Locally delivered GLP-1 analogues liraglutide and exenatide enhance microvascular perfusion in individuals with and without type 2 diabetes

Abstract

Aims/hypothesis: Glucagon-like peptide-1 (GLP-1) analogues reduce the risk of macrovascular disease in diabetes; however, little is known about their microvascular effects. This research examined the microvascular actions of the GLP-1 analogues liraglutide and exenatide in individuals with and without type 2 diabetes (study 1). It also explored the involvement of the GLP-1 receptor (study 2) and the nitric oxide pathway in mediating the microvascular effects of the analogues.

Methods: Trial design: Studies 1 and 2 had a randomised, controlled, double-blind study design. Study 1 participants, intervention and methods: three participant groups were recruited: individuals with well-controlled type 2 diabetes, and obese and lean individuals without diabetes (21 participants per group). Liraglutide (0.06 mg), exenatide (0.5 μg) and saline (154 mmol/l NaCl; 0.9%) control were microinjected into separate sites in the dermis (forearm) in a randomised order, blinded to operator and participant. Skin microvascular perfusion was assessed by laser Doppler perfusion imaging. Outcomes were stabilised response (mean skin perfusion between 7.5 and 10 min post microinjection) and total response (AUC, normalised for baseline perfusion). Perfusion response to GLP-1 analogues was compared with saline within each group as well as between groups. Study 2 participants, intervention and methods: in healthy individuals (N = 16), liraglutide (0.06 mg) and saline microinjected sites were pretreated with saline or the GLP-1 receptor blocker, exendin-(9,39), in a randomised order, blinded to participant and operator. Outcomes were as above (stabilised response and total perfusion response). Perfusion response to liraglutide was compared between the saline and the exendin-(9,39) pretreated sites. In vitro study: the effects of liraglutide and exenatide on nitrate levels and endothelial nitric oxide synthase phosphorylation (activation) were examined using human microvascular endothelial cells.

Results: Study 1 results: both analogues increased skin perfusion (stabilised response and total response) in all groups (n = 21 per group, p < 0.001), with the microvascular responses similar across groups (p ≥ 0.389). Study 2 results: liraglutide response (stabilised response and total response) was not influenced by pretreatment with exendin-(9,39) (70 nmol/l) (N = 15, one dataset excluded) (p ≥ 0.609). Liraglutide and exenatide increased nitrate production and endothelial nitric oxide synthase (eNOS) phosphorylation (p ≤ 0.020).

Conclusions/interpretation: Liraglutide and exenatide increased skin microvascular perfusion in individuals with and without well-controlled diabetes, potentially mediated, at least in part, by NO.

Trial registration: ClinicalTrials.gov NCT01677104.

Funding: This work was supported by Diabetes UK (grant numbers: 09/0003955 and 12/0004600 [RW and JM Collins Legacy, Funded Studentship]).

Keywords: Exenatide; Glucagon-like peptide-1 analogues; Liraglutide; Microvascular perfusion; Obesity; Type 2 diabetes.

Figures

Fig. 1

Representative skin perfusion response to microinjection of the GLP-1 analogues exenatide and liraglutide, compared with saline control, in a lean individual; the graph represents a typical pattern of response that was observed across all participants (n=63). The arrow denotes time of microinjection

Fig. 2

Skin perfusion response to microinjection of exenatide, liraglutide, ACh and saline in the lean, obese and type 2 diabetes groups (n=21 in each group). (a) Stabilised response and (b) total perfusion response to saline (squares), exenatide (circles), liraglutide (diamonds) and ACh (triangles). Data are presented as median (25th–75th percentile). The saline response was significantly lower than the responses to exenatide, liraglutide and ACh, respectively (stabilised response [a] and total response [b]) in all participant groups (***p<0.001, Wilcoxon signed rank tests). There was no difference in the response to exenatide, liraglutide or ACh between the participant groups. T2DM, type 2 diabetes

Fig. 3

GLP-1R inhibition does not alter the skin microvascular response to liraglutide in healthy individuals. Each participant (n=15) had 4 treatment sites, each receiving two microinjections: saline site (saline followed by saline); exendin-(9,39) (exendin-(9,39) followed by saline); liraglutide (saline followed by liraglutide); exendin-(9,39) liraglutide (exendin-(9, 39) followed by liraglutide). (a) Stabilised response, data presented as median (25th–75th percentile). (b) Total response, data presented as mean (SD). The skin perfusion response at the saline site was significantly lower than the response (stabilised and total) at the liraglutide site (**p<0.01 by Wilcoxon signed rank test for stabilised response; by paired t test for total response). Pretreatment by microinjection of exendin-(9,39) did not alter the microvascular response to liraglutide (liraglutide site vs exendin-9,39 liraglutide site, p≥0.609 by Wilcoxon signed rank test for stabilised response; by paired t test for total response)

Fig. 4

Exenatide and liraglutide increase eNOS phosphorylation and nitrate levels. (a) eNOS phosphorylation: after initial starvation, human microvascular endothelial cells (HCMEC/D3s) were treated with exenatide and liraglutide for 10 min. Controls were treated with 0.1% BSA medium only. Phosphorylation data were normalised to total eNOS (n=8). (b) Nitrate levels: HCMEC/D3s were treated with exenatide and liraglutide for 24 h (n=9). Controls were treated with medium only. For both (a) and (b), data are expressed as percentage of control, with control set at 100%, and are presented as median (25th–75th percentile). *p<0.05, **p<0.01 vs control, Wilcoxon sign rank test.