Genes Potentially Associated with Resistance of Lung Cancer Cells to ...

2 downloads 0 Views 167KB Size Report
mechanisms of lung cancer to paclitaxel. ... tivity of lung cancer cells to paclitaxel and the expres sion of a ..... Alley, M.C., Scudiero, D.A., Monks, A., et al., Cancer.
ISSN 16076729, Doklady Biochemistry and Biophysics, 2011, Vol. 437, pp. 105–108. © Pleiades Publishing, Ltd., 2011. Original Russian Text © K.N. Kashkin, E.A. Musatkina, A.V. Komelkov, D.A. Sakharov, E.V. Trushkin, E.A. Tonevitsky, T.V. Vinogradova, E.P. Kopantzev, M.V. Zinovyeva, O.V. Kovaleva, K.A. Arkhipova, I.B. Zborovskaya, A.G. Tonevitsky, E.D. Sverdlov, 2011, published in Doklady Akademii Nauk, 2011, Vol. 437, No. 6, pp. 837–841.

BIOCHEMISTRY, BIOPHYSICS AND MOLECULAR BIOLOGY

Genes Potentially Associated with Resistance of Lung Cancer Cells to Paclitaxel K. N. Kashkina, E. A. Musatkinab, A. V. Komelkovb, D. A. Sakharovc, E. V. Trushkinc, E. A. Tonevitskyc, T. V. Vinogradovaa, E. P. Kopantzeva, M. V. Zinovyevaa, O. V. Kovalevab, K. A. Arkhipovab, I. B. Zborovskayab, Corresponding Member of the RAS A. G. Tonevitskyc, and Academician E. D. Sverdlova Received December 24, 2010

DOI: 10.1134/S1607672911020153

Paclitaxel is a natural substance from the group of taxanes and one of the most widely used drugs in che motherapy of malignant tumors, including the non small cell lung cancer (NSCLC). Treatment of patients with advanced NSCLC with paclitaxel alone or in combination with other drugs led to complete or partial tumor regression in at most 41% of patients [1]. Individual selection of drugs that are effective for a particular patient is one of the hottest approaches in the strategy to improve the results of chemotherapeu tic treatment of cancer patients. This work is dedicated to finding new markers and studying the resistance mechanisms of lung cancer to paclitaxel. The main mechanism of action of paclitaxel is the blockade of cell division due to its specific binding to β3 tubulin and stabilization of microtubules. An increased expression of some βtubulin isotypes corre lates with resistance to paclitaxel and the degree of malignancy of tumors of the lung, prostate, ovary and other tissues. The same result can be caused by muta tions of βtubulins, although the clinical significance of mutations is controversial (see reviews [2, 3]). It is known that the resistance of tumor cells to paclitaxel may be associated with an increased expres sion or polymorphism of some ABCtransporters responsible for multidrug resistance [4] as well as with activity of growth factor receptors (EGFR and HER 2), proteins LIMK1, LIMK2, TGFB1, STMN1, spin dle checkpoint proteins, apoptosis regulators, and others [2, 3].

a

Shemyakin–Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, ul. MiklukhoMaklaya 16/10, Moscow, 117997 Russia b Blokhin Cancer Research Center, Russian Academy of Medical Sciences, Kashirskoe sh. 24, Moscow, 115478 Russia c AllRussia Institute of Physical Culture and Athletics, Elizavetinskii per. 10, Moscow, 105005 Russia

The use of DNA microchips [5–10] made it possi ble to reveal the sets of genes whose expression corre lates with chemoresistance of lung cancer cells. How ever, the sets of genes that, with a certain probability, may be involved in resistance development, which were proposed by different authors, differ. In addition, the genes detected in cell cultures may be uninforma tive for predicting the response of patients to chemo therapy [11]. At the same time, studies of clinical specimens only may give ambiguous results due to polymorphisms, which lead to a high interindividual variability of systems involved in the development of chemoresistance of cancer patients, as well as because of chemotherapy regimens that usually include several drugs. To search for new informative markers and to reveal the molecular mechanisms of drug resistance in tumors, we studied the relationship between the sensi tivity of lung cancer cells to paclitaxel and the expres sion of a wide range of genes using the new microchip platform Affymetrix Human GeneChip ST1.0, which contained probes for over 28000 human mRNAs. As a biological model, we used lung cancer cells of six lines derived from ATCC (A549, NCIH292, NCI H460, and NCIH1299) and ECACC (NCIH322 and NCIH358). Although the sensitivity of certain lines to various drugs is known (http://dis cover.nci.nih.gov/cellminer/), we redefined the IC50 of paclitaxel for all cells using the MTT technique [12] in order to eliminate the effect of subculturing of cells on their sensitivity to the drug. For each cell line, we analyzed the results of at least three significant mea surements (Fig. 1). Hybridization of labeled samples prepared from total cell RNA with microchips was performed using the equipment and method of Affymetrix. The results of hybridization were processed by the RMA algo rithm using the xps library (Christian Stratowa; http://www.bioconductor.org) in R system. Fluores cence signals were filtered using I/INI algorithm [13]; signals with random variation were ignored.

105

106

KASHKIN et al. Cells with intermediate sensitivity

Paclitaxel, IC50, µM

0.04

Resistant cells

Sensitive cells

0.03 0.02 0.01 0

A549

NCIH1299NCIH358 NCIH322 NCIH292 NCIH460

IC50 concentrations of paclitaxel for lung cancer cells. The mean IC50 value (diamond sign), the standard error of the mean (SEM), and conditional separation of cells into groups according to sensitivity to paclitaxel are shown. The abscissa axis shows cell lines.

According to the results of IC50 measurements, cells were divided into three groups: resistant cells (lines A549, NCIH1299, and NCIH358), cells of intermediate sensitivity (line H322), and sensitive cells (lines NCIH292 and NCIH460). According to the results of hybridization with the microchip, we selected the probes for which the difference between the threshold fluorescence values (expression index) in the first and third groups differed by at least twice (Table 1). In addition, we determined the Pearson cor relation coefficient (R) for the gene expression rate in cells of all six lines with IC50 and selected genes with R ≥ 0.9 and R ≤ –0.9. Having combined the genes selected by the two methods, we obtained two sets of genes whose expression is associated with paclitaxel resistant (56 genes) and paclitaxelsensitive (52 genes) phenotypes. The obtained panels of genes were used to pre dict possible mechanisms that determine the sensi tivity or resistance of tumor cells to paclitaxel. For this purpose, we identified the biological pro cesses that involve the identified genes using NCI Pathway Interaction Databases (NCI PID, http://pid.nci.nih.gov/index.shtml), Gene Ontology database (http://www.geneontology.org), and the Table 1. Genes differing in the expression level between groups of resistant and sensitive cells* Gene predominantly ex pressed in resistant cells XAGE1D, FLJ35848, UCP2, VCAN, PEG10, GEM, TNNT1, ARL4C, CCDC144A, C8orf37

Gene predominantly ex pressed in sensitive cells SERPINB2, CDH13, NID2, IL1A, SORBS2, FAM134B, C4orf18, ASS1, SMOC1, ADAMTS1

* Ten genes that most significantly differed in the level of expression are shown.

Kyoto Encyclopedia of Genes and Genomes (KEGG, http://www.genome.jp/kegg/kegg2.html) (Table 2). The study of the processes showed that the resistant cells actively expressed the genes that are involved in signal transduction cascades mediated by SMAD2/3, TGFβ, and CMYC. These signaling pathways are responsible for the passage of cellcycle checkpoints by the cell in the G1 phase and are not associated with interleukins. Probably, these genes and signal trans duction pathways allow cells to stop in the G1 phase and survive at low concentrations of paclitaxel [14]. Note an increased expression of certain target genes of CMYC, a transcriptional regulator of a broad spec trum of genes that control pivotal cell functions, including DNA replication and repair and the cell cycle passage. In addition, paclitaxelresistant cells are characterized by an increased expression of some genes whose products are involved in signal transduc tion and genes encoding proteins with different trans port functions: ENY2 (a transcription activator that ensures the transport of mRNA to the cytoplasm), SMAD3 (protein involved in nuclear translocation of βcatenin), GABRB3 (a member of the GABAA family receptors, which are components of ion chan nel), and UCP2 (mitochondrial regulator of proton transport). The specific role of these processes in drug resistance of cells should be investigated in further studies. An increased expression of genes encoding inter leukins and other regulators of apoptosis was detected in the paclitaxelsensitive cells. This indicates a possi ble role of interleukindependent signaling cascades in cell response to paclitaxel and their relation to certain diseases and interleukinmediated reactions of the body (e.g., prion diseases, inflammation, etc.). This opens up new ways to boost cell response to the drug. In particular, it was shown that the prion protein PrPc interacts with Pglycoprotein during the development of multidrug resistance and that it is involved in the regulation of doxorubicininduced apoptosis by mod ulating the expression of Bcl2 and Bax genes [15]. Interestingly, the group of genes identified in the pacli taxelsensitive cells includes both the genes preventing cell proliferation (ADAMTS1 and CDH13) and the genes with the opposite function, which inhibit apop tosis (SERPINB2/PAI2 and SERPINB9/PI9). Apparently, the paclitaxelsensitive cells also differ from the paclitaxelresistant cells in some metabolic characteristics (oxidation–reduction, fatty acid metabolism, and the respiratory electron transport chain), which should also be studied in the future. One of the key regulators of the cell cycle and apo ptosis that may determine the degree of cell resistance to drugs, in particular to antimitotic drugs, is p53 [3]. We noted that the group of paclitaxelsensitive cell lines included the cell lines with the wildtype p53 gene (p53wt, lines NCIH292 and NCIH460) and that the group of paclitaxelresistant cell lines included two cell lines with the p53 gene deletion

DOKLADY BIOCHEMISTRY AND BIOPHYSICS

Vol. 437

2011

GENES POTENTIALLY ASSOCIATED WITH RESISTANCE

107

Table 2. Biological processes associated with different sensitivity of lung cancer cells to paclitaxelTable 2. Pathways and in teractions associated with different sensitivity of lung cancer cells to paclitaxel Gene KPNA2, SMAD3 PPP2CB, SMAD3 PEG10, SMAD3 ADRA1B, F2RL1, GABRB3 ENY2, GABRB3, SMAD3, UCP2 ADRA1B, PPP1R14A IL1A, IL1B IL1A, IL1B, SERPINB2, SERPINB9 ADAMTS1, CDH13, IL1A, IL1B HADH, SDHD, PRDX6 BACE2, IL1B, SDHD IL1A, IL1B

Process* Paclitaxel resistant phenotype Regulation of cytoplasmic and nuclear SMAD2/3 signaling TGFbeta receptor signaling Validated targets of CMYC transcriptional activation GO:0007165 Signal transduction; ko04080 Neuroactive ligandreceptor interaction GO:0006810 Transport (of molecules and ions) ko04270 Vascular smooth muscle contraction Paclitaxel sensitive phenotype IL1mediated signaling events GO:0006916 Antiapoptosis GO:0008285 Negative regulation of cell proliferation ko01100 Metabolic pathways ko01110 Biosynthesis of secondary metabolites ko05010 Alzheimer’s disease ko04210 Apoptosis; ko04010 MAPK signaling pathway; ko04060 Cytokinecytokine receptor interaction; ko05332 Graftversushost disease; ko04640 Hematopoietic cell lineage; ko04940 Type I diabetes mellitus; ko05140 Leishmaniasis; ko05020 Prion diseases

* Processes without indexes are specified by NCI PID (if two or more genes from the set are involved), with “GO” index by Gene Ontol ogy (if three or more genes from the set are involved, Hyp*