Influence of Maternal Milk on Functional Activation of -Opioid Receptors in Postnatal Rats

Transcription

1 /01/ $3.00 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 296, No. 3 Copyright 2001 by The American Society for Pharmacology and Experimental Therapeutics 3164/ JPET 296: , 2001 Printed in U.S.A. Influence of Maternal Milk on Functional Activation of -Opioid Receptors in Postnatal Rats ROBIN JAMES GOODY and IAN KITCHEN Pharmacology Group, School of Biomedical and Life Sciences, University of Surrey, Guildford, Surrey, United Kingdom Received July 27, 2000; accepted November 10, 2000 This paper is available online at ABSTRACT Weaning rat pups at day 21 activates a -opioid receptor that mediates swim-stress-induced analgesia (swim SIA). We have addressed the possibility that removal of maternal milk is the stimulus for the weaning-induced -receptor activation by studying the effect of lactating and nonlactating surrogate mothers and two milk substitutes (casein-rich and casein-free) on opioid receptor control of swim SIA. The -receptor antagonist naltrindole (1 mg/kg) significantly antagonized swim SIA in 25-day-old weaned rats, in rats provided with a nonlactating surrogate, and those provided with casein-free milk substitute. Stress-induced analgesia (SIA) can be divided into opioid and nonopioid types (on the basis of naloxone reversibility) and is dependent on the nature of the stressful stimulus (Bodnar, 1990). SIA has been observed in young rats in response to warm water (20 C) swimming. Swimming periods of 3 min produce an opioid form of SIA that is dependent on the hypothalamic-pituitary-adrenal axis (Jackson and Kitchen, 1989). Studies with naloxone and the -selective antagonists ICI and naltrindole have shown that the opioid form of SIA is mediated by -receptors in 20-day-old rat pups, but by day 25 it is predominantly -sites that operate this behavior (Jackson and Kitchen, 1989; Kitchen and Pinker, 1990). We have shown that weaning plays a critical role in this transition. Swim SIA responses in pups weaned at day 21 are partially reversed following pretreatment with the -selective antagonist naltrindole, indicating an active -opioid receptor component. However, pups housed with their mother until 25 days beyond the normal day of weaning continue to exhibit -receptor-mediated swim SIA, insensitive to naltrindole (Muhammad and Kitchen, 1993). Delay of weaning was able to halt the - to -opioid receptor transition in the mediation of swim SIA by 5 to 10 days (Muhammad and Kitchen, 1993). The molecular mechanisms underlying the effect of weaning have been addressed by homogenate binding and autoradiography and these have suggested weaning activates a subpopulation of -receptors This work was supported by the Biomed 2 program of the European Community: EC BMH4-CT ABBREVIATION: SIA, stress-induced analgesia. 744 Naltrindole had no effect in nonweaned pups, pups given a casein-rich substitute, or in pups from litters provided with a lactating surrogate from day 21 to day 25. Weaning-induced activation of -receptors involved in mediating swim SIA appears to be dependent on the loss of dietary casein, which is known to produce peptide fragments that can exert opioid activity. The data suggest that exposure to exogenous opioid peptides can influence the ontogenesis of - and -opioid receptors. predominantly located in cortical structures (Kitchen et al., 1995). The precise physiological mechanisms underlying this receptor transition remain to be elucidated and we have now examined the behavioral trigger for this weaning-induced effect. We report here that provision of casein-rich milk for pups is the critical component in maintaining -receptormediated swim SIA up to day 25. Materials and Methods Animals and Experimental Conditions. Wistar albino rats (University of Surrey bred) of either sex were used and maintained in litters of 10 pups. Animals were cross-fostered at birth and allowed free access to water and rat chow. Other conditions and assignment to experimental groups were as previously described (Muhammad and Kitchen, 1993). Weaned, nonweaned, lactating surrogate mother, nonlactating surrogate mother, casein-rich, casein-free milk substitute, and casein-spiked, casein-free milk substitute experimental groups were studied for SIA responses at 25 days of age. Weaned, nonweaned, lactating surrogate mother, and caseinrich milk substitute groups were also tested for SIA responses at 40 days of age. Weaned groups were separated from their mother at day 21 (where day of birth is day 0) and housed in groups of 10 without their mother. In groups where animals were not weaned (nonweaned), the mother remained with the litter at all times except during drug administration, swimming, and analgesic testing. Litters provided with surrogate mothers were separated from their birth mother at day 21 and immediately replaced by a lactating or nonlactating surrogate. Lactating surrogate mothers were taken from a litter

2 containing the same number of pups that had been delivered on the same day and were housed with pups in the same manner as birth mothers. Nonlactating surrogates had not given birth to or been housed with a litter for a minimum of 60 days before transfer. All groups provided with milk substitute had their birth mother removed at day 21 and replaced by a bottle containing milk formula in addition to continued provision of solid rat chow and water. Casein-rich milk substitute (Welpi; Petlife International Ltd., Bury St. Edmunds, UK) was made up as 1 part formula to 2 parts water and provided to weaned pups ad libitum from day 21 until the completion of analgesia studies on either day 25 or day 40. Caseinfree formula milk was provided by Special Diets Services, Cambridge, UK. This custom formula had overall protein levels maintained by an increase in whey protein composition and was made up and provided as described for the casein-rich formula. Casein-spiked milk formula was comprised of casein-free milk formula spiked with 20% w/w casein (Sigma, Poole, UK) to provide casein at levels approximating those in the casein-rich milk formula and was provided to pups in the same way as described for casein-rich and casein-free milk formulas. All rats were maintained at 21 1 C in a constant 12-h light/dark cycle (lights on at 7:00 AM) and experimental procedures were carried out in a quiet, windowless, air-controlled laboratory. All procedures were carried out between 11:00 AM and 4:00 PM to minimize diurnal variation in nociceptive responses. All pups were weighed before 9:00 AM on the day of swim-stress studies for determination of drug dosing volumes. Swim-Stress Procedures and Analgesic Testing. Animals were divided into treatment groups so that analgesic testing took place for saline and drug-treated rats on separate days to minimize interday and interlitter variation. Analgesic responses were recorded immediately before drug administration by use of the tail immersion test at 50 C as described previously (Kitchen et al., 1984). Naltrindole (1 mg/kg) was administered in 0.9% saline i.p. in dose volumes no greater than 0.2 ml, 10 min before swimming stress. Animals were stressed by placing them individually in a plastic water tank ( cm deep) containing water at 20 1 C for a 3-min period as previously described for young rats (Jackson and Kitchen, 1989). At the end of the swimming period, rats were removed from the water, dried, and returned to the home cage before subsequent analgesic testing. Tail immersion responses were measured at 1, 5, 10, 15, and 30 min following swimming stress. Drugs and Statistical Procedures. Naltrindole was purchased from Sigma and 0.9% w/v saline was obtained from Phoenix Pharmaceuticals (Gloucester, UK). Nociceptive treatment groups were compared using repeated measures ANOVA and post hoc comparisons made by Duncan s multiple range test using the Superanova software package for the Macintosh. Repeated measures ANOVA was carried out to determine an overall effect between groups. Where significant differences were observed by repeated measures ANOVA individual post hoc comparison were then carried out across treatment groups at individual time points. Results Naltrindole (1 mg/kg) had no significant effect on nociceptive latencies in unstressed 25-day-old animals from any experimental groups studied (Figs. 1 and 2). The magnitude of SIA in response to swimming was similar in all groups (latency 5 7 s; Figs. 1 and 2). Naltrindole (1 mg/kg) did not significantly alter swim SIA in 25-day-old nonweaned animals or in pups from litters provided with a lactating surrogate mother (Fig. 1, b and d). In weaned pups and those housed with a nonlactating surrogate, naltrindole significantly reduced swim SIA (Fig. 1, a and c). Swim SIA was unaffected by naltrindole (1 mg/kg) when the birth mother was replaced at day 21 by a bottle containing casein-rich milk substitute (Fig. 2a). However, swim SIA Maternal Milk and -Opioid Receptor Ontogeny 745 Fig. 1. Effect of naltrindole (1 mg/kg) on analgesia induced by a 3-min swim in 25-day-old weaned rats (a), 25-day-old nonweaned rats (b), 25- day-old pups transferred to a nonlactating surrogate at day 21 (c), and 25-day-old pups transferred to a lactating surrogate at day 21 (d). Values represent mean S.E.M. for 8 to 10 animals. Saline-injected, unstressed (E); saline-injected, swim-stressed (F); naltrindole (1 mg/kg), unstressed ( ); naltrindole (1 mg/kg), swim-stressed (f). Pre, pretest response 10 min before swim stress. *P 0.05 swim-stressed versus appropriate unstressed control; P 0.05 saline-treated, swim-stressed versus naltrindole-treated, was significantly reduced following naltrindole (1 mg/kg) pretreatment in 25-day-old pups provided with a bottle of caseinfree milk formula from day 21 to day 25 (Fig. 2b). Analgesic responses in pups provided with a bottle of casein-spiked milk formula were unaffected by pretreatment with naltrindole (1 mg/kg), corresponding to nonweaned, lactating surrogate mother, and casein-rich milk substitute-fed groups (Fig. 2c). Swim SIA was significantly antagonized in 40-day-old weaned and lactating surrogate-switched rats and completely blocked in nonweaned rats by naltrindole (1 mg/kg) pretreatment, whereas swim SIA in 40-day-old casein-rich milk-fed rats was unaffected by naltrindole (1 mg/kg) (Fig. 3). At 25 days of age weaned animals and those provided with a nonlactating surrogate had significantly lower body weights than nonweaned groups, those provided with a lactating surrogate, and those fed milk substitutes. Weaned rats also had the lowest body weights at 40 days of age (Table 1). Discussion In common with our previous studies of delayed weaning, swim SIA was partially blocked by naltrindole (1 mg/kg) in 25-day-old weaned rats but was unaffected in nonweaned pups (Muhammad and Kitchen, 1993). This dose of naltrindole is selective for antagonism of -sites at this age (Kitchen and Pinker, 1990; Crook et al., 1992). Previous studies in this laboratory have demonstrated complete reversal of the swim SIA response in weaned animals by pretreatment with 5

3 746 Goody and Kitchen Fig. 2. Effect of naltrindole (1 mg/kg) on analgesia induced by a 3-min swim in 25-day-old casein-rich milk-fed rats (a), 25-day-old casein-free milk-fed rats (b), and 25-day-old casein-spiked milk-fed rats (c). Values represent mean S.E.M. for 8 to 10 animals. Saline-injected, unstressed (E); saline-injected, swim-stressed (F); naltrindole (1 mg/kg), unstressed ( ); naltrindole (1 mg/kg), swim-stressed (f). Pre, pretest response 10 min before swim stress. *P 0.05 swim-stressed versus appropriate unstressed control; P 0.05 saline-treated, swim-stressed versus naltrindole-treated, Fig. 3. Effect of naltrindole (1 mg/kg) on analgesia induced by a 3-min swim in 40-day-old weaned rats (a), 40-day-old nonweaned rats (b), 40- day-old rats transferred from mother to a lactating surrogate at 21 days (c), and 40-day-old casein-rich milk-fed rats (d). Values represent mean S.E.M. for 8 to 10 animals. Saline-injected, unstressed (E); saline-injected, swim-stressed (F); naltrindole (1 mg/kg), unstressed ( ); naltrindole (1 mg/kg), swim-stressed (f). Pre, pretest response 10 min before swim stress. *P 0.05 swim-stressed versus appropriate unstressed control; P 0.05 saline-treated, swim-stressed versus naltrindoletreated, TABLE 1 The effect of weaning, maternal surrogates, and milk feeding on the body weight of rat pups at 25 and 40 days of age Body weight of animals on the day of performing swim-stress studies. Weaned groups were separated from their mother at postnatal day 21. Nonweaned litters were housed with birth mother throughout experiments. Surrogate mother groups had their birth mother removed at day 21 and replaced with either a lactating or nonlactating surrogate mother; casein-rich and casein-free milk-substitute-fed litters were provided with a bottle of the respective milk formula on removal of birth mother at day 21. Casein-spiked formula comprised casein-free formula with 20% w/w casein added. Values are means S.E.M. for 16 to 40 determinations. Experimental Group Bodyweight (g) 25 Days 40 Days Weaned Nonweaned a Lactating surrogate a,b Nonlactating surrogate Casein-rich milk-fed a,b,c Casein-free milk-fed a,b,c Casein-spiked casein-free a,b,c Effects of weaning, surrogates, and milk-substitutes: significant difference from corresponding experimental group a P 0.01 in comparison with weaned pups; b P 0.01 in comparison with pups provided with a nonlactating surrogate; c P 0.01 in comparison with nonweaned pups (one-factor ANOVA followed by Duncan s new multiple range post hoc test). mg/kg naltrindole, suggesting exclusive -mediation of this response. However, at this dose naltrindole may exert some -antagonist activity and therefore the lower concentration was used for the current experiments to address only the -receptor component (Muhammad and Kitchen, 1993). Replacement of the mother with a lactating surrogate produced swim SIA responses similar to those of nonweaned pups, i.e., unaffected by naltrindole. In contrast, replacement with a nonlactating adult female resulted in responses comparable to those of weaned groups. Despite the likely absence of milk, pups are still able to suckle, although previous studies have demonstrated pups over 2 weeks of age choose to suckle, at a lactating nipple over nonlactating (Kenny et al., 1979; Blass, 1990). Our results therefore indicate that the opioid receptor switch in the mediation of swim SIA responses is not influenced by the psychological stimulus of maternal deprivation but is due to reduction in the suckling stimulus and/or dietary modification. Replacement of the mother at the time of weaning with a casein-rich milk formula prevents the - to -receptor switch in mediation of swim SIA at 25 days of age, whereas pups fed a similar composition casein-free formula exhibit -mediation. These findings suggest loss of suckling is not the stimulus for the - to -receptor transition observed in weaned and nonlactating surrogate groups. Furthermore, our results clearly show that the presence of casein in the milk substitute is the regulatory component for maintenance of -receptor mediation of swim SIA. Spiking of casein into the caseinfree formula results in maintained -mediated swim SIA, confirming casein is the critical component. Because confirmation of -receptor mediation was not carried out in caseintreated groups it is not possible to be 100% sure that casein treatment does not impact on other receptor systems. Although there is the possibility that dietary alteration has a nonopioid receptor component we observed no difference in the magnitude of non- -mediated analgesia nor in the duration of analgesia. It therefore seems likely that what is retained is indeed -receptor mediated. Casein is a milk-derived protein that has been shown to

4 contain peptide fragments that can exert opioid activity (Brantl et al., 1979; Brantl, 1984; Teschemacher et al., 1997). The most widely studied group of these peptides is the -casomorphins that have been identified in bovine, ovine, and human -casein (Teschemacher et al., 1997). Interestingly, -casomorphins had previously been undetected in -casein from rat and mouse (Teschemacher et al., 1990). Penetration of the blood-brain barrier is likely to be less restricted in 25-day-old rat pups (Ermisch et al., 1985; Reid and Hubbell, 1994; Teschemacher et al., 1997) and it is therefore possible that milk-derived -casomorphin-like peptides could elicit centrally mediated behavioral effects (Blass and Blom, 1996). The effect of casein upon opioid receptor development persists to day 40 when the blood-brain barrier should be fully developed in the rat and we must therefore entertain the possibility that the effect of these peptides may be exerted locally or peripherally rather than centrally. An alternative possibility is that casein contains a novel component distinct from the -casomorphins that is responsible for the regulation of opioid receptor development. Protein and carbohydrate levels in rat milk do not change significantly between days 5 and 20 after birth of the litter although fat levels decrease over this time (Godbole et al., 1981). However, by 25 days of age the time spent drinking and feeding by pups reaches adult levels and there is very little suckling (Thiels et al., 1990) evidenced by the low quantities of milk found in rat stomach at this age (Henning, 1981). The absolute amount of casein ingested from around the second week when they begin to eat solid food is thus likely to decrease and by 40 days of age rat pups are likely to be totally reliant upon rat chow diet and water as a possible food source. The critical component in casein is therefore diminishing after around 14 days of age. The normal process of weaning takes pups from a high protein-high fat diet to a high carbohydrate-low protein diet (Jenness, 1974). The groups used in these experiments demonstrated significant differences in body weight at 25 days. Both weaned and nonlactating surrogate groups had lower body weights than nonweaned, lactating surrogate, and milk substitute-fed groups. There was no difference in body weight observed between animals fed on the three milk substitutes, which suggests the weaning-induced - to -transition is not the result of a gross nutritional change. At the normal age of weaning (21 days) around 75% of the total adult levels of -receptors are developed in the rat brain, whereas - and -sites are fully mature (Spain et al., 1985; McDowell and Kitchen, 1986; Kitchen et al., 1992). Membrane homogenate binding and autoradiographic studies have demonstrated that weaning stimulates the development of a subpopulation of -receptors recognized by [ 3 H]deltorphin I. These differences may indicate the expression of a -receptor subtype encoded for by a different gene, or that a subpopulation develops after weaning, which couples to a different G protein and that these coupled sites are recognized by [ 3 H]deltorphin I (Kitchen et al., 1995; Kelly et al., 1998). This subpopulation of receptors may be responsible for the -receptor-mediated swim SIA observed in weaned, nonlactating surrogate mother, and casein-free milk substitutefed groups. By day 30 the effect of delayed weaning is lost because naltrindole is able to reverse swim SIA regardless of extended housing with their mother (Muhammad and Kitchen, Maternal Milk and -Opioid Receptor Ontogeny ). Naltrindole partially reversed swim SIA in 40-day-old weaned and lactating surrogate-switched pups and completely reversed the response in 40-day-old nonweaned pups but swim SIA in animals provided with casein-rich milk substitute was unaffected by naltrindole. This suggests provision of dietary casein in a milk formula is able to delay the weaning transition beyond that achieved by extended housing with the mother. It therefore seems possible to provide prolonged alteration to brain opioid systems by dietary manipulation. Thiels et al. (1990) have demonstrated that suckling can continue up to day 34 although self-weaning can occur long before this and would suggest nonweaned pups would no longer be obtaining significant quantities of milk from their mother at day 40. The complete reversal of the swim SIA response in nonweaned 40-day-old rats by naltrindole compared with partial blockade in weaned and lactating surrogate groups may be linked to differences in body weight and therefore nutritional intake between these experimental groups. Nonweaned pups exhibit the highest body weights at 40 days of age and this may influence the complete transition from - to -receptor mediation of swim SIA responses observed in these animals. In conclusion, removal of casein from the neonatal rat diet at the time of weaning results in a - to -transition in the mediation of swim SIA. This transition can be delayed up to at least 40 days of age by continued dietary provision of casein-rich milk substitute but not by extended housing with the mother. References Blass EM (1990) Suckling: Determinants, changes, mechanisms, and lasting impressions. Dev Psychol 26: Blass EM and Blom J (1996) -Casomorphin causes hypoalgesia in 10-day-old rats; evidence for central mediation. Pediat Res 39: Bodnar RJ (1990) Effects of opioid peptides on peripheral stimulation and stressinduced analgesia in animals. Crit Rev Neurobiol 6: Brantl V (1984) Novel opioid peptides derived from human -casein: Human -casomorphins. Eur J Pharmacol 106: Brantl V, Teschemacher H, Henschen A and Lottspeich F (1979) Novel opioid peptides derived from casein ( -casomorphins): I. Isolation from bovine casein peptone. Hoppe-Seyler s Z Physiol Chem 360: Crook TJ, Kitchen I and Hill RG (1992) Effects of the -opioid receptor antagonist naltrindole on antinociceptive responses to selective -agonists in post-weanling rats. Br J Pharmacol 107: Ermisch A, Ruhle H-J, Landgraf R and Hess J (1985) Blood-brain barrier and peptides. J Cereb Blood Flow Metab 5: Godbole VY, Grundleger M, Pasquine T and Thene S (1981) Composition of rat milk from day 5 to 20 of lactation and milk intake of lean and preobese Zucker pups. J Nutr 111: Henning SJ (1981) Postnatal development: Coordination of feeding, digestion and metabolism. Am J Physiol 241:G199 G214. Jackson HC and Kitchen I (1989) Swim-stress-induced antinociception in young rats. Br J Pharmacol 96: Jenness R (1974) Biosynthesis and composition of milk. J Invest Dermatol 63: Kelly MDW, Hill RG, Borsodi A, Toth G and Kitchen I (1998) Weaning-induced development of -opioid receptors in rat brain: Differential effects of guanine nucleotides and sodium upon ligand-receptor recognition. Br J Pharmacol 125: Kenny JT, Stoloff ML, Bruno JP and Blass EM (1979) The ontogeny of preference for the nutritive over nonnutritive suckling in albino rats. J Comp Physiol Psychol 93: Kitchen I, Kelly M and Hunter JC (1992) Ontogenic studies with [ 3 H]-CI-977 reveal early development of -opioid sites in rat brain. Mol Neuropharmacol 2: Kitchen I, Leslie FM, Kelly M, Barnes R, Crook TJ, Hill R, Borsodi A, Toth G, Melchiorri P and Negri L (1995) Development of delta-opioid receptor subtypes and the regulatory role of weaning: Radioligand binding, autoradiography and in situ hybridisation. J Pharmacol Exp Ther 275: Kitchen I, McDowell J, Winder C and Wilson J (1984) Low-level lead exposure alters morphine antinociception in neonatal rats. Toxicol Lett 22: Kitchen I and Pinker SR (1990) Antagonism of swim-stress-induced antinociception by the -opioid receptor antagonist naltrindole in adult and young rats. Br J Pharmacol 100: McDowell J and Kitchen I (1986) Ontogenesis of -opioid receptors in rat brain using [ 3 H][D-Pen 2, D-Pen 5 ]enkephalin as a binding ligand. Eur J Pharmacol 128: Muhammad BY and Kitchen I (1993) Effect of delayed weaning on opioid receptor

5 748 Goody and Kitchen control of swim stress-induced antinociception in the developing rat. Br J Pharmacol 109: Reid LD and Hubbell CL (1994) An assessment of the addiction potential of the opioid associated with milk. J Dairy Sci 77: Spain JW, Roth BL and Coscia J (1985) Differential ontogeny of multiple opioid receptors (,, and ). J Neurosci 5: Teschemacher H, Brantl V, Henschen A and Lottspeich F (1990) -Casomorphins- -casein fragments with opioid activity: Detection and structure, in -Casomorphins and Related Peptides (Nyberg F and Brantl V eds) pp 9 14, Fyris-Tryck AB, Uppsala, Sweden. Teschemacher H, Koch G and Brantl V (1997) Milk protein-derived opioid receptor ligands. Biopolymers 43: Thiels E, Alberts JR and Cramer CP (1990) Weaning in rats: II. Pup behaviour patterns. Dev Psychobiol 23: Send reprint requests to: Professor I. Kitchen, School of Biomedical and Life Sciences, University of Surrey, Guildford, Surrey, GU2 7XH, UK. i.kitchen@surrey.ac.uk