| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
Corixa Corp., Seattle, Washington 98104 [J. X., M. K., J. A. S., E. J. Z., X. Z., R. L. H., S. G. R.]; Hospital Aristides Maltez, Salvador 40125-001, Bahia, Brazil [A. M. F., M. N., R. B.]; and Department of Pathobiology, University of Washington, Seattle, Washington 98195 [S. G. R.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Numerous approaches have been used to identify cancer-specific and cancer-associated markers, including expressed sequence tag sequencing (2 , 3) , serial analysis of gene expression (4 , 5) , and differential display PCR (6) . Expression cloning using sera (7) or T cells (8) from cancer patients has been used to identify a panel of genes that are immunologically relevant and may be used as potential cancer vaccines and markers. Each of the above-mentioned approaches requires a tremendous amount of effort to identify a limited number of antigens. Furthermore, success depends on the availability of clinical reagents, and many antigens identified using these approaches are not cancer or tissue specific. Finally, none of the above-mentioned techniques provide a complete, systematic, and reliable comparison of the gene expression differences between two tissue types.
We have recently described a novel genome-based approach for the identification and characterization of tumor-specific proteins. This approach involves an initial cDNA library subtraction, followed by high-throughput microarray screening for tissue- and/or tumorspecific gene expression (9) . In this report, we describe the application of this approach to isolate a novel prostate tissue-specific protein, prostein. The results presented in this report demonstrate that prostein is a prostate-specific marker with potential clinical utility for the diagnosis and treatment of prostate cancer.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
For the androgen stimulation experiments, LNCaP cells were plated at 1.5 x 106 cells/T75 flask (for RNA isolation) or 3 x 105 cells/well of a 6-well plate (for FACS3 analysis) and grown overnight in RPMI 1640 containing 10% charcoal-stripped FCS (Life Technologies, Inc.). Cell culture was continued for an additional 72 h in RPMI 1640 containing 10% charcoal-stripped FCS, with 1 nM of the synthetic androgen methyltrienolone (R1881; New England Nuclear) added at various time points. Cells were then harvested for RNA isolation and FACS analysis at 0, 1, 2, 4, 8, 16, 24, 48, and 72 h after androgen addition.
For tissue sample RNA isolation, tissues were frozen in liquid nitrogen and homogenized with a Polytron (Kinematica), and total RNA was prepared using Trizol reagent (Life Technologies, Inc.). Polyadenylated RNA was isolated using a Qiagen oligotex spin column mRNA purification kit. For cell line RNA isolation, total RNA was prepared using Trizol reagent (Life Technologies, Inc.).
cDNA Library Subtraction.
cDNA library subtraction was performed as described by Hara et
al. (10)
, with modifications (9)
.
Microarray.
mRNA expression of prostein was determined using a microarray assay as
described previously (9)
.
Northern Blot Analysis.
Northern blot analysis was performed as described previously
(9)
using 32P-labeled prostein cDNA
probe.
Quantitative Real-Time PCR.
Quantitative real-time PCR assay was performed as described previously
(9)
using 300 nM each of prostein forward
(5'-CCCAGGACCTTGGAAATTCTACT-3') and reverse
(5'-ACCTTCCTTCAACACCCTAACCT-3') primers.
Bioinformatic Analysis.
Transmembrane domains and protein localization of prostein were
predicted by the PSORT algorithm using the prostein amino acid
sequence.
FACS Analysis.
A prostein-specific mouse monoclonal antibody 10E3-G4-D3 that
recognizes an intracellular epitope of prostein was generated by
standard hybridoma technologies and is described in detail
elsewhere.4
For FACS analysis, cells were harvested using trypsin (which was shown
not to affect the antibody epitope), fixed in 2% formaldehyde,
permeabilized in 0.5% saponin, and incubated with either the
anti-prostein antibody 10E3-G4-D3 or an isotype-matched control
antibody (purified anti-trinitrophenol murine IgG2a; PharMingen)
at 1 µg/ml, followed by incubation with a FITC-conjugated antimouse
immunoglobulin secondary antibody (PharMingen) at 5 µg/ml. After
washes, the cells were analyzed using a FACSCalibur instrument (Becton
Dickinson).
Immunohistochemical Studies.
Immunohistochemical staining was performed on formalin-fixed,
paraffin-embedded tissues by QualTek Molecular Laboratories using the
prostein-specific mouse monoclonal antibody 10E3G4-D3.
Chromosome Localization.
The GeneBridge 4 Radiation Hybrid panel (Research Genetics) was used to
determine the chromosomal location of prostein. Prostein primers
5'-ACTATGGTCCAGAGGCTGTG-3' and 5'-AGAGGCGGCACATAGGTGAT-3' were used in
PCR reactions with DNA pools from the hybrid panel according to the
manufacturer’s instructions. After 38 cycles of amplification, the
reaction products were separated on a 1.2% agarose gel, and the
results were analyzed through the Whitehead Institute/Massachusetts
Institute of Technology Center for Genome Research web
server5
to determine the probable chromosomal location.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Multiple stop codons in frame with the prostein ORF were identified
in the prostein cDNA upstream of the presumptive initiator ATG,
indicating that the entire prostein ORF had been obtained. The
full-length prostein sequence was used to search GenBank DNA and
protein databases, and no significant homologies were detected. The
closest homology (28% identity and 44% similarity at the amino acid
level) was with the human AIM-1 protein (GenBank ID, 5802879), a shared
melanoma antigen recognized by HLA-A*0201-restricted T cells identified
with the use of an in vitro immunoselected tumor line. Based
on the PSORT algorithm, prostein is predicted to be a type IIIa plasma
membrane protein with 11 potential transmembrane spans and a cleavable
signal sequence. The putative transmembrane regions and signal peptide
are indicated in Fig. 1
.
|
The Prostein Transcript Is Uniquely Expressed in Normal Prostate
and Prostate Tumor Tissues.
The tissue expression profile of the prostein transcript was determined
using three independent approaches. Northern blot analysis was
initially used to determine the expression of prostein in prostate
tumors, normal prostate, BPH, and a panel of normal tissues including
colon, kidney, liver, lung, pancreas, skeletal muscle, brain, stomach,
testis, small intestine, and bone marrow. As shown in Fig. 2
, a 3.8-kb transcript was detected in four of four prostate tumors, one
normal prostate, and one BPH sample; expression was undetectable in the
other normal tissues tested. The prostate-specific expression was
further analyzed by microarray analysis on a larger panel of tissues
(Fig. 3)
. Similar to the results from Northern blot analysis, prostein mRNA
expression was shown to be restricted to prostate tissues. Prostein
expression levels in prostate tissues ranged from 2- to 66.5-fold
higher than that of the paired normal tissues, with an average
overexpression of 23.3-fold in prostate tissues compared to other
tissues tested. Quantitative real-time PCR, a more sensitive and
quantitative assay, was also used to examine prostein mRNA expression.
As shown in Fig. 4A
, prostein mRNA could be detected in 23 of 23 prostate
tumors, 3 of 3 BPH samples, and 3 of 3 normal prostate samples. No
significant expression was detected in any of the other normal tissues
tested. In summary, the data generated from each of these three
independent methodologies demonstrate that the prostein transcript is
expressed in a prostate-specific manner by the majority of normal and
malignant prostate tissues.
|
|
|
, 3
, and 4A
, prostein expression is not
tumor specific because normal prostate, BPH, and prostate tumors all
express prostein message. Prostein mRNA expression was further examined
on a Gleason panel consisting of six normal prostate tissues, six BPH
samples, and a panel of prostate tumors of various Gleason grades. As
shown in Fig. 4B
, prostein mRNA was expressed in all normal
prostate, BPH, and prostate tumors of various Gleason grades.
Additionally, no correlation was observed between the level of prostein
mRNA expression and the Gleason grade of the tumors. PSA mRNA
expression levels were also assessed on the same Gleason panel; similar
to prostein, PSA expression levels were not Gleason grade dependent
(data not shown).
We have also examined the expression of prostein in a cell line panel
consisting of in vitro established tumor cell lines and
pooled normal and tumor tissues. Prostein expression is specific
for prostate and is not detected in other tumor cell lines, including
breast, lung, ovarian, and colon tumor cell lines. As shown in Fig. 4C
, with the exception of a single breast tumor cell line,
MDA-MB-415, which expresses a low level of prostein, all other tumor
cell lines tested did not express significant amounts of prostein.
Furthermore, prostein was not detected in pooled breast, lung, ovary,
and colon tumor and normal tissues.
Prostein mRNA and Protein Are Expressed in Prostate Tumor Cell
Lines.
The expression of prostein mRNA and protein in prostate tumor cell
lines was determined using quantitative reverse transcription-PCR and
FACS analysis. As shown in Fig. 5A
, prostein-specific transcripts could be detected in normal
prostate and prostate tumors as well as in LNCaP cells and were
detected at a very low level in PC-3 but not in DU-145 prostate
tumor cells. These data also demonstrate that normal prostate and
prostate tumor tissues express significantly higher levels of prostein
transcript than the prostate tumor cell lines tested. The FACS analysis
shown in Fig. 5B
confirmed the real-time PCR analysis
because the prostein-specific monoclonal antibody 10E3-G4-D3
specifically stained LNCaP cells and also stained PC-3 cells at a low
level but failed to react with DU-145 cells. Because LNCaP cells
expressed higher transcript levels and stained more intensely with the
anti-prostein monoclonal antibody when compared with PC-3 and DU145
cells, these results demonstrate that prostein mRNA levels correlate
with protein expression levels.
|
. Expression was detected in all prostatic glandular cells.
Furthermore, expression levels were not tumor grade related because
similar staining intensity was seen in tumors of different grades.
Prostate tumor tissues metastatic to lymph node, bone, and liver also
expressed high levels of prostein (data not shown). Prostein expression
was not detected in normal heart, kidney, liver, lung, or colon. The
immunohistochemistry analysis also showed a punctate plasma membrane
staining pattern. In many areas, the staining appeared to be clustered
within the cytoplasm in a perinuclear location. The staining was
typically polarized and located between the nucleus and the lumen in
normal and benign prostate epithelial cells. In prostate carcinomas,
the polarity of the staining was completely random from cell to cell
but usually remained polarized within each cell.
|
Expression of Prostein Is Up-Regulated by Androgen.
To determine whether prostein expression was affected by androgen,
LNCaP cells were grown overnight in androgen-depleted media and then
supplemented with the synthetic androgen methyltrienolone. Cells were
harvested after androgen stimulation for various time periods, and
prostein expression was measured both at the mRNA level using Northern
blot analysis and at the protein level using FACS analysis. As shown in
Fig. 7A
, culture of LNCaP cells in the absence of androgen for
72 h (0 h time point) reduced prostein mRNA levels. Culture of
LNCaP cells in the presence of androgen resulted in the induction of
prostein mRNA, which was detectable as early as 2 h after androgen
treatment and increased through 48–72 h of culture in the presence of
androgen. As shown in Fig. 7B
, prostein protein levels were
also affected by the presence of androgen. Culture of LNCaP in the
absence of androgen for 72 h (0 h time point) resulted in a loss
of detectable prostein protein expression. Culture of LNCaP cells in
the presence of androgen resulted in the induction of prostein protein,
which was detectable as early as 8 h after androgen treatment and
increased through 72 h of culture in the presence of androgen.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sequence homology comparisons suggest that the prostein gene product does not belong to a known family of proteins. Unlike a number of previously described prostate-specific genes such as PSA, glandular kallikrein, and the newly discovered serine protease prostase (11) , prostein does not appear to be a serine protease and shows no homology to kallikrein family members. The closest identifiable homology to a known protein is to the human AIM-1 protein. Although the mechanisms for prostate-specific expression and the biological function of prostein are unknown, localization of the prostein gene to a region of chromosome 1 reported to be linked to prostate cancer susceptibility in hereditary prostate cancer (12 , 13) suggests that the prostein gene may play a role in prostate cancer malignancy. The observation that the strongest homology of prostein was with a previously identified tumor antigen also provides the possibility that these molecules may be involved in the malignant phenotype.
The data presented in this report do not demonstrate correlation of prostein gene expression and prostate tumor Gleason grade. However, due to the cellular heterogeneity of prostate tumor tissue samples, it is likely that the percentage of actual cancer cells in each sample varies, and this may affect the levels of prostein gene expression detected. Immunohistochemical staining of a large panel of prostate tumors of various Gleason grades will further address the issue of prostein gene expression and Gleason grade.
Prostein appears to be expressed in an exclusively prostate-specific pattern because prostein mRNA and protein were not detected in any of the non-prostatic tissues tested. Recent reports have demonstrated similarities in gene expression between prostate and breast tissues (14) . However, as shown in this manuscript, prostein mRNA expression could not be detected in normal and cancer breast tissues. Therefore, prostein could be useful in clinical diagnosis as an additional marker to both enhance the accuracy and reduce the false positive and negative rates of the currently used PSA test. Use of prostein in this setting may reduce or eliminate the issues of PSA cross-reactivity with other closely related serine proteases such as glandular kallikreins 1 and 2. Both protein-based assays (FACS, cytostain) and mRNA-based assays (reverse transcription-PCR) are currently being evaluated to determine the potential of prostein as a marker to detect prostate cancer cells circulating in the blood stream, which in turn may provide stage-related diagnostic and prognostic information. We anticipate that further characterization of the prostein genomic sequence will allow us to determine whether prostein is amplified in prostate cancers. The potential amplification of the prostein genomic sequence may be an important tool for prostate cancer diagnosis. Furthermore, the identification and characterization of the prostein promoter sequence may provide important information for the development of gene therapy approaches.
Expression of prostein on the cell surface was initially predicted by the bioinformatic algorithm PSORT. In support of this observation, immunohistochemical staining revealed a punctate surface-staining pattern. FACS analysis using prostein peptide-specific polyclonal antibodies to a predicted surface-expressed epitope showed surface staining of LNCaP cells.4 FACS analysis using the monoclonal antibody 10E3-G4-D3 only stained permeabilized prostate tumor cell lines, suggesting that the epitope recognized by 10E3-G4-D3 is intracellular. The potential expression of prostein on the cell surface suggests that prostein may be an excellent target for therapeutic antibodies. Antibody-mediated therapies toward cell surface proteins such as CD20 and HER2/neu are being used as treatments for non-Hodgkin’s lymphoma (15) and breast cancer (16) , respectively. In prostate, antibodies against prostate-specific membrane antigens are being evaluated as a therapeutic approach to treat prostate cancers (17) . Recently, a new prostate-specific cell surface antigen, STEAP, has been described (18) . However, the expression of STEAP is not entirely specific for prostate because it has been found both at message and at protein levels in other tissues, such as normal bladder and colon and other types of tumors. In contrast, prostein is very specifically and highly expressed in almost all prostate tumors and normal prostate tissues. It is highly expressed in metastatic prostate cancers and is likely to be expressed on the cell surface. These features of prostein make it an ideal target for prostate cancer antibody therapy. Furthermore, the homogeneous expression pattern of prostein on all prostate glandular cells indicates that therapeutic regimens that use prostein could be very efficient and powerful because all prostatic cells can be targeted. Finally, prostein expression could potentially be used as a marker for in vivo imaging diagnosis to detect metastatic prostate cancer cells.
In addition to diagnostic and antibody therapeutic uses, immunotherapy strategies involving a vaccine targeting prostein are being evaluated. The immunological relevance of prostein and its potential as a vaccine candidate have been demonstrated by in vitro T-cell priming and stimulation experiments.4 The homology of prostein with a shared melanoma antigen demonstrated to elicit immune response provides further evidence for the potential of prostein to serve as a tumor antigen. The development of adjuvant and delivery systems to generate and expand strong immune responses to prostein could allow this protein to be developed as an effective prostate cancer-specific vaccine.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported in part by National Cancer Institute
Grant CA80518. 
2 To whom requests for reprints should be
addressed, at Corixa Corporation, 1124 Columbia Street, Suite 200,
Seattle, WA 98104. Phone: (206) 754-5798; Fax: (206) 754-5715;
E-mail: xu{at}corixa.com
3 The abbreviations used are: FACS,
fluorescence-activated cell-sorting; ORF, open reading frame; BPH,
benign prostate hyperplasia; PSA, prostate-specific antigen.
4 Michael Kalos, Thomas S. Vedvick, Steven G.
Reed, David H. Devsing, and Gary R. Fouger. Characterization of
10E3-G4-D3, a murine monoclonal antibody specific for prostein.
Manuscript in preparation.
5 http://www-genome.wi.mit.edu/cgi-bin/contig/rhmapper.pl.
Received 8/ 7/00. Accepted 12/13/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
in a mast cell-derived interleukin-4-dependent cell line.. Blood, 84: 189-199, 1994.
chain locus.. Proc. Natl. Acad. Sci. USA, 97: 9437-9442, 2000.This article has been cited by other articles:
|
|
 |
|
  W. H Walker and J. Cheng FSH and testosterone signaling in Sertoli cells Reproduction, July 1, 2005; 130(1): 15 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Maeda, S. Nagata, C. D. Wolfgang, G. L. Bratthauer, T. K. Bera, and I. Pastan The T Cell Receptor {gamma} Chain Alternate Reading Frame Protein (TARP), a Prostate-specific Protein Localized in Mitochondria J. Biol. Chem., June 4, 2004; 279(23): 24561 - 24568. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Duggan, J. J. McKnight, K. E. Williamson, M. Loughrey, D. O'Rourke, P. W. Hamilton, S. R. Johnston, C. C. Schulman, and A. R. Zlotta The Need to Embrace Molecular Profiling of Tumor Cells in Prostate and Bladder Cancer Clin. Cancer Res., April 1, 2003; 9(4): 1240 - 1247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Herness and R. K. Naz A Novel Human Prostate-specific Gene-1 (HPG-1): Molecular Cloning, Sequencing, and Its Potential Involvement in Prostate Carcinogenesis Cancer Res., January 15, 2003; 63(2): 329 - 336. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Xia, D. Robinson, A.-H. Ma, H.-C. Chen, F. Wu, Y. Qiu, and H.-J. Kung Identification of Human Male Germ Cell-associated Kinase, a Kinase Transcriptionally Activated by Androgen in Prostate Cancer Cells J. Biol. Chem., September 13, 2002; 277(38): 35422 - 35433. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
