| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Short Communications |


From the Departments of Pathology*
and
Molecular Oncology,
Genentech, Inc., South San
Francisco, California
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Targeted therapy for prostate cancer is a highly pursued avenue of cancer research with the goal to identify proteins that either show a tissue- or tumor-specific pattern of expression. Prostate stem cell antigen (PSCA) may represent such a protein because it has been reported to be highly expressed in a large percentage of human prostate tumors, including advanced and metastatic cases.4,5 Structurally, PSCA is a GPI-linked membrane glycoprotein with homology to members of the Thy-1/Ly-6 family of proteins. It is most closely related to Sca2, a cell surface marker for immature lymphocytes.6 Expression of human PSCA RNA in normal tissues was originally described not only as tissue-specific, but essentially limited to prostatic basal cells. It was inferred from this latter observation and from the structural homology to Sca2 that PSCA may represent a stem cell antigen.
Animal models are of crucial importance for understanding pathogenetic mechanisms and for the development of novel therapeutics. Thus, it is critical to analyze the expression of the murine orthologues of human genes of interest for cancer therapy. Murine and human PSCA are 70% identical at the nucleotide and the amino acid level.5 Although there are significant anatomical differences in the mouse prostate compared to that of the human there are inherent common characteristics including secretory function and hormonal regulation.5 One well-established and characterized animal model for human prostate cancer is the transgenic adenocarcinoma of the mouse prostate (TRAMP).7,8 In the TRAMP model, expression of the SV40 large T antigen is targeted to the prostatic epithelium by the minimal promoter of the rat probasin gene. All male TRAMP mice progress to prostatic intraepithelial neoplasia (PIN) or well-differentiated adenocarcinoma of the prostate usually by 8 to 12 weeks of age.8 The autochthonous TRAMP model is particularly relevant to human prostate carcinoma in that the development of the cancer is specific to the prostatic epithelium and is initially regulated by androgens. Furthermore, metastases to distant sites eventually occur.9,10
Here we report the pattern of expression of murine PSCA in fetal and adult tissues, as well as in a murine model for human prostate adenocarcinoma using a sensitive and stringent in situ hybridization technique. Our results suggest that the pattern of expression of murine PSCA in normal tissues and in tissues of murine prostatic adenocarcinoma is different from published data on human normal prostate, normal extra-prostatic tissues, and prostatic adenocarcinomas. The potential benefits of a mouse model for the evaluation of therapeutic targets for prostate cancer are discussed.
| Materials and Methods |
|---|
|
|
|---|
Male and female adult CD-1 mice between the ages of 10 and 24 weeks were obtained form Charles River Laboratories, Wilmington, MA. Pregnant mice were ordered for the collection of embryos. The day of isolation with copulatory plug was considered day 1 of embryonic development. Fetal tissues examined by in situ hybridization included: E10, E13, E14, E15, E17, and E18. Adult tissues examined included: tissues of the male urogenital system, female urinary bladder and kidney, pubescent mammary gland, 14-day pregnant mammary gland, lactating mammary gland, liver, heart, skin, and intestine. All tissues were fixed in 4% formalin and paraffin-embedded. Dr. Norman Greenberg (Baylor College of Medicine, Houston, TX) kindly provided breeder pairs of TRAMP transgenic mice. Urogenital tissues from wild-type littermate C57BL/6 were taken at age 12 and 24 for comparison to age-matched TRAMP transgenic mice. By the age of 12 weeks, TRAMP mice have typically progressed to PIN and or well-differentiated tumors. Nine TRAMP male mice between the ages of 12 to 39 weeks of age were sacrificed and tissues were collected for routine histology and in situ hybridization studies. The histological grade of prostate cancer was determined in accordance to previously published studies in the TRAMP model.10
Cloning of the Murine PSCA Orthologue
Full-length cDNA for murine PSCA was amplified from mouse 17-day embryo Marathon-Ready cDNA (Clontech, Palo Alto, CA). Primers were designed based on EST sequences present in GenBank. Sense primer (5' ACT ATG AAG CTT TGC AGC TCA TCC CTT CAC AAT CG 3') and anti-sense primer in the 3'-untranslated region (5' GAA TTC GGA TCC ACC ATG AAG ACC GTC TTC TTT CTC CTG CTG 3') were used resulting in a 420-bp fragment that was subsequently cloned into the polymerase chain reaction (PCR) subcloning vector PCR2.1TOPO (Invitrogen, Carlsbad, CA). Clones were confirmed by DNA sequencing.
In Situ Hybridization
PCR primers (upper: 5' CCT GCT GGC CAC CTA CT 3' and lower: 5' CCT TCA CAA TCG GGC TAT 3') were designed to amplify a 388-bp fragment of murine PSCA. Primers included extensions encoding 27 nucleotide T7 or T3 RNA polymerase initiation sites to allow in vitro transcription of sense or antisense probes, respectively, from the amplified products.11 Five-µm-thick sections were deparaffinized, deproteinated in 4 µg/ml of proteinase K for 30 minutes at 37°C, and further processed for in situ hybridization. 33P-UTP-labeled sense and antisense probes were hybridized to the sections at 55°C overnight. Unbound probe was removed by incubation in 20 mg/ml RNase A for 30 minutes at 37°C, followed by a high stringency wash at 55°C in 0.1x standard saline citrate for 2 hours, and dehydration through graded concentrations of ethanol. The slides were dipped in NBT2 nuclear track emulsion (Eastman Kodak, Rochester, NY), exposed in sealed plastic slide boxes containing desiccant for 4 weeks at 4°C, developed, and counterstained with hematoxylin and eosin. The in situ hybridization was routinely performed on duplicate sections with sense and antisense probes. Sections hybridized with the sense probe routinely did not show a specific hybridization signal.
| Results |
|---|
|
|
|---|
PSCA mRNA expression was examined in multiple stages of embryonic
development from E10 to E18. No expression was detected in any
histological structure at E10. At E14, PSCA expression was detected in
the epithelial lining of the urogenital sinus. Strong expression in the
superficial epithelium of the urogenital sinus was seen to be
maintained throughout the rest of fetal development (Figure 1, A and B)
. Expression was detected in
the adluminal epithelial cells of the renal pelvis (data not shown).
Expression was also observed in the fetal skin from E15 to E17, but
neither before or after this time window. Expression within the fetal
skin was confined to the outermost epithelial cell layer (Figure 1, C and D)
. PSCA expression was detected in the mucosal epithelium of the
gastrointestinal tract. Particularly high levels of expression were
observed in the anal canal (Figure 1, E and F)
; focal expression was
detected in the colon. Again, the hybridization signal in these sites
was restricted to the epithelial lining. Significant PSCA expression
was observed in the mucosa of the oral cavity, extending into the
oropharynx and into the esophagus (Figure 1, G and H)
. The epithelial
cells lining the inner aspect of the amnionic membrane were also
positive for PSCA (data not shown). In summary, PSCA expression was
limited to epithelial tissues. In multilayered epithelia, PSCA was
expressed in the most superficial cell layers.
|
The expression of PSCA in the adult was more restricted
compared with the fetus and was essentially limited to urogenital
tissues in the male and the urinary tract in the female. Unlike the
human gland the murine prostate is composed of several, histologically
distinct lobes (anterior, dorsal-lateral, and ventral). Murine PSCA was
focally expressed in all three prostatic lobes, with some acini being
completely negative and others showing scattered, positive cells. PSCA
was detected in the luminal cells (Figure 2, A and B)
; we found no evidence of PSCA
expression in the basal cell population (Figure 2A
, inset). Strong
expression was seen at the juncture between the large prostatic ducts
and the urethra. Individual luminal cells within the prostatic acini
did not show unique morphological features indicative of expression of
PSCA RNA. Seminal vesicle, vas deferens, and testis were completely
negative for PSCA expression.
|
Expression of PSCA in Prostatic Adenocarcinoma of TRAMP Mice
We examined murine PSCA expression in different stages of
progression of prostatic neoplasia in the TRAMP model. We evaluated
nine TRAMP animals, all showed morphological evidence of prostatic
neoplasia at various stages of tumor progression. The expression of
PSCA was increased in the neoplastic areas in all of the animals
examined (Table 1)
. A marked increase in
PSCA expression was detected in foci of PIN. These foci were
characterized by epithelial crowding or the formation of papillary
tufts lined by a mitotically active epithelium with an increased
nuclear:cytoplasmic ratio and mild to moderate nuclear pleomorphism
(Figure 3A)
. However, the increased
expression of PSCA was not uniformly observed in these foci of PIN
(Figure 3B)
. Areas of invasive adenocarcinoma also demonstrated a
significantly increased expression of PSCA compared to normal prostate.
Similar to the foci of PIN, the areas of increased expression followed
a patchy distribution within an individual tumor, with some areas
devoid of any expression (Figure 3, C and D)
. Cases of PIN and
well-differentiated adenocarcinoma showed no obvious correlation
between expression of PSCA and morphology of the malignant cells. Areas
with identical cellular morphology varied significantly in the level of
PSCA expression (Figure 3, E and F)
. Expression of PSCA was
undetectable in the two poorly differentiated cancers analyzed. Within
the PSCA-negative areas, glands involved by PIN were observed to have
high levels of PSCA expression (Figure 3, G and H)
indicating that the
lack of a signal in the poorly differentiated tumor was not because of
RNA degradation.
|
|
| Discussion |
|---|
|
|
|---|
The strongest expression of PSCA is detected in the urothelium of the developing and mature urinary tract. In the adult organism the expression appears continuous from the renal pelvis throughout the urethra. Prostatic expression is patchy, but increases noticeably at the juncture of the large prostatic ducts and urethra. It is tempting to speculate, whether this finding is related to urinary reflux into these structures providing a potential clue to the function of this protein. A common feature of the epithelial surfaces exhibiting PSCA expression is the fact that they line structures that are in continuous contact with fluids, mucous or secretions. This is obviously the case for urothelium, however, this notion also holds true for the observation that significant PSCA expression is present in the amnionic membrane, anal canal, oropharynx, esophagus, and skin of the developing fetus. Additional denominators must be at play, because other fluid-exposed epithelial tissues do not express PSCA, such as the fetal tracheobronchial tree. Whether the expression in fetal skin is temporally regulated is difficult to determine at this point as the number of embryos evaluated is small. However it seems that expression of PSCA in skin is no longer required in postnatal, extrauterine life. The pattern of expression within multilayered epithelial surfaces suggests that PSCA may function as a barrier or buffer against external insults. Based on the morphological evaluation it is difficult to determine whether PSCA would be primarily a barrier against chemical, physical, or microbial insults. Further studies, particularly the generation of PSCA knockout mice should clarify the primary function of PSCA in normal tissue.
Whether PSCA plays a role during carcinogenic events is purely speculative at this point. One could consider that expression of PSCA is turned on in proliferative epithelium during tissue repair and incidentally increased in the context of proliferative activity in transformed cells. All of the TRAMP tumors evaluated in this study show, when compared with normal prostate, an increased proportion of PSCA-positive cells. The functional state of the scattered PSCA-positive cells in normal murine prostate is unclear as they are morphologically indistinguishable from adjacent, PSCA-negative cells.
Increased expression of members of the Ly-6 family has been correlated to enhanced malignant potential in some murine tumor cells.12 Another member of the Ly-6 family, the E48 antigen, may have a role in cell adhesion in both normal and malignant squamous epithelia.13,14 However, the specific function of these genes in carcinogenesis has yet to be established. The correlation between the levels of PSCA expression and tumor stage and grade for prostate and bladder in human patients is presently controversial. Our own results, which are based on a small number of animals seem to suggest a negative correlation between expression of PSCA and advanced tumor stage in this murine model of prostatic carcinogenesis. Down-regulation of PSCA has been reported for poorly differentiated bladder cancers whereas high-grade and metastatic prostate carcinoma showed increased PSCA expression.4,15,16 In humans, PSCA is located on chromosome 8q, which is often amplified in metastatic and recurrent prostatic adenocarcinoma and considered to indicate a poor prognosis.17-19 Interestingly, PSCA is in close proximity to the c-myc proto-oncogene, which is amplified in >20% of recurrent and metastatic prostate cancers.20 Thus, overexpression of PSCA may not necessarily be causative in prostate carcinogenesis, but may occur incidentally in the context of amplification of contributing oncogenes, such as c-myc. Most likely, the differences in PSCA expression between the TRAMP tumors and human prostatic adenocarcinoma either indicate species-specific regulation of PSCA expression or are related to the type of carcinogenic event at play.
The rather selective expression of PSCA in fetal and adult tissues makes the mouse a suitable model to determine the physiological function of this protein. Although PSCA is lost in the advanced tumor stages the TRAMP model may represent a potentially useful tool to study PSCA as a potential target for prostate cancer. This model would allow investigators to study potential toxic effects of therapeutics targeting genes that are expressed not only in malignant prostate epithelium, but also in normal epithelial tissues of the urogenital tract or any other organ system.
| Acknowledgements |
|---|
| Footnotes |
|---|
Accepted for publication December 4, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. P. Tran, C. Lin, J. Yamashiro, and R. E. Reiter Prostate Stem Cell Antigen Is a Marker of Late Intermediate Prostate Epithelial Cells Mol. Cancer Res., December 1, 2002; 1(2): 113 - 121. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Jain, A. Lam, I. Vivanco, M. F. Carey, and R. E. Reiter Identification of an Androgen-Dependent Enhancer within the Prostate Stem Cell Antigen Gene Mol. Endocrinol., October 1, 2002; 16(10): 2323 - 2337. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Ross, S. D. Spencer, I. Holcomb, C. Tan, J. Hongo, B. Devaux, L. Rangell, G. A. Keller, P. Schow, R. M. Steeves, et al. Prostate Stem Cell Antigen as Therapy Target: Tissue Expression and in Vivo Efficacy of an Immunoconjugate Cancer Res., May 1, 2002; 62(9): 2546 - 2553. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Dubey, H. Wu, R. E. Reiter, and O. N. Witte Alternative Pathways to Prostate Carcinoma Activate Prostate Stem Cell Antigen Expression Cancer Res., April 1, 2001; 61(8): 3256 - 3261. [Abstract] [Full Text] |
||||
![]() |
T. Watabe, M. Lin, H. Ide, A. A. Donjacour, G. R. Cunha, O. N. Witte, and R. E. Reiter Growth, regeneration, and tumorigenesis of the prostate activates the PSCA promoter PNAS, January 8, 2002; 99(1): 401 - 406. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |