Original Paper

Acute regulation of pancreatic islet microcirculation and glycaemia by telmisartan and ramipril: discordant effects between normal and Type 2 diabetic rats

Anna Olverling, Zhen Huang, Thomas Nyström, Åke Sjöholm

Abstract

Diabetic patients are often treated with an ACEi (angiotensin-converting enzyme inhibitor) or angiotensin receptor antagonist against hypertension or albuminuria. These drugs also have a positive impact on glucose tolerance, but the mechanism for this remains elusive. Hypothesizing a positive non-additive effect, we studied whether the angiotensin receptor antagonist telmisartan or the ACEi ramipril acutely influence insulin secretion and glycaemia in vivo in healthy and Type 2 diabetic rats through effects on islet blood perfusion. Telmisartan and ramipril were injected intravenously into anaesthetized non-diabetic Wistar rats or Type 2 diabetic GK (Goto–Kakizaki) rats. In non-diabetic Wistar rats, neither whole PBF (pancreatic blood flow) nor IBF (islet blood flow) were significantly influenced by telmisartan and ramipril, alone or in combination. Renal blood flow was enhanced significantly by telmisartan and ramipril when used in combination, whereas ABF (adrenal blood flow) was not affected by any of the drugs. Telmisartan and ramipril both significantly increased serum insulin levels, but did not influence glycaemia. In Type 2 diabetic GK rats, both whole PBF and IBF were significantly decreased by telmisartan and ramipril, but only when used in combination. Renal blood flow was enhanced significantly by telmisartan and ramipril alone, but not when used in combination, whereas ABF was not affected by any of the drugs. Telmisartan and ramipril both significantly decreased serum insulin levels, and non-additively elevated blood glucose levels. In conclusion, the present study suggests that a local pancreatic RAS (renin–angiotensin system), sensitive to acute administration of telmisartan and ramipril, controls pancreatic IBF and insulin secretion and thereby has an impact on glucose tolerance. Our findings indicate unexpected significant differences in the effects of these agents on islet microcirculation, in vivo insulin secretion and glycaemia between healthy and Type 2 diabetic rats.

  • angiotensin receptor antagonist
  • blood glucose levels
  • glycaemia
  • renin–angiotensin system

CLINICAL PERSPECTIVES

  • Diabetic patients are often treated with an ACEi or angiotensin receptor antagonist against hypertension or albuminuria. These drugs also have a positive impact on glucose tolerance, but the mechanism for this remains elusive.

  • Our present findings suggest that a local pancreatic RAS, sensitive to acute administration of telmisartan and ramipril, controls pancreatic IBF and insulin secretion and thereby has an impact on glycaemia. Unexpectedly, the findings indicate significant differences in the effects of these agents on islet microcirculation, in vivo insulin secretion and glycaemia between healthy and Type 2 diabetic rats that could be related to impairments of microvascular function conferred by diabetes.

  • These differences may be relevant for understanding variability in clinical responses to RAS-interfering drugs between non-diabetic and diabetic subjects.

INTRODUCTION

The systemic RAS (renin–angiotensin system) has long been known to play a crucial role in the regulation of vascular tone and arterial BP (blood pressure). More recently, it has become increasingly clear that local RAS also exist in various tissues, for example, the pancreas, implying that high local levels of AngII (angiotensin II) might exert paracrine influences on neighbouring cells [13]. For instance, angiotensinogen, renin and AngII have been detected in the pancreas of several species [46]. AngII has been shown to negatively affect pancreatic and IBF (islet blood flow) through vasoconstrictive effects [7]. It is thus possible that pancreatic AngII, locally produced by intrinsic RAS, negatively influences insulin secretion in vivo by restraining islet blood perfusion, and, conversely, that blocking this system positively affects islet perfusion and function. This may be conspicuously important in diabetic subjects since hypertension is a common co-morbidity and several islet RAS components seem to be up-regulated by hypertension and hyperglycaemia [13]. Hence, many diabetic patients are treated with ACEi (angiotensin-converting enzyme inhibitors) or angiotensin receptor antagonists against their hypertension or as part of a renal protection strategy. Interestingly, many RAS-interfering agents have been reported to decrease the risk of incident diabetes in large clinical trials in which they are chronically administered [8]. However, the mechanisms behind these antidiabetic effects remain elusive. In the present paper, we have investigated the acute influence of the angiotensin receptor antagonist telmisartan and the ACEi ramipril, alone and in combination, on PBF (pancreatic blood flow) and IBF, as well as on blood glucose and insulin concentrations, in healthy and Type 2 diabetic rats. We hypothesized that these agents would, in a non-additive fashion, have a positive impact on glycaemia through stimulatory effects on islet microcirculation.

MATERIALS AND METHODS

Animals and drugs

Male Wistar rats (ScanBur) and GK (Goto–Kakizaki) rats [obtained from our KISÖS breeding colony (Animal Department, Stockholm South Hospital, Stockholm, Sweden)], weighing 300–350 g, were used in all experiments. The Stockholm GK rat colony [9] was established at KISÖS in 2004 from breeding pairs graciously donated by Professor Robert V. Farese (Gladstone Institute of Cardiovascular Disease, San Francisco, CA, U.S.A.). The animals had free access to pelleted food (Type R34; ScanBur) and tap water at all times. All experiments were approved by the local animal ethics committee. Telmisartan and ramipril were provided by Boehringer Ingelheim Pharmaceuticals.

Blood flow measurements

The experiments were performed according to a protocol previously described in detail [1018]. Non-fasting rats were anaesthetized with an intraperitoneal injection of thiobutabarbital sodium (120 mg/kg of body weight; Inactin™, Research Biochemicals International), and placed on a heated operating table to maintain the body temperature. Polyethylene catheters were inserted into the ascending aorta, via the right common carotid artery, and into the left femoral artery. The catheter in the aorta was connected to a pressure transducer (model PDCR 75/1; Druck) to allow constant monitoring of the mean arterial BP. After the BP became stable, the animals were acutely injected with a single i.v. (intravenous) bolus of 1 ml of saline, 1 ml of telmisartan (1 mg/kg of body weight), 1 ml of ramipril (0.4 mg/kg of body weight) or 1 ml of a combination thereof. Both of these substances were dissolved in saline. These treatment conditions (time period and concentrations) were chosen based upon previous publications [4,7,1118,24] and recommendations from the drug provider. An advantage of this acute setting, over chronic treatment, is that non-specific changes due to alterations in, for example, body weight does not confound the outcome.

At 10 min after the intravenous drug injections, 1.5–2.0×105 non-radioactive microspheres (IMT; Stason Labs), with a mean diameter of 10 μm, were injected in 10 s via the catheter with its tip located in the ascending aorta. An arterial blood sample was collected from the catheter in the femoral artery 5 s before the microsphere injection, and this process continued for a total of 60 s.

The exact withdrawal rate in each experiment was determined by weighing the sample. Additional arterial blood samples were obtained and later analysed for haematocrit, blood glucose and serum insulin concentrations (see below). After the animals were killed by cervical dislocation, the whole pancreas and both adrenal glands, as well as a 100-mg slice of the left kidney (including both cortex and medulla) were collected. The microsphere contents in these organs were determined separately. The organs were treated with a freeze–thawing technique [10], which enabled the visualization and localization of the microspheres from either the endocrine or the exocrine parenchyma of the pancreas. This was achieved by using a microscope (Zeiss MB6; Leica AB) equipped with both bright and dark field illumination. The former type of illumination allowed us to count the microspheres, whereas the latter enabled localization of the microspheres from either the endocrine or exocrine parenchyma [10]. The number of microspheres in the islets and exocrine tissue was counted as previously described in detail [1018]. The microsphere contents of the adrenal glands were used as a control to confirm an even distribution of the microspheres in the arterial circulation. The microsphere content of each of the arterial reference samples was determined by transferring the samples to glass microfibre filters and counting the microspheres with a stereomicroscope.

The blood flow rates were calculated according to the equation: Embedded Image(1) where Qorg denotes organ blood flow (ml/min), Qref denotes withdrawal rate of the reference sample (ml/min), Norg denotes number of microspheres in the organ and Nref denotes the number of microspheres in the reference sample.

Measurement of glucose and insulin concentrations

Non-fasting blood glucose concentrations were measured using a glucometer (OneTouch® Ultra® 2; LifeScan) and serum insulin concentrations with an ELISA kit (Rat Insulin ELISA; Mercodia).

Statistical analysis

All values are given as means±S.E.M. Statistical comparisons were made with two-way ANOVA (SigmaStat; SSPD). P<0.05 was deemed statistically significant.

RESULTS

In non-diabetic Wistar rats, neither whole PBF (pancreatic blood flow) nor IBF was significantly influenced by telmisartan and ramipril, alone or in combination (Figure 1). Renal blood flow was enhanced significantly by telmisartan and ramipril when used in combination (Table 1), whereas ABF (adrenal blood flow) was not affected by any of the drugs (Table 1). Telmisartan and ramipril both significantly increased serum insulin levels (Figure 2A), but did not influence glycaemia (Figure 2B).

Figure 1 Effects of ramipril and telmisartan on PBF in non-diabetic Wistar rats

Following i.v. injection of saline (control), telmisartan, ramipril or the combination thereof (R+T) into normal rats, rates of blood perfusion in the whole pancreas (A) and pancreatic islets (B) were measured using a microsphere technique. Bars represent means±S.E.M. for six independent experiments.

Figure 2 Effects of ramipril and telmisartan on serum insulin levels and glycaemia in non-diabetic Wistar rats

Following i.v. injection of saline (control), telmisartan, ramipril or the combination thereof (R+T) into normal rats, non-fasting serum insulin concentrations (A) were measured with ELISA and blood glucose levels (B) with test reagent strips. Bars represent means±S.E.M. for six independent experiments. *P<0.05, **P<0.01 and ***P<0.001 compared with controls using ANOVA.

View this table:
Table 1 Effects of ramipril and telmisartan on renal and ABF in non-diabetic Wistar rats and in Type 2 diabetic GK rats

Values represent means±S.E.M. for six independent experiments; R+T, ramipril+telmisartan. *P<0.05 and †P<0.01 compared with controls using ANOVA.

In Type 2 diabetic GK rats, both whole PBF and IBF were significantly decreased by telmisartan and ramipril, but only when used in combination (Figure 3). Renal blood flow was enhanced significantly by telmisartan and ramipril alone, but not when used in combination (Table 1), whereas ABF was not affected by any of the drugs (Table 1). Telmisartan and ramipril both significantly decreased serum insulin levels (Figure 4A), and non-additively elevated blood glucose levels (Figure 4B).

Figure 3 Effects of ramipril and telmisartan on PBF in Type 2 diabetic GK rats

Following i.v. injection of saline (control), telmisartan, ramipril or the combination thereof (R+T) into GK rats, rates of blood perfusion in the whole pancreas (A) and pancreatic islets (B), were measured using a microsphere technique. Bars represent means±s.e.m. for six independent experiments.*P<0.05 compared with controls using ANOVA.

Figure 4 Effects of ramipril and telmisartan on serum insulin levels and glycaemia in Type 2 diabetic GK rats

Following i.v. injection of saline (control), telmisartan, ramipril or the combination thereof (R+T) into GK rats, non-fasting serum insulin concentrations (A) were measured with ELISA and blood glucose levels (B) with test reagent strips. Bars represent means±S.E.M. for six independent experiments.*P<0.05 and **P<0.01 compared with controls using ANOVA.

DISCUSSION

Type 2 diabetes is increasing in prevalence globally and is seen in ever-younger age groups [8]. The disease is characterized not only by hyperglycaemia, but also insulin resistance with attendant dyslipidaemia and hypertension. Most diabetic patients are thus being treated with one or more antidiabetic drugs, a lipid-lowering statin and an ACEi or angiotensin receptor antagonist against hypertension and/or albuminuria. In clinical trials, it has been repeatedly observed that many of these drugs also influence glucose tolerance and the rate of incident diabetes [8]. Consistent with this, we have shown previously that RAS-interfering drugs exert substantial short-term impact on islet blood perfusion, insulin secretion and glucose tolerance in vivo in the normal and diabetic rats [15,17,18].

Telmisartan is an angiotensin receptor antagonist which, unlike others in its class, also acts as an activator of PPARγ (peroxisome-proliferator-activated receptor γ) [19]. Since activation of PPARγ results in increased insulin sensitivity, many [20]–but not all [21]–reports indicate that telmisartan improves glucose tolerance. Ramipril is an ACEi, which has also been reported in clinical trials to improve glucose tolerance [22,23].

However, the mechanisms underlying the glycaemic effects of these RAS-interfering agents are unknown. Since insulin secretion in vivo can be rapidly tuned by changes in pancreatic microcirculation, we evaluated the acute influence of administration of telmisartan and ramipril on IBF and dynamic changes in insulin secretion and glycaemia in healthy and diabetic rats, hypothesizing a positive non-additive acute effect of the drugs on these parameters. In contrast with this hypothesis, however, our findings unexpectedly reveal important quantitative and qualitative differences between healthy and Type 2 diabetic rats in the acute effects of these agents. In non-diabetic Wistar rats, neither whole PBF nor IBF was significantly influenced by telmisartan and ramipril, alone or in combination. Telmisartan and ramipril both significantly increased serum insulin levels, but did not influence glycaemia. Thus it seems that both of these agents–when given acutely–transiently induced a state of insulin resistance. This resembles previous findings in both male [18] and female [17] non-diabetic rats in which the ACEi captopril and the angiotensin receptor antagonist irbesartan produced similar results, albeit also enhancing islet microcirculation.

In Type 2 diabetic GK rats, both whole PBF and IBF were unexpectedly decreased acutely by telmisartan and ramipril, but only when used in combination (although a non-significant trend was noticed for ramipril). Telmisartan and ramipril both significantly decreased serum insulin levels, and non-additively elevated blood glucose levels. Such findings are compatible with a scenario not of increased insulin sensitivity, but of IBF being preferentially reduced, resulting in decreased insulin secretion (impaired β-cell function) and hence hyperglycaemia. Again, these findings contrast with previous results in which both captopril and irbesartan were found to increase IBF in GK rats [15]. The reasons for the unexpected differences are unclear, but could conceivably relate to differences in selectivity, potency and pharmacokinetics between the different drugs. The reasons for the marked differences between healthy and diabetic rats in acute vascular and glycaemic responses to telmisartan and ramipril currently noted also remain elusive, but are nonetheless consistent with previous findings [24]. The mechanisms underlying the unexpected differences between the control and Type 2 diabetic rats will have to await future studies and could only be speculated upon at this time. One possible explanation could be that endothelial function in the diabetic rats is compromised in such a way as to alter the polarity of the drugs’ effects.

When making comparisons with the diabetes preventive actions of RAS-interfering drugs–including telmisartan and ramipril–observed in clinical studies [8,20,22,23,25], it should be kept in mind that our present results reflect only very acute effects of the substances tested. Previous findings on long-term effects of such drugs on IBF and function consistently indicate beneficial effects. For instance, both ACEi and angiotensin receptor antagonists have been shown to ameliorate islet fibrosis and dysfunction during long-term oral treatment of Zucker diabetic fatty rats and other animal models of Type 2 diabetes [2631]. Specifically, telmisartan effectively prevented the development of glucose intolerance and diabetes in spontaneously non-obese Type 2 diabetic rats, possibly by down-regulating islet RAS components contributing to oxidative stress [32]. Mechanistically, telmisartan was found to attenuate lipotoxicity-induced islet formation of ROS (reactive oxygen species), PKC (protein kinase C) activation and NADPH oxidase activity [33], thereby improving islet cell viability and function. Similar findings were reported for ramipril, which reduced islet fibrosis along with decreasing expression of TGFβ (transforming growth factor β) [34]. However, it is unclear whether this was a direct islet protective effect or secondary to the alleviated insulin resistance by the weight-reducing effect of the drug. As to the mechanisms that might underlie differences between acute and chronic effects of these drugs, it is conceivable that chronic administration allows the accumulation of drug metabolites that have different effects than the drug itself. However, it is clear that further mechanistic investigations are needed in future work. We could not observe any correlations between changes in insulin levels and microcirculation, suggesting effects of the drugs that are not coherent in different physiological contexts.

In conclusion, our present findings suggest that a local pancreatic RAS, sensitive to acute administration of telmisartan and ramipril, controls pancreatic IBF and insulin secretion and thereby has an impact on glycaemia. Unexpectedly, the findings indicate significant differences in the effects of these agents on islet microcirculation, in vivo insulin secretion and glycaemia between healthy and Type 2 diabetic rats that could be related to impairments of microvascular function conferred by diabetes.

AUTHOR CONTRIBUTION

Anna Olverling and Zhen Huang performed the experiments and analysed the data. Thomas Nyström assisted in revising the paper. Åke Sjöholm conceived the hypothesis, planned the experimental design, wrote and edited the paper.

FUNDING

This work was supported by Boehringer Ingelheim Pharmaceuticals

Abbreviations: ABF, adrenal blood flow; ACEi, angiotensin-converting enzyme inhibitor; AngII, angiotensin II; BP, blood pressure; GK, Goto–Kakizaki; IBF, islet blood flow; i.v., intravenous; PBF, pancreatic blood flow; PPARγ, peroxisome-proliferator-activated receptor γ; RAS, renin–angiotensin system

References

View Abstract