83

Brain Research, 584 (1992) 83-91 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00

BRES 17888

Analgesic effects of intraventricular and intrathecal injection of morphine and ketocyclazocine in the infant rat Gordon A. Barr a,b, Dorene Y. Miya a and William Paredes c Biopsychology Doctoral Program, City University of New York, Department of Psychology, Hunter College, City University of New York, New York, NY 10021 (USA), b Department of Developmental Psychobiology, New York State Psychiatric Institute, Columbia University College of Physicians and Surgeons, New York, NY 10032 (USA) and c Department of Psychiatry, Albert Einstein Collegeof Medicine, Room G49, Bronx, NY 10469 (USA) a

(Accepted 11 February 1992)

Key words: Development; Ontogeny; Analgesia; Antinociception; Noxious stimuli; Pain; Morphine; Ketocyclazocine; Opiate

Little is known of the neural bases of analgesia in immature animals. This experiment examined the effects of intracerebroventricular (i.c.v.) and intrathecal (i.t.) administration of morphine or ketocyclazocine in tests of antinociception in rats aged 3 to 14 days of age. Analgesia tests were conducted using both thermal and mechanical (pressure) noxious stimuli applied to the forepaw, hindpaw or tail. In the 3-day-old morphine-injected i.c.v, produced analgesia in the forepaws when either the mechanical or thermal noxious stimulus was used. There was no effect when the hindpaw or tail was tested. At 10 days of age, when the mechanical stimulus was used, morphine was analgesic in tests on all three appendages but was only effective in the forepaw when the thermal stimulus was used. Morphine was fully effective in all tests with both stimuli at 14 days of age. Ketocyclazocine had no consistent effect when given i.c.v. When injected i.t., morphine produced analgesia in the forepaws in the thermal test at 4 days of age and in all appendages by 10 days. When the mechanical test was used, morphine was effective in all appendages at all ages tested. Ketocyclazocine was analgesic at all appendages for the mechanical stimulus at all ages but was only transiently effective in the thermal test. The results demonstrate differential development of analgesia mediated at different levels of the neural axis and are consistent with the development of descending inhibitory that may mediate analgesia induced by i.c.v, injections of morphine. Neural mechanisms that are involved in the analgesic effects of these drugs against the two types of stimuli are also developmentally distinct.

INTRODUCTION

Opiate drugs produce analgesia through distinct classes of opioid receptors at different levels of the neuroaxis. How the neural mechanisms that underlie analgesia mediated by specific opioid receptors are organized is not clear. It appears that opiates act predominantly through /z and ~ opioid receptors at midbrain and hindbrain sites and the spinal cord 44'45'60-62, and that K opioid agonists act primarily through receptors within the spinal cord and may not be fully effective when administered to the b r a i n 19'21'27'40'43'47'48'51'60 (but see refs. 35,56). The actions of /~ and 8 opioid receptors in brain can be either through the activation of spinopetal inhibitory systems that block the rostral transmission of pain signals or by alteration of the processing of nociceptive signals in the brain.

Although the pharmaco-dynamics of opioid-induced antinociception have been intensely investigated in the adult, fewer studies have addressed the maturation of pain modulatory systems in the developing animal. Since the first reports demonstrating that opiates induced analgesia in infants a°, many studies have demonstrated an antinociceptive effect in immature animals (for reviews see refs. 6,39), including human infants 64. Studies that have examined analgesia at early ages have found that rat and mouse pups become more responsive to opiate-induced analgesia from birth to the end of the preweaning p e r i o d 3,5,9,17,42,5°,65. Although opiates produce analgesia in the young infant, the organization of the neural circuitry that mediates antinociception in infant animals is an important yet unstudied question. It is not known at what level of the neural axis analgesic drugs have their antinociceptive effects. At least two groups have

Correspondence: G.A. Barr, Department of Psychology, Room 611N, Hunter College, 695 Park Avenue, New York, NY 10021, USA.

84 demonstrated analgesia after intracerebroventricular (i.c.v.) injection of opiates in infant rats. Kehoe and Blass 29 and Pasternak and co-workers 42, reported that morphine, fl-endorphin and D-alaZ-metS-enkepha linamide injected directly to the lateral ventricle or cerebral cistern produced analgesia using tests of thermal or mechanical pain. Both/.L and r opioid receptors exist in the spinal cord of the early postnatal rat pup 2'4 and thus it would be expected that tx and K opiates would produce analgesia when administered intrathecally. U69593, a K opiate, produced dose-dependent antinociception to mechanical pressure 2. The effects of U69593 were naloxone-reversible. In adult animals, the arylacetamide ligands such as U69593 do not produce analgesia due to the apparent absence of those binding sites (termed K1) in spinal cord. In contrast the spinal cord of the infant has a significant concentration of K] receptors which are functional both in vivo and in vitro 2. No studies have directly injected other opiates to the spinal cord. The first goal of these studies was to determine whether opiates administered to the brain or spinal cord would produce analgesia in the developing animal. Opioid receptor ontogenesis, spinopetal monoamine projections, and dorsal horn spinal circuitry mature and reorganize largely postnatally. For example, /x opioid receptors in brain are present at birth albeit in low numbers. They develop linearly during the first 2-3 weeks of life peaking around day 18-219,49,52,59 • In contrast 6 opioid receptors are not detectable using specific ligands until the middle of the second w e e k 34'36'38'39'49'52. Likewise, analgesia following systemic administration of different opiates occurs at different ages. Peripheral administration of ketocTclazocine produces analgesia at an early age; the onset of morphine analgesia occurs later. Although these two drugs are not fully selective for/x and K opioid receptors, the different patterns of analgesia, the lack of cross-tolerance, the inability of a ~5 opioid receptor antagonist to block their effects, and the different potencies of naloxone to antagonize each effect suggest that the different developmental patterns of analgesia are due to the differential development of /x and K opioid receptor types 9']7. A second goal of the present studies was to determine if morphine or ketocyclazocine have different effects when injected into the brain or spinal cord of the infant rat. In studies using peripheral administration of morphine, analgesia first occurred when the noxious stimulus was applied to the forepaws, appearing at later ages in the hindpaws and tail 9']7. Depending on the parameters of the noxious stimulus, the onset of analgesia can be quite abrupt 9'5°. Morphine produced antinocicep-

tion in the forepaws at 3 days of age but did not affect hindpaw or tail nocisponsive sites until later, tz Opioid receptors, perhaps located in the periaqueductal gray among other sites, may act via descending spinal monoaminergic system to demonstrate the general rostral to caudal maturation. K Opioid receptors may act predominantly at the local spinal level 6 and ketocyclazocine's effects appear more uniformly in development. This developmental model is consistent with theories of others H but has not been investigated specifically in the infant. A third goal of these studies was to study whether the rostral to caudal maturation of analgesia that occurs after peripheral administration of drug also occurs following i.c.v, or intrathecal administration of opiates. The vast majority of studies have used noxious thermal stimuli; a number of authors have suggested that different opioid systems may inhibit specific types of pain 1'22"47'48'55'57'58. We have reported similar findings in the infant rat 17'18'24. Millan 4° in addition reported that /x and K opiates differ in their ability to block responses to noxious stimuli of different intensities. Ix and K Opiates blocked responding to moderate intensity thermal and mechanical (pressure) stimuli. K Opiate drugs were sensitive to the intensity of the thermal stimulus whereas /x drugs were not. Neither class of opiate agonists were sensitive to the intensity of the mechanical stimulus, and neither drug blocked vocalization to a painful electrical stimulus applied to the tail. It is not known whether that specificity (either stimulus type or intensity) is due to the loci of the relevant receptor (e.g. brain or spinal cord) or to characteristics of the different receptors at the same locus. The final goal of this report is to examine the specificity of morphine and ketocyclazocine in the brain and in the spinal cord for these two types of noxious stimuli. Parts of this work have been presented previously 6,8,10. MATERIALS AND METHODS Subjects Offspring of L o n g - E v a n s hooded rats (Rattus norvegicus) bred and reared in the H u n t e r College and the Albert Einstein colonies were used as subjects. Pregnant rats were checked for newborn pups twice daily. Pups found during these checks were considered day 0. D a m s and their litters were housed in 40 × 20 × 24 cm plastic terraria in a temperature- and humidity-controlled room. The l i g h t / d a r k regimen was 12:12 hours. Food (Purina pellets 5001) and water were constantly available. Three days following birth, litters were culled to 12 pups and otherwise left undisturbed except for routine cleaning until time of surgery. Both sexes were tested. Litters were treated as experimental units 14 except where noted.

lntraventricular injection procedure Subjects aged 3, 7, 10, 14 and 21 days were removed from their dams and placed in a warm incubator (33-35°C). T h e procedure for

85 i.c.v, injection was adapted from Ellis et al? 6. Briefly, drug was loaded into a 1-~1 syringe with a 30-gauge needle that was fitted with a guard. The length that the needle extended beyond the guard corresponded to the depth of the lateral ventricle at each age. The pup was hand-held and gently restrained. The skull was punctured perpendicular to the dorsal surface and 1 /~1 of drug, mixed with a drop of Cresyl violet (0.05%), was injected over a 1-min period. The pups were replaced into the incubator until testing. Following testing, the pups were overdosed and the brain inspected for even and complete spread of the dye through the ventricular system. Data from any pup with inadequate spread of marker was discarded. We have done the required control experiments to determine that the dye by itself or in interaction with the drugs had no effect on the measured behaviors. In the young pups ( < 14 days of age) bregma was visible through the skin, and was used as a landmark, and the hit rate was greater than 95%. In the 14-day-olds, the landmarks were less visible. For these older animals the hit rate was lower ( ~ 75%).

Intrathecal injection procedure Subjects aged 3, 9 and 13 days were implanted with intrathecal catheters. The details of the construction of the dialysis catheters and the implanting, injection and histology can be found in previous work 7'24'41. The time elapsed between injection and testing was determined for each drug by consulting previous intrathecal studies and by performing a time course experiment whereby young rats were administered a single dose of drug and tested in the manner described below at regular intervals after injection. The time during which a maximum increase in the analgesic response occurred represented the post-injection period for that drug. This procedure was used to determine the post-injection period for all drugs used in the current study. Moreover, during surgery, the rupture of small blood vessels surrounding the spinal cord is unavoidable. Because, in the young, the blood-brain barrier is immature 46, it is possible that drug could spread to supraspinal or non-neural sites through a systemic route to produce analgesia. To address this possibility, pups were implanted with cannulae in the jugular vein as describedz° with modifications for preweaning animals. Selected doses of drugs at selected ages were given i.v. in the same volume and dose (highest only) as the intrathecal study to test whether this route of administration produced analgesia that is as potent as that following intraspinal injection. In prior studies 24 and with a limited number of subjects given morphine, we have failed to note any analgesia following i.v. administration at doses and volumes that produce robust antinociception effects intrathecally. Therefore we have some confidence that the results from our intrathecal studies are not due to systemic diffusion of the drug.

withdrawal latency of the forepaw, hindpaw and tail of each animal from noxious thermal and mechanical stimuli was measured to 0.01 s and recorded. Testing was done both prior to, and following the administration of drugs. The order of presentation of stimuli to each limb was randomly determined for each animal. The order of mechanical and thermal stimulus presentation was also randomly determined for each pup. Thermal and mechanical tests were done on limbs on the same side of each animal. Each pup was tested only one time before and one time following drug administration. A 10-s cut-off latency was used for both tests after which the stimulus was discontinued even if there had been no response. The intensity of stimuli were chosen to be the lowest intensity that produced reliable responding at all three ages. Tissue damage does not occur with these intensities. In prior work the latency to withdraw each appendage was approximately equal at all ages and thus we have considered them to be roughly equivalent intensity stimuli although they differ on many other parameters (e.g. number of dermatomes stimulated). In the intrathecal experiment, for reasons that are not clear, the baseline latencies for the mechanical stimuli were longer than for the thermal stimuli. Preliminary experiments showed that, for each stimulus type, the baseline latencies and the analgesic response to peripherally administered opiates were not altered if either stimulus was presented first or second. Limbs ipsilateral to the catheter injection site occasionally demonstrated some loss of motor function following surgery and therefore appendages contralateral to the side of the cord into which the catheter was inserted were tested. For all studies, breaths per minute and righting response latencies were recorded for each animal just prior to baseline testing and prior to experimental testing.

Spinal vs. supraspinal responding As has been pointed out by Jensen and Yaksh 26, the neural organization of the withdrawal response to noxious stimulation may be at spinal sites, supraspinal sites or both. For example, they argue on the basis of spinal cord section and intrathecal drug administration studies that the tail-flick response is a strict spinal reflex but withdrawal of the forepaw from a hotplate requires supraspinal processing. To determine if the forepaw, hindpaw and tail responses in the infant are organized likewise we modified 'cold block' procedures to functionally but reversibly lesion the spinal cord at various levels to determine the fate of the withdrawal response. A thin length of metal tubing was shaped into a shallow inverted U-shape such that when placed perpendicular to the dorsal surface of the lower thoracic vertebral column of a rat pup, it lay in contact with the vertebrae. Liquid isopentane was then perfused through the metal tubing cooling the cord. All voluntary motor movement ceased distal to the cooling, and the tail or hindpaws were tested with the mechanical or thermal noxious stimuli as described.

Behavioral tests All tests were conducted blind with respect to dose of the drug including vehicle. For all experiments, subjects were exposed to both thermally and mechanically noxious stimuli presented in a random order. Testing methods were described previously 24. We provide only an abbreviated description here. For the mechanical test, a 64-gram weight with a 0.2 cm diameter flat surface was gently applied in turn to the dorsal surface of the forepaw, hindpaw, and tail. Contact was made between the 0.2 cm surface and each limb before allowing the weight to be released in order to assure that withdrawal responses were not made to tactile non-noxious stimuli. Three-day-old pups were not always able to remove fully their limbs or tail from the weight. Therefore, at this age withdrawal iatencies were determined to be the first attempt to remove the limb or tail from the weight even if this attempt was not successful. For the thermal test, pups were suspended above a water bath maintained at 47°C. The distal 2/3 of the forepaw, hindpaw, and tail were sequentially submerged in the bath. The latency for removal of the limb or tail from the weight or the bath was recorded using a timer (Lafayette instruments) operated by a foot pedal. The timer was activated when the weight was placed upon the limb or tail or when the limb or tail hit the water surface and was terminated when the animal removed the limb or tail from below the weight or from the water bath. The

Statistics Drug, dose, limb, stimulus and age effects were compared by means of a factorial analysis of variance for all experiments. In the i.c.v, study, for the 3- and 10-day-olds, repeated measures were done for limb and dose as all three limbs tested were from the same animal and all four doses were administered to the same litter. For the 14-day-old pups, the greater miss rate precluded a repeated measures design for the dose effect because typically we missed in at least one pup in the litter. This age was analyzed separately using a between subjects analysis for the dose effect. Posthoc tests for simple main effects were used when applicable to determine the significance of response differences in control animals exposed to a thermal vs. mechanical stimulus. For the intrathecal data, three analyses were performed. First, an overall factorial analysis of variance using limb, dose and stimulus type as within subjects variables and drug and age as between subjects variables was performed. In this analysis each animal was tested with both the mechanical and thermal stimuli but only at 4 and 14 days of age. Because there were substantially more pups tested with the thermal stimulus and because tests were conducted at 10 days of age analyses were performed separately for the thermal and the mechanical stimuli.

86 Respiration rates and latencies to right were analyzed by factorial analyses of variance. Separate analyses were performed for morphine and ketocyclazocinefor each measurement. Follow-up analyses were standard posthoc tests and correlations of each with the analgesia data. RESULTS

Cold block The tail-flick response remained intact and latencies did not differ from control animals treated similarly but not cooled (data not shown). This suggests that in the infant, as in the adult, the tail-flick response is a spinal reflex. In contrast, the hindlimb response was eliminated by spinal cooling. Even at high intensity stimulation, there was no limb withdrawal. Hindlimb withdrawal returned following warming of the spinal column and cord. This suggests that the hindlimb response requires supraspinal input. We were unable to test the forelimb response because cooling at levels sufficiently rostral to eliminate forepaw movement invariably disrupted respiration and thus the data for the forepaw using this method was compromised. On the basis of the hindlimb data and on studies of the forepaw withdrawal response in adults, it is likely that this response is also mediated by supraspinal sites.

l.c. v. administration

compared to mechanical stimulus. The morphine data suggest that the descending system that mediates the analgesic actions of morphine has matured to innervate the spinal cord segments that receive noxious sensory input from the forepaws only. At 10 days of age, morphine produced analgesia in all three appendages when the mechanical stimulus was applied but was effective only in the forepaw test using the thermal stimulus. At 14 days of age morphine produced analgesia to both stimuli in all appendages. These data argue for at least partially separate descending systems mediating thermal and mechanical pain. These descending systems apparently develop differentially. Ketocyclazocine produced a less potent and non-dose-dependent effect. Ketocyclazocine had no effect on any appendage for either test at 10 or 14 days of age. (The ketocyclazocine data are not shown.)

Respiration rate and the righting reflex Morphine reduced the respiration rate and lengthened the righting response. The effects on both were greater in the 3-day-old pups than the 10-day-old pups. Ketocyclazocine increased the respiration and righting responses only slightly and much less than did morphine. KC had no effect on either response at ten days of age. The data are presented in Table I.

Intraspinal administration

Analgesia As can be seen in Fig. 1, i.c.v, morphine produced profound analgesia in the forepaws of 3-day-old pups but only minimal effects in the hindpaws and tail. The effect was slightly greater for the thermal stimulus 3-DIly Old

Morphine Thermal analgesia. The analysis of variance showed that morphine produced analgesia at four days of age, lO-Day Old

14-Dlly Ola

°L..J,,

ffl 'U

t, e.

s

,,

Dose (pg), log scale Fig. 1. This figure illustrates the effects of lateral ventricle injections of morphine on the withdrawal latencies following thermal (filled circles) and mechanical (open squares) noxious stimulation. At 3 days of age, morphine was effective in the forepaw only, with some greater potency when the noxious thermal stimulus was used. In 10-day-old pups morphine remained more effective in the forepaw immersion test but was effective in the hindpaw and tail tests only when the mechanical stimulus was used. The opiate was effective in all body parts with both stimuli only at 14 days of age.

87 4-oay Otd

10-~y Old

°L, X °L,°° ,oL,/. ,oLo

I I"1 Mechmn~l ~ ]

w

,

"0 tO

14-OayOld

.

o

.E

,

~

.~

Veh. "~

10

30

>, o c .J

., •

10

30

• ~'

I 10

I 30

Dose (pg), log scale Fig. 2. This figure shows the effects of intrathecal administration of morphine on withdrawal latency from the thermal (filled circles) or mechanical (open squares) noxious stimuli. Morphine was marginally effective at 4 days and very potent at 10 days of age when tested with the thermal stimulus but at 14 days, it was ineffective. Morphine produced robust analgesia when the mechanical stimulus was used.

the earliest tested. The effect was modest and present only in the forepaw in the 4-day-old and in all three appendages in the 10-day-old, where it was most potent in the tail-flick test. At 15 days of age, morphine was totally without effect (Fig. 2). Mechanical analgesia. Morphine was effective at both ages and in all three appendages when the noxious stimulus was mechanical. At the younger age, morphine was particularly effective in the hindpaw test; by 14 days of age it was maximally effective in the forepaw and hindpaw tests and near-maximally effective in the tail-flick tests (Fig. 2). Respiration rate and the righting reflex. Morphine reduced the breathing rate and lengthened the rightingresponse latencies. The effect was greater in the older animals. These data are shown in Table I. For both responses, both doses of morphine were significantly greater than the vehicle. Correlations between the effects of morphine on righting and on respiration with its effects on analgesia were largely non-significant. Using one-tailed tests of significance (a = 0.05; predicted negative correlations for respiration and positive correlations for righting) no consistent pattern of correlations was seen. Further, although the effects of morphine were greater on respiration and righting as the animal matured, this relationship did not hold for analgesia.

Ketocyclazocine Thermal analgesia. In the 4-day-old, KC was effective when the stimulus was applied to the forepaw and

hindpaw but not for the spinally mediated tail-flick. However, by 10 and 14 days of age, KC was virtually without analgesic effect in any appendage (Fig. 3). Mechanical analgesia. The results using the test of mechanically induced pain showed a different pattern of results. Both doses of KC were more effective in the 4-day-olds than the 14-day-olds. Unlike the thermal test, however, there was no main effect or interaction effect of the appendage tested. This suggests that KC's antinociceptive action occurs more uniformly at the different levels of the spinal cord than does the effect of morphine (Fig. 3). Respiration rate and the righting reflex. Ketocyclazocine reduced the breathing rate and lengthened the righting-response latencies and the effect was constant across ages. Those data are shown in Table I. For the righting response, both the low and higher doses were significantly different than the vehicle but did not differ from each other. For the respiration rate data, only the high dose was significantly different from the vehicle. Correlations between the effects of ketocyclazocine on righting and on respiration with its effects on analgesia were also largely non-significant. As with morphine no consistent pattern of correlations was seen. DISCUSSION The results of these experiments demonstrate that in the infant as well as the adult, the antinociceptive effects of opiates are dependent on the site at which

88 4-Day Old 10

o

10-Day Old

Mechanlcal

14-Day Old

lo

lO

g "0

\

tO o G) (n

._c

I

\

gs

i

~

~.

s

>,

¢J t,--

m

.J

•

10

X

30

.

X

10

30

Veh.

\

10

30

Dose (pg), log scale Fig. 3. This figure demonstrates the effects of intrathecal administration of ketocyclazocine in tests with the mechanical and thermal noxious stimuli. KC was somewhat effective at 4 days of age with the thermal stimulus but mostly effective against the mechanical stimulus. The legend is as for Fig. 2.

TABLE I

Respiration rates and righting latency for i.c.v, and intrathecal opiates Lc.v. Saline Morphine 3-day-old Respiration Righting reflex 10-day-old Respiration Righting reflex KC 3-day-old Respiration Righting reflex 10-day-old Respiration Righting reflex

1.0

3.0

10.0

62.45- 4.9 8.25:4.0

67.05-5.0 14.2+5.0

45.05-8.0 20.15:4.2

26.7_+ 3.6 26.35- 3.4

73.6 5- 1.9 0.75:0.1

74.0 5- 3.8 1.05:0.2

67.3 5:3.3 2.05:0.5

64.0 5- 4.6 26.45:10.4

77.15:12.9 4.75- 1.4

86.9+1.4 7.55:3.9

81.25-3.1 6.4_+2.8

56.0+12.3 11.25:6.5

82.05- 2.6 1.25:0.2

82.05-2.5 0.7+0.0

88.05:3.7 0.85-0.1

83.05:4.8 0.75:0.0

Intrathecal Saline Morphin~ 4-day-old Respiration Righting reflex 10-dayzold Respiration Righting reflex KC 4-day-old Respiration Righting reflex 10-day-old Respiration Righting reflex

10. 0

30. 0

110.5 + 1.7 1.4 5- 0.4

105.5 _+1.2 4.4 + 1.2

103.5 + 1.1 5.14- 1.2

107.4 + 0.9 0.55- 0.0

100.4 + 5.1 2.5+0.8

76.65- 7.4 9.95-0.1

110.45:1.8 1.6 5 : 0 . 2

109.35:1.3 5.5 5:1.0

99.35:3.7 7.5 5:1.2

109.05:2.2 0.7 5:0.01

109.25:1.1 2.7 5:1.5

101.75-2.1 4.15- 1.9

Note: cell entries for the respiration rates differ for the two injections. For i.c.v, injections it is the number of breaths per 30 s whereas for the i.t. injections it is per min. Entries are means 5: S.E.M.

the opiate is administered and the type of opiate given. Some have suggested that there is specificity of opiates for different types or intensities of noxious stimuli 15,21'32'33'48'55. That finding is replicated here. Furthermore, as in earlier work, we have found that in the immature animal, the development of analgesia differs for the forepaw, hindpaw and tail. These differences likely reflect developmental changes in the neural organization of pain processing a n d / o r its inhibition in infant rats.

Behavioral specificity Even though both opiates affected respiration rate intrathecal injection, the effects were small compared to those of morphine following i.c.v, administration. It is not likely that the changes in analgesia were secondary to the respiratory depression or increase in righting latency for either route of administration. First, there were few significant correlations between the analgesic response and the breathing rate or righting response for either drug. Second, the analgesic actions of the two drugs differed for the two stimuli. For example, at 14 days of age, when morphine and ketocyclazocine administered intrathecally were most potent in inhibiting the righting reflex and depressing respiration, and analgesia to the mechanical stimulus was potent, there was no effect of these two drugs on the response to the thermal stimulus. Third, the effects on analgesia were somatotopically specific and it is difficult to envision how more generalized effects (e.g. respiratory depression) could affect a single ap-

89 pendage. As an example, in the 3-day-old morphine effectively reduced respiration rates and increased righting latencies, induced analgesia in the forepaw, but had only minimum effects in the hindpaw and tail. Thus it is not likely that the analgesia seen is due to the result of general debilitation of the animal.

Anatomical specificity Intraventricular administration bypasses the spinal opioid systems and the resultant analgesia seen can be attributed either to the activation of descending spinal pathways that synapse at the appropriate spinal site(s) or tO inhibition of the ascending pain signal at supraspinal sites. Spread to the spinal cord is possible but unlikely. First, injection of dye into the lateral ventricle never spread to the spinal cord, although the diffusion of dye is only a very rough approximation of the diffusion of drug. Second, if diffusion of drug to the spinal cord were occurring then it would be expected that the pattern of analgesia recorded after spinal administration would be similar. That did not occur. Finally, the diffusion of drug to the spinal cord would of necessity result in substantial dilution of the drug. Thus it would be expected that the dose-response for i.c.v, treatment would be shifted to the right substantially compared to i.t. administration. That also did not happen. The i.c.v, data, therefore, likely reflect actions of morphine and KC in the brain. Drug administered intraspinally could also spread to hindbrain areas through the CSF and act at this level rather than at the level of the spinal cord. This is unlikely although direct controls have not been performed. The volume that we inject is small and directed in a caudal direction. Because CSF flows down the spinal cord, it is unlikely that significant amounts of drug reaches the brain. Dye injections rarely reached above midcervical regions and we discarded data from subjects with spread farther rostral. Further, i.c.v, injections differed markedly from i.t. injections in their behavioral and physiological effects. In the absence of direct injections into possible caudal brainstem sites, however, we cannot totally rule out this possibility. Intrathecal administration r Opioids injected directly to the spinal cord caused analgesia in the immature animal when a pressure stimulus was used 2. For the thermal stimulus, however, KC was not effective after 4 days of age, and the rostral to caudal pattern of analgesia seen following i.c.v, treatment did not occur following i.t. administration. Morphine generally increased its effectiveness in both tests except for the puzzling lack of effect against the thermal stimulus in 14-day-olds; KC generally be-

came less effective with age. In in vitro isolated spinal cord preparations, a number of opiates depress the response to noxious stimulation 13'25'63. In the neonate, /z opiate receptor agonists such as [D-AIa2,NMe Phe4,Gly-ol]-enkephalin (DAMGO) are most effective and more potent than K opiate receptor preferring compounds such as U69569, PDl17302 or U5048813'25. Morphine itself is effective but not as potent as DAMGO or the x agonists in these preparations 25. That opiates are effective both when administered to the spinal cord and to the brain might explain some discrepancies in the developmental literature (for example see refs. 17,23), since peripheral administration of opiates targets opioid receptors both in brain and in the spinal cord. The primary site(s) at which opiates act following systemic injection likely depends on the site and route of injection and on the stimulus characteristics and test paradigms. For example, s.c. injection just above the spinal cord might favor distribution of drug to spinal cord sites but i.p. injections may not. Whether stimulus characteristics (e.g. intensity or type) are affected by activation of opioid receptors that are specific for brain or spinal cord is an empirical question not yet addressed.

I.c.v. administration There is a large amount of literature that demonstrates that opiates can act at both spinal and supraspinal opioid receptors to produce analgesia and i.c.v, administration of morphine in the 10-day-old was reported to produce analgesia to a mildly noxious thermal stimulus applied to the forepaw 29. In the present study the analgesic properties of morphine administered i.c.v, demonstrated a rostral to caudal pattern of analgesia such that analgesia was seen only in the forepaw in 3-day-old pups and appeared in the hindpaw and tail only at 10 to 14 days of age for the mechanical noxious stimulus. The same caudal progression of analgesia was found following stimulation of the periaqueductal gray of the midbrain s3 (see ref. 54, accompanying paper). For the thermal stimulus, analgesia was never seen in the hindpaws or tail. One mechanism by which opiates act in brain is by activation of descending inhibitory projections that block the transmission of noxious signals at the level of the spinal cord 11. The analgesia produced by i.c.v. administration of morphine in the developing rat is also likely mediated, at least in part, by descending pathways from the brain to the spinal cord. Two possible neurotransmitters that might act to inhibit processing of noxious stimuli are serotonin and norepinephrine. Because there are no serotonin or noradrenergic perikarya in the spinal cord, each monoamine

90 originates from brain and descends to innervate the spinal cord. Anatomical data suggest that the innervation of the cord by either occurs largely postnatally. Serotonergic innervation of the spinal cord develops in a rostral to caudal and ventral-dorsal pattern 12. The noradrenergic system also develops postnatally, is present throughout the cord by 10 days of age but does not demonstrate a clear rostral to caudal development. Whether the different maturational time course of these monoamines is the mechanism by which opiates inhibit pain is not known, but it is possible to speculate that the stimulus difference reflects inhibition by specific descending neurotransmitters. Both in the adult 32'33 and in the infant 18'24 analgesia to the thermal stimulus may be mediated by descending serotonin pathways and that to the mechanical stimulus is coded by descending noradrenergic pathways. Other mechanisms are possible. For example, morphine could be acting at different brain sites that show differential maturation of opioid receptors. The two possibilities are by no means mutually exclusive. Based on prior work in the adult, an important site in the brain that is involved in the descending control of analgesia is the periaqueductal gray of the midbrain 37. The following paper represents initial steps to investigate these possibilities. Acknowledgements. This work was supported in part by NIDA Grant DA-152131, PSC-CUNY grants, and funds from the Biopsychology Doctoral Program, CUNY. REFERENCES 1 Abbott, F.V. and Melzack, R., Brainstem lesions dissociate neural mechanisms of morphine analgesia in different kinds of pain, Brain Res., 251 (1982) 149'-155. 2 Allerton, C.A., Smith, J.A.M., Hunter, J.C., Hill, R.G. and Hughes, J., Correlation of ontogeny with function of [3H]U69593 labelled K opioid binding sites in the rat spinal cord, Brain Res., 502 (1989) 149-157. 3 Alleva, E. and Laviola, G., Short-term and delayed behavioral effects of pre- and post-weaning morphine in mice, PharmacoL Biochem. Behav., 26 (1987) 539-542. 4 Attali, B., Saya, D. and Vogel, Z., Pre- and postnatal development of opiate receptor subtypes in rat spinal cord, Dev. Brain Res., 53 (1990) 97-102. 5 Auguy-Valette, A., Cros, J., Gouarderes, C., Gout, R. and Pontonnier, G., Morphine analgesia and cerebral opiate receptors: a developmental study, Br. J. Pharmacol., 63 (1978) 303-308. 6 Barr, G.A., The behavioral effects of opiates during development. In M.W. Miller (Ed.), Development of the Central Nervous System: Effects of Alcohol and Opiates, Wiley, New York, 1992, pp. 221-254. 7 Barr, G.A., Neuropharmaco-ontogeny: concepts and methods of study. In H. Shair, G.A. Barr and M.A. Hofer (Eds.), Developmental Psychobiology: New Methods and Changing Concepts, Oxford University Press, New York, 1991, pp. 321-341. 8 Barr, G.A., Miya, D. and Paredes, W. (1989) Differences in analgesia induced by intrathecal and intracerebral morphine and ketocyclazocine in 3- and 14-day-old rat pups, Paper presented to the International Narcotic Research Conference, 1989.

9 Barr, G.A., Paredes, W., Erickson, K.L. and Zukin, R.S., re opioid receptor mediated analgesia in the developing rat, Des. Brain Res., 29 (1986) 145-152. 10 Barr, G.A., Miya, D. and Paredes, W., Changing patterns of analgesia induced by lateral ventricle or spinal injections of morphine or ketocyclazocine in developing rats. Neurosci. Abstr.. 13 (1987) 1001. 11 Bashaum, A.I. and Fields, H.L., Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry., Annu. Ret,. Neurosci., 7 (1984) 309-338. 12 Bregman, B.S., Development of serotonin immunoreactivity in the rat spinal cord and its plasticity after neonatal spinal cord lesions, Brain Res., 34 (1987) 245-263. 13 Briggs, I. and Barnes J.C., Actions of opioids on the dorsal root potential of the isolated spinal cord preparation of the neonate rat, Neuropharmacol., 26 (1987) 469-475. 14 Denenberg, V.H., Assessing the effects of early experience. In R.D. Myers (Ed.) Methods in Psychobiology: Advanced Laboratory Techniques in Neuropsychology and Neurology (Vol. 3), Academic Press, New York, 1977, pp. 127-147. 15 Dykstra, L.A., Behavioral and pharmacological factors in opioid analgesia. In L.S. Seiden and R.L. Balster (Eds.), Behavioral Pharmacology The Current Status, Liss, New York, 1985, pp. 111-129. 16 Ellis, S., Axt, K. and Epstein, A.N., The arousal of ingestive behaviors by chemical injection into the brain of the suckling rat, J. Neurosci., 4 (1983) 945-955. 17 Giordano, J. and Barr, G.A., Morphine- and ketocyclazocine-induced analgesia in the developing rat: differences due to type of noxious stimulus and body topography, Det~. Brain Res., 32 (1987) 246-253. 18 Giordano, J. and Barr, G.A., Effects of neonatal spinal cord serotonin depletion on opiate-induced analgesia in tests of thermal and mechanical pain, Dee. Brain Res., 41 (1988) 121-127. 19 Han, J.S., Xie, G.X. and Goldstein, A., Analgesia induced by intrathecal injection of dynorphin B in the rat, Life Sci., 34 (1984) 1573-1579. 20 Harms, P.G. and Ojeda, S.R., A rapid and simple procedure for chronic cannulation of the rat jugular vein, J. Appl. Physiol., 36 (1974) 391-392. 21 Hayes, A.G., Skingle, M. and Tyers, M.B., Antinociceptive profile of dynorphin in the rat, Life Sci., 33 Suppl. 1 (1983) 657-660. 22 Hayes, A.G., Sheehan, M.J. and Tyers, M.B., Differential sensitivity of models of antiinociception in the rat, mouse and guineapig to mu- and kapp-opioid receptor agonists, Br. J., Pharmacol., 91 (1987) 823-832. 23 Helmstetter, F.J., Calcagnetti, D.J., Cramer, C.P. and Fanselow, M.S., Ethylketocyclazocine and bremazocine analgesia in neonatal rats, PharmacoL Biochem. Behav., 30 (1988) 817-821. 24 Hughes, H.E. and Barr, G.A., Analgesic effects of intrathecally applied noradrenergic compounds in the developing rat: differences due to thermal vs. mechanical nociception, Dev. Brain Res., 41 (1988) 109-120. 25 James, I.F., Bettaney, J., Perkins, M.N., Ketchum, S.B. and Dray, A., Opioid receptor ligands in the neonatal rat spinal cord: binding and in vitro depression of the nociceptive responses, Br. J. Pharmacol., 99 (1990) 503-508. 26 Jensen, T.S. and Yaksh, T.L., Comparison of the antinociceptive action of mu and delta opioid receptor ligands in the periaqueductal grey matter, medial and paramedial ventral medulla in the rat as studied by the microinjection technique. Brain Res., 372 (1986) 301-312. 27 Jhamandes, K., Sutak, M. and Lemaire, S., Comparative spinal analgesic action of dynorphin 1-8, dynorphin 1-13 and a kappareceptor agonist, U50-488, Can. J. PhysioL Pharmacol., 64 (1986) 263 -268. 28 Johannesson, T. and Becker, B.A., Morphine analgesia in rats at various ages, Acta PharmacoL Toxicol., 33 (1973) 429-441. 29 Kehoe, P. and Blass, E.M., Opioid-mediation of separation distress in 10-day-old rats; reversal of stress with maternal stimuli, Dev. Psychobiol., 19 (1986) 385-398. 30 Kupferberg, H.J. and Way, E.L., Pharmacologic basis for the

91 increased sensitivity of the newborn rat to morphine, J. Pharmacol. Exp. Ther., 141 (1963) 105-112. 31 Kuraishi, Y , Fukui, K., Shiomi, H., Akaike, A. and Takagi, H., Microinjections of opioids into the nucleus reticularis gigantocellularis of the rat: analgesia and increase in the normetanephrine level in the spinal cord, Biochem. Pharmacol., 27 (1978) 27562758. 32 Kuraishi, Y., Harada, Y., Aratani, S., Satoh, M. and Takagi, H., Separate involvement of the spinal noradrenergic and serotonergic systems in morphine analgesia. The differences in mechanical and thermal algesic tests, Brain Res., 273 (1983) 245-252. 33 Kuraishi, Y., Hirota, N., Satoh, M. and Takagi, H., Antinociceptive effects of opioids, noradrenaline and serotonin in rats: mechanical and thermal algesic tests, Brain Res., 326 (1985) 168-171. 34 Laughlin, S.E., Massamiri, T.R., Kornblum, H.I. and Leslie, F.M., Postnatal development of opioid systems in rat brain, Neuropeptides, 5 (1985) 469-472. 35 Leighton, G.E., Rodriquez, R.E., Hill, R.G. and Hughes, J., Kappa-opiate agonists produce antinociception after i.v. and i.c.v. but not intrathecal administration in the rat. Br. J. Pharmacol., 93 (1988) 553-560. 36 Leslie, F.M., Tso, S. and Hurlbut, D.E., Differential appearance of opiate receptor subtypes in neonatal rat brain, Life Sci., 31 (1982) 1393-1396. 36 Lewis, V.A. and Gebhart, G.F., Evaluation of the periaqueductal gray (PAG) as a morphine-specific locus of action and examination of morphine-induced and stimulation-produced analgesia at coincident PAG loci, Brain Res., 124 (1977) 283-303. 38 McDowell, J. and Kitchen, I., Ontogenesis of &opioid receptors in rat brain using [3H]D-Pen2,D-PenS]enkephalin as a binding ligand, Eur. J. Pharmacol., 128 (1986) 287-289. 39 McDowell, J. and Kitchen, I., Development of opioid systems peptides receptors and pharmacology, Brain Res. Rev., 12 (1987) 397-421. 40 Millan, M.J., Kappa-opioid receptor-mediated antinociception in the rat. I. Comparative actions of mu- and kappa-opioids against noxious thermal, pressure and electrical stimuli, J. Pharmacol. Exp. Ther., 251 (1989) 334-341. 41 Paredes, W., Hughes, H.E., Giordano, J. and Barr, G.A., Methods of injecting drugs directly into the spinal cord of neonatal rats, Lab. Animal, 19 (1990) 39-41. 42 Pasternak, G.W., Zhang, A. and Tecott, L., Developmental differences between high and low affinity opiate binding sites: their relationship to analgesia and respiratory depression, Life Sci., 27 (1980) 1185-1190. 43 Piercy, M.F., Lahti, R.A., Schroeder, L.A., Einspahr, F.I. and Barshrun, C., U-50,488H, a pure kappa receptor agonist with spinal analgesic loci in the mouse, Life Sci., 31 (1982) 1197-1200. 44 Porreca, P., Mosberg, H.I., Hurst, R., Hruby, V.J. and Burks, T.F., Roles of mu delta and kappa opioid receptors in spinal and supraspinal mediation of gastrointestinal transit effects and hotplate analgesia in the mouse, Pharmacol. Exp. Ther., 230 (1984) 341-348. 45 Porreca, F., Heyman, J.S., Mosberg, H.I., Omnas, J.R and Vaught, J.L., Role of mu and delta receptors in the supraspinal and spinal analgesic effects of (D-Pen 2, D-PenS)enkephalin in the mouse, J. Pharmacol. Exp. Ther., 241 (1987) 393-400. 46 Risau, W. and Wolberg, H., Development of the blood-brain-barrier, Trends Neurosci., 13 (1990) 174-178. 47 Schmauss, C., Spinal kappa-opioid receptor-mediated antinociception is stimulus-specific, Eur. J., Pharmacol., 137 (1987) 197-

205. 48 Schmauss, C. and Yaksh, T.L., In vivo studies on spinal opiate receptor system mediating antinociception. III. Pharmacological profiles suggesting a differential association of mu, delta and kappa receptors with visceral chemical and cutaneous thermal stimuli in the rat, J. Pharmacol. Exp. Ther., 228 (1984) 1-12. 49 Spain, J.W., Roth, B.L. and Coscia, C.J., Differential ontogeny of multiple opioid receptors (p., ~ and K), J. Neurosci., 5 (1985) 584-588. 50 Spear, L.P., Enters, E.K., Aswad, M.A. and Louzan, M., Drug and environmentally induced manipulations of the opiate and serotonergic systems alter nociception in neonatal rat pups, Behay. Neural. Biol., 44 (1985) 1-20. 51 Stevens, C.W. and Yaksh, T.L., Dynorphin A and related peptides administered intrathecally in the rat: a search for putative kappa opiate receptor activity, J. Pharmacol. Exp. Ther., 238 (1986) 833-838. 52 Tavani, A., Robson, L.E. and Kosterlitz, H.W., Differential postnatal development of/~-, iS- and r-opioid binding sites in mouse brain, Dev. Brain Res., 23 (1985) 306-309. 53 Tive, L.A., Maturational changes in analgesia produced by glutamate and morphine stimulation of the PAG, Unpublished Doctoral Dissertation, Dissertation Abstr. 49B (1991) 3493. 54 Tive, L.A. and Barr, G.A., Analgesia from the periaqueductal gray in the developing rat: focal injections of morphine or glutamate and effects of intrathecal injection of methysergide or phentolamine, Brain Res., 584 (1992) 92-109. 55 Tyers, M.B., A classification of opiate receptors that mediate antinociception in animals, Br. J. Pharmacol., 69 (1980) 503-512. 56 Unterwald, E., Sasson, S. and Kornetsky, C., Evaluation of the supraspinal analgesic activity and abuse liability of ethylketocyclazocine, Eur. J. Pharmacol., 133 (1987) 275-281. 57 Upton, N., Sewell, R.D. and Spencer, P.S., Differentiation of potent mu and kappa-opiate agonists using heat and pressure antinocieeption profiles and combined potency analysis, Eur. J. Pharmacol., 78 (1982) 421-429. 58 Ward, S.J. and Takemori, A.E., Relative involvement of mu, kappa and delta receptor mechanisms in opiate-mediated antinociception in mice, J. Pharmacol. Exp. Ther., 224 (1983) 525-530. 59 Wohltmann, M., Roth, B.L and Coscia, C.J., Differential postnatal development of mu and delta opiate receptors, Deu. Brain Res., 3 (1982) 679-684. 60 Wood, P.L., Tackham, A. and Richard, J., Spinal analgesia: comparison of the mu agonist morphine and the kappa agonist ethylketazocine, Life Sci., 28 (1981) 2119-2125. 61 Yaksh, T.L., Analgesic action of intrathecal opiates in cat and primate, Brain Res., 153 (1978) 205-210. 62 Yaksh, T.L., Yeung, J.C. and Rudy, T.A., Systematic examination in the rat of brain sites sensitive to direct application of morphine. Observation of differential effects within the periaqueductal gray, Brain Res., 114 (1976) 83-103. 63 Yanagisawa, M., Murakoshi, T., Tamai, S. and Otsuka, M., Tailpinch method in vitro and effects of some antinociceptive compounds, Eur. J. Pharmacol., 106 (1984) 231-239. 64 Yaster, M., Analgesia and anesthesia in neonates, J. Pediat., 111 (1987) 394-396. 65 Zhang, A.-Z. and Pasternak, G.W., Ontogeny of opioid pharmacology and receptors. High and low affinity site differences, Eur. J. Pharmacol., 73 (1981) 29-40.