0022-534 7/92/1482-0441$03.00/0 Vol. 148, 441-445, August 1992
THE JOURNAL OF UROLOGY
Copyright© 1992 by AMERICAN UROLOGICAL ASSOCIATION, lNC.
Printed in US.A.
ESTABLISHMENT AND CHARACTERIZATION OF DOXORUBICINRESISTANT HUMAN BLADDER CANCER CELL LINE, KK47/ADM KOHICHI KIMIYA, SEIJI NAITO,* TSUKASA SOEJIMA, NAOTAKA SAKAMOTO, SHUJI KOTOH, JOICHI KUMAZAWA AND TAKASHI TSURUO From the Department of Urology, Faculty of Medicine, Kyushu University, Fukuoka, and Institute of Applied Microbiology, University of Tokyo, Tokyo, Japan
ABSTRACT
A human bladder cancer cell line resistant to doxorubicin, KK47/ADM has been established in vitro by exposing KK4 7 parent cells to progressively higher concentrations of the drug over a period of 16 months. The KK47/ADM was 271 times more resistant to doxorubicin than the KK47 parent. The KK4 7/ ADM exhibited cross-resistance to doxorubicin derivatives (pirarubicin, epirubicin), vinca alkaloids (vinblastine, vincristine) and etoposide, but not to cisplatin, carboplatin, mitomycin C, peplomycin and methotraxate. Unlike the KK4 7 parent, about 70% of the KK4 7/ ADM cells showed a positive reaction with monoclonal antibody against P-glycoprotein, MRK16. Uptake and efflux studies with [14 C]doxorubicin indicated that the resistance exhibited by the KK4 7/ ADM line was mainly due to a lower accumulation of the drug caused by an increased active efflux, and these were reversed in the presence of verapamil. Although verapamil enhanced doxorubicin sensitivity of KK4 7/ ADM, a complete overcoming of the resistance could not be obtained. These two lines with different chemosensitivity are thus considered to be a useful model for developing new chemotherapeutic strategies against multidrug resistant bladder cancer, KEY WORDS: bladder neoplasms, doxorubicin, chemotherapy
Chemotherapy has been widely used for the management of bladder cancer. Some combination chemotherapies, such as MVAC,1 are effective for advanced bladder cancer, Intravesical instillation chemotherapy is also useful for preventing recurrence of superficial bladder cancer. 2 However, the responses in advanced bladder cancer are often of relatively short duration and the recurrences of superficial bladder cancer are still frequent in spite of intravesical chemotherapy. One of the most significant problems in such chemotherapy is the emergence of drug resistance, which often leads to therapeutic failure. It is important to clarify the mechanisms of drug resistance and develop a means to overcome such resistance. To date, many drug resistant tumor cell lines have been established in vitro as models for investigating drug resistance. 3- 7 However, most of these tumor lines have been either leukemias or sarcomas. 3-6 There is little information in the literature regarding drug resistant human bladder cancer, 7 We have developed a doxombicin (ADM)-resistant human bladder cancer cell line, KK 47/ADM and characterized it in terms of morphology, growth kinetics, cross-resistance to other anticancerous agents, pharmacokinetics of ADM, expression of P-glycoprotein which is closely related to the multidmg resistance (MDR) phenotype, 8- 10 and a reverse of ADM resistance by one of the calcium channel blockers, verapamil. MATERIALS AND METHODS
Cell culture and establishment of KK47 cells resistant to ADM, The KK47 used as a parent line was established from transitional cell carcinoma of the bladder (grade 1, pTa) obtained by a transurethral resection from a 50 year old male in the Department of Urology, Kanazawa University, Japan 11 and kindly provided by Dr. H. Hisazumi. The patient had undergone neither systemic nor intravesical chemotherapy prior to the surgery. The cells were grown as a monolayer culture in complete MEM (Eagle's minimum essential medium supplemented with 10% fetal calf serum, sodium pyruvate, non-essential Accepted for publication February 10, 1992. * Requests for reprints: Department of Urology, Kyushu University, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812, Japan.
441
amino acids, L-glutamine and two-fold vitamine solution) (Gibco) at 37C in a humidified atmosphere of 5% CO 2 and subcultured usually once a week with 0,25% trypsin-0.02% EDTA solution. At first, the KK47 cells at more than 100 passages were frozen and stored in liquid nitrogen as the parent. The resistant subline, KK47/ADM was established by continual exposure of the KK47 parent cells to the culture media containing ADM over a 16-month period. Initially, KK47 cells were exposed to ADM at a concentration of 1.7 x 10-3 µg./ml., which represented the concentration of the drug needed to inhibit cell growth by 50% (IC50). Cells were maintained at this drug concentration over a three month period until their growth almost approached that of the untreated KK4 7 parent cells. The concentration of ADM was gradually increased until the concentration finally reached 3x10- 1 µg./mL The cells were kept for three months at this concentration of ADM. Then, the cells were expanded in an ADM free medium, and stored in liquid nitrogen. This required about 50 passages by this point, For the following experiments, the cells from both KK4 7 parent and KK4 7/ ADM lines were thawed, subcultured twice and then used, The KK 4 7/ADM cells were subcultured usually once a week as well as the parent KK47 cells. All the experiments were performed within 6 passages after thawing the cells, A K562 human myelogenous leukemia cell line and its ADMresistant subline, K562/ ADM established by one of the authors (Tsuruo, T.), 6 were maintained in suspension in plastic dishes in RPMI 1640 (Gibco) supplemented with 10% fetal calf serum. Drugs. The anticancerous agents used were ADM, pirarubicin, epirubicin, cisplatin, carboplatin, vinblastine, vincristine, peplomycin, mitomycin C and methotrexate, obtained from commercial sources, and either dissolved or diluted immediately before use except for ADM, which was stored at -BOC in stock solutions of 1 mg./ml. for periods of up to 2 weeks. Working concentrations were prepared in complete MEM; control medium contained equal volumes of drug-free diluent. Verapamil was purchased from Eisai Co., Ltd. [14 C]ADM (95 µCi/mg.) was purchased from Amersham Japan Ltd., Tokyo. Growth kinetics. Tumor cells (5 X 104 ) from the different lines were seeded into 50-mm. tissue culture dishes. Every 24
442
KIMIYA AND ASSOCIATES
hours, cells from triplicate dishes were harvested with 0.25% trypsin-0.02% EDTA solution, and the cell number was determined with Coulter counter. A growth curve was constructed and cell doubling time was calculated. Drug sensitivity. Experiments on the effects of the drugs were performed according to the manner described by Tsuruo et al. 12 Cells in the logarithmic phase of monolayer growth were harvested from the different lines with 0.25% trypsin-0.2% EDTA solution and tumor cell suspensions were adjusted to a concentration of 3 x 104 /ml. of complete MEM. Portions (100 ,ul.), containing 3 X 103 cells, were seeded to the wells of 96-well microtiter plates. The plates were incubated for 24 hours at 37C in air with 5% CO2 and then 100 ,ul. of varying concentrations of anticancerous agents was added to each well. For examining the effect of verapamil on .ADM . sensitivity, 100 µl. of varying concentrations of ADM plus verapamil was added, as well as ADM alone. Triplicate wells were used for each drug concentration. The cultures were then incubated for a further 72 hours in the aforementioned condition. After this exposure, cells were harvested from each well with 0.25% trypsin-0.02% EDT A solution and counted with a Coulter counter. The results were expressed as a percentage of the controls. IC50s of drugs were determined by plotting the logarithm of the drug concentration versus the growth rate (percentage of control) of the treated cells. The assays were repeated at least twice for each drug. Immunohistochemical determination of P-glycoprotein expression. The expression of P-glycoprotein of cells from different lines was examined immunohistochemically using a monoclonal antibody MRK16, preparation and characterization of which have been previously described in detail. 10 In brief, cells were suspended in a phosphate buffered saline solution (PBS) and cytospun resulting in a monolayer cell smear on the slides. After air drying, the cells were fixed with 4 % paraformaldehyde in PBS for 30 min at 23C. Smears were washed with PBS and preincubated in 10% normal goat serum for 10 minutes. Then, the primary monoclonal antibody, MRK16 (10 ,ug./ml.) was applied overnight at 4C. The dilutions and conditions for the subsequent peroxidase irnmunohistochemical steps have been described previously. 13 K562 and its ADM-resistant subline, K562/ADM were used as negative and positive control, respectively.13· 14 ADM uptake and efflux studies. This experiment was performed basically according to the method of Giavazzi et al. 4 Tumor cell suspensions from the different lines were prepared as described above and adjusted in complete MEM to give 2.5 X 105 viable cells/ml. Two hundred-,ul. aliquots of these cell suspensions were seeded to the wells of 96-well microtiter plate and the plate was incubated overnight at 37 in air with 5% CO2. For uptake assays, the culture supernatant was aspirated off; 100 ,ul. aliquots of [14 C]ADM (2 ,ug./ml. in complete MEM) in either the absence or presence of verapamil (10 ,ug./ml.) were added to the wells; and plates were incubated at 37C. At the appropriate time points after adding the [14 CJADM, the drug solution was aspirated off, the cells were washed once with PBS free of calcium and magnesium [PBS(-)), and the remaining adherent cells were lysed with 100 ,ul. of 0.2% sodium dodecyl sulfate. The lysates were harvested and added to 5 ml. of scintillation fluid in glass vials, and the radioactivity was monitored in a Beckman ,6-scintillation counter. The cpm obtained by loading the cells with [14 C]ADM for only 5 seconds was quite low and almost the same as that obtained by loading the cells for 10 seconds, 30 seconds or one minute regardless of the lines and the absence or presence of verapamil. Therefore, we considered that this value could be used as the background, and this value was subtracted from all subsequent values. To determine the cell number, cells from 3 wells of each cell line were harvested with 0.25% trypsin-0.02% EDTA solution and
counted with a Coulter counter. The results were expressed as cpm/10 6 cells. For the efflux assays, the culture supernatants were aspirated off, and 100 ,ul. aliquots of [14 C]ADM (2 ,ug/ml in complete MEM) were added to each culture, and the cells were incubated for 45 minutes at 37C. In KK4 7/ ADM, verapamil was also added at a concentration of 10 ,ug./ml. to obtain a sufficiently high intracellular concentration of ADM. Following this incubation, the drug solution was aspirated off; the cells were washed once with PBS(-); and 100 ,ul. aliquots of complete MEM in either the absence or presence of verapamil (10 ,ug./ ml.) were added to the wells. At the time indicated in figure 6, the medium was aspirated off. Then the adherent cells were lysed and harvested as described for the uptake studies and assaved for the retained radioactivitv. The results were expres~ed in terms of the percentage of radioactive drug retained; 100% values were obtained by allowing the medium to remain for only 5 seconds. RESULTS
Characteristics of KK47/ADM. The growth inhibition curves for both KK47 and KK47/ADM cells treated with different concentrations of ADM are shown in figure 1. The IC50 of KK4 7/ ADM was 4.6 x 10-1 ,ug./ml. and the relative resistance, which was expressed as a ratio of the IC50 concentration for the KK47/ADM versus KK47 cells, was 271. The KK47/ADM cells maintained almost the same degree of resistance to ADM for at least 4 weeks in the drug-free medium. The KK4 7/ ADM line was morphologically distinguishable from the KK47 parent line. The resistant cells were slightly expanded in size and weak in potential of cell to cell attachment as compared to KK47 parent cells (figure 2). Consequently the KK47/ADM line was easily converted to a single cell suspension by trypsinization. When cultured in the drug-free medium, KK4 7/ ADM cells showed a somewhat slower growth rate than
100
,--._
....0
+-'
C
0 (.)
'+-
0
~ "--'
50
..c +-'
3: ....0O')
THE JOURNAL OF UROLOGY
Copyright© 1992 by AMERICAN UROLOGICAL ASSOCIATION, lNC.
Printed in US.A.
ESTABLISHMENT AND CHARACTERIZATION OF DOXORUBICINRESISTANT HUMAN BLADDER CANCER CELL LINE, KK47/ADM KOHICHI KIMIYA, SEIJI NAITO,* TSUKASA SOEJIMA, NAOTAKA SAKAMOTO, SHUJI KOTOH, JOICHI KUMAZAWA AND TAKASHI TSURUO From the Department of Urology, Faculty of Medicine, Kyushu University, Fukuoka, and Institute of Applied Microbiology, University of Tokyo, Tokyo, Japan
ABSTRACT
A human bladder cancer cell line resistant to doxorubicin, KK47/ADM has been established in vitro by exposing KK4 7 parent cells to progressively higher concentrations of the drug over a period of 16 months. The KK47/ADM was 271 times more resistant to doxorubicin than the KK47 parent. The KK4 7/ ADM exhibited cross-resistance to doxorubicin derivatives (pirarubicin, epirubicin), vinca alkaloids (vinblastine, vincristine) and etoposide, but not to cisplatin, carboplatin, mitomycin C, peplomycin and methotraxate. Unlike the KK4 7 parent, about 70% of the KK4 7/ ADM cells showed a positive reaction with monoclonal antibody against P-glycoprotein, MRK16. Uptake and efflux studies with [14 C]doxorubicin indicated that the resistance exhibited by the KK4 7/ ADM line was mainly due to a lower accumulation of the drug caused by an increased active efflux, and these were reversed in the presence of verapamil. Although verapamil enhanced doxorubicin sensitivity of KK4 7/ ADM, a complete overcoming of the resistance could not be obtained. These two lines with different chemosensitivity are thus considered to be a useful model for developing new chemotherapeutic strategies against multidrug resistant bladder cancer, KEY WORDS: bladder neoplasms, doxorubicin, chemotherapy
Chemotherapy has been widely used for the management of bladder cancer. Some combination chemotherapies, such as MVAC,1 are effective for advanced bladder cancer, Intravesical instillation chemotherapy is also useful for preventing recurrence of superficial bladder cancer. 2 However, the responses in advanced bladder cancer are often of relatively short duration and the recurrences of superficial bladder cancer are still frequent in spite of intravesical chemotherapy. One of the most significant problems in such chemotherapy is the emergence of drug resistance, which often leads to therapeutic failure. It is important to clarify the mechanisms of drug resistance and develop a means to overcome such resistance. To date, many drug resistant tumor cell lines have been established in vitro as models for investigating drug resistance. 3- 7 However, most of these tumor lines have been either leukemias or sarcomas. 3-6 There is little information in the literature regarding drug resistant human bladder cancer, 7 We have developed a doxombicin (ADM)-resistant human bladder cancer cell line, KK 47/ADM and characterized it in terms of morphology, growth kinetics, cross-resistance to other anticancerous agents, pharmacokinetics of ADM, expression of P-glycoprotein which is closely related to the multidmg resistance (MDR) phenotype, 8- 10 and a reverse of ADM resistance by one of the calcium channel blockers, verapamil. MATERIALS AND METHODS
Cell culture and establishment of KK47 cells resistant to ADM, The KK47 used as a parent line was established from transitional cell carcinoma of the bladder (grade 1, pTa) obtained by a transurethral resection from a 50 year old male in the Department of Urology, Kanazawa University, Japan 11 and kindly provided by Dr. H. Hisazumi. The patient had undergone neither systemic nor intravesical chemotherapy prior to the surgery. The cells were grown as a monolayer culture in complete MEM (Eagle's minimum essential medium supplemented with 10% fetal calf serum, sodium pyruvate, non-essential Accepted for publication February 10, 1992. * Requests for reprints: Department of Urology, Kyushu University, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812, Japan.
441
amino acids, L-glutamine and two-fold vitamine solution) (Gibco) at 37C in a humidified atmosphere of 5% CO 2 and subcultured usually once a week with 0,25% trypsin-0.02% EDTA solution. At first, the KK47 cells at more than 100 passages were frozen and stored in liquid nitrogen as the parent. The resistant subline, KK47/ADM was established by continual exposure of the KK47 parent cells to the culture media containing ADM over a 16-month period. Initially, KK47 cells were exposed to ADM at a concentration of 1.7 x 10-3 µg./ml., which represented the concentration of the drug needed to inhibit cell growth by 50% (IC50). Cells were maintained at this drug concentration over a three month period until their growth almost approached that of the untreated KK4 7 parent cells. The concentration of ADM was gradually increased until the concentration finally reached 3x10- 1 µg./mL The cells were kept for three months at this concentration of ADM. Then, the cells were expanded in an ADM free medium, and stored in liquid nitrogen. This required about 50 passages by this point, For the following experiments, the cells from both KK4 7 parent and KK4 7/ ADM lines were thawed, subcultured twice and then used, The KK 4 7/ADM cells were subcultured usually once a week as well as the parent KK47 cells. All the experiments were performed within 6 passages after thawing the cells, A K562 human myelogenous leukemia cell line and its ADMresistant subline, K562/ ADM established by one of the authors (Tsuruo, T.), 6 were maintained in suspension in plastic dishes in RPMI 1640 (Gibco) supplemented with 10% fetal calf serum. Drugs. The anticancerous agents used were ADM, pirarubicin, epirubicin, cisplatin, carboplatin, vinblastine, vincristine, peplomycin, mitomycin C and methotrexate, obtained from commercial sources, and either dissolved or diluted immediately before use except for ADM, which was stored at -BOC in stock solutions of 1 mg./ml. for periods of up to 2 weeks. Working concentrations were prepared in complete MEM; control medium contained equal volumes of drug-free diluent. Verapamil was purchased from Eisai Co., Ltd. [14 C]ADM (95 µCi/mg.) was purchased from Amersham Japan Ltd., Tokyo. Growth kinetics. Tumor cells (5 X 104 ) from the different lines were seeded into 50-mm. tissue culture dishes. Every 24
442
KIMIYA AND ASSOCIATES
hours, cells from triplicate dishes were harvested with 0.25% trypsin-0.02% EDTA solution, and the cell number was determined with Coulter counter. A growth curve was constructed and cell doubling time was calculated. Drug sensitivity. Experiments on the effects of the drugs were performed according to the manner described by Tsuruo et al. 12 Cells in the logarithmic phase of monolayer growth were harvested from the different lines with 0.25% trypsin-0.2% EDTA solution and tumor cell suspensions were adjusted to a concentration of 3 x 104 /ml. of complete MEM. Portions (100 ,ul.), containing 3 X 103 cells, were seeded to the wells of 96-well microtiter plates. The plates were incubated for 24 hours at 37C in air with 5% CO2 and then 100 ,ul. of varying concentrations of anticancerous agents was added to each well. For examining the effect of verapamil on .ADM . sensitivity, 100 µl. of varying concentrations of ADM plus verapamil was added, as well as ADM alone. Triplicate wells were used for each drug concentration. The cultures were then incubated for a further 72 hours in the aforementioned condition. After this exposure, cells were harvested from each well with 0.25% trypsin-0.02% EDT A solution and counted with a Coulter counter. The results were expressed as a percentage of the controls. IC50s of drugs were determined by plotting the logarithm of the drug concentration versus the growth rate (percentage of control) of the treated cells. The assays were repeated at least twice for each drug. Immunohistochemical determination of P-glycoprotein expression. The expression of P-glycoprotein of cells from different lines was examined immunohistochemically using a monoclonal antibody MRK16, preparation and characterization of which have been previously described in detail. 10 In brief, cells were suspended in a phosphate buffered saline solution (PBS) and cytospun resulting in a monolayer cell smear on the slides. After air drying, the cells were fixed with 4 % paraformaldehyde in PBS for 30 min at 23C. Smears were washed with PBS and preincubated in 10% normal goat serum for 10 minutes. Then, the primary monoclonal antibody, MRK16 (10 ,ug./ml.) was applied overnight at 4C. The dilutions and conditions for the subsequent peroxidase irnmunohistochemical steps have been described previously. 13 K562 and its ADM-resistant subline, K562/ADM were used as negative and positive control, respectively.13· 14 ADM uptake and efflux studies. This experiment was performed basically according to the method of Giavazzi et al. 4 Tumor cell suspensions from the different lines were prepared as described above and adjusted in complete MEM to give 2.5 X 105 viable cells/ml. Two hundred-,ul. aliquots of these cell suspensions were seeded to the wells of 96-well microtiter plate and the plate was incubated overnight at 37 in air with 5% CO2. For uptake assays, the culture supernatant was aspirated off; 100 ,ul. aliquots of [14 C]ADM (2 ,ug./ml. in complete MEM) in either the absence or presence of verapamil (10 ,ug./ml.) were added to the wells; and plates were incubated at 37C. At the appropriate time points after adding the [14 CJADM, the drug solution was aspirated off, the cells were washed once with PBS free of calcium and magnesium [PBS(-)), and the remaining adherent cells were lysed with 100 ,ul. of 0.2% sodium dodecyl sulfate. The lysates were harvested and added to 5 ml. of scintillation fluid in glass vials, and the radioactivity was monitored in a Beckman ,6-scintillation counter. The cpm obtained by loading the cells with [14 C]ADM for only 5 seconds was quite low and almost the same as that obtained by loading the cells for 10 seconds, 30 seconds or one minute regardless of the lines and the absence or presence of verapamil. Therefore, we considered that this value could be used as the background, and this value was subtracted from all subsequent values. To determine the cell number, cells from 3 wells of each cell line were harvested with 0.25% trypsin-0.02% EDTA solution and
counted with a Coulter counter. The results were expressed as cpm/10 6 cells. For the efflux assays, the culture supernatants were aspirated off, and 100 ,ul. aliquots of [14 C]ADM (2 ,ug/ml in complete MEM) were added to each culture, and the cells were incubated for 45 minutes at 37C. In KK4 7/ ADM, verapamil was also added at a concentration of 10 ,ug./ml. to obtain a sufficiently high intracellular concentration of ADM. Following this incubation, the drug solution was aspirated off; the cells were washed once with PBS(-); and 100 ,ul. aliquots of complete MEM in either the absence or presence of verapamil (10 ,ug./ ml.) were added to the wells. At the time indicated in figure 6, the medium was aspirated off. Then the adherent cells were lysed and harvested as described for the uptake studies and assaved for the retained radioactivitv. The results were expres~ed in terms of the percentage of radioactive drug retained; 100% values were obtained by allowing the medium to remain for only 5 seconds. RESULTS
Characteristics of KK47/ADM. The growth inhibition curves for both KK47 and KK47/ADM cells treated with different concentrations of ADM are shown in figure 1. The IC50 of KK4 7/ ADM was 4.6 x 10-1 ,ug./ml. and the relative resistance, which was expressed as a ratio of the IC50 concentration for the KK47/ADM versus KK47 cells, was 271. The KK47/ADM cells maintained almost the same degree of resistance to ADM for at least 4 weeks in the drug-free medium. The KK4 7/ ADM line was morphologically distinguishable from the KK47 parent line. The resistant cells were slightly expanded in size and weak in potential of cell to cell attachment as compared to KK47 parent cells (figure 2). Consequently the KK47/ADM line was easily converted to a single cell suspension by trypsinization. When cultured in the drug-free medium, KK4 7/ ADM cells showed a somewhat slower growth rate than
100
,--._
....0
+-'
C
0 (.)
'+-
0
~ "--'
50
..c +-'
3: ....0O')
