Engineered cell lines for insulin replacement in diabetes: current status and future prospects.

The recently completed diabetes complications and control trial has highlighted the need for improvement of insulin delivery systems for treatment of insulin-dependent diabetes mellitus. Despite steady improvement in methods for islet and whole pancreas transplantation over the past three decades, the broad-scale applicability of these approaches remains uncertain due in part to the difficulty and expense associated with procurement of functional tissue. To address this concern, we and others have been using the tools of molecular biology to develop cell lines with regulated insulin secretion that might serve as a surrogate for primary islets or pancreas tissue in transplantation therapy. This article seeks to provide a brief summary of the current status of this growing field, with a particular emphasis on progress in producing cell lines with appropriate glucose-stimulated insulin secretion.

Novel insulinoma cell lines produced by iterative engineering of GLUT2, glucokinase, and human insulin expression.

Cellular engineering studies in our group are directed at creating insulin-secreting cell lines that simulate the performance of the normal islet beta-cell. The strategy described in this article involves the stepwise stable introduction of genes relevant to beta-cell performance into the RIN 1046-38 insulinoma cell line, a process that we term "iterative engineering." RIN cells stably engineered to contain multiple copies of the human insulin gene exhibit a large increase in insulin content, such that they approach the content of human islets assayed in parallel. Analysis by high-performance liquid chromatography demonstrates that these engineered cell lines process human proinsulin to mature insulin with high efficiency. Cell lines that are further engineered to express the GLUT2 and glucokinase genes demonstrate stable expression of the three transgenes for the full lifetime of the lines produced to date (6 months to 1 year in continuous culture). Transplantation of the engineered cell lines into nude rats reveals that stably integrated genes are expressed at constant levels in the in vivo environment over the full duration of experiments performed (48 days). Several endogenous genes expressed in normal beta-cells, including rat insulin, amylin, sulfonylurea receptor, and glucokinase, are stably expressed in the insulinoma lines during these in vivo studies. Endogenous GLUT2 expression, in contrast, is rapidly extinguished during in vivo passage. The loss of GLUT2 is overcome in engineered cell ines in which transporter expression is provided by a stably transfected transgene. These results suggest that a potential advantage of the iterative engineering approach may be to preserve stability of function and phenotype, particularly in the in vivo setting.

Transfection of AtT-20ins cells with GLUT-2 but not GLUT-1 confers glucose-stimulated insulin secretion. Relationship to glucose metabolism.

Glucose is thought to stimulate insulin release from islet beta-cells through generation of metabolic signals. In the current study we have introduced the genes encoding the facilitated glucose transporters known as GLUT-1 and GLUT-2 into AtT-20ins cells to assess their impact on glucose-stimulated insulin release and glucose metabolism. We find that transfection of AtT-20ins cells with GLUT-2, but not GLUT-1, confers glucose-stimulated insulin release in both static incubation and perifusion studies. Cells transfected with GLUT-1 have a Km for 3-O-methyl glucose uptake of 4 mM and a Vmax of 5-6 mmol/min/liter cell space. These values are increased compared to untransfected AtT-20ins cells (Km = 2 mM; Vmax = 0.5 mmol/min/liter cell space), but are less than observed in GLUT-2-transfected lines (Km = 16-17 mM; Vmax = 17-25 mmol/min/liter cell space). Despite these dramatic differences in glucose transport affinity and capacity, the rates of [5-3H]glucose usage are not different in the control and transfected lines over a range of glucose concentrations from 10 microM to 20 mM. We conclude that the specific effect of GLUT-2 on glucose-stimulated insulin release in AtT-20ins cells is not related to changes in the overall rate of glucose metabolism and may instead involve physical coupling of GLUT-2 with cellular proteins and/or structures involved in glucose signaling.

Engineering of glucose-stimulated insulin secretion and biosynthesis in non-islet cells.

The high-capacity glucose transporter known as GLUT-2 and the glucose phosphorylating enzyme glucokinase are thought to be key components of the "glucose-sensing apparatus" that regulates insulin release from the beta cells of the islets of Langerhans in response to changes in external glucose concentration. AtT-20ins cells are derived from anterior pituitary cells and are like beta cells in that they express glucokinase and have been engineered to secrete correctly processed insulin in response to analogs of cAMP, but, unlike beta cells, they fail to respond to glucose and lack GLUT-2 expression. Herein we demonstrate that stable transfection of AtT-20ins cells with the GLUT-2 cDNA confers glucose-stimulated insulin secretion and glucose regulation of insulin biosynthesis and also results in glucose potentiation of the secretory response to non-glucose secretagogues. This work represents a first step toward creation of a genetically engineered "artificial beta cell."

Expression of normal and novel glucokinase mRNAs in anterior pituitary and islet cells.

The glucose-phosphorylating enzyme glucokinase likely plays an important role in regulating glucose-stimulated insulin secretion from the islets of Langerhans and has previously been thought to be expressed only in that tissue and in liver. In this study, we demonstrate high levels of glucokinase mRNA in the anterior pituitary cell line AtT20ins, which has been engineered to secrete correctly processed insulin, as well as in primary anterior pituitary tissue. Unlike islet or liver cells, expression of glucokinase mRNA in anterior pituitary cells was not accompanied by expression of the high Km glucose transporter (GLUT-2) mRNA. The glucokinase transcript in anterior pituitary cells was similar in size to islet glucokinase mRNA, which has a unique, elongated 5'-end relative to the liver glucokinase message. Amplification and sequence analysis of the glucokinase mRNA expressed in islets, RIN1046-38 cells, and anterior pituitary cells confirmed that the glucokinase transcripts in these cell types contain the same 5'-sequence. In addition, a novel alternative transcript was identified that contains a 52-nucleotide deletion and that predicts a 58-amino acid peptide as a result of a frame shift. Both the deleted and undeleted transcripts were found in islets, RIN cells, and AtT20ins cells, whereas only the deleted product was identified in primary anterior pituitary tissue. An antibody prepared against a peptide found at the N terminus of the islet isoform of glucokinase easily detected a protein with a size predicted by the undeleted transcript in extracts prepared from islets, RIN1046-38 cells, and AtT20ins cells. Since both the glucokinase protein and mRNA are naturally expressed in AtT20ins and RIN1046-38 cells, we compared the effect of varying concentrations of glucose on insulin secretion from the two lines. Insulin secretion from RIN1046-38 cells was stimulated by glucose in a dose-dependent manner over the range 0-2.5 mM, where it reached a maximum. AtT20ins cells, in contrast, exhibited no response to glucose at any concentration tested, despite the fact that insulin secretion from both cell lines was stimulated by incubation with dibutyryl cAMP. We conclude that glucokinase expression in AtT20ins cells may be necessary, but is not sufficient to confer glucose-stimulated insulin secretion.

Analysis of the protein products encoded by variant glucokinase transcripts via expression in bacteria.

Five variant transcripts of the single rat glucokinase gene have been described that are naturally expressed in islets of Langerhans, liver and anterior pituitary. Four of these were prepared as cDNA and expressed in bacteria in order to begin to address their physiological roles. Expression of constructs pGKB1 (normal islet/pituitary glucokinase) and pGKL1 (normal liver glucokinase) resulted in a glucose-dependent, glucokinase-like activity, 7-fold and 45-fold, respectively, above background. Expression of pGKB3 (variant islet/pituitary glucokinase) and pGKL2 (variant liver glucokinase) in contrast, did not result in any glucokinase-like activity.

Differential expression and regulation of the glucokinase gene in liver and islets of Langerhans.

Glucokinase, a key regulatory enzyme of glucose metabolism in mammals, provides an interesting model of tissue-specific gene expression. The single-copy gene is expressed principally in liver, where it gives rise to a 2.4-kilobase mRNA. The islets of Langerhans of the pancreas also contain glucokinase. Using a cDNA complementary to rat liver glucokinase mRNA, we show that normal pancreatic islets and tumoral islet cells contain a glucokinase mRNA species approximately 400 nucleotides longer than hepatic mRNA. Hybridization with synthetic oligonucleotides and primer-extension analysis show that the liver and islet glucokinase mRNAs differ in the 5' region. Glucokinase mRNA is absent from the livers of fasted rats and is strongly induced within hours by an oral glucose load. In contrast, islet glucokinase mRNA is expressed at a constant level during the fasting-refeeding cycle. The level of glucokinase protein in islets measured by immunoblotting is unaffected by fasting and refeeding, whereas a 3-fold increase in the amount of enzyme occurs in liver during the transition from fasting to refeeding. From these data, we conclude (i) that alternative splicing and/or the use of distinct tissue-specific promoters generate structurally distinct mRNA species in liver and islets of Langerhans and (ii) that tissue-specific transcription mechanisms result in inducible expression of the glucokinase gene in liver but not in islets during the fasting-refeeding transition.