JPM Vol. 28, No. 1 August 1992:9-14

Quantitative Colorimetric Assay for Basic Fibroblast Growth Factor Using Bovine Endothelial Cells and Heparin T o m o k o Kajio, Kenji K a w a h a r a , and Koichi K a t o Biotechnology Research Laboratories, Research and Development Division, Takeda Chemical Industries, Ltd., Yodogawa-ku, Osaka 532, Japan

An accurate and reproducible colorimetric assay was established to determine the concentration of basic fibroblast growth factor (bFGF) or bFGF-like activity in culture media and biological fluids. Fetal bovine heart endothelial cell line ATCC CRL 1395 was used as the bFGF-dependent cell line. The proliferation-stimulating activity of bFGF was determined by measuring the amount of formazan formed by the mitochondrial enzymes from a tetrazolium salt, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium'bromide (MTT), instead of counting the viable cell numbers or measuring the incorporation of [3H]-thymidine. The addition of 250 ng/mL of heparin to the culture medium resulted in about a tenfold increase in the proliferation-stimulating activity of bFGF and allowed the detection of as low as 10 pg/mL of bFGF. Heparin also resulted in much smaller inter- and intraassay variations. The bFGF concentrations determined by this colorimetric assay correlated well with those determined by both the [3H]-thymidineincorporation assay using BALB/c 3T3 fibroblast cells (r = 0.998) and the cell number count assay (r = 0.996). This assay can be adapted to quantify bFGF or bFGF-like activity in tissue culture media and biological fluids such as plasma and organ extracts.

Keywords: Basic fibroblast growth factor; MTT; Quantitative assay; Heparin

Introduction Basic fibroblast growth factor (bFGF) is a mitogen for various types of mesoderm- and neuroectodermderived cells. It is multifunctional as evidenced by its ability to stimulate proliferation and differentiation of a wide range of responding cells (Gospodarowicz et al., 1987; Burgess and Maciag, 1989). It is also a potent angiogenic factor (Folkman and Klagsbrun, 1987). Escherichia coli-derived recombinant bFGF and its genetically engineered mutein CS23 (Seno et al., 1988) have been demonstrated to accelerate significantly the healing of chronic duodenal ulcers produced by cysteamine in rats (Szabo et al., 1991). Oral administration of recombinant human bFGF (rbFGF) CS23 caused a tenfold increase in angiogenesis in the ulcer bed. These findings demonstrate the important role of bFGF in

Address reprint requests to T. Kajio, Biotechnology Research Laboratories, Research and Development Division, Takeda Chemical Industries, Ltd., Yodogawa-ku, Osaka 532, Japan. Received January 29, 1992; revised and accepted May 5, 1992. Journal of Pharmacological and Toxicological Methods 28, 9-14 (1992) © 1992 Elsevier Science Publishing Co., Inc., 655 Avenue of the Americas, New York, NY 10010

ulcer healing and imply a novel pharmacological modulation of duodenal ulcers. bFGF activity has been measured by a cell proliferation assay with endothelial cells (Baird et al., 1986; Gospodarowicz and Cheng, 1986) or fibroblast cells (Gospodarowicz and Moran, 1974: Klagsbrun et al., 1987) as an indicator. The method using endothelial cells requires that cells be detached by trypsinization before counting the cell number and, therefore, is laborious. In the bioassay with endothelial cells, bFGF must be added every other day during the culture because bFGF is labile. This makes the assay laborious and inaccurate. When using fibroblasts, one must decrease the serum concentration to make the cells quiescent; the use of radioisotopes, such as [3H]-labeled thymidine, is also required to quantify the mitogenicity. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) has been used to develop a quantitative colorimetric assay for cell proliferation (Mosmann, 1983; Tada et al., 1986). We have established an accurate colorimetric assay (MTT assay) of bFGF that involves measuring the proliferation of endothelial

1056-8719/92/$5.00

10

JPM Vol. 28, No. 1 August 1992:9-14

cells with MTT. This assay does not require the use of hazardous radioisotopes and is simple to perform. We report here that fetal bovine heart endothelial (FBHE) cells, ATCC CRL 1395, were stimulated to replicate in the presence of bFGF, and that the addition of heparin to the assay medium greatly increased the proliferation-stimulating activity of bFGF. We also compared the amount of bFGF measured by our method with those obtained by the [3H]-thymidine incorporation assay and the cell number count assay. Materials and Methods

bFGF Recombinant human bFGF-CS23 is a mutein of human bFGF derived from E. coli in which Cys 7° and Cys 88 of the original bFGF have both been replaced with serine residues (Seno et al., 1988). Since its in vitro biological activities have been revealed to be identical to those of the original rbFGF (Seno et al., 1988), rbFGF-CS23 expressed in E. coli and purified to homogeneity was used as bFGF throughout the work presented in this paper. For comparative study, the original rbFGF was also expressed in E. coli (Iwane et al., 1987) and purified to homogeneity as previously described (Sakaguchi et al., 1988). Cell Culture F B H E cells, ATCC CRL 1395, were purchased from American Type Culture Collection at a passage number of 12. The cells were cultured in Dulbecco's modified Eagle's medium (DMEM) (Flow Laboratories, Irvine, Scotland) supplemented with bFGF (2 ng/ mL), 10% heat-inactivated fetal calf serum (FCS) (MA Bioproducts, MD, USA), 20 units/mL penicillin G, and 100 I~g/mL streptomycin under an atmosphere of 10% CO2 and at 37°C. The plastic flasks (Iwaki Glass Co., Ltd., Tokyo, Japan) were coated with 0.1% (w/v) gelatin as described (Folkman et al., 1979). The cells were grown in culture flasks and used for the assay at passage numbers less than 23 because a slight but significant autocrine growth was observed after more than 25 passages. Mouse BALB/c 3T3 fibroblast cells were cultured in DMEM supplemented with 10% heat-inactivated calf serum, 20 units/mL penicillin G, and 100 i~g/mL streptomycin in a plastic dish under an atmosphere of 5% CO2 and at 37°C.

Determination o f b F G F by M T T Assay with F B H E Cells MTT (Sigma Chemical Co., St. Louis, MO, USA) was dissolved in phosphate-buffered saline (PBS) at a concentration of 5 mg/mL, filtered to sterilize, and

stored at 4°C (Tada et al., 1986). The bFGF standard and test samples were serially diluted twofold with an assay medium (50 i~L/well), DMEM supplemented with 10% heat-inactivated FCS, 20 units/mL penicillin G, 100 txg/mL streptomycin, and 250 ng/mL of heparin (Roussel Uclaf, Paris, France) on a 96-well plate (Nunc, Roskilde, Denmark). Then, 4 × 103 F B H E cells suspended in 50 I~L of the assay medium were added to each well. After 3 days, 40 I~L of the MTT stock solution was added to each well, and the plates were incubated at 37°C for 6 hr under an atmosphere of 10% CO2. Then, 100 ~L of 10% sodium dodecyl sulfate (SDS)-0.01N HCI was added to each well to dissolve the MTT formazan formed, and the plates were incubated overnight under the same conditions described above. The absorbance at 590 n m (A59o) was measured with a multiwell scanning spectrophotometer (Titertek Multiskan, Flow Laboratories Inc., PA, USA) using distilled water as the blank. The A59o corrected for the background ( A A 5 9 o ) w a s used to calculate the bFGF concentration.

Viable Cell N u m b e r s For counting the viable cell numbers, 500 I~L of the bFGF standard and test samples in the assay medium were plated on a 24-well plate (Flow Laboratories Inc.). Then, 3 × 10 4 F B H E cells suspended in 500 ~L of the assay medium were added to each well. After 3 days, the medium was removed, the wells were trypsinized, and the number of cells were counted with a Coulter counter (Coulter Electronics, Inc., FL, USA) as described (Sakaguchi et al., 1988).

Incorporation o f [3H]-Thymidine Proliferative activity of bFGF was assayed with mouse BALB/c 3T3 ceils by [3H]-labeled thymidine incorporation assay (Gospodarowicz et al., 1974). BALB/c 3T3 cells (n = 2000) suspended in 75 ~L of DMEM supplemented with 5% FCS were added to each well of a 96-well plate, and the plates were incubated at 37°C under an atmosphere of 5% CO2. After 1 day, the medium was removed, 200 ~L of DMEM supplemented with 0.6% FCS was added, and the culture was continued for an additional 72 hr. The bFGF standard and test samples (50 t~L) diluted with DMEM supplemented with 0.1% bovine calf serum albumin were then added to the well. After 18 hr, 37 kBq of [3H]-thymidine (Amersham Japan, Tokyo, Japan) were added to each well. After an additional 6 hr of incubation, the medium was removed, and the cells were washed with PBS. Then, the cells were lysed by adding I00 i~L/well of 5% SDS, and the radioactivity incorporated was measured by using a scintillation counter.

11

KAJIO ET AL. QUANTITATIVE COLORIMETRIC ASSAY FOR bFGF USING BOVINE ENDOTHELIAL CELLS AND HEPARIN

Results 1.0

Effects of Heparin on bFGF-Dependent Growth of FBHE Cells To study the influence of heparin on the growth of FBHE cells, we cultured the cells with various concentrations of heparin ranging from 10 to 1000 ng/mL. In the presence of bFGF, the A59o increased with the heparin concentration, and it reached a plateau at a heparin concentration of 250 ng/mL. Meanwhile, in the absence of bFGF, no increase was observed at any concentrations of heparin under the experimental conditions. The optimal heparin concentration was determined to be 250 ng/mL. Heparin (250 ng/mL), however, did not restore the activity of heat- (e.g., pH 7.4, 70°C, 30 min) or acid- (e.g., pH 3.0, 37°C, 20 hr) inactivated bFGF under these conditions. The subsequent assays for bFGF activity were conducted in the presence of 250 ng/mL heparin unless otherwise noted. The kinetics of MTT formazan production was studied with different bFGF concentrations. The production of formazan gradually increased with time and reached a plateau at 8 hr.

Comparison of the Proliferation Curves with and Without Heparin The MTT assay was performed with or without heparin. As shown in Figure 1, when heparin was added, the maximal value of A59ocorrected for the background (AA59o) increased, and the concentration-response curve became steeper. In the presence of heparin, as little as 10 pg/mL of bFGF was detected, and the concentration-response curve reached a maximal value at a bFGF concentration of 1.25 ng/mL. In the absence of heparin, it reached a plateau at a bFGF concentration of 100 ng/mL. The A A59ovalues generated from twofold serial dilutions of a bFGF standard solution (2.5 ng/mL) were calculated as a percentage of maximal AA59oand plotted on a probability (probit) chart (Figure 2). The straight line crosses 50% of the maximal value on the Y axis. At this point, the concentration can be defined from the X axis dilution coordinate. This concentration was designated EDso. The EDso for bFGF with and without heparin was calculated to be 90 and 1024 pg/ mL, respectively. This shows that the proliferationstimulating activity of bFGF on FBHE cells was potentiated by approximately tenfold in the presence of heparin. bFGF concentration in a test sample was calculated based on the ratio of dilution giving 50% of the maximal AA59o of the sample and that of bFGF standard. The EDso for the original rbFGF was identical to that obtained for rbFGF-CS23 in this assay (Figure 2). The inter- and intraassay percent coefficients of

0.8

0.6 t~

0.4

0.2

0

I

I

9.8

I

I

39

I

I

156

I

J

625

bFGF CONCENTRATION

I

I

llll

2500

i

t

200000

(pg/ml)

Figure 1. Comparison of the proliferation curves with and without heparin, bFGF was added to the well on a 96-well plate followed by twofold serial dilution with the assay medium containing 250 ng/mL of heparin (0) or without heparin (©). FBHE cells were plated at a density of 4 × 103 cells/ well and cultured for 3 days. The background A59ovalue in the presence or absence of heparin was determined without bFGF, and used for calculating AA59o values. Each point represents the mean of quintuplicate determinations. The bars indicate the standard deviations.

variation were determined using a single I0 ng/mL bFGF sample followed by twofold serial dilutions. In the presence of heparin, the interassay percent coefficient of variation for the bFGF sample was 7.3%, and the intraassay percent coefficient of variation was 7.4%, whereas those obtained in the absence of heparin were 14.6% and 14.0%, respectively.

Comparison of bFGF Concentrations Determined by MTT Assay and Those Determined by [3H]-Thymidine Incorporation Assay or Cell Number Count Assay To examine the correlation between the bFGF concentrations determined by this MTT assay using FBHE cells and those determined by the [3H]-thymidine incorporation assay using BALB/c 3T3 fibroblast cells, we subjected test samples containing different amounts of bFGF to the two assay methods. The results [Figure 3(A)] indicate that there is a good correlation between the values determined by this MTT assay (X) and those

12

JPM Vo[. 28, No. 1 August 1992:9-14

T a b l e 1. R e c o v e r y

99

95 tt~ . l-