INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.
ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, Mi 48106-1346 USA 800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EFFECTS OF NORMAL AND NEOPLASTIC CANINE PROSTATE TISSUE ON
BONE FORMATION AND INVESTIGATIONS ON THE ORIGIN OF CANINE
PROSTATE CARCINOMA
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
by
Bruce E. LeRoy, DVM
* * * * *
The Ohio State University
2002
Dissertation Committee: Approved by Thomas J. Rosol, Adviser
Robert R. Bahnson
William C. Kisseberth Department of Veterinary Biosciences
Gary J. Kociba
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3072913
______® UMI
UMI Microform 3072913 Copyright 2003 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT
Prostate cancer is a common cause of cancer in older men. Prostate cancer
commonly metastasizes to bone and induces new woven bone formation (osteoblastic
metastases), likely caused by paracrine effects of factors produced by prostate cancer
cells and released from the bone matrix. There are few models of prostate cancer
osteoblastic metastases. Canine prostate cancer is a good model for human disease. To
develop a new model of osteoblastic metastases, we investigated the effect of
hyperplastic canine prostate tissue on bone formation in the calvarium (skull cap) of
nude mice. Canine prostate induced new bone formation with minor bone lysis and no
fractures. Osteoclasts were increased. To identify factors involved in the new bone
formation, we treated organ cultures of rat calvaria with canine prostate homogenate.
Prostate homogenate stimulated calvarial alkaline phosphatase activity 4 to 6-fold
greater than controls. Alkaline phosphatase stimulation was prevented by pretreatment
with endothelin receptor antagonists, but not anti-parathyroid hormone-related protein
or indomethacin. The effect of neoplastic prostate tissue on bone was investigated using
a canine prostate carcinoma xenograft. The xenografts caused bone formation, bone
lysis, and lung metastasis in nude mice. The tumor cells expressed PTHrP, cathepsin K,
and keratin 7. We also investigated keratin 7 and arginine esterase as candidate markers
to distinguish canine prostate carcinoma from bladder carcinoma. Keratin 7 expression ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was very similar in neoplastic prostate and bladder tissues. Arginine esterase enzyme
activity was present in high levels in normal and some neoplastic prostate tissue, but
low in normal and neoplastic bladder tissue. Arginine esterase may be a useful marker
to distinguish these carcinomas.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dedicated to my wife Susan and daughter Madison, the lights of my life
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGMENTS
As with any large project, this work could not have been performed without the help of many people. However, the contributions of my adviser, Dr. Thomas J. Rosol,
simply cannot be overstated. Dr. Rosol has been supportive, encouraging, motivating, innovative, and most importantly, generous with his time. He has been, for me, the perfect graduate adviser. He has given me a great gift - he has shown me how to succeed. I am eternally in his debt, and my goal is for my future contributions to veterinary pathology and biomedical research to be a worthy legacy for him. I would like to thank my committee members and Rosol Lab Mates for their
time, support, encouragement, and companionship. I would also like to thank Drs. Nate Collins and Eric Blomme. I was lucky enough to enter the graduate program while they were still here, and they provided an
invigorating melange of rigorous scientific thought, humor, and friendship that changed
my life. Other people I would like to acknowledge are Anne Saulsbery, Evelyn Handley, and Andrea Levine. Kathy Hopwood and Christie Newland provided excellent animal care, and Tim Vojt gave outstanding assistance with my visual data.
Finally, thank you Maxey, for board preparation help above and beyond the call of duty, but mostly thanks just for being you. I shall never forget your friendship.
V
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA
March 3, 1963...... Born - Orlando, Florida, USA
1985 ...... Bachelor of Science (Agriculture) University of Georgia Athens, GA
1989 ...... Doctor of Veterinary Medicine University of Georgia Athens, GA
1995-1998 ...... Residency in Veterinary Clinical Pathology, The Ohio State University, Columbus, OH
1998-present ...... Graduate Research Associate Department of Veterinary Biosciences The Ohio State University
2000...... Diplomate, American College of Veterinary Pathologists
PUBLICATIONS
Research Publications
1. B.E. LeRoy, R.R. Bahnson, and T. J. Rosol, “Canine Prostate Induces New Bone Formation in Mouse Calvaria: A Model of Osteoinduction by Prostate Tissue.” The Prostate, 50, 104, (2002).
2. N.C. Collins and B.E. LeRoy, “Artifactually Increased Serum Bicarbonate Values in Two Horses and a Calf with Severe Rhabdomyolysis.” Veterinary Clinical Pathology, 27, 85, (1998).
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. B.M. Pressler, D.S. Rotstein, J.M. Law, T.J. Rosol, B.E. LeRoy, B.W. Keene, and M.W. Jackson, “Hypercalcemia and High Parathyroid Hormone-Related Protein Concentration Associated with Malignant Melanoma in a Dog.” J Am Vet Med Assoc., 221, 263, (2002).
FIELDS OF STUDY
Major Field: Veterinary Biosciences
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS
Page Abstract...... ii
Dedication ...... iv
Acknowledgments ...... v
V ita...... vi
List of Tables...... x
List of Figures...... xi
Chapters:
1. Canine Prostate Induces New Bone Formation in Mouse Calvaria ...... 1
Abstract...... Introduction ...... 2 Materials and Methods ...... 3 Results...... 5 Discussion ...... 7 Literature Cited ...... 12
2. Canine Prostate Carcinomas Express Markers of Urothelial and Prostatic Differentiation ...... 21
Abstract...... 21 Introduction ...... 22 Materials and Methods ...... 25 Results...... 31 Discussion ...... 37 Literature Cited ...... 41
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. Canine Prostate Stimulates Osteoblast Function Using Endothelin Receptors.. 59
Abstract...... 59 Introduction ...... 60 Materials and Methods ...... 62 Results...... 68 Discussion ...... 72 Literature Cited ...... 78
4. New Bone Formation and Osteolysis by a Metastatic, Highly Invasive Canine Prostate Carcinoma Xenograft ...... 90
Abstract...... 90 Introduction ...... 90 Materials and Methods ...... 92 Results...... 99 Discussion ...... 103 Literature Cited ...... 106
Bibliography...... 122
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES
Table Page
2.1 Arginine esterase nested RT-PCR primer sequences and the predicted amplicon product size ...... 57
2.2 Histologic type, keratin 7 expression, and arginine esterase activity of 4 canine prostate carcinomas ...... 58
X
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES
Figure Page
1.1 Prostate implanted calvarium with periosteal new bone formation and bone resorption ...... 17
1.2 Control and prostate-implanted calvarium with new bone formation and calcein fluorescence ...... 18
1.3 Periosteal new bone formation and calcein fluorescence in undecalcified glycol methacrylate...... 19
1.4 Box plot of calvarial thickness ...... 20
1.5 Box plot of osteoclasts per mm ...... 2 1
2.1 Keratin 7 staining of normal canine prostate ...... 44
2.2 Canine prostate carcinoma ...... 45
2.3 Keratin 7 staining of prostate carcinoma ...... 46
2.4 Keratin 7 staining of normal canine bladder urothelium ...... 47
2.5 Canine transitional cell carcinoma ...... 48
2.6 Keratin 7 staining of transitional cell carcinoma ...... 49
2.7 Bar chart showing distribution of keratin 7 staining in 17 prostate and 19 transitional cell carcinomas ...... 50
2.8 Bar chart showing intensity of keratin 7 staining in 17 prostate and 19 transitional cell carcinomas ...... 51
2.9 Heterogenous appearance of canine prostate carcinoma ...... 52
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.10 Bar chart of Northern blot expression of arginine esterase mRNA in normal and neoplastic prostate ...... 53
2.11 Expression of arginine esterase mRNA by nested RT-PCR in prostate and transitional cell carcinomas ...... 54
2.12 Line graph demonstrating differences in arginine esterase activity between normal prostate and bladder homogenates ...... 55
2.13 Bar chart of differences in arginine esterase activity in normal and neoplastic bladder and prostate tissues ...... 56
3.1 Northern blot demonstrating endothelin-1 mRNA expression by canine prostate epithelial cells...... 81
3.2 Endothelin-1 protein levels in prostate cell conditioned medium and tissue homogenates...... 82
3.3 Dose-response curve of alkaline phosphatase activity following treatment of rat calvaria with canine prostate homogenate ...... 83
3.4 Time course study of alkaline phosphatase activity following treatment of rat calvaria with canine prostate homogenate ...... 84
3.5 Box plot demonstrating alkaline phosphatase activity of calvarial osteoblasts following 24 hrs of treatment with homogenate from 3 different dogs ...... 85
3.6 Bar chart of efficacy of various compounds for preventing stimulation of calvarial alkaline phosphatase activity by canine prostate homogenate ...... 86
3.7 Dose-response of endothelin AB receptor antagonist prevention of stimulation of alkaline phosphatase by canine prostate homogenate ...... 87
4.1 Histologic appearance of canine prostate carcinoma used to establish ACE-1 xenograft...... 110
4.2 Histology of ACE-1 xenograft in nude mouse ...... I ll
4.3 Subcutaneous cyst in ACE-1 xenograft ...... 112
4.4 Heterogenous morphology of ACE-1 xenograft...... 113
4.5 Lung metastasis of ACE-1 xenograft...... 114
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.6 Rib from ACE-1 xenograft tumor with periosteal new bone ...... 115
4.7 Normal mouse vertebrae and vertebrae with new bone formation and bone destruction from an ACE-1 xenograft ...... 116
4.8 Ultrastructural appearance of normal prostate and ACE-1 xenograft ...... 117
4.9 Radiograph of new bone formation and bone lysis in a mouse with intra- tibial injection of ACE-1 cells ...... 118
4.10 Plasma total calcium levels from mice with ACE-1 xenograft ...... 119
4.11 Densitometric analysis of Northern blot of arginine esterase mRNA expression by normal prostate, primary carcinoma, and ACE-1 xenograft ...... 120
4.12 Arginine esterase activity in normal prostate, primary carcinoma, and ACE-1 xenograft...... 121
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1
INTRODUCTION
ABSTRACT
Osteoblastic metastases are common in patients with advanced prostate cancer.
The pathophysiology of the new bone formation at metastatic sites is not currently
known, but it is hypothesized that growth factors secreted by the prostate may be
involved. Unfortunately, most rodent models of prostate cancer with metastasis to bone
are osteolytic and not osteoblastic. Significant osteolysis by tumor cells at metastatic
sites also may lead to fractures or bone instability. Misinterpretation of new periosteal
bone due to bone instability as tumor-cell osteoinduction is another disadvantage of the
osteolytic models. To circumvent these problems, we have developed a model system of
new bone formation in the calvaria of nude mice stimulated by normal canine prostate
tissue. Collagenase-digested normal prostate tissue was implanted adjacent to the
calvaria of nude mice. Calvaria were examined at 2 weeks post-implantation for
changes in the bone microenvironment by histology, calcein uptake at sites of bone
mineralization, and tartrate-resistant acid phosphatase staining for osteoclasts. The
prostate tissue remained viable and induced abundant new woven bone formation on the
adjacent periosteal surface. In some cases new bone formation also was induced on the
distant or concave calvarial periosteum. The new bone stained intensely with calcein,
l
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. which demonstrated mineralization of the bone matrix. The new bone formation on
prostate-implanted calvaria significantly increased (2.0-fold) the thickness of the
calvaria compared with control calvaria. New bone formation was not induced in
calvaria of mice implanted with normal canine kidney, urinary bladder, spleen, or
skeletal muscle tissue, or mice with surgically-induced disruption of the periosteum.
Osteoclast numbers in the medullary spaces and periosteum of calvaria were mildly
increased (61%) in mice with implanted prostate tissue. This animal model will be
useful for investigating the roles of prostate-derived growth factors on new bone
formation in vivo.
A unique characteristic of the skeletal metastases of human prostate cancer is
their ability to induce osteoblastic lesions, which are predominated by new woven bone
formation on medullary trabeculae and variable degrees of osteoclastic bone resorption
[1]. The bone metastases of other carcinomas, such as breast, colon, and lung
carcinomas, usually manifest as primarily osteolytic lesions, although a small
percentage of mammary gland carcinoma metastases also may contain an osteoblastic
component [2]. The tissue-specific mechanisms responsible for the unique
osteosclerotic appearance of prostate carcinoma metastases are not known. It has been
hypothesized [3;4] that at metastatic sites there is a complex paracrine relationship
between the prostate carcinoma, osteoblasts and osteoclasts, and bone-matrix associated
growth factors leading to a final common pathway of osteoblast activation and new
woven bone formation. Numerous factors produced by prostate cells, both normal and
neoplastic, have the potential to stimulate new bone formation by effects on both
osteoblasts and osteoclasts. Included in this group are parathyroid hormone-related
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. protein (PTHrP) [5], transforming growth factor-P (TGF-P) [6], urokinase-type
plasminogen activator [7], bone morphogenetic proteins [8;9], and endothelin-1 [10]. In
addition, tumor-induced osteolysis of the local bone microenvironment at metastatic
sites may result in the release of bone matrix-associated growth factors, such as TGF-P
and insulin-like growth factor [3]. Both of these factors are capable of influencing new
woven bone formation by stimulating mitosis and bone formation by osteoblasts, as
well as providing a paracrine source of mitogens for metastatic prostate carcinoma cells.
The production of bioactive factors by prostate carcinoma cells, coupled with the
release of growth factors during osteolysis, likely results in a milieu of bioactive
molecules capable of stimulating new woven bone formation by osteoblasts, and
possible suppression of osteoclastic activity, such as by the C-terminus of PTHrP [11].
The precise roles of various factors in promoting osteoblast stimulation and/or
osteoclast inhibition on the production of the new woven bone associated with prostate
cancer metastatic sites have not been elucidated. Understanding the pathogenetic
mechanisms will identify novel targets for pharmacologic intervention aimed at
relieving the debilitating complications of pain and spinal cord compression associated
with osteoblastic metastases.
Most mouse models of prostate cancer rarely exhibit metastasis to bone, and the
metastases are primarily osteolytic [12]. Only a few models, including the WISH cell
line, [13] and direct bone injection of prostate carcinoma cell lines [14] demonstrate
new bone formation. To understand the relative contributions of prostate- and bone-
derived paracrine factors in producing the osteoblastic metastases associated with
prostate cancer, we have developed a mouse model that results in new woven bone
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. formation. The purpose of this manuscript is to report the morphologic characterization
of new bone formation resulting from the interaction between canine prostate tissue and
the bone microenvironment of the calvarium in nude mice. Prostate-implanted calvaria
were evaluated using histopathology, histomorphometry of calcein-labeled mineralized
new bone production, and enumeration of tartrate-resistant acid phosphatase (TRAP)-
positive osteoclasts. Our results indicate that bioactive factors produced by normal
prostate tissue are capable of reproducing the osteoblastic effect of prostate carcinoma if
placed in close proximity to bone.
MATERIALS AND METHODS
Animal Experiments:
Adult male nude mice were obtained from Harlan Sprague-Dawley
(Indianapolis, IN). Mice were housed under constant humidity and temperature, with
12-hour light and dark cycles. They were allowed ad libitum access to standard
irradiated rodent chow and water. All experimental procedures were approved by the
Institutional Laboratory Animal Care and Use Committee of The Ohio State University.
For implantation of prostate tissue, mice were sedated with 2-3 mg/mouse ketamine and
0.1-0.2 mg/mouse xylazine administered subcutaneously. A transverse 2-mra skin
incision was made at the caudal aspect of the calvaria. The subcutaneous tissue rostral
to the skin incision was gently dissected free from the periosteum using fine forceps.
Prostate tissue was introduced into the subcutaneous space using a 2-mm metal or
plastic cannula attached to a sterile lcc syringe. Following prostate tissue implantation,
the cannula was withdrawn and the skin incision closed with surgical skin staples. Mice
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were humanely sacrificed at 14 days post-implantation by inhalation of >70% CO 2. To
fluorescently label sites of new bone formation, some mice were injected
intraperitoneally with 0.3mg calcein (Sigma Chemical Co., St Louis, MO) 3 days prior
to sacrifice as previously described by our laboratory [18],
Canine prostate tissue preparation:
Canine prostate tissue was obtained from adult male mongrel dogs immediately
following euthanasia. The bladder was exteriorized, clamped, and the prostate excised
from the urethra. The prostate was removed en bloc, and stored in ice-cold Dulbecco’s
Modified Eagle/Fl2 Medium (Life Technologies) containing 10 (ig/ml gentamicin
sulfate. Prostate tissues were washed 3X in medium. Sharp dissection was used to
remove periprostatic connective tissue, the fibrous capsule of the prostate, and the
prostatic urethra. The prostate tissue was minced into 1-mm3 pieces, and digested in
collagenase type 1 in basal DMEM/F12 media (200 mg/ml, Worthington Biochemical
Corp., Lakewood, NJ) for 4-6 hrs at 37°C. The digested prostate tissue was washed
once with DMEM/F12 and 5% fetal bovine serum, 3X with basal DMEM/F12, and
implanted subcutaneously as described above. Controls (n=9) consisted of nude mice
implanted with normal canine skeletal muscle, kidney, bladder, and spleen following
tissue processing as for the prostate. Two control mice had their periosteum surgically
scraped, but no tissue implanted. Control tissues were obtained from 3 of the 31 dogs.
Most of the dogs appeared to be sexually mature, young adults. All of the dogs were
sexually intact. Most were medium-sized dogs, estimated to weigh between 40 and 60
lbs.
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Histopathology:
Following sacrifice, calvaria of nude mice were excised and fixed in 10%
buffered formalin solution at 4°C for 24-48 hrs. Calvaria were decalcified (10% EDTA,
pH 7.4, for 24 hrs at 4°C), paraffin-embedded, 4 pm sections cut, mounted on glass
slides, and stained with hematoxylin and eosin (H&E). New bone formation was
evaluated using H&E-stained slides and/or fluorescent microscopy of undecalcified
slides from mice administered calcein. Calvaria thickness and length were determined
using a stage micrometer. Tissue viability was assessed by observation of preservation
of the normal tissue architecture, maintenance of differential staining, and lack of light
microscopic evidence of necrosis or apoptosis.
Calcein and Tartrate-Resistant Acid Phosphatase (TRAP) staining:
Calvaria were fixed in cold (4°C) 10% buffered formalin for 24-48 hrs,
infiltrated with JB-4 glycol methacrylate (JB-4 Embedding Medium, Polysciences, Inc.,
Warrington, PA) (2 changes 2-3 days apart), and embedded in JB-4 at 4°C. Sections
(4pm) were cut and one section was stained for TRAP in osteoclasts using a
commercial kit (Leukocyte Acid Phosphatase, Sigma Chemical Co, St. Louis, MO), and
one unstained section was evaluated for calcein uptake of mineralized bone using
fluorescence microscopy. Osteoclasts were counted on the endosteal and periosteal
surfaces of implanted calvaria, which ranged from 3.5 mm to 12.5 mm in length (frontal
and sagittal sections, respectively).
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Statistical methods and data analysis:
An unpaired t test (GraphPad InStat, version 3.01) was used to compare
calvarial thickness and osteoclast numbers between prostate-implanted mice and control
nude mice.
RESULTS
A total of 78 nude mice were implanted with prostate tissue from 31 dogs.
Viable prostate tissue was present in 54 of the 78 implanted mice at 2 weeks. The
implanted canine prostate tissues retained a glandular morphology (Fig. 1.1). There was
moderate stromal cell proliferation encircling the implanted prostate tissue (Fig. 1.1).
Implanted prostates also exhibited varying degrees of tissue degeneration, cystic
dilatation, squamous metaplasia, and infiltration of the tissue by neutrophils,
macrophages, and lymphocytes.
Most mice (35/54) with implanted prostate tissue had periosteal new bone
formation (Figs. 1.1,1.2, and 1.3). The periosteal new bone was hypercellular compared
to the pre-existing mature calvarial bone. The periosteal new bone had a disorganized
appearance, in contrast to the regular, lamellar appearance of the pre-existing calvaria
(Fig. 1.2). Also, in contrast to the mature bone, the new woven bone was not
birefringent when viewed with polarized light. The number of vascular spaces was
increased in the periosteal new bone compared to the mature bone. Most of the new
bone formation was on the convex surface of the calvaria, but occasionally new bone
formation was also present on the concave surface of the calvaria (Fig. 1.1). In some
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. calvaria with extensive new bone formation, osteolysis of the pre-existing calvaria had
occurred (Fig. 1.1), but in no case was there fracture of the calvarium.
Thickness of the entire calvarium with new bone formation in mice with
implanted prostate tissue was significantly increased (p < 0.05) compared to controls
(331 ± 39 pm vs. 192 ± 19 pm) (Fig. 1.4). Control mice (n=9) implanted with normal
canine muscle, kidney, bladder, spleen or with surgically-induced trauma (scraping) of
the periosteum did not develop periosteal new bone formation. A stromal cell response
did not develop in mice implanted with control tissues, as in the prostate-implanted
mice. Control tissues were viable at 2 weeks. Dogs that donated control tissues also had
prostate tissue that stimulated new bone formation.
All of the implanted prostate tissues had mild to moderate benign prostatic
hyperplasia (BPH), and was considered to be a normal finding in young, sexually intact
male dogs [19]. Prostate cancer was not present in any prostate tissue sections from
donor dogs. None of the implanted prostate tissue had evidence of neoplasia. Mild
neutrophilic and/or lymphoplasmacytic prostatitis was present in 6 of the prostates that
stimulated new bone formation. The prostatitis was very mild in 4 of the 6 prostates,
and was not considered to be an important factor in the new bone formation.
Calcein uptake by calvarial bone (evaluated by UV fluorescent microscopy)
effectively delineated mineralized new bone formation in the prostate-implanted
calvaria (Figs. 1.2, 1.3). Calvaria from control mice did not have calcein-labeled
periosteal surfaces, but there was mild staining of surfaces in the medullary endosteum.
TRAP staining was effective at demonstrating osteoclasts in the periosteum and
endosteum of marrow spaces in calvaria (Fig. 1.1, inset). There were significantly
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. increased (p < 0.05) numbers of osteoclasts per mm of bone surface (periosteal and
endosteal) in the prostate-implanted calvaria, compared to control calvaria (Fig. 1.5).
DISCUSSION
We have demonstrated that non-neoplastic canine prostate tissue induces new
woven bone formation when implanted adjacent to the calvarium of the nude mouse.
The new woven bone formation resulted in marked thickening of the calvaria in prostate
tissue-implanted mice, compared to controls. One of the most important advantages of
this model is the low degree of osteolysis at sites of prostate implantation. In our
system, implantation of normal canine prostate tissue near bone mimics osteoblastic
metastases of prostate cancer by producing new woven bone without the confounding
influence of bone invasion by cancer cells. The predominantly osteoblastic reaction
with a minor osteolytic component (analogous to the morphology of osteoblastic
prostate cancer metastases in humans) makes pathologic fractures at the implantation
sites extremely unlikely. Models of bone metastasis with prostate carcinoma cell lines
often induce marked osteolysis that can lead to the development of bone loss and
pathologic fractures. The presence of fractures makes it difficult to properly interpret
new bone formation at metastatic sites as due to the de novo synthesis of new bone
(callus formation) in response to osteolysis and cortical bone disruption [20]. Another
important advantage of this model is the rapidity of the new woven bone formation.
New woven bone formation was evident by 14 days post-implantation. This model also
utilizes benign prostate tissue and not cell lines, and may more closely represent
osteoblastic bone metastasis. The ability of prostate carcinoma metastases to stimulate
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. new bone formation may not be dependent upon neoplastic transformation since
factor(s) produced by the normal prostate stimulate osteoblastic bone formation. In
addition, a recent report demonstrating new woven bone formation by cholesterol-
lowering medications has demonstrated the value of the rodent calvarial model for
investigations of new bone formation [21].
Canine prostate tissue was selected for our model system for several reasons.
The canine prostate gland and its diseases, including benign prostatic hyperplasia,
prostatic intra-epitheiial neoplasia, and prostate cancer, have similarities with diseases
of the human prostate gland [IS], Canine prostate tissue also produces bioactive factors
known to be produced by human prostate and thought to be important in the
pathogenesis of the new bone formation at metastatic sites, such as PTHrP [16] and
endothelin-1 [17]. In addition, canine prostate tissue is widely available and can be
obtained rapidly for experimental manipulation, while human tissue is often only
available in limited quantities and with a significant delay in implantation time.
Factors produced by the prostate with potential effects on bone cells include
PTHrP, insulin-like growth factors, transforming growth factors, bone morphogenetic
proteins, plasminogen activator, and endothelins, among others [22]. In humans, it has
been reported [23] that prostate-specific antigen (PSA) cleaves the N-terminus of
PTHrP in vitro and destroys the osteoclast-stimulating, PTH-like region of the
molecule. The C-terminus of PTHrP has been shown to decrease bone resorption and
stimulate bone formation [24]. These mechanisms could potentially result in PTHrP-
mediated new bone formation at sites of prostate cancer metastasis sites [20]. The
biologic relevance of this proposed mechanism is uncertain in the dog, since canine
1 0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. prostate does not produce PSA. However, the dog prostate produces an abundant
trypsin-like serine protease, arginine esterase [25].
The pathogenesis of the new bone formation stimulated by canine prostate tissue
is not known. Due to the large number of bioactive factors produced by the prostate
gland, and those present in bone, it is likely that the new bone formation in our model is
the result of interactions between the prostate tissue, the bone microenvironment, and
possibly host-derived effects from the nude mouse. An important advantage of using
dog prostate in our model is the production of factors such as PTHrP and endothelin-1
by the canine prostate gland. These factors have been identified as important molecules
in the development of human bone metastases [3; 10]. Production of PTHrP by
metastatic prostate carcinoma cells could stimulate osteoclastic bone resorption, thereby
freeing growth factors stored in the bone matrix. Bone has large stores of TGF-P, and
TGF-P has been shown to increase PTHrP production by prostate epithelial cells.
Therefore, TGF-P may act as a growth factor for the metastatic prostate cells, although
PTHrP is not an autocrine growth factor for normal prostate epithelial cells [26], In
addition, TGF-P has been reported to increase endothelin-l mRNA transcription in
endothelial cells [27] and a human prostate cancer cell line [28], Up-regulation of
endothelins by TGF-P could represent another potential mechanism of osteoblast
stimulation and new woven bone formation at bone metastasis of prostate cancer.
Endothelin-l is produced by breast cancer cell lines producing osteoblastic metastasis
[29], and endothelin-l production by the WISH cell line results in increased new bone
formation at sites of existing bone [13]. Future experiments by our laboratory will
evaluate the effect of specific blockade of the endothelin A and B receptors on the new
l i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. bone formation observed in the canine model.
The glandular morphology of the implanted canine prostate tissue was
maintained for the 2-week implantation period. This finding was consistent with
previous reports demonstrating the value of implanted non-neoplastic human prostate
tissue into nude mice as useful models for investigations of human prostate diseases
[30;31]. The histopathologic changes observed in the implanted canine prostate tissue
(cystic dilatation, squamous cell metaplasia, foci of tissue degeneration, and leukocyte
infiltration) were similar to those observed in implanted human tissues.
The stromal tissue present around the implanted prostate tissue may represent
stimulation of murine fibrocytes by prostate-derived growth fact