Abstract
As shown previously doxorubicin (1 μM) plus sulindac (50 μM) reduced the expression of ABCB1 (ATP-binding cassette, sub-family B (MDR/TAP), member 1) mRNA in HeLa cells and this effect was accompanied by increased apoptosis. The aim of this study was to define if the decrease of ABCB1 expression or blocking of P-glycoprotein (P-gp) can affect the expression of the apoptotic genes determined with use of quantitative real time polymerase chain reaction (qRT-PCR). Western blot was used for visualization of chosen pro- and antiapoptotic proteins. Doxorubicin was the main compound which affected the apoptotic genes. The effectiveness of the drugs in reducing of P-gp function has been shown as not being related to the regulation of apoptotic gene transcription. In this experimental scheme, regulation of apoptotic gene transcription depended on the kind of P-gp modulator.
Non steroidal anti-inflammatory drugs (NSAIDs) are cyclooxygenase inhibitors; they are analgesic and anti-inflammatory (1). Epidemiological and experimental studies have shown that cyclooxygenase-2 (COX2) inhibitors such as NSAIDs are also effective chemopreventive agents helping to reduce the risks of many types of tumor, including colon, lung, prostate and gastric cancer (2).
The mechanisms underlying the antitumor activity of COX-2 inhibitors are thought to involve inhibition of COX-2 enzyme activity and induction of apoptosis, genetically controlled mechanisms of cell death regulating tissue homeostasis (2). There are many studies which show synergistic activity of the NSAIDs, such as sulindac, meloxicam, rofecoxib, indomethacin and anthracyclines in various types of cancer cells (3-6). Therefore, NSAIDs should be investigated as a treatment supplementary to chemotherapy.
We previously demonstrated that doxorubicin (1 μM) and sulindac (50 μM) given simultaneously reduced the mRNA expression of ABCB1 in HeLa cells and this effect was accompanied by increased apoptosis as compared to the cells exposed only to doxorubicin (7).
Sulindac may affect apoptosis indirectly through the increase of intracellular level of doxorubicin due to quenching of ABCB1 but there is also another possibility that such a drug combination as doxorubicin plus sulindac can change the expression of apoptosis proteins at the mRNA level, as was noted by the other authors in the case of treatment of the cells with sulindac alone (8-10). Overexpression of P-gp encoded by ABCB1 mostly reduces apoptosis. Among possible mechanisms of this reduction are: efflux of apoptogens from the cells, regulation of the cytosolic levels of apoptotic mediators, such as caspase-8 or FADD. It remains possible that P-gp may inhibit caspase activation, in particular caspase-3, by regulating chloride channel activity to stabilize intracellular K+ and Na+ concentrations, thereby inhibiting the early steps in apoptosis and subsequent caspase activation. It has been also hypothesized that P-gp may act as a primary anti-apoptotic molecule by reducing the pool of plasma membrane sphingomyelin which is hydrolysed to apoptogenic ceramide (11, 12). Liu et al. (13) reported P-gp/ABCB1 overexpression as associated with increased survivin transcription. P-gp may also regulate the expression of miRNA16 and BCL2 but the regulation is believed to be associated with the efflux of P-gp substrates which may likely be important for P-gp function in gene expression (14). Tsang et al. (14) mentioned that the mechanism by which P-gp regulates gene expression is not clear. There is mounting evidence that P-gp cannot protect against caspase-independent death stimuli (14).
The aim of this study was to define if the decrease in ABCB1 expression by SUL or blocking of P-gp by verapamil can affect the expression of the apoptosis genes.
Materials and Methods
Drugs. Doxorubicin was purchased from medac (Hamburg, Germany); sulindac and verapamil from Sigma-Aldrich Chemie GmbH (Steinheim, Germany).
Cells. The human cervix adenocarcinoma cell line – HeLa was obtained from the American Type Culture Collection (Manassas, VA, USA). Cells were grown in Minimum Essential Medium (MEM) supplemented with 10% fetal serum and penicillin (5mg/ml), streptomycin (5 mg/ml), amphotericin B (12.5 mg/ml) (Lonza, Walkersville, MD, USA). All cell cultures were mycoplasma-free.
Treatment protocol. For the studies the 24-h cell cultures were co-incubated for 24 h with doxorubicin at 1 μM with/without sulindac at 50 μM with/without verapamil at 40 μM (7). The dose of verapamil was chosen as the highest non-toxic dose for HeLa cells which was confirmed with enzymatic assay based on the conversion of hydrogenated tetrazolium salt into coloured compound (Gruber et al., unpublished data).
Flow cytometry. The cells were exposed to the drugs as described above, then scraped very gently and centrifuged at 1000 × g at 4°C for 3 min. The resulting pellets were resuspended in Hank's buffer. The cell suspensions were removed to cytometry tubes and analyzed for cellular doxorubicin fluorescence by flow cytometry. Cytometric data were measured using a BD FACSCalibur flow cytometer (BD Biosciences, San Jose CA, USA), analyzed by CellQuest software (BD Biosciences and WinMIDI 2.9 Joseph Trotter).
RNA isolation and qRT-PCR. RNA isolation was performed as two independent repeats for each sample and was carried out as described previously (7). qRT-PCR was performed with Prime Script RT Reagent Kit (TAKARA BIO INC, Japan) using a total RNA of 500 ng application per sample including incubation at 37°C for 30 min and stopping with heat treatment (85°C; 5 s).
Gene expression was assessed using a two-step qRT–PCR, including reverse transcription and quantitative PCR analyses and was performed with Mx3005P qPCR System (Stratagene, La Jolla, CA, USA). qPCR was carried out with SYBR Premix Ex Taq (TAKARA) with the application of 400 ng cDNA template into a sample-specific working solution prepared for 96-well StellArray plate according to Lonza Inc. (US) recommendations (15). Cycling parameters included preincubation at 50°C for 2 min, initial polymerase activation at 95°C for 30 s and amplification of 40 cycles (15 s at 95°C and 1 min at 60°C). For evaluation of specificity of the PCR products, a melting curve analysis of the amplification products was carried out by additional cycle at the end of amplification process (1 min at 95°C, 30 s at 55°C and 30 s at 95°C). Before StellArrays, the amount of cDNA was normalized with that for β-actin (ACTB) gene. The primer sequences for ACTB gene were designed by Beacon Designer v.7.7 delivered by Beacon Designer 7 Software (PREMIER Biosoft International, Palo Alto, CA, USA) and included sequences: forward ACTB: 5’-TCGTGCGTGACATTAAGGAG-3’; and reverse ACTB: 5’-GAAGGAAGGCTGGAAGAGTG-3’. A variation between the CT values for ACTB, i.e. max. 1.0 was acceptable for the cDNA samples.
Gene expression. Evaluation of the results was carried out with Global Pattern Recognition™ Software (Bar Harbor Biotechnology, Inc. Trenton, ME, USA). For each gene tested a Global Pattern Recognition (GPR) fold change was calculated. The results with p-value ≤0.05 were accepted as being significant.
The panel of the genes analyzed in this study is presented in Table I.
Western blot. The cytoplasmic extracts preparation and western blot were proceeded as described by Gruber et al. (16). The assessed proteins were detected with use of primary monoclonal rabbit Ab for: BAX, 21 kDa (1:500, Epitomics, Burlingame, CA, USA); BCL-2, 26 kDa (1:250, Epitomics); Survivin, 16 kDa (1:250, Epitomics); primary monoclonal mouse Ab for β-actin, 42 kDa (1:1500, Abcam, Cambridge, UK) and with secondary goat anti-rabbit Ab (1:500, Epitomics) and goat anti-mouse Ab (1:500, Sigma, Saint Louis, MO, USA).
Results
Flow cytometry. To study the effectiveness of sulindac and verapamil as P-gp down-regulators, doxorubicin retention within cells was analyzed. When doxorubicin, as a fluorescent substrate, diffuses into the cells, P-gp actively pumps out the fluorochrome. Use of inhibitor along with the fluorescent marker increases the intensity of fluorescence which can be quantitatively analyzed. As shown in Figure 1, verapamil was the most effective for reducing P-gp function. The highest doxorubicin retention, reported as the highest value of fluorescence, was observed in the cells co-incubated with verapamil at 40 μM. Similar doxorubicin retention was shown after combination with doxorubicin (1 μM) plus verapamil (40 μM) plus sulindac (50 μM). The lowest doxorubicin retention was observed in the cells treated with sulindac 50 μM alone.
qRT-PCR. As shown in Table II combined treatment of HeLa cells with doxorubicin at 1 μM alone or in combination with sulindac at 50 μM and/or verapamil at 40 μM affected some genes related directly to apoptosis. All mRNA levels refer to the mRNA level determined in the intact cells.
24-hours exposure of HeLa cells to doxorubicin at 1 μM caused significant up-regulation of five out of 96 tested genes related to apoptosis, namely, BCL6 (ca. 30-fold), CD40 (ca. 60-fold); FASLG (above 230-fold), TNF (ca. 60-fold), TNFAIP3 (called also A20) (160-fold) (Table II). Sulindac at 50 μM alone slightly induced overexpression of TNF gene only (more than 25-fold), similarly to verapamil at 40 μM (ca. 30-fold). Verapamil also induced A20 expression by over 50-fold. Sulindac plus verapamil led to overexpression of three genes: FASLG (20-fold), TNF (ca. 70-fold) and A20 (over 20-fold).
Combined treatment of HeLa cells with doxorubicin (1 μM) plus sulindac (50 μM) or with doxorubicin (1 μM) plus verapamil (40 μM) significantly induced the same genes as doxorubicin, although doxorubicin plus sulindac and doxorubicin plus verapamil additionally induced GADD45G (over 130-fold and 60-fold, respectively) and LTA (ca. 30-fold and 25-fold respectively). It is worth noting that doxorubicin plus verapamil were weaker inducers of mRNA expression than doxorubicin plus sulindac. Under doxorubicin plus verapamil only BCL6 and LTA expressed similar mRNA levels as those affected by doxorubicin plus sulindac. Both drug combinations strongly induced TNF and A20 expression (over 200- and ca. 1800-fold for doxorubicin plus sulindac or ca. 150 and over 1300-fold for doxorubicin plus verapamil, respectively).
Doxorubicin (DOX) retention in HeLa cells measured with flow cytometry. 24-H cell cultures were treated for 24 h with: culture medium (control); sulindac (SUL) 50 μM; verapamil (VER) 40 μM; SUL 50 μM + VER 40 μM; DOX 1 μM; DOX 1 μM + SUL 50 μM; DOX 1 μM + VER 40 μM; DOX 1 μM + SUL 50 μM + VER 40 μM.
Simultaneous treatment of the cells with all three drugs significantly affected the same genes, except LTA, which was induced by doxorubicin plus sulindac and doxorubicin plus verapamil. The level of mRNA expression after exposure to the triple drug combination was the highest for A20 gene (over 2000-fold) as compared to the control. In the case of the other genes, combination of three drugs was shown to have similar effect to doxorubicin plus verapamil. None of the drugs caused any significant decreases in any gene mRNA expression. All results presented in Table II were statistically significant (p<0.05).
Western blot. As was shown in Figure 2, exposure of cells to tested drugs, did not change the expression levels of BAX and BCL2. Most significant changes were noted for survivin levels: the highest expression of this protein was observed in the cells exposed to doxorubicin (1 μM) plus sulindac (50 μM) plus verapamil (40 μM), with significant increase also occurring after treatment with sulindac at 50 μM alone, after doxorubicin combined with sulindac at 50 μM and verapamil at 40 μM separately.
Discussion
On the basis of gene expression profiles tested in this study, it can be said that doxorubicin at 1 μM generates apoptosis through death receptors. This drug causes overexpression of genes encoding FASLG, CD40 and TNF. All three proteins belong to the TNF receptor superfamily members, which points to the extrinsic pathway of apoptosis being induced by doxorubicin. As was noted, doxorubicin at 1 μM after 24-h treatment strongly induced also the expression of A20, which is connected with proapoptotic function through NFκB inhibition or antiapoptotic through, among others, inhibition of p53 or BAX expression (17). Overexpression of A20 upon treatment with doxorubicin at 1 μM, which was rather strong (ca. 160-fold) seems to be related to induction of apoptosis noted in HeLa cells (7) as well as that of FASLG gene, which reached a much higher level after doxorubicin treatment than A20 (above 230-fold). On the other hand, stimulation of A20 mRNA by non-apoptogenic agents verapamil or verapamil plus sulindac is rather related to cytoprotective effect of A20 induced upon TNF and FASLG stimulation. As was presented by Verstrepen et al. (17), the role of A20 in tumorigenesis might be cell type- and apoptogen-dependent. As was shown by the authors, this gene protected endothelial cells against apoptosis induced by TNF and FAS triggering, whereas in A549, HepG2 and HeLa cells, it was not able to block TNF-induced apoptosis.
Panel of 96 genes analyzed with StellArray.
Western blot. Cytoplasmic extract. 24-H cultured HeLa cells were exposed for 24 h as follows: Lane 1: control medium; lane 2: sulindac (SUL) 50 μM; lane 3: verapamil (VER) 40 μM; lane 4: doxorubicin (DOX) 1 μM; lane 5: DOX 1 μM + VER 40 μM; lane 6: DOX 1 μM + SUL 50 μM; lane 7: SUL 50 μM + VER 40 μM; lane 8: DOX 1 μM + SUL 50 μM + VER 40 μM; lane 9: Mw (kDa).
Apoptosis gene expression up–regulated by 24-h treatment of HeLa cells with sulindac (SUL) at 50 μM, verapamil (VER) at 40 μM alone or doxorubicin (DOX) 1 μM alone and in their different combinations.
The results obtained with doxorubicin in qRT-PCR and in western blot are in agreement with Suzuki et al. (18), who noted that doxorubicin did not significantly change the intracellular concentrations of BCL2, BAX and BAD in HL-60 cells, although the treatment time in that case was much shorter (6 h) than in the present study (24 h). Lüpertz et al. (19) reported the increase of BAX in colon cancer cells under treatment with doxorubicin at 1 μM but given as a bolus (3 h). The authors suggested that doxorubicin is able to induce cell death by apoptosis only at particular doses and under specified treatment conditions, and imply a completely different cellular response following bolus or continuous exposure to DOX.
As was shown earlier (7), combined treatment of HeLa cells with doxorubicin plus sulindac enhanced apoptotic processes and this event is reflected in the gene expression profiles. Doxorubicin plus sulindac caused strong induction of proapoptotic TNF and GADD45G genes. A20 gene expression also reached a much higher level (ca. 1800-fold) as compared to the mRNA level in untreated cells. mRNA fold change of A20 was positively correlated with apoptosis observed in HeLa cells treated with doxorubicin and doxorubicin plus sulindac (7). It suggests a proapoptotic function of A20, which may be the specific response of HeLa cells to doxorubicin or doxorubicin plus sulindac exposure. Increases in A20 mRNA expression alongside P-gp modulator addition indicate the dependence of the level of mRNA expression of the genes affected on intracellular doxorubicin concentration.
The slight influence of sulindac on the transcription of the genes related to apoptosis is consistent with the effects reported by Gruber et al. (7), i.e. sulindac alone at the concentration of 50 μM was not apoptogenic in HeLa cells and overexpression of TNF which has been evoked as the effect of sulindac has no significance. Most literature data indicate the apoptogenic functions of sulindac. Liu et al. (20) noted such a process in HepG2 cells, Park et al. (21) in HT-29 cells and Scheper et al. (22) in oral squamous carcinoma (SCC) cells. In turn, Wu et al. (23) noted apoptosis induced by sulindac depended on the differentiation of the cells. Han et al. (24) showed that induction of apoptosis by sulindac is cell dependent. The authors treated breast cancer cells with sulindac at 1600 μM for 48 h and no apoptotic processes were reported. Worth noting is the fact that the gene expression profile depends on the dose and time of exposure of the cells to the drug. In our study, sulindac was used at 50 μM for 24 h and under such conditions, this drug was not apoptogenic. On the other hand, sulindac used by the other authors at the concentration range of: 100-4000 μM for 48 or 72 h did affect antiapoptotic and proapoptotic genes and these effects were reflected in apoptosis induction in neoplastic cells by sulindac alone (8, 9, 22, 23). The effect of sulindac frequently pointed out by the other authors, i.e. down-regulation of survivin gene (9, 10, 22) was not noted in our study. On the contrary, western blot showed strong activation of this protein under non-apoptogenic dose of sulindac which suggests some posttranslational changes at the protein level but this is unclear. As was shown in a previous study (7), sulindac significantly attenuated the expression of ABCB1 and the same P-gp membrane pump. The are some reasons for which we suggest that intensified apoptosis noted under treatment with doxorubicin plus sulindac and the influence of these drug combination on apoptosis genes expression may be related rather to doxorubicin retention inside the cells than with the direct correlation between P-gp modulation and apoptotic genes expression. These reasons are: the lack of significant influence of sulindac on the mRNA levels of apoptosis genes; the fact that verapamil as the most effective down-regulator of P-gp slightly affected the apoptosis gene profile; sulindac or verapamil alone, or the combination of these drugs did not affect the same genes; doxorubicin plus verapamil were weaker inducers of mRNA expression than doxorubicin plus sulindac. The results obtained in this study are inconsistent with these of Tsang et al. (14), who noted that ABCB1 expression correlated with BCL2 in hepatocellular carcinoma cells. Notarbartolo et al. (25) and Liu et al.(26) observed the decrease of survivin (BIRC5) and IAP mRNA under verapamil treatment. None of these effects were noted in this study.
In this study, the main compound which affected the apoptosis genes was doxorubicin. Sulindac and verapamil only modified the level of the mRNA expression in the same genes. The effectiveness of the drugs in reducing P-gp function does not appear to be related to the regulation of transcription of apoptosis genes.
It can be concluded that in this experimental scheme, the efficiency of P-gp modulators such as sulindac and verapamil in regulation of transcription of apoptosis genes depends on the intracellular apoptogen retention which is directly related to the the kind of the P-gp modulator.
- Received February 27, 2012.
- Revision received April 13, 2012.
- Accepted April 17, 2012.
- Copyright © 2012 The Author(s). Published by the International Institute of Anticancer Research.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.