| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
The Lundberg Laboratory for Diabetes Research (G.M., M.S., L.S., S.G., P.L., P.-A.J.), Center of Excellence for Cardiovascular and Metabolic Research, Department of Molecular and Clinical Medicine, The Sahlgrenska Academy at Göteborg University, SE 405 30 Göteborg, Sweden; Department of Internal Medicine (G.M.), Section of Internal Medicine, Endocrine and Metabolic Sciences, Perugia University, I-06122, Perugia, Italy; and Department of Medicine (L.L.), Uppsala University Hospital, SE-431 80 Mölndal, Sweden
Address all correspondence and requests for reprints to: G. Murdolo, M.D., Ph.D., Department of Internal Medicine, Section of Internal Medicine, Endocrine and Metabolic Sciences, Perugia University, Via Enrico Dal Pozzo, I-06122, Perugia, Italy. E-mail: tugiuseppe.murdolo{at}medic.gu.seut and tugmurdolo{at}tiscalinet.itut.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of this study was to evaluate whether metacholine (MCh), a nitric oxide vasodilator, potentiates muscle capillary recruitment and forearm glucose uptake (FGU) during physiological hyperinsulinemia.
Design: The double-forearm technique [i.e. infused vs. control (Ctrl) forearm] was combined with im microdialysis during an oral glucose tolerance test in 15 nondiabetic, obese subjects divided into a group of insulin-resistant (IR) (n = 7) and insulin-sensitive (n = 8) individuals.
Results: After the oral glucose tolerance test, forearm blood flow in the Ctrl forearm was unchanged, whereas it increased about 3-fold (P < 0.0001 vs. baseline) in response to MCh. Capillary permeability surface area product for glucose (PSglu) (capillary recruitment), FGU, and interstitial insulin concentrations increased significantly over time (P < 0.001) in both forearms. Compared with insulin-sensitive, the IR subjects exhibited lower PSglu (P < 0.001) and FGU (P < 0.01) in the Ctrl arm, whereas this difference was insignificant in the MCh arm despite the blunted forearm blood flow increase. Moreover, in IR individuals MCh significantly (P < 0.05) ameliorated the delayed onset of insulin action, i.e. the FGU response to hyperinsulinemia. Finally, we found PSglu to be a strong and independent predictor of FGU response (adjusted R2 0.72; P < 0.0001).
Conclusions: MCh-induced vasodilation may improve the microvascular and metabolic responses to physiological hyperinsulinemia in obese, IR individuals. Further studies are required to unravel whether stimulation of nitric oxide production in skeletal muscle may represent an attractive therapeutic approach to bypassing cellular resistance to glucose disposal.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The notion that in resting muscle exists a "reserve" of unperfused capillaries that can be recruited by insulin or exercise has recently been questioned by other authors that, by intravital microscopy, showed red blood cells flow in the majority of muscle capillaries (12). These latter findings prompted to speculate that the increased substrate delivery to skeletal muscle seems to occur within already flowing capillaries, rather than through the recruitment of unperfused vessels, and the number/velocity of red blood cells within the muscle microvasculature may critically modulate the overall metabolic response.
Our laboratory has previously validated the use of im microdialysis to estimate directly, in vivo, the capillary permeability surface area product for glucose (PSglu), a composite measure that describes the capacity of a substance to reach the (i)fluid in relation to the extent of capillary recruitment (11, 13, 14). In those studies we reported that, under hyperinsulinemic conditions, glucose uptake is closely linked with the PSglu rather than with the forearm muscle flow, and the PSglu is further reduced among type 2 diabetic patients as compared with nondiabetic, obese subjects (11, 13). In harmony with these findings, insulin resistance to glucose uptake in obesity and type 2 diabetes has strongly been associated with impairment of insulin-mediated capillary recruitment (13, 14, 15). Interestingly, the transcapillary delivery of insulin to the muscle interstitium and the onset of insulin action to stimulate glucose uptake are equally delayed among obese, insulin-resistant (IR) individuals (14), supporting the hypothesis that, beyond the signaling defects, the dysfunctional capillary recruitment in skeletal muscle may further attenuate the overall action of insulin on glucose disposal.
Although the mechanisms underlying insulin-mediated microvascular recruitment remain unclear, compelling circumstantial data suggest nitric oxide (NO) as an important modulator (1, 5, 16, 17, 18, 19, 20, 21). Metacholine (MCh), a muscarinic receptor agonist also known as a potent stimulator of NO production in the forearm vascular bed (22), emerges as a suitable pharmacological tool in evaluation of the insulin-mediated metabolic and vascular effects in skeletal muscle (5, 18, 19, 20, 21, 22). Remarkably, in humans, local vasodilation with MCh has increased the forearm glucose uptake (FGU) either in the postabsorptive state (5) or after systemic insulin infusion (18, 19), implying that MCh affects both muscle capillary recruitment and glucose disposal over a wide range of levels of ambient insulinemia.
The objective of the present study was to evaluate whether MCh-induced vasodilation potentiates insulin-mediated capillary recruitment and ameliorates the FGU in obese, IR individuals during physiological hyperinsulinemia.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
There were 15 nondiabetic, obese subjects selected from individuals recruited through an advertisement in a local newspaper. The participants were enrolled if they met the following criteria: 1) abdominal obesity, defined by both a body mass index of more than 30 and less than 40 kg/m2, and a waist circumference more than or equal to 102 cm in male or more than or equal to 88 cm in female subjects; 2) a healthy state, as determined by medical history, physical examination, and screening laboratory evaluations; 3) a fasting plasma glucose concentration less than 5.6 mmol/liter; and 4) no current regular medication. Based on the homeostasis model assessment of insulin resistance (HOMA-IR) at the screening visit, the subjects were divided into a group of IR (HOMA-IR 
2.6; seven males) and insulin-sensitive (IS) (HOMA-IR < 2.6; seven men and one woman) individuals. Clinical characteristics of the study subjects are described in Table 1
.
|
Study procedures
The investigation started at 0800 h after an overnight fast, with the subjects lying supine in a quiet room kept at a constant temperature (25 C). Under local anesthesia (Xylocaine 2%; AstraZeneca, Molndal, Sweden), two catheters were inserted retrogradely into a deep antecubital vein in each forearm. Another cannula was inserted into the brachial artery of the dominant arm for the regional MCh infusion and arterial blood sampling. Hereinafter, the forearm instrumented with the arterial catheter will be referred to as the "MCh forearm" and the contralateral forearm as the "control (Ctrl) forearm." Another catheter was inserted into an antecubital vein of the Ctrl forearm for the inulin infusion.
After the cannulation procedures were completed, basal blood samples were withdrawn. Thus, a bolus injection of inulin (50 mg/kg) (Inutest 25%; Fresenius Kabi GmbH, Graz, Austria) was administered, followed by a constant iv infusion (100 mg/min for 300 min) (23). Moreover, three microdialysis catheters (CMA Microdialysis AB; CMA, Stockholm, Sweden) were inserted through the steel mandarin of a 20-gauge cannula into the brachioradialis muscle in both forearms. For insulin and inulin measurements, two catheters with a 12 x 0.5-mm dialysis membrane and a 100-kDa molecular mass cutoff were used, while for glucose and lactate, another catheter with a 16 x 0.5-mm 20-kDa molecular mass cutoff was applied. Finally, two mercury in-silastic strain gauges (Hokanson, Inc., Bellevue, WA) were placed on the upper third of both forearms for forearm blood flow (FBF) measurements by venous occlusion plethysmography (5, 22).
Study protocol
The study design is depicted in Fig. 1. After the baseline determinations, the volunteers underwent a standard oral glucose tolerance test (OGTT) (75 g). MCh chloride (0.1 mg/ml) was dissolved in isotonic saline to a concentration of 4 µg/ml, and infused into the brachial artery at 2 µg/min–1 for 60 min. This dosage was designed to induce an approximate 3-fold increase of FBF in obese individuals.
|
Muscle microdialysis
The principle of muscle microdialysis has previously been described in detail (11, 13, 14, 23). Calibration of the catheters for glucose and lactate measurements was performed using urea as an internal reference (24). Measurements of (i)insulin were taken according to the external reference calibration technique (13, 23, 25). In this study the mean relative recovery of inulin (dialysate inulin/plasma inulin) in vivo was 6.7 ± 0.4%, and the mean calculated in vivo recovery of insulin was 3.0 ± 0.4%.
Calculations
The permeability surface area product (PS) was calculated using the method for estimating the extraction fraction:
 |
Plasma flow rate was calculated by multiplying the FBF by hematocrit (%)/100. Finally, FGU was estimated using Fick’s principle, whereby: FGU = blood flow x (A – V) (blood concentrations).
Analytical methods
Glucose, lactate, and urea concentrations in the dialysate and plasma fractions were determined using a colorimetric (glucose and lactate) and UV (urea) method on a CMA 600 microdialysis analyzer (CMA Microdialysis AB).
In plasma and microdialysates, insulin concentrations were measured by an ultrasensitive assay (Ultra-Sensitive Insulin ELISA; Mercodia AB, Uppsala, Sweden), with a detection limit of 0.4 pmol/liter, and intraassay and interassay coefficients of variation of 5.3 and 2.7%, respectively. HOMA-IR was calculated as: fasting insulin (mU/liter) x fasting glucose (mmol/liter)/22.5. Finally, inulin concentrations were evaluated photometrically (26).
Statistical analysis
Comparisons within and between the subject groups in the two forearms over time were performed using a two-way ANOVA for repeated measures, with Geisser-Greenhouse adjustment for nonsphericity and Bonferroni’s post hoc comparison, where appropriate. Before statistical analysis, the Box-Cox transformation was used to normalize approximately the skewed distributed variables (Shapiro-Wilk test). Area under the curves (AUCs) were calculated by the trapezoidal integration method. The correlation between pairs of variables was assessed by a simple linear regression analysis, whereas multivariate relationships were analyzed using a general linear model.
Data are shown as means ± SEM for parametric and as medians (25th percentile; 75th percentile) for nonparametric variables, respectively. A P value less than 0.05 was regarded as statistically significant. Statistical analysis was performed using the NCSS software (Kaysville, UT).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The postabsorptive insulin concentrations in arterial plasma (P = 0.01) as well as in the veins of both the forearms (P < 0.05) were higher in IR than IS subjects (data not shown). After the OGTT, arterial and venous plasma insulin levels increased from 15 min onward (P < 0.0001 vs. basal), and the response was larger in IR than IS subjects (Table 1
).
The fasting (i)insulin concentrations were similar in the two forearms, and after the glucose challenge, both increased significantly (P < 0.0001 vs. baseline) (Fig. 2, panels A and B). Notably, in the MCh arm, the increase of (i)insulin occurred in both subject groups from 45 min onward (P < 0.0001 vs. baseline). However, the (i)insulin levels in the Ctrl forearm significantly increased from 45 and 105 min (P < 0.0001) onward in IS and IR subjects, respectively. Moreover, the (i)insulin concentrations in MCh [116.0 (68.0; 169.0) vs. 75.0 (37.0; 119.0) pmol/liter; P < 0.001] and Ctrl [121.0 (87.5; 203.5) vs. 84.0 (51.0; 117.0) pmol/liter; P < 0.001] arms, as well as the arterial plasma insulin levels [374.4 (211.8; 483.6) vs. 258.7 (190.4; 328.9) pmol/liter; P < 0.001] were higher in IR than IS subjects. Finally, the arterial-to-interstitial concentration gradient (AIG) for insulin increased after the OGTT (P < 0.0001 vs. baseline), but it did not differ between the groups and across the forearms.
|
In the fasting state, (i)glucose levels were significantly lower (P < 0.0001) than arterial plasma glucose (Fig. 3). The postabsorptive AIG for glucose increased (P < 0.0001) after the OGTT, but it was similarly maintained in the two forearms and between the groups.
|
). After the glucose load, arterial plasma lactate was unchanged in the IR subjects, whereas it increased significantly among the IS individuals from 45 min onward (P < 0.0001 vs. baseline). Accordingly, in response to the glucose challenge, a significant increase (P < 0.0001 vs. basal) in (i)lactate was seen, and the (i)lactate formation in IR subjects was lower than in IS subjects in both the MCh [0.87 (0.53; 1.06) vs. 1.04 (0.79; 1.52); P < 0.01] and Ctrl [(0.89 (0.65; 1.39) vs. 1.19 (0.82; 1.53); P = 0.01] arms.
|
At baseline the FBF in the two forearms was similar between the subject groups. After the OGTT, in both groups, the FBF rate in the Ctrl forearm remained stable, whereas it significantly increased in response to MCh at 15, 30, 45, and 60 min (P < 0.0001 vs. baseline and vs. Ctrl forearm) (Fig. 5, panels A and B).
|
).
|
In the postabsorptive state, neither the PSglu nor the PS for insulin (PSins) or the FGU differed between the forearms. In both groups, PSglu (P < 0.001 vs. baseline) and FGU (P < 0.0001 vs. baseline) increased significantly after the glucose challenge from 30 min onward, and the increase was observed in both forearms (Fig. 5, panels C–F). However, in the Ctrl arm, the IR individuals exhibited lower PSglu (P < 0.001) and FGU (P < 0.01) AUCs compared with the IS subjects, whereas in the MCh arm, the differences between the groups did not reach significance (Table 2). On the other hand, compared with the Ctrl forearm, the PSglu and the FGU AUCs in the MCh arm tended to be more pronounced in IR subjects and slightly reduced in IS individuals, respectively.
To more formally ascertain MCh effect on insulin action, we characterized the slope of the tangent to the initial activation curve of FGU (14, 27). In IR subjects, the FGU activation slope was reduced by approximately 55% (P < 0.05) in the Ctrl compared with the MCh arm, whereas in IS subjects the kinetic delay in activation between the forearms was not significant (Fig. 6, panels A and B).
|
AUC) of the FGU was strongly correlated with the AUC PSglu, both in the IR and IS individuals (Fig. 6C). Multiple regression analysis revealed that PSglu was a strong and independent predictor of FGU (adjusted R2 0.72; P < 0.0001), and the addition of other potential confounders did not significantly improve the prediction by PSglu alone, explaining about 8% additional variance (Table 3).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In obesity, although the diminished glucose uptake and utilization in the skeletal muscle have been attributed to a cellular defect of the insulin signaling network (28, 29), it is still debated whether the impairments of insulin-mediated actions on muscle vasculature may affect overall glucose disposal (1, 2, 3, 4, 8, 9, 10, 11, 13). Human studies suggest that insulin may differentially induce total muscle flow and capillary recruitment (4, 8, 9, 10), with important consequences for the glucose disposal rate. Indeed, insulin-stimulated capillary recruitment (independent of total blood flow) has been suggested to account for more than 50% of insulin-mediated glucose uptake (30). Moreover, even in the absence of changes in total muscle flow, increments in plasma insulin within the physiological range (4, 8, 10) or local insulin administrations (4, 9) enhance muscle capillary recruitment and concurrently increase glucose uptake. The "dissociation" between the microvascular and bulk blood flow responses to insulin has been explained by the higher "sensitivity" and the more rapid kinetics of the precapillary arterioles, which regulate the number of perfused capillaries, as compared with the larger resistance vessels, which modulate overall muscle flow (8, 10, 31).
In line with these observations, our present finding demonstrating a significant increase in both the PSglu and the FGU in the Ctrl forearm, in which FBF remained unchanged, agrees with the concept that, under physiological hyperinsulinemia, changes in glucose disposal rate closely follow the increase in microvascular recruitment rather than variations in FBF (10, 31). In addition, in line with previous results in lean IS volunteers (11), the present data in obese individuals show PSglu as a better predictor than FBF of the glucose uptake rate, especially when obesity is accompanied by insulin resistance. Consequently, inasmuch as an optimal metabolic response to insulin in muscle requires adequate "capillarization," the blunted capillary recruitment and the perturbed expansion of the permeability surface area appear to be a prerequisite for the relative resistance of the muscle cell to insulin action on glucose metabolism. However, from the present data, we cannot conclude as to whether the reduced PSglu, observed in the IR subjects’ Ctrl forearm, was due to a lower number of recruitable capillaries or was related to a blunted recruitment signal, or both (11, 13, 14).
So far, the bulk of experimental evidence indicates that the dynamics of muscle glucose uptake correlate closely to those of (i), rather than plasma insulin, implicating transcapillary insulin delivery as a rate-limiting step for insulin-mediated glucose disposal (32). Insulin availability to the skeletal muscle is virtually limited by either transport across endothelium or preferential distribution of microvascular flow between the "nutritive" and the "non-nutritive" route. Insofar as the permeability characteristics of muscle endothelium do not permit the transcellular transport of most macromolecules (such as insulin), insulin transfer to the interstitium largely occurs at the level of capillaries (30, 33). However, in IR states, transendothelial transport of insulin is rather "insufficient," and the rate of passage of insulin from the bloodstream to the muscle (i)space regulates the rate of activation of glucose disposal (14, 15). Therefore, delayed transcapillary transport of insulin may induce kinetic defect of insulin action. Indeed, the delayed distribution of insulin-to-muscle (i)fluid and the slower activation of the glucose uptake in obesity-linked insulin resistance occur concomitantly with the presence of a blunted insulin-mediated capillary recruitment (14, 15).
This study provides the first evidence that, in the obese IR state, intrabrachial MCh may circumvent the blunted PSglu and, importantly, ameliorate the slower onset of insulin action, as well as the delay of insulin transcapillary delivery in response to physiological hyperinsulinemia. In both groups the rate of FGU was largely accounted for (i.e. 
70% of its variance) by the changes in PSglu rather than by variations in FBF. Altogether, these data are consistent with the hypothesis that in obesity linked insulin resistance, MCh may attenuate the hindered glucose uptake by virtue of its action on the microvasculature. Indeed, in vivo studies have demonstrated that, in contrast to the endothelium-independent vasodilators, MCh enhances the FGU by augmenting capillary recruitment (1, 5, 18, 19, 20, 21). Accordingly, in human forearm vascular bed MCh has substantially increased the production/release of NO (22), a critical mediator of the insulin-stimulated hemodynamic actions (1, 5, 16, 17, 18, 19, 20, 21). It is also known that the NO pathway is defective in insulin resistance (34), and that inhibition of the NO synthase blocks microvascular recruitment and concomitantly blunts muscle glucose uptake in response to insulin (35). Together, the expansion of the endothelial surface area and the increase of the FGU in response to MCh may well be explained by the enhancement of capillary recruitment through the increased NO formation. Interestingly, whereas in IR subjects the magnitude of FBF increase after MCh infusion was roughly proportional to the increase of PSglu, in IS individuals MCh tended to decrease PSglu and FGU. This observation may imply that most of the NO produced in response to MCh was released in feed and resistance arteries outside the nutritive capillaries, and blood flow was partly shunted via the nonnutritive route, bypassing the muscle fibers. Alternatively, the more prominent FBF response to MCh in IS subjects may also explain such an "opposite" effect because the PS for a small molecule like glucose appears critically dependent on blood flow variations.
However, other mechanisms may underlie the microvascular and the metabolic response to MCh. First, in isolated rat skeletal muscle, NO per se has increased glucose transport through a mechanism that is distinct from the insulin pathway (16, 17). Second, by considering that the method we used to measure PSglu does not discriminate between microvascular recruitment and increase in the diameter of already open capillaries, we cannot exclude that MCh may have enhanced the number/velocity of red blood cells within the nutritive capillaries via a vasodilation of the upstream terminal arterioles (12). Third, in human forearm vasculature, cholinergic-induced vasodilation is predominantly mediated by the M3 receptor subtype (36, 37), which, in turn, induces the vascular effects via NO-dependent and NO-independent endothelial hyperpolarizing factor-mediated pathways (38). Finally, it is also possible that muscle capillary recruitment could be regulated by lactate (23, 39, 40, 41). In this study we found lower (i)lactate in IR than IS subjects after the OGTT, and the reduced ability to produce lactate in response to glucose and insulin may harmonize with the concept that the nonoxidative glucose metabolism modulates capillary recruitment (40). Furthermore, hyperinsulinemia failed to increase muscle (i)lactate in type 2 diabetes patients (13), and lactate has elsewhere exerted vascular effects independently of insulin-induced NO synthesis (40). Consequently, we cannot exclude that in obesity linked insulin resistance, reduced lactate formation may have exerted a more important effect on capillary recruitment and glucose uptake than insulin and NO did.
Another relevant finding in this study was the increase in the AIG for insulin and glucose after the OGTT, which was accompanied by a reduction in the transcapillary gradient for lactate. These data again concur with the hypothesis that, in obesity, the capillary wall appears partly rate limiting for muscle insulin delivery and glucose metabolism (14, 39, 40). However, in both the forearms, (i)insulin concentrations were higher in IR than IS subjects, implicating that the impairment of FGU was not due to failing capillary delivery of insulin but, likely, to the kinetic defect in delivery. Therefore, in addition to the intracellular defects, the increased (i)insulin concentrations, along with the delayed transcapillary insulin transport, may well contribute to the impairment of insulin-mediated glucose uptake by desensitizing insulin receptor signaling (42) and slowing the onset of insulin action (14), respectively.
Regarding the study limitations, it needs to be mentioned that this work was based on a relatively low number of participants. Thus, we may have missed significant differences within the forearms in each of the groups. In addition, it is important to emphasize that caution must prevail when extrapolating our data to the whole body either because of heterogeneity of the human muscles in intermediate metabolism or due to the limitations of Fick’s principle for calculation of net substrate uptake under "non-steady-state" conditions.
In summary, the present data show that local vasodi-lation with MCh may attenuate the impairment of insulin actions on muscle capillary recruitment and concurrently improve the glucose disposal rate in obese, IR individuals. Because NO availability drives both the vascular and metabolic responses induced by MCh, further studies are warranted to establish whether therapies that target the NO production/release in skeletal muscle may represent a strategy to bypassing the cellular resistance to insulin-mediated glucose disposal.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: G.M., L.S., S.G., P.L., and P.-A.J. have nothing to disclose. M.S. and L.L. are employed by AstraZeneca R&D, Sweden.
First Published Online May 6, 2008
1 G.M. and M.S. contributed equally to the work. 
Abbreviations: AIG, Arterial-to-interstitial concentration gradient; AUC, area under the curve; AUC0–120min, area under the curve during the oral glucose tolerance test; Ctrl, control; FBF, forearm blood flow; FGU, forearm glucose uptake; HOMA-IR, homeostasis model assessment of insulin resistance; (i), interstitial; IR, insulin-resistant; IS, insulin-sensitive; MCh, metacholine; NO, nitric oxide; OGTT, oral glucose tolerance test; PS, permeability surface area product; PSglu, permeability surface area product for glucose; PSins, permeability surface area product for insulin.
Received December 12, 2007.
Accepted April 29, 2008.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. G. Clark Impaired microvascular perfusion: a consequence of vascular dysfunction and a potential cause of insulin resistance in muscle Am J Physiol Endocrinol Metab, October 1, 2008; 295(4): E732 - E750. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, P < 0.05 vs. IS subjects, 
, P < 0.0001 vs. baseline in arterial plasma. *, P < 0.0001 vs. baseline in the MCh forearm. 
, P < 0.0001 vs. baseline in the Ctrl forearm.
