Int. J. Cancer: 52, 105-109 (1992) 0 1992 Wiley-Liss, Inc.
Publicationof the InternationalUnion Against Cancer Publication de I'Union lnternationaleContre le Cancel
SECRETION OF AUTOCRINE GROWTH-PROMOTING ACTIVITY BY RENAL-CARCINOMA CELLS TREATED WITH 5-FLUOROURACIL Naoki NISHIMURA, Shigeru KANDA,Yasuo YOGI,Masaya KAWAMURA, Mikio NAWURA, Hiroyuki HYAKUTAKE, Hiroshi KANETAKE and Yutaka SAITO' Department of Urology, Nagasaki University School of Medicine, Nagasaki 852, Japan. A study was made of the auto-proliferative activity of human renal-carcinoma cells in the supernatant from a carcinoma-cell culture in serum-free medium to which an anticancer agent had been added (5-FU). The human renal-cancer cells used in this study were of 3 strains: ACHN, VMRC-RCW and NT. When each line was cultured in medium containing no 5-FU, the supernatant showed almost no activity for stimulating DNA synthesis. However, when the line was cultured in the presence of 5-FU, the supernatant showed autocrine growth-promoting activity which strongly stimulated DNA synthesis of 3 renalcancer cell lines in a dose-dependent manner. Activity could be detected in a 4- to 6-kDa fraction by gel filtration. This fraction increased the DNA-synthesis-promoting activity of epidermal growth factor, transforming growth factor+, basic fibroblast growth factor and insulin, and was acid- and heat-stable. It was also stable against pepsin and dithiothreitol. DNA synthesis in BALB/c 3T3. adult rat hepatocytes and rabbit renal tubular cells was not affected by this fraction which was thus considered not to affect non-cancerouscells. Renal-cell carcinoma responds poorly to anticancer agents, and the autocrine activity of the fraction may possibly be a factor accounting for this resistance.
c 1992 Wilq-Liss, Inc. Growth factors play an important role in regulating the growth of normal and cancer tissues. Growth factors can be classified into 3 groups on the basis of their mode of secretion and action; endocrine, paracrine or autocrine. Autocrine secretion is frequently detected in cancer cells (Sporn and Roberts, 1985; Heldin and Westermark, 1989), but is also present in non-transformed cells (Walsh-Reitz et al., 1985; Vlodavsky et al., 1987). Renal-cell carcinoma, which usually originates in proximal tubular cells (Miettinen et al., 1983), is resistant to anticancer agents and thus very difficult to treat (Buzaid and Todd, 1989; Elson et al., 1988). The reason for this may be that a multi-drug resistance gene is involved in renal cancer (Fojo et al., 1987). There is also a possibility that the autocrine growth factor plays an important role. The authors thus determined the autocrine growth activity in the conditioned medium of each of 3 cell lines of renal-cell carcinoma (ACHN, VMRC-RCW and NT) in the presence of 5-fluOrouracil (5-FU), an anticancer agent for treatment of renal-cell carcinoma. Autocrine growth-stimulating activity was detected in the conditioned medium obtained from cells grown in the presence of 5-FU. It was not found when each cell line was cultured without 5-FU. MATERIAL AND METHODS
Ma terial Epidermal growth factor was purified from the submaxillary glands of male mice as described by Savage and Cohen (1972). Bovine basic fibroblast growth factor was purchased from Toyobo (Osaka, Japan) and insulin from Sigma (St. Louis, MO). Human transforming growth factor-p was obtained from Wako Pure chemicals (Osaka, Japan). Dulbecco's modified Eagle's medium (DMEM) was from Nissui Pharmaceuticals (Tokyo, Japan). lzSI-deoxyuridine (2200 Ci/mmol) was from New England Nuclear (Boston, MA).
Cell lines and culture methods The following human-renal carcinoma cell lines were used. ACHN cells (Boden et al., 1982), purchased from Dainippon
Pharmaceuticals (Osaka, Japan), VMRC-RCW from the Japanese Cancer Research Resource Bank, and NT cells (Tsuda et al., 1980) produced in our laboratory. BALB/c 3T3 cells of a fibroblast cell line were kindly provided by Prof. A. Ichihara (Enzyme Research Center, Tokushima University, Japan). The following media were used: Dulbecco's modified Eagle's medium supplemented with 10% FCS for ACHN and N T and BALB/c 3T3; Dulbecco's modified Eagle's medium supplemented with 10% FCS and 1% non-essential amino acids for VMRC-RCW. Each cell line was cultured at 37"C, under normal atmospheric pressure, in 5% COz and 95% air.
Isolation and culture of adult rat hepatocytes The methods have been reported elsewhere (Saha et al., 1990). Isolation and culture of rabbit renal cotfical tubular cells The methods have been reported elsewhere (Kanda et al., 1989). Assay of D N A synthesis The cells of each line were seeded into wells of 24-well culture plates at a density of 5 x lo4cells/cm2 and incubated at 37°C for 24 hr. The culture medium was replaced by fresh serum-free medium and the cells were cultured again for another 24 hr for NT and VMRC-RCW and 48 hr for ACHN. The medium was then replaced by fresh serum-free medium, and the samples were added to the cells. After 20 hr, 12s"Ideoxyuridine (1 Ci/ml) was added to the cells and the incorporation of iododeoxyuridine into D N A was measured with a gamma-ray counter (Kanda et al., 1989). BALB/c 3T3 cells were seeded into the wells at a density of 1.5 x lo4 cells/cm2 and, 24 hr later, the medium was replaced by fresh DMEM containing 0.1% bovine serum albumin and cultured for another 72 hr. After the medium had been discarded, fresh serum-free DMEM and the samples were added to the cells and culturing was continued for another 22 hr. 'z'I-deoxyuridine (1 p Ci/ml) was added to the cells and, after 2 hr, radioactivity was measured. Assay of D N A synthesis in primary cultured adult-rat hepatocytes and rabbit renal-cortical tubular cells The methods have been reported elsewhere (Saha et al., 1990; Kanda et al., 1989). Collection of medium conditioned by renal carcinoma cells with or without 5-FU After renal-carcinoma cells had attained confluence, the medium supplemented with fetal calf serum was removed and the cells were washed and cultured for 24 hr in a serum-free medium containing 5-FU at an adequate concentration. The medium was then collected and centrifuged at 2,000 g for 15 min. For removal of 5-FU, the supernatant was dialyzed against the respective medium for 24 hr using dialysis tubing 'To whom correspondence and reprint requests should be addressed. Received: February 3, 1992 and
in
revised form March 31. 1992.
106
NISHIMURA ETAL.
with a MW cut-off value of 3,500. The supernatant was then stored at -20°C. The medium containing no 5-FU was similarly collected and stored following dialysis. Molecular-sieve chromatography The medium was thawed, ultrafiltered with an Amicon YM-5 filter, and concentrated about 200-fold. A portion of the concentrate was dialysed against fresh medium for 12 hr and applied to a Bio Gel P-60 column (1.6 X 65 cm) equilibrated with 10 mM N-(2-hydroxyethyl)-piperazine-N'-2-ethanesulfonic acid (HEPES) containing 0.15 M NaCl (pH 7.2). It was then eluted at room temperature at a flow rate of 15 ml/hr and collected in 3-ml portions. Treatment of active fractions Active fractions from the Bio Gel P-60 column were heated at 100°C for 5 min or treated with acetic acid (0.2 M at 4°C for 16 hr) or dithiothreitol (SO mM at 24°C for 2 hr) or pepsin (10 pg/ml, 37"C, 2 hr). After treatment with dithiothreitol, 2.5 mg/ml BSA were added to each sample which was dialyzed against fresh culture medium. Pepsin digestion was terminated by neutralizing the sample with NaOH. Analysis of 5-FU content in the autocrine activity secreted by N T cells Standard 5-FU solution or the autocrine activity of NT cells obtained by gel filtration was extracted with ethyl acetate, dissolved in a 0.02 M phosphate buffer, p H 6.8 containing 2% methanol, applied to a YMC A-312 column equibrated in the
same buffer and eluted at a flow rate of 1 ml/min. The absorbance of the eluate was monitored at 264 nm. The data obtained were analyzed with the Shimazu chromatopac C-R4. RESULTS
Each renal-carcinoma cell line was cultured to a confluent monolayer and 5-FU was added at various concentrations. The supernatant was collected and autocrine activity determined. As shown in Figure 1, strong activity which enhanced DNA synthesis was detected and reached its maximum at a 5-FU concentration of 30-50 kg/ml. When the cells were observed under a phase-contrast microscope, virtually no detached cells or lysed cells could be detected even at a 5-FU concentration of 100 pg/ml. The supernatant of each renal-carcinoma cell culture with (30 pg/ml) or without 5-FU was added at various concentrations to the cells and DNA synthesis was assessed. The results are shown in Figure 2. Following the addition of the supernatant from a culture containing no 5-FU, D N A synthesis for each carcinoma-cell line was found not to increase. When the supernatant from a culture containing 5-FU was added, DNA synthesis of all 3 cell lines clearly and dose-dependently increased. Thus, the addition of 5-FU to medium appears to induce a substance which can stimulate D N A synthesis. The conditioned medium in the presence or absence of 5-FU (30 pg/ml) was concentrated 200-fold by ultrafiltration and fractionated by column chromatography using Bio Gel P-60, then
M
0
0
-
-
5 - FU (Cg/ml)
5 FU ( l g / m l )
5 FU ( l g / m l )
FIGURE1 - Dose-response effect of 5-FU on the DNA synthesis of renal-carcinoma cell lines. Experimental conditions are indicated in the text. ( 0 ) ACHN; (b) VMRC-RCW; (c) NT. Values are means from triplicate experiments. c,
L 2.0 -
B
A
7
X
a 0 v
ln
Q
z
n 0
50
100
conditioned medium ( % )
50
100
conditioned medium ( % )
0
50
1 O(
conditioned medium ( % )
FIGURE2 - Dose-dependent effects of the conditioned medium from 3 renal-carcinoma cell lines with or without 5-fluorouracil on DNA synthesis of their own cells. Experimental conditions are indicated in the text. Open circles, conditioned medium without 5-FU; closed circles, conditioned medium with 5-FU (30 pgiml). (a) ACHN; (b) VMRC-RCW; ( c ) NT. Values are means from triplicate experiments.
107
5-FU-INDUCED AUTOCRINE GROWTH ACTIVITY Non -Treated Condltloned Medlum
Vo 1 0
25K
4
C I
3.0
13.7K
I
5- FU Treated Conditloned Medlum
25K
Vo
Vt
13.7K
Vt
I 1.0 3.0-
-1
0 . 5 1.5-
-0
0 20
0 Vo
25K
60
40
13.7K
20
I
I F 1
60
40
0
60
40
25K
Vo
21 10. .050;
0
20
0
Vt
13.7K
I
Vt
4
k 20
40
60
Fractions ( 1 5 m l / t & )
FIGURE 3 - Molecular sieve chromatography of the conditioned medium from renal-carcinoma cell lines on a Bio Gel P-60 column. Concentrated conditioned medium from each renal-carcinoma cell was applied to the column and eluted as described in the text. Closed circles, DNA synthesis ofrenal-carcinoma cells; continuous line, absorbance at 280 nm. (a) and (b)ACHN; (c) and (d) VMRC-RCW; (e) and (f) NT. (a), (c) and (e), non-treated conditioned medium; (h), (d) and 5-FU treated conditioned medium. Markers were: chymotripsinogen A, MW 25 kDa; ribonuclease A, MW 13.7 kDa.
u),
active fraction
(a/ well)
active fraction (@/well )
active fraction (@/ well)
FIGURE 4 - Dose-dependent effect of autocrine growth activity from the medium conditioned by renal-carcinoma cell lines with 5-FU. Each conditioned mediumwas concentrated about 200-fold and applied to a Bio Gel P-60 column. The eluted active fractionswere added to each cell line and DNA synthesis was measured. (u) ACHN; (b) VMRC-RCW; (c) NT. Values are expressed as means from triplicate experiments.
I08
N I S H I M U R A ET A L .
its activity for increasing DNA synthesis was determined. The results are shown in Figure 3. In none of the cell lines did any fraction of the conditioned medium without 5-FU show DNA synthesis-stimulating activity. When 5-FU was added to the medium, fractions containing a substance with a molecular weight of 4-6 kDa showed activity that strongly enhanced DNA synthesis, i.e., autocrine activity. When the active fraction was added to cells at different concentrations, DNA synthesis of each carcinoma cell line increased dose-dependently. The results are shown in Figure 4. Moreover, DNA synthesis of carcinoma cell lines other than the cell line
none
insulin ~o-’M
b-FGF zng/m0
TGF-P
EGF
zngfmo
iongfmo
yielding that fraction increased (results not shown). DNA synthesis of the 3 renal-carcinoma cell lines reached a maximum at lo-’ M for insulin, 2 ng/ml for b-FGF, 2 ng/ml for TGF-P and 10 ngiml for EGF (Nishimura, 1991). The fraction showing autocrine growth activity was added along with these growth factors to N T cells, and changes in DNA synthesis were analyzed. Autocrine activity induced with 5-FU increased markedly following the addition of insulin, b-FGF, TGF-P and E G F (Fig. 5). Similar results were obtained with ACHN and VMRC-RCW. To investigate the nature of this autocrine activity, the active fraction was subjected to various treatments. As shown in Table I, the active substance was stable toward acids and heat, as well as pepsin and dithiothreitol. This substance may thus be a non-peptide. The effects of autocrine activity on normal, non-cancer cells were also investigated. Table I1 shows the effect due to the addition of a fraction with autocrine activity obtained from NT cells on DNA synthesis of 3T3 cells (mouse fibroblast cells), primary cultures of rat hepatocytes and rabbit renal-cortical tubular cells. The fraction with autocrine activity did not affect DNA synthesis of normal cells and was thus considered to specifically affect cancer cells. Finally, autocrine activity was analysed by high-performance liquid chromatography to determine the presence of 5-FU. Figure 6 shows that no 5-FU could be detected in the autocrine activity. The autocrine activity is thus not due to 5-FU itself.
IO%FCS
FIGURE 5 -Additive effect of the autocrine activity of NT cells on the various growth factors. Pooled fractions (30 pl/well) were added to NT cells with the indicated growth factors. Open bars, growth factor only; closed bars, growth factor with active fractions. The open bar in “none”, indicates no addition; closed bar in “none”; indicates active fraction only. Values are means from triplicate experiments. TABLE I
~
EFFECT OF VARIOUS TREATMENTS ON THE AUTOCRINE ACTIVITY OF ACHN CELLS ~~
~~~~
~
Residual activity
Treatment
5
0
10
15
(%\
None 10o”C,5 min 0.2 M acetic acid, 24”C, 2 hr 50 mM dithiothreitol, 24“C, 2 hr 10 pgiml pepsin, 3 7 T , 2 hr
100 94 92 95 97
Experimental conditions were as described in the text. Values are means from triplicate experiments.
time ( min )
0
5
10
time ( rnin 1
FIGURE6 -Analysis of 5-FU content in autocrine activity of NT cells eluted from Bio Gel P-60 column by high-performance liquid chromatography. (a) Standard 5-FU solution 2.5 pgiml; (6) autocrine activity. Height of peak 1 in (a) was 15107 and in (b),15. 5-FU was not detected in the autocrine activity of NT cells (the concentration of 5-FU was below 3 ngiml).
TABLE 11 EFFECT OF AUTOCRINE ACTIVITY ON T H E DNA SYNTHESIS O F BALBic 3T3 CELLS, PRIMARY RAT HEPATOCYTES AND RABBIT RENAL-CORTICAL TUBULAR CELLS ~
DNA synthesis (cpm) Addition NT
15
BALB’c 3T3
Rat hepatocytes
Rabbit tubular cells
625 r 54 501 2 21 1,450 2 30 20,334 2 2,398 None 520 t 104 1,580 & 65 564 t 122 Autocrine activity (50 pl) 92,912 2 1,146 3,589 f 394 4,675 2 225 ND ND Insulin lO-’M + EGF 10 ngirnl ND ND 45,262 5 742 34,626 t 2,209 10% FCS
ND; not determined. Experimental conditions were as described in the text. Values are means 2 SD from triplicate experiments.
5-FU-INDUCED AUTOCRINE GROWTH ACTIVITY
DISCUSSION
The present study reports the addition of 5-FU to cultured renal-carcinoma cells to induce the secretion of a substance possessing autocrine activity which enhances D N A synthesis. This activity was induced by 5-FU in all 3 carcinoma cell lines and was detected in the fraction of the culture medium containing a substance with a molecular weight of 4-6 kDa. The fraction showing autocrine activity from each carcinoma cell line also enhanced D N A synthesis of the other 2 lines. Bascd on the additive effects observed with other growth factors and chemical properties, the active fractions from the 3 renal-carcinoma cell lines may b e considered to induce the same active substance. The active substance was stable toward acids, heat, pepsin and dithiothreitol, and thus may be a low-molecular-weight substance bound to a secreted protein, rather than a peptide growth factor. Renal-cell carcinoma had been shown to secrete autocrine growth factors. Nakamoto et al. (1988) reported the secretion of a factor having affinity for heparin from a renal-cell carcinoma cell line. Burton et al. (1990) reported a parathyroid-hormone-related peptide as an autocrine growth factor on renal-cell carcinoma with hypercalcemia. Miki et al. (1989) observed interleukin-6 to function as the autocrine growth factor. However, the factor showing autocrine activity, found in the present study, showed characteristics differing from those reported for any of the above factors. Our factor thus appears to be completely different from other known factors. 5-FU is metabolized in cells to yield FUTP and FdUMP, which cause R N A impairment and inhibition of D N A synthesis and subsequent cell toxicity (Cadman et al., 1981). 5-FU added to cells of renal-cell
109
carcinoma may thus possibly lead to changes in carcinoma cells to induce the secretion of a substance which enhances cell growth. In the present study, the active fraction showing autocrine growth activity did not affect D N A synthesis of fibroblast or epithelial cells, both normal cells. The multi-drugresistance gene may be involved in the resistance of renal-cell carcinoma toward anticancer agents (Fojo et al., 1987), but many points remain unclear in this respect. If the induction of secretion of the autocrine growth factor by 5 F U , detected here in vitro, is also present in vivo, the administration of an anticancer agent would enhance the growth of the carcinoma. This may be related to the resistance of renal-cell carcinoma to anticancer agents. At present, it is not clear whether an occurrence similar to that reported in the present paper also occurs in other cancer cells, or whether drugs other than 5-FU cause such an occurrence. In the future, we will purify the active fraction and attempt to determine whether secretion of the autocrine growth factor occurs in vivo. Whether the activity of this factor can be used as a tumor marker is a point that will also be studied.
ACKNOWLEDGEMENTS
W e thank Dr. K. Tanaka (Enzyme Research Center, Tokushima University, Japan) for helpful discussions and T. Shimogama, M. Yoshimoto, S. Honda and E. Suda for excellent technical assistance. This work was partially supported by a Grant-in-Aid for Scientific Research, Ministry of Education, Science and Culture, Japan.
REFERENCES BODEN,E.C., HOGAN,T.F. and VOELKEL.J.G.. Comparative antiprolif- functions as an in vitro autocrine growth factor in renal cell carcinoma. erative activity in vitro of natural interferons 01 and p for diploid and FEES Lett., 250,607-610 (1989). transformed human cells. Cancer Res., 42,49484953 (1982). NAKAMOTO, T., Usui, A., OSHIMA,K., IKEMOTO,H.. MITANI. S. and BURION,P.B., MONIZ,C. and KNIGHT, D.E., Parathyroid hormone USUI, T., Analysis of growth factors in renal cell carcinoma. Biochem. related peptide can function as autocrine growth factor in human renal hiophys. Res. Commun., 153,818-824 (1988). cell carcinoma. Biochem. Biop/zys. Res. Commun., 167, 1134-1 138 NISHIMURA, N., The effect of epidermal growth factor, basic fibroblast ( 1990). growth factor, transforming growth factor-p and insulin on the DNA BUZAID.A.C. and TODD,M.B., Therapeutic options in renal cell synthesis of renal carcinoma cell lines. Acta med. Nagasaki. 36, 47-51 carcinoma. Semin. onto/., 16,12-19 (1989). (1991). CADMAN. E.. HEIMER.R . and BENZS.C.. The influence of methotrexSAHA, P.K., KANDA, S., MORIMITSU. H., NISHIMURA, N., KANETAKE, H. ate pretreatment on 5-fluorouracil metabolism in L 1210 cells. J. bid. and SAITO,Y., Transforming growth factor p-like activity in human Chem., 256, 1695-1704 (1981). hydrocele fluid. Urol. Res., 18,295-298 (1990). ELSON,P.J., W i n e , R.S. and TRUMP, D.L.. Prognostic factors for SAVAGE,C.R. and COHEN,S., Epidermal growth factor and a new survival in patients with recurrent or metastatic renal cell carcinoma. derivative. Rapid isolation procedures and biological and chemical Cancer Res.. 48,7310-7313 (1988). characterization. J. biol. Chem.. 247,7609-761 1 (1972). FOJO,A.T., SHEN.D.W. and MICKLEY, L.A., Intrinsic drug resistance SPORN,M.B. and ROBERTS,A.B., Autocrine growth factors and in human kidney cancer is associated with expression of a human cancer. Nature (Lond.), 313,745-747 (1985). multi-drug resistance gene. J. clin. Oncol., 5, 1922-1927 (1987). HELDIN. C.-H. and WES-rERMARK, B., Growth factors as transforming TSUDA,N., ARAKI,J., YUSHITA,Y., TAKAGI,T., KISHIKAWA, M.. proteins. Europ. J . Biochem., 184,487-496 (1989). I.. SAITO,Y., SHINDO,K. and KANETAKE.H., UltrastrucNISHIMORI, KANDA,S., NOMATA,K., SAHA,P.K., NISHIMURA, N., YAMADA,J., tural study of a continuous cell line from human renal cell carcinoma, KANETAKE, H. and SAITO,Y., Growth regulation of the renal cortical J. elin. Electron Microsc., 13,5-6 (1980). tubular cells by epidermal growth factor, insulin-like growth factor-I, VLODAVSKY, I., FRIDMAN, R., SULLIVAN, R., SASSE,J. and KLAGSBRUN, acidic and basic fibroblast growth factor, and transforming growth M., Aortic endothelial cells synthesize basic fibroblast growth factor factor-p in serum-free culture. Cell. Biol. int. Rep.. 13,687-689 (1989). which remains cell associated and platelet-derived growth factor-like MIETTINEN. H.H.. PAASIVUO, R., LEHTO,V.-P., LINDER,E., ALFTHAN, protein which is secreted. J. cell. Physiol., 131,402408 (1987). 0 . and VIRTANEN, I., Cellular origin and differentiation of renal WALSH-REITZ,M.M., GLUCK,S.L., WAACK, S. and TOBACK. G., carcinomas. Lab. Invest., 49,317-326 (1983). Lowering extracellular Na+ concentration releases autocrine growth MIKI,S.. IWANO,M., MIKI,Y., YAMAMOTO, M.. TANG, B., YOKOKAWA. factors from renal epithelial cells. Proc. nat. Acad. Sci. (Wash.), 83, K., SONODA.T.. HIRANO,T. and KISHIMOTO, T., Interleukin-6 (IL-6) 47644768 (1986).
Publicationof the InternationalUnion Against Cancer Publication de I'Union lnternationaleContre le Cancel
SECRETION OF AUTOCRINE GROWTH-PROMOTING ACTIVITY BY RENAL-CARCINOMA CELLS TREATED WITH 5-FLUOROURACIL Naoki NISHIMURA, Shigeru KANDA,Yasuo YOGI,Masaya KAWAMURA, Mikio NAWURA, Hiroyuki HYAKUTAKE, Hiroshi KANETAKE and Yutaka SAITO' Department of Urology, Nagasaki University School of Medicine, Nagasaki 852, Japan. A study was made of the auto-proliferative activity of human renal-carcinoma cells in the supernatant from a carcinoma-cell culture in serum-free medium to which an anticancer agent had been added (5-FU). The human renal-cancer cells used in this study were of 3 strains: ACHN, VMRC-RCW and NT. When each line was cultured in medium containing no 5-FU, the supernatant showed almost no activity for stimulating DNA synthesis. However, when the line was cultured in the presence of 5-FU, the supernatant showed autocrine growth-promoting activity which strongly stimulated DNA synthesis of 3 renalcancer cell lines in a dose-dependent manner. Activity could be detected in a 4- to 6-kDa fraction by gel filtration. This fraction increased the DNA-synthesis-promoting activity of epidermal growth factor, transforming growth factor+, basic fibroblast growth factor and insulin, and was acid- and heat-stable. It was also stable against pepsin and dithiothreitol. DNA synthesis in BALB/c 3T3. adult rat hepatocytes and rabbit renal tubular cells was not affected by this fraction which was thus considered not to affect non-cancerouscells. Renal-cell carcinoma responds poorly to anticancer agents, and the autocrine activity of the fraction may possibly be a factor accounting for this resistance.
c 1992 Wilq-Liss, Inc. Growth factors play an important role in regulating the growth of normal and cancer tissues. Growth factors can be classified into 3 groups on the basis of their mode of secretion and action; endocrine, paracrine or autocrine. Autocrine secretion is frequently detected in cancer cells (Sporn and Roberts, 1985; Heldin and Westermark, 1989), but is also present in non-transformed cells (Walsh-Reitz et al., 1985; Vlodavsky et al., 1987). Renal-cell carcinoma, which usually originates in proximal tubular cells (Miettinen et al., 1983), is resistant to anticancer agents and thus very difficult to treat (Buzaid and Todd, 1989; Elson et al., 1988). The reason for this may be that a multi-drug resistance gene is involved in renal cancer (Fojo et al., 1987). There is also a possibility that the autocrine growth factor plays an important role. The authors thus determined the autocrine growth activity in the conditioned medium of each of 3 cell lines of renal-cell carcinoma (ACHN, VMRC-RCW and NT) in the presence of 5-fluOrouracil (5-FU), an anticancer agent for treatment of renal-cell carcinoma. Autocrine growth-stimulating activity was detected in the conditioned medium obtained from cells grown in the presence of 5-FU. It was not found when each cell line was cultured without 5-FU. MATERIAL AND METHODS
Ma terial Epidermal growth factor was purified from the submaxillary glands of male mice as described by Savage and Cohen (1972). Bovine basic fibroblast growth factor was purchased from Toyobo (Osaka, Japan) and insulin from Sigma (St. Louis, MO). Human transforming growth factor-p was obtained from Wako Pure chemicals (Osaka, Japan). Dulbecco's modified Eagle's medium (DMEM) was from Nissui Pharmaceuticals (Tokyo, Japan). lzSI-deoxyuridine (2200 Ci/mmol) was from New England Nuclear (Boston, MA).
Cell lines and culture methods The following human-renal carcinoma cell lines were used. ACHN cells (Boden et al., 1982), purchased from Dainippon
Pharmaceuticals (Osaka, Japan), VMRC-RCW from the Japanese Cancer Research Resource Bank, and NT cells (Tsuda et al., 1980) produced in our laboratory. BALB/c 3T3 cells of a fibroblast cell line were kindly provided by Prof. A. Ichihara (Enzyme Research Center, Tokushima University, Japan). The following media were used: Dulbecco's modified Eagle's medium supplemented with 10% FCS for ACHN and N T and BALB/c 3T3; Dulbecco's modified Eagle's medium supplemented with 10% FCS and 1% non-essential amino acids for VMRC-RCW. Each cell line was cultured at 37"C, under normal atmospheric pressure, in 5% COz and 95% air.
Isolation and culture of adult rat hepatocytes The methods have been reported elsewhere (Saha et al., 1990). Isolation and culture of rabbit renal cotfical tubular cells The methods have been reported elsewhere (Kanda et al., 1989). Assay of D N A synthesis The cells of each line were seeded into wells of 24-well culture plates at a density of 5 x lo4cells/cm2 and incubated at 37°C for 24 hr. The culture medium was replaced by fresh serum-free medium and the cells were cultured again for another 24 hr for NT and VMRC-RCW and 48 hr for ACHN. The medium was then replaced by fresh serum-free medium, and the samples were added to the cells. After 20 hr, 12s"Ideoxyuridine (1 Ci/ml) was added to the cells and the incorporation of iododeoxyuridine into D N A was measured with a gamma-ray counter (Kanda et al., 1989). BALB/c 3T3 cells were seeded into the wells at a density of 1.5 x lo4 cells/cm2 and, 24 hr later, the medium was replaced by fresh DMEM containing 0.1% bovine serum albumin and cultured for another 72 hr. After the medium had been discarded, fresh serum-free DMEM and the samples were added to the cells and culturing was continued for another 22 hr. 'z'I-deoxyuridine (1 p Ci/ml) was added to the cells and, after 2 hr, radioactivity was measured. Assay of D N A synthesis in primary cultured adult-rat hepatocytes and rabbit renal-cortical tubular cells The methods have been reported elsewhere (Saha et al., 1990; Kanda et al., 1989). Collection of medium conditioned by renal carcinoma cells with or without 5-FU After renal-carcinoma cells had attained confluence, the medium supplemented with fetal calf serum was removed and the cells were washed and cultured for 24 hr in a serum-free medium containing 5-FU at an adequate concentration. The medium was then collected and centrifuged at 2,000 g for 15 min. For removal of 5-FU, the supernatant was dialyzed against the respective medium for 24 hr using dialysis tubing 'To whom correspondence and reprint requests should be addressed. Received: February 3, 1992 and
in
revised form March 31. 1992.
106
NISHIMURA ETAL.
with a MW cut-off value of 3,500. The supernatant was then stored at -20°C. The medium containing no 5-FU was similarly collected and stored following dialysis. Molecular-sieve chromatography The medium was thawed, ultrafiltered with an Amicon YM-5 filter, and concentrated about 200-fold. A portion of the concentrate was dialysed against fresh medium for 12 hr and applied to a Bio Gel P-60 column (1.6 X 65 cm) equilibrated with 10 mM N-(2-hydroxyethyl)-piperazine-N'-2-ethanesulfonic acid (HEPES) containing 0.15 M NaCl (pH 7.2). It was then eluted at room temperature at a flow rate of 15 ml/hr and collected in 3-ml portions. Treatment of active fractions Active fractions from the Bio Gel P-60 column were heated at 100°C for 5 min or treated with acetic acid (0.2 M at 4°C for 16 hr) or dithiothreitol (SO mM at 24°C for 2 hr) or pepsin (10 pg/ml, 37"C, 2 hr). After treatment with dithiothreitol, 2.5 mg/ml BSA were added to each sample which was dialyzed against fresh culture medium. Pepsin digestion was terminated by neutralizing the sample with NaOH. Analysis of 5-FU content in the autocrine activity secreted by N T cells Standard 5-FU solution or the autocrine activity of NT cells obtained by gel filtration was extracted with ethyl acetate, dissolved in a 0.02 M phosphate buffer, p H 6.8 containing 2% methanol, applied to a YMC A-312 column equibrated in the
same buffer and eluted at a flow rate of 1 ml/min. The absorbance of the eluate was monitored at 264 nm. The data obtained were analyzed with the Shimazu chromatopac C-R4. RESULTS
Each renal-carcinoma cell line was cultured to a confluent monolayer and 5-FU was added at various concentrations. The supernatant was collected and autocrine activity determined. As shown in Figure 1, strong activity which enhanced DNA synthesis was detected and reached its maximum at a 5-FU concentration of 30-50 kg/ml. When the cells were observed under a phase-contrast microscope, virtually no detached cells or lysed cells could be detected even at a 5-FU concentration of 100 pg/ml. The supernatant of each renal-carcinoma cell culture with (30 pg/ml) or without 5-FU was added at various concentrations to the cells and DNA synthesis was assessed. The results are shown in Figure 2. Following the addition of the supernatant from a culture containing no 5-FU, D N A synthesis for each carcinoma-cell line was found not to increase. When the supernatant from a culture containing 5-FU was added, DNA synthesis of all 3 cell lines clearly and dose-dependently increased. Thus, the addition of 5-FU to medium appears to induce a substance which can stimulate D N A synthesis. The conditioned medium in the presence or absence of 5-FU (30 pg/ml) was concentrated 200-fold by ultrafiltration and fractionated by column chromatography using Bio Gel P-60, then
M
0
0
-
-
5 - FU (Cg/ml)
5 FU ( l g / m l )
5 FU ( l g / m l )
FIGURE1 - Dose-response effect of 5-FU on the DNA synthesis of renal-carcinoma cell lines. Experimental conditions are indicated in the text. ( 0 ) ACHN; (b) VMRC-RCW; (c) NT. Values are means from triplicate experiments. c,
L 2.0 -
B
A
7
X
a 0 v
ln
Q
z
n 0
50
100
conditioned medium ( % )
50
100
conditioned medium ( % )
0
50
1 O(
conditioned medium ( % )
FIGURE2 - Dose-dependent effects of the conditioned medium from 3 renal-carcinoma cell lines with or without 5-fluorouracil on DNA synthesis of their own cells. Experimental conditions are indicated in the text. Open circles, conditioned medium without 5-FU; closed circles, conditioned medium with 5-FU (30 pgiml). (a) ACHN; (b) VMRC-RCW; ( c ) NT. Values are means from triplicate experiments.
107
5-FU-INDUCED AUTOCRINE GROWTH ACTIVITY Non -Treated Condltloned Medlum
Vo 1 0
25K
4
C I
3.0
13.7K
I
5- FU Treated Conditloned Medlum
25K
Vo
Vt
13.7K
Vt
I 1.0 3.0-
-1
0 . 5 1.5-
-0
0 20
0 Vo
25K
60
40
13.7K
20
I
I F 1
60
40
0
60
40
25K
Vo
21 10. .050;
0
20
0
Vt
13.7K
I
Vt
4
k 20
40
60
Fractions ( 1 5 m l / t & )
FIGURE 3 - Molecular sieve chromatography of the conditioned medium from renal-carcinoma cell lines on a Bio Gel P-60 column. Concentrated conditioned medium from each renal-carcinoma cell was applied to the column and eluted as described in the text. Closed circles, DNA synthesis ofrenal-carcinoma cells; continuous line, absorbance at 280 nm. (a) and (b)ACHN; (c) and (d) VMRC-RCW; (e) and (f) NT. (a), (c) and (e), non-treated conditioned medium; (h), (d) and 5-FU treated conditioned medium. Markers were: chymotripsinogen A, MW 25 kDa; ribonuclease A, MW 13.7 kDa.
u),
active fraction
(a/ well)
active fraction (@/well )
active fraction (@/ well)
FIGURE 4 - Dose-dependent effect of autocrine growth activity from the medium conditioned by renal-carcinoma cell lines with 5-FU. Each conditioned mediumwas concentrated about 200-fold and applied to a Bio Gel P-60 column. The eluted active fractionswere added to each cell line and DNA synthesis was measured. (u) ACHN; (b) VMRC-RCW; (c) NT. Values are expressed as means from triplicate experiments.
I08
N I S H I M U R A ET A L .
its activity for increasing DNA synthesis was determined. The results are shown in Figure 3. In none of the cell lines did any fraction of the conditioned medium without 5-FU show DNA synthesis-stimulating activity. When 5-FU was added to the medium, fractions containing a substance with a molecular weight of 4-6 kDa showed activity that strongly enhanced DNA synthesis, i.e., autocrine activity. When the active fraction was added to cells at different concentrations, DNA synthesis of each carcinoma cell line increased dose-dependently. The results are shown in Figure 4. Moreover, DNA synthesis of carcinoma cell lines other than the cell line
none
insulin ~o-’M
b-FGF zng/m0
TGF-P
EGF
zngfmo
iongfmo
yielding that fraction increased (results not shown). DNA synthesis of the 3 renal-carcinoma cell lines reached a maximum at lo-’ M for insulin, 2 ng/ml for b-FGF, 2 ng/ml for TGF-P and 10 ngiml for EGF (Nishimura, 1991). The fraction showing autocrine growth activity was added along with these growth factors to N T cells, and changes in DNA synthesis were analyzed. Autocrine activity induced with 5-FU increased markedly following the addition of insulin, b-FGF, TGF-P and E G F (Fig. 5). Similar results were obtained with ACHN and VMRC-RCW. To investigate the nature of this autocrine activity, the active fraction was subjected to various treatments. As shown in Table I, the active substance was stable toward acids and heat, as well as pepsin and dithiothreitol. This substance may thus be a non-peptide. The effects of autocrine activity on normal, non-cancer cells were also investigated. Table I1 shows the effect due to the addition of a fraction with autocrine activity obtained from NT cells on DNA synthesis of 3T3 cells (mouse fibroblast cells), primary cultures of rat hepatocytes and rabbit renal-cortical tubular cells. The fraction with autocrine activity did not affect DNA synthesis of normal cells and was thus considered to specifically affect cancer cells. Finally, autocrine activity was analysed by high-performance liquid chromatography to determine the presence of 5-FU. Figure 6 shows that no 5-FU could be detected in the autocrine activity. The autocrine activity is thus not due to 5-FU itself.
IO%FCS
FIGURE 5 -Additive effect of the autocrine activity of NT cells on the various growth factors. Pooled fractions (30 pl/well) were added to NT cells with the indicated growth factors. Open bars, growth factor only; closed bars, growth factor with active fractions. The open bar in “none”, indicates no addition; closed bar in “none”; indicates active fraction only. Values are means from triplicate experiments. TABLE I
~
EFFECT OF VARIOUS TREATMENTS ON THE AUTOCRINE ACTIVITY OF ACHN CELLS ~~
~~~~
~
Residual activity
Treatment
5
0
10
15
(%\
None 10o”C,5 min 0.2 M acetic acid, 24”C, 2 hr 50 mM dithiothreitol, 24“C, 2 hr 10 pgiml pepsin, 3 7 T , 2 hr
100 94 92 95 97
Experimental conditions were as described in the text. Values are means from triplicate experiments.
time ( min )
0
5
10
time ( rnin 1
FIGURE6 -Analysis of 5-FU content in autocrine activity of NT cells eluted from Bio Gel P-60 column by high-performance liquid chromatography. (a) Standard 5-FU solution 2.5 pgiml; (6) autocrine activity. Height of peak 1 in (a) was 15107 and in (b),15. 5-FU was not detected in the autocrine activity of NT cells (the concentration of 5-FU was below 3 ngiml).
TABLE 11 EFFECT OF AUTOCRINE ACTIVITY ON T H E DNA SYNTHESIS O F BALBic 3T3 CELLS, PRIMARY RAT HEPATOCYTES AND RABBIT RENAL-CORTICAL TUBULAR CELLS ~
DNA synthesis (cpm) Addition NT
15
BALB’c 3T3
Rat hepatocytes
Rabbit tubular cells
625 r 54 501 2 21 1,450 2 30 20,334 2 2,398 None 520 t 104 1,580 & 65 564 t 122 Autocrine activity (50 pl) 92,912 2 1,146 3,589 f 394 4,675 2 225 ND ND Insulin lO-’M + EGF 10 ngirnl ND ND 45,262 5 742 34,626 t 2,209 10% FCS
ND; not determined. Experimental conditions were as described in the text. Values are means 2 SD from triplicate experiments.
5-FU-INDUCED AUTOCRINE GROWTH ACTIVITY
DISCUSSION
The present study reports the addition of 5-FU to cultured renal-carcinoma cells to induce the secretion of a substance possessing autocrine activity which enhances D N A synthesis. This activity was induced by 5-FU in all 3 carcinoma cell lines and was detected in the fraction of the culture medium containing a substance with a molecular weight of 4-6 kDa. The fraction showing autocrine activity from each carcinoma cell line also enhanced D N A synthesis of the other 2 lines. Bascd on the additive effects observed with other growth factors and chemical properties, the active fractions from the 3 renal-carcinoma cell lines may b e considered to induce the same active substance. The active substance was stable toward acids, heat, pepsin and dithiothreitol, and thus may be a low-molecular-weight substance bound to a secreted protein, rather than a peptide growth factor. Renal-cell carcinoma had been shown to secrete autocrine growth factors. Nakamoto et al. (1988) reported the secretion of a factor having affinity for heparin from a renal-cell carcinoma cell line. Burton et al. (1990) reported a parathyroid-hormone-related peptide as an autocrine growth factor on renal-cell carcinoma with hypercalcemia. Miki et al. (1989) observed interleukin-6 to function as the autocrine growth factor. However, the factor showing autocrine activity, found in the present study, showed characteristics differing from those reported for any of the above factors. Our factor thus appears to be completely different from other known factors. 5-FU is metabolized in cells to yield FUTP and FdUMP, which cause R N A impairment and inhibition of D N A synthesis and subsequent cell toxicity (Cadman et al., 1981). 5-FU added to cells of renal-cell
109
carcinoma may thus possibly lead to changes in carcinoma cells to induce the secretion of a substance which enhances cell growth. In the present study, the active fraction showing autocrine growth activity did not affect D N A synthesis of fibroblast or epithelial cells, both normal cells. The multi-drugresistance gene may be involved in the resistance of renal-cell carcinoma toward anticancer agents (Fojo et al., 1987), but many points remain unclear in this respect. If the induction of secretion of the autocrine growth factor by 5 F U , detected here in vitro, is also present in vivo, the administration of an anticancer agent would enhance the growth of the carcinoma. This may be related to the resistance of renal-cell carcinoma to anticancer agents. At present, it is not clear whether an occurrence similar to that reported in the present paper also occurs in other cancer cells, or whether drugs other than 5-FU cause such an occurrence. In the future, we will purify the active fraction and attempt to determine whether secretion of the autocrine growth factor occurs in vivo. Whether the activity of this factor can be used as a tumor marker is a point that will also be studied.
ACKNOWLEDGEMENTS
W e thank Dr. K. Tanaka (Enzyme Research Center, Tokushima University, Japan) for helpful discussions and T. Shimogama, M. Yoshimoto, S. Honda and E. Suda for excellent technical assistance. This work was partially supported by a Grant-in-Aid for Scientific Research, Ministry of Education, Science and Culture, Japan.
REFERENCES BODEN,E.C., HOGAN,T.F. and VOELKEL.J.G.. Comparative antiprolif- functions as an in vitro autocrine growth factor in renal cell carcinoma. erative activity in vitro of natural interferons 01 and p for diploid and FEES Lett., 250,607-610 (1989). transformed human cells. Cancer Res., 42,49484953 (1982). NAKAMOTO, T., Usui, A., OSHIMA,K., IKEMOTO,H.. MITANI. S. and BURION,P.B., MONIZ,C. and KNIGHT, D.E., Parathyroid hormone USUI, T., Analysis of growth factors in renal cell carcinoma. Biochem. related peptide can function as autocrine growth factor in human renal hiophys. Res. Commun., 153,818-824 (1988). cell carcinoma. Biochem. Biop/zys. Res. Commun., 167, 1134-1 138 NISHIMURA, N., The effect of epidermal growth factor, basic fibroblast ( 1990). growth factor, transforming growth factor-p and insulin on the DNA BUZAID.A.C. and TODD,M.B., Therapeutic options in renal cell synthesis of renal carcinoma cell lines. Acta med. Nagasaki. 36, 47-51 carcinoma. Semin. onto/., 16,12-19 (1989). (1991). CADMAN. E.. HEIMER.R . and BENZS.C.. The influence of methotrexSAHA, P.K., KANDA, S., MORIMITSU. H., NISHIMURA, N., KANETAKE, H. ate pretreatment on 5-fluorouracil metabolism in L 1210 cells. J. bid. and SAITO,Y., Transforming growth factor p-like activity in human Chem., 256, 1695-1704 (1981). hydrocele fluid. Urol. Res., 18,295-298 (1990). ELSON,P.J., W i n e , R.S. and TRUMP, D.L.. Prognostic factors for SAVAGE,C.R. and COHEN,S., Epidermal growth factor and a new survival in patients with recurrent or metastatic renal cell carcinoma. derivative. Rapid isolation procedures and biological and chemical Cancer Res.. 48,7310-7313 (1988). characterization. J. biol. Chem.. 247,7609-761 1 (1972). FOJO,A.T., SHEN.D.W. and MICKLEY, L.A., Intrinsic drug resistance SPORN,M.B. and ROBERTS,A.B., Autocrine growth factors and in human kidney cancer is associated with expression of a human cancer. Nature (Lond.), 313,745-747 (1985). multi-drug resistance gene. J. clin. Oncol., 5, 1922-1927 (1987). HELDIN. C.-H. and WES-rERMARK, B., Growth factors as transforming TSUDA,N., ARAKI,J., YUSHITA,Y., TAKAGI,T., KISHIKAWA, M.. proteins. Europ. J . Biochem., 184,487-496 (1989). I.. SAITO,Y., SHINDO,K. and KANETAKE.H., UltrastrucNISHIMORI, KANDA,S., NOMATA,K., SAHA,P.K., NISHIMURA, N., YAMADA,J., tural study of a continuous cell line from human renal cell carcinoma, KANETAKE, H. and SAITO,Y., Growth regulation of the renal cortical J. elin. Electron Microsc., 13,5-6 (1980). tubular cells by epidermal growth factor, insulin-like growth factor-I, VLODAVSKY, I., FRIDMAN, R., SULLIVAN, R., SASSE,J. and KLAGSBRUN, acidic and basic fibroblast growth factor, and transforming growth M., Aortic endothelial cells synthesize basic fibroblast growth factor factor-p in serum-free culture. Cell. Biol. int. Rep.. 13,687-689 (1989). which remains cell associated and platelet-derived growth factor-like MIETTINEN. H.H.. PAASIVUO, R., LEHTO,V.-P., LINDER,E., ALFTHAN, protein which is secreted. J. cell. Physiol., 131,402408 (1987). 0 . and VIRTANEN, I., Cellular origin and differentiation of renal WALSH-REITZ,M.M., GLUCK,S.L., WAACK, S. and TOBACK. G., carcinomas. Lab. Invest., 49,317-326 (1983). Lowering extracellular Na+ concentration releases autocrine growth MIKI,S.. IWANO,M., MIKI,Y., YAMAMOTO, M.. TANG, B., YOKOKAWA. factors from renal epithelial cells. Proc. nat. Acad. Sci. (Wash.), 83, K., SONODA.T.. HIRANO,T. and KISHIMOTO, T., Interleukin-6 (IL-6) 47644768 (1986).
