruction of antisense Bmi-1 expression plasmid and its inhibitory effect on K562 cells proliferationChin Med J (Engl)118134613502005 MehtaRGMurilloGNaithaniRPengXCancer chemoprevention by natural products: how far have we come?Pharm Res279509612010 MengZYangPShenYBeiWZhangYGeYNewmanRACohenLLiuLThorntonBChangDZLiaoZKurzrockRPilot study of huachansu in patients with hepatocellular carcinoma, nonsmall-cell lung cancer, or pancreatic cancerCancer115530953182009 KawazoeNAiuchiTMasudaYNakajoSNakayaKInduction of apoptosis by bufalin in human tumor cells is associated with a change of intracellular concentration of Na+ ionsJ Biochem1262782861999 YuCHKanSFPuHFJea ChienEWangPSApoptotic signaling in bufalin- and cinobufagin-treated androgen-dependent and -independent human prostate cancer cellsCancer Sci99246724762008 ZhuZSunHMaGWangZLiELiuYLiuYBufalin induces lung cancer cell apoptosis via the inhibition of PI3K/Akt pathwayInt J Mol Sci13202520352012 SunLChenTWangXChenYWeiXBufalin induces reactive oxygen species dependent bax translocation and apoptosis in ASTC-a-1 cellsEvid Based Complement Alternat Med20112490902011 WangYLQianJLinJYaoDMQianZZhuZHLiJYMethylation status of DDIT3 gene in chronic myeloid leukemiaJ Exp Clin Cancer Res29542010 QianJChenZLinJWangWCenJDecreased expression of CCAAT/enhancer binding protein zeta (C/EBPzeta) in patients with different myeloid diseasesLeuk Res29143514412005 FriedmanADGADD153/CHOP, a DNA damage-inducible protein, reduced CAAT/enhancer binding protein activities and increased apoptosis in 32D c13 myeloid cellsCancer Res56325032561996 MatsumotoMMinamiMTakedaKSakaoYAkiraSEctopic expression of CHOP (GADD153) induces apoptosis in M1 myeloblastic leukemia cellsFEBS Lett3951431471996 LuMChenFWangQWangKPanQZhangXDownregulation of inhibitor of growth 3 is correlated with tumorigenesis and progression of hepatocellular carcinomaOncol Lett447522012 GhoshAKSteeleRRayRBFunctional domains of c-myc promoter binding protein 1 involved in transcriptional repression and cell growth regulationMol Cell Biol19288028861999 RayRMillerDMCloning and characterization of a human c-myc promoter-binding proteinMol Cell Biol11215421611996 ChaudharyDMillerDMThe c-myc promoter binding protein (MBP-1) and TBP bind simultaneously in the minor groove of the c-myc P2 promoterBiochemistry34343834451995 RayRBInduction of cell death in murine fibroblasts by a c-myc promoter binding proteinCell Growth Differ6108910961995 GhoshAKSteeleRRyerseJRayRBTumor-suppressive effects of MBP-1 in non-small cell lung cancer cellsCancer Res6611907119122006 KimMKJeonBNKohDIParkSYYunCOHurMWRegulation of the cyclin-dependent kinase inhibitor 1A gene (CDKNIA) by the repressor BOZF1 through inhibition of p53 acetylation and transcription factor Sp1 bindingJ Biol Chem288705370642013 LaityJHLeeBMWrightPEZinc finger proteins: new insights into structural and functional diversityCurr Opin Struct Biol1139462001 WangYMedvidRMeltonCJaenischRBlellochRDGCR8 is essential for microRNA biogenesis and silencing of embryonic stem cell self-renewalNat Genet393803852007 MorelloLGColtriPPQuaresmaAJSimabucoFMSilvaTCSinghGNickersonJAOliveiraCCMooreMJZanchinNIThe human nucleolar protein FTSJ3 associates with NIP7 and functions in pre-rRNA processingPLoS One6e291742011 TsuneokaMKodaYSoejimaMTeyeKKimuraHA novel myc target gene, mina53, that is involved in cell proliferationJ Biol Chem27735450354592002 EilbrachtJKneisselSHofmannASchmidt-ZachmannMSProtein NO52-a constitutive nucleolar component sharing high sequence homologies to protein NO66Eur J Cell Biol842792942005 TanXPZhangQDongWGLeiXWYangZRUpregulated expression of Mina53 in cholangiocarcinoma and its clinical significanceOncol Lett3103710412012 BedfordMTClarkeSGProtein arginine methylation in mammals: who, what, and whyMol Cell331132009 GuZLiYLeePLiuTWanCWangZProtein arginine methyltransferase 5 functions in opposite ways in the cytoplasm and nucleus of prostate cancer cellsPLoS One7e440332012 BaoXZhaoSLiuTLiuYLiuYYangXOverexpression of PRMT5 promotes tumor cell growth and is associated with poor disease prognosis in epithelial ovarian cancerJ Histochem Cytochem612062172013 GallantPControl of transcription by Pontin and ReptinTrends Cell Biol171871922007 HuberOMénardLHaurieVNicouATarasDRosenbaumJPontin and reptin, two related ATPases with multiple roles in cancerCancer Res68687368762006 JhaSDuttaARVB1/RVB2: running rings around molecular biologyMol Cell345215332009 TaniueKOdaTHayashiTOkunoMAkiyamaTA member of the ETS family, EHF, and the ATPase RUVBL1 inhibit p53-mediated apoptosisEMBO Rep126826892011 ItoSIshiiAKakushoNTaniyamaCYamazakiSFukatsuRSakaue-SawanoAMiyawakiAMasaiHMechanism of cancer cell death induced by depletion of an essential replication regulatorPLoS One7e363722012 MontagnoliATencaPSolaFCarpaniDBrothertonDAlbaneseCSantocanaleCCdc7 inhibition reveals a p53-dependent replication checkpoint that is defective in cancer cellsCancer Res64711071162004 TudzarovaSTrotterMWWollenschlaegerAMulveyCGodovac-ZimmermannJWilliamsGHStoeberKMolecular architecture of the DNA replication origin activation checkpointEMBO J29338133942010 ImJSLeeJKATR-dependent activation of p38 MAP kinase is responsible for apoptotic cell death in cells depleted of Cdc7J Biol Chem28325171251772008 KimJMKakushoNYamadaMKanohYTakemotoNMasaiHCdc7 kinase mediates Claspin phosphorylation in DNA replication checkpointOncogene27347534822008 TanYRaychaudhuriPCostaRHChk2 mediates stabilization of the FoxM1 transcription factor to stimulate expression of DNA repair genesMol Cell Biol27100710162007 DailyDVlamis-GardikasAOffenDMittelmanLMelamedEHolmgrenABarzilaiAGlutaredoxin protects cerebellar granule neurons from dopamine-induced apoptosis by dual activation of the ras-phosphoinositide 3-kinase and jun n-terminal kinase pathwaysJ Biol Chem27621618216262001 EnokssonMFernandesAPPrastSLilligCHHolmgrenAOrreniusSOverexpression of glutaredoxin 2 attenuates apoptosis by preventing cytochrome c releaseBiochem Biophys Res Commun3277747792005 DiotteNMXiongYGaoJChuaBHHoYSAttenuation of doxorubicin-induced cardiac injury by mitochondrial glutaredoxin 2Biochim Biophys Acta17934274382009 LilligCHLonnMEEnokssonMFernandesAPHolmgrenAShort interfering RNA-mediated silencing of glutaredoxin 2 increases the sensitivity of HeLa cells toward doxorubicin and phenylarsine oxideProc Natl Acad Sci USA10113227132322004 WuHXingKLouMFGlutaredoxin 2 prevents H(2)O(2)-induced cell apoptosis by protecting complex I activity in the mitochondriaBiochim Biophys Acta1797170517152010 Conde-PérezprinaJCLeón-GalvánMÁKonigsbergMDNA mismatch repair system: repercussions in cellular homeostasis and relationship with agingOxid Med Cell Longev20127284302012 SirivoluVRVernekarSKMarchandCNaumovaACherguiARenaudAStephenAGChenFShamYYPommierYWangZ5-Arylidenethioxothiazolidinones as inhibitors of tyrosyl-DNA phosphodiesterase IJ Med Chem55867186842012 AlagozMGilbertDCEl-KhamisySChalmersAJDNA repair and resistance to topoisomerase I inhibitors: mechanisms, biomarkers and therapeutic targetsCurr Med Chem19387438852012 RizoAOlthofSHanLVellengaEde HaanGSchuringaJJRepression of BMI1 in normal and leukemic human CD34(+) cells impairs self-renewal and induces apoptosisBlood114149815052009 BhattacharyyaJMiharaKYasunagaSTanakaHHoshiMTakiharaYKimuraABMI-1 expression is enhanced through transcriptional and posttranscriptional regulation during the progression of chronic myeloid leukemiaAnn Hematol883333402009 DattaSHoenerhoffMJBommiPSaingerRGuoWJDimriMBandHBandVGreenJEDimriGPBmi-1 cooperates with H-Ras to transform human mammary epithelial cells via dysregulation of multiple growth-regulatory pathwaysCancer Res6710286102952007 HoenerhoffMJChuIBarkanDLiuZYDattaSDimriGPGreenJEBMI1 cooperates with H-RAS to induce an aggressive breast cancer phenotype with brain metastasesOncogene28302230322009 MerkerovaMBruchovaHKracmarovaAKlamovaHBrdickaRBmi-1 over-expression plays a secondary role in chronic myeloid leukemia transformationLeuk Lymphoma487938012007 DaniszKBlasiakJRole of anti-apoptotic pathways activated by BCR/ABL in the resistance of chronic myeloid leukemia cells to tyrosine kinase inhibitorsActa Biochim Pol605035142013 YangJChaiLLiuFFinkLMLinPSilbersteinLEAminHMWardDCMaYBmi-1 is a target gene for SALL4 in hematopoietic and leukemic cellsProc Natl Acad Sci USA10410494104992007 GodlewskiJNowickiMOBroniszAWilliamsSOtsukiANuovoGRaychaudhuryANewtonHBChioccaEALawlerSTargeting of the Bmi-1 oncogene/stem cell renewal factor by microRNA-128 inhibits glioma proliferation and self-renewalCancer Res68912591302008 WuXMLiuXBuYQSenguptaJCuiHJYiFPLiuTYuanCFShiYYSongFZRNAi-mediated silencing of the Bmi-1 gene causes growth inhibition and enhances doxorubicin-induced apoptosis in MCF-7 cellsGenet Mol Biol326977032009 VenkataramanSAlimovaIFanRHarrisPForemanNVibhakarRMicroRNA 128a increases intracellular ROS level by targeting Bmi-1 and inhibits medulloblastoma cancer cell growth by promoting senescencePLoS One5e107482010 NowakKKerlKFehrDKrampsCGessnerCKillmerKSamansBBerwangerBChristiansenHLutzWBMI1 is a target gene of E2F-1 and is strongly expressed in primary neuroblastomasNucleic Acids Res34174517542006

(A) Concentration- and time-dependent effects of bufalin-induced apoptosis of K562 cells. The cells were treated with 2.0, 1.0, 0.5, 0.1, 0.050 or 0.025 μmol/l bufalin for 48 h or 72 h. The inhibition rate was examined by MTT assay. Data were derived from three independent experiments. (B) Control K562 cells cultured for 48 h without bufalin. (C) K562 cells treated with 0.5 μmol/l bufalin for 48 h.

K562 cell early apoptosis induced by bufalin was analyzed by flow cytometry. K562 cells were cultured with (A) 0, (B) 0.025, (C) 0.05, (D) 0.5 or (E) 1.0 μmol/l bufalin for 48 h. Bufalin induced early apoptosis of the cells in a dose-dependent manner. The inhibition rate of 0.5 μmol/l bufalin was 39%.

(A) Expression levels of BMI-1, p16INK4a and p14ARF in K562 cells cultured with 0.5 μmol/l bufalin by qPCR analysis. p16INK4a expression levels began to increase as levels of BMI-1 expression decreased, and the rise in p14ARF expression levels lagged behind those of p16INK4a. (B) Bufalin downregulated BCR/ABL expression levels in a time-dependent manner, and the expression levels decreased 6.73-fold at 48 h following bufalin treatment.

K562 cells treated with 0.025, 0.05, 0.5 or 2.0 μmol/l bufalin, culture solution or no additional solution for 24 h. BMI-1 expression levels in the untreated control K562 cells were higher than those in the normal control bone marrow mononuclear cells, and were downregulated in drug-treated cells.

Most upregulated functional annotation clusters of bufalin-induced apoptosis of K562 cells.

Category Term Count % P-value
GOTERM_BP_FAT GO:0006350; transcription 258 19.06874 1.94 E-21
GOTERM_BP_FAT GO:0045449; regulation of transcription 297 21.95122 3.82 E-20
GOTERM_BP_FAT GO:0006355; regulation of transcription, DNA-dependent 204 15.07761 2.26 E-13
GOTERM_BP_FAT GO:0051252; regulation of RNA metabolic process 206 15.22542 5.77 E-13

Annotation Cluster 1. Enrichment Score, 16.253795458938388.

Most downregulated functional annotation clusters of bufalin-induced K562 cell apoptosis.

A, Annotation cluster 1a
Category Term Count % P-value
GOTERM_BP_FAT GO:0034660; ncRNA metabolic process 50 2.9036 6.25 E-11
GOTERM_BP_FAT GO:0034470; ncRNA processing 44 2.555168 7.16 E-11
GOTERM_BP_FAT GO:0006399; tRNA metabolic process 26 1.509872 3.38 E-06
GOTERM_BP_FAT GO:0008033; tRNA processing 20 1.16144 3.94 E-06
GOTERM_BP_FAT GO:0042254; ribosome biogenesis 26 1.509872 6.34 E-06
GOTERM_BP_FAT GO:0022613; ribonucleoprotein complex biogenesis 33 1.916376 8.37 E-06
GOTERM_BP_FAT GO:0006364; rRNA processing 21 1.219512 2.12 E-05
GOTERM_BP_FAT GO:0016072; rRNA metabolic process 21 1.219512 4.07 E-05
B, Annotation cluster 2b
Category Term Count % P-value
GOTERM_BP_FAT GO:0006259; DNA metabolic process 60 3.484321 0.001126
GOTERM_BP_FAT GO:0006281; DNA repair 38 2.206736 0.001353
GOTERM_BP_FAT GO:0006974; response to DNA damage stimulus 45 2.61324 0.003657
GOTERM_BP_FAT GO:0033554; cellular response to stress 53 3.077816 0.121174

Enrichment Score, 6.320442125275494;

Enrichment Score: 2.29263824337844.

Results of qPCR confirmation of differentially-expressed genes compared with microarray data.

Fold upregulation
Gene qPCR Microarray
BMI-1 1.63 1.11
E2F-1 4.89 1.45
MDM2 1.04 1.00

qPCR, quantitative polymerase chain reaction.