Glonal Insulinoma Cell Line That Stably Maintains Correct Glucose Responsiveness David Knaack, Deborah M. Fiore, Manju Surana, Margarita Leiser, Megan Laurance, David Fusco-DeMane, Orion D. Hegre, Norman Fleischer, and Shimon Efrat A number of pancreatic P-tumor cell (PTC) lines have been derived from insulinomas arising in transgenic mice expressing the SV40 T antigen gene under control of the insulin promoter. Some of these lines secrete insulin in response to physiological glucose concentrations. How- ever, this phenotype is unstable. After propagation in culture, these nonclonal lines become responsive to sub- physiological glucose levels and/or manifest reduced in- sulin release. Here we report the use of soft-agar cloning to isolate single-cell clones from a PTC line, which give rise to sublines that maintain correct glucose responsive- ness and high insulin production and secretion for >55 passages (over a year) in culture. One of these clonal lines, denoted PTC6-F7, was characterized in detail. (JTC6-F7 cells expressed high glucokinase and low hexo- kinase activity, similarly to normal islets. In addition, they expressed mRNA for the GLUT2 glucose transporter isotype and no detectable GLUT1 mRNA, as is character- istic of normal P-cells. These results demonstrate that transformed P-cells can maintain a highly differentiated phenotype during prolonged propagation in culture, which has implications for the development of continuous P-cell lines for transplantation therapy of diabetes. Dia- betes 1413-1417, 1994 P ancreatic p-cell lines derived from rodents have been extensively used to study the regulation of insulin production and secretion. A number of continuous P-cell lines have been generated from insulinomas and hyperplastic islets arising in mice express- ing a transgene encoding the SV40 T antigen oncogene under control of the insulin promoter (RIP-Tag) (1-6). Several of these lines displayed insulin secretion characteristics similar to those observed in intact adult islets, in particular the response to glucose concentrations in the physiological range (5-15 mmol/1). However, a common problem encoun- tered with all these cell lines is their phenotypic instability. After prolonged propagation in culture, these cells become From CytoTherapeutics (D.K., D.M.F., M. Laurance, O.D.H.), Providence, Rhode Island, and the Departments of Medicine (M.S., M. Leiser, N.F.) and Molecular Pharmacology (D.F.-D., S.E.), Albert Einstein College of Medicine, Bronx, New York. Address correspondence and reprint requests to Dr. Shimon Efrat, Department of Molecular Pharmacology, Albert Einstein College of Medicine, 1300 Morris Park Ave., Bronx, NY 10461. Present addresses: M. Laurance, Oregon Health Sciences University, Portland, OR; O.D.H., Neocrin Inc., 31 Technology Drive, Irvine, CA. Received for publication 9 April 1994 and accepted in revised form 21 July 1994. PTC, p-tumor cell; KRB, Krebs-Ringer buffer; RIA, radioimmunoassay; IBMX, isobutylmethylxanthine; BSA, bovine serum albumin; PCR, polymerase chain reaction. responsive to subphysiological concentrations of glucose and/or manifest diminished insulin output (4,6-9). A similar instability has been observed with P-cell lines derived by other methods (10-12). The triggering of insulin secretion by glucose requires glucose metabolism in p-cells (13). It has been proposed that glucose phosphorylation to glucose-6-phosphate, which con- stitutes the rate-limiting step in glycolysis, represents the "glucose sensor" of P-cells (14). This reaction is catalyzed in P-cells primarily by the high-if m enzyme glucokinase (15), while most other cells use \ow-K m hexokinases (16). In the case of one insulinoma cell line, P-tumor cell (PTC)7, the development of responsiveness to subphysiological glucose concentrations has been clearly associated with an increase in hexokinase activity (6). Like most other reported p-cell lines, the PTC7 line is a nonclonal cell population. Therefore, it remains unclear whether the observed changes represent an adaptation of the entire cell population to growth in culture or a selection of a subpopulation. A phenotypically stable cell line manifesting correct glu- cose responsiveness would be of considerable value for the study of P-cell function and regulation. Here we report the use of soft-agar cloning to establish correctly regulated p-cell clonal sublines from a poorly regulated parental PTC cell population. One of these sublines was studied in detail during prolonged propagation in culture. It has maintained high insulin output and correct glucose regulation for >55 passages. RESEARCH DESIGN AND METHODS Cell culture. (3TC cells were grown according to Efrat et al. (2) in Dulbecco's modified Eagle's medium (GIBCO) containing 25 mmol/1 glucose and supplemented with 15% horse serum and 2.5% fetal bovine serum (GIBCO) (growth medium). Cells were passaged twice a week using a brief trypsinization in calcium-free Hank's balanced salt solution containing 0.05% trypsin and 0.53 mmol/1 EDTA. Except in the case of expanding cultures, cells were plated at 50,000 cells/ml in Falcon tissue culture dishes. Soft-agar cloning. Approximately 60,000 PTC6 cells at passage 18 were suspended in 9 ml prewarmed growth medium with 10% conditioned medium from (3TC3 cells (complete medium). The cell suspension was examined to assure that it consisted of well-dispersed individual cells; 0.5 ml of Matrigel (Collaborative Research) and 1 ml 3% agar (Batco) were added to the cell suspension, and 1.5 ml of the mixture was then placed in each well of a six-well plate. Cells were fed with complete medium twice a week and were allowed to grow in agar for 3 weeks. Individual cell clusters were then harvested from the agar by pipette, and each cell cluster was placed in a single well of a 96-well plate. Clusters were disrupted mechanically, fed three times a week with complete medium, and split as required. Each clone was tested in a single well for insulin secretory response to glucose. The selected clones were then expanded in complete medium. Insulin release. Initial screening of clones for glucose responsiveness was performed in 24-well plates. Cells were rinsed several times with DIABETES, VOL. 43, DECEMBER 1994 1413