Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2011 Feb 28;589(Pt 9):2401–2414. doi: 10.1113/jphysiol.2010.201814

Changes in muscle sympathetic nerve activity and vascular responses evoked in the spinotrapezius muscle of the rat by systemic hypoxia

Steven Hudson 1, Christopher D Johnson 1, Janice M Marshall 1
PMCID: PMC3098710  PMID: 21486771

Non-technical summary

Hitherto, activity in sympathetic vasoconstrictor nerves that supply arterial vessels of skeletal muscle has been deduced from recordings made on mixed nerves that may be supplying skin, muscle or other tissue. We describe new methodology that allows direct recordings from nerve fibres on the surface of arterial vessels of muscle. The impulses occurred irregularly, but with rhythms that reflected heart rate and breathing rate. When oxygen levels in the blood were progressively lowered (hypoxia), muscle sympathetic nerve activity progressively increased from an average impulse frequency of 0.2 to one of 0.62 per second. Simultaneously, heart rate and respiratory rate increased, while our recordings of blood flow into the muscle showed that the arterial vessels dilated. We deduce that hypoxia causes an increase in the activity of the sympathetic nerves that supply muscle blood vessels that is mainly dependent on the stimulus for breathing, but their normal vasoconstrictor effect is overcome by local dilator influences.

Abstract

Abstract

Responses evoked in muscle sympathetic nerve activity (MSNA) by systemic hypoxia have received relatively little attention. Moreover, MSNA is generally identified from firing characteristics in fibres supplying whole limbs: their actual destination is not determined. We aimed to address these limitations by using a novel preparation of spinotrapezius muscle in anaesthetised rats. By using focal recording electrodes, multi-unit and discriminated single unit activity were recorded from the surface of arterial vessels. This had cardiac- and respiratory-related activities expected of MSNA, and was increased by baroreceptor unloading, decreased by baroreceptor stimulation and abolished by autonomic ganglion blockade. Progressive, graded hypoxia (breathing sequentially 12, 10, 8% O2 for 2 min each) evoked graded increases in MSNA. In single units, mean firing frequency increased from 0.2 ± 0.04 in 21% O2 to 0.62 ± 0.14 Hz in 8% O2, while instantaneous frequencies ranged from 0.04–6 Hz in 21% O2 to 0.09–20 Hz in 8% O2. Concomitantly, arterial pressure (ABP), fell and heart rate (HR) and respiratory frequency (RF) increased progressively, while spinotrapezius vascular resistance (SVR) decreased (Spinotrapezius blood flow/ABP), indicating muscle vasodilatation. During 8% O2 for 10 min, the falls in ABP and SVR were maintained, but RF, HR and MSNA waned towards baselines from the second to the tenth minute. Thus, we directly show that MSNA increases during systemic hypoxia to an extent that is mainly determined by the increases in peripheral chemoreceptor stimulation and respiratory drive, but its vasoconstrictor effects on muscle vasculature are largely blunted by local dilator influences, despite high instantaneous frequencies in single fibres.

Introduction

The pattern of respiratory and cardiovascular responses evoked by systemic hypoxia reflects a complex interaction of reflex, neurally- and hormonally- mediated influences superimposed upon local effects of hypoxia. In skeletal muscle, local vasodilator influences and effects of circulating adrenaline on β2-adrenoceptors overcome the vasoconstrictor influence of increased muscle sympathetic nerve activity (MSNA; Marshall, 1994). But, direct evidence for the latter comes from relatively few studies in which MSNA has been directly recorded (see Smith & Muenter, 2000).

Thus, in studies on humans, graded levels of systemic hypoxia induced by breathing 14–8% O2, evoked graded increases in MSNA (Saito et al. 1988; Rowell et al. 1989; Somers et al. 1989b). The extent of the increase in MSNA was attributable to the increase in central respiratory drive (CRD) caused by peripheral chemoreceptor stimulation and by barorecptor unloading caused by the hypoxia-induced fall in arterial blood pressure (ABP), but was limited by pulmonary stretch receptor stimulation and hypocapnia secondary to hyperventilation (Saito et al. 1988; Somers et al. 1989a; Guynet 2000). The only studies performed on animals showed that in anaesthetised, artificially ventilated cats, a single level of systemic hypoxia increased MSNA (Gregor & Janig, 1977; Blumberg et al. 1980); no study was made of MSNA during spontaneous respiration.

Apart from their paucity, these recordings of MSNA in hypoxia are limited because they were made on fibres in the mixed nerve supply of the limb that were judged to be sympathetic and destined for skeletal muscle vasculature on the basis of their firing characteristics (see Janig, 1988); their actual, final destination could not be established. Further, little attempt was made to analyse activity in single nerve fibres discriminated from the multiunit activity. This is important, firstly because an increase in multiunit activity cannot differentiate between an increase in activity in individual fibres and recruitment of fibres, and secondly because the patterning of activity in individual fibres plays crucial roles in determining the relative importance of the sympathetic co-transmitters noradrenaline, ATP and neuropeptide Y (NPY) in the evoked vasoconstrictor response (Huidoboro-Toro & Donoso, 2004).

The focal recording technique introduced by Johnson & Gilbey (1994), which allows sympathetic nerve recording from fibres running over the caudal ventral artery (CVA) of the rat tail by application of glass suction electrodes, apparently holds the potential to overcome some of these limitations (Johnson & Gilbey, 1994, 1996). Indeed, we recently used the technique to follow changes evoked by graded systemic hypoxia in single unit activity in sympathetic nerves on the CVA (Johnson et al. 2007). Thus, a primary aim of the present study was to establish whether we could use the technique to record MSNA from individual vessels of the spinotrapezius muscle of the spontaneously breathing rat, a preparation we have used to investigate responses evoked in the vasculature by changes in sympathetic activity and systemic hypoxia (e.g. Marshall, 1982; Mian & Marshall, 1991a,b,c;).

Having established that we could, we investigated the effects of graded levels of systemic hypoxia upon multiunit and single unit activity and modified the preparation so that we could continuously record blood flow to the spinotrapezius muscle whilst simultaneously recording femoral artery blood flow. These recordings, together with ABP, heart rate and respiration, allowed us to interpret the changes in MSNA and spinotrapezius vascular resistance in light of the factors that contribute to vascular responses evoked in hindlimb muscle of the rat by graded systemic hypoxia (e.g. Marshall & Metcalfe, 1988a,b, 1989).

Methods

Experiments were performed on 26 adult male Wistar rats (300–350 g body weight) under terminal anaesthesia, in accordance with the Animals (Scientific Procedures) Act 1986. At the end of the experiments, each rat was killed with an overdose of pentobarbitone sodium (60 mg ml−1; Sagatal, Rhone Merieux). For the experiments, anaesthesia was induced and maintained as we have previously described (e.g. Johnson et al. 2001, 2007). Briefly, the rat was placed in a Perspex box to allow induction of anaesthesia with 3% halothane in O2 delivered into the box at 3 l min−1. When the righting reflex was lost the rat was transferred to an operating table and anaesthesia was maintained via a face-mask with halothane in O2 at 3 l min−1. Meanwhile a jugular vein was cannulated so that anaesthesia could be maintained by continuous infusion of Saffan (Schering-Plough Animal Health, UK; 9 mg ml−1 alphaxalone/3 mg ml−1 alphadalone, diluted with 0.9% saline to 4 mg ml−1 total steroids) delivered by an infusion pump at 12–20 mg kg−1 h−1. Throughout the experiment, the depth of anaesthesia was adjusted so that the pedal withdrawal reflex was absent and arterial pressure (ABP) remained constant between experimentally induced stimuli.

The rat was then prepared to allow recording of physiological variables and to expose the spinotrapezius muscle as we have described before; the recording equipment we used was also essentially the same as we have recently described (see Mian & Marshall, 1991a; Johnson et al. 2001, 2007). In brief, a stainless steel T-shaped cannula was placed in the trachea so that tracheal pressure could be recorded via a pressure transducer and amplifier attached to a fine tube placed in the main shaft of the cannula. This signal was processed to allow respiratory frequency (RF) to be continuously recorded. The side arm of the cannula was connected via tubing to a rotameter system and thence to N2 and O2 cylinders. The rat breathed 21% O2 in N2 throughout, except during experimentally induced periods of hypoxia when the gas mixture was adjusted to 12, 10, or 8% O2 in N2 (see below). A cannula was placed in the left brachial artery so that samples (150 μl) of blood could be taken for analysis of the arterial partial pressures of O2 and CO2 (Inline graphic and Inline graphic, respectively) and pH, by using a blood gas analyser (IL 1640 or Nova phOX plus L, Nova Biomedical, Waltham, MA, USA). The left femoral artery was cannulated to allow ABP to be continuously recorded and the left vein was cannulated so that pharmacological agents could be given. Finally, a temperature probe was inserted into the rectum and connected to a thermostatically controlled blanket on which the animal was placed; core temperature was maintained between 36.5 and 37.5°C.

With the rat lying on its left side on a purpose-built Perspex board, the right spinotrapezius muscle was exposed via a cutaneous incision. The muscle was carefully separated from overlying and underlying tissue and fine silk threads were tied into the narrow strip of muscle that adjoins the lateral border of the spinotrapezius muscle proper at its rostral and caudal ends. The spinotrapezius was then arranged ventral surface uppermost on a Perspex column set into the board, the threads being used to secure the spinotrapezius muscle under tension to pins that were set into the board in a horseshoe shape around the column. In this way the spinotrapezius was adjusted to more or less its in situ size with intact circulation and nerve supply. With the aid of an operating microscope, fine forceps were then used to remove the fascia overlying the arterial vessels at the rostral end of the muscle so as to allow good access to the arteries and their innervation. A square piece of transparent polyvinyl film (Saran Wrap, Dow Chemical, Indianapolis, IN, USA), which is virtually impermeable to water, O2, and CO2, was arranged over the spinotrapezius muscle, adjacent exposed muscles of the shoulder and back and over the tethering threads (see above). The area overlying the spinotrapezius was traced with a marker pen and cut out. The spinotrapezius was then covered with a small piece of saline-soaked gauze whilst a bath was created to allow recording of activity from nerve fibres. Briefly, the Saran wrap was secured to the narrow muscle strip bordering the spinotrapezius by using super glue (UHU, UK, Middlesex, UK). Dental impression material (President light body, Coltène, Altstätten, Switzerland) was then piped on top of the Saran wrap and narrow muscle strip to create a fluid-tight seal around the muscle. Further impression material was piped onto the Saran wrap around the spinotrazius, outwards from the two ends of the seal that had already been created, to form three sides of rectangular wall 10–20 mm high. The fourth side of the rectangle was formed by the dorsal surface of the animal. Thus the walls were continued to abut the Saran wrap-covered muscle. The bath was filled with 0.9% saline in experiments in which nerve recordings were made and with a modified Krebs solution when blood flow was recorded (see below).

Blood flow recordings

In some experiments, blood flow was recorded from the femoral artery (femoral blood flow; FBF) and/or from the main artery supplying the spinotrapezius (spinotrapezius blood flow; SBF). For FBF, with the animal lying on its right side as described above, the right hindlimb was lifted, held in position by means of a stout ligature tied around the right ankle and secured to a clamp. The right femoral artery was exposed through a skin incision. The overlying inguinal fat pad was ligated and separated and the artery was placed in the trough of a transonic flow probe (0.5 V, Transonic Systems Inc., Ithaca, NY, USA); the probe was filled with acoustic coupling to facilitated recording. To record SBF, a 10–15 mm length of the main feed artery (thoracodorsal artery) was exposed just proximal to where it branches to give rise to the main artery that supplies the rostral end of the spinotrapezius (see Marshall, 1982). The artery was separated from the adjacent vein and placed on a transonic flow probe as described above. In each case the probe was connected to a dual channel flow meter (T206, Transonic Systems).

Nerve recordings

Recordings were made with glass suction electrodes pulled from borosilicate capillary tubing (GC150T-10, ID = 1.17 mm, OD = 1.5 mm, Harvard Apparatus, Holliston, MA, USA) by using a vertically mounted electrode puller (Ealing Beck Ltd, Watford, Herts, UK, UK). With the aid of a microscope, the tip was broken back to an internal diameter of 20–40 μm and then smoothed by advancing a heating coil towards it. A silver wire was inserted into the shank of the electrode and the proximal end was inserted into a Perspex holder (MEH2SW15, WPI, Sarasota, FL, USA). A collar sealed between the glass electrode and holder and connected to a 10 ml syringe allowed the pipette to be filled with saline and, when appropriate, suction to be applied (see below). The electrode holder was mounted on a micromanipulator (Prior Scientific Instruments, Fulbourn, UK) and the electrode was connected to a head stage pre-amplifier (NL 100AK Neurolog, Digitimer Ltd, Welwyn Garden City, UK). A silver reference electrode was placed near the recording site with its tip in the bath.

To record nerve activity, the electrode was moved over the surface of the spinotrapezius muscle with the aid of the micromanipulator whilst under observation through an operating microscope. The electrode was lowered until its tip lay on the surface of one of the arterial vessels that supplies the muscle from its rostral end. Successful placement of the electrode on a nerve fibre was confirmed by deflections that appeared to be action potentials; ongoing activity was simultaneously made audible via a speaker (D130, Neurolog, Digitimer). Light suction was then applied by means of the syringe as described above, to improve the signal to noise ratio and to try to ensure that the electrode remained in contact with the vessel wall even during changes in perfusion pressure or vessel tone.

Data acquisition

All data were collected to a computer (Sony Vaio PCV RX-204) via a lab interface 1401 micro (CED Ltd, Cambridge, UK). For cardiorespiratory variables, sampling frequencies were: ABP: 100 Hz, tracheal pressure: 50 Hz, FBF and SBF 100 Hz. On- and off-line analyses of these recordings were carried out by using Spike2 software (CED). Scripts were written for Spike2 to allow heart rate (HR) to be computed from the ABP recording and mean ABP (mABP), mean FBF (mFBF) and mean SBF (mSBF) to be computed from the raw signals. The latter recordings were used to compute femoral and spinotrapezius vascular resistance (FVR and SVR respectively) as mFBF and mSBF divided by mABP.

Nerve activity was sampled at 10,000 Hz, and signals were amplified (100,000–200,000 times) and filtered (15–5000 Hz). The raw multiunit recordings were analysed with the New Wave Mark facility of Spike2 software (v4.11) to show individual action potentials and to allow counts per unit time to be recorded. Single units were discriminated on the basis of shape, amplitude and duration by comparing the actual recordings of individual action potentials against templates for spikes that were constructed off-line for each data set. The number of data points needed to match the template was set at 65–80% and when at least eight action potentials met this criterion, the template was confirmed as one that described a unit. Action potentials that had been discriminated into separate templates were then re-drawn on a separate channel and checked for errors by using the New Wave Mark Editing facility of the Spike2 software. All multiunit and single unit activity was analysed off-line by using cross-correlation histogram analysis to check for cardiac- and respiratory-related rhythmicity (100 bins, width 10 ms and 50 ms, respectively, for each cardiac and respiratory cycle of the 60 s periods of analysis, see below). To this end, the Spike2 software was used to generate trigger signals from the nerve action potentials, and the systolic pressure peak of the ABP and expiratory peak of the tracheal pressure waveforms.

Protocols

In each rat, an attempt was made to record nerve activity from the surface of the main artery as it enters the rostral end of the muscle, or from one of its branches, the primary arterioles, vessels collectively ranging from 120 to 40 μm diameter (see Marshall, 1982). Attempts were also made to record nerve activity from the surface of the corresponding primary venules and main veins, collectively of ∼60–140 μm diameter. When a stable nerve recording had been achieved, one of the following protocols (Groups 1, 2 and 4) was performed. All of these protocols were performed on nerve activity that had cardiac- and respiratory-related rhythmicity (see Results). Within an individual experiment, if the recording from a particular site was lost before or during a protocol, a search was made for another site at which ongoing activity could be recorded. In practice, ongoing nerve activity with appropriate rhythmicities could not be recorded from ∼50–60% of rats. In these animals, experiments were performed on SBF and FBF (Group 3).

Group 1

In six rats (305 ± 8 g), when stable activity had been recorded, the effect of baroreceptor unloading was tested. After a 2 min baseline period, a bolus injection of the NO donor sodium nitroprusside (SNP, 60 μg kg−1i.v.) was given to produce a fall in ABP. In two of these rats, the same nerve recording was held long enough to test the response evoked by baroreceptor stimulation: a bolus dose of the α-adrenoreceptor agonist phenylephrine (PE, 10 μg kg−1i.v.) was given to produce a rise in ABP. In five nerve recordings made from these same rats, the effect of autonomic ganglion blockade was tested by administration of trimetaphan (50 mg kg−1, i.v.).

Group 2

In six rats (307 ± 12 g), tests were made of the effects of progressive graded systemic hypoxia on MSNA, and cardiovascular and respiratory variables. To this end, following a 2 min baseline period, the inspirate was switched in turn from 21% O2 to 12, 10 and then to 8% O2 each for 2 min, followed by return to 21% O2. Prior to induction of hypoxia and in the final minute of 8% O2, arterial blood was removed for blood gas analysis.

Group 3

In 10 rats (332 ± 8 g) in which recordings of MSNA were not possible, simultaneous recordings of SBF and FBF were made before and during progressive graded hypoxia as described for Group 2 above.

Group 4

In four rats (337 ± 7 g), MSNA and FBF were recorded before and during an extended period of systemic hypoxia: 8% O2 for 10 min, with arterial samples taken before and in the 10th minute of hypoxia.

Chemicals

Phenylephrine hydrochloride and SNP were obtained from Sigma-Aldritch Co. (Poole, UK); they were each stored frozen as a stock solution and diluted before the experiment with 0.9% saline. Trimetaphan camsylate was obtained from Cambridge Laboratories Ltd (Newcastle, UK; it was stored at 4°C and diluted with 0.9% saline. In the experiments in which SBF was recorded, the spinotrapezius muscle was superfused with a modified Krebs solution: 131.9 mm NaCl, 4.7 mm KCl, 1.17 mm MgSO4.7H2O, 2 mm CaCl2.2H2O, 22 mm NaHCO3, pH 7.35–7.45 (Marshall, 1982).

Data analyses

All data are expressed as means ± SEM. In Group 1, baseline frequencies of multi- and single unit MSNA and level of ABP over a 10 s period before injection of SNP were compared with those recorded over 10 s when the depressor response evoked by SNP reached its maximum and when ABP had stabilised during recovery, by using one-way ANOVA. For responses evoked by hypoxia in Groups 2, 3 and 4, recordings were analysed over the final 1 min of breathing 21% O2, in the second minute of each level of hypoxia and in the second minute following return to 21% O2: MSNA recordings were assessed for cardiac and respiratory-related rhythmicities, multi- and single unit activity was analysed for mean frequency and the latter for instantaneous frequency, while respiratory and cardiovascular variable were averaged. In Group 4, recordings were also analysed in the ninth minute of 8% O2. Repeated measures ANOVA was used to compare values recorded during each level of hypoxia with baseline, Fisher's post hoc test being used when P < 0.05. In all cases, P < 0.05 was considered statistically significant.

Results

General observations on nerve activity

Ongoing nerve activity recorded from the spinotrapezius was generally multiunit, containing two to three units of varying amplitudes and durations; on two occasions, a single unit only was apparent. Successful recording sites were on the surface of arterial vessels, usually towards the lateral edge of the vessel or over branching sites. Although attempts were made to record ongoing activity from the surface of venous vessels, none of these was successful. In most preparations, placement of the electrode also gave recordings of electrical activity that were much greater in amplitude than those normally recorded. These were often associated with twitching of the muscle fibres. Off-line analysis of these recordings showed action potentials whose amplitude and shape differed from the action potentials judged to be sympathetic in origin (see below) and they did not have cardiac- or respiratory-related rhythmicity. They may well have been recorded from somatic motor or sensory fibres.

Typical examples of the nerve recordings, described in more detail below, are shown in Fig. 1, which shows ongoing nerve activity together with recordings of respiratory air-flow and ABP (Fig. 1A) and the outcome of cross-correlation histogram analysis of nerve activity against ABP and respiration respectively (Fig. 1B and C). Such cardiac- and respiratory-related rhythmicity indicated that the activity could be regarded as MSNA (see Discussion): all recordings described below had these rhythmicities. As can be seen in Fig. 1D, action potentials in single units that were discriminated from multiunit activity were generally triphasic; sometimes they were biphasic as shown in the lower set of recordings. Irrespective, single unit activity also had strong cardiac- and respiratory-related rhythmicity (see Fig. 4). The mean frequency of the single unit activity and the instantaneous frequencies recorded in single units in normoxia and at different levels of hypoxia are shown in Table 2.

Figure 1. Nerve activity recorded from the surface of an arterial vessel of spinotrapezius under normoxic conditions.

Figure 1

Open in a new tab

A, raw multi-unit nerve activity. MSNA, respiratory air –flow; EXP, expiration; INS, inspiration; ABP, arterial pressure. B and C, examples of ABP- and tracheal pressure-triggered cross-correlation histograms showing the cardiac- and respiratory-related rhythmicities, respectively, of the multi-unit activity (histogram triggers: 480 and 144, respectively). D, multi-unit activity recorded from another animal under normoxic conditions. Below, 4 different examples of 4 or 5 superimposed action potentials discriminated from multi-unit activity.

Figure 4. Typical example of rhythms recorded in single unit activity in normoxia (21% O2) over 1 min before and over the second 1 min of progressive graded levels of hypoxia (12, 10, 8% O2).

Figure 4

Open in a new tab

A, systolic pressure-triggered cross-correlations (100 bins, width 0.01 s) shown above ABP recordings (histogram triggers: 348 and 363, 396, 480 from top down). B, expiratory-triggered cross correlations (100 bins, width 0.05 s) shown above recordings of respiratory airflow (histogram triggers: 96 and 114, 116, 120 from top down). NB, respiratory and cardiac-related rhythmicities present during each inspirate, but cardiac-related and respiratory-related rhythmicity became less, and more distinct, respectively, in severe hypoxia.

Table 2.

Mean firing frequencies and instantaneous firing frequencies of single unit activity, in normoxia (21% O2) and in the final minute of 12, 10 and 8% O2 in Group 2, and in the 2nd and 9th min of 8% O2 in Group 4

Mean firing frequency (Hz) Mean instantaneous firing frequency (Hz)
Group 2
 21% O2 0.2 (0.10–0.31) 0.23 (0.04–6)
 12% O2 0.28 (0.13–0.38) 0.3 (0.07–6.6)
 10% O2 0.33 (0.27–0.45) 0.35 (0.1–20)
 8% O2 0.62 (0.36–1.2) 1.29 (0.07–20)
Group 4
 21% O2 0.28 (0.21–0.35) 0.3 (0.09–10)
 8% O2 2rd min 0.68 (0.57–0.83) 1.5 (0.09–20)
 8% O2 10th min 0.46 (0.45–0.46) 0.73 (0.06–6.6)
Open in a new tab

In each case, the mean is shown, together with range in parentheses.

Group 1

As expected, in the six rats that were given bolus injection of SNP, ABP fell and HR showed a baroreceptor reflex-mediated tachycardia from 399 ± 11 to 418 ± 13 beats min−1. Over the following 1–1.5 min ABP and HR returned to control levels (Fig. 2). Meanwhile, the frequency of multiunit and single unit activity increased (Fig. 2). In the two animals in which responses evoked by PE were tested, ABP increased from 131 and 115 to 162 and 144 mmHg, respectively, accompanied by baroreceptor-mediated reflex decrease in HR from 431 to 331 and from 419 to 321 beats min−1, respectively. There was cessation of MSNA while ABP was increasing to its highest level, MSNA returning with restoration of ABP.

Figure 2. Responses evoked in MSNA by bolus injection of sodium nitroprusside (SNP).

Figure 2

Open in a new tab

Original recordings above: MSNA (analysed as spikes s−1), heart rate (HR; in beats min−1, b.p.m.), raw multi-unit activity and arterial pressure (ABP). Histograms below show mean ± SEM of mean ABP (mABP), multi-unit activity and single unit activity discriminated from multi-unit activity, over three 10 s periods, before (baseline), at the peak of the response evoked by SNP (SNP) and in recovery. ***P < 0.001, baseline vs. SNP (n = 6 for number of rats and for number of single units, i.e. one was analysed in each animal).

In the five rats that were given trimetaphan, ABP fell from 129 ± 11 to 70 ± 4 mmHg in ∼10 s, HR decreased from 385 ± 18 to 301 ± 18 beats min−1, with no change in RF. In each case there was an almost immediate cessation of ongoing nerve activity. In three of the animals, nerve activity recovered to the baseline level over a period of 13–15 min while ABP was recovering. In the other two animals, ABP and HR recovered, but nerve activity did not.

Group 2

In these animals, progressive graded hypoxia evoked a graded increase in respiration, with an increase in RF that reached a plateau during 10 and 8% O2 (Fig. 3A). This hyperventilation was accompanied by a graded increase in the frequency of augmented breaths: an additional inspiratory effort at the end of normal inspiration (see Fig. 3A and C; Marshall & Metcalfe, 1988a). Not surprisingly, the hyperventilation was associated with a fall in Inline graphic and in Inline graphic and an increase in arterial pH, as measured in the last minute of 8% O2 (Table 1). Concomitantly, ABP fell in a graded manner, reaching its lowest levels during 10 and 8% O2, while HR increased to reach its peak in 8% O2 (Fig. 3A and B).

Figure 3. Respiratory and cardiovascular responses evoked by progressive graded hypoxia.

Figure 3

Open in a new tab

A, original recordings showing responses evoked by successive 2 min periods of breathing 12, 10 and 8% O2 beginning at arrows; downward arrow indicates return to 21% O2. B, mean systemic cardiovascular and respiratory responses and changes in MSNA evoked by progressive, graded hypoxia. Values are shown as means ± SEM in final 1 min before, and in the second 1 min at each level of hypoxia and after return to 21% O2. RF, respiratory frequency; other abbreviations as in Fig. 2. ***P < 0.001, **P < 0.01, *P < 0.05, for 12, 10 or 8% O2vs. 21% O2 by post hoc analysis (n = 6). C, original recordings showing changes in MSNA and ABP associated with 2 augmented breaths (•): additional inspiratory effort at end of normal inspiration. NB, MSNA ceases before augmented breath.

Table 1.

Arterial blood gases and pH values in normoxia (21% O2), in the final minute of progressive hypoxia when breathing 8% O2 in Groups 2 and 3 and in the 9th min of a 10 min period of breathing 8% O2 in Group 4

Inline graphic (mmHg) Inline graphic (mmHg) pHa
Group 2
 21% O2 77 ± 4 30.4 ± 2.6 7.39 ± 0.01
 8% O2 29.2 ± 0.7*** 21.6 ± 1.5*** 7.52 ± 0.02***
Group 3
 21% O2 77.8 ± 3 34.9 ± 1.8 7.41 ± 0.02
 8% O2 31 ± 1.3*** 24 ± 0.8** 7.53 ± 0.03**
Group 4
 21% O2 71.1 ± 1.1 37.7 ± 1.5 7.46 ± 0.02
 8% O2 28.4 ± 2.4*** 17.9 ± 0.5*** 7.63 ± 0.03**
Open in a new tab

All values are means ± SEM.

***

P < 0.001

**

P < 0.01, normoxia vs. hypoxia.

During these responses, the recorded nerve activity ceased temporarily during each augmented breath, the onset of this preceding the augmented breath and transient fall in ABP (Fig. 3C). However, there was an overriding, progressive increase in the frequency of multiunit activity during progressive hypoxia, and in the frequency of single unit activity; both multiunit and single unit activity achieved their maximum values during 8% O2 (Fig. 3A and B, Table 2). The multiunit and single unit activity continued to show cardiac- and respiratory-related rhythmicities, although cardiac-related activity became less pronounced during 10 and 8% O2, whereas respiratory-related activity became more distinct (Fig. 4).

Considering the patterning of the single unit activity, in normoxia (breathing 21% O2), the mean firing frequency was low. Impulses therefore appeared to occur singly; couplets, with an interpulse interval of <0.1 s, seldom occurred, i.e. the instantaneous firing frequencies were low (Table 2). During progressive hypoxia when the mean firing frequency increased by up to 3-fold, couplets were still unusual, but instantaneous firing frequency increased substantially such that during 10 and 8% O2, successive action potentials occurred at up to 20 Hz (Table 2).

Group 3

In this group, both FVR and SVR decreased during graded hypoxia, FVR reaching its lowest levels at 10 and 8% O2, and SVR reaching its lowest level at 8% O2 (Fig. 5), indicating net vasodilatation in hindlimb and spinotrapezius muscle.

Figure 5. Muscle blood flow and vascular resistance changes evoked by progressive, graded hypoxia.

Figure 5

Open in a new tab

A, original recordings of with the addition of spinotrapezius and femoral blood flow (SBF and FBF, respectively) and spinotrapezius and femoral vascular resistance (SVR and FVR, respectively) computed as ABP/SBF or FBF. Other abbreviations, as in Fig. 1. Arrows indicate onset of each inspirate for 2 min each. B, mean systemic cardiovascular, respiratory and regional vascular responses evoked by progressive, graded hypoxia. Values are shown as means ± SEM in final 1 min before, and in the second 1 min at each level of hypoxia and after return to 21% O2. ***P < 0.001, **P < 0.01, *P < 0.05, for 12, 10 or 8% O2vs. 21% O2 by post hoc analysis (n = 10).

Group 4

During the extended period of hypoxia (8% O2 for 10 min), the respiratory and cardiovascular variables initially showed changes comparable to those recorded in Groups 2 and 3 when 8% O2 was breathed for 2 min in the final stage of progressive hypoxia (Fig. 6). Further, as in Group 2, single unit MSNA increased significantly in the first 2 min of breathing 8% O2.

Figure 6. Mean systemic cardiovascular, respiratory and femoral vascular responses and changes in MSNA evoked by 10 min period of breathing 8% O2.

Figure 6

Open in a new tab

Abbreviations as in Figs 1 and 5. Values are shown as means ± SEM in final 1 min before, in the 2nd and 9th min of hypoxia and in the 2nd min after return to 21% O2. ***P < 0.001, **P < 0.01, *P < 0.05, for 8% O2vs. 21% O2. †P < 0.05, 2nd min vs. 9th min (n = 4).

However, during the later part of the 10 min stimulus, RF returned towards baseline and the tachycardia waned, whereas the fall in ABP was well maintained. In individual animals, HR returned towards or fell below the baseline as we described before (e.g. Thomas et al. 1994) and so the variance was large in the ninth minute (Fig. 6). The Inline graphic achieved in the final minute of breathing 8% O2 for 10 min was similar to that reached in the second minute of breathing 8% O2 for 2 min in Group 2 (Table 1). The mean frequency of single unit MSNA seemed to follow the secondary waning of RF and HR, decreasing towards the baseline level by the ninth minute, as did the maximum instantaneous frequency, from 20 Hz to 6.6 Hz (Fig. 6; Table 2). FVR at the ninth minute was variable between animals, two showing a waning of the decrease in FVR and the other two showing a maintained decrease in FVR which may indicate competing vasodilator and vasoconstrictor influences (see Marshall & Metcalfe, 1998b and Discussion), but in all of them FVR was below baseline indicating net vasodilatation. Further, by the second minute after return to 21% O2, FVR fell to well below the baseline value in all four animals indicating further vasodilatation. By contrast, ABP, HR, RF and the frequency of MSNA returned towards or reached baseline (Fig. 6).

Discussion

In the present study, we used the focal recording technique described by Johnson & Gilbey (1994, 1996) to record nerve activity from the surface of arterial vessels in the rat spinotrapezius muscle, the preparation we used for intravital microscopy of muscle vasculature during systemic hypoxia (Mian & Marshall, 1991a,b,c;). The nerve activity had clear cardiac- and respiratory-related rhythmicities, increased when baroreceptors were unloaded during the depressor response evoked by SNP, and ceased during pressor responses evoked by PE. Thus, it fulfilled the criteria considered essential for activity recorded in the whole nerve supplying the hindlimb of the rat, cat or humans to be designated MSNA; Janig, 1988; Habler et al. 1994; Macefield et al. 1994). That the activity ceased during autonomic ganglion blockade induced by trimetaphan and returned in 3 of 5 animals when ABP and HR returned to their original levels strengthens the conclusion that it was MSNA. Failure of activity to return in two cases could be explained if the arterial vessel moved away from the electrode during the fall in perfusion pressure induced by ganglion blockade.

To record MSNA, relatively little modification was required to the spinotrapezius preparation (see Marshall, 1982). The main difficulty was creating a fluid-filled bath around the muscle. The recording sites that yielded MSNA were generally towards the sides of arterial vessels and over branching points, which suggests we generally recorded from the perivascular nerve network. Perivascular nerve fibres with varicosities run alongside and over main arteries and primary arterioles as they course into the muscle, often crossing over at branching points (Marshall, 1982). That the success rate was low (∼40–50% of preparations), compared with that achieved on the CVA (60–70%, Johnson & Gilbey, 1994), is not surprising. We were limited by the number of arterial vessels lying over, rather than under, the muscle fibres and accessible to the electrode, and by the sympathetic innervation density, which is much lower in the spinotrapezius than CVA (Marshall, 1982; Sittiracha et al. 1987; Saltzman et al. 1992).

Brock & Cunnane (1992) who used the focal recording technique on vas deferens deduced that triphasic action potentials were ones that passed through the nerve fibre below the electrode, whereas biphasic ones occurred when the fibre terminated at the recording site. Since we mainly recorded triphasic action potentials, it is likely we generally recorded from nerve fibres supplying more distal arterioles and that just a few terminated at the arterial vessel under the electrode. As sympathetic fibres release transmitters ‘en passant’ from varicosities (Brock & Cunnane, 1993), the fibres we recorded from would have released transmitters locally as well as more distally. The fact that we failed to record MSNA from venous vessels accords with evidence that venous vessels of spinotrapezius and other skeletal muscle have no sympathetic innervation (Marshall, 1982; Hainsworth 1986).

Responses evoked by systemic hypoxia

The respiratory and cardiovascular responses evoked when the level of hypoxia was made progressively more severe in 2 min stages were comparable to those seen when 15, 12, 8 or 6% O2 were given in separate 3 min periods (Marshall & Metcalfe, 1988b). Not only did FVR decrease in a graded manner indicating vasodilatation in the hindlimb (see Marshall & Metcalfe, 1988b), as others have shown in human limb muscle (e.g. Leuenberger et al. 1999), but so too did SVR. Indeed, a second innovation we introduced was to record spinotrapezius blood flow via a transonic probe on the feed artery. The decrease in SVR is consistent with the direct observation that the majority of spinotrapezius arterioles dilate during hypoxia, even though a small proportion constrict (Mian & Marshall, 1991a,b;).

MSNA in hypoxia

The hyperventilation induced by graded hypoxia was modulated by an increase in the frequency of augmented breaths (see Marshall & Metcalfe, 1988a), which are initiated by rapidly adapting, pulmonary irritant receptors and facilitated by peripheral chemoreceptors (Widdicombe, 1982). We previously deduced that these receptors trigger reflex vasodilatation in skeletal muscle whose onset often precedes the augmented breath (Marshall & Metcalfe, 1988a); dilatation was also directly observed in skeletal muscle arterioles in association with augmented breaths (Yu et al. 1990). We now show that a transient cessation of MSNA occurs before each augmented breath and so provide direct evidence of the mechanism underlying the muscle vasodilatation.

Despite these transient perturbations, the mean frequency of spinotrapezius MSNA increased in a graded manner with graded hypoxia as reported for multiunit MSNA in humans (Saito et al. 1988; Rowell et al. 1989; Somers et al. 1989b). That mean frequency increased in single units supplying spinotrapezius muscle demonstrates conclusively that hypoxia increases the activity in sympathetic fibres that have tonic activity in normoxia; an increase in multiunit unit activity cannot distinguish between this and recruitment of more fibres.

We can then consider the factors that contributed to this increase in MSNA in hypoxia. The fall in Inline graphic in 12 and 10% O2 clearly stimulated peripheral chemoreceptors enough to increase RF (see Fig. 5). That RF did not increase further in 8% O2, and that MSNA did not increase substantially until the inspirate was 8% O2, might suggest that the increase in MSNA was largely a primary response to chemoreceptor stimulation per se and that the facilitatory effect of increased CRD on MSNA (see Guynet, 2000) reached a plateau at 12–8% O2. However, we previously showed that tidal volume (VT) increased progressively when the inspirate was changed from 12 to 8 and 6% O2, indicating a graded increase in CRD (Marshall & Metcalfe, 1988b). It seems likely this also happened in the present study, for the hypoxia-induced tachycardia reached a maximum in 8% O2, and in the rat, this largely reflects the influence of CRD overcoming the primary bradycardia of chemoreceptor stimulation (Marshall & Metcalfe, 1989; Thomas & Marshall, 1994). Thus, we deduce that the increase in CRD evoked by peripheral chemoreceptor stimulation made a major contribution to the increase in MSNA in systemic hypoxia. Consistent with this, the respiratory-modulation of MSNA progressively increased as the level of hypoxia became more severe (see Fig. 4 and Guynet, 2000).

Given that MSNA in the spinotrapezius was baroreceptor-modulated, the increase in MSNA may have been partly due to baroreceptor unloading caused by the hypoxia-induced fall in ABP. However, as ABP did not fall further when the inspirate was changed from 10 to 8% O2 (see Fig. 3), the further increase in MSNA frequency in 8% O2 cannot be ascribed to greater baroreceptor unloading. Indeed, as the cardiac rhythmicity of MSNA became less defined as the level of hypoxia was increased, it seems that even in 10% O2 they were nearing their maximum excitatory influence on MSNA in systole as well as diastole.

On the other hand, hypocapnia secondary to hyperventilation can have an inhibitory effect on MSNA, mainly mediated by central chemoreceptors (Lioy & Trzebski, 1984; Millhorn 1986; see Guynet et al. 2010). Indeed, in spontaneously breathing humans, hypocapnia limited the increase in MSNA evoked by systemic hypoxia (Somers et al. 1989a). Further, hypocapnia secondary to hyperventilation can induce reflex vasodilatation in skeletal muscle (see Marshall, 1994). However, in the rat, hypocapnia secondary to hypoxia-induced hyperventilation made little contribution to the vasodilatation in hindlimb muscle during 12 or 8% O2 (Marshall & Metcalfe, 1989). Similarly, the present results suggest that any inhibitory effect of hypocapnia on MSNA is small in the rat. For the level of hypocapnia was probably greater during 8 than 12% O2 (see Marshall & Metcalfe, 1989) and yet the increase in MSNA frequency was greater in 8% O2. Thus, at most it can be argued that the increase in MSNA in 8% O2 might have been even greater had it not been for the hypocapnia. Similar arguments can be made for pulmonary stretch receptor stimulation, for any inhibitory effect they had on MSNA would have been greater in 8 than 12% O2, particularly if the increase in VT were greater in 8% O2 (see above). Certainly, reflex vasodilatation in muscle caused by pulmonary stretch receptor stimulation (Marshall, 1994) is weak in the rat and makes little contribution to muscle vasodilatation in systemic hypoxia (Marshall & Metcalfe, 1989). Clearly, in future studies, both VT and RF, and also blood gases, should be assayed at each level of hypoxia to properly test these proposals.

The responses evoked by 10 min periods of 8% O2 similarly suggest that the changes evoked in MSNA by systemic hypoxia largely reflect the changes in CRD. By the ninth minute, the increases in RF, HR and MSNA waned such that all three variables fell towards or below baseline. In our previous studies, VT, rather than RF, showed a secondary fall (see Thomas & Marshall, 1994; Thomas et al. 1994). There is no obvious explanation for this disparity for all the experiments were performed on male Wistar rats of similar age and weight, under the same anaesthetic regime. However, secondary waning of VT, or RF has been reported in rats, cats and humans, and is largely due to hypoxia acting on central respiratory neurones, rather than a secondary decline in peripheral chemoreceptor input (see Thomas & Marshall, 1994; Teppema & Dahan, 2010). Thus, the secondary fall in MSNA is most readily explained by removal of the facilitatory influence of CRD on MSNA (Guynet, 2000).

Patterning of MSNA and vascular responses

In normoxia, the mean frequency of the MSNA in single units on spinotrapezius arterial vessels was low (∼0.2–0.3 Hz), but comparable to that recorded in single fibres judged to be MSNA and supplying hindlimb in rats under anaesthesia (0.3 Hz; Habler et al. 1994) or leg muscle in conscious humans (0.47 Hz; Macefield et al. 1994). The range of firing frequencies in different fibres was substantial: 0.1–0.35 Hz in normoxia in the present study, 0.3–2.4 Hz in rats (Habler et al. (1994) and 0.09–0.69 Hz in humans (Macefield et al. 1994). The present results indicate that this irregularity continues in hypoxia, but the range of instantaneous frequencies increased from 0.04 to 10 Hz in normoxia, to 0.07 to as high as 20 Hz in 8% O2, even though mean frequency increased to only 0.68 Hz.

In arteries in vitro (e.g. Nilsson et al. 1985) and in rat hindlimb muscle and tail in vivo (Johnson et al. 2001), irregular activity was far more effective in evoking vasoconstriction than action potentials at constant frequency. The instantaneous frequencies we recorded in MSNA in normoxia and hypoxia were within the range that evoked substantial vasoconstriction in rat hindlimb muscle when applied in normoxia to the lumbar sympathetic chain (Johnson et al. 2001). Thus, the patterning of MSNA in normoxia is consistent with the significant sympathetic vasoconstrictor tone in arterioles of spinotrapezius (see Mian & Marshall, 1991a).

The higher instantaneous frequencies in systemic hypoxia might therefore have been expected to evoke greater vasoconstriction. However, the fact that gross SVR decreased, indicating vasodilatation, accords with evidence that sympathetic vasoconstriction in rat skeletal muscle is blunted by local factors in systemic hypoxia (Coney & Marshall, 2003) and that locally released adenosine and circulating adrenaline induce active muscle vasodilatation (Mian et al. 1990; Mian & Marshall, 1991b,c; Ray et al. 2002). It may be that the muscle arterioles that constrict particularly in severe hypoxia (Mian & Marshall, 1991a,b,c;) are less influenced by local factors, or their sympathetic fibres have very high instantaneous frequencies. For, sympathetic stimulation with short bursts at 20 or 40 Hz evoked vasoconstriction in skeletal muscle that was attributable to NPY and that persisted even during severe hypoxia when the noradrenaline and ATP components were absent (Coney & Marshall, 2007).

In summary, we describe a new preparation of rat spinotrapezius muscle that allows MSNA to be directly recorded from arterial vessels: multiunit and single unit activity that has baroreceptor and respiratory rhythmicity. We show for the first time in the rat that graded systemic hypoxia evokes a graded increase in multiunit and single unit MSNA that is most readily explained by the facilitatory influence of peripheral chemoreceptor stimulation and increased CRD. The mean frequency of single unit MSNA activity increased modestly, from 0.2 Hz in normoxia to only 0.68 Hz in severe hypoxia, but instantaneous frequencies increased up to 20 Hz. Moreover our novel recordings of blood flow from the feed artery showed that graded hypoxia evokes a graded decrease in spinotrapezius vascular resistance, indicating net vasodilatation. Together, these findings confirm that the vasoconstrictor influence of increased MSNA is generally blunted by dilator influences in systemic hypoxia, although high instantaneous frequencies in single fibres may explain vasoconstriction observed in individual arterioles (Mian & Marshall, 1991a,b;) and the vasoconstrictor influence of the sympathetic co-transmitter NPY even in severe hypoxia (Coney & Marshall, 2007). This preparation should pave the way for future studies on the relationship between changes in MSNA and responses in muscle vasculature.

Acknowledgments

This study was generously supported by the British Heart Foundation.

Glossary

Abbreviations

ABP

arterial blood pressure

CRD

central respiratory drive

CVA

caudal ventral artery

FBF

femoral blood flow

HR

heart rate

MSNA

muscle sympathetic nerve activity

NPY

neuropeptide Y

SBF

spinotrapezius blood flow

SVR

spinotrapezius vascular resistance

Author contributions

SH: performed all experiments with guidance of CDJ and JMM and played major role in analysis of data and preparing the draft manuscript, CDJ: made important contributions to analysis and interpretation of nerve recordings and in revising the manuscript, JMM: played a major role in drawing material together and writing material. All authors contributed to design of study and all approved the final version.

References

  1. Blumberg H, Janig W, Rieckmann C, Szulczyk P. Baroreceptor and chemoreceptor reflexes in postganglionic neurones supplying skeletal muscle and hairy skin. J Auton Nerv Syst. 1980;2:223–240. doi: 10.1016/0165-1838(80)90013-2. [DOI] [PubMed] [Google Scholar]
  2. Brock JA, Cunnane TC. Impulse conduction in sympathetic nerve terminals in the guinea-pig vas deferens and the role of the pelvic ganglia. Neuroscience. 1992;47:185–196. doi: 10.1016/0306-4522(92)90131-k. [DOI] [PubMed] [Google Scholar]
  3. Brock JA, Cunnane TC. Neurotransmitter release mechanisms at the sympathetic neuroeffector junction. Exp Physiol. 1993;78:591–614. doi: 10.1113/expphysiol.1993.sp003709. [DOI] [PubMed] [Google Scholar]
  4. Coney AM, Marshall JM. Contribution of adenosine to the depression of sympathetically evoked vasoconstriction induced by systemic hypoxia in the rat. J Physiol. 2003;549:613–623. doi: 10.1113/jphysiol.2003.042267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Coney AM, Marshall JM. Contribution of α2 adrenoceptors and NPY Y1 receptors to the blunting of sympathetic vasoconstriction induced by systemic hypoxia in the rat. J Physiol. 2007;582:1349–1359. doi: 10.1113/jphysiol.2007.132563. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Gregor M, Janig W. Effects of systemic hypoxia and hypercapnis on cutaneous muscle vasoconstrictor neurones to the cat's hindlimb. Pflugers Arch. 1977;368:71–81. doi: 10.1007/BF01063457. [DOI] [PubMed] [Google Scholar]
  7. Guynet PG. Neural structures that mediate sympathoexcitation during hypoxia. Resp Physiol. 2000;121:147–162. doi: 10.1016/s0034-5687(00)00125-0. [DOI] [PubMed] [Google Scholar]
  8. Guynet PG, Stornetto RL, Abbott SBG, Depuy SD, Fortuna MG. Central chemoreception and integrated neural mechanisms of cardiovascular and respiratory control. J Appl Physiol. 2010;108:995–1002. doi: 10.1152/japplphysiol.00712.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Habler H-J, Janig W, Krummel M, Peters OA. Reflex patterns in postganglionic neurons supplying skin and skeletal muscle of the rat hindlimb. Neurophysiol. 1994;72:2222–2236. doi: 10.1152/jn.1994.72.5.2222. [DOI] [PubMed] [Google Scholar]
  10. Hainsworth R. Vascular capacitance: its control and importance. Rev Physiol Biochem Pharmacol. 1986;105:101–173. doi: 10.1007/BFb0034498. [DOI] [PubMed] [Google Scholar]
  11. Huidoboro-Toro JP, Donoso MV. Sympathetic co-transmission: the co-ordinated action of ATP and noradrenalone and their modification by neuropeptide Y in human vascular neuroeffector junctions. Eur J Pharmacol. 2004;500:27–35. doi: 10.1016/j.ejphar.2004.07.008. [DOI] [PubMed] [Google Scholar]
  12. Janig W. Pre- and postganglionic vasoconstrictor neurones: differentiation, types and discharge properties. Annu Rev Physiol. 1988;50:525–539. doi: 10.1146/annurev.ph.50.030188.002521. [DOI] [PubMed] [Google Scholar]
  13. Johnson CD, Gilbey MP. Sympathetic activity recorded from the caudal ventral artery in vivo. J Physiol. 1994;476:437–442. doi: 10.1113/jphysiol.1994.sp020145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Johnson, Gilbey On the dominant rhythm in the discharges of sympathetic neurons innervating rat tail artery and vein. Am J Physiol Regul Integr Comp Physiol. 1996;275:R942–R941. doi: 10.1152/ajpregu.1998.275.4.R942. [DOI] [PubMed] [Google Scholar]
  15. Johnson CD, Coney AM, Marshall JM. Roles of norepinephrine and ATP in sympathetically evoked vasoconstriction in rat tail and hindlimb in vivo. Am J Physiol Heart Circ Physiol. 2001;281:H2432–H2440. doi: 10.1152/ajpheart.2001.281.6.H2432. [DOI] [PubMed] [Google Scholar]
  16. Johnson C, Hudson S, Marshall J. Responses evoked in single sympathetic nerve fibres of the rat tail artery by systemic hypoxia are dependent on core temperature. J Physiol. 2007;584:221–233. doi: 10.1113/jphysiol.2007.139360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Leuenberger UA, Gray K, Herr MD. Adenosine contributes to hypoxia-induced vasodilatation in humans. J Appl Physiol. 1999;87:2218–2224. doi: 10.1152/jappl.1999.87.6.2218. [DOI] [PubMed] [Google Scholar]
  18. Lioy F, Trzebski A. Pressor effect of CO2 in the rat: different thresholds of the central cardiovascular and respiratory responses to CO2. J Auton Nerv Syst. 1984;10:43–54. doi: 10.1016/0165-1838(84)90066-3. [DOI] [PubMed] [Google Scholar]
  19. Macefield VG, Wallin BG, Vallbo AB. The discharge of single sympathetic vasoconstrictor motoneuroes in human nerves. J Physiol. 1994;481:799–809. doi: 10.1113/jphysiol.1994.sp020482. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Marshall JM. Peripheral chemoreceptors and cardiovascular regulation. Physiol Rev. 1994;74:543–594. doi: 10.1152/physrev.1994.74.3.543. [DOI] [PubMed] [Google Scholar]
  21. Marshall JM. The influence of the sympathetic nervous system on individual vessels of the microcirculation of skeletal muscle of the rat. J Physiol. 1982;332:160–186. doi: 10.1113/jphysiol.1982.sp014408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Marshall JM, Metcalfe JD. Cardiovascular changes associated with augmented breaths in normoxia and hypoxia in the rat. J Physiol. 1988a;400:15–27. doi: 10.1113/jphysiol.1988.sp017107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Marshall JM, Metcalfe JD. Analysis of the cardiovascular changes induced in the rat by graded levels of systemic hypoxia. J Physiol. 1988b;407:385–403. doi: 10.1113/jphysiol.1988.sp017422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Marshall JM, Metcalfe JD. Influences on the cardiovascular response to graded levels of systemic hypoxia of the accompanying hypocapnia in the rat. J Physiol. 1989;410:381–384. doi: 10.1113/jphysiol.1989.sp017539. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Mian R, Marshall JM, Kumar P. Interactions between K+ and β2-adrenoreceptors in determining muscle vasodilatation induced in the rat by systemic hypoxia. Exp Physiol. 1990;75:407–410. doi: 10.1113/expphysiol.1990.sp003416. [DOI] [PubMed] [Google Scholar]
  26. Mian R, Marshall JM. Responses observed in individual arterioles and venules of rat skeletal muscle during systemic hypoxia. J Physiol. 1991a;436:485–497. doi: 10.1113/jphysiol.1991.sp018562. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Mian R, Marshall JM. The roles of catecholamines in responses evoked in arterioles and venules of rat skeletal muscle by systemic hypoxia. J Physiol. 1991b;436:499–510. doi: 10.1113/jphysiol.1991.sp018563. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Mian R, Marshall JM. The role of adenosine in dilator responses induced in arterioles and venules of rat skeletal muscle in systemic hypoxia. J Physiol. 1991c;443:499–511. doi: 10.1113/jphysiol.1991.sp018847. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Millhorn DE. Neural respiratory and cardiovascular interaction during chemoreceptor stimulation and cooling of ventral medulla in cats. J Physiol. 1986;370:217–231. doi: 10.1113/jphysiol.1986.sp015931. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Nilsson H, Ljung B, Sjoblom N, Wallin BG. The influence of the sympathetic impulse pattern on contractile responses of rat mesenteric arteries and veins. Acta Physiol Scand. 1985;123:303–309. doi: 10.1111/j.1748-1716.1985.tb07592.x. [DOI] [PubMed] [Google Scholar]
  31. Ray CJ, Abbas MR, Coney AM, Marshall JM. Interactions of adenosine, prostaglandins and nitric oxide in hypoxia-induced vasodilatation: in vivo and in vitro studies. J Physiol. 2002;544:195–209. doi: 10.1113/jphysiol.2002.023440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Rowell LB, Johnson DG, Chase PB, Comess KA, Seals DR. Hypoxemia raises sympathetic activity but not norepinephrine in resting humans. J Appl Physiol. 1989;66:1736–1743. doi: 10.1152/jappl.1989.66.4.1736. [DOI] [PubMed] [Google Scholar]
  33. Saito M, Mano T, Iwase S, Koga K, Abe H, Yamazaki Y. Responses inmuscle sympathetic activity to acute hypoxia in humans. J Appl Physiol. 1988;65:1548–1552. doi: 10.1152/jappl.1988.65.4.1548. [DOI] [PubMed] [Google Scholar]
  34. Saltzman D, Delano FA, Schmidschonbein GW. The microvasculature in skeletal muscle adrenergic innervation of arterioles in normotensive and spontaneously hypertensive rats. Microvasc Res. 1992;44:263–273. doi: 10.1016/0026-2862(92)90086-5. [DOI] [PubMed] [Google Scholar]
  35. Sittiracha T, McLachan EM, Bell C. The innervation of the caudal artery of the rat. Neuroscience. 1987;21:647–659. doi: 10.1016/0306-4522(87)90150-3. [DOI] [PubMed] [Google Scholar]
  36. Smith ML, Meunter NK. Effects of hypoxia on sympathetic neural control in humans. Resp Physiol. 2000;121:163–171. doi: 10.1016/s0034-5687(00)00141-9. [DOI] [PubMed] [Google Scholar]
  37. Somers VK, Mark AL, Zaval DC, Abboud FM. Influence of ventilation and hypocapnia on sympathetic nerves responses to hypoxia in humans. J Appl Physiol. 1989a;67:2095–2100. doi: 10.1152/jappl.1989.67.5.2095. [DOI] [PubMed] [Google Scholar]
  38. Somers VK, Mark AL, Zaval DC, Abboud FM. Contrasting effects of hypoxia and hypercapnia on ventilation and sympathetic activity in humans. J Appl Physiol. 1989b;67:2101–2106. doi: 10.1152/jappl.1989.67.5.2101. [DOI] [PubMed] [Google Scholar]
  39. Teppema L, Dahan A. The ventilatory response to hypoxia in mammals: mechanisms, measurement, and analysis. Physiol Rev. 2010;90:675–754. doi: 10.1152/physrev.00012.2009. [DOI] [PubMed] [Google Scholar]
  40. Thomas T, Marshall JM. Interdependence of respiratory and cardiovascular changes induced by systemic hypoxia in the rat: the roles of adenosine. J Physiol. 1994;480:627–636. doi: 10.1113/jphysiol.1994.sp020389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Thomas T, Elnazir B, Marshall JM. Differentiation of the peripherally mediated from the centrally mediated influences of hypoxia in the rat during systemic hypoxia. Exp Physiol. 1994;79:809–822. doi: 10.1113/expphysiol.1994.sp003809. [DOI] [PubMed] [Google Scholar]
  42. Widdicombe JG. Pulmonary and respiratory-tract receptors. J Exp Biol. 1982;100:41–57. doi: 10.1242/jeb.100.1.41. [DOI] [PubMed] [Google Scholar]
  43. Yu Y, Roberts AM, Joshua IG. Lung inflation evokes reflex vasodilation of microvessels in rat skeletal muscle. Am J Physiol Heart Circ Physiol. 1990;285:H939–H945. doi: 10.1152/ajpheart.1990.258.4.H939. [DOI] [PubMed] [Google Scholar]

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

RESOURCES