l
0022-534 7/90/1435-1049$02.00/0 Vol. 143, May Printed in U.S.A.
THE JOURNAL OF UROLOGY Copyright© 1990 hy AMERICAN UROLOGICAL ASSOCIATION, INC.
CLONAL GROWTH REQUIREMENTS OF HUMAN BLADDER TUMOR CELL LINES HARVEY B . NIELL,* KIM WEBSTER, DANA RADEMACKER
AND
MAURIZIO BRAUSI
From the Veterans Administration Research Service, V. A. Medical Center, Memphis, Tennessee
ABSTRACT
We have serially investigated the role of nutrients and growth factors for their ability to support the clonal growth of human bladder tumor cell lines. Five established human bladder tumor cell lines were selected for study of their requirements for clonogenic growth in semisolid cultures. We identified an "optimal medium" to be Alpha MEM medium, 15% horse serum, five µg./ml. of transferrin, five µg.jml. of insulin and 30 ng./ml. of epidermal growth factor . The "optimal assay" enhanced the growth of human bladder tumor cell lines 2.5 times compared to the standard clonogenic assay. The resulting medium is less complicated to maintain, less expensive and supports the growth of human bladder tumor cell lines better than the standard clonogenic assay. (J. Ural., 143: 1049-1052, 1990) Human bladder tumor cell lines (HBTCL) serve as valuable models for studying bladder cancer in vitro. They frequently are used in studies- of basic tumor biology as well as for determining patterns of anticancer agent activity in bladder cancer. A broad range of HBTCL have been established in a variety of selective media and most are now commercially available.1• 2 HBTCL have been studied extensively as indicator cells for identifying potentially useful agents for patients with human bladder cancer. 3 - 7 The most commonly used assay for the assessment of antineoplastic drug activity has been the clonogenic assay of Hamburger and Salmon. 8 Using this assay, several laboratories have reported that the clonogenic assay is predictive of clinical response to anticancer agents in man.9 - 15 The clonogenic assay has also been shown to be a useful tool for pre-screening investigational compounds for activity in both HBTCL and human tumor explants.16• 17 The clonogenic assay is a complicated medium to prepare and includes a variety of nutrients that have been selected for the clonal growth for a wide variety of tumors. Little attention has been given to developing selective media specifically for the clonal growth of bladder tumor cells. We have serially investigated the role of nutrients and growth factors in supporting the clonal growth of HBTCL. Using a panel of five HBTCL, we have identified a medium with the potential of enhancing the clonal growth of HBTCL to a greater extent than the standard clonogenic assay. The medium is less complicated and less expensive than the standard clonogenic assay. MATERIALS AND METHODS
McCoy's 5A medium, CMRL 1066 medium, Iscove's modified Dulbecco's medium, Nutrient Mixture f-12(HAM) medium, MEM Alpha medium, Leibovitz L-15 medium and RPMI 1640 were all purchased at GIBCO Laboratories (Grand Island, New York). Bactoagar Difeo was obtained at Difeo Laboratories (Detroit, Michigan) . Horse serum and fetal calf serum were obtained at GIBCO. Sodium pyruvate, L-serine, glutamine, asparagine and ascorbic acid were obtained at GIBCO . Penicillin-Stretomycin-Fungizone mixture 100 x was obtained at Wittaker M.A. Bioproducts (Walkersville M.A.). Tryptic soybroth was purchased at Difeo Laboratories, DEAE Dextran at Pharmacia Fine Chemicals (Uppsala, Sweden), transferrin and CaCl2 at Sigma Chemical Co., regular insulin at Eli Lilly and Company (Indianapolis, Ind.) and epidermal growth factor at Accepted for publication December 15, 1989. * Requests for reprints: Dept. of Medicine, Coleman Res. Bldg., 956 Court Ave., Memphis TN 38163,
Collaborative Research Inc. The hormones 17-Beta estradiol, triodothyronine, testosterone and progesterone were all obtained at Sigma Chemical Co. Tumor cells. All the HBTCL used for the present study are commercially available and were obtained from the American Type Culture Collection (Rockville, Maryland). The cell lines used for this study were CUB-2(1), RT4(18), TCCSUP(19), T24(20), SCaBER(21), J82(22), 5637(23), UM-UC-3(24), HT1197(25) and HT-1376(25). Each cell line was maintained in monolayer culture in the recommended medium prior to tumor cloning experiments. Preparation of single cell suspensions. Tumor cells were harvested from monolayer culture by incubation for 10 minutes at 37C in 0.25% trypsin in Hank's balanced salt solution (GIBCO Laboratories). All drug studies were done on cells that were in the log growth phase (2/3d confluent on monolayer culture). Agar culture systems. The methods for the clonogenic assay have been published. 8 Briefly, the underlayer consisted of one ml. of enriched McCoy's medium in 0.5% agar with 15% fetal calf serum plated on 35 mm. scored petri dishes (LUX, Miles Laboratories, Naperville, Il). Tumor cells were suspended in the upper layer in 0.3% agar in enriched CMRL 1066 medium and 15% horse serum. BALB/c mouse conditioned medium and 2-mercaptoethanol were not used. A series of experiments were performed to sequentially test additives for their ability to support the clonal growth of HBTCL and these experiments are described in the results section. Tumor cells were usually plated at a concentration of 10 to 20,000 cells/plate. Cultures were incubated at 37C in 6% CO2 and 100% humidified atmosphere. Cultures were examined with an Olympus CK inverted microscope at X40 and XlOO. Final colony counts were made from 14 to 21 days after plating. Aggregates of 50 or more cells were scored as colonies. In all experiments six plates were run per experiment and each experiment was run twice and averaged. Experimental design for improving the culture conditions. The sequence of experiments to define the better culture conditions occured in the order presented in the results section. We used an initial panel of five tumor cells for the screening studies (CUB -2, J82, SCaBER, T24, and RT4). Three major components were investigated for their role in supporting clonal growth: the culture medium, serum and defined growth factors . Prior to this sequential study, we performed studies identifying factors in the clonogenic assay that might inhibit colony formation. In each experiment, the component that best supported clonal growth was identified and used as the control for the next set of experiments. Using this step wise method of inves-
1049
1050
NIELL AND ASSOCIATES
TABLE
1. Sequence and results of experiments eliminating elements of
the clonogenic assay Sequence and Additives
CUB-2
1. Clonogenic Assay (Control) 2. Clonogenic Assay without DEAEdextran 3. McCoy's+ 5% HS + 15% FCS + TBS/ CMRL + 15% HS + insulin + transf. 4. McCoy's + 5% HS + 15% FCS/CMRL + 15% HS 5. McCoy's + 5% HS/ CMRL + 15% HS + insulin + transf. 6. McCoy's + 5% HS/ CMRL + 15% HS
J-82
SCaBER
RT4
T24
Total 5
0.66
1.26
1.43
1.14
0.34
4.8
1.34
0.97
1.14
1.09
0.23
4.8
en
-~ C
0
0 0 1.91
1.23
1.20
1.24
0.20
5.7
1.65
1.45
1.44
1.12
1.07
6.7
1.82
1.31
0.86
0.95
0.16
5.1
All numbers represent the colonies on 6 plates in each of two separate experiments divided by the colonies from control plates. Horse serum-HS; Fetal calf serum-FCS; Transferrin-transf.
TABLE 2 .
Experiments evaluating alternate nutrien t media
Sequence and Additives 1. McCoy's + 5% HS/ CMRL + 15% HS+ insulin + transf. (controll 2. Medium substituted (both layers) RPMI 1640 3. Medium substituted (both layers) Iscove's modified Dulbecco's 4. Medium substituted (both layers) Nutrient mixture f-12 (HAM) 5. Medium substituted (both layers) Leibovitz L-15 6. Medium substituted (both layers) MEM Alpha
CUB-2
J-82
RT4
T24
1
SCaBER 1
1
1
Total 5
1.08
1.22
3.05
1.18
0.88
7.4
0.82
1.10
3.64
1.04
0.78
7.4
0.43
0.68
0.73
1.92
0.83
4.6
0.00
0.36
0.00
1.16
0.29
1.8
1.19
1.39
2.82
2.94
1.83
10.2
0.5
5
10
20
µg/ml of transferrin FIG. 2. Experiments designed to evaluate optimal level of transferrin required to support clonal growth of HBTCL. All numbers represent actual colony counts from 6 plates in each of two separate experiments. +-CUB-2 -J82 •-SCaBER D-RT4 ,.__T24.
f
70
40 en
-~ C
0
0 0
Each medium was substituted for the nutrient media used in the control culture conditions. All numbers represent the colonies on 6 plates in each of two separate experiments divided by the colonies from control plates.
20·---~ 30
l___o
10°
•
~
1---"'+-1-----------__;; D
0
1 00
90 80
0 .5
----O---o
70
20
µg/ml of insulin FIG. 3. Experiments designed to evaluate optimal level of insulin required to support clonal growth of HBTCL. All numbers represent actual colony counts from 6 plates in each of two separate experiments. + - CUB-2 - J82 • - SCaBER D- RT4 .._- T24.
TABLE 3 .
en
10
5
-o
Experiments evaluating epidermal growth factor
0022-534 7/90/1435-1049$02.00/0 Vol. 143, May Printed in U.S.A.
THE JOURNAL OF UROLOGY Copyright© 1990 hy AMERICAN UROLOGICAL ASSOCIATION, INC.
CLONAL GROWTH REQUIREMENTS OF HUMAN BLADDER TUMOR CELL LINES HARVEY B . NIELL,* KIM WEBSTER, DANA RADEMACKER
AND
MAURIZIO BRAUSI
From the Veterans Administration Research Service, V. A. Medical Center, Memphis, Tennessee
ABSTRACT
We have serially investigated the role of nutrients and growth factors for their ability to support the clonal growth of human bladder tumor cell lines. Five established human bladder tumor cell lines were selected for study of their requirements for clonogenic growth in semisolid cultures. We identified an "optimal medium" to be Alpha MEM medium, 15% horse serum, five µg./ml. of transferrin, five µg.jml. of insulin and 30 ng./ml. of epidermal growth factor . The "optimal assay" enhanced the growth of human bladder tumor cell lines 2.5 times compared to the standard clonogenic assay. The resulting medium is less complicated to maintain, less expensive and supports the growth of human bladder tumor cell lines better than the standard clonogenic assay. (J. Ural., 143: 1049-1052, 1990) Human bladder tumor cell lines (HBTCL) serve as valuable models for studying bladder cancer in vitro. They frequently are used in studies- of basic tumor biology as well as for determining patterns of anticancer agent activity in bladder cancer. A broad range of HBTCL have been established in a variety of selective media and most are now commercially available.1• 2 HBTCL have been studied extensively as indicator cells for identifying potentially useful agents for patients with human bladder cancer. 3 - 7 The most commonly used assay for the assessment of antineoplastic drug activity has been the clonogenic assay of Hamburger and Salmon. 8 Using this assay, several laboratories have reported that the clonogenic assay is predictive of clinical response to anticancer agents in man.9 - 15 The clonogenic assay has also been shown to be a useful tool for pre-screening investigational compounds for activity in both HBTCL and human tumor explants.16• 17 The clonogenic assay is a complicated medium to prepare and includes a variety of nutrients that have been selected for the clonal growth for a wide variety of tumors. Little attention has been given to developing selective media specifically for the clonal growth of bladder tumor cells. We have serially investigated the role of nutrients and growth factors in supporting the clonal growth of HBTCL. Using a panel of five HBTCL, we have identified a medium with the potential of enhancing the clonal growth of HBTCL to a greater extent than the standard clonogenic assay. The medium is less complicated and less expensive than the standard clonogenic assay. MATERIALS AND METHODS
McCoy's 5A medium, CMRL 1066 medium, Iscove's modified Dulbecco's medium, Nutrient Mixture f-12(HAM) medium, MEM Alpha medium, Leibovitz L-15 medium and RPMI 1640 were all purchased at GIBCO Laboratories (Grand Island, New York). Bactoagar Difeo was obtained at Difeo Laboratories (Detroit, Michigan) . Horse serum and fetal calf serum were obtained at GIBCO. Sodium pyruvate, L-serine, glutamine, asparagine and ascorbic acid were obtained at GIBCO . Penicillin-Stretomycin-Fungizone mixture 100 x was obtained at Wittaker M.A. Bioproducts (Walkersville M.A.). Tryptic soybroth was purchased at Difeo Laboratories, DEAE Dextran at Pharmacia Fine Chemicals (Uppsala, Sweden), transferrin and CaCl2 at Sigma Chemical Co., regular insulin at Eli Lilly and Company (Indianapolis, Ind.) and epidermal growth factor at Accepted for publication December 15, 1989. * Requests for reprints: Dept. of Medicine, Coleman Res. Bldg., 956 Court Ave., Memphis TN 38163,
Collaborative Research Inc. The hormones 17-Beta estradiol, triodothyronine, testosterone and progesterone were all obtained at Sigma Chemical Co. Tumor cells. All the HBTCL used for the present study are commercially available and were obtained from the American Type Culture Collection (Rockville, Maryland). The cell lines used for this study were CUB-2(1), RT4(18), TCCSUP(19), T24(20), SCaBER(21), J82(22), 5637(23), UM-UC-3(24), HT1197(25) and HT-1376(25). Each cell line was maintained in monolayer culture in the recommended medium prior to tumor cloning experiments. Preparation of single cell suspensions. Tumor cells were harvested from monolayer culture by incubation for 10 minutes at 37C in 0.25% trypsin in Hank's balanced salt solution (GIBCO Laboratories). All drug studies were done on cells that were in the log growth phase (2/3d confluent on monolayer culture). Agar culture systems. The methods for the clonogenic assay have been published. 8 Briefly, the underlayer consisted of one ml. of enriched McCoy's medium in 0.5% agar with 15% fetal calf serum plated on 35 mm. scored petri dishes (LUX, Miles Laboratories, Naperville, Il). Tumor cells were suspended in the upper layer in 0.3% agar in enriched CMRL 1066 medium and 15% horse serum. BALB/c mouse conditioned medium and 2-mercaptoethanol were not used. A series of experiments were performed to sequentially test additives for their ability to support the clonal growth of HBTCL and these experiments are described in the results section. Tumor cells were usually plated at a concentration of 10 to 20,000 cells/plate. Cultures were incubated at 37C in 6% CO2 and 100% humidified atmosphere. Cultures were examined with an Olympus CK inverted microscope at X40 and XlOO. Final colony counts were made from 14 to 21 days after plating. Aggregates of 50 or more cells were scored as colonies. In all experiments six plates were run per experiment and each experiment was run twice and averaged. Experimental design for improving the culture conditions. The sequence of experiments to define the better culture conditions occured in the order presented in the results section. We used an initial panel of five tumor cells for the screening studies (CUB -2, J82, SCaBER, T24, and RT4). Three major components were investigated for their role in supporting clonal growth: the culture medium, serum and defined growth factors . Prior to this sequential study, we performed studies identifying factors in the clonogenic assay that might inhibit colony formation. In each experiment, the component that best supported clonal growth was identified and used as the control for the next set of experiments. Using this step wise method of inves-
1049
1050
NIELL AND ASSOCIATES
TABLE
1. Sequence and results of experiments eliminating elements of
the clonogenic assay Sequence and Additives
CUB-2
1. Clonogenic Assay (Control) 2. Clonogenic Assay without DEAEdextran 3. McCoy's+ 5% HS + 15% FCS + TBS/ CMRL + 15% HS + insulin + transf. 4. McCoy's + 5% HS + 15% FCS/CMRL + 15% HS 5. McCoy's + 5% HS/ CMRL + 15% HS + insulin + transf. 6. McCoy's + 5% HS/ CMRL + 15% HS
J-82
SCaBER
RT4
T24
Total 5
0.66
1.26
1.43
1.14
0.34
4.8
1.34
0.97
1.14
1.09
0.23
4.8
en
-~ C
0
0 0 1.91
1.23
1.20
1.24
0.20
5.7
1.65
1.45
1.44
1.12
1.07
6.7
1.82
1.31
0.86
0.95
0.16
5.1
All numbers represent the colonies on 6 plates in each of two separate experiments divided by the colonies from control plates. Horse serum-HS; Fetal calf serum-FCS; Transferrin-transf.
TABLE 2 .
Experiments evaluating alternate nutrien t media
Sequence and Additives 1. McCoy's + 5% HS/ CMRL + 15% HS+ insulin + transf. (controll 2. Medium substituted (both layers) RPMI 1640 3. Medium substituted (both layers) Iscove's modified Dulbecco's 4. Medium substituted (both layers) Nutrient mixture f-12 (HAM) 5. Medium substituted (both layers) Leibovitz L-15 6. Medium substituted (both layers) MEM Alpha
CUB-2
J-82
RT4
T24
1
SCaBER 1
1
1
Total 5
1.08
1.22
3.05
1.18
0.88
7.4
0.82
1.10
3.64
1.04
0.78
7.4
0.43
0.68
0.73
1.92
0.83
4.6
0.00
0.36
0.00
1.16
0.29
1.8
1.19
1.39
2.82
2.94
1.83
10.2
0.5
5
10
20
µg/ml of transferrin FIG. 2. Experiments designed to evaluate optimal level of transferrin required to support clonal growth of HBTCL. All numbers represent actual colony counts from 6 plates in each of two separate experiments. +-CUB-2 -J82 •-SCaBER D-RT4 ,.__T24.
f
70
40 en
-~ C
0
0 0
Each medium was substituted for the nutrient media used in the control culture conditions. All numbers represent the colonies on 6 plates in each of two separate experiments divided by the colonies from control plates.
20·---~ 30
l___o
10°
•
~
1---"'+-1-----------__;; D
0
1 00
90 80
0 .5
----O---o
70
20
µg/ml of insulin FIG. 3. Experiments designed to evaluate optimal level of insulin required to support clonal growth of HBTCL. All numbers represent actual colony counts from 6 plates in each of two separate experiments. + - CUB-2 - J82 • - SCaBER D- RT4 .._- T24.
TABLE 3 .
en
10
5
-o
Experiments evaluating epidermal growth factor
