cAMP-responsive element-binding protein regulates vascular endothelial growth factor expression: implication in human prostate cancer bone metastasis.

Aberrant expression of vascular endothelial growth factor (VEGF) is associated with human prostate cancer (PCa) metastasis and poor clinical outcome. We found that both phosphorylation of cyclic AMP-responsive element-binding protein (CREB) and VEGF levels were significantly elevated in patient bone metastatic PCa specimens. A PCa ARCaP progression model demonstrating epithelial-to-mesenchymal transition exhibited increased CREB phosphorylation and VEGF expression as ARCaP cells became progressively more mesenchymal and bone-metastatic. Activation of CREB induced, whereas inhibition of CREB blocked, VEGF expression in ARCaP cells. CREB may regulate VEGF transcription via a hypoxia-inducible factor-dependent mechanism in normoxic conditions. Activation of CREB signaling is involved in the coordinated regulation of VEGF and may pre-dispose to PCa bone metastasis.

Variation in dermcidin expression in a range of primary human tumours and in hypoxic/oxidatively stressed human cell lines.

Dermcidin acts as a survival factor in a variety of cancer cell lines under hypoxia or oxidative stress. The aim of this study was to evaluate dermcidin expression in cell lines following simulation of tumour microenvironmental conditions and in a range of primary tumours. Tumour tissues were collected from patients with oesophageal (28 samples), gastric (20), pancreatic (five), bile duct (one) and prostatic (52) carcinomas as well as 30 benign tissue samples, for assessment of dermcidin mRNA levels using real-time PCR. Dermcidin expression was assessed in prostatic and pancreatic cancer cell lines, with and without induction of hypoxia or oxidative stress. Dermcidin mRNA expression was very low or absent in both unstressed and stressed prostate cell lines. None of the primary prostate tissue, benign or malignant, expressed dermcidin mRNA. Only two (4%) of the gastro-oesophageal cancer samples expressed moderate quantities of dermcidin mRNA. However, three (60%) of the pancreatic cancer samples and the single cholangiocarcinoma specimen had moderate/high levels of dermcidin expression. Of the two pancreatic cancer cell lines, one expressed dermcidin moderately but neither showed a response to hypoxia or oxidative stress. Expression of dermcidin in human primary tumours appears highly variable and is not induced substantially by hypoxia/oxidative stress in cell line model systems. The relationship of these findings to dermcidin protein levels and cell survival remains to be determined.

Micro-engineered remote palpation device for assessing tissue compliance.

This paper concerns the operation of the actuator for a prototype micro-engineered mechanical palpation device for deployment via a cystoscope to measure the dynamic mechanical properties of the prostate gland in vivo. The subassembly consists of a 400x200 microm silicon (Si) piston manufactured using deep reactive ion etching (DRIE) housed within an anodically bonded glass-Si-glass sandwiched housing. The micro-channel on the Si layer was formed by powder blasting and contains the micro-piston with one end pointing to the side of the housing and the other facing a via hole leading to a capillary tube. The opening on the side of the housing was sealed by a 5 microm thick silicone membrane which acts to retain the micro-piston and act as a return spring. A 320 microm diameter capillary forms the connection between the micro-channel and a micro-syringe which is operated by a programmable syringe pump to produce a reciprocating action. A pressure sensor is connected along the capillary tube to measure the dynamic pressure within the system. The micro-piston has already been used, separately actuated to measure the dynamic mechanical properties of known viscoelastic materials and prostate tissue. The purpose of the present work is to assess the functionality of the actuator assembly.

A variant epidermal growth factor receptor protein is similarly expressed in benign hyperplastic and carcinomatous prostatic tissues in black and white men.

BACKGROUND AND OBJECTIVE: Anti-epidermal growth factor receptor strategies are now established in cancer treatment We have recently described the presence of EGFRvIII (a variant EGFR) in prostatic tumours from UK white men and this is now a target for anti-prostate cancer treatments. However, there has been no report on the expression of this abnormal protein in black men. MATERIALS AND METHODS: We determined EGFRvIII expression in sections of normal, benign hyperplastic (BPH) and carcinomatous (CaP) prostatic archival tissues from Nigerian men and UK white men using streptavidin immunohistochemical techniques. The EGFRvIII immunoreactivity was scored visually using a semi-quantitative method and the results compared statistically. RESULTS: EGFRvIII expression increased with increasing malignancy in both study populations (CaP > BPH > Normal p, <0.0001). Furthermore, EGFRvIII expression was similar in both BPH and CaP tissues in black and white men (p, 0.86 and 0.31 respectively). CONCLUSION: These results demonstrate that EGFRvIII immunoreactivity in prostatic tumours in black men is similar to that in white men. Anti-cancer treatments directed at the EGFRvIII should be equally effective in men from both subpopulations.

In-vitro dynamic micro-probing and the mechanical properties of human prostate tissues.

In vitro macro- and micro-indentation test systems have been designed to measure the dynamic micro-mechanical properties of human prostate tissues at actuation frequencies between 5 Hz and 30 Hz, and 0.5 Hz and 20 Hz, respectively. The development of in vitro test systems was aimed at assessing the capacity of such an in vivo medical probe to provide information useful for the diagnosis of various prostate diseases. The macro-indentation test system is an established one, which we have used to determine structure-property relationships in human and canine prostate tissues and here we use it to validate a newly-developed micro-indentation test system using a tissue phantom. Mechanical testing was also carried out on sections of prostate tissue harvested from cystectomy and radical prostatectomy, diagnosed with bladder cancer and benign prostatic hyperplasia. Dynamic probing under displacement control was carried at pre-strains between 5% and 8% for macro-probing and at 5% pre-strain for micro-probing, and the general effect of pre-strain on the dynamic mechanical properties (described by the amplitude ratio between stress and strain, and the phase lag between strain and stress) of phantom and prostate tissues is presented. Specific point probing on epithelial and stromal histological components was also carried out showing a significant difference between the amplitude ratios of epithelial and stromal components for actuation frequencies exceeding 5 Hz. However, no significant difference was found between phase lags for epithelial and stromal tissues.

Characterization of the retinol and retinoic acid binding proteins in the human prostate.

A retinol-binding protein has been detected in the cytosol of human prostates with benign hyperplasia. The binding was of high affinity and specific for retinol (Kd = 35 nM), with other retinoids such as trans-retinoic acid, retinal, and the synthetic analogues, all-trans-9-(4-methoxy-2,3,6-trimethylphenyl)-3,7-dimethyl-2,4,6,8-nona tetraenoic acid and p-[(E)-2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)-1-propenyl] benzoic acid, showing little or no competition. The retinol binding, which sedimented as a 2S component on sucrose density gradients, was also unaffected by the addition of excess unlabeled steroid hormones. Furthermore, pretreatment of the cytosol proteins with heat and/or trypsin totally abolished the retinol binding. Parallel experiments with trans-retinoic acid suggest that the hyperplastic prostate possesses a second retinoid-binding site which is specific for retinoic acid and distinct from the retinol-binding component. Experiments with serum from patients with benign prostate hyperplasia revealed no binding at the 2S sedimentation position; this suggests that the retinoid-binding proteins were exclusively associated with prostatic tissue and were not therefore derived from serum.

Malignant transformation of human prostatic epithelium is associated with the loss of androgen receptor immunoreactivity in the surrounding stroma.

The cellular pathways involved in the pathogenesis of hormone resistance remain unclear. Studies evaluating the role of changes in human androgen receptor (hAR) expression in the progression of prostatic tumors have been inconclusive. Androgenic influence over prostatic growth is mediated via the regulation of interactions between stromal and epithelial cells. We hypothesized that neoplastic transformation of the prostate would be associated with alterations in hAR expression in the adjacent stroma. Using immunohistochemical techniques, we determined hAR positivity in the epithelium and adjacent stroma of sections from 17 benign and 39 malignant prostatic glands. We found that whereas the expression of the receptor decreased in both cellular compartments as the tissues dedifferentiated, the depletion was more pronounced in the stromal nuclei (P<0.0001). However, in sections from both untreated and hormone-resistant prostate cancer tissues, although heterogeneity of hAR expression in malignant epithelia was increased, there appeared to be a unique field effect around the cancerous prostate glands that resulted in a decreased expression of the receptor in the adjacent benign glands and its total loss in the surrounding stroma. The loss of hAR in the stroma adjacent to malignant prostatic epithelium may play an important role in prostate cancer progression. Furthermore, the similarity of the lack of stromal hAR expression in newly diagnosed and hormone-resistant prostate cancer (P = 0.85) may be an indication that the mechanisms responsible for the acquisition of hormone independence are established early in the malignant transformation process.

In vitro expression of endothelin-1 (ET-1) and the ETA and ETB ET receptors by the prostatic epithelium and stroma.

RT-PCR analysis of total RNA prepared from the prostate cancer cell lines DU145 and PC3 and from primary epithelial cells indicated the presence of endothelin-1 (ET-1) messenger RNA (mRNA). Neither the LNCaP cell line nor primary prostatic stromal cells possess ET-1 mRNA transcripts. Seventy-two-hour-conditioned media derived from DU145, PC3, and primary epithelia contain immunoreactive ET concentrations equivalent to 0.814 +/- 0.048, 0.330 +/- 0.050, and 0.856 +/- 0.055 fmol/mL/10(6) cells after 72 h, respectively. Basal immunoreactive ET secretion was exhibited by LNCaP (0.029 +/- 0.009 fmol/mL/10(6) cells after 72 h) and stromal cells (0.067 +/- 0.007 fmol/ mL/10(6) cells after 72 h). Examination of ETA and ETB gene expression by RT-PCR demonstrates that ET receptor mRNA is almost completely undetectable in the prostate cancer cell lines. Both ETA and ETB mRNAs are detectable in primary cultures of prostatic epithelia and stroma. Competitive binding studies demonstrate a single class of binding site in both primary benign epithelia (dissociation constant = 1.85 x 10(-10) mol/L; maximal binding capacity = 2.7 x 10(4) binding sites/cell), and stroma (dissociation constant = 1.93 x 10(-10) mol/L; maximal binding capacity = 3.7 x 10(5) binding sites/cell). Use of selective ET receptor antagonists confirmed that the predominant stromal receptor subtype expressed in vitro is ETB. This receptor seems not to be coupled to mitogenic pathways because no growth response to exogenous ET-1 or cooperation between ET-1 and bFGF could be observed. Similarly, no effect of ET-1 or the ET-converting enzyme inhibitor, phosphoramidon, on benign epithelial cells could be observed over a 4-day period.

The insulin-like growth factor type I receptor stimulates growth and suppresses apoptosis in prostatic stromal cells.

Stromal cells derived from benign prostatic hyperplasia synthesize and secrete measurable levels of insulin-like growth factor (IGF). Seventy-two-hour conditioned medium obtained from these cells contains IGF-II at levels ranging from 125-177 ng/mL.10(6) cells. IGF-I is almost undetectable. RT-PCR analysis has demonstrated that the genes for both the type I IGF receptor (IGF-IR) and the type II IGF receptor (IGF-IIR) are expressed by benign stromal cells in vitro. Competition binding analysis for IGF-I and IGF-II confirmed the existence of binding sites for both ligands with respective Kd and binding capacities of 4.9 x 10(-9) mol/L and 6.6 x 10(5) sites/cell and 4.7 x 10(-9) mol/L and 3.8 x 10(6) sites/cell. Under serum-free conditions, IGF-I and IGF-II at 500 ng/mL induce 80% and 113% increases in stromal cell density, respectively, over a 96-h period. Incubation with the IGF-IR-neutralizing antibody alpha IR3 (50 micrograms/mL) reduces the rate of stromal cell proliferation by approximately 60-80% even in the presence of stimulatory concentrations of IGFs. Camptothecin-induced apoptosis is inhibited by the addition of IGF-I and -II (500 ng/mL). alpha IR3 suppresses these survival signals and itself induces cell death in the prostatic stroma. The data suggest that IGF-IR is a pivotal molecule in prostatic stromal cell maintenance, and that specific antagonism may offer a novel means of controlling the fibromuscular expansion characteristic of benign prostatic hyperplasia.

Evidence that human prostatic 5 alpha-reductase is located exclusively in the nucleus.

The 5 alpha-reductase activities in human prostatic nuclei and microsomes were compared. The activities in both subcellular fractions were identical with respect to pH dependence, heat inactivation and Michaelis constants for NADPH and testosterone. Subcellular distribution studies using DNA and nicotinamide mononucleotide adenyl transferase as nuclear markers showed that the amount of 5 alpha-reductase present in the microsomes was directly proportional to the amount of nuclear contamination. These results indicate that human prostatic tissue contains only one form of 5 alpha-reductase, which is located exclusively in the nucleus. This finding has important implications for the mechanism of steroid action in the prostate.

Levels of circulating intercellular adhesion molecule-1 in patients with metastatic cancer of the prostate and benign prostatic hyperplasia.

Current reports suggest a role for intercellular adhesion molecule-1 (ICAM-1) in the progression of malignancy. The availability of a new antibody makes it possible to measure circulating ICAM-1 (cICAM-1) in human body fluids including serum; this might help in monitoring tumour burden and in providing additional prognostic information. In this study, serum levels of cICAM-1 were measured by an ELISA assay in patients with benign prostatic hyperplasia (BPH; n = 20) and metastatic cancer of the prostate (CaP; n = 25). Serum ICAM-1 concentrations were also measured in a group of healthy men (n = 8). The mean +/- S.E.M. cICAM-1 level for BPH was 339.52 +/- 15.30 ng/ml compared with 263.55 +/- 18.54 ng/ml for CaP. Even though the difference between the two groups was significant (P < 0.005), there was a marked overlap between the individual values in both groups, thus minimising the prognostic value of these measurements in prostate cancer. Endocrine therapy had no notable effect on the serum levels of cICAM-1. The mean +/- S.E.M. cICAM-1 concentrations in serum from a younger group of healthy volunteers was 204.1 +/- 10.38 ng/ml, and this value was significantly lower than that measured in serum from either BPH or CaP. We also undertook some immunohistochemical studies to examine the distribution of ICAM-1 in prostate tissue. We observed focal epithelial cell membrane staining which was exceedingly patchy in both the BPH and cancer specimens. On the basis of these studies, we suggest that cICAM-1 levels do not provide additional information on patients with metastatic CaP.

A case for synchronous reduction of testicular androgen, adrenal androgen and prolactin for the treatment of advanced carcinoma of the prostate.

The present study was undertaken mainly to investigate whether prolactin manipulation combined with maximal androgen blockage improves the effectiveness of treatment in advanced prostatic cancer. The efficacy of oral hydrocortisone as an alternative to commercial anti-androgens in reducing the adrenal androgens, and of bromocriptine in reducing the prolactin level were also examined. A consecutive series of 30 patients with untreated and advanced prostatic cancer were entered into a three-arm prospective randomised trial. 10 patients received subcapsular orchiectomy alone (arm 1), another 10 had subcapsular orchiectomy plus flutamide (arm 2), and the remaining 10 had subcapsular orchiectomy plus oral hydrocortisone and bromocriptine (arm 3). Clinical and biochemical parameters, including trans-rectal ultrasound-determined prostatic volumes, hormonal profiles and radionuclide bone scan were evaluated at regular intervals. At 12 months, serum testosterone was reduced by more than 90% in all arms, however, maximum suppression of androstenedione, prolactin, and reduction of prostatic volumes were only observed in arm 3; this was reflected by the significant improvement in clinical response in arm 3 compared with other arms. This study suggests that a combined maximal suppression of androgens and prolactin offers a significant improvement in response over conventional treatments without prolactin suppression in the treatment of advanced prostatic cancer. Importantly, a better clinical outcome in arm 3 was still apparent at the end of 36 months.

Isolation and characterization of a cyclic hydroxamic acid from a pollen extract, which inhibits cancerous cell growth in vitro.

One fraction, designated FV-7, in the water soluble ingredient of the pollen extract Cernilton was found to be inhibitory to the growth of a prostate cancer cell line. Characterization of FV-7 by high-resolution mass spectrometry and nuclear magnetic resonance identified the fraction as hydroxamic acid, 2,4-dihydroxy-2H-1,4-benzoxazin-3(4H)-one (DIBOA). To confirm this further, we synthesized an authentic sample of DIBOA and found subsequently that the synthetic DIBOA was structurally indistinguishable from FV-7. Furthermore, in a separate experiment we compared the in vitro effects of FV-7 and DIBOA on the growth of a prostate cancer cell line and found that in both cases the effect was inhibitory and that the inhibition curves obtained for both compounds were virtually identical.

Characterization of the prolactin receptor in human prostate.

The uptake of 125I-labelled human prolactin by subcellular fractions obtained from the human hyperplastic prostate was investigated and the specific binding sites were characterized. The binding was both time- and temperature-dependent with maximum specific uptake achieved after incubation for 20 h at 4 degrees C (dissociation constant = 2.4 x 10(-10) mol/l). The present studies also indicate that the bulk of the specific prolactin binding was confined to the 105 000 and 15 000 g membrane fractions whilst the cytosol and nuclear pellet exhibited a lower capacity for the peptide hormone. Attempts to optimize the binding revealed that pretreatment of the subcellular fractions with 4 mol MgCl2/l quadrupled the binding sites; a similar effect was produced after pretreatment with dextran-coated charcoal, but the changes were less pronounced. Furthermore, our studies suggest that the binding was specific for the human prolactin, with other hormones such as ovine prolactin, human LH, human FSH and human GH showing little or no competition. The addition of magnesium and copper ions to the incubation medium also markedly increased the specific binding, whereas calcium, manganese and EDTA inhibited the prolactin uptake. Freezing and storage of tissue at -70 degrees C did not greatly affect the binding sites. The results suggest that we are clearly dealing with a specific prolactin-binding protein.

Interaction between prolactin and zinc in the human prostate gland.

The interaction between prolactin and zinc was examined in vitro in the human prostate gland. The results indicated that prolactin did not modulate the acute uptake of zinc into benign prostatic hypertrophy tissue whereas zinc, in contrast, increased the uptake of prolactin into the prostate gland. Our study further showed that the augmented uptake of prolactin by zinc was partly due to an increase in the non-specific binding properties of the peptide hormone. We were also able to demonstrate that the specific binding of 125I-labelled human prolactin to the receptor was reduced in the presence of zinc by a competitive mechanism.

Nuclear retinoic acid binding protein in human prostate adenomas.

A retinoic acid binding protein has been detected in salt extracts of nuclei obtained from human prostate adenoma. The binding was characterized by competition experiments, temperature/time studies and saturation analysis. Substantial binding was only observed after sonication of nuclei and charcoal-pretreatment of a salt extract. The binding of radiolabelled all-trans-retinoic acid was displaced by all-trans-retinoic acid, retinol and to a lesser extent retinal and two synthetic retinoids, RO 10-1670 and RO 13-7410. Testosterone and dihydrotestosterone, at a 100-fold excess, had little effect on the binding. The association between retinoic acid and nuclear protein was both temperature and time dependent. At 37 degrees C, equilibrium was rapidly reached (30 min) whereas at 4 and 25 degrees C, ligand binding occurred at a slower rate. Saturation analysis performed under steady-state conditions yielded a dissociation constant of 15 +/- 2 nmol/l. Metabolism studies failed to show conversion of either radiolabelled all-trans-retinol or [3H]retinoic acid in vitro; these data suggest that both acid and alcohol forms of vitamin A are recognized by the extracted nuclear protein. The effect of three enzyme inhibitors on [3H]retinoic acid binding was studied. Binding was unaltered in the presence of aprotinin and phenylmethylsulphonyl fluoride but sodium molybdate (10 mmol/l) increased binding by 18%. The presence of a specific retinoid binding protein in prostate nuclei suggests that retinoids may play some role in the function of the gland.

Effects of tamoxifen on the binding and metabolism of testosterone by human prostatic tissue and plasma in vitro.

The in-vitro metabolism of testosterone in benign and malignant prostatic tissue was examined and distinct quantitative differences between the two types of specimens were observed. The major metabolite of testosterone in the hyperplastic prostate was 5 alpha-dihydrotestosterone and a high 3 alpha(beta)-hydroxysteroid dehydrogenase activity was also detected. In the malignant tissue, 5 alpha-reductase activity was considerably reduced and there was little or no androstanediol formed; the 17 beta-dehydrogenase activity was, however, higher than in the benign tissue. The decrease in 5 alpha-reductase was always followed by a compensatory change in the 3 alpha(beta)-hydroxysteroid dehydrogenase of the malignant prostate. The present study revealed that the ratio of the mean activities of 5 alpha-reductase to 3 alpha(beta)-hydroxysteroid dehydrogenase in the two types of specimen always remained a constant. Although the antioestrogen, tamoxifen, induced an inhibitory effect on the activities of 5 alpha-reductase and 17 beta-hydroxysteroid dehydrogenase in the gland, the present investigation also suggested that tamoxifen stimulated the activity of 3 alpha(3 beta)-hydroxysteroid dehydrogenase. In blood, the action of tamoxifen appeared to be confined to the displacement of androgens from the binding sites on the sex hormone binding globulin.

Androgen metabolism in the epithelial and stromal components of the human hyperplastic prostate.

The reduced and oxidized metabolites of testosterone and dihydrotestosterone were measured in the stromal and epithelial components of 23 human hyperplastic prostates. Our studies indicate differences in the hormonal metabolic patterns of the stroma and epithelium of the resected specimens when compared with tissues obtained retropubically. Testosterone 5 alpha-reductase was evenly distributed between the two components of the specimens obtained retropubically whereas the 3 alpha (beta)-hydroxysteroid dehydrogenase was predominantly located in the stroma. The measurements on the resected specimens suggest, on the other hand, that the bulk of the 5 alpha reductase and 3 alpha (beta)-hydroxysteroid dehydrogenase activities were confined to the stroma although these activities were considerably lower than those measured in the corresponding components of the retropublically obtained specimens. The conversion of testosterone to androstenedione was negligible in all the samples analysed. We therefore conclude that the stroma is the main site for the transformation of dihydrotestosterone to the androstanediol epimers and that the asymmetric distribution of the 3 alpha (beta)-hydroxysteroid dehydrogenase may be instrumental in the development of hyperplasia in the prostate gland. Furthermore, the results of this study indicate that electroresection impairs the enzymatic activities of the tissue.

Localization and characterization of epidermal growth factor receptors on human testicular tissue by biochemical and immunohistochemical techniques.

In the present study attempts were made to characterize the epidermal growth factor (EGF) receptor on human testicular tissue. A radioligand exchange assay with 125I-labelled EGF was used to detect a high affinity, low capacity, single binding site in the 105,000 g particulate fraction of human testicular tissue. Binding was optimal at 32 degrees C following a 40-min incubation with a mean (+/- S.D.) dissociation constant of 327 +/- 59 pmol/l (d.f.9). The number of binding sites ranged from 0.07 to 0.21 pmol/mg protein. Competition studies with other peptide hormones including LH, FSH, prolactin, insulin-like growth factor-I, fibroblast growth factor and nerve growth factor have confirmed the specificity of EGF for its receptor. The receptor was also found to be heat-labile and sensitive to trypsinization. Cross-linking experiments using disuccinimidyl suberate revealed major binding species at the 125 kDa region and this is thought to represent a proteolysed form of the receptor. Immunohistochemical localization of the receptors demonstrated their presence in the interstitial tissue and not within the seminiferous tubules. The presence of specific EGF binding in the interstitial tissue suggests that EGF may play some role in testicular steroidogenesis.

Localization of epidermal growth factor receptors in the human prostate by biochemical and immunocytochemical methods.

The receptor for epidermal growth factor (EGF) was characterized in the particulate fraction from human benign prostatic hyperplasia (BPH) and was present in 85% of tissues analysed. The uptake of 125I-labelled EGF by BPH was dependent on both time and temperature, with maximum specific uptake achieved after incubation for 90 min at 37 degrees C. Binding characteristics revealed two classes of binding sites of higher (mean dissociation constant (Kd) +/- S.D. = 0.8 +/- 0.2 nmol/l) and lower (Kd = 7.6 +/- 2.8 nmol/l) affinities. Competition studies demonstrated the specificity of the receptor assay since the binding of labelled EGF was abolished with excess unlabelled EGF but not with excess unlabelled human GH, human insulin, venom nerve growth factor, human FSH, human LH and human prolactin. There was a complex biphasic relationship between specific binding and protein concentration in the range 0.1-8 mg/ml. Subcellular fractionation of BPH homogenates demonstrated that the bulk of the specific binding was confined to the 800 g (crude heavy pellet) and 15,000 g (mitochondrial pellet) fractions. The 105,000 g (microsomal pellet) and the 105,000 g (cytosol fraction) exhibited low and variable binding capacities for the growth factor. The presence of EGF receptor was also confirmed by immunocytochemical staining of frozen sections from BPH using monoclonal antibody specific for EGF receptors. A positive correlation between 125I-labelled EGF binding and the intensity of staining was found. The presence of a specific EGF-binding receptor protein in human BPH tissues suggests that EGF may play a role in the pathogenesis of human BPH.