Exposure to a cool environment (body cooling) causes an increase in sympathetic nerve fiber activity to the human cutaneous circulation (Chap. 4). This increase in sympathetic discharge or nerve fiber activity is similar in control subjects and individuals with RP [32]. However, the constrictor response to that sympathetic stimulation [and to the sympathetic neurotransmitter norepinephrine [48]] is increased in RP compared to control individuals, and this increased reactivity is thought to be responsible for the enhanced cold-induced vasoconstriction summarized in section “Cutaneous Vascular Responses to Cold Exposure in RP.”

Increased reactivity of the RP cutaneous circulation to sympathetic stimulation is especially evident in the important regulation of nutritional blood flow. Increased sympathetic activity during body cooling does not impact finger nutritional blood flow in control subjects, but causes a significant reduction in individuals with RP (section “Cutaneous Vascular Responses to Cold Exposure in RP”). This reduction in RP nutritional blood flow was prevented and nutritional flow restored to normal by intra-arterial treatment with low doses of reserpine, which blocks the release of norepinephrine from sympathetic nerve fibers [19]. When control subjects were exposed to the same cool environment, there was no change in nutritional blood flow and no significant effect of inhibiting norepinephrine-induced vasoconstriction on nutritional flow (inhibition of α₁-ARs or α₂-ARs) [21]. Therefore, sympathetic-mediated constriction expands from predominantly AVAs in normal individuals to additionally encompass the nutritional vasculature in individuals with RP. Indeed, nutritional blood vessels of individuals with secondary RP (SSc) have a marked increase in constrictor activity of smooth muscle α₂-ARs, but no change in activity of α₁-ARs [36].

The digital artery vasospasm occurring in RP (section “Cutaneous Vascular Responses to Cold Exposure in RP”) is also dependent on sympathetic nerve activity. Reflex increases in sympathetic activity in response to body cooling can precipitate vasospasm in the absence of local cooling, especially in those with secondary disease [12]. However, the vasospastic potential of sympathetic activity in RP is dramatically increased by local finger cooling, and is more marked in individuals with secondary compared to primary forms of the disorder [12]. In contrast, when sympathetic nerve activity is decreased by body warming, the constrictor response to local finger cooling in RP is dramatically reduced [12]. Therefore, vasospasm of digital arteries in RP appears to reflect an increased activity of these blood vessels to sympathetic stimulation and also an increased cold-induced amplification of the response, compared to control subjects [12]. Local cooling of the fingers also amplified sympathetic-mediated constriction of hand blood flow (in response to neck cooling) to a significantly greater degree in RP compared to control subjects, which the authors proposed was mediated by a greater cold-induced amplification of α-AR constriction in RP [67].

Within the local sympathetic process, key mediators of cold-induced amplification are α2-ARs located on the cutaneous smooth muscle cells, which have increased constrictor activity at cool temperatures (Chap. 4). However, assessing smooth muscle constrictor activity of α₂-ARs within the intact cardiovascular system is a daunting task. Because of inhibitory prejunctional α2-ARs on vascular sympathetic nerves, α₂-AR inhibition will augment sympathetic transmission, increase the release of norepinephrine and promote vasoconstriction (see Chap. 4). Likewise, α₂-AR agonists will reduce sympathetic transmission and promote vasodilatation. These responses are the opposite of effects resulting from the inhibition or activation of smooth muscle α₂-ARs. This divergence is especially evident following systemic administration of α₂-AR ligands. α₂-ARs located in the central nervous system powerfully inhibit the sympathetic outflow [15]. Indeed, the preferential α₂-AR agonist clonidine was used clinically to treat hypertension, and reduce blood pressure by inhibiting sympathetic activity [110]. Likewise, oral clonidine was evaluated as a potential therapy for RP resulting from hand arm vibration syndrome (HAVS), with the goal of reducing the heightened sympathetic activity in digital arteries [112]. However, clonidine actually provoked RP in some individuals [112]. The observed effects of α₂-AR agonists and antagonists will be dependent on the activity of sympathetic nerves, and on whether they are administered locally or systemically. Smooth muscle constrictor activity of α2-AR agonists is best observed when sympathetic activity is low. Likewise, inhibition of sympathetic constriction by α₂-AR antagonists will occur only if the response is mediated predominantly, if not exclusively by α₂-ARs [39]. The analysis is further complicated by potential deficiencies in the selectivity and activity of the agonists or antagonists. For example, clonidine is a partial agonist with weak activity at α₂-ARs and very weak activity at α_2C-ARs [36]. Where possible, high efficacy agonists should be used (e.g., brimonidine—UK14,304) [36].

Local selective inhibition of α₂-ARs abolished cold-induced vasospastic attacks in individuals with primary RP, whereas inhibition of α₁-ARs had no effect [45]. In primary RP, the low finger blood flow during exposure to a cool environment was dramatically increased by local selective inhibition of α₂-ARs (sevenfold) or combined non-selective inhibition of α₁-ARs and α₂-ARs (sixfold), whereas local selective inhibition of α1-ARs had a much smaller effect (twofold) [20]. In a separate study, under thermoneutral conditions, α₁-AR inhibition had a greater effect to increase blood flow in controls compared to primary RP, such that the difference in blood flow between controls and RP actually increased [22]. In contrast, α₂-AR inhibition caused greater dilatation in RP compared to controls and therefore diminished the differences in blood flow between these groups [22]. In individuals with primary RP, inhibition of α₁-ARs or α₂-ARs prevented vasoconstriction in finger blood flow to moderate local cold exposure [22]. This contrasts with other studies where α₁-AR blockade did not prevent cold-induced vasoconstriction in human finger blood flow in normal or RP subjects [18, 30]. In individuals with secondary RP (HAVS), cold-induced vasoconstriction was abolished after local selective inhibition of α₂-ARs [85].

Molecular expression of α-ARs on RP cutaneous arteries has not been determined. However, expression of α₂-ARs on circulating platelets is increased in individuals with primary RP compared to controls [28, 69]. Furthermore, the constrictor activity of smooth muscle α₂-ARs is dramatically and selectively increased in RP (SSc) compared to control arteries, which would be consistent with increased receptor expression [36] (Fig. 5.3). Similarly, when administered intra-arterially in a thermoneutral setting, the preferential α₂-AR agonist clonidine caused vasoconstriction in human fingers that was significantly increased in primary RP compared to control subjects [20] (Fig. 5.3). In contrast, constriction to the preferential α₁-AR agonist phenylephrine was similar between the two groups [20] (Fig. 5.3). Other studies confirmed this increased responsiveness to clonidine in primary RP compared to control subjects, although they also observed increased responsiveness to phenylephrine [49, 50]. When assessed during body heating (to reduce sympathetic activity), local cooling increased the constriction of finger blood flow to the preferential α₂-AR agonist clonidine in individuals with primary RP, whereas in control subjects local cooling actually inhibited responses to the agonist [49]. Local cooling did not affect constriction to the preferential α₁-AR agonist phenylephrine in controls or RP subjects [49]. In contrast to these analyses of finger blood flow, when constrictor activity was assessed directly in non-glabrous skin, norepinephrine or the α₂-AR agonist BHT933 evoked cutaneous vasoconstriction that was similar in control and secondary RP individuals (HAVS), whereas phenylephrine caused constriction that was reduced in RP (HAVS) [29].

Despite the caveats surrounding analysis of smooth muscle α₂-ARs, the ability of selective α₂-AR antagonists to prevent the vasospastic attacks of RP and to inhibit cold-induced vasoconstriction of finger blood flow in RP strongly implicates increased activity or expression of smooth muscle α₂-ARs in the pathogenesis of this disorder. This conclusion is reinforced by the increased constrictor activity of the preferential α₂-AR agonist clonidine in individuals with primary RP and increased contractile response to the highly selective α₂-AR agonist UK14,304 in secondary RP arterioles (SSc). Increased activity of α₂-ARs would explain the increased constrictor activity of the sympathetic nervous system (e.g., in response to body cooling), and the increased cold-induced amplification of that constriction in RP (Fig. 5.4). A key component of RP pathogenesis is the expansion of powerful sympathetic-mediated vasoconstriction from predominantly AVAs in control subjects to digital arteries and arterioles in individuals with RP (Fig. 5.5). This expansion of sympathetic constriction, which is also likely to be mediated by increased activity of smooth muscle α₂-ARs, explains the interruption in nutritional blood flow that occurs in primary RP and to a greater degree in secondary RP (SSc) (Fig. 5.5) (see also section “Endothelial NO”).

Endothelial cells respond to numerous stimuli by increasing production of the powerful vasodilator and protective mediator NO, as well as secondary vasodilators such as prostacyclin (Chap. 4) (Fig. 5.4). Indeed, even norepinephrine acting on endothelial α₂-ARs can initiate endothelium-dependent vasodilatation [40]. Although endothelial α₂-ARs are most active in the coronary circulation, they have the potential to initiate dilatation in cutaneous arteries [23, 40]. The physical action of the blood stream is a key endothelial activator, with increasing blood flow causing increased production of endothelial dilators and endothelium-dependent dilatation [26, 44].

Endothelial NO activity is generally reduced in vascular disease resulting in a diminution in its protective activity and precipitating pathological functional and structural changes in the blood vessel wall, including enhanced constriction, vascular cell death (apoptosis), vascular inflammation, and vascular remodeling [34, 43, 84, 140]. Most studies have found that endothelium-dependent dilatation is normal in individuals with primary RP, including in response to increased flow (flow-mediated dilatation) [2, 73, 88, 114]. However, some reports have reported increased and decreased endothelial responses [71, 72], although this variation could reflect in part differences in gender and/or age between the experimental groups. In individuals with RP secondary to SSc, studies have consistently demonstrated impaired vasodilatation to endothelial activators including flow-mediated vasodilatation, and that this dysfunction occurs early and worsens as the disease progresses [1, 2, 46, 47, 80, 82, 83, 88, 93, 116, 117] (Fig. 5.4). Vasodilatation to endothelium-independent, direct smooth muscle dilators (NO-mimics: nitrovasodilators) may be preserved [46, 47, 128] or reduced in SSc [2, 80, 82, 83, 93, 117]. Therefore, the decrease in SSc endothelial dilatation may progress from dysfunction localized to the endothelium, to pathological changes throughout the blood vessel wall including structural changes that limit vasodilatation. These pathological changes in the SSc cutaneous circulation are part of a widespread systemic vasculopathy [4, 6, 25, 132, 133] (section “Structural Changes in Cutaneous Circulation”) (Fig. 5.4).

Endothelial dilatation is likely to have an especially important role in glabrous skin including the finger circulation, where high blood flow through AVAs will expose proximal digital arteries to high shear stress and endothelial activation. Interestingly, RP is restricted to areas that are rich in AVA, with digital artery vasospasm occurring over a similar temperature range as cold-induced closure of AVAs (see section “Cutaneous Vascular Responses to Cold Exposure in RP”). The closure of AVAs, by reducing shear stress-induced activation of digital artery endothelium, may shift the balance towards sympathetic vasoconstrictor activity. Although diminution in shear stress-induced endothelial activity would also occur in control subjects, they would continue to be protected from vasospasm by a reduced sympathetic stimulus for digital artery constriction [12, 141] (Fig. 5.5).

Alterations in endothelial-dependent dilatation also likely contribute to the differing influence of cold exposure on nutritional blood flow in primary RP and secondary RP (SSc). As discussed above, during cold exposure, nutritional blood flow is markedly disrupted in secondary RP (SSc) compared to the minor impairment observed in primary RP. Endothelial flow-mediated vasodilatation enables dilation in distal compartments of the circulation (e.g., terminal arterioles) to be conducted upstream and provide graded and targeted increases in blood flow (see Chap. 4). Therefore, despite the remarkable whitening in primary RP, metabolic vasodilation (Chap. 4) in nutritional arterioles will provide an endothelial dilator stimulus to maintain nutritional blood flow (albeit somewhat restricted because of profound arterial constriction). This low-level blood flow will quickly exit the constricted venous system and therefore not influence the pallor of the skin (see section “Vascular Mechanisms Underlying the Characteristic Color Changes of RP”). In contrast to primary RP, there is severe endothelial dysfunction in SSc. Therefore, although metabolic activity may dilate the local SSc nutritional arterioles, reduced flow-mediated dilatation will prevent this dilation from being conducted upstream to proximal arteries, which will continue to markedly restrict blood flow. Therefore, unlike primary RP, there will be a more profound interruption in SSc nutritional blood flow, precipitating SSc tissue injury.

A reduction in endothelial NO activity would also be expected to amplify vasoconstrictor episodes in SSc arteries, and to promote further pathological changes in the vascular system including thrombotic and inflammatory signaling and vascular remodeling. Indeed, expression of cold-sensitive α_2C-ARs in human cutaneous smooth muscle cells is amplified by inflammatory stress [15–17].

Endothelin-1 (ET1) is a powerful vasoconstrictor, inflammatory and fibrotic mediator that can be produced and stored (as its precursor, Big ET1) within the vascular endothelium [56]. Following appropriate stimulation, endothelial cells quickly release stored proteins including Big ET1 by exocytosis [56]. Big ET1 is rapidly converted to ET1 during this exocytotic process [56]. There has been considerable interest in the potential role of ET1 in contributing to the vasospastic attacks of RP and to pathological vascular and tissue remodeling in SSc. Indeed, endothelial exocytosis can apparently be activated by exposure to cool temperatures [152]. However, under normal physiological conditions, the endothelium neither synthesizes nor releases sufficient quantities of ET1 to initiate vasoconstriction [56]. Furthermore, NO is a powerful endogenous inhibitor of ET1 expression and of endothelial exocytosis [56]. Although early research suggested that circulating ET1 levels were higher in individuals with primary RP compared to controls, and were further increased by local cold exposure [153], subsequent studies revealed conflicting results and ET1 is not thought to be involved in the vasospastic episodes of RP [27, 125, 146].

Endothelial exocytosis mediates the release of other stored proteins including ULVWF (ultra large von Willebrand factor), which promotes hemostasis and thrombosis [13, 56]. ULVWF is unfurled by blood flow and is subsequently cleaved by an endothelial surface protease ADAMTS13 to generate (circulating) VWF fragments with reduced thrombotic potential [13]. Release of ULVWF occurs in parallel with the release/generation of ET1 [56]. In controls and in individuals with primary RP, there is minimal evidence for endothelial exocytosis of ULVWF in cutaneous blood vessels [75]. In contrast, there is exuberant release of ULVWF in SSc superficial skin blood vessels with concomitant loss of ULVWF storage in the associated endothelium [75] (Fig. 5.4). The extent of endothelial exocytosis is related to SSc disease progression [75]. A similar pattern is evident with the circulating level of VWF, which is normal in primary RP but significantly increased in individuals with SSc and correlates with disease severity [9, 57, 64, 89, 90]. The circulating level of VWF is not influenced by cold exposure [9]. Increased exocytosis of ULWF, which will be a powerful stimulus for thrombosis and vascular inflammation, may reflect diminished activity of NO in SSc (Fig. 5.4). The contents of endothelial storage granules can be regulated to increase expression of pathological mediators, including Big ET1. Indeed, the expression and storage of Big ET1 is dramatically increased in vascular aging [56]. There is increased presence of ET1 in cutaneous blood vessels of SSc subjects compared to control individuals, and as occurred with ULVWF release, ET1 is localized predominantly to superficial blood vessels [138]. Therefore, as occurs during aging, SSc endothelium likely has increased expression of ET1 precursors, with ET1 being formed during endothelial exocytosis (Fig. 5.4). Indeed, individuals with secondary RP (including SSc) have markedly increased circulating levels of both VWF and ET1, with close correlation between them [119, 139]. Short-term treatment with bosentan, which antagonizes ET_A (mediates smooth muscle constriction) and ET_B receptors (mediates endothelial generation of NO, and sometimes smooth muscle constriction), has shown promise in significantly reducing the formation of new ulcers in SSc subjects, although it does not significantly influence vasospastic attacks of RP [76, 92, 106]. This may reflect an increased role of ET1 in the nutritional microcirculation rather than in proximal digital arteries, which is consistent with the increased prominence of endothelial exocytosis in more superficial blood vessels.

There have been promising preliminary results obtained with botulinum toxin A (botox) in treating severe RP (including SSc), reportedly improving pain, perfusion and ischemic lesions [51, 104, 105, 121, 135, 137]. It remains unclear how this agent might mediate beneficial effects in this disease spectrum. Botulinum toxins specifically inhibit the molecular machinery involved in exocytosis, targeting macromolecular “SNARE” complexes involved in vesicle fusion with the plasma membrane [97, 144]. They can therefore inhibit the exocytotic release of neurotransmitters from nerve fibers. Botulinum toxin A comprises a heavy chain (Hc) and light chain (Lc), with the Hc involved in the binding and internalization of the toxin and the Lc subsequently degrading the SNARE protein SNAP25 [97, 144]. Cell surface receptors for the toxin comprise high affinity proteins (SV2, synaptic vesicle protein 2) and low affinity glycolipids called gangliosides [97, 144]. The toxin is thought to have preferential activity to inhibit cholinergic neurotransmission (including the neuromuscular junction in skeletal muscle) because of a higher expression of SV2 receptors [97, 144]. Indeed, when administered in low doses to human non-glabrous skin (forearm), botulinum toxin A selectively inhibited the vasodilatation and sudomotor (sweating) responses caused by heat stress-induced activation of the sympathetic cholinergic system [70, 122] (see Chap. 4 for description). In contrast, it had no effect on the sympathetic adrenergic cutaneous vasoconstriction caused by body cooling or the neuropeptide-mediated vasodilation evoked by local warming [70, 122]. However, when used in higher concentrations, the toxin can access sympathetic nerves to cause proteolytic degradation of SNAP25 and inhibition of sympathetic neurotransmission [98, 99, 126]. Indeed, botulinum toxin A also powerfully inhibits endothelial exocytosis [108], and will therefore inhibit the release of ULVWF and ET1. Although cold-induced translocation of smooth muscle α₂-ARs involves fusion of transport vesicles with the plasma membrane [100], it is not clear if this pathway involves SNARE complexes. The potential therapeutic effects of botulinum toxin A will therefore be determined by the dose and route of administration, but also the ability of the toxin to access target cells and the role of SNAP25 in mediating vesicle fusion in those cells. Based on available data, the most likely therapeutic targets include sympathetic-mediated vasoconstriction and endothelial exocytosis of ET1 and ULVWF. Endothelial production of NO is not dependent on exocytosis and should not be influenced by the toxin. However, the toxin may have a negative impact on responses to body warming (cholinergic dilatation, sweating) and local skin warming (neuropeptide dilatation).

Although the vasospastic episodes of RP reflect abnormal thermoregulatory responses, we lack insight into thermoregulation in these individuals. During cold exposure, processes are initiated to restrict heat loss (cutaneous vasoconstriction) and to generate heat (thermogenesis) (see Chap. 4). Analysis of the RP population has so far been restricted to the process of cutaneous vasoconstriction and reduction in heat loss, with no studies assessing thermogenesis (including shivering and activation of brown adipose tissue, BAT). Activation of thermogenesis has the potential to diminish the role of heat conservation and cutaneous vasoconstrictor mechanisms. Indeed, if perivascular adipose tissue (PVAT, Chap. 4) of cutaneous blood vessels can participate in thermogenesis, then such local heat production could diminish the influence of local cold exposure on cutaneous vasoconstrictor mechanisms. Despite the exuberant cold-induced cutaneous vasoconstriction in RP, core temperature was actually lower in RP in a thermoneutral environment and fell more during a reduction in ambient temperature, when compared to control subjects [58, 59]. However, core temperatures of control and RP individuals were similar in a warm environment [58]. This preliminary analysis suggests that thermogenic responses may be impaired in RP.

The density of CGRP-containing nerve fibers in finger skin (non-glabrous) is reported to be reduced in individuals with primary RP and further decreased in SSc [11]. Although CGRP is a vasodilator and has specific efficacy in terminating ET1 activity [81, 94, 95], no studies have provided evidence that a deficiency in this mediator contributes to heightened vasoconstriction in RP. As discussed in Chap. 4, activation of CGRP-containing nerve fibers appears to be involved in the immediate cutaneous vasodilatation to local warming [68]. However, cutaneous blood vessels of RP subjects dilate more to local warming than control subjects, with SSc subjects displaying an intermediate or reduced response [141, 147]. Therefore, there appears to be no functional deficit in the neuropeptide response to warming in primary RP, and the dysfunction present in SSc may represent a structural limitation of the SSc vasculature (see section “Structural Changes in Cutaneous Circulation”).

Increased activity of reactive oxygen species (ROS) may play an important role in acute vasospastic episodes of RP and in the vascular and tissue remodeling occurring in SSc [1, 62, 146]. Increased ROS activity contributes to cold-induced cutaneous vasoconstriction by stimulating smooth muscle α_2C-AR mobilization and potentially by inhibiting NO dilatation, impairs the protective role of the endothelium, and participates in vascular and tissue fibrosis [5, 35, 62, 146].