A tissue culture cytotoxicity test for large-scale cancer chemotherapy screening.

Abstract

IRVING TOPLIN (John L. Smith Memorial for Cancer Research, Chas. Pfizer & Co., Inc., Maywood, N.J,) Recent reviews of the role of tissue culture in cancer chemotherapy screening (5, 6) have empha- sized the need for a more extensive trial of cell cultures as a screening tool for the detection of anti-tumor substances. Eagle and Foley (~--4) have reported the cyto- toxic activity against human cell cultures of more than 180 selected compounds. These compounds had previously been tested against at least three experimental animal tumors. It was found that 79 per cent of the substances active against two or more experimental tumors showed cell culture ac- tivity at 10 -4 gm/ml or less. This degree of sensi- tivity was achieved at the cost of ~1 per cent of the tumor-negative compounds also exhibiting cell culture activity at the arbitrary cut-off point of 10 --4 gm/ml or less. It was also observed that there was no regular difference in the susceptibility to the various drugs of the cell lines derived from malignant and normal tissue. Eagle and Foley concluded that a large-scale investigation of the usefulness of human cell cultures as a cancer chemotherapy screening procedure was warranted. Nitta (7-9) has tested 38 antibiotics and syn- thetics of varying anti-tumor activity against HeLa cell cultures. It was found that, in general, the strong tumor-inhibitory antibiotics were de- structive to HeLa cells at the lowest concentra- tions, while those antibiotics inactive against ani- mal tumors were least injurious to the cell cul- tures. Nitta concluded that the test method with the use of HeLa cells can be practically applied to the screening or assay of the anti-tumor activity of substances. The large-scale screening program of the Cancer Chemotherapy National Service Center (CCNSC) with experimental animal tumors affords a unique opportunity to test concomitantly the reliability * These studies were carried out under Contract SA-43-ph- 1926 between the National Cancer Institute and Chas. Pfizer & Co., Inc. Presented in part at the 50th Annual Meeting of tile American Association for Cancer Research, Atlantic City, April 1~, 1959. Received for publication May ll, 1959. and sensitivity of a cell culture screen in detecting anti-tumor agents from a large random group of test candidates. This report describes a simple, rapid, and inexpensive tissue culture cytotoxicity test used in this laboratory to supplement the ex- perimental tumors in the screening of fermenta- tion broth filtrates, natural products, and syn- thetics. Results obtained by this technic for a group of compounds of known tumor activity are also presented. Some preliminary observations are made on the applications and limitations of the test in random screening and in following the fractionation and concentration of anti-tumor agents from crude preparations. MATERIALS AND METHODS The tissue culture test used in these studies included three essential steps: 1. Dispensing of graded doses of test solution directly into disposable plastic containers by a microburet technic. ~. Addition of a HeLa or other human cell sus- pension standardized at 10,000-15,000 cells/ml in growth medium to a total volume of 1 ml/con- tainer. 3. Microscopic evaluation of cytotoxicity after 5 days' incubation by a prescribed cytotoxicity rating system. Details of the procedure were as follows: Preparation of cell suspension.--Stock cultures of the HeLa and other human cell strains were grown on the glass surface of Blake bottles in Eagle's medium (1) with 10 per cent pooled human serum and 50 units/ml penicillin and 50 ~g/rnl dihydro- streptomycin. Stock cultures were refcd with fresh medium every ~ days. A suspension of mainly single cells was obtained from a 3-5-day culture by scraping or trypsinizing 1 the cell sheets off the glass and dispersing the cells with repeated pipetting. Cells grown in fluid suspension were also used suc- cessfully. In either case, the cell suspension was diluted in fresh medium to a final concentration of 10,000-15,000 cells/ ml based on a hemocytometer cell count at about the ~00,O00 cells/ml level. The cell suspension was placed in a 500-ml. Erlenmeyer flask equipped with a suspended Teflon-coated magnet and with glass and rubber tubing connections leading to a 1-ml. Cornwall syringe (Chart 1). A 3-way adapter placed 10.05-0.1 per cent trypsin (Difco 1:~50) in Hanks' salt solution minus calcium, magnesium, or phosphate. 959