ENDOTHELIAL CONTROL OF LOCAL CONDUCTANCE

Assessment of endothelial control of local conductance

Reactive hyperaemia

If the arterial inflow to a tissue is occluded for more than a few seconds, the conduit and microcirculatory arterial vessels dilate downstream of the occlusion. As a result, release of the occlusion is accompanied by a temporary elevation of local blood flow above the pre-occlusion level, termed reactive hyperaemia. The reactive hyperaemic responses of both conduit and microcirculatory vessels are attenuated after administration of substances that prevent NO production, suggesting that both involve EDRFs and might, therefore, be utilized as indices of endothelial dilator function. It is uncertain whether EDRFs other than NO are implicated.

Reactive hyperaemia is usually elicited in the forearm with occlusion of the brachial artery above the elbow for a period of between 2 and 3 min. In these circumstances, the peak flow immediately after release of occlusion is typically two to four times the pre-occlusion value and returns to this resting level over 30–50 s. Conduit vessel dilation is assessed using ultrasound imaging of brachial artery diameter, and the reactive hyperaemic response is referred to as flow-mediated dilation. Microcirculatory dilation is usually assessed either by measurement of blood flow through forearm or finger using venous occlusion plethysmography (Fig. 6.1) or by laser-Doppler measurement of forearm skin flow.

Figure 6.1 Venous occlusion plethysmographic record of forearm blood flow immediately following cessation (at red arrow) of 3 min brachial arterial occlusion. The red lines indicate rates of rise of tissue circumference during each period of venous occlusion. Note the high initial reactive hyperaemia and the subsequent decline in blood flow over the succeeding minute.

From Barry (2003) with permission of the author.

Comparative studies show poor correlation between the magnitudes of large and small vessel responses in different subject populations (Hansell et al 2004), which is perhaps not surprising if we consider the mechanisms likely to underlie dilation in the two vessel types. Assuming that all endothelial cells release EDRFs continuously, local accumulation of these factors during the period of arterial occlusion should have similar dilator consequences in large and small vessels once the normal pressure gradient is re-established. In conduit vessels, however, the increased volume flow will be accompanied by a high rate of shear stress on the endothelium that will increase local EDRF release further. As flow velocity is much lower in the microcirculation, shear stress is not likely to generate factor release at that site. A second complicating factor arises from the fact that cessation of blood flow through a tissue results in local accumulation of dilator metabolites (see Chapter 6, p. 67). Depending on the metabolic rate of the tissue and the duration of arterial occlusion, metabolic dilation in the microcirculation may, therefore, act additively with EDRFs.

It is not certain at present whether reactive hyperaemia in large or in small arteries provides the better index of physiological endothelial behaviour. Purely from the point of view that the microcirculation contributes to peripheral resistance most, it seems logical that this site should be used if the investigators’ concern is the implications of endothelial function for blood pressure control. If one adopts this position, then it is clearly essential to avoid interference by endothelium-independent metabolic dilation. Given the fact that skin metabolism is negligible unless there is active sweating, using skin rather than whole forearm blood flow may provide more accurate data.

Endothelial turnover

The endothelium is turning over continually, with dying cells being shed into the circulation and being replaced by new cells that originate from circulating bone marrow-derived progenitors. These processes may provide additional indices by which the functional status of the endothelium can be assessed at a whole-body level. On the one hand, large numbers of dislodged endothelial cells in the blood stream could indicate reduced viability, consistent with reduced EDRF function. A number of studies have reported elevated numbers of circulating endothelial cells (CECs) in various disease states associated with other evidence of reduced endothelial dilator capacity, including hypertension and diabetes. Conversely, it could be argued that rapid cell turnover would lower the average age of the endothelial population and so reduce the proportion of old, metabolically sub-optimal cells. This possibility is consistent with observations that CECs rise in parallel with endothelial dilator capacity in healthy subjects during aerobic training (see Chapter 11, p. 134).

CEC numbers alone, therefore, provide only ambiguous information on endothelial status and additional markers of endothelial viability must be used concomitantly in order to decide whether high CEC values denote healthy or unhealthy endothelium. One marker that may be useful is Von Willebrand factor, which is released from endothelial cells in response to inflammatory stimuli. Von Willebrand factor levels appear to be elevated in at least most pathological situations where CEC numbers are increased. By contrast, the one study that has compared these parameters with fitness levels in healthy subjects (O’Sullivan 2003) found that the rise in CECs associated with training was not linked to any change in plasma Von Willebrand levels.

METABOLIC CONTROL OF LOCAL CONDUCTANCE

All known cellular metabolites, including adenosine nucleotide degradation products, CO₂, phosphates, lactate and protons, have dilator actions on vascular smooth muscle. This provides a mechanism by which local vascular conductance can, in theory, be matched automatically to tissue metabolic demands. At least some of the activity of certain metabolites (for example adenine molecules) is produced via their stimulating release of NO and other endothelial dilator factors; other metabolites, such as protons, appear to act directly on the vascular smooth muscle. Since the balance between different factors depends on the precise metabolic pathways utilized in a tissue, the dilator contributions of each metabolite vary between different organ systems. Identifying the relative importance of different factors in particular situations may be important since variations in those factors that are most potent, or in highest concentration, may be the most reliable index of hyperaemia. Because of this, much debate has taken place in the research literature as regards the roles of specific factors in, for example skeletal muscle (Clifford & Hellsten 2004).

In terms of functional significance, on the other hand, the evidence to date suggests that no single identified factor is overwhelmingly important in any tissue (see Functional hyperaemia during exercise, p. 72, for further discussion). This does not necessarily imply that another unknown factor is the main contributor. The non-specificity of functional dilatation can be ascribed partly to the additive dilator effects of all metabolites present. It also reflects the fact that absolute interstitial osmolality is a powerful vasodilator influence regardless of the identity of the osmotically active particles. In addition, it is possible that certain metabolites may have far greater biological efficacy when present at the outer surface of the vascular media than when presented in the bloodstream as is common in experimental studies.

The effect of metabolites is amplified in some tissues by several other vasodilator influences that are proportional in intensity to tissue activity. The most widespread of these is mediated by a population of ion channels on vascular smooth muscle cells that open in response to locally reduced dissolved oxygen tension; that is, that function as hypoxia receptors (Quayle et al 2006). Opening of these channels results in cell hyperpolarization and relaxation, so falls in interstitial oxygenation following increased tissue oxygen consumption have a similar effect to that of increased metabolite accumulation. Interestingly, the hypoxia receptors in pulmonary arterioles produce local vasoconstriction rather than dilatation, a feature that is very important in regulation of gas exchange in the lung (see Chapter 8, p. 96).

Additional dilator factors are liberated during metabolic activity from particular tissue cell types. These include potassium ions, NO and PGs in the case of striated muscle, short-chain peptides termed kinins in the case of exocrine glands and secretomotor peptide hormones in the case of the gastrointestinal tract. Further information on these factors and how they may interact during the circulatory adjustments to exercise will be found in the section Functional hyperaemia during exercise, p. 72.

MYOGENIC CONTROL OF LOCAL CONDUCTANCE

Stretching a vascular smooth muscle cell causes graded depolarization of the cell membrane and this opens voltage-dependent calcium channels. As a result, the level of free intracellular Ca²⁺ rises and the cell contracts; a phenomenon known as the myogenic response. Experimental studies using isolated arterial segments suggest that the amount of membrane stretch in arterioles due to normal levels of intravascular pressure depolarizes the cells by 20–30 mV, which is sufficient to cause 40–50% constriction relative to their passive diameter (Dora 2005). Thus, a substantial proportion of ongoing total peripheral resistance related to blood pressure maintenance is due to the myogenic response, independent of vasoconstrictor influences by sympathetic nerves or circulating hormones.

As well as contributing to resting peripheral resistance, the myogenic response is elicited whenever intravascular pressure changes. An increase in arterial perfusion pressure will result in depolarization and arteriolar constriction and, conversely, decreased perfusion pressure will cause hyperpolarization and dilation. The molecular details of the process that monitors membrane distortion are still unclear. The end-result, however, is that the contractile state of the cell does not alter to maintain a stable vessel diameter in the face of altered perfusion pressure. Rather, increased pressure results in a net decrease in diameter and decreased pressure in a net increase. This implies that what is being monitored is not membrane distortion itself, but wall stress.

AUTOREGULATION

From the pressure/flow relationships discussed in Chapter 5, you would expect changes in perfusion pressure gradient through a regional vascular bed to produce approximately proportionate changes in blood flow. In skeletal and cardiac muscles, brain, gastrointestinal tract and kidney, however, this occurs only with extremely low or high pressures. Over the range typical of most normal circumstances (from somewhere around 60–70 mmHg up to around 150 mmHg) flow changes transiently when pressure is altered but returns over 20–60 s to its previous value, so that flow is effectively independent of pressure over this so-called autoregulatory pressure range (Fig. 6.2).

Figure 6.2 In a microcirculatory vascular bed that exhibits autoregulation, flow remains constant over a range of perfusion pressures that is typically around 60–150 mmHg (A). The absolute value at which flow is regulated is dependent on tissue metabolic rate, with moderately increased metabolism causing increased flow without altering the autoregulatory range (B). At pressures outside the autoregulatory range, the pressure–flow relationship approaches that in rigid tubes (C), unless pressure falls below the critical closing pressure (CCP).

Clearly, maintenance of flow in the face of altered pressure requires adjustment of vascular resistance. The relationships between perfusion pressure and release of endothelial dilator factors, tissue metabolite production and myogenic responses, as discussed above, suggest that any or all of these processes could provide the basis for the autoregulatory response.

In cardiac and skeletal muscles, brain and gastrointestinal tract, changes in local metabolite concentration appear to be the dominant mechanism. At any time, absolute microvascular resistance reflects the ambient metabolite concentration. This is a balance between metabolite production and the rate at which these molecules are being washed out of the tissue by the blood flow. If perfusion pressure is altered, the rate of washout will change, resulting in accumulation of metabolites and reduced vascular resistance if perfusion pressure falls or depletion of metabolites and vasoconstriction if pressure rises.

It is uncertain to what extent endothelial and myogenic processes participate in the autoregulatory response in muscle, brain and digestive tract in vivo. Both are well suited to the task: the myogenic response involves identical effects of perfusion pressure on vascular resistance, as does autoregulation, and the scenario outlined above for the effects of altered blood flow on metabolite washout could be postulated to apply equally to washout of endothelium-derived dilator factors, such as NO. These processes may contribute to the rapidity of autoregulatory adjustments, but are unlikely to be essential to autoregulatory responses in the tissues listed above, since these all have relatively high metabolic rates and, therefore, the interstitial levels of metabolites are subject to rapid change in the face of altered perfusion.

The renal circulation also exhibits powerful autoregulation. Here, however, autoregulation is needed not to maintain metabolism, but because a high blood flow is needed for continual removal of waste products from the bloodstream. Since this rate of blood flow is far in excess of the metabolic needs of the renal tissue, metabolites could not transduce the autoregulatory signal. It is thought that renal autoregulation relies primarily on the myogenic response and, consistent with this, the latency of renal resistance responses to altered perfusion pressure is considerably faster (less than 20 s) than is seen typically in other autoregulating vascular beds.

VASOMOTION

Although autoregulation can maintain stable total organ blood flow despite altered perfusion pressure, provided that metabolic rate remains constant, this does not necessarily mean that the organ is perfused evenly. When the tissue cells are inactive, as in resting skeletal muscles, blood flow may be so low that intravascular pressure within the microcirculation is not sufficient to prevent some arteriolar vessels collapsing under the constraints of Laplace’s law (see Chapter 5, p. 52). Under these circumstances there is intermittent perfusion of neighbouring portions of the capillary bed, a process known as vasomotion.

The phenomenon of vasomotion relies on the fact that absolute critical closing pressure (CCP) is inversely related to the local concentration of interstitial metabolites. The sequence of events involved is most easily visualized by simplifying the vasculature to one arteriole giving rise to two sequential metarterioles and capillaries (Fig. 6.3) (in real life, three or more sequential metarterioles would be present). Imagine that, initially, the interstitial metabolite concentration is homogenous so that both metarterioles have similar CCPs. That which is more distally located along the parent arteriole will be most likely to close, since perfusion pressure must be lower in that region. A situation will, therefore, be created where the upstream capillary is perfused while the more distal capillary has no flow (Fig 6.3A). This will progressively wash metabolites out of the interstitium around the perfused capillary and allow metabolite accumulation around the non-perfused capillary (Fig. 6.3B). In parallel, the relaxant effect of the metabolites will change so that CCP of the perfused metarteriole rises and that of the non-perfused vessel falls. Eventually, CCP in the non-perfused vessel falls below the perfusion pressure so that flow is re-established, while CCP in the perfused vessel will rise above its perfusion pressure and so that metarteriole will shut (Fig. 6.3C–D).

Figure 6.3 Part of a nutritional vascular bed, with two metarterioles branching off an arteriole and each giving rise to a capillary. (A) The interstitial fluid throughout the region initially contains an even, low concentration of metabolites (shading). With this degree of metabolic dilation of the metarteriolar smooth muscle the absolute input pressure to metarteriole 2 is less than its critical closing pressure, so capillary 2 is not perfused. (B) As a consequence, the local metabolite concentration rises, while metabolites are progressively washed out from the interstitium around capillary 1. (C) When local metabolite concentrations change sufficiently, critical closing pressure in metarteriole 2 becomes less than the input pressure and capillary 2 opens, while metarteriole 1 shuts as its critical closing pressure exceeds the input pressure. (D) As a result, local metabolites begin to accumulate around capillary 1 and are washed out from the area around capillary 2, beginning the sequence again.

This intermittency of local perfusion takes place typically with a cycle time of 10–30 s, depending on the absolute perfusion pressure and the absolute tissue metabolic rate. Consequently, only around 30% of a resting skeletal muscle is perfused at any one time. Only when metabolic rate rises so far that CCP in all the metarterioles falls below arteriolar perfusion pressure will the entire capillary bed of the tissue be perfused and only under those circumstances of full capillary recruitment can we assume with accuracy that there is a near-linear relationship between blood flow and delivery of blood-borne solutes to specific regions of the tissue. In some vascular beds, however, this may still not mean that there is a proportionate relationship between perfusion pressure and organ blood flow, due to the presence of autoregulation.

It can be envisaged that metarteriolar muscle tone may be subject to tonic dilator influence of EDRFs as well as that of extravascular metabolites. Under these circumstances, EDRFs would accumulate in vessels without blood flow and be depleted by washout in vessels with flow, with the same end result of fluctuating critical closing pressure as has been discussed above. It is unknown whether endothelial factors do contribute significantly to vasomotion, but the relatively long duration of the oscillations could be argued as being more consistent with varying metabolite build-up than with varying EDRF clearance.

It is worth remembering that when vascular smooth muscle is maximally relaxed, CCP approaches zero. Thus, during exercise, the metarterioles in active skeletal muscles will all be open despite the presence of sympathetic vasoconstriction in the arteriolar bed further upstream. As is discussed in Chapter 8 (p. 94), this allows a substantial increase in nutritional muscle perfusion without the need for abolition of local sympathetic tone.

FUNCTIONAL HYPERAEMIA DURING EXERCISE

Skeletal muscle circulation

During exercise, nutritional blood flow through the active skeletal muscles rises by 10–20-fold from its resting value of 2–5 ml/min/100 g. This functional hyperaemia reaches a plateau within 2 s of muscle contraction beginning and is maintained throughout the exercise. Although a number of dilator metabolites are known to be produced within contracting muscles, including protons, adenine breakdown products and lactate, none of these substances alone appears to be the primary mediator of the functional hyperaemia, since in no case does the timecourse and concentration profile of metabolite appearance in the interstitium parallel exactly the timecourse and magnitude of hyperaemia (Clifford & Hellsten 2004).

It is likely, therefore, that the full hyperaemic effect represents contributions from a number of metabolites, at least partly through their combined effect on interstitial osmolality. Not even this, however, can explain the rapidity with which blood flow rises at the start of exercise, since no significant release of metabolites has been detected over the first few seconds of contraction. This early phase of hyperaemia seems to be due entirely to release of potassium ions associated with each muscle action potential. Such an effect may seem counter-intuitive since increased extracellular potassium is traditionally thought of as causing depolarization and activation of muscle cells. At concentrations up to around four times the normal extracellular level of 4 mMol; however, potassium selectively relaxes vascular muscle cells by activating a population of inwardly rectifying (K_IR) channels that cause hyperpolarization.

Other sets of factors also help bring about hyperaemia in exercising muscle, although it is not possible to estimate how much each contribute to blood flow in different patterns of exercise. One is circulating adrenaline (epinephrine), which may activate dilator β-adrenoceptors on the arteriolar muscle, although the effect of this is limited by the presence also of α-adrenoceptors (see Chapter 7, p. 85). As well, contraction leads to release from muscle cells of NO and PGs, more prominently in oxidative than in glycolytic fibres. Thirdly, there is a direct relaxant effect on vascular smooth muscle of the increased tissue temperature associated with increased metabolism (McMeeken & Bell 1990).

These effects on arteriolar tone are supplemented by dilation of the vessels immediately upstream, including the most distal conducting arteries. Despite the arterioles contributing most to regional vascular resistance, arterial dilation reduces regional resistance somewhat and so enhances absolute tissue perfusion. The dilation of conducting arteries appears to involve retrograde propagation of a relaxing signal from the arteriolar level by direct coupling between cells in the vessel wall. The mechanisms underlying this conducted (or propagated) dilation are not well-defined, but may involve endothelial NO and ATP-sensitive potassium channel activation (Takano et al 2005). Finally, the increased blood flow due to all these factors will increase endothelial shear stress in the conduit arteries supplying the active muscles, producing local release of NO and PGs, and reducing overall regional vascular resistance further.

Here is our flow chart (Fig. 6.4) of the matrix of responses to acute exercise, updated to include the effects of these local vasodilator processes.

Figure 6.4 Expansion of Figure 4.2, with the factors discussed in this chapter denoted in red.

Questions for revision