Molecular Cancer Research Funding CR Podcast
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
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Spence, J.
Right arrow Articles by Mercola, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Spence, J.
Right arrow Articles by Mercola, D.
Molecular Cancer Research 4:47-60 (2006)
© 2006 American Association for Cancer Research

Signaling and Regulation

Messenger RNAs under Differential Translational Control in Ki-ras–Transformed Cells

Jean Spence1, Brendan M. Duggan2, Colleen Eckhardt1,2, Michael McClelland1,3 and Dan Mercola1,4

1 Sidney Kimmel Cancer Center, San Diego, California; 2 BioMedical Genomics Microarray Facility, Department of Medicine and 3 The Moores Cancer Center, University of California at San Diego, La Jolla, California; and 4 Pathology and Laboratory Medicine Department, University of California at Irvine, Irvine, California

Requests for reprints: Jean L. Spence, Omnitron Biosciences, 3090 Admiral Avenue, San Diego, CA 92123. Phone: 858-576-0877. E-mail: jlspence{at}san.rr.com


   Abstract
Top
 Notes
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Microarrays have been used extensively to identify differential gene expression at the level of transcriptional control in oncogenesis. However, increasing evidence indicates that changes in translational control are critical to oncogenic transformation. This study identifies mRNA transcripts that are differentially regulated, primarily at the level of translation, in the immortalized human embryonic prostate epithelial cell line 267B1 and the v-Ki-ras–transformed counterpart by comparing total mRNA to polysome-bound mRNA by using Affymetrix oligonucleotide microarrays. Among the transcripts that were identified were those encoding proteins involved in DNA replication, cell cycle control, cell-to-cell interactions, electron transport, G protein signaling, and translation. Many of these proteins are known to contribute to oncogenesis or have the potential to contribute to oncogenesis. Differential expression of RNA-binding proteins and the presence of highly conserved motifs in the 5' and 3' untranslated regions of the mRNAs are consistent with multiple pathways and mechanisms governing the changes in translational control. Although Alu sequences were found to be associated with increased translation in transformed cells, an evolutionarily conserved motif was identified in the 3' untranslated regions of ephrinB1, calreticulin, integrin{alpha}3, and mucin3B that was associated with decreased polysome association in 267B1/Ki-ras. (Mol Cancer Res 2006;4(1):47–60)


   Introduction
Top
 Notes
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Prostate cancer is the leading form of cancer among men. Mutations in ras occur in 5% to 30% of prostate cancer cases (1) with the incidence being higher in Japanese men (2, 3). Mutations in the tumor suppressor PTEN are genetically linked to the incidence of prostate cancer and result in activation of major downstream effectors of the ras pathway (4), further increasing the number of cases where ras is involved. Activation of the ras pathway and its downstream effectors, the mitogen activated protein kinase (MAPK) pathway, and AKT, are important factors in the late-stage or androgen-independent prostate cancer, which is incurable (5, 6). Ras controls three major signaling pathways: the RalGEF pathway, the Raf/MEK/ERK pathway, and the phosphatidylinositol 3-kinase pathway (7). Of these pathways, the phosphatidylinositol 3-kinase pathway targets the translational machinery by increasing cap-dependent translation and the translation of mRNAs encoding the translational machinery (reviewed in ref. 8). The downstream effectors of the phosphatidylinositol 3-kinase pathway that regulate translation include p70S6 kinase, which is activated, and 4E-BP1, which is inactivated by phosphorylation (9). Activation of p70S6 kinase by phosphorylation results in the phosphorylation of ribosomal protein S6 and increased translation of mRNA transcripts that encode ribosomal proteins, translation factors, and other proteins involved in translation (10). Phosphorylation of 4E-BP1 inhibits its binding to the cap-binding protein, eIF4E (11). When 4E-BP1 binds to eIF4E, it physically prevents the binding of eIF4G and the formation of the cap recognition complex, which is required for the binding of the 40S ribosomal subunit to the mRNA (12). Inhibition of 4E-BP1 binding results in a global increase in translation as all nuclear-encoded mRNAs have an m7GpppN cap (where m is a methyl group and N is any nucleic acid; ref. 13).

In addition to 4E-BP1 and ribosomal protein S6, a number of other components of the translational machinery are involved in various cancers. Examples include eIF4E that has been shown to be oncogenic when overexpressed in cells with activated ras (14). Overexpression of the initiation factors eIF4G1, eIF-2a, eIF4A, eIF5A2, eIR3a, eIF3b, eIF3c, and eIF3e, and elongation factor EEF1A2 has been associated with various cancers (reviewed in refs. 15-17). In particular, the p40 subunit of eIF3 has been associated genetically with advanced-stage prostate cancer (18). In addition to components of the translational machinery, a growing list of RNA-binding proteins are misregulated in various cancers, including HuR (19), members of the heterogenous nuclear ribonucleoprotein family (hnRNP; ref. 20), and the signal transduction and activation of RNA family of K-homologous (KH) domain–containing RNA-binding proteins (21, 22). Furthermore, several RNA-binding proteins, including c-myc mRNA coding region instability determinant binding protein (CRD-BP), KH domain-containing protein overexpressed in cancer (KOC), and hnRNPD, induce tumors when overexpressed in mice (23-25).

The recent introduction of oligonucleotide microarrays has been exploited extensively to identify mRNA transcripts that are differentially expressed in various cancers. In the majority of publications, total mRNA is analyzed, which does not reflect the level of translation of a given transcript. However, experiments in yeast indicate that there is little correlation between mRNA abundance and protein levels (26). Microarray analysis of polysome-bound mRNAs has been used to identify internal ribosome entry site–containing mRNAs (27), mRNAs encoding membrane and secreted proteins (28), and mRNAs under translational control following cytokine stimulation (29) and rapamycin treatment (30). More recently, microarray studies have shown that translational control is an important contributor to the transformed phenotype. Using microarray analysis of polysome-associated and total mRNA, Rajasekhar et al. (31) showed that the immediate response to signaling by oncogenic Ras and Akt is the recruitment of specific mRNA species to polysomes. The effect of Ras and Akt on translation precedes induction of mRNA synthesis by transcription and includes an increase in translation of mRNAs encoding transcription factors, indicating that translational control is a primary effector of transformation by Ki-ras and AKT and that translational control precedes transcriptional control (31).

This article identifies mRNAs under differential translational control during the maintenance phase of oncogenic transformation by the v-Ki-ras oncogene, which is a model system for the human oncogenic Ki-ras mutation. The immortalized human embryonic prostate epithelial cell line 267B1 and the v-Ki-ras–transformed derivative, 267B1/Ki-ras, were used as the source of mRNA for comparison (32). The mRNAs that were identified by this protocol encode proteins that are known to contribute to the transformed phenotype and other proteins that may potentially contribute to the transformed phenotype by altering cell cycle control, cell adhesion, apoptosis, and metabolism. Moreover, genes under differential transcriptional but not translational control were identified that encode proteins involved in translational control. Transcripts encoding RNA-binding proteins that are induced in various cancers are increased and RNA-binding proteins and ribosomal proteins involved in apoptosis and p53 signaling are decreased in 267B1/Ki-ras. Finally, known motifs and unique evolutionarily conserved novel motifs have been identified in the 5' and 3' untranslated regions (UTRs) of the mRNAs under differential translational control.


   Results
Top
 Notes
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 Acknowledgements
 References
 
Experimental Design
The goal of these experiments was to identify mRNA transcripts that are differentially regulated by translational control in transformed cells. mRNAs were identified that were increased or decreased in the polysomes of either 267B1 or 267B1/Ki-ras but which remained constant in the total mRNA preparations and in the opposite polysome-associated mRNA preparation. 267B1 is immortalized by the SV40 T antigen and is nontumorigenic in nude mice and does not form colonies in soft agar (33). 267B1/Ki-ras, the v-Ki-ras–transformed derivative, aggressively forms tumors in nude mice and colonies in soft agar (32, 33). Briefly, mRNA was isolated from a total cell extract prepared from 267B1 and 267B1/Ki-ras. In parallel, mRNA associated with large polysomes was purified from these cell lines on sucrose gradients and used to prepare probes for analysis on microarray chips. The fractions of the sucrose gradient used to prepare polysome-associated mRNA are shown in Fig. 1A . Two independent experiments were done on the 267B1 and three on 267B1/Ki-ras total mRNAs and polysome-associated mRNAs. Between 43% and 51% of the oligonucleotide sites on the chip were recognized in the experiments. Probes prepared from mRNA isolated from v-Ki-ras–transformed cells recognized 4% to 5% more features on the chip than probes from the immortalized cells. It is possible that other factors, such as changes in the cytoskeleton, membrane-bound polysomes, and sequence-specific RNases, may have also contributed to the changes observed in this study. For simplicity of presentation, it will be assumed that the differences are due to differences in ribosome loading.


Figure 1
View larger version (52K):
[in this window]
[in a new window]
  		FIGURE 1. A. Purification of polysome-associated mRNA. Polysomes were separated by sucrose gradient. Top, UV profile recorded as the gradient was fractionated. Bottom, aliquots of the fractions were run on an agarose gel stained with ethidium bromide. The large polysomes (fractions 1, 2, and 3) or bottom third of the gradient were used to prepare mRNA for probe preparation and analysis on Affymetrix microarrays. These polysomes were isolated from 267B1 cells; similar results were obtained with 267B1/Ki-ras. B. Average lengths of mRNAs, and 5' and 3' UTRs. Columns: gray, 267B1 mRNAs under positive translational control; diagonal lines, 267B1 mRNAs under negative translational control; black, 267B1/Ki-ras mRNAs under positive translational control; white, 267B1/Ki-ras mRNAs under negative translational control.

 
The Affymetrix microarray results were statistically analyzed by the Data Mining Tool program (Affymetrix). Polysome-associated mRNAs of a given cell type were compared with total mRNAs and the alternate polysome-associated mRNA. mRNAs under differential translational control were selected as being low or high in a given polysome-associated preparation but similar in the chip results for total mRNA for 267B1 and 267B1/Ki-ras and the alternate polysome-associated mRNA. A probe list was constructed from the polysome-associated mRNA probes by applying Student's t test and selecting the probes with P ≤ 0.005. This probe list was intersected with a probe list of Affymetrix signals that were consistently high or low for polysome-associated mRNA in all comparisons. The final collection of probes with a P ≤ 0.005 and that were consistently high or low was then clustered by self-organizing maps with the Genesis program (34).

mRNAs Were Identified That Were Significantly Increased or Decreased in Either 267B1 or 267B1/Ki-ras Polysomes
Of the 5,907 transcripts present in 267B1/Ki-ras total mRNA preparations, 5,446 (92%) are present in the polysome-associated fractions. Of the 5,370 transcripts present in both chips of 267B1 total mRNA, 4,788 (89%) are present in both polysome-associated mRNA microarray results. Eighty-six percent (267B1/Ki-ras) and 84% (267B1) of the mRNA species found in the total mRNA preparations do not show a statistically different level in the polysome-bound RNA fractions. This suggests that ~15% to 25% of the mRNA species are regulated by translational control as determined by differential association with polysomes, which is consistent with observations of other investigators (35).

Four classes of mRNA under translational control were identified: mRNAs present but with decreased association with polysomes from 267B1 or from 267B1/Ki-ras, and mRNAs with increased association with polysomes from 267B1 or from 267B1/Ki-ras. The average lengths of the total mRNA, the 5' UTRs, and the 3' UTR were determined for each group (Fig. 1B). Notably, the lengths of the total mRNA and the 3' UTR of the transcripts under negative translational control (reduced polysome association) are approximately twice as long as those from the transcripts that showed an increase in the polysome-bound fractions. This is consistent with observations in yeast where longer mRNAs show a reduced loading with polysomes compared with shorter transcripts with more abbreviated 5' and 3' UTRs (36).

Transcripts under Positive Translational Control in 267B1/Ki-ras
Transcripts that are under positive translational control according to the expression profile in v-Ki-ras–transformed cells are shown in Fig. 2 . Of the mRNA transcripts that increase in the large polysome fractions in transformed cells, 9 of 41 of the transcripts are involved in DNA replication and/or cell cycle control. These include thymidylate synthetase (nucleotide synthesis); RFC2, POLB, and CDC45L (DNA replication); UbcH10, Ubc9, E2F5, and BTG3 (cell cycle control); POLB and Ubc9 (DNA repair); and MYST1 (chromatin remodeling). Strikingly, there are two ubiquitin-conjugation enzymes that are required for cell cycle progression, the cyclin-selective UbcH10 and the SUMO-specific Ubc9.


Figure 2
View larger version (47K):
[in this window]
[in a new window]
  		FIGURE 2. Profile of mRNA transcripts under differential positive translational control in 267B1/Ki-ras compared with the immortalized parent, 267B1. The color of each block represents the normalized relative expression of that gene in that microarray compared with the same gene in the other microarrays. Red, increased expression; green, reduced expression (according to the color bar at the top of the diagram); black, median values. The order of columns is as follows: 1, 267B1/Ki-ras total mRNA 1; 2, 267B1/Ki-ras total mRNA 2; 3, 267B1/Ki-ras total mRNA 3; 4, 267B1/Ki-ras polysome-associated mRNA 1; 5, 267B1/Ki-ras polysome-associated mRNA 2; 6, 267B1/Ki-ras polysome-associated mRNA 3; 7, 267B1 total mRNA 1; 8, 267B1 total mRNA 2; 9, 267B1 polysome-associated mRNA 1; 10, 267B1 polysome-associated mRNA 2. Right, accession numbers, probability as determined by Student's t test, gene names, and gene descriptions.

 
Of these transcripts, UbcH10, folypolyglutamyl synthetase, homeobox B5, homeobox B7, casein kinase 2, and D2LIC are commonly misregulated in cancer. Interestingly, several of the transcripts encode proteins that function in the same pathway. For example, folypolyglutamyl synthetase and thymidylate synthetase are both involved in folate synthesis and are targeted by the antifolate cancer drugs (37).

Transcripts under Negative Translational Control in v-Ki-ras–Transformed 267B1 Cells
Transcripts that are transcribed but that show reduced association with polysomes in v-Ki-ras–transformed 267B1 are shown in Fig. 3 . This list is notable because of the number of transcripts encoding proteins involved in cell adhesion. Prominent among these are integrin{alpha}3, integrinß5 (ITGB5), cadherinH, and cadherinEGF. Integrin{alpha}3 is down-regulated in a subset of human cancers, including lung and prostate cancer (38, 39), but it is increased in other cancers (40). Integrin expression has been reported to be regulated by the ras pathway (41, 42). Some of the transcripts encode proteins that are increased in transformed cells or tumors, including basigin and CD44. Just as transcripts encoding homeobox proteins HOXA5 and HOXB7 are increased in 267B1/Ki-ras polysomes, transcripts encoding the homeobox protein HOXB13 seem to be under negative translational control.


Figure 3
View larger version (44K):
[in this window]
[in a new window]
  		FIGURE 3. Profile of mRNA transcripts under differential negative translational control in 267B1/Ki-ras compared with the immortalized parent, 267B1, determined by statistical analysis and clustering. Red, increased expression; green, reduced expression (according to the color bar at the top of the diagram); black, median values. The order of columns is as follows: 1, 267B1/Ki-ras total mRNA 1; 2, 267B1/Ki-ras total mRNA 2; 3, 267B1/Ki-ras total mRNA 3; 4, 267B1/Ki-ras polysome-associated mRNA 1; 5, 267B1/Ki-ras polysome-associated mRNA 2; 6, 267B1/Ki-ras polysome-associated mRNA 3; 7, 267B1 total mRNA 1; 8, 267B1 total mRNA 2; 9, 267B1 polysome-associated mRNA 1; 10, 267B1 polysome-associated mRNA 2. Right, accession numbers, P values, gene names, and gene descriptions.

 
Transcripts under Differential Translational Control in 267B1
mRNAs that are either under differential increased or decreased translational control in 267B1 are presented in Fig. 4 . Transcripts that are under differential negative translational control in 267B1 are, therefore, differentially increased in 267B1/Ki-ras. Similar to the previous results, transcripts encoding proteins involved in cell cycle regulation (NEDD9, mTOR, and CDC2L5), chromatin homeostasis (SMARCC1), and DNA repair (PRKDC, mTOR) are under negative translational control in 267B1 and, therefore, increased in 267B1/Ki-ras. In addition, there are a number of transcripts encoding ras-related or G proteins, including RIT1, GNA11, and RAC2, which are differentially increased in 267B1 polysomes, and GNG7 and RAB28, which are differentially decreased in 267B1 polysomes. Many of these transcripts are commonly misregulated in cancer, including TOP2, PRK, RAB28, GNA11, and SULTA1 (43-45).


Figure 4
View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 4. Profile of mRNA transcripts under differential translational control in 267B1. Red, increased expression; green, reduced expression (according to the color bar at the top of the diagram). Transcripts under differential positive control are at the top and differential negative translational control are below. Black, median values. The order of columns is as follows: 1, 267B1/Ki-ras total mRNA 1; 2, 267B1/Ki-ras total mRNA 2; 3, 267B1/Ki-ras total mRNA 3; 4, 267B1/Ki-ras polysome-associated mRNA 1; 5, 267B1/Ki-ras polysome-associated mRNA 2; 6, 267B1/Ki-ras polysome-associated mRNA 3; 7, 267B1 total mRNA 1; 8, 267B1 total mRNA 2; 9, 267B1 polysome-associated mRNA 1; 10, 267B1 polysome-associated mRNA 2. Right, accession numbers, P values, gene names, and gene descriptions.

 
Quantitative PCR Confirmation of Affymetrix Results
A subset of the transcripts was chosen for confirmation of the results by real-time or quantitative PCR in independent experiments. Initially, we compared the relative amounts of a transcript that were associated with polysomes with the amount of transcript in the total cell lysate by adjusting the levels to the housekeeping mRNA glyceraldehyde-3-phosphate dehydrogenase (GAPDH). As shown in Fig. 5A , relatively lower amounts of the total mRNA encoding calreticulin, integrinß5, and integrin{alpha}3 were associated with polysomes in 267B1/Ki-ras–transformed cells than with the parental cell line (Fig. 5A, a, b, c). Conversely, there is a relative increase in the transcripts encoding Ubc9, MYST, and thymidylate synthetase in 267B1/Ki-ras–transformed cells (Fig. 5A, d, e, f).


Figure 5
View larger version (35K):
[in this window]
[in a new window]
  		FIGURE 5. A. Quantitative PCR confirmation of relative amounts of differentially translated mRNAs. mRNAs were prepared from total cell lysates and large polysomes as described in Materials and Methods. The quantitative PCR results were normalized against levels of GAPDH for each sample. Subsequently, the relative amount of mRNA represented in the large polysomes was determined by dividing the relative RNA units for polysome-associated mRNA by the mRNA units in the total cell lysate for each cell type. a. Calreticulin. b. ITGB5. c. Integrin{alpha}3. d. Thymidylate Synthetase. e. Ubc9. f. MYST. B. The relative fractions of mRNA species in the RNPs and small polysomes (top two thirds of the gradient) versus the large polysomes (bottom third of the gradient) were determined for 267B1 and 267B1/Ki-ras by quantitative PCR. The relative RNA units were determined and the results are plotted as percentage of total for 267B1 and 267B1/Ki-ras RNPs and small polysomes versus large polysomes. g. Ubc9. h. Thymidylate Synthetase. i. Bet1. j. UbcH10. k. Integrin{alpha}3. l. GAPDH. Results are the average of two independent experiments.

 
Differential changes in the level of association of an mRNA species with polysomes could have multiple causes, including retention in the ribonucleoprotein particles, changes in mRNA stability, or changes in intracellular localization resulting in differential loss during cell lysis and mRNA isolation. To determine if these mRNAs are differentially associated with ribosomes, the fraction of transcripts in the RNPs and small polysomes was determined and compared with the fraction in the large polysomes (four or more ribosomes). By this analysis, there was an increase in polysome-associated Ubc9, thymidylate synthetase, Bet1, and UbcH10 mRNAs in 267B1/Ki-ras cells and a corresponding increased presence of these transcripts in the small polysomes and RNPs in 267B1 (Fig. 5B, g, h, i, j). The pattern was reversed with integrin{alpha}3 mRNA, which was retained in 267B1/Ki-ras RNPs but associated with rapidly translating polysomes in 267B1 cells (Fig. 5B, k). By contrast, the mRNA for the housekeeping protein, GAPDH, was associated with large polysomes in both cell types (Fig. 5B, l). A possible complication of this experiment is that the mRNAs in the RNPs and small polysomes may be subject to degradation during the running of the sucrose gradient. Total mRNA is isolated using a much shorter procedure wherein chaotropic agents quickly inactivate RNases. However, this difference in isolation time will simply reduce the apparent proportion of transcripts in the RNPs and small ribosomes. Comparison of this fraction between the 267B1 and 267B1/Ki-ras cell lines will still identify differential association with ribosomes.

Transcripts under Differential Transcriptional, but not Translational, Control Encoding Proteins Affecting Translation
If translation is a contributing factor to oncogenesis by Ki-ras, transcripts encoding proteins that are involved in RNA metabolism or translation should be misregulated by transcription. To identify these transcripts, mRNAs were identified that were increased or decreased by at least 2-fold and had a P ≤ 0.005 for chance observation in a comparison of the combined total and polysome-bound experiments. This list of transcripts includes primarily RNA-binding proteins and RNA processing enzymes with biological functions that are relevant to cancer. RNAs encoding members of two families of RNA-binding proteins, the KH domain–containing RNA-binding proteins (RBPMS, KHDRBS3, and KOC) and the 2',5'-oligoadenylate synthetases, are increased in 267B1/Ki-ras. Overexpression of KOC and KHDRBS3 is associated with a variety of cancers so consistently that KOC is considered to be a tumor marker (21, 46, 47). These proteins control translation through binding to UTRs in the mRNA and are thought to function as intermediaries between signaling pathways and translation (22). OAS1 is an RNA-modifying enzyme that activates RNaseL through the synthesis of oligoadenylate in response to the presence of double-stranded mRNA, resulting in apoptosis (48). Mutations in RNaseL are genetically linked to prostate cancer (49) but the possible involvement of OAS1 in prostate cancer has not been reported previously.

The transcripts differentially decreased in 267B1/Ki-ras cells encode proteins involved in many functions, including apoptosis. Although GADD45B is not a ribosomal protein, it is included in this group because its homology to ribosomal protein L7Ae suggests that it may interact with components of the protein synthesis machinery. Both the ribosomal proteins death-associated protein 3 (DAP3) and RPS3A that are reduced in 267B1/Ki-ras are proapoptotic (50-52). Other transcripts that are transcribed at reduced levels in 267B1/Ki-ras encode TRAM1 and TRAM2. These proteins are required for translocation of secretory proteins into the endoplasmic reticulum through recognition of the signal sequence (53). The lower levels of these proteins may account for the reduction of some of the transcripts encoding secretory and membrane proteins in 267B1/Ki-ras. These results suggest that the changes in translational control that accompany oncogenic transformation facilitate the transformed phenotype by enhancing growth and inhibiting apoptosis.

Conserved Motifs in the 5' UTRs of mRNAs under Differential Translational Control
Translational control is usually mediated by sequences in the UTRs of mRNAs (54). Therefore, the 3' UTR sequences, from the stop codon to the poly(A) tails, and the 5' UTR sequences up to the AUG start codons were selected from the four groups of mRNAs under differential translational control. This list was expanded to include the top 85 candidates in each group with P ≤ 0.005. These sequences were scanned by the UTR website for known regulatory regions (55). Consistent with the literature, there was a 1.5-fold increase in the association of internal ribosome entry site–containing mRNAs and a 3-fold increase in the association of mRNAs with 5' TOP sequences with the polysomes from 267B1/Ki-ras.

In addition to the UTR website, the 5' and 3' UTRs were scanned for common motifs with multiple EM for motif enhancement (MEME; ref. 56). MEME detected highly conserved motifs in the 5' UTRs of a small number of mRNAs. The RNA motifs with the lowest probability for chance observation are presented in Fig. 6 . The first motif is shared between ANAX10 and ZNF266 and, to a lesser degree, with SKALP (Fig. 6A, motif 1). ANAX10 mRNA is increased in 267B1/Ki-ras polysomes (Fig. 3), and ZNF266 and SKALP are decreased in 267B1 polysomes and, therefore, differentially increased in 267B1/Ki-ras. This motif has 50% identity in ANAX10 and ZNF266. The 13 conserved residues in the 43 nucleotide motif were used to search the 5' and 3' UTR libraries derived from the genes represented on the Affymetrix chips. This search identified ANAX10, ZNF266, the genomic clone CTA-221G9, and the genomic clone encoding hypothetical protein FLJ13946; indicating this sequence is not widely represented. The second motif, which is shared between the 5' UTR of ITGB5 and the 5' UTR and the 5' coding sequences of CLPTM1, is highly GC rich, also has 50% homology and has the potential to form well-defined hairpins (Fig. 6A, motif 2). An upstream AUG in the 5' UTR of ITGB5 is three nucleotides upstream in the motif from the start codon of CLPTM1. If translated, it would encode a short peptide of 12 amino acids. A Motif Alignment and Search Tool search of a human expressed sequence tag library recognized the original UTRs but did not uncover additional homologues.


Figure 6
View larger version (98K):
[in this window]
[in a new window]
  		FIGURE 6. Motifs in the 5' and 3' UTRs of differentially translated mRNAs. The most statistically significant motifs in the UTRs were identified from computer analysis by MEME. A. Motif 1 is in the 5' UTR of ANAX10, ZNF266, and SKALP, which are differentially increased in 267B1/Ki-ras polysomes. Gray boxes, nucleotides that are shared between two mRNAs in the region shared between all three sequences; black boxes, common nucleotides in the longer motif shared by ANAX10 and ZNF266. Motif 2 is in the 5' UTR of ITGB5 and the 5' UTR and 5' coding sequence of CLPTM1. The start codons for the upstream open reading frame in ITGB5 and the coding sequence of CLPTM1 and the stop codon for the upstream open reading frame in ITGB5 are indicated. The multiple probabilities are a result of the way MEME calculates probabilities based on identity of all motifs, including those of low probability, that are related. B. Conservation of 3' UTR motif 3 in ephrinB1 (EFNB1; rat, mouse, and human), calreticulin (CALR; human, mouse, and cow), integrin{alpha}3 (ITGA3; human, mouse, and chicken), and mucin3B (human) across species. This portion of the 3' UTR of mucin3B is not present in the chimpanzee and canine transcripts deposited in the National Center for Biotechnology Information database. Identical bases in six or more transcripts are highlighted in black. Positions with only two different bases present in 9 of the 10 sequences or with 5 identical residues in one position are highlighted in gray. Motifs were identified by MEME and aligned by CLUSTALW. The motif extends for an additional 90 bases in the upstream direction in CALR and EFNB1. The probability for the motif in integrin{alpha}3, mucin3B, and calreticulin (all species) is e–22.34. The probability for the long motif conserved between ephrinB1 and calreticulin (all shown species) is e–47.81.

 
Conserved Motifs in the 3' UTRs of mRNAs under Differential Translational Control
Not surprisingly, the most pronounced motif in the 3' UTRs was that of Alu sequences. There is, however, a dramatic 6-fold increase in the number of long Alu sequences (80% or more of AluJ) in the 3' UTRs of mRNAs undergoing increased translation in 267B1/Ki-ras (12 mRNAs in the population of mRNAs undergoing increased translation in 267B1/Ki-ras compared with two mRNAs increased in 267B1 polysomes). Notably, two of the mRNAs under increased translational control, IL18 and RFC4, have very short 3' UTRs that are primarily composed of a single Alu sequence (P < 0.0001). An exhaustive search for mRNAs with 3' UTRs that are <500 bases and that contain at least an 80% complete Alu sequence in the "plus" orientation identified only two additional candidates: inactivation escape 1 (INE1) and eukaryotic initiation factor S2 or TGFß receptor-interacting protein 1 (eIFS2). INE1 is expressed at very low levels, whereas eIFS2 is increased in the polysomes from 267B1 and 267B1/Ki-ras in duplicate sites on the chip and across all 10 microarray experiments.

Notably, 36% of the 3' UTRs of the transcripts differentially decreased in 267B1/Ki-ras polysomes have two or more 15-LOX DICE sequences, almost twice as many as the other groups. DICE sequences mediate translational silencing by binding to hnRNPE1 and hnRNPK (57). At least two of the minimal LOX DICE motifs (CCCCPuCCCUCUU) must be present for translational silencing (57).

Forty-four percent of the 3' UTRs of the mRNAs that show increased association with polysomes in transformed cells do not share common, statistically significant motifs. In contrast, six mRNAs that are under differential negative translational control in 267B1/Ki-ras share multiple motifs resulting in a combined P value of <e–200. Multiple shared motifs were found in the 3' UTRs of seven genes: protein O-fucosyltransferase 2 (POFUT2), mucin3B (MUC3B), integrin{alpha}3 (ITGA3), ephrinB1 (EFNB1), calreticulin (CALR), the {alpha}-1 subunit of L-type calcium channel subunit (CACNAIC), and cadherin EGF (CELSR2). Calreticulin is included because it has shared motifs with the other mRNAs with low P values but does not have an overall P value <e–200 because of its smaller size. These mRNAs encode proteins involved with calcium homeostasis and cell surface to cytoplasm signaling. To define which sequences are most likely to be regulatory motifs, the 3' UTRs from human, mouse, and other species were obtained from the National Center for Biotechnology Information website and compared by MEME. Strong regions of homology were found over ~400 bases between the 3' UTRs of calreticulin and ephrinB1 (≥50% homology). A smaller overlapping region of homology was found in calreticulin (human, cow, and mouse), Mucin3B (human but not the shorter dog or chimpanzee genes), and integrin{alpha}3 (mouse, human, and chicken) with 39.9% homology. The overlapping regions from these motifs are shown in Fig. 6B.

A Motif Alignment and Search Tool search for homologues did not identify additional transcripts that shared common motifs from either the UTR library based on the genes represented on the U95Av microarray chips or a human expressed sequence tag library.


   Discussion
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 Acknowledgements
 References
 
To determine if translational control contributes significantly to oncogenic transformation by the v-Ki-ras oncogene, mRNAs that are candidates for translational control were identified by using Affymetrix microarrays to compare total mRNA to mRNA species associated with large polysomes between 267B1 and 267B1/Ki-ras. As in the study by Rajasekhar et al. (31), the majority of the genes identified by this protocol encode known proteins. Moreover, many of these proteins are known to contribute or have the potential to contribute to the transformed phenotype. Altered association of transcripts encoding homeobox transcription factors, proteins involved in DNA replication and repair, cell cycle control, proliferation, and cytoskeleton assembly were found in polysomes isolated from v-Ki-ras–transformed 267B1 cells. Perhaps the most striking collection of transcripts was the group that is differentially increased in polysomes isolated from 267B1/Ki-ras compared with polysomes from 267B1 cells. These transcripts encode proteins involved in DNA replication (CDC45L, replication factor C, subunit 2, thymidylate synthetase, and topoisomerase2A), cell cycle control (mTOR, UBCH10, and UBC9), DNA repair (POLB and PRKDC), and chromatin remodeling (MYST and SMARCC1). Transcripts down-regulated by translational control in 267B1/Ki-ras included DDB2 (the genetic locus for Xeroderma pigmentosum E), SP100, and ZWINT. Although this collection of the most significantly differentially regulated transcripts is relatively small, many of the proteins are targeted by anticancer drugs, including folylpolyglutamyl synthetase, which is targeted by the antifolate class of drugs (58); thymidylate synthetase, which is targeted by 5-fluorouracil (59); and mTOR, which is targeted by rapamycin (60). Notably, polymorphisms in the 3' and 5' UTRs of thymidylate synthetase are associated with increased risk of squamous cell carcinoma (61). In contrast to the mRNAs that are differentially increased in 267B1/Ki-ras polysomes compared with polysomes of 267B1 cells, the mRNAs that are differentially reduced encode a number of cell adhesion proteins, including integrinß5 and integrin{alpha}3 and cadherins H and EGF. Loss of contact inhibition, which is mediated by cadherins and integrins and the regulator, calreticulin, is considered to be a critical step for metastasis to occur. Integrinß3/ß5 knockout mice, which show increased tumorigenesis with enhanced angiogenesis, illustrate the importance of integrin expression in oncogenesis (62). Integrin signaling has been shown previously to be controlled by the ras pathway (41, 42) and misregulation of integrin expression contributes to the metastasis of prostate cancer to bone (63, 64). The results observed here suggest that specific differential translation may contribute to the misregulation of integrin expression in transformed cells.

To determine the possible mechanisms for the differential translation, transcripts that encode proteins involved in control of translation and that are regulated by transcription but not translation were identified (Table 1 ). There are four groups of transcripts: (a) mRNAs encoding KH domain proteins that are increased or decreased in v-Ki-ras–transformed cells; (b) transcripts encoding proteins involved in spermatogenesis, which are more abundantly translated in transformed cells; (c) transcripts encoding proteins involved in apoptosis, which are less abundantly translated in transformed cells; and (d) transcripts encoding proteins involved in viral replication, which are more abundantly translated in 267B1/Ki-ras. The most decisive correlation to cancer is in the increase in the KH-domain RNA-binding proteins, KOC and KHDRBS3. These proteins have SHC2 and SHC3 domains and are proposed to link signaling pathways to RNA metabolism (22). The overexpression of OAS1 as a consequence of ras transformation may be applicable to the etiology of prostate cancer, as oligoadenylate activates RNaseL, a familial prostate cancer gene (49, 65). Reduction in the transcripts encoding ribosomal protein S3a and the mitochondrial ribosomal protein DAP3 in 267B1/Ki-ras cells may contribute to oncogenesis, as these are both involved in apoptosis (50-52).


View this table:
[in this window]
[in a new window]
  		Table 1. mRNA Transcripts Encoding Proteins Involved in Translation that Are under Differential Transcriptional Control

 
The variety of conserved motifs in the UTRs of the differentially expressed mRNAs is consistent with the contribution of multiple mechanisms to the differences in translation. Consistent with published results, there was an increase in the association of 5' internal ribosome entry site–containing sequences with polysomes in v-Ki-ras–transformed 267B1 cells. Internal ribosome entry site sequences regulate the translation of apoptosis regulatory proteins, such as X-linked inhibitor of apoptosis (66) and proteins involved in cell cycle progression (67). There is also a ≥2-fold increase in the number of transcripts with two or more LOX-DICE sequences in transcripts under negative translational control in 267B1/Ki-ras. These sequences are involved in negative translational control and their very high incidence (over 40% of transcripts under negative translational control) suggests that this may be an important pathway in the changes in translation that accompany oncogenesis.

Regulatory sequences in the UTRs would be expected to coordinate the synthesis of interacting proteins. The three transcripts that have an Alu sequence comprising the majority of the 3' UTR (IL18, RFC2, and eIFS2) function in growth and cytokine signaling through the TGF-ß pathway (68, 69). A regulatory role has been defined for Alu sequences located in the 3' UTRs of the low-density lipoprotein receptor and MnSOD mRNAs (70, 71). More significantly, a mobilization of Alu-containing mRNAs to polysomes has been observed in kidney tumors and transformed cell lines (72).

The proteins encoded by the transcripts represented in Fig. 6B are involved in signaling from the extracellular matrix to the cytoplasm and in calcium homeostasis. Integrins and cadherins regulate cell growth through attachment to the extracellular matrix and misregulation of this signaling pathway is involved in the loss of contact inhibition during carcinogenesis and metastasis. Calreticulin and members of the ephrin family regulate signaling through cadherins and integrins (73, 74). Although not directly involved in integrin signaling, POFUT2 glycosylates membrane proteins by the addition of fucose, which can change the properties of membrane proteins and alter the behavior of cells (75). Interestingly, calreticulin has been found to physically interact with integrin{alpha}3 and subsequently activate matrix metalloprotease-2, thereby regulating cancer invasion and metastasis (76). This suggests a possible function for these motifs—regulation of timing and location of mRNA translation, thereby colocalizing interacting proteins.

Our results complement the results found by Rajeshakar et al. (31) in indicating an important role for translational control in v-Ki-ras–induced oncogenic transformation. However, our results with these established cell lines indicate that both transcriptional and translational control contribute to the transformed phenotype. The contribution of translational control to transformation by v-Ki-ras is highlighted by the known roles of the proteins encoded by these transcripts in oncogenic transformation and by the changes in transcriptional regulation of RNA metabolism proteins that are known to, or have the potential to, contribute to the transformed phenotype. The translational control may be mediated through highly conserved motifs in the 5' and 3' UTRs.


   Materials and Methods
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
Acknowledgements
 References
 
Cells and Cell Culture Reagents
The human embryonic prostate epithelial cell lines 267B1 and 267B1/Ki-ras were obtained from J. Rhim (32). Cells were maintained in RPMI with 10% FCS, 100 units of penicillin per milliliter, 0.1 mg of streptomycin per milliliter in a 37°C incubator with water saturation, and 5% CO2. Medium was prepared from premixed powder and supplemented with NaHCO3 and glutamine according to the specifications of the manufacturer.

Isolation of Polyribosomes and RNA
An adaptation of the protocol of Taha et al. (77) was used for the preparation of polyribosome-bound mRNA. All solutions were prepared with diethyl pyrocarbonate–treated water and ultrapure, RNase-free reagents. Ten to 20 plates (100 mm) of 267B1/Ki-ras or 20 to 40 plates (100 mm) of 267B1 cells were harvested at 40% to 70% confluency on the second day after transfer. Cells were harvested by scraping with a rubber policeman, pelleted at 1,000 rpm for 5 minutes, and then washed once with ice-cold PBS with 100 µg/mL cycloheximide. The cell pellet was lysed by resuspension in 500 µL to 1 mL of polysome lysis buffer [100 mmol/L KCl, 5 mmol/L MgCl2, 10 mmol/L HEPES (pH 7.4), 100 µg/mL cycloheximide, 0.05% NP40, and 100 units/mL RNasin] followed by one passage (if 1 mg/mL heparin was added) or five passages through a 27 Ga needle. The nuclei and cell debris were removed by centrifugation for 2 minutes at 17,000 rpm followed by recentrifugation of the supernatant at 17,000 rpm for 10 minutes. The clarified cell extract was loaded onto a 15% to 45% (w/v) sucrose gradient prepared in polysome gradient buffer [100 mmol/L KCl, 5 mmol/L MgCl2, 100 µg/mL heparin, and 10 mmol/L HEPES (pH 7.4)] and centrifuged for 4 hours at 26,000 rpm in an AH627 rotor (Sorval, Asheville, NC). The gradient profile was monitored by A254 as the gradient was fractionated. The fractions containing the large polysomes (approximately five ribosomes and over per transcript, bottom one-third of the gradient) were extracted using phenol/chloroform/isoamyl alcohol. Ammonium acetate was added to a final concentration of 0.3 mol/L and the RNA was precipitated by the addition of an equal volume of isopropanol. Total mRNA was purified from cells lysed with a commercial buffer (Qiagen, Valencia, CA). The lysate was passed over a QiaShredder column according to the directions of the manufacturer to remove DNA and then passed over a RNEasy column. RNA was recovered according to the protocol of the manufacturer. Aliquots were monitored for degradation on an agarose gel.

Microarray Analysis
Preparation and analysis of cDNA preparations on Affymetrix U95-Av microarrays were executed according to instructions of the manufacturer (www.affymetrix.com). Five to 20 µg of RNA were used to prepare the first strand of DNA using Superscript II reverse transcriptase and a primer that linked the T7 promoter to poly(dT). The second strand was synthesized by DNA polymerase I in the presence of E. coli RNase H and DNA ligase. The reaction was terminated by the addition of 1/15 volume of 0.5 mol/L EDTA (pH 8.0) followed by addition of an equal volume of phenol/chloroform/isoamyl alcohol. The aqueous phase was separated from the organic phase by centrifugation in a phase lock gel tube. The DNA was precipitated by the addition of 0.6 volume of 5 mol/L ammonium acetate and 2.5 volumes of ethanol. Linear amplification of biotin-labeled cRNA by T7-RNA polymerase was done using the ENZO kit according to the kit instructions. The RNA was purified by passage over a Qiagen RNEasy column and an aliquot examined by agarose gel electrophoresis for degradation. The biotin-labeled cRNA was fragmented and hybridized overnight to the Hu-95 Av human genome microarray chips according to standard protocol (EukGE-WS2v4). The chips were scanned with a GeneArray Scanner (Agilent, Palo Alto, CA) and read according to the instructions of the manufacturer. The Affymetrix controls and background were within the recommended ranges. The images were visually inspected for defects before analysis. Probe sets were scaled to 250 and analyzed by the MAS5 software. Results were then "published" to the Data Mining Tool for statistical analysis. Probe sets were selected that were consistently increased or decreased in the relevant polysome-associated mRNA with P ≤ 0.005. These probe sets were then clustered using self-organizing maps as implemented by the Genesis software (34).

Quantitative PCR
Quantitative PCR was done on total cell lysates and sucrose gradient samples representing large polysomes (fractions 1 to 3 of nine, starting with the bottom fraction) and small polysomes and ribonucleotide particles (fractions 4 to 8 of nine). An equal amount of 267B1 or 267B1/Ki-ras RNA was loaded onto the gradient in each experiment. The RNA was extracted with phenol/chloroform/isoamyl alcohol, and then precipitated with the addition of 0.6 volume of 5 mol/L ammonium acetate and an equal volume of isopropanol. The precipitate was resuspended and further purified on a RNEasy column (Qiagen) according to the instructions of the manufacturer. One microgram of RNA was used to prepare cDNA using the Superscript II RNase H– Reverse Transcriptase kit (Invitrogen, San Diego, CA). The cDNA products were diluted equally across the samples to give 5 to 50 ng per reaction. The quantitative PCR was done on a Model 7900 instrument (Applied Biosystems, Foster City, CA). Duplicate reactions were done using SYBR Green PCR Master Mix (PE Applied Biosystems). Relative amounts of product were calculated based on standard curves of serially diluted substrate. Primers were designed using Primer Express software (Applied Biosystems). The following primers were used: Bet1 forward (5'-TCTCTCCTCTGGCAGGATGAG-3') and Bet1 reverse (5'-AGCATAGCCATAGTTCCCATAGTTG-3'); Ubc9 forward (5'-GGCACGATGAACCTCATGAA-3') and Ubc9 reverse (5'-TCCCACGGAGTCCCTTTCT-3'); MYST forward (5'-AGATCTACCGCAAGAGCAACATC-3') and MYST reverse (5'-GCAGACACAGGTTCTGACAGTAAATC-3'); ITGB5 forward (5-AGCCGTC TGCAAGGAGAAG-3') and ITGB5 reverse (5'-CACATGTTGTGAACACCAGC-3'); Calreticulin forward (5'-ACGTGCTGATCAACAAGGACA-3') and Calreticulin reverse (5'-TGGCCGCACAATCAGTGTGTA-3'); ITGA3 forward (5'-TGGCTCTGCCTTTGGTTTATCT-3') and ITGA3 reverse (5'-CAGCAATATCCTGAAATCCATCCT-3'). Uniformity of the amplicon melting curve and analysis of PCR products by agarose gel electrophoresis were used to ascertain the absence of primer dimers and other artifacts. To determine the relative amount of the total mRNA associated with polysomes (Fig. 5A), the relative RNA units for total and polysome-associated mRNA, as determined by the standard curve, were normalized to GAPDH. The polysome-associated RNA units were divided by the total RNA units for each cell type and plotted. In Fig. 5B, the total RNA units were determined for the top two thirds of the gradient (RNPs and small polysomes) and the bottom third of the gradient (large polysomes) were plotted as percent of total.

Identification of Motifs by Computer Analysis
The collection of mRNAs from each group was expanded to include the most statistically significant 75 mRNAs with P < 0.005. The 5' and 3' UTRs were collated from public information for each gene as accessed through the Affymetrix website. Seventy-five sequences from each group had a 3' UTR but not all mRNAs had a 5' UTR available through the site. The number of 5' UTRs examined were 53 for the group of mRNAs decreased in 267B1/Ki-ras polysomes; 67 for the group of mRNAs increased in 267B1/Ki-ras polysomes; 67 sequences for the mRNAs decreased in 267B1 polysomes; and 69 for the mRNAs increased in 267B1 polysomes. Sequences were analyzed in the UTR website (http://bighost.ba.itb.cnr.it/UTR/) and by MEME (http://meme.sdsc.edu/meme/website/meme.html). The sequences derived from the group of mRNAs that were increased in 267B1/Ki-ras were pooled with sequences from the mRNAs that had decreased association with polysomes in 267B1 for analysis by MEME as both mRNAs exhibit increased translation in 267B1/Ki-ras. Similarly, the populations of mRNAs that were relatively decreased in 267B1/Ki-ras were pooled. Motifs with low probabilities of random occurrence were cross-checked with other expression groups and with Blast (NIH) for uniqueness.

To identify highly conserved motifs, the UTRs from two vertebrate species were obtained from the National Center for Biotechnology Information database and compared with human sequences by MEME. Highly conserved sequences were compared by Motif Alignment and Search Tool (http://meme.sdsc.edu/meme/website/mast-intro.html) to a human genome–expressed sequence tag library and to libraries constructed from the genes represented on the Hu-95 Av microarray chip by using open reading frames as a reference for 5 and 3' UTRs. The library of 5' UTRs contained 8,590 entries >20 nucleotides long and 952,242 characters. The 3' UTR library was formatted by size to include sequences of ≤8,000 nucleotides and contained 10,182 entries and 16,100,382 characters. Additional homologues to mRNA cohorts were identified by using a motif-matching program written in PERL to select mRNAs with common motifs from the UTR libraries. Both Motif Alignment and Search Tool and the PERL motif-matching program successfully identified the initial transcripts containing a given motif. Motifs used to identify potential homologues were as follows: ANNANNNNTN{10 or 11 times}TANCNNTNNNCNNNANNNCNTNCT for ANAX10/SKALP/ZNF266 and GCGNNCGGGNANGCNGCT and GGANGNTGGCGGCGGNGC for CLPTM1 and ITGB5.


   Acknowledgements
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
References
 
We thank J.S. Rhim (Center for Prostate Disease Research, Uniformed Services University of the Health Sciences, Bethesda, MD) for the generous gift of the 267B1 and 267B1/Ki-ras cells; Dr. David Rothwarf, Marilyn Morgan, and Eileen Adamson for reviewing the manuscript; YiPeng Wang for technical advice; and J. Spence thanks J. Leek for support and encouragement.


   Notes
Top
 Notes
Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Grant support: NIH grants 04084107 (D. Mercola) and UO1 CA084998 (D. Mercola).

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.

Received 11/ 5/04; revised 11/30/05; accepted 12/ 7/05.


   References
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 

  1. Castrillon DH, DePinho RA. Modeling prostate cancer in the mouse. Adv Cancer Res 2001;82:187–204.[Medline]
  2. Konishi N, Hiasa Y, Tsuzuki T, Tao M, Enomoto T, Miller GJ. Comparison of ras activation in prostate carcinoma in Japanese and American men. Prostate 1997;30:53–7.[CrossRef][Medline]
  3. Shiraishi T, Muneyuki T, Fukutome K, et al. Mutations of ras genes are relatively frequent in Japanese prostate cancers: pointing to genetic differences between populations. Anticancer Res 1998;18:2789–92.[Medline]
  4. Li J, Yen C, Liaw D, et al. PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science 1997;275:1943–7.[Abstract/Free Full Text]
  5. Daaka Y. G proteins in cancer: the prostate cancer paradigm. Sci STKE 2004;216:re2.
  6. Weber MJ, Gioeli D. Ras signaling in prostate cancer progression. J Cell Biochem 2004;91:13–25.[CrossRef][Medline]
  7. Hamad NM, Elconin JH, Karnoub AE, et al. Distinct requirements for Ras oncogenesis in human versus mouse cells. Genes Dev 2002;16:2045–57.[Abstract/Free Full Text]
  8. Fingar DC, Blenis J. Target of rapamycin (TOR): an integrator of nutrient and growth factor signals and coordinator of cell growth and cell cycle progression. Oncogene 2004;23:3151–71.[CrossRef][Medline]
  9. Miron M, Lasko P, Sonenberg N. Signaling from Akt to FRAP/TOR targets both 4EBP and S6K in Drosophila melanogaster. Mol Cell Biol 2003;23:9117–26.[Abstract/Free Full Text]
  10. Meyuhas O. Synthesis of the translational apparatus is regulated at the translational level. Eur J Biochem 2000;267:6321–30.[Medline]
  11. Gingras AC, Gygi SP, Raught B, et al. Regulation of 4E-BP1 phosphorylation: a novel two-step mechanism. Genes Dev 1999;13:1422–37.[Abstract/Free Full Text]
  12. Marcotrigiano J, Gingras AC, Sonenberg N, Burley SK. Cap-dependent translation initiation in eukaryotes is regulated by a molecular mimic of eIF4G. Mol Cell 1999;3:707–16.[CrossRef][Medline]
  13. Calero G, Wilson KF, Ly T, Rios-Steiner JL, Clardy JC, Cerione RA. Structural basis of m7GpppG binding to the nuclear cap-binding protein complex. Nat Struct Biol 2002;9:912–7.[CrossRef][Medline]
  14. Lazaris-Karatzas A, Smith MR, Frederickson RM, et al. Ras mediates translation initiation factor 4E-induced malignant transformation. Genes Dev 1992;6:1631–42.[Abstract/Free Full Text]
  15. Anand N, Murthy S, Amann G, et al. Protein elongation factor EEF1A2 is a putative oncogene in ovarian cancer. Nat Genet 2002;31:301–5.[Medline]
  16. Rajasekhar VK, Holland EC. Postgenomic global analysis of translational control induced by oncogenic signaling. Oncogene 2004;23:3248–64.[CrossRef][Medline]
  17. Ruggero D, Pandolfi PP. Does the ribosome translate cancer? Nat Rev Cancer 2003;3:179–92.[CrossRef][Medline]
  18. Saramaki O, Willi N, Bratt O, et al. Amplification of EIF3S3 gene is associated with advanced stage in prostate cancer. Am J Pathol 2001;159:2089–94.[Abstract/Free Full Text]
  19. Lopez de Silanes I, Fan J, Yang X, et al. Role of the RNA-binding protein HuR in colon carcinogenesis. Oncogene 2003;22:7146–54.[CrossRef][Medline]
  20. Ostareck-Lederer A, Ostareck DH, Cans C, et al. c-Src-mediated phosphorylation of hnRNP K drives translational activation of specifically silenced mRNAs. Mol Cell Biol 2002;22:4535–43.[Abstract/Free Full Text]
  21. Mueller F, Bommer M, Lacher U, et al. KOC is a novel molecular indicator of malignancy. Br J Cancer 2003;88:699–701.[CrossRef][Medline]
  22. Vernet C, Artzt K. STAR, a gene family involved in signal transduction and activation of RNA. Trends Genet 1997;13:479–84.[CrossRef][Medline]
  23. Tessier CR, Doyle GA, Clark BA, Pitot HC, Ross J. Mammary tumor induction in transgenic mice expressing an RNA-binding protein. Cancer Res 2004;64:209–14.[Abstract/Free Full Text]
  24. Wagner M, Kunsch S, Duerschmied D, et al. Transgenic overexpression of the oncofetal RNA binding protein KOC leads to remodeling of the exocrine pancreas. Gastroenterology 2003;124:1901–14.[CrossRef][Medline]
  25. Gouble A, Grazide S, Meggetto F, Mercier P, Delsol G, Morello D. A new player in oncogenesis: AUF1/hnRNPD overexpression leads to tumorigenesis in transgenic mice. Cancer Res 2002;62:1489–95.[Abstract/Free Full Text]
  26. Gygi SP, Rochon Y, Franza BR, Aebersold R. Correlation between protein and mRNA abundance in yeast. Mol Cell Biol 1999;19:1720–30.[Abstract/Free Full Text]
  27. Johannes G, Carter MS, Eisen MB, Brown PO, Sarnow P. Identification of eukaryotic mRNAs that are translated at reduced cap binding complex eIF4F concentrations using a cDNA microarray. Proc Natl Acad Sci U S A 1999;96:13118–23.[Abstract/Free Full Text]
  28. Diehn M, Eisen MB, Botstein D, Brown PO. Large-scale identification of secreted and membrane-associated gene products using DNA microarrays. Nat Genet 2000;25:58–62.[CrossRef][Medline]
  29. Zong Q, Schummer M, Hood L, Morris DR. Messenger RNA translation state: the second dimension of high-throughput expression screening. Proc Natl Acad Sci U S A 1999;96:10632–6.[Abstract/Free Full Text]
  30. Grolleau A, Bowman J, Pradet-Balade B, et al. Global and specific translational control by rapamycin in T cells uncovered by microarrays and proteomics. J Biol Chem 2002;277:22175–84.[Abstract/Free Full Text]
  31. Rajasekhar VK, Viale A, Socci ND, Wiedmann M, Hu X, Holland EC. Oncogenic Ras and Akt signaling contribute to glioblastoma formation by differential recruitment of existing mRNAs to polysomes. Mol Cell 2003;12:889–901.[CrossRef][Medline]
  32. Parda DS, Thraves PJ, Kuettel MR, et al. Neoplastic transformation of a human prostate epithelial cell line by the v-Ki-ras oncogene. Prostate 1993;23:91–8.[Medline]
  33. Baron V, Duss S, Rhim J, Mercola D. Antisense to the early growth response-1 gene (Egr-1) inhibits prostate tumor development in TRAMP mice. Ann N Y Acad Sci 2003;1002:197–216.[CrossRef][Medline]
  34. Sturn A, Quackenbush J, Trajanoski Z. Genesis: cluster analysis of microarray data. Bioinformatics 2002;18:207–8.[Abstract/Free Full Text]
  35. Jechlinger M, Grunert S, Tamir IH, et al. Expression profiling of epithelial plasticity in tumor progression. Oncogene 2003;22:7155–69.[CrossRef][Medline]
  36. Arava Y, Wang Y, Storey JD, Liu CL, Brown PO, Herschlag D. Genome-wide analysis of mRNA translation profiles in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 2003;100:3889–94.[Abstract/Free Full Text]
  37. Zhao R, Goldman ID. Resistance to antifolates. Oncogene 2003;22:7431–57.[CrossRef][Medline]
  38. Adachi M, Taki T, Huang C, et al. Reduced integrin {alpha}3 expression as a factor of poor prognosis of patients with adenocarcinoma of the lung. J Clin Oncol 1998;16:1060–7.[Abstract]
  39. Schmelz M, Cress AE, Scott KM, et al. Different phenotypes in human prostate cancer: {alpha}6 or {alpha}3 integrin in cell-extracellular adhesion sites. Neoplasia 2002;4:243–54.[CrossRef][Medline]
  40. Tsuji T, Hakomori S, Osawa T. Identification of human galactoprotein ß3, an oncogenic transformation-induced membrane glycoprotein, as VLA-3 {alpha} subunit: the primary structure of human integrin {alpha}3. J Biochem 1991;109:659–65.[Abstract/Free Full Text]
  41. Kinbara K, Goldfinger LE, Hansen M, Chou FL, Ginsberg MH. Ras GTPases: integrins' friends or foes? Nat Rev Mol Cell Biol 2003;4:767–76.[Medline]
  42. Schramm K, Krause K, Bittroff-Leben A, Goldin-Lang P, Thiel E, Kreuser ED. Activated K-ras is involved in regulation of integrin expression in human colon carcinoma cells. Int J Cancer 2000;87:155–64.[CrossRef][Medline]
  43. Asada K, Miyamoto K, Fukutomi T, et al. Reduced expression of GNA11 and silencing of MCT1 in human breast cancers. Oncology 2003;64:380–8.[CrossRef][Medline]
  44. Wang Y, Spitz MR, Tsou AM, Zhang K, Makan N, Wu X. Sulfotransferase (SULT) 1A1 polymorphism as a predisposition factor for lung cancer: a case-control analysis. Lung Cancer 2002;35:137–42.[CrossRef][Medline]
  45. Wong CF, Liyou N, Leggett B, Young J, Johnson A, McManus ME. Association of the SULT1A1 R213H polymorphism with colorectal cancer. Clin Exp Pharmacol Physiol 2002;29:754–8.[CrossRef][Medline]
  46. Gu L, Shigemasa K, Ohama K. Increased expression of IGF II mRNA-binding protein 1 mRNA is associated with an advanced clinical stage and poor prognosis in patients with ovarian cancer. Int J Oncol 2004;24:671–8.[Medline]
  47. Zhang J, Chan EK. Autoantibodies to IGF-II mRNA binding protein p62 and overexpression of p62 in human hepatocellular carcinoma. Autoimmun Rev 2002;1:146–53.[Medline]
  48. Silverman RH. Implications for RNase L in prostate cancer biology. Biochemistry 2003;42:1805–12.[CrossRef][Medline]
  49. Carpten J, Nupponen N, Isaacs S, et al. Germline mutations in the ribonuclease L gene in families showing linkage with HPC1. Nat Genet 2002;30:181–4.[CrossRef][Medline]
  50. Miyazaki T, Reed JC. A GTP-binding adapter protein couples TRAIL receptors to apoptosis-inducing proteins. Nat Immunol 2001;2:493–500.[CrossRef][Medline]
  51. Mukamel Z, Kimchi A. DAP3 localizes to the mitochondria and is involved in the process of mitochondrial fragmentation during cell death. J Biol Chem 2004;279:36732–8.[Abstract/Free Full Text]
  52. Song D, Sakamoto S, Taniguchi T. Inhibition of poly(ADP-ribose) polymerase activity by Bcl-2 in association with the ribosomal protein S3a. Biochemistry 2002;41:929–34.[CrossRef][Medline]
  53. Voigt S, Jungnickel B, Hartmann E, Rapoport TA. Signal sequence-dependent function of the TRAM protein during early phases of protein transport across the endoplasmic reticulum membrane. J Cell Biol 1996;134:25–35.[Abstract/Free Full Text]
  54. Wilkie GS, Dickson KS, Gray NK. Regulation of mRNA translation by 5'- and 3'-UTR-binding factors. Trends Biochem Sci 2003;28:182–8.[CrossRef][Medline]
  55. Pesole G, Liuni S, Grillo G, et al. UTRdb and UTRsite: specialized database of sequences and functional elements of 5' and 3' untranslated regions of eukaryotic mRNAs. Update. Nucleic Acids Res 2002;30:335–40.[Abstract/Free Full Text]
  56. Bailey TL, Gribskov M. Combining evidence using P-values: application to sequence homology searches. Bioinformatics 1998;14:48–54.[Abstract/Free Full Text]
  57. Reimann I, Huth A, Thiele H, Thiele BJ. Suppression of 15-lipoxygenase synthesis by hnRNP E1 is dependent on repetitive nature of LOX mRNA 3'-UTR control element. J Mol Biol 2002;315:965–74.[CrossRef][Medline]
  58. Gangjee A, Dubash NP, Zeng Y, McGuire JJ. Recent advances in the chemistry and biology of folypoly-{gamma}-glutamate synthetase substrates and inhibitors. Curr Med Chem Anti-Canc Agents 2002;2:331–55.
  59. Yoshinare K, Kubota T, Watanabe M, et al. Gene expression in colorectal cancer and in vitro chemosensitivity to 5-fluorouracil: a study of 88 surgical specimens. Cancer Sci 2003;94:633–8.[Medline]
  60. Neshat MS, Mellinghoff IK, Tran C, et al. Enhanced sensitivity of PTEN-deficient tumors to inhibition of FRAP/mTOR. Proc Natl Acad Sci U S A 2001;98:10314–9.[Abstract/Free Full Text]
  61. Zhang Z, Shi Q, Sturgis EM, Spitz MR, Hong WK, Wei Q. Thymidylate synthase 5'- and 3'-untranslated region polymorphisms associated with risk and progression of squamous cell carcinoma of the head and neck. Clin Cancer Res 2004;10:7903–10.[Abstract/Free Full Text]
  62. Reynolds LE, Wyder L, Lively JC, et al. Enhanced pathological angiogenesis in mice lacking ß3 integrin or ß3 and ß5 integrins. Nat Med 2002;8:27–34.[CrossRef][Medline]
  63. Sikes RA, Nicholson BE, Koeneman KS, et al. Cellular interactions in the tropism of prostate cancer to bone. Int J Cancer 2004;110:497–503.[CrossRef][Medline]
  64. Slack-Davis JK, Parsons JT. Emerging views of integrin signaling: implications for prostate cancer. J Cell Biochem 2004;91:41–6.[CrossRef][Medline]
  65. Xiang Y, Wang Z, Murakami J, et al. Effects of RNase L mutations associated with prostate cancer on apoptosis induced by 2',5'-oligoadenylates. Cancer Res 2003;63:6795–801.[Abstract/Free Full Text]
  66. Holcik M. Targeting translation for treatment of cancer—a novel role for IRES? Curr Cancer Drug Targets 2004;4:299–311.[CrossRef][Medline]
  67. Sachs AB. Cell cycle-dependent translation initiation: IRES elements prevail. Cell 2000;101:243–5.[CrossRef][Medline]
  68. Choy L, Derynck R. The type II transforming growth factor (TGF)-ß receptor-interacting protein TRIP-1 acts as a modulator of the TGF-ß response. J Biol Chem 1998;273:31455–62.[Abstract/Free Full Text]
  69. Wald D, Commane M, Stark GR, Li X. IRAK and TAK1 are required for IL-18-mediated signaling. Eur J Immunol 2001;31:3747–54.