Interleukin 1 reduces opioid binding in guinea pig brain

Peptides, Vol. 6, pp. 114%1154, 1985. ¢ AnkhoInternationalInc. Printed in the U.S.A.

01%-9781/85 $3.00 + .00

Interleukin 1 Reduces Opioid Binding in Guinea Pig Brain M. S. A H M E D , * J. L L A N O S - Q . , t C. A. D I N A R E L L O $ A N D C. M. B L A T T E I S t 1

Departments of *Biochemistry and of tPhysiology and Biophysics University of Tennessee Center fi~r the Health Sciences, Memphis, TN 38163 and ?-Department of Medicine, Tufts University School of Medicine, Boston, MA 02111 R e c e i v e d 19 F e b r u a r y 1985 AHMED, M. S., J. LLANOS-Q., C. A. DINARELLO AND'C. M. BLATTEIS. lnterleukin I reduces opioid binding in guinea pig brain. PEPTIDES 6(6) 114%1154, 1985.--lnterleukin 1 (ILl) is a macrophage-derived polypeptide which signals neurons in the preoptic-anterior hypothalamus to initiate fever and the acute-phase glycoprotein response. Recently, increases in cerebrospinal fluid and hypothalamic levels of/3-endorphin have been reported during endotoxin (LPS)- and ILl-induced fevers, suggesting that this opioid may participate in the modulation of ILl effects in the CNS. In this study, we investigated whether purified (human) ILl influences the specific binding of three prototypic opioid agonists (2-Dalanine-5-L-methionineamide, DAME; (-)-ethylketocyclazocine, EKC; dihydromorphine, DHM) and one antagonist (naloxone) to opioid receptor-enriched membrane preparations in cerebral cortex, hypothalamus, midbrain, pons, medulla, and cerebellum of guinea pig brain. ILl reduced the binding of these ligands to their receptors during a 30-min incubation. The extent of ILl inhibition of a given ligand for its binding sites varied according to the brain region; within some regions, the extent of this inhibition also varied with the four ligands tested. But in cortex the effect of ILl on the specific binding of DHM is dose-dependent. Similar results were obtained with crude homologous ILl. S. enteritidis endotoxin, suspended in pyrogen-free saline at concentrations from 4 to 36/zg/ml, did not inhibit the binding of these opioid ligands to their receptors in any brain region. These results indicate that ILl interacts with the opiate receptors in guinea pig brain. This interaction, moreover, is not limited to the hypothalamus alone, the primary site of the pyrogenic action of ILl, but also occurs in other brain regions. Fever

Endotoxin

Hypothalamus

Brain regions

I N T E R L E U K I N 1 (ILl), a 17.5 K-dalton protein produced principally by macrophages/monocytes in response to diverse stimuli, is the endogenous mediator of fever and of various inflammatory and immunological host defense responses which characteristically accompany it (reviewed by [16, 24, 32]). ILl evokes fever and the acute-phase glycoprotein response by activating, in an as yet unknown manner, neuronal mechanisms located primarily in the preopticanterior hypothalamus (POAH) [6,7]. Certain extra-POAH regions also may be sensitive to locally-applied ILl (reviewed by [4,36]). Circulating ILl may enter the brain interstitium through the circumventricular organs, e.g., the organum vasculosum laminae terminalis [6]. Alternatively, ILl may be synthesized within the brain [20]. It is assumed that ILl initiates its multiple effects by binding to specific receptors, but their precise localization and nature remain unknown. Various neurochemical mediators have been implicated as mediators of the ILl effects in the CNS, but none have as yet been unequivocally identified (reviewed by [14]). Recently, increases in cerebrospinal fluid and hypothalamic levels of/3-endorphin have been reported during endotoxinand ILl-induced fevers, including in naloxone-treated animals [8,30], suggesting that this opioid peptide may participate in the febrigenic action of ILl. Indeed, low doses of

Opioid agonists

Naloxone

Receptors

/3-endorphin, morphine, and various other agonists are generally hyperthermogenic when microinjected intracerebrally (including into the POAH) (reviewed by [%12]). Naloxone, however, does not antagonize the febrile responses to endotoxin and ILl [2, 8, 13, 25]. Hence, the possible role of endogenous opioids in the central control of ILl-induced fever remains equivocal. Whether the opioids may be involved in the modulation of the nonfebrile effects of ILl has not yet been investigated. Since ILl and opioids induce in vivo apparently similar effects on the endocrinologic, immunologic, and autonomic systems (reviewed by [4, 16, 24, 28, 29, 32, 37]), and since both affect common brain regions (reviewed by [5,35]), we elected to determine whether the actions of ILl may, in fact, be modulated through the opioid system. We report here on the influence of ILl on the in vitro binding of opioids to their receptor subtypes in different regions of the guinea pig brain, using prototypic 8-, K-, and /z-selective agonists and one antagonist; the brain regions were chosen on the basis of previous observations indicating that they may contain ILl-sensitive sites [3]. METHOD

Animals Fifty-three adult, male, Hartley guinea pigs were used in

~Requests for reprints should be addressed to Clark M. Blatteis, Ph.D., Department of Physiology and Biophysics, University of Tennessee Center for the Health Sciences, 894 Union Avenue, Memphis, TN 38163.

1149

AHMED ET AL.

1150 4O

0f ~

4

Cortex

Cortex

0

4

,~

8( ~

Cortex

liraC IL 1

:Oof.. .....u oil-~ Ot

':E

Midbrein~2000 ~

':l L

Hypothal

. Ions

Midbrain

~ 40

I 0

JL"d"" ,:L

'°°e

C°rtex

,oo

..d.,. o[L

|.

40

Midbrain

f 0

Medulla

,o.

..du,,.

FIG. I. Effects of human purified ILl (5 txl/0.5 ml of medium) on the specific binding of: (A) the 8-1igand aH-2-D-alanine-5-Lmethionineamide (DAME, 1.7 nM) in 5 guinea pig brain regions-results (means_+SEM) of 3 experiments; (B) the K-ligand 3H-(-)ethylketocyclazocine (EKC, 3.2 nM) in 6 guinea pig brain regions-results of 4 experiments; (C) the p.-ligand 3H-dihydromorphine (DHM, 1.8 nM) in 5 guinea pig brain regions--results of 3 experiments. C=control. *p<0.05.

FIG. 2. Effects of human purified ILl (5 ~zl/0.5ml of medium) on the binding of '~H-(-)-naloxone (NAL, 2.0 nM) in 6 guinea pig brain regions--results of 3 experiments. *p<0.05.

DHM ~

these experiments. They were purchased from Murphy Breeding Labs, Plainfield, IN at weights of 300-349 g. This species was chosen because a broad data base already exists regarding its responses to ILl [4]. The animals were quarantined, three to a cage, for at least 4 and usually 10 days prior to use. They were provided Ralston-Purina Guinea Pig Chow and tap water, ad lib; fresh cabbage supplement was offered every other day. The ambient temperature of the animal room was near 22°C; light and darkness were alternated, with light on from 0600 to 1800 hr. The animals were routinely inspected for signs of disease or any untoward conditions, and only healthy guinea pigs were used in these experiments. T o - o b v i a t e possible superimposed effects of circadian rhythm, all the brains were obtained at 0930 hr.

Opioid Receptors Membrane Preparations The animals were decapitated and their brains rapidly removed and dissected on an ice-cooled plate according to the method of Glowinski and Iversen [22]. Membrane-bound opioid receptor preparations were obtained from the cerebral cortex, hypothalamus, midbrain, pons, medulla, and cerebellum. The tissues were homogenized in 0.32 M sucrose in 0.01 M Tris HCI buffer (pH 7.4, 4°C) and centrifuged at 1,000 g for 10 rain; the pellet was discarded. The supernatant was recentrifuged at 48,000 g for 10 min at 4°C, and the resulting pellet was washed twice by resuspending in the buffer solution and recentrifuging at 48,000 g. The final pellet was resuspended in buffer at a concentration of 1-2 mg of protein/ml and stored at - 15°C until needed; the maximum interval until the binding assay was one week. Protein was estimated using the Bio-Rad Protein Assay Kit with bovine serum albumin as the standard.

E

|

DAME '

$o Cortex

Milbr

FIG. 3. Effects of crude homologous ILl (=endogenous pyrogen [EP], 50/zl/0.5 ml of medium) on the binding of DHM (I .9 nM) in guinea pig cortex, and of DAME (I .6 nM) in guinea pig midbrain-results of 2 experiments each. *p<0.05.

Binding Assay Aliquots of the membrane homogenates containing 400/.Lg of protein were incubated for 30 min at 27°C with 1-4 nM of tritiated opioid ligand, in the presence or absence of I p.M of nonlabelled ligand or levorphanol. The final volume of the assay mixture was 0.5 ml. The binding assay was terminated by centrifuging at 48,000 g for 10 min at 4°C. The pellet was washed without resuspending, using 1 ml of ice-cold buffer, and then resuspended in 1% Triton XI00 solution. Scintillation fluid was added and the samples were counted. The following ligands (purchased from New England Nuclear, Boston, MA) were used: the 8-agonist 3H-2-D-alanine-5-Lmethionineamide (DAME, 39.4 Ci/mmol), the K-agonist 3H(-)-ethylketocyclazocine (EKC, 21.9 Ci/mmol), the /z-agonist 3H-dihydromorphine (DHM, 79.0 Ci/mmol), and the antagonist 3H-(-)-naloxone (NAL, 25.3 Ci/mmol). Specific binding was calculated as the difference between the counts of radioligand bound in the presence and absence of nonlabelled ligand or levorphanol. Each binding assay was performed in quadruplicates. The amount of ligand specifically bound represented from 20 to 60% of the total ligand bound, depending on the ligand and region assayed. The effect of I L l on the binding of the opioids was de-

O P I O I D S - - I N T E R L E U K I N I INTERACTIONS

1151

TABLE 1 EFFECTS OF HUMANPURIFIEDILl ON THE SPECIFICBINDINGOF 40PIOID LIGANDS TO DIFFERENTGUINEAPIG BRAINREGIONS Brain Regions Ligands DAME

Cortex

Hypothalamus

Midbrain

0

60

24

Pons

Medulla

Cerebellum

0

40

NA*

EKC

0

60

0

0

28

69

DHM

67

78

70

84

0

NA*

NAL

0

24

27

0

0

45

Values are percent inhibition in binding of the ligand in the presence of ILl; 0=no significant effect. Summary of I l experiments. *NA=not assayed.

4O

Cortex

Hypothel

*J

Pons

~20

O

10

\

0 CILI1

3 S 7 • 1 LPS

9

i J L 0 24

L 8

I 16 ILl

32 dose

'

()Jl)

FIG. 4. Effects of human purified ILl (5/~1/0.5 ml of medium) and of S. enteritidis endotoxin (LPS, suspended in pyrogen-free saline) in the concentration range from 2 to 18/xg/0.5 ml of medium (I to 9 p.l) on the binding of DHM (1.1 nM) in 3 guinea pig brain regions-results of 3 experiments each. *p<0.05.

FIG. 5. Effects of 5 doses (2 to 50/~1/0.5 ml of medium) of human purified ILl on the binding of DHM (2.0 nM) in guinea pig cortex-results of 3 experiments. The original ILl preparation was diluted 1:10.

termined by adding 5/xl of human purified ILl to the binding assay tubes, in triplicates. The mixture was incubated for 30 min at 27°C, then centrifuged, washed, and counted as before. Specific radioligand binding remaining was evaluated by deducting the counts in the presence of nonlabelled ligand or levorphanol alone from those in the presence of added ILl. The purification and characteristics of ILl have been published previously [18]. For these studies, an improved immunoadsorbant was employed, followed by gel-filtration and ion-exchange chromatography. The ILl lymphocyte proflieration assay was used to monitor each fraction. Active fractions were pooled and concentrated by dialysis against polyethylene glycol (MW 6-8,000). The concentrate was then dialyzed against phosphate-buffered saline. It contained 10,000 LAF units/ml; the intravenous injection of 100/~1 of this material produced a I°C fever in rabbits. Its pyrogenicity was further verified by microinjecting 1/xl bilaterally into the POAH of six guinea pigs; it produced a characteristic monophasic fever of 1.1-0.2°C (mean_+SD). The effects of opioid binding of 50/~1 of crude homologous (guinea pig) ILl and of S. enteritidis en~iotoxin (LPS, Boivin preparation, batch No. 651628, Difco, suspended in pyrogen-free saline at concentrations from 4 to 36 p~g/ml) were assayed similarly. The crude ILl was prepared from

guinea pig adherent blood cells activated by LPS, as described previously [6]. In addition, the cell-free solution of crude ILl was passed through an HPLC column to insure that low molecular weight opioid peptides that might be contained in the solution would not interfere with the assays. Finally, to assess the nature of the possible interaction between ILl and opioid receptors, the specific binding of DHM (83.3 Ci/mmol) was measured in a single brain region, the cortex, using 5 concentrations of purified human ILl. To this end, the original ILl preparation was diluted l:10. Statistics

The amounts of radioligand specifically bound in the presence or absence of ILl or LPS were compared statistically by Student's paired t-test. The null hypothesis was rejected at the 5% level. RESULTS Receptors of all three subtypes were abundantly distributed in the cortex, hypothalamus, and midbrain (Fig. l, solid bars). By contrast, the pons and medulla were relatively poor in 8- and/~-receptors but contained a moderate amount of K-receptors. Overall, EKC binding was highest in all the

1152

AHMED ET AL.

brain regions, while DAME and DHM bindings were lower and approximately equal. The cerebellum was not assayed for DAME and DHM bindings. The effects of human purified I L l on the specific binding of the various opioid agonists are presented in Fig. 1, hatched bars. Overall, I L l inhibited the binding of DHM more potently than that of DAME or EKC, in that order, in all the brain regions. I L l significantly attenuated DAME binding in the hypothalamus, midbrain, and medulla, but not in the cortex and pons. It also reduced EKC binding in the hypothalamus, medulla, and cerebellum, but not in the cortex, midbrain, and pons. In addition, it inhibited DHM binding in the cortex, hypothalamus, midbrain, and pons, but not in the medulla. For all the ligands, the attenuation was most prominent in the hypothalamus, intermediate in the midbrain, and variable in the cortex and medulla (Table 1). The specific binding of naloxone was highest in the hypothalamus and midbrain, intermediate in the cortex, medulla and cerebellum, and absent in the pons. It was depressed sparingly by I L l in the hypothalamus and midbrain and more so in the cerebellum (Fig. 2 and Table 1). Crude, homologous I L l attenuated the binding of both DAME and DHM in the cortex and midbrain in a manner similar to the effect of human purified I L l (Fig. 3). But LPS did not affect the binding of DHM in any of the brain regions (Fig. 4). Pyrogen-free saline, the vehicle for the crude I L l , did not significantly affect the binding of any of the agonists. The effects of 5 doses of human purified I L l on the specific binding of DHM in cortical membranes are shown in Fig. 5. It is apparent that the amount of this agonist specifically bound is dependent on the dose of I L l added to the medium. DISCUSSION

The results of the present study indicate that I L l reduces the in vitro binding of these prototypic ~ (DAME)-, K (EKC)-, and/x

(DHM)-selective ligands to their respective opioid receptor subtypes in synaptosomal membrane preparations from the guinea pig brain. It is interesting to note that this inhibitory effect of I L l is not limited to the hypothalamus, which contains the primary site mediating the febrigenic and certain acute-phase actions of I L l [7], but also occurs in other brain regions. This result is consistent with observations that the cortex, midbrain, pons, and medulla also may contain I L l sensitive sites (reviewed by [3,36]). Indeed, the microinjection of I L l into the midbrain reticular formation of rabbits [34] and into the pons and medulla of guinea pigs [4] induces fever. However, it is not yet known whether I L l injected into these extra-hypothalamic sites also evokes any acutephase effects. While injections of I L l into the cortex generally do not produce fever [3], tissue from this region has been shown to synthesize prostaglandin E2, a putative mediator of fever, when incubated with I L l in vitro [17]. In light of the present results, this suggests that the cortex may be involved in the mediation of nonfebrile effects of I L l . The finding that I L l depresses EKC binding in the cerebellum is of interest; no studies as yet have been made on the possible modulation of I L l effects by this brain region. Taken together, these data support the notion that central opioids may modulate other, as yet unknown, effects of I L l in the CNS. The finding that naloxone binding in the hypothalamus was only moderately inhibited in the presence of I L l would be consistent with the earlier observation that the hypothalamic level offl-endorphin is increased in sheep following the administration of LPS and not depressed by pre-treatment

with naloxone [8]. Thus, the interaction between ILl and opioid receptors may require that these exist in their agonistic form. The reduction caused by I L l in the specific binding of naloxone in the cerebellum is as yet difficult to interpret, since 84% of opioid binding sites in the guinea pig cerebellum is of the K-subtype [27], which has low affinity for naloxone. In view of the limited availability of purified I L l , we also determined whether crude I L l might similarly interact with opioid receptors. The results obtained were identical with those from pure I L l . Since low molecular weight peptides were excluded from the crude preparation of I L l , i.e., no opioid peptides were present, it is probable that the observed effects were due to I L l . Moreover, the possibility that endotoxin contaminating the crude I L l preparation might be responsible for the effect was ruled out by the demonstration that pure endotoxin, in the concentration range examined, did not inhibit the binding of DHM to its receptors in any brain region. This result suggests that the reported elevation of cerebral fl-endorphin levels during endotoxic fever [8,30] is consequent to the I L l generated by macrophages in response to the LPS and not a direct effect of the LPS on central neurons. The extent of I L l inhibition of a given opioid binding to its receptors varied according to the brain region; moreover, within some regions, the interaction between I L l and the three agonists and one antagonist tested also differed. A portion of this variability could be due to the lack of specificity of the agonists used to a single receptor subtype and our dependence on a single dose of I L l . For that matter, the different potency of I L l found in the different brain regions may reflect the different proportions of p., 8, and r-sites present, but poorly differentiated due to the lack of selectivity of the agents used. Another part could be due to the unavoidable variations in the dissections of individual brain regions, particularly in the brainstem. However, true quantitative differences in the inhibitory effect of I L l on the binding of opioid ligands to their receptor subtypes may also exist in different brain regions. This would be consistent with the differential distribution of the opioid peptides and of their corresponding receptors as well as with their differential effects on neurons in different regions of the brain [l, 15, 19]. This possibility could be elucidated by using receptorspecific, radiolabelled opioids in the presence of masking concentrations of other ligands specific to different receptor subtypes, and by studying more anatomically discrete regions. Nevertheless, the finding that the attenuation of the specific binding of DHM in cortex is dependent on the dose of I L l indicates that the observed interaction of this compound with opioid receptors is competitive in nature. To the best of our knowledge, no data have been reported previously on the regional distribution of opioid binding sites in the guinea pig brain. In general, the various receptor subtypes, as characterized by the present selected agonists, would appear to occur in the different regions of the guinea pig brain in differential patterns and densities similar to those of rats [I, 15, 19, 31], with the exception of the cerebellum which, in rats, lacks opioid receptors [31]. The present results also conform to earlier findings [21, 23, 26, 27, 33] that K-receptors are particularly abundant in guinea pig whole brain homogenates and are predominant in the cerebellum. In conclusion, these results show that I L l reduces the specific binding of opioid agonists and antagonists to their corresponding receptor subtypes in different regions of the guinea pig brain. They do not indicate what functions of ILl might be affected in this way. Since the induction of ILl

OPIOIDS--INTERLEUKIN

1 INTERACTIONS

1153

f e v e r d o e s not a p p e a r to b e m o d u l a t e d t h r o u g h n a l o x o n e s e n s i t i v e opioid r e c e p t o r s [2, 8, 13, 25], it is p o s s i b l e t h a t the

role o f the opioids m a y b e in the c e n t r a l c o n t r o l o f the nonfebrile effects o f I L l .

ACKNOWLEDGEMENTS This study was supported in part by NSF grants No. BNS 8308257 and PCM 83-17217, NIH AI15614, and the Dean's Special Education Fund in the College of Medicine. J. Llanos-Q. was an International Research Fellow of the Fogarty International Center (Grant No. FOS TWO 3099-0151). He is presently in the Department of Physiology, Faculty of Medicine, National University of Trujillo, Trujillo, Peru.

REFERENCES 1. Atweh, S. F. and M. J. Kuhar. Autoradiographic localization of opioid receptors in rat brain. I. Spinal cord and lower medulla. Brain Res 124: 53-67, 1977. 2. Bl~isig, J., V. B/iuerle and A. Herz. Endorphin-induced hyperthermia: characterization of the exogenously and endogenously induced effects. Naunyn Schmiedebergs Arch Pharmacol 309: 137-143, 1979. 3. Blatteis, C. M. Functional anatomy of the hypothalamus from the point of view of temperature regulation. In: Contributions to Thermal Physiology, edited by Z. Szelrnyi and M. Szrkely. Budapest: Akadrmiai Kiado, 1980, pp. 3-12. 4. Blatteis, C. M. Endogenous pyrogen: Fever and associated effects--an update. In: Thermal Physiology, edited by J. R. S. Hales. New York: Raven Press, 1984, pp. 539-546. 5. Blatteis, C. M. Neural mechanisms in the pyrogenic and acutephase responses to interleukin 1. In: Proc. Ist International Workshop on Neuroimmunomodulation, edited by H. N. Spector, New York: Gordon and Breach, in press. 6. Blatteis, C. M., S. L. Bealer, W. S. Hunter, J. Llanos-Q., R. A. Ahokas and T. A. Mashburn, Jr. Suppression of fever after lesions of the anteroventral third ventricle in guinea pigs. Brain Res Bull 11: 51%526, 1983. 7. Blatteis, C. M., W. S. Hunter, J. Llanos-Q., R. A. Ahokas and T. A. Mashburn, Jr. Activation of acute-phase responses by intrapreoptic injections of endogenous pyrogen in guinea pigs. Brain Res Bull 12: 68%695, 1984. 8. Carr, D. B., R. Bergland, A. Hamilton, H. Blume, N. Kasting, M. Arnold, J. B. Martin and M. Rosenblatt. Endotoxinstimulated opioid peptide secretion: two secretory pools and feedback control in vivo. Science 217: 845-848, 1982. 9. Clark, W. G. Influence of opioids on central thermoregulatory mechanisms. Pharmacol Biochem Behav 10: 60%613, 1979. 10. Clark, W. G. Changes in body temperature after administration of amino acids, peptides, dopamine, neuroleptics and related agents. Neurosci Biobehav Rev 3: 17%231, 1979. 11. Clark, W. G. Effects of opioid peptides on thermoregulation. Fed Proc 40: 2754--2759, 1981. 12. Clark, W. G. and Y. L. Clark. Changes in body temperature after administration of acetylcholine, histamine, morphine, prostaglandins and related agents. Neurosci Biobehav Rev 42: 175-240, 1980. 13. Clark, W. G. and N. F. Harris. Naloxone does not antagonize leukocytic pyrogen. Eur J Pharmacol 49: 301-304, 1978. 14. Cox, B. and T. F. Lee. Role of central neurotransmitters in fever. In: Pyretics and Antipyretics, edited by A. S. Milton. Berlin: Springer Verlag, 1982, pp. 125-150. 15. Della Bella, D., F. Casacci and A. Sassi. Opiate receptors: different ligand affinity in various brain regions. In: Advances in Biochemical Psychopharmacology, vol 18, edited by E. Costa and M. Trabucchi. New York: Raven Press, 1978, pp. 271-277. 16. Dinarello, C. A. Interleukin-1. Rev Infect Dis 6: 51-95, 1984. 17. Dinarello, C. A. and H. A. Bernheim. The ability of human leukocytic pyrogen to stimulate brain prostaglandin synthesis in vitro. J Neurochem 37: 702-708, 1981.

18. Dinarello, C. A., L. Renfer and S. M. Wolff. Human leukocytic pyrogen: purification and development of a radioimmunoassay. Proc N a t / A c a d Sci USA 74: 4624-4627, 1977. 19. Duka, Th., P. Schubert, M. Wiister, R. Stoiber and A. Herz. A selective distribution pattern of different opiate receptors in certain areas of rat brain as revealed by in vitro autoradiography. Neurosci Lett 21: 11%124, 1981. 20. Fontana, A., E. Weber and J. M. Dayer. Synthesis of interleukin 1/endogenous pyrogen in the brain of endotoxin-treated mice: a step in fever induction? J lmmuno/ 133: 1696-1698, 1984. 21. Foote, R. W. and R. Maurer. Kappa opiate binding sites in the substantia nigra and bulbus olfactorius of the guinea pig as shown by in vitro autoradiography. Life Sci 33: 243-246, 1983. 22. Glowinski, J. and L. L. Iversen. Regional studies of catecholamines in the rat brain. I. The disposition of [3H]norepinephrine, [3H]dopamine, and [3H]DOPA in various regions of the brain. J Neurochem 13: 655-669, 1966. 23. Itzhak, Y., J. M. Hiller and E. J. Simon. Solubilization and characterization of/z, 8, and K opioid binding sites from guinea pig brain: physical separation of r receptors. Proc Natl Acad Sci USA 81: 4217-4221, 1984. 24. Kampschmidt, R. F. The numerous postulated biological manifestations of interleukin-l. J Leukocyte Bio! 36: 341-355, 1984. 25. Kandasamy, S. B. and B. A. Williams. Central effects of some peptide and non-peptide opioids and naloxone in thermoregulation in the rabbit. In: Environment, Drugs, and Thermoregulation. edited by P. Lomax and E. Schrnbaum. Basel: Karger, 1983, pp. 98-100. 26. Kosterlitz, H. W., S. J. Paterson and L. E. Robson. Characterization at the K-subtype of the opiate receptor in the guinea-pig brain. Br J Pharmacol 73: 93%949, 1981. 27. Magnan, J., S. J. Paterson, A. Tavani and H. W. Kosterlitz. The binding spectrum of narcotic analgesic drugs with different agonist and antagonist properties. Naunyn Schmiedebergs Arch Pharmacol 319: 197-205, 1982. 28. Martin, W. R. Pharmacology of opioids. Pharmacol Rev 35: 283-323, 1984. 29. Morley, J. E. Neuroendocrine effects of endogenous opioid peptides in human subjects: a review. Psychoneuroendocrinology 8: 361-379, 1983. 30. Murphy, M. T., J. I. Koenig and J. M. Lipton. Changes in central concentration offl-endorphin in fever. Fed Proc 42: 464, 1983. 31. Pert, C. B., M. J. Kuhar and S. H. Snyder. Opiate receptor: autoradiographic localization in rat brain. Proc Nat! Acad Sci USA 73: 372%3733, 1976. 32. Powanda, M. C. and W. R. Beisel. Hypothesis: Leukocyte endogenous mediator/endogenous pyrogen/lymphocyte-activating factor modulates the development of nonspecific and specific immunity and affects nutritional status. Am J Clin Nutr 35: 762-768, 1982.

1154

33. Robson, L. E., R. W. Foote, R. Maurer and H. W. Kosterlitz. Opioid binding sites of the K-type in guinea-pig cerebellum. Neuroscience 12: 621-627, 1984. 34. Rosendorff, C. and J. J. Mooney. Central nervous system sites of action of a purified leukocyte pyrogen. Am J Physiol 220: 597-603, 1971. 35. Spector, H. J. The central state of the hypothalamus in health and disease. Old and new concepts. In: Handbook of the Hypothalamus, Vol 2, edited by P. J. Morgane and J. Panksepp. New York: Dekker, 1980, pp. 453-517.

AHMED ET AL. 36. Stitt, J. T. Neurophysiology of fever. Fed Proc 40: 2835-2842, 1981. 37. Weber, R. J. and C. B. Pert. Opiatergic modulation of the immune system. In: Central and Peripheral Endorphins: Basic and Clinical Aspects, edited by E. E. MOiler and A. R. Genazzani. New York: Raven Press, 1984, pp. 35-42.