Microvascular perfusion heterogeneity contributes to peripheral vascular disease in metabolic syndrome

Abstract

A major challenge facing public health is the increased incidence and prevalence of the metabolic syndrome, a clinical condition characterized by excess adiposity, impaired glycaemic control, dyslipidaemia and moderate hypertension. The greatest concern for this syndrome is the profound increase in risk for development of peripheral vascular disease (PVD) in afflicted persons. However, ongoing studies suggest that reductions in bulk blood flow to skeletal muscle may not be the primary contributor to the premature muscle fatigue that is a hallmark of PVD. Compelling evidence has been provided suggesting that an increasingly spatially heterogeneous and temporally stable distribution of blood flow at successive arteriolar bifurcations in metabolic syndrome creates an environment where a large number of the pre‐capillary arterioles have low perfusion, low haematocrit, and are increasingly confined to this state, with limited ability to adapt perfusion in response to a challenged environment. Single pharmacological interventions are unable to significantly restore function owing to a divergence in their spatial effectiveness, although combined therapeutic approaches to correct adrenergic dysfunction, elevated oxidant stress and increased thromboxane A₂ improve perfusion‐based outcomes. Integrated, multi‐target therapeutic interventions designed to restore healthy network function and flexibility may provide for superior outcomes in subjects with metabolic syndrome‐associated PVD.

Abbreviations

LZR

lean Zucker rat

OZR

obese Zucker rat

PGE₂

prostaglandin E₂

PGI₂

prostacyclin

PVD

peripheral vascular disease

TxA₂

thromboxane

Introduction

A defining characteristic of most developed economies world‐wide is an increasing incidence and prevalence of a combined constellation of pathologies that has become commonly referred to as the ‘metabolic syndrome’. While defining the specific constituents of the metabolic syndrome differ between governmental and scientific agencies, a reasonably complete definition of this clinical condition would include the combined presentation of three of: obesity, atherogenic dyslipidaemia, impaired glycaemic control, hypertension and the altered systemic conditions that develop in parallel with these, including pro‐oxidant, pro‐inflammatory and pro‐thrombotic states (Grundy et al. 2004; Ford et al. 2008). What makes a clinical diagnosis of metabolic syndrome compelling is that these systemic conditions result in an elevated risk for the development of overt cardiovascular disease that is in excess of what would be predicted based on a cumulative clustering of independent effects (Grundy et al. 2004). As such, the increased incidence and prevalence of the metabolic syndrome has profound implications for public health, including staggering direct and indirect economic costs to societies (Tamariz et al. 2010), its well‐documented effects on patient mortality and morbidity/quality of life, and the increasingly concerning interrelationships between the elevated cardiovascular disease risk associated with the metabolic syndrome and the patient outcomes associated with cognitive and behavioural impairments (Dearborn et al. 2014). Clearly, given that the most significant health outcomes of the metabolic syndrome may be the progressive decay in cardiovascular function and its impacts on end organ damage, breaking the connections between the progression of the metabolic syndrome and the negative cardiovascular disease and disease risk outcomes that it engenders represents a critical area of translational research facing us all.

The defining characteristic of peripheral vascular disease (either atherosclerotic or non‐atherosclerotic) is a progressive inability of the vascular system and networks to deliver blood to working tissues at a level and in a manner that is appropriate to maintain, or effectively meet, metabolic need. As such, in times of elevated metabolic demand, tissue and organ function becomes prematurely compromised owing to a growing perfusion:demand mismatch, and an early decay in function rapidly ensues. If sufficiently severe, this mismatch can create a compromised situation even at resting metabolic demands and is associated with chronic limb/organ discomfort and poor wound healing, among other related outcomes (Bild et al. 1989). Given this, the overwhelming majority of research into peripheral vascular disease has focused on altered patterns of vascular reactivity (either dilator or constrictor), with focus on both the vascular smooth muscle and endothelial cells (Muniyappa et al. 2013; Jia et al. 2014), their intrinsic function (Barrett & Liu, 2013; Musa et al. 2014), and alterations to the patterns of control over them exerted by extrinsic systems (e.g. adrenergic tone; Lambert & Lambert, 2011; Smith & Minson, 2012). However, all too often, our focus on vascular reactivity, its specific sites of altered (compromised) function and potential means through which it can be restored, or at least forced into a condition wherein bulk perfusion could be rescued, may have narrowed our vision such that the actual links between vascular reactivity and organ performance have been somewhat overlooked, or at least have been based on the fundamental assumption that improving bulk perfusion will, almost by definition, improve skeletal muscle performance.

In recent years, we have determined that this is not necessarily the case in the obese Zucker rat (OZR), possibly the most translationally relevant rodent model of the metabolic syndrome in human subjects (Aleixandre de Artiñano & Miguel Castro, 2009; Fellmann et al. 2013). The remainder of this Symposium Review will discuss this broader avenue of investigation, the new insights that it has generated, and potential avenues for future interrogation of the negative cardiovascular outcomes in the metabolic syndrome and how these could lead us to a greater understanding of system behaviour and the most potentially beneficial avenues for realistic therapeutic options.

The obese Zucker rat (OZR) model of the metabolic syndrome

The OZR model of the metabolic syndrome is one based on a chronic elevation in dietary food intake. Owing to a specific missense mutation, OZRs possess a leptin receptor that is largely non‐functional (Fellmann et al. 2013). As a result of this disruption in the negative feedback loop, the animal is severely hyperleptinaemic and has a minimal satiety reflex, with the ensuing chronic hyperphagia. In this phenotype both copies of the gene coding for the leptin receptor need to be dysfunctional, as heterozygotes for this gene are phenotypically normal and not distinguishable from the lean Zucker rat (LZR) control strain. With the chronic hyperphagia, OZRs rapidly become overweight and develop morbid obesity, with the ensuing development of worsening insulin resistance over the subsequent weeks (Aleixandre de Artiñano & Miguel Castro, 2009; Fellmann et al. 2013). While OZRs will eventually progress to type II diabetes mellitus, this takes considerable additional time, and true fasting glucose levels will remain largely normal for many weeks. Developing in parallel with the insulin resistance is a progressive dyslipidaemia, dominated by hypertriglyceridaemia. While there is certainly an elevation in plasma cholesterol levels, this is more moderate in nature and not exceptionally striking (Goodwill et al. 2009). As the insulin resistance worsens and the obesity becomes more severe, OZRs also develop a moderate hypertension that appears to be predominantly of adrenergic origin (Stepp & Frisbee, 2002). This tends not to be exceptionally large (on the order of 25–30 mmHg), but it is consistent. Developing in parallel with all of these pathologies is a steady elevation in the pro‐oxidant (Henriksen et al. 2011), pro‐inflammatory (Naka et al. 2004) and pro‐thrombotic (Shang et al. 2014) state of the OZR, and this can readily be identified through standard assays for the appropriate plasma and vascular biomarkers of reactive oxygen species, inflammatory cytokines and adipokines, and intermediates in the coagulation cascade. Taken together, it is evident that the OZR represents an excellent model for the metabolic syndrome in human subjects and not only manifests the appropriate pathologies, but also to an appropriate level of severity.

Previous efforts investigating vascular dysfunction in obese Zucker rats

The OZR has traditionally been used to interrogate alterations to endocrine and metabolic systems and control, given its rapid and predictable development of the metabolic syndrome. More recently, this model has become increasingly utilized for investigation of the cardiovascular outcomes of these conditions of elevated cardiovascular disease risk, especially as initial observations into the OZR model revealed a widespread dysfunction within the dilator reactivity of most vascular beds (Jin & Bohlen, 1997; Frisbee & Stepp, 2001; Bohlen & Nase, 2002; Lu et al. 2005; Zhao et al. 2005), with the predominant observation being that of a generalized compromised behaviour of the endothelial lining focusing around nitric oxide production and bioavailability (Jin & Bohlen, 1997; Frisbee & Stepp, 2001; Bohlen & Nase, 2002). However, alterations to other metabolic pathways contributing to vascular reactivity were also identified in the vasculature of the OZR, including those focusing around the balance between prostacyclin (PGI₂), prostaglandin E₂ (PGE₂) and thromboxane (TxA₂) production and action (Xiang et al. 2006, 2007, 2008), other pathways of arachidonic acid metabolism (Zhao et al. 2005) and haeme‐oxygenase‐dependent pathways (Johnson et al. 2006) among others. There have been no consistent reports of alterations to vascular smooth muscle dilator responses themselves.

With specific attention to the skeletal muscle resistance vasculature, we and others have consistently identified that the dilator responses of the resistance arterioles at multiple longitudinal positions, and in ex vivo, in situ and in vivo preparations, are significantly impaired in OZRs as compared to LZRs. While the alterations to dilator reactivity are predominantly endothelium dependent, this general impairment is present for a wide array of both physiological (e.g. low oxygen tension: Frisbee, 2003 a; shear stress: Frisbee & Stepp, 2001; Bouvet et al. 2007) and pharmacological stimuli (e.g. acetylcholine: Frisbee & Stepp 2001; Frisbee, 2003 a; arachidonic acid: Lu et al. 2005; Zhou et al. 2005). However, work done by Hester's group also identified the importance of the elevated production and actions of TxA₂ in contributing to vascular impairments in the skeletal muscle of OZR (Xiang et al. 2006, 2007, 2008; Hodnett et al. 2009), highlighting the need for targeted investigation. More importantly, this work was some of the first efforts to truly begin to link actual perfusion implications with altered vascular reactivity in OZRs as these authors demonstrated impairments to arteriolar blood flow patterns as a result of the actions of TxA₂, although the nature of their experimental preparation did not allow for an assessment of either bulk blood flow or skeletal muscle performance. However, to that point, there had been very little attempt to truly link alterations to vascular reactivity with perfusion outcomes to metabolic demand and muscle performance in the OZR.

Efforts at linking altered vascular reactivity to skeletal muscle performance in obese Zucker rats

One of the confounding variables in the interrogation of the links between vascular flow control and muscle performance is that the actual skeletal muscle fibres begin to change their phenotype with advanced duration of the metabolic syndrome towards the phenotype that is less oxidative, although this requires the development of overt type II diabetes mellitus (Adachi et al. 2007). This appears to be associated with the progressive decline in daily physical activity of the OZR, which begins to be identifiable after approximately 16 weeks of age (Stepp et al. 2004). As such, many investigators in this area have focused their attention on OZRs prior to 18–20 weeks of age to minimize fibre type shifts as a source of confounding influences. Further, given the inherent variability in the severity and composition of the metabolic syndrome at a specific age in OZR, the patterns of vascular reactivity or dysfunction, the perfusion and hyperaemic responses and the muscle performance, it is imperative that experiments addressing these issues be performed in individual animals to minimize the potential for drawing inferences and determining correlations inappropriately, as altered vascular reactivity is not necessarily a consistently strong predictor of either muscle hyperaemic responses or fatigue resistance.

In OZRs prior to 17 weeks of age, we have demonstrated that muscle fatigue resistance to high metabolic rate isometric tetanic (Frisbee, 2003 a) and graded metabolic rate isometric twitch (Frisbee, 2004; Frisbee et al. 2011) contractions is impaired relative to that in LZRs. Initial experiments indicated that this was associated with a reduction in the magnitude of the functional hyperaemic response to elevated metabolic demand and did not appear to be associated with altered skeletal muscle myocyte behaviour per se as single twitch and tetanic maximum tension levels were identical between LZRs and OZRs and half‐relaxation times of individual ex vivo skeletal muscles was not different (Frisbee, 2004). While confounding influences from alterations to microvascular network structure were present that had the potential to restrict perfusion responses at higher levels of metabolic demand and contribute to poor muscle fatigue resistance (Frisbee, 2003 b, 2005; discussed below), these results clearly demonstrated that the ability of the skeletal muscle resistance vasculature to respond appropriately to elevated metabolic demand was impaired in OZRs manifesting the full metabolic syndrome.

A follow‐up study designed to further interrogate the relationships between blood flow, oxygen extraction and uptake (${\dot{V}}_{{\mathrm{O}}_2})$and muscle performance revealed a separation between our traditional measurements of vascular reactivity/function and muscle perfusion/performance outcomes in OZRs (Frisbee et al. 2011). In this study, it was determined that despite the reduction in bulk blood flow to the in situ gastrocnemius muscle of OZRs contracting with elevated metabolic demand, oxygen extraction was reduced across the muscle and that the associated ${\dot{V}}_{{\mathrm{O}}_2}$ was lower than expected, suggesting that the oxygen that was being delivered to the muscle was not being effectively extracted into the surrounding skeletal muscle tissue, and that this may represent an important contributing factor for the impaired muscle performance (Frisbee et al. 2011). Treating the muscle with phentolamine to alleviate any potential constraint on perfusion and performance from an altered adrenergic constraint on vascular tone increased bulk perfusion and hyperaemic responses to the muscle, but had minimal impact on either oxygen extraction or muscle performance. Conversely, treatment of the animal with a combined cocktail to reduce the impact of elevated oxidant stress (TEMPOL) and the actions of TxA₂ (SQ‐29548) resulted in no change to bulk perfusion or hyperaemic responses or on muscle fatigue resistance, yet resulted in a substantial improvement to oxygen extraction across the gastrocnemius (Frisbee et al. 2011). The perplexing implication of this result was that a condition was generated where the amelioration of endothelial dysfunction (treatment with TEMPOL–SQ‐29548) in OZRs did not result in an improvement to muscle performance outcomes even though vascular reactivity was largely normalized. Taken together, these results suggested a ‘divergence of dysfunction’ wherein dysfunction in adrenergic constraint on perfusion restricted primarily bulk blood flow to the skeletal muscle within the proximal microcirculation, while the endothelial dysfunction (although ubiquitously present) primarily manifested its impacts in the more distal microcirculation, where blood/erythrocyte distribution was most critical to matching perfusion with local metabolic demand and improving oxygen extraction. The overall interpretation of these studies was that only when both adrenergic derangement and endothelial dysfunction were simultaneously rectified would a true recovery in both microvascular network perfusion and performance be realized within the contracting skeletal muscle of OZRs (Frisbee et al. 2011).

In a roughly parallel series of experiments, it was determined that one of the fundamental differences that define the presence of the metabolic syndrome in the skeletal muscle microcirculation of the OZR was an increasing heterogeneity of flow and erythrocyte distribution at successive arteriolar bifurcations (Frisbee et al. 2009). This distribution parameter, denoted as ‘γ’, describes the fractional flow down each daughter arteriole arising from a single parent (Bassingthwaighte et al. 1994; Frisbee et al. 2009). Obviously, γ of 0.5 reflects homogeneous flow distribution at a bifurcation, while values deviating from 0.5 reflect increasingly heterogeneous perfusion distribution across any arteriolar bifurcation. In this initial study, it was determined that at isolated arteriolar bifurcations, γ was elevated in OZRs versus that in LZRs and that this appeared to an integrated outcome of alterations to both adrenergic control over arterioles and the widespread endothelial dysfunction (Frisbee et al. 2009).

Building on this, we engaged in a detailed study designed to provide answers to three specific questions: (1) is a shift in γ present throughout the skeletal muscle microcirculation of OZRs; (2) what are the major contributing mechanisms to a shift in γ in the skeletal muscle of OZRs; (3) are those mechanisms spatially heterogeneous? Using the traditional in situ cremaster muscle of OZRs it was determined that the changes in γ that were previously identified in a limited number of bifurcations were consistently present throughout the microcirculation and spanned the range from 1–2A arteriolar bifurcations in the proximal microcirculation down to 4–5A bifurcations in the distal, pre‐capillary microcirculation (Frisbee et al. 2011). Further, it was determined that the previous observations that integrated contributions of altered adrenergic control and endothelial dysfunction (primarily reactive oxygen specieis (ROS)‐induced effects on TxA₂ production/action) were the major mechanistic players for the shift in γ in OZR, but that this is not spatially consistent (Fig. 1). In the proximal microcirculation, adrenergic dysfunction appears to dominate in terms of modifying γ, as pharmacological blockade with phentolamine restored normal levels of γ, while treatment against endothelial dysfunction was largely without effect (Frisbee et al. 2011). However, the reverse was true in the more distal microcirculation, as treatment with TEMPOL–SQ‐29548 normalized γ, and blockade of adrenergic dysfunction was without significant effect.

Figure 1. Perfusion Distribution at Arteriolar Bifurcations .

Microvascular perfusion distribution (γ) at arteriolar bifurcations spanning 1A (parent) and 2A (daughter) arterioles (panel A) and 3A (parent) and 4A (daughter) arterioles (panel B) within in situ cremaster muscle. Data are presented as means ± SEM for LZRs under control conditions and in OZRs under control conditions and following treatment with the adrenoreceptor antagonist phentolamine, the antioxidant TEMPOL, the PGH₂/TxA₂ receptor blocker SQ‐29548, or the combination of these agents. All data are adapted from Frisbee et al. (2011) with permission. *P < 0.05 vs. LZR; ^† P < 0.05 vs. OZR.

Using a simple simulation based on the experimentally determined values for γ, dichotomous branching and an 8‐bifurcation model for the skeletal muscle microvascular networks, these shifts in γ had the effect of radically changing perfusion distribution at the level of the 5A (pre‐capillary) arterioles in OZRs. In LZRs, terminal arteriolar perfusion distribution was largely Gaussian in nature, with relatively equal sharing of perfusion at the level of 5A arterioles, with an expected distribution of vessels receiving ‘lower than average’ blood flow and ‘higher than average’ blood flow. However, in OZRs, this distribution was severely shifted such that a small number of terminal arterioles received a very large, disproportionate perfusion at the expense of a large number of arterioles that received a lower than normal (ischaemic) blood flow (Frisbee et al. 2011). Treatment with either phentolamine or TEMPOL–SQ‐29548 exerted only a minimal impact on simulated terminal arteriolar perfusion distribution in OZRs, although combined treatment with both largely restored the Gaussian distribution predicted in LZRs (Frisbee et al. 2011). While this pattern of microvascular perfusion dysfunction and outcomes following intervention was a compelling explanation for the more integrated responses determined in the contracting gastrocnemius muscle presented earlier, it left open the question of multi‐scale validity.

To address the issue of multi‐scale validity, we used a single tracer washout technique in the in situ gastrocnemius muscle of LZRs and OZRs to determine the extent to which perfusion heterogeneity was present within the microvascular networks of this muscle (Wu et al. 2011). The results from these experiments demonstrated that perfusion heterogeneity was significantly increased in OZR skeletal muscle as compared to LZR muscle, exhibiting both accelerated, rapid pathways through the muscle, and very slow pathways that can retain the tracer marker for considerable periods of extra time, an observation that supports the results from the simulation efforts described above. However, at the conclusion of these experiments, there were two significant questions that needed to be addressed. The first of these was relevant to system compensation against dysfunction, and the second was relevant to the cumulative outcome for the actual processes of mass transport and exchange in the distal microcirculation.

The failure of system compensation against increased perfusion heterogeneity

One of the truisms of all normally functioning biological systems is that they have compensatory mechanisms in place that can minimize loss of function when portions of the system are compromised as a result of an exogenous stressor(s). When the system fails to find an appropriate or effective compensation, it must inevitably suffer a decline in performance to a new ‘set point’ that is consistent with the compromised elements. If, in the current case, the change in γ reflects a loss of normal function to the microvascular networks within skeletal muscle in OZR, this would represent a failure in the spatial distribution of blood flow within the network. The most likely source of compensation would be temporal in nature, and has at times been referred to as ‘switching’ (Wagner et al. 1999; Baumgartner et al. 2004) of blood flow within daughter branches arising from a parent arteriole. It is logical to assume that an increased rate of temporal switching would occur in the microcirculation of OZRs as a means of compensating for the increased spatial heterogeneity as a result of the shift in γ. This would be manifested as an increase in the rate at which the parallel daughter branch with the lower γ would switch to a higher flow, potentially becoming the branch with the higher γ, and vice versa as a means for reducing the downstream dilator signal that arises as a result of the relative ischaemia. Regardless of the stimulus for the downstream signals, parenchymal tissue metabolic end‐products, conducted responses from specific ischaemic/low flow regions, increased offloading of ATP from erythrocytes with lower than expected partial pressure of O₂ ($P_{{\mathrm{O}}_2})$ values, etc., this signal for temporal switching should be elevated in OZRs versus LZRs as a means of compensation.

This was not borne out by experimental determination (Fig. 2). When the temporal activity of proximal and distal arteriolar bifurcations within the in situ cremaster muscle of OZRs was tracked, it was determined that the degree of temporal switching at both 1A–2A (proximal) and 3A–4A (distal) bifurcations was significantly reduced in OZRs compared to LZRs (Butcher et al. 2013). While these relationships could be modified by pharmacological intervention, the central lesson taught by this was that there does not appear to be a significant compensation for the increased spatial heterogeneity in perfusion distribution (γ) in OZRs via an increased temporal activity. Rather, the temporal switching in γ appears to be reduced in OZRs compared to LZRs and this may actually serve to exacerbate the spatial heterogeneity and lock in the compromised microvascular function to a state for which there may be minimal compensation and minimal flexibility.

Figure 2. Statistical Distribution and Temporal Behavior of Gamma .

Frequency distribution and representative sample of the temporal changes in γ in 1A–2A (panels A and B, respectively) and 3A–4A (panels C and D, respectively) arteriolar bifurcations in LZRs and OZRs under control conditions. For data in panels B and D, γ is determined every 20 s throughout a 5 min collection window. All data are reprinted from Butcher et al. (2013) with permission.

Ultimately, in order for these alterations to perfusion distribution and control to be of functional significance, there has to be a compromised behaviour of oxygen transport and exchange in the distal microcirculation and capillary networks. To address this, a multiple tracer washout protocol was adapted to the in situ rat gastrocnemius preparation (Overholser et al. 1994; Wu et al. 2011) in order to determine the aggregate microvascular haematocrit within the microcirculation. As a parallel series of experiments designed to provide data at the highest relevant level of resolution, the tube haematocrit within individual capillaries of in situ cremaster muscle of LZRs and OZRs was determined using the established method of cell counting (Desjardins & Duling, 1987). When taken as an aggregate of ‘microvascular haematocrit’ using tracer washout, values in OZRs were significantly reduced as compared to LZRs, although combined acute interventions to remove adrenergic dysfunction and improve endothelial function were successful at largely restoring microvascular haematocrit in OZRs to levels determined in LZRs (Butcher et al. 2014). When given individually, the interventions were less successful at improving microvascular haematocrit alone in OZRs. Interestingly, the differences in mean capillary haematocrit using direct cell counting were not evident between LZRs and OZRs, although the variability in tube haematocrit between microvessels within a strain was significantly elevated in OZRs (Butcher et al. 2014), and the combined intervention used above was most effective at restoring normal levels of variability. Using haemodynamic and blood gas data from the in situ gastrocnemius muscle preparation, these results also allowed for a determination that perfusive and diffusive oxygen fluxes in OZR skeletal muscle were significantly reduced as compared to LZR muscle.

Taken together, the implications of this are considerable in terms of the loss of system flexibility for the skeletal muscle microcirculation. If we assume that the microcirculation, and more specifically arteriolar bifurcations, can normally assume a range of states with regard to γ, where γ can be highly variable around a central mean and rapidly responsive to stresses (e.g. elevated metabolic demand, haemorrhage, ischaemic insult or reperfusion injury, etc.), this affords a maximum degree of system flexibility in order to optimally meet the perfusion demands required by such imposed challenges (Fig. 3). With the development of the metabolic syndrome, the responsiveness of the system is severely abrogated, such that perfusion within the microvascular network becomes increasingly stable, with increasingly fixed perfusion distributions, with an increasingly variable and stable microvascular haematocrit. As this loss of system flexibility prevents the microvascular networks from effectively adapting their function in response to imposed stimuli, this will become manifested as a compromised fatigue resistance of the skeletal muscle they serve.

Figure 3. Temporal Behavior of Gamma Following Interventions .

Presentation of the attractor describing the overall spatial–temporal behaviour of γ at 1A–2A arteriolar bifurcations in LZRs (blue) and OZRs (red) under control conditions (panel A). The attractors are presented as iterated maps, where the respective value for γ is presented at multiple successive time points (T ₁ and T ₂) within that condition. In subsequent panels, the control data are greyed (light for LZR, dark for OZR) and the effects of the interventions on the attractor are presented in black. This is done for treatment with phentolamine (panel B), TEMPOL and/or SQ‐29548 (panel C) and phentolamine and TEMPOL and/or SQ‐29548 (panel D). See Butcher et al. (2013) for additional details (figure reprinted with permission).

The impact of skeletal muscle microvascular rarefaction on outcomes

Without question, the progressive impairments to the spatial and temporal control over perfusion distribution at skeletal muscle arteriolar bifurcations in OZRs compared to LZRs represents a significant contribution to the poor fatigue resistance of this tissue in response to elevated metabolic demand. Most importantly, this accumulating ‘error’ at successive bifurcations results in an extremely variable haematocrit within the capillary networks, and one which is also exceptionally stable and resistant to situation‐specific modifications. While this is clearly a significant problem for mass transport and exchange within the skeletal muscle microcirculation of OZRs, the negative effects of this mal‐distribution of blood and erythrocyte flow is compounded by the fact that it is emptying into a capillary network for which microvessel density is reduced by as much as 25% as compared to LZRs (Frisbee, 2003 b, 2005). This microvascular rarefaction under conditions of the metabolic syndrome is not limited to the OZR or to rodents in general, but has been repeatedly identified in human subjects suffering from obesity‐induced insulin resistance and other co‐morbidities of the metabolic syndrome. The obvious implication of this rarefaction for skeletal muscle performance is the impact on the volume and radial dimensions of the tissue cylinder that must be supplied by individual capillaries as a result of the reduced microvessel density. As has been clearly demonstrated and modelled/simulated extensively in the literature (Stainsby et al. 1988; Dah et al. 2010), the compromised delivery of substrate and blood gases, combined with the potential for impaired removal and disposal of metabolic end‐products will act as a major contributor to compromised tissue function – manifested as an increased rate of fatigue development in the skeletal muscle.

Future directions

Over the past few years, it has become increasingly apparent that in order for us to develop effective interventions against the negative microvascular outcomes associated with the development of the metabolic syndrome, this will require a greater understanding of the integrated function (dysfunction) within these networks in our experimental models. While the obvious, and frequently most effective, treatment for the negative health outcomes in the metabolic syndrome is a reduction in the severity of cardiovascular disease risk factors through lifestyle alterations (e.g. increased physical activity, altered diet, etc.), the ability of many subjects to immediately engage in this altered behaviour can frequently be somewhat limited. As such, we must begin to take a more sophisticated approach to the amelioration of the poor vascular function in metabolic syndrome. It is our contention that we must develop a more accurate understanding of what microvascular dysfunction truly is under the conditions of the metabolic syndrome, as it is clearly more than simply altered endothelial function, vascular reactivity or changes to hyperaemic responses. Information relevant to the multi‐scale behaviour of the microvascular networks and, increasingly, the level of synchronicity and communication between elements of the network may yield results and insights that are equally important putative mechanistic agents contributing to compromised outcomes. Clearly, the function of the integrated network is impaired in the metabolic syndrome, and this occurs at a level that is broader than simple indices of reactivity or predictive biomarkers of dysfunction. We must also determine the differences between interventions that can fully or partially restore system function and those that simply force an overriding condition on the microcirculation that can improve some indices of function (e.g. nitric oxide bioavailability), but does not effectively improve integrated perfusion or muscle performance outcomes. Most importantly, we must target the development of integrated therapies that will not only improve endothelial function through a reduction in levels of vascular oxidant stress and a normalization of the patterns of arachidonic acid metabolism to improve distal perfusion:demand matching as well as a normalization of adrenergic constriction to improve bulk hyperaemic responses. However, a more detailed understanding of the process of microvascular rarefaction in the metabolic syndrome must also be incorporated in order to maximally improve mass transport and exchange parameters at the capillary levels. While current knowledge in this area remains superficial, a higher resolution understanding of the temporal development of rarefaction in the metabolic syndrome may be particularly useful in terms of gaining mechanistic insight into this devastating process.

Conclusions

Over the course of the last 20 years, the OZR model of the metabolic syndrome has proven to be an exceptionally useful model for interrogating health outcomes associated with a constellation of elevated cardiovascular disease risk factors. While impairments to the integrated patterns of vascular reactivity have been extremely well established, simply assigning translational relevance to these measurements as being directly associated with impaired perfusion and, by extension, muscle performance, does not appear to be well justified. As such, simply correcting vascular dysfunction does not necessarily improve muscle perfusion and performance outcomes. More detailed investigation has determined that the importance of the alterations to vascular reactivity is that it engenders a shift in haemodynamic behaviour at arteriolar bifurcations, leading to an increasingly heterogeneous distribution of blood flow at each successive bifurcation, with the end result being an extremely heterogeneous distribution of perfusion at the pre‐capillary level. Further, this increasingly heterogeneous distribution of perfusion becomes increasingly stable, such that the flexibility within the microvascular networks of OZRs is severely attenuated and the ability of the networks to adapt their perfusion patterns to imposed challenges becomes increasingly abrogated. Ultimately, the functional consequences of this accumulating error at successive bifurcations leads to a microvascular haematocrit that is somewhat lower in OZRs than LZRs, but exceptionally variable and stable. When combined with the significant rarefaction of the microvascular networks of skeletal muscle in OZRs, this represents a series of significant contributing impairments to the ability of skeletal muscle to maintain performance with elevated metabolic demand; a defining feature of peripheral vascular disease in afflicted patients. Future experimentation and development of clinical therapeutic efforts could do well to focus on interventions that target the restoration of normal patterns of microvascular perfusion in human subjects suffering from the metabolic syndrome.

Additional information

Competing interests

None declared.

Funding

This work has been supported by the National Institutes of Health and the American Heart Association.

Biography

Jefferson Frisbee, PhD, is a Professor in the Department of Physiology and Pharmacology at the West Virginia University School of Medicine, and serves as the current Director for the Center for Cardiovascular and Respiratory Sciences. His research on microvasculopathy with the metabolic syndrome has received funding from the National Institutes of Health and the American Heart Association. He has served as the President of the Microcirculatory Society and is the current Editor‐in‐Chief of Microcirculation.