Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2011 Aug 1;589(Pt 18):4555–4564. doi: 10.1113/jphysiol.2011.215020

Effects of cyclooxygenase inhibition on vascular responses evoked in fingers of men and women by iontophoresis of α1- and α2-adrenoceptor agonists

Amar Srinivasa 1, Janice M Marshall 1
PMCID: PMC3208224  PMID: 21807614

Non-technical summary

Sympathetic nerve fibres in the fingers release noradrenaline, which has the potential to act on α1- or α2-adrenoreceptors on the blood vessels. We used iontophoresis, which generates a tiny electrical charge, to push selective α1- or α2-receptor agonists through the skin and recorded changes in finger blood flow before and after aspirin, which inhibits cyclooxygenase (COX), an enzyme which synthesises vasodilator and vasoconstrictor prostaglandins. Our results yielded the novel findings that finger vasoconstriction produced by α1-adrenoceptors is blunted by locally synthesised vasodilator COX products in young men and in young women in the high but not the low oestrogen phase of the menstrual cycle. By contrast, finger vasoconstriction evoked by α2-adrenoceptors is largely attributable to vasoconstrictor COX products in young men, but overcome by vasodilator COX products in young women. This provides a foundation for testing whether COX products similarly modify vasoconstriction evoked by changes in sympathetic nerve activity.

Abstract

Abstract

In 10 men and nine women aged 20–23 years, we aimed to establish whether endogenous prostanoids synthesised by cyclooxygenase (COX) affect responses evoked in the finger by α1- or α2-adrenoceptor agonists. Cutaneous red cell flux (cRCF) was recorded in dorsal finger during iontophoresis of phenylephrine (PE) or clonidine (0.5 mm, seven 0.1 mA pulses followed by one 0.2 mA pulse: 30 s each at 60 s intervals) before and after the COX inhibitor aspirin (600 mg p.o.). In men, PE evoked a biphasic mean increase/decrease in cRCF before but a monophasic mean decrease in cRCF of 30–40% after aspirin (P < 0.05). In women in the low oestrogen (E2) phase of the menstrual cycle, PE evoked a decrease in cRCF (30–40%; P < 0.05) that was unchanged by aspirin, whereas in the high E2 phase, PE evoked no change before but a graded decrease in cRCF (30–40%; P < 0.05) after aspirin. Clonidine evoked a decrease in cRCF (∼30%; P < 0.05) in men before, but not after, aspirin. Clonidine evoked both increases and decreases in cRCF before and after aspirin in women in the low and high E2 phases (P > 0.05). We propose that finger vasoconstriction evoked by extraluminal α1-adrenoceptor stimulation is blunted by vasodilator COX products in young men and overcome by their action in women in the high, but not low E2, phase of the menstrual cycle. By contrast, α2-adrenoceptor stimulation evokes finger vasoconstriction that is mediated by vasoconstrictor COX products in young men, but evokes no consistent response in women in the low or high E2 phases of the menstrual cycle.

Introduction

There is evidence that the resting level of cutaneous blood flow in the hand is much lower in young women than young men. This has been attributed to greater tonic influence of the sympathetic noradrenergic fibres in young women. (Cooke et al. 1990). It has also been reported that the reflex decrease in finger blood flow evoked by application of a cold stimulus to the neck is of similar magnitude in young women and men (Freedman et al. 1987). However, primary Raynaud's disease, which is characterised by digital vasospasm evoked by cold or emotional stress, is far more common in young women than men and the severity of the disease decreases after the menopause (Cooke & Marshall, 2005). Thus, it seems that under some circumstances, ovarian hormones may promote vasoconstriction in the fingers, even though oestrogen (E2) is considered to play a protective role in vascular function.

Post-junctional adrenoceptors on vascular smooth muscle that respond to sympathetically released noradrenaline are predominantly of the α1-subtype, while α2-adrenoceptors are considered to be pre-junctional and to inhibit transmitter release. However, the preponderance of postsynaptic α2-adrenoceptors increases from the radial artery, to the digital arteries (Flavahan et al. 1987). Indeed, vasoconstriction was evoked in the finger both by α1- and by α2-adrenoceptor agonists when applied iontophoretically to the skin, or infused intra-arterially (Linblad & Ekenvall, 1986; Coffman & Cohen, 1988). Further, intra-arterial infusion of selective antagonists showed that vasoconstrictor tone in the finger of normal men and women is dependent on both α1-adrenoceptor, and α2-adrenoceptor stimulation (Cooke et al. 1997). Thus, an obvious question arises as to whether vasoconstriction evoked in the finger by α1- and/or α2-adrenoceptor stimulation is greater in women than men.

In previous studies, intra-arterial infusion of α1- or α2-agonists into the hand evoked considerably smaller vasoconstrictor responses in the finger of women than men (Freedman et al. 1987). These findings are difficult to interpret in relation to sympathetically mediated vasoconstriction because responses evoked by intraluminally administered α-agonists may be more readily modulated by endothelium-dependent vasodilator and vasoconstrictor substances than noradrenaline released from sympathetic nerves on the extra-luminal surface. Nevertheless, infusion of phenylephrine (PE), a selective α1-adrenoceptor agonist, into the hand of women, evoked greater finger vasoconstriction in the luteal phase of the menstrual cycle when oestrogen (E2) levels are relatively high, whereas the α2-adrenoceptor agonist clonidine evoked greater vasoconstriction in the follicular phase, when E2 levels are low. These results suggest that the influences of E2 may enhance α1- but limit α2-evoked finger vasoconstriction (Freedman & Girgis, 2000).

Correspondingly, contraction evoked by α1-adrenoceptor stimulation in aorta of female rats, was attenuated by ovariectomy, but restored by chronic administration of E2. Moreover, in aorta of intact female rats, contraction evoked by α1-adrenoceptor stimulation was attenuated by endothelium removal, cyclooxygenase (COX) inhibition, or antagonism of the Thromboxane receptor (TP) receptor that binds both PGH2 and thromboxane (TXA2), which is generated from PGH2 by TXA2 synthase. It was therefore proposed that in the presence of E2, stimulation of α1-adrenoceptor on the endothelium causes synthesis and release of PGH2/TXA2, which facilitates contraction evoked directly by α1-adrenoceptor stimulation (Fulton & Stallone, 2002).

There are also endothelial α2-adrenoceptors whose stimulation increases nitric oxide (NO) synthesis and limits vasoconstriction evoked by stimulation of α2-adrenoceptors on the vascular smooth muscle (Angus et al. 1986; Bockman et al. 1996). Given the endothelium generates vasodilator, as well as vasoconstrictor COX products and E2 increases the activity and expression of NO synthase and COX (Chambliss & Shaul, 2002), the cyclical variation in α2-adrenoceptor evoked finger vasoconstriction noted by Freedman & Girgis (2000) may reflect greater attenuation by NO and vasodilator COX products when E2 levels are high.

In view of these findings, the present study was performed to elucidate vascular responses evoked in the finger of young men and women by iontophoresis of selective α1- and α2-adrenoceptor agonists, such that they had access from the extraluminal surface like sympathetically released noradrenaline; we tested responses before and after the COX inhibitor aspirin. We hypothesised that (i) α1-mediated vasoconstriction is greater in women than men, particularly in the high E2 phase of the menstrual cycle, due to α1-evoked release of vasoconstrictor COX products and (ii) α2-mediated vasoconstrictor responses are smaller in women than men, particularly in the high E2 phase, reflecting action of vasodilator COX products. Some of these findings have been reported in brief (Srinivasa & Marshall, 2009).

Methods

We used a modification of the iontophoresis protocol by which we applied noradrenaline, the mixed α1- and α2-adrenoceptor agonist, to the finger of young men (Hendry & Marshall, 2004). This, in turn, was based on the protocol originally described for acetylcholine by Morris & Shore (1996) and used by us (Hendry & Marshall, 2004; Easter & Marshall, 2005), in which successive pulses of the agonists are applied so as to generate a cumulative vascular response that reaches a plateau. In preliminary studies on young men, not reported here, we established the concentration of the α1- and α2-adrenorecptor agonist, the pulse width and the interval between pulses that allowed a reproducible vascular response to develop such that the cRCF value achieved following the penultimate pulse appeared to have reached a maximum response to that agonist; the final pulse was delivered at twice the current with the intention of ensuring a maximum was reached (see Morris & Shore, 1996 and below). The same parameters were used for all studies on young men and women reported below.

The studies were carried out on 10 young men (age 21.1 ± 0.13 years; mean ± SEM) and nine young women (age 21.4 ± 0.38 years) all of whom were in good health and had no known cardiovascular or other disorder. A questionnaire was used in order to gain information on medical history along with height and weight, which showed that none of the subjects was obese or underweight (Table 1). None of the volunteers was a smoker, or taking vasoactive medication during the study or in the week preceding it. This includes all drugs that fall into the NSAIDs category, as well as substances such as vitamin C, which might influence oxidant status and thereby vascular responses. None of the women was taking an oral contraceptive pill or using any other medication known to alter levels of reproductive hormones, but they all had regular menstrual cycles, enabling them to predict when menstruation would begin with a confidence of 1–2 days. All subjects abstained from alcohol and strenuous exercise for at least 24 h before the study. They were also asked not to have a heavy meal or consume caffeine within 2 h of any experiment. All subjects gave informed consent. The study was approved by the Local Research Ethics Committee of the South Birmingham Health Authority.

Table 1.

Baseline values of mean ABP, pulse rate, height, weight, BMI and skin temperature on Day 1

ABP (mmHg) HR (beats min−1) Height (cm) Weight (kg) BMI Skin temp. (°C)
Males 86.7 ± 4.0 68.3 ± 2.7 1176.8 ± 1.8 73.6 ± 3.0 23.4 ± 0.6 35.2 ± 0.4
Females (high E2) 81.5 ± 3.3 68.6 ± 3.1 1162.5 ± 1.4* 59.8 ± 1.8* 22.9 ± 0.4 33.8 ± 0.4*
Females (low E2) 82.4 ± 2.9 69.5 ± 3.2 34.1 ± 0.3
Open in a new tab

Values are means ± SEM.

*

P < 0.05, male vs. female subjects by Student's t test.

Comparisons were also made in female subjects between the high and low E2 phases, but they did not reach statistical significance.

All experiments were carried out in a windowless, temperature-controlled room kept at 23–25°C. Extraneous sound and all distraction within the room was minimised so as to limit the novel or noxious stimuli to which the subjects were exposed. This helped ensure that steady, basal levels of cardiovascular variables were obtained. On arrival, subjects washed their hands in warm water so as to ensure good resting perfusion to the fingers and to facilitate the expected agonist-induced constrictor responses. Each subject was seated throughout the protocol, with both arms and hands supported at heart level by cushions to maximise comfort. Arterial blood pressure (ABP) and heart rate (HR) were recorded from the right hand, while iontophoresis was applied to the left hand (see below), which was placed on an electric heating blanket so as to maintain the hand skin temperature at 33–35°C. Skin temperature was recorded continuously via a thermistor that was attached to the dorsal surface of the left hand with surgical tape. In our previous study, which was carried out under exactly the same conditions except that we did not use a heating blanket, control cRCF was ∼30 perfusion units (PU) in young male subjects (Hendry & Marshall, 2004) compared with ∼50–60 PU in the present study (Table 2). Thus, it is reasonable to assume that we induced tonic vasodilatation.

Table 2.

Baseline values of mean RCF recorded in male and female subjects at start of each iontophoresis protocol

Control (PU) 30 min post-aspirin (PU) 180 min post-aspirin (PU) 24 h post-aspirin (PU)
Males
 Phenylephrine 66.9 ± 12.4 62.8 ± 11.6 56.8 ± 7.4 55.0 ± 9.1
 Clonidine 54.4 ± 10.1 61.8 ± 12.2 58.9 ± 7.1 57.0 ± 7.8
Females
 Low E2
 Phenylephrine 73.8 ± 16.9 44.3 ± 5.7 64.6 ± 14.7 59.6 ± 15.1
 Clonidine 47.1 ± 9.5 62.0 ± 10.5 46.7 ± 8.4 41.5 ± 6.9
 High E2
 Phenylephrine 41.8 ± 10.2 56.3 ± 8.9 64.4 ± 12.4 58.5 ± 12.4
 Clonidine 43.2 ± 6.5 58.5 ± 15.8 55.2 ± 8.1 67.5 ± 11.9
Open in a new tab

Values are means ± SEM. PU: perfusion units. Comparisons were also made between females during the high and low E2 phases: they did not reach statistical significance.

After a 30 min acclimatisation period, when the subjects had adjusted to the temperature of the room and the electric blanket, the cuff of a Finapres device (2300; Ohmeda, Stirling, UK) was placed on the middle finger of the subject's right hand. This allowed continuous recordings of ABP and HR throughout the protocol. The dorsal surface of the subject's left hand was then cleaned with the aid of a sterile alcohol wipe in preparation for iontophoresis to be carried out on the proximal phalynx of the index and middle fingers for phenylephrine (PE) and clonidine, respectively. Iontophoresis was performed and cutaneous red cell flux (cRCF) was recorded as described before (Hendry & Marshall, 2004). Briefly, an iontophoresis chamber comprising a Perspex ring containing the anode was attached to the dorsal surface of the finger, while the cathode was fastened onto the dorsal surface of the hand. Both electrodes were connected to a battery powered iontophoresis controller (MIC 1; Moor Instruments, Axminster, UK). The iontophoresis chamber (8 mm inner diameter) was filled with 0.5 ml of the relevant drug. A laser Doppler probe (DPIT-V2; Moor Instruments) was then placed into the well and connected to a laser Doppler perfusion (DRT 4; Moor Instruments), so as to provide a continuous recording of cRCF. The outputs of the iontophoresis controller, cRCF, ABP and HR were recorded on a PC (Gateway GP6-400) via MacLab hardware (MacLab/400; ADInstruments, Hastings, UK).

Protocols

Day 1

When a steady baseline cRCF had been maintained for at least 5 min, the iontophoresis protocol for either 0.5 mm PE or 0.5 mm clonidine was begun: seven pulses of 0.1 mA for 30 s, followed by a final pulse of 0.2 mA for 30 s with 60 s intervals between pulses. After an interval of 5 min, this protocol was repeated with the other drug. The order of applying PE and clonidine was randomised.

Day 2

On the following day the subject was given 600 mg of soluble aspirin dissolved in water to take orally. The equipment was connected as described above and the iontophoresis protocol just described was repeated at 30 min post-aspirin and at 3–4 h post-aspirin for PE and for clonidine, each being applied at the same site as on Day 1. This dose given orally can be expected to provide 86% inhibition of prostacyclin (PGI2) synthesis attributable to endothelial COX at 30 min, 70% inhibition at 90 min and full recovery at 6 h, whereas 99% inhibition of TXA2 synthesis by platelets is achieved from 30 min to 6 h due to their inability to synthesise new COX (Heavey et al. 1985).

Day 3

Approximately 24 h post-aspirin, a final repeat of the iontophoresis protocol was performed for both PE and clonidine each at the same site as on the previous days.

Exactly the same protocol was used on the male and female subjects. However, female subjects attended for two complete runs of the protocol, the first in the low E2 phase of the menstrual cycle (days 0–5) and the second in the high E2 phase (days 9–12).

Water control

The iontophoresis protocol described above was performed on five male subjects, with sterile water in the chamber.

Drugs

Phenylephrine hydrochloride and clonidine hydrochloride (Sigma, Poole, Dorset) were dissolved in sterile water for injection (Steri-Amp Water for injection BP; Norton, Runcorn, Cheshire, UK) to create stock solutions of 1 mm of PE and clonidine which were kept refrigerated at –3°C. These stock solutions were used to create the 0.5 mm solutions for iontophoresis on the day of the experiment by dilution with sterile water. New stock solutions were made up weekly. All solutions of PE were wrapped in aluminium foil due its photosensitivity and both test solutions were kept on ice until use during the experiment. Introduction of the ice-cold solution of PE or clonidine into the iontopheresis chamber did not result in any significant change in cRCF until the iontophoresis currents were applied.

Analysis of results

All results are expressed as means ± SEM. In each iontophoresis protocol, the responses evoked by successive pulses were calculated as the mean cRCF over the last 20 s of the interval between successive current pulses, so that the change in RCF in response to each pulse could be calculated (Hendry & Marshall, 2004). As can be seen from Table 2, there was variation between subjects within each group in the baseline level of cRCF. Therefore, in order to normalise the cRCF data, the values recorded during iontophoresis were expressed as the percentage change from the baseline recorded before iontophoresis. Changes in cRCF within each iontophoresis protocol for PE and for clonidine in the absence of and at different time intervals after aspirin were analysed by using ANOVA for repeated measures with Scheffé's post hoc test to detect the time within the protocol at which a change reached significance. Student's paired and unpaired t test was used to compare baseline values without and after aspirin within and between groups. Statistical significance was taken as P < 0.05.

Results

The height and weight of the subjects and the mean values of ABP and HR in males and in females in the low and high E2 phases of the menstrual cycle are shown in Table 1. The male subjects weighed more than the females and had higher skin temperature than the females in the high E2 phase. There were no significant differences between groups or between experimental days within groups (data not shown). Within each group, ABP did not change significantly during the iontophoresis protocol under any condition. Therefore the changes in cRCF that occurred during iontophoresis can be taken as an index of the vascular response in the cutaneous circulation of the finger.

Vascular reponses evoked in male subjects

Baseline RCF values in the absence of and at different times after aspirin were not significantly different for the PE or clonidine protocol (see Table 2).

Phenylephrine

In males, the control response to successive pulses of PE comprised a biphasic change in mean cRCF: cRCF increased in 5 of the 10 subjects in response to pulses 1–4 and decreased thereafter, whereas in the remaining subjects, cRCF decreased throughout (see Fig. 1A). The response over the entire iontophoresis protocol was statistically significant by ANOVA, but changes evoked by individual pulses did not reach significance by post hoc test. However, at 30 and at 180 min post-aspirin on Day 2, PE evoked constriction in all subjects (Fig. 1B and C), the changes from baseline reaching significance by post hoc test at the sixth to eighth pulses at 180 min after aspirin, (Fig. 1C). At 24 h post-aspirin on Day 3, the mean response evoked by PE was again biphasic and similar to the control response on Day 1 with 5/10 subjects showing dilatation to pulses 1–4 (Fig. 1D),

Figure 1. Cutaneous vascular responses evoked in finger of young men by iontophoresis of phenylephrine before and at different times after aspirin.

Figure 1

Open in a new tab

In each graph, RCF is shown as percentage change from baseline (mean ± SEM) following each iontophoretic current pulse, as shown on the abscissa. A, responses evoked by PE before aspirin (control response). B–D, responses evoked 30 min, 180 min, 24 h post-aspirin, respectively. n = 10. †, ††, †††: P < 0.05, 0.01, 0.001 respectively by ANOVA. *P < 0.05 vs. baseline by post hoc Scheffé's test.

Clonidine

The control response to successive pulses of clonidine was a decrease in cRCF (see Fig. 2A). By contrast, at 30 and 180 min post-aspirin, clonidine evoked no significant change in cRCF (Fig. 2B and C). However, at 24 h post-aspirin the response evoked by clonidine was again vasoconstriction comparable to the control response (Fig. 2D).

Figure 2. Cutaneous vascular responses evoked in finger of young men by iontophoresis of clonidine before and at different times after aspirin.

Figure 2

Open in a new tab

In each graph, RCF is shown as percentage change from baseline (mean ± SEM) following each iontophoretic current pulse, as shown on the abscissa: A, responses evoked by clonidine before aspirin (control response). B–D, responses evoked 30 min, 180 min, 24 h post-aspirin respectively. n = 10. †, ††: P < 0.05, 0.01 respectively by ANOVA.

Vascular reponses evoked in female subjects

Baseline cRCF values were not significantly different in the absence of or at different times after aspirin for the PE or clonidine protocols in the low or high E2 phases of the menstrual cycle (see Table 2).

Phenylephrine

In the low E2 phase of the menstrual cycle, the control response to PE was a decrease in cRCF (see Fig. 3A, upper row). This response was not obviously altered by aspirin (Fig. 3B and C upper row), although the response evoked at 24 h post-aspirin did not reach statistical significance (P = 0.07), reflecting dilator responses to the second to fifth pulses of PE in three subjects.

Figure 3. Cutaneous vascular responses evoked in finger of young women by iontophoresis of phenylephrine in the low (above) and high (below) E2 phases of the menstrual cycle, before and at different times after aspirin.

Figure 3

Open in a new tab

In each graph, RCF is shown as percentage change from baseline (mean ± SEM) following each iontophoretic current pulse, as shown on the abscissa: A, responses evoked by PE before aspirin (control response). B–D, responses evoked 30 min, 180 min, 24 h post-aspirin respectively. n = 9. †P < 0.05, ††P < 0.01,†††P < 0.001 by ANOVA. *P < 0.05 vs. baseline by post hoc Scheffé's test.

By contrast, in the high E2 phase of the cycle, the control response to PE was very variable between individuals in the direction and magnitude of the changes in cRCF (Fig. 3A, lower row). However, at 30 and 180 min post-aspirin, PE evoked significant changes in cRCF, vasoconstriction occurring by the sixth to eighth pulses (Fig. 3B and C lower row). At 24 h post-aspirin, PE again evoked no significant change in cRCF (Fig. 3D).

Clonidine

In the low and high E2 phases of the cycle, responses evoked by clonidine were very variable both with and between individuals such that there were no significant changes in cRCF either before or at different times post-aspirin (Fig. 4). Visual inspection of the site of iontophoresis when the chamber had been removed generally showed a circular halo of pale skin around patchy flushing which seemed to be located around hair follicles. Thus, it appeared that clonidine evoked competing vasoconstrictor and dilator responses. This contrasted with the general situation in men and women in that the skin usually appeared uniformly pale when PE or clonidine evoked a net decrease in cRCF and flushed when there was a net increase in cRCF.

Figure 4. Cutaneous vascular responses evoked in finger of young women by iontophoresis of clonidine in the low (above) and high (below) E2 phases of the menstrual cycle, before and at different times after aspirin.

Figure 4

Open in a new tab

In each graph, RCF is shown as percentage change from baseline (mean ± SEM) following each iontophoretic current pulse, as shown on the abscissa. A, responses evoked by clonidine before aspirin (control response). B–D, responses evoked 30 min, 180 min, 24 h post-aspirin respectively. n = 9. None of these responses was statistically significant at P < 0.05.

Water

Iontophoresis of water with the same protocol as used for PE and clonidine evoked no significant changes in cRCF (see Fig. 5).

Figure 5. Cutaneous vascular responses evoked in finger of young men by iontophoresis of water.

Figure 5

Open in a new tab

RCF is shown as percentage change from baseline (mean ± SEM) following each iontophoretic current pulse, as shown on the abscissa.

Discussion

In the present study, iontophoresis of successive pulses of the α1-adrenoceptor agonist PE evoked biphasic dilatation/constriction in the finger of young men, but vasoconstriction after COX inhibition with aspirin. By contrast, in women in the low E2 phase of the menstrual cycle, PE evoked finger vasoconstriction that was not altered by aspirin, whereas in the high E2 phase, PE evoked no response until aspirin was given; then it evoked finger vasoconstriction. Further, the α2-adrenoceptor agonist clonidine evoked vasoconstriction in the finger of men that was unchanged by aspirin, but no net response in the finger of women in the low or high E2 phases, whether or not aspirin had been taken. These findings contrast in several respects with our working hypotheses (see Introduction), but before interpreting them we need to consider the conditions under which our study was done and its limitations.

In previous studies, resting blood flow in the hand was lower in young men than women (Cooke et al. 1990). By contrast, in the present study, baseline levels of cRCF in the finger were similar in men and women, presumably because we warmed the hand to facilitate vasoconstrictor responses to α-agonists. This would have achieved a relatively low level of sympathetic vasoconstrictor tone and possibly active vasodilatation mediated by dilator nerve fibres and local heating, both of which involve NO (Kellogg, 2006). Our results must therefore be considered in this context (see below). In order to compare responses evoked in the low and high E2 phases of the menstrual cycle, we performed experiments on days 0–5 and on days 9–12, respectively. The women who volunteered had regular cycles, but we acknowledge uncertainty, particularly in estimating the high E2 phase. This may explain some of the variability in the evoked changes in cRCF in the high E2 phase (see below).

We used aspirin at 600 mg p.o. because this dose achieves a maximum inhibitory effect on COX activity in nucleated cells at 30 min, with recovery by 6 h (Heavey et al. 1985). In men and in women in the low and high E2 phase of the menstrual cycle, aspirin had no effect on baseline levels of cRCF suggesting that tonically synthesised vasoconstrictor or dilator COX products do not contribute to finger vascular tone. This accords with previous findings when COX inhibitors were given systemically (e.g. Morris & Shore, 1996; Khan et al. 1997; Hendry & Marshall, 2004; Easter & Marshall, 2005). Further, when the control response evoked by α1- or α2-adrenoceptor stimulation was changed at 30 and 180 min post-aspirin, the response evoked at 24 h was generally comparable to the control response. Thus, we can reasonably attribute such changes to inhibition of COX, rather than to time, or inherent variability within or between subjects. Given iontophoresis of water evoked no change in cRCF, we can attribute the substantial changes in cRCF evoked by PE and clonidine to α1- or α2-adrenoceptor stimulation rather than to the iontophoresis current per se.

Responses evoked by α1-adrenoceptor stimulation

On the basis of previous studies, we hypothesised that iontophoresis of α1-agonist would evoke greater finger vasoconstriction in women than men, particularly in the high E2 phase of the menstrual cycle due to α1-evoked release of vasoconstrictor COX products. In fact, iontophoresis of PE evoked pure finger vasoconstriction only in women in the low E2 phase.

The increases in cRCF evoked by the early pulses of PE in young men were converted to decreases after aspirin such that pure vasoconstriction occurred, cRCF decreasing by 30–40%. Assuming the concentration of PE in superficial skin gradually increases during successive pulses and that PE gradually diffuses more deeply, it is likely the initial dilator responses reflected α1-stimulated release of a dilator COX product from vascular smooth muscle or other cells close to the extraluminal surface of blood vessels, although we cannot exclude the possibility of diffusion to endothelium. Vascular smooth muscle cells and endothelial cells express COX-1 and COX-2 in aortae of male and female rats and both cell types synthesise dilator and constrictor COX products (Shepherd & Katusic, 1991; Li et al. 2008). Thus, irrespective of the cells involved, we can propose that in young men, COX-dependent vasodilator products may blunt finger vasoconstriction evoked by α1-adrenoceptors.

Finger vasoconstriction evoked by PE in women in the low E2 phase was not affected by aspirin. Thus, there is no reason to propose it was partly mediated by vasoconstrictor COX products as we hypothesised from in vitro studies (Fulton & Stallone, 2002). That the evoked change in cRCF did not reach statistical significance at 24 h post-aspirin in the low E2 phase may be explained if E2 level was rising in some women by this time (see above). Indeed, our finding that in the high E2 phase, PE had no significant effect on cRCF in the absence of aspirin, but evoked finger vasoconstriction at 30 and 180 min post-aspirin suggests that α1-stimulation released vasodilator COX products that overwhelmed the α1-evoked constriction. This proposal accords with evidence that E2 increases the activity and expression of COX (Orshal & Khalil, 2003). The finger vasoconstriction evoked by PE when COX was blocked, was comparable in the high E2 and low E2 phases, cRCF decreasing by 30–40%. Thus, even though E2 increases endothelial NO synthase (eNOS) activity and expression (Chambliss & Shaul, 2002) and NO blunts α1-evoked vasoconstriction (Zhang & Davidge, 1999; Tuttle & Fallone, 2001), there seems no reason to suppose that NO has a menstrual phase-related blunting effect on finger vasoconstrictor responses to α1-stimulation.

Responses evoked by α2-adrenoceptor stimulation

There are three subtypes of α2-adrenoceptor: α2A-, α2B- and α2C-receptors (Bylund et al. 1994). The α2C-adrenoceptor has been specifically implicated in the vasoconstrictor response to cold in extremities such as fingers and mouse-tail. In contrast to vasoconstriction evoked by α1-, and other subtypes of α2-receptors, that evoked by α2C-receptors is augmented by cooling because these receptors are translocated to the plasma membrane by cooling (Chotani et al. 2000). Also, E2 increases the expression of α2C-receptors in vascular smooth muscle (Eid et al. 2007). However, we maintained the hand at 35–37°C when α2C-receptors are silent (Chotani et al. 2000). Thus, iontophoretically applied clonidine probably had access only to α2A- and α2B-receptors.

We hypothesised that α2-mediated finger vasoconstriction evoked by iontophoresis of clonidine would be smaller in women than men, particularly in the high E2 phase, reflecting the action of vasodilator COX products. However, our finding that iontophoresis of α2-agonist failed to evoke an overriding decrease in cRCF in women in the low or high E2 phases of the menstrual cycle irrespective of whether COX was inhibited is consistent with an overriding tonic or α2-stimulated influence of NO and does not implicate COX products at all. That α2-stimulation did evoke an underlying vasoconstriction is supported by the visual observation of skin blanching (see Results). It may be that the patchy flushing around hair follicles reflects their ability to give easier access of the agonist to endothelial α2-adrenoceptors.

By contrast, in men, α2-stimulation evoked graded vasoconstriction that was attenuated by aspirin indicating it was largely attributable to vasoconstrictor COX products. To our knowledge, this is a novel finding. However, α2-stimulation evoked contraction in endothelium-intact rat mesenteric arteries that was partly attributable to a COX-dependent vasoconstrictor product and attenuated by castration (Tejera et al. 1999). Further, in pig coronary artery, testosterone facilitated release of an endothelium-derived vasoconstrictor that was probably a COX product (Farhat et al. 1995), while testosterone increased TP receptor expression in vascular smooth muscle (Masuda et al. 1991). As indicated above COX-1 and COX-2 are expressed in vascular smooth muscle and endothelium (Li et al. 2008). Thus, it may be that diffusion of α2-agonist from the extraluminal surface evokes finger vasoconstriction in young men mediated by COX products that are released from vascular smooth muscle or endothelium, whose action on TP receptors is facilitated by testosterone.

This proposal is consistent with our finding that vasoconstrictor COX products limit vasodilatation evoked by iontophoresis of ACh in the finger of young men, but not young women (Hendry & Marshall, 2004; Easter & Marshall, 2005). Indeed, our findings directly contrast with the proposal that E2 potentiates vasoconstriction evoked by a range of agonists even in normal females, by releasing vasoconstrictor COX products that act on TP receptors (see Sellars & Stallone, 2007). Rather, our results indicate that in the finger, release of such a substance is a characteristic of healthy young men, but not healthy young women, even in the high E2 phase of the menstrual cycle. That a vasoconstrictor COX product limits ACh-evoked finger vasodilatation in healthy, post-menopausal women and in pre- and post-menopausal women with primary Raynaud's disease (Easter & Marshall, 2005), suggests that such substances become important in women in the absence of E2, or pathologically.

In summary, our results allow the novel proposal that under conditions of low sympathetic vasoconstrictor tone, iontophoresis of α1-adrenoceptor agonist into the finger, giving access from the extraluminal surface like sympathetically released noradrenaline, evokes finger vasoconstriction that is of similar magnitude in young men and women in the low and high E2 phases of the menstrual cycle, decreasing cRCF by 30–40%. However, vasodilator COX products overcome the vasoconstriction evoked by low concentrations of α1 agonist in men, and completely attenuate finger vasoconstriction in women in the high E2 phase of the menstrual cycle, whilst having no modulatory effect in the low E2 phase. They also allow the novel proposal that iontophoresis of α2 agonist evokes finger vasoconstriction in young men that is largely mediated by vasoconstrictor COX products. By contrast, in young women, in both the low and high E2 phases of the menstrual cycle, any finger vasoconstrictor response to α2-stimulation is overcome by vasodilator influences that probably include NO, but not COX products. It remains to be determined whether vasodilator and vasoconstrictor COX products similarly modulate α1 and α2 components of vasoconstriction evoked in fingers by sympathetic nerve activity.

Acknowledgments

This study was generously supported by the Raynaud's and Scleroderma Association.

Glossary

Abbreviations

COX

cyclooxygenase

cRCF

cutaneous red cell flux

PE

phenylephrine

Author contributions

J.M.M. conceived the study and planned it. A.S. performed the experiments and analysed the data under the guidance of J.M.M. A.S. produced a first draft of the paper. J.M.M. set the study in context and played the major role in completing the paper. Both authors have approved the final version.

References

  1. Angus JA, Cocks TM, Satoh K. The α adrenoceptors on endolethelial cells. Fed Proc. 1986;45:2355–2359. [PubMed] [Google Scholar]
  2. Bockman CS, Gonzalez-Cabrera I, Abel PW. Alpha-2 adrenoceptor subtype causing nitric oxide-mediated vascular relaxation in rats. J Pharmacol Exp Ther. 1996;278:1235–1243. [PubMed] [Google Scholar]
  3. Bylund DB, Eikenberg DC, Hieble JP, Langer SZ, Lefkowitz RJ, Minneman KP, Molinoff PB, Ruffulo RR, Trendelenburg U. International Union of Pharmacology nomenclature of adrenoceptors. Pharmacol Rev. 1994;46:121–136. [PubMed] [Google Scholar]
  4. Chambliss KL, Shaul PW. Estrogen modulation of endothelial nitric oxide synthase. Endocr Rev. 2002;23:665–686. doi: 10.1210/er.2001-0045. [DOI] [PubMed] [Google Scholar]
  5. 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]
  6. Coffman JD, Cohen RA. Role of alpha-adrenoceptor subtypes mediating sympathetic vasoconstriction in human digits. Eur J Clin Invest. 1988;18:309–313. doi: 10.1111/j.1365-2362.1988.tb01264.x. [DOI] [PubMed] [Google Scholar]
  7. Cooke JP, Creager MA, Scales KM, Ren KM, Tsapaisaris NP, Beetham WP, Creager MA. Role of digital artery adrenoceptors in Raynaud's disease. Vasc Med. 1997;2:1–7. doi: 10.1177/1358863X9700200101. [DOI] [PubMed] [Google Scholar]
  8. Cooke JP, Creager MA, Osmundson PJ, Shepherd JT. Sex differences in control of cutaneous blood flow. Circulation. 1990;82:1607–1615. doi: 10.1161/01.cir.82.5.1607. [DOI] [PubMed] [Google Scholar]
  9. Cooke JP, Marshall JM. Mechanmisms of Raynaud's Disease. Vasc Med. 2005;10:293–307. doi: 10.1191/1358863x05vm639ra. [DOI] [PubMed] [Google Scholar]
  10. Easter MJ, Marshall JM. Contribution of COX products to endothelium-dependent vasodilatation in the digital circulation of women with primary Raynaud's disease. Clin Sci. 2005;109:45–54. doi: 10.1042/CS20040262. [DOI] [PubMed] [Google Scholar]
  11. Eid AH, Maiti K, Mitra S, Chotani MA, Flavahan S, Bailey SR, Thompson-Torgerson CS, Flavahan NA. Estrogen increases smooth muscle expression of α2C-adrenoceptors and cold-induced constriction of cutaneous arteries. Am J Physiol Heart Circ Physiol. 2007;293:H1955–H1961. doi: 10.1152/ajpheart.00306.2007. [DOI] [PubMed] [Google Scholar]
  12. Ekenvall L, Lindblad LE, Verbeuren TJ, Norbeck O, Etzell BM. α-Adrenoceptors and cold induced vasoconstriction in human finger skin. Am J Physiol Heart Circ Physiol. 1988;255:H1000–H1003. doi: 10.1152/ajpheart.1988.255.5.H1000. [DOI] [PubMed] [Google Scholar]
  13. Farhat MY, Wolfe R, Vargas R, Foegh ML, Ramwell PW. Effect of testosterone treatment on vasoconstrictor response of left anterior descending coronary artery in male and female pigs. J Cardiovasc Pharmacol. 1995;25:495–500. doi: 10.1097/00005344-199503000-00023. [DOI] [PubMed] [Google Scholar]
  14. Flavahan NA, Cooke JP, Shepherd JT, Vanhoutte PM. Human post-junctional alpha-1 and alpha-2 adrenoceptors: differential distribution in arteries of the limbs. J Pharmacol Exp Ther. 1987;241:361–365. [PubMed] [Google Scholar]
  15. Freedman RR, Girgis R. Effects of menstrual cycle and race on peripheral vascular α-adrenergic responsiveness. Hypertension. 2000;35:795–799. doi: 10.1161/01.hyp.35.3.795. [DOI] [PubMed] [Google Scholar]
  16. Freedman RR, Sabbarwal SC, Desia N. Sex differences in peripheral vascular adrenergic receptors. Circ Res. 1987;61:581–585. doi: 10.1161/01.res.61.4.581. [DOI] [PubMed] [Google Scholar]
  17. Fulton CT, Stallone JN. Sexual dimorphism in prostanoid-potentiated vascular contraction: roles of endothelium and ovarian steroids. Am J Physiol Heart Circ Physiol. 2002;283:H2062–H2073. doi: 10.1152/ajpheart.00099.2002. [DOI] [PubMed] [Google Scholar]
  18. Heavey DJ, Barrow SE, Hickling N, Ritter JM. Aspirin causes short-lived inhibition of bradykinin-stimulated prostacyclin production in man. Nature. 1985;318:186–188. doi: 10.1038/318186a0. [DOI] [PubMed] [Google Scholar]
  19. Hendry RG, Marshall JM. Vasoconstrictor products of cyclooxygenase activity limit acetylcholine-induced cutaneous vasodilatation in young men. Clin Sci. 2004;107:323–330. doi: 10.1042/CS20040077. [DOI] [PubMed] [Google Scholar]
  20. Kellogg DL. In vivo mechanisms of cutaneous vasodilation and vasoconstriction in humans during thermoregulatory challenges. J Appl Physiol. 2006;100:1709–1718. doi: 10.1152/japplphysiol.01071.2005. [DOI] [PubMed] [Google Scholar]
  21. Khan F, Davidson NC, Littleford RC, Litchfield SJ, Struthers AD, Belch JJF. Cutaneous vascular responses to acetylcholine are mediated by a prostanoid-dependent mechanism in man. Vasc Med. 1997;2:82–86. doi: 10.1177/1358863X9700200202. [DOI] [PubMed] [Google Scholar]
  22. Li M, Kuo L, Stallone JN. Estrogen potentiates constrictor prostanoid function in female rat aorta by upregulation of cyclooxygenase-2 and thromboxanes pathway expression. Am J Physiol Heart Circ Physiol. 2008;294:H2444–H2455. doi: 10.1152/ajpheart.01121.2007. [DOI] [PubMed] [Google Scholar]
  23. Linblad LE, Eckenvall L. Alpha-adrenoceptors in the vessels of human finger skin. Acta Physiol Scand. 1986;128:219–222. doi: 10.1111/j.1748-1716.1986.tb07969.x. [DOI] [PubMed] [Google Scholar]
  24. Masuda A, Mathur R, Halushka PV. Testosterone increases thromboxane A2 receptors in cultured rat aortic smooth muscle cells. Circ Res. 1991;69:638–643. doi: 10.1161/01.res.69.3.638. [DOI] [PubMed] [Google Scholar]
  25. Morris SJ, Shore AC. Skin blood flow responses to the iontophoresis of acetylcholine and sodium nitroprusside in man: possible mechanisms. J Physiol. 1996;496:531–542. doi: 10.1113/jphysiol.1996.sp021704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Orshal JM, Khalil RA. Gender, sex hormones and vascular tone. Am J Physiol Regul Integr Comp Physiol. 2003;286:R233–R249. doi: 10.1152/ajpregu.00338.2003. [DOI] [PubMed] [Google Scholar]
  27. Sellars MM, Stallone JN. Sympathy for the role of the devil: the role of thromboxanes in the regulation of vascular tone and blood pressure. Am J Physiol Heart Circ Physiol. 2007;294:H1978–H1986. doi: 10.1152/ajpheart.01318.2007. [DOI] [PubMed] [Google Scholar]
  28. Shepherd JT, Katusic ZS. Endothelium-derived vasoactive factors. I. Endothelium-dependent relaxation. Hypertension. 1991;18(Suppl):III76–III85. doi: 10.1161/01.hyp.18.5_suppl.iii76. [DOI] [PubMed] [Google Scholar]
  29. Srinivasa A, Marshall JM. The role of cyclooxygenase (COX) products in cutaneous vascular responses evoked by α1- and α2- adrenoceptor stimulation in male and female subjects. Microcirculation. 2009;16:444–486. [Google Scholar]
  30. Tejera N, Balfagon G, Marin J, Ferrer M. Gender differences in the endothelial regulation of α2-adrenoceptor-mediated contraction in the rat aorta. Clin Sci. 1999;97:19–25. [PubMed] [Google Scholar]
  31. Tuttle JL, Fallone JC. Nitric oxide release during α1-adrenoceptor-mediated contraction of arterioles. Am J Physiol Heart Circ Physiol. 2001;281:H873–H881. doi: 10.1152/ajpheart.2001.281.2.H873. [DOI] [PubMed] [Google Scholar]
  32. Zhang Y, Davidge ST. Effect of estrogen replacement on vasoconstrictor responses in rat mesenteric arteries. Hypertension. 1999;34:1117–1122. doi: 10.1161/01.hyp.34.5.1117. [DOI] [PubMed] [Google Scholar]

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

RESOURCES