Introduction

Traditional insulin therapy uses insulin injection to prevent blood glucose levels from soaring. However, injecting too much insulin would excessively lower blood glucose levels and cause hypoglycaemia. Therefore it is essential that the proper amount of insulin in injected to avoid side effects and complications.

In our previous gene therapy approach where insulin production is driven by a Tet-promoter, blood glucose levels of the patient would still need to be regularly checked to determine the amount of doxycycline needed to induce insulin production within a sensible range.

We came up with a better solution - transfecting cells with a glucose-regulated system that automatically monitors glucose levels and produces insulin whenever blood glucose is too high. For this we need a glucose-sensitive promoter that is only activated under high glucose levels. Extensive literature research revealed several promising candidates. Among them we chose a promoter of microbial origin, the S. cerevisiae Hexose Transporter 1 (HXT-1) promoter .

With this promoter as a glucose sensor upstream of our modified insulin gene, our system would be able to synthesize insulin when glucose levels are high, and stop automatically when glucose levels are low.

The HXT-1 Promoter

The HXT-1 promoter is a glucose-sensitive promoter derived from region -750 bp to -161 bp upstream of the S. cerevisiae Hexose Transporter 1(HXT-1) gene. In yeast, HXT-1 expression is positively regulated by glucose, and Ferrer-Martinez et al. characterized a minimal sequence between -750 bp and -161 bp upstream of the HXT-1 gene as the promoter region sufficient for glucose-sensitivity^[1] (Henceforth referred to as the HXT-1 promoter).

The function of the HXT-1 promoter relies on a specific glucose-sensing pathway in yeast, and although there is no known mammalian homologue, HXT-1 promoter has been shown to retain its glucose sensitivity in human fibroblast cells^[1].

In practice, the HXT-1 promoter works best as a hybrid promoter with a minimal-CMV (MIN) promoter attached downstream.

Constructing a Glucose Sensor

We were able to clone the HXT-1 promoter (586 bp) from the S. cerevisiae genome, and ligate it to a 206 bp minimal CMV promoter. Together they compose a glucose-sensitive promoter capable of functioning in both mammalian cells and S. cerevisiae cells.

We were surprised to learn there was no glucose-sensing promoter documented in the iGEM Parts Registry, therefore we submitted our glucose sensor as BioBrick BBa_K1328000 to the parts registry, hoping that other teams would benefit from our work.

Characterizing Glucose-Sensitivity

Before we put our system to producing insulin in mammalian cells, we first tested its glucose-sensitivity in yeast and mammalian cells by using EGFP as a reporter, by culturing transfected mammalian/yeast cells in culture media with different glucose concentrations. Fluorescence microscopy and flow cytometry analysis would then reveal any differences between the experiment groups.

Yeast

To fully characterize glucose-sensitivity, yeast cells transfected with the HXT1-CMVmini-EGFP plasmid was cultured in glucose-free media (with 3% glycerol as carbon source). They were then incubated with different concentrations of glucose (0, 5 mM, 10 mM, 15 mM, 20 mM, 25 mM) for different durations (0.5 h, 1 h, 8 h).

Figure 1. EGFP fluorescence observed in yeast cells cultured in a) 2% glucose, and b) 3% glycerol. c) Flow cytometry analysis: Cell count vs. EGFP intensity. Red: 2% glucose; Cyan: 3% glycerol.

Both fluorescence microscopy (Fig. 1) and flow cytometry analysis (Fig. 2) showed a positive correlation between EGFP fluorescence and glucose concentrations, demonstrating high glucose-sensitivity (lower than 5 mM). Time course studies revealed no significant difference between longer and shorter incubation durations, suggesting a fast response rate (less than 30 min).

Figure 2. Mean EGFP fluorescence measured by flow cytometry (FACS) in yeast cells, a) induced by different levels of glucose, and b) different incubation durations.

HEK293 Cells

HEK293 cells were transiently transfected with the HXT1-CMVmini-EGFP glucose-sensor/reporter plasmid, and cultured in glucose-deficient media. Different levels of glucose was then added to the cell cultures to induce EGFP expression. Cells were collected and subjected to FACS analysis. No significant EGFP signal change was observed along with increasing glucose levels, however GFP fluorescence was clearly visible under microscopic observation in all cases. We suspect this is due to HEK293 cells having a much lower glucose sensing threshold than yeast cells, and also caused by trace amount of glucose in the glucose-deficient media. This experiment will be repeated with glucose-free cell cultures and a weaker promoter than CMV-mini (which will be replaced with an even shorter 128 bp version.)

Figure 3. Flow cytometry analysis of glucose-induced EGFP expression in HEK293 cells cultured in various glucose concentrations and citrate (as non-glucose carbon source). Plotted is frequency vs. EGFP intensity.

Glucose-Regulated Insulin Production

After demonstrating the function of our glucose-sensor system, we will proceed to testing its capabilities in regulating insulin production in human cell lines. An Sensor-insulin plasmid has been constructed and we will be detecting insulin production and secretion with western blots against insulin.

Return to: Project

Reference

[1] Ferrer-Martı́nez, Andreu, et al. "A glucose response element from the S. cerevisiae hexose transporter HXT1 gene is sensitive to glucose in human fibroblasts." Journal of molecular biology 338.4 (2004): 657-667.
[2] Strathdee, Craig A., Marilyn R. McLeod, and Jennifer R. Hall. "Efficient control of tetracycline-responsive gene expression from an autoregulated bi-directional expression vector." Gene 229.1 (1999): 21-29.
[3] Baron, Udo, et al. "Co-regulation of two gene activities by tetracycline via a bidirectional promoter." Nucleic acids research 23.17 (1995): 3605.