Antinociceptive effects of intrathecal 5-HT agonists in sheep

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

Antinociceptive effects of intrathecal 5-HT agonists in sheep B R Roberts BSc, PhD Zeneca Pharmaceuticals, Alderly Edge, Maccles¢eld, Cheshire SK10 4TG, UK

A Livingston BVetMed, BSc, PhD Western College of Veterinary Medicine, University of Saskatchewan, 52 Campus Drive, Saskatoon S7N 5B4, Canada

A E Waterman-Pearson BVSc, PhD, DVA, DipECVA Department of Clinical Veterinary Science, School of Veterinary Science, Langford House, Langford, Bristol BS40 5TD, UK

Correspondence: A Livingston,Western College of Veterinary Medicine, University of Saskatchewan, 52 Campus Drive, Saskatoon S7N 5B4, Canada.

Abstract Objectives To examine the role of spinal 5-hydroxytryptamine (5-HT) binding sites in nociceptive processing in conscious sheep and to study the role of 5-HT agonists in mediating analgesia. Study design Prospective controlled study. Animals Nine adult healthy female sheep (Swaledale, Swaledale-cross or Clun Forest) weighing 45^ 65 kg. Methods Intrathecal (IT) catheters were implanted at the cervical (n ˆ 5) or lumbar (n ˆ 4) level of the spinal cord under general anaesthesia. At least 1week later, and at 1week intervals thereafter, the e¡ects of intrathecal Ringer's solution (control), xylazine (100 mg), 5-HT creatinine sulphate (200, 400 and 800 mg), RU24969 (200 mg), aMethyl-5-HT and 1-(3-Chlorophenyl)-biguanide (CPBG) on the mechanical nociceptive threshold (MT) were studied. Results were plotted as mean variable versus time curves. Areas under portions of the curves (0^30 and 0^60 minutes) were measured and expressed as mean 2 standard error. Differences between values for control and drug trials were examined using the two-tailed Student's ttest. Results Baseline values of MT were lower on the hind limbs than on the forelimbs. Intrathecal

Ringer's solution did not alter MT in the cervical or lumbar region. Xylazine (100 mg) produced a characteristic elevation in MT between 5 and 60 ‡ minutes. Lumbar IT injection of 5-HT (800 mg) raised the MT more than cervical injection, while cervical injection of RU24969 (200 mg) raised the MT more than lumbar administration. Cervical IT injection of a-Me-5-HT (500 mg) produced a marked and signi¢cant increase in MT while lumbar application had no e¡ect. CPBG (500 mg) injection caused no signi¢cant e¡ect on MT with either cervical or lumbar applications. Conclusions The activation of 5-HT1 and 5-HT 2 receptors particularly at the cervical level appears to be involved in spinal nociceptive processing in the sheep. Clinical relevance These e¡ects, which lasted about 60 minutes, may have an implication in the development of new analgesic strategies for animals. Keywords 5-HT receptors, intrathecal, nociception. Introduction The technique of using a chronic intrathecal (IT) catheter to administer drugs directly onto the spinal cord of sheep has been used in a range of studies (Eisenach & Dewan 1990, Eisenach & Tong 1991; Waterman et al. 1988; Ley et al. 1989; Kyles 73

Spinal 5-HT receptors and nociception in sheep B R Roberts et al.

et al. 1991, 1992, 1995). Combined with nociceptive threshold measurement, it has been used to identify the neurotransmitters involved in the processing of nociceptive information in the sheep. Ley et al. (1989) showed that IT injections of low doses of the a2-adrenoceptor agonists xylazine (5^ 50 mg) and clonidine (3^35 mg) produced a dosedependent elevation in mechanical nociceptive threshold (MT) in sheep, which was reversed by idazoxan (100 mg kg ÿ1 IV). In contrast, even high doses of the k-and m-opioid agonists U50488H (350^ 2000 mg) and fentanyl (5^100 mg), respectively, produced no signi¢cant alteration in threshold. Morphine (500^3000 mg) produced only a slight increase in threshold. The antinociceptive e¡ect of intravenous midazolam was mimicked by high doses (1000 mg) given IT, suggesting that the elevation of mechanical threshold in sheep by midazolam is due, at least in part, to benzodiazepine receptors in the spinal cord (Kyles et al. 1995). Dopamine 2 receptors have been shown to potentiate fentanyl analgesia; following IT injections of 7.5 mg droperidol, 5 mg kg ÿ1 fentanyl IV produced a profound analgesia (Kyles et al. 1991). The method developed by Nolan et al. (1987) to measure mechanical nociceptive thresholds in conscious, unrestrained sheep has been used to study the e¡ects of various drug types on nociception and has been used in this study to measure the e¡ects of intrathecally applied 5-HT receptor agonists. In this study, however, readings were made on both the front and hind legs to compare the e¡ects of cervical and lumbar drug application. Studies of peripheral, but not spinal cord 5-HT receptors have been performed in sheep. For example, contraction of ovine umbilical veins and arteries induced by 5-HT, are mediated through action at 5-HT2 receptor subtypes, with no involvement of 5-HT1 or 5-HT 3 receptors (Zhang & Dyer 1990a, 1990b). In ovine cerebral arteries, 5-HTinduced constriction was mediated in part by 5HT2 receptors, shown by blockade of the e¡ect by ketanserin. There was also a 5-HT1 mediated constriction of these vessels, induced by sumatriptan. However, 5-HT3 receptors were not involved (Gaw et al. 1990). Webber et al. (1990) studied the 5-HTinduced contractions of tracheal smooth muscle and tracheal blood vessels. They found that 5-HT1 receptors mediated the contraction of tracheal muscle and the dilatation of blood vessels, 5-HT 2 receptors constricted blood vessels and 5-HT 3 receptors had no e¡ect.

Nine adult healthy female sheep (Swaledale, Swaledale-cross or Clun Forest) weighing 45^65 kg with no history or symptoms of painful lesions or conditions, were used in the study. Intrathecal catheterization of sheep was performed by the method described by Kyles et al. (1992). Following overnight starvation, each sheep was anaesthetised with 5% halothane in oxygen, administered via a face mask. An endotracheal tube was inserted and the sheep was maintained with 2% halothane. Anaesthetic depth was assessed using the corneal re£ex. In ¢ve cases, sheep were placed in left lateral recumbency with the head held at approximately  90 to the neck and a 5-cm midline skin incision was made from the occipital protuberance, which extended caudally. A 16G Tuohy needle (Jencons Ltd., Leighton Buzzard, UK) with the bevel pointing caudally was inserted into the atlanto-occipital junction, and advanced slowly for 1^3 cm until a back £ow of cerebrospinal £uid (CSF) was obtained. An 18G Portex catheter (Jencons Ltd., Leighton Buzzard, UK) was inserted throughtheTuohy needle into the subarachnoid space and advanced until the tip reached the cervical enlargement of the spinal cord, which is approximately level with the ¢fth cervical vertebra where nerves to the brachial plexus originate. The length of catheter required for this was determined from a lateral cervical radiograph. The needle was then withdrawn whilst holding the catheter ¢rmly in position. The catheter was ¢lled with 350 mL of the radiographic contrast medium iopamidol (Niopam 300, Merck, Rahway, NJ, USA) to assess the exact position of the catheter tip and catheter integrity. It was a predetermined requisite that the tip be located over the ¢fth cervical vertebra. The proximal end of the catheter was passed via a subcutaneous tunnel to the caudal cervical

74

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

Earlier studies using both ligand binding and autoradiographic techniques have shown that 5HT binding sites exist in the spinal cords of sheep (Roberts & Livingston 1991, 1992). The aim of this study was to establish if the observed 5-HT binding sites played a role in nociceptive processing at the spinal level, and to ascertain if 5-HT agonists might have a role in animal analgesia. A preliminary report of these studies has been made (Livingston et al. 1996). Materials and methods

Spinal 5-HT receptors and nociception in sheep B R Roberts et al.

region, where a Luer lock connecter was attached and secured by suturing a zinc oxide tape butter£y to the skin. The skin incision was closed with interrupted 2/0 mono¢lament nylon sutures and a neck bandage applied. Four other sheep were prepared with catheters positioned in the lumbar region of the spinal cord. Animals were anaesthetized, positioned in left lateral recumbency and the spine arched to allow easy insertion of the Tuohy needle between the last lumbar vertebra and the ¢rst sacral vertebra. The catheter was passed up the subarachnoid space to the lumbar enlargement of the spinal cord around the junction of the fourth and ¢fth lumbar vertebrae, where hind limb innervation occurs. The subcutaneous portion of the catheter was looped to avoid the catheter being displaced by movement at the proximal Luer lock connecter (Jencons Ltd, Leighton Buzzard, UK). The connecter was covered in this case by a padded dressing (12 12 cm) sutured to the skin. Antibiotics (penicillin/streptomycin) were given for 3 days following catheterization. Seven days after catheter insertion (both cervical and lumbar), 100 mg of the a2-adrenoceptor agonist xylazine was administered as a positive control in a volume of 100 mL of Ringer's solution through the catheter and £ushed with a further 350 mL (catheter volume) of Ringer's solution. The mechanical threshold for nociception was measured at10 minute intervals. When not in use, the catheter patency was maintained by £ushing it with Ringer's solution each week. The nociceptive tests were based on those described by Nolan et al. (1987). The area of skin covering the radius (cervical catheters) or metatarsus (lumbar catheters) was shaved and a mechanical pressure device secured to the leg with cable ties and rubber foam padding. A blunt pin in the leg device protruded when a hand lever was squeezed, so that an increasing force was applied to the sheep's leg at that point. The force on the pin was measured by a transducer in the leg device and recorded on readout at the control box, in arbitrary units. By calibrating the device, it was possible to ascertain the force applied to the leg in Newtons. To establish the mechanical nociceptive threshold, a steadily increasing force was applied to the sheep's leg until a response was observed (lifting the leg) at which point the force was removed and the threshold level noted. A cut-o¡ force of 18 N was used to prevent tissue damage during periods of analgesia,

at which point the force was removed even if no response was observed. The threshold was measured in one forelimb or one hind limb depending on whether the animal had been given an injection via the cervical or lumbar catheter. All experiments were performed in the same laboratory under identical conditions. The animals were allowed to settle for 60 minutes in the laboratory before experiments commenced. Initially, nociceptive thresholds were recorded at ¢ve minute intervals until ¢ve consistent control values were obtained. Control experiments were also done by injecting 100 mL Ringer's solution followed by 350 mL £ush of Ringer's solution. The drugs, dissolved in 100 mL of Ringer's solution, were then injected through the catheter, followed by a 350-mL £ush of Ringer's solution. The nociceptive threshold was then recorded at 5-minute intervals for 60 minutes, or at10-minute intervals if the threshold was raised very markedly, to reduce the risk of tissue damage. Along with measurements of nociceptive thresholds, behavioural changes associated with drug administration were also noted. The following drugs and doses were administered as the salt indicated: (a) xylazine HCl (Rompun, Bayer, Bury St. Edmunds, UK) 100 mg ^ as a known analgesic a2 agonist test; (b) 5-hydroxytryptamine creatinine sulphate complex, 5-HT (Sigma, St Louis, MO, USA) 200, 400 and 800 mg ^ a 5-HT agonist; (c) RU24969 (Hoechst Marion Roussel, Uxbridge, UK) 200 mg ^ a 5-HT1B/1A agonist; (d) a-Methyl-5hydroxytryptamine maleate, a-Me-5-HT (Cookson Chemicals, Southampton, UK) 500 mg, ^ a 5-HT 2 agonist); (e) 1-(3-Chlorophenyl)-biguanide hydrochloride, CPBG (Cookson Chemicals, Southampton, UK) 500 mg ^ a 5-HT 3 agonist. The time interval between administration of di¡erent drugs was 7 days. The results were displayed as mean variable versus time curves for a visual display of drug e¡ect. The areas under certain portions of these curves (0^30 and 0^60 minutes) were measured using the InPlot programme (GraphPad Software Inc., UK) and expressed as mean 2 standard error. Any signi¢cant di¡erences between Ringer's solution values and drug values were then examined using the two-tailed Student's t-test. Probability ( p) values <0.05 were considered to be signi¢cant.

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

75

Results The e¡ects of Ringer's solution on mechanical thresholds (MT) in the sheep, when administered

Spinal 5-HT receptors and nociception in sheep B R Roberts et al. 20 18 16

Threshold (n )

14 12 10 8 6 4 2 0 – 30

– 20

–10

0

10

20

30

40

50

60

Time from injection (minutes)

either in the cervical (n ˆ 5) or lumbar (n ˆ 4) regions of the spinal cord, showed that the baseline values of MT were slightly lower on the hind limbs than on the forelimbs. This was consistent across the entire 60-minute evaluation period. The IT injection of Ringer's solution did not alter the MT when given either in the cervical or lumbar region. An example of the e¡ects of IT xylazine (100 mg) is given in Fig. 1, which shows the characteristic elevation in MT between 5 and 60 ‡ minutes. The same pro¢le was seen with both cervical (n ˆ 5) and

Figure 1 The e¡ect of intrathecal injections of 100 mg xylazine, in both cervical (forelimb) (*, n ˆ 5) and lumbar (hind limb) ( & , n ˆ 4) regions, on the time course of mechanical nociceptive threshold in sheep.Values are mean 2SEM

lumbar (n ˆ 4) IT injections. This was done as a positive control experiment. Figure 2 shows that 5-HT (200, 400 and 800 mg) produced a small increase in the area under the MT curve, when applied in the cervical region (Fig. 2) which was only signi¢cant at the highest dose and only in the 0^30 minute time band. The time to onset of the e¡ect was 5 minutes, peaking at 10 and returning to control levels by 40 minutes (n ˆ 5). With a lumbar injection, 5-HT appeared to have a greater a¡ect on the MT than in the case of a cervical injection (Fig. 3). The two higher doses both pro-

14 12

Threshold (n )

10 8 6 4 2 0 – 30

– 20

–10

0

10

20

30

Time from injection (minutes)

76

40

50

60

Figure 2 The e¡ect of intrathecal injections of 5-HT, in the cervical region, on the time course of forelimb mechanical nociceptive threshold in sheep. 200 mg *, 400 mg &, 800 mg ~.Values are mean 2SEM..

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

Spinal 5-HT receptors and nociception in sheep B R Roberts et al.

14 12

Threshold (n )

10 8 6 4

Figure 3 The e¡ect of intrathecal injections of 5-HT, in the lumbar region, on the time course of hind limb mechanical nociceptive threshold in sheep. 200 mg *, 400 mg &, 800 mg ~.Values are mean 2SEM.

2 0 –30

–20

–10

0

10

20

30

40

50

60

Time from injection (minutes)

ducing a signi¢cant increase in the area under the curve in the 0^30 minute time band, and the middle dose still signi¢cant in the 0^60 minute time band. The pro¢le of the response was similar to that with a cervical injection (n ˆ 4). No obvious behavioural side-e¡ects were observed following 5-HT administration. RU24969 (200 mg) produced an increase in MT when applied onto the cervical spinal cord (Fig. 4). The increased area under the curve was signi¢cant in the time band of 0^60 minutes; the onset of the e¡ect being by ¢ve minutes and lasting to approxi-

mately 50 minutes, with the peak at about 15 minutes (n ˆ 5). A small increase in MT was also seen with lumbar application, although this was not shown to be signi¢cant (n ˆ 3). No other behavioural side-e¡ects were observed. Cervical IT injection of a-Me-5-HT (500 mg) produced a marked and signi¢cant increase in MT, in both time bands (Fig. 5). Onset and peak took place within 5 minutes and the threshold remained elevated until about 50 minutes (n ˆ 3). Lumbar application of a-Me-5-HT had no e¡ect on the MT (n ˆ 3). In one animal a-Me-5-HT (lumbar) appeared to

10

Threshold (n )

8

Figure 4 The e¡ect of intrathecal injections of 200 mg RU24969, in both cervical (forelimb) (*, n ˆ 5) and lumbar (hind limb) ( & , n ˆ 3) regions, on the time course of mechanical nociceptive threshold in sheep.Values are mean 2SEM

6

4

2

0 –30

–20

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

–10

0

10

20

30

40

50

60

Time from injection (minutes)

77

Spinal 5-HT receptors and nociception in sheep B R Roberts et al. 20 18 16

Threshold (n )

14 12 10 8 6

Figure 5 The e¡ect of intrathecal injections of 500 mg a-Me-5HT, in both cervical (forelimb) (*, n ˆ 3) and lumbar (hind limb) ( & , n ˆ 3) regions, on the time course of mechanical nociceptive threshold in sheep.Values are mean 2SEM

4 2 0 – 30

– 20

–10

0

10

20

30

40

50

60

Time from injection (minutes)

cause some degree of discomfort, inferred from arching of the back, scratching of its £ank with its hind legs, licking its lips and general restlessness. In another sheep, a transient hind limb paralysis was seen 12 hours after a lumber IT injections of aMe-5-HT. CPBG (500 mg) injection caused no signi¢cant e¡ect on MT with either cervical or lumbar application (n ˆ 4). A small increase was observed with cervical application, at 5 minutes, but was not shown to be signi¢cant. No other behavioural sidee¡ects were observed. In order to compare the data from the di¡erent agonists, the areas under the time ^ response

curves were calculated and the results are shown for the cervical intrathecal injections in Table1 and for the lumbar intrathecal injections in Table 2. Discussion The mechanical nociceptive test used in the current study discriminates between drugs producing sedation and those producing analgesia. Nolan et al. (1987) showed that a sedative dose of acepromazine did not alter MT values recorded in the test, indicating that the response is produced by a spinal re£ex with little or no integration within higher centres. Therefore, it is unlikely that the apparent antinoci-

Table 1 Areas under the curves from the mechanical threshold/time graphs for cervical intrathecal injections in sheep Drug

Dose (mg)

Dose (nmol)

n

AUC (0±30 min)

p

AUC (0±60 min)

p

Ringer's solution Xylazine 5-HT

(450ml) 100 200 400 800 200 500 500

(450ml) 4546 516 1033 2065 696 1632 2015

5 5 3 5 5 5 3 4

150.9 ‹ 10.6 355.7 ‹ 49.1 151.9 ‹ 16.8 207.9 ‹ 28.4 229.8 ‹ 31.2 226.6 ‹ 31.2 377.3 ‹ 51.9 157.7 ‹ 25.6

± < 0.01 ± ± < 0.05 ± < 0.001 ±

299.0 ‹ 21.0 738.0 ‹ 101.8 ± 393.5 ‹ 53.8 393.3 ‹ 53.5 474.9 ‹ 65.4 653.1 ‹ 90.0 321.0 ‹ 52.1

± < 0.01 ± ± ± < 0.05 < 0.01 ±

RU24969 a-Me-5-HT CPBG

The p column shows the p-value, where a signi®cant dif ference from control (Ringer's solution) was observed using the Student's two-tailed t-test. AUC refers to area under the curve. Values are mean ‹ SEM

78

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

Spinal 5-HT receptors and nociception in sheep B R Roberts et al. Table 2 Areas under the curves from the mechanical threshold/time graphs for lumbar intrathecal injections in sheep Drug

Dose (mg)

Dose (nmol)

n

AUC (0±30min)

p

AUC (0±60min)

p

Ringer's solution Xylazine 5-HT

(450ml) 100 200 400 800 200 500 500

(450ml) 4546 516 1033 2065 696 1632 2015

4 4 3 4 3 3 3 4

124.9 ‹ 19.1 415.4 ‹ 34.1 250.7 ‹ 63.6 316.5 ‹ 42.8 253.7 ‹ 51.5 143.4 ‹ 20.8 181.4 ‹ 33.7 120.9 ‹ 21.9

± < 0.001 ± ± < 0.01 < 0.05 ± ±

244.5 ‹ 37.5 738.0 ‹ 77.2 416.7 ‹ 105.7 536.2 ‹ 72.5 445.3 ‹ 90.5

± < 0.001 ± < 0.05 ± ± ± ±

RU24969 a-Me-5-HT CPBG

336.4 ‹ 62.4 251.7 ‹ 45.7

The p column shows the p-value, where a signi®cant difference from control (Ringer's solution) was observed using the Student's two-tailed t-test. AUC refers to area under the curve. Values are mean ‹ SEM

ceptive actions of some of the compounds used in this study were due to a sedative action. Nolan et al. (1987) also found that the test produced consistent results with consecutive readings made over long periods of time, suggesting that minimal peripheral tissue damage, in£ammation and central sensitization are induced by the testing procedure. If present, these phenomena would cause an apparent hyperalgesia seen by decreasing nociceptive thresholds. The xylazine pro¢le observed in this study is consistent in sheep and has been studied extensively (Ley et al. 1989). The pH of the 10 mg mL ÿ1 5-HT solution was 4.4. To check that pH had no e¡ect on nociception, a solution of citric acid (pH 4.4) was tested for e¡ect in the mechanical test. It produced no change in the MT. None of the other solutions had a pH lower than that of 5-HT. Negative control experiments using 100 mL IT Ringer's solution showed no e¡ects, whilst the positive control experiments using 100 mg xylazine produced a profound analgesia without supraspinal e¡ects such as sedation, when the catheter was intact and correctly positioned (Waterman et al. 1988). Autoradiographic studies using [3 H] clonidine have shown that injection of a similar dose in these volumes produces a spread of drug localized to approximately ¢ve vertebral segments around the catheter tip (Waterman et al. 1988). Intrathecal injections of 5-HT produced a signi¢cant elevation of the MT and seemed to be more e¡ective in the lumbar region than the cervical region. Both RU24969 and a-Me-5-HT elevated the MT signi¢cantly when given cervically, but had no

signi¢cant e¡ect in the lumbar region. Activation of 5-HT 3 receptors by CPBG appeared to have no e¡ect on MT in either region. These results suggest a possible di¡erence in the involvement of 5-HT receptor subtypes in the processing of nociceptive information, between the cervical and lumbar levels of the ovine spinal cord. Activation of 5-HT1B/1A and 5HT 2 receptors only appeared to induce antinociceptive activity in the cervical region. It is possible that these observations result from physical di¡erences between spinal levels which altered the access of the compounds to receptors, since we have found the distribution of dorsal horn 5-HT1 and 5-HT 2 binding sites to be relatively consistent between cervical and lumbar level (Roberts & Livingston 1992). KjÖrsvik et al. (1995) have reported di¡erences between lumbosacral and upper thoracic e¡ect of IT DOI (5-HT2 agonist) on nociception in rats. However, they found that DOI was pronociceptive in their tests. They proposed regional di¡erences in the 5-HT2A /5-HT2C ratio in the cord as a possible explanation for the regional di¡erences in agonist e¡ect.We did not di¡erentiate 5-HT 2A to 5-HT 2C receptors in our study, so this could also be true in the sheep. The apparent di¡erences in potency between 5-HT and the more speci¢c agonists between the cervical and lumbar applications re£ects the potential complexity of the situation. It may be that the interaction of multiple receptors to 5-HT at one time contributes to these observations. Since 5-HT-induced constriction of ovine cerebral arteries is mediated in part by 5-HT 2 and 5HT1 receptors (Gaw et al. 1990) it is possible that

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

79

Spinal 5-HT receptors and nociception in sheep B R Roberts et al.

activation of these receptors following IT injections of agonists could reduce blood £ow in the spinal region. Any resulting hypoxaemia for spinal or primary a¡erent neurones could feasibly alter nociceptive transmission. Spinal blood £ow was not investigated in this study. We, and other workers, have found that IT administration of RU24969 in rats actually facilitates responses to noxious thermal stimuli to the tail (Solomon & Gebhart 1988) but does not alter mechanical nociceptive thresholds (Roberts & Livingston, unpublished). However, in mice, the same compound inhibited responses to noxious colorectal distension (Danzebrink & Gebhart 1991) reduced behavioural responses to IT substance-P (Eide 1992) and increased the tail £ick latency (Eide et al. 1990). Therefore, it would appear that the e¡ects of RU24969 on spinal nociceptive processing are dependent upon both the species and the nociceptive test employed. The results with IT a-Me-5-HT and CPBG in the sheep agree well with our ¢ndings in the rat, that is, a-Me-5-HT elevated mechanical thresholds, whereas CPBG had no e¡ect (Roberts & Livingston, unpublished). Occasional behavioural e¡ects with a-Me-5-HT resulted in the limitation of these observations to three animals. In conclusion, this study shows that 5-HT receptor subtypes in the sheep spinal cord are capable of inhibiting the re£ex response to a noxious mechanical stimulus, as has been shown in rats (Danzebrink & Gebhart 1991). However, the analgesic potency of the 5-HT agonists tested appears to be much less than that of the a2-adrenoceptor agonist xylazine and, at this time, their role as sole analgesic agents would not appear to be of any signi¢cance, but their use in combination with other analgesic agents remains to be investigated. Although the agents tested here were applied intrathecally, in order to evaluate the signi¢cance of their action on spinal neurones, their administration by more traditional peripheral routes should be expected to produce a similar analgesic response.

with the surgery and Mr Simon Leader for the dayto-day care of the animals. References

This study was supported by the Medical Research Council and the Wellcome Trust. The authors wish to thank Dr Andrew Kyles for considerable help

Danzebrink RM, Gebhart GF (1991) Evidence that spinal 5-HT1, 5-HT2, and 5-HT3 receptor subtypes modulate responses to noxious colorectal distension in the rat. Brain Res 538, 64^75. Eide PK (1992) Stimulation of 5-HT1 receptors in the spinal cord changes substance P-induced behaviour. Neuropharmacology 31, 541^545. Eide PK, Joly NM, Hole K (1990) The role of spinal cord 5HT1A and 5-HT1B receptors in the modulation of a spinal re£ex. Brain Res 536, 195^200. Eisenach JC, Dewan DM (1990) Intrathecal clonidine in obstetrics: sheep studies. Anesthesiology 72, 663^668. Eisenach JC, Tong C (1991) Site of hemodynamic e¡ects of intrathecal alpha2 adrenergic agonists. Anesthesiology 74, 766^771. Gaw AJ, Wadsworth RM, Humphrey PPA (1990) Pharmacological characterization of post-junctional receptors in cerebral arteries from the sheep. Eur J Pharmacol 179, 35^44. KjÖrsvik A, StÖrkson R,TjÖlsen A, et al. (1995) Di¡erential e¡ects of activation of lumbar and thoracic 5-HT2A/2C receptors on nociception in rats. Pharmacol Biochem Behav 56, 523^527. Kyles AE, Waterman AE, Livingston A (1991) The e¡ects of low dose droperidol on fentanyl analgesia in the sheep. ActaVet Scand 87 (Suppl.), 176^178. Kyles AE, Waterman AE, Livingston A (1992) Chronic intrathecal catheterization in the sheep. J Pharmacol Toxicol Methods 27, 177^183. Kyles AE,Waterman AE, Livingston A (1995) Antinociceptive activity of midazolam in sheep. J Vet Pharmacol Ther18, 54^60. Ley S, Dash A, Waterman AE, et al. (1989) A comparison of the analgesic e¡ects of intrathecal a2-adrenoceptor agonists and opioids in conscious unrestrained sheep. Adv Biosciences 75, 495^498. Livingston A, Roberts BR, Waterman AE (1996) Antinociceptive e¡ects of intrathecal serotonin agonists in sheep.Vet Surg 25, 183. Nolan A, Livingston A, Morris R, et al. (1987) Techniques for the comparison of thermal and mechanical nociceptive stimuli in the sheep. J Pharmacol Methods 17, 39^ 49. Roberts BR, Livingston A (1991) A comparison of 5-HT binding in the spinal cords of rats and sheep. Br J Pharmacol104, 368P. Roberts BR, Livingston A (1992) Autoradiographic localization of 5-HT binding in the sheep spinal cord. Neuroscience Lett 42, 516. Solomon RE, Gebhart GF (1988) Mechanisms of e¡ects of intrathecal serotonin on nociception and blood pressure in rats. J Pharmacol Exp Ther 245, 906^912.

80

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

Acknowledgements

Spinal 5-HT receptors and nociception in sheep B R Roberts et al. Waterman AE, Livingston A, Bouchenafa O (1988) Analgesic e¡ects of intrathecally applied a2 adrenoceptor agonists in conscious unrestrained sheep. Neuropharmacology 27, 213^216. Webber SE, Salonen RO, Widdicombe JG (1990) Receptors mediating the e¡ects of 5-hydroxytryptamine on the tracheal vasculature and smooth muscle of the sheep. Br J Pharmacol 99, 21^26.

Zhang L, Dyer DC (1990a) Receptor mechanisms for 5hydroxytryptamine in isolated umbilical vein. Eur J Pharmacol184, 281^293. Zhang L, Dyer DC (1990b) Characterization of serotonergic receptors mediating contraction of ovine umbilical artery. J Pharmacol Exp Ther 255, 233^239.

VeterinaryAnaesthesia and Analgesia, 2000, 27, 73^81

81

Received 26 January 2000; Accepted 2 February 2000.