| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Department of Pediatrics, Oregon Health Sciences University, Portland, Oregon 97201
Address all correspondence and requests for reprints to: Ron G. Rosenfeld, M.D., Department of Pediatrics, Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, Oregon 97201. E-mail: rosenfer{at}ohsu.edu
Abstract
Neuroendocrine cells have been implicated in many cancers, including small cell lung, cervical, breast, and prostate carcinomas. The increase in neuroendocrine cell number in prostate cancer has been reported to correlate with poor prognosis, progressive tumors, and androgen insensitivity. The mechanisms involved in this differentiation remain unknown. IGF-binding protein-related protein 1 is a member of the IGF-binding protein superfamily and has recently been shown to exhibit differentiation and tumor suppression activity in prostate cancer cell lines stably overexpressing IGF-binding protein-related protein 1. From a yeast two-hybrid screen, a novel IGF-binding protein-related protein 1-interacting protein was identified. Immunocytochemical techniques indicate that this protein, 25.1, and intracellular IGF-binding protein-related protein 1 colocalize in the nucleus. When 25.1 is transiently expressed in a stable prostate cancer cell line overexpressing IGF-binding protein-related protein 1, cells assume a neuritic-like morphology with long dendritic-like processes and express the neuroendocrine markers chromogranin A and neuron-specific enolase. We propose that 25.1 (neuroendocrine differentiation factor) together with IGF-binding protein-related protein 1 can induce neuroendocrine cell differentiation in prostate cancer cells.
IN PATIENTS WITH prostate cancer, an increase in the neuroendocrine (NE) cell population correlates with an advanced prostate cancer that is androgen insensitive and, consequently, resistant to traditional modes of chemotherapy. NE cells have morphological similarities to neuronal cells and may secrete various bioactive neuropeptides, including bombesin, serotonin, TSH-like peptide, PTH-like peptide, and calcitonin-like peptide (1, 2). NE cells are, therefore, hypothesized to exert paracrine effects on the proliferation of neighboring cancer cells. In addition to the neuropeptides, NE cells secrete specific neuronal markers, such as chromogranin A (CgA) and neuron-specific enolase.
The progenitor of NE cells in prostate carcinoma appears to be an epithelial stem cell, based partly on the observation that there are basal, luminal, and intermediate NE cells types (3, 4). The mechanism(s) by which differentiation into NE cells occurs, however, is not well understood. Studies using xenograft models (5) as well as the androgen-sensitive human prostate cancer cell line, LNCaP (6), indicate that androgen withdrawal can induce NE differentiation. Further, in LNCaP cells, NE differentiation can be induced by epidermal growth factor (7) and IL-6 (8). The signaling pathway requires a functional ErbB2 (9) and activation of MAPK via elevation of cAMP (7). In support of these observations, treatment of cells with cAMP analogs, which activate cAMP-dependent protein kinase (PKA), induced NE differentiation (7, 10, 11), thus confirming one pathway involved in NE differentiation. The downstream effectors of this process, however, have not been identified either in vivo or in vitro.
Recently, a human prostate cancer cell line overexpressing IGF-binding protein-related protein 1 (IGFBP-rP1) was described as presenting NE-like cell morphology (12). The cDNA for IGFBP-rP1, a member of the IGFBP superfamily with significant N-terminal similarity to the IGFBPs (13), was first identified and cloned from normal leptomeningial cells (14). The biological function of IGFBP-rP1 is unknown, but it is capable of binding IGFs and insulin with low affinity (15, 16) and has also been implicated in senescence and tumor suppression (12, 17). Stable transfection of the cDNA for IGFBP-rP1 into malignant prostate epithelial cells results in decreased cell growth, increased apoptosis, poor growth in soft agar, and suppression of tumor formation in nude athymic mice (12).
In this report we introduce a novel protein, 25.1 [neuroendocrine differentiation factor (NEDF)], that interacts specifically with IGFBP-rP1 and leads to NE differentiation of the prostate cancer cell line overexpressing IGFBP-rP1 protein.
Materials and Methods
Materials and cell lines
Hs578T, MCF-7, and COS-7 cell lines (American Type Culture Collection, Manassas, VA) were maintained in DMEM-high glucose
with 10% FCS. All tissue culture media are from Life Technologies, Inc.-BRL (Grand Island, NY) unless otherwise
stated. human mammary epithelial cells (Clonetics, San Diego, CA) were
maintained as recommended by the manufacturer. M12:pcDNA3.1 and
M12:IGFBP-rP1 stable cell lines (gifts from Dr. Stephen Plymate,
University of Washington, Seattle, WA) (12) were
maintained in RPMI 1640 supplemented with 10 ng/ml epidermal growth
factor, 0.02 mM dexamethasone, 5 µg/ml insulin, 5 µg/ml
transferrin, 5% FCS, and 200 µg/ml G418 (Mediatech, Inc., Herndon,
VA); supplements were purchased from Sigma-Aldrich Corp.
(St. Louis, MO). Cell lines were incubated at 37 C in 5%
CO2. The 
-neuron-specific enolase and -Cg A
monoclonal antibodies were purchased from NeoMarkers, Inc. (Fremont,
CA); -hemagglutinin (-HA) and -M2 (against the
FLAG epitope) monoclonal antibodies were obtained from Roche Molecular Biochemicals (Indianapolis, IN); normal rabbit IgG was
purchased from Sigma-Aldrich Corp.
Yeast two-hybrid screen
Total RNA (5 µg) from MCF-7 breast cancer cells was employed to generate a cDNA library, using the Match Maker Yeast Two Hybrid cDNA Construction Kit (CLONTECH Laboratories, Inc., Palo Alto, CA). The bait was the full-length IGFBP-rP1 cDNA fused to the GAL4 DNA binding domain (Match Maker Yeast two-hybrid screening kit). Eighty-eight colonies were obtained from the initial screen, and of these, 65 were positive, with all 65 encoding the same cDNA. The cDNA contained approximately 650 bp, and a BLAST search of known databases (www.ncbi.nlm.nih.gov) indicated the sequence to be novel. The full-length clone, provisionally designated 25.1, was subsequently generated by rapid amplification of 5'-cDNA ends (Life Technologies, Inc.), and the resultant 850-bp cDNA was cloned and sequenced.
Northern blot
A multihuman tissue mRNA blot, containing 2 µg polyadenylase
RNA/lane from the indicated tissue, was purchased from CLONTECH Laboratories, Inc. The 25.1 cDNA was radiolabeled with
[-32P]deoxy-CTP (NEN Life Science Products, Boston, MA), using a random DNA priming kit (Prime It,
Stratagene, La Jolla, CA). The blot was hybridized with
Rapid Hybridization Solution (Amersham Pharmacia Biotech,
Piscataway, NJ) at 65 C overnight. The blot was washed as suggested by
the manufacturer and exposed to BioMax MS film (Eastman Kodak Co., Rochester, NY).
Transient transfections
For transient transfections, pcDNA3.1 (Invitrogen, San Diego, CA)-based plasmids with the indicated cDNA were transiently transfected into the various mammalian cell lines using the Fugene 6 reagent (Roche Molecular Biochemicals) according to the manufacturer’s suggestions. Plasmid concentrations, in the case of single or dual transfections, were standardized with control vector, pcDNA3.1. For neuroendocrine differentiation studies, the transfections were performed in six-well dishes with a maximum of 8 µg/well plasmid DNA for 6–18 h and then trypsinized and seeded into new dishes for subsequent experiments.
Generation of polyclonal 25.1 antibody
The cDNA of 25.1 was C-terminally epitope tagged with the DNA sequence for the HA nanopeptide (amino acids YPYDVPDYA) using PCR amplification. The 25.1HA protein was expressed in Escherichia coli BL21 cells as a fusion protein with glutathione-S-transferase (GST Gene Fusion System, Pharmacia Biotech). Crude bacterial lysates containing the fusion protein were prepared by sonication, proteins were solubilized in the presence of 1% Triton-X, and a reducing agent, dithiothreitol, was added. The fusion protein was purified using a glutathione-Sepharose column. The 25.1HA-fused peptide was released by thrombin protease digestion (there is a thrombin protease site at the fusion junction).
Polyclonal antibodies against 25.1 was generated in rabbits. Each of three New Zealand White female rabbits were sc injected with 100 µg purified protein mixed with an equal volume of Freund’s complete adjuvant (preimmune sera was collected 5 d before immunization). At 1-month intervals, two booster injections were preformed with the same concentration of protein in incomplete Freund’s adjuvant. The rabbits were killed 1 month after the last booster, and serum was collected from whole blood. A 25.1-cyanogen bromide-activated Sepharose column (Pharmacia Biotech) was employed to further purify the IgG fraction of specific anti-25.1 antibody.
Immunoprecipitation and Western immunoblot
Cells lysates for coimmunoprecipitation experiments were
prepared by sonication, and insoluble material was removed by
centrifugation. Protein complexes were immunoprecipitated (Ip) from
total cell lysates with primary antibody [-25.1 or -IGFBP-rP1
polyclonal antibodies (18), normal rabbit IgG, or as
indicated] preincubated with protein A-Sepharose beads (Amersham Pharmacia Biotech). The protein complexes were eluted from the
beads by boiling in the presence of reducing SDS-PAGE sample buffer. In
other experiments, total cell lysates were prepared in RIPA buffer
[150 mM NaCl, 20 mM HEPES (pH 7.5), 1%
Triton-100, 1% sodium deoxycholate, 0.1% SDS, and a cocktail of
protease inhibitors]. Equal protein concentrations were resolved on
SDS-PAGE, transferred to nitrocellulose membrane, and immunoblotted
with the indicated antibody.
For Western immunoblots, proteins were separated on 12% SDS-PAGE and transferred to nitrocellulose membranes (Hybond ECL, Amersham Pharmacia Biotech). Membranes were blocked with 4% nonfat evaporated milk in Tris-buffered saline-0.1% Tween-20 and incubated with the indicated primary antibody. After overnight incubation at 4 C, blots were washed and incubated for 1 h at 22 C with a 1:3000 dilution of horseradish peroxidase-linked antirabbit IgG (or antimouse IgG) secondary antibodies. Specific proteins were detected with enhanced chemiluminescence reagents according to the manufacturer’s protocol (Renaissance, NEN Life Science Products).
Immunocytochemistry
For immunofluorescence studies, cells were grown in 24-well
dishes for the indicated times and with the indicated treatments. The
cells were fixed with 4% paraformaldehyde for 10 min at room
temperature and permeabilized with 50% methanol/50% acetone for 1–2
min at room temperature. The cells were blocked in 0.25% normal goat
serum in PBS and 0.1% Triton X-100. Incubations with the respective
primary antibodies (-25.1, -HA, or -IGFBP-rP1) were performed
overnight at 4 C. Each well was rinsed three times in PBS, and cells
incubated with fluorescent-conjugated secondary antibodies (goat
antirabbit IgG or goat antimouse IgG conjugated to either fluorescein
or Texas Red, Molecular Probes, Inc., Eugene, OR) and
Hoechst stain (1 µg/ml) for 1–2 h at room temperature. Each well was
washed three times with PBS and viewed under a Nikon
inverted phase, fluorescent microscope with an analog camera
(Nikon, Melville, NY).
Results
Identification of a novel IGFBP-rP1-interacting protein, 25.1
The yeast two-hybrid system was employed to identify
IGFBP-rP1-interacting proteins. A cDNA library from MCF-7 breast cancer
cells was screened, with IGFBP-rP1 cDNA as bait. One novel cDNA was
subsequently identified and initially designated 25.1. The 850-bp cDNA
encodes a peptide of 223 amino acids, which contains a putative nuclear
localization (Fig. 1A
). Hybridization of
human multitissue Northern blots with the 25.1 cDNA indicates
ubiquitous expression of 25.1 mRNAs, with three distinct transcripts
(1.1, 3.4, and 8.5 kb; Fig. 1B). Expression of the three transcripts
appears to be differentially regulated, with considerable tissue to
tissue variability. The GenBank accession number for 25.1 (NEDF, see
below) is AF219226.
|
, the affinity-purified -25.1 antibody recognizes E.
coli-generated 25.1 (20 ng) as a single protein of 29 kDa, and
baculovirus-generated 25.1 (cell lysate) as a doublet. In Western
immunoblots of cell lysates from breast and prostate cancer cell lines,
the 25.1 protein appears as a doublet of about 29 and 31 kDa (Fig. 1C).
Immunoblotting with antibody preabsorbed with E.
coli-generated 25.1 protein demonstrated both bands to be 25.1
specific (data not shown). Interaction between IGFBP-rP1 and 25.1
Coimmunoprecipitation experiments were employed to
confirm the interaction between IGFBP-rP1 and 25.1 (Fig. 2). To determine the specificity of
coimmunoprecipitation, cell lysates from COS-7 cells cotransfected with
both pcDNA3.1:25.1HA and
pcDNA3.1:IGFBP-rP1FLAG were Ip with normal rabbit
IgG or with -25.1 and immunoblotted with -M2 (anti-FLAG). As
expected, FLAG-tagged IGFBP-rP1 protein coimmunoprecipitated only when
-25.1, but not normal rabbit IgG, was used for Ip (Fig. 2A).
|
). Expression of recombinant proteins in transiently
transfected COS-7 cells was confirmed by Ip with -IGFBP-rP1 antibody
and immunoblotting with -M2 (Fig. 2B, lanes c' and d', upper
panel) or Ip with -25.1 antibody and immunoblotting with -HA
(Fig. 2B, lanes b' and d', lower panel). Results indicate
that coexpression of IGFBP-rP1FLAG and
25.1HA (lane d') was equivalent to the expression
of either protein alone (lanes b' and c'). Cell lysates were then
analyzed for the presence of IGFBP-rP1–25.1 complexes; -IGFBP-rP1
antibody or -25.1 antibody was used for Ip, and the precipitate was
immunoblotted with -HA or -M2, respectively. In the absence of
either IGFBP-rP1FLAG or
25.1HA recombinant proteins in the total cell
extracts, specific immunoreactive bands were not detected with either
-M2 or -HA antibodies (lane a). Endogenous IGFBP-rP1 permitted
co-Ip of transiently expressed 25.1HA, detected
with -HA (lane b); endogenous 25.1 protein permitted coIp of
transiently expressed IGFBP-rP1FLAG, detected
with -M2 (lane c); concurrent transient expression of
25.1HA and IGFBP-rP1FLAG
permitted co-Ip of both 25.1HA and
IGFBP-rP1FLAG by either antibody (lane d).
Similar results were obtained in other cell lines (Hs578T or MCF-7
cells) transiently transfected with the same panel of plasmids (data
not shown).
Interestingly, transiently expressed 25.1HA
protein is detected as a doublet, consistent with the doublet protein
bands corresponding to endogenous 25.1 (Fig. 1C). Furthermore, it
appears that IGFBP-rP1, both endogenous and the recombinant FLAG-tagged
protein, coprecipitates preferentially with the larger band of the 25.1
doublet.
Cellular localization of IGFBP-rP1 and 25.1
Immunofluorescent studies of endogenous 25.1 and IGFBP-rP1
proteins in mammary and prostate cells indicate that cellular
localization of both proteins is predominantly nuclear (Fig. 3). The immunofluorescent signals were
reduced to background when the respective antibodies were first
preabsorbed with either 25.1HA protein or
IGFBP-rP1FLAG protein (data not shown),
indicating that localization was specific.
|
-IGFBP-rP1 showed a punctate nuclear
distribution of the protein (Fig. 3A). In prostate cells overexpressing
IGFBP-rP1 (M12:IGFBP-rP1) and in prostate cells expressing very little
secreted IGFBP-rP1 (M12:pcDNA3.1), nuclear distribution was again noted
(Fig. 3A). Similar to IGFBP-rP1, the signal for 25.1 was nuclear and
punctate (Fig. 3A). No cytoplasmic staining was detectable, and 25.1
protein was not detected by immunoblotting techniques in conditioned
medium from any breast or prostate cell line tested to date. To further
substantiate that 25.1 can translocate to the nucleus, Hs578T cells, an
E-nonresponsive breast cancer cell line, were transiently transfected
with pcDNA3.1:25.1HA, and trafficking of
25.1HA protein was followed. Immunofluorescence
of these transfected cells showed that transiently expressed
25.1HA protein clearly translocates to the
nucleus, as demonstrated with both -HA and -25.1 antibodies (Fig. 3B). At 0 h, only endogenous 25.1 was detectable with -25.1
antibody (secondary antibody was goat antirabbit IgG conjugated to
fluorescein) and was localized to the nucleus. The transiently
expressed 25.1HA protein, was detectable at
6 h, and by 12 h had clearly translocated from the cytoplasm
to the nucleus (-HA, with goat antimouse IgG conjugated to Texas Red
as the secondary antibody). Overexpression of IGFBP-rP1 and 25.1 in prostate cancer cells induces neuroendocrine differentiation
To investigate the biological role of 25.1, we used the prostate epithelial cancer cell line, M12, stably overexpressing IGFBP-rP1 (M12:IGFBP-rP1) (12). In comparison to wild-type M12 cells or to M12 cells stably transfected with vector alone, M12:IGFBP-rP1 cells have been shown to acquire an elongated morphology with multiple projections (12). When grown in defined medium, they displayed a significantly reduced growth rate compared with that of the parental cell line transfected with vector, pcDNA3.1 (M12:pcDNA3.1). Recent data further demonstrated that in nude athymic mice, the IGFBP-rP1-overexpressing stable cells have significantly reduced tumorigenic potential compared with control transfected cells (12).
The mechanism for these morphological and biological changes in
M12:IGFBP-rP1 stable cells is unclear, but could involve an interaction
between IGFBP-rP1 and endogenous 25.1. To investigate this possibility,
M12:pcDNA3.1 and M12:IGFBP-rP1 cells were transiently transfected with
25.1HA cDNA and examined by immunofluorescence
(Fig. 4A). M12:pcDNA cells overexpressing
25.1HA protein demonstrated minimal morphological
changes. In contrast, M12:IGFBP-rP1 cells exhibited further changes
when overexpressing 25.1HA protein. Long,
dendritic-like processes developed, and cells assumed a neuritic
appearance, suggesting that M12:IGFBP-rP1 had further differentiated
into neuroendocrine-like cells upon transient overexpression of the
25.1 protein.
|
). This phenomenon thus appears
to require both IGFBP-rP1 and 25.1 to be present and to be
overexpressed.
To determine whether M12:IGFBP-rP1 cells transiently transfected with
25.1 cDNA are phenotypically NE cells, they were examined for
expression of the NE cell marker CgA. In immunofluorescent studies
employing anti-CgA monoclonal antibodies, only cells overexpressing
25.1 were also positive for CgA (Fig. 5).
Similar results were obtained with the NE marker, neuron-specific
enolase (data not shown).
|
Cellular differentiation involves a cascade of events that results in changes in cellular morphology and a reduced rate of proliferation. The M12:IGFBP-rP1 prostate cancer cells stably overexpressing IGFBP-rP1 exhibit many of the early stages of NE differentiation. Employing a yeast two-hybrid screen, we have now identified a novel IGFBP-rP1-interacting protein, 25.1, which mediates the conversion to a prostate neuroendocrine cell phenotype. On this basis, we propose the designation of 25.1 as NEDF.
Endogenous NEDF/25.1 appears to be expressed in mammalian cells as a doublet of 29 and approximately 31 kDa. The same doublet was detected when HA-tagged NEDF/25.1 cDNA was transiently expressed in mammalian cells, but not when expressed in E. coli, suggesting posttranslational modification of NEDF/25.1 in mammalian systems. The nature of the modification is unknown. In baculovirus, NEDFFLAG was also expressed as a doublet, although the larger of the two protein bands differed from that observed in the mammalian system. This would be consistent with the different posttranslational modification machinery between insect and mammalian systems.
Interactions between IGFBP-rP1 and NEDF/25.1 were readily demonstrable
by coimmunoprecipitation. The interaction was specific, and when both
IGFBP-rP1FLAG and
NEDF/25.1HA proteins were overexpressed, a 3-fold
increase in the IGFBP-rP1FLAG-
NEDF/25.1HA complex was detected in cell lysates
immunoprecipitated with -IGFBP-rP1 antibody. Surprisingly, a similar
increase was not observed when immunoprecipitating with -25.1
antibody, and, in fact, there appeard to be a decrease in detectable
IGFBP-rP1FLAG- NEDF/25.1HA
complexes. It is not clear why this is the case, although it is
possible that the polyclonal antibody generated against the recombinant
NEDF/25.1HA protein can disrupt
IGFBP-rP1-NEDF/25.1HA interactions. Not all the
expressed recombinant proteins (<50% by densitometric determination)
participate in the interaction, suggesting that other parameters, such
as correct cellular colocalization of the two proteins, are involved.
Optimal interaction may also be a regulated process (see below).
Interestingly, IGFBP-rP1 (native and transiently expressed FLAG-tagged
protein) appears to interact preferentially with the larger, presumably
posttranslationally modified, form of NEDF/25.1.
The potential biological effect of interactions between IGFBP-rP1 and NEDF was observed in M12 prostate cancer cells. Overexpression of both IGFBP-rP1 and NEDF results in NE terminal differentiation of M12 cells, with many features characteristic of neuronal cells, including dendritic arms that promote cell to cell contact and the presence of neuron-specific markers, such as CgA and neuron-specific enolase. The mechanism(s) by which IGFBP-rP1 and NEDF interact in this cell background to promote NE differentiation is unclear. Both proteins localize to the nucleus, although IGFBP-rP1 is also clearly a secreted protein (15). Newly synthesized NEDF protein is immediately targeted to the nucleus. In contrast, the sequence of events leading to nuclear localization of IGFBP-rP1 is unclear and may include initial secretion of IGFBP-rP1 into the extracellular milieu, with subsequent uptake and transport to the nucleus. However, the fact that secreted IGFBP-rP1 is not detectable in some cells, such as M12:pcDNA3.1 and MCF-7 cells, suggests that another mechanism(s) for nuclear targeting of IGFBP-rP1 exists. Analysis of the amino acid sequences of IGFBP-rP1 and NEDF proteins suggest the presence of putative nuclear localization signals (K89SRKRRKGK97 in IGFBP-rP1; K45RSVKDAAKK54 in NEDF). Whether these are functional nuclear localization signal sequences has yet to be demonstrated. The colocalization of IGFBP-rP1 and NEDF in the nucleus provides opportunities for the two proteins to interact. Further investigations are necessary to determine whether the interaction is a regulated process that results in NE differentiation.
All of the cell lines examined to date demonstrate the presence of both nuclear IGFBP-rP1 and NEDF. It is therefore intriguing that the distinctive morphological effects of NE differentiation were observed only in M12:IGFBP-rP1 cells, not in control M12:pcDNA3.1 cells, COS-7 cells, or the breast cancer cell lines Hs578T and MCF-7 (data not shown). Hs578T cells, in particular, secrete IGFBP-rP1 protein at levels at least equivalent to those seen in CM of M12:IGFBP-rP1 cells (Wilson, E. M., unpublished). We, therefore, speculate that in M12:IGFBP-rP1 cells, an interaction is specifically triggered between IGFBP-rP1 and a limited amount of endogenous NEDF, resulting in growth suppression and partial activation of the differentiation pathway. The transient overexpression of NEDF further amplifies terminal NE differentiation in these prostate cells. Hence, the presence of both IGFBP-rP1 and NEDF appears to result in synergistic activation of NE differentiation. As both IGFBP-rP1 and NEDF can be found in the nucleus, it is possible that the resulting complex could be involved in transcriptional regulation of the genes necessary for NE differentiation. It is of note that several of the high affinity IGF binders, the IGFBPs, which share the amino-terminal motif found in the IGFBP-rPs, have also been shown to translocate to the nucleus (19, 20, 21, 22).
The morphological effects of transiently overexpressing NEDF/25.1 in M12:IGFBP-rP1 could also be a result of indirect effects. For example, excess NEDF/25.1 in the cytoplasm may perturb the stability of proteins involved in modulating signaling pathways critical for determination of cell morphology, such as the focal adhesion kinase pathways (23); it is noted that excess cytoplasmic GFP does not cause the same effect. Excess NEDF/25.1 in cytoplasm of other cell lines tested, including Hs578T cells, does not generate the same morphological changes. Simultaneously, it remains unclear why NE differentiation of M12 cells requires both overexpression of NEDF and IGFBP-rP1. Clearly, the specificity and mechanism(s) of NEDF NE differentiation action warrant further investigation.
It has been proposed that all three cell types comprising the prostate epithelium (epithelial, basal, and neuroendocrine cells) have a common endodermal origin. Alternative models suggest a neurogenic origin for prostatic NE cells. An increase in NE cell number in prostatic tumors has been shown to correlate with progressive cancer, poor prognosis, and androgen insensitivity (1). A transgenic mouse model, generated using the mouse cryptdin-2 gene to direct expression of simian virus 40 T antigen in a subset of prostatic NE cells, resulted in a 100% incidence of metastatic prostate cancer independent of androgen activity (24). Terminal NE differentiation has been induced in two prostatic cell lines, LNCaP and PC-3M, by increasing intracellular levels of cAMP, possibly signaling through the MAPK pathway (7, 10, 11, 25). In our studies in prostatic M12 cells stably transfected with IGFBP-rP1, coexpression of NEDF resulted in pronounced NE differentiation, with the characteristic morphological changes and positive staining for neuron-specific markers. Further studies will be required to determine whether the IGFBP-rP1-NEDF pathway is involved in cAMP-stimulated NE differentiation. It is of interest that NE cells have been noted in many other cancers, such as lung, breast, and cervical (1), and IGFBP-rP1-NEDF/25.1 interactions in these systems need to be further investigated. Given the striking effects of IGFBP-rP1 in prostate cancer (12) and the identification of a novel interacting protein involved in the neuroendocrine differentiation of prostate cancer cells, new insights concerning the pathophysiology, diagnosis, and treatment of prostate cancer should emerge.
Acknowledgments
Footnotes
This work was supported by NIH Grants CA-58110 and DK-51513 and U.S. Army Grants DAMD 17-96-2604 and DAMD 17-97-17204.
Abbreviations: CgA, Chromogranin A; GFP, green fluorescent protein; IGFBP-rP1, IGF-binding protein-related protein 1; HA, hemagglutinin; Ip, immunoprecipitated, immunoprecipitation; NE, neuroendocrine; NEDF, neuroendocrine differentiation factor.
Received August 7, 2000.
Accepted May 15, 2001.
References
QL)-mediated signaling increases
invasiveness and tumorigenicity of PC-3M prostate cancer cells. Oncogene 18:3376–3382[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  M. Kutsukake, R. Ishihara, M. Yoshie, H. Kogo, and K. Tamura Involvement of insulin-like growth factor-binding protein-related protein 1 in decidualization of human endometrial stromal cells Mol. Hum. Reprod., October 1, 2007; 13(10): 737 - 743. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Salido, J. L. Gonzalez, and J. Vilches Loss of mitochondrial membrane potential is inhibited by bombesin in etoposide-induced apoptosis in PC-3 prostate carcinoma cells Mol. Cancer Ther., April 1, 2007; 6(4): 1292 - 1299. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Fang, E. M. Kofoed, B. M. Little, X. Wang, R. J. M. Ross, S. J. Frank, V. Hwa, and R. G. Rosenfeld A Mutant Signal Transducer and Activator of Transcription 5b, Associated with Growth Hormone Insensitivity and Insulin-Like Growth Factor-I Deficiency, Cannot Function as a Signal Transducer or Transcription Factor J. Clin. Endocrinol. Metab., April 1, 2006; 91(4): 1526 - 1534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Hwa, B. Little, E. M. Kofoed, and R. G. Rosenfeld Transcriptional Regulation of Insulin-like Growth Factor-I by Interferon-{gamma} Requires STAT-5b J. Biol. Chem., January 23, 2004; 279(4): 2728 - 2736. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Diaz, Y. H. Yang, T. Ferreira, K. C. Loh, Y. Okazaki, Y. Hayashizaki, M. Tessier-Lavigne, T. P. Speed, and J. Ngai Analysis of gene expression in the developing mouse retina PNAS, April 29, 2003; 100(9): 5491 - 5496. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
