CrossTalk proposal: De novo capillary recruitment in healthy muscle is necessary

Perfusion is a principal determinant of muscle function, as is evident from impaired myocardial contraction during ischaemia. Perfusion determines solute exchange between blood and tissues, and is tightly regulated by local and central mechanisms. Here, we discuss how exercise and insulin regulate muscle perfusion, focusing on the control of microvascular surface area or ‘recruitment’. Where Ca, Cv and Ci and are the arterial, venous and interstitial concentrations of solute, F is flow, P is endothelial permeability, and S is the total endothelial surface area. Blood flow, concentration gradient, endothelial permeability and endothelial surface area are regulated variables. Recognizing that increasing blood flow or endothelial permeability will also increase muscle substrate uptake, we focus here on control of the endothelial surface area in response to exercise or insulin. Early, epoch-making studies by August Krogh introduced the concept that within unstimulated muscle the distribution of blood flow is uneven (Krogh, 1919), with some capillaries not receiving significant flow while others are well perfused. Much later, Honig et al. (1980) demonstrated that >50% of capillaries were unperfused at the time of sampling from resting muscle and that contraction rapidly (<5 s) expanded the number of erythrocyte-perfused capillaries. Central to our hypothesis is that the capacity of insulin and muscle contraction to increase microvascular surface area, i.e. microvascular recruitment, increases insulin and nutrient delivery. The term ‘capillary recruitment’, however, has become the topic of a vigorous debate. Capillary surface area can be recruited by de novo perfusion of previously temporarily non-perfused capillaries. Interestingly, microvascular blood flow is known to undergo rhythmical variations or ‘flowmotion’, which switches flow between vessels in cycles between 10 and 100 s duration. This flowmotion can be observed with laser Doppler flowmetry and contrast-enhanced ultrasound (CEU). Its amplitude is affected by insulin (Newman et al. 2009) and closely relates to microvascular recruitment in humans (Boer et al. 2014). Accommodating vasomotion with microvascular recruitment may involve insulin- or contraction-induced lengthening of perfusion intervals throughout the muscle microcirculation. We became interested in whether exercise or insulin affected its own delivery and that of glucose to muscle. This was part of an effort to understand the determinants of insulin transit from plasma to muscle interstitium, which appears to be a limiting step for insulin-stimulated glucose disposal. We developed two methods, one based on the single-pass extraction of 1-methylxanthine (Rattigan et al. 1997) and the second using CEU (Coggins et al. 2001). We have observed comparable effects of insulin using both methods, and findings were consistent with insulin acting to ‘recruit’ submaximally perfused microvasculature. We have used CEU more extensively for the following reasons: (i) it can be used in human studies; (ii) it provides measures of both the volume perfused and the rate or velocity of flow to the region of interest; (iii) it can be performed repeatedly; and (iv) it can be used for cardiac (Wei et al. 1998) as well as skeletal muscle. Extensive efforts have been made to use intravital microscopy to examine the same question. Studies of this process in the very thin muscles amenable to video microscopy have provided data for and against microvascular recruitment in response to exercise (Clark et al. 2008 ). This may in part be due to differences between very thin muscles and the larger, three-dimensional muscles that support the bulk of body musculature or to the clear methodological differences imposed by the invasive nature of intravital microscopy. These issues have been discussed previously (Clark et al. 2008 ). Non-invasive visualization of individual capillaries is available using nail-fold capillaroscopy in humans (Serne et al. 2002; de Jongh et al. 2004). Using this direct capillary visualization, insulin and ischaemic reperfusion induced capillary recruitment, clearly showing that, at least in skin, a microvascular reserve volume is available. Interestingly, in studies using insulin as a stimulus, there was good correlation between microvascular recruitment within skeletal muscle as measured by CEU and insulin-dependent microvascular recruitment within the nail-fold viewed directly by capillaroscopy (Meijer et al. 2012). In a series of studies in rodents and humans, we observed that insulin expands microvascular volume at physiologically relevant concentrations and exposure times. This effect occurs at lower insulin concentrations (Vincent et al. 2002; Zhang et al. 2004) and at earlier times (Vincent et al. 2002) than the effect of insulin to promote simultaneously measured limb blood flow. In addition, the microvascular flow velocity (measured as the rate at which microbubbles fill a region of interest within muscle) is not increased by physiological doses of insulin in either rodents or humans. The concordant results observed between the 1-methyxanthine and CEU methods add confidence to the conclusion that insulin acts specifically to expand microvascular volume. Using CEU, we subsequently examined the effect of muscle contraction. Interestingly, there was congruence between what we observed with insulin and what occurred with very modest levels of exercise (Vincent et al. 2006; Inyard et al. 2007). In both cases, the earliest and most sensitive changes were increases in microvascular volume with no change in microvascular flow velocity and no net change in total limb blood flow either to the human arm or to the rodent leg (Inyard et al. 2007). Even a modest hand-grip contraction (25% of maximum) expanded the microvascular volume by ∼2.5-fold without increasing total flow. Increasing contraction intensity to 80% of maximal hand-grip further increased microvascular volume minimally, but increased total forearm blood flow and microvascular flow velocity. In the rat leg, we examined the effect of contraction frequency varying from 0.1 to 10 Hz on microvascular perfusion and total limb flow. The lowest-frequency contraction again increased microvascular volume with no change in femoral blood flow, while with increasing contraction frequency total limb flow eventually increased, again with minimal or no further increases of perfused volume. It is straightforward to see advantages to this two-step vascular response by skeletal muscle microvasculature when challenged by either exercise or insulin (a surrogate for feeding). First, as indicated by eqn 1, the delivery of nutrients via exchange surface will increase when endothelial surface area expands, i.e. the efficiency of nutrient delivery is enhanced. Second, flow redistribution does not necessitate an increase in overall cardiac output. Finally, from the standpoint of the myocyte, the expanded distribution of perfusion shortens the pathway for nutrient delivery after leaving the blood compartment. This co-operative co-ordination between the microvasculature and the myocytes is highlighted by recent fluorescence and electron microscopy studies showing the intimate contact between skeletal muscle capillaries and the myocyte surface. The capillaries at many sites nestle into channels composed of invaginations of the myocyte membrane. Interestingly, myocyte mitochondria appear to aggregate around these sites of capillary proximity to the myocytes, presumably facilitating oxygen delivery and CO2 removal from the terminal elements of the respiratory chain (Glancy et al. 2014). If indeed blood flow distribution within skeletal muscle is regulated, what are the key regulatory elements? Our studies suggest that the redistribution of flow triggered by insulin depends critically on the action of insulin to enhance the activity of endothelial nitric oxide synthase (eNOS). Specifically blocking nitric oxide synthase (NOS) activity eliminates insulin-induced microvascular recruitment (Vincent et al. 2003). For contraction/exercise, the regulation is less clear, but blocking NOS has little or no effect (Inyard et al. 2007 ). Perhaps this is not surprising, because there are clear differences between the effects of contraction and that of insulin. The former is a much more potent and rapidly acting stimulus. The effects of insulin are typically characterized by a 40–70% increase in microvascular volume which requires 10–20 min to develop, whereas contraction acts within 5–10 s and can yield a 3-fold increment in microvascular volume. Over decades, it has proved difficult to elucidate the signalling pathways involved in exercise-stimulated increases of total limb flow, because there appears to be considerable redundancy, and blockade of one pathway is compensated by others. Dissecting out the specific pathways and mediators involved in exercise-triggered microvascular blood flow redistribution may likewise prove challenging. Insulin is not unique among hormones in enhancing muscle microvascular perfusion. Adiponectin (Zhao et al. 2013), Glucagon-like peptide-1 (Chai et al. 2012) and angiotensin II (via its type 2 receptor; Chai et al. 2010) each increase microvascular perfused volume and enhance skeletal muscle nutrient and oxygen delivery. Activation of NOS is required for each effect. The cellular pathways for activating eNOS differ between these several stimuli (Dong et al. 2013) and, perhaps importantly, do not involve the phosphatidylinositol 3-kinase pathway used by insulin. Adiponectin (Meijer et al. 2012) or 5-Aminoimidazole-4-carboxamide ribonucleotide (Bradley et al. 2010) activates AMP-kinase, Glucagon-like peptide-1 (Chai et al. 2014) acts on cAMP-dependent kinase, while angiotensin II directly enhances Ca2+ influx to activate eNOS. These microvascular actions remain intact in insulin-resistant states, while phosphatidylinositol 3-kinase activity is decreased. Interestingly, the effect of insulin to recruit microvasculature does not appear to be restricted to skeletal muscle; microvascular flow reserve within the myocardium (Liu, 2007) and adipose tissue (Sjøberg et al. 2011) are likewise ‘recruited’ by hyperinsulinaemia. This effect again appears absent in the myocardium of patients with type 2 diabetes (Scognamiglio et al. 2005). The findings in aggregate argue for an important role of microvascular recruitment for the delivery of nutrients and oxygen and the removal of waste products from muscle tissue in settings of enhanced contraction or altered nutrient availability. Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief (250 word) comment. Comments may be submitted up to 6 weeks after publication of the article, at which point the discussion will close and the CrossTalk authors will be invited to submit a ‘Last Word’. Please email your comment to journals@physoc.org. Eugene Barrett, MD, PhD is Professor of Medicine and directs the University of Virginia Diabetes Center. He completed medical and graduate training at the University of Rochester and endocrinology/diabetes training at Yale where he was on faculty from 1980 to 1991. He has served as vice president and president of the Am. Diabetes Assn. His research focuses on the relationship between insulin's vascular and metabolic actions. He is a member of ASCI and AAP and has chaired both the Metabolism and Clinical Integrative Diabetes and Obesity NIH study sections. Grants from the NIH and the ADA support his research program. Michelle A Keske, Ph.D., is a Senior research fellow at the Menzies Research Institute Tasmania, University of Tasmania. Her research has shown that i) microvascular function in muscle plays an important role in blood glucose regulation and ii) microvascular dysfunction in muscle can cause insulin resistance. Her current research focuses on interventions to prevent or reverse insulin resistance by regulating microvascular blood flow within muscle. Stephen Rattigan, Ph.D., is a Professor and Deputy Director at the Menzies Research Institute Tasmania, University of Tasmania. His research interests have focussed on the control of skeletal muscle metabolism. Collaborative studies with investigators in the USA, Denmark and the Netherlands have elucidated the important role that the vascular system plays in regulating muscle glucose uptake. He has pioneered novel methods to investigate microvascular blood flow in vivo in experimental animals and humans. Etto C. Eringa, Ph.D., is assistant professor at the Institute for Cardiovascular Research of the VU University Medical Centre in Amsterdam, the Netherlands. His research interests are control of tissue perfusion by adipose tissue and its contribution insulin sensitivity. He is board member of the Dutch Society for Microcirculation and Vascular Biology and is supported by the Netherlands Heart Foundation, the Netherlands Kidney Foundation and the Netherlands Organisation for Scientific Research. Disclaimer: Supporting information has been peer-reviewed but not copyedited. Last Word from David C. Poole Last Word from Eugene J. Barrett, Michelle A. Keske, Stephen Rattigan and Etto C. Eringa Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. None declared. NIDDDK 057878