Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2006 May 25;574(Pt 3):849–857. doi: 10.1113/jphysiol.2006.109884

The involvement of nitric oxide in the cutaneous vasoconstrictor response to local cooling in humans

Gary J Hodges 1, Kun Zhao 1, Wojciech A Kosiba 1, John M Johnson 1
PMCID: PMC1817728  PMID: 16728451

Abstract

Cutaneous vascular conductance (CVC) declines in response to local cooling (LC). Previous work indicates that at least part of the vasoconstrictor response to LC may be through an inhibitory effect on nitric oxide synthase (NOS) activity. In this study we further tested that notion. A total of eight (6 male, 2 female) subjects participated (Part 1 n = 7; Part 2 n = 5, 4 of whom participated in Part 1). Skin blood flow was monitored by laser-Doppler flowmetry. Control of local skin and body temperatures was achieved with Peltier cooler/heater probe holders and water perfused suits, respectively. Microdialysis fibres were inserted aseptically. Saline, l-NAME (20 mm; to inhibit NOS activity) and sodium nitroprusside (SNP 10 μm) were infused by microdialysis. Bretylium tosylate (BT), to block adrenergic function, was administered by iontophoresis. CVC was calculated from blood flow and blood pressure. Part 1 was designed to determine the relative roles of the NO and the adrenergic systems. The infusion of l-NAME elicited a 35 ± 4% decrease in CVC at the l-NAME and BT +l-NAME sites (P < 0.05); subsequent slow LC (34–24°C) for 35 min caused a significant (P < 0.05) decrease in CVC at control sites (68 ± 4%) and at the BT treated sites (39 ± 5%). LC caused a further 23 ± 5% of initial baseline decrease in CVC at the l-NAME treated sites (P < 0.05). Importantly, CVC at the BT +l-NAME sites was unaffected by LC (P > 0.05). Part 2 was designed to test whether LC influences were specific to the NOS enzymes. Two sites were pretreated with both BT and l-NAME. After 50 min, SNP was added as an NO donor to restore baseline CVC at one site. The same LC process as in Part 1 was applied. There was a 24 ± 10% decrease (P < 0.05) in CVC at sites with baseline CVC restored, while, as in Part 1, there was no change (P > 0.05) at sites treated with BT +l-NAME only. These data suggest that the vasoconstriction with slow LC is due to a combination of increased noradrenaline release and decreased activity of both NOS per se and of process(es) downstream of NOS.

Under resting conditions, reflex modulation of blood flow to non-glabrous skin is regulated by the adrenergic vasoconstrictor system (Brengelmann & Savage, 1997; Stephens et al. 2001). Cutaneous vasoconstriction can also be induced by direct local cooling, and it is known that this response is in part adrenergically dependent (Ekenvall et al. 1988; Pérgola et al. 1993; Johnson et al. 2005; Yamazaki et al. 2006). Nevertheless, presynaptic blockade of noradrenaline release from vasoconstrictor nerve fibres does not completely eliminate the vasoconstriction elicited by local cooling (Pérgola et al. 1993; Johnson et al. 2005; Yamazaki et al. 2006), showing that there are also non-adrenergic mechanisms involved in the vasoconstrictor response to local cooling.

Venturini et al. (1999) reported that the in vitro activity of the nitric oxide synthase (NOS) enzymes (particularly neuronal NOS and inducible NOS) was reduced by mild cooling. Yamazaki et al. (2006) suggested that the non-neural portion of the vasoconstrictor response to local cooling might include a temperature-dependent inhibition of basal NOS activity. Their data indicated that temperature effects on basal NOS function might indeed be responsible – at least in part – for the remaining vasoconstriction; however, the fast local cooling used in their protocol (−4°C min−1) produced an initial and transitory vasodilatation, which hindered somewhat the interpretation of the mechanisms involved in the subsequent vasoconstriction. In vitro work indicates that the form of the NOS enzyme usually associated with tonic nitric oxide (NO) production (endothelial NOS) does not appear to be as temperature sensitive as the other forms (Venturini et al. 1999). This would suggest either that the NOS system is not involved in the non-adrenergic portion of the cutaneous vasoconstrictor response to local cooling, or that some point in the NO production scheme other than NOS per se is temperature sensitive.

Therefore, the aim of the present study was to determine the involvement of basal NOS activity and of enhanced adrenergic function on the vasoconstrictor response to local cooling in human forearm skin. To do this we performed two studies. In Part 1 we used a slow local cooling protocol, which does not produce a transient vasodilatation (Yamazaki et al. 2006), and assessed the response to local cooling in the presence of NOS inhibition, with and without vasoconstrictor nerve inhibition by BT pretreatment. In Part 2, we tested whether local cooling has a temperature-dependent effect specifically on the NOS enzymes or if processes in the NO vasodilator pathway downstream of NOS might be involved. We did this by restoring baseline skin blood flow during NOS inhibition with an NO donor, sodium nitroprusside (SNP). In essence, this approach restores the tonic vasodilator effect of basal NO levels. We then applied slow local cooling. The hypotheses being tested were (1) that removal of NO production via the inhibition of NOS would eliminate the non-adrenergic component of the vasoconstrictor response to local cooling, and (2) that following the restoration of cutaneous vascular conductance (CVC) with SNP during NOS-inhibition, local cooling would cause a vasoconstriction by affecting process(es) downstream of NOS in the NO vasodilator pathway.

Methods

Subjects

All studies were approved by the local Institutional Review Board and conformed to The Declaration of Helsinki, and all subjects were fully informed of the methods and risks before consent was obtained. A total of eight subjects participated (6 men and 2 women; age range 20–42 years; Part 1, 5 men and 2 women; Part 2, 5 men, 4 of whom participated in Part 1). All were healthy non-smokers; not taking medications, and all refrained from alcoholic and caffeinated beverages for at least 12 h prior to the study. The menstrual cycle was not considered in the experiments as in previous studies the vasoconstrictor response to local cooling was unaffected by reproductive hormone status (Charkoudian et al. 1999). Responses by men and women did not differ perceptively and their results were combined. Methods of measurement and the local cooling protocols, described below, were the same in all cases.

Instrumentation

Subjects had microdialysis probes placed intradermally on the ventral aspect of the left forearm as previously described (Crandall et al. 1997; Kellogg et al. 1999). These probes, consisted of 1 cm of microdialysis tubing (inner diameter 200 μm, 18 kDa nominal molecular weight cutoff) attached at each end to polyimide tubing. Before implantation, the area of skin was temporarily anaesthetized by the application of an ice pack for 5 min. A 25-gauge needle was introduced aseptically for ∼2.5 cm into the skin before exiting. The microdialysis probe and the connecting tubing were introduced into the skin via the lumen of the needle, and then the needle was removed, leaving the probe in place. All probes were placed in this manner, and ∼1.5 h was allowed for the effects of the insertion trauma to subside (Anderson et al. 1994). The different probes were placed 3–5 cm apart. The subject was supine during this procedure.

Measurements

All measurements were performed with the subjects resting in the supine posture. Skin blood flow was measured from the ventral aspect of the forearm by laser-Doppler flowmetry (Moor Instruments Inc., Axminster, UK), and expressed as laser-Doppler flow (LDF) (Johnson, 1990; Öberg, 1990). LDF measures are exclusive to the skin and are not contaminated by underlying skeletal muscle blood flow (Saumet et al. 1988). Local temperature control was achieved with custom-built Peltier cooling/heating metal probe holders (Johnson et al. 2005; Yamazaki et al. 2006); these controlled surface temperature over an area of 6.3 cm2 with the exception of a small aperture (0.28 cm2) in the centre of the holder to enable placement of the laser-Doppler probe. A thermocouple between the skin surface and the probe holder enabled local skin temperature assessment and control. Local skin temperature can be precisely maintained within 0.1°C. Blood pressure was recorded non-invasively and continuously by the Penaz method (Parati et al. 1989) from the left middle finger (Finapres, Ohmeda, Madison, WI, USA). Mean arterial pressure was obtained from the electrical integration of the continuous blood pressure signal. CVC, in arbitary units, was calculated as the ratio of LDF to mean arterial pressure. Whole body skin temperature was recorded as the weighted mean from six thermocouples placed on the body surface and controlled by the use of a water-perfused suit (Taylor et al. 1989). The suit covered the entire body surface apart from the head, hands, feet and the forearm used for the blood flow measurement. Whole body skin temperature was maintained at 34°C (thermoneutral), with the exception of a 3 min period of whole body cooling (30°C) to test for adequate vasoconstrictor nerve blockade (Kellogg et al. 1989). All variables were collected at 1 s intervals and stored as 20 s averages.

Vasoconstrictor nerve blockade was achieved by the iontophoresis of bretylium tosylate (BT; Schweizerhall, South Plainfield, NJ, USA) at a concentration of 10 mm at 250 μA for 10 min, to a 0.64 cm2 area of skin (Kellogg et al. 1989). Administration of BT causes a selective and localized blockade of transmitter release by the cutaneous vasoconstrictor nerves (Haeusler et al. 1979; Kellogg et al. 1989).

NOS activity was inhibited by the infusion of N-nitro-l-arginine methyl ester (l-NAME; Sigma Chemical Co., St Louis, MO, USA) via microdialysis. A solution at a concentration of 20 mm of l-NAME in sterile saline was perfused at 4 μl min−1, which after 50 min, has been previously reported to produce a complete NOS inhibition under resting conditions (Yamazaki et al. 2006).

Protocols

Part 1 involved the determination of the relative roles of NOS and the adrenergic system in the vasoconstrictor response to slow local cooling. In seven subjects, four sites were each prepared with microdialysis probes, and two of these sites were pretreated with BT. Whole body and local temperatures were maintained at 34°C (thermoneutral). The protocol began with all four microdialysis probes perfused with sterile saline solution for 15 min at 4 μl min−1 (baseline measures). The perfusate for two probes (one at a site previously treated with BT) was then changed to 20 mml-NAME, with saline solution continuing at the other two sites. As indicated in Fig. 1, this arrangement provided one site with NOS inhibition, one site with adrenergic inhibition, one site with both systems inhibited and a fourth site with both systems intact. After 50 min, when the new baseline following l-NAME administration was stable, slow local cooling (−0.33°C min−1) commenced at all sites. The temperature at all sites was reduced from 34 to 24°C over a 30 min period and maintained at the final temperature for 5 min (see Fig. 1). All sites were then rewarmed to 34°C. Once the original baseline values for CVC were restored, whole body cooling was performed (34–30°C) for 3 min to test the adequacy of the BT vasoconstrictor nerve blockade (Kellogg et al. 1989). A reduction in CVC at the BT-treated sites of > 10% was taken to indicate inadequate blockade. There were no instances of failure to achieve a blockade in this study.

Figure 1. Outline of Part 1.

Figure 1

Open in a new tab

Four sites were prepared with microdialysis probes, and two of these sites were pretreated with BT. All probes were perfused with saline for 15 min. The perfusate for two sites (one pretreated with BT) was then changed to 20 mm l-NAME, providing one site (site 3) with NOS inhibition, one with adrenergic inhibition (site 2), one with both systems inhibited (site 4) and a fourth site with both systems intact (site 1). After 50 min, slow local cooling (LC) was performed at those sites reducing temperature from 34 to 24°C over a 30 min period and 5 min were allowed for stabilization of CVC. All sites were re-warmed (RW) to 34°C. Whole body cooling (WBC) was performed (34 to 30°C) for 3 min to test the adequacy of vasoconstrictor nerve blockade by BT.

Part 2 involved the identification of whether the effects of slow local cooling were solely affecting NOS function per se or had an inhibitory effect at a stage of the NO system downstream from NOS. In five subjects, four of whom had participated in Part 1, two sites were prepared with microdialysis fibres. Both sites were pretreated with BT. Whole body and local temperatures were maintained at 34°C. As illustrated by Fig. 2, the protocol began with both microdialysis probes perfused with sterile saline solution for 15 min at 4 μl min−1 (baseline measures). The perfusate for both probes was then changed to a 20 mm solution of l-NAME (as in Part 1). After 50 min, one site was perfused with a combination of 20 mml-NAME and 10 μm SNP (Sigma Chemical Co.). Pilot studies had shown this concentration of SNP to be sufficient to restore baseline CVC following NOS inhibition; the other site continued to be perfused with l-NAME only. Once baseline CVC values were restored (∼20 min), slow local cooling to 24°C as in Part 1 was applied to both sites for 35 min (see Fig. 2). All sites were then rewarmed to 34°C. Once original baseline values for CVC were reached, whole body cooling was performed (34–30°C) for 3 min to test the adequacy of the blockade of vasoconstrictor nerves by BT; there were no instances of failure to achieve a blockade in this study.

Figure 2. Protocol for Part 2.

Figure 2

Open in a new tab

Two microdialysis sites were pretreated with BT. Both probes were perfused with sterile saline solution for 15 min. The perfusate for both probes was changed to 20 mm of l-NAME (as in Part 1). After 50 min, site 2 was perfused with a combination of 20 mm l-NAME and 10 μm SNP; l-NAME continued at site 1. Once baseline CVC values were restored (∼20 min), slow local cooling (LC) to 24°C as in Part 1 was applied to both sites for 35 min. Finally, whole body cooling was used to test the adequacy of the blockade of vasoconstrictor nerves by BT.

Data analysis

Data for CVC from Parts 1 and 2 were analysed for the final 5 min of each section (see Figs 1 and 2). CVC was expressed relative to the baseline values for each site. Absolute values for baseline CVC were used to determine whether sites were at unusually high or low values as a result of the pharmacological background or the preparatory procedures. There were no such cases in this study. Analysis was by paired statistics or, when appropriate, repeated-measures ANOVA. Statistical significance was assumed when P < 0.05.

Results

Part 1

Figure 3 shows the responses in CVC, normalized to baseline, from a representative subject for the protocol in Part 1. One site was treated with BT, the second site was treated with l-NAME, the third site was treated with both BT and l-NAME, and the fourth site was an untreated control site (saline only). Note that the large vasoconstriction with local cooling at the control site is reduced both at the site treated with BT and at the site treated with l-NAME. Importantly, note the absence of a response to local cooling at the site that had been treated with both BT and l-NAME. This was seen for the group as a whole. Figure 4 shows the average responses in CVC in Part 1 from all seven subjects. During 50 min of l-NAME infusion, CVC at the l-NAME and BT +l-NAME sites decreased significantly (P < 0.05) to 64 ± 8 and 66 ± 8%baseline, respectively. CVC levels at the control and BT-only sites (saline) were unchanged (99 ± 16 and 104 ± 17%baseline). The subsequent slow local cooling caused a significant (P < 0.05) reduction in CVC at the control site and at the BT-only site, with CVC decreasing to 32 ± 6 and 60 ± 7%baseline, respectively. Local cooling elicited a further significant (P < 0.05) reduction at the l-NAME site, with CVC decreasing to 41 ± 10%baseline, a further 23 ± 7% of baseline from the l-NAME phase. At the BT +l-NAME site, CVC was not significantly affected by the slow local cooling, from 66 ± 8 precooling to 64 ± 9%baseline at the end of local cooling (P > 0.05).

Figure 3. CVC responses – normalized to baseline – from a representative subject in Part 1.

Figure 3

Open in a new tab

One site was treated with BT (▵), the second site with l-NAME (•), the third site with both BT and l-NAME (▵), and the fourth site was untreated (saline only; ○). Note the large vasoconstriction with local cooling at the control site is reduced at the site treated with BT and the site treated with l-NAME. Importantly, note the absence of a response to local cooling at the site treated with both BT and l-NAME.

Figure 4. Responses in CVC, normalized to baseline, for all 7 subjects in Part 1.

Figure 4

Open in a new tab

l-NAME infusion caused a reduction in CVC at l-NAME only and BT +l-NAME sites to 64 ± 8 and 66 ± 8%baseline, respectively. Slow local cooling caused a significant reduction in CVC at the control site, the BT-only site, and the l-NAME-only site. At the BT +l-NAME site, CVC was not affected by the slow local cooling (n = 7). *P < 0.05 from baseline; **P < 0.05 from saline/l-NAME (ANOVA). CT = control.

Comparison by ANOVA of changes in response to local cooling relative to the original baseline, revealed all sites to differ in response. The decrease in CVC caused by local cooling at the control site (67 ± 11%) was significantly (P < 0.05) greater than the decreases in CVC at the BT-only (44 ± 12%) and l-NAME (23 ± 7%) sites. The decreases in CVC in response to local cooling at the BT-only and the l-NAME sites were statistically different (P < 0.05). The responses in CVC to local cooling at the control, BT-only, and l-NAME sites were all significantly greater than that observed at the BT +l-NAME site (2 ± 5%; P < 0.05). This suggests that the adrenergic component accounts for one-third of the response to local cooling, and the NOS component, the remaining two-thirds.

Part 2

Figure 5 shows the responses in CVC from a representative subject to the protocol for Part 2. Both sites were pretreated with BT. Figure 6 shows the average responses in CVC for Protocol 2 from all five subjects. Following the 50 min of l-NAME perfusion, CVC at both sites had decreased significantly (P < 0.05). At site 1 CVC decreased to 53 ± 6 and at site 2 to 55 ± 5%baseline. Both sites then continued to be perfused with 20 mm of l-NAME; however, site 2 also had 10 μm SNP added to the perfusate to restore CVC to the original baseline values. After ∼20 min, CVC at site 1 was unchanged (57 ± 11%baseline, P > 0.05), whereas CVC at site 2 had increased (P < 0.05) to 103 ± 20% baseline and was not significantly different from the original baseline (P > 0.05). The same slow local cooling process as in Part 1 was then applied. As in Part 1, CVC at site 1 (treated with BT +l-NAME) was unaffected (P > 0.05) by the slow local cooling. However, with the addition of SNP to restore baseline CVC at site 2, slow local cooling decreased CVC significantly (P < 0.05) to 76 ± 16% baseline. Despite the cooling-induced vasoconstriction, CVC at that site was significantly higher than after l-NAME before SNP and greater than CVC at site 1 following local cooling (P < 0.05).

Figure 5. Responses in CVC, normalized to baseline, from a representative subject in Part 2.

Figure 5

Open in a new tab

Both sites were treated with BT and l-NAME. One site (•) was further treated with SNP to restore baseline CVC. Note the absence of a response to local cooling at the BT +l-NAME site only, whereas the site with the restored baseline has a reduced CVC to a level intermediate between baseline and l-NAME treatment. SNP = sodium nitroprusside; RW = rewarming; WBC = whole body cooling.

Figure 6. Average responses in CVC from the 5 subjects participating in Part 2.

Figure 6

Open in a new tab

Both sites were pretreated with BT. l-NAME perfusion caused CVC at both sites to fall (P < 0.05). Both sites continued to be perfused with 20 mm of l-NAME; however, one site (○) also had 10 μm of SNP added in the perfusate to restore CVC to the baseline values. Slow local cooling (LC) was then applied. As in Part 1, the site treated with BT +l-NAME was unaffected by slow LC, whereas the site with the addition of SNP to restore baseline had a significant reduction in CVC to a level intermediate between baseline and that achieved with l-NAME (n = 5) *P < 0.05 from baseline; **P < 0.05 from l-NAME administration; †P < 0.05 from baseline, l-NAME and l-NAME + SNP (ANOVA).

Changes relative to baseline in response to the local cooling were assessed by ANOVA. The decrease in CVC at the BT +l-NAME + SNP site in response to the slow local cooling was significantly (P < 0.05) greater than that at the BT +l-NAME site which, as previously stated, was unaffected by local cooling (P > 0.05), suggesting that processes downstream of NOS account for ∼50% of the NO-dependent vasoconstriction to local cooling.

Discussion

These data suggest strongly that the vasoconstrictor response to this level of local cooling can be explained by the contribution of an increased release of noradrenaline from the vasoconstrictor nerves and an inhibition of the NO vasodilator system. They further suggest that local cooling acts on at least two places in the NO system: an inhibition of the NOS enzyme(s), per se, and at a site or sites downstream from NOS.

A role for the vasoconstrictor nerves in the cutaneous vasoconstrictor response to direct local cooling has been noted previously (Ekenvall et al. 1988; Pérgola et al. 1993; Johnson et al. 2005; Yamazaki et al. 2006). Indeed, Ekenvall et al. (1988) noted that the vasoconstrictor response to short-term direct local cooling of the finger was eliminated by α2-adrenergic blockade, but not by α1 blockade, indicating the α2 subtype involvement to be important in the local cooling response. A series of important studies on the specific α-adrenoceptor subtype involvement led to the proposal that local cooling elicits an increase in reactive oxygen species activity from vascular smooth muscle mitochondria, which results in activation of the RhoA and Rho-kinase systems, inducing a translocation of the α2C-adrenoceptor subtype from the Golgi apparatus to the plasma membrane (Jeyaraj et al. 2001; Bailey et al. 2004, 2005; Chotani et al. 2004). This postsynaptic receptor mobilization is complemented by an enhanced release of noradrenaline through a local reflex stimulated by local cooling (Pérgola et al. 1993; Johnson et al. 2005). Other elements of the noradrenergic system may, in fact, be inhibited by local cooling (e.g. noradrenaline synthesis) (Vanhoutte, 1980).

In human hairy skin, elimination of the noradrenergic pathway does not completely eliminate cooling-induced vasoconstriction (Pérgola et al. 1993; Johnson et al. 2005; Yamazaki et al. 2006). In an earlier study, it was suggested that decreased NOS activity caused by local cooling might be a major component of that residual vasoconstriction (Yamazaki et al. 2006); the combination of vasoconstrictor nerve blockade and NOS inhibition did indeed appear to substantially inhibit the ultimate vasoconstrictor response in CVC to rapid local cooling. However, rapid local cooling (−4°C min−1) is attended by a large, but transient vasodilator component early in the cooling process (Pérgola et al. 1993; Johnson et al. 2005; Yamazaki et al. 2006), which made final analysis of the impact of NOS inhibition on the late vasoconstriction somewhat unclear. That initial vasodilator component is revealed by presynaptic blockade of vasoconstrictor nerves, postsynaptic blockade of adrenergic receptors or local blockade of sensory nerves (Johnson et al. 2005). Although the mechanism is unknown, it is no longer evident when the local cooling is performed at a slower rate (Yamazaki et al. 2006). We took advantage of this latter observation and applied it to the same set of antagonists used earlier (Yamazaki et al. 2006). Under these conditions, it was quite clear that the inhibition of NOS activity significantly reduced the response in CVC to local cooling (Figs 3 and 4), as previous data had indicated. When noradrenaline release was also blocked, the entire vasoconstrictor response in CVC to local cooling was eliminated (Figs 5 and 6). Thus, it would appear that the vasoconstriction in response to local cooling can be explained entirely on the basis of these two systems.

The inhibition of NOS with l-NAME removes that system in its entirety from participating in vasomotor control. The NOS enzymes, however, may not be the only site of action of local cooling in the NO system. Thus local cooling may have an inhibitory role downstream from the NOS enzymes. We were able to address that question in the present investigation by re-supplying NO through SNP perfusion. In that scenario, despite the absences of functional NOS and adrenergic function, a vasoconstriction is again seen in response to local cooling. We take this as good evidence that the influence of local cooling also affects the NO system at a site(s) downstream of NOS. The data do not allow a further discrimination of those possible sites of action. Therefore, our current data suggest that local cooling has at least two sites of action in the NO pathway. One is an inhibition of the NOS enzyme itself, the other is the downstream site (or sites) discussed above. We draw this conclusion because in Part 1 when only vasoconstrictor nerves are inhibited (BT-only site), a 44% reduction in CVC is observed (Fig. 4). In Part 2, when vasoconstrictor nerves and NOS are both inhibited, and NO levels are restored by SNP infusion, local cooling reduced CVC by 24% (Fig. 6). This suggests that some of the reduction in CVC at the former sites (BT only; Fig. 4) was due to local cooling affecting both NOS enzymes and processes downstream of NOS – hence the larger decrease in CVC – with the smaller decrease at the BT +l-NAME + SNP site (Fig. 6) due to local cooling affecting only function of the NO system downstream of the NOS enzymes. In vitro studies show nNOS and iNOS to be particularly temperature sensitive over the range of temperatures used in this study (Venturini et al. 1999). We do not know if that means that eNOS has sufficient thermal sensitivity to provide the responses seen here, or if another of the NOS enzymes is important in the response to local cooling.

The level of local cooling used in the present study causes approximately a 70% reduction in CVC (Fig. 4). The present data indicate that roughly one-third of this response is due to local excitation of the vasoconstrictor nerves, with the remaining two-thirds dependent on an inhibition of the NO system. These data further suggest that local cooling effects on the NOS enzyme account for approximately 50% of the non-neural component, with the remaining ∼50% of the NO-dependent portion due to local cooling effects downstream of NOS (Fig. 6). Thus, of the reduction in CVC at the control sites, we estimate that sympathetic activation, NOS inhibition and processes downstream from NOS each contribute 33%.

The first finding of this investigation, that the cutaneous vasoconstrictor response to local cooling is dependent on an intact adrenergic system and functional NOS, can be seen in Figs 3 and 4. These clearly show no response in CVC at the BT +l-NAME sites to local cooling, whereas all other sites showed a decline. The argument could be made that the absence of a decline at this point at the site with combined blockade may have been due to the reduced baseline following the infusion of l-NAME. However, this argument fails on two counts. Firstly, the l-NAME-only site can be seen to decline further in the presence of local cooling (64 ± 8 to 41 ± 10%baseline) (Figs 3 and 4). Secondly, by removal of basal vasoconstrictor activity, skin treated with BT (i.e. no basal adrenergic function) typically displays a CVC twice that of non-BT treated areas (Kellogg et al. 1989); therefore CVC at the BT sites will actually be higher than those with functional vasoconstrictor nerves, the latter showing a vasoconstrictor response to local cooling.

It is perhaps of some interest that the cutaneous vasodilator response to local warming also has an important contribution to the NO system (Minson et al. 2001; Kellogg et al. 1999). Given the present results, it would be of interest to know whether that contribution was confined to an effect of warming on the NOS enzyme or if, as in cooling, an important part is due to thermal sensitivity at some other point in the NO vasodilator pathway. The remaining part of the vasodilator response to local warming is due to sensory nerves, as it is blocked by local anaesthetic. In contrast to local cooling, it does not appear that sensory nerves are acting on sympathetic nerves to cause this vasodilatation as the increase in CVC is far greater than could be evoked by vasoconstrictor nerve inhibition (Kellogg et al. 1999; Minson et al. 2001). Indeed, in the case of local warming, sympathetic nerves are thought to inhibit the axon reflex (Houghton et al. 2005), whereas in local cooling the sensory nerves are thought to be excitatory to the sympathetic vasoconstrictor nerves (Pérgola et al. 1993; Johnson et al. 2005).

Figure 7 shows our working model of the mechanisms of the cutaneous vasoconstrictor response to local cooling based on this study and previous investigations in our laboratory (Pérgola et al. 1993; Johnson et al. 2005; Yamazaki et al. 2006). As mentioned, Pérgola et al. (1993) reported that the vasoconstrictor response at the beginning of local cooling required functional adrenergic nerve terminals and noradrenaline release, and that the noradrenaline release can be elicited locally. There was a non-neural element to the response to prolonged local cooling. More recently Johnson et al. (2005) found that early initial vasoconstrictor response required not only intact sympathetic function, but also intact sensory function, indicating the probability of an axon reflex involvement. Yamazaki et al. (2006) found that this early vasodilator response to local cooling was not dependent on NOS, as inhibition of NOS failed to affect this phenomenon. At the present time, speculation as to what elicits this early vasodilatation is difficult. The early vasodilatation is typically observed when sympathetic or sensory function is inhibited and the skin is exposed to rapid local cooling. Slow local cooling as applied in this investigation does not cause this early vasodilatation.

Figure 7. Model of the factors contributing to the cutaneous vascular responses to local cooling.

Figure 7

Open in a new tab

Neural mechanisms are initiated by stimulation of cold receptors, which stimulate sympathetic vasoconstrictor nerves to release noradrenaline. This pathway can be interrupted by sensory nerve blockade, by sympathetic presynaptic blockade, or by sympathetic postsynaptic blockade of α- and β-adrenergic receptors. Such blockade reveals a non-neural early vasodilatation, which is succeeded by a later vasoconstriction. The mechanism for the early vasodilatation is unknown. The later vasoconstriction is via the nitric oxide vasodilator pathway and through continued excitation of vasoconstrictor nerves by an axon reflex. Not shown are the effects of local cooling on noradrenaline release and synthesis, adrenergic receptors, or vascular smooth muscle function. + Excitatory stimuli; − inhibitory stimuli

The effects illustrated in Fig. 7 are in addition to the well-accepted effects of local cooling on postsynaptic adrenergic receptors, vascular smooth muscle, and noradrenaline synthesis and release. Cooling augments the sensitivity of α2-adrenoreceptors to noradrenaline as well as increasing the number of the α2-adrenoreceptors (particularly the α2C subtype), by a translocation via a Rho/Rho kinase pathway, while reducing that of α1-adrenoreceptors, the latter effect being buffered in some tissues through receptor reserve (Vanhoutte et al. 1985; Harker & Vanhoutte, 1988; Lindblad et al. 1990; Chotani et al. 2000; Bailey et al. 2004).

Recent work by Durand et al. (2005) confirmed that, as in other vascular beds, NO can attenuate adrenergic vasoconstriction in the skin (Costa et al. 2001; Chavoshan et al. 2002; Kolo et al. 2004). However, the concentrations (8.4 × 10−3 and 8.4 × 10−4m) used by Durand et al. (2005) that caused an attenuation were high in comparison to what is present at basal levels, whereas at lower concentrations, they found no effect. In Part 2 we restored baseline CVC with the NO donor SNP following NOS inhibition with l-NAME. We found that 10 × 10−6m was sufficient to fully restore baseline CVC (Fig. 6) and thus can be viewed as restoring basal NO function. Therefore, it is unlikely under basal conditions that NO is affecting the adrenergic vasoconstrictor response in the cutaneous vasculature. Our restoration of baseline with the administration of SNP would not have affected any adrenergic function present as BT was used at all sites with SNP added, eliminating adrenergic function.

In summary, we identified that a complete vasoconstrictor response to local cooling is dependent on a functional adrenergic system and the NO system. Furthermore, local cooling appears to affect both the activity of the NOS enzyme(s), as well as process(es) downstream of the NOS enzymes in the NO vasodilator pathway. However, the data do not allow a further discrimination of those possible sites of action.

Acknowledgments

The authors thank the volunteers for their time and effort. This study was supported by the National Heart, Lung and Blood Institute grant HL-059166.

References

  1. Anderson C, Andersson T, Wardell K. Changes in skin circulation after insertion of a microdialysis probe visualized by laser Doppler perfusion imaging. J Invest Dermatol. 1994;102:807–811. doi: 10.1111/1523-1747.ep12378630. [DOI] [PubMed] [Google Scholar]
  2. Bailey SR, Eid AH, Mitra S, Flavahan S, Flavahan NA. Rho kinase mediates cold-induced constriction of cutaneous arteries: role of α2C-adrenoceptor translocation. Circ Res. 2004;94:1367–1374. doi: 10.1161/01.RES.0000128407.45014.58. [DOI] [PubMed] [Google Scholar]
  3. Bailey SR, Mitra S, Flavahan S, Flavahan NA. Reactive oxygen species from smooth muscle mitochondria initiate cold-induced constriction of cutaneous arteries. Am J Physiol Heart Circ Physiol. 2005;289:H243–H250. doi: 10.1152/ajpheart.01305.2004. [DOI] [PubMed] [Google Scholar]
  4. Brengelmann GL, Savage MV. Temperature regulation in the neutral zone. Ann N Y Acad Sci. 1997;813:39–50. doi: 10.1111/j.1749-6632.1997.tb51670.x. [DOI] [PubMed] [Google Scholar]
  5. Charkoudian N, Stephens DP, Pirkle KC, Kosiba WA, Johnson JM. Influence of female reproductive hormones on local thermal control of skin blood flow. J Appl Physiol. 1999;87:1719–1723. doi: 10.1152/jappl.1999.87.5.1719. [DOI] [PubMed] [Google Scholar]
  6. Chavoshan B, Sander M, Sybert TE, Hansen J, Victor RG, Thomas GD. Nitric oxide-dependent modulation of sympathetic neural control of oxygenation in exercising human skeletal muscle. J Physiol. 2002;540:377–386. doi: 10.1113/jphysiol.2001.013153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chotani MA, Flavahan S, Mitra S, Daunt D, Flavahan NA. Silent α2C-adrenergic receptors enable cold-induced vasoconstriction in cutaneous arteries. Am J Physiol Heart Circ Physiol. 2000;278:H1075–H1083. doi: 10.1152/ajpheart.2000.278.4.H1075. [DOI] [PubMed] [Google Scholar]
  8. Chotani MA, Mitra S, Su BY, Flavahan S, Eid AH, Clark KR, Montague CR, Paris H, Hardy DE, Flavahan NA. Regulation of alpha2–adrenoceptors in human vascular smooth muscle cells. Am J Physiol Heart Circ Physiol. 2004;286:1156–1167. doi: 10.1152/ajpheart.00268.2003. [DOI] [PubMed] [Google Scholar]
  9. Costa F, Christensen NJ, Farley G, Biaggioni I. NO modulates norepinephrine release in human skeletal muscle: implications for neural preconditioning. Am J Physiol Regul Integr Comp Physiol. 2001;280:R1494–R1498. doi: 10.1152/ajpregu.2001.280.5.R1494. [DOI] [PubMed] [Google Scholar]
  10. Crandall CG, Etzel RA, Johnson JM. Evidence of functional β-adrenoceptors in the cutaneous vasculature. Am J Physiol. 1997;273:H1038–H1043. doi: 10.1152/ajpheart.1997.273.2.H1038. [DOI] [PubMed] [Google Scholar]
  11. Durand S, Davis SL, Cui J, Crandall CG. Exogenous nitric oxide inhibits sympathetically mediated vasconstriction in human skin. J Physiol. 2005;562:629–634. doi: 10.1113/jphysiol.2004.075747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ekenvall L, Lindblad LE, Norbeck O, Etzell BM. α-Adrenoceptors and cold-induced vasoconstriction in human finger skin. Am J Physiol. 1988;255:H1000–H1003. doi: 10.1152/ajpheart.1988.255.5.H1000. [DOI] [PubMed] [Google Scholar]
  13. Haeusler G, Haefely W, Huerlimann A. On the mechanism of the adrenergic nerve blocking action of bretylium. Naunyn Schmiedebergs Arch Pharmacol. 1979;265:260–277. doi: 10.1007/BF01002340. [DOI] [PubMed] [Google Scholar]
  14. Harker CT, Vanhoutte PM. Cooling the central ear artery of the rabbit: myogenic and adrenergic responses. J Pharmacol Exp Ther. 1988;245:89–93. [PubMed] [Google Scholar]
  15. Houghton BL, Meedering JR, Wong BJ, Minson CT. Nitric oxide and noradrenaline contribute to the temperature threshold of the axon reflex response to graded local heating in human skin. J Physiol. 2006;572:811–820. doi: 10.1113/jphysiol.2005.104067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jeyaraj SC, Chotani MA, Mitra S, Gregg HE, Flavahan NA, Morrison KJ. Cooling evokes redistribution of α2C-adrenoceptors from Golgi to plasma membrane in transfected human embryonic kidney 293 cells. Mol Pharmacol. 2001;60:1195–1200. doi: 10.1124/mol.60.6.1195. [DOI] [PubMed] [Google Scholar]
  17. Johnson JM. The cutaneous circulation. In: Shepherd AP, Öberg PA, editors. Laser-Doppler Blood Flowmetry. 1990. pp. 121–139. [Google Scholar]
  18. Johnson JM, Yen TC, Zhao K, Kosiba WA. Sympathetic, sensory, and nonneuronal contributions to the cutaneous vasoconstrictor response to local cooling. Am J Physiol Heart Circ Physiol. 2005;288:H1573–H1579. doi: 10.1152/ajpheart.00849.2004. [DOI] [PubMed] [Google Scholar]
  19. Kellogg DL, Jr, Johnson JM, Kosiba WA. Selective abolition of adrenergic vasoconstrictor responses in skin by local iontophoresis of bretylium. Am J Physiol. 1989;257:H1599–H1606. doi: 10.1152/ajpheart.1989.257.5.H1599. [DOI] [PubMed] [Google Scholar]
  20. Kellogg DL, Jr, Liu Y, Kosiba IF, O'Donnell D. Role of nitric oxide in the vascular effects of local warming of the skin in humans. J Appl Physiol. 1999;86:1185–1190. doi: 10.1152/jappl.1999.86.4.1185. [DOI] [PubMed] [Google Scholar]
  21. Kolo LL, Westfall TC, Macarthur H. Nitric oxide decreases the biological activity of norepinephrine resulting in altered vascular tone in the rat mesenteric arterial bed. Am J Physiol Heart Circ Physiol. 2004;286:H296–H303. doi: 10.1152/ajpheart.00668.2003. [DOI] [PubMed] [Google Scholar]
  22. Lindblad LE, Ekenvall L, Klingstedt C. Neural regulation of vascular tone and cold induced vasoconstraction in human finger skin. J Auton Nerv Syst. 1990;30:169–173. doi: 10.1016/0165-1838(90)90141-5. [DOI] [PubMed] [Google Scholar]
  23. Minson CT, Berry LT, Joyner MJ. Nitric oxide and neurally mediated regulation of skin blood flow during local heating. J Appl Physiol. 2001;91:1619–1626. doi: 10.1152/jappl.2001.91.4.1619. [DOI] [PubMed] [Google Scholar]
  24. Öberg PA. Laser-Doppler flowmetry. Crit Rev Biomed Eng. 1990;18:125–163. [PubMed] [Google Scholar]
  25. Parati G, Casadei R, Groppelli A, Di Rienzo M, Mancia G. Comparison of finger and intra-arterial blood pressure monitoring at rest and during laboratory testing. Hypertension. 1989;13:647–655. doi: 10.1161/01.hyp.13.6.647. [DOI] [PubMed] [Google Scholar]
  26. Pérgola PE, Kellogg DL, Jr, Johnson JM, Kosiba WA, Solomon DE. Role of sympathetic nerves in the vascular effects of local temperature in human forearm skin. Am J Physiol. 1993;265:H785–H792. doi: 10.1152/ajpheart.1993.265.3.H785. [DOI] [PubMed] [Google Scholar]
  27. Saumet JL, Kellogg DL, Jr, Taylor WF, Johnson JM. Cutaneous laser-Doppler flowmetry: influence of underlying muscle blood flow. J Appl Physiol. 1988;65:478–481. doi: 10.1152/jappl.1988.65.1.478. [DOI] [PubMed] [Google Scholar]
  28. Stephens DP, Aoki K, Kosiba WA, Johnson JM. Nonnoradrenergic mechanism of reflex cutaneous vasoconstriction in men. Am J Physiol Heart Circ Physiol. 2001;280:H1496–H1504. doi: 10.1152/ajpheart.2001.280.4.H1496. [DOI] [PubMed] [Google Scholar]
  29. Taylor WF, Johnson JM, Kosiba WA, Kwan CM. Cutaneous vascular responses to isometric handgrip exercise. J Appl Physiol. 1989;66:1586–1592. doi: 10.1152/jappl.1989.66.4.1586. [DOI] [PubMed] [Google Scholar]
  30. Vanhoutte PM. Physical factors of regulation. In: Bohn DF, Somlyo AP, Sparks HV Jr, editors. Handbook of Physiology, section 2, The Cardiovascular System. vol. II. Bethesda MD: American Physiological Society; 1980. pp. 443–474. Vascular Smooth Muscle Chap. 16, pp. [Google Scholar]
  31. Vanhoutte PM, Cooke JP, Lindblad LE, Shepherd JT, Flavahan NA. Modulation of postjunctional α-adrenergic responsiveness by local changes in temperature. Clin Sci. 1985;68(Suppl. 10):121s–123s. doi: 10.1042/cs068s121. [DOI] [PubMed] [Google Scholar]
  32. Venturini G, Colasanti M, Fioravanti E, Bianchini A, Ascenzi P. Direct effect of temperature on the catalytic activity of nitric oxide synthases types I, II, and III. Nitric Oxide. 1999;3:375–382. doi: 10.1006/niox.1999.0250. [DOI] [PubMed] [Google Scholar]
  33. Yamazaki F, Sone R, Zhao K, Alvarez GE, Kosiba WA, Johnson JM. Rate dependency and role of nitric oxide in the vascular response to direct cooling in human skin. J Appl Physiol. 2006;100:42–50. doi: 10.1152/japplphysiol.00139.2005. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES

OSZAR »