inductive signals of growth and differentiation from dihydrotestosterone in a manner similar to that observed in the in vivo condition. These results offer an ...
In VitroCell.Dev.Biol. Animal33:375380,May1997 © 1997Societyfor In VitroBiology 1071-2690/97 $05.00+0.00
ESTABLISHMENT OF A THREE-DIMENSIONAL HUMAN PROSTATE ORGANOID COCULTURE U N D E R MICROGRAVITY-SIMULATED CONDITIONS: EVALUATION OF A N D R O G E N - I N D U C E D GROWTH A N D PSA EXPRESSION HAIYEN E. ZHAU,1 THOMASJ. GOODWIN,SHI-MINGCHANG, TACEY L. BAKER, ANDLELANDW. K. CHUNG Molecular Urology and Therapeutics Program, Box 422, Department of Urology, University of Virginia, Charlottesville, Virginia 22908 (H. E. Z., S.-M. C., L. W. K. C.); NASA Johnson Space Center, Houston, Texas 77058 (T. J. G., T L. B.)
SUMMARY A novel in vitro human prostate cancer model was established by using a cocuhure technique in which isolated human prostate fibroblasts were observed to grow as a mixed culture with isolated human prostate cancer cells (LNCaP) on microcarrier beads under microgravity-sinmlated conditions. This model appears to be promising and deserves further exploration because: (a) cocultured human prostate fibroblasts and cancer epithelial cells appear to undergo patterns of histogenesis similar to those observed in human prostate tumors and (b) unlike the conventional cell culture on plastic dishes, cocultnred human prostate fibroblasts and LNCaP cells in microgravity-simulated conditions responded to the inductive signals of growth and differentiation from dihydrotestosterone in a manner similar to that observed in the in vivo condition. These results offer an opportunity to examine molecular mechanisms of cellular signaling in response to androgen stimulation during normal and aberrant human prostate development. The mierogravity-simulated three-dimensional prostate epithelial cell culture with prostate fibroblasts can be further explored as an ideal in vitro model for the study of normal and neoplastic prostate development. This model could also be adopted as a drug screening program for the discovery of novel therapeutic agents in the treatment of human prostate cancer and benign hyperplastie growth. Key words: stromal-epithelial interaction; mierogravity-sinmlated prostate cell culture; three-dimensional prostate cell culture; dihydrotesterone-induced prostate cancer growth and differentiation; prostate-specific antigen expression.
(Chung et al., 1991), as well as the resulting osteoblastic reaction commonly observed in the bone as a result of osseous metastasis (Jacob, 1983). Clearly, one of the difficulties that impedes progress in precisely defining the molecular basis of stromal-epithelial interaction is the lack of an adequate in vitro androgen- responsive system in which the roles of growth-promoting and inhibitory mediators can be precisely delineated. In this communication, we describe the use of a microgravity-simulated cell culture system to define growth- and differentiation- associated events during coculture of prostate stromal and epithelial cells on microcarrier beads, wbich were maintained under microgravity-simulated conditions. This method has been used previously for the establishment of three-dimensional models of cell cultures for bladder (Prewett et al., 1993), ovarian (Becker et al., 1993), and colon (Jessup et al., 1993) cancers and the normal chondrocyte (Freed et al., 1994) and cardiac and skeletal muscle cells (Schroedl et al., 1994). In the present study, we examine the effects of androgen on growth and PSA expression by hmnan prostate cancer cells (LNCaP) either grown alone or as a mixed culture with human prostate fibroblasts. Results of this study show that microgravity-simulated cell culture technology closely mimics the physiologic responses of the prostate gland to androgen and may be a useful and promising toot for investigating the biology and mechanism of stromal-epithelial interaction in vitro.
INTRODUCTION Stromal-epithelial interaction is recognized as being of fundamental importance in hormonal-regulated prostate growth, development, and differentiation. Unfortunately, most previous studies have focused on defining the action of hormones with the assumption that hormones act directly on target epithelial cells without the involvement of neighboring stroma. Recently, howevm, a nmnber of studies have explored the regulatory influence of stroinal growth mediators on prostate epithelium. These have involved organ culture explants (Lasnitzki and Mizuno, 1980), tissue recmnbinants grown as subcutaneous or subrenal capsular grafts (Cunha and Chung, 1981), direct introduction of embryonic tissues (Chung et al., 1984) or stromal growth factors (Marengo and Chung, 1994) in situ as growth inductors, and coinoculating (Chung et al., 1989; Camps et al., 1990) or coculturing (Chang and Chung, 1989) interactive prostatic stromal and epithelial cells in vivo or in vitro. These approaches have led to the delineation of the possible roles of mesenchymal or stromal mediators in normal and tumor epithelial growth and differentiation. These approaches also have provided evidence that reciprocal cellular interaction between stromal and epithelial cells mediated by soluble or matrix-associated molecules could be responsible for regulating prostate growth, differentiation, and cancer metastasis
MATERIALSAND METHODS Conventional cell culture onplastic dishes. LNCaP cells (5 X 104 cells per 60-ram dish) were seeded at Day 0 in 5 ml ofT-medium containing 5% fetal
/To whom con'espondence should be addressed. 375
376
ZHAU ET AL.
bovine serum (FBS) (Chang and Chung, 1989). On Day 1, tissue culture media were replaced by T-medium containing 3% TCM either in the presence or absence of a potent androgen, dihydrotestosterone (100 pM, Sigma Chemical Co., St. Louis, MO) according to a previously described procedure (Chang and Chung, 1989). Total number of cells per dish was counted at Days 3, 5, and 7. In parallel studies, an aliquot of the conditioned media harvested at Day 3, 5, and 7 was subjected to PSA analysis. Because human prostatic fibroblasts, when coeuhured with LNCaP on plastic dishes, tend to overgrow the LNCaP (unpublished observations), cocuhure study examining interactions between prostatic fibroblasts and LNCaP using conventional cell culture procedures was not performed. Microgravity-simulated cell culture in rotary vessels. One hundred and ten milliliter Slow-Turning Lateral Vessel (STLV) models of the Rotating-Wall Vessel (Synthecon, Inc., Friendswood, TX) were prepared and seeded as previously described (Goodwin et al., 1992; Schwarz et al., 1992; Prewett et al., 1993). Briefly, vessels were autoclaved at standard temperature and pressure, cooled, and filled with growth medium [T-medium containing 5% FBS (Chang and Chung 1989)]. Cytodex-3 microcarriers (Pharmacia, Piscataway, NJ) were seeded into the vessels at a density of 5 mg/ml, and cells were seeded at a density of 2 x 10s cells/ml (10 cells/microcarrier). Vessel rotation was initiated at 25-30 qom and increased as required to maintain cells in suspension without failing through the fluid. Cultures were monitored daily for pH, dissolved CO2 and 02, and glucose utilization and were refed accordingly. One set of experiments was initiated with STLV filled with LNCaP alone while another set of experiments was initiated by first filling STLV with normal human prostatic fibroblasts, followed by the LNCaP 3 d later. For time-point comparisons, the inoculation of the LNCaP for both the monocuhm'es and the cocuhures is designated Day 0. Cultures were initiated in T- medium supplemented with 5% FBS and switched to serum-free T- medium supplemented with 3% TCM at Day 4. At Day 4, the media were replaced with Tmedium containing 3% TCM, a serum substitute (Celox Corp., Hopkins, MN), either in the presence or absence of DHT (100 pM). Total cell counts and media PSA were determined every 48 h. Prostate Specific Antigen was assayed by an IMAX kit kindly provided by Abbott Laboratories (Abbott Park, IL). Cell number was determined visually by observation under a hemocytometer. Human prostate fibroblasts were obtained from a retropubic prostectomy specimen (kindly provided by Dr. Patricia Troncoso); the contralateral disease-free region of a pathologically defined prostate cancer region was used for derivation of normal human prostate fibrablasts. Fibroblasts at passage number 25 was chosen for the present study. Histology. Microca~'rier beads harboring prostatic cells were collected and fixed in neutralized formalin as previously described. Five micron sections of the fixed and processed specimens were subjected to routine histologic analysis with sections stained by hematoxylin and eosin (Zhau et al., 1992). Some comparable specimens were stained with PSA antibody as described previously (Gleave et al., 199]). Northern hybridization. Microcarrier beads containing prostate cancer cells were subjected to RNA extraction using the guanidinium-isothiocyanate method (Cbomczynski and Sacchi, 1987). Northern analysis was perfbrmed by loading 20 lag total RNA onto 1% agarose-formaldehyde gels and, after electrophoresis, gels were transferred to a Hybound N membrane (Anmrsham CorP., Arlington Heights, IL) and hybridized with a random primer 32p-labeled PSA cDNA probe (Zhau et al., 1992). RESULTS The growth of LNCaP, under conventional cell culture conditions on plastic dishes, was found to be induced by the presence of 100 pM dihydrotesterone (DHT). Fig. 1 shows that DHT increased the growth of LNCaP in a time-dependent manner. By Day 7, DHT induced the growth of LNCaP by threefold. DHT was also found to stimulate PSA production by the LNCaP cultured on plastic dishes (Fig. 2). Under these culture conditions, DHT was found to stimulate PSA synthesis and secretion between Days 3 and 7. The levels of PSA stimulation reached a plateau at Day 7 at a concentration of PSA at 450 ng/ml. By comparison, vehicle-treated control cultures yielded steady-state concentrations of PSA in the range of 5 to 80 ng/ml.
LNCaP Conventional Cell Culture on Plastic Dishes 25
20
+DHT
j
15 ,,o
l0
Z
Days in Culture FIG. 1. Time-dependent increases of human prostate cancer cells (LNCaP) growth in the presence or absence of dihydrotesterone (DHT, 100 pM) under conventional cell culture on plastic dishes.
L N C a P Conventional Cell Culture on Plastic Dishes 500
400
