Engineering an L-cell line that expresses insulin under the control of the glucagon-like peptide-1 promoter for diabetes treatment

Abstract

Background: Diabetes mellitus is a complicated disease with a pathophysiology that includes hyperinsulinemia, hyperglycemia and other metabolic impairments leading to many clinical complications. It is necessary to develop appropriate treatments to manage the disease and reduce possible acute and chronic side effects. The advent of gene therapy has generated excitement in the medical world for the possible application of gene therapy in the treatment of diabetes. The glucagon-like peptide-1 (GLP-1) promoter, which is recognised by gut L-cells, is an appealing candidate for gene therapy purposes. The specific properties of L-cells suggest that L-cells and the GLP-1 promoter would be useful for diabetes therapy approaches.

Results: In this study, L-cells were isolated from a primary intestinal cell line to create suitable target cells for insulin expression studies. The isolated cells displayed L-cell properties and were therefore used as an L-cell surrogate. Next, the isolated L-cells were transfected with the recombinant plasmid consisting of an insulin gene located downstream of the GLP-1 promoter. The secretion tests revealed that an increase in glucose concentration from 5 mM to 25 mM induced insulin gene expression in the L-cells by 2.7-fold. Furthermore, L-cells quickly responded to the glucose stimulation; the amount of insulin protein increased 2-fold in the first 30 minutes and then reached a plateau after 90 minutes.

Conclusion: Our data showed that L-cells efficiently produced the mature insulin protein. In addition, the insulin protein secretion was positively regulated with glucose induction. In conclusion, GLP-1 promoter and L-cell could be potential candidates for diabetes gene therapy agents.

Figure 1

Construction and restriction maps of recombinant plasmids. (A) Schematic diagram of GLP-1/Ins/pBud plasmid: GLP-1 promoter (GLP-1 pro, blue part) and insulin gene (red part) in the pBudCE4.1 vector. The pBudCE4.1 vector contains the zeocin resistant gene. (B) Schematic diagram of GLP-1/Neo/pBlu plasmid: GLP-1 promoter (GLP-1 pro, blue part) and neomycin resistant gene (Neo) in the pBluescript-II-SK vector. The pBluescript-II-SK vector contains the ampicillin (Amp) resistant gene.

Figure 2

An MTT assay was used to check geneticin and zeocin cytotoxicity level. The viability of STC-1 cells was studied in the presence of 0 to 1 mg/ml geneticin (diamond) and zeocin (square) antibiotics. 400 μg/ml of geneticin and 500 μg/ml of zeocin were the lethal concentrations that kill all the STC-1 cells. Error bars indicate standard deviation, p < 0.05.

Figure 3

The results of RT-PCR analysis of expression of mouse GLP-1 and mouse β-actin mRNA. The five isolated clones from the STC-1 cell line (1-5) were subjected to RT-PCR. The PCR products of GLP-1 and β-actin were 250 bp and 578 bp, respectively, when compared with a 100-bp DNA ladder. All the five clones express mouse GLP-1 mRNA as well as mouse β- actin.

Figure 4

Analysis of GLP-1 mRNA expressions using quantitative-PCR. GLP-1 expressions in five isolated L-cells and STC-1 cells (as a control) were measured. All the isolated cells expressed more GLP-1 mRNA than the control cells. The BioRad CFX Manager software was employed to analyse GLP-1 expression data. The mouse β-actin and β-2 microglobulin (β2 m) mRNA levels were used to normalize the values. Error bars indicate standard deviation, p < 0.05.

Figure 5

The results of RT-PCR analysis of expressions of human insulin and mouse β-actin mRNA. Five stable clones in the antibiotic condition were analysed for insulin gene expression. The PCR products of the insulin and β-actin genes were 150 bp and 578 bp, respectively, when compared with a 100-bp DNA-ladder. All the five clones express human insulin mRNA as well as mouse β- actin.

Figure 6

The expression of human insulin protein was analysed using western blotting and immunocytochemistry assays. (A) The result of western blotting shows that mature human insulin was secreted by engineered L-cells. Total protein was extracted from L-1 and L-2. (B) Immunocytochemistry assay confirmed insulin expression in the L-cells. The green sections show that the cytoplasm of the cells expresses human insulin, and the blue sections are nuclei stained by DAPI.

Figure 7

Glucose responsive insulin secretion was analysed in L-cells. Insulin expressions in five isolated cells were studied by ELISA. The samples were collected during secretion tests. Insulin expression in all the isolated clones increased significantly with 25 mM glucose induction. Error bars indicate standard deviation, one-way ANOVA with unequal variances p < 0.05.

Figure 8

The relationship between the insulin secretion and time was investigated by ELISA. Clone number 1 and 2 were used for this experiment. The samples was collected every 30 min after inducing with glucose. The values were normalized to the amount of insulin secreted under basal conditions.