- Email: [email protected]
Production of epidermal growth factor in human prostatic cells cultured in vitro
ANNALS OF ANATOMY
Production of epidermal growth factor in human prostatic cells cultured in vitro Sergio De Angeli, Sabrina Buoro*, Andrea Fandella**, Giuseppe Anselmo**, Giorgio Pal~*, Roberto Mingrino***, and Pier Paolo Parnigotto*** Cell Culture Laboratory of the Transfusion Center, Hospital of Treviso, Piazzale dell' Ospedale, 1-31100 Treviso, Italy, *Microbiology Institute, University of Padua, Italy, **Division of Urology, Hospital of Treviso, Italy, and ***Department of Pharmaceutical Science, University of Padua, Italy
Summary. The Epidermal Growth Factor (EGF) plays an important role in the regulation of in vitro growth of prostate cells inducing a strong mitogenic effect. Nevertheless in our previous study we observed that the treatment of human hypertrophic prostate cell line U285 with exogenous E G F produces a restricted effect on the cellular growth rate. This phenomenon could be due to the capacity of the cells to produce E G E In this study we aimed to verify this hypothesis by evaluating the presence of mRNA of E G F and E G F receptor (EGF-R) and of their translation products in U285 cells, before and after the treatment with suramin and exogenous E G E Moreover we studied the effects exerted by these substances on the proliferative rate of the cells U285 after different treatment protocols. The presence in the cells of mRNA for E G F and EGF-R and of their translation products was demonstrated by means of reverse transcription polymerase chain reaction (RT-PCR) and immunocytochemical methods respectively. The modification of growth rate induced by these drugs was studied by F R A M E Cytotoxicity Test. The operative modalities adopted to carry out these growth assays tended to 1)focus the effects of suramin in relation to in vitro cellular growth phase; 2) verify the reversibility of its effects; 3)ascertain if it was possible to antagonize the action of suramin by adding exogenous E G E The results obtained from the RTPCR showed the presence, in the control cells and in the treated ones, of mRNA coding for E G F and EGF-R. The immunocytochemical analysis indicated that 20% of the control cells are E G F positive, and 83% are E G F - R positive, confirming the results obtained with RT-PCR. More-
Correspondence to: S. De Angeli
I
Ann Anat (2000) 182:249-258 © Urban & Fischer Verlag
http:llwww.urbanfischer.deljournalslannanat
over, these stainings showed that the treatment with E G F does not significantly modify the percentage of cells marked by the anti-EGF antibody, while treatments with suramin and suramin plus E G F double this percentage. None of the treatments modifies the percentage of EGFR positive cells. The growth assays showed that the exposition to highest doses of suramin in the first 24 h of cultures causes a decrease (p<0.05) of the cellular proliferation during the following 48 h and 72 h and that these effects are irreversible. Moreover, a contemporaneous exposition of the cells to E G F and suramin at seeding strengthens the cytotoxic action of the last drug. To sum up, the demonstration of the presence in the U285 cells of mRNA coding for E G F and EGF-R and of the corresponding proteins, confirms the hypothesis that these cells can produce E G E Moreover, the cytotoxicity experiments allowed a focusing of the role of the endogenous E G F in the regulation of the U285 cells proliferation and confirmed the importance of biological events that take place in U285 cells during the first 24 h of culture.
Key words: Human prostate U285 cell line - Suramin Epidermal growth factor - Autocrine mechanism
Introduction It is generally accepted that the Epidermal Growth Factor (EGF) plays an important role in regulating in-vitro growth of prostate cells (McKeehan 1991; Ilio et al. 1995; Sherwood and Lee 1995; Byrne et al. 1996; Culig et al. 1996). In primary cultures the adding of exogenous 0940-9602/2000/182/3-249 $12.00/0
E G F induces a strong mitogenic effect on the normal prostatic epithelium of rat (McKeehan et al. 1984; M c K e e h a n et al. 1987), dog (Eaton et al. 1988) and man (Peehl and Stamey 1986; Peehl et al. 1989). The action of this growth factor on stabilized neoplastic cell lines is very complex. In fact, Connolly and Rose (1989, 1990) demonstrated the capacity of cells DU145 and L N C a P to produce high quantities of E G F and the presence in these cells of autocrine growth loop for this growth factor. In hypertrophic tissue cultures, however, the modalities of E G F action are less well defined and the mitogenic effects exerted by its exogenous supply are controversial. According to Chopra et al. (1996), the presence of exogenous E G F in the nutritive medium is essential for the in-vitro growth of the hypertrophic epithelial cells. O n the other hand, Janssen et al. (1997), observing 3 0 p r i m a r y cultures of hypertrophic tissue noted that the administration of this growth factor induces a variable proliferative response. In the hypertrophic line BPH-1, Dahiya et al. (1996) detected transcripts only for E G F and not for E G F - R . In our recent study we observed that treatment of the h u m a n hypertrophic prostate cell line U285 with exogenous E G F produces a restrictive effect on the cellular growth rate, shortening the Lag phase, and regulates the subsequent logarithm growth phase (Log phase) (De Angeli et al. 1997). These observations do not exclude the possibility that the U285 cell line is able to produce enough E G F to affect cell growth. In order to verify this hypothesis, we used suramin, a potent anti-mitogenic drug, for its ability to inhibit cell proliferation by interfering with paracrine/autocrine growth loops, as demonstrated by Yayon and Klagsbrun (1990), Kim et al. (1991) and Mukaida et al. (1994). In this study we therefore evaluated: a) the presence of m R N A for E G F and E G F receptor ( E G F - R ) and of their translation products in the U285 cells before and after the treatment with suramin and E G F exogenous; b) the effects exerted by these substances on the proliferative rate of the U285 cells after different treatment protocols.
Identification of m R N A transcripts for EGF and EGF-R by means of RT-PCR
Cells were treated at 24 hr after seeding with 200 gM and 400 gM of suramin, 10 ng/ml of EGF and suramin plus EGF at the concentration of 200 gM and 10 ng/ml and 400 gM and 10 ng/ ml respectively. As control, cellular aliquots were cultured in the culture medium only. Analysis was carried out at 72 hr after seeding. RNA was extracted from 1 x 107 cells with 90 gl of RNAzol B (Cinnabiotex) and 100 ~1 of chloroform. After extraction, RNA was precipitated with isopropanol and washed with ethanol 70%. Extracted RNA was resuspended in 200 gl of water and treated for 30' at 37 °C with 1.5 U DNasi/RNasi free (Boeringher Manheim) in a 10 mM solution of Mg2SO4. After incubation, RNA was precipitated with isopropanol and sodium acetate 3 M and washed with ethanol 70%. Extracted RNA was resuspended in 100 ~tl of diethylpyrocarbonate treated water for eDNA synthesis. Genomic sequences of EGF and EGF-R (EGF-R) were obtained from GeneBank database (Access numbers AF023155 and X00588). We chose the primers using OLIGO Software Analysis. The RT-PCR resulted in the production of either an EGF 168bp (base pairs) fragment (EGFI: 5' actgttgggagaggaatcgtat 3', EGF2: 5' acaattcacagagtttaacagc 3') or EGF-R 269bp RT-PCR product fragment (EGF-RI: 5' acaggacggggaccagacaact 3', EGF-R2: 5' gccaccaccagcagcaagagga 3'). The complete absence of DNA contaminants was confirmed by PCR, amplifying 200 ng of total RNA with human fl-actin primers (Walthr et al. 1994). For cDNA synthesis and PCR amplification, we used a modified protocol by Perkin-Elmer Cetus (GeneAmp RNA PCR kit). Extracted total RNA, 1 pg, was heat denatured at 70 °C for 3' and reverse transcribed at 42 °C for 50' in tubes containing 2.5 gl of 10 x buffer, 4.5 ~tl of MgC1225 mM, 4 ~tl of deoxyribonucleoside triphosphates (dNTPs)10 mM, 1 gl (20 U) of RNase inhibitor, 1 gl (50 U) of Murine Leukemia Virus Reverse Transcriptase (MuLV), 1.5 gl of random primer 50 mM, final volume 25 gl. cDNA product was denatured at 95 °C for 5', then cooled at 5 °C for 5' and used for amplification. Amplification was performed in PCR buffer 1 x, with 2 mM of MgC12, 0.25 gM of EGF1 and EGF2 primers or EGF-R1 and EGF-R2 primers, 2.5 U AmpliTaq DNA Polymerase, total reaction volume 100 ~tl. For both steps the reaction mixture was subjected to 40 rounds of amplification with the following temperatures: denaturation 94°C, annealing 50°C, elongation 72°C, 1 rain for each step. For each round of RT-PCR we have introduced the negative (H20) control. After amplification, 10 gl of the RTPCR product were loaded onto 2% Nu-Sieve 3:1 agarose gel (FMC Bioproducts), electrophoresed in TBE buffer l x (0.045 mMTris borate, 0.001MEDTA, pH8.8), stained with ethydium bromide and analyzed by an image apparatus, Molecular Analyst (Biorad).
Materials and methods Cell line
Identification of EGF and EGF-R in the U285 cells by means of immunocytochemical methods
U285 cell line was isolated through a primary culture of human prostate hypertrophic tissue by means of the primary explant technique already reported (Fresheny 1987; De Angeli et al. 1995; De Angeli et al. 1996). Tissue fragments were cultivated in TV1 medium supplemented with 5% fetal calf serum and 10% horse serum. The secondary culture was obtained using the Leibovitz' method (1986). During further serial passages, cell propagation was achieved through the non-enzymatic K-Passing method or the Pet-Passing method according to Lechner et al. (1980).
The cells underwent the same treatment protocol used for the RT-PCR. Immunocytochemical methods were carried out on cytocentrifugated preparations fixed with a 1:1 methanol acetone solution. EGF identification was made by the ABC immunocytochemical method with the EGF-Immunohistochemistry System kit (Oncogene Science), while the EGF-Rs were highlighted with the immunoperoxidase (IP) technique by Makin et al. (1984) using as primary antibody a monoclonal anti-EGF-R (monoclonal clone F4, Sigma) and as secondary antibody a goatanti mouse IgG peroxidase-conjugated antibody (Sigma). Posi-
250
tive controls were done on tissue section of normal human skin for EGF and on A431cell samples for EGF-R. Negative controls (blanks) were done on U285 cell samples. The latter were carried out to verify the deactivation of endogenous peroxidases and to control the specificity of secondary antibodies. In order to obtain statistically representative data, the surface of the coverslip in each preparation was divided by means of a grid with 36 squares and 4 mm sides. Ten were chosen randomly from among these by a compatible PC IBM program. The number of marked cells was determined on at least 300 cells for each preparation.
Evaluation of the proliferative rate by means of the FRAME Cytotoxicity Test The effects exerted by the suramin and by the exogenous EGF were analyzed by means of the FRAME Cytotoxicity Test (FCT) according to Borenfreund and Puerner (1985) and Knox et al. (1986). Cells were seeded at the concentration of 1 x 104 cells/ml in 24-well plates and treated with 200 gM and 400 gM of suramin according to the following five trials: a) treatment with suramin at seeding and determinations at 24 hr, 48 hr and 72 hr after seeding (Trial A); b) treatment with suramin from seeding till 24 hr and subsequent substitution with a new medium, determinations at 48 hr and 72 hr after seeding (Trial B); c) treatment with suramin from 24 hr till 48 h after seeding and subsequent substitution with a new medium, determinations at 72 hr after seeding (Trial C); d) treatment with suramin at seeding adding at the same time 10 ng/ml of EGF, determinations at 24 hr, 48 hr, 72 hr after seeding (Trial D); e) treatment with suramin at seeding adding 10 ng/ml of EGF after 24 hr, determinations at 48 hr, 72 hr after seeding (Trial E). Optic Densities (O. D.) of neutral red incorporated into lysosomes as well as of kenacid blue bound to cellular proteins were read using Kontron UVICON 930 spectrophotometers respectively at 540 and 577 nm of wave length. All FCTs were repeated five times, testing each dose in duplicate. The linearity of absorbency of neutral red and kenacid blue over a range of i x 10 4 and 2 x 105 cells/ml (1 x 104, 2 x 104, 3 x 104, 4 x 104, 5 × 104 and 1 x 105 cell/ml) was established by determining the linear correlation coefficient (r). In the neutral red assay this coefficient resulted as equaling 0.98 and in the kenacid blue assay it was equaled to 0.94. Considering these results, suramin activity was studied by comparing the mean O. D. values of controls with those of treated cultures without referring back to the corresponding cell concentration values.
Statistical analysis The results of the FRAME Cytotoxieity Test were analyzed by one-way ANOVA and by Tukey posttest. The results obtained from the immunocytochemical staining were analyzed by Z2 test. p < 0.05 was assumed as being significant.
Results Identification o f m R N A transcripts for E G F and E G F - R by means of RT-PCR I n this research we investigated the presence of m R N A for E G F a n d E G F - R , both in the control cells either in
Fig. 1. a) The expression of EGF mRNA transcripts in U285 cell line using RT-PCR of 168bp is shown. Lanes 1 and 2 are the transcripts of cells treated with 400 gM of suramin plus 10 ng/ml of EGF and 200gM of suramin plus 10ng/ml of EGF respectively. Lane 3 is the transcript of cells treated with only 10 ng/ml of EGF. Lanes 4 and 5 are the transcripts of cells treated only with suramin 400 gM and 200 gM respectively. Lane 6 is the transcript of control cells. Lane 7 is the negative control and Lane 8 is the Molecular Weight Marker VIII (l114bp, 900bp, 692bp, 501/ 489bp, 404bp, 320bp, 242bp, 190bp, 147bp, 124bp, 110bp). b) The expression of EGF-R mRNA transcripts in U285 cell line using RT-PCR of 269bp is shown. Lane 1 is the Molecular Weight Marker VIII. Lanes 2 and 3 are the transcripts of cells treated with 400~tM of suramin plus 10ng/ml of EGF and 200 gM of suramin plus 10 ng/ml of EGF respectively. Lane 4 is the transcript of cells treated with only 10 ng/ml of EGE Lanes 5 and 6 are the transcripts of ceils treated only with suramin 400 gM and 200 gM respectively. Lane 7 is the transcript of control cells. Lane 8 represents the negative control.
251
: Fig. 2
4 Fig. 3 252
those treated with EGF, suramin and suramin/EGE The RT-PCR for the identification of these transcripts followed the evaluation of the total amount of R N A extract and of the absence of contamination with genomic D N A by amplifying with PCR and with RT-PCR of the housekeeping/~-actin gene. The negative and positive results of this analysis (no showed data) allowed up to proceed with the study of the cell transcripts. The products of RT-PCR obtained from the processing of the U285 control cells and those exposed to different pharmacological treatments were submitted to migration in agarose gel. The migration of these highlighted two bands of 168bp and 268bp respectively, which correspond to the E G F and E G F - R amplifying m R N A (Fig. l a and b).
Identification of EGF and EGF-receptor in the U285 cells by means of immunocytochemical methods The microscopic observations of samples, tested with the kit tor the identification of EGF, revealed out the presence of marked cells in all the immunocytochemical preparations apart from the treatment (see Fig. 2 and 3 a and 3 b). The marker was localized both on the cellular membrane and in the cytoplasm as a coarse precipitate, the intensity of which varied in relation to the type of treatment and to suramin concentration. The percentages of E G F positive cells expressed in the different treatments are listed in Table 1. In the control preparation and in those treated with EGF, the percentage of marked cells is respectively 20.58% and 22%. The treatment ot the U285 cells with suramin and suramin/EGF increases the percentage of E G F positive cells, ranging between 38.88% and 45.57%. Statistical analysis carried on with the 3(2 test did not show any significant difference (p > 0.05) between the percentage ot marked cells in the E G F treated preparations and in those of control. Moreover, no significant difference (p > 0.05) was observed comparing the cells exposed to suramin (doses 200 gM and 400 gM) with those simultaneously exposed to suramin and E G F (suramin 200 gM + E G F 10 ng/ml and suramin 400 gM + E G F 10 ng/ml). On the contrary, the increase in the number of marked cells in the preparations exposed to suramin
and suramin/EGF, was statistically significant (p < 0.05) when compared to the controls and to the cells treated with E G E In the samples tested with the antibody antiEGF-R, positive cells were marked uniformly and the stain was localized, mostly on the cell membrane (see Figure 3c and 3d) in these preparations. Marked cell percentages in the control and in the different treatments, reported in Table1, ranged from 83.77% to 88.67% and statistical analysis did not highlight any significant variation (p > 0.05).
FRAME cytotoxicity test T R I A L A. In this trial we studied the short-term effects of the suramin on the proliferative rate of the U285 line, by exposing the cells to suramin starting from the seeding up until the neutral red labeling. The results obtained are reported in Figures 4 a, b. In the neutral red assay, the mean O. D. values detected at 24 h from the seeding are 0.2083 + 0.0500 in the control cultures, 0.2146 + 0.0533 in the cultures exposed at 200 ~tM suramin and 0.193+ 0.0633 in those treated with 400 ~tM of suramin. The statistical analysis carried out with one-way A N O V A did not detect any significant difference (p > 0.05) between these values. Also the mean O.D. detected at 48 h (0.3757 + 0.1039) and 72 h (0.6108 + 0.2710) of culture in the cells exposed to 200 gM do not differ (p > 0.05) from those detected in the corresponding controls (48h: 0.3762+ 0.0791; 72 h: 0.6204 + 0.1383). The cultures treated with 400 gM show instead a considerable decrease (p < 0.05) of the mean values of the O.D. at these two end points (48 h: 0.3283 + 0.1242; 72 h: 0.5416 + 0.1820) with respect to the controls. The determination of the kenacid blue confirms the results obtained with the neutral red assay. At 2 4 h of culture, the statistical analysis does not indicate any variation (p > 0.05) of the mean O.D. of the treated cultures (200 ~tM: 0.7922 + 0.1987; 400 gM: 0.7192 + 0.2426) with respect to the control one (0.8104 _+ 0.1495). At 48 h and 72 h from seeding, the mean O.D. values of the controls are respectively 1.3680 + 1882 and 2.3270 + 0.4210, while those of the cells exposed to a concentration of 200 gM are 1.1970+0.2551 and 2.0000+
Fig. 2. Immunocytochemical analysis of the interactions between suramin and EGF performed by the EGF-lmmunohistochemistry System kit (Oncogene Science). a) Negative control (blank) carried out to verify the deactivation of endogenous peroxidases (objective 40 ×). b) Control cells: 20.58% of cells are positive to anti-EGF antibody (objective 40 ×). c) CeIIs exposed to 10 ng/ml of EGF: 22% of cells are positive to anti-EGF antibody (objective 40 x). d) Cells exposed to 400 laM of suramin: 38.88% of cells are positive to anti-EGF antibody (objective 40 x). Fig. 3. Immunocytochemical analysis of the interactions between suramin and EGF (a and b) performed by the EGF-Immunohistochemistry System kit (Oncogene Science) and between suramin and EGF-R (c and d) carried out with the immunoperoxidase technique by Makin et al. using as primary antibody a monoclonal anti-EGF-R (monoclonal done 29.1, Sigma). a) Ceils exposed to 200 ~tM of suramin and 10 ng/ml of EGF: 45.18% of cells are positive to anti-EGF antibody (objective 40 ×). b) Cells exposed to 400 ~tM of suramin and 10 ng/ml of EGF: 45.57% of cells are positive to anti-EGF antibody (objective 40 x). c) Control cells: 83.77% of cells are positive to anti-EGF-R antibody (objective 100 x). d) Cells exposed to 400 gM of suramin: 87.12% of cells are positive to anti-EGF-R antibody (objective 100 x).
253
Table 1. Results •ftheimmun•cyt•chemica•ana•ysis•ftheinteracti•nsbetweensuraminandEGFandbetweensuraminandEGF-R
a b c d e f
Control EGF Sura 200 gM Sura 400 gME Sura 200 gM + EGF 10 ng/ml Sura 400 gM + EGF 10 ng/ml
EGF % positive cells
% negative cells
20.58 22.00 45.19 38.88 45.18 45.57
79.42 78.00 54.81 61.12 54.82 54.43
a) Trial A: neutral red
t .01
2411
4811
72h
culture time
control, ~
200 FM, ~
400 v.M, ~ p<0.05, "[" standard deviation
b) Trial A: kenacid blue
d 6
24h
48h
72h
culture time
control, rn-nn 200 I~M, ~
400 llM, ~ p<0.05, I
standard deviation
Fig. 4. Effects of treatment with suramin at seeding (TRIAL A) on neutral red and kenacid blue assays. Values are expressed as mean O. D. _+standard deviation of five experiments.
p < 0.05
EGF-receptor % positive cells
% negative cells
a/c, b/c a/d, b/d a/e, b/e a/f, b/f
83.77 85.52 87.37 87.12 86.71 88.67
16.23 14.48 12.63 12.88 13.29 11.33
p < 0.05
0.6609, and those of the cells treated with 400 gM are 1.1150 + 0.4267 and 1.8960 + 0.7842. These data show that both the suramin concentrations we tested induce decreases of the mean values of O. D. of the kenacid blue. In any case, the results of the statistical analysis show that the observed decrease is statistically significant (p < 0.05) only for the 400 gM dose. T R I A L B. This trial was performed in order to verify the recovery of cell growth after removal of the drug. The results obtained by exposing the cells to suramin only in the first 24 hr of culture are summarized in Figures 5 a and 5 b. In the neutral red assay the mean O.D. values detected at 48 h and 72 h of seeding are respectively 0.5206+0.0410 and 1.0210+0.0763 in the control cultures, 0.4988+0.0514 and 0.9896 +0.0890 in those exposed to 200 pM, 0.4258 _+ 0.0933 and 0.8438 + 0.1575 in those treated with 400 pM suramin. In the kenacid blue the mean O. D. values detected at 48 h and 72 h of seeding are respectively 1.3990 +_ 0.3022 and 1.7960 + 0.1598 in the control cultures, 1.2690+0.3740 and 1.7880+ 0.1926 in those exposed at 200 gM, 0.9478 + 0.2826 and 1.5720 + 0.2847 in those treated with 400 gM suramin. T R I A L C. In this trial, the results of which are reported in Figure 6 a and 6 b, we further investigated the recovery of the cells growth after exposition to suramin, carrying out the treatment of cultures between 24 h and 48 h from seeding and the determination of the neutral red and of the kenacid blue at 72 h. Both the suramin concentrations induce a slight increase of the mean O. D. either of the neutral red (200gM: 1.2190+0.8545; 400gM: 1.2190+0.0751) and of the kenacid blue (200 ktM: 2.0880 + 0.5082; 400 gM: 2.1320 + 0.4575) with respect to the corresponding controls (neutral red: 1.2020 + 0.1129; kenacid blue: 1.9190 + 0.2853). In any case, the results of the statistical analysis show that these increases are not statistically significant (p > 0.05). T R I A L D. The results of this trial, the operative modalities of which (contemporary exposition of the cells U285 to suramin and E G F since seeding) were intended to ascertain if it was possible to antagonize the action of suramin by adding exogenous EGF, are reported in Figure 7 a and 7b. Against any expectation, both the growth assays showed that the exogenous E G F does not exert any protective action towards the suramin, and
254
makes the U285 line more sensitive to this drug. In fact, neutral red determinations showed that at 24 hr and 48 hr after seeding, the m e a n O. D. values of the cultures treated with 400 gM of suramin (respectively of 0.2528 + 0.0210 and 0.6079+0.0428) were significantly lower (p < 0.05) than those of the control (24 h: 0.2889 + 0.0146; 48h: 0.7935_+0.0253). A t 7 2 h r of incubation, this p h e n o m e n o n appeared with both doses tested (control: 1.3350 + 0.0590; 200 gM: 1.1240 _+0.0549; 400 gM: 0.9580 _+0.0414). Kenacid blue determinations confirmed neutral red data. The m e a n O. D. values, detected at 24 h, 48 h, 72 h after seeding, in the control culture are respectively 0.7878 _+0.0301, 1.5770 _+0.0879 and 2.2020 _+ 0.0627. In culture treated with 200 gM the m e a n O. D. values are respectively 0.6925 _+ 0.0540, 1.4530 _+ 0.1416 and 2.0940 +_ 0.0945. 0.5949 _+ 0.0947. In culture treated with 400 gM of suramin the m e a n O. D. values are respectively
0.8958 + 0.0922 e 1.2910 _+ 0.0837. The statistical comparisons performed between these data highlight the same significances as evidenced in the neutral red assay. T R I A L E. This trial has also been performed in order to understand better the capacities of the exogenous E G F to antagonize the effect of suramin. The results of the treatment with suramin at seeding, adding 10 ng/ml of E G F 24 hr later, are summarized in Figure 8 a and 8b. For both the markers, statistical analysis revealed a relevant decrease (p < 0.05) of the mean O. D. values only in the cultures treated with 400 gM doses at 48 hr (neutral red: 0.2851 + 0.0162; kenacid blue: 0.9100 + 0.0333) and 72 hr (neutral red: 0.5847 + 0.0348; kenacid blue: 1.4550 + 0.0825) with respect to the corresponding controls (neutral red: 0.4344 _+ 0.0326, 0.7841 + 0.0570; kenacid blue: 1.4350 + 0.0585, 2.3380 _+ 0.1705).
a) Trial C: neutral red 1.50-
a) Trial B: neutral red
1.25.
n.~
n.s.
1.251.00. I .O00.75.
d
6 0.50'
0.75-
0.50-
I
i I Ii
0.25-
0.25-~
::::=: .............
i
i
i 0.00
0.00 24h
48h
24h
72h
~ i J I
48h
72h
culture time
colture time control, rTTT~ 200 p.M, ~
i
400 p-M, ~r p<0.05, "[" standard deviation
control, t lull i 200 I~M, ~
b) Trial B: kenacid blue
400 p.M, n.¢ p>0.05, ]~
standard deviation
b) Trial C: kenacid blue
2.0n.s.
n,s.
1.52
d
6
o
t.0-
0.5-
0.0.
24
control, itplrl 200 ~M, ~
48h colture time 400 ~M, *
24h
72h
48h
72h
culture time p<0.05, T standard deviation
control, rTT'nn 200 FM, ~
Fig. 5. Effects of treatment with suramin from seeding up to 24hr and subsequent substitution with a new medium (TRIAL B), on neutral red and kenacid blue assays. Values are expressed as mean O.D. + standard deviation of five experiments. 255
400 p.M, n,s. p>O.05, "]" standard deviation
Fig. 6. Effects of treatment with suramin from 24hr up to 48h after seeding and subsequent substitution with a new medium (TRIAL C), on neutral red and kenacid blue assays. Values are expressed as mean O.D. + standard deviation of five experiments.
Discussion
mechanism. This line could be made up of two functionally distinct cell populations: the former, numerically restricted (20%), producing EGF, and the latter, prevalent (80%), characterized by the expression of EGF-R. In this case the in-vitro growth ot the U285 cells would depend upon a paracrine regulation mechanism based on one cellular population that produces E G F and of another that employs it. As an alternative to the hypothesis of a paracrine growth control, we could presume the existence of an autocrine regulation system as described for the prostatic cell lines LNCaP (Connolly and Rose 1990) and D U 145 (Connolly and Rose 1989; Connolly and Rose 1994). In this case the discrepancy between the percentages of E G F and E G F - R cells positive could be justified by the difference in velocity between the production and secretion processes and that of E G F clearance. It is well known that by linking to its receptor, the E G F triggers a complex series of events such as the clustering on the
In this study we demonstrated the expression of codifying m R N A for E G F and E G F receptor in U285 cells by means of RT-PCR techniques. Immunocytochemical analysis, identifying their translation products, confirmed these data: in the assays tested with the kit for the EGF, the percentage of marked cells is 20%, while in those stained with an anti-EGF-R antibody this percentage is 80%. These results, therefore, explain the apparent insensitivity of our line to the action of this growth factor during the logarithmic growth phase, as reported previously (De Angeli et al. 1997). However, the immunocytochemical data reported in this study were determined by means of single labeling method, and were therefore insufficient to decide wether the E G F regulates the growth of U285 cells through a paracrine or autocrine
a) Trial E: neutral red
a) Trial D: neutral red 1.00" 1.50-
1.25-
iiiiii:!i
*
0.75-
1.00-
0
d d o.so-
0.75-
iiliii!
0.50-
0.00
48h
24h
:[ii =i ~::.::-:
ii!iiiiii IZ:2 !!!!!!!!~ 0.25-
!!!!!!!!!
iii~iiiii 2121 O.OC 72h
72h
48h
24h
culture time
culture time
control, ~
200 ~tM, ~
400 p.M, '/¢ p
standard deviation
control, rcr/5~ 200 I~M, ~\\\~ 400 FM, 9¢ p<0.05, 1" standard deviation
b) Trial E: kenacid blue
b) Trial D: kenacid blue 2.5-
iii!iiii!i!i
2.0-
=:
iiiiiiiiili! :!!=i!:::i iiiiiiiiiii ::=:::x ......:::::
1.5-
0 1.0"
0.0
24h
li 48h
•::.=:::: :: :::= =::..x x::::=::: :: ::x: ~iii~i!i ii;i;i!;~i
~iilZliiiiii lii!iiii!iii ..........
iiiii
0
72h
i 24h
culture time control, l l l t l i 200 pM, ~
48h culture time
...........I control, ftlrl ] 200 FM, ~
400 pM, 9¢ p<0.05, "[" standard deviation
Fig. 7. Effects of treatment with suramin at seeding adding, at the same time, 10 ng/ml of EGF (TRIAL D), on neutral red and kenacid blue assays. Values are expressed as mean O. D. + standard deviation of five experiments.
400 I~M, ~
iii 72h
p
Fig. 8. Effects of treatment with suramin at seeding adding after 24 hr 10 ng/ml of EGF (TRIAL E), on neutral red and kenacid blue assays. Values are expressed as mean O.D. -+ standard deviation of five experiments.
256
membrane cell surface of the EGF-receptor complexes and their internalization by endocytosis (Carpenter 1985; Carpenter and Cohen 1990). The recycling on the cell membrane of the complex EGF-receptors or the degradation of the E G F itself at a lysosomal level follows this, by a long metabolic pathway (Sorking et al. 1997). In cultures treated with suramin, the increase of the number of E G F positive cells, in the absence of a significant percentage variation in the number of cells with EGF-R, can be easily explained by the action mechanism of this drug (Larsen 1993; Stein 1993) and is compatible with both hypotheses we formulated. For this reason in a future study it would be interesting to verify the contemporary expression of E G F in the U285 line cells, and of its receptor by means of double labeling techniques. Cytotoxicity experiments allowed us to focus the role of the endogenous E G F in the regulation of the U285 cells proliferation. Trials A, B, e C showed, in agreement with the results of a previous research (De Angeli et al. 1994), that in the first 24 hours of culture the suramin interacts with an inductive mechanism, the inhibition of which causes a decrease in cell proliferation during the following 48 h and 72 h. Even the enhancement of the cytostatic action of the suramin, induced by the exogenous E G F when both these substances are given together at seeding (Trials D and E), confirms the importance of biological events that take place in U285 cells during the first 24 hours of culture. We have previously demonstrated that in this period of time, corresponding to the Lag phase, similar to that described by Connolly and Rose (1989, 1990) in the prostatic lines DU145 and LNCaR our cells are responsive to the stimulation of exogenous EGF, probably due to their low concentration of seeding. Taken overall, these data indicate that the endogenous E G F is responsible for the reactivation of the U285 cells and for their passage to the following logarithmic growth phase. Many important aspects of this phenomenon, such as the pathway through which cell reactivation take place, the presence of an inductive phenomenon of the receptor, the role played by the complex EGF-receptor in the down regulation of proliferation, remain to be explained and will be the objectives of further research.
Acknowledgements. This work was partially supported by grant 434/01/94 from the Regione Veneto, Giunta Regionale - Health Research - Venice - Italy.
References Borrenfreund E, Puerner JA (1985) Toxicity determined in vitro by morphological alterations and neutral red absorption. Toxicology Lett 24:11%124 Byrne RL, Leung H, Neal DE (1996) Peptide growth factors in the prostate as mediators of stromal epithelial interaction. Br J Urol 77:627~533 Carpenter G (1985) Epidermal growth factor: biology and receptor metabolism. J Cell Sci 3:1-9
Carpenter G, Cohen S (1990) Epidermal growth factor. J Biol Chem 265:7709-7712 Chopra DR Grignon D J, Joiakin A, Mathieu PA, Mohamed A, Sakr WA, Powell IJ, Sarkar FH (1996) Differential growth factor responses of epithelial cultures derived from normal human prostate, benign prostatic hyperplasia, and primary prostate carcinoma. J Cell Physiol 169:269-280 ConnollyJM, Rose DP (1989) Secretion of epidermal growth factor and related polypeptides by the DU145 human prostate cancer cell line. Prostate 15:177-186 Connolly JM, Rose DP (1990) Production of epidermal growth factor and transforming growth factor-a by androgen-responsive LNCaPprostate cancer cell line. Prostate 16:209-218 ConnoUyJM, Rose DP (1994) Regulation of DU145 human prostate cancer cell proliferation by insulin-like growth factors and its interaction with the epidermal growth factor autocrine loop. Prostate 24:167-175 Culig Z, Hobisch A, Cronauer MV, Radmayr C, Hittmair A, Zhang J, Thurnher M, Bartsch G, Klocker H (1996) Regulation of prostatic growth and function by peptide growth factors. Prostate 28:392-405 Dahiya R, Lee C, Haughney PC, Chui R, Ho R, Deng G (1996) Differential gene expression of transforming growth factors alpha and beta, epidermal growth factor, keratinocyte growth factor, and their receptors in fetal and adult human prostatic tissues and cancer cell lines. Urology 48:963-970 De Angeli S, Fandella A, Conconi MT, Anselmo G, Parnigotto PP (1994) Growth morphology and morphometry of human hypertrophic prostate cells treated with suramin in vitro. Prostate 25:117-124 De Angeli S, Valenti F, Durante E, Bassani V, Parnigotto PP (1995) Primary cultures of human hypertrophic prostate tissue in WAJC 404 medium: a study of cell morphology and kinetics. Ann Anat 177:185-192 De Angeli S, Valenti F, Buoro S, Conconi MT, Parnigotto PP (1996) Morphological features and kinetics of primary cultures of BPH tissue supported by TV1 medium. Ann Anat 178: 4148 De Angeli S, Favretti C, Buoro S, Fandella A, Anselmo G, Parnigotto PP (1997) Effect of DHT and EGF on human hyperplastic prostate stern cells cultured in vitro growth morphology and phenotype characterization. Ann Anat 179:255-264 Eaton CL, Davies P, Phillips MEA (1988) Growth factor involvement and oncogene expression in prostatic tumours. J Steroid Biochem 30:341-345 Fresheny RI (1987) Culture of animal cells: a manual of basic technique. Alan R Liss Inc. NY, pp 107-126 Ilio KY, Sensibar JA, Lee C (1995) Effect of TGF-beta 1, TGFalpha, and EGF on cell proliferation and cell death in rat ventral prostatic epithelial cells in culture. J Androl 16:482-490 Janssen T, Petein M, van Velthoven R, De Decker R, Assenmacher C, Corbusier A, Pasteeels JL, Kiss R, Schnlman C (1997) Coregulatory effects of epidermal growth factor, dehydrotestosterone, and prolactin on human prostatic hyperplasia tissue culture proliferation. Prostate 30:47-52 Kim JH, Sherwood ER, Sutkowski DM, Lee C, Kozlowski JM (1991) Inhibition of prostatic tumor cell proliferation by surarain: alterations in TGF alpha-mediated autocrine growth regulation and cell cycle distribution. J Urol 146:171-176 Knox P, Uphill PF, Fry JR, BenfordJ, BallssM (1986) The FRAME multicenter project on in vitro cytotoxicology. Fd Chem Toxic 24:457-463 Larsen AK (1993) Suramin. An anticancer drug with unique biological effects. Cancer Chem Pharm 32:96-98
257
Lechner JF, Babcock MS, Marnell M, Narayan S, Kaighn ME (1980) Normal human prostate epithelial cell cultures. Meth Cells Biol 21B: 191-224 Leibovitz A (1986) Development tumor cells lines. Cancer Genet Cytogenet 19:11-19 Makin CA, Bobrow LG, Bodmer WF (1984) Monoelonal anditody to cytokeratin for use in routine histopathology. J Clin Pathol 37:975-983 McKeehan WL, Adams PS, Rosser PM (1984) Direct mitogenic effects of insulin, EGF, glucoeorticoid, cholera toxin, unknown pituitary factors and possibly prolactin but not androgen on normal rat prostate epithelial cells in serum-free, primary cell culture. Cancer Res 44:1998-2010 McKeehan WL, Adams PS, Fast D (1987) Different hormonal requirement for androgen-independent growth of normal and turnout epithelial cells from rat prostate. In Vitro Cell Dev Biol 23:147-152 McKeehan WL (1991) Growth factor receptors and prostate cell growth. Canc Surv 11:165-175 Mukaida H, Hirabayashi N, Hirai T, Yamashita Y, Iwata T, Saeki S, Toge T (1994) The inhibitory effect caused by suramin on paracrine growth of human cancer cells and fibroblasts. Surg Today 24:234-240
Peehl D, Stamey TA (1986) Serum free growth of adult human prostatic epithelial cells. In Vitro Cell Dev Biol 22:82-90 Peehl D, Wong ST, Bazinet M, Stamey TA (1989) In vitro studies of human prostatic cells: attempts to identify distinguishing features of malignant cells. Growth Factors 1:237-250 Sherwood ER, Lee C (1995) Epidermal growth factor-related peptides and the epidermal growth factor receptor in normal and malignant prostate. World J Urol 13:290-296 Stein CA (1993) Suramin: a novel antineoplastic agent with multiple potential mechanism of action. Cancer Res 53:2239-2248 Sorkin A, Kornilova E, Teslenko L, Sorokin A, Nikolsky N (1997) Recycling of epidermal growth factor-receptor complex in A431 cells. Mol Cell Endocrinol 126:185-192 Walthr W, Stein U, Eder C (1994) RNA analysis using miniprep RNA in reverse transcription PCR. Biotechniques 17:674-675 Yayon A, Klagsbrun M (1990) Autocrine trasformation by chimeric signal peptide-basic fibroblast growth factor: reversal by suramin. Proc Nat Acad Sci USA 87:5346-5350
Accepted October 5, 1999
258
Production of epidermal growth factor in human prostatic cells cultured in vitro Sergio De Angeli, Sabrina Buoro*, Andrea Fandella**, Giuseppe Anselmo**, Giorgio Pal~*, Roberto Mingrino***, and Pier Paolo Parnigotto*** Cell Culture Laboratory of the Transfusion Center, Hospital of Treviso, Piazzale dell' Ospedale, 1-31100 Treviso, Italy, *Microbiology Institute, University of Padua, Italy, **Division of Urology, Hospital of Treviso, Italy, and ***Department of Pharmaceutical Science, University of Padua, Italy
Summary. The Epidermal Growth Factor (EGF) plays an important role in the regulation of in vitro growth of prostate cells inducing a strong mitogenic effect. Nevertheless in our previous study we observed that the treatment of human hypertrophic prostate cell line U285 with exogenous E G F produces a restricted effect on the cellular growth rate. This phenomenon could be due to the capacity of the cells to produce E G E In this study we aimed to verify this hypothesis by evaluating the presence of mRNA of E G F and E G F receptor (EGF-R) and of their translation products in U285 cells, before and after the treatment with suramin and exogenous E G E Moreover we studied the effects exerted by these substances on the proliferative rate of the cells U285 after different treatment protocols. The presence in the cells of mRNA for E G F and EGF-R and of their translation products was demonstrated by means of reverse transcription polymerase chain reaction (RT-PCR) and immunocytochemical methods respectively. The modification of growth rate induced by these drugs was studied by F R A M E Cytotoxicity Test. The operative modalities adopted to carry out these growth assays tended to 1)focus the effects of suramin in relation to in vitro cellular growth phase; 2) verify the reversibility of its effects; 3)ascertain if it was possible to antagonize the action of suramin by adding exogenous E G E The results obtained from the RTPCR showed the presence, in the control cells and in the treated ones, of mRNA coding for E G F and EGF-R. The immunocytochemical analysis indicated that 20% of the control cells are E G F positive, and 83% are E G F - R positive, confirming the results obtained with RT-PCR. More-
Correspondence to: S. De Angeli
I
Ann Anat (2000) 182:249-258 © Urban & Fischer Verlag
http:llwww.urbanfischer.deljournalslannanat
over, these stainings showed that the treatment with E G F does not significantly modify the percentage of cells marked by the anti-EGF antibody, while treatments with suramin and suramin plus E G F double this percentage. None of the treatments modifies the percentage of EGFR positive cells. The growth assays showed that the exposition to highest doses of suramin in the first 24 h of cultures causes a decrease (p<0.05) of the cellular proliferation during the following 48 h and 72 h and that these effects are irreversible. Moreover, a contemporaneous exposition of the cells to E G F and suramin at seeding strengthens the cytotoxic action of the last drug. To sum up, the demonstration of the presence in the U285 cells of mRNA coding for E G F and EGF-R and of the corresponding proteins, confirms the hypothesis that these cells can produce E G E Moreover, the cytotoxicity experiments allowed a focusing of the role of the endogenous E G F in the regulation of the U285 cells proliferation and confirmed the importance of biological events that take place in U285 cells during the first 24 h of culture.
Key words: Human prostate U285 cell line - Suramin Epidermal growth factor - Autocrine mechanism
Introduction It is generally accepted that the Epidermal Growth Factor (EGF) plays an important role in regulating in-vitro growth of prostate cells (McKeehan 1991; Ilio et al. 1995; Sherwood and Lee 1995; Byrne et al. 1996; Culig et al. 1996). In primary cultures the adding of exogenous 0940-9602/2000/182/3-249 $12.00/0
E G F induces a strong mitogenic effect on the normal prostatic epithelium of rat (McKeehan et al. 1984; M c K e e h a n et al. 1987), dog (Eaton et al. 1988) and man (Peehl and Stamey 1986; Peehl et al. 1989). The action of this growth factor on stabilized neoplastic cell lines is very complex. In fact, Connolly and Rose (1989, 1990) demonstrated the capacity of cells DU145 and L N C a P to produce high quantities of E G F and the presence in these cells of autocrine growth loop for this growth factor. In hypertrophic tissue cultures, however, the modalities of E G F action are less well defined and the mitogenic effects exerted by its exogenous supply are controversial. According to Chopra et al. (1996), the presence of exogenous E G F in the nutritive medium is essential for the in-vitro growth of the hypertrophic epithelial cells. O n the other hand, Janssen et al. (1997), observing 3 0 p r i m a r y cultures of hypertrophic tissue noted that the administration of this growth factor induces a variable proliferative response. In the hypertrophic line BPH-1, Dahiya et al. (1996) detected transcripts only for E G F and not for E G F - R . In our recent study we observed that treatment of the h u m a n hypertrophic prostate cell line U285 with exogenous E G F produces a restrictive effect on the cellular growth rate, shortening the Lag phase, and regulates the subsequent logarithm growth phase (Log phase) (De Angeli et al. 1997). These observations do not exclude the possibility that the U285 cell line is able to produce enough E G F to affect cell growth. In order to verify this hypothesis, we used suramin, a potent anti-mitogenic drug, for its ability to inhibit cell proliferation by interfering with paracrine/autocrine growth loops, as demonstrated by Yayon and Klagsbrun (1990), Kim et al. (1991) and Mukaida et al. (1994). In this study we therefore evaluated: a) the presence of m R N A for E G F and E G F receptor ( E G F - R ) and of their translation products in the U285 cells before and after the treatment with suramin and E G F exogenous; b) the effects exerted by these substances on the proliferative rate of the U285 cells after different treatment protocols.
Identification of m R N A transcripts for EGF and EGF-R by means of RT-PCR
Cells were treated at 24 hr after seeding with 200 gM and 400 gM of suramin, 10 ng/ml of EGF and suramin plus EGF at the concentration of 200 gM and 10 ng/ml and 400 gM and 10 ng/ ml respectively. As control, cellular aliquots were cultured in the culture medium only. Analysis was carried out at 72 hr after seeding. RNA was extracted from 1 x 107 cells with 90 gl of RNAzol B (Cinnabiotex) and 100 ~1 of chloroform. After extraction, RNA was precipitated with isopropanol and washed with ethanol 70%. Extracted RNA was resuspended in 200 gl of water and treated for 30' at 37 °C with 1.5 U DNasi/RNasi free (Boeringher Manheim) in a 10 mM solution of Mg2SO4. After incubation, RNA was precipitated with isopropanol and sodium acetate 3 M and washed with ethanol 70%. Extracted RNA was resuspended in 100 ~tl of diethylpyrocarbonate treated water for eDNA synthesis. Genomic sequences of EGF and EGF-R (EGF-R) were obtained from GeneBank database (Access numbers AF023155 and X00588). We chose the primers using OLIGO Software Analysis. The RT-PCR resulted in the production of either an EGF 168bp (base pairs) fragment (EGFI: 5' actgttgggagaggaatcgtat 3', EGF2: 5' acaattcacagagtttaacagc 3') or EGF-R 269bp RT-PCR product fragment (EGF-RI: 5' acaggacggggaccagacaact 3', EGF-R2: 5' gccaccaccagcagcaagagga 3'). The complete absence of DNA contaminants was confirmed by PCR, amplifying 200 ng of total RNA with human fl-actin primers (Walthr et al. 1994). For cDNA synthesis and PCR amplification, we used a modified protocol by Perkin-Elmer Cetus (GeneAmp RNA PCR kit). Extracted total RNA, 1 pg, was heat denatured at 70 °C for 3' and reverse transcribed at 42 °C for 50' in tubes containing 2.5 gl of 10 x buffer, 4.5 ~tl of MgC1225 mM, 4 ~tl of deoxyribonucleoside triphosphates (dNTPs)10 mM, 1 gl (20 U) of RNase inhibitor, 1 gl (50 U) of Murine Leukemia Virus Reverse Transcriptase (MuLV), 1.5 gl of random primer 50 mM, final volume 25 gl. cDNA product was denatured at 95 °C for 5', then cooled at 5 °C for 5' and used for amplification. Amplification was performed in PCR buffer 1 x, with 2 mM of MgC12, 0.25 gM of EGF1 and EGF2 primers or EGF-R1 and EGF-R2 primers, 2.5 U AmpliTaq DNA Polymerase, total reaction volume 100 ~tl. For both steps the reaction mixture was subjected to 40 rounds of amplification with the following temperatures: denaturation 94°C, annealing 50°C, elongation 72°C, 1 rain for each step. For each round of RT-PCR we have introduced the negative (H20) control. After amplification, 10 gl of the RTPCR product were loaded onto 2% Nu-Sieve 3:1 agarose gel (FMC Bioproducts), electrophoresed in TBE buffer l x (0.045 mMTris borate, 0.001MEDTA, pH8.8), stained with ethydium bromide and analyzed by an image apparatus, Molecular Analyst (Biorad).
Materials and methods Cell line
Identification of EGF and EGF-R in the U285 cells by means of immunocytochemical methods
U285 cell line was isolated through a primary culture of human prostate hypertrophic tissue by means of the primary explant technique already reported (Fresheny 1987; De Angeli et al. 1995; De Angeli et al. 1996). Tissue fragments were cultivated in TV1 medium supplemented with 5% fetal calf serum and 10% horse serum. The secondary culture was obtained using the Leibovitz' method (1986). During further serial passages, cell propagation was achieved through the non-enzymatic K-Passing method or the Pet-Passing method according to Lechner et al. (1980).
The cells underwent the same treatment protocol used for the RT-PCR. Immunocytochemical methods were carried out on cytocentrifugated preparations fixed with a 1:1 methanol acetone solution. EGF identification was made by the ABC immunocytochemical method with the EGF-Immunohistochemistry System kit (Oncogene Science), while the EGF-Rs were highlighted with the immunoperoxidase (IP) technique by Makin et al. (1984) using as primary antibody a monoclonal anti-EGF-R (monoclonal clone F4, Sigma) and as secondary antibody a goatanti mouse IgG peroxidase-conjugated antibody (Sigma). Posi-
250
tive controls were done on tissue section of normal human skin for EGF and on A431cell samples for EGF-R. Negative controls (blanks) were done on U285 cell samples. The latter were carried out to verify the deactivation of endogenous peroxidases and to control the specificity of secondary antibodies. In order to obtain statistically representative data, the surface of the coverslip in each preparation was divided by means of a grid with 36 squares and 4 mm sides. Ten were chosen randomly from among these by a compatible PC IBM program. The number of marked cells was determined on at least 300 cells for each preparation.
Evaluation of the proliferative rate by means of the FRAME Cytotoxicity Test The effects exerted by the suramin and by the exogenous EGF were analyzed by means of the FRAME Cytotoxicity Test (FCT) according to Borenfreund and Puerner (1985) and Knox et al. (1986). Cells were seeded at the concentration of 1 x 104 cells/ml in 24-well plates and treated with 200 gM and 400 gM of suramin according to the following five trials: a) treatment with suramin at seeding and determinations at 24 hr, 48 hr and 72 hr after seeding (Trial A); b) treatment with suramin from seeding till 24 hr and subsequent substitution with a new medium, determinations at 48 hr and 72 hr after seeding (Trial B); c) treatment with suramin from 24 hr till 48 h after seeding and subsequent substitution with a new medium, determinations at 72 hr after seeding (Trial C); d) treatment with suramin at seeding adding at the same time 10 ng/ml of EGF, determinations at 24 hr, 48 hr, 72 hr after seeding (Trial D); e) treatment with suramin at seeding adding 10 ng/ml of EGF after 24 hr, determinations at 48 hr, 72 hr after seeding (Trial E). Optic Densities (O. D.) of neutral red incorporated into lysosomes as well as of kenacid blue bound to cellular proteins were read using Kontron UVICON 930 spectrophotometers respectively at 540 and 577 nm of wave length. All FCTs were repeated five times, testing each dose in duplicate. The linearity of absorbency of neutral red and kenacid blue over a range of i x 10 4 and 2 x 105 cells/ml (1 x 104, 2 x 104, 3 x 104, 4 x 104, 5 × 104 and 1 x 105 cell/ml) was established by determining the linear correlation coefficient (r). In the neutral red assay this coefficient resulted as equaling 0.98 and in the kenacid blue assay it was equaled to 0.94. Considering these results, suramin activity was studied by comparing the mean O. D. values of controls with those of treated cultures without referring back to the corresponding cell concentration values.
Statistical analysis The results of the FRAME Cytotoxieity Test were analyzed by one-way ANOVA and by Tukey posttest. The results obtained from the immunocytochemical staining were analyzed by Z2 test. p < 0.05 was assumed as being significant.
Results Identification o f m R N A transcripts for E G F and E G F - R by means of RT-PCR I n this research we investigated the presence of m R N A for E G F a n d E G F - R , both in the control cells either in
Fig. 1. a) The expression of EGF mRNA transcripts in U285 cell line using RT-PCR of 168bp is shown. Lanes 1 and 2 are the transcripts of cells treated with 400 gM of suramin plus 10 ng/ml of EGF and 200gM of suramin plus 10ng/ml of EGF respectively. Lane 3 is the transcript of cells treated with only 10 ng/ml of EGF. Lanes 4 and 5 are the transcripts of cells treated only with suramin 400 gM and 200 gM respectively. Lane 6 is the transcript of control cells. Lane 7 is the negative control and Lane 8 is the Molecular Weight Marker VIII (l114bp, 900bp, 692bp, 501/ 489bp, 404bp, 320bp, 242bp, 190bp, 147bp, 124bp, 110bp). b) The expression of EGF-R mRNA transcripts in U285 cell line using RT-PCR of 269bp is shown. Lane 1 is the Molecular Weight Marker VIII. Lanes 2 and 3 are the transcripts of cells treated with 400~tM of suramin plus 10ng/ml of EGF and 200 gM of suramin plus 10 ng/ml of EGF respectively. Lane 4 is the transcript of cells treated with only 10 ng/ml of EGE Lanes 5 and 6 are the transcripts of ceils treated only with suramin 400 gM and 200 gM respectively. Lane 7 is the transcript of control cells. Lane 8 represents the negative control.
251
: Fig. 2
4 Fig. 3 252
those treated with EGF, suramin and suramin/EGE The RT-PCR for the identification of these transcripts followed the evaluation of the total amount of R N A extract and of the absence of contamination with genomic D N A by amplifying with PCR and with RT-PCR of the housekeeping/~-actin gene. The negative and positive results of this analysis (no showed data) allowed up to proceed with the study of the cell transcripts. The products of RT-PCR obtained from the processing of the U285 control cells and those exposed to different pharmacological treatments were submitted to migration in agarose gel. The migration of these highlighted two bands of 168bp and 268bp respectively, which correspond to the E G F and E G F - R amplifying m R N A (Fig. l a and b).
Identification of EGF and EGF-receptor in the U285 cells by means of immunocytochemical methods The microscopic observations of samples, tested with the kit tor the identification of EGF, revealed out the presence of marked cells in all the immunocytochemical preparations apart from the treatment (see Fig. 2 and 3 a and 3 b). The marker was localized both on the cellular membrane and in the cytoplasm as a coarse precipitate, the intensity of which varied in relation to the type of treatment and to suramin concentration. The percentages of E G F positive cells expressed in the different treatments are listed in Table 1. In the control preparation and in those treated with EGF, the percentage of marked cells is respectively 20.58% and 22%. The treatment ot the U285 cells with suramin and suramin/EGF increases the percentage of E G F positive cells, ranging between 38.88% and 45.57%. Statistical analysis carried on with the 3(2 test did not show any significant difference (p > 0.05) between the percentage ot marked cells in the E G F treated preparations and in those of control. Moreover, no significant difference (p > 0.05) was observed comparing the cells exposed to suramin (doses 200 gM and 400 gM) with those simultaneously exposed to suramin and E G F (suramin 200 gM + E G F 10 ng/ml and suramin 400 gM + E G F 10 ng/ml). On the contrary, the increase in the number of marked cells in the preparations exposed to suramin
and suramin/EGF, was statistically significant (p < 0.05) when compared to the controls and to the cells treated with E G E In the samples tested with the antibody antiEGF-R, positive cells were marked uniformly and the stain was localized, mostly on the cell membrane (see Figure 3c and 3d) in these preparations. Marked cell percentages in the control and in the different treatments, reported in Table1, ranged from 83.77% to 88.67% and statistical analysis did not highlight any significant variation (p > 0.05).
FRAME cytotoxicity test T R I A L A. In this trial we studied the short-term effects of the suramin on the proliferative rate of the U285 line, by exposing the cells to suramin starting from the seeding up until the neutral red labeling. The results obtained are reported in Figures 4 a, b. In the neutral red assay, the mean O. D. values detected at 24 h from the seeding are 0.2083 + 0.0500 in the control cultures, 0.2146 + 0.0533 in the cultures exposed at 200 ~tM suramin and 0.193+ 0.0633 in those treated with 400 ~tM of suramin. The statistical analysis carried out with one-way A N O V A did not detect any significant difference (p > 0.05) between these values. Also the mean O.D. detected at 48 h (0.3757 + 0.1039) and 72 h (0.6108 + 0.2710) of culture in the cells exposed to 200 gM do not differ (p > 0.05) from those detected in the corresponding controls (48h: 0.3762+ 0.0791; 72 h: 0.6204 + 0.1383). The cultures treated with 400 gM show instead a considerable decrease (p < 0.05) of the mean values of the O.D. at these two end points (48 h: 0.3283 + 0.1242; 72 h: 0.5416 + 0.1820) with respect to the controls. The determination of the kenacid blue confirms the results obtained with the neutral red assay. At 2 4 h of culture, the statistical analysis does not indicate any variation (p > 0.05) of the mean O.D. of the treated cultures (200 ~tM: 0.7922 + 0.1987; 400 gM: 0.7192 + 0.2426) with respect to the control one (0.8104 _+ 0.1495). At 48 h and 72 h from seeding, the mean O.D. values of the controls are respectively 1.3680 + 1882 and 2.3270 + 0.4210, while those of the cells exposed to a concentration of 200 gM are 1.1970+0.2551 and 2.0000+
Fig. 2. Immunocytochemical analysis of the interactions between suramin and EGF performed by the EGF-lmmunohistochemistry System kit (Oncogene Science). a) Negative control (blank) carried out to verify the deactivation of endogenous peroxidases (objective 40 ×). b) Control cells: 20.58% of cells are positive to anti-EGF antibody (objective 40 ×). c) CeIIs exposed to 10 ng/ml of EGF: 22% of cells are positive to anti-EGF antibody (objective 40 x). d) Cells exposed to 400 laM of suramin: 38.88% of cells are positive to anti-EGF antibody (objective 40 x). Fig. 3. Immunocytochemical analysis of the interactions between suramin and EGF (a and b) performed by the EGF-Immunohistochemistry System kit (Oncogene Science) and between suramin and EGF-R (c and d) carried out with the immunoperoxidase technique by Makin et al. using as primary antibody a monoclonal anti-EGF-R (monoclonal done 29.1, Sigma). a) Ceils exposed to 200 ~tM of suramin and 10 ng/ml of EGF: 45.18% of cells are positive to anti-EGF antibody (objective 40 ×). b) Cells exposed to 400 ~tM of suramin and 10 ng/ml of EGF: 45.57% of cells are positive to anti-EGF antibody (objective 40 x). c) Control cells: 83.77% of cells are positive to anti-EGF-R antibody (objective 100 x). d) Cells exposed to 400 gM of suramin: 87.12% of cells are positive to anti-EGF-R antibody (objective 100 x).
253
Table 1. Results •ftheimmun•cyt•chemica•ana•ysis•ftheinteracti•nsbetweensuraminandEGFandbetweensuraminandEGF-R
a b c d e f
Control EGF Sura 200 gM Sura 400 gME Sura 200 gM + EGF 10 ng/ml Sura 400 gM + EGF 10 ng/ml
EGF % positive cells
% negative cells
20.58 22.00 45.19 38.88 45.18 45.57
79.42 78.00 54.81 61.12 54.82 54.43
a) Trial A: neutral red
t .01
2411
4811
72h
culture time
control, ~
200 FM, ~
400 v.M, ~ p<0.05, "[" standard deviation
b) Trial A: kenacid blue
d 6
24h
48h
72h
culture time
control, rn-nn 200 I~M, ~
400 llM, ~ p<0.05, I
standard deviation
Fig. 4. Effects of treatment with suramin at seeding (TRIAL A) on neutral red and kenacid blue assays. Values are expressed as mean O. D. _+standard deviation of five experiments.
p < 0.05
EGF-receptor % positive cells
% negative cells
a/c, b/c a/d, b/d a/e, b/e a/f, b/f
83.77 85.52 87.37 87.12 86.71 88.67
16.23 14.48 12.63 12.88 13.29 11.33
p < 0.05
0.6609, and those of the cells treated with 400 gM are 1.1150 + 0.4267 and 1.8960 + 0.7842. These data show that both the suramin concentrations we tested induce decreases of the mean values of O. D. of the kenacid blue. In any case, the results of the statistical analysis show that the observed decrease is statistically significant (p < 0.05) only for the 400 gM dose. T R I A L B. This trial was performed in order to verify the recovery of cell growth after removal of the drug. The results obtained by exposing the cells to suramin only in the first 24 hr of culture are summarized in Figures 5 a and 5 b. In the neutral red assay the mean O.D. values detected at 48 h and 72 h of seeding are respectively 0.5206+0.0410 and 1.0210+0.0763 in the control cultures, 0.4988+0.0514 and 0.9896 +0.0890 in those exposed to 200 pM, 0.4258 _+ 0.0933 and 0.8438 + 0.1575 in those treated with 400 pM suramin. In the kenacid blue the mean O. D. values detected at 48 h and 72 h of seeding are respectively 1.3990 +_ 0.3022 and 1.7960 + 0.1598 in the control cultures, 1.2690+0.3740 and 1.7880+ 0.1926 in those exposed at 200 gM, 0.9478 + 0.2826 and 1.5720 + 0.2847 in those treated with 400 gM suramin. T R I A L C. In this trial, the results of which are reported in Figure 6 a and 6 b, we further investigated the recovery of the cells growth after exposition to suramin, carrying out the treatment of cultures between 24 h and 48 h from seeding and the determination of the neutral red and of the kenacid blue at 72 h. Both the suramin concentrations induce a slight increase of the mean O. D. either of the neutral red (200gM: 1.2190+0.8545; 400gM: 1.2190+0.0751) and of the kenacid blue (200 ktM: 2.0880 + 0.5082; 400 gM: 2.1320 + 0.4575) with respect to the corresponding controls (neutral red: 1.2020 + 0.1129; kenacid blue: 1.9190 + 0.2853). In any case, the results of the statistical analysis show that these increases are not statistically significant (p > 0.05). T R I A L D. The results of this trial, the operative modalities of which (contemporary exposition of the cells U285 to suramin and E G F since seeding) were intended to ascertain if it was possible to antagonize the action of suramin by adding exogenous EGF, are reported in Figure 7 a and 7b. Against any expectation, both the growth assays showed that the exogenous E G F does not exert any protective action towards the suramin, and
254
makes the U285 line more sensitive to this drug. In fact, neutral red determinations showed that at 24 hr and 48 hr after seeding, the m e a n O. D. values of the cultures treated with 400 gM of suramin (respectively of 0.2528 + 0.0210 and 0.6079+0.0428) were significantly lower (p < 0.05) than those of the control (24 h: 0.2889 + 0.0146; 48h: 0.7935_+0.0253). A t 7 2 h r of incubation, this p h e n o m e n o n appeared with both doses tested (control: 1.3350 + 0.0590; 200 gM: 1.1240 _+0.0549; 400 gM: 0.9580 _+0.0414). Kenacid blue determinations confirmed neutral red data. The m e a n O. D. values, detected at 24 h, 48 h, 72 h after seeding, in the control culture are respectively 0.7878 _+0.0301, 1.5770 _+0.0879 and 2.2020 _+ 0.0627. In culture treated with 200 gM the m e a n O. D. values are respectively 0.6925 _+ 0.0540, 1.4530 _+ 0.1416 and 2.0940 +_ 0.0945. 0.5949 _+ 0.0947. In culture treated with 400 gM of suramin the m e a n O. D. values are respectively
0.8958 + 0.0922 e 1.2910 _+ 0.0837. The statistical comparisons performed between these data highlight the same significances as evidenced in the neutral red assay. T R I A L E. This trial has also been performed in order to understand better the capacities of the exogenous E G F to antagonize the effect of suramin. The results of the treatment with suramin at seeding, adding 10 ng/ml of E G F 24 hr later, are summarized in Figure 8 a and 8b. For both the markers, statistical analysis revealed a relevant decrease (p < 0.05) of the mean O. D. values only in the cultures treated with 400 gM doses at 48 hr (neutral red: 0.2851 + 0.0162; kenacid blue: 0.9100 + 0.0333) and 72 hr (neutral red: 0.5847 + 0.0348; kenacid blue: 1.4550 + 0.0825) with respect to the corresponding controls (neutral red: 0.4344 _+ 0.0326, 0.7841 + 0.0570; kenacid blue: 1.4350 + 0.0585, 2.3380 _+ 0.1705).
a) Trial C: neutral red 1.50-
a) Trial B: neutral red
1.25.
n.~
n.s.
1.251.00. I .O00.75.
d
6 0.50'
0.75-
0.50-
I
i I Ii
0.25-
0.25-~
::::=: .............
i
i
i 0.00
0.00 24h
48h
24h
72h
~ i J I
48h
72h
culture time
colture time control, rTTT~ 200 p.M, ~
i
400 p-M, ~r p<0.05, "[" standard deviation
control, t lull i 200 I~M, ~
b) Trial B: kenacid blue
400 p.M, n.¢ p>0.05, ]~
standard deviation
b) Trial C: kenacid blue
2.0n.s.
n,s.
1.52
d
6
o
t.0-
0.5-
0.0.
24
control, itplrl 200 ~M, ~
48h colture time 400 ~M, *
24h
72h
48h
72h
culture time p<0.05, T standard deviation
control, rTT'nn 200 FM, ~
Fig. 5. Effects of treatment with suramin from seeding up to 24hr and subsequent substitution with a new medium (TRIAL B), on neutral red and kenacid blue assays. Values are expressed as mean O.D. + standard deviation of five experiments. 255
400 p.M, n,s. p>O.05, "]" standard deviation
Fig. 6. Effects of treatment with suramin from 24hr up to 48h after seeding and subsequent substitution with a new medium (TRIAL C), on neutral red and kenacid blue assays. Values are expressed as mean O.D. + standard deviation of five experiments.
Discussion
mechanism. This line could be made up of two functionally distinct cell populations: the former, numerically restricted (20%), producing EGF, and the latter, prevalent (80%), characterized by the expression of EGF-R. In this case the in-vitro growth ot the U285 cells would depend upon a paracrine regulation mechanism based on one cellular population that produces E G F and of another that employs it. As an alternative to the hypothesis of a paracrine growth control, we could presume the existence of an autocrine regulation system as described for the prostatic cell lines LNCaP (Connolly and Rose 1990) and D U 145 (Connolly and Rose 1989; Connolly and Rose 1994). In this case the discrepancy between the percentages of E G F and E G F - R cells positive could be justified by the difference in velocity between the production and secretion processes and that of E G F clearance. It is well known that by linking to its receptor, the E G F triggers a complex series of events such as the clustering on the
In this study we demonstrated the expression of codifying m R N A for E G F and E G F receptor in U285 cells by means of RT-PCR techniques. Immunocytochemical analysis, identifying their translation products, confirmed these data: in the assays tested with the kit for the EGF, the percentage of marked cells is 20%, while in those stained with an anti-EGF-R antibody this percentage is 80%. These results, therefore, explain the apparent insensitivity of our line to the action of this growth factor during the logarithmic growth phase, as reported previously (De Angeli et al. 1997). However, the immunocytochemical data reported in this study were determined by means of single labeling method, and were therefore insufficient to decide wether the E G F regulates the growth of U285 cells through a paracrine or autocrine
a) Trial E: neutral red
a) Trial D: neutral red 1.00" 1.50-
1.25-
iiiiii:!i
*
0.75-
1.00-
0
d d o.so-
0.75-
iiliii!
0.50-
0.00
48h
24h
:[ii =i ~::.::-:
ii!iiiiii IZ:2 !!!!!!!!~ 0.25-
!!!!!!!!!
iii~iiiii 2121 O.OC 72h
72h
48h
24h
culture time
culture time
control, ~
200 ~tM, ~
400 p.M, '/¢ p
standard deviation
control, rcr/5~ 200 I~M, ~\\\~ 400 FM, 9¢ p<0.05, 1" standard deviation
b) Trial E: kenacid blue
b) Trial D: kenacid blue 2.5-
iii!iiii!i!i
2.0-
=:
iiiiiiiiili! :!!=i!:::i iiiiiiiiiii ::=:::x ......:::::
1.5-
0 1.0"
0.0
24h
li 48h
•::.=:::: :: :::= =::..x x::::=::: :: ::x: ~iii~i!i ii;i;i!;~i
~iilZliiiiii lii!iiii!iii ..........
iiiii
0
72h
i 24h
culture time control, l l l t l i 200 pM, ~
48h culture time
...........I control, ftlrl ] 200 FM, ~
400 pM, 9¢ p<0.05, "[" standard deviation
Fig. 7. Effects of treatment with suramin at seeding adding, at the same time, 10 ng/ml of EGF (TRIAL D), on neutral red and kenacid blue assays. Values are expressed as mean O. D. + standard deviation of five experiments.
400 I~M, ~
iii 72h
p
Fig. 8. Effects of treatment with suramin at seeding adding after 24 hr 10 ng/ml of EGF (TRIAL E), on neutral red and kenacid blue assays. Values are expressed as mean O.D. -+ standard deviation of five experiments.
256
membrane cell surface of the EGF-receptor complexes and their internalization by endocytosis (Carpenter 1985; Carpenter and Cohen 1990). The recycling on the cell membrane of the complex EGF-receptors or the degradation of the E G F itself at a lysosomal level follows this, by a long metabolic pathway (Sorking et al. 1997). In cultures treated with suramin, the increase of the number of E G F positive cells, in the absence of a significant percentage variation in the number of cells with EGF-R, can be easily explained by the action mechanism of this drug (Larsen 1993; Stein 1993) and is compatible with both hypotheses we formulated. For this reason in a future study it would be interesting to verify the contemporary expression of E G F in the U285 line cells, and of its receptor by means of double labeling techniques. Cytotoxicity experiments allowed us to focus the role of the endogenous E G F in the regulation of the U285 cells proliferation. Trials A, B, e C showed, in agreement with the results of a previous research (De Angeli et al. 1994), that in the first 24 hours of culture the suramin interacts with an inductive mechanism, the inhibition of which causes a decrease in cell proliferation during the following 48 h and 72 h. Even the enhancement of the cytostatic action of the suramin, induced by the exogenous E G F when both these substances are given together at seeding (Trials D and E), confirms the importance of biological events that take place in U285 cells during the first 24 hours of culture. We have previously demonstrated that in this period of time, corresponding to the Lag phase, similar to that described by Connolly and Rose (1989, 1990) in the prostatic lines DU145 and LNCaR our cells are responsive to the stimulation of exogenous EGF, probably due to their low concentration of seeding. Taken overall, these data indicate that the endogenous E G F is responsible for the reactivation of the U285 cells and for their passage to the following logarithmic growth phase. Many important aspects of this phenomenon, such as the pathway through which cell reactivation take place, the presence of an inductive phenomenon of the receptor, the role played by the complex EGF-receptor in the down regulation of proliferation, remain to be explained and will be the objectives of further research.
Acknowledgements. This work was partially supported by grant 434/01/94 from the Regione Veneto, Giunta Regionale - Health Research - Venice - Italy.
References Borrenfreund E, Puerner JA (1985) Toxicity determined in vitro by morphological alterations and neutral red absorption. Toxicology Lett 24:11%124 Byrne RL, Leung H, Neal DE (1996) Peptide growth factors in the prostate as mediators of stromal epithelial interaction. Br J Urol 77:627~533 Carpenter G (1985) Epidermal growth factor: biology and receptor metabolism. J Cell Sci 3:1-9
Carpenter G, Cohen S (1990) Epidermal growth factor. J Biol Chem 265:7709-7712 Chopra DR Grignon D J, Joiakin A, Mathieu PA, Mohamed A, Sakr WA, Powell IJ, Sarkar FH (1996) Differential growth factor responses of epithelial cultures derived from normal human prostate, benign prostatic hyperplasia, and primary prostate carcinoma. J Cell Physiol 169:269-280 ConnollyJM, Rose DP (1989) Secretion of epidermal growth factor and related polypeptides by the DU145 human prostate cancer cell line. Prostate 15:177-186 Connolly JM, Rose DP (1990) Production of epidermal growth factor and transforming growth factor-a by androgen-responsive LNCaPprostate cancer cell line. Prostate 16:209-218 ConnoUyJM, Rose DP (1994) Regulation of DU145 human prostate cancer cell proliferation by insulin-like growth factors and its interaction with the epidermal growth factor autocrine loop. Prostate 24:167-175 Culig Z, Hobisch A, Cronauer MV, Radmayr C, Hittmair A, Zhang J, Thurnher M, Bartsch G, Klocker H (1996) Regulation of prostatic growth and function by peptide growth factors. Prostate 28:392-405 Dahiya R, Lee C, Haughney PC, Chui R, Ho R, Deng G (1996) Differential gene expression of transforming growth factors alpha and beta, epidermal growth factor, keratinocyte growth factor, and their receptors in fetal and adult human prostatic tissues and cancer cell lines. Urology 48:963-970 De Angeli S, Fandella A, Conconi MT, Anselmo G, Parnigotto PP (1994) Growth morphology and morphometry of human hypertrophic prostate cells treated with suramin in vitro. Prostate 25:117-124 De Angeli S, Valenti F, Durante E, Bassani V, Parnigotto PP (1995) Primary cultures of human hypertrophic prostate tissue in WAJC 404 medium: a study of cell morphology and kinetics. Ann Anat 177:185-192 De Angeli S, Valenti F, Buoro S, Conconi MT, Parnigotto PP (1996) Morphological features and kinetics of primary cultures of BPH tissue supported by TV1 medium. Ann Anat 178: 4148 De Angeli S, Favretti C, Buoro S, Fandella A, Anselmo G, Parnigotto PP (1997) Effect of DHT and EGF on human hyperplastic prostate stern cells cultured in vitro growth morphology and phenotype characterization. Ann Anat 179:255-264 Eaton CL, Davies P, Phillips MEA (1988) Growth factor involvement and oncogene expression in prostatic tumours. J Steroid Biochem 30:341-345 Fresheny RI (1987) Culture of animal cells: a manual of basic technique. Alan R Liss Inc. NY, pp 107-126 Ilio KY, Sensibar JA, Lee C (1995) Effect of TGF-beta 1, TGFalpha, and EGF on cell proliferation and cell death in rat ventral prostatic epithelial cells in culture. J Androl 16:482-490 Janssen T, Petein M, van Velthoven R, De Decker R, Assenmacher C, Corbusier A, Pasteeels JL, Kiss R, Schnlman C (1997) Coregulatory effects of epidermal growth factor, dehydrotestosterone, and prolactin on human prostatic hyperplasia tissue culture proliferation. Prostate 30:47-52 Kim JH, Sherwood ER, Sutkowski DM, Lee C, Kozlowski JM (1991) Inhibition of prostatic tumor cell proliferation by surarain: alterations in TGF alpha-mediated autocrine growth regulation and cell cycle distribution. J Urol 146:171-176 Knox P, Uphill PF, Fry JR, BenfordJ, BallssM (1986) The FRAME multicenter project on in vitro cytotoxicology. Fd Chem Toxic 24:457-463 Larsen AK (1993) Suramin. An anticancer drug with unique biological effects. Cancer Chem Pharm 32:96-98
257
Lechner JF, Babcock MS, Marnell M, Narayan S, Kaighn ME (1980) Normal human prostate epithelial cell cultures. Meth Cells Biol 21B: 191-224 Leibovitz A (1986) Development tumor cells lines. Cancer Genet Cytogenet 19:11-19 Makin CA, Bobrow LG, Bodmer WF (1984) Monoelonal anditody to cytokeratin for use in routine histopathology. J Clin Pathol 37:975-983 McKeehan WL, Adams PS, Rosser PM (1984) Direct mitogenic effects of insulin, EGF, glucoeorticoid, cholera toxin, unknown pituitary factors and possibly prolactin but not androgen on normal rat prostate epithelial cells in serum-free, primary cell culture. Cancer Res 44:1998-2010 McKeehan WL, Adams PS, Fast D (1987) Different hormonal requirement for androgen-independent growth of normal and turnout epithelial cells from rat prostate. In Vitro Cell Dev Biol 23:147-152 McKeehan WL (1991) Growth factor receptors and prostate cell growth. Canc Surv 11:165-175 Mukaida H, Hirabayashi N, Hirai T, Yamashita Y, Iwata T, Saeki S, Toge T (1994) The inhibitory effect caused by suramin on paracrine growth of human cancer cells and fibroblasts. Surg Today 24:234-240
Peehl D, Stamey TA (1986) Serum free growth of adult human prostatic epithelial cells. In Vitro Cell Dev Biol 22:82-90 Peehl D, Wong ST, Bazinet M, Stamey TA (1989) In vitro studies of human prostatic cells: attempts to identify distinguishing features of malignant cells. Growth Factors 1:237-250 Sherwood ER, Lee C (1995) Epidermal growth factor-related peptides and the epidermal growth factor receptor in normal and malignant prostate. World J Urol 13:290-296 Stein CA (1993) Suramin: a novel antineoplastic agent with multiple potential mechanism of action. Cancer Res 53:2239-2248 Sorkin A, Kornilova E, Teslenko L, Sorokin A, Nikolsky N (1997) Recycling of epidermal growth factor-receptor complex in A431 cells. Mol Cell Endocrinol 126:185-192 Walthr W, Stein U, Eder C (1994) RNA analysis using miniprep RNA in reverse transcription PCR. Biotechniques 17:674-675 Yayon A, Klagsbrun M (1990) Autocrine trasformation by chimeric signal peptide-basic fibroblast growth factor: reversal by suramin. Proc Nat Acad Sci USA 87:5346-5350
Accepted October 5, 1999
258