Altered Levels of Angiopoietin 1 and Tie 2 Are Associated with Androgen-Regulated Vascular Regression and Growth in the Ventral Prostate in
http://www.100md.com
内分泌学杂志 2005年第8期
Department of Medical Bioscience, Pathology (A.J., P.W., A.B.), and Department of Surgical and Perioperative Sciences, Urology and Andrology (S.H.R.), Ume? University, 901 85 Ume?, Sweden
Address all correspondence and requests for reprints to: Anders Bergh, Department of Medical Bioscience, Pathology, Ume? University, 901 85 Ume?, Sweden. E-mail: anders.bergh@medbio.umu.se.
Abstract
The involution of the rat ventral prostate gland after castration could be caused by primary changes in the vasculature. To explore the mechanisms, we studied the effects of castration and testosterone treatment on the vasculature in the ventral prostate in adult rats and mice. Androgen receptor expression, vascular morphology, and the expression of angiopoietin (ang) 1 and 2 and their receptor tie 2 were examined 1, 3, and 7 d after castration and after testosterone treatment of castrated animals using stereological methods, immunohistochemistry, laser capture microdissection, and Western blotting. One day after castration, the percentage of blood vessels covered with smooth muscle actin, endothelial cell proliferation, and vascular volume had decreased, whereas endothelial cell apoptosis had increased. Simultaneously, ang 1 and tie 2 protein levels decreased. Nuclear expression of androgen receptor was observed not only in glandular and stroma smooth muscle cells but also in the mural cells of prostate arteries and veins and was markedly down-regulated already 1 d after castration. Testosterone administration of castrated mice and rats reversed all the observed effects. At the mRNA level, tie 2 was exclusively, but ang 1 predominantly, expressed in the stroma, compared with the epithelial compartment. Local delivery of soluble tie 2 during testosterone-stimulated growth, inhibited vascular maturation and increased vascular volume and leukocyte infiltration compared with controls. We conclude that androgens may regulate the prostate vasculature by direct effects on mural vascular cells and by influencing the secretion of the angiopoietins, in above all, the stroma cells.
Introduction
ANDROGEN ABLATION, a standard therapy for advanced human prostate cancer, results in a decrease in tumor epithelial cell proliferation and increased tumor cell death in some, but unfortunately not all tumors (1, 2). The mechanisms involved are unknown, but observations of the ventral prostate (VP) in rodents suggest that glandular involution due to hormonal withdraw could be caused by a decrease in blood flow (3, 4) and altered paracrine signaling of survival factors and/or death signals from the prostate stroma cells (5, 6).
In the androgen-sensitive human prostate cancer cell line (LnCaP), and in an androgen-dependent breast tumor (Shionogi), castration induces vascular regression before tumor epithelial apoptosis (7, 8). In a rat prostate cancer model, the androgen-sensitive and highly differentiated Dunning R3327 PAP tumor responds to castration with a temporarily inhibited tumor cell proliferation and growth (9). However, castration does not influence blood flow significantly (3) nor does it increase tumor cell or endothelial cell apoptosis (10). Vascular endothelial growth factor (VEGF), a well-known survival factor for endothelial cells, is decreased after castration both in the normal VP (11) and in LnCaP tumors (7). In addition, neutralization of VEGF bioactivity with a soluble chimeric VEGF receptor protein inhibits testosterone-stimulated regrowth of the prostate in castrated mice (12). We have shown previously that castration-induced reduction of VEGF is not seen in advanced human prostate cancer (13). Down-regulation of VEGF appeared vital for induction of tumor cell apoptosis after castration, and the authors speculated that a poor response in VEGF was associated with advanced prostate cancer. Together, these observations indicate that the magnitude of castration-induced glandular cell death could be related to the magnitude of the response in the adjacent vasculature. However, the kinetics of the VEGF response (14) and the complicated process of coordinating vascular and tissue regression in the normal prostate after castration suggests that other factors are also of importance.
In other tissues, the angiopoietins are important mediators of vascular growth and stabilization (15, 16). Angiopoietin (ang) 1 stabilizes blood vessels by recruiting pericytes and smooth muscle cells to the endothelial cell wall by activating the tie 2 receptor (15, 16, 17). Tie 2 signaling also promotes endothelial cell survival through the phosphoinositide 3-kinase/Akt pathway (18, 19) and it down-regulates leukocyte-endothelial adhesion (20, 21, 22). Ang 2, which is an antagonist to ang 1, destabilizes the vasculature, and stimulates either angiogenesis or vessel regression depending on high and low levels of VEGF, respectively (23, 24). Collectively, previous studies indicate that a well-pronounced vascular regression is crucial to induce a sustained tumor regression. Therefore, we wanted to explore whether the angiopoietin system could serve as regulators of castration-induced regression and testosterone-stimulated growth of the prostate gland by affecting the vasculature.
Materials and Methods
Animals and treatments
Adult male Sprague Dawley rats and C57 black mice (B&K, Stockholm, Sweden) were sedated with pentobarbital (60 mg/kg), castrated, and examined 1, 3, and 7–10 d after castration. In addition, 7-d castrated rats were treated with daily sc injections of long-acting testosterone esters (Sustanon, 10 mg/kg; Organon, Oss, The Netherlands) and were studied 1, 2, or 3 d thereafter. Intact animals were used as controls. One hour before being killed, the animals were injected with bromodeoxyuridine (BrdU, 50 mg/kg ip; Sigma-Aldrich, St. Louis, MO) to label proliferating cells. After sedation, some of the animals were fixed by vascular perfusion with Bouins fluid and the VP was removed, weighed, and post fixed with Bouins solution for 24 h, and then embedded in paraffin, as previously described (12, 25, 26). For Western blot and quantitative RT-PCR, the VP was removed and frozen in liquid nitrogen. The animal experimentation ethics committee in Ume? approved the design of this study.
In a separate experiment, C57 black mice were castrated and divided into groups with five to seven animals in each group. Seven days after castration, one group was given an intraprostatic injection of 10 μg soluble tie 2 receptor (tie 2-Fc mouse chimera; R&D Systems, Abingdon, UK) delivered in 5 μl of basement membrane extract (Matrigel; BD Biosciences, Stockholm, Sweden). The second group was given Matrigel loaded with PBS using a 10-μl Hamilton syringe. The third group was not operated on. These three groups were given sc injections of Sustanon (10 mg/kg) at d 7, 8, and 9 after castration. At d 10, the animals were injected with BrdU as described above, and the VP was taken out from all animals, carefully weighed, fixed by immersion in formalin solution for 24 h, and then embedded in paraffin.
Stereology
The volume density of blood vessels, i.e. percentage of prostate volume composed of factor VIII (Dako, Stockholm, Sweden) stained blood vessels (vascular lumina and vascular walls) were assessed according to standard procedures (26). In the immersion-fixed tissues from the tie 2-Fc experiment, the volume density of leukocytes in prostate blood vessels was also measured as described earlier (25).
Apoptosis
The percentage of apoptotic cells (apoptotic index) was determined in sections immunostained for activated caspase 3 (Cell Signaling Technology, Beverly, MA). Formalin sections were heated in target antigen retrieval solution with high pH (Dako) and then incubated with the antibody (1/200) overnight. The immunoreaction was visualized using the Envision kit (Dako). Approximately 1000 epithelial cells were assessed in each animal at x400 magnification. The number of apoptotic endothelial cells per blood vessel profile was also measured as previously described (12, 26).
Cell proliferation
Sections were immunostained with an antibody against BrdU (Dako). The percentage of BrdU-positive epithelial cells was measured (about 1000 cells were scored in each prostate at x400 magnification), and the number of BrdU-labeled endothelial cells per vascular profile was measured (12, 26).
Androgen receptor (AR)
Sections were immunostained for AR using an antibody from Upstate (Lake Placid, NY) as previously described (12).
Factor VIII and smooth muscle actin
Sections were double-stained with antibodies against factor VIII (Dako) staining endothelial cells, and smooth muscle actin (Sigma, Stockholm, Sweden), staining mural vascular cells (27). Briefly, the sections were first incubated with a primary antibody for smooth muscle actin overnight, followed by a 30-min incubation with the secondary antibody Envision AP Mouse (Dako). The sections were developed using Fast Red (Dako). The following day, the same sections were incubated with an antibody toward factor VIII, followed by secondary antibody incubation with Envision HRP Rabbit for 30 min. The slides were then developed using diaminobenzidine (Dako).
Protein extraction and Western blot
The VP was homogenized using a Micro Dismembrator (B. Braun Biotech International GmbH, Melsungen, Germany) at 2000 rpm for 45 sec. Tissues were then added to an adequate amount of lysis buffer containing either 0.5% Nonidet P-40 (NP-40; mice tissues), 0.5% NaDOC, 0.1% SDS, 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA (pH 8.0), 1 mM NaF, and Complete Protease Inhibitor Cocktail (Roche Molecular Biochemicals, Mannheim, Germany), or alternatively, 7 M urea (rat tissues), 2 M thio-urea, 4% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate, and 30 mM Tris (pH 8.5). Samples were mixed and incubated on ice for 30 min. For urea-extracted proteins, an appropriate amount of benzonase (Merck KGaA, Darmstadt, Germany) was added to the mixture to reduce viscosity caused by nucleic acids. The supernatants were isolated after centrifugation (20,000 x g, 4 C, 30 min), and the protein concentration was determined using BCA Protein assay reagent kit (Pierce Chemical Co., Rockford, IL). Concentrations of urea-extracted protein samples were determined using the 2-D Quant Kit (Amersham Biosciences, Uppsala, Sweden) according to protocol. The samples were separated by 7.5–13% SDS-PAGE under reducing conditions and subsequently transferred to a Hybond-P polyvinylidene difluoride membrane (Amersham Biosciences). The membrane was blocked in 5% milk followed by primary [ang 1 and ang 2 (US Biological, Swampscott, MA) tie 2 and VEGF (Santa Cruz Biotechnology, Inc., Santa Cruz, CA)] and secondary [antirabbit IgG, Amersham Biosciences; and antigoat, Jackson Laboratory, Pierce Biotechnology (Rockford, IL)] antibody incubations. Protein expression was visualized using enhanced chemiluminescence plus (Amersham Biosciences). The experiment was repeated several times with no significant difference (data not shown). To verify that the bands shown corresponded to the proteins of interest, blocking experiments were performed. Briefly, control peptide and intact VP protein from mice or rats were separated by 10% SDS-PAGE under reducing conditions and subsequently transferred to a Hybond-P polyvinylidene difluoride membrane. The membranes were treated as previously described (above) except that one half of the filter was incubated with an antibody:peptide solution (1:40 wt/wt) and the other half was incubated with antibody as previously described.
Laser capture microdissection (LCM)
Intact Sprague Dawley rats (n = 2) were sedated, and the VP was removed and immediately frozen in liquid nitrogen. Five-micrometer thin sections were stained with Mayer’s Hematoxylin (Sigma-Aldrich) and Eosin B (Sigma Diagnostics, St. Louis, MO) and dehydrated according to protocol provided by Arcturus (Mountain View, CA). Using a laser capture microscope (PixCell II; Arcturus) cells from selected areas were dissected with a laser diameter of 15 μm and a power of 47 mW and captured on special caps (HS CapSure; Arcturus). Five hundred shots of epithelial cells and 500 shots of stroma cells were isolated from each animal.
RNA extraction from LCM-captured cells and cDNA synthesis
RNA was prepared according to instructions in the PicoPure RNA isolation Kit (Arcturus). Total RNA was then used in a first-strand cDNA synthesis using Superscript II (Invitrogen, Stockholm, Sweden) in a 10-μl reaction according to protocol.
Quantitative RT-PCR
Quantification of the mRNA coding for ang 1 and 2 were performed by real-time quantitative PCR using the Light Cycler SYBR Green 1 technology (Roche Diagnostics, Bromma, Sweden). RT-PCR was performed in a 20-μl reaction volume, using 0.5 μM primers for ang 1 (forward, 5'- GTC GGA GAT GGC CCA G-3'; reverse, 5'-CTG TGA GCT TTC TGG TC-3'), ang 2 (forward, 5'-CAG CCA ACC AGG AAG TGA TTT-3'; reverse, 5'-GAG CAT CTG GGA ACA CTT GCA-3'), and tie 2 (forward, 5'-ATT GAC GTG AAG ATC AAG AAT GCC-3'; and reverse, 5'-AGT CAT TCC AGC CGA CCC AAG-3'), 3–4 mM MgCl2 and 2 μl cDNA (12 shots per microliter), according to protocol.
The PCR was initiated with a 10 min hotstart at 95 C, followed by amplification of 45 cycles. Each amplification cycle was composed of denaturation at 95 C for 15 sec, annealing at 60–65 C for 10 sec (optimized for each primer) and elongation at 72 C for 10 sec. To verify PCR specificity, the PCR product was subjected to melting curve analysis. Each experimental sample was run in duplicates together with positive and negative controls. Data were analyzed using the LightCycler Software 3.5.3 (Roche Molecular Biochemical, Bromma, Sweden). Glyceraldehyde-3-phosphate dehydrogenase (forward, 5'-TGC ACC ACC AAC TGC TTA GC –3'; and reverse, 5'-GGC ATG GAC TGT GGT CAT GAG-3') mRNA was analyzed in all samples to control for equal loading. All samples were analyzed on a 3% agarose gel.
Statistics
Groups were compared using the Mann-Whitney U test. Values are given as means ± SD; a P value less than 0.05 was considered significant.
Results
Castration down-regulates nuclear AR in the normal epithelium, stroma, and vasculature in the rat VP
To study how androgen ablation influences AR signaling in different prostate tissue compartments, VP sections were stained to visualize AR expression. When the AR is activated and bound to a ligand it is localized to the nucleus, whereas the inactivated form is localized to the cytoplasm (28, 29). In intact rats, a strong nuclear immunostaining for the AR was seen in most epithelial cells, in the stroma smooth muscle cells, and in fibroblasts. Some mural cells in arterioles and venules showed nuclear AR staining, whereas endothelial cells were generally unstained. Already 1 d after castration there was a marked decrease in nuclear AR staining in epithelial, stroma, and mural vascular cells (Fig. 1, A and B). AR expression was restored after 2 d of testosterone treatment (Fig. 1C).
FIG. 1. Immunohistochemical analysis of AR expression (A–C) and vessel stability in terms of factor VIII/smooth muscle actin (D–F) staining in control animals, 1 d after castration, and 2 d after testosterone replacement. In addition, immunohistochemical analysis of vessel stability in terms of factor VIII/smooth muscle actin (G and H), BrdU labeling, and the presence of intravascular leukocytes (I and J) in animals treated with PBS vs. tie 2-Fc during testosterone-stimulated prostate growth. Mark the intense AR staining of the nucleus in epithelial, stroma, and mural vascular cells of the arteries and venules in the intact animals (arrows) and the decreased nuclear staining of AR in the same cells, 1 d after castration. Note the sustained staining of AR in the cytoplasm of epithelial cells (B, arrowheads). Testosterone treatment restores the levels of AR (C). D–F, Costaining of smooth muscle cell actin (red) and factor VIII (brown, defined as a stabilized blood vessel), shows a dark brown and reddish color, whereas only factor VIII staining appears more brownish. Castration appeared to cause a decrease in smooth muscle actin staining in venules (E, arrowheads), in particular. A venule with costaining of dark brown and red is enlarged in the control animal (D) and a venule showing only brown staining is enlarged in E. Testosterone treatment restored vessel stabilization (F), and this was prevented by treatment with soluble tie 2 (arrows, G and H). A significant increase in the number of intravascular leukocytes (arrowheads) was seen in animals treated with tie 2-Fc compared with PBS (I and J). Proliferating epithelial and endothelial cells (arrows) are stained brown (I and J). a, Artery; v, vein; c, capillary; and ge, glandular epithelium.
Castration causes a major decrease in the glandular and vascular volumes in the rat VP
As previously described, castration induces a gradual decrease in the VP lobe weight. This decrease is accompanied by a decrease in epithelial cell proliferation (BrdU labeling index) and an increase in apoptosis (Table 1). However, here we show that the blood vessels are also affected. The number of proliferating endothelial cells decreased rapidly, and the endothelial cell apoptosis increased (Table 1 and Fig. 2). The volume density of blood vessels (percentage of prostate volume occupied by blood vessels) was initially decreased, but normalized at d 7. Treatment of 7-d castrated rats with testosterone rapidly increased endothelial cell proliferation and vascular volumes and decreased endothelial apoptosis as described earlier (26).
TABLE 1. Summarized VP lobe weight (g) and stereological and morphological data in control, 1, 3, 7, and 9 d after castration of rats (%)
FIG. 2. Graphical view of vascular regression shown as endothelial cell proliferation index (%, dotted line), endothelial cell apoptosis index (%, dashed-dot line), and volume density of blood vessels (%, bold line) in intact animals and 1, 3, and 7 d after castration. Seven-day castrated testosterone-treated animals during 2 d are shown as d 9. Values are given as means with n = 5–8. a, Significantly different from controls; b, significantly different from d 7; P < 0.05.
Castration down-regulates factors known to inhibit inflammation and to stabilize blood vessels
Having found that the vasculature shrinks after castration in the rat VP, it was of interest to study the mechanisms possibly involved. The protein levels of the tie 2 receptor, which is known to mediate both vascular stabilization by enhancing endothelial-pericyte interactions, and down-regulate inflammation, was decreased as early as 1 d after castration in rats and even more pronounced in mice (Fig 3, A and B). The drop in tie 2 was more pronounced than the drop in vascular volume (Table 1) indicating that the decrease in tie 2 is not only a matter of endothelial cell apoptosis. Ang 1 was decreased in the same manner (Fig. 3, A and B). No obvious differences were observed for ang 2 in rats (Fig. 3, A and B) when proteins were extracted with urea. However, when the same experiment was repeated and the proteins extracted with 0.5% NP-40, ang 2 expression was increased after castration (d 1–3) and subsequently decreased (data not shown). There were no differences in ang 1 or tie 2 expression levels detected between urea-extracted and 0.5% NP-40-extracted rat protein samples. Protein from mice tissues were extracted in all cases with 0.5% NP-40, and castration had no clear-cut effect on ang 2 protein levels. In line with a previous finding (14), VEGF was decreased after castration in rats, but this was evident somewhat later than the change of the angiopoietins, namely 3 d after castration (Fig. 3A). Testosterone administration to 7-d castrated rats over 1–3 d reversed these findings and resulted in an increase in ang 1, tie 2, and VEGF levels (Fig. 3, A and B). In addition, tie 2 and ang 1 expression was examined in castrated rats that had not received any testosterone administration at 7, 10, and 14 d after castration. In these animals, there was no up-regulation of either ang 1 or tie 2 (data not shown). To verify that the bands shown were correct, control Western blots with peptide and intact VP protein extracts from mice or rats were analyzed (Fig. 3C). Both peptide and the corresponding bands in the tissue samples were detected with the antibody, and both bands could be blocked by preincubating the antibody with peptide before usage. The faint unspecific bands were not blocked by this treatment.
FIG. 3. Western blot results of the angiopoietins and VEGF by 7.5–13% SDS-PAGE under reducing conditions in rats (A, urea-extracted proteins) and mice (B, 0.5% NP-40-extracted samples). Each group includes five animals showing intact VP (lane 1), and castrated VP 1, 3, and 7 d after treatment (lanes 2, 3, and 4, respectively). Lanes 5–7, Seven-day castrated testosterone-treated animals during 1, 2, and 3 d. Actin was used as a control for equal loading. The experiment was repeated twice using different animals with no apparent change (data not shown). For tie 2, the same results were obtained using a different antibody (data not shown). Western blots with positive and negative controls in terms of control tissues with preblocked antibody (see Materials and Methods) and control peptides (R&D Systems) for ang 2 (C) and ang 1 (D). Ang 1 gave a somewhat larger band than previously reported by other researchers. This could possibly be due to differences in glycosylation. Occasionally a 55-kDa band was identified in intact mouse VP tissue by the antibody against ang 2 (C, lane 2), but in most cases we only observed the 75-kDa band.
Castration induces detachment and/or dedifferentiation in the mural cell layer in prostate blood vessels
To examine the possibility that the decreased ang 1 and tie 2 levels may as in other tissues cause endothelial-mural cell detachment and/or dedifferentiation, the rat prostate vasculature was double stained for factor VIII related antigen and -smooth muscle actin (staining pericytes and vascular smooth muscle cells). This showed that the percentage of blood vessels with a detectable smooth muscle actin covering was decreased already 1 d after castration (Figs. 1, D and E, and4). Testosterone treatment of castrated animals increased the percentage of blood vessels covered with smooth muscle actin (Fig. 1F).
FIG. 4. Graphic picture showing the decrease and increase in vessel stabilization in percent () and relative protein values of ang 1 () and tie 2 () in control rats and 1, 3, 7, and 9–10 d after castration. Protein levels were obtained by scanning Western blots with pooled samples of five animals in each group, using Flour-S Multi Imager (Bio-Rad, Hercules, CA) and relating all values to their intact value, after corrections for corresponding actin levels. Note the high level of ang 1 at d 10. Vessel stabilization was determined by counting vessels with a smooth muscle cells coat vs. those without. Values are means, n = 5–8. a, Significantly different from controls, P < 0.05. The SD varied from 13–24%.
Administration of soluble tie 2 receptor blocks testosterone-stimulated vascular maturation
To test the hypothesis that angiopoietin signaling is of importance for testosterone-stimulated vascular growth and maturation, we treated 7-d castrated mice with a local deposition of 10 μg tie 2-Fc or PBS in Matrigel. The use of Matrigel as a vehicle for local administration of growth factors to the prostate has been described earlier (30). After injection, the Matrigel formed a gel inside and partly around the prostate lobe. These mice were then treated with daily injections of testosterone at d 7, 8, and 9 and examined at d 10 after castration. Castrated mice without Sustanon treatment and castrated mice that had been given 10 mg/kg Sustanon were used as controls. We found that blockage of angiopoietin signaling inhibited vascular maturation in terms of fraction of blood vessels covered with smooth muscle actin (Table 2, P = 0.004; Fig. 1, G and H) compared with controls. In addition, the volume density of factor VIII stained blood vessels was significantly increased (Table 2, P = 0.01) and the apparent dilated blood vessels contained more leukocytes than mice injected with PBS alone (Table 2, P = 0.02; Fig. 1, I and J). However, treatment with tie 2-Fc did not influence endothelial or epithelial cell proliferation (Table 2; Fig. 1, I and J). When animals injected with PBS and Matrigel were compared with noninjected animals it was evident that Matrigel, which contains a cocktail of growth factors, stimulated endothelial cell proliferation and possibly vascular volume (Table 2). The Matrigel itself did not however show any effect on any of the other variables studied. In addition, no signs of tissue damage, due to the amount of matrigel injected, could be detected by microscopy.
TABLE 2. Endothelial and epithelial BrdU labeling index (%), percentage of blood vessels covered with smooth muscle actin (SMA %), volume density of blood vessels (Vv FVIII), and volume density (Vv) of leukocytes in VP blood vessels (%) after 3 d testosterone (T) treatment of castrated (C) mice
The angiopoietins are expressed in the epithelium and stroma in rats
Finally, it was of interest to know where the angiopoietins are localized and act. Thus, rat VP from two intact animals was removed, and LCM technology was used to obtain pure fractions of epithelial cells and stroma cells. The mRNA expression of ang 1 and ang 2 was detected in both the stroma and epithelial cells, whereas tie 2 was exclusively expressed in the stroma fraction (Fig. 5). Similar results were seen in mice (data not shown) and semiquantitatively, there was apparently a higher ang 1 expression in the stroma compartment compared with the epithelial fraction (Fig. 5).
FIG. 5. Levels of ang 1, ang 2, and tie 2 mRNA, in the epithelial compartment and the stroma compartment in the VP of one rat, analyzed on a 3% agarose gel. Lane 1 shows a positive control and lane 4 shows a negative control. Glyceraldehyde-3-phosphate dehydrogenase was analyzed in all samples to ensure equal loading. mRNA data from mice showed similar results (data not shown).
Discussion
In the normal prostate, earlier studies have shown that castration-induced prostate regression is preceded by endothelial cell death and a decrease in blood flow (3, 4, 28). We previously suggested that these effects were caused by a drop in VEGF (11, 12, 26), but our present observations indicate that the angiopoietin signaling pathway is also involved. The regulation of ang 1 (but apparently not ang 2) and tie 2 by testosterone, the observation of vascular changes that have been related to angiopoietins in other tissues, and the results of local treatment with tie 2 decoy receptor during testosterone stimulation clearly indicate that changes in ang 1 signaling are of importance for vascular growth and regression in the prostate gland.
Vascular destabilization, often measured as a decrease in the percentage of blood vessels covered with smooth muscle actin, is assumed to be caused by mural cell detachment and is seen in other tissues after decreased ang 1 or increased ang 2 signaling (16, 17). However, as some mural vascular cells in the prostate have AR, the reduced smooth muscle staining seen after castration could also be a direct sign of vascular cell dedifferentiation (loss of smooth muscle actin expression but not necessarily detachment), due to absence of androgens.
Castration causes a rapid influx of leukocytes into the VP (31). The mechanism regulating this has not been studied, but in other tissues, leukocyte migration is started by up-regulation of leukocyte adhesion molecules on the endothelium. In several other tissues, leukocyte adhesion is inhibited by ang 1 (20, 21, 22) suggesting that the leukocyte accumulation seen in VP already at 24 h after castration (31) could be related to the decrease in ang 1 and tie 2 expression. In line with this, treatment with tie 2-Fc increased the number of leukocytes present in prostate blood vessels compared with controls.
Another novel observation in this study is that vascular mural cells, mainly smooth muscle cells in prostate arterioles and venules, express AR and that this expression is rapidly down-regulated after castration. Arteriolar and venular smooth muscle cells in the VP are apparently direct target cells for androgens. In contrast, previous studies based on a different AR antibody (3) did not show any expression in the vasculature. However, using the same antibody as in the present study, Prins et al. (32) also detected AR in vascular smooth muscle cells in the rat VP. Therefore, it cannot be excluded that the decrease in prostate blood flow (1, 2), total vascular volume, and loss of smooth muscle actin staining can be caused by direct effects. In line with several previous studies (12, 28, 32) prostate endothelial cells do not express AR, suggesting that the decrease in endothelial cell proliferation and increase in endothelial cell apoptosis, seen mainly in micro vessels, must be indirectly caused by altered signaling from surrounding AR-positive mural cells. More specifically, because these cells in general were not covered by smooth muscle cells, the signals may come from more remote cells such as stromal or glandular epithelial cells.
To understand how castration-induced vascular regression is coordinated with glandular and stroma involution, the cellular sites of angiopoietin synthesis and action should be considered. In a human prostate study, ang 1 was found to be localized mainly in basal epithelial cells, whereas ang 2 was found both in smooth muscle cells and epithelial cells (33). In this study of rodents, ang 1 and ang 2 were found both in the prostate stroma and glandular compartments, although ang 1 showed a relatively higher expression in the stroma than in the epithelial fraction. This is in line with other tissues where ang 1 is produced principally by pericytes and other vascular mural cells. The difference in ang 1 expression in humans and rodents could reflect a species difference (for example; basal epithelial cells are considerably more common in humans than in rodents). Tie 2 mRNA were exclusively found in the stroma, as expected, because numerous studies have concluded tie 2 to be expressed on endothelial cells (15, 16, 17, 18, 19) specifically. In addition, we have detected ang 1 and ang 2 protein in a rat prostate endothelial cell line (YPEN-1; our unpublished observations), suggesting a possible autocrine tie 2 stimulation in rats. Therefore, androgens may regulate the prostate vasculature by direct effects on mural vascular cells and or indirectly by influencing the secretion of paracrine regulators such as ang 1.
Testosterone stimulates growth of the prostate gland by affecting epithelial and endothelial proliferation and apoptosis (12, 26). By increasing VEGF synthesis in prostate epithelial cells glandular growth is coordinated with that of the vasculature (12). In this study, we show that testosterone-stimulated prostate vascular growth is accompanied by increased levels of ang 1 and tie 2 and by using a tie 2 decoy receptor administered in Matrigel we could examine the functional role of angiopoietin signaling during testosterone-stimulated prostate growth. Interestingly, and as could be predicted from the effects of castration, reduced angiopoietin signaling resulted in a delay in vascular maturation (fraction of vessels with smooth muscle actin covering) and an increased number of intravascular leukocytes. This finding is line with studies showing that inhibition of tie 2 signaling increase local inflammation in tumors (34). The blood vessels were significantly larger in the tie 2-Fc group, probably due to the lack of the stabilizing effect of ang 1, and the presence of high VEGF levels in the growing prostate. This effect is supported by several other studies that also have shown large dilated blood vessels when ang 1 signaling has been perturbed in knockout experiments (35) or by using nonfunctional tie 2 receptors (36, 37). The angiopoietins have not been shown to exert a proliferative effect on endothelial cells, and in line with this, we did not see any difference in proliferation between the tie 2-Fc group and the PBS group, although the Matrigel itself seemed to have a stimulatory effect on the proliferation of endothelial cells. Whether the vessel phenotype described should impair testosterone-stimulated growth of the prostate due to nonfunctional blood vessels, or the opposite, namely support growth of the prostate, due to an increased angiogenesis/vascular volume, is not known. A possible increased prostate weight could also be an indirect effect of leaky blood vessels due to the accumulation of fluids. Long-term inhibition of tie 2 signaling decreases growth of malignant human astrocytoma in sc and intracranial xenografts (36) and in a murine breast cancer cell line when injected into nude mice (37). In our study, we weighed the prostate after 3 d of testosterone stimulation but found no significant changes in prostate weight. Further studies blocking angiopoietin signaling for a longer time are probably needed to elucidate whether this, as blocking VEGF (12), is a way to perturb testosterone-stimulated prostate growth.
In conclusion, the present study suggests that androgen-regulated levels of ang 1/tie 2 signaling mediate vascular regression and growth in the prostate. This, in turn, precedes hormonally induced regression and growth of prostate glands. Interestingly, similar results implicated a role for the angiopoietins in ACTH-regulated vascular growth and regression of the adrenal gland (38). This raises the possibility of a general mechanism for hormones to coordinate growth and regression of various organs by regulating the vasculature through the angiopoietins.
Acknowledgments
We thank Mrs. Birgitta Ekblom, Mrs. Pernilla Andersson, Mrs. Sigrid Kilter, Mrs. ?sa Skytt, Mrs. Elisabeth Dahlberg, and Mrs. Ulla-Stina Spetz for their professional technical assistance and contribution to this paper.
References
Westin P, Stattin P, Damber J-E, Bergh A 1995 Castration therapy rapidly induces apoptosis in a minority and decreases cell proliferation in a majority of human prostatic tumors. Am J Pathol 146:1368–1375
Ohlson N, Wikstrom P, Stattin P, Bergh A 2005 Cell proliferation and apoptosis in prostate tumors and adjacent non-malignant prostate tissue in patients at different time-points after castration treatment. Prostate 62:307–315
Lek?s E, Johansson M, Widmark A, Bergh A, Damber J-E 1997 Decrement of blood flow precedes the involution of the ventral prostate in the rat after castration. Urol Res 25:309–314
Shabsigh A, Chang DT, Heitjan DF, Kiss A, Olsson CA, Puchner PJ, Buttyan R 1998 Rapid reduction in blood flow to the rat ventral prostate gland after castration: preliminary evidence that androgens influence prostate size by regulating blood flow to the prostate gland and prostatic endothelial cell survival. Prostate 36:201–206
Kurita T, Wang YZ, Donjacour AA, Zhao C, Lydon JP, O'Malley BV, Isaacs JT, Dahiya R, Cunha GR 2001 Paracrine regulation of apoptosis by steroid hormones in the male and female reproductive systems. Cell Death Differ 8:192–200
Cooke PS, Buchanan DL, Young P, Setiawan T, Brody J, Korach KS, Taylor J, Lubahn DB, Cunha GR 1997 Stromal estrogen receptors mediate mitogenic effects of estradiol on uterine epithelium. Proc Natl Acad Sci USA 94:6535–6540
Jain RK, Safabakhsh N, Sckell A, Chen Y, Jiang P, Benjamin L, Yuan F, Keshet E 1998 Endothelial cell death, angiogenesis, and micro vascular function after castration in an androgen-dependent tumor: role of vascular endothelial growth factor. Proc Natl Acad Sci USA 95:10820–10825
Stewart RJ, Panigrahy D, Flynn E, Folkman J 2001 Vascular endothelial growth factor expression and tumor angiogenesis are regulated by androgens in hormone responsive human prostate carcinoma: evidence for androgen dependent destabilization of vascular endothelial growth factor transcripts. J Urol 165:688–693
Westin P, Bergh A, Damber J-E 1993 Castration rapidly results in a major reduction in epithelial cell number in the rat prostate, but not in the highly differentiated Dunning R3327 prostatic adenocarcinoma. Prostate 22:65–74
Brandstrom A, Westin P, Bergh A, Cajander S, Damber JE 1994 Castration induces apoptosis in the ventral prostate but not in an androgen-sensitive prostatic adenocarcinoma in the rat. Cancer Res 54:3594–3601
Haggstrom S, Wikstrom P, Bergh A, Damber JE 1998 Expression of vascular endothelial growth factor and its receptors in the rat ventral prostate and Dunning R3327 PAP adenocarcinoma before and after castration. Prostate 36:71–79
Lissbrant IF, Hammarsten P, Lissbrant E, Ferrara N, Rudolfsson SH, Bergh A 2004 Neutralizing VEGF bioactivity with a soluble chimeric VEGF-receptor protein flt(1–3) IgG inhibits testosterone stimulated prostate growth in castrated mice. Prostate 58:57–65
H?ggstr?m S, Stattin P, Wikstr?m P, Bergh A, Damber JE 2001 Castration-induced reduction of vascular endothelial growth factor expression in benign human prostate tissue is lost in advanced prostate cancer. BJU Int 88:110–116
Burchardt M, Burchardt T, Chen MW, Hayek OR, Knight C, Shabsigh A, de la Taille A, Buttyan R 2000 Vascular endothelial growth factor-A expression in the rat ventral prostate gland and early effects of castration. Prostate 43:184–194
Thurston G, Suri C, Smith K, McClain J, Sato TN, Yancopoulos GD, McDonald DM 1999 Leakage resistant blood vessels in mice transgenically over expressing angiopoietin 1. Science 286:2511–2514
Thurston G, Rudge JS, Ioffe E, Zhou H, Ross L, Croll SD, Glazer N, Holash J, McDonald DM, Yancopoulos GD 2000 Angiopoietin 1 protects the adult vasculature against plasma leakage. Nat Med 6:460–463
Papapetropoulos A, Garcia-Cardena G, Dengler TJ, Maisonpierre PC, Yancopoulos GD, Sessa WC 1999 Direct actions of angiopoietin 1 on human endothelium: evidence for network stabilization, cell survival, and interaction with other angiogenic growth factors. Lab Invest 79:213–223
Kwak HJ, So JN, Lee SJ, Kim I, Koh GY 1999 Angiopoetin 1 is an apoptosis survival factor for endothelial cells. FEBS Lett 448:249–253
Papapetropoulos A, Fulton D, Mahboubi K, Kalb RG, O’Connor DS, Li F, Altieri DC, Sessa WC 2000 Angiopoietin 1 inhibits endothelial cell apoptosis via the Akt/survivin pathway. J Biol Chem 275:9102–9105
Gamble JR, Drew J, Trezise L, Underwood A, Parsons M, Kasminkas L, Rudge J, Yancopoulos G, Vadas MA 2000 Angiopoietin 1 is an antipermeability and anti-inflammatory agent in vitro and targets cell junctions. Circ Res 29: 603–607
Pizurki L, Zhou Z, Glynos K, Roussos C, Papapetropoulos A 2003 Angiopoietin 1 inhibits endothelial permeability, neutrophil adherence and IL-8 production. Br J Pharmacol 139:329–336
Haggstrom Rudolfsson S, Johansson A, Franck Lissbrant I, Wikstrom P, Bergh A 2003 Localized expression of angiopoietin 1 and 2 may explain unique characteristics of the rat testicular microvasculature. Biol Reprod 69:1231–1237
Holash J, Wiegand SJ, Yancopoulos GD 1999 New model of tumor angiogenesis: dynamic balance between vessel regression and growth mediated by angiopoietins and VEGF. Oncogene 18:5356–5362
Lobov IB, Brooks PC, Lang RA 2002 Angiopoietin 2 displays VEGF dependent modulation of capillary structure and endothelial cell survival in vivo. Proc Natl Acad Sci USA 99:11205–11210
Bergh A, Widmark A, Damber JE, Cajander S 1986 Are leukocytes involved in the human chorionic gonadotropin-induced increase in testicular vascular permeability? Endocrinology 119:586–590
Franck-Lissbrant I, Haggstrom S, Damber JE, Bergh A 1998 Testosterone stimulates angiogenesis and vascular re-growth in the ventral prostate in castrated adult rats. Endocrinology 139:451–456
Wikstrom P, Lissbrant IF, Stattin P, Egevad L, Bergh A 2002 Endoglin (CD105) is expressed on immature blood vessels and is a marker for survival in prostate cancer. Prostate 51:268–275
Buttyan R, Ghafar MA, Shabsigh A 2000 The effects of androgen deprivation on the prostate gland: cell death mediated by vascular regression. Curr Opin Urol 10:415–420
Zhou ZX, Sar M, Simental JA, Lane MV, Wilson EM 1994 A ligand-dependent bipartite nuclear targeting signal in the human androgen receptor. J Biol Chem 269:13115–13123
Marengo SR, Chung LW 1994 An orthotopic model for the study of growth factors in the ventral prostate of the rat: effects of epidermal growth factor and basic fibroblast growth factor. J Androl 15:277–286
Wilson MJ, Whitaker JN, Sinha AA 1991 Immunocytochemical localization of cathepsin D in rat ventral prostate: evidence for castration-induced expression of cathepsin D in basal cells. Anat Rec 229:321–333
Prins GS, Birch L, Greene GL 1991 Androgen receptor localization in different cell types of the adult rat prostate. Endocrinology 129:3187–3199
Lind AJ, Wikstrom P, Granfors T, Egevad L, Stattin P, Bergh A 2004 Angiopoietin 2 expression is related to histological grade, vascular density, metastases, and outcome in prostate cancer. Prostate 62:394–399
Melani C, Stoppacciaro A, Foroni C, Felicetti F, Caré A, Colombo MP 2004 Angiopoietin decoy secreted at tumor site impairs tumor growth and metastases by inducing local inflammation and altering neoangiogenesis. Cancer Immunol Immunother 53:600–608
Suri C, Jones PF, Patan S, Bartunkova S, Maissonpierre PC, Davis S, Sato TN, Yancopoulos GD 1996 Requisite role of angiopoietin 1, a ligand for the tie 2 receptor, during embryonic angiogenesis. Cell 87:1171–1180
Zadeh G, Qian B, Okhowat A, Sabha N, Kontos CD, Guha A 2004 Targeting the tie 2/tek receptor in astrocytomas. Am J Pathol 164:467–476
Stratmann A, Acker T, Burger AM, Amann K, Risau W, Plate KH 2001 Differential inhibition of tumor angiogenesis by tie 2 and vascular endothelial growth factor receptor-2 dominant negative receptor mutants. Int J Cancer 273–282
Féraud O, Mallet C, Vilgrain I 2003 Expressional regulation of the angiopoietin 1 and 2 and the endothelial-specific receptor tyrosine kinase Tie 2 in adrenal atrophy: a study of adrenocorticotropin-induced repair. Endocrinology 144:4607–4615(Anna Johansson, Stina H?g)
Address all correspondence and requests for reprints to: Anders Bergh, Department of Medical Bioscience, Pathology, Ume? University, 901 85 Ume?, Sweden. E-mail: anders.bergh@medbio.umu.se.
Abstract
The involution of the rat ventral prostate gland after castration could be caused by primary changes in the vasculature. To explore the mechanisms, we studied the effects of castration and testosterone treatment on the vasculature in the ventral prostate in adult rats and mice. Androgen receptor expression, vascular morphology, and the expression of angiopoietin (ang) 1 and 2 and their receptor tie 2 were examined 1, 3, and 7 d after castration and after testosterone treatment of castrated animals using stereological methods, immunohistochemistry, laser capture microdissection, and Western blotting. One day after castration, the percentage of blood vessels covered with smooth muscle actin, endothelial cell proliferation, and vascular volume had decreased, whereas endothelial cell apoptosis had increased. Simultaneously, ang 1 and tie 2 protein levels decreased. Nuclear expression of androgen receptor was observed not only in glandular and stroma smooth muscle cells but also in the mural cells of prostate arteries and veins and was markedly down-regulated already 1 d after castration. Testosterone administration of castrated mice and rats reversed all the observed effects. At the mRNA level, tie 2 was exclusively, but ang 1 predominantly, expressed in the stroma, compared with the epithelial compartment. Local delivery of soluble tie 2 during testosterone-stimulated growth, inhibited vascular maturation and increased vascular volume and leukocyte infiltration compared with controls. We conclude that androgens may regulate the prostate vasculature by direct effects on mural vascular cells and by influencing the secretion of the angiopoietins, in above all, the stroma cells.
Introduction
ANDROGEN ABLATION, a standard therapy for advanced human prostate cancer, results in a decrease in tumor epithelial cell proliferation and increased tumor cell death in some, but unfortunately not all tumors (1, 2). The mechanisms involved are unknown, but observations of the ventral prostate (VP) in rodents suggest that glandular involution due to hormonal withdraw could be caused by a decrease in blood flow (3, 4) and altered paracrine signaling of survival factors and/or death signals from the prostate stroma cells (5, 6).
In the androgen-sensitive human prostate cancer cell line (LnCaP), and in an androgen-dependent breast tumor (Shionogi), castration induces vascular regression before tumor epithelial apoptosis (7, 8). In a rat prostate cancer model, the androgen-sensitive and highly differentiated Dunning R3327 PAP tumor responds to castration with a temporarily inhibited tumor cell proliferation and growth (9). However, castration does not influence blood flow significantly (3) nor does it increase tumor cell or endothelial cell apoptosis (10). Vascular endothelial growth factor (VEGF), a well-known survival factor for endothelial cells, is decreased after castration both in the normal VP (11) and in LnCaP tumors (7). In addition, neutralization of VEGF bioactivity with a soluble chimeric VEGF receptor protein inhibits testosterone-stimulated regrowth of the prostate in castrated mice (12). We have shown previously that castration-induced reduction of VEGF is not seen in advanced human prostate cancer (13). Down-regulation of VEGF appeared vital for induction of tumor cell apoptosis after castration, and the authors speculated that a poor response in VEGF was associated with advanced prostate cancer. Together, these observations indicate that the magnitude of castration-induced glandular cell death could be related to the magnitude of the response in the adjacent vasculature. However, the kinetics of the VEGF response (14) and the complicated process of coordinating vascular and tissue regression in the normal prostate after castration suggests that other factors are also of importance.
In other tissues, the angiopoietins are important mediators of vascular growth and stabilization (15, 16). Angiopoietin (ang) 1 stabilizes blood vessels by recruiting pericytes and smooth muscle cells to the endothelial cell wall by activating the tie 2 receptor (15, 16, 17). Tie 2 signaling also promotes endothelial cell survival through the phosphoinositide 3-kinase/Akt pathway (18, 19) and it down-regulates leukocyte-endothelial adhesion (20, 21, 22). Ang 2, which is an antagonist to ang 1, destabilizes the vasculature, and stimulates either angiogenesis or vessel regression depending on high and low levels of VEGF, respectively (23, 24). Collectively, previous studies indicate that a well-pronounced vascular regression is crucial to induce a sustained tumor regression. Therefore, we wanted to explore whether the angiopoietin system could serve as regulators of castration-induced regression and testosterone-stimulated growth of the prostate gland by affecting the vasculature.
Materials and Methods
Animals and treatments
Adult male Sprague Dawley rats and C57 black mice (B&K, Stockholm, Sweden) were sedated with pentobarbital (60 mg/kg), castrated, and examined 1, 3, and 7–10 d after castration. In addition, 7-d castrated rats were treated with daily sc injections of long-acting testosterone esters (Sustanon, 10 mg/kg; Organon, Oss, The Netherlands) and were studied 1, 2, or 3 d thereafter. Intact animals were used as controls. One hour before being killed, the animals were injected with bromodeoxyuridine (BrdU, 50 mg/kg ip; Sigma-Aldrich, St. Louis, MO) to label proliferating cells. After sedation, some of the animals were fixed by vascular perfusion with Bouins fluid and the VP was removed, weighed, and post fixed with Bouins solution for 24 h, and then embedded in paraffin, as previously described (12, 25, 26). For Western blot and quantitative RT-PCR, the VP was removed and frozen in liquid nitrogen. The animal experimentation ethics committee in Ume? approved the design of this study.
In a separate experiment, C57 black mice were castrated and divided into groups with five to seven animals in each group. Seven days after castration, one group was given an intraprostatic injection of 10 μg soluble tie 2 receptor (tie 2-Fc mouse chimera; R&D Systems, Abingdon, UK) delivered in 5 μl of basement membrane extract (Matrigel; BD Biosciences, Stockholm, Sweden). The second group was given Matrigel loaded with PBS using a 10-μl Hamilton syringe. The third group was not operated on. These three groups were given sc injections of Sustanon (10 mg/kg) at d 7, 8, and 9 after castration. At d 10, the animals were injected with BrdU as described above, and the VP was taken out from all animals, carefully weighed, fixed by immersion in formalin solution for 24 h, and then embedded in paraffin.
Stereology
The volume density of blood vessels, i.e. percentage of prostate volume composed of factor VIII (Dako, Stockholm, Sweden) stained blood vessels (vascular lumina and vascular walls) were assessed according to standard procedures (26). In the immersion-fixed tissues from the tie 2-Fc experiment, the volume density of leukocytes in prostate blood vessels was also measured as described earlier (25).
Apoptosis
The percentage of apoptotic cells (apoptotic index) was determined in sections immunostained for activated caspase 3 (Cell Signaling Technology, Beverly, MA). Formalin sections were heated in target antigen retrieval solution with high pH (Dako) and then incubated with the antibody (1/200) overnight. The immunoreaction was visualized using the Envision kit (Dako). Approximately 1000 epithelial cells were assessed in each animal at x400 magnification. The number of apoptotic endothelial cells per blood vessel profile was also measured as previously described (12, 26).
Cell proliferation
Sections were immunostained with an antibody against BrdU (Dako). The percentage of BrdU-positive epithelial cells was measured (about 1000 cells were scored in each prostate at x400 magnification), and the number of BrdU-labeled endothelial cells per vascular profile was measured (12, 26).
Androgen receptor (AR)
Sections were immunostained for AR using an antibody from Upstate (Lake Placid, NY) as previously described (12).
Factor VIII and smooth muscle actin
Sections were double-stained with antibodies against factor VIII (Dako) staining endothelial cells, and smooth muscle actin (Sigma, Stockholm, Sweden), staining mural vascular cells (27). Briefly, the sections were first incubated with a primary antibody for smooth muscle actin overnight, followed by a 30-min incubation with the secondary antibody Envision AP Mouse (Dako). The sections were developed using Fast Red (Dako). The following day, the same sections were incubated with an antibody toward factor VIII, followed by secondary antibody incubation with Envision HRP Rabbit for 30 min. The slides were then developed using diaminobenzidine (Dako).
Protein extraction and Western blot
The VP was homogenized using a Micro Dismembrator (B. Braun Biotech International GmbH, Melsungen, Germany) at 2000 rpm for 45 sec. Tissues were then added to an adequate amount of lysis buffer containing either 0.5% Nonidet P-40 (NP-40; mice tissues), 0.5% NaDOC, 0.1% SDS, 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA (pH 8.0), 1 mM NaF, and Complete Protease Inhibitor Cocktail (Roche Molecular Biochemicals, Mannheim, Germany), or alternatively, 7 M urea (rat tissues), 2 M thio-urea, 4% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate, and 30 mM Tris (pH 8.5). Samples were mixed and incubated on ice for 30 min. For urea-extracted proteins, an appropriate amount of benzonase (Merck KGaA, Darmstadt, Germany) was added to the mixture to reduce viscosity caused by nucleic acids. The supernatants were isolated after centrifugation (20,000 x g, 4 C, 30 min), and the protein concentration was determined using BCA Protein assay reagent kit (Pierce Chemical Co., Rockford, IL). Concentrations of urea-extracted protein samples were determined using the 2-D Quant Kit (Amersham Biosciences, Uppsala, Sweden) according to protocol. The samples were separated by 7.5–13% SDS-PAGE under reducing conditions and subsequently transferred to a Hybond-P polyvinylidene difluoride membrane (Amersham Biosciences). The membrane was blocked in 5% milk followed by primary [ang 1 and ang 2 (US Biological, Swampscott, MA) tie 2 and VEGF (Santa Cruz Biotechnology, Inc., Santa Cruz, CA)] and secondary [antirabbit IgG, Amersham Biosciences; and antigoat, Jackson Laboratory, Pierce Biotechnology (Rockford, IL)] antibody incubations. Protein expression was visualized using enhanced chemiluminescence plus (Amersham Biosciences). The experiment was repeated several times with no significant difference (data not shown). To verify that the bands shown corresponded to the proteins of interest, blocking experiments were performed. Briefly, control peptide and intact VP protein from mice or rats were separated by 10% SDS-PAGE under reducing conditions and subsequently transferred to a Hybond-P polyvinylidene difluoride membrane. The membranes were treated as previously described (above) except that one half of the filter was incubated with an antibody:peptide solution (1:40 wt/wt) and the other half was incubated with antibody as previously described.
Laser capture microdissection (LCM)
Intact Sprague Dawley rats (n = 2) were sedated, and the VP was removed and immediately frozen in liquid nitrogen. Five-micrometer thin sections were stained with Mayer’s Hematoxylin (Sigma-Aldrich) and Eosin B (Sigma Diagnostics, St. Louis, MO) and dehydrated according to protocol provided by Arcturus (Mountain View, CA). Using a laser capture microscope (PixCell II; Arcturus) cells from selected areas were dissected with a laser diameter of 15 μm and a power of 47 mW and captured on special caps (HS CapSure; Arcturus). Five hundred shots of epithelial cells and 500 shots of stroma cells were isolated from each animal.
RNA extraction from LCM-captured cells and cDNA synthesis
RNA was prepared according to instructions in the PicoPure RNA isolation Kit (Arcturus). Total RNA was then used in a first-strand cDNA synthesis using Superscript II (Invitrogen, Stockholm, Sweden) in a 10-μl reaction according to protocol.
Quantitative RT-PCR
Quantification of the mRNA coding for ang 1 and 2 were performed by real-time quantitative PCR using the Light Cycler SYBR Green 1 technology (Roche Diagnostics, Bromma, Sweden). RT-PCR was performed in a 20-μl reaction volume, using 0.5 μM primers for ang 1 (forward, 5'- GTC GGA GAT GGC CCA G-3'; reverse, 5'-CTG TGA GCT TTC TGG TC-3'), ang 2 (forward, 5'-CAG CCA ACC AGG AAG TGA TTT-3'; reverse, 5'-GAG CAT CTG GGA ACA CTT GCA-3'), and tie 2 (forward, 5'-ATT GAC GTG AAG ATC AAG AAT GCC-3'; and reverse, 5'-AGT CAT TCC AGC CGA CCC AAG-3'), 3–4 mM MgCl2 and 2 μl cDNA (12 shots per microliter), according to protocol.
The PCR was initiated with a 10 min hotstart at 95 C, followed by amplification of 45 cycles. Each amplification cycle was composed of denaturation at 95 C for 15 sec, annealing at 60–65 C for 10 sec (optimized for each primer) and elongation at 72 C for 10 sec. To verify PCR specificity, the PCR product was subjected to melting curve analysis. Each experimental sample was run in duplicates together with positive and negative controls. Data were analyzed using the LightCycler Software 3.5.3 (Roche Molecular Biochemical, Bromma, Sweden). Glyceraldehyde-3-phosphate dehydrogenase (forward, 5'-TGC ACC ACC AAC TGC TTA GC –3'; and reverse, 5'-GGC ATG GAC TGT GGT CAT GAG-3') mRNA was analyzed in all samples to control for equal loading. All samples were analyzed on a 3% agarose gel.
Statistics
Groups were compared using the Mann-Whitney U test. Values are given as means ± SD; a P value less than 0.05 was considered significant.
Results
Castration down-regulates nuclear AR in the normal epithelium, stroma, and vasculature in the rat VP
To study how androgen ablation influences AR signaling in different prostate tissue compartments, VP sections were stained to visualize AR expression. When the AR is activated and bound to a ligand it is localized to the nucleus, whereas the inactivated form is localized to the cytoplasm (28, 29). In intact rats, a strong nuclear immunostaining for the AR was seen in most epithelial cells, in the stroma smooth muscle cells, and in fibroblasts. Some mural cells in arterioles and venules showed nuclear AR staining, whereas endothelial cells were generally unstained. Already 1 d after castration there was a marked decrease in nuclear AR staining in epithelial, stroma, and mural vascular cells (Fig. 1, A and B). AR expression was restored after 2 d of testosterone treatment (Fig. 1C).
FIG. 1. Immunohistochemical analysis of AR expression (A–C) and vessel stability in terms of factor VIII/smooth muscle actin (D–F) staining in control animals, 1 d after castration, and 2 d after testosterone replacement. In addition, immunohistochemical analysis of vessel stability in terms of factor VIII/smooth muscle actin (G and H), BrdU labeling, and the presence of intravascular leukocytes (I and J) in animals treated with PBS vs. tie 2-Fc during testosterone-stimulated prostate growth. Mark the intense AR staining of the nucleus in epithelial, stroma, and mural vascular cells of the arteries and venules in the intact animals (arrows) and the decreased nuclear staining of AR in the same cells, 1 d after castration. Note the sustained staining of AR in the cytoplasm of epithelial cells (B, arrowheads). Testosterone treatment restores the levels of AR (C). D–F, Costaining of smooth muscle cell actin (red) and factor VIII (brown, defined as a stabilized blood vessel), shows a dark brown and reddish color, whereas only factor VIII staining appears more brownish. Castration appeared to cause a decrease in smooth muscle actin staining in venules (E, arrowheads), in particular. A venule with costaining of dark brown and red is enlarged in the control animal (D) and a venule showing only brown staining is enlarged in E. Testosterone treatment restored vessel stabilization (F), and this was prevented by treatment with soluble tie 2 (arrows, G and H). A significant increase in the number of intravascular leukocytes (arrowheads) was seen in animals treated with tie 2-Fc compared with PBS (I and J). Proliferating epithelial and endothelial cells (arrows) are stained brown (I and J). a, Artery; v, vein; c, capillary; and ge, glandular epithelium.
Castration causes a major decrease in the glandular and vascular volumes in the rat VP
As previously described, castration induces a gradual decrease in the VP lobe weight. This decrease is accompanied by a decrease in epithelial cell proliferation (BrdU labeling index) and an increase in apoptosis (Table 1). However, here we show that the blood vessels are also affected. The number of proliferating endothelial cells decreased rapidly, and the endothelial cell apoptosis increased (Table 1 and Fig. 2). The volume density of blood vessels (percentage of prostate volume occupied by blood vessels) was initially decreased, but normalized at d 7. Treatment of 7-d castrated rats with testosterone rapidly increased endothelial cell proliferation and vascular volumes and decreased endothelial apoptosis as described earlier (26).
TABLE 1. Summarized VP lobe weight (g) and stereological and morphological data in control, 1, 3, 7, and 9 d after castration of rats (%)
FIG. 2. Graphical view of vascular regression shown as endothelial cell proliferation index (%, dotted line), endothelial cell apoptosis index (%, dashed-dot line), and volume density of blood vessels (%, bold line) in intact animals and 1, 3, and 7 d after castration. Seven-day castrated testosterone-treated animals during 2 d are shown as d 9. Values are given as means with n = 5–8. a, Significantly different from controls; b, significantly different from d 7; P < 0.05.
Castration down-regulates factors known to inhibit inflammation and to stabilize blood vessels
Having found that the vasculature shrinks after castration in the rat VP, it was of interest to study the mechanisms possibly involved. The protein levels of the tie 2 receptor, which is known to mediate both vascular stabilization by enhancing endothelial-pericyte interactions, and down-regulate inflammation, was decreased as early as 1 d after castration in rats and even more pronounced in mice (Fig 3, A and B). The drop in tie 2 was more pronounced than the drop in vascular volume (Table 1) indicating that the decrease in tie 2 is not only a matter of endothelial cell apoptosis. Ang 1 was decreased in the same manner (Fig. 3, A and B). No obvious differences were observed for ang 2 in rats (Fig. 3, A and B) when proteins were extracted with urea. However, when the same experiment was repeated and the proteins extracted with 0.5% NP-40, ang 2 expression was increased after castration (d 1–3) and subsequently decreased (data not shown). There were no differences in ang 1 or tie 2 expression levels detected between urea-extracted and 0.5% NP-40-extracted rat protein samples. Protein from mice tissues were extracted in all cases with 0.5% NP-40, and castration had no clear-cut effect on ang 2 protein levels. In line with a previous finding (14), VEGF was decreased after castration in rats, but this was evident somewhat later than the change of the angiopoietins, namely 3 d after castration (Fig. 3A). Testosterone administration to 7-d castrated rats over 1–3 d reversed these findings and resulted in an increase in ang 1, tie 2, and VEGF levels (Fig. 3, A and B). In addition, tie 2 and ang 1 expression was examined in castrated rats that had not received any testosterone administration at 7, 10, and 14 d after castration. In these animals, there was no up-regulation of either ang 1 or tie 2 (data not shown). To verify that the bands shown were correct, control Western blots with peptide and intact VP protein extracts from mice or rats were analyzed (Fig. 3C). Both peptide and the corresponding bands in the tissue samples were detected with the antibody, and both bands could be blocked by preincubating the antibody with peptide before usage. The faint unspecific bands were not blocked by this treatment.
FIG. 3. Western blot results of the angiopoietins and VEGF by 7.5–13% SDS-PAGE under reducing conditions in rats (A, urea-extracted proteins) and mice (B, 0.5% NP-40-extracted samples). Each group includes five animals showing intact VP (lane 1), and castrated VP 1, 3, and 7 d after treatment (lanes 2, 3, and 4, respectively). Lanes 5–7, Seven-day castrated testosterone-treated animals during 1, 2, and 3 d. Actin was used as a control for equal loading. The experiment was repeated twice using different animals with no apparent change (data not shown). For tie 2, the same results were obtained using a different antibody (data not shown). Western blots with positive and negative controls in terms of control tissues with preblocked antibody (see Materials and Methods) and control peptides (R&D Systems) for ang 2 (C) and ang 1 (D). Ang 1 gave a somewhat larger band than previously reported by other researchers. This could possibly be due to differences in glycosylation. Occasionally a 55-kDa band was identified in intact mouse VP tissue by the antibody against ang 2 (C, lane 2), but in most cases we only observed the 75-kDa band.
Castration induces detachment and/or dedifferentiation in the mural cell layer in prostate blood vessels
To examine the possibility that the decreased ang 1 and tie 2 levels may as in other tissues cause endothelial-mural cell detachment and/or dedifferentiation, the rat prostate vasculature was double stained for factor VIII related antigen and -smooth muscle actin (staining pericytes and vascular smooth muscle cells). This showed that the percentage of blood vessels with a detectable smooth muscle actin covering was decreased already 1 d after castration (Figs. 1, D and E, and4). Testosterone treatment of castrated animals increased the percentage of blood vessels covered with smooth muscle actin (Fig. 1F).
FIG. 4. Graphic picture showing the decrease and increase in vessel stabilization in percent () and relative protein values of ang 1 () and tie 2 () in control rats and 1, 3, 7, and 9–10 d after castration. Protein levels were obtained by scanning Western blots with pooled samples of five animals in each group, using Flour-S Multi Imager (Bio-Rad, Hercules, CA) and relating all values to their intact value, after corrections for corresponding actin levels. Note the high level of ang 1 at d 10. Vessel stabilization was determined by counting vessels with a smooth muscle cells coat vs. those without. Values are means, n = 5–8. a, Significantly different from controls, P < 0.05. The SD varied from 13–24%.
Administration of soluble tie 2 receptor blocks testosterone-stimulated vascular maturation
To test the hypothesis that angiopoietin signaling is of importance for testosterone-stimulated vascular growth and maturation, we treated 7-d castrated mice with a local deposition of 10 μg tie 2-Fc or PBS in Matrigel. The use of Matrigel as a vehicle for local administration of growth factors to the prostate has been described earlier (30). After injection, the Matrigel formed a gel inside and partly around the prostate lobe. These mice were then treated with daily injections of testosterone at d 7, 8, and 9 and examined at d 10 after castration. Castrated mice without Sustanon treatment and castrated mice that had been given 10 mg/kg Sustanon were used as controls. We found that blockage of angiopoietin signaling inhibited vascular maturation in terms of fraction of blood vessels covered with smooth muscle actin (Table 2, P = 0.004; Fig. 1, G and H) compared with controls. In addition, the volume density of factor VIII stained blood vessels was significantly increased (Table 2, P = 0.01) and the apparent dilated blood vessels contained more leukocytes than mice injected with PBS alone (Table 2, P = 0.02; Fig. 1, I and J). However, treatment with tie 2-Fc did not influence endothelial or epithelial cell proliferation (Table 2; Fig. 1, I and J). When animals injected with PBS and Matrigel were compared with noninjected animals it was evident that Matrigel, which contains a cocktail of growth factors, stimulated endothelial cell proliferation and possibly vascular volume (Table 2). The Matrigel itself did not however show any effect on any of the other variables studied. In addition, no signs of tissue damage, due to the amount of matrigel injected, could be detected by microscopy.
TABLE 2. Endothelial and epithelial BrdU labeling index (%), percentage of blood vessels covered with smooth muscle actin (SMA %), volume density of blood vessels (Vv FVIII), and volume density (Vv) of leukocytes in VP blood vessels (%) after 3 d testosterone (T) treatment of castrated (C) mice
The angiopoietins are expressed in the epithelium and stroma in rats
Finally, it was of interest to know where the angiopoietins are localized and act. Thus, rat VP from two intact animals was removed, and LCM technology was used to obtain pure fractions of epithelial cells and stroma cells. The mRNA expression of ang 1 and ang 2 was detected in both the stroma and epithelial cells, whereas tie 2 was exclusively expressed in the stroma fraction (Fig. 5). Similar results were seen in mice (data not shown) and semiquantitatively, there was apparently a higher ang 1 expression in the stroma compartment compared with the epithelial fraction (Fig. 5).
FIG. 5. Levels of ang 1, ang 2, and tie 2 mRNA, in the epithelial compartment and the stroma compartment in the VP of one rat, analyzed on a 3% agarose gel. Lane 1 shows a positive control and lane 4 shows a negative control. Glyceraldehyde-3-phosphate dehydrogenase was analyzed in all samples to ensure equal loading. mRNA data from mice showed similar results (data not shown).
Discussion
In the normal prostate, earlier studies have shown that castration-induced prostate regression is preceded by endothelial cell death and a decrease in blood flow (3, 4, 28). We previously suggested that these effects were caused by a drop in VEGF (11, 12, 26), but our present observations indicate that the angiopoietin signaling pathway is also involved. The regulation of ang 1 (but apparently not ang 2) and tie 2 by testosterone, the observation of vascular changes that have been related to angiopoietins in other tissues, and the results of local treatment with tie 2 decoy receptor during testosterone stimulation clearly indicate that changes in ang 1 signaling are of importance for vascular growth and regression in the prostate gland.
Vascular destabilization, often measured as a decrease in the percentage of blood vessels covered with smooth muscle actin, is assumed to be caused by mural cell detachment and is seen in other tissues after decreased ang 1 or increased ang 2 signaling (16, 17). However, as some mural vascular cells in the prostate have AR, the reduced smooth muscle staining seen after castration could also be a direct sign of vascular cell dedifferentiation (loss of smooth muscle actin expression but not necessarily detachment), due to absence of androgens.
Castration causes a rapid influx of leukocytes into the VP (31). The mechanism regulating this has not been studied, but in other tissues, leukocyte migration is started by up-regulation of leukocyte adhesion molecules on the endothelium. In several other tissues, leukocyte adhesion is inhibited by ang 1 (20, 21, 22) suggesting that the leukocyte accumulation seen in VP already at 24 h after castration (31) could be related to the decrease in ang 1 and tie 2 expression. In line with this, treatment with tie 2-Fc increased the number of leukocytes present in prostate blood vessels compared with controls.
Another novel observation in this study is that vascular mural cells, mainly smooth muscle cells in prostate arterioles and venules, express AR and that this expression is rapidly down-regulated after castration. Arteriolar and venular smooth muscle cells in the VP are apparently direct target cells for androgens. In contrast, previous studies based on a different AR antibody (3) did not show any expression in the vasculature. However, using the same antibody as in the present study, Prins et al. (32) also detected AR in vascular smooth muscle cells in the rat VP. Therefore, it cannot be excluded that the decrease in prostate blood flow (1, 2), total vascular volume, and loss of smooth muscle actin staining can be caused by direct effects. In line with several previous studies (12, 28, 32) prostate endothelial cells do not express AR, suggesting that the decrease in endothelial cell proliferation and increase in endothelial cell apoptosis, seen mainly in micro vessels, must be indirectly caused by altered signaling from surrounding AR-positive mural cells. More specifically, because these cells in general were not covered by smooth muscle cells, the signals may come from more remote cells such as stromal or glandular epithelial cells.
To understand how castration-induced vascular regression is coordinated with glandular and stroma involution, the cellular sites of angiopoietin synthesis and action should be considered. In a human prostate study, ang 1 was found to be localized mainly in basal epithelial cells, whereas ang 2 was found both in smooth muscle cells and epithelial cells (33). In this study of rodents, ang 1 and ang 2 were found both in the prostate stroma and glandular compartments, although ang 1 showed a relatively higher expression in the stroma than in the epithelial fraction. This is in line with other tissues where ang 1 is produced principally by pericytes and other vascular mural cells. The difference in ang 1 expression in humans and rodents could reflect a species difference (for example; basal epithelial cells are considerably more common in humans than in rodents). Tie 2 mRNA were exclusively found in the stroma, as expected, because numerous studies have concluded tie 2 to be expressed on endothelial cells (15, 16, 17, 18, 19) specifically. In addition, we have detected ang 1 and ang 2 protein in a rat prostate endothelial cell line (YPEN-1; our unpublished observations), suggesting a possible autocrine tie 2 stimulation in rats. Therefore, androgens may regulate the prostate vasculature by direct effects on mural vascular cells and or indirectly by influencing the secretion of paracrine regulators such as ang 1.
Testosterone stimulates growth of the prostate gland by affecting epithelial and endothelial proliferation and apoptosis (12, 26). By increasing VEGF synthesis in prostate epithelial cells glandular growth is coordinated with that of the vasculature (12). In this study, we show that testosterone-stimulated prostate vascular growth is accompanied by increased levels of ang 1 and tie 2 and by using a tie 2 decoy receptor administered in Matrigel we could examine the functional role of angiopoietin signaling during testosterone-stimulated prostate growth. Interestingly, and as could be predicted from the effects of castration, reduced angiopoietin signaling resulted in a delay in vascular maturation (fraction of vessels with smooth muscle actin covering) and an increased number of intravascular leukocytes. This finding is line with studies showing that inhibition of tie 2 signaling increase local inflammation in tumors (34). The blood vessels were significantly larger in the tie 2-Fc group, probably due to the lack of the stabilizing effect of ang 1, and the presence of high VEGF levels in the growing prostate. This effect is supported by several other studies that also have shown large dilated blood vessels when ang 1 signaling has been perturbed in knockout experiments (35) or by using nonfunctional tie 2 receptors (36, 37). The angiopoietins have not been shown to exert a proliferative effect on endothelial cells, and in line with this, we did not see any difference in proliferation between the tie 2-Fc group and the PBS group, although the Matrigel itself seemed to have a stimulatory effect on the proliferation of endothelial cells. Whether the vessel phenotype described should impair testosterone-stimulated growth of the prostate due to nonfunctional blood vessels, or the opposite, namely support growth of the prostate, due to an increased angiogenesis/vascular volume, is not known. A possible increased prostate weight could also be an indirect effect of leaky blood vessels due to the accumulation of fluids. Long-term inhibition of tie 2 signaling decreases growth of malignant human astrocytoma in sc and intracranial xenografts (36) and in a murine breast cancer cell line when injected into nude mice (37). In our study, we weighed the prostate after 3 d of testosterone stimulation but found no significant changes in prostate weight. Further studies blocking angiopoietin signaling for a longer time are probably needed to elucidate whether this, as blocking VEGF (12), is a way to perturb testosterone-stimulated prostate growth.
In conclusion, the present study suggests that androgen-regulated levels of ang 1/tie 2 signaling mediate vascular regression and growth in the prostate. This, in turn, precedes hormonally induced regression and growth of prostate glands. Interestingly, similar results implicated a role for the angiopoietins in ACTH-regulated vascular growth and regression of the adrenal gland (38). This raises the possibility of a general mechanism for hormones to coordinate growth and regression of various organs by regulating the vasculature through the angiopoietins.
Acknowledgments
We thank Mrs. Birgitta Ekblom, Mrs. Pernilla Andersson, Mrs. Sigrid Kilter, Mrs. ?sa Skytt, Mrs. Elisabeth Dahlberg, and Mrs. Ulla-Stina Spetz for their professional technical assistance and contribution to this paper.
References
Westin P, Stattin P, Damber J-E, Bergh A 1995 Castration therapy rapidly induces apoptosis in a minority and decreases cell proliferation in a majority of human prostatic tumors. Am J Pathol 146:1368–1375
Ohlson N, Wikstrom P, Stattin P, Bergh A 2005 Cell proliferation and apoptosis in prostate tumors and adjacent non-malignant prostate tissue in patients at different time-points after castration treatment. Prostate 62:307–315
Lek?s E, Johansson M, Widmark A, Bergh A, Damber J-E 1997 Decrement of blood flow precedes the involution of the ventral prostate in the rat after castration. Urol Res 25:309–314
Shabsigh A, Chang DT, Heitjan DF, Kiss A, Olsson CA, Puchner PJ, Buttyan R 1998 Rapid reduction in blood flow to the rat ventral prostate gland after castration: preliminary evidence that androgens influence prostate size by regulating blood flow to the prostate gland and prostatic endothelial cell survival. Prostate 36:201–206
Kurita T, Wang YZ, Donjacour AA, Zhao C, Lydon JP, O'Malley BV, Isaacs JT, Dahiya R, Cunha GR 2001 Paracrine regulation of apoptosis by steroid hormones in the male and female reproductive systems. Cell Death Differ 8:192–200
Cooke PS, Buchanan DL, Young P, Setiawan T, Brody J, Korach KS, Taylor J, Lubahn DB, Cunha GR 1997 Stromal estrogen receptors mediate mitogenic effects of estradiol on uterine epithelium. Proc Natl Acad Sci USA 94:6535–6540
Jain RK, Safabakhsh N, Sckell A, Chen Y, Jiang P, Benjamin L, Yuan F, Keshet E 1998 Endothelial cell death, angiogenesis, and micro vascular function after castration in an androgen-dependent tumor: role of vascular endothelial growth factor. Proc Natl Acad Sci USA 95:10820–10825
Stewart RJ, Panigrahy D, Flynn E, Folkman J 2001 Vascular endothelial growth factor expression and tumor angiogenesis are regulated by androgens in hormone responsive human prostate carcinoma: evidence for androgen dependent destabilization of vascular endothelial growth factor transcripts. J Urol 165:688–693
Westin P, Bergh A, Damber J-E 1993 Castration rapidly results in a major reduction in epithelial cell number in the rat prostate, but not in the highly differentiated Dunning R3327 prostatic adenocarcinoma. Prostate 22:65–74
Brandstrom A, Westin P, Bergh A, Cajander S, Damber JE 1994 Castration induces apoptosis in the ventral prostate but not in an androgen-sensitive prostatic adenocarcinoma in the rat. Cancer Res 54:3594–3601
Haggstrom S, Wikstrom P, Bergh A, Damber JE 1998 Expression of vascular endothelial growth factor and its receptors in the rat ventral prostate and Dunning R3327 PAP adenocarcinoma before and after castration. Prostate 36:71–79
Lissbrant IF, Hammarsten P, Lissbrant E, Ferrara N, Rudolfsson SH, Bergh A 2004 Neutralizing VEGF bioactivity with a soluble chimeric VEGF-receptor protein flt(1–3) IgG inhibits testosterone stimulated prostate growth in castrated mice. Prostate 58:57–65
H?ggstr?m S, Stattin P, Wikstr?m P, Bergh A, Damber JE 2001 Castration-induced reduction of vascular endothelial growth factor expression in benign human prostate tissue is lost in advanced prostate cancer. BJU Int 88:110–116
Burchardt M, Burchardt T, Chen MW, Hayek OR, Knight C, Shabsigh A, de la Taille A, Buttyan R 2000 Vascular endothelial growth factor-A expression in the rat ventral prostate gland and early effects of castration. Prostate 43:184–194
Thurston G, Suri C, Smith K, McClain J, Sato TN, Yancopoulos GD, McDonald DM 1999 Leakage resistant blood vessels in mice transgenically over expressing angiopoietin 1. Science 286:2511–2514
Thurston G, Rudge JS, Ioffe E, Zhou H, Ross L, Croll SD, Glazer N, Holash J, McDonald DM, Yancopoulos GD 2000 Angiopoietin 1 protects the adult vasculature against plasma leakage. Nat Med 6:460–463
Papapetropoulos A, Garcia-Cardena G, Dengler TJ, Maisonpierre PC, Yancopoulos GD, Sessa WC 1999 Direct actions of angiopoietin 1 on human endothelium: evidence for network stabilization, cell survival, and interaction with other angiogenic growth factors. Lab Invest 79:213–223
Kwak HJ, So JN, Lee SJ, Kim I, Koh GY 1999 Angiopoetin 1 is an apoptosis survival factor for endothelial cells. FEBS Lett 448:249–253
Papapetropoulos A, Fulton D, Mahboubi K, Kalb RG, O’Connor DS, Li F, Altieri DC, Sessa WC 2000 Angiopoietin 1 inhibits endothelial cell apoptosis via the Akt/survivin pathway. J Biol Chem 275:9102–9105
Gamble JR, Drew J, Trezise L, Underwood A, Parsons M, Kasminkas L, Rudge J, Yancopoulos G, Vadas MA 2000 Angiopoietin 1 is an antipermeability and anti-inflammatory agent in vitro and targets cell junctions. Circ Res 29: 603–607
Pizurki L, Zhou Z, Glynos K, Roussos C, Papapetropoulos A 2003 Angiopoietin 1 inhibits endothelial permeability, neutrophil adherence and IL-8 production. Br J Pharmacol 139:329–336
Haggstrom Rudolfsson S, Johansson A, Franck Lissbrant I, Wikstrom P, Bergh A 2003 Localized expression of angiopoietin 1 and 2 may explain unique characteristics of the rat testicular microvasculature. Biol Reprod 69:1231–1237
Holash J, Wiegand SJ, Yancopoulos GD 1999 New model of tumor angiogenesis: dynamic balance between vessel regression and growth mediated by angiopoietins and VEGF. Oncogene 18:5356–5362
Lobov IB, Brooks PC, Lang RA 2002 Angiopoietin 2 displays VEGF dependent modulation of capillary structure and endothelial cell survival in vivo. Proc Natl Acad Sci USA 99:11205–11210
Bergh A, Widmark A, Damber JE, Cajander S 1986 Are leukocytes involved in the human chorionic gonadotropin-induced increase in testicular vascular permeability? Endocrinology 119:586–590
Franck-Lissbrant I, Haggstrom S, Damber JE, Bergh A 1998 Testosterone stimulates angiogenesis and vascular re-growth in the ventral prostate in castrated adult rats. Endocrinology 139:451–456
Wikstrom P, Lissbrant IF, Stattin P, Egevad L, Bergh A 2002 Endoglin (CD105) is expressed on immature blood vessels and is a marker for survival in prostate cancer. Prostate 51:268–275
Buttyan R, Ghafar MA, Shabsigh A 2000 The effects of androgen deprivation on the prostate gland: cell death mediated by vascular regression. Curr Opin Urol 10:415–420
Zhou ZX, Sar M, Simental JA, Lane MV, Wilson EM 1994 A ligand-dependent bipartite nuclear targeting signal in the human androgen receptor. J Biol Chem 269:13115–13123
Marengo SR, Chung LW 1994 An orthotopic model for the study of growth factors in the ventral prostate of the rat: effects of epidermal growth factor and basic fibroblast growth factor. J Androl 15:277–286
Wilson MJ, Whitaker JN, Sinha AA 1991 Immunocytochemical localization of cathepsin D in rat ventral prostate: evidence for castration-induced expression of cathepsin D in basal cells. Anat Rec 229:321–333
Prins GS, Birch L, Greene GL 1991 Androgen receptor localization in different cell types of the adult rat prostate. Endocrinology 129:3187–3199
Lind AJ, Wikstrom P, Granfors T, Egevad L, Stattin P, Bergh A 2004 Angiopoietin 2 expression is related to histological grade, vascular density, metastases, and outcome in prostate cancer. Prostate 62:394–399
Melani C, Stoppacciaro A, Foroni C, Felicetti F, Caré A, Colombo MP 2004 Angiopoietin decoy secreted at tumor site impairs tumor growth and metastases by inducing local inflammation and altering neoangiogenesis. Cancer Immunol Immunother 53:600–608
Suri C, Jones PF, Patan S, Bartunkova S, Maissonpierre PC, Davis S, Sato TN, Yancopoulos GD 1996 Requisite role of angiopoietin 1, a ligand for the tie 2 receptor, during embryonic angiogenesis. Cell 87:1171–1180
Zadeh G, Qian B, Okhowat A, Sabha N, Kontos CD, Guha A 2004 Targeting the tie 2/tek receptor in astrocytomas. Am J Pathol 164:467–476
Stratmann A, Acker T, Burger AM, Amann K, Risau W, Plate KH 2001 Differential inhibition of tumor angiogenesis by tie 2 and vascular endothelial growth factor receptor-2 dominant negative receptor mutants. Int J Cancer 273–282
Féraud O, Mallet C, Vilgrain I 2003 Expressional regulation of the angiopoietin 1 and 2 and the endothelial-specific receptor tyrosine kinase Tie 2 in adrenal atrophy: a study of adrenocorticotropin-induced repair. Endocrinology 144:4607–4615(Anna Johansson, Stina H?g)