Dec 18, 1982 - Wilms' tumors from seven patients were dissociated by me chanical and .... Epithelial tumors, both primary and metastatic, attained max.
[CANCER RESEARCH 42, 5262-5264, 0008-5472/82/0042-OOOOS02.00
December 1982]
Clonogenic Assay for Wilms' Tumor: Improved Technique for Obtaining Single-Cell Suspensions and Evidence for Tumor Cell Specificity1 Lois W. Dow,2 Manoo Bhakta, and Judith Wilimas Division of Hematology-Oncology,
St. Jude Children's Research Hospital, Memphis, Tennessee 38101
ABSTRACT Wilms' tumors from seven patients were dissociated
by me
chanical and enzymatic means; this technique resulted in sin gle-cell suspensions for five specimens and a few aggregates for two. By dye exclusion, cell viability ranged from 56 to 100% (median, 92%). All seven preparations produced more than five colonies/2 x 105 cells plated. Forty-three colonies grown from cells of a glucose-6-phosphate dehydrogenase hétéro zygote were of the same glucose-6-phosphate dehydrogenase isoenzyme type as the original tumor, indicating that the assay is specific for tumor cells. We attribute the high rate of colony formation to an improved method of cell preparation (combined mechanical and enzymatic dissociation of tumors) which may be applicable to other primary human tumors assayed in the soft agar system. INTRODUCTION Hamburger and Salmon (7) have established short-term cul tures of clonogenic cells from a variety of solid tumors, notably adult tumors such as ovarian carcinoma and multiple myeloma (5-8). The results of drug sensitivity testing with the Ham burger-Salmon assay are potentially relevant to the drug re sponsiveness of tumors in vivo (14, 19, 20), yet clinical appli cation of this human tumor stem cell assay remains limited, partly because of low colony yields. We report an improved method of cell preparation for cloning studies and the specific ity of the Hamburger-Salmon assay for Wilms' tumor cells, as determined by G-6-PD3 isoenzyme typing. MATERIALS
AND METHODS
Laboratories, Detroit, Mich.) for 10 min at 37°, after which an equal volume of McCoy's Medium 5A with 20% FCS was added, and the cells were washed 3 times with McCoy's Medium 5A with 10% FCS. Suspensions were then reinspected for clumps. Cells were counted with a hemocytometer and viability was determined by erythrosin B dye exclusion. Cell Culture. Cells were suspended at a concentration of 2 x 105 cells/ml
in enriched
CMRL-1066
tissue culture
medium containing
0.3% agar, as described by Hamburger and Salmon (7). One ml of this suspension was plated in quadruplicate in 35-mm Petri dishes over a 0.5% agar layer containing enriched McCoy's medium, with or without conditioned medium from spleen cells of oil-primed BALB/c mice. Cultures were incubated at 37°in a humidified (6% CO2) incubator and observed at 2-day intervals from Day 1 to Day 21. Groups of 30 cells or more were considered colonies. Colony Characterization. Colonies were characterized morpholog ically by Papanicolaou staining of the intact soft agar layer as described by Salmon and Buick (13). Nonspecific esterase and periodic acidSchiff staining was performed on cells of colonies transferred to slides with a fine capillary pipet. Mitotic figures within colonies were demon strated by the method of Trent (18). G-6-PD Isoenzyme Studies. The one black girl in our study was a G-6-PD hétérozygote,which allowed us to study the G-6-PD isoenzymes of individual colonies. The isoenzyme types in skin and tumor samples were determined in Dr. Philip Fialkow's laboratory at the University of Washington in Seattle. Cells from this tumor were cultured in our laboratory. For studies on Days 6 through 19, individual colonies were plucked with a fine capillary pipet from the agar after they had been identified by inverted microscopy. Each colony was applied to a cellulose acetate strip that had been soaked in electrophoretic buffer containing NADP* (17). RBC from a known G-6-PD hétérozygotewere used as a control. Colony cells were lysed by freezing and thawing on dry ice. The lysates were then subjected to electrophoresis and stained for G-6-PD activity with NADP*, glucose 6-phosphate, 3-t4,5-dimethylthiazol-2-yl}-2,5-diphenyl tetrazolium
bromide, and phenazine methosulfate
(17).
Tumor Samples. Tumor samples were obtained at surgery from 7 successive patients with either primary or metastatic Wilms' tumor; there were 4 primary epithelial tumors, 2 metastatic epithelial tumors, and one primary sarcomatous tumor. Preparation of Cell Suspensions. The samples were transported to the laboratory in McCoy's Medium 5A with 10% FCS. Tumors were mechanically dissociated with scissors in a sterile Petri dish. After all large clumps had settled at unit gravity, supernatant was collected and examined by inverted microscopy. If clumps of tumor cells were pres ent, the suspensions were passed through a series of needles (21gauge, 23-gauge, 25-gauge). Any suspensions that still contained tumor cell clumps were enzymatically dissociated. Cells were centrifuged and resuspended
in 1 ml of 0.25%
crystalline
trypsin (Difco
' Supported by Clinical Cancer Education Grant CA23944 and Cancer Center Support (CORE) Grant CA21765 from the National Cancer Institute. Institutional Grant IN-99G from the American Cancer Society, and by ALSAC. Presented in part at the 72nd Annual Meeting of the American Association for Cancer Re search, Washington, D. C., April 28, 1981
