Gene therapy for diabetes: metabolic effects of helper-dependent adenoviral exendin 4 expression in a diet-induced obesity mouse model

Abstract

Exendin 4 (Ex4) is a glucagon-like peptide-1 receptor (GLP- 1R) agonist which is available as a short-acting injectable treatment for type 2 diabetes. Our aim was to characterize the long-term effects of elevated steady-state levels of Ex4 provided by in vivo gene therapy. We constructed a helper-dependent adenoviral (HDAd) vector for long-term expression of Ex4 in vivo. A high-fat diet (HFD)-induced obesity (DIO) mouse model was chosen to approximate the metabolic derangements seen in obese patients. Mice were treated with a single injection of HDAd-Ex4 and were monitored for 15 weeks. Both hepatic Ex4 RNA and plasma Ex4 were detectable at the end of the study. HDAd-Ex4 treatment improved glucose homeostasis without increasing insulin levels. However, there was evidence of enhanced insulin action and decreased gluconeogenic enzyme expression. HDAd-Ex4 caused decreased weight gain without detectable changes in food intake, in part, due to increases in energy expenditure (EE). HDAd-Ex4 DIO mice also had reduced hepatic fat and an improved adipokine profile. In the liver, there was decreased expression of genes that were involved in de novo fatty acid synthesis. These observations are important in considering the development of longer acting GLP-1R agonists for the treatment of type 2 diabetes.

Figure 1. The exendin 4 (Ex4) cassette expresses immunoreactive and biologically active secreted Ex4

(a) Schematic diagram of the Ex4 expression cassette. (b) Transduction of Hep3B cells with HDAd-Ex4 (100 viral particles/cell) leads to detection of Ex4 in the cell lysate and the media after 48 hours as measured by enzyme immunoassay. (c) Rat 832/13 insulinoma cells were plated in 48-well dishes and cotransfected with 10 ng of pcDNA3.1 or pcDNA3.1-Ex4 (CMV-Ex4) and 400 ng of cyclic adenosine monophosphate–responsive element–driven luciferase reporter plasmid. After 48 hours, Ex4 peptide (10 nmol/l) or forskolin (10 μmol/l) was added in the indicated wells for an additional 6 hours before harvesting (*P < 0.05 compared to control). (d) 832/13 cells were transfected in quadruplicate with 10 ng of pcDNA3.1 or pcDNA3.1-Ex4. After 48 hours of expression, the media was removed and cells were treated with Hanks’ Balanced Salt Solution (HBSS) containing low (2 mmol/l) or high (15 mmol/l) glucose for 4 hours, with or without Ex4 peptide (10 nmol/l) as indicated. Immunoreactive insulin secreted into the HBSS was measured by enzyme-linked immunosorbent assay and normalized to total cell protein for each well (*P < 0.05 indicated by brackets). LUC, luciferase.

Figure 2. HDAd-Ex4 decreases weight gain and fat mass in diet-induced obesity (DIO) mice without detectable changes in food intake

The % weight change from baseline is shown with time from helper-dependent adenovirus (HDAd) injection (14 weeks of age) is shown for (a) DIO (n = 10 in each group) and (b) chow-fed mice (n = 5 in each group). The change was significant for the DIO mice (P < 0.05) at all time points. The net change in (c) fat mass and the (d) lean mass was measured by magnetic resonance imaging (*P < 0.05 HDAd-Ex4 compared to HDAd-0). (e) Twenty-four-hour food intakes were measured for 5 days at baseline (DIO, n = 20; chow, n = 10) and at 2 and 14 weeks after HDAd injection (DIO groups, n = 10; chow groups, n = 5) and are shown as kilocalories per gram of mouse per 24 hours. (f) Acute food intake was also measured after a 24-hours fast (n = 5 each group). Values are mean ± SD. Ex4, exendin 4.

Figure 3. HDAd-Ex4 treatment alters metabolic rate

Indirect calorimetry was measured for singly housed mice 4 weeks post-treatment for 22 hours from 3 PM to 1 PM the next day. The dark cycle was from 8 PM to 6 AM. (a) VO₂ (ml/kg/min) of diet-induced obesity (DIO) mice over the light and dark cycle (n = 7 each group). (b) Average VO₂ (ml/kg/min) of chow and DIO mice (c) VO ₂ (ml/kg/min) of chow-fed mice over the light and dark cycle (n = 4 each group). (d) Calculated energy expenditure (EE). (e) The respiratory quotient (RQ) was calculated from the VCO ₂-to-VO₂ ratio. The average VO₂ and calculated energy expenditure (kJ/kg/min) is compared among the different treatment groups for the total 22-hours light and dark cycle as well as the 10-hour dark cycle alone (*P < 0.05 HDAd-Ex4 compared to HDAd-0). (f) Activity measurements. Values indicate mean ± SD. Ex4, exendin 4; HDAd, helper-dependent adenovirus.

Figure 4. Glucose homeostasis is improved by HDAd-Ex4

Intraperitoneal (IP) glucose tolerance tests were performed (1.5 mg glucose/g mouse IP) for (a) diet-induced obesity (DIO) mice (n = 10, each group) and (b) chow-fed mice (n = 5, each group) at 15 weeks post-treatment. The area under the curve (AUC) for (c) blood glucose and (d) plasma insulin was calculated (*P < 0.05 HDAd-Ex4 compared to helper-dependent adenovirus (HDAd)-0 at the same time point). Acute insulin secretion (3.0 mg glucose/g mouse IP) was performed for at 15 weeks post-HDAd injection for DIO HDAd-treated mice (n = 10, each group), chow-fed mice (n = 5) and DIO mice injected with 2.4 nmol/kg mouse of commercial exendin 4 (Ex4) (n = 5) for measurement of (e) blood glucose and (f) plasma insulin levels (*P < 0.05 HDAd-0 compared to HDAd-Ex4; **P <0.05 for Ex4 peptide injection compared to both HDAd-Ex4 and HDAd-0). (g) The product of the fasting insulin × fasting glucose was calculated (*P < 0.05). (h) An insulin tolerance test was performed with 1.0 unit/kg mouse insulin IP. (i) Relative liver RNA levels of glucose-6-phosphatase (G6P) and PEPCK were measured by quantitative PCR and normalized to cyclophilin. The HDAd-0 values were set at 1.0 for comparison with HDAd-Ex4 (*P < 0.05). Values are mean ± SD.

Figure 5. Adipokine levels are altered by HDAd-Ex4 treatment

(a) Fasting (24 hours) plasma levels of adiponectin (Adpn; μg/ml), leptin (ng/ml), and resistin (ng/ml) are shown at 10 weeks after helper-dependent adenovirus (HDAd) injection for diet-induced obesity (DIO) mice (n = 10 each group; *P < 0.05 for HDAd-Ex4 compared to HDAd-0). Values are mean ±SD. Scatter plots of (b) resistin versus % fat and (c) adiponectin versus % fat are shown (HDAd-0, white circles; HDAd-Ex4, black circles). Ex4, exendin 4.

Figure 6. Hepatic fat is decreased in HDAd-Ex4-treated mice

(a) Liver tissue from mice treated for 15 weeks with helper-dependent adenovirus (HDAd)-0 or HDAd-Ex4 were fixed in phosphate-buffered formalin, paraffin sectioned, and hematoxylin and eosin staining was performed. Mice were fasted for 4 hours before killing. Representative sections from three separate mice for each HDAd treatment group are shown. Bars = 50 μm. (b) Lipid and triglyceride content of the liver was measured (n = 5/group; *P < 0.05 compared to baseline and brackets indicate P < 0.05 HDAd-Ex4 compared to HDAd-0). (c) The relative liver RNA expression of levels of key lipogenic gene products were measured by quantitative PCR normalized to cyclophilin (*P < 0.05 for HDAd-Ex4 compared to HDAd-0). Mice were fasted for 4 hours before killing. Values are mean SD. Ex4, exendin 4. Fasn, fatty acid synthase; Scd1, stearoyl coA desaturase 1; Acc, acetyl CoA carboxylase; Dgat, diacylglycerol acetyltransferase; Mgat, monoacylglycerol acetyltransferase; Fabp1, Fatty acid–binding protein 1; PPAR, peroxisome proliferator–activated receptor; SREBP-1c, sterol regulatory-element-binding protein-1c.