| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Lineberger Comprehensive Cancer Center [D. J. T., J. P. M., C. C., D. T. B., J. P-Y. T.], Department of Microbiology and Immunology [J. P. M., C. C., J. P-Y. T.], and Curriculum in Genetics and Molecular Biology [D. T. B., J. P-Y. T.], University of North Carolina at Chapel Hill, North Carolina 27599-7295
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B by several anticancer drugs (1)
. A more recent finding is the induction of MEK/Erk3
by chemotherapeutic agents. Identifying the antiapoptotic signals that are induced by specific chemotherapy, intervening with these signals by the inclusion of additional drugs, and identifying new molecular targets of combination drug treatment are at the forefront of innovative strategies for cancer treatment.
Paclitaxel (Taxol) is a front-line antineoplastic agent that is efficacious in the treatment of several malignancies, including ovarian, breast, lung, and prostate cancers, although its usage in other cancers has been less promising. Paclitaxel exerts its effect through stabilization of microtubules, cell cycle arrest in G2-M, and activation of proapoptotic signaling (2
, 3) . Defined pathways associated with apoptosis by paclitaxel include phosphorylation of Bcl-2 (4)
and activation of p53, cyclin dependent kinases, and the c-Jun NH2-terminal kinase/stress-activated protein kinase signaling pathway (5
, 6)
. At high concentrations, paclitaxel also stimulates the release of tumor necrosis factor 
and IL-1 in murine macrophages (7)
and activates the expression of IL-8 in ovarian and lung carcinoma cells via activator protein-1 and nuclear factor-B promoter sites (8
, 9)
. Paradoxically, paclitaxel can induce activation of the MEK/Erk pathway, which is frequently associated with cell survival (10
, 11)
. We and others proposed that this may cause suboptimal efficacy of paclitaxel in the treatment of many malignancies. Using a rational chemotherapeutic strategy, we combined paclitaxel with the MEK/Erk inhibitor U0126, demonstrating greatly enhanced induction of cell death in breast, ovarian, and lung carcinomas (12)
. Others have expanded these studies in additional transformed cell lines (13
, 14)
.
The MEK/Erk pathway is activated in response to multiple extracellular stimuli by a cascade of phosphorylation events downstream of the Ras proto-oncogene. Constitutive activation of MEK results in transformation (15) . A pharmacological MEK inhibitor is effective in the treatment of colon tumors in mice, and clinical trials are under way to define its efficacy in humans with promising preliminary data (16 , 17) . MEK inhibitors also have been combined effectively with retinoids and 1-ß-D-arabinofuranosylcytosine (18) , the Bcl-2 inhibitor HA-14-1 (19) , cisplatin (20) , and UCN-01 (21) .
Toward understanding the molecular basis for the combined effects of paclitaxel and U0126, we analyzed transcriptional profiles of >12,000 gene sequences, using Affymetrix chips. Genes were identified with defined roles in cell cycle, apoptosis, proliferation, and invasiveness. Expression was compared in ovarian and melanoma cell lines, and the melanoma metastasis marker, MGSA, was shown to be a target of both paclitaxel and U0126. TOP3B also was shown to be a potential therapeutic target and tumor biomarker. These results provide a molecular mechanism for the enhanced effects of paclitaxel and MEK inhibitors.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
DNA Microarray Analysis.
H157 cells were treated for 6 h with 250 nM paclitaxel, 10 µM U0126, or a combination of 250 nM paclitaxel and 10 µM U0126. Because DMSO was used as a solvent, a DMSO-only sample served as a control. Ten µg of RNA (RNeasy; Qiagen) were reverse-transcribed using Superscript II (Stratagene), labeled using the Enzo Bioarray High Yield RNA Transcript Labeling kit (Enzo Diagnostics), and analyzed on Human Cancer Array or HGU95V2 chips at the UNC Genomics Facility according to standard Affymetrix protocols. Replicates were processed using preparations harvested on different days to control for biological variations. RNA quality was confirmed by spectrophotometric examination, electrophoresis on formaldehyde gels, and by assessing 5'/3' ratios of control genes provided on the HGU95V2 chips. Expression was normalized to an average of 2500, using Affymetrix Microarray Suite 4.0 default scaling, and data sheets were imported into GeneSpring 4.1 (Silicon Genetics). Genes dynamically regulated by paclitaxel or U0126, alone or in combination, were filtered by the following criteria: 
2-fold modulation in both experimental replicates; 
0.4 SE between replicates; and 1000 average for the highest signal. A total of 159 of 12,627 gene entries met these criteria for at least one drug treatment. Two genes (NFIX and COMP) were eliminated based on expression patterns that deviated from other genes and could not be confirmed by real-time PCR. The expression patterns of the remaining 157 dynamically regulated genes were normalized to a median of 1.0, and negative values were adjusted to 0. Genes were clustered by applying a K-means algorithm (Genespring; Silicon Genetics). Functional categorization of genes was based on ontological designations in CGAP, gene descriptions in OMIM, and journal citations in PubMed.
Northern Analyses.
Northern analyses were performed as described previously (23)
. Probes (300–600 bp) were PCR-amplified, isolated on Qiagen gel purification columns, and labeled (Prime-it II; Stratagene). Primers for PCR amplification were as follows:
CIAP-2, 5'-AAGCGCCAACACGTTTGAAC-3'/5'-TGGAGTTTACAGGATTTGATGG-3';
CYP1ß1, 5'-GGTCACATAATTTAAAGCTTGG-3'/5'-CAAGATTGGTCTCCCATATG-3';
HB-EGF, 5'-ATCCCTTGGTGGTACTTGAG-3'/5'-ATGACTAATTCCCACTGAGAG-3';
MGSA, 5'-AAAGAGAGACACAGCTGCAGA-3'/5'-GCATGTTGCAGGCTCCTCAG-3';
EPH-A2, 5'-AGGTGACGCTGTAGACAATG-3'/5'-TACGAGAAGGTGGAGGATGC-3';
PLAUR, 5'-GAACCACATTGATGTCTCCTG-3'/5'-TAATAACAACAACACAACAGCG-3';
TGFßR2, 5'-ACAGGCAGCAGGTTAGGTCG-3'/5'-CACGTGTGCCAACAACATCAAC-3'.
GAPDH and IL-8 probes were made by labeling gel purified full-length gene fragments.
Real-Time PCR.
cDNA was synthesized from total RNA by use of random hexomers and M-MLV reverse transcriptase (Life Technologies, Inc.). Real-time PCR was performed using Platinum Quantitative Supermix-UDG (Invitrogen) and the ABI Prism 7900 sequence detection system (Perkin-Elmer). Values were calculated based on standard curves generated for each gene and normalized to 18 s rRNA for each sample. Probe/primer sets included the following:
MGSA, 5'-6FAM-CGCCCAAACCGAAGTCATAGCC-TAMRA-3'/5'-TCCAAAGTGTGAACGTGAAGTC-3'/5'-AAGCTTTCCGCCCATTCT-3';
TOP3B, 5'-6FAM-ACATCCAGGCCAAGCCAAGC-TAMRA-3'/5'-TCTCACGCTGTGGGAAGT-3'/5'-GGGAGCGTGAGGTCTC-A-3';
18s rRNA, 5'-6TET-CAAATTACCCACTCCCGACCCG-TAMRA-3'/5'-GCTGCTGGCACCAGACTT-3'/5'-CGGCTACCACATCCAAGG-3'.
Cell Death and Proliferation Assays.
Cell death ELISAs were performed as described previously (12)
, using the Cell Death Detection ELISAPLUS assay (Roche). [3H]Thymidine incorporation assays were performed using 2500 cells/well. Cells were cultured in the absence of FCS for 64 h with 30 nM paclitaxel, 10 µM U0126, or 10 µg/ml anti-MGSA neutralizing antibody mAb275 (R&D Systems, Inc.) as indicated. Before harvesting, cells were pulsed with 1 µCi of [3H]thymidine for 18–20 h.
Isolation of RNA from Matched Lung Carcinoma/Normal Tissue Pairs.
Human tissue sample pairs from non-small cell lung carcinoma and adjacent lung tissue were obtained from the UNC Lineberger tissue procurement facility. RNA was isolated by dounce homogenization and purification with an SV Total RNA Isolation kit (Promega; 175 µl of lysis buffer per 30 mg of tissue).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. As observed previously, 250 nM paclitaxel induced low-level apoptosis in H157 cells (3.5-fold induction; Fig. 1
, Lane 3), and although having negligible effects on its own, U0126 greatly enhanced apoptosis by paclitaxel (17.9-fold induction; Fig. 1
, Lane 5). The addition of actinomycin D, although having little or no effect on untreated cells or on cells treated with paclitaxel or U0126 alone, reduced induction of apoptosis for cells treated with paclitaxel and U0126 from 17.9-fold to 7.4-fold (Fig. 1
, Lane 6). These results suggest that de novo gene synthesis is required for the enhanced levels of cell death induced by paclitaxel and U0126.
|
. Clusters 1, 2, and 3 (52 gene sequences total) represented the most prominent expression patterns. Expression of genes in these clusters was reduced by U0126 regardless of the presence of paclitaxel. Cluster 4 (19 genes) also was well conserved, but with the converse pattern-induced expression by U0126 regardless of paclitaxel. Although smaller in scope and less uniform, clusters 5 and 6 showed patterns of induction or reduction, respectively whenever paclitaxel was present. Cluster 7 was reduced by all drug treatments, and cluster 8 was induced by all drug treatments.
|
Assessment of Gene Targets for Paclitaxel- and U0126-mediated Cell Death.
The genes from each of the clusters were classified by function and are presented in Table 1
. Among the genes identified are several cell cycle and cell death regulatory genes. Cyclin D1 (cluster 1) is a well-characterized target of MEK activation and was identified in a previous study of Raf targets (24)
. Our results also link MEK inhibition for the first time to a decrease of a putative G0-G1 switch gene, G0S2 (Ref. 25
; cluster 2) and an increase of cyclin G2 (cluster 4). MEK inhibition led to a decrease of several cell death regulatory genes, including the Bcl-2 family member Mcl-1, the tumor necrosis factor receptor-associated protein Traf1, the serine/threonine kinase DRAK1, and the immediate early gene IEX1 (clusters 1 and 3). The reduction by U0126 of the glucose transporter GLUT3 (cluster 1) and the glucose-dependent enzyme GFAT2 (cluster 2) has interesting implications given the proposed correlation between glucose metabolism and apoptosis (26)
. Interestingly, paclitaxel and U0126 in combination caused a reduction in TOP3B (27)
and a regulator of the calcium signaling pathway, inositol 1,4,5-trisphosphate-3-kinase C (cluster 9). All of the latter findings are novel and could contribute to apoptosis by paclitaxel and U0126 in combination.
|
Other findings of our study include the reduction of TGFß receptor II (cluster 3) by MEK inhibitor and the TGFß superfamily member PLAB (cluster 6) by paclitaxel. The cell adhesion proteins intercellular adhesion molecule 1and integrin ß5 (cluster 1) are down-regulated by MEK inhibitor, whereas mucin 2 (cluster 5) and mucin 1(cluster 12) are up-regulated by paclitaxel. Effects on DNA topology and transcriptional control also are suggested by reductions in HMG2a by U0126 (cluster 3) and CHAF1bby paclitaxel (cluster 6). The latter findings all are novel.
Paclitaxel and U0126 Modulate Expression of Groups of Genes with Common Function.
To further understand relevant functional pathways contributing to paclitaxel and U0126 activity in cancer, Affymetrix was repeated a third time, using Human Cancer chips comprising 
1700 genes. A lower stringency was used for filtering to include interesting genes that were missed by the more stringent screening methods used to generate Table 1
. Genes with differences of at least 1.8-fold and a minimum signal of 300 were grouped by functional or signaling pathways (Fig. 3)
. This included an array of genes with diverse functions and expanded on many of the genes in Table 1
. DUSP 5and 10were added to the previously identified DUSPsto form a group of four that were U0126-modulated (Fig. 3
, panel 1). Interestingly, a member of the IAP family of caspase inhibitors, CIAP-2 (34)
, was identified with Mcl-1 and IEX-1 as an additional antiapoptosis gene inhibited by U0126 (Fig. 3
, panel 3). CIAP-2 was missed in previous Raf target analyses (24
, 28)
, probably because its expression is tissue-restricted (34)
. In addition to IL-8, paclitaxel was found to slightly induce, and U0126 to reduce, expression of MGSA, a related gene implicated in the melanoma growth and transformation (Refs. 35
, 36
; Fig. 3
, panel 5). To our knowledge, this is the first demonstration that paclitaxel induces MGSA. A further link to the TGFß pathway was supported by the identification, along with TGFß receptor II, of a downstream gene, TSC-22 (Ref. 37
; Fig. 3
, panel 6). We also demonstrated coordinate down-regulation of four EGF-like growth factors by U0126, with HB-EGF being dramatically induced by paclitaxel (Fig. 3
, panel 7). Four members of the plasminogen activator pathway (38)
, four integrins, three Ets family transcription factors, and three calcium regulation genes displayed a similar pattern of suppression by U0126 (Fig. 3
, panels 8–11). Cytochrome P450 subunits displayed an opposite trend, exhibiting enhancement by U0126 (Fig. 3
, panel 12). The identification of diverse sets of functionally relevant genes that are regulated in parallel provided further evidence for the coordinated targeting of genes by paclitaxel and U0126 and was consistent with the global effects of these drugs on the transcriptional program of a tumor cell.
|
. A comparison of results from Affymetrix and Northern analyses confirmed that, although levels of activation may fluctuate, patterns of expression determined by the two techniques are quite similar.
|
. In addition to the 250 nM paclitaxel dose used for DNA microarray analysis, cells were treated with 3 µM paclitaxel to examine dose dependence. These concentrations are within the clinically achievable range (0.1–10 µM; Ref. 3
). An alternative MEK inhibitor, PD98059, was tested as an additional control. Consistent with the Affymetrix results for H157 cells, U0126 induced expression of CYP1B1 in OVCA 194 cells and reduced expression of all other genes in both OVCA 194 and Hs294T cells (Fig. 4B
, Lanes 4 and 5). Modulation by PD98059 (Fig. 4B
, Lane 6) paralleled modulation by U0126, with the exception of the CYP1B1 gene in OVCA 194 cells. CYP1B1 previously was identified as a Raf-responsive gene (24)
. This discrepancy therefore likely reflects the lower potency of PD98059 compared with U0126 in inhibiting MEK activity (22)
. Several of the genes tested were induced by paclitaxel (Fig. 4B
, Lanes 2 and 3), including CIAP-2 in OVCA 194 cells; EPH-A2, IL-8, and TGFBR2 in Hs294T cells; and MGSA and PLAUR in both cell lines.
MGSA Is Induced by Paclitaxel in Selected Melanoma Cell Lines and Enhances Proliferation of C8161 Cells.
To assess implications of these data beyond lung carcinoma, we focused on MGSA/GRO1. MGSA is expressed in 70% of human melanomas, and aberrant overexpression is implicated in melanoma progression (35
, 36)
. Melanomas are typically poorly responsive to paclitaxel treatment, and improvement in their response to paclitaxel would be highly desirable. To test whether MGSA is induced by paclitaxel in alternate human melanoma cells, Hs294T and three additional melanoma cell lines were treated with paclitaxel, and RNA was harvested after a time course of drug addition. Northern analysis demonstrated that paclitaxel induced MGSA in C8161 cells, with two peaks of expression at 1.5 and 10 h; and in Hs294T cells, with peak expression at 1.5–3 h (Fig. 5A)
. Expression in RPMI 7951 and SK-MEL-2 cells was minimal. We have observed variable induction levels in the C8161 and Hs294T cell lines, from 2- to 10-fold, perhaps because of effects on basal MGSA levels by cell density or additional unknown factors. However, a bimodal peak in C8161 cells was reproducible and also was observed for the mouse melanoma cell line B16.5
Taken together, these findings indicate that paclitaxel induces MGSA in a subset of melanoma cells, with various time courses of expression.
|
, Lanes 3 and 4), confirming that MEK inhibition reverses MGSA induction. The p38 inhibitor SB202190 also reduced expression slightly, indicating some overlap with the p38 pathway (Fig. 5B
, Lane 5). Recently, we showed that paclitaxel and U0126 synergistically block PI3K (39)
. The PI3K inhibitor wortmannin, however, did not reduce expression (Fig. 5B
, Lane 6), indicating that U0126 does not reduce MGSA induction by paclitaxel indirectly through inhibition of the PI3K pathway.
To determine whether induction of MGSA by paclitaxel is dose-dependent, we performed real-time PCR on C8161 cells treated with a range of paclitaxel doses (Fig. 5C)
. The results demonstrated detectable induction at doses as low as 3 nM, with expression increasing with increasing paclitaxel. This suggests that MGSA is induced by even very low concentrations of paclitaxel and that induction is dose-dependent.
The tumorigenic effect of MGSA involves, in part, stimulation of cell proliferation (40)
. To determine whether induction of MGSA expression in C8161 cells enhances proliferation, we assessed [3H] incorporation rates in the presence or absence of a specific neutralizing antibody to MGSA, mAb275 (R&D Systems, Inc.). We used 30 nM paclitaxel because this dose, although inducing near-maximum MGSA levels (Fig. 5C)
, showed minimal cytotoxicity in our assay. In the absence of paclitaxel, mAb275 had no detectable effect on [3H]thymidine incorporation (Fig. 5C
, Lane 2 versus Lane 1). In contrast, in the presence of 30 nM paclitaxel, mAb275 decreased proliferation (Fig. 5C
, Lane 4 versus Lane 3). Similar results were observed in a cell-staining-based cytotoxicity assay (not shown). These findings suggest a mechanism whereby paclitaxel induces a gene that could compromise its efficacy by promoting proliferation and whereby MEK inhibitor reverses this effect. Taken in the context that the clinical efficacy of paclitaxel in melanomas is poor, these findings provide for the intriguing possibility of combining the two drugs for chemotherapy of selected melanomas.
TOP3B Represents a Potential Gene Target for Treatment of Lung Carcinoma with Paclitaxel and MEK Inhibitor.
TOP3B (cluster 9) displays the interesting expression pattern in which reduction is achieved only by a combination of paclitaxel and U0126 (Fig. 3
; Table 1
). TOP3B, although not extensively studied, belongs to a class of topoisomerases that associates with disease-related genes and can rescue a slow-growth phenotype in yeast (27
, 41)
. Recent studies also demonstrated a role for this gene in viability (42)
. Real-time PCR verified that the TOP3B transcript is reduced specifically on combination treatment (Fig. 6A)
. To determine whether TOP3B may represent a new biotarget for chemotherapy, its expression was analyzed in eight primary lung carcinomas (Fig. 6B)
. Adjacent unaffected tissues were used as controls. The results demonstrated that TOP3B is expressed >2-fold higher in lung carcinoma samples in seven of eight sample pairs with an average expression versus normal tissue of 5.7. To our knowledge this represents the first correlation of TOP3B expression with cancer. Although further studies will be necessary to define the precise mechanisms of TOP3B in cell death mediated by paclitaxel and U0126, these findings suggest that TOP3B may be a potential tumor marker as well as a therapeutic target.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To provide extensive substantiation of our findings, we built several levels of controls into our analysis. Microarray analysis was performed in duplicate or in triplicate, and Northern analysis or real-time PCR was performed to confirm expression patterns for selected genes. Similar expression profiles in ovarian and melanoma cells support the contention that many of our results may have application to other types of malignancies as well. This is particularly interesting for melanomas, because clinical trials using chemotherapy to treat melanomas have not significantly improved patient prognosis (43) . Our results indicate several unintended effects of paclitaxel that may negate its efficacy, most specifically for melanomas, including the enhancement of MGSA. MGSA is significantly reduced by MEK inhibitor, which argues for a combination of paclitaxel and MEK inhibitors as a candidate chemotherapeutic strategy for melanomas.
The combination drug treatment likely affects the expression of an array of genes, which then leads to the final outcome. This is underscored by the clustering analysis, which identified multiple genes that have distinct profiles of modulation by paclitaxel and U0126. Many genes can be grouped by function, suggesting that cellular pathways are targeted. This conservation in modulation profile of sets of genes provides additional substantiation of these results and is consistent with global effects on transcription that in combination might lead to enhanced cell death.
U0126 primarily reduced, rather than induced, gene expression. This finding is consistent with the ability of the specific targets of MEK, Erk 1 and Erk 2, to phosphorylate and activate multiple transcriptional activators, including ets and the activator protein-1 family activators, creb and c-myc (30
, 44)
. The few genes that were increased by U0126 might be attributable to a limited number of activators that are repressed by Erk phosphorylation (30
, 45
, 46)
or to association of transcriptional activators with histone deacetylases or other corepressors, as observed recently for Elk-1 after prolonged activation (47)
. Paclitaxel, in contrast, elicited relatively few changes in gene expression, suggesting a predominantly post-transcriptional mechanism for this drug. The observation that actinomycin D inhibits enhanced cell death in combination-treated cells but does not effect the levels of cell death elicited by paclitaxel alone (Fig. 1)
also is consistent with a predominantly post-transcriptional mechanism for paclitaxel. Paclitaxel has been reported to regulate several genes at the transcriptional level that were not found in the present study (48, 49, 50)
. Our failure to identify these genes could be a result of our comparatively stringent criteria for inclusion of genes, to our use of a lower more clinically relevant concentration of paclitaxel, or to differences in biological models.
Several modulations in expression were identified that are consistent with the cell cycle block and cell death induced by paclitaxel and U0126. Novel findings include modulation by U0126 of cyclin G2, the putative G0-G1 switch gene, G0S2 (25)
, and a member of the inhibitor of apoptosis family, CIAP-2 (34)
. GLUT3 and GFAT2 are down-regulated by U0126, consistent with the apoptotic role of other glucose metabolism genes, including GLUT1, GLUT4, hexokinase, and phosphofructokinase 2(26)
. Genes involved in calcium signaling may contribute to apoptosis as well. Inositol 1,4,5-trisphosphate kinase-C functions along a calcium signaling pathway activated by PI3K (51)
, and its specific reduction by the two drugs in combination is interesting in view of our findings that PI3K activity is reduced by paclitaxel and U0126 in combination (39)
. TOP3B represents an additional gene down-regulated specifically by paclitaxel and U0126 in combination. Overexpression in seven of eight lung carcinoma patient samples suggests that TOP3B represents a potential tumor marker as well as a chemotherapeutic target (Fig. 6B)
. Although TOP3B has been proposed to have a role in cell growth and viability (27
, 42)
, the precise mechanism of this gene as it relates to paclitaxel- and U0126-mediated cell death is unclear. Ongoing experiments indicate that TOP3B expression reduces basal cell death; however, TOP3B alone cannot overcome the intense effects of paclitaxel and U0126 and may require the coexpression of additional genes. These findings support a hypothesis that the drug combination has a multiplicity of effects that culminate in the significant enhancement of cell death.
In addition to cell cycle control and antiapoptosis genes, we have identified multiple other pathways modulated by U0126 and paclitaxel with relevance to adhesion, invasion, and metastasis. MGSA, intercellular adhesion molecule 1, and four integrins are down-regulated by U0126, as well as several genes along the plasminogen activator pathway, which is involved in angiogenesis, and whose expression predicts poor patient prognosis (38)
. The EGF family of growth factors are known to cause tumor cell proliferation, and their down-regulation by MEK inhibition could be advantageous in the treatment of metastases (24)
. This is of particular relevance for combination treatment because paclitaxel activates expression of HB-EGF (Fig. 3)
.
Recently, targets for Raf-1 were identified by subtractive hybridization in Rat-1 fibroblasts (28) and by microarray analysis of epithelial cells exogenously expressing Raf-1 (24) . The primary effect of Raf-1 is the activation of MEK1 and -2. Interestingly, fewer than 20 genes found in these other studies overlap with our analyses. Differences could be attributable to the introduced expression of Raf-1 in these other systems. Our model has the advantage that the MEK pathway was specifically interrupted in its natural context and as induced by the front-line therapeutic agent paclitaxel. The modulation of gene expression in the context of a clinically relevant chemotherapeutic agent has significant implications for the pharmacologic intervention of cancer by MEK inhibition. Differences in cell types or time courses of exposure or our use of a chip with nearly twice as many genes as the previous microarray study (24) also could account for differences in target identification. In addition, some of the differences may be genes that are Raf-dependent but MEK/Erk-independent.
In summary, this work shows the global effects of paclitaxel and U0126, either alone or in combination. A myriad of genes with known functions in apoptosis, cell cycle, cell growth, and metastasis are affected by this combination. Functional genomic analysis identified several new targets, including MGSA and TOP3B. The extension of these findings to other cancer cell types and to primary cancer cells provides a comprehensive analysis of the potentials for this new combination drug treatment.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by NIH Grant CA-58223. 
2 To whom requests for reprints should be addressed, at Lineberger Comprehensive Cancer Center, Campus Box 7295, University of North Carolina, Chapel Hill, NC 27599. Phone: (919) 966-5538; Fax: (919) 966-8212; E-mail: panyun{at}med.unc.edu
3 The abbreviations used are: MEK, mitogen-activated protein kinase kinase; ERK, extracellular signal-regulated kinase; IL, interleukin; MGSA, melanoma growth stimulatory activity; TOP3B, topoisomerase IIIß; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; 6FAM, 6-carboxyfluorescein; TAMRA, 6-carboxytetramethylrhodamine; TET, tetrachloro-6-carboxyfluorescein; DUSP, dual specificity phosphatase; TGF, transforming growth factor; EGF, epidermal growth factor; PI3K, phosphatidylinositol 3'-kinase kinase.
4 D. T. Bergstralh, unpublished observations.
5 D. J. Taxman, unpublished observations.
Received 3/ 4/03. Revised 5/13/03. Accepted 6/13/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B. J. Clin. Investig., 107: 241-246, 2001.[Medline]
and interleukin-1 in macrophages. J. Leukoc. Biol., 52: 119-121, 1992.[Abstract]
B- and AP-1-dependent mechanism. Cancer Immunol. Immunother., 49: 78-84, 2000.[Medline]
/DOß heterodimer. J. Immunol., 165: 1410-1416, 2000.. Biochem. Biophys. Res. Commun., 206: 792-798, 1995.[Medline]
. Nucleic Acids Res., 29: 3172-3180, 2001.
. Science (Wash. DC), 274: 2100-2103, 1996.This article has been cited by other articles:
|
|
 |
|
  H. Hu, G.-x. Li, L. Wang, J. Watts, G. F. Combs Jr., and J. Lu Methylseleninic Acid Enhances Taxane Drug Efficacy against Human Prostate Cancer and Down-Regulates Antiapoptotic Proteins Bcl-XL and Survivin Clin. Cancer Res., February 15, 2008; 14(4): 1150 - 1158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Vijapurkar, W. Wang, and R. Herbst Potentiation of Kinesin Spindle Protein Inhibitor-Induced Cell Death by Modulation of Mitochondrial and Death Receptor Apoptotic Pathways Cancer Res., January 1, 2007; 67(1): 237 - 245. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Moscova, D. J. Marsh, and R. C. Baxter Protein Chip Discovery of Secreted Proteins Regulated by the Phosphatidylinositol 3-Kinase Pathway in Ovarian Cancer Cell Lines Cancer Res., February 1, 2006; 66(3): 1376 - 1383. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Hu, C. Jiang, C. Ip, Y. M. Rustum, and J. Lu Methylseleninic Acid Potentiates Apoptosis Induced by Chemotherapeutic Drugs in Androgen-Independent Prostate Cancer Cells Clin. Cancer Res., March 15, 2005; 11(6): 2379 - 2388. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.-C. Huang, S.-Y. Liu, M.-H. Lin, Y.-Y. Kuo, and Y.-C. Liu The Synergistic Cytotoxicity of Cisplatin and Taxol in Killing Oral Squamous Cell Carcinoma Jpn. J. Clin. Oncol., September 1, 2004; 34(9): 499 - 504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Whitsett, C. J. Bachurski, K. C. Barnes, P. A. Bunn Jr., L. M. Case, D. N. Cook, D. Crooks, M. W. Duncan, L. Dwyer-Nield, R. C. Elston, et al. Functional Genomics of Lung Disease Am. J. Respir. Cell Mol. Biol., August 1, 2004; 31(2/S1): S1 - S81. [Full Text] [PDF]
|
|
|
|
 |
|
 X. Lu, W. E. Burgan, M. A. Cerra, E. Y. Chuang, M.-H. Tsai, P. J. Tofilon, and K. Camphausen Transcriptional signature of flavopiridol-induced tumor cell death Mol. Cancer Ther., July 1, 2004; 3(7): 861 - 872. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
indicates cell cultures to which 5 ng/ml actinomycin D (ActD) also was added. All drug treatments were for 24 h before analysis by ELISA. Bars, SD.
, DMSO;
, 250 mM pacilitaxel; 
, 10 mM U0126.
