expansion and long-term culture of differentiated normal rat urothelial cells in vitro

In Vitro Cell. Dev. Biol.~Animal 37:419M,29, July/August 2001 �9 2001 Society for In Vitro Biology i071-2690/0i $10.00+0.00

EXPANSION AND LONG-TERM CULTURE OF DIFFERENTIATED NORMAL RAT UROTHELIAL CELLS IN VITRO

YUAN YUAN ZHANG, 1 BARBARA LUDWIKOWSKI, ROBERT HURST, AND PETER FREY

Department of Urology, The University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma (g E Z., R. H.) and Pediatric Urology Research Laboratory, Center Hospital Universitaire Vaudois, Lausanne, Switzerland (B. L., P. F.)

(Received 11 September 2000; accepted 31 May 2001)

SUMMARY

The objective of this study is to establish a reliable cell culture system for the long-term culture of rat urothelial cells (RUC), in which the cells multiply in vitro and form stratified polarized urothelium. Urothelial cells were harvested by the enzymatic digestion of the urothelium exposed by the eversion of resected rat bladders. Primary cultures were initiated in keratinocyte serum-free medium (KSFM) for selective proliferation of urothelial cells. Subsequently, the cells were propagated in a mixture of conditioned medium (CM) derived from Swiss 3T3 cell culture supernatant and KSFM (CM- KSFM). Mean population doubling time was 13.8 -4- 0.9 h. RUC were successfully maintained for 18 passages over a period of 4 -5 too. Detailed investigations of culture conditions showed that CM-KSFM yielded a differentiated muhilayer structure. The stratified urothelial sheets measuring 4 • 6 cm 2 could be formed and then detached using dispase. Cytokeratin pattern in both the cultured urothelial monolayer and engineered stratified layers was similar to those seen in vivo, as assessed with monoclonal antibody against cytokeratin 17. Uhrastructural morphology showed microvilli, basal cell layer, and desmosomes between adjacent cells in the stratified urothelium.

Key words: rat; urothelial cells; cell culture; urothelial stratification.

INTRODUCTION

Culture of urothelial cells provides an in vitro model system to help advance our understanding of the cellular mechanisms of urothelial development and epithel ial-stromal interaction in urinary tract reconstruction and regeneration, bladder cancer, and cystitis. The rat is a commonly used animal model; however, the use of cultured rat urothelial cells (RUC) has been limited due to the difficulty of their isolation and maintenance in a long-term culture (Chlapowski, 1989; Noguchi et al., 1990; Leighton, 1992). The main approach currently used for the culture of RUC is to plate the cells in a serum-containing medium on a collagen-coated culture dish or on a feeder layer of lethally irradiated mouse fibroblasts that support growth of cultured ceils in vitro. However, with these techniques, RUC culture is often contaminated with rat bladder stromal cells such as fibroblasts, smooth muscle cells, or capillary endothelium. The stromal cells overgrow and finally replace the urothelial cells in the serum-containing medium after only a few passages.

Based on our previous experience with cultured urothelial cells from humans and pigs (Ludwikowski et al., 1999; Zhang et al., 1999, 2000; Sugasi et al., 2000), we developed a simple technique to isolate rat urothelium by enzymatic release of RUC from an 'everted bladder ' and to maintain the primary culture in keratinocyte serum-free medium (KSFM). In this study, the conditions for the growth of RUC in long-term culture were investigated system- atically. As a result, we show that a mixture of conditioned medium

1 To whom correspondence should be addressed at E-mail: [email protected]

(CM) obtained from mouse 3T3 fibroblasts and KSFM (CM-KSFM) yielded large quantities of normal RUC without stromal cell contamination. RUC could be subcultured up to 18 times in CM-KSFM during an observation period of up to 5 mo. Additionally, rat urothelium stratification can be induced with or without a feeder layer in vitro, which provides potential application as an autograft for urothelial replacement in bladder augmentation in a rat model ( G u a n e t al., 1990), and serves as a tool for research on various bladder diseases.

MATERIALS AND METHODS

Isolation of RUC. Following anesthetization of the rats with an intraperitoneal injection of sodium pentobarbital, 22 male Wistsr rats weighing 540--650 g were sacrificed. After the whole bladder was excised, a modified Roszell's procedure (Roszell et al., 1977) was applied to evert the bladder to expose the urothelial surface. The bladder neck was reinserted into tile lumen, and surgically closed to form an 'everted bladder ball' exposing only the urothelial surface. In order to choose the optimal method for the isolation of RUC, the everted bladder was immersed either in 4 ml of 1% collagenase IV (Worthington Biochemical Co., Lakewood, NJ) at 37 ~ C on a shaker for 60 min, or in 10 nfl of 0.1% disodium ethylene diaminetetraacetic acid (EDTA) (Sigma Chemical Co., St. Louis, MO) at 4 ~ C for 4 h. The bladder mucosa was gently scraped from the muscle tissue following digestion using a forceps with coarse tips. The urothelial cells were collected, washed, and plated on T25 Primuria | cell culture flasks at a cell density of 5 • 10 ~ cells/nil in 5-ml KSFM (GIBCO-BRL/Life Technologies, Basel, Swit- zerland) with 0.09 mM calcium. The culture was incubated in a humidified at- mosphere of 5% CO2 at 37 ~ C. This sermn-fi'ee medium was supplemented with epidermal growth factor (EGF) (5 ng/ml; GIBCO-BRL), bovine pituitary extract (BPE) (50 Ixg/ml; GIBCO-BRL), cholera toxin (CT) (GIBCO-BRL) at a final concentration of 30 ng/ml, penicillin (100 U/ml, GIBCO-BRL), and streptomycin (1 txg/nfl, GIBCO-BRL).

Swiss albino-transformed mouse fibroblasts (3T3 cells) were kindly pro-

4 1 9

420

FIG. 1. Day-3 primary culture of RUC in KSFM showing an island of growing RUC.

ZHANG ET AL.

FIG. 2. Growth cmwes of passage 2 normal RUC in various culture media.

60

50 m

r 4O

m

�9 - - 3 0

o

o 20

Z 10

_- KSFM - -O- - KSFM+DMEM (5%FBS)

T DMEM+ 10%FBS - - v - - CM (10%FBS)

_- KSFM+CM (5%FBS) - - o - - KSFM+5%FBS - -O- - KSFM+10%FBS

i i i i i

6 8 10 12 14

Days

RAT UROTHELIAL CELL CULTURE 421

FIG. 3. RUC grown for 14 d in (A) CM KSFM and (B) KSFM alone. Note that the cells in KSFM alone failed to achieve confluence.

FIG. 3. Continued.

422 ZHANG ET AL.

FIG. 4. Detached stratified rat urothelial sheet (• Urothelial stratification was induced in CM-KSFM at 2.5 mM calcium level within 1 wk.

vided by Dr. M. Benathan (Center Hospital Universitaire Vaudois, Lausanne, Switzerland). These 3T3 cells and rat bladder smooth muscle cells were also cultured in KSFM, and monitored tbr over 3 wk to determine their potential to contaminate RUC culture.

3T3feeding layers and CM. Swiss albino transformed mouse fibroblasts (3T3 cells) were initially" grown in Dulbeceo modified Eagle medium (DMEM) (GIBCO- BRL) supplemented with 10% fetal bovine serum (FBS) (GIBCO-BRL), 1% sodium pyruvate (GIBCO-BRL), and 1% penicillin-streptomycin. For the prepa- ration of the feeder layers, confluent cultures of 3T3 cells containing 3 • I(Y' cells/T 75 flask were lethally irradiated with a dose of 4500 fads for 35 min from a cobalt source. After irradiation, 3T3 cells were trypsinized and replated at 1/10 dilution in DMEM with 3 x lO t eellsff 75 flask. RUC secondary cultures in keratinocyte medium were inoculated on the lethally irradiated 3T3 cells within 2 h. This medium contained 3/4 DMEM, 1/4 Ham~ F12 (GIBCO-BRL), 10% FBS, 0.4 mg/nd hydroeortisone, 10 ,o M CT, 5 mg/ml insulin (GIBCO-BRL), 1.2 mg/ml adenine (GIBCO-BRL), 2.5 mg/ml transfemn (GIBCO-BRL) plus 0.136 mg/ml 3,3',5-ttiiodo-L-thyronine (GIBCO-BRL), 10 mg/ml EGF, and 1% penicillin-s'treptomycin.

Since RUC proliferated rapidly on irradiated 3T3 feeding layers, we further investigated the growth potential of RUC in CM derived from 3T3 cell cultures. The CM was prepared by incubating cultures of 90% confluent 3T3 cells in DMEM for 24 h. The supernatant media were collected and filter- sterilized with a nitrocellulose membrane filter with 0.22 Ixm pore size (Mil- lex-GP, Singapore).

RUC cultures and growth curves. Various media formulations were tested to select the optimal medium for RUC adhesion, proliferation, and stratification. After primary culture, the urothelial cells were cultured in the following seven different media: (1) KSFM, (2) KSFM with 5% FBS, (3) KSFM with 10% FBS, (4) a l : i mixture of KSFM and DMEM (5% FBS, KSFM- DMEM), (5) DMEM with 10% FBS and (6) a 1:1 mixture of CM (prepared originally to contain 10% FBS) and KSFM, and (7) CM (Table 1).

Growth curves were established on freshly isolated urothelial cells from passage 2 seeded in 24-well plates (5 • 104 cells/well) in each of the seven media. On the following d, cells in three wells were detached and pooled, and the cells were counted. The procedure was repeated daily for the first

TABLE 1

SEEDING EFFICIENCY, POPULATION DOUBLING TIME, AND SATURATION DENSITY ON RAT UROTHELIAL CELLS IN THE

VARIOUS CULTURE MEDIA

Original Saturation FBS density

content Seeding (x 104 Culture media (%)" efficiency (%) PDT j' (h) cells/era 2)

1. KSFM None 28.3 - 6.0 27.2 _+ 2.2 8.4 -+ 0.2 2. KSFM 5 34.6 -+ 1.5 27.7 -+ 0.4 13.7 -+ 0.8 3. KSFM 10 26.5 -+ 1.1 32.8 -+ 0.7 15.8 -+ 0.3 4. KSFM-DMEM (1:1) 5 51.7 -+ 9.2 17.1 -+ 1.8 16.3 -+ 0.7 5. DMEM 10 11.1 -+ 1.8 25.1 _+ 1.4 5.7 -+ 0.4 6. KSFM-CM (1:1) 5 57.8 - 8.9 13.8 - 0.9 25.4 -+ 1.0 7. CM 10 35.0 -+ 5.0 14.9 -+ 0.8 14.5 -+ 0.3

For media 6 and 7, this represents the FBS content prior to culture of the 3T3 cells.

h PDT: population doubling time; data are mean _+ SE.

three d and then every second d for a period of 13 d. The results were plotted semilogarithmieally.

Seeding efficiency is expressed as the percentage of RUC attached to the plastic well in relation to the total number of RUC inoculated. Population doubling time is evaluated to quantify the response of the RUC growth in the various media. In addition, saturation density indicates the maxinmm RUC concentration achieved in the plateau phase of the growth curve.

In primary euhures, RUC were allowed to grow in KSFM only for 3 wk to obtain a pure urothelial cell culture and also to inhibit stromal cell contamination. For establishing subeuhures, the cells were detaehed with trypsin- EDTA (0.05-0.02%). Trypsin activity was stopped by adding soybean trypsin inhibitor (GIBCO-BRL). The cells were maintained in CM-KSFM.


FIG. 5. Urothelial sheet cultured in CM-KSFM. (A) • showing the stratified structure and (B) X 160, showing three distinct layers.

FIG. 5. Continued.

424 ZHANG ET AL.

FIG. 6. Transmission electron micrograph of an engineered rat urothelial sheet showing polarization and desmosomes. (A) Microvilli seen on the top of the growing layer (arrow), (B) several examples of desmosomes (arrow), (C) basal cells on the bottom of the layer and desmosome (arrow), x37,500.

FIG. 6. Continued.

RAT UROTHELIAL CELL CULTURE 4 2 5

FIG. 6. Continued.

Induction of urothelial stratification. Urothelial stratification could be induced by culturing urothelial cells either on irradiated 3T3 cells or in CM- KSFM alone for 1 wk. For the detachment of the muhilayered urothelial sheet, 1% enzyme dispase enzyme-solution (Boehringer, Mannheim, Ger- many) was added to the culture and incubated at 37 ~ C for 30 rain in accordance with the techniques previously used in the procedure for skin cell culture (Green et al., 1979). After detachment from the culture flask, the stratified urothelial sheet was washed twice with serum-free medium.

Histology and immunohistochemistry. The detached urothelial sheet was fixed in 10% formalin and subsequently enclosed between two pieces of waxed papeL The stratified urothelial samples were dehydrated through a graded series of ethanol to xylene, embedded in paraffin, and cut vertically for hematoxylin-eosin staining. A small portion of the normal fl'esh bladder wall isolated from the santo animal was used as a control.

Immunohistochemical studies were performed on cultured monolayers at passage 3 on formalin-fixed, paraffin-embedded bladder tissue samples of normal bladder wall and detached engineered urothelial sheets. Indirect immunofluorescence staining was performed using monoclonal antibody against cytokeratin peptide 17 (E3) (Sigma, BioSciences, St. Louis, MO) combined with anti-mouse IgG fluorescein isothiocyanate (FITC) conjugate. Positive staining was scored (0, +, + +, + + + scale) by one investigator (Y. Y. Z). The viability of cultured stratified urothelial sheets following detachment was evaluated with the catcein-ethidiumdimer-rhodamine reaction (live/dead eukoLight viability/Cyto-toxicity kit, AA, Leiden, Netherlands).

Transmission electron microscopy of urothelial sheets. Stratified urothelial sheets and the control normal rat bladder mucosa were fixed with 2% glu- taraldehyde, rinsed with S~rensen solution, postfixed with 2% osmium te- troxide, embedded in agarose, dehydrated in ethanol, and fixed in Epon Ar- aldite resin. Uhrathin (70-nm-thick) sections were cut and stained for 10 min each in saturated uranyl acetate and lead citrate. Specimens were viewed with a transmission electron microscope with a voltage of 80 kV.

RESULTS

Isolation ofRUC. A consistently high yield of urothelial ceils was achieved from the everted bladder treated with 1% collagenase IV

for 1 h. The collagenase dispersed the stromal cells but left the

urothelial cells remaining in small clusters that easily attached to

plastic culture flasks, and subsequently allowed rapid cell growth.

In contrast, 50% fewer ceils survived after treatment with 0.1%

EDTA for 4 h.

Primary culture of RUC. As early as 3 d in prima1~r culture,

urothelial cells in KSFM grew as closely packed cobblestone-like

colonies. The cells showed a morphology characterized by a small

and homogeneous cell type (Fig. 1). When primai3~ culture exceeded

25 d, RUC growth slowed, the ceils spread diffusely with a sub- sequent increase in size, and the cell number gradually diminished.

However, nei ther rat bladder stromal cells nor 3T3 fibroblasts pro-

liferated on the plastic flasks 3 wk after plating in KSFM. Stromal

cells occasionally attached to the culture flask during an earlier

phase but they showed a shrunken cytoplasm, indicating nonvia-

bility in the serum-free medium.

Expansion and differentiation of RUC. The cell proliferation and

differentiation of RUC were greatly dependent on the culture me-

dium used. Table 1 compares the seeding efficiency, population

doubling time, and saturation density of different treatments. CM-

KSFM provided the most effective attachment, highest cell density,

and shortest doubling time. The cells cultured in CM-KSFM showed

a fourfold increase in seeding efficiency and grew in half the pop-

ulation doubling time compared to those in KSFM. The highest

number of cells harvested was observed in CM-KSFM at various

periods of time. Cell number increased exponentially between days

1 and 7, from 2.6 • 104 to 48 • 104/well in CM-KSFM compared

to 1.4 • 104 to 10.1 • 1@/well in KSFM (Fig. 2). Figure 3 com-

pares the cells at saturating density in CM-KSFM with those grown

4 2 6 ZHANG ET AL.

FIG. 7. Immunofluorescence using anticytokeratin 17 with FITC-conjugated antibody. (A) control rat bladder, (B) transverse section of cultured urothelial sheet, (C) individual cells grown on slides taken from above.

FIG. 7. Continued.


FIG. 7. Continued.

in KSFM alone. In fact, only in CM-KSFM did the cells even achieve confluence. In contrast, cells grown in other media only grew as a single layer without evidence of stratification and failed to reach confluence. The objective was to identify a medium pro- ducing a stratified multilayer; the growth conditions not meeting that basic condition are not illustrated.

Figure 2 compares the growth curves with different media. The different conditions fell into three groups regarding cell proliferation on day 13. The first group included only CM-KSFM, and was the only medium in which RUC formed a muhilayer of cells at the peak of the cell growth curve. To date, urothelial cells derived from rat bladder have been successfully maintained up to passage 18 in CM- KSFM over a period of 5 too. The second included KSFM-DMEM, KSFM + 5%FBS, and KSFM + 10% FBS. Cells grew moderately well in this group but fail to achieve confluence or to produce a multilayer. The third group consisted of CM (enriched with 10% FBS), KSFM, and DMEM + 10% FBS. Cells grew poorly in this media and then began to die after a few d. Interestingly, the urothelial cell cultures yielded higher cell numbers on day 13 when KSFM was combined with other media or with serum than in any single medium. RUC grew poorly in any single medium, expressing numerous granules in the cytoplasm and increasing in cell size.

When plated on an irradiated 3T3 cell feeder layer, urothelial cells appeared as colonies among the 3T3 cells. The urothelial cell colonies exnanded on the 3T3 cells and eventually completely over- grew them. It was possible to maintain RUC for up 10 passages in cocuhure with lethally irradiated 3T3 cells.

Stratification of urothelial cultures. Within 1 wk, stratified urothelial cultures could be formed on the irradiated 3T3 cell feeder

layers as well as in CM-KSFM. When detached from the culture flask by enzymatic treatment with dispase, the urothelial sheets con- tracted and the sheet shrunk from 25 to 6 cm 2 within 30 min at 37 ~ C, forming a mucosa-like structure with slightly rounded edges (Fig. 4). The stratified urothelial cells were closely packed and found to be in 2M0 layers as demonstrated by HE staining of a cross-section (Fig. 5). No photograph is provided for cells growing in media other than CM-KSFM because they failed to achieve confluence, and therefore could not be detached as a layer suitable for analysis. As seen from electron microscopy, the superficial layer demonstrated microvilli (Fig. 6A), and the lateral membrane of the apical cells showed desmosomes (Fig. 6B). The basal layers, the latter attaching to the culture flask, resembled that of normal bladder in vivo (Fig. 6C).

The E3 antibody reacted specifically with the stratified layers of native rat urothelium, cultured urothelial cells, and stratified urothelium induced in vitro (Fig. 7). Assessed with the living/dead kit | over 95% of the area of stratified urothelial sheet showed viable cells on both surfaces.

DISCUSSION

A method for the long-term culture of rat urothelium will facili- tate the investigation of the cellular mechanisms in bladder regeneration and bladder diseases such as cystitis and carcinoma in a rat model. However, it is difficult to culture pure RUC and maintain them in long-term culture using conventional methods. When RUC were cultured in standard medium containing serum, the cells showed low plating efficiency, poor growth characteristics, a limited

428 ZHANG ET AL.

potential for cell division, and failed to differentiate in vitro. Serum includes many factors that have a potent mitogenie effect on fibroblasts or smooth muscle cells and tends to inhibit urothelium proliferation by inducing terminal differentiation. The life span of RUC strains is short according to previous studies. Normal primary- RUC usually seneseed after 3-4 passages (Johnson et al., 1985). RUC cultures could be maintained maximally for 14-60 d (Chlapowski et al., 1989). We report here, a method for culturing rat urothelium in CM-KSFM for up to 18 passages over a period of 5 mo and inducing an engineered urothelial sheet.

Three methods to isolate urothelial cells from human or animal bladders have been reported. Mechanical scraping techniques (Ata- la et at., 1993; Baskin et al., 1993; Guhe et al., 1994; Fujiyama et al., 1995) are more suitable for human and large animal bladders. Harvesting of urothelial eells from urine can result in contamination with kidney, prostate, or vaginal ceils (Herz et al., 1985). Enzyme digestion separates the urothelial cells from the bladder muscle layers with collagenase or trypsin in EDTA or pancreatin in trypsin solution (Roszell et al., ]977; Howlett et al., 1986; Hutton et al., 1993; Southgate et al., 1994). This is the most suitable method for small animal bladders. Our study showed that collagenase digestion is simple and effective for separating RUC from the bladder wall since the interface of the submucosal tissue is predominately based on collagen. The cell suspensions harvested by collagenase digestion contained basal, intermediate, and superfieial cells that were morphologically well preserved and had minimal eontamination with stromal cells, stone of these proliferated rapidly.

Normal RUC, in primary culture, are able to grow in KSFM for a period of 3 wk. Stromal cells and fibroblasts in the control flasks could, however, neither proliferate nor survive in this serum-free medium with a low calcium concentration within the same period of time. Clearly, KSFM provides an optimal medium to separate urothelial cells selectively from the other types of cells. However, unlike human urothelial cells (Hutton et al., 1993; Southgate et al., 1994), RUC cannot he maintained in KSFM alone long term. The cells euhured in KSFM failed to achieve confluence, and after 25 d showed gradual senescence. For this reason, we only cultured RUC in KSFM alone tor a maximum of 3 wk and then replated them in CM-KSFM to initiate long-term cultures.

The present study concurs with those of previous investigators (Howlett et al., 1986) and confirmed that irradiated 3T3 ceils ac- tively interact with RUC and support cell growth. When RUC were seeded onto an irradiated 3T3 cell teeder layel, the cells maintained normal patterns of stratified urothelial growth and differentiation in vitro, indicating that the cells require a continuous input of exog- enous stromal signals for the maintenance of normal morphologie and functional characteristics. However, the feeder fibroblasts can be potential cytotoxic agents as well as confounding factors in epithelial cell cultures utilized to study gene transformation in vitro. Additionally, a recent report (Hultman et al., 1996) showed that foreign 3T3 fibroblasts used as a feeder layer could initiate an iln- munogenic reaction, and induce destruction of cultured epidermal autografts. Therefore, use of the conditioned media ti'om 3T3 cells provides the benefit of RUC adhesion and proliferation while avoid- ing the potential hazards of 3T3 cells. However, previous studies showed that the RUC grew poorly in conditioned media derived either from cultured 3T3 cells (Howlett et al., 1986) or rat stromal cells (Noguchi et al., 1990). Our data showed that the 3T3 cell CM alone also failed to promote further differentiation in RUC culture,

but CM-KSFM significantly accelerated the proliferation of urothelial cells and induced formation of a stratified urothelium sheet within 1 wk.

CM-KSFM promotes growth of RUC probably by several mechanisms. Extracellular matrix proteins in CM enhance attachment to the flask. Peptide growth factors secreted by the feeder may also enhance growth. KFSM with EGF, BPE, and CT have been reported to be critical for urothelial growth (Hutton et al., 1993; Southgate et al., 1994). Clearly, both media combined improved RUC proliferation, induction of stratification, and maintenance of markers of eyto-differentiation.

Over the whole period in culture, all urothelial cells remained eytokeratin 17-positive, demonstrating a stable epithelial phenotype with control bladder tissue, serial passages, after induction of stratification. We attempted RUC staining with antibodies against cy- tokeratins 7, 8, 13, and 18. These demonstrated moderate cross- reactivity with the rat urothelium and SMC in the control bladder tissue, despite previous studies that showed these to react selectively with human and pig urothelial cells (Southgate et ah, 1994; Ludwikowski et al., 1999; Zhang et al., 1999).

Urothelial cells lining the lower urinary tract are specialized to produce an impermeable layer that protects the underlying stromal tissue fiom urine. Tight junction, gap junctions, and desmosomes present between the adjacent urothelial cells of rat bladder (Pete1, 1978) modulate the permeability and electrophysiological properties of the urothelium (Negrete et al., 1996). In the present study, in vitro tissue-engineered stratified rat urothelial sheet displayed tight junctions and desmosomes. In addition, the urothelial sheet demonstrated cell polarization with microvilli and basal cells that cor- relate to the normal rat bladder mucosa (Malczak, 1980). Further- more, the engineered urothelial sheet demonstrated elasticity, con- tracting to 30% of the initial size 30 min after detachment. This was in accordance with the findings of a morphological study of epithelial cells from human skin and oral mucosa seeded on 3T3 feeder layers (Tomson et al., 1995). This phenomenon seems to be due to the contraction of actin bundles in the basal cortex and cytokeratin bundles in all cell layers.

In summary, our study describes a simple technique to achieve large quantities of pure urothelial cell populations without fibroblast contamination. A combination of KSFM and CM, the medium conditioned fi'om mouse 3T3 fibroblasts, provides optimal cell growth medium for RUC culture. During long-term culture in CM-KSFM, RUC maintains some important features of epithelial cell differentiation such as eytokeratin expression, stratification with anchoring and communicating junction formation as well as functional elasticity. Histology, immunohistoehmistry, and transmission electron microscopy showed that the cuhured urothelial cells have sinfilar anatomic structures to those tbund in the normal rat bladder. This culture method may provide a useful system to study the physiology of RUC in vitro, and epithelial-stromal communication in bladder regeneration. Furthermore, the pathophysiology of bladder diseases such as bladder carcinoma and cystitis can be evaluated. This culture technique is also thought to be applicable to other primary tissue culture systems where potential contamination and subse- quent overgrowth with fibroblasts remain a problem.

ACKNOWLEDGMENTS

This study was supported by a grant from Swiss National Science Foun- dation, 3200-049740.96.


REFERENCES

Atala, A.; Freeman, M. R.; Vacanti, J. P.; Shepard, J.; Retik, A. B. Implan- tation in vivo and retrieval of artificial structures consisting of rabbit and human urothelium and human bladder muscle. J. Urol. 150:608- 612; 1993.

Baskin, L.; Howard, P. S.; Macarak, E. Effect of physical forces on bladder smooth muscle and nrothelium. J. Urol. 150:601-607; 1993.

Chlapowski, E J. Long term growth and maintenance of stratified rat urothelium in vitro. Cell Tissue Kinet. 22:245-257; 1989.

Fujiyama, C.; Masaki, Z.; Sugihara, H. Reconstruction of the urinary bladder mucosa in three-dimensional collagen gel cultures fibroblast-extracellular matrix interactions on the difterentiation of transitional epithelial cells. J. Urol. 153:2060-2067; 1995.

Green, H.; Kehinde, O.; Thomas, J. Growth of cultured human epidermal cells into multiple epithelia suitable for grafting. Proc. Natl. Acad. Sci. USA 76:5665-5668; 1979.

Guan, Z. C.; Ricard, G.; Chal~st-Boule, L.; Neilson, K.; Kiruluta, G. Aug- mentation cystoplasty in rats: development of an animal model. J. Urol. 144:461-465; 1990.

Guhe, C.; Folhnann, W. Growth and characterization of porcine urinary bladder epithelial cells in vitro. Am. J. Physiol. 266:F298-F308; 1994.

Herz, E; Gazivoda, P.; Papenhausen, P. R.; Katsuyama, J.; Koss, L. G. Normal human urothelial cells in culture. Subculture procedure, flow cyto- metric and chromosomal analyses. Lab. Invest. 53:571-574; 1985.

Howlett, A. R.; Hodges, G. M.; Rowlatt, C. Epithelial-stromal interactions in the adult bladder: urothelial growth, differentiation, and maturation on culture facsimiles of bladder stroma. Dev. Biol. 118:403-415; 1986.

Hultman, C. S.; Brinson, G. M.; Siltharm, S.; deSerres, S.; Cairns, B. A.; Peterson, H. D.; Meyer, A. A. Allogeneic fibroblasts used to grow cultured epidermal autografts persist in vivo and sensitize the graft recipient for accelerated second-set rejection. J. Trauma 41:51-58; 1996.

Hntton, K. A. R.; T~jdosiewicz, L. K.; Thomas, D. E M.; Southgate, J. Uroth- elial tissue culture for bladder reconstruction: an experimental study. J. Urol. 150:721-725; 1993.

Johnson, M. D.; Bryan, G. T.; Reznikoft, C. A. Serial cultivation of normal rat bladder epithelial cells in vitro. J. Urol. 133:1076-1081; 1985.

Leighton, J. Structural biology of epithelial tissue in histophysiologic gradient culture. In Vitro Cell. Dev. Biol. 28A:482492; 1992.

Ludwikowski, B.; Zhang, Y. Y.; Frey, P. Long-term culture of porcine urothelial cells and induction of urothelial stratification. Br. J. Urol. 84: 507-514; 1999.

Malczak, H. T. The urinary bladder epithelium of the rat. Scanning electron- microscopic study. Acta Anat. 108:89-93; 1980.

Negrete, H. O.; Lavelle, J. P.; Berg, J.; Lewis, S. A.; Zeidel, M. L. Perme- ability properties of the intact mammalian bladder epithelium. Am. J. Physiol. 271:F886-F895; 1996.

Noguchi, S.; Yura, Y.; Sherwood, E. R.; Kakinuma, H.; Kashihara, N.; Oyasu, R. Stimulation of stromal cell growth by normal rat urothelial cell- derived epidermal growth factor. Lab. Invest. 62:538-544; 1990.

Peter, S. The junctional connections between the cells of the urinary bladder in the rat. Cell Tissue Res. 187:439-448; 1978.

Roszell, A. J.; Douglas, C. J.; Irving, C. C. Polyamine-stinmlated growth of rat urinary bladder epithelial cells. Cancer Res. 37:239-243; 1977.

Southgate, J.; Hutton, K. A. R.; Thomas, D. E M.; Trejdosiewicz, L. K. Meth- ods in laboratory investigation. Normal human urothelial cells in vitro: proliferation and induction of stratification. Lab. Invest. 71:583-594; 1994.

Sugasi, S.; Lesbros, Y.; Bisson, I.; Zhang, Y. Y.; Kucera, P.; Frey, P. In vitro engineering of human stratified urothelium: analysis of its morphology and function. J. Urol. 164:951-957; 2000.

Tomson, A. M.; Schohna, J.; Blaauw, E. H. The effect of detachment of cultured epithelial sheets on cellular uhrastructure and organisation. Ep- ithelial Cell Biol. 4:43-51; 1995.

Zhang, Y. Y.; Kropp, B.; Moore, P.; Cowan, R.; Furness III, P. M.; Kolligian, M. E.; Frey, P.; Cheng, E. Y. Co-culture of bladder urothelial and smooth muscle cells on small intestinal submucosa (SIS): potential applications for tissue engineering technology. J. Urol. 164:928-935; 2000.

Zhang, Y. Y.; Leuba, A.; Frey, P. Tissue-engineered human urothelium for urinary tract reconstruction. J. Urol. 161(4) (Suppl.):238; 1999.

expansion and long-term culture of differentiated normal rat urothelial cells in vitro

Documents