Advertisement
Molecular Biology and Genetics

Molecular Profiling of Transformed and Metastatic Murine Squamous Carcinoma Cells by Differential Display and cDNA Microarray Reveals Altered Expression of Multiple Genes Related to Growth, Apoptosis, Angiogenesis, and the NF-κB Signal Pathway

Gang Dong, Elena Loukinova, Zhong Chen, Lisa Gangi, Tatyana I. Chanturita, Edison T. Liu and Carter Van Waes
DOI:  Published June 2001

Abstract

To identify changes in gene expression with transformation and metastasis, we investigated differential gene expression in a squamous carcinoma model established in syngeneic mice. We used mRNA differential display (DD) to detect global differences and cDNA arrays enriched for cancer-associated genes using mRNA from primary keratinocytes, transformed Pam 212 squamous carcinoma cells, and metastases of Pam 212. After DD, 72 candidate cDNAs expressed primarily in transformed and metastatic cells were selected and cloned. Fifty-seven were detected, and 32 were confirmed to be differentially expressed by Northern blot analysis. mRNA expression profiles were also generated using a mouse cDNA array composed of 4000 elements representing known genes and expressed sequence tags plus the 57 DD candidate cDNAs detected by Northern analysis to facilitate data validation. cDNA array detected 76.9% of the differentially expressed mRNAs selected from DD and confirmed by Northern blot, whereas low-abundance mRNAs did not reach the threshold for detection by the lower-sensitivity array method. Clustering analysis of DD and array results from transformed and metastatic cells identified genes that exhibited decreased or increased expression with transformation and metastasis. Alterations in the expression of several genes detected during tumor progression were consistent with their functional activities involving growth (p21, p27, and cyclin D1), resistance and apoptosis (glutathione-S-transferase, cIAP-1, PEA-15, and Fas ligand), inflammation and angiogenesis [chemokine growth-regulated oncogene 1 (also called KC)], and signal transduction (c-Met, yes-associated protein, and syk). Strikingly, 10 of 22 genes in the cluster expressed in metastases have been associated with activation of the nuclear factor (NF)-κB signal pathway. The NF-κB-inducible cytokine Gro-1 was recently shown to promote tumor growth, metastasis, and angiogenesis of squamous cell carcinomas in vivo (Loukinova et al., Oncogene, 19: 3477–3486, 2000). The results demonstrate that early response genes related to NF-κB contribute to metastatic tumor progression. Comparison of cell lines and tumor tissue revealed a concordance of ∼50% by array, and 70% for Northern-confirmed, metastasis-related genes. Functional genomic approaches comparing expression among cell lines and tumor tissue may promote a better understanding of the genes expressed by malignant and host cells during tumor progression and metastasis.

INTRODUCTION

The prevalence of SCCs 3 of the skin, upper aerodigestive tract, lungs, and cervix leads to significant cancer morbidity and mortality worldwide, and aggressive local and metastatic disease is the most common cause of death in patients with SCC, as with other cancers. Tumor development and metastasis is a complex process that includes transformation, proliferation, invasion, neovascularization, and metastatic spread (1) . Studies to identify molecules involved in tumor development and progression of SCC by us and others have resulted in the identification of a number of individual candidates that function as cell growth factors (2) , 4 adhesion molecules (3 , 4) , proteases (5) , and cytokines that promote inflammation and angiogenesis (6 , 7) . Interestingly, several of these candidates have been shown to be regulated by early injury response signal pathways and transcription factors. We and others have demonstrated that several early response transcription factors, such as NF-κB, AP-1, and signal transducer and activator of transcription-3, are constitutively activated in SCC (8, 9, 10) . Activation of NF-κB, AP-1, and signal transducer and activator of transcription-3 have been found to contribute to tumor cell transformation, proliferation, migration, survival, and resistance to chemo- and radiation therapy in a variety of different cancers (9, 10, 11, 12, 13, 14, 15, 16) . However, there has been limited characterization of the global changes in expression of genes and their relationship to activation of specific signal pathways during transformation and metastasis of SCC. Such studies have been limited by the lack of well-defined models of metastasis and technical obstacles to the comparative analysis of large numbers of genes in multiple samples.

To recapitulate the stages of transformation and metastatic tumor progression of SCC, we have developed a syngeneic model of SCC that includes the spontaneously transformed BALB/c keratinocyte line Pam 212 (17) , and rare lymph node and lung metastases (LY and LU) of Pam 212 (18) . The metastatic Pam LY and Pam LU cell lines were found to form tumors and metastases at a higher rate than the parental Pam 212 tumor line in vivo (18) in association with increased expression of a repertoire of proinflammatory and proangiogenic factors (19 , 20) that are regulated by transcription factor NF-κB (21) . The development of mRNA DD and, more recently, cDNA microarray technology has made it feasible to analyze the expression of large numbers of genes and their potential relationship to activation of signal pathways such as NF-κB during the stages of transformation and metastasis.

In the present study, we used mRNA DD to detect global differences and cDNA microarrays enriched to detect cancer-associated genes to examine stable differences in mRNA expression between primary keratinocytes, transformed Pam 212 squamous carcinoma cells, and metastases of Pam 212. Microarray analysis was also used to compare mRNA from tumor cells grown in vitro and in vivo to obtain an estimate of differences in mRNA between cell lines and heterogeneous tumor tissue. cDNAs differentially expressed between the nontransformed and transformed cell lines and between the transformed and metastatic variant cell lines were identified, including genes with functional activities involved in growth, resistance and apoptosis, inflammation and angiogenesis, and signal transduction. Microarray analysis identified a cluster of genes showing increased expression in metastatic lines, 10 of which were related to the NF-κB pathway. The results of genomic analysis along with recent functional analysis of several candidate genes in this murine SCC model provide evidence for the importance of the expression of NF-κB-related genes in the metastatic tumor progression of SCC. Comparison of cell lines and tumor tissue revealed a concordance of ∼50% by array, and 70% for Northern-confirmed, metastasis-related genes. Functional genomic approaches comparing expression among cell lines and tumor tissue may promote a better understanding of genes expressed by malignant and host cells during tumor progression and metastasis.

MATERIALS AND METHODS

Cell Culture and Tumor Growth in BALB/c Mice.

Primary keratinocytes were isolated from the skin of male BALB/c neonates as described previously (17) and grown either in Eagle’s minimal essential media plus 10% fetal bovine serum (thus designated as high Ca2+) or the same media with the calcium concentration reduced to 0.02 mm (low Ca2+; Ref. 19 ). The PAM 212 cell line is a spontaneously transformed cell line derived from neonatal BALB/c skin keratinocytes in vitro (17) , which forms SCCs in vivo but rarely metastasizes when s.c. inoculated in normal BALB/c mice (18) . The metastatic LY and LU lines were isolated from lymph node and lung metastases of PAM 212 tumor implants in BALB/c mice (18 , 19) . The Pam 212, LY, and LU cell lines used were confirmed previously to retain keratin and integrin markers of squamous epithelial origin (18) . All tumor lines were grown in Eagle’s minimal essential media plus 10% fetal bovine serum and penicillin, streptomycin, and glutamine. To grow tumors from cell lines in syngeneic hosts, 5 × 106 PAM 212 and LY cells isolated during the log-phase of growth were injected s.c. into male BALB/c mice. Animals were euthanized when tumors reached ∼1 cm in diameter, and tumors were dissected free of skin, underlying tissue, and capsule and snap frozen and stored at −80°C.

RNA Isolation, mRNA Differential Display, and Northern Blot Analysis.

Total RNA was isolated from cultured primary keratinocytes, Pam 212, LY-1, LY-2, LY-8, and LU-1 cells using Trizol reagent (Life Technologies, Inc., Gaithersburg, MD). DD was performed with RNAimage kits (GenHunter Co., Brookline, MA) following the manufacturer’s instructions, as described previously (19) . Anchor primers H-T11C and H-T11A were used for reverse transcription of mRNAs, and the resulting cDNAs were amplified by PCR using these anchor primers and random primer sets H-AP1–8 and H-AP17–32. To confirm the results of DD, Northern blotting analysis was performed using 20 μg of total RNA from the above cell lines with 32P-labeled cDNA probes from DD, as described (19) . Each blot was then stripped and rehybridized with a glyceraldehyde-3-phosphate dehydrogenase probe to confirm equal loading. After sequencing of cloned DD products using the Dye Terminator ABI PRISM Kit (Perkin-Elmer, Foster City, CA), the alignment of insert sequences was performed with MacDNASIS (Hitachi Software Engineering America, Ltd., San Bruno, CA), and the BLAST program was used for the homology search of GenBank.

cDNA Arrays, Probes, Hybridization, and Scanning.

Poly(A)+ RNA or total RNA was isolated from primary cultured murine keratinocyte, PAM 212, LY-1, and LY-2 cells and tumors of PAM 212, LY-1, and LY-2 using the FastTrack mRNA isolation kit (Invitrogen, Carlsbad, CA) or Trizol reagent, respectively (Life Technologies, Inc.). mRNA from primary mouse keratinocytes, tumor line Pam 212, and metastatic lines LY-1 and LY-2 were used to generate Cy3- or Cy5-labeled first-strand cDNA by reverse transcription, and the cDNA products were used to hybridize array slides. To examine similarities and differences between cultured cells and tumors arising from these lines in vivo, mRNA was also isolated from PAM 212, metastatic LY-1, and LY-2 tumors grown in BALB/c mice and labeled for array analysis. mRNA from Pam 212 cells was labeled with Cy5 dye and paired with the rest of samples that were labeled with Cy3 dye to allow a direct comparison between primary keratinocytes and PAM 212 tumor line or PAM 212 line and metastatic lines. To make fluorescence-labeled cDNA targets by reverse transcription, 1–2 μg of poly(A)+ RNA or 50–100 μg of total RNA was incubated in a cocktail containing Cy3 or Cy5-dUTP (Amersham Pharmacia Biotech Inc., Piscataway, NJ) and SuperScript II RT (Life Technologies, Inc.). Labeled targets were purified using a Microcon column (Millipore, Bedford, MA). The appropriate Cy3 and Cy5 targets were combined, along with 2 μl (20 μg) of mouse COT-1 DNA, 1 μl (8–10 μg) of poly(A), 2.6 μl of 20× SSC, and 0.45 μl of 10% SDS in a final volume of 15 μl. After denaturation, labeled targets were added to processed National Cancer Institute mouse array slides which were then placed in hybridization chambers and incubated overnight (10–16 h) at 65°C. The next day, slides were washed for 1 min in 1× SSC, for 1 min in 0.2× SSC, for 10 s in 0.05× SSC, then spin dried. Fluorescence images were captured using a Genepix 4000 (Axon Instruments, Inc., Foster City, CA).

Statistical and Cluster Analysis of Array Data.

Both image and signal intensity data were stored in a database supported by the Center for Information Technology of NIH. Cy3:Cy5 intensity ratios from each gene were calculated and subsequently normalized to ratios of overall signal intensity from the corresponding channel in each hybridization. In all of the data sets included for the data analysis, ratios of overall signal intensity ranged from 0.7 to 1.25. The ratio distribution extracted from microarray images exhibited a normal distribution, constant coefficient-of-variation, and sufficient positivity (we used 3-fold above background as the intensity cutoff in current experiments), satisfying the statistical conditions necessary to determine the significance of ratio measurements by the confidence interval generated using Array Suite software developed at the National Human Genome Research Institute (Bethesda, MD; Ref. 22 ). Under our experimental conditions, a 99% confidence interval defined ratios of 2 and larger or 0.5 and smaller as indicators of significantly different gene expression levels between two samples hybridized to the same array spot. These ratios were designated as positives. A set of 287 cDNAs was extracted from the database by filtering cDNAs with at least two positive ratios among 10 hybridizations.

For clustering analysis, we used Cluster and Treeview array data clustering and visualization software developed at Stanford University (23) . Log2-converted expression data from the set of 287 cDNAs measured across 10 hybridizations were subjected to one-dimensional hierarchical clustering, and we used this 287 × 10 (cDNA versus specimens) expression table in Cluster to generate gene dendrograms based on the pair-wise calculation of the Pearson correlation coefficient of normalized fluorescence ratios as measures of similarity and average linkage clustering. The reordered table from Cluster was imported into Treeview and displayed by a graded color scheme, representing ratios for each cell line in the expression table.

RESULTS

Comparison of mRNA Expression in Mouse Keratinocytes, PAM 212, and PAM Metastatic Variant Cells by mRNA DD and cDNA Microarray.

To obtain a molecular profile of differential gene expression that is associated with transformation and metastatic tumor progression, we used the random primer-based mRNA DD method to detect global differences and cDNA microarrays enriched to detect cancer-associated genes (Fig. 1) ⇓ . We compared mRNA expression between primary keratinocytes, the spontaneously transformed squamous carcinoma line PAM 212, and PAM 212-derived LY and LU cell lines. The PAM 212 line used for these studies is tumorigenic and forms SCCs, whereas the PAM LY and LU cell lines reisolated from rare metastases exhibit more rapid tumor growth and a high rate of metastasis in vivo (18) . As a means to compare the two methods for detection of differentially expressed mRNAs, Northern blot analysis was used to confirm differential expression of candidates identified by DD, and cDNAs of these differentially expressed genes were included in cDNA microarrays for validation (Fig. 1) ⇓ .

Fig. 1.
Fig. 1.

Diagram of multistage PAM 212 tumor model and methods used for identification of differentially expressed genes using DD and microarray analysis. RNA was harvested from primary BALB/c keratinocytes, PAM 212 cells, and metastatic PAM LY and LU cells and used for DD and microarray as described in “Materials and Methods.” Representative mRNA DD results (left panel) were obtained using different H-T primers for reverse transcription and H-AP random primer pairs for PCR. Lane 1, BALB/c keratinocytes; Lane 2, Pam 212; Lane 3, LY1; Lane 4; LY-2; and Lane 5, LY-8. For validation of these two methods, 57 cDNAs identified by DD and confirmed by Northern blot analysis were printed onto a mouse array developed at the National Cancer Institute containing 4000 cDNAs. Representative array results (right panel) were obtained comparing arrays hybridized with cDNA from keratinocytes, Pam 212, and Pam metastatic cell lines. The differentially expressed genes identified by DD were confirmed by reverse transcription-PCR and Northern blot analysis and sequenced for comparison with GeneBank.

Cluster Analysis of Genes Detected by mRNA Differential Display.

DD was performed using 240 primer pairs. Approximately 12,000 cDNA bands were generated, and >97% were displayed in all cell types, consistent with a common syngeneic and epithelial origin of the nonmalignant and malignant cell lines. Four hundred and sixty-three bands (<3% of the mRNA population) exhibited differential expression patterns among the cell lines analyzed, and >90% of those differentially expressed fell into one of five categories, as summarized in Fig. 2 ⇓ (left panel). These included bands detected (a) mostly in primary or transformed keratinocytes, and which exhibited a decrease in expression with tumor progression and metastasis; (b) mostly in PAM 212 cells; (c) in all tumor lines; (d) in metastatic lines only; and (e) in primary keratinocyte and metastatic lines. The percentage in each category are shown in Fig. 2 ⇓ (right panel). The major differences in cDNA fragments detected were seen between primary keratinocytes and the five tumor lines, which comprised 76.8% of all cDNA candidates identified, and 75 (16.1%) of the cDNA fragments were detected uniquely in metastatic lines. The three fragments (0.7%) detected in both primary keratinocytes and metastatic lines and those expressed only in one or two LY or LU lines (not shown) were not considered for additional study.

Fig. 2.
Fig. 2.

Summary of differential gene expression detected by DD. Using 240 pairs of PCR primers, ∼12,000 expressed total cDNA fragments and 463 differentially expressed cDNAs were detected. Gene expression patterns detected by autoradiography were confirmed by prolonged exposure of sequencing gels, and bands exhibiting differences in intensity between cell lines were scored. The percentage in each category was calculated and shown as column labels.

Northern Blot Analysis and Identification of Selected cDNAs Detected by DD.

Because a main objective for our present study was to examine the diversity of genes activated after transformation or metastasis, the majority of DD candidates we selected for analysis were primarily those candidate cDNAs which exhibited increased signal intensity in all tumor or all metastatic lines. We also included several cDNAs that showed decreased levels in tumor lines when compared with cultured normal keratinocytes. One hundred such candidate cDNA fragments were selected for additional analysis. We used cDNA fragments from DD gels after PCR amplification and cloning to avoid potential multiple cDNA species present in the original samples recovered from DD gels. Seventy-two bands were successfully amplified, cloned, and found to be suitable for analysis. To verify whether the candidate cDNAs generated by DD detected changes in gene expression in cell lines of different malignant potential, Northern blot analysis was carried out using poly(A)+ or total RNA from the cell lines used in DD. All Northern blots were repeated at least twice using different RNA preparations to check for reproducibility. Overall, we detected signals in 57 of 72 Northern blots performed, and 32 showed differential expression of at least one mRNA species. Fig. 3 ⇓ shows the images of all nonredundant positive blots obtained in the study. All expression patterns showed consistency with DD experiments except band G1, which appeared in metastatic lines by DD but was detected in all tumor lines by Northern analysis. To control for loading, all blots used were reprobed with 32P-labeled cDNA fragment of glyceraldehyde-3-phosphate dehydrogenase and showed similar intensity of the signals (Fig. 3 ⇓ , lower panels).

Fig. 3.
Fig. 3.

Northern blot analysis to confirm differential expression of cDNA clones from DD. Twenty μg of total RNA or 1 μg of poly(A)+ RNA was used from each sample for Northern blotting. DNA fragments amplified from DD experiments were labeled with 32P and used as probes. Each Northern blotting was repeated with cDNA after they were cloned and sequenced to confirm the identity. Shown are the light-exposed images of Northern blotting. Probes are designated by their code names used in DD. Kerat, keratinocytes. Low, calcium level used for culturing primary keratinocytes is below 0.07 mm. High, 2 mm calcium.

The clones identified by DD and confirmed by Northern blot analysis were sequenced and compared with those listed in GenBank. Of the 32 DD cDNA clones confirmed by Northern analysis and which demonstrated >95% homology with sequences in GenBank, 8 were found twice, 3 clones matched with ESTs, and 21 are known genes. In Table 1 ⇓ , the known genes identified by DD are underlined and classified according to their expression pattern with tumor progression and putative function as reported in previous studies. DD and Northern analysis revealed increased expression of candidates associated with inflammation and immune function (e.g., chemokine growth-regulated oncogene 1 [also called KC]; a proteasome component involved in antigen processing, LMP7, and lymphocyte antigen 6 [ 24, 25, 26, 27] ); signal transduction and transcription (c-Met, HGM-I, Rap 1b, and Yes-associated protein 65; Refs. 28, 29, 30, 31 ); growth and apoptosis (gas5 and Nedd-8; Refs. 32 , 33 ); adhesion/migration (ADAMTS-1; Ref. 34 ); metabolism and drug resistance (e.g., glutathione S-transferase; Ref. 35 ) and structure (e.g., Capping protein; Ref. 36 ). Of the 24 unique genes, 8 showed increased expression in the metastatic lines, including Gro-1, c-Met, LMP-7, gas5, capping protein, spermidine synthase (a pseudogene), methlmalonyl-CoA mutase, sperm protein p179, and novel clone O34. The four candidates selected that exhibited decreased expression (thymosin β4, actin- and spectrin-binding proteins, and tropomyosin α) primarily have been associated with functions involved in cell structure and motility.

View this table:
Table 1

Altered gene expression during tumor progression

Cluster Analysis of Genes Detected by cDNA Microarray.

We performed cDNA microarray analysis in the same model using a mouse cDNA microarray containing four thousand genetic elements enriched for cancer-associated genes (Fig. 1) ⇓ , as described in “Materials and Methods.” The 57 cDNAs detected by both DD and Northern blot were included as a measure of validation. A cluster analysis of 287 genes that showed a >2-fold difference in intensity is shown in Fig. 4 ⇓ . At least five gene expression patterns were detected and can be discerned as clusters, according to the relative levels of mRNA gene expression in each cluster. Genes in cluster 1, keratinocytes ≫ PAM 212 and LY, and cluster 2, keratinocytes > PAM 212 > LY, exhibited a decrease in expression with transformation and metastasis; genes in cluster 3, keratinocytes < PAM 212 > LY, cluster 4, keratinocytes < PAM 212 and LY, and cluster 5, keratinocytes < PAM 212 < LY, exhibited increased expression in tumor cells. Thus, among the five clusters, the majority of differences in gene expression were found to reside between keratinocytes and tumor lines, consistent with the results by DD (Fig. 2) ⇓ . The results using both methods demonstrated apparent differences in expression profiles that were associated with the stepwise changes observed in the phenotypic behavior of the cell lines after transformation in vitro and acquisition of a metastatic phenotype in vivo.

Fig. 4.
Fig. 4.

Hierarchical clustering of gene expression data from mouse arrays. Shown is the single dimension clustered gene expression measured from 10 mouse 4K array analyses obtained from 10 samples of normal and malignant keratinocytes. Each row represents a separate cDNA on the microarray, and each column represents the cell source in which expression is compared. Expression among cDNAs was compared from PAM 212 and LY-1 and LY-2 cells grown in vitro and from whole tumors established from the cell lines in vivo (PAM 212*, LY-1*, and LY-2*). Log ratios of hybridized fluorescent cDNA probes from each experiment are depicted according to the color scale shown at the bottom. Color schemes for probe labeling are reflected by colored sample names, top left. Scores underneath sample names represent relative levels of gene expression deduced from ratios of hybridization in each row.

As a measure of the validity of data obtained from microarray, we examined the array results for 57 spots containing cDNAs detected by DD and Northern blot analysis. Fig. 5A ⇓ shows that 26 of 39 (66.7%) clones exhibiting differential expression by DD were detected by microarray. When compared with Northern blot, differential expression of 20 of 26 (76.9%) genes was detected by both methods, with 6 each detected by one method but not the other (Fig. 5B) ⇓ . Because two clones represented the same genes, there were 24 unique genes included. Microarray analysis was also used to compare mRNA from tumor cells grown in vitro and in vivo to obtain an estimate of differences in mRNA between cell lines and heterogeneous tumor tissue. We compared expression of mRNA detected in PAM 212 and LY cell lines grown in vitro and tumors established from PAM 212 and LY in mice (PAM 212*, LY-1*, and LY-2*), as shown in Figs. 4 ⇓ 5 ⇓ 6 ⇓ . We observed a >50% concordant pattern of gene expression overall between in vitro and in vivo samples in cluster analysis (Fig. 5A) ⇓ . For genes detected by DD and Northern blot, Fig. 5A ⇓ shows an ∼70% concordance in the pattern of expression detected in Pam 212 and LY cell lines and Pam 212* and LY* tumors. These results comparing expression among cell lines and tumor tissue provide evidence for differences in the pattern of gene expression between cultured tumor cells in vitro and tumor and host cells in vivo.

Fig. 5.
Fig. 5.

Detection of DD- and Northern-identified genes by microarray. A, clustering analysis performed with cDNA clones identified by DD after data were filtered to select clones, with at least two samples showing a ratio difference by 2-fold or larger. B, results of Northern blotting were compared with array analyses based on the criteria described in A. ND, not determined.

Fig. 6.
Fig. 6.

Microarray detection of genes up-regulated in the PAM tumor progression model. Using the same filtering method as that described in the legend to Fig. 4 ⇓ , we clustered clones that showed predominant up-regulation in metastatic LY cell lines as well as in LY tumors. Their association with NF-κB is based on experimental results as well as on a PubMed search. All putative NF-κB-related cDNAs are indicated in the right column.

Differential Expression of Genes Detected by Microarray Analysis.

Candidates detected by cDNA microarray were classified according to expression in malignant cells and putative function and are listed in Table 1 ⇓ without underlining. The cDNA microarrays enriched for cancer-associated genes detected differences in the expression of multiple genes with diverse functions. The differential pattern of expression and function of several of the candidates detected by cDNA microarray are consistent with those reported for various cancers. For example, decreased expression of growth-modulatory cell cycle-dependent kinase inhibitors cdk inhibitor 1A p21 (CIP1/WAF1) and cdk 4 inhibitor p27 (KIP1) and increased expression of cyclin D1, cdc 25, and PCNA have been implicated frequently in the proliferation of various cancers, including SCC (37, 38, 39, 40) . Increased expression of genes involved in signal transduction and transcription, including the protein tyrosine phosphatase, Myc-associated zinc finger protein, Janus kinase 3, and E2F pathways, and decreased expression of syk have also been reported (41, 42, 43, 44, 45, 46) . Increased expression of PDGF-α and FasL has also been detected in SCC (47 , 48) . The pattern of expression and function of these candidates and others not well studied in SCC remains to be confirmed in this model and a wider panel of SCCs.

Metastasis-associated Genes Exhibit Diverse Functions and Include Multiple Genes Related to the NF-κB Signal Pathway.

We examined further the cluster of genes exhibiting increased expression in metastatic lines and tumors, and the pattern of expression and identity of these genes is shown in Fig. 6 ⇓ . Thirty-four genes were detected in this group, and three included in duplicate for quality assurance appeared twice. Among these genes, 22 are known genes, 1 is a putative gene, and 8 are ESTs. The known genes include candidates with a wide range of putative functions, and several have been shown previously to be expressed in various cancers including SCC. The greatest expression was detected in metastasis for several genes that function in immunological, inflammatory, and angiogenesis responses, including Gro-1 (KC; Refs. 24 and 25 ), complement component 3 (49) , IL-12B (50) , CSF-1 (51 , 52) , and OPN (53, 54, 55) . Several candidates involved in signal transduction and regulation of gene expression and DNA replication were detected, including c-Met (56) , neurotrophic receptor tyrosine kinase (57) , HMG-1(Y) (29) , and replication protein A (Mr 14,000 subunit; Ref. 58 ). Three candidates detected have been reported to be expressed and involved in the modulation of apoptosis in cancer: cIAP-1 encodes a cellular inhibitor for caspase 3 (59 , 60) ; FasL encodes a membrane protein that activates a TNF family receptor that contains a death-domain (48 , 61) ; and PEA-15 encodes a recently cloned Mr 15,000 death effector domain-containing protein (62) . Capping protein α 2 encodes an actin-binding protein that may be involved in cell shape and mobility (36) .

Recently, we demonstrated that increased levels of the Gro-1 (KC) and other cytokines detected in metastatic Pam LY and LU lines is attributable to increased activation of transcription factor NF-κB (21) . We analyzed whether gene expression in the metastasis cluster includes an enriched representation for other NF-κB-related genes by a search of PubMed. Among the 22 known genes, 10 have been reported to be associated with the NF-κB signal pathway (Fig. 6 ⇓ , right column, solid lines and their association,). The 10 NF-κB-related genes identified may be classified into two groups. One group includes candidates involved in the regulation of NF-κB activation. For example, OPN encodes a secreted phosphoprotein that can promote the activation of NF-κB by binding to integrins (63 , 64) . HMG-1(Y) is required for NF-κB-dependent induction of Gro-1 (29) . Protein tyrosine phosphatase has been reported to modulate NF-κB by interacting with IκBα (65) . Ubiquitin-activating enzyme E1 participates in protein ubiquitination of IκB proteins involved in the activation of NF-κB (66 , 67) . The second group includes NF-κB target genes, such as Gro-1/KC, complement component 3, CSF-1, IL-12B, cIAP-1, and FasL. All of these have been shown to be genes regulated by NF-κB in various cell types (21 , 49, 50, 51, 52 , 59, 60, 61) . Thus, the cDNA array results are consistent with the enriched expression of NF-κB-related genes in metastatic SCC lines that exhibit increased activation of NF-κB.

DISCUSSION

In this study, we examined gene expression in a multistage murine model of SCC consisting of nonmalignant, transformed, and metastatic cells using DD and cDNA microarray (Fig. 1) ⇓ . Clusters of differentially expressed mRNAs associated with these stages were detected by both the random primer-based DD method and arrays containing a large number of genes associated with oncogenesis (Fig. 2 ⇓ and 4 ⇓ ). The majority of DD candidates confirmed by Northern analysis (Fig. 3) ⇓ and included in the microarray for comparison were also detected by microarray (Fig. 5) ⇓ . The candidates detected by both methods included novel ESTs and known genes that function in inflammation/immunity, angiogenesis, cell growth and death, signal transduction, metabolism, and cell structure (Table 1) ⇓ . We identified a small group of 26 metastasis-related genes representing 11–16% of all differentially expressed genes identified in the current study (Fig. 6) ⇓ . Among these, 10 of 22 known genes are related to the NF-κB signal transduction pathway. We have examined the functional importance of NF-κB and the NF-κB-inducible gene Gro-1/KC while these gene discovery studies were in progress. We found increased Gro-1 (KC) mRNA and protein expression in the metastases, and we found that Gro-1 (KC) promotes inflammation, angiogenesis, increased tumor growth, and metastasis after enforced expression in the low Gro-1 (KC)-expressing parental PAM 212 SCC in vivo (25) . Conversely, inhibition of NF-κB resulted in inhibition of proangiogenic chemokine expression, angiogenesis, and tumor growth of Pam LY-2 (68) and human UM-SCC-9 cells (9) . These results provide evidence for the functional importance of the NF-κB signal pathway and NF-κB induced gene(s) detected by DD and cDNA microarray in this study.

The development of DD and microarray technology has enabled the study of global expression profiles of random or disease-associated genes. The combination of DD and microarray used in the current study allowed the detection of 30–50% of the total genes that are estimated to be differentially expressed in a given cell type (69 , 70) . In comparing 32 cDNA clones detected by DD, Northern blot, and cDNA array analyses, a 76.9% overlap was obtained. When we reviewed the expression of mRNAs detected by the 12 cDNA clones that showed a discrepancy, all were found to be of relatively low abundance when compared with other positive clones. The low expression levels of these genes either fell into the nonlinear detection range on the array, and failed to pass the filtering process, or were excluded during clustering analysis. We take these results as evidence that the fluorescence dye-based microarray and cluster analysis methods used in the current study may be slightly less sensitive than the PCR-based DD method in detection of low-abundance mRNA species in complex murine and human genomes.

To obtain an estimate of differences in mRNA expression that would be detected in cells in culture and in whole tumors, we included mRNA from Pam 212 and LY cell lines and tumors in microarray analysis (Fig. 4 ⇓ 5 ⇓ 6 ⇓ ). We observed a >50% concordant pattern of gene expression overall between in vitro and in vivo samples in cluster analysis (Fig. 5A) ⇓ and >70% concordance in genes confirmed by DD and Northern analysis (Fig. 5A) ⇓ . Although these results provide evidence that changes observed in cultured cell lines can also be detected in specimens obtained from whole tumors in vivo, they also indicate that there are important differences in expression in mRNA from tumors. Such differences could arise because of the presence of a variety of stromal cells, or because of inducible responses in tumor cells arising from interactions of tumor cells with the host environment. The advent of laser microdissection and other technologies can facilitate future studies to examine which of these differences are attributable to heterogeneity in tumor or stromal cells and activation of genes after exposure of tumor cells to the host environment (71 , 72) .

Among the genes detected using both DD and microarray methods, only a relatively small number of candidate genes identified were found to be related to metastasis (30–50 genes). The detection of a finite number of candidates whose expression and function can be systematically studied underscores the potential usefulness of genomic approaches such as DD and cDNA microarray in the study of gene expression in tumor progression. The number and potential functional relatedness of the candidates detected in this study are consistent with recent results of Clark et al. (73) , whose genomic analysis in murine B16F0 melanoma lung metastasis and human melanotic tumor A375 also detected a relatively small number of candidates and a statistical association between genes detected with tumor progression.

The detection of the differential expression of genes involved in modulation of the NF-κB signal transduction pathway is consistent with our hypothesis and previous experimental data indicating that constitutive activation of NF-κB signal transduction pathway is important in the tumor progression and metastasis of SCC (8 , 9 , 68) . Here we found evidence that increased activation of the NF-κB pathway in metastatic SCC cells is accompanied by the altered expression of several genes that can regulate NF-κB responses. In this group, ubiquitin-activating enzyme E1 participates in the initial steps of protein ubiquitination necessary for degradation of IκBα, an inhibitor of NF-κB activation (66 , 67) . Nedd-8 is another ubiquitin-like protein that can be activated by ubiquitin-activating enzyme E1 (74) , functioning in modification of the Cul-1 component in a specific Skp1, Cdc53 or Cullin, and F-box complex, SCFβTrCP, a ubiquitin-ligase enzyme complex which carries out IκBα ubiquitination (33) . Protein tyrosine phosphatase 1B has been speculated to modulate NF-κB by physically interacting with IκBα, resulting in dephosphorylation of tyrosine in IκBα (75) . LMP-7 is a component of the 26S proteasome that participates in the final stage of degradation of IκB proteins and antigen processing (26) . HMG-I(Y) was originally identified as a basic, nonhistone, chromosomal binding protein that regulates gene expression by relieving histone H1-mediated transcriptional repression and is required for NF-κB-dependent induction of cytokines Gro-1 (KC), and granulocyte macrophage CSF (29 , 76) , which are overexpressed in metastatic PAM cells (20) . Taken together, ubiquitin-activating enzyme E1, Nedd-8, protein tyrosine phosphatase 1B, and LMP-7 are involved in degradation of IκB and NF-κB activation, and HMG-I(Y) acts as functional enhancer for NF-κB transcription activity.

In addition to genes that are related to NF-κB signal pathway, differential expression of components involved in several other signal transduction pathways were identified in transformed or metastatic PAM cells. We detected altered expression of the hepatocyte growth factor/scatter factor receptor c-Met, epidermal growth factor receptor and Ras-associated kinase Rap-1b, PDGF- and the PDGF receptor-associated Yes-associated protein (YAP65), neurotrophic receptor tyrosine kinase, protein tyrosine phosphatase 1B, and syk. The expression patterns of several of these candidates are consistent with alterations detected in other cancers. Activation or elevated expression of c-Met receptor, EGF-R, and PDGF receptor pathway genes have been detected and shown to be important in tumor cell proliferation, survival, angiogenesis and/or metastasis of SCC (28 , 56 , 77, 78, 79, 80, 81) . We have completed studies confirming expression and function of the EGF, c-Met, and PDGF receptors and their ligands in PAM 212 and/or human SCC cell lines. 4 ,5 ,6 We found that activation of epidermal growth factor receptor and c-Met induced expression of angiogenesis factors IL-8 and VEGF in human HNSCC tumor cells. 4 ,5 Expression of neurotrophic receptor has been reported in thyroid papillary carcinoma (82) , and decreased expression of syk has been reported in breast cancer (46) . The relative importance of the altered expression of the genes involved in these other pathways to the malignant phenotype of PAM and other SCC remains to be determined.

We detected altered expression of several genes involved in inflammation and angiogenesis in transformed and metastatic SCC, including Gro-1 (KC), ADAMTS-1, and OPN. In addition to the direct evidence already obtained regarding the role of Gro-1 (KC) in angiogenesis, growth, and metastasis in the Pam model, ADAMTS-1 and OPN are attractive candidates. In contrast to Gro-1, ADAMTS-1 expression was decreased in all three metastatic cell lines. ADAMTS-1 is a mouse metalloproteinase and disintegrase with thrombospondin motifs. Its human ortholog, METH-1, was identified as a member of a new family proteins with antiangiogenic activity (83, 84, 85) . OPN is a phosphoprotein expressed in primary and metastatic carcinomas, including SCC, and has been reported to promote endothelial survival through binding to integrin αvβ3 and activation of NF-κB in a ras- and src-dependent manner (53, 54, 55 , 63 , 64) . OPN is a strong stimulator for endothelial cell migration and is functionally synergistic with VEGF (64) . Taken together, our experimental data provide evidence that differences in expression of multiple genes that can modulate angiogenesis accompany tumor progression and metastasis of SCC.

Several differentially expressed genes that we identified regulate cell growth and apoptosis and are critically altered in tumor development and progression. Cell cycle-dependent kinase inhibitors cdk inhibitor 1A p21 (CIP1/WAF1) and cdk 4 inhibitor p27 (KIP1) have been reported to be down-regulated, and cyclin D1, cdc 25, and PCNA have been shown to be up-regulated and involved in the proliferation of various cancers (37, 39, 40) . cIAP-1, FasL, PEA-15, and Nedd-8 genes are involved in modulating apoptosis. cIAP-1, which exhibits increased expression in Pam and LY SCC cells, has been shown to mediate the NF-κB-mediated resistance of cancer cells to apoptosis, including that induced by TNF, FasL, cytotoxic chemotherapy agents, and radiation (11, 12, 13) . cIAP-1 is a potent inhibitor of cell death proteases, especially caspase-3 and 7, thereby inhibiting apoptosis in tumor cells (59 , 60) . The expression pattern of cIAP-1 observed is consistent with our observations in murine and human SCC that increased resistance to TNFα and radiation-induced apoptosis accompanies activation of NF-κB during tumor progression and metastasis (86 , 87) . 6 Increased expression of PEA-15 in tumor cells has been shown to mediate increased resistance to Fas as well as TNF-induced apoptosis in tumor cells by inhibiting caspase 8 activity (62) . Increased FasL and decreased Fas expression have been detected in human HNSCC, and such altered gene expression reportedly promotes the resistance of tumor cells to FasL-mediated apoptosis while inducing apoptosis and suppression of T cell-mediated immunity (48) .

In conclusion, using genomic approaches, we have detected global alterations in the expression profile of genes associated with transformation and metastatic tumor progression in a murine SCC model and obtained evidence that these changes involve NF-κB and potentially other signal pathways. Activation of these signal transduction pathways is associated with an altered expression of genes involved in angiogenesis, growth, and apoptosis, which can contribute to a more aggressive malignant phenotype, including metastasis. Examination of the function of several of the preferentially expressed genes identified from this study are underway in both animal tumor models and in human HNSCC, and this information may be useful in molecular diagnosis, selection, and targeting of therapy.

Acknowledgments

We thank John Powell, John Greene, Esther Asaki and other members of BioInformatics and Molecular Analysis Section, Center for Information Technology, National Institutes of Health for their invaluable effort to build and maintain the database for all our array studies. Review of the manuscript and helpful suggestions of Dr. Louis Staudt of the National Cancer Institute and Dr. James F. Battey of the National Institute on Deafness and Other Communication Disorders were appreciated.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • ↵1 Supported by National Institute on Deafness and Other Communication Disorders Intramural Project DC-00016.

  • ↵2 To whom requests for reprints should be addressed, at Head and Neck Surgery Branch, National Institute on Deafness and Other Communication Disorders/NIH, Building 10, Room 5D55, Bethesda, MD 20892.

  • ↵3 The abbreviations used are: SCC, squamous cell carcinomas; NF-κB, nuclear factor-κB; AP-1, activator protein-1; DD, differential display; Gro-1, chemokine growth-regulated oncogene 1; cdk, cyclin-dependent kinase; OPN, osteopontin; FasL, Fas ligand; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor; PDGF, platelet-derived growth factor; HNSCC, head and neck squamous cell carcinoma; CSF, colony-stimulating factor; EST, expressed sequence tag; IL, interleukin.

  • ↵4 G. Dong, Z. Chen, Z. Y. Li, N. T. Yeh, C. C. Bancroft, and C. Van Waes. Hepatocyte growth factor/scatter factor-induced activation of MEK and PI3K pathways contributes to expression of proangiogenic cytokines IL-8, and VEGF in head, and neck squamous cell carcinoma, submitted for publication.

  • ↵5 C. C. Bancroft, U. Yeh, Z. Chen, U. B. Sunwoo, G. Dong, C. Park, N. Yeh, S. Jackson, and C. Van Waes. Epidermal growth factor receptor coactivates the NF-κB and AP-1 signal pathways and expression of angiogenesis factors IL-8 and VEGF in human head and neck squamous cell carcinoma lines, manuscript submitted.

  • ↵6 Z. Chen, unpublished data.

  • Received January 23, 2001.
  • Accepted April 12, 2001.

References

  1. ↵
    Liotta L. A., Steeg P. S., Stetler-Stevenson W. G. Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation. Cell, 64: 327-336, 1991.
    OpenUrlCrossRefPubMed
  2. ↵
    Grandis J. R., Melhem M. F., Gooding W. E., Day R., Holst V. A., Wagener M. M., Drenning S. D., Tweardy D. J. Levels of TGF-α and EGFR protein in head and neck squamous cell carcinoma and patient survival. J. Natl. Cancer Inst., 90: 824-832, 1998.
    OpenUrlFREE Full Text
  3. ↵
    Van Waes C., Kozarsky K. F., Warren A. B., Kidd L., Paugh D., Liebert M., Carey T. E. The A9 antigen associated with aggressive human squamous carcinoma is structurally and functionally similar to the newly defined integrin α6 β4. Cancer Res., 51: 2395-2402, 1991.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Van Waes C., Surh D. M., Chen Z., Rhim J. S., Brager R., Roy S. B., Kashima H., Wolf G. T., Carey T. E. Increase in suprabasilar integrin adhesion molecule expression in human epidermal neoplasms accompanies increased proliferation occurring with immortalization and tumor progression. Cancer Res., 55: 5434-5444, 1995.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Andreasen P. A., Egelund R., Petersen H. H. The plasminogen activation system in tumor growth, invasion, and metastasis. Cell. Mol. Life Sci., 57: 25-40, 2000.
    OpenUrlCrossRefPubMed
  6. ↵
    Chen Z., Colon I., Ortiz N., Callister M., Dong G., Pegram M. Y., Arosarena O., Strome S., Van Waes C. Effects of IL-1α, IL-1RA and neutralizing antibody on proinflammatory cytokine expression by human squamous cell carcinoma lines. Cancer Res., 58: 3668-3676, 1998.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Chen Z., Malhotra P. S., Thomas G. R., Ondrey F. G., Duffey D. C., Enamorada I., Mousa S., Van Waes C. Expression of proinflammatory and proangiogenic cytokines in human head and neck patients. Clin. Cancer Res., 5: 1369-1379, 1999.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Ondrey F., Sunwoo J., Dong G., Chen Z., Bancroft C., Van Waes C. Constitutive expression of proinflammatory cytokines and survival in head and neck squamous cell carcinoma cell lines. Mol. Carcinog., 26: 119-129, 1999.
    OpenUrlCrossRefPubMed
  9. ↵
    Duffey D., Chen Z., Dong G., Ondrey F. G., Wolf J., Brown K., Siebenlist U., Van Waes C. Expression of a dominant negative inhibitor factor-κBα in a human head and neck squamous cell carcinoma inhibits NF-κB, proinflammatory cytokine expression and tumor growth in vivo. Cancer Res., 59: 3468-3474, 1999.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Song J. I., Grandis J. R. Stat-3 STAT signaling in head and neck cancer. Oncogene, 19: 2489-2495, 2000.
    OpenUrlCrossRefPubMed
  11. ↵
    Barkett M., Gilmore T. D. Control of apoptosis by Rel/NF-κB transcription factors. Oncogene, 18: 6910-6924, 1999.
    OpenUrlCrossRefPubMed
  12. ↵
    Mayo M. W., Baldwin A. S. The transcription factor NF-κB: control of oncogenesis and cancer therapy resistance. Biochim. Biophys. Acta., 1470: M55-M62, 2000.
    OpenUrlPubMed
  13. ↵
    Rayet B., Gelinas C. Aberrant rel/nfkb genes and activity in human cancer. Oncogene, 18: 6938-6947, 1999.
    OpenUrlCrossRefPubMed
  14. ↵
    Li J-J., Westergaard C., Ghosh P., Colburn N., H. Inhibition of both nuclear factor-κB and activator protein-1 activation block the neoplastic transformation response. Cancer Res., 57: 3569-3576, 1997.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Li J-J., Rhim J. S., Schlegel R., Vousden K. H., Colburn N. H. Expression of dominant negative Jun inhibits elevated AP-1 and NF-κB transactivation and suppresses anchorage independent growth of HPV immortalized human keratinocytes. Oncogene, 16: 2711-2722, 1998.
    OpenUrlCrossRefPubMed
  16. ↵
    Grandis J. R., Drenning S. D., Zeng Q., Watkins S. C., Melhem M. F., Endo S., Johnson D. E., Huang L., He Y., Kim J. D. Constitutive activation of Stat3 signaling abrogates apoptosis in squamous cell carcinogenesis in vivo. Proc. Natl. Acad. Sci. USA, 97: 4227-4232, 2000.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Yuspa S. H., Hawley-Nelson P., Koehler B., Stanley J. R. A survey of transformation markers in differentiating epidermal cell lines. Cancer Res., 40: 4694-4703, 1980.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Chen Z., Smith C. W., Kiel D., Van Waes C. Metastatic variants derived following in vivo tumor progression of an in vitro transformed squamous cell carcinoma line acquire a differential growth advantage requiring tumor-host interaction. Clin. Exp. Metastasis, 15: 527-535, 1997.
    OpenUrlCrossRefPubMed
  19. ↵
    Dong G., Loukinova E., Smith C. W., Chen Z., Van Waes C. Genes differentially expressed with malignant transformation and metastatic tumor progression of murine squamous cell carcinoma. J. Cell. Biochem. Suppl., 28/29: 90-100, 1997.
  20. ↵
    Smith C. W., Chen Z., Dong G., Loukinova E., Pegram M. Y., Nicholas-Figueroa L., Van Waes C. The host environment promotes the development of primary and metastatic squamous cell carcinoma that constitutively express proinflammatory cytokines, IL-1α, IL-6, GM-CSF and KC. Clin. Exp. Metastasis, 16: 655-664, 1998.
    OpenUrlCrossRefPubMed
  21. ↵
    Dong G., Chen Z., Kato T., Van Waes C. The host environment promotes the constitutive activation of NF-κB and proinflammatory cytokine expression during metastatic tumor progression of murine squamous cell carcinoma. Cancer Res., 59: 3954-3504, 1999.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Chen Y., Dougherty E. R., Bittner M. L. Ratio-based decisions and the quantitative analysis of cDNA microarray images. J. Biomed. Optics, 2: 364-374, 1997.
    OpenUrlCrossRef
  23. ↵
    Eisen M. B., Spellman P. T., Brown P. O., Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. USA, 95: 14863-14868, 1998.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Oquendo P., Alberta J., Wen D., Graycar J. L., Derynck R., Stiles C. D. The platelet-derived growth factor-inducible KC gene encodes a secretory protein related to platelet α-granule proteins. J. Biol. Chem., 264: 4133-4137, 1989.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Loukinova E., Dong G., Chen Z., Sunwoo J., Van Waes C. Constitutive activation of a C-X-C chemokine KC promotes metastatic potential in murine squamous cell carcinoma. Oncogene, 19: 3477-3486, 2000.
    OpenUrlCrossRefPubMed
  26. ↵
    Zheng P., Guo Y., Niu Q., Levy D. E., Dyck J. A., Lu S., Sheiman L. A., Liu Y. Proto-oncogene PML controls genes devoted to MHC class I antigen presentation. Nature (Lond.), 396: 373-376, 1998.
    OpenUrlCrossRefPubMed
  27. ↵
    Eshel R., Zanin A., Sagi-Assif O., Meshel T., Smorodinsky N. I., Dwir O., Alon R., Brakenhoff R., van Dongen G., Witz I. P. The GPI-linked Ly-6 antigen E48 regulates expression levels of the FX enzyme and of E-selectin ligands on head and neck squamous carcinoma cells. J. Biol. Chem., 275: 12833-12840, 2000.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Jeffers M., Rong S., Vande Woude G. F. Hepatocyte growth factor/scatter factor-Met signaling in tumorigenicity and invasion/metastasis. J. Mol. Med., 74: 505-513, 1996.
    OpenUrlCrossRefPubMed
  29. ↵
    Wood L. D., Farmer A. A., Richmond A. HMGI(Y) and sp1 in addition to NF-κB regulate transcription of the MGSA/GRO α gene. Nucleic Acids Res., 23: 4210-4219, 1995.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Bos J. L. All in the family? New insights and questions regarding interconnectivity of Ras, Raps1 and Ral. EMBO J., 17: 6776-6782, 1998.
    OpenUrlAbstract
  31. ↵
    Sudol M., Bork P., Einbond A., Kastury K., Druck T., Negrini M., Huebner K., Lehman D. Characterization of the mammalian YAP (Yes-associated protein) gene and its role in defining a novel protein module, the WW domain. J. Biol. Chem., 270: 14733-14741, 1995.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Smith C. M., Steitz J. A. Classification of gas5 as a multi-small-nucleolar-RNA (snoRNA) host gene and a member of the 5′-terminal oligopyrimidine gene family reveals common features of snoRNA host genes. Mol. Cell. Biol., 18: 6897-6909, 1998.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Read M. A., Brownell J. E., Gladysheva T. B., Hottelet M., Parent L. A., Coggins M. B., Pierce J. W., Podust V. N., Luo R. S., Chau V., Palombella V. J. Nedd8 modification of cul-1 activates SCF[β(TrCP)]-dependent ubiquitination of IκBα. Mol. Cell. Biol., 20: 2326-2333, 2000.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Vazquez F., Hastings G., Ortega M., Lane T. F., Oikemus S., Lombardo M., Iruela-Arispe M. L. METH-1 a human ortholog of ADAMTS-1, and METH-2 are members of a new family of proteins with angio-inhibitory activity. J. Biol. Chem., 274: 23349-23357, 1999.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Nishimura T., Newkirk K., Sessions R. B., Andrews P. A., Trock B. J., Rasmussen A. A., Montgomery E. A., Bischoff E. K., Cullen K. J. Immunohistochemical staining for glutathione S-transferase predicts response to platinum-based chemotherapy in head and neck cancer. Clin. Cancer Res., 2: 1859-1865, 1996.
    OpenUrlAbstract
  36. ↵
    Hart M. C., Korshunova Y. O., Cooper J. A. Mapping of the mouse actin capping protein α subunit genes and pseudogenes. Genomics, 39: 264-270, 1997.
    OpenUrlCrossRefPubMed
  37. ↵
    Sherr C. J. The Pezcoller lecture: cancer cell cycles revisited. Cancer Res., 60: 3689-3695, 2000.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Sandhu C., Slingerland J. Deregulation of the cell cycle in cancer. Cancer Detect. Prev., 24: 107-118, 2000.
    OpenUrlPubMed
  39. ↵
    Boulaire J., Fotedar A., Fotedar R. The functions of the cdk-cyclin kinase inhibitor p21WAF1. Pathol. Biol. (Paris), 48: 190-202, 2000.
    OpenUrlPubMed
  40. ↵
    Lloyd R. V., Erickon L. A., Jin L., Kulig E., Qian X., Cheville J. C., Scheithauer B. W. P27kip1: a multifunctional cyclin-dependent kinase inhibitor with prognostic significance in human cancers. Am. J. Pathol., 154: 313-323, 1999.
    OpenUrlCrossRefPubMed
  41. ↵
    Parsons R. Phospahatases and tumorigenesis. Curr. Opin. Oncol., 10: 88-91, 1998.
    OpenUrlCrossRefPubMed
  42. ↵
    Zornig M., Evan G. I. Cell cycle: on target with Myc. Curr. Biol., 6: 1553-1556, 1996.
    OpenUrlCrossRefPubMed
  43. ↵
    Bossone S. A., Asselin C., Patel A. J., Marku K. B. MAZ, a zinc finger protein binds to c-Myc and C2 gene sequences regulating transcriptional initiation and termination. Proc. Natl. Acad. Sci. USA, 89: 7452-7456,
  44. ↵
    Weber-Nordt R. M., Mertelsmann R., Finke J. The JAK-STAT pathway: signal transduction involved in proliferation, differentiation and transformation. Leuk. Lymphoma, 28: 459-467, 1998.
    OpenUrlPubMed
  45. ↵
    Balasubramian S., Ahmad N., Kukhtar H. Upregulation of E2F transcription factors in chemically induced mouse skin tumors. Int. J. Oncol., 15: 387-390,
  46. ↵
    Coopman P. J., Do M. T., Barth M., Bowden E. T., Hayes A. J., Basyuk E., Blancato J. K., Vezza P. R., McKlesky S. W., Mangeat P. H., Mueller S. C. The syk tyrosine kinase suppresses malignant growth of human breast cancer cells. Nature (Lond.), 406: 742-747, 2000.
    OpenUrlCrossRefPubMed
  47. ↵
    Thomas G. R., Tubb E. E., Sessions R. B., Cullen K. J. Growth factor mRNA expression in human head and neck malignant tumors. Arch. Otolaryngol. Head Neck Surg., 119: 1247-1252, 1993.
  48. ↵
    Gastman B. R., Atarashi Y., Reichert T. E., Saito T., Balkir L., Rabinowich H., Whiteside T. L. Fas ligand is expressed on human squamous cell carcinomas of the head and neck, and it promotes apoptosis of T lymphocytes. Cancer Res., 59: 5356-5364, 1999.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    Moon M. R., Parikh A. A., Pritts T. A., Fischer J. E., Cottongim S., Szabo C., Salzman A. L., Hasselgren P. O. Complement component C3 production in IL-1β-stimulated human intestinal epithelial cells is blocked by NF-κB inhibitors and by transfection with ser 32/36 mutant IκBα. J. Surg. Res., 82: 48-55, 1999.
    OpenUrlCrossRefPubMed
  50. ↵
    Du C., Sriram S. Induction of interleukin-12/p40 by superantigens in macrophages is mediated by activation of nuclear factor-κB. Cell. Immunol., 199: 50-57, 2000.
    OpenUrlCrossRefPubMed
  51. ↵
    Suzuki M., Tamura N., Kobayashi H., Ohwada M., Terao T., Sato I. Clinical significance of combined use of macrophage colony-stimulating factor and squamous cell carcinoma antigen as a selective diagnostic marker for squamous cell carcinoma arising in mature cystic teratoma of the ovary. Gynecol. Oncol., 77: 405-409, 2000.
    OpenUrlCrossRefPubMed
  52. ↵
    Sapi E., Flick M. B., Tartaro K., Kim S., Rakhlin Y., Rodov S., Kacinski B. M. Effect of all-trans-retinoic acid on c-fms proto-oncogene [colony-stimulating factor 1 (CSF-1) receptor] expression and CSF-1-induced invasion and anchorage-independent growth of human breast carcinoma cells. Cancer Res., 59: 5578-5585, 1999.
    OpenUrlAbstract/FREE Full Text
  53. ↵
    Crawford H. C., Matrisian L. M., Liaw L. Distinct roles of osteopontin in host defense activity and tumor survival during squamous cell carcinoma progression in vivo. Cancer Res., 58: 5206-5215, 1998.
    OpenUrlAbstract/FREE Full Text
  54. ↵
    Tuck A. B., Arsenault D. M., O’Malley F. P., Hota C., Ling M. C., Wilson S. M., Chambers A. F. Osteopontin induces increased invasiveness and plasminogen activator expression of human mammary epithelial cells. Oncogene, 18: 4237-4246, 1999.
    OpenUrlCrossRefPubMed
  55. ↵
    O’Regan A. W., Hayden J. M., Berman J. S. Osteopontin augments CD3-mediated interferon-γ and CD40 ligand expression by T cells, which results in IL-12 production from peripheral blood mononuclear cells. J. Leukoc. Biol., 68: 495-502, 2000.
    OpenUrlAbstract/FREE Full Text
  56. ↵
    Marshall D. D., Kornberg L. J. Overexpression of scatter factor and its receptor (c-met) in oral squamous cell carcinoma. Laryngoscope, 108: 1413-1417, 1998.
    OpenUrlCrossRefPubMed
  57. ↵
    Klein R., Nanduri V., Jing S. A., Lamballe F., Tapley P., Bryant S., Cordon-Cardo C., Jones K. R., Reichardt L. F., Barbacid M. The trkB tyrosine protein kinase is a receptor for brain-derived neurotrophic factor and neurotrophin-3. Cell, 66: 395-400, 1991.
    OpenUrlCrossRefPubMed
  58. ↵
    Walter J., Newport J. Initiation of eukaryotic DNA replication: origin unwinding and sequential chromatin association of Cdc45, RPA, and DNA polymerase α. Mol. Cell, 5: 617-627, 2000.
    OpenUrlCrossRefPubMed
  59. ↵
    Roy N., Deveraux Q. L., Takahashi R., Salvesen G. S., Reed J. C. The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J., 16: 6914-6925, 1997.
    OpenUrlAbstract
  60. ↵
    Yang Y., Fang S., Jensen J. P., Weissman A. M., Ashwell J. D. Ubiquitin protein ligase activity of IAPs and their degradation in proteasomes in response to apoptotic stimuli. Science (Wash. DC), 288: 874-877, 2000.
    OpenUrlAbstract/FREE Full Text
  61. ↵
    Harwood F. G., Kasibhatla S., Petak I., Vernes R., Green D. R., Houghton J. A. Regulation of FasL by NF-κB and AP-1 in Fas-dependent thymineless death of human colon carcinoma cells. J. Biol. Chem., 275: 10023-10029, 2000.
    OpenUrlAbstract/FREE Full Text
  62. ↵
    Condorelli G., Vigliotta G., Cafieri A., Trencia A., Andalo P., Oriente F., Miele C., Caruso M., Formisano P., Beguinot F. PED/PEA-15: an anti-apoptotic molecule that regulates FAS/TNFR-1 induced apoptosis. Oncogene, 18: 4409-4415, 1999.
    OpenUrlCrossRefPubMed
  63. ↵
    Scatena M., Almeida M., Chaisson M. L., Fausto N., Nicosia R. F., Giachelli C. M. NF-κB mediated αvβ3 integrin-induced endothelial cell survival. J. Cell Biol., 141: 1083-1093, 1998.
    OpenUrlAbstract/FREE Full Text
  64. ↵
    Shijubo N., Uede T., Kon S., Maeda M., Segawa T., Imada A., Hirasawa M., Abe S. Vascular endothelial growth factor and osteopontin in stage I lung adenocarcinoma. Am J. Respir. Crit. Care Med., 160: 1269-1273, 1999.
    OpenUrlCrossRefPubMed
  65. ↵
    Perez M., Haschke B., Donato N. J. Differential expression and translocation of protein tyrosine phophatase 1B-related proteins in ME-180 tumor cells expressing apoptotic sensitivity and resistance to tumor necrosis factor: potential interaction with epidermal growth factor receptor. Oncogene, 18: 967-978, 1999.
    OpenUrlCrossRefPubMed
  66. ↵
    Chen Z. J., Parent L., Maniatis T. Site-specific phosphorylation of IκBα by a novel ubiqutination-dependent protein kinase activity. Cell, 84: 853-862, 1996.
    OpenUrlCrossRefPubMed
  67. ↵
    Kornitzer D., Ciechanover A. Modes of regulation of ubiquitin-mediated protein degradation. J. Cell. Physiol., 182: 1-11, 2000.
    OpenUrlCrossRefPubMed
  68. ↵
    Sunwoo J. B., Chen Z., Dong G., Yeh N., Bancroft C. C., Sausville E., Adams J., Van Waes C. Novel proteasome inhibitor PS-341 inhibits activation of NF-κB, cell survival, tumor growth and angiogenesis in squamous cell carcinoma. Clin. Cancer Res., 7: 1419-1428, 2001.
    OpenUrlAbstract/FREE Full Text
  69. ↵
    Adams M. D., Soares M. B., Kerlavage A. R., Fields C., Venter J. C. Rapid cDNA sequencing (expressed sequence tags) from a directionally cloned human infant brain cDNA library. Nat. Genet., 4: 373-380, 1993.
    OpenUrlCrossRefPubMed
  70. ↵
    Jo H., Zhang H., Zhang R., Liang P. Cloning oncogenic ras-regulated genes by differential display. Methods, 16: 365-372, 1998.
    OpenUrlCrossRefPubMed
  71. ↵
    Leethanakul C., Patel V., Gillespie J., Pallente M., Ensley J. F., Koontongkaew S., Liotta L. A., Emmert-Buck M., Gutkind J. S. Distinct pattern of expression of differentiation and growth-related genes in squamous cell carcinomas of the head and neck revealed by the use of laser capture microdissection and cDNA arrays. Oncogene, 19: 3220-3224, 2000.
    OpenUrlCrossRefPubMed
  72. ↵
    Shillitoe E. J., May M., Patel V., Lethanakul C., Ensley J. F., Strausberg R. L., Gutkind J. S. Gene expression profiles in squamous cell carcinomas of the oral cavity: use of laser capture microdissection for the construction and analysis of stage specific cDNA libraries. Oral Oncol., 36: 474-483, 2000.
    OpenUrlCrossRefPubMed
  73. ↵
    Clark E. A., Golub T. R., Lander E. S., Hynes R. O. Genomic analysis of metastasis reveals an essential role for RhoC. Nature (Lond.), 406: 532-535, 2000.
    OpenUrlCrossRefPubMed
  74. ↵
    Whitby F. G., Xia G., Pickart C. M., Hill C. P. Crystal structure of the human ubiquitin-like protein NEDD8 and interactions with ubiquitin pathway enzymes. J. Biol. Chem., 273: 34983-34991, 1998.
    OpenUrlAbstract/FREE Full Text
  75. ↵
    Maekawa K., Imagawa N., Naito A., Harada S., Yoshie O., Takagi S. Association of protein-tyrosine phosphatase PTP-BAS with the transcription-factor-inhibitory protein IκBα through interaction between the PDZ1 domain and ankyrin repeats. Biochem. J., 337: 179-184, 1999.
  76. ↵
    Shang C., Attema J., Cakouros D., Cockerill P. N., Shannon M. F. Nuclear factor of activated T cells contributes to the function of the CD28 response region of the granulocyte macrophage-colony stimulating factor promoter. Int. Immunol., 11: 1945-1956, 1999.
    OpenUrlAbstract/FREE Full Text
  77. ↵
    Cortesina G., Martone T., Galeazzi E., Olivero M., De Stefani A., Bussi M., Valente G., Comoglio P. M., Di Renzo M. F. Staging of head and neck squamous cell carcinoma using the MET oncogene product as marker of tumor cells in lymph node metastases. Int. J. Cancer, 89: 286-292, 2000.
    OpenUrlCrossRefPubMed
  78. ↵
    Di Renzo M. F., Olivero M., Martone T., Maffe A., Maggiora P., Stefani A. D., Valente G., Giordano S., Cortesina G., Comoglio P. M. Somatic mutations of the MET oncogene are selected during metastatic spread of human HNSC carcinomas. Oncogene, 19: 1547-1555, 2000.
    OpenUrlCrossRefPubMed
  79. ↵
    Huang S. M., Bock J. M., Harari P. M. Epidermal growth factor receptor blockade with C225 modelates proliferation, apoptosis, and radiosensitivity in squamous cell carcinomas of the head and neck. Cancer Res., 59: 1935-1940, 1999.
    OpenUrlAbstract/FREE Full Text
  80. ↵
    Stiles C. D. The biological role of oncogenes—insights from platelet-derived growth factor. Cancer Res., 45: 5215-5218, 1985.
    OpenUrlAbstract/FREE Full Text
  81. ↵
    Heldin C. H., Betsholtz C., Johnsson A., Nister M., Ek B., Ronnstrand L., Wasteson A., Westermark B. Platelet-derived growth factor: mechanism of action and relation to oncogenes. J. Cell Sci. Suppl., 3: 65-76, 1985.
    OpenUrlPubMed
  82. ↵
    Melillo R. M., Barone M. V., Lupoli G., Cirafici A. M., Carlomagno F., Visconti R., Matoskova B., Di Fiore P. P., Vecchio G., Fusco A., Santoro M. Ret-mediated mitogenesis requires Src kinase activity. Cancer Res., 59: 1120-1126, 1999.
    OpenUrlAbstract/FREE Full Text
  83. ↵
    Kuno K., Kanada N., Nakashima E., Fujiki F., Ichimura F., Matsushima K. Molecular cloning of a gene encoding a new type of metalloproteinase-disintegrin family protein with thrombospondin motifs as an inflammation associated gene. J. Biol. Chem., 272: 556-562, 1997.
    OpenUrlAbstract/FREE Full Text
  84. ↵
    Kuno K., Iizasa H., Ohno S., Matsushima K. The exon/intron organization and chromosomal mapping of the mouse ADAMTS-1 gene encoding an ADAM family protein with TSP motifs. Genomics., 46: 466-471, 1997.
    OpenUrlCrossRefPubMed
  85. ↵
    Kuno K., Terashima Y., Matsushima K. ADAMTS-1 is an active metalloproteinase associated with the extracellular matrix. J. Biol. Chem., 274: 18821-18826, 1999.
    OpenUrlAbstract/FREE Full Text
  86. ↵
    Duffey D., Crowl Bancroft C. V., Chen Z., Ondrey F., Nejhad-Sattari M., Dong G., Van Waes C. Inhibition of transcription factor nuclear factor-κB by a mutant inhibitor-κBα attenuates resistance of human head and neck squamous cell carcinoma to TNF-α caspase-mediated cell death. Br. J. Cancer, 83: 1367-1374, 2000.
    OpenUrlCrossRefPubMed
  87. ↵
    Kato T., Duffey D., Ondrey F. G., Dong G., Chen Z., Cook J. A., Mitchell J. B., Van Waes C. Cisplatin and radiation sensitivity are independently modulated by glutathione and transcription factor nuclear factor-κB in head and neck squamous carcinomas. Head Neck, 22: 748-759, 2000.
    OpenUrlCrossRefPubMed
View Abstract
Back to top
Cancer Research: 61 (12)
June 2001
Volume 61, Issue 12

Sign up for alerts

View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
Molecular Profiling of Transformed and Metastatic Murine Squamous Carcinoma Cells by Differential Display and cDNA Microarray Reveals Altered Expression of Multiple Genes Related to Growth, Apoptosis, Angiogenesis, and the NF-κB Signal Pathway
Gang Dong, Elena Loukinova, Zhong Chen, Lisa Gangi, Tatyana I. Chanturita, Edison T. Liu and Carter Van Waes
Cancer Res June 15 2001 (61) (12) 4797-4808;
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement