Growth of human small cell (oat cell) carcinoma of the lung in serum-free growth factor-supplemented medium.

A cell culture line of human small cell (oat cell) carcinoma of the lung (NCI-H69) was studied for growth factor requirements for replication in serum-free medium. By a series of addition experiments, the combination of selenium (3x10~8 M), hydrocortisone (1CT8 M), insulin (5 fig/ml), transferrin (100 fig/ml), and 17ß-estradiol (10~8 M) (SHITE) added to Roswell Park Memorial Institute Medium 1640 without serum was found to allow optimum replication when cells were transferred from serum-containing medium. NCI-H69 cells have replicated con tinuously in SHITE-supplemented Roswell Park Memorial Insti tute Medium 1640 for periods of more than 12 months. The cells replicate with approximately the same doubling times as in medium supplemented with 10% fetal calf serum, but they exhibited a lower saturation density and longer lag phase in SHITE-supplemented medium compared to serum-supple mented medium. When individual components of SHITE were deleted, immediate and dramatic differences in growth were seen only with deletion of transferrin while minimal decreases were noted with deletion of the other factors. After 6 months of growth in SHITE, deletion of insulin or transferrin from the formula resulted in cessation of cell growth in serum-free medium while deletion of hydrocortisone or estradiol (alone or together) did not. A variety of other hormones and growth factors did not add significantly to SHITE in promoting repli cation. In addition, a series of hormones was found to inhibit replication or saturation density achieved in serum-free, hor mone-supplemented medium including epidermal growth fac tor, luteinizing hormone-releasing factor, parathyroid hormone, blood meal, and the tripeptide glycyl-L-histidyl-L-lysine acetate. Six other small cell carcinoma lines were tested and also were able to replicate in SHITE-supplemented serum-free medium indicating the generality of this formula for small cell carcinoma, NCI-H69 cells cultured for 4 months in SHITE medium retained their amine precursor uptake and decarboxylation properties, characteristic small cell carcinoma histology by light micros copy, dense core (neurosecretory) granules by electron mi croscopy, and tumorigenicity in nude mice. The identification of this growth formula should allow selection of tumor cells directly from patient samples, study of hormone production and regulation by these cell lines, and identification of new growth and therapeutic agents, and it should provide a way of selecting for normal amine handling cells related histogenetically to small cell carcinoma. INTRODUCTION Small cell carcinoma of the lung represents 20 to 25% of lung cancer occurring in the United States and has distinct clinical, biological, and morphological properties (10, 12). This tumor type, in contrast to other lung cancers, responds dra matically to chemotherapy and radiotherapy and in fact may be cured by these modalities when in a surgically unresectable state (23). In addition, it is often associated with paraneoplastic syndromes and the production of a variety of hormones includ ing ACTH,2 AVP, and calcitonin (27, 29). Small-cell carcinomas are believed to be derived from cells of the diffuse neuroen docrine system portion of the APUD (24, 25, 31). In trying to develop new methods of treating this tumor, we have derived continuously replicating, clonable cell lines and characterized them as small cell carcinomas (17). The identification of hor mones and growth factors required for the replication of such cells would be of potential therapeutic use and of interest in the study of the biology of the APUD system. We have followed the lead of Hayashi and Sato (20) and studied the replication of our small cell carcinoma of the lung cell lines in serum-free media supplemented with a variety of growth factors and hor mones. In this paper, we report that the combination SHITE will permit the continuous replication for over 12 months of small cell carcinoma of the lung in vitro, as well as the continued expression of differentiated APUD functions. MATERIALS AND METHODS Cell Lines. The human small cell carcinoma of the lung tissue culture lines were adapted to culture by us either directly from our patients (H designation, e.g., NCI-H69) or from transplantable nude-mouse tumors (N designation, e.g., NCI-N230). The details of their establishment and characterization are reported elsewhere (17). All of the lines came from patients with a histological and clinical diagnosis of small cell carcinoma of the lung, were tumorigenic in nude mice, had human enzymes on starch gel electrophoresis (19), had high specific activities of L-aromatic amino acid decarboxylase [L-dopa decarboxylase (EC 4.1.1.2.8)], demonstrated formaldehyde-induced fluores cence, showed neurosecretory granules by electron micros copy, and had histology and cytology typical of small cell carcinoma. They were continuously growing cell lines that ' To whom requests for reprints should be addressed, at National Cancer Institute-Veterans Administration Medical Oncology Branch, Veterans Adminis tration Medical Center. 50 Irving Street. N. W . Washington, D. C. 20422. Received May 1. 1980; accepted July 10. 1980. ' The abbreviations used are: ACTH, adrenocorticotrophin; AVP, arginine vasopressin; APUD, amine precursor uptake and decarboxylation series; SHITE, selenium:hydrocortisone:insulin:transferrin:1 7/<-estradiol; RPMI, Roswell Park Memorial Institute; FCS. fetal calf serum; T3, triidothyronine; LHRF, luteinizing hormone-releasing factor; PTH. parathyroid hormone; TRH, thyrotropin-releasing hormone; EGF, epidermal growth factor; NGF. nerve growth factor; IT. insulin: transferrin; HIT, hydrocortisone:insulin:transferrin; ITE. insulin:transferrin:17/iestradiol; HITE, hydrocortisone:insulin:transferrin:17/j-estradiol. 4356 CANCER RESEARCH VOL. 40 on April 14, 2017. © 1980 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from Defined Medium for Small Cell Carcinoma could be cloned in microwells and in 0.3% agarose. Stock cultures of NCI-H69 were typical of these, being a floating line which grows in balls, clumps, and chains in a suspension culture. They were maintained in RPMI Medium 1640 supple mented with 10% (v/v) PCS in 75-sq cm Falcon flasks. It was derived from the malignant pleural effusion of a male patient and required conditioned medium from another small cell car cinoma line (NCI-H45) for its establishment. After 6 months in conditioned medium, it could be shifted to unconditioned me dium and were maintained in this prior to the current study. The subline of NCI-H69 used in the current study was free of mycoplasma contamination by bacteriological tests done at Microbiologie Associates, Bethesda, Md. and was aneuploid by flow microfluorometric analysis with a DMA index of 1.75 (compared to a diploid human lymphocyte standard) by Dr. P. Bunn of our laboratory. RPMI Medium 1640 and modified F12 medium (called F14 medium) (13) were prepared from powder obtained from Grand Island Biological Co., and liquid RPMI Medium 1640 from Grand Island Biological Co. was also used. Water for tissue culture medium was prepared using a Millipore "Super-Q" system. Falcon and Costar plastic ware was used, and cultures were maintained in a water-jacketed incubator with 100% humidity and 95% air:5% CO2 atmosphere. Growth Factor and Hormones. Growth factors and hor mones were obtained from the following sources. From Sigma Chemical Co., St. Louis, Mo. were: ACTH, porcine, Grade 1 (Lot 37C-0138); 17/S-estradiol (Lot 178-0287); follicle-stimu lating hormone, porcine (Lot 106C-0244); crystalline glucagon, bovine and porcine pancreas (Lot 77C-0320); hydrocortisone (Lot 71C-0020); insulin, bovine pancreas crystalline (Lots 121C-1350 and 66C-0395, 26.4 lU/mg); progesterone (Lot 95C-0320); prolactin (luteotrophic hormone) from sheep pitui tary (Lot 127C-0182); putrescine dihydrochloride (Lot P 7505); T3 (Lot 95C-0388); testosterone acetate (Lot 73C-1210); transferrin, human over 90% iron free (Lot 96C-0107); and Pitressin. From Calbiochem-Behring Corp., La Jolla, Calif, were: calcitonin, porcine (Lot 720117); glycyl-L-histidyl-L-lysine acetate, Grade A, (Lot 600700); luteinizing hormone, equine, Grade B (Lot 602250, 1.26 Armour units/mg); LHRF, synthetic (Lot 760003); PTH, bovine (Lot 700667); somatostatin, synthetic (Lot 628094); and TRH, synthetic (Lot 600924). From Collaborative Research, Waltham, Mass, were: EGF, mouse submaxillary gland (Lot 794-5); fibroblast growth factor, bovine pituitary gland (Lot 794-67D); multiplication-stimulating activity, rat liver cell (Lot 734-27); and NGF, mouse submaxil lary gland (Lot 794-70). From other sources, we had gifts of: growth hormone, human (National Pituitary Agency, distributed by NIAMDD, Lot HS2160E); thyrotropin, bovine (Miles Labo ratories, Elkhart, Ind., via Dr. L. Kohn, NIH Lot 33); and blood meal, bovine, crude extract (Dr. G. Sato). Stock solutions of the hormones and growth factors were prepared at a concen tration of 100 to 2500 times the final concentration. The procedure of preparing these hormone concentrates varies with each compound. The steroid hormones were dissolved in ethanol and then diluted in 0.9% NaCI solution while glucagon, T3, and insulin required adjustments of the pH to go into solution. The highly concentrated stock solutions were then filter sterilized (0.2-/iM Nalgene filter) and stored at -20° in 1to 3-ml aliquots in glass vials. ACTH and insulin were stored at 4°.For each experiment, the appropriate factors were thawed and diluted to 100 times their final concentration in serum-free medium, and then, appropriate volumes of the 100-fold con centrates were mixed together to make the various experimen tal combinations at a final 2-fold concentration. One ml of the 2-fold-concentrated medium was then plated into each well of a Costar 24-well plate according to the experimental design, and the plates were placed in a tissue culture incubator to equilibrate temperature and pH prior to addition of 1 ml of cell suspension. Growth Curve Experiments. Growth curves were performed in duplicate in 24-well (16 mm diameter) cluster plates. A cell suspension was prepared by collecting cells by centrifugation from stock medium containing cells growing in suspension in active growth phase. The cells were washed twice with serumfree medium (either RPMI Medium 1640 or F14) and sus pended by trituration in a 10-ml Falcon tissue culture plastic pipet. Viable cell count was determined by trypan blue exclu sion in a hemocytometer, and appropriate dilutions of viable cells were made to give a final concentration of 5 x 10" viable cells/ml. One ml of this cell suspension was then added to each well into which previously had been placed 1 ml of the test medium with a 2-fold concentration of growth factors or hormones so that the final concentration was as reported in the results. Wells were fed with 0.5 ml of medium containing the final hormone concentration every 3 to 4 days, and after the first feeding, 0.5 ml of spent medium was removed from each well prior to feeding. Cell counts were performed at the indicated intervals by removing the entire contents of one well with a Pasteur pipet, washing the well once with phosphatebuffered saline (Grand Island Biological Co., Grand Island, N. Y.), centrifuging the cells, discarding the supernatant and resuspending the cells in a known volume of either RPMI Medium 1640 or F14 medium, and counting the suspension in a hemocytometer. The data represent the average of duplicates of total cell numbers per well determined by multiplying the cell concentration times the volume resuspended. For long-term growth after 24 days, cells were transferred to new wells in the appropriate test medium at a concentration of 5 to 10 x 104, or into larger flasks. Other Methods. Glutaraldehyde-fixed cell pellets were ex amined by electron microscopy for dense core granules as described (17). L-Aromatic amino acid decarboxylase (L-dopa decarboxylase, EC 4.1.1.28) activity was determined by Dr. S. Baylin (Johns Hopkins Cancer Center) as described (2). Form aldehyde-induced fluorescence was determined by growing replicate cultures in SHITE medium, SHITE medium supple mented with 5-hydroxy-L-tryptophan (100 /ig/ml), or 1-/ÃŽ-3.4dihydroxyphenylalanine for 18 hr. as described (16). Tumorigenicity was determined by injecting 107 cells into each of 3 athymic nude mice (nu/nu), BALB/c background (ARS/ Sprague-Dawley, Madison, Wis.).