This study has characterized constrictions of small cutaneous arteries in the guinea pig ear in response to electrical stimulation of the cervical sympathetic nerve (SNS) in vivo. Video microscopy and on-line image analysis were used to examine diameter changes of ear arteries (80–140 μm resting diameter) in anesthetized guinea pigs. Trains of 50–300 impulses, but not single pulses or short trains, produced frequency-dependent (2–20 Hz) constrictions. The purinoceptor antagonist suramin (30 μM) greatly reduced constrictions produced by exogenous ATP but did not affect constrictions produced by SNS at 10 Hz or exogenous norepinephrine. The α2-adrenoceptor antagonist yohimbine (1 μM) enhanced the peak amplitude of sympathetic constrictions at lower stimulation frequencies (1–5 Hz). The amplitude of constrictions to SNS at 10 Hz was reduced, and the latency of constrictions was increased by the α1-adrenoceptor antagonist prazosin (1 μM). Constrictions to SNS at 10 Hz remaining after prazosin treatment were reduced in amplitude by dihydroergotamine (2 μM) and were attenuated further by the neuropeptide Y Y1-receptor antagonist 1229U91 (0.3 μM). Thus norepinephrine and neuropeptide Y act as cotransmitters to mediate sympathetic constriction of small ear arteries at higher stimulation frequencies (10 Hz), but ATP does not seem to contribute directly to these constrictions.
- cutaneous vasculature
- adenosine 5′-triphosphate
- neuropeptide Y
sympathetic vasoconstrictor neurons typically utilize two or more cotransmitters. The most common cotransmitters are norepinephrine (NE), ATP, and neuropeptide Y (NPY) (16, 19, 25). There is good evidence that the roles of these cotransmitters vary considerably between different regions of the vasculature and with different types of sympathetic stimulation (16, 19, 25).
Cutaneous sympathetic neurons play an important role in regulating blood flow to the skin to conserve body heat in a cold environment (9,10) or as part of the alerting responses (29). Different populations of sympathetic neurons innervate proximal arteries compared with more distal cutaneous arteries and can be activated selectively by specific physiological stimuli (14). These subpopulations of sympathetic neurons can contain characteristic combinations of cotransmitters (8). Furthermore, the distribution of postsynaptic receptors mediating vasoconstriction in response to exogenous agonists varies throughout the cutaneous vascular tree (22, 23).
The high degree of heterogeneity in innervation pattern and receptor distribution throughout the cutaneous vasculature suggests that sympathetic neurotransmission may also vary between vascular segments. Indeed, a recent electrophysiological study has demonstrated clear differences in neurotransmission between the main ear artery and small distributing arteries isolated from the guinea pig ear (26). Nevertheless, that study indicated that the cotransmitters NE and ATP mediated membrane potential changes in response to electrical field stimulation in both proximal and distal ear arteries. Despite pharmacological demonstration of a class of unusual adrenoceptors and NPY Y1 receptors on small ear arteries (22, 23), there was no evidence that these receptors contributed to membrane potential changes produced by field stimulation (26). Furthermore, in small ear arteries the excitatory junction potentials (EJPs) mediated by ATP did not sum to produce active membrane responses and smooth muscle contraction, as occurred in the main ear artery (26). These results suggest that NE acting on unusual adrenoceptors and NPY acting on Y1receptors do not contribute to sympathetic vasoconstriction of small ear arteries. Alternatively, they may do so without producing changes in smooth muscle membrane potential. Furthermore, the electrophysiological results suggest that although ATP is released from sympathetic nerve terminals to produce EJPs, these membrane responses may not contribute to sympathetic vasoconstriction.
The aim of the present study was to examine sympathetic vasoconstriction of identified small ear arteries in vivo by directly observing the changes in real time of the arterial diameter produced by electrical stimulation of the cervical sympathetic chain in anesthetized guinea pigs and to compare these responses with the previous electrophysiological results. Specifically, the roles of ATP, NE, and NPY in sympathetic vasoconstriction were determined by local application of adrenoceptor antagonists, purinoceptor antagonists, and the NPY Y1-receptor antagonist 1229U91 to the ear vasculature.
MATERIALS AND METHODS
White-eared guinea pigs of either sex (Hartley-IMVS strain, 260–320 g body wt, n = 45) were injected with atropine (0.14 mg/kg sc) and were anesthetized initially with a mixture of ketamine (40 mg/kg im), xylazine (8 mg/kg im), and diazepam (2.5 mg/kg ip). Animals were ventilated (without paralysis) with room air plus oxygen via a tracheal cannula, and arterial blood pressure of each guinea pig was monitored continuously via a catheter in the femoral artery. Arterial blood gases were monitored at 2-h intervals, and ventilation parameters were adjusted to maintain at 35–45 mmHg, at 90–110 mmHg, and pH at 7.35–7.50. The level of anesthesia was maintained by infusion of ketamine (30–40 mg ⋅ kg−1 ⋅ h−1), and core temperature was kept constant at 37°C by a homeothermic blanket with a rectal temperature probe. The left vagus and sympathetic nerve trunks were transected at midcervical level, and a bipolar platinum stimulating electrode was placed around the sympathetic nerve trunk distal to the transection. After hair was removed from the ears with the use of a depilatory cream (active ingredient thioglycollate), selected regions of the dorsal vasculature of the left ear were exposed by gently peeling off small areas of skin. Animals were placed on the stage of a Leitz Laborlux microscope, and the left ear was placed in a Perspex bath for transillumination (Fig.1). Ears were superfused with HEPES-buffered balanced salt solution (composition in mM: 146 NaCl, 4.7 KCl, 0.6 MgSO4, 1.6 NaHCO3, 0.13 NaH2PO4, 2.5 CaCl2, 7.8 glucose, 20 HEPES, and 0.1 ascorbic acid) at 28°C, the surface temperature of the ear. The reservoir of HEPES-buffered solution, but not the ear bath, was bubbled with 100% O2 to ensure adequate oxygenation of large subdermal vessels after removal of overlying skin and associated microvasculature. After section of the cervical sympathetic nerve trunk, ear arteries typically maintained a resting diameter of 75–100% of their maximum diameter (23).
Vessels were imaged with a Nikon ×10 water immersion lens, and images were captured with a Panasonic CCD WV-600 video camera connected to a high-resolution monitor and to a PC-386 personal computer. The internal diameter of identified arteries with a branch order of 3 or 4 (see Ref. 22) was determined on-line using the DIAMTRAK DTVIVO program (developed and supplied by Dr. T. O. Neild). This program determines the internal diameter by tracking the edges of the dark column of blood, contrasted against the lighter arterial wall and surrounding tissue (see Fig. 2
The cervical sympathetic nerve trunk was stimulated with single pulses or trains of pulses of 0.1- to 0.3-ms duration and supramaximal voltage, delivered at frequencies ranging from 1 to 40 Hz by a Grass S11 stimulator. Agonist drugs were delivered close to the adventitial surface of short arterial segments in small volumes (50–75 nl) of known concentration via a micropipette using a motorized injector (Nanoject, Drummond, Broomall, PA) as described previously (22, 23). Antagonists were added to the reservoir of the superfusate and were applied to the adventitial surface of exposed regions of the ear vasculature for 10–30 min before agonists or electrical stimulation were retested. In many experiments the cumulative effects of more than one receptor antagonist were examined, although full pharmacological characterization of responses was not possible in any single experiment. Changes in arterial diameter before and after successive antagonists were included in the quantitative analysis only if resting arterial diameter between stimuli varied by <20% throughout each experiment. At the conclusion of the experiments animals were given a lethal dose of ketamine and exsanguinated via the femoral artery catheter. These experiments have been approved by the Flinders University Animal Welfare Committee.
Drugs used were ATP, benextramine hydrochloride, dihydroergotamine tartrate, guanethidine sulfate, norepinephrine hydrochloride (arterenol), prazosin hydrochloride, suramin, tetrodotoxin, yohimbine hydrochloride (all obtained from Sigma, St. Louis, MO), NPY (porcine sequence, supplied by Dr. T. O. Neild and sequenced by Dr. Roger Murphy), and 1229U91 (synthesized by Dr. Roger Murphy, Department of Pharmacology, University of Melbourne; Refs. 5 and 18).
Group data are expressed as means ± SE. Data were analyzed using ANOVA (SPSS for Windows Release 6.1), repeated-measures ANOVA followed by contrast analysis (SAS, Release 6.04, SAS Institute, Cary, NC), or paired t-tests (SPSS).
Stimulation of the preganglionic nerve fibers in the cervical sympathetic nerve trunk with single pulses or short-pulse trains (2–20 pulses) did not change the diameter of small cutaneous arteries (80–140 μm resting diameter). Stimulation with trains of 50–300 pulses produced constrictions that were relatively uniform along an arterial segment (Fig. 2). Constrictions were frequency dependent (Table 1), and for a given number of pulses the maximum decrease in diameter occurred with stimulation frequencies of 10–20 Hz (Fig.3). The latency of these constrictions ranged from 4 to 8 s, and the maximum decrease in diameter ranged from 20 to 70%. Constrictions in response to electrical stimulation were reduced by 80–100% after the addition of tetrodotoxin (n = 3) or guanethidine (n = 4) to the solution superfusing the ear. With well-maintained anesthesia, arterial blood pressure did not change in response to SNS. Furthermore, arterial pressure was not affected by the addition of tetrodotoxin or guanethidine to the solution superfusing the ear; thus pharmacological agents applied in this way did not appear to enter the circulation in amounts sufficient to affect central cardiovascular parameters. Occasionally, small increases in arterial diameter were apparent following constrictions produced by SNS.
Effects of purinoceptor blockade. Suramin (30 μM) was tested for its ability to antagonize constrictions produced by exogenous agonists or SNS. Brief local application of ATP (0.1–10 mM) produced rapid dose-dependent vasoconstrictions (n = 4), which were unaffected by guanethidine or tetrodotoxin. Suramin (30 μM) produced a large reduction in the amplitude of constrictions produced by a submaximal concentration of ATP (0.3–1 mM;n = 11; Fig.4; Table 2). In contrast, suramin did not reduce constrictions produced by SNS with 300 pulses at 10 Hz (n = 4; Fig. 4; Table2). Furthermore, suramin did not affect constrictions produced by local application of a submaximal concentration of NE (10 μM;n = 4; Table 2). In a separate series of experiments (n = 3), α,β-methylene-ATP (10 μM) was applied to the reservoir of the solution bathing the ear. Large constrictions (90% reduction in arterial diameter) were produced that desensitized rapidly, but the arteries remained slightly constricted (10–20% reduction in resting diameter) in the continued presence of α,β-methylene-ATP. Constrictions in response to SNS (10 Hz for 30 s) were not affected by this α,β-methylene-ATP treatment.
Effects of adrenoceptor antagonists. Yohimbine (1 μM) consistently enhanced the peak amplitude of constrictions produced by SNS with 200 pulses at 1–5 Hz (n = 4; Table 3), and sometimes yohimbine also enhanced constrictions at 10 Hz (2 of 4 experiments). Constrictions produced by SNS with one or more frequencies in the range of 2–20 Hz always were reduced in amplitude after the addition of prazosin (0.3–1 μM) to the superfusate, in the presence (n = 2) or absence (n = 17) of yohimbine. Treatment with prazosin produced a 30–40% decrease in the amplitude of constrictions produced by SNS with 300 pulses at 10 Hz and a significant increase in the latency of constrictions from 4 to 7 s (Fig. 5). Subsequent treatment with yohimbine (1 μM) enhanced the peak amplitude of constrictions remaining after prazosin by up to 50% (n = 3).
Constrictions to SNS remaining in the presence of prazosin alone, or prazosin and yohimbine together, were not reduced further by treatment with the noncompetitive α-adrenoceptor antagonist benextramine (20 μM; n = 4). However, dihydroergotamine (1–10 μM, n = 10) consistently produced an additional reduction in the peak amplitude of sympathetic constrictions remaining after prazosin treatment (Fig. 5). A small constriction (10–25% reduction in diameter) to SNS at ≥5 Hz was still apparent after combined treatment with prazosin and dihydroergotamine (Figs. 5 and 6).
Effects of Y1-receptor antagonist 1229U91.
Addition of 1229U91 (0.3 μM) to the solution bathing the ear produced a transient constriction in some experiments (5 of 13). However, arterial diameter returned to resting levels within 10 min of the addition of the antagonist. Constrictions produced by topical application of NPY (10 μM) were abolished or greatly reduced in the presence of 1229U91 (0.3 μM, n = 4; Fig. 6; Table 4). Constrictions to SNS at 10 Hz remaining in the presence of prazosin (1 μM) and dihydroergotamine (2 μM) were reduced significantly in amplitude after addition of 1229U91 to the bathing solution (n = 5; Fig. 6; Table 4). The same concentration of 1229U91 did not affect constrictions produced by exogenous NE (n = 4; Table4).
This study has utilized a valuable in vivo preparation in anesthetized guinea pigs for direct observation and real-time analysis of constrictions of identified cutaneous arteries in response to electrical stimulation of preganglionic sympathetic neurons. This preparation is relatively noninvasive, yet agonist and antagonist drugs can be applied to the adventitial surface of only small regions of cutaneous arteries, thus isolating the drug actions from the systemic circulation. The ability of tetrodotoxin or guanethidine applied in this way to substantially reduce or abolish sympathetic constrictions demonstrates that the observed changes in diameter were due almost entirely to transmitter release from sympathetic axons innervating the exposed cutaneous arteries.
Stimulation parameters producing constrictions. Electrical stimulation of the cervical sympathetic nerve trunk with single pulses or trains of up to 300 pulses delivered at constant low frequencies (≤2 Hz) failed to change the diameter of small cutaneous arteries. In contrast, single pulses and short pulse trains of field stimulation produce biphasic membrane depolarization of smooth muscle cells in isolated guinea pig small ear arteries (26), demonstrating that these stimulation parameters can release neurotransmitters from local sympathetic nerve terminals to act on postjunctional receptors. This difference could be explained by a lower number of varicosities releasing neurotransmitters in response to transganglionic activation of postganglionic sympathetic neurons in vivo compared with a higher number of release sites directly depolarized by electrical field stimulation in vitro. The inability of EJPs in small ear arteries to initiate active membrane responses in the smooth muscle cells (26) may contribute to the absence of constrictions to low levels of sympathetic stimulation in vivo. This contrasts with the rat tail artery, where an EJP in response to a single impulse can trigger an active membrane response and associated smooth muscle contraction (4, 25).
Many studies of cutaneous sympathetic nerves and other vasoconstrictor nerves have demonstrated clearly that under resting conditions, single units fire with a mean frequency ≤1 Hz, although short bursts of firing of up to 20 Hz have been recorded. During reflex activation, the mean firing rate can increase 10-fold but usually still stays within the 1- to 3-Hz range with instantaneous frequencies up to 35 Hz (11,14, 15, 28). These data suggest that electrical stimulation of sympathetic nerves at a constant rate of 1–2 Hz should produce vasoconstriction. However, many studies now have shown that irregular patterns of stimulation, such as those that occur in vivo (11, 21), are more effective in producing vasoconstriction than stimulation at a regular rate with the same number of pulses and average frequency (1,2, 20, 27). Furthermore, the ratio of sympathetic cotransmitters released by irregular stimulation can differ from that in response to a regular stimulation rate (20). Nevertheless, it seems clear that all cotransmitters can be released by regular stimulation at higher frequencies (16, 20, 25). Thus, although it is difficult to equate the frequencies used in the present study with average sympathetic nerve activity in conscious animals or humans, the stimulation regime used here is likely to reveal the full repertoire of neurotransmitters contributing to sympathetic vasoconstriction of small ear arteries in guinea pigs during reflex activation of cutaneous vasoconstrictor neurons. However, the possibility remains that anesthesia of animals in the present study, a factor not present in in vitro studies or studies in conscious animals and humans, may have affected the actions of neurotransmitters or pharmacological agents.
Neurotransmitters producing sympathetic vasoconstriction. Suramin-sensitive purinoceptors mediated constriction of small cutaneous arteries to exogenous ATP in the present study. Furthermore, ATP can produce fast depolarizations (summed EJPs) in isolated arteries in response to field stimulation with trains of impulses at frequencies up to 20 Hz (26). However, ATP did not contribute to the sympathetic constriction of small arteries in the guinea pig ear in response to trains of stimuli at 10 Hz. This difference probably is due to the failure of summed EJPs to produce the active membrane responses required for contraction in isolated small arteries (26). This contrasts with the guinea pig main ear artery examined in vitro, where transmural stimulation at 10 Hz produces active membrane responses and contractions, which were partially blocked by suramin (24, 26). Similar stimulation parameters also produce a suramin-sensitive component of vasoconstriction in guinea pig submucosal arterioles (6). The failure of ATP-mediated EJPs to produce sufficient depolarization to trigger active membrane responses and consequent contractions may be a consequence of the relatively low density of innervation of small ear arteries (4, 26). Nevertheless, conditions may exist in conscious intact animals that would increase the probability of ATP-mediated depolarizations triggering active membrane responses and smooth muscle contraction.
NE released from postganglionic neurons after electrical stimulation of preganglionic sympathetic nerves with trains of pulses at frequencies ≥2 Hz clearly acts on both postjunctional adrenoceptors mediating constriction of small ear arteries and prejunctional adrenoceptors mediating inhibition of transmitter release. Although detailed pharmacological characterization of the postjunctional adrenoceptors was not possible with this in vivo preparation, the reduction in constrictions produced by prazosin indicates that α1-adrenoceptors mediated at least 30–40% of the constrictions at higher frequencies (5–10 Hz). This is consistent with conclusions drawn from a previous study using these same relatively high concentrations of prazosin and yohimbine in vivo, where contractions of small ear arteries >80 μm diameter produced by exogenous NE were reduced in amplitude by prazosin but not by yohimbine (23). The remaining constrictions were not substantially reduced by the nonselective adrenoceptor antagonist benextramine but were attenuated by dihydroergotamine. Dihydroergotamine was shown in a previous study to abolish constrictions of small ear arteries produced by locally applied NE, either in the presence or absence of prazosin and yohimbine, and lead to the proposal that part of the constriction produced by NE was mediated by a class of unusual adrenoceptors (23). It is not yet clear whether these adrenoceptors are similar to the non-α, non-β-adrenoceptors reported in previous studies (3, 7, 13, 17). However, there does not seem to be a significant change in membrane potential associated with the dihydroergotamine-sensitive component of the sympathetic vasoconstriction (26).
Even after blockade of adrenoceptors with prazosin and dihydroergotamine, up to 30–40% of the sympathetic vasoconstrictions to higher frequency stimulation remained. This residual response was reduced significantly by a newly developed antagonist for NPY Y1 receptors 1229U91 (5, 12, 18), indicating that NPY can be released from sympathetic nerve terminals to contribute to the vasoconstrictions. It is most likely that neuronally released NPY acted on Y1 receptors located postjunctionally on the vascular smooth muscle cells. Thus NPY probably acts as a cotransmitter with NE to produce sympathetic constriction of small cutaneous arteries through a mechanism involving little or no change in smooth muscle membrane potential (cf. 26).
In conclusion, this in vivo study has demonstrated that the sympathetic constriction of small cutaneous arteries in response to stimulation with trains of impulses is mediated by NE acting on α1-adrenoceptors and a population of dihydroergotamine-sensitive adrenoceptors and by NPY acting on Y1 receptors. However, the present study provides no evidence for direct participation by neuronally released ATP in these sympathetic vasoconstrictions in response to trains of pulses delivered at higher frequencies, despite the evidence that ATP released from sympathetic nerve terminals can depolarize smooth muscle cells in isolated small ear arteries. These findings emphasize the variable roles of the cotransmitters NE, ATP, and NPY in mediating sympathetic constriction of different segments of the vasculature. Although the precise combination of neurotransmitters regulating resistance in the cutaneous vasculature in intact animals during tonic or reflex activation of sympathetic pathways remains to be determined, it is likely that the cotransmitters and the receptors mediating sympathetic constriction will vary with both the level of sympathetic activity and the exact location in the vascular bed.
I am extremely grateful to Prof. J. Angus, Dr. R. Murphy, and Dr. M. Lew of the Department of Pharmacology, University of Melbourne, for the generous gift of 1229U91, and to Dr. T. O. Neild for development and supply of DIAMTRAK DTVIVO and for supplying crude neuropeptide Y. Rachel Perry and Pat Vilimas are thanked for expert technical assistance. Prof. Ian Gibbins is thanked for assistance with statistical analysis and for valuable comments on the paper.
Address for reprint requests and other correspondence: J. L. Morris, Dept. of Anatomy & Histology, Flinders Univ. of South Australia, GPO Box 2100, Adelaide, S.A. 5001, Australia (E-mail:).
This study was supported by grants from the National Health & Medical Research Council of Australia (NH&MRC, Reg. Key 950012), Flinders Medical Centre Foundation, Clive and Vera Ramaciotti Foundation, and Flinders University. J. L. Morris is a Senior Research Fellow of the NH&MRC.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 1999 the American Physiological Society