| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Article |
From the Laboratory of Cancer Genetics,*
Institute of
Medical Technology, and the Department of Clinical Genetics and
Laboratory of Cancer Genetics,¶
Tampere University
Hospital, Tampere, Finland; the Institute for
Pathology
and Urologic Clinics
BS/BL,
University of Basel, Basel,
Switzerland; the Department of Surgery,
Helsingborg Hospital, Helsingborg, Sweden; and the Cancer Genetics
Branch,||
National Human Genome Research Institute,
National Institutes of Health, Bethesda, Maryland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Using CGH, we and others have previously shown that one of the most
common genetic aberrations in advanced prostate cancer is the gain of
the long arm (q-arm) of chromosome 8.3-7
It is found in
up to 80% of hormone-refractory tumors and distant metastases but only
in 
5% of untreated primary prostate carcinomas.3,4
In
the prostatectomy-treated patients, the gain of 8q seems to be
associated with advanced stage and poor prognosis.7,10
In
addition to prostate cancer, gain of 8q is commonly found in several
other malignancies, such as breast, bladder, and ovarian
cancers.11
For example, almost half of the breast
carcinomas contain gain of 8q. And, the gain seems to be associated
also with poor survival.12
In most of the prostate tumors gain of 8q comprises the whole q-arm.
However, CGH studies have indicated that there are, at least two
independently amplified subchromosomal regions, 8q21 and 8q23-q24,
suggesting the presence of several target genes.4,5
The
well-known oncogene, MYC, located at 8q24.1, is considered to be a
putative target gene for the gain.4
To identify other
possible target genes, we recently used the subtraction hybridization
technique to clone overexpressed genes in breast and prostate
tumors.13
We found that EIF3S3, located at 8q23, was
amplified and overexpressed in approximately one-third of the
hormone-refractory prostate carcinomas.13,14
The EIF3S3
gene encodes for the p40 subunit of the eukaryotic translation
initiation factor 3 (eIF3). eIF3 is the largest (600 kd) translation
initiation factor protein complex, which has a central role in the
initiation of translation. It binds to 40S ribosomal subunits in the
absence of other initiation factors and preserves the dissociated state
of 40S and 60S ribosomal subunits. It also stabilizes
eIF2 · GTP · Met-tRNA binding to 40S subunits and mRNA binding to
ribosomes.15
However, very little is known about the p40
subunit itself and how it could be functionally involved in
tumorigenesis.13,16
There are, however, suggestions that
aberrant regulation of translation could be important in the
development of cancer. For example, overexpression of initiation
factors EIF4E and EIF4G1 has been shown to transform normal
cells.17,18
In addition, amplification of EIF4G1, and
overexpression of EIF4E have been found in squamous cell lung and
breast carcinomas, respectively.19,20
Also, allelic
imbalance of INT6, which encodes for p48 subunit of eIF3, has been
detected in breast cancer.21
Recently, a new gene EIF5A2,
encoding a putative translation initiation factor was cloned and shown
to be amplified in the subset of ovarian cancer.22,23
The aim of this study was to investigate the frequency of the EIF3S3 amplification in different stages of prostate cancer and to study the co-amplification of EIF3S3 and MYC. In addition, prognostic utility of the EIF3S3 amplification was evaluated. The analyses were done using FISH and new tissue microarray technology allowing large number of tumors to be rapidly analyzed.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The material consisted of three sets of prostate tumors. Group I included 21 benign prostatic hyperplasias, 42 prostatic intraepithelial neoplasias, 183 radical prostatectomy specimens, 20 Tru-Cut needle biopsy specimens of stage T3/T4 prostate tumors, 95 hormone refractory prostate tumors, and 39 distant metastases (obtained from University of Basel and Tampere University Hospitals). Fifty-four untreated local lymph node metastases were obtained from Lund University Hospital. Clinical stage and Gleason score of the tumors were available.
Group II included 112 incidentally found T1a/b tumors from transuretral resections for benign prostatic hyperplasia obtained from the University of Basel. The age of the patients at the time of diagnosis varied between 58 and 94 years with a mean of 76 years. FISH analysis was successful in 105 of 112 specimens. Of those 105, there were 20 Gleason 2-4, 60 Gleason 5-7, and 21 Gleason 8-10 tumors. Gleason score of four tumors was not available. The patients had been treated with standard therapies. Overall and prostate cancer-specific survival data were available.
Group III included 145 radical prostatectomy specimens from the University of Basel. The age of the patients at time of diagnosis was between 45 and 82 years with a mean of 65.4 years. The TNM stage distribution of the successfully hybridized cases was: 1 T1N0M0, 26 T2N0M0, 46 T3N0M0, 4 T2N1M0, 11 T3N1M0, and 6 T3N2M0. The TNM distribution was not available for 41 tumors. The Gleason score distribution was 100 Gleason 5-7, 33 Gleason 8-10 tumors, and 2 unknown. The progression-free time of the patients was available. The progression was defined either by increase in prostate-specific antigen levels (86% of cases), a positive finding in bone scan (11% of cases), or by biopsy proven local recurrence (3% of cases). The average recurrence-free time was 4.5 years (range, 0.6 to 15.1 years).
FISH
Multitissue blocks were made from the original formalin-fixed paraffin-embedded tumor blocks according to published guidelines.24 Routine hematoxylin and eosin-stained slides were used to evaluate the representativeness of the samples. For the FISH analyses 5-µm sections from the multitissue blocks were either cut onto SuperFrost Plus slides (Menzel-Gläser, Braunschweig, Germany) and baked overnight, or an adhesive-coated tape sectioning system (Instrumedics, Hackensack, NJ) was used. A locus-specific PAC probe for EIF3S3 or MYC13 and pericentromeric probe for chromosome 8 (pJM128) were labeled by nick translation with digoxigenin (locus-specific probes) and fluorescein-isothiocyanate (centromere-specific probe). The deparaffinized slides were treated with 1 mol/L NaSCN for 10 minutes at 80°C, followed by incubation in 4 mg/ml pepsin (P-7012, in 0.9% NaCl, pH 1.5; Sigma Chemical Co., St. Louis, MO) for 15 minutes at 37°C. The slides were then washed in H2O and 2x standard saline citrate, followed by dehydration in an ethanol series, and air-dried. The probes were applied on the slides in a hybridization mix (50% formamide, 10% dextran sulfate in 1x standard saline citrate, pH 7) and then co-denatured with the samples at 80°C for 8 minutes. After hybridization for 2 to 3 days in a humid chamber, the slides were washed and the locus-specific probes were detected immunohistochemically by anti-digoxigenin rhodamine. The slides were counterstained with 0.1 mol/L 4,6-diamidino-2-phenylindole in Vectashield anti-fade solution (Vector Laboratories Inc., Burlingame, CA).
The FISH signals were scored from nonoverlapping epithelial cells using
an Olympus BX50 epifluorescence microscope (Tokyo, Japan). A
Photometrics charge-coupled device camera (Photometrics, Tucson, AZ)
and IPLab software program (Scananalytics Inc., Fairfax, VA) were used
to capture images. The previously published criteria for
amplification13
were slightly modified because
tissue sections, instead of isolated nuclei, were analyzed
here. Briefly, the tumors were classified into three groups:
nonamplified (no increase in EIF3S3 or c-myc copy number),
low-level amplification (3 to 5 copies per cell), and high-level
amplification (
5 copies of the genes per cell). Tumors that showed
>10% of malignant cells with increased copy number of either EIF3S3
or c-myc were considered to have copy number alterations.
Statistical Analysis
Statistical analysis of the data were done using BMDP Statistical Software Package.25 Pearson chi-square test was used to evaluate the associations of the gene copy number and tumor type, clinical stage, and Gleason score. The survival differences of patients were evaluated by Kaplan-Meier method, and the statistical significance of survival differences between the patient groups was determined with Mantel-Cox and Breslow tests.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
demonstrates a FISH
analysis of a hormone-refractory prostate tumor with high-level
amplification of EIF3S3.
|
summarizes the frequency
of the EIF3S3 amplification in prostate tumors. No copy number changes
were found in nonmalignant (benign prostatic hyperplasia) or
premalignant (prostatic intraepithelial neoplasia) prostate. The
high-level amplification of EIF3S3 was found only in <10% of local
(pT1 and pT2) prostate cancers, whereas hormone-naïve lymph
node metastases as well as hormone-refractory tumors showed the
amplification in 20 to 50% of the cases. Gain (or low-level
amplification) of EIF3S3 was found in 30 to 50% of the prostate
cancers. The amplification of the gene was statistically significantly
associated with advanced stage of disease (P <
0.001) and high Gleason score (P < 0.001).
|
. The difference
was not quite statistically significant (P =
0.084). In the disease-specific survival analysis (Figure 2A)
, the increased copy number (either
low- or high-level amplification) of EIF3S3 was associated with poor
survival (P = 0.023).
|
|
. However, the difference
was not statistically significant (P = 0.380).
By combining the data from all above-mentioned samples
(n = 461), the co-amplification of EIF3S3 and
c-myc in almost all cases became evident (Table 3)
. There was only one case with
high-level EIF3S3 amplification with two copies of MYC.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
8) than moderately or
well-differentiated tumors (Gleason score 
7). Altogether, the
results suggest that the amplification of EIF3S3 is involved in the
late progression of prostate cancer. The FISH analyses were done using tissue sections making the evaluation of actual copy number of the gene difficult. However, in the case of amplification, typically 5 to 10 signals per nucleus were seen indicating that the level of amplification was quite moderate. The finding is consistent with the earlier CGH studies, which most often have shown gain of the whole q-arm of the chromosome, and only rarely high-level regional amplification.3-5,9 This feature makes the gain of 8q clearly different from, for example gain of Xq, the second most common gain in hormone-refractory prostate cancers. According to FISH analysis, the amplification of the target gene of the Xq gain, androgen receptor gene, typically consists of 10 or more copies of the gene.26 However, we have previously shown that even the moderately increased copy number of EIF3S3 may lead to overexpression of the gene in prostate cancer.13 Therefore, it is possible that amplification of lower level but larger chromosomal region harboring many genes in 8q is selected for during the progression of prostate cancer.
Another putative target gene for the gain of 8q in prostate cancer is MYC located at 8q24.1.4 We have earlier shown that in a subset of breast carcinomas EIF3S3 and MYC are not co-amplified.14 Here, in this large series of prostate tumors, we found only one case in which EIF3S3 was amplified without MYC amplification, whereas all cases with MYC amplification had also EIF3S3 amplification. The finding suggests that both EIF3S3 and MYC may be important in the progression of prostate cancer, and therefore they are equally selected for. In addition to MYC and EIF3S3 there are other putative target genes for 8q gain in prostate cancer as well. These include recently cloned prostate stem-cell antigen, and GC79 encoding a zinc-finger protein, both located in the 8q23-24 region.27,28 The other minimal commonly amplified region in prostate cancer is 8q21 of which target genes are still not known.4,5,9 Because there are likely to be numerous putative target genes, it will be important to compare the alterations of all of the different putative target genes in large tumor materials as done here for MYC and EIF3S3.
The prognostic significance of the EIF3S3 amplification was
retrospectively studied here in two sets of tumors. In the cases of
incidental prostate cancers, the amplification was found in only 5%
of cases. The number of tumors with the amplification was too small for
prognostic analyses. However, by combining the groups of low- and
high-level amplification, we found that the increased copy number of
EIF3S3 was associated with poor disease-specific survival. Thus, it may
be that the copy number alteration of EIF3S3 could be useful in
predicting which incidental cancers are clinically significant.
Evidently larger studies are needed to confirm the finding.
In the second set of cases, 135 prostatectomy specimens, from patients
whose progression-free survival data were available, were analyzed.
Approximately 55% of the patients with the amplification but only
30% of those without amplification experienced progression.
Although the progression-free time curves showed a worse prognosis for
the patients with amplification of EIF3S3 than for patients without the
amplification, the difference was not statistically significant. This
is in some contrast to the findings of Sato and
co-workers,29
who have previously suggested that MYC
amplification is associated with poor survival in prostatectomy treated
stage C disease. Because MYC and EIF3S3 were almost always
co-amplified, it was evident that MYC did not have prognostic value in
our material either (data not shown). The major difference between the
studies by Sato and colleagues29
and by us is that Sato
and co-workers analyzed high-grade, stage pT3 tumors, whereas our
material consisted of both moderately and poorly differentiated pT1-pT3
tumors. Thus, for example, the frequency of 8q gain was higher in the
study by Sato and colleagues29
than in our study (54.2%
versus 30.4%). The definition of high-level amplification
was also different in the two studies. Sato and
co-workers29
found an "additional increase" of MYC in
19.4% of cases, whereas high-level amplification of EIF3S3 (and MYC)
was found in only 8% of our prostatectomy series. It may also be that
because the cases in our study were lower stage and grade, the
prostatectomy had removed these tumors before the effects of EIF3S3
(and/or MYC) on progression had have enough time to affect.
In this study, the FISH analyses were performed in a multitissue section format. The use of multitissue blocks instead of original single tumor blocks has several advantages. It allowed us to screen a large number of tumors in a relatively short period of time. Altogether 609 specimens were analyzed. Because the hybridizations were done on a few slides (five slides per gene), the slide-to-slide variation could also be expected to be low. An additional advantage of multitissue slides is that, at least in our hands, the FISH analyses seemed to work better in this than in the traditional one tumor section per slide format. This may well be because of the fact that in the multitissue blocks the tissue samples are small and equal in size. Therefore, the pretreatment of the slide, which is the most important variable in the FISH analysis, probably affects equally to each specimen on the slide. The disadvantage of the multitissue technology is that only a small proportion of the tumor is analyzed. Because of the known intratumoral heterogeneity of prostate cancer, the absolute frequency of aberrations may thus be somewhat underestimated. However, this possible underestimation of absolute frequencies should not affect the clinicopathological associations based on a large number of specimens, because all specimens on a tissue microarray are subjected to the same sampling limitations.
In conclusion, we have shown here that the high-level amplification of EIF3S3 gene is associated with advanced stage, androgen independence, and poor differentiation of prostate cancer. The gene is in most of the cases co-amplified with MYC. Both association with advanced stage and preliminary prognostic analyses suggest that amplification of EIF3S3 might be important for the progression of prostate cancer. Further studies are warranted to evaluate the function of the gene as well as possible prognostic utility of the amplification.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by grants from the Academy of Finland, the Cancer Society of Finland, the Reino Lahtikari Foundation, the Medical Research Fund of Tampere University Hospital, the Sigrid Juselius Foundation, and the CaP CURE.
Accepted for publication August 17, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. Zhang, Z. Smit-McBride, X. Pan, J. Rheinhardt, and J. W. B. Hershey An Oncogenic Role for the Phosphorylated h-Subunit of Human Translation Initiation Factor eIF3 J. Biol. Chem., August 29, 2008; 283(35): 24047 - 24060. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Abdulghani, L. Gu, A. Dagvadorj, J. Lutz, B. Leiby, G. Bonuccelli, M. P. Lisanti, T. Zellweger, K. Alanen, T. Mirtti, et al. Stat3 Promotes Metastatic Progression of Prostate Cancer Am. J. Pathol., June 1, 2008; 172(6): 1717 - 1728. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-H. Tan, A. Dagvadorj, F. Shen, L. Gu, Z. Liao, J. Abdulghani, Y. Zhang, E. P. Gelmann, T. Zellweger, Z. Culig, et al. Transcription Factor Stat5 Synergizes with Androgen Receptor in Prostate Cancer Cells Cancer Res., January 1, 2008; 68(1): 236 - 248. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Dagvadorj, S. Collins, J.-B. Jomain, J. Abdulghani, J. Karras, T. Zellweger, H. Li, M. Nurmi, K. Alanen, T. Mirtti, et al. Autocrine Prolactin Promotes Prostate Cancer Cell Growth via Janus Kinase-2-Signal Transducer and Activator of Transcription-5a/b Signaling Pathway Endocrinology, July 1, 2007; 148(7): 3089 - 3101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Schleicher, C. Poremba, H. Wolters, K.-L. Schafer, N. Senninger, and M. Colombo-Benkmann Gain of Chromosome 8q: A Potential Prognostic Marker in Resectable Adenocarcinoma of the Pancreas? Ann. Surg. Oncol., April 1, 2007; 14(4): 1327 - 1335. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zhang, X. Pan, and J. W. B. Hershey Individual Overexpression of Five Subunits of Human Translation Initiation Factor eIF3 Promotes Malignant Transformation of Immortal Fibroblast Cells J. Biol. Chem., February 23, 2007; 282(8): 5790 - 5800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Scoles, W. H. Yong, Y. Qin, K. Wawrowsky, and S. M. Pulst Schwannomin inhibits tumorigenesis through direct interaction with the eukaryotic initiation factor subunit c (eIF3c) Hum. Mol. Genet., April 1, 2006; 15(7): 1059 - 1070. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Spence, B. M. Duggan, C. Eckhardt, M. McClelland, and D. Mercola Messenger RNAs under Differential Translational Control in Ki-ras-Transformed Cells Mol. Cancer Res., January 1, 2006; 4(1): 47 - 60. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T.-H. Kim, B.-H. Kim, A. Yahalom, D. A. Chamovitz, and A. G. von Arnim Translational Regulation via 5' mRNA Leader Sequences Revealed by Mutational Analysis of the Arabidopsis Translation Initiation Factor Subunit eIF3h PLANT CELL, December 1, 2004; 16(12): 3341 - 3356. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Rubin, S. Varambally, R. Beroukhim, S. A. Tomlins, D. R. Rhodes, P. L. Paris, M. D. Hofer, M. Storz-Schweizer, R. Kuefer, J. A. Fletcher, et al. Overexpression, Amplification, and Androgen Regulation of TPD52 in Prostate Cancer Cancer Res., June 1, 2004; 64(11): 3814 - 3822. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Wang, C. Brennan, M. Rook, J. L. Wolfe, C. Leo, L. Chin, H. Pan, W.-H. Liu, B. Price, and G. M. Makrigiorgos Balanced-PCR amplification allows unbiased identification of genomic copy changes in minute cell and tissue samples Nucleic Acids Res., May 21, 2004; 32(9): e76 - e76. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Wang, J. Xu, O. Saramaki, T. Visakorpi, W. M. Sutherland, J. Zhou, B. Sen, S. D. Lim, N. Mabjeesh, M. Amin, et al. PrLZ, a Novel Prostate-Specific and Androgen-Responsive Gene of the TPD52 Family, Amplified in Chromosome 8q21.1 and Overexpressed in Human Prostate Cancer Cancer Res., March 1, 2004; 64(5): 1589 - 1594. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Linja, K. P. Porkka, Z. Kang, K. J. Savinainen, O. A. Janne, T. L. J. Tammela, R. L. Vessella, J. J. Palvimo, and T. Visakorpi Expression of Androgen Receptor Coregulators in Prostate Cancer Clin. Cancer Res., February 1, 2004; 10(3): 1032 - 1040. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.A. Schulz, M. Burchardt, and M.V. Cronauer Molecular biology of prostate cancer Mol. Hum. Reprod., August 1, 2003; 9(8): 437 - 448. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. M. Ghadimi, M. Grade, T. Liersch, C. Langer, A. Siemer, L. Fuzesi, and H. Becker Gain of Chromosome 8q23-24 Is a Predictive Marker for Lymph Node Positivity in Colorectal Cancer Clin. Cancer Res., May 1, 2003; 9(5): 1808 - 1814. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J M M Tan, E P C Tock, and V T K Chow The novel human MOST-1 (C8orf17) gene exhibits tissue specific expression, maps to chromosome 8q24.2, and is overexpressed/amplified in high grade cancers of the breast and prostate Mol. Pathol., April 1, 2003; 56(2): 109 - 115. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Tsuchiya, Y. Kondo, A. Takahashi, H. Pawar, J. Qian, K. Sato, M. M. Lieber, and R. B. Jenkins Mapping and Gene Expression Profile of the Minimally Overrepresented 8q24 Region in Prostate Cancer Am. J. Pathol., May 1, 2002; 160(5): 1799 - 1806. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
