Identification Of Common Gene Expression Patterns Of Cytotoxic Antitumor Agents By Microarray Analysis

Gene expression microarray is a newly-developed biotechnology in recent years, permits the simultaneous and rapid measurement of the expression levels of thousands of genes, and allows the analysis of molecular events of drug response and potentially identifying novel drug targets. It is also suitable for high-throughput drug screening with multi-targets and evaluation of drug activity and toxicology.Cytotoxic anticancan drugs inhibit cell proliferation, disturb cellcycle and induce cell apoptosis after binding to diverse molecular targets. The growing numbers of documents have shown the key regulatory molecules and cellular pathways control cell apoptosis and the cell cycle. However, the connectivity of known mechanisms with other regulators or pathways remains unclear. Although the anticancer drugs have diverse initial molecular targets, cellular phenotypic outcomes are similar for different drug treatments. Therefore, it appears reasonable to assume that some of genetic alternations among the different drug responses might be likely shared as common key effects of the drug-target interactions. Analysis of these intersection genes may provide new insights into the molecular mechanisms of the antitumor agents and discover new potential target molecules for the development of novel antineoplastic drugs.This study was designed to test above hopothesis with microarray analysis of the sensitive early gene expression profiles in QGY-7703 hepatoma cells and HL-60 leukemia cells stimulated with the chosen representative drug paclitaxel (Taxol), dactinomycin (ACD), camptothecin (CPT), and homoharringtonine (HHRT), based on their identical cell growth inhibition actions. Drug response genes were then clustered and cellular pathways were analyzed by Genmapp anlysis software and Microsoft Ascess software. Realtime PCR was also conducted for some interest genes to comfirm our microarray results. Results are shown as follows:1. To analyze the sensitive early gene expression profiles, cell phenotypes of growth inhibition, apoptosis and cell cycle in response to drug treatment were determined by MTT assay and flow cytometry. 50% of inhibitory concentration (IC50) of Taxol, HHRT, CPT and ACD was 0.025,320,79.85,2.67,μmol/mL and 2.26,52.77,19.09,0.48,μmol/mL respectively in HL-60 leukemia cells and QGY-7703 hepatoma cells after 24 cluture. Morphological changes were not observed by microscope after 1 h incubation of drugs with both cell lines, wheras the decreased cell numbers and cell pieces were seen in both cells treated with the chosen drugs for 24 h. Flow cytometry results showed that treatment of paclitaxel and homoharringtonine for 24 h markedly accelerated leukaemia and heptoma cells through G1/S interface resulting G2/M arrest and apoptotic cell death. These results are consitent with the reported actions of these drugs.2. Total RNA was isolated from drug treated cells and from control (equal volume of drug solvent) cells after 1 h incubation. Labeling of cDNA, microarray hybrization and fluorescence detection was performed as standard protocol of Unite Gene Corpoation. Genes were identified as differentially expressed if the absolute value of the natural logarithm of the average ratios of three independent experiments was > 0.69. The number of regulated genes differed significantly among drugs, with paclitaxel and homoharringtonine causing the strongest responses in HL-60 and QGY7703 cells respectively.3. Genmapp software was used to gene cluster and pathway analysis of the drug response genes. A number of cellular pathways, functional molecules and cellular components were found as classied by the used software. For example, paclitaxel reglulated mRNA processing; homoharringtonine inhibited the expression of protesome component PSMC3,PSMD2 and PSMD11;camptothecin affected G protein signaling pathways and dactinomycin modulated MAPK pathways and TNF-NFkB pathways. These results indicate initial tagets are connected with many known pathways responsible for cell proliferation and apoptosis, which may clearly explain actions of these antitumor agents. In addition, cytotoxic agents also regulated several groups of genes specific to internal organs, endocrine and CNS as well as embryo stem cells, suggesting these genes may be related to the cytotoxicities of antitumor drugs.4. Microsoft office Access software was used to sort common drug response genes and common cellular pathways. Our results demonstrated that antitumor agents shared many cellular pathways and genes with the same regulated direction in the transcriptional levels. The largest overlap was found between two agents. Common response genes and pathways were also identified among three drugs, even in all four four drugs. For exaple, 17 and 40 common response genes and 7 and 9 pathways were obtained in HL-60 cell and in QGY7703 cells respectively. There were 6 common response genes including granulin (GRN),myeloid/lymphoid or mixed-lineage leukemia 3 (MLL3),prosaposin (variant Gaucher disease and variant metachromatic leukodystrophy, PSAP),endoplasmic reticulum protein 29 (ERP29),low density lipoprotein receptor-related protein associated protein 1 (LRPAP1),ST3 beta-galactoside alpha-2,3-sialyltransferase 4 (ST3GAL4) and 2 pathways (regulation of Actin Cytoskeleton and focal adhesion) shared by all drugs in both cell lines. These results provided evidence for common down stream effects of topoisomerase inhibition and microtubular dynamics or ribosome function interference as well as DNA template corruption, which may result in final similar phenotypic outcomes.5. The common response genes may be classied into four categrories including growth and proliferation related genes, apoptosis related genes, carbohydrate metabolism related genes and unkown function genes. The biochemical functions of intersection response genes are diverse and include cell cycle regulators, proapoptotic and antiapoptotic factors, growth factors, signal mediators, metabolic regulators, proteases and glycosyltransferases, and proteins of miscellaneous or unknown function. (1) Chemotherapeutical agents-treated cells decreased expression of a cluster of genes involved in cell proliferation. For example, a broad range of genes is associated with cell cycle and mitosis (QSCN6,PPP2R1A),gene transcription [general transcription factor IIIC (GTF3C5) , activating transcription factor 5 (ATF5)],RNA processing[heterogeneous nuclear ribonucleoprotein A/B (HNRPAB),PRP6 pre-mRNA processing factor 6 homolog (S. cerevisiae) (PRPF6),nucleolin (NCL)] and protein synthesis[eukaryotic translation elongation factor 1 delta EEF1D, Eukaryotic translation initiation factor 4A, isoform 1,EIF4A1]。In addition, tumor associated genes granulin,myeloid/lymphoid or mixed-lineage leukemia 3 were also inhibited upon multiple treatment.The alterations in these specific genes functionally related to cell proliferation might determine tumor cell growth arrest, even ultimate cell death, correlating well with the cytotoxic agents-mediated cytotoxicity. (2) Treatment of antitumor agents induced proapototic genes phorbol-12-myristate- 13-acetate-induced protein 1(PMAIP1) and decreased antipoptotic gene prosaposin (PSAP), suggesting the involvement of cell apoptotic mechanisms in these drug actions. (3) Regulation of endoplasmic reticulum (ER) and mitochondrial components was shown by drug treatment, for example, ERP29, a disulfide isomerase (PDI)-like proteina, is involved in oxidative stress and regulates protein unfolding and facilitates the transport of secretory proteins. ARF5 is a member of ADP-ribosylation factors necessary for the formation of clathrin- and coatomer protein (COP) I-coated vesiclesare, and isivovled in protein transport between ER and Golgi compartments. Inhibition of tese ER associated genes by antitumor agents may result in ER stress leading to cell apoptosis. Moreover, the expression of mitochondrial components was also observed in response to antitumor drugs. Isocitrate dehydrogenase 2 participates in tricarboxylic acid cycle and solute carrier family 25 (SLC25A6) is adenine nucleotide translocator. Thus the decresed energy production and transport may be responsible for cytotoxic agents-mediated apoptosis. (4) Cells exposed to cytotoxic agents resulted in the down regulation of a cluster of Ubiquitin/proteasome degradation pathways (UPDP) related genes, such as PSMC3, VCP and SAE1. PSMC3 is a core component of the 19S regulatory particle of the 26S proteasome, VCP helps dislocateing misfolded proteins from the ER membrane into the cytosol for proteasomal degradation, and SAE1/SAE2 is the activating enzyme E1 of small ubiquitin-related modifier (SUMO) responsible for post-translational modifications of proteins. Since ubiquitin/proteasome degradation pathways have function in regulating cell proliferation and apoptosis, the decraesed expression of UPDP related genes may be contributed to cytotoxic effects of antitumor drugs. (5) Genes functioned in the intracellular glycoprotein glycosylation pathway such as sialyltransferase 4C and SLC35D1expression were reduced in drug responses, which may imply the modulating effects of these antitumor drugs on the tumor invasion and metastasis due to the involvement of specific carbohydrate structures in cellular recognition processes and cancer metastasis.6. Some common response genes have been documented to be the targets for drug design and screening. For example, proteasome becomes a new target for novel drug therapy. Based on a modelled structure of PP2A, a new class of protein phosphatase inhibitors, cantharimides, are designed and exhibit broad-spectrum anti-cancer activity. Suppression of sialyltransferases by chemicals has been shown to reduce the potentiality of growth and metastasis of various tumors. These suggest that another unexplored drug targets might be discovered from these intersection genes. In addition, these common response genes may also be used for screening and evaluation of antitumor agents as made of multiple-target microarray chip.Taken togethe, determination of the common response genes with microarrays and multiple tool agents is of great value for our deciphering complex regulatory pathways and discovering new potential target molecules.