Sensory and sympathetic nerve contributions to the cutaneous vasodilator response from a noxious heat stimulus

Exp

erim

enta

lPhy

siol

ogy

1208 Exp Physiol 96.11 pp 1208–1217

Research PaperResearch Paper

Sensory and sympathetic nerve contributionsto the cutaneous vasodilator response from a noxiousheat stimulus

Stephen J. Carter and Gary J. Hodges

Department of Kinesiology, The University of Alabama, Tuscaloosa, AL 35401, USA

We investigated the roles of sensory and noradrenergic sympathetic nerves on the cutaneousvasodilator response to a localized noxious heating stimulus. In two separate studies, fourforearm skin sites were instrumented with microdialysis fibres, local heaters and laser-Dopplerprobes. Skin sites were locally heated from 33 to 42◦C or rapidly to 44◦C (noxious). In the firststudy, we tested sensory nerve involvement using EMLA cream. Treatments were as follows:(1) control 42◦C; (2) EMLA 42◦C; (3) control 44◦C; and (4) EMLA 44◦C. At the EMLA-treatedsites, the axon reflex was reduced compared with the control sites during heating to 42◦C(P < 0.05). There were no differences during the plateau phase (P > 0.05). At both the sitesheated to 44◦C, the initial peak and nadir became indistinguishable, and the EMLA-treated siteswere lower compared with the control sites during the plateau phase (P < 0.05). In the secondstudy, we tested the involvement of noradrenergic sympathetic nerves in response to the noxiousheating using bretylium tosylate (BT). Treatments were as follows: (1) control 42◦C; (2) BT 42◦C;(3) control 44◦C; and (4) BT 44◦C. Treatment with BT at the 42◦C sites resulted in a markedreduction in both the axon reflex and the secondary plateau (P < 0.05). At the 44◦C sites, therewas no apparent initial peak or nadir, but the plateau phase was reduced at the BT-treated sites(P < 0.05). These data suggest that both sympathetic nerves and sensory nerves are involvedduring the vasodilator response to a noxious heat stimulus.

(Received 9 June 2011; accepted after revision 19 August 2011; first published online 2 September 2011)Corresponding author G. J. Hodges: Department of Kinesiology, The University of Alabama, 133 Russell Hall,Tuscaloosa, AL 35401, USA. Email: [email protected]

In humans, both neurogenic reflexes and local factorscontribute to the highly responsive nature of the cutaneouscirculation (Taylor et al. 1984; Pergola et al. 1993;Hodges et al. 2008). During periods of heat stress,sensory nerves and two separate sympathetic nervouspathways, a noradrenergic vasoconstrictor system anda cholinergic active vasodilator system, are responsiblefor regulating the cutaneous vasculature (Kellogg et al.1995; Wong & Minson, 2006). Additionally, local tissuetemperature is a potent factor that contributes tothe extent of vasodilatation achieved. Johnson et al.(1986) demonstrated that skin warming localized to thesite of measurement produces a graded vasodilatationin accordance with local tissue temperature, reachingmaximal levels in skin heated to 42◦C for 20–40 min.Collectively, these mechanisms offer a tremendous range

of blood flow patterns, making the cutaneous vasculature ahighly adaptable thermoregulatory system (Rowell, 1974;Kellogg, 2006).

In non-glabrous (hairy) skin, local heating produces acharacteristic biphasic hyperaemia (Kellogg et al. 1999;Minson et al. 2001). There is a transient initial peakfollowed by a nadir, which is succeeded by a prolongedplateau phase. Current data suggest that the initial peakand nadir of the vasodilator response are mediated byan axon reflex (Minson et al. 2001; Tew et al. 2011a).Additionally, the initial peak has been shown to have nitricoxide and noradrenergic sympathetic nerve components(Minson et al. 2001; Houghton et al. 2006; Hodges et al.2008, 2009b; Tew et al. 2011b). Kellogg et al. (1999)determined that the plateau phase of the cutaneousvasodilator response is ∼70% dependent on NO and

DOI: 10.1113/expphysiol.2011.059907 C© 2011 The Authors. Journal compilation C© 2011 The Physiological Society

Exp Physiol 96.11 pp 1208–1217 Sensory and sympathetic contributions to noxious heat stimulus 1209

have recently shown that the vasodilatation in response tolocal skin heating is produced by endothelial NO synthase(eNOS; Kellogg et al. 2008, 2009).

Interestingly, NOS inhibition does not alter thesecondary plateau when a perception of mild pain isreported during the local heating stimulus. Both Kellogget al. (1999) and Minson et al. (2001) reported similarfindings regarding this phenomenon when local skintemperature was heated to >41◦C in conjunction witha brief sensation of pain (<5–10 s). Both groups observedan abolition of the biphasic increase in skin bloodflow (SkBF), which ultimately rendered the vasodilatorresponse insensitive to NOS inhibition. Sensory-nerve-mediated effects are thought to be produced via calcitoningene-related peptide (CGRP) and substance P, and sinceboth are colocalized in nerve terminals in human skin(Wallengren, 1997), it is believed that upon stimulationthey are released to increase vascular permeability anddilate adjacent blood vessels (Drummond, 2009). Thus, apainful stimulus might invoke a sensory-nerve-mediatedvasodilator response in an attempt to reduce thermaldamage to the vasculature. Presynaptic blockade ofsympathetic adrenergic nerves (Houghton et al. 2006;Hodges et al. 2009) or postsynaptic antagonism ofα-, β- and Y1 receptors (Hodges et al. 2008; Tew et al.2011) have been shown to reduce the plateau phase of thethermal hyperaemic response. These data suggest that theplateau phase is largely dependent on NO and an intactsympathetic noradrenergic system.

Therefore, we sought to determine the roles of sensorynerves and sympathetic nerves to a noxious heat stimulus(44◦C) in vivo in human skin. Two studies were performed,during which local heating was applied to four skin sites;two were heated to 42◦C (non-noxious) and two rapidly to44◦C (noxious). These heating protocols were performedin the presence of local sensory nerve blockade viatopical EMLA cream (study 1) and local sympathetic nerveblockade with bretylium tosylate (BT) via microdialysis(study 2). Owing to the element of pain from the noxiousheating protocol, we hypothesized that sensory nerveswould exhibit a marked involvement in the exaggerated,NO-independent vasodilator response to noxious localskin heating. However, upon finding only a modest rolefor sensory nerves, we performed a second study toexamine whether sympathetic nerves were involved inthe NO-independent vasodilator response to noxious skinheating.

Methods

Ethical approval

This study was approved by the local InstitutionalReview Board. Participants were fully informed of themethods and risks prior to participation. Verbal and

written informed consent was obtained from eachparticipant, and all protocols conformed to the guidelinesset forth by the Declaration of Helsinki.

Participants

Ten participants (six men and four women, 24 ± 2 yearsold, 174 ± 5 cm tall and weighing 70 ± 2 kg) volunteeredfor studies 1 and 2. All participants were healthynon-smokers, not taking any prescription medicationapart from oral contraceptives. All participants weremaintaining moderately active lifestyles and free from anycardiovascular disease. Participants were asked to refrainfrom consuming caffeine and alcohol for 12 h prior to thestudy. The menstrual status of the female participants wasrecorded; nevertheless, their responses did not differ fromthe men and their results were combined. All experimentswere performed in the morning hours.

Instrumentation and measurements

All experiments were performed in an environmentalchamber with ambient temperature controlled at22 ± 0.5◦C. All measurements were performed with theparticipants resting in the supine posture. Participantshad microdialysis probes placed intradermally onthe ventral aspect of the forearm as previouslydescribed (Hodges et al. 2006). These custom-builtprobes were 10 mm of microdialysis tubing (regeneratedcellulose, inner diameter 200 μm, 18 kDa nominalmolecular mass cut-off) attached at each end topolyimide tubing. Before implantation, the area of skinwas temporarily anaesthetized with the application of anice pack for 5 min (Hodges et al. 2009a). A 22 gaugeneedle was inserted aseptically into the dermal layer∼2.5 cm before exiting. The microdialysis probe was thenthreaded through the lumen of the needle, which wasthen removed, leaving the probe in place. All probes wereinserted in this manner. To allow for the effects ofthe insertion trauma to subside, we waited 90 min beforebeginning any experimental procedures (Anderson et al.1994). The different probes were placed 3–5 cm apart.Skin blood flow was measured from the ventral aspectof the forearm by laser-Doppler flowmetry (Moor VMS;Moor Instruments, Axminster, UK) and expressed as laser-Doppler flow (LDF; Johnson, 1990; Oberg, 1990). TheLDF measures are limited to the skin and are not affectedby underlying skeletal muscle blood flow (Saumet et al.1988). To control local skin temperature, local heatingdevices (SH02; Moor Instruments) were placed directlyover the microdialysis probes. Blood pressure was recordedevery 5 min by auscultation from the contralateral arm.Cutaneous vascular conductance (CVC) was calculated asthe ratio of LDF to mean arterial pressure and expressed in

C© 2011 The Authors. Journal compilation C© 2011 The Physiological Society

1210 S. J. Carter and G. J. Hodges Exp Physiol 96.11 pp 1208–1217

arbitrary units. Whole-body skin temperature (T sk) wasrecorded as the weighted mean from six thermocouplesplaced on the body surface (Taylor et al. 1989) andcontrolled by a water-perfused suit (Allen Vanguard,Ottawa, Ontario, Canada) covering the entire body surfaceapart from the head, hands and feet and the forearm usedfor the blood flow measurements. All variables werecollected at 50 Hz and stored for offline analysis (MP150;BioPac, Goleta, CA, USA).

Drugs

In study 1, sensory nerve blockade was achieved with theuse of a topical local anaesthetic cream (EMLA; Fougera,Melville, NY, USA) consisting of lidocaine (2.5%) andprilocaine (2.5%). The EMLA cream was applied to anarea of 8 cm2 under an occlusive dressing and left for aminimum of 2 h (Johnson et al. 2005; Hodges et al. 2007;Tew et al. 2011a). The area was tested for sensory loss bothbefore the study began and at its completion by pin-prickand pinching of the skin (Hodges et al. 2007; Tew et al.2011a).

In study 2, blockade of the transmitter release fromnoradrenergic sympathetic nerves was achieved by usingbretylium tosylate (US Pharmacopeia, Rockville, MD,USA). The BT was administered via microdialysis ata concentration of 10 mM for 60 min at a rate of4 μl min−1 (Hodges et al. 2009b; Tew et al. 2011b). Theadequacy of the BT blockade was tested at the end of theexperiment by 3 min of whole-body cooling, reducing T sk

from 34 to 31◦C (Kellogg et al. 1989; Hodges et al. 2006).In studies 1 and 2, completion of the heating protocol

was followed by administration of 10 mM of noradrenaline(NA) and 1 mg ml−1 of ascorbic acid (NA preservative;Sigma Aldrich, St Louis, MO, USA) via microdialysisat a rate of 4 μl min−1 for 5 min to determine whetherthe cutaneous blood vessels exposed to the noxious heatstimulus were still reactive to vasomotor influences.

Finally, all sites were heated to 44◦C, and 58 mM ofsodium nitroprusside (SNP; Tocris, Bristol, UK) wasinfused at 4 μl min−1 for 35 min to elicit maximalcutaneous vasodilatation (Kellogg et al. 2009, 2010, 2011).

Protocols

Study 1 was designed to test whether the abolition of thecharacteristic biphasic SkBF response and the removalof the NO dependence of the plateau phase during thevasodilator response to noxious skin heating (Kellogget al. 1999; Minson et al. 2001) were due to theinvolvement of cutaneous sensory nerves. Four skin siteson the ventral aspect of the forearm were instrumentedwith microdialysis fibres, LDF probes and heater probeholders as previously described. The treatments were as

follows: sites 1 and 3, untreated; and sites 2 and 4, treatedfor sensory nerve blockade with EMLA cream (Fig. 1).The protocol began with the local temperature heldat 33◦C for 10 min at all sites (baseline measures).Sites 1 and 2 were subjected to local heating to 42◦C (non-noxious), and sites 3 and 4 to rapid local heating to 44◦C(noxious), at which temperature all participants reporteda brief, mild sensation of pain (lasting approximately5–10 s). This enabled the measurement of cutaneousvascular responses to both skin heating protocols with andwithout blockade of sensory nerve function. Thereafter,all skin sites were perfused with NA (and ascorbic acid)administered via microdialysis to confirm that bloodvessels were still reactive to vasomotor influences followingnoxious heating. Finally, to achieve maximal cutaneousvasodilatation, SNP was administered to all sites, andsites 1 and 2 were heated from 42 to 44◦C at 0.5◦C (10 s)−1.Rapid local (noxious) heating involved the local heaterprobe holders being immediately changed from 33 to 44◦C.The probe holders took an average of 128 ± 12 s to reachthe desired 44◦C. For the ‘standard’ local heating protocol,probe holders were heated to 42◦C, which was performedat a rate of 0.5◦C (10 s)−1 and took 180 s in each case.

Study 2 was performed to examine the involvementof cutaneous sympathetic nerves during noxious heatexposure. Again, four sites on the ventral aspect of theforearm were instrumented with microdialysis fibres, LDFprobes and heater probe holders as previously described.The treatments were as follows: sites 1 and 3, untreated;and sites 2 and 4, treated for sympathetic nerve blockadevia microdialysis with BT (Fig. 1). Procedures for timing,sequence and heating protocols were the same as describedfor study 1. Rapid local (noxious) heating to 44◦C took anaverage of 127 ± 14 s for study 2.

Statistical analysis

Cutaneous vascular conductance was expressed relative tothe maximal values for each site as achieved in responseto 44◦C and 58 mM SNP, and all data are presented asmeans ± SEM. Owing to the rapid and transient nature ofthe initial peak response, stable 30 s periods of CVC wereused for analysis. For the secondary plateau and maximalCVC phases, stable 5 min periods of CVC were used foranalysis. Finally, a stable 1 min plateau in response to NAinfusion was used to determine whether the cutaneousvasculature was still responsive (see Fig. 1). Absolutevalues for baseline CVC were used to determine whethersites were at unusually high or low values as a resultof the pharmacological background or the preparatoryprocedures; there were no such cases present in thisstudy. Initial peaks, plateau phases and responses to NAwere analysed by repeated-measures ANOVA. Statisticalsignificance was assumed when P < 0.05.



Results

Sensory nerve involvement

Figure 2 shows the results from a representative participantin study 1, wherein we tested the involvement of cutaneoussensory nerves in the vasodilator response during anoxious heat exposure. In response to local heating to42◦C, a blunted axon reflex was observed at skin sitestreated with EMLA (47 ± 5%CVCmax) compared withthe untreated control sites (70 ± 7%CVCmax; P < 0.05;Fig. 2A versus B). In the presence of noxious heating(rapid increase to 44◦C), the characteristic biphasicresponse was abolished at both control and EMLA-treated skin sites. Plateau values reached during sustained

local skin heating to 42◦C heating did not differbetween control (88 ± 3%CVCmax) and EMLA-treatedsites (85 ± 3%CVCmax; P > 0.05; Fig. 3). In response tonoxious heating, the vasodilator response at the controlsite was increased (94 ± 1%CVCmax; P < 0.05 comparedwith all other skin sites). However, at the skin sites treatedwith EMLA, the vasodilator response to 44◦C heatingwas significantly lower than at the corresponding controlskin site (86 ± 5%CVCmax; P < 0.05; Fig. 3). There wereno differences between EMLA-treated sites regardless oflocal temperature differences (P > 0.05). Apart from theEMLA-treated sites (sites 2 and 4), all participants reporteda brief, mild sensation of pain during the rapid localheating to 44◦C (noxious) but not to 42◦C. Additionally,

Site 1: rol

rolSite 3:

3

3

Tloc

Baseline

10 min

Local warming

35 min

4

Site 4:

Site 2:

Tskin

NA

10 min

Tloc

Tskin

Tloc

Tskin

Tloc

Tskin

4

3

3

4 4

3

3

4 4

3

3

4 4

3

3

3

Tloc

Baseline

10 min

Drug infusion

60 minLocal warming

35 min

4

4

3

3 4

3

3

3

Tskin

4

NA

10 min

Tloc

Tskin

Tloc

Tskin

Tloc

Tskin

4

4

4

4

Site 1: 4

Site 3: 44

Site 4:

Site 2:

SNP

35 min

4

4

SNP

35 min

4

4

Stu

dy

1S

tud

y 2

Figure 1. Schematic diagram of the protocol for studies 1 and 2In both studies, four skin sites were instrumented with microdialysis fibres. In study 1, EMLA cream was thenapplied at two random sites for ∼2 h to block sensory nerves; these sites were then designated as sites 2 and 4.In study 2, ∼2 h were allotted for resolution of the insertion trauma. Subsequently, laser-Doppler probes and localheating probe holders were placed at each skin site. Whole-body skin temperature (T sk) was held at 34◦C andlocal skin temperature (T loc) at 33◦C. In both studies, 10 min baseline was recorded, with all sites perfused withsaline. In study 2, bretylium tosylate (BT) was infused via microdialysis for 60 min to achieve sympathetic nerveblockade. In both studies, local heating was performed for 35 min to either 42 (standard) or 44◦C (noxious). Thiswas followed by infusion of noradrenaline (NA) to test vasomotor function following the noxious heating protocoland finally, to achieve maximal cutaneous vasodilatation, all skin sites were infused with sodium nitroprusside(SNP) and skin sites 1 and 2 were slowly heated to 44◦C.



all four sites responded in a similar manner to NA infusionfollowing skin heating (�%CVCmax 40 ± 3, 40 ± 2, 41 ± 3and 43 ± 3 at sites 1–4, respectively), indicating that thecutaneous vasculature had not been damaged by thepainful temperature and was still vasoactive.

Sympathetic nerve involvement

In study 2, we examined the role of sympathetic nervesin the vasodilator response to local heating to 42◦C(non-noxious) and 44◦C (noxious). Figure 4 shows datafrom a representative participant in study 2. Again, notethe absence of a characteristic biphasic response withseparate initial peak and nadir and the exaggeratedplateau phase during noxious heating (Fig. 4C versusD). Axon reflexes were significantly lowered at skin sitestreated with BT (59 ± 6%CVCmax) compared with controlsites (71 ± 5%CVCmax) during local heating to 42◦C(P < 0.05; Fig. 4). The plateau phase differed significantlybetween the untreated control sites (86 ± 4%CVCmax)and the BT-treated sites (78 ± 3%CVCmax) during heating

100

80

6042˚C Control 42˚C EMLA 44˚C Control 44˚C EMLA

70

90

*

CV

C (

%m

ax

)

Figure 3. The plateau phase vasodilatation in cutaneousvascular conductance (CVC) in response to the local skinheating in study 1Blockade of sensory nerves with EMLA does not alter the vasodilatorresponse to 42◦C skin heating. However, EMLA treatment significantlydecreases the response to rapid skin heating to 44◦C.

∗P < 0.05

compared with both 42◦C conditions; †P < 0.05 compared with 44◦Ccontrol.

0

20

40

60

80

100

0 5 10 15 20 25 30 35 40 45 50 55

0

20

40

60

80

100

0 5 10 15 20 25 30 35 40 45 50 55

0

20

40

60

80

100

0 5 10 15 20 25 30 35 40 45 50 55

A B

DC

CV

C (

%m

ax

)

Time (min)

0

20

40

60

80

100

0 5 10 15 20 25 30 35 40 45 50 55

Figure 2. A representative trace from a single participant in study 1A, control 42◦C. B, EMLA 42◦C. C, control 44◦C. D, EMLA 44◦C. First vertical line indicates starting of local heatingand second vertical line indicates start of infusion of noradrenaline. As previously shown, there was a reducedinitial peak with EMLA during heating at 42◦C and no difference in the plateau phase (Minson et al. 2001; Tewet al. 2011a). During conditions of rapid local heating to 44◦C at both untreated and EMLA-treated sites therewas no initial peak. The plateau phase at the EMLA-treated site rapidly heated to 44◦C was lower than that at thecontrol site heated to 44◦C. The plateau phase was increased at the 44◦C control site relative to both 42◦C sites.



to 42◦C (P < 0.05; Fig. 5). In the control conditions,the introduction of 44◦C heat yielded a significantincrease in CVC compared with 42◦C heat exposure(95 ± 3%CVCmax); this value was higher than all otherconditions (P < 0.05). In conditions of BT administration,the degree of vasodilatation achieved in response tonoxious heating (79 ± 3%CVCmax) was significantly lowerthan at the control site (P < 0.05), and the vasodilatorresponse did not differ significantly from the plateau phaseachieved in response to 42◦C heating (P > 0.05; Fig. 5).Again, all four skin sites responded in a similar manner toNA infusion (�%CVCmax 42 ± 3, 38 ± 4, 43 ± 7, 38 ± 3at sites 1–4, respectively), verifying that the cutaneousblood vessels exposed to the painful stimuli had not been‘damaged’.

Discussion

Data from previous research have shown that the plateauphase of the vasodilator response is heavily dependent(∼70%) on NOS activity (Kellogg et al. 1999). However,

100

80

6042˚C Control 42˚C BT 44˚C Control 44 C BT

70

90

CV

C (

%m

ax

)

* *

Figure 5. The plateau phase vasodilatation in cutaneousvascular conductance (CVC) in response to the local skinheating in study 2Blockade of noradrenergic sympathetic nerves with BT significantlyreduced the vasodilator response to 42◦C skin heating. This decreasedvasodilatation was also observed during BT treatment with rapid skinheating to 44◦C.

∗P < 0.05 compared with control conditions;

†P < 0.05 compared with all other conditions.

0

20

40

60

80

100

0 5 10 15 20 25 30 35 40 45 50 55

0

20

40

60

80

100

0 5 10 15 20 25 30 35 40 45 50 55

0

20

40

60

80

100

0 5 10 15 20 25 30 35 40 45 50 55

A B

DC

0

20

40

60

80

100

0 5 10 15 20 25 30 35 40 45 50 55

CV

C (

%m

ax

)

Time (min)

Figure 4. A representative trace from a single participant in study 2A, control 42◦C. B, bretylium tosylate (BT) 42◦C. C, control 44◦C. D, BT 44◦C. First vertical line indicates startingof local heating and second vertical line indicates start of infusion of noradrenaline. As previously shown, theinitial peak with BT was somewhat reduced during heating to 42◦C and the plateau phase was depressed. Duringconditions of rapid local heating to 44◦C (noxious) at both the untreated and the BT-treated site, there was noinitial peak. The plateau phase at the BT-treated site rapidly heated to 44◦C was lower than that at the control siteheated to 44◦C. The plateau phase was increased at the 44◦C control site relative to both 42◦C sites.



it has been reported that in the presence of a noxiousheat stimulus the plateau phase is preserved despite NOSinhibition (Kellogg et al. 1999; Minson et al. 2001). Tothis end, we sought to examine these previous findings inorder to investigate what mechanisms were contributing tothe sustained hyperaemia in response to noxious localizedskin heating. The major findings of this investigation arethat localized exposure to a rapid, noxious (mild sensationof pain) heating (44◦C) abolishes the classic biphasicincrease in SkBF. Additionally, sensory nerve blockadewith topical analgesic cream (EMLA) significantly reducedthe vasodilator response, suggesting a role for sensorynerves during a noxious heat exposure. Furthermore,the infusion of bretylium tosylate via microdialysisfibres significantly attenuates the plateau phase duringthe cutaneous hyperaemia, suggesting a pronouncedsympathetic nerve involvement.

Sensory nerves

Among its many functions, the integumentary system is aremarkably effective organ to regulate body temperature.During exposure to a host of stimuli, an array ofbiological mediators (e.g. bradykinin and prostaglandins)are secreted by resident skin cells and, together withsensory nerves, provide an integrated defence againstpotential threats (Metz & Maurer, 2009; Galli & Tsai, 2010).Through a series of experiments using a standardized localheating protocol, Minson et al. (2001, 2002) witnesseda transitory vasodilator response (e.g. initial peak andnadir). However, this biphasic response was significantlyreduced (>50%) following the topical application ofEMLA cream. This notable feature suggested a significantrole for thermally sensitive afferents in the forearm dermis.More recently, Wong & Fieger (2010) determined thattransient receptor potential vanilloid type 1 (TRPV1)channels localized within sensory afferents contributegreatly to the initial peak and nadir of the hyperaemicresponse. Much of the work investigating TRP channelshas identified a subfamily of vanilloid receptors withpain- and heat-sensing capabilities, yet their individualactivation thresholds exhibit clear distinctions (Numazaki& Tominaga, 2004; Willis, 2009). More specifically, TRPV1channels have an established high-temperature thresholdthat works to depolarize its sensory afferents directly inresponse to >42◦C heat (Mandadi et al. 2004; Numazaki& Tominaga, 2004; Caterina, 2007).

With the exception of EMLA-treated sites in the presentstudy, we found that the initiation of noxious heatingresulted in a mild sensation of pain lasting 5–10 s inall participants. Interestingly, despite sensory blockadethe localized noxious heat eliminated the characteristicbiphasic increase in SkBF, similar to reports from others(Kellogg et al. 1999; Minson et al. 2001). Althoughthe present study did not directly investigate TRP

channels, they are considered to be heat sensing and,more importantly, pain sensing. Recently, Wong & Fieger(2010) reported that TRPV1 channels contributed to thecutaneous thermal hyperaemia response in skin. Thus, apainful heat stress localized to a small area of skin, as usedin the present study, might induce an autonomic responsevia TRPV1 stimulation, causing a rapid and exaggeratedvasodilator response.

While various cutaneous neuropeptides and theirfunctions have been identified, the most commonlycited to have roles in sensory-nerve-mediated effectsare CGRP and substance P (Schmelz & Petersen, 2001).As both CGRP and substance P are colocalized innerve terminals in human skin (Wallengren, 1997), itis believed that upon stimulation they are released toincrease vascular permeability and dilate adjacent bloodvessels (Drummond, 2009). Likewise, these qualitiesseem to mimic the vasculature responses reportedby Drummond (2009, 2010a), where the resultantvasodilatation of cutaneous blood vessels or nearby ‘flare’(5–10 mm from stimulation site) ultimately provides anopportunity to disperse harmful threats that may disrupthealthy functioning. In humans, the axon-reflex-mediatedcomponent of the cutaneous hyperaemia is believed toinvolve CGRP and substance P (Sauerstein et al. 2000).Recently, Wong & Minson (2011) discovered a regulatoryrole for neurokinin-1 (NK1) receptors during thecutaneous hyperaemia from localized heating. They foundthat pretreatment with substance P desensitized the NK1

receptors (Wong & Minson, 2006), and this desensitizationaltered the initial peak and nadir but did not affectthe plateau phase of local heating hyperaemia (Wong& Minson, 2011). Given that substance P preferentiallybinds to NK1 receptors, these data strongly support a rolefor substance P in the cutaneous vasodilator response tolocal skin heating. In addition to the effects of CGRP andsubstance P, on vascular smooth muscle and endothelialcells, these neurotransmitters communicate with residentskin cells (Drummond, 2010b). A resultant ‘two-way’ linkis established between resident skin cells and sensorynerves. Accordingly, fine-tuned responses can be madewhile simultaneously interacting with the central nervoussystem; a feature that readily underscores the redundancyof the cutaneous vasculature and its ability to respondquickly to a variety of environmental stressors.

Sympathetic nerves

While it may seem counter-intuitive to assign a role forsympathetic nerves to cutaneous vasodilatation duringskin heating, growing evidence exists in support of thisassertion (Houghton et al. 2006; Hodges et al. 2008,2009b; Tew et al. 2011b). In this context, we sought toexamine whether sympathetic nerves were involved inthe augmented cutaneous vasodilator response to noxious



skin heating. We found, in conditions of sympatheticnerve blockade, that the biphasic rise in SkBF wasabsent during rapid noxious local heating. Ultimatelythough, the plateau phase of the vasodilator responseproduced in conditions of sympathetic blockade didnot differ between the two heating protocols. Likewise,Charkoudian et al. (2002) examined SkBF responsesto local heating in participants who had previouslyundergone surgical sympathectomy. Interestingly, thecharacteristic initial peak and nadir were absent, whereasthe plateau phase seemed unaffected. These observationsappear to indicate a sympathetic contribution during theaxon-reflex-mediated portion of the hyperaemic response.Additionally, it could be that the deficient sympatheticinvolvement led to an earlier than normal influence of theNO system, thereby obscuring the biphasic hyperaemicresponse (Johnson & Kellogg, 2010).

Our data suggest that the sympathetic nerves are highlyinvolved in the hyperaemic response to noxious heating.In the presence of a painful heating stimulus, the biphasicpattern in SkBF is abolished and leads to higher bloodflow (%CVCmax) compared with 42◦C local heating.These data suggest that the ‘pain’ element associatedwith the heating protocol was the critical elementthat markedly increased SkBF levels. Along these lines,nociceptor-mediated vasodilatation has been theorized tobe the most powerful vasodilator mechanism in the skin(Habler et al. 1997). As local heating becomes relatively‘painful’ (>40◦C), vasoactive peptides are released in largequantities by nociceptive primary afferents, resulting ina significant increase in SkBF (Magerl & Treede, 1996).The distinctive SkBF patterns witnessed in the presentstudy seem to support claims made by Minson et al.(2002), who suggested that a smaller initial peak mightbe associated with a greater risk of local tissue damagein response to direct heat. From our observations, thisexplanation seems plausible and positively supports whyan initial peak was not seen during the 42◦C heating trials.Accordingly, the slow rate of heating may not have been‘deemed’ a threat to the underlying tissues, making thenoxious heat directly responsible for the drastic increasein SkBF. From a teleological perspective, the hyperaemic‘surge’ may serve as a safeguard, thereby maximizing theconvective properties for a localized area to overcome agiven thermal load.

It is possible that the drugs used (EMLA and BT) arenot acting only in the manner described and expected inthis investigation. While we have had success previouslywith these pharmacological agents in blocking sensory andsympathetic nerve function in the cutaneous vasculature,we have not used these drugs in conjunction with a noxiousheating stimulus. Our tests of the efficacy of the blockadesappeared to confirm their desired action, but we cannotrule out potential additional effects these agents might behaving that may alter the primary findings of this study.

In summary, we found that the blockade of sensoryand sympathetic nerves significantly attenuated the fullexpression of cutaneous vasodilatation in the presenceof a noxious heat stimulus (rapid, localized skinheating to 44◦C). These data suggest that sensoryand sympathetic nerves contribute to the resultantcutaneous vasodilatation. Furthermore, our data affirmthat a greater portion of this mechanism arises fromsympathetic influence. More importantly, the collectivecontributions of the neural pathways appear responsiblefor sufficiently increasing cutaneous blood flow. We cantherefore reasonably speculate that the redundant rolesshared by the sensory and sympathetic nerves work tosupport the cutaneous vasculature should it experience asudden hyperthermic stress, ultimately acting as a practicalsafeguard against possible tissue damage.

References

Anderson C, Andersson T & Wardell K (1994). Changes in skincirculation after insertion of a microdialysis probe visualizedby laser Doppler perfusion imaging. J Invest Dermatol 102,807–811.

Caterina MJ (2007). Transient receptor potential ion channelsas participants in thermosensation and thermoregulation.Am J Physiol Regul Integr Comp Physiol 292, R64–R76.

Charkoudian N, Eisenach JH, Atkinson JL, Fealey RD & JoynerMJ (2002). Effects of chronic sympathectomy on locallymediated cutaneous vasodilation in humans. J Appl Physiol92, 685–690.

Drummond PD (2009). Alpha-1 adrenoceptor stimulationtriggers axon-reflex vasodilatation in human skin. AutonNeurosci 151, 159–163.

Drummond PD (2010a). Inflammatory consequences ofcutaneous stimulation. Exp Neurol 222, 181–183.

Drummond PD (2010b). Sensory disturbances in complexregional pain syndrome: clinical observations, autonomicinteractions, and possible mechanisms. Pain Med 11,1257–1266.

Galli SJ & Tsai M (2010). Mast cells in allergy and infection:versatile effector and regulatory cells in innate and adaptiveimmunity. Eur J Immunol 40, 1843–1851.

Habler HJ, Wasner G & Janig W (1997). Interaction ofsympathetic vasoconstriction and antidromic vasodilatationin the control of skin blood flow. Exp Brain Res 113,402–410.

Hodges GJ, Chiu C, Kosiba WA, Zhao K & Johnson JM (2009a).The effect of microdialysis needle trauma on cutaneousvascular responses in humans. J Appl Physiol 106, 1112–1118.

Hodges GJ, Kosiba WA, Zhao K & Johnson JM (2008). Theinvolvement of norepinephrine, neuropeptide Y, and nitricoxide in the cutaneous vasodilator response to local heatingin humans. J Appl Physiol 105, 233–240.

Hodges GJ, Kosiba WA, Zhao K & Johnson JM (2009b). Theinvolvement of heating rate and vasoconstrictor nerves in thecutaneous vasodilator response to skin warming. Am JPhysiol Heart Circ Physiol 296, H51–H56.



Hodges GJ, Traeger JA 3rd, Tang T, Kosiba WA, Zhao K &Johnson JM (2007). Role of sensory nerves in the cutaneousvasoconstrictor response to local cooling in humans. Am JPhysiol Heart Circ Physiol 293, H784–H789.

Hodges GJ, Zhao K, Kosiba WA & Johnson JM (2006). Theinvolvement of nitric oxide in the cutaneous vasoconstrictorresponse to local cooling in humans. J Physiol 574,849–857.

Houghton BL, Meendering JR, Wong BJ & Minson CT (2006).Nitric oxide and noradrenaline contribute to thetemperature threshold of the axon reflex responseto gradual local heating in human skin. J Physiol 572,811–820.

Johnson JM, ed. (1990). the cutaneous circulation. In:Laser-Doppler Blood Flowmetry, (eds Shepherd AP & ObergPA), pp. 121–139. Springer, New York.

Johnson JM & Kellogg DL Jr (2010). Local thermal control ofthe human cutaneous circulation. J Appl Physiol 109,1229–1238.

Johnson JM, O’Leary DS, Taylor WF & Kosiba W (1986). Effectof local warming on forearm reactive hyperaemia. ClinPhysiol 6, 337–346.

Johnson JM, Yen TC, Zhao K & Kosiba WA (2005).Sympathetic, sensory, and nonneuronal contributionsto the cutaneous vasoconstrictor response to local cooling.Am J Physiol Heart Circ Physiol 288,H1573–H1579.

Kellogg DL Jr (2006). In vivo mechanisms of cutaneousvasodilation and vasoconstriction in humans duringthermoregulatory challenges. J Appl Physiol 100,1709–1718.

Kellogg DL Jr, Johnson JM & Kosiba WA (1989). Selectiveabolition of adrenergic vasoconstrictor responses in skin bylocal iontophoresis of bretylium. Am J Physiol Heart CircPhysiol 257, H1599–H1606.

Kellogg DL Jr, Liu Y, Kosiba IF & O’Donnell D (1999).Role of nitric oxide in the vascular effects of localwarming of the skin in humans. J Appl Physiol 86,1185–1190.

Kellogg DL Jr, Pergola PE, Piest KL, Kosiba WA, Crandall CG,Grossmann M & Johnson JM (1995). Cutaneous activevasodilation in humans is mediated by cholinergic nervecotransmission. Circ Res 77, 1222–1228.

Kellogg DL Jr, Zhao JL & Wu Y (2008). Endothelial nitric oxidesynthase control mechanisms in the cutaneous vasculature ofhumans in vivo. Am J Physiol Heart Circ Physiol 295,H123–H129.

Kellogg DL Jr, Zhao JL & Wu Y (2009). Roles of nitric oxidesynthase isoforms in cutaneous vasodilation induced by localwarming of the skin and whole body heat stress in humans.J Appl Physiol 107, 1438–1444.

Kellogg DL Jr, Zhao JL, Wu Y & Johnson JM (2010).VIP/PACAP receptor mediation of cutaneous activevasodilation during heat stress in humans. J Appl Physiol109, 95–100.

Kellogg DL Jr, Zhao JL, Wu Y & Johnson JM (2011).Antagonism of soluble guanylyl cyclase attenuates cutaneousvasodilation during whole body heat stress and localwarming in humans. J Appl Physiol 110, 1406–1413.

Magerl W & Treede RD (1996). Heat-evoked vasodilatation inhuman hairy skin: axon reflexes due to low-level activity ofnociceptive afferents. J Physiol 497, 837–848.

Mandadi S, Numazaki M, Tominaga M, Bhat MB, Armati PJ &Roufogalis BD (2004). Activation of protein kinase C reversescapsaicin-induced calcium-dependent desensitization ofTRPV1 ion channels. Cell Calcium 35, 471–478.

Metz M & Maurer M (2009). Innate immunity and allergy inthe skin. Curr Opin Immunol 21, 687–693.

Minson CT, Berry LT & Joyner MJ (2001). Nitric oxide andneurally mediated regulation of skin blood flow during localheating. J Appl Physiol 91, 1619–1626.

Minson CT, Holowatz LA, Wong BJ, Kenney WL & WilkinsBW (2002). Decreased nitric oxide- and axonreflex-mediated cutaneous vasodilation with age during localheating. J Appl Physiol 93, 1644–1649.

Numazaki M & Tominaga M (2004). Nociception and TRPchannels. Curr Drug Targets CNS Neurol Disord 3, 479–485.

Oberg PA (1990). Laser-Doppler flowmetry. Crit Rev BiomedEng 18, 125–163.

Pergola PE, Kellogg DL Jr, Johnson JM, Kosiba WA & SolomonDE (1993). Role of sympathetic nerves in the vascular effectsof local temperature in human forearm skin. Am J PhysiolHeart Circ Physiol 265, H785–H792.

Rowell LB (1974). Human cardiovascular adjustments toexercise and thermal stress. Physiol Rev 54, 75–159.

Sauerstein K, Klede M, Hilliges M & Schmelz M (2000).Electrically evoked neuropeptide release and neurogenicinflammation differ between rat and human skin. J Physiol529, 803–810.

Saumet JL, Kellogg DL Jr, Taylor WF & Johnson JM (1988).Cutaneous laser-Doppler flowmetry: influence of underlyingmuscle blood flow. J Appl Physiol 65, 478–481.

Schmelz M & Petersen LJ (2001). Neurogenic inflammation inhuman and rodent skin. News Physiol Sci 16, 33–37.

Taylor WF, Johnson JM, Kosiba WA & Kwan CM (1989).Cutaneous vascular responses to isometric handgripexercise. J Appl Physiol 66, 1586–1592.

Taylor WF, Johnson JM, O’Leary D & Park MK (1984). Effectof high local temperature on reflex cutaneous vasodilation. JAppl Physiol 57, 191–196.

Tew GA, Klonizakis M, Moss J, Ruddock AD, Saxton JM &Hodges GJ (2011a). Role of sensory nerves in the rapidcutaneous vasodilator response to local heating in young andolder endurance-trained and untrained men. Exp Physiol 96,163–170.

Tew GA, Saxton JM, Klonizakis M, Moss J, Ruddock AD &Hodges GJ (2011b). Aging and aerobic fitness affect thecontribution of noradrenergic sympathetic nerves to therapid cutaneous vasodilator response to local heating. J ApplPhysiol 110, 1264–1270.

Wallengren J (1997). Vasoactive peptides in the skin. J InvestigDermatol Symp Proc 2, 49–55.

Willis WD Jr (2009). The role of TRPV1 receptors in painevoked by noxious thermal and chemical stimuli. Exp BrainRes 196, 5–11.

Wong BJ & Fieger SM (2010). Transient receptor potentialvanilloid type-1 (TRPV-1) channels contribute to cutaneousthermal hyperaemia in humans. J Physiol 588, 4317–4326.



Wong BJ & Minson CT (2006). Neurokinin-1 receptordesensitization attenuates cutaneous active vasodilatation inhumans. J Physiol 577, 1043–1051.

Wong BJ & Minson CT (2011). Altered thermal hyperaemia inhuman skin by prior desensitization of neurokinin-1receptors. Exp Physiol 96, 599–609.

Acknowledgements

The authors would like to thank the participants forvolunteering. This study was supported by The University ofAlabama.


Documents

Sensory and sympathetic nerve contributions to the cutaneous vasodilator response from a noxious heat stimulus