Subcellular compartmentation of opioid receptors: modulation by enkephalin and alkaloids

Delta-opioid receptor endocytosis in spinal cord after dermenkephalin activation

BACKGROUND

The delta(delta)-opioid receptors belong to the G protein-coupled receptors and in vitro studies have shown that delta-opioid receptors undergo an internalization process in response to agonist stimulation. The immediate consequence is the disappearance of receptors from the plasma membrane. This adaptation process reveals the cell's capacity to desensitize after a strong agonist stimulus. This process, if it occurs in vivo, could contribute to the tolerance phenomenon observed after opiate treatment. To study the mechanisms underlying regulation of the delta-opioid receptors in vivo, the effects of an application of the drug dermenkephalin, a potent and selective agonist of the delta-opioid receptor, were analysed in the rat spinal cord.

RESULTS

Using immunocytochemistry and electron microscopy, we observed in control rats that membrane labelling was strictly localized at the interface between two neurites. Fifteen minutes after dermenkephalin stimulation, the plasma membrane labelling was associated with invaginated areas. Thirty minutes after stimulation, labelled vesicles were found in the cytoplasm confirming the internalization process.

CONCLUSIONS

The present findings support the view that delta-opioid receptors are internalized in response to prolonged exposure to dermenkephalin in vivo and confirm the presynaptic localization of delta-opioid receptors in the dorsal horn of the rat spinal cord.

Agonist-induced desensitization and down-regulation of delta opioid receptors alter the levels of their 125I-beta-endorphin cross-linked products in subcellular fractions from NG108-15 cells

The delta opioid binding sites in subcellular fractions from NG108-15 cells were characterized with respect to their relative molecular size and levels under conditions of receptor adaptation. 125I-beta-Endorphin was cross-linked to preparations enriched in plasma membranes (P20), nuclear membranes or nuclear matrices. Five cross-linked bands appear in all subcellular fractions. The largest molecular size reaction product in nuclear matrix preparations (approximately 72 kDa) differed from that in the other two fractions-(approximately 83 kDa). Immunoblot analyses with an antibody to the delta opioid receptor gave a P20 band pattern similar to that for the corresponding cross-linked products. To determine which cross-linked products in P20 are glycoproteins, labeled membranes were solubilized and purified by wheat germ agglutinin chromatography. The absence of a approximately 36 kDa band after purification suggests that this product is not a glycoprotein. The remaining four bands were present in N-acetyl-D-glucosamine eluates, although their % distribution changes in favor of the largest molecular size band (approximately 83 kDa). Immunoblotting of the eluate gave a single diffuse band at approximately 73 kDa, suggesting the native glycoprotein has a molecular size in the 70-80 kDa range. Etorphine-induced desensitization of cell surface receptors increased the amount of some cross-linked products associated with nuclear membranes. The same treatment did not affect the relative density of the four larger molecular size bands in P20, but increased the density of the approximately 26 kDa product two fold. Etorphine-induced down-regulation evoked an elevation of cross-linked products in nuclear matrix preparations, while all band densities of P20 were diminished. These results suggest that nuclear matrix associated opioid binding sites represent internalized, truncated forms of the glycosylated delta opioid receptor found in P20.

Electron microscopic localization of photoaffinity-labelled delta opioid receptors in the neostriatum of the rat

The distribution of delta opioid receptors, selectively labelled in vitro with the photoaffinity probe monoiodo azido-DTLET ([D-Thr2,pN3Phe4, Leu5]enkephaly-Thr6), was analyzed by light and electron microscopic radioautography in sections from rat neostriatum. Preliminary experiments indicated that up to 65% of specific 125I-azido-DTLET binding to rat striatal sections was still detectable following prefixation of the brain with 0.5% glutaraldehyde. These experiments also showed that up to 20-30% of the specifically bound radioactivity was covalently linked following ultraviolet irradiation and was thereby retained in tissue during subsequent postfixation and dehydration steps. Accordingly, the topographic distribution of the covalently attached azido-DTLET molecules was similar to that seen in fresh frozen sections and characteristic of that previously described for delta sites. Light and electron microscopic examination of the label in prefixed, striatal sections irradiated with ultraviolet light revealed that a significant proportion of specifically bound 125I-azido-DTLET molecules was intraneuronal. Specifically, 16% of the labelled binding sites were found in dendrites, 12% in perikarya and 4% in axon terminals. These results suggest that an important proportion of delta opioid binding sites labelled in the neostriatum correspond to receptors that are undergoing synthesis, transport and/or recycling. They also imply that a major fraction of delta sites are associated with intrastriatal neurons, as opposed to afferent axons. Approximately 44% of the labelled binding sites were associated with neuronal plasma membranes. Although most of these were found at the level of axodendritic (20%) and dendrodendritic (7%) appositions, comparison of the labelling incidence of these two compartments with their frequency of occurrence in tissue suggested that delta sites are fairly widely dispersed along neuronal plasma membranes. Only a small proportion (smaller than that of mu or kappa sites labelled in the same region) was associated with synaptic specializations. These results support the concept that delta receptors correspond to molecular entities that are distinct from mu and kappa sites and suggest that delta ligands act primarily nonjunctionally on the plasma membrane of striatal neurons.

Ultrastructural localization of enkephalin-like immunoreactivity in developing rat cerebellum

Methionine enkephalin, an endogenous opioid peptide, participates in the regulation of growth in the developing brain. In the present study, enkephalin-like immunoreactivity was localized in the cerebellum of developing and adult rats by immunoelectron microscopy. In 10-day-old animals, enkephalin-like immunoreactivity was found in the somata of proliferating, migrating and differentiating neural cells, and was associated with the plasma membrane, microtubules, filaments, mitochondria, endoplasmic reticulum and nuclear envelope. Both neurons and glia in the cerebellum of the preweaning rat displayed a similar profile of immunoreactivity. Reaction product was also detected in the dendrites and dendritic spines of Purkinje cells where it was concentrated in postsynaptic densities. The majority of internal granule neurons in 10-day-old animals were not immunoreactive, nor were axons, glial processes and postsynaptic elements (with the exception of mossy fiber terminals). At weaning (Day 21), enkephalin-like immunoreactivity was confined primarily to the somata of Purkinje, basket and stellate neurons, and to Purkinje cell dendrites and synaptic spines. Adult rats (day 75) exhibited no enkephalin-like immunoreactivity. These results establish that enkephalin or an enkephalin-like substance can be detected during the ontogeny of both neurons and glia in the cerebellar cortex, and appears to be associated with certain structural elements.

Differentiation of human neuroblastoma cells: marked potentiation of prostaglandin E-stimulated accumulation of cyclic AMP by retinoic acid

Neuroblastoma cells in culture contain low levels of cyclic AMP, a second messenger which plays a major role in neuronal maturation. In this study, human neuroblastoma cells, SK-N-SH-SY5Y, were induced to differentiate by treatment with either nerve growth factor (50 ng/ml), retinoic acid (10 microM), dibutyryl cyclic AMP (1 mM), or 12-O-tetradecanoylphorbol-13-acetate (0.1 microM), and the ability of several neurotransmitters or hormones to stimulate adenylyl cyclase was tested. Although all four differentiation factors caused morphological changes towards a neuronal phenotype, only retinoic acid dramatically enhanced cyclic AMP accumulation, specifically upon stimulation with prostaglandin E1 (PGE1). PGE2 was also active, but less potent, than PGE1, whereas the other cyclic AMP-stimulating agents tested were largely unaffected. Further, the rapid desensitization of the PGE1-cyclic AMP response observed in control cells after 20 min of PGE1 exposure did not occur in retinoic acid-treated cells, and the EC50 values for PGE1 were reduced from approximately 240 to 14 nM after retinoic acid treatment. The increased sensitivity to PGE was associated with an increase of high-affinity PGE1 binding sites, whereas the Gs coupling proteins and adenylyl cyclase were not measurably affected. A similar enhancement of the PGE1-cyclic AMP response by retinoic acid was also observed in two additional human neuroblastoma cell lines tested, Kelly and IMR-32, suggesting that up-regulation of the prostaglandin response by retinoic acid is common among neuroblastoma cells.(ABSTRACT TRUNCATED AT 250 WORDS)