Ventral Prostate Epithelial Cells Research Articles

Does the lack of regression-associated mRNA expression render a rat ventral prostate epithelial cell line androgen independent?

A series of rapidly dividing epithelial (RDE) cell lines have been isolated from primary cultures of rat ventral prostate (RVP) epithelial cells. Unlike androgen-dependent secretory epithelial cells, the RDE cells in culture do not express the androgen-dependent secretory proteins, nor do they express the androgen-repressed cell death sequences (TRPM-2) found in the epithelial cells during prostatic regression. Screening of a cDNA clone library established from RDE cell mRNA has yielded a number of RDE cell-specific sequences. One of these, RDE-.25 is a 250-base mRNA. The sequence of RDE-.25 shows considerable homology with the rat growth hormone gene and two murine oncogene sequences. We believe that the absence of androgen-repressed cell death sequence expression confers androgen independence for survival and growth, while the expression of RDE-.25 may represent an autocrine growth stimulus which greatly increases the rate of cell division in these cells.

Separation of mature rat ventral prostate epithelial and fibroblast cells

Several techniques for the separation of rat ventral prostate cells, using density gradient centrifugation or mechanical means have been published, yielding fibroblast and epithelial cell populations of varying purities and viability. These techniques are often tedious, yield relatively limited numbers of cells, demand considerable technical expertise, and result in the isolation of cells of limited viability. Two techniques for the isolation and establishment of epithelial and fibroblast cell cultures from mature rat ventral prostate are described here. Collagenase/trypsin digestion of the tissue yields a single-cell suspension of both cell types that are separated on Percoll isopycnic centrifugation gradients. A continuous gradient system allows for the separation of a greater number of cells with very high degrees of purity. A second technique based on a step-gradient system produces reproducible subfractionation of the epithelial cell component of the prostate within a considerable shorter period of time. An improved medium for plating epithelial and fibroblast cells has also been developed. The separated cells are plated on collagen and/or fibronectin-coated dishes in a serum-free plating medium that is later replaced with a serum containing growth medium. The plating medium greatly increases the plating efficiency of the isolated cell types, particularly the epithelial cells.

Estrogen metabolism by primary cultures of rat ventral prostate epithelial and stromal cells

Estrogen metabolism was examined in primary cultures of rat ventral prostate epithelial and stromal cells developed from young (~3 weeks old) animals. Supraphysiologic concentrations (50 nM) of tritium-labelled estradiol (E 2) and estrone (E 1) were incubated separately with each cell type and the metabolites formed were measured at selected time points over a 24 h period. The metabolites were analyzed using high performance liquid chromatography. Epithelial cells exhibited an equal capability to interconvert E 2 and E 1 thus demonstrating the presence of similar oxidative and reductive activities for 17β-hydroxysteroid oxidoreductase (17β-HSOR) [0.45 and 0.40 pmol/3 h/μg DNA respectively]. In contrast, stromal cells showed a 6-fold lower rate of oxidation of E 2 to E 1 (0.08 pmol/3 h/μg DNA) but exhibited an approx 5-fold higher rate of reduction of E 1 to E 2 (1.81 pmol/3 h/μg DNA). Estriol (E 3) formation from either substrate was not detected in the two cell types. The results demonstrate that rat ventral prostate epithelial cells have similar capabilities to form or remove biologically active E 2. In contrast, prostate stromal cells exhibited a preferential capability to form and possibly maintain high levels of biologically active E 2. These findings are discussed with reference to the actions of estrogens on prostate epithelial-stromal cellular interactions.

Effect of a 4-methyl-4-aza steroid on androgen metabolism by rat ventral prostate epithelial and stromal cell cultures: selective inhibition of 5 alpha-reductase activity.

The effect of a potent steroid metabolic inhibitor, 17 beta-N,N-diethylcarbamoyl-4-methyl-4-aza-5 alpha-androstan-3-one (DMAA), on androgen metabolism was investigated in primary monolayer cultures of rat ventral prostate epithelial and stromal cells. Using testosterone (T) as substrate, 5 alpha-reductase (5 alpha-R) activity in both cell types was inhibited by greater than 98% at an inhibitor concentration of 1000 nM. The concentrations required to produce a 50% inhibition (IC50) were 7.4 and 9.0 nM for epithelial and stromal cells, respectively. To examine the specificity of this compound, its effect on other steroid-metabolic enzymes was examined. DMAA at a concentration of 1,000 nM had no effect on 3 alpha-hydroxysteroid oxidase (3 alpha-HSORox), 3-ketosteroid reductase (3 alpha-HSORred), and 6/7-hydroxylase (6/7-HSH) activities in both cell types; 17 beta-hydroxysteroid oxidase (17 beta-HSORox) activity, located primarily in epithelial cells, also was not influenced by DMAA. In contrast, epithelial 3 beta-hydroxysteroid oxidase (3 beta-HSORox) and 3-ketosteroid reductase (3 beta-HSORred) activities were inhibited by 65% (P less than .001) and 58% (P greater than .05), respectively, albeit the latter result was not statistically significant. Stromal 3 beta-HSORox and 3 beta-HSORred activities were negligible; hence the effect of the inhibitor of these enzymes could not be assessed. In conclusion, DMAA is a relatively selective and potent inhibitor of 5 alpha-R activity in primary cultures of rat ventral prostate epithelial and stromal cells and should be a useful compound for antagonizing androgen-mediated actions in the prostate and other androgen target tissues.

Ultrastructural features of the ventral prostate epithelial cells in the Australian plains rat, Pseudomys australis.

The fine structure of epithelial cells of the small ventral prostate of Pseudomys australis males was studied. The cells were sometimes binucleated, exhibited relatively little granular endoplasmic reticulum, generally few secretory granules (probably reflecting a low proteinaceous secretory activity), but had abundant agranular endoplasmic reticulum (AER), similar to that in steroidogenic cells. Some of the mitochondria also showed tubular cristae. Furthermore, most cells had some irregular dense bodies in the cytoplasm which may have developed, by a gradual transformation process, from the membranes of AER, mitochondria and other organelles; they could be the product of degenerative changes in these organelles. These findings indicate a significant difference in the structure of these cells from those present in the ventral prostate of the hopping mouse, Notomys alexis. It is speculated that this gland secretes relatively little protein but perhaps more lipid or cholesterol-derived substance.

Distribution of microtubules and microfilaments in exocrine (ventral prostatic epithelial cells and pancreatic exocrine cells) and endocrine cells (cells of the adenohypophysis and islets of Langerhans). The relationship between cytoskeletons and epithelial-cell polarity.

We investigated the distribution of microtubules and microfilaments in some exocrine and endocrine cells in rats. Microtubules were stained by applying an immunofluorescent technique using antibodies against beta-tubulin, while microfilaments were stained with rhodamine-phalloidin, which binds selectively to polymerized actin filaments. In the cytoplasm of some exocrine cells (pancreatic acinar cells and ventral prostatic epithelial cells), the microtubules were distributed longitudinally from the apical region to the basal region, but no microtubules were found in the nuclear region. In exocrine cells, most of the microfilaments were localized beneath the apical plasma membrane. In some endocrine cells (those of the adenohypophysis and the islets of Langerhans), the microtubules exhibited a radial or reticular distribution in the cytoplasm, and intense fluorescence was observed in the perinuclear region. The immunofluorescence produced by the antibodies against beta-tubulin was more intense in endocrine cells than in exocrine cells. The microfilaments observed in the endocrine cells studied were homogenously distributed beneath the plasma membrane. Dot-like rhodamine-phalloidin staining was often observed in the cytoplasm of both the exocrine and endocrine cells. The present study clearly demonstrated marked differences in the distribution of cytoskeletal elements in exocrine and endocrine cells, and these may reflect differences in the secretory direction of such cells as well as in epithelial-cell polarity.

Response of glandular versus basal rat ventral prostatic epithelial cells to androgen withdrawal and replacement.

Light microscope autoradiography was utilized to investigate the separate proliferative activities of the basal and glandular cell types during androgen-mediated restoration of the ventral prostatic epithelium in castrated rats. Our objective was to determine whether the proliferative population of the prostatic epithelium is restricted to the basal cell subtype under these experimental conditions. In the intact animal, substantial numbers of basal as well as glandular cells were found to be proliferating. The basal cell labeling index (LI) was 3.75% +/- 0.3% or 1.85 +/- 0.4 x 10(5) labeled cells per prostate. For the glandular cells the comparable figures were 0.38% and 3.32 +/- 1.5 x 10(5). Seven days after surgical castration, both subtypes persisted. However, only the numbers of glandular cells were found to be significantly reduced (66% decrease). When androgens were replaced by daily subcutaneous injection (testosterone propionate, 2 mg/0.2 ml sesame oil), both cell types responded with significantly increased proliferation. The temporal pattern of response displayed by each cell type was very similar. A greater proportion of basal than glandular cells responded, ie, 28% vs 11% LI. However, in terms of approximate numbers of cells labeled during the peak response, the value for the glandular cells was larger, ie, 2.4 vs 1.1 x 10(6) labeled cells. These data clearly indicate that under normal conditions as well as during testosterone stimulation, basal cells are not the sole population of proliferating cells in the prostatic epithelium. Thus, basal cells, as a distinct morphologic population, are not homogeneously undifferentiated stem cells which give rise to differentiated and nonproliferating glandular epithelial cells. Aspects of these data can nevertheless be interpreted to support the idea that the prostatic epithelium is maintained by a single cell lineage system in which the stem cells comprise a subpopulation of the basal cells. Alternatively, the two populations may be separate and distinct as no data are presented which unequivocally links the two cell types in a precursor-progeny relationship.

Primary culture of epithelial cells derived from the rat ventral prostate: formation of three-dimensional acinus-like structure in collagen gel.

Rat ventral prostate epithelial cells were cultured in collagen gel after collagenase digestion. The primary cultures were mainly composed of single and spherical cells. After 10 days incubation in growth medium containing insulin, transferrin, and cholera toxin, there was a 3.8-fold increase in cell numbers, aggregates of which formed three-dimensional acinus-like structures. These structures consisted of one layer of cells surrounding the lumen. The cells were joined together with a junctional complex and had microvilli on the luminal surface and secretory vacuoles in the cytoplasm facing the lumen. The ultrastructural features of the cells were not altered by growth medium containing steroids. This culture system may prove to be very useful in elucidating proliferation, organization, and differentiation of prostatic epithelial cells in relation to the extracellular matrix and stromal cells.

Androgen metabolism and actions in rat ventral prostate epithelial and stromal cell cultures.

The rat ventral prostate requires androgens for normal development, growth, and function. To investigate the relationship between androgen metabolism and its effects in the prostate and to examine differences between the epithelial and stromal cells, we have established a system of primary cell cultures of immature rat ventral prostate cells. Cultures of both cell types after reaching confluency (6-7 days) actively metabolized 3H-labelled testosterone (T), 5 alpha-dihydrotestosterone (5 alpha-DHT), 5 alpha-androstane-3 alpha,17 beta-diol, and 5 alpha-androstane-3 beta,17 beta-diol. The epithelial cells actively reduced T to 5 alpha-DHT and formed significant amounts of 5 alpha-androstane-3,17-dione from T, 5 alpha-DHT, and 5 alpha-androstane-3 alpha,17 beta-diol. All substrates were converted to significant amounts of C19O3 metabolites. The stromal cells also metabolized all substrates, but very little 5 alpha-androstane-3,17-dione was formed. The metabolism studies indicate that both cell types have delta 4-5 alpha-reductase, 3 alpha- and 3 beta-hydroxysteroid oxidoreductase and hydroxylase activities. The epithelial cells have significant 17 beta-hydroxysteroid oxidoreductase activity. The epithelial cells cultures grown in the presence of T have higher acid phosphatase (AP) contents (demonstrated histochemically and by biochemical assay). Tartrate inhibition studies indicate that the epithelial cells grown in the presence of T are making secretory AP. Stromal cell AP is not influenced by T. The results indicate that the cultured cells maintain differentiated prostatic functions: ability to metabolize androgens and, in the case of the epithelial cells, synthesize secretory AP.

In vitro stimulation by prostate extracts of rat ventral prostate stromal and epithelial cell division.

We examined the effect of prostate cell extracts on the replication of highly enriched rat ventral prostate stromal and epithelial cells cultured in RPMI-1640. Extracts from normal rat prostates completely inhibited cell division in both fractions, while a 10% (v/v) extract of Dunning R-3327G adenocarcinoma inhibited replication of stromal cells but permitted that of epithelial cells. The cytotoxic effect of prostate extracts was dialyzable, heat stable, unaffected by proteases, soluble in acid/alcohol, and insoluble in chloroform. These properties are consistent with those of polyamines incubated in the presence of fetal calf serum. Dialyzed extracts from Dunning adenocarcinomas, and from human benign hypertrophic and carcinomatous prostates stimulated rat prostate stromal and epithelial cell division, in keeping with other reports of "growth factors" in prostate tissue. This mitogenic activity was stable to temperature (70 degrees C, 4 hours), and marginally if at all affected by exposure to trypsin or to pronase coupled to Sephadex. Short-term culture of separated prostate cells should provide a useful assay system for detecting putative autocrine or paracrine stromal and/or epithelial cell growth factors and identifying suspected homo- or heterotypic cellular interactions in normal and diseased prostates.

Immunofluorescent localization of an androgen-dependent isoenzyme of prostatic acid phosphatase in rat ventral prostate.

Isoenzymes of rat ventral prostate (RVP) acid phosphatase were isolated and partially purified by ultracentrifugation, Sephadex G-100 column chromatography, and isoelectric focusing. Antisera were raised to the isoenzymes of prostatic acid phosphatase by immunization of New Zealand white rabbits. Rabbit antisera reacting specifically to homologous but not heterologous isoenzymes of acid phosphatase were then reacted with a variety of tissues using indirect immunofluorescence. The tissues included prostate, spleen, bone marrow, liver, kidney, salivary gland complex, small intestine, and adrenal glands. An antiserum against a RVP acid phosphatase isoenzyme with a pI of 4.5 (A-PAP) localized acid phosphatase only in the supranuclear region of rat ventral prostate epithelial cells, and did not react with acid phosphatase in any of the other organs tested. A-PAP did not localize acid phosphatase in the ventral prostate from rats 14 days after castration. A-PAP did localize acid phosphatase in the ventral prostate from castrated animals that were treated with testosterone. These results indicate the A-PAP localized an androgen-dependent isoenzyme of acid phosphatase in RVP epithelial cells that may be secretory in nature. This antiserum should prove to be an ideal marker for studies involving hormonal regulation of prostatic epithelial function in vivo and in vitro.

Characterization of rat ventral prostatic epithelial cells in collagen gel culture

AbstractThe characterization of rat ventral prostatic epithelial cells grown in collagen gel culture was undertaken to determine the usefulness of this culture technique for the study of prostatic growth and differentiation in vitro. The results of these studies demonstrate that embedding prostatic epithelial cells in a matrix of collagen gel permitted rapid proliferation and maintenance of the cells for a prolonged period of time in a state that showed remarkable structural resemblance and physiological responsiveness to that found in vivo. Three‐dimensional outgrowths from dissociated single epithelial cells and/or small aggregates of epithelial cells were obtained which resembled structures actually found within the prostate gland, as determined by histological and visual examination of the cellular growths present within the gels. In addition, the responsiveness of the cells in the collagen gel cultures to various hormonal additives, including insulin, testosterone, prolactin, and estradiol correlated closely to the manner in which these cells have been shown to respond in vivo. The general characteristics of growth of the cells in the collagen gel cultures, the effect of various sera and different pH levels upon growth, and the ability of the cells to produce acid phosphatase were also investigated.

Androgen 5 alpha-reductase and 3 alpha-hydroxysteroid dehydrogenase activities in ventral prostate epithelial and stromal cells from immature and mature rats.

To study androgen-mediated differentiation in the rat ventral prostate, we separated the two principal cell types (epithelial and stromal) derived from prostates of immature and mature rats on two continuous Percoll gradients. Cells were immediately placed in culture medium. Testosterone metabolism by the two prostatic cell types was evaluated using [3H]testosterone and quantifying the formation of 5 alpha-[3H]dihydrotestosterone (5 alpha-DHT) and 5 alpha-[3H]androstane-(3 alpha or 3 beta), 17 beta-diols. In epithelial cells from both immature and mature rat prostates the major testosterone metabolites were 5 alpha-DHT and 5 alpha-androstane-3 alpha, 17 beta-diol. Stromal cells metabolized less testosterone than did the epithelial cells. Differences in the relative levels of the various metabolites were observed for the two age groups. To examine in more detail the changes in testosterone metabolism observed in vitro both types of cells and unfractionated cells from immature and mature rat prostates were assayed for testosterone 5 alpha-reductase (using testosterone as substrate) and 3 alpha-hydroxysteroid dehydrogenase (using 5 alpha-DHT as substrate) activities (expressed as pmol substrate reduced/min per 10(6) cells). In immature rats both 5 alpha-reductase and 3 alpha-hydroxysteroid dehydrogenase activities were localized in the epithelial cell fraction (17 and 52 respectively); stromal cells showed lower 5 alpha-reductase and 3 alpha-hydroxysteroid dehydrogenase activity (4 and 4). Relative to epithelial cells from immature rats epithelial cells from mature rats showed a decrease in 5 alpha-reductase (7) and an increase in 3 alpha-hydroxysteroid dehydrogenase (160) activity while stromal 5 alpha-reductase showed little change (3) and 3 alpha-hydroxysteroid dehydrogenase increased to 22.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparative C19‐Radiosteroid Metabolism in Primary Monolayer Cultures of Epithelial Cells and Fibroblasts from Rat Ventral Prostate, Canine Prostate, and Rat Lung

Stromal‐epithelial cell interactions may be important for the regulation of normal and abnormal prostatic growth. While androgen transformation by rat ventral prostate and canine prostate has been investigated in vivo and in organ culture, little is known about metabolic pathways in cultures of epithelial cells and fibroblasts. Metabolism of radioisotope‐labeled 17β‐hydroxy C19‐steroids was studied in primary cultures of highly‐enriched rat ventral prostate and canine prostate epithelial cells and fibroblasts isolated by selective attachment procedures. The fibroblasts contained little testosterone 5α‐reductase in contrast to high activity in epithelial cells. We found high levels of 5α‐androstane‐3α,17β‐diol (3α‐diol) dehydrogenase and the terminal 5α‐androstane‐3β,17β‐diol (3β‐diol) hydroxylases in both cell types; 3α‐diol was a more effective precursor of 5α‐dihydrotestosterone than was testosterone. For prostatic fibroblasts these pathways seem to be differentiated functions, since rat‐lung fibroblasts converted 3β‐diol to 5α‐dihydrotestosterone and 3α‐diol. We conclude that epithelial cells and fibroblasts make interactive contributions to prostatic androgen metabolism.

Early effects of castration and replacement of androgen on the expression of retrovirus-like particles in rat ventral prostate epithelial cells.

AbstractPrevious work has shown the presence of C-type retrovirus-like particles in normal prostate epithelial cells from Sprague–Dawley rats. We now report attempts to alter retrovirus-like particle expression by castration and replacement of androgen, i.e., testosterone propionate (TP). The frequency of C-type viral-like particle expression in the ventral prostates from the following groups of animals was estimated by a semiquantitative electron microscopic assay: (a) normal rats, (b) normal rats that received TP for 3 days, (c) rats 3 or 6 days after castration, and (d) rats that received TP for 3 days, beginning 3 or 6 days after castration, to induce cellular hypertrophy or hyperplasia, respectively. Three or six days following castration, the frequency of C-type particle expression in the regressed ventral prostates was markedly reduced. Typical C-type viral-like particle morphology was retained for those few particles detected. In contrast, prostates with hypertrophic epithelium from rats treated w...

Modified nutrient medium MCDB 151, defined growth factors, cholera toxin, pituitary factors, and horse serum support epithelial cell and suppress fibroblast proliferation in primary cultures of rat ventral prostate cells

Nutrient medium WAJC 401 containing 5 micrograms/ml insulin. 10 ng/ml EGF, 10 ng/ml cholera toxin, 50 micrograms/ml pituitary extract, 1 microgram/ml prolactin, 1 microM dexamethasone, and 5% horse serum supports the rapid proliferation of rat ventral prostate epithelial cells in primary culture. The same medium suppresses the growth of prostate fibroblasts.

Densitometric and morphometric evaluation of growth in primary cultures of rat ventral prostate epithelial cells.

This study describes a method for establishing primary cultures in 24-well culture vessels and evaluating the effects of serum, hormones, and other factors on cell growth using densitometry. Primary cultures of rat ventral prostate epithelial cells were grown in 24-well culture vessels containing F12K culture medium supplemented with various concentrations of the following substances: fetal bovine serum (FBS), horse serum (HS), testosterone (T), dihydrotestosterone (DHT), hydrocortisone (HC), zinc (Zn), transferrin (TR), cysteine (CYS), glutamine (GLT), selenium (SEL), and ascorbic acid (AC). The effect of each supplement on cell growth was evaluated on fixed and stained cultures using a photovoltaic cell densitometer designed to read the total culture surface of a 16-mm well. Increasing concentrations of both HS and FBS resulted in an increase in cell growth. T, Zn, HC, TR, and AC each had a stimulatory effect on cell growth. CYS, GLT, and DHT had little effect on cell growth, while SEL was inhibitory to cell growth. This data compares favorably with that obtained by other methods, such as morphometric analysis and ornithine decarboxylase production. These results indicate that densitometry is a useful method for determining the effect of supplements on cell growth in primary culture.

Freeze-fracture study of rat ventral prostate: secretory mechanisms in the epithelial cell.

Membrane events during apocrine and merocrine secretion of rat ventral prostate epithelial cells were analyzed by freeze-fracture. Early morphological manifestation of both secretory modes involves focal clearance of microvilli characterized by their progressive inclination and attenuation over the apical membrane. Merocrine secretion includes invasion of the apical microfilament web by the secretory vesicle and clearance of membrane particles at the site of its interaction with the apical plasmalemma. During apocrine secretion, a portion of the plasma membrane cleared of microvilli projects into the lumen. Growth of this projection is accompanied by a progressive, but partial, clearing of membrane particles and results in the formation of a large bleb containing dilated endoplasmic reticulum cisternae. Completion of the process involves "degeneration" of the bleb and its release by constriction of a neck, and possibly, fusion of vesicular or tubular structures. Swelling and blebbing of microvilli are shown to be preparation artifacts.

The culture of hormone-dependent epithelial cells from the rat ventral prostate

This paper describes a method for obtaining cultures of rat ventral prostate epithelial cells. The prostate is first perfused with a collagenase solution before removal from the animal; subsequent mincing and incubation in vitro produces a suspension of alveolar cell clumps. Upon incubation, these clumps attach to the surface of the culture dish and spread into discrete epithelial cell colonies, which both retain differentiated morphology, and secrete a species of plasminogen activator that is characteristic of prostatic tissue. These properties were not observed in cultures prepared from single cell suspensions of the same organ. Maintenance of epithelial colony integrity and secretory activity specifically required the continued presence of stromal cells, glucocorticoids and insulin. Androgenic steroids were much less effective than glucocorticoids in stimulating plasminogen activator secretion and in maintaining colony integrity, in spite of the well-established androgen dependence of prostatic tissue morphology in vivo and in organ culture. Furthermore, no effects of prolactin were observed, either when this hormone was tested alone or in conjunction with steroid hormones. Of 3 retinoids tested, retinal was highly cytotoxic at concentrations in the range of 1 μM, whereas retinol and retinoic acid were without detectable effect.

Isolation of viable rat ventral prostate epithelial and nonepithelial cells.

A procedure is described for the dispersion, partial purification, and identification of epithelial and nonepithelial cells from the rat ventral prostate, the latter based in part upon the ability of cells to bind androgen. Since initial efforts to mechanically disrupt prostatic tissue at 2 C lysed many cells, minced rat prostates were exposed to collagenase at 37 C and further dispersed by repetitive pipetting and passage through a tissue sieve. Washed cells in culture medium were centrifuged through an isokinetic Ficoll gradient from which three visible fraction (1, 2, and 4) and a less definite fraction 3 were harvested. Characterically, fraction 1 contained cellular debris; fraction 2 contained round nucleated cells, fewer elliptically shaped cells, red blood cells, and rare free cell nuclei; fraction 3 contained somewhat larger elliptical and round cells; and fraction 4 contained larger round and elliptically shaped cells. These were epithelial cells, as judged by electron microscopy. Isolated prostate cells from rats castrated for 24 h were incubated with [3H]testosterone; 80-90% of the retained radioactivity, the majority of which was dihydrotestosterone, was associated with washed cells from fraction 4. Similar results were obtained after in vivo administration of labeled androgen and subsequent analysis of radioactivity in cells from these fractions. Histochemically and enzymatically demonstrable formalin-insensitive acid phosphatase was increased in bands 3 and 4, which were enriched in epithelial cells. Many cells in fractions 2-4 were viable before and after exposure to Ficoll, as estimated by their ability to exclude trypan blue, incorporate radioactive uridine into RNA, generate cell monolayers during 1-4 weeks of culture, and actively metabolize androgens. Compared to fraction 4, cells from fraction 2, considered to be enriched in nonepithelial cells, actively metabolized but bound much less [3H]testosterone and its metabolites. A number of epithelial cells in fraction 4 isolated from prostates dissociated at 2 C were associated with typical-C-type RNA viruses.