Endogenous Opiates and Behavior (a lot of Theorie)

This paper is the twenty-ninth consecutive installment of the annual review of research concerning the endogenous opioid system, now spanning thirty years of research. It summarizes papers published during 2006 that studied the behavioral effects of molecular, pharmacological and genetic manipulation of opioid peptides, opioid receptors, opioid agonists and opioid antagonists. The particular topics that continue to be covered include the molecular-biochemical effects and neurochemical localization studies of endogenous opioids and their receptors related to behavior (Section 2), and the roles of these opioid peptides and receptors in pain and analgesia (Section 3); stress and social status (Section 4); tolerance and dependence (Section 5); learning and memory (Section 6); eating and drinking (Section 7); alcohol and drugs of abuse (Section 8); sexual activity and hormones, pregnancy, development and endocrinology (Section 9); mental illness and mood (Section 10); seizures and neurological disorders (Section 11); electrical-related activity and neurophysiology (Section 12); general activity and locomotion (Section 13); gastrointestinal, renal and hepatic functions (Section 14); cardiovascular responses (Section 15); respiration and thermoregulation (Section 16); and immunological responses (Section 17).

2. Endogenous Opioids and Receptors

2a. Molecular-biochemical effects

This sub-section will review current developments in the molecular and biochemical characteristics of opioid peptides and receptors by subtypes: mu agonists and receptors (2a-i), delta agonists and receptors (2a-ii), kappa agonists and receptors (2a-iii), and OFQ/N and the ORL-1 receptor (2a-iv).

2a-i. Mu agonists and receptors 

Endocytosis of the MOR-1D splice variant as well as DOR and the CB1 receptor is mediated by an agonist-independent and constitutive PLD2 activation (604). Separation of MOR desensitization and internalization effects was demonstrated with endogenous receptors in primary neuronal LC cultures (40). Exons 11 and 1 promoters of the MOR gene were characterized in transgenic mice (1249). The splice variants of MOR, SV1 and SV2 do not exhibit binding to [3H] diprenorphine (212). Five single nucleotide polymorphisms were identified for the MOR promoter, and no differences in construct activity were found in control and morphine-treated animals (297). MOR-effector coupling and trafficking occurred in DRG neurons with DAMGO producing greater internalization in MOR/partial differential opioid receptors (1180). MOR-DOR functional interactions occur through receptor-G (i1) alpha fusion (1051). The poly C binding protein 1 is a regulator of the proximal promoter of the mouse MOR gene (716). There is interplay between Sps and poly C binding protein 1 on MOR gene expression (960). The neuron-restrictive silencer factor interacts with Sp3 to synergistically repress the MOR gene (577). Mitochondrial damage decreases MOR, but not DOR function in neuronal SK-N-SH cells (941). Diffusion of MOR at the surface of human neuroblastoma SH-SY5Y cells is restricted to permeable domains (990). Differences in the intracisternal A-particle element in the 3′ noncoding region of the MOR gene in CXBK mice appear to cause this relatively insensitive phenotype (446). Although 14-methoxymetopon displayed similar binding affinities for multiple splice variants of the MOR gene, its potency varied widely for the same splice variants (713). MOR activation of ERK ½ is GRK3 and arrestin dependent in striatal neurons (712). Introduction of a hydroxyl group in position 2 of the cyclohexyl residue of spiropiperidine decreased the binding affinity of mu receptors and OFQ/N (177). Morphine displays PKC-dependent and DAMGO displays GRK2-dependent mechanisms of MOR desensitization in human embryonic kidney 293 cells (529). Null and operational models of mu opioid binding in the mouse vase deferens revealed that DAMGO and DALDA were full agonists while morphine and endomorphin derivatives acted as partial agonists (967). There is direct nose-to-brain transfer of morphine after nasal administration to rats (1216). Transport is not rate-limiting in morphine glucuronidation in the single-pass perfused liver preparation (291). Greater in vitro inhibition of M6G relative to M3G formation from morphine occurred following treatment with (R)- and (S)-methadone and structurally related opioids (804). M6G was identified in chromaffin cell secretory granules (411). Mu opioid receptors are activated by Vitex agnus-castus methanol extracts (1204). Binding of [(35) S] GTPgammaS is stimulated by endomorphin-2 and morphiceptin analogs (343). An aequorin luminescence-based calcium assay gives pharmacologically-relevant data for mu and delta opioid agonists without involving radioactivity or animal tissues (344). The latter assay also indicated that D-1-Nal or D-2-Nal substitutions in position 4 of endomorphin-2 produced mu receptor antagonists, whereas substitution in position 3 produced partial agonists (346). The tripeptides, Tyr-Pro-Ala-NH2 and Tyr-Pro-Ala-OH, which do not bind to mu receptors, are potent inhibitors of endomorphin degrading enzymes in the brain (345). A partial agonistic effect of 9-hydroxycorynatheidine was observed on MOR in the guinea pig ileum (747). Synthesis, radio labeling and receptor binding was accomplished for [3H] [(1S, 2R) ACPC2] endomorphin-2 (569).

MOR, DOR and KOR binding properties were observed for 8-[N-(4′-phenyl)-phenethyl) carboxyamido] analogues of cyclazocine and ethylketocyclazocine (1214). Pharmacokinetics of morphine was slower in flounder than in trout (837). A direct and sensitive analysis of morphine, codeine and other bioactive drugs in human urine was performed by cation-selective exhaustive injection and sweeping micellar electrokinetic chromatography (671). Alkyl chain length altered mu opioid activity of 3, 6-bis {H-Tyr/H-Dmt-NH (CH2) m, n]-2(1H) pyrazinone derivatives (1031). Determination of human urinary opiates occurred with application of poly (methacrylic acid-ethylene gycol dimethacrylate) monolith micro extraction coupled with capillary zone electrophoresis (1207). Urinary excretion of morphine and codeine occurred following the administration of single and multiple doses of opium preparations prescribed in Taiwan as “brown mixture” (680). There are natural heroin impurities derived from tetrahydrobenzylisoquinoline alkaloids (1128). [Sar2] endomorphin-2 was almost equipotent to the parent peptide in mu opioid receptor binding, and was highly resistant to enzymatic degradation (512). Morphine-3-O-propionyl-6-O-sulfate had four times greater affinity than morphine at the MOR (248). Mu opioid binding can be detected in a competitive displacement assay using a beta-imager (916). Identification of potent phenyl imidazoles such as 4-aminocarbonyl-2, 6-dimethyl-Phe as opioid receptor agonists was completed (133). A class of 4-substituted-8-(2-phenyl-cyclohexyl)-2-8-diaza-spiro[4,5]decan-1-one inhibitors was designed achieving antagonist selectivity against the mu opioid and ORL1 receptors (15) that show improved metabolic stability (16) and superior pharmacological and pharmacokinetic properties (17). Chimeric peptides containing a mu opioid receptor ligand and an ORL-1 receptor ligand, Ac-RYYRIK-amide were synthesized (560). Derivatives of 14-aminomorphinones with substitution in the aromatic ring of cinnamoylaminomorphinones and codeinones produced potent mu agonists (844).

A novel mu opioid antagonist was created by replacement of the N-terminal tyrosine residue in opioid peptides containing 3-(2, 6-dimethyl-4-carbamoylphenyl) propanoic acid (Dcp) results in a novel mu opioid antagonist (701). Lysine at the C-terminus of the Dmt-Tic opioid pharmacophore produced new lead compounds in the formation of opioid peptidomimetics (57). Multiple ligands from Dmt-Tic and morphinan pharmacophores produced agonists with MOR, KOR and DOR properties (836). Whereas maximal stimulation of [35S]GTPgammaS binding decreased in canine thalamic and spinal cord homogenates for mu > ORL1 > kappa > delta opioid compounds, cortical homogenates showed an affinity order of kappa > ORL1 > delta > mu (650). Morphine increased 5HT, Na (+), K (+)-ATPase activity differentially as a function of adulthood and adolescence (432), and reduced human 5-HT3A receptors in an ondansetron assay (1225). Bivalent ligands containing homo- and heterodimeric pharmacophores at mu, delta and kappa opioid receptors were developed and synthesized in an in vitro assay (895). Potent and highly selective chiral tri-amine and tetra-amine MOR ligands were identified by lead optimization using mixture-based libraries (831). Opioid and CCK receptors also have overlapping pharmacophores required for binding affinity and biological activity (8). QSAR studies were performed upon 4-phenylpiperidine derivatives as mu opioid agonists by a neural network method (1194). A HPLC assay for morphine was developed in small plasma samples (293). Morphine concentrations in bone marrow paralleled plasma morphine concentration for up to 14 days after death in rabbits (178). Unbound concentrations of oxycodone were three times higher in brain than in blood, indicating active influx at the blood-brain barrier (118). An implantable buprenorphine delivery system was characterized using in vitro and in vivo techniques (594). Capillary electrophoresis contributed to the identification of HMOR-3O-glucide and N-oxides in hydromorphone metabolism in humans (58). STW 5 (Iberoglast) binds to intestinal 5-HT, muscarinic M3 and opioid receptors (1038). A member of the heat shock protein 40 family, hlj1, binds to the carboxyl tail of the human MOR (31).

Pre-POMC cDNA from the ostrich pituitary gland was sequenced revealing the positions for gamma-MSH, ACTH, AMSH, gamma-LPH, beta-MSH and BEND (824). PC1/3 and PC/2 gene expression and post-translational endoproteolytic POMC processing is regulated by photoperiod in the seasonal Siberian hamster (465). BEND analysis in the hypothalamus or PAG is not affected by diurnal variation (348). BEND is present in the stallion testis, but appears to be derived from POMC gene expression in the pituitary (1057). Three POMC subtype genes and subsequent peptides were identified in the pars distalis and intermedia of the barfin flounder pituitary (1101). TAN-821 and TAN-1014 have been developed respectively as agonists and antagonists of the putative epsilon receptor (368).

Histamine was released during constant rate infusion of morphine in dogs (423). Cholinergic nicotinic stimulation of endogenous morphine release occurs from lobster nerve cord (1327). Dextrorotatory morphinans inhibited alpha3beta4 nicotinic Ach receptor subunit cRNAs-induced inward currents in the presence of Ach in Xenopus oocytes (639). Myristoylated G proteins (Galphai1 and GalphaoA) were maximally activated by DAMGO, Menk and Lenk, whereas endomorphin-1 and –2 as well as BEND produced strong, but not maximal responses. Morphine, methadone, fentanyl and buprenorphine produced statistically significant activation (983). 3-aminopropionyl substituted fentanyl analogs were synthesized and displayed opioid activity (906). New hybrid derivatives of fentanyl were found to be active at the MOR and I2-imidazoline binding sites (261). Buprenorphine and norbuprenorphine display in vivo glucuronidation as determined by liquid chromatography-electrospray ionization-tandem mass spectrometry (488). An immunalysis microplate ELISA was validated for the detection of buprenorphine and its metabolite norbuprenorphine in urine (778). Opioid disposition occurs in human sweat after controlled oral codeine administration (1003). CYP2D6-dependent formation of morphine does not exclusively explain the central effects of codeine; codeine-6-glucuronide is an additional active moiety (696). Pharmacokinetic modeling of oxycodone in sheep displayed delayed equilibration between brain and blood levels that would be affected by changes in both cerebral blood flow and blood brain barrier permeability (1164). The principal metabolic pathway of oxycodone in humans is CYP3A-mediated N-demethylation, but circulating oxidative and reductive metabolites provide a negligible contribution to central opioid effects (627). The use of PTX-insensitive Galpha mutants revealed that the potency for mu agonists was highest for cells expressing Galpha (i3) and Galpha (o) and lowest with Galpha (i1) and Galpha (i2) (228). NalBzOH displays agonist activity at MOR, DOR and KOR, but not ORL-1 receptors expressed either in a heterologous cell system or in a native environment (862). FK33–824, a mu opioid agonist decreases enzymatic activity of PKC, adenylate cyclase and PKA in porcine granulosa cells (546). The anti-opioid actions of NPVF, a NPFF agonist was observed in a decrease to opioids to voltage-gated (N-type Ca2+ currents and enhancement of muscarinic-induced intracellular Ca2+ release in SH-SY5Y cells as they do in neurons (570).

Naltrexone increased PKCepsilon, ERK and integrin alpha 7 in SH-SY5Y neuroblastoma cells (857). Long-acting naltrexone had a 1-month pharmacokinetic activity in plasma that was proportional to dose and number of treatments (302). The dissociation of [3H] naloxone was four times faster under displacement than under infinite dilution conditions, demonstrating the retention effect of receptors confined in space (1061). Naltrexone release from biodegradable microspheres produced constant rates of release culminating in 80% over 2 months (691). Two novel tripartite codrugs of naltrexone and 6beta-naltrexol with hydroxybupropion were synthesized as potential alcohol abuse and smoking cessation agents (440). Transdermal delivery of 6-beta-naltrexol was enhanced by a codrug linked to hydroxybupropion (588). MOR antagonists were synthesized and evaluated from novel octahydro-1H-pyrido [1, 2-a] pyrazine (633) and from N-substituted trans-3, 4-dimethyl-4-(3-hydroxyphenyl) piperadine (634). The permeation of nalmefene hydrochloride was better across the middle turbinate, posterior septum and superior turbinate mucosa (301).

2a-ii. Delta agonists and receptors 

Knock-in mice expressing fluorescent DOR uncovered G protein-coupled receptor dynamics in vivo (996). Simultaneous activation of DOR and the sensory neuron-specific receptor-4 hetero-oligomer by the mixed bivalent agonist BAM-22 activates the latter, but inhibits the former (132). Morphine and pain-related stimuli enhance cell surface availability of somatic DOR in rat dorsal root ganglia (393). Delta and mu opioid receptors were cloned, heterologously expressed and pharmacologically characterized from the brain of a urodele amphibian, the rough-skinned newt, Taricha granulosa (127). Adenosine A1 and A2 receptor agonists increase c-fos in striatal GABAergic Enk, but not GABAergic DYN striatal neurons as well as increasing striatal PreEnk (553). The enrichment by 18-fold with FACS robustly increased beta-arrestin-1-GFP expression associated with strong human DOR desensitization (10). A new duplicate DOR in zebra fish was characterized (911). Enk and SP are found in reduced concentrations in primary culture striatal neurons relative to adult preparations (330). The mGluRI antagonists, LY367385 and MPEP reduce NMDA-induced expression of the Pro-Enk gene in neocortex, yet enhance AMPA-induced expression of the neocortical Pro-Enk gene (674). Pro-enk was identified in the hypothalamus and striatum using a method allowing selective isolation of neuropeptides of murine brains lacking carboxypeptidase E (268). ProEnk A 119–159 is a stable ProEnk A precursor fragment identified in human circulation (323). Only the C-terminus glucose conjugate of Lenk showed transport by glucose transporters and hPepT1 (1237). Whereas only a few samples of adherent human fetal chromaffin cells expressed Menk early in vitro, almost all of the neurosphere-like colonies appearing later expressed Menk (1321). Human Menk binds to anionic phosphatidylserine in high preference to zwitterionic phosphatidylchlorine in large unilamellar vesicles (586). Lenk binds in a turn confirmation to DOR, but it is not a (1–4) beta-turn (108). Nasal administration of Lenk is facilitated by a thiolated polycarbophil that slows Lenk degradation (82), and reversible lipidization is preferable for the oral delivery of Lenk (1185). 6-N, N-dimethylamino-2, 3-napthalimide was identified as a new environment-sensitive probe in delta- and mu-selective opioid peptides (1149). The novel, orally active, DOR agonist, RWJ-394674 is biotransformed to the potent MOR agonist RWJ-413216 (229). Highly potent and selective phenylmorphan-based inverse agonists of the DOR were developed (1117). AUF-1 is expressed in the developing brain, binds to AT-rich double-stranded DNA, and regulates Enk gene expression (287). Design and synthesis were completed of novel hydrazide-linked bifunctional peptides as delta/mu opioid receptor agonists and CCK1/CCK2 receptor antagonists (641). Three types of latex nanoparticles carrying NTI derivatives were identified as possessing high DOR affinity (453). A new method using HPLC with electrochemical detection was developed allowing for the simultaneous measurement of Menk, Lenk, endomorphin-1 and endomorphin-2 (677).

Selective alkylation of delta-2 opioid receptors occurred following NAC 5′Nti-isothiocyanate in the NAC and ventral caudate; beta-chlornaltrexamine was nonselective for MOR and DOR in the same site (734). Distinct subcellular localization for constitutive and agonist-modulated palmitoylation of the human DOR was described (900). Pertussis-toxin abolished DPDPE-induced inhibition of forskolin-stimulated intracellular cAMP production that was rescued by Galpha (i2), but not Galpha (i3) or Galpha (o) mutants; the former, but not latter mutants co-precipitated with DOR. Long-term DPDPE treatment allowed pertussis toxin-induced elimination of naloxone-induced superactivation of adenylyl cyclase activity, an effect again rescued by Galpha (i2) mutants (1311). DPDPE, acting through Gi-coupled DOR receptors mediate phosphorylation of CPI-17 and MLC20 through preferential activation of the PI3K/ILK pathway (486). Down-regulation of the glutamate transporter EAAC1 occurred following expression and activation of the DOR and its agonist, DPDPE (1240). Genetically-engineered human mesenchymal stem cells produce Menk at augmented higher levels in vitro (1078). Lenk caused membrane currents across lipid membranes through adsorption, transportation and desorption, effects confirmed using fluorescence spectrometry and confocal laser scanning microscopy (687). DADL induces a reversible hibernation-like state in HeLa cells (1152). DADL molecules enter the cytoplasm and nucleus of LNCap cells that are devoid of opioid receptors, and binds to perichromatin fibrils where transcription and early splicing of pre-mRNAs and pre-rMRNAs occur, thereby resulting in decreased transcription and proliferation without apoptosis (59). AVP enhances PAG synthesis and secretion of Enk and BEND, but not DYN (1263).

2a-iii. Kappa agonists and receptors 

KOR activation of p38 MAPK is GRK3- and arrestin-dependent in neurons and astrocytes (141). Netrin-1 signaling regulates de novo protein synthesis of KOR by facilitating polysomal partition of its mRNA (1133). Naloxone and NBNI, but not selective mu or delta antagonists increased, whereas DYN decreased the cell surface level of the human KOR by activation-induced down-regulation and pharmacological chaperone-mediated enhancement (200). KOR affinity in a human embryonic kidney cell system indicated a rank order of cyclazocine, naltrexone, SKF10047, xorphanol, WIN44441, nalorphine, butorphanol, nalbuphine, lofentanil, dezocine, metazocine, morphine, hydromorphone and fentanyl (399). A cDNA encoding KOR in zebra fish has been cloned and characterized (27). GEC1 interacts with the kappa opioid receptor and enhances its expression (188). Big DYN, a 32-amino acid Pro-DYN peptide consisting of DYN A and DYN B showed similar selectivity for human KOR as DYN A, was less selective than DYN A for human MOR, DOR and ORL1, but activated G proteins more potently than DYN A; DYN B was less potent and selective (771). Big DYN and DYN A, but not DYN B causes leakage effects in large unilamellar phospholipid vesicles thereby causing perturbations in the lipid bilayer (490). DYN A inserts its N-terminus into the bilayer of the bicelle, whereas DYN B resides on the surface of the bilayer (673). Bikunin was identified as an endogenous inhibitor of DYN convertase in human cerebrospinal fluid (1077). Although selective opioid agonists developed for mammalian opioid receptors did not fully recognize opioid binding sites in zebra fish brain, DYN A showed good affinities in the nanomolar range (409). Pro-DYN cDNA’s were cloned in eels and tilapia (24).

NalBzOH binding is abolished in triple MOR/KOR/DOR mice, and in vivo changes reflect that of a nonselective opiate drug (235). N-substituted 4beta-methyl-5-(3-hydroxyphenyl)-7alpha-amidomorphans are potent, selective kappa opioid receptor antagonists (171). New C (4)-modified salvinorin A analogues were synthesized as KOR agonists (636). Salvinorin A analogues were also synthesized following effects of configuration at C (2) and substitution at C (18) (73) along with those isolated from Salvia divinorum (1120). Salvinorin A, a selective KOR agonist was used in a combined ligand-based and target-based drug design approach for G-protein coupled receptors (1041). A unique binding epitope was found for salvinorin A, a non-nitrogenous KOR agonist (548). Bioisosteric modification of the C-2 thioacetate isoester of salvinorin A produced a potent and selective KOR agonist (1070). The substitution of a tert-butyl with a cyclobutyl moiety in buprenorphine created a highly selective KOR agonist with low addictive potential and dependence liability (877). 3D-QSAR studies were performed upon orvinol analogs as kappa-opioid agonists (656). The extraction fraction for NAC DA in the mouse is increased following NBNI treatment (186). The application of Schild-analysis to the antagonism of U50488H by NBNI provides pharmacological evidence for KOR in Planaria (934).

2a-iv. OFQ/N and ORL-1 receptor 

OFQ/N binding to ORL-1 receptors triggers internalization of these components into vesicular compartments that is dependent on PKC and occurs selectively for N-type calcium channels (25). OFQ/N, but not partial agonists induce concentration-dependent endocytosis and recycling of the human ORL1 receptor, enhancing up regulation of adenylyl cyclase activity (1059). OFQ/N differentially activates ERK, p38 and JNK MAPK to contribute to potentiations of prostaglandin vasoconstriction after fluid percussion brain injury (38). The induction of Pro-OFQ/N mRNA by cAMP appears to be mediated by a cAMP-response element, histone acetylation and through CREB (1301). The ORL-1 receptor utilizes both G (oA) and G (oB) for signal transduction (1136). OFQ/N and its agonist, Ro64-6198 display great regional similarities in binding assays using [(35) S]-GTPgammaS binding (390). OFQ/N and DYN A analogues with Dmp substituted for N-terminal aromatic residues show ORL-1 and opioid receptor preferences (989). A novel D-proline amide class of spiropiperidines was developed as an effective ORL-1 antagonist (410). OFQ/N (1–13) NH2 analogues modified in the 9 and/or 13 positions were synthesized and showed biological activity (828). An enzymatically-resistant OFQ/N peptide containing a carbamic acid residue was synthesized (665). Indole derivatives were designed, synthesized and biologically evaluated as novel OFQ/N receptor antagonists (1079). Potential ORL-1 ligands were synthesized from spiro-[piperidine-4,2′(1′H)-quinazolin]-4′ (3′H)-ones and spiro-[piperidine-4,5′ (6′H)-[1,2,4]triazolo[1,5-c]quinazolines] (809). Trap-101, an achiral analogue of J-113397 was identified as a potent ORL-1 antagonist (1132). Both 3-(4-Piperidinyl) indoles and 3-(4-piperidinyl) pyrrolo-[2, 3-b] pyridines are ligands for the ORL-1 receptor (96). A series of hexapeptides with a general formula of Ac-RYY-R/K-W/I-R/K-NH (2) were identified as novel partial agonists for the ORL-1 receptor (428).

BDNF induced expression of Pro-OFQ/N mRNA’s that in turn induced expression of immediate early genes in hippocampal cultures. OFQ/N, but not nocistatin increased both neurite length and number in hippocampal cultures (954). Mature OFQ/N and nocistatin were identified in human brain and CSF (534).

2b. Neuroanatomical localization

This sub-section will review current neuroanatomical studies indicating localization of opioid peptides and receptors by subtypes: mu agonists and receptors (2b-i), delta agonists and receptors (2b-ii), kappa agonists and receptors (2b-iii), OFQ/N and ORL-1 receptor (2b-iv), and opioid-related neuroanatomy (2b-v).

2b-i. Mu agonists and receptors 

Immunohistochemical labeling of the MOR carboxy terminal splice variant mMOR-1B4 was observed in the olfactory bulb, cerebral cortex, BNST, hippocampus, habenula, amygdala, thalamus, hypothalamus, median eminence, SN, VTA, oculomotor nucleus, red nucleus raphe nuclei, PAG, LC, trigeminal nucleus reticular formation, area postrema and Purkinje and deep cerebellar nuclei (1314). Whereas acute morphine decreased MOR gene expression in the PAG, chronic morphine decreased PAG and striatal MOR and KOR gene expression in female rats (1113). MOR, KOR and DOR receptors were found in disc-shaped and stellate cells of the inferior colliculus with MOR and GABA receptors co-localized in the central nucleus, dorsal cortex and external cortex of the inferior colliculus. BEND and GABA neurons were in close proximity to each other (1126). MOR is extensively co-localized with parvalbumin, but not somatostatin, in the hippocampal dentate gyrus (299). There is a distinct distribution of DA D1 receptor and MOR-1 receptor immunoreactivities in the amygdala and interstitial nucleus of the posterior limb of the anterior commissure (508). Reciprocal connections of endomorphin-1- and endomorphin-2-containing neurons were observed between the tubural and lateral hypothalamus on the one hand, and the medial, commissural, lateral and gelatinous parts of the NTS on the other (492). Mu opioid receptors were found in nerve fibers within the bovine pineal gland, whereas delta opioid receptors were observed on nerve fibers in the pineal perivascular space and intraparenchema. DOR, MOR and Lenk were co localized in some nerve fibers (908). Whereas BEND and TH immunoreactive axon terminals of MPOA cells that project to the SFO were predominantly axo-somatic, NPY terminals were more axo-dendritic (559). POMC co-localizes with CRF in axon terminals of the noradrenergic LC (948). Fluoxetine increased mu opioid receptor expression in obese Zucker rat C/P, dentate gyrus, lateral septum, amygdala, and frontal, parietal and piriform cortices (222).

MOR possessed little overlap with preprotachykinin B in the dorsal horn of the spinal cord (917). The seizure-resistant rodents, Proechimys guyannesis, display high DAMGO binding in the medial amygdala, dorsal dentate gyrus and PAG, and decreased DAMGO binding in the anterior olfactory tubercule, cingulate cortex, thalamus, basolateral amygdala, SN and dorsal hippocampus (963). Whereas lower medial septal, but not MPOA MOR densities were lower in older reproductively senescent male Japanese quail than females or young males, DOR densities in both the medial septum and MPOA were lower in this group (871).

Heroin decreased NPFF-immunoreactivity in the NTS and reduced NPFF fibers in the median eminence, pituitary stalk and neurohypophysis, but increased NPY neurons and fibers in the thalamic PVN and BNST (279). Both morphine and methadone increased c-Fos expression in the striatum and NAC (1111). Morphine produced up-regulation of the functional expression of neurokinin-1 receptors in cortical neurons (1181). Morphine respectively reduced and increased ascorbic acid release in the striatum and NAC of rats (421). Morphine increased fMRI signals in the NAC, extended amygdala, orbitofrontal cortex, hippocampus, PAG and hypothalamus and decreased fMRI signals in the anterior cingulated gyrus in the brains of healthy volunteers (70).

2b-ii. Delta agonists and receptors 

Enk afferents from the posterior BNST and the basomedial and cortical amygdaloid nuclei innervate the medial amygdaloid nuclei whereas Enk afferents from the anterolateral BNST as well as the cortical, medial and basomedial amygdaloid nuclei innervate the central amygdaloid nucleus (920). Striatal dendrites display co-localization of DOR and dopamine D1 receptor immunoreactivity (29). Striatal projection neurons to the SN mostly possess Enk-only and SP-only cells; Enk/SP co localized neurons primarily projecting to the SN do not appear confined to striosomes (1183). Enk is found in overlapping distributions with VP, VIP, CRF, gastrin releasing peptide, calbindin and calcretin in both mouse and rat (801). SP was more robust than Enk in their broad distribution in song control nuclei of 11 different oscine species (653). Four different Enks, but not DYN is found in synovium, bone marrow, periosteum and juxta-articular bone in rat joints (79). DOR immunoreactivity was evenly distributed throughout the neuropil of the feline reticular formation, and was very dense in the ventral and central areas of cytoplasm and medium to large dendrites of the parabrachial nuclei and LC (28). L-enk and the DOR were found most predominantly in the inner spiral bundle of the guinea pig cochlea, whereas DYN and KOR were found in both the inner and outer spiral bundles (533). Menk is found in small diameter and diffusely localized neurons throughout the superficial grey and stratum opticum of the superior colliculus of the camel (767). Enk is localized in the dorsolateral septal nucleus of the fire-bellied toad, Bombina orientalis (318).

2b-iii. Kappa agonists and receptors 

Although KOR was often co-localized with MOR, small ovoid KOR-containing cells were observed in close apposition with larger MOR-containing cells in medial thalamus, PAG and brainstem (415). KOR and GAD67 immunoreactivity have been found in OFF and NEUTRAL cells in the RVM (1222). Axonal mRNA transport and localized translational regulation of KOR occur in primary neurons of the DRG (90). KORS have been localized in the lipid rafts which are micro domains of plasma membranes rich in cholesterol and sphingolipids (1250). Unprocessed DYN was much greater than co-existent DYN peptides in the axon terminals of Pro-DYN neurons projecting to the CA3 region of the hippocampus and in the striatal projections to the VTA (1253). DYN and neurokinin B are co expressed in hypothalamic arcuate neurons extending to the ME and periventricular zone with these fibers projecting to the PVN, anterior hypothalamus, MPOA, median preoptic nucleus, anteroventral periventricular nucleus and BNST (152). DYN and neurokinin B immunoreactivity are co localized in the arcuate nucleus and median eminence of the sheep (352). Dopamine D1 receptors have subcellular distributions that are conducive to interactions with Pro-DYN in the NAC shell (449). Moreover, CART is co localized with pro-DYN and DA D1 receptors in the NAC (489). Pro-DYN mRNA labeling is in the mouse olfactory tubercule, lateral septum, C/P, central amygdala, PVN, SON, LHA, VMH, lateral reticular nucleus and NTS; it is also co-localized with NPY mRNA in the arcuate nucleus (669). Long-day exposure produces greater DYN A1–17 expression in the hypothalamo-pituitary axis, ependymal cells, subcomissural organ and lateral and third brain ventricle choroid plexus in the Siberian hamster than short-day exposure (776).

2b-iv. OFQ/N and the ORL-1 receptor 

OFQ/N was immunolocalized in the rat cochlea in tunnel fibers as well as inner and outer hair cells (571). The slow rate of divergence in the amino acid of OFQ/N precursor sequences in lungfish was analyzed (638).

2b-v. Opioid-related neuroanatomy

3. Pain and Analgesia

This section has four major parts examining recent advances in: a) pain responses especially as they may relate to opioid function, b) opioid analgesia organized as a function of receptor subtypes, c) sex, age and genetic differences in opioid analgesic responses, and d) opioid mediation of other analgesic responses. 

3a. Pain responses

The following sub-sections examine work done on spinal (3a-i) and supraspinal (3a-ii) circuits respectively.

3a-i. Spinal circuits 

Nociceptive behaviors (8–24 h) outlasted spinal c-Fos increases (2 h) following skin-muscle incisions, and after 24 h, surgically-treated animals were no different from controls in c-Fos-induced increases induced by thermal stimulation. Morphine was more effective in blocking nociceptive behaviors relative to c-Fos-induced increases (1325). Hindlimb and abdominal stretches occurred during labor, and were increased by systemic oxytocin. Morphine decreased oxytocin-induced stretching responses without altering labor duration, and decreased labor-induced c-Fos elevations in lumbosacral spinal segments (175). Although MOR mechanisms are not essential in the processing of acute noxious mechanical and electrical stimuli by WDR neurons, they may play an important role in endogenous inhibitory mechanisms that regulate the development of spinal neuronal sensitization during windup (422). Absence of Reelin results in significant reductions in mechanical sensitivity and a pronounced thermal hyperalgesia and aberrant neuronal positioning of Dab-1 immunoreactivity in the dorsal spinal cord (1163). Chronic constriction injury of the saphenous nerve in rats produce significant allodynia and hyperalgesia that is sensitive to morphine gabapentin, amitryptaline and WIN55212-2 administration, that increases c-Fos as well as MOR and CB-1 receptor expression in the spinal cord, and that leads to decreased functional receptive fields downstream to the injury accompanied by A-fiber ectopic discharges (1174). Spinalized rats laminectomized in the lumbo-sacral region displayed increased multi-unit efferent discharges from the ventral root following mechanical von Frey hair mechanical stimulation of the hindpaw. During discharges and after discharges were reduced by resiniferatoxin, a potent capsaicin analogue, whereas only after-discharges were reduced by morphine, ketamine and ezlopitant (1255). Partial injury of tail-innervating nerves creates both mechanical allodynia and non-allodynia in different subgroups of rats; the former group shows greater losses of spinal MOR, whereas the latter group displays more augmented mechanical allodynia following peripheral naloxone and spinal CTOP (49).

3a-ii. Supraspinal circuits 

A rodent pain model of gynecologic surgery involving laparotomy with tonic distension of the cervix and lower uterine segment was shown to be sensitive to systemic and intrathecal morphine (1125). There are strong pharmacological correlation between the formalin test and the neuropathic pain behavior in different species with chronic constriction injury (1166). An animal model of chronic inflammatory pain using large volumes of adjuvant to the intra-articular space of the rat knee showed sensitivity to morphine, dexamethasone and ibuprofen, but different temporal sensitivity from acute models (1221). Experimental pancreatitis produced visceral pain and increased thoracic DYN content with lidocaine injections into the RVM or nucleus gracilis or spinal DYN antisera reversing its effects (1154). Transgenic mice with early-stage pancreatic cancer displayed visceral pain that was enhanced by centrally-acting, but not peripherally-acting opioid antagonists (1008). An abdominal withdrawal reflex model of visceral pain due to colorectal distension was established in a behavioral assessment in rats (1264). NRM on cells excited by colorectal distension facilitate responses to colorectal distension itself, which in turn augments excitation of NRM off cells that then act to suppress cutaneous nociception (134). Anterior cruciate ligament transaction produced experimental osteoarthritis and joint pain accompanied by increased nitrite and inducible NO synthase, effects reduced by morphine, L-NAME, indomethacin and meloxicam (174). One can differentiate between capsaicin-induced allodynia and hyperalgesia using a thermal operant assay; both effects were blocked by morphine (835). The orofacial formalin test in the mouse appears to be a strong behavioral model for studying physiology and modulation of trigeminal nociception; morphine inhibits formalin responses in both phases (702). Formalin injection directly into the knee joints of rats produced the characteristic two phases of nociception; morphine blocked both phases in a naloxone-reversible manner, whereas diclofenac and midazolam reduced only the second phase, whereas meclizine and loratadine increased the second phase (735). Experimental pain-related responses are induced by a scorpion Buthus martensi Karsch sting (52). Zymosan applied to the tibial-tarsal joint produces hypernociception elicited by articular dorsal flexion movement, an effect blocked by morphine and indomethacin (430). Animals with chronic constriction injury had free access to a light: dark chamber, and chose the less-preferred light part of the chamber following stimulation of the injured paw in the more-preferred dark chamber. Whereas morphine, gabapentin, duloxetine and 8-OH-DPAT reinstated preferences for the dark side of the chamber, gaboxadol and WIN55,212–2 did not (890). The use of calibrated forceps are a reliable tool for pain, morphine analgesia and morphine tolerance (706). Ultrasonic vocalizations occur primarily during the interphase of the formalin test, and are suppressed by morphine in a naloxone-reversible manner (863). Hairless mice fed a special diet develop atoplic-like dry skin inducing prolonged scratching that is dose-dependently blocked by naloxone, but not H1 or 5HT(1/2) antagonists (369). Administration of 2,4,6-Trinitrochlorobenzene to hairless mice produced a naltrexone- and L-NAME-reversible itch response (1138).

3b. Opioid analgesia

The following sub-sections examine advances in our understanding of opioid-mediated analgesia in the past year especially as they pertain to the opioid receptor subtypes and their genes: i) mu agonists and receptors, ii) delta agonists and receptors, iii) kappa agonists and receptors, and iv) OFQ/N and the ORL-1 receptor. A large number of studies examine either knockout or knockdown techniques to indicate roles of the receptors, and potential splice variants in opioid analgesic function. Separate paragraphs are devoted to studies in which other transmitter and peptide systems affect opioid analgesia; the effects of opioid manipulations upon analgesia induced by other peptides and transmitters are covered in Section 3d. Finally, human studies related to opioid and particularly mu receptor-mediated analgesia is covered at the end of Section 3b-v.

3b-i. Mu agonists and receptors 

MorphineA rank-order analgesic potency of morphine was observed on the paw pressure, hot-plate and tail-withdrawal tests with the lowest on the formalin test, effects independent of stimulus intensity on all tests. A within-subject cumulative dosing procedure resulted in lower ED50 values than a between-subject procedure, and the ED50 of morphine progressively increased when the cut-off value was adjusted from 4 to 5, 6, 7 and 8 standard deviations above the mean (796). Nociceptive stimulus modality-related differences occurred in pharmacokinetic-pharmacodynamic modeling of morphine in the rat such that drug effects were more effective for mechanical relative to thermal stimuli (1015). Sub-analgesic doses of DAMGO or fentanyl enhanced morphine analgesia through induction of MOR endocytosis (454). Opiate receptor-dependent analgesia is associated with several compounds bound to Gbetagamma subunits (110). The analgesic, tolerant, dependent, rewarding and sensitizing effects of morphine, but not kappa agonists were lost in mice lacking adenylyl cyclase type 5 that is highly enriched in striatum (582). Mice lacking phospholipase Cbeta1 displayed hypersensitivity to pain and reduced morphine analgesia and tolerance formation (685).

Morphine administered systemically or directly into the ACC produced a naloxone-sensitive reduction in pain affect, as demonstrated by decreased aversiveness of noxious cutaneous stimulation in nerve-damaged animals with no concomitant alteration of response to mechanical stimulation (623). NAC morphine-induced increases in thermal and mechanical hindpaw latencies were blocked by naloxone, BFNA and NBNI, but not NTI (1244). Morphine in the thalamic nucleus submedius reduced the second phase of formalin-induced evoked spinal dorsal horn nociceptive responses in a naloxone-reversible manner (1317). Microinjections into the thalamic nucleus submedius of morphine, endomorphin-1 or DADL, but not the kappa agonist, spiradoline, dose-dependently inhibited mechanical and cold allodynia induced by spinal nerve ligation, effects blocked by general and mu, but not delta antagonism (1187). Similarly, microinjections into the ventrolateral orbital cortex of morphine, endomorphin-1 or DADL, but not the kappa agonist, U-62066, dose-dependently inhibited mechanical and cold allodynia induced by spinal nerve ligation, effects blocked by general and mu, but not delta antagonism (1318).

Morphine, gabapentin and lamotrigine produced analgesia of similar potencies in attenuating capsaicin and nerve-injury-induced mechanical hypersensitivity (535). Morphine modulated potentiation of capsaicin-evoked TRPV1 responses through a cyclic AMP-dependent PKA pathway (1160). Intrathecal naloxone or CTAP blocked systemic morphine- and fentanyl-induced analgesia on mechanical measures of nociception as well as thermal responses to the hindpaw, but not forepaw (198). However, noncompetitive interactions between the opioid antagonists, CTAP and naltrexone were observed in the rat tail-withdrawal assay (1175). Morphine-associated pruritus occurred after a single extradural injection in a horse (150).

Two studies independently demonstrated that morphine-induced decreases in intraocular pressure and pupil diameter in rabbits was blocked by naloxone and partially reversed by L-NAME or L-glutathione, indicating NO involvement (111, 296). Low-dose morphine-induced hyperalgesia was blocked by PLC inhibitors (U73122) and blockers (calphostin C), also by ketamine and MK-801, and finally by AS against PLCbeta3 (378). Acute and chronic morphine increases Phase II pain-associated behaviors on the formalin test in neonatal rats after four days of abstinence (1332).

Mu Opioid Agonists 

Human opiorphin, a physiological inhibitor of Enk-inactivating zinc endopeptidases produced potent analgesia in chemical and mechanical pain models (1223). Peripherally-mediated antinociception of the mu-opioid receptor agonist 2-[(4,5alpha-epoxy-3-hydroxy-14beta-methoxy-17-methylmorphinan-6beta-yl)amino]acetic acid (HS-731) occurred after subcutaneous and oral administration in rats with carrageenan-induced hindpaw inflammation (100).

Mu Opiate AgonistsL-methadone, but not D-methadone produced stronger anti-allodynic effects in the spinal nerve ligation model relative to morphine, oxycodone and methadone itself (646). 6-acetyl-codeine analgesia was blocked by AS probes targeting Exons 1 and 2 of the MOR gene (780). The ability of buprenorphine to increase tail-flick latencies correlated with the extent and duration of reduced numbers of mu opioid receptor binding; however, residual analgesia after 8 h was not accompanied by mu binding changes (320). Ventricular and intrathecal buprenorphine analgesia on the formalin test was blocked by ventricular and intrathecal naloxone and the ORL-1 antagonist, J113397. Methadone and morphine were more effective than buprenorphine in increasing thermal and pressure thresholds in cats (1067). Systemic buprenorphine analgesia was blocked by systemic or intrathecal, but not ventricular naloxone, and was enhanced by systemic or ventricular J113397. Systemic, but not ventricular or intrathecal buprenorphine decreased formalin-induced Fos activity in the lumbar dorsal horn (1257). Thienorphine produced a similar analgesic duration as buprenorphine, but was longer than the latter in antagonizing morphine-induced lethality (1279).

The analgesic efficacy of such partial opioid agonists as buprenorphine or tramadol was increased in mice with targeted inactivation of the alpha2A-adrenoceptor gene or in wild-type mice treated with atipamezole and yohimbine (874). Tramadol produced analgesia in MOR KO mice that was blocked by alpha-2 adrenergic antagonists alone or with naloxone, but not by methysergide (499). Isobolographic analysis revealed dual-site synergism in the systemic and intraplantar antinociceptive response of tramadol in the formalin test in rats (921). Oral and intrathecal, but not ventricular oxycodone, but not morphine produced analgesia in diabetic mice. Whereas, mu but not kappa or delta antagonism blocked oxycodone analgesia in non-diabetic mice, kappa, but not mu or delta antagonists blocked oxycodone analgesia in diabetic mice (852). Whereas subcutaneous oxycodone and oxymorphone produced potent analgesia in the hot-plate and paw pressure tests, intrathecal oxymorphone produced far more potent effects than intrathecal oxycodone (645). Liposome-encapsulated hydromorphone provided extended analgesia in a rat model of neuropathic pain (1047). Analgesic effects were observed after epidural administration of hydromorphone in horses (822), and similar effects were observed following subarachnoid hyperbaric morphine, buprenorphine and methadone (823). Systemic and intrathecal remifentanil produced naltrexone-reversible analgesia on the acetic acid test in Rana pipiens (792). Gabapentin activated the NO-cyclic GMP-K+ channels pathway in inhibiting formalin-induced pain (869). Spinal alpha-2 adrenergic and muscarinic receptors as well as the NO release cascade mediated the supraspinally-produced effectiveness of gabapentin in decreasing mechanical hypersensitivity in mice after partial nerve injury (1103).

Endomorphins 

Intrathecal endomorphin-2, but not endomorphin-1 or DAMGO analgesia was prevented by antisera directed against DYN A(1–17), but not DYN B(1–13); this effect was reversed by pretreatment with ultra-low doses of BFNA or naloxonazine that were ineffective at mu opioid receptors (788). Endomorphin-1 and endomorphin-2 produced analgesia when injected directly into an inflamed hindpaw, effects blocked by mu, but not kappa antagonists, and with delta antagonism blocking the latter but not former effect. Antibodies raised against BEND, Lenk or Menk failed to affect endomorphin-induced analgesia (622). Endomorphin-1 produced longer-lasting analgesia when injected in nanoparticles coated with polysorbate 80 (679). The analogues, guanidine-[d-Ala(2), pCl-Phe(4)] endomorphin-1, the four D-Ala-containing tetrapeptides and a chloro-halogenated d-Pro-Gly-containing pentatpetides of endormorphin-1 showed potent and prolonged analgesic activity (681). Strong opioid receptor binding and analgesic activity were found with analogues of endomorphin-2 and morphiceptin with phenylalanine mimics in position 3 or 4 (384). The novel analogue, [Dmt1]endomorphin-1 produced spinal and supraspinal analgesia that was blocked by naloxone and BFNA, but not naloxonazine, indicating mu-2 action (527). Tyr-Pro produced naloxone-reversible analgesia on the tail-flick test, and [3H]-Tyr-Pro could be found in endomorphin-related tetra- and tri-peptides in brain (968). The modulation of endomorphin-1 and –2 analgesia by L-NAME was inhibited and reversed by L-Arg and naloxone respectively (204).

BEND 

A BEND anti-serum blocks thermal analgesia noted at the initial stages of a murine osteosarcoma (47). Intraplantar BEND respectively suppressed and potentiated concanavalin A-induced paw edema at high and low doses in delta- and kappa-selective antagonist fashion in two inbred strains that differed in BEND-induced modulation of phagocytosis and NO production (1065).

Manipulations affecting Mu Analgesia 

Dextro- and levo-naloxone each reversed the attenuation of morphine analgesia induced by lipopolysaccharides in the mouse spinal cord through a non-opioid mechanism (1233). Bicuculline or picrotoxin administered into the ventrolateral orbital cortex enhanced the ability of morphine to inhibit tail-flick latencies in this region, whereas muscimol and THIP attenuated the morphine-induced response (930). Gabapentin enhances the analgesic response to morphine in an acute model of pain in male rats (775). SB203580, a p38 MAPK inhibitor reversed antianalgesia induced by dextro-morphine or morphine in the mouse spinal cord (1234).

Endothelin ETA receptor blockade potentiates morphine analgesia but does not affect gastrointestinal transit in mice (748). Mice lacking P-glycoprotein displayed enhanced analgesic effects to morphine and fentanyl, but not meperidine. Morphine and fentanyl stimulated P-glycoprotein ATPase activity as well (439). Mice lacking one allele of glial cell line-derived neurotrophic factor displayed enhanced morphine analgesia and enhanced NAC DA output (11). Enhancement and restoration of the analgesic efficacy of codeine and morphine occurred following delta9-THC (1219). Systemic morphine produced antinociception mediated by spinal 5-HT7, but not 5-HT1A and 5-HT2 receptors (290). The I2-imidazoline binding site ligand, phenyzoline enhanced morphine analgesia on the tail-flick and hot-plate tests, whereas its ortho-phenyl derivative decreased this response (394). Synergistic analgesia occurs between the phosphodiesterase inhibitor, zaprinast and morphine in the spinal cord of rats on the formalin test (1276), whereas riboflavin enhanced this measure as well (84). Blockers of the gap junction channel, carbenoxolone and Gap27 reduced morphine analgesia after intrathecal administration (1090). Co-administration of the non-competitive NMDA antagonist, MK-801 and morphine attenuated neuropathic pain (441). [Ser1]-histogranin, a peptide NMDA receptor antagonist enhanced morphine analgesia in the formalin test (438). Naloxone pretreatment and post treatment blocked the morphine-induced and dipyrone-induced analgesic potentiation in the tail-flick test in rats (467).

Hemokinin-1, a mammalian tachykinin peptide, markedly potentiated the antinociceptive effects of morphine administered at the peripheral and supraspinal level (364). Enhanced antinociceptive effects of morphine were noted in histamine H2 receptor gene KO mice (790). Lithium potentiated morphine analgesia in a time-sensitive manner near the end of the light phase of the light:dark cycle (549). Ketamine enhanced the analgesic, cataleptic and hypoactive effects of morphine in a naloxone-reversible manner, with the first effect reduced by yohimbine, but not glibenclamide (163). BIBP3226, a NPY Y1 receptor antagonist, prevented the ability of neuropeptide FF to reduce morphine analgesia on the tail-flick test (331). Rats over expressing beta-arrestin 2 in the PAG display decreased morphine analgesia (522). The analgesic effects of the CB2 receptor agonist AM1241, but not morphine were absent in CB2 KO mice (497). Animals allowed wheel-running activity displayed significantly less morphine- and M6G-induced analgesia elicited from the vlPAG than inactive rats (744). Both GABA-B and GABA-A agonists reduced morphine-induced Straub tail responses in mice (1295). Resiniferatoxin, a potent capsaicin analogue decreased individual, but not combined immunoreactivities of mu opioid receptors and transient receptor potential vanilloid type-1 neurons in the DRG and dorsal horn. This was accompanied by enhancements in the analgesic magnitude and duration following intrathecal DAMGO and morphine as well as systemic morphine; the number, but not affinity of spinal [³H-DAMGO] binding was also decreased (197). The methanolic extract of Polygala telephiodes antagonized morphine analgesia on the hot-plate test, improved morphine-induced memory impairments on the elevated plus maze, and suppressed naloxone-precipitated jumping behaviors in morphine-dependent mice (309).

Mu receptors 

The 3′ non-coding region of MOR is important is morphine analgesia, and appears to be the reason for a disrupted response in CXBK mice (555). MOR KO mice displayed spinal morphine analgesia at higher doses that was blocked by naloxone and NBNI, but not NTI. Spinal U50488H, but not DPDPE analgesia was observed in MOR KO mice (1254).

3b-ii. Delta agonists and receptors 

There is a crucial role of delivery and trafficking of DOR in opioid analgesia and tolerance (1313). Chronic pain-induced emotional dysfunction is associated with astrogliosis due to cortical DOR dysfunction (815). Protease activated receptor 2 agonists activate trigeminal nociceptors and induce functional competence in DOR (885). Dextro- and levo-morphine attenuated delta and kappa opioid-mediated analgesia in MOR KO mice (1235). The mechanical and thermal allodynic as well as the thermal hyperalgesic effects of neuropathic pain were enhanced in DOR KO mice (810). The peripheral delta agonist, [dVal(L)2,Ala(L)5]Enk produced naloxone-reversible and NTI-reversible analgesia induced by osteosarcoma cells injected into mouse femur with greater potency than morphine (131). Dmt-Tic-CH2-Bid, a potent DOR receptor agonist produced NTI-reversible analgesia and reduced immobility during forced swimming (1155). Chimeric glycosylated peptides of Menk and FMRFamide, like [O-Glu-Ser5]Yfa, display increased levels of analgesia and increased bioavailability (737).

3b-iii. Kappa agonists and receptors 

The antinociceptive and hypothermic effects of the KOR agonist, Salvinorin A are abolished in a novel strain of KOR-1 KO mice, indicating an effect mediated by KOR-1 and not KOR-2 receptors (33). The antinociceptive effects of salvinorin A in mice were blocked by NBNI, but not by BFNA or Nti (528). Elevated levels of DYN through the activation of bradykinin receptors contribute to the maintenance of neuropathic pain, an effect blocked by bradykinin B1 and B2 receptors only when elevated DYN is present (626; comment by (26)). Spinal cord DYN expression increases, but does not drive microglial prostaglandin production or mechanical hypersensitivity after incisional surgery in rats (1329). U50488H administered into the contralateral hindpaw of animals exposed to unilateral arthritis induction reduced hindpaw oedema, ankle joint inflammation, pain behaviors, inflammatory severity in both hind paws and decreased cartilage oligomeric matrix protein, effects reversed by the peripheral opioid antagonist, naloxone methiodide (99). Whereas morphine was effective in treatment of herpetic pain, the KOR agonist, nalfurafine, suppressed herpetic and post-herpetic pain to similar degrees following intrathecal administration. Spinal MOR was down regulated in the post-herpetic period (1103). Whereas stimulation of MOR and DOR induced hyperalgesia, stimulation of KOR induced analgesia in the hot-plate test in the naked mole-rat (1129). DYN A protein levels, increased in peripheral nerves and footpads of STZ diabetic mice, produces analgesia with activation of KOR and hyperalgesia with activation of NMDA receptors in these mice (531). Salvinorin A produced short-lived (10 min) analgesia on the tail-flick, hot-plate and acetic acid writhing tests, effects blocked by NBNI (755). Nalfurafine, a KOR agonist, inhibited scratching behavior secondary to cholestasis induced by chronic ethynylestradiol injections in rats (502). KT-95, an agonist binding to mu, delta and kappa receptors induced analgesia on the acetic acid writhing test that was blocked by NBNI, but not naloxone. KT-95-induced amelioration of scopolamine-induced memory impairments was unaffected by NBNI, but blocked by sigma antagonism (473).

3b-iv. OFQ/N and ORL-1 receptor 

OFQ/N continues to present a complex picture concerning its role in pain responses producing both “pro-nociceptive” and “anti-nociceptive” actions depending on such factors as site of administration, dose and time course. This section therefore presents these data separately.

Pro-nociceptive actions 

Administration of the ORL-1 antagonist, UFP-101 produced analgesia on the formalin test following ventricular administration and hyperalgesia following intrathecal administration. Mice receiving systemic administration of the ORL-1 antagonist, J-113397 or ORL-1 KO mice displayed increased nociceptive behaviors on the formalin test (961). OFQ/N administered into the vicinity of the knee caused a shift in weight bearing to the contralateral leg in rats, and reduced paw withdrawal thresholds and latencies to the affected foot, effects independent of mast cell activation (756). OFQ/N and PGE2 induced allodynia that was blocked by the ORL-1 antagonist, JTC-801 and that was absent in Pro-OFQ/N KO mice. OFQ/N-induced allodynia was unaffected by inhibition of PG production by indomethacin, and the PGE receptor agonist, AE1-329 stimulated OFQ/N from spinal slices and induced allodynia (860).

Both partial sciatic nerve transaction and administration of CFA up regulated OFQ/N and the ORL-1 receptor in the dorsal root ganglion; these effects were accompanied by increases in activating transcription factor 3, a neuronal marker of nerve injury (202). Hyperalgesia induced by OFQ/N was blocked by melatonin that in turn produced analgesia that was blocked by naloxone and the melatonin receptor antagonist, luzindole (1188). Two novel ORL-1 peptide analogues, Peptide-1 and –2 produced pro- and anti-nociceptive effects respectively with the latter block the effects of the former. Peptide-1 produced greater decreases in MAP than OFQ/N, and the latter’s hypotensive effect was blocked by Peptide-2 (896). The OFQ/N agonist Ac-RYYRIK-ol produced hyperalgesia following supraspinal administration and analgesia following intrathecal administration as well as producing decreased locomotion and increased food intake (427).

Antinociceptive actions 

Intrathecal OFQ/N produced analgesia in monkeys that was blocked by the ORL-1 antagonist, (+)J-113397, but not naltrexone with its metabolite OFQ/N(2–17) present at 1.5 but not 4.5 hours later (599). Intrathecal OFQ/N as well as electro acupuncture attenuated in an ORL-1-sensitive manner the hyperalgesia induced by CFA with greater effects observed when OFQ/N and electro acupuncture were combined (365).

3b-v. Human studies 

Volunteers 

The high-estrogen state was associated with regional increases in baseline MOR availability in women and a greater activation of endogenous opioid neurotransmission during the pain stressor (1050). Using [18F]-fluorodiprenorphine PET in healthy volunteers, it was found that heat pain stimulation reduced diprenorphine binding in limbic and paralimbic areas including the NAC and insula (1062). The 118A>G single nucleotide polymorphism of the MOR appears important in morphine- and M6G-induced pain control in healthy human volunteers (695) as well as reducing alfentanil-induced analgesia and respiratory depression (856). Endogenous pain-inhibitory systems activated by spatial summation are opioid-mediated given naloxone’s effectiveness in volunteer male and female subjects (538). Pain thresholds and tolerance were higher in men at high risk for hypertension, with greater ratings of pain and salivary cortisol following naltrexone treatment (13).

Oxycodone was more effective than morphine in a multimodal, tissue-differentiated experimental pain model (1064). CRF increased tolerance to pressure, but not heat pain in a naloxone-insensitive manner; the CRF-induced increases in plasma BEND and BLPH did not correlate with the pain tolerance thresholds (743). Both alfentanil and morphine reduced human experimental muscle pain (1002). Transdermal fentanyl was better tolerated than transdermal buprenorphine in terms of skin irritation as measured by erythema (997).

Dental Pain 

Both BEND and somatostatin levels were higher in exposed and inflamed dental pulps than in uninjured pulps (806).

Chronic Pain 

A meta-analysis (373) revealed that weak and strong opioids outperformed placebo for pain and function in chronic noncancer pain patients. Withholding of opioid therapy should not be a consequence of concern of hastening death in a large hospice population (919). The frequency of the minor allele of the functional A118G polymorphism of the human MOR gene was lower in patients with acute and chronic pain (514). Parental history of chronic pain appears to be correlated with impairments in endogenous opioid analgesic systems (142). Intrathecal morphine produced greater pain relief in patients with non-cancerous chronic back pain, an effect accompanied by increased pruritus, nausea and vomiting (937). Intrathecal fentanyl and sufentanil for the treatment of chronic pain was found to be potent and efficacious (1173). Intravenous morphine was effective in acute pain without much incidence of nausea or vomiting (128), but was marginally better than placebo treatment for acute abdominal pain (381). Chronic pain patients administered oral morphine for one month became hyperalgesic and tolerant on the cold presser, but not a heat model of pain (217). Chronic pain patients administered intrathecal morphine displayed significantly less OFQ/N in cerebrospinal fluid that controls (938).

Effective treatment of chronic moderate-to-severe non-malignant pain was observed with polymer-coated extended-release morphine sulfate capsules (840). Ganglionic local opioid application (GLOA) was effective for treatment of chronic headache and facial pain (451). Use of morphine and other opiates was associated with greater risk of fracture due to falls related to dizziness (1158). Opioid switching from oral slow release morphine to oral methadone improves pain control in chronic non-malignant pain over a nine-month follow-up study (359). Administration of naloxone following long-term morphine treatment produced withdrawal and reinstatement of lower morphine doses to produce analgesia in a selected group of chronic pain patients (693). CJC-1008, a long-acting parenteral opioid analgesic was effective in the treatment of postherpetic neuralgia (1177). Hydrocodone produced greater amounts of abuse liability than NSAIDs or tramadol in a female group of patients treated for chronic non-cancer pain (6). Intrathecal hydromorphone reduced intractable nonmalignant pain (304), and was shown to be safe and efficacious in acute emergency-related pain (180). Hydromorphone can be produced as a minor metabolite of morphine within minutes in chronic pain patients (234).

Daily-extended release hydromorphone was effective in patients with persistent moderate to severe chronic pain (1211). Oxytrex, a combination of oxycodone and an ultra-low naltrexone dose, minimized physical dependence while providing effective analgesia in patients with low back pain (1205). Extended release oxymorphone appeared effective in pain relief for osteoarthritis of the hip or knee (593), and was more effective and less costly than an oxycodone-acetaminophen combination for osteoarthritis pain of the hip or knee (732). A weight-based dose of oxycodone without adjustment for age between 6 months and 7 years is valuable for evaluating dosing schedules and routes (316). Although rofecoxib was more effective in ameliorating chronic osteoarthritis pain in the knee than acetaminophen, the latter reduced plasma BEND levels (1025). The analgesic effect of pamidronate is not caused by the elevation of BEND level in Paget’s disease (77). Serum BEND and IgE are useful biomarkers for itch and disease severity in patients with atopic dermatitis (635). Electro acupuncture was effective in the treatment of postthoracotomy pain (1226). Opioid expenditures and utilization increased in the Medicaid system between 1998 and 2003, particularly for morphine derivatives (135).

Cancer Pain 

The use of low morphine doses in opioid-naïve cancer patients with pain appeared to be a reliable method in pain treatment (769). Epidural morphine produced greater analgesia than intravenous morphine for oral cancer surgery with pectoralis major myocutaneous flap reconstruction (1042). Oxycodone was as efficacious and tolerable as morphine for cancer pain treatment (945). Three-cycle fentanyl patch system significantly improves pain control in gynecologic cancer (547). Even though oral morphine is not contra-indicated, there is a trend to use a fentanyl patch as the first-choice strong opioid in cancer patients undergoing titration, in the presence of intractable pain and in the absence of dysphagia (956). S-methadone had a shorter half-life and shorter population pharmacokinetics than R-methadone in hospice patients with cancer pain (45). There was a beneficial effect of low-dose Ketamine addition to epidural administration of morphine-bupivacaine mixture for cancer pain (1007). Midazolam was an effective adjunct therapy to morphine in the alleviation of severe dyspnea perception in patients with advanced cancer (827).

Surgical Pain 

The A118G single nucleotide polymorphism of MOR appears to moderate, but does not mediate the effects of the anger-out trait on postoperative pain responses (144). Age and the prior use of psychotropic agents before surgery were positively associated with a higher rate of postoperative morphine consumption (240). Intrathecal morphine was similar to epidural morphine during liver resection surgery for postoperative pain relief (276), but improved pain relief after abdominal aortic surgery (105), and is effective in treating elderly patients in a regular post surgical ward (667). Intrathecal morphine combined with bupivacaine reduced pain scores following dorsal rhizotomy in children (469). Morphine and Entonox were not very effective analgesics in children undergoing chest drain removal (140). Rectal morphine administered in solution or gel form administered as premedications were effective analgesics (707). Following laparoscopic cholecystectomy surgery, children with sickle-cell disease consume twice as much morphine, report more pain and have longer hospital stays than non-sickle cell children (246).

No adverse events in pediatric ketamine sedations occurred with morphine pretreatment (1200). There was an association of MOR gene polymorphism (A118G) with increased morphine consumption for analgesia after total knee arthroplasty (214). Both perioperative and PCA morphine were effective analgesics for total knee arthroplasty (1153). DepoDur, an epidural extended-release morphine formulation produced extended pain relief after knee arthroplasty (452), and high doses of oral morphine were effective after total hip arthroplasty (720). Addition of morphine and ketolorac to ropivacine enhanced pain levels in patients with arthroscopic knee surgery (838). The level of post-operative pain after hip or knee arthroplasty correlated with plasma levels of ACTH, BLPH and BEND (741). Moreover, there was an inverse correlation between the level of post-operative pain and pre-operative levels of CSF BLPH, but not ACTH or BEND (742).

Intraoperative administration of tramadol for postoperative nurse-controlled analgesia resulted in earlier awakening and less sedation than morphine in children after cardiac surgery (218).

Tramadol is comparable to morphine during adenotonsillectomy for obstructive sleep apnea in children (494). Tramadol with lornoxicam had the least immunity depression during postoperative analgesia relative to morphine or tramadol alone (1196). Postoperative cognitive function and pain relief was more effective with tramadol as compared to fentanyl patient-controlled analgesia (839). Preemptive gabapentin reduces postoperative pain and opioid demand following thyroid surgery (23), but a meta-analysis still finds its clinical significance undetermined (1005).

Acute opioid tolerance developed during infusion of remifentanil for pediatric scoliosis surgery (245). A combination of low ropivacine and morphine doses was not more effective than a higher ropivacaine dose in post-operative analgesia for arterial bypass surgery of the lower extremities (354), and ropivacaine had opiate sparing effects following total knee arthroplasty (1004). The use of continuous epidural infusion or PCA of levobupivacaine produced comparable analgesia in terms of postoperative morphine consumption (278). Caudal bupivacine into the sacral hiatus was more effective than morphine in attenuating intraoperative and postoperative pain and stress responses in children undergoing abdominal surgery (1116). Oxycodone absorption is similar after buccal and sublingual instillation during surgery in children (605). Hydrocodone and acetaminophen combinations produced similar pain relief to rofecoxib in functional endoscopic sinus surgery (221), whereas acetaminophen and ketoprofen produced pain relief in children after soft tissue or orthopedic surgery (471).

Remifentanil is a reasonable alternative to fentanyl when using a target controlled propofol infusion in patients undergoing craniotomy for supratentorial lesions (271). Rofecoxib decreased pain scores and morphine consumption after orthopedic, breast and spine surgery (951). Propofol prevented post-operative pruritus induced by epidural morphine with ropivacaine (607). Intravenous bolus of ultra-low-dose naloxone added to morphine does not enhance analgesia in emergency department patients (97). Preoperative flurbiprofen axetil reduced postoperative pain and morphine needs for spinal fusion surgery (1258). Intraoperative infusion of dexmedetomidine reduces postoperative analgesic requirements (429), and can substitute for fentanyl for bariatric surgery (340).

Caesarean Pain 

Oral morphine was better than intravenous morphine PCA in controlling pain after cesarean delivery (264). Chloroprocaine failed to affect epidural morphine for post cesarean delivery analgesia (468). Droperidol, allzapride and propofol, but not promethazine had prophylactic effects on spinal morphine-induced pruritus following caesarean section (480). However, ondansetron and tropisetron did not prevent intraspinal morphine- and fentanyl-induced pruritus in elective cesarean delivery (988). Sufentanil and morphine added to hyperbaric bupivacaine in supraspinal anesthesia for caesarean section improved pain relief (551). Naloxone is rarely used to reverse opiate effects in newborn infants, and its use can be curtailed further without increasing respiratory morbidity (121).

Pre-operative intrathecal morphine enhanced the quality of postoperative analgesia for total abdominal hysterectomy (552). Human opioid receptor A118G polymorphism increased intravenous PCA morphine consumption after total abdominal hysterectomy (213). Intrathecal bupivacaine, clonidine and morphine were more effective than general anesthesia for pain relief and functional status after vaginal hysterectomy (1063). Lornoxicam and ketoprofen reduced post-operative pain and morphine consumption following hysterectomy (550). Women undergoing hysterectomy with a genetic polymorphism of the interleukin IL-1Ra receptor, but not polymorphisms in interleukin-1 displayed greater variability in post-operative morphine consumption for pain (85). Gabapentin attenuated late, but not acute pain after abdominal hysterectomy (335). Intrathecal adenosine was not efficacious for postoperative pain relief following abdominal hysterectomy (1023).

3c. Sex, age and genetic differences

So-called organismic variables play vital roles in the mediation of opioid analgesic responses, and continue to attract a great deal of attention. The first variable is the sex or gender of the animal, and opioid receptor subtypes appear to play important roles in whether there is a sex differences in analgesia and whether it is mediated by circulating gonadal hormones. Test-specific and route-specific variables also come into play that not only informs us about sex differences in opioid analgesic function, but other functions covered in other parts of the review. Aging factors, and especially genetic factors from both inbred and transgenic models continue to play a vital role in our understanding of opioid analgesic function.

3c-i. Sex 

Female rats have more PAG-RVM output neurons than males, but males have more activated PAG-RVM cells than females during inflammatory pain. Systemic morphine significantly suppressed CFA-induced Fos in males only (700). Testosterone produces anti-hyperalgesic effects during development in male animals that is maintained during adulthood. Morphine analgesia inhibits inflammation-induced pain in adult gonadectomized, but not neonatally-gonadectomized animals (114). Male rodents and nonhuman primates display greater analgesic responses than females following mu and kappa opioids, particularly those that display low efficacy such that they appear to act as full agonists in males, but antagonists in females. These effects interact with drug history, genotype and the modality, duration and intensity of the nociceptive stimulus (67). Whereas estrogen attenuates OFQ/N analgesia elicited from the spinal cord in the female, testosterone is required for OFQ/N analgesia in the male rat (227). Male rats displayed greater analgesic and antihyperalgesic effects of morphine than females on CFA-induced pain over a 3-week time course (1192). Male rats displayed greater analgesia on a visceromotor pain test following systemic loperamide and inventricular, but not intrathecal morphine than female rats (521). Female rats, and particularly ovariectomized females displayed more marked leftward shifts in morphine’s dose-response curve following the barbiturate, pentobarbital (243). Testosterone levels that affected male sexual behavior and reproductive physiology did not similarly alter basal nociception or morphine analgesia (1085).

Chronic sucrose intake reduced the antagonist effect of BFNA on morphine-induced analgesia in female, but not male rats (241). Whereas female Fischer rats showed greater enhancements of contact hypersensitivity following spiradoline, but not SNC80, both sexes showed comparable analgesic and antidiuretic effects following both compounds (314). Neurokinin-1 receptor antagonism decreased mu and kappa opioid-induced enhancement of contact hypersensitivity in females but not males (315). Female rats displayed greater analgesia following oxycodone and U50488H than males, and male rats showed greater hyperalgesia following low oxycodone and U50488H doses (477).

Gender differences in sensitivity to short duration cold pain were associated with variations in the genes for transient receptor potential A subtype 1, catecol-O-methyltransferase and fatty acid amide hydroxylase (579). Placebo analgesia occurred in males with female experimenters, suggesting the importance of the social context in which the pain is recorded (350). Women undergoing dental procedures displayed greater nalbuphine analgesia than men with men displaying a late onset anti-analgesia following nalbuphine. Pretreatment with chlorpromazine or haloperidol enhanced nalbuphine analgesia eliminating any sex difference, and abolished the male anti-analgesia (389).

3c-ii. Aging 

Morphine analgesia did not differ between young and aged rats using tail-flick latencies, but older animals showed a decreased analgesic response using a method measuring tail-flick thresholds as well as escape thresholds from electrical stimulation of the mesencephalic reticular formation (249).

3c-iii. Genetic differences 

Opioid-induced hyperalgesia that occurs after repeated morphine was strongly strain-dependent in 15 inbred groups of mice with genetic analysis identifying a genetic locus near the beta-2 adrenergic receptor gene. The selective beta2-adrenergic antagonist, butoxamine blocked opioid-induced hyperalgesia (662). Low doses of naltrexone enhanced morphine analgesia and attenuated the development of morphine tolerance in Sprague-Dawley and Long-Evans rat strains, but not in F344 or Lewis rat strains (1115). Fischer rats displayed greater carrageenan-induced swelling and pain as well as lower circulating levels of TNF-alpha and higher levels of IL-6 as compared to Lewis rats. Although morphine reduced carrageenan-induced behaviors in both strains, low morphine doses produced a mechanical allodynia and hyperalgesia in the noninflamed paw of Fischer, but not Lewis rats (337). Lewis rats displayed lower thresholds for CRF-induced analgesia, and higher doses of the CRF antagonist, astressin to produce hyperalgesia that was blocked by morphine pretreatment (1167).

3d. Opioid mediation of other analgesic responses

This section summarizes studies that indicate that analgesia elicited by a wide range of peptides and transmitters can alternatively and respectively be sensitive (3d-i) or insensitive (3d-ii) to opioid manipulations using agonists, antagonists and transgenic knockouts.

3d-i. Opioid-sensitive analgesic responses 

Central galanin-induced analgesia was reduced in morphine-tolerant rats, and these animals displayed an up-regulation of galanin-like immunoreactivity in the hypothalamic arcuate nucleus (1238). The neurotensin agonist NT1 displayed analgesia and tolerance on the hot-plate test that was not cross-tolerant with morphine, but was sensitive to naloxone on the jumping, but not licking response. The neurotensin-2 receptor antagonist blocked neurotensin, but not morphine analgesia on the hot-plate test (130). Intrathecal naloxone inhibited but was less effective than an Oxy antagonist in blocking the ability of Oxy or PVN stimulation to reduce the withdrawal responses to mechanical and cold stimuli in sciatic nerve-ligated rats (782). KOR and DOR antagonists block the antinociceptive effect of Oxy in formalin-induced pain responses in mice (945). VP-induced analgesia elicited from the PAG was blocked by naloxone and V2, but not V1 receptor antagonists (1260). Ghrelin inhibited inflammatory pain induced by carrageenan that was reversed by naloxone (1037). The COX-2 inhibitor, SC236 produced hypoalgesic effects in carrageenan-treated rats that were blocked by naltrexone and absent in morphine-tolerant animals (357). The PKC inhibitor, chelerythrine blocked basic spinal c-Fos expression to formalin pain as well as the naloxone-induced increases in spinal c-Fos expression to formalin pain (842). The analgesic effects of fatty acid amide hydrolase inhibition with URB597 in a rat model of neuropathic pain were blocked by naloxone and CB-1 antagonism (520). Intraplantar injection of CXCL2/3, but not CXCL12 elicited naloxone-reversible mechanical and thermal analgesia that was abolished by systemic polymorphonuclear cell depletion (958). Central melatonin produced antihyperalgesic, but antiallodynic actions in sciatic nerve-injured mice, effects blocked by naloxone and L-arginine (1140), and naloxone blocked melatonin-induced analgesia (723). The SSRI, fluvoxamine produced analgesia that was blocked by intrathecal and systemic administration of general, mu and delta, but not kappa antagonists in sciatic nerve-injured mice (851).

Fluoxetine-induced analgesia on the tail immersion and hot-plate assays was significantly less potent in STZ-induced diabetic mice, and reversed by opioid and muscarinic, but not alpha-2 adrenergic antagonists (32). Inhibition of the jaw opening reflex and single neurons in the trigeminal subnucleus caudalis by activation of striatal D2 dopamine receptors was suppressed by striatal quinpirole and reversed by systemic naloxone (991). Desimiprimine and trimipramine-induced analgesia were blocked by Nti, but only by high doses of naltrexone (875). A chlorinated chimeric peptide of Menk and FMRFa, [p-Cl Phe(4)], produced a naloxone-reversible analgesia (447). Analgesia induced by AMPA receptor antagonism in the spinal cord was blocked by mu and delta opioid antagonists (606). L-arginine and a NO donor each produced analgesia that were blocked by naloxone and NBNI, and that was more greatly reduced by antisera directed against DYN(1–13) than against Menk (219). Nitrous oxide produced analgesia on the acetic acid writhing test that was markedly reduced in mice lacking the ORL-1 receptor; acetic acid induced greater increases in plasma ACTH in this KO mice (472). Ketamine produced analgesia in the electrical stimulation test in singly-housed, but not group-housed rats. Whereas ketamine decreased mu opioid binging in the hippocampus in group-housed animals, it increased mu opioid binding in the frontal cortex and hippocampus of singly-housed animals (72). Transfer of CD4+ T-cell lymphocytes blocked visceral hyperalgesia in mice that was reversed by naloxone methiodide (1156).

Intrathecal Tyr-d-Arg-Phe-Sar produced analgesia that was blocked by BFNA, naloxonazine and NBNI, but not Nti as well as by an antiserum directed against DYN B, but not DYN A, alpha-neo-endorphin, Menk or Lenk. Pro-DYN KO mice also showed a lesser analgesic response as well (786). Intrathecal Tyr-d-Arg-Phe-Sar produced analgesia that was blocked by selective mu-1, but not mu-2 opioid peptide antagonists (787). Inhibition of fatty acid amide hydrolase produced naloxone-reversible analgesia on the spinal nerve ligation and mild thermal injury tests (181). Naloxone-sensitive analgesia on the acetic acid and formalin tests was observed following (+/−)-cis-(6-ethyl-tetrahydropyran-2-yl)-formic acid (728). Analgesic effects induced by automatically controlled rotating acupuncture in rats were blocked by naloxone (583). Acupuncture inhibited in a naloxone-reversible manner the ability of noxious tooth pulp stimulation to induce Fos expression in the trigeminal nucleus subcaudalis, the transitional region between the subnucleus caudalis and subnucleus interpolaris, the inferior olivary nucleus, the ventrolateral and centrolateral thalamic nuclei, SON, PVN, NTS and rostral ventrolateral medulla (539). Analgesia, but not freezing behaviors elicited by electrical stimulation of the ventrolateral PAG was blocked by naltrexone, whereas midazolam blocked both behaviors (272). The NE and 5HT reuptake inhibitor, venlafaxine produced analgesia that was blocked by high naloxone doses, but not by N-omega-nitro-L-arginine (426). Mu, but not delta and kappa opioid receptor antagonists block interferon-alpha-induced analgesia from the thalamic nucleus submedius (1186). Tolerance to nicotine-induced analgesia occurred faster in MOR KO mice, nicotine-induced tolerance decreased MOR in C/P and the core and shell of the NAC, but not the spinal cord (377).

The visceral analgesic effects of the anti-viral drug, ribavirin were partially blocked by naloxone, enhanced by D2 receptor antagonists, and unaffected by alpha-2 NE or Ach antagonism (1). Whereas carrageenan produced hyperalgesia in control animals, this procedure produced naloxone-reversible analgesia in rats deficient in polyamine (325). NSAID drugs produce analgesia in sheep that are blocked by intrathecal naloxone or atipamezole (692). Artemin reduces herpes-related pain responses in mice inoculated with herpes simplex, an effect associated with decreased spinal DYN, but not BEND (41). Analgine, ketrolac and xefocam each produced analgesia that developed tolerance and cross-tolerance with morphine, but was not blocked by naloxone (1135).

The pyrazolyl-thiazole derivative (B50) produced naloxone-sensitive analgesia on the writhing, but not tail immersion test with removal of the methyl group or substitution of a bromo group resulting in loss of analgesia (924). Moderate hypercapnia reduced nociceptive behaviors on the formalin test in a naloxone-reversible manner, but failed to alter arcuate BEND (372). Exposure to an extremely low frequency magnetic field for four days produced analgesia and increased hypothalamic BEND and SP as well as brainstem 5HT (62).

Mu opioid antagonism is observed for the analgesic and GI transit inhibition induced by 7-hydroxymitragynine, isolated from Thai herbal medicine Mitragyna speciosa (746).

Analgesia induced by the ethanolic extract from the flowers of Combretum leprosum was observed on formalin-, capsaicin- and glutamate-induced pain tests, effects reversed by naloxone, 5HT1A and 5HT2A antagonists, but not by NO precursors or 5HT3 antagonism (909). The antinociceptive effects of Thymus broussonetli Boiss extracts on the formalin test in mice and rats were blocked by naloxone (310). GDHCF, the active ingredient of the stem bark of Hintonia standleyana increases hot-plate latencies and decreases acetic acid-induced writhing, effects blocked by naloxone, L-NAME and glibenclamide (270). Analgesia on the formalin and hot-plate tests induced by an extract of echium amonenum Fisch & C.A. Mey was blocked by naloxone (460).

The tonic and phasic analgesic, anti-depressant and hypothermic effects of the aqueous root extract of Securidaca longepedunculata were blocked by naloxone pretreatment (7). The antinociceptive effect of Hyptis pectinata leaves extract was blocked by naloxone (676). Attenuation of visceral nociception by alpha- and beta-amyrin, a triterpenoid mixture isolated from the resin of Protium heptaphyllum produced naloxone-reversible actions in mice (668). Further, oleanolic acid, a pentacyclic triterpene attenuated mustard oil-induced colonic nociception, an effect blocked by naloxone, but not yohimbine (714). Mechanical allodynia and thermal hyperalgesia caused by chronic constriction injury were blocked by processed Aconiti tuber in a NBNI-sensitive, but naloxone-insensitive manner (1247).

A recombinant herpes simplex vector encoding human Pro-Enk to macaque feet induced Enk peptide production in the application area that resulted in a long-acting anti-hyperalgesic and analgesic response to C-fiber stimulation of the application area (1271). Irradiation of osteolytic sarcoma cells improved pain-related behaviors in a manner similar to ketorolac; decreases in spinal pain mediators including DYN and COX-2 were noted (1168).

3d-ii. Opioid-insensitive analgesic responses 

Intrathecal, but not ventricular administration of neuropeptide W-23 or Neuropeptide B, endogenous ligands of GPR7 reduced mechanical allodynia induced by sciatic nerve ligation in a naloxone-insensitive manner (1256). Intrathecal CART (55–102) attenuated hyperalgesia and allodynia in as mouse model of neuropathic, but not inflammatory pain in a naloxone-independent manner (259). Intrathecal sensory neuron-specific receptor agonists, bovine adrenal medulla 8–22 and (Tyr6)-gamma2-MSH-6-12, inhibit formalin-induced nociception and neuronal Fos-like immunoreactivity in the spinal cord in a naloxone-insensitive manner (199). Whereas mGlu1 and mGlu5 antagonist co treatment enhanced inhibition of formalin-induced pain, neither antagonist paired with morphine showed enhanced effects on this measure (1009). L-glutamate sodium in the PVN enhanced acupuncture analgesia, an effect abolished by VP antisera, but only mildly affected by naloxone. PVN L-glutamate sodium increased VP, but not OXY, Lenk, BEND or DYN A(1–13) concentrations (1262).

Analgesic activity of a non-peptide imidazolidinedione somatostatin agonist was not reversed by naloxone (521). Although morphine and the 5HT3 agonist, m-CPBG, each reduced phase 1 and phase 2 nociceptive responses on the formalin test, the former was unaffected by intrathecal 5HT3 antagonism, and the latter was unaffected by intrathecal naloxone (1275). The 5HT1A receptor antagonist, WAY-100635 administered systemically or in the RVM attenuated mechanical hypersensitivity in rats with spinal nerve injury in a 5HT-1A antagonist sensitive, but not naloxone-sensitive manner (1208). The SSRI, fluvoxamine produced stronger analgesic effects than fluoxetine or citalopram, but all were naloxone-insensitive (1000). The antinociceptive activity of chemical congeners of improgan optimizing side chain length was nonpioid (484). Delta9-THC and R-methanandamide produced analgesia on the phenyl-p-quinone stretch test in mice that was not blocked by opioid antagonism (437). Analgesia mediated by the TRPM8 cold receptor in chronic neuropathic pain is mediated by Group II/III metabotropic glutamate, but not opioid receptors (925). Vinpocetine and piracetam exerted antinociceptive effects in visceral pain that was potentiated by naloxone (2).

The antinociceptive effects of the flavenoid myricitrin following glutamate, phorbol myriusate acetate, bradykinin and acetic acid, but not epinephrine or prostaglandin E(2), were blocked by the NO precursor, L-arginine, but not naloxone (768). A nonpioid pharmacological profile was observed for the novel analgesic M58996 in rat models of persistent and neuropathic pain (12). The antinociceptive effect of thalidomide on zymosan-induced experimental articular incapacitation was insensitive to naloxone antagonism (1142). Mice deficient in CD26, a multifunctional cell surface glycoprotein showed decreases in endopeptidase activity, increased SP, normal endomorphin-2, and short nociceptive latencies that were normalized by a SP NK1 antagonist or dipeptidyl-peptidase activity (424). A long-form alpha-neurotoxin from cobra venom produced potent opioid-independent analgesia (203). Analgesia induced by crotoxin isolated from crotalus durissus terrificus venom was opiate- and Ach-independent (1305). 12-acetyoxyhawtriwaic acid lactone, a diterpene from Egletes viscose, attenuated capsaicin-induced ear edema and hindpaw nociception in mice through a naloxone-independent response (764). The anti-inflammatory and analgesic effects of bee venom acupuncture were blocked by yohimbine, but not naloxone (50). Two compounds derived from bis selenide increased pain thresholds in the acetic acid, capsaicin and tail-flick tests in a naloxone-insensitive manner (993).

4. Stress and Social Status

In comparison with previous years, the amount of studies investigating the role of stressors in analgesic responses has dramatically declined as the use of selective opioid agonists and antagonists as well as the use of animals with knockouts of opioid receptor genes in analgesic processes has dramatically increased. This prompted us to include Section 3b in the present review. This section will continue to examine the phenomenon of stress-induced analgesia (4a), emotional responses in opioid-mediated behaviors (4b), and opioid involvement in stress response regulation (4c). 

4a. Stress-induced analgesia

One major theme of stress-induced analgesia is to examine its role vis a vis the opioid system. Ecologically relevant, parametric and sex/age variables continue to play an important role in the neurobiological substrates mediating these responses. Female rats placed on an exercise schedule on running wheels displayed decreased sensitivity to analgesia elicited by morphine, levorphanol, buprenorphine and butorphanol with the sensitivity restored upon return to sedentary conditions (1048). Mice placed on a restricted feeding schedule displayed more potent analgesic effects to morphine during that time in the light:dark schedule in which feeding was entrained; this effect was not observed in adrenalectomized mice, suggesting a role for endogenous glucocorticoid secretion (1277).

Increased opioid mediated warm water swim analgesia, but no changes in CWS analgesia were observed in mice lacking the proneuropeptide convertase PC2 (247).

CWS, but not fentanyl analgesia was blocked by an anti-neural-cell-adhesion molecule that in cultured DRG neurons, was reduced by BEND in a naloxone-reversible manner (485). Immobility induced by clamping the neck of mice produced a naloxone-reversible analgesia on a tonic pain test, the acetic acid writhing test (781). Rats repeatedly exposed to swim stress developed thermal hyperalgesia that was blocked by naloxone and naloxonazine, but not NTI or NBNI only if the antagonist was administered before each forced swimming episode (1075). Chronic restraint stress induced increased responsiveness in formalin-induced pain in the temporomandibular joint and decreases in the effectiveness of morphine analgesia. In contrast, fluoxetine produced greater analgesic effects in the stressed group (382). Swim stress-induced analgesia on the hot-plate test observed in wild-type male and female mice was reversed in MOR-KOR-DOR KO mice and in MOR-DOR KO females. Swim stress-induced analgesia delayed front- and hindpaw behaviors on the hot plate test, increased tail-flick latencies and increased ACTH and corticosterone levels in both wild-type and KO females (236).

4b. Emotional responses in opioid-mediated behaviors

Anxiogenic responses were elicited by CFA or sciatic nerve ligation as well as by ventricular administration of mu or delta antagonists or DYN. CFA or neuropathic pain decreased [35S]GTPgammaS binding induced by DAMGO or SNC80 in amygdala neurons, but increased this response following kappa agonists in the amygdala (812). Ventricular morphine produced anxiolytic effects on the elevated plus maze that was mediated by vasopressinergic system and nitric oxide pathways (544). Prenatal exposure to unpredictable shock enhanced subsequent morphine-induced CPP and greater depressive behaviors on the forced swim test; these behaviors were blocked by postnatal exposure to an enriched environment (1261). Morphine and naloxone administered into the inferior colliculus respectively decreased and increased defensive attention, immobility and escape behavior thresholds. Whereas morphine stimulated bicuculline-induced increases in fear thresholds when administered into the inferior colliculus, naloxone reduced those thresholds (161).

Lateral, but not medial septal injections of morphine produced naloxonazine-sensitive reductions in open-arm exploration in the plus maze, and reduced head-dipping frequency in the hole board, effects associated with decreased Fos in ventral septum, dorsal hippocampus and anterior hypothalamus (647). Morphine, U50488H and SNC80 decreased rearing and climbing activity in the staircase paradigm with naloxone blocking the first and second, but not third agonist. Morphine produced analgesia on the hot-plate test, but failed to alter anxiety-related behaviors on the four-plate test. In contrast, alprazolam and diazepam reduced anxiogenesis, but failed to alter hot-plate latencies (957). CRF produced anxiogenesis, locomotion and grooming behaviors that were reduced by low doses of chronic morphine and increased by high doses of chronic morphine (104). Naltrexone microinjections into the central, but not the basolateral amygdala blocked the anxiolytic effects of diazepam in the plus maze (151).

There was an atypical anxiolytic-like response to naloxone, but not selective opioid receptor subtype antagonists in benzodiazepine-resistant 129S2/SvHsd mice relative to C57BL/6JOlaHsd mice (965). Whereas Nti attenuated diazepam-induced anxiolytic effects in rats over expressing Pro-Enk in the amygdala, neither Nti nor BFNA altered diazepam-induced anxiolytic effects in normal animals (923). Social defeat stress-induced behavioral responses were mediated by the endogenous kappa opioid system such that NBNI and lack of the pre-pro-DYN gene blocked this immobility and subsequent cocaine-induced preferences in stressed mice (759). OFQ/N produces anxiolytic-like effects in the elevated plus maze and in the conditioned defensive burying test in rats that was blocked by the ORL-1 antagonist, UFP-101 (1169). Coupling of ORL-1 receptors to G proteins was decreased in the NAC of anxious relative to non-anxious mice (644). The anxiogenic, hyperactive and impairments in spontaneous alternation induced by buprenorphine were impaired by high doses of clorazepate; buprenorphine-induced memory impairments were unaffected (643).

Combinations of nalorphine and naloxone reduced climbing and rearing as well (779). SNC80 produced anxiolytic effects that were blocked by NTI, whereas Nti produced anxiogenic behaviors (898). Aged mice displayed increased anxiety-related behaviors on the light-dark and elevated plus-maze tests that were accompanied by increased astrocytes in the cingulated cortex due to dysfunction of DOR systems (816). Naltrexone reduced the reinforcing effects of ultraviolet preference in subjects who were frequent tanners (558).

4c. Opioid involvement in stress response regulation

Centrally-administered OFQ/N activates the hypothalamic-pituitary axis through and ORL-1 receptor-mediated up-regulation of CRF and POMC mRNA and stimulation of corticosterone release in rats (642). Prenatal morphine exposure blocks prenatal stress-induced up-regulation of mineralcorticoid and glucocorticoid receptors in the hippocampus in male, but not female rats. However, prenatal morphine exposure blocks prenatal stress-induced decreases in mineralcorticoid binding in the hippocampus of diestrous females, and glucocorticoid binding in the hypothalamus of estrous females (953). Blood-brain barrier increases in protein tracers were observed by the second day of morphine withdrawal that was accompanied by stress symptoms, whereas acute methamphetamine produced extravasation of endogenous serum protein (1020); the former effect was blocked by the 5HT modulator, AP-267 (1021). Naloxone and CRF increased deoxycorticosterone levels in cynomogolus monkeys (918).

Both in vivo restraint in carp and in vitro administration of carp interleukin-1beta stimulated AMSH and n-acetylated BEND from the pituitary (774). Graded exercise increased plasma BEND levels in both young and old horses, whereas it failed to alter plasma cortisol levels in old horses (717). Whereas baseline Enk and DYN levels in the hippocampal dentate gyrus were lower in the Flinders Sensitive relative to the Flinders Resistant line of mice, running failed to affect opioid levels in the former group, but decreased dentate DYN mRNA in the latter group (102). The increases in ProEnk mRNA in developing Syrian hamsters by glucocorticoids are dependent on age and adrenal versus brain tissue (358).

Plasma BEND, ACTH and cortisol responses were all higher and similar for arm and leg exercises at two exercise intensities in untrained subjects (725). Human exercise increased plasma Pro-Enk peptide F immediately and 15 min later, with peptide content in white and red blood cell layers increased after 15 min (153). Naloxone enhanced the ability of heat to induce exhaustion during high intensity exercise, an effect accompanied by increased in plasma BEND and ACTH. In contrast, cold attenuated these naloxone-induced effects (39). A polymorphism in MOR at A118G enhanced increased cortisol responses to naloxone, but lowered increased cortisol responses to social stress in volunteer subjects (211).

Greater levels of basal cortisol, but not ACTH were found in healthy subjects with a 118G allele in exon 1 of the MOR gene (68). Male subjects had greater HPA axis responses to a psychological stressor, whereas female subjects had greater cortisol reactivity to a naloxone challenge (1139). Interference with pre-pro-orexin, but not co-localized pre-pro-DYN in the perifornical hypothalamus reduced rapid eye movement sleep (192). Basal release of opioids appears not to affect eNOS expression in normal in vitro culture conditions, but might do so under stress (1006). Preoperative extradural bupivacaine and morphine lowered the cortisol stress response in dogs undergoing femoral-tibial joint surgery (1036). The ability of methanolic, ethanolic and aqueous extracts of Spondias mombin L to reduce novelty-induced rearing behaviors was insensitive to naltrexone (46).

5. Tolerance and Dependence

The most-often studied variables in the functional analysis of opioid-mediated responses next to analgesic processes are the underlying neurobiological roles of tolerance and dependence. This has continued unabated through the years, and continues to be a focus in this review. Developments will be reviewed for animal models in tolerance (Section 5a), and animal models in dependence and withdrawal responses (Section 5b).

5a. Animal models in tolerance

This section will be divided into the following sub-sections: (i) cellular effects on morphine tolerance, (ii) organismic effects on morphine tolerance, (iii) opioid effects on morphine tolerance, (iv) peptide-transmitter effects on morphine tolerance, and (v) other forms of opioid tolerance.

5a-i. Cellular effects on morphine tolerance 

A review (402) analyzes the post-opioid receptor adaptations to chronic morphine focusing on altered functionality and associations with signaling molecules. Another review (53) indicates a crucial role for PKC in MOR desensitization and morphine tolerance. Adenylyl cyclase superactivation induced by long-term morphine, etorphine or methadone is dependent on receptor localized within lipid rafts and is independent of receptor internalization (1315). A proteonomic analysis of rat cerebral cortex, hippocampus and striatum after pellet exposure to morphine revealed 26 distinct proteins that were differentially expressed (94). DeltaFosB over expression in the mouse NAC produced faster development of morphine tolerance and dependence, less morphine analgesia and increased sensitivity to morphine-induced reward through a DYN-sensitive system (1284).

Chronic treatment with fentanyl produces a different pattern from morphine in altering internalization or resensitization of MORs in the spinal cord under a pain-like state (501). Whereas morphine and buprenorphine produce tolerance and locomotor sensitization with non-internalization of MOR, etonitazene, which produces rapid endocytosis of MOR shows weak tolerance and a lack of locomotor sensitization (416). Acute and chronic morphine respectively increased and decreased neuroglycan C levels in the NAC, striatum, hippocampus, VTA and amygdala (505). Increased phosphorylation of neurogranin, protein kinase C and Ca2+/calmodulin dependent protein kinase II occurs in opioid tolerance and dependence (1035).

Chronic morphine treatment increased the expression of the neural cell adhesion molecule in the dorsal horn of the mouse spinal cord (1091). Chronic morphine-mediated adenylyl cyclase superactivation is attenuated by the Raf-1 inhibitor, GW5074 (1283). The Gbetagamma that interacts with adenylyl cyclase in morphine tolerance originated from a Gs protein (1184). Chronic morphine treatment also increased the expression of vesicular glutamate transporter 1 in the mouse spinal cord (1092). A subpopulation of dorsal horn neurons displays enhanced NMDA receptor function after chronic morphine exposure (1316). Chronic morphine, but not D-amphetamine or nicotine, elevated autoantibodies to the MOR-DOR receptor and to the AMPA Glu1, but not NMDA NR2 subunit receptors (414). Dexamethasone mimics the inhibitory effect of chronic pain on the development of tolerance to morphine analgesia and compensates for morphine induced changes in G protein gene expression (518). Long-term exposure of SH-SY5Y cells to morphine revealed variations in 45 proteins using a whole cell proteomic analysis (830). Inhibition of neuronal nitric oxide synthase antagonizes morphine antinociceptive tolerance by decreasing the activation of p38 MAPK in the spinal microglia (689).

Chronic morphine decreased proDYN gene expression in the striatum, but not NAC or hippocampus, whereas chronic oxycodone-6-oxime increased proDYN gene expression in the striatum and hippocampus, but not NAC. Chronic 14-methoxymetopon increased proDYN gene expression in the NAC and hippocampus, but not striatum (589). Chronic morphine decreases MOR signaling and reduces Ca2+ current density in sensory neurons (530). Chronic morphine increased lactate and myo-inositol, but decreased glutamate in thalamus and somatosensory cortex using nuclear magnetic resonance spectroscopy (1242). Intermittent, but chronic morphine alters protein expression in the NAC (654).

Whereas chronic morphine inhibited spinal GABA release, recurrent morphine withdrawal reversed this effect and elevated spinal glutamate levels. In contrast, acute morphine withdrawal increased both spinal GABA and glutamate release (303). Whereas chronic etorphine, but not morphine increased spinal G-protein-coupled receptor kinase 2, dynamin II, beta-arrestin 2 and phosphorylated-conventional PKC, chronic morphine, but not etorphine increased spinal glial fibrillary acidic protein. Manipulations that modulate glial fibrillary protein altered morphine-induced, but not etorphine-induced analgesic tolerance (820).

Whereas acute morphine increases adenylate cyclase II activity by Galphas or PKC stimulation regulated by Gbetagamma subunits, chronic morphine decreases adenylate cyclase II activity upon stimulation of the Galphas, but not PKC pathway (994). Down-regulation in dopamine transporter function in striatum occurred during acute morphine addiction and its abstinence in rhesus monkey (1243).

5a-ii. Organismic effects on morphine tolerance 

Antinociceptive tolerance was observed following cumulative intracranial microinjections of morphine into the PAG in the rat (795). Morphine analgesic tolerance in 129P3/J and 129S6/SvEv mice was largely absent because of differences in tail-flick responsivity between these strains and C57BL/6J mice (148). Acute morphine tolerance was observed in proestrous, but not ovariectomized female rats; chronic estradiol in ovariectomized animals reinstated acute morphine tolerance. Both forms of tolerance were blocked by the NMDA antagonist, memantine (1024). Chronic morphine drinking establishes morphine tolerance, but not addiction in Wistar rats (101). PKC and PKA inhibitors reinstate morphine-induced behaviors in morphine tolerant mice (1045).

Lithium chloride reduced the development and expression of morphine-induced tolerance, but only the development of morphine-induced dependence in isolated guinea pig ileum (18). Continuous naltrexone opioid receptor antagonism abolished continuous morphine analgesia and potentiated morphine hyperalgesia with the latter, but not former effect blocked by the NMDA antagonist, MK-801. After subsiding, morphine hyperalgesia could be reinstated by a low, but not high morphine dose as well as the morphine metabolite, M3G (541). The ability of processed Aconiti tuber to increase the duration of repeated morphine analgesia before inducing tolerance was prevented by both mu (C-CAM) and kappa (NBNI) opioid antagonists (1034).

5a-iii. Opioid effects on morphine tolerance 

Chronic morphine pretreatment reduced subsequent morphine-induced, but not heroin-induced, overdose lethality, and failed to form toxic concentrations of M6G (1072). OFQ/N KO mice or normal mice treated with the ORL-1 antagonist J113397 fail to display morphine tolerance when morphine is administered at the same dose. Escalating doses of morphine produce tolerance but not withdrawal symptoms in OFQ/N KO mice (220). Pro-Enk KO mice displayed a blunted morphine analgesic tolerance, but not changes in morphine-induced CPP or morphine-induced motor sensitization (731). Morphine-tolerant rats displayed increased DOR levels in varicosities that appose the postsynaptic membrane and interact with increased GAD. Delta agonists reduced morphine tolerance and significantly inhibited presynaptic GABA release (710). Attenuation of morphine tolerance and withdrawal syndromes was observed by co administration of nalbuphine (513).

Intrathecal administration of BAM 22 produced analgesia in both opiate-naïve and morphine-tolerant rats, resumed morphine analgesia in morphine-tolerant rats, but produced hyperalgesia following chronic administration (522).

5a-iv. Peptide-transmitter effects on morphine tolerance 

Calmodulin-stimulated adenylyl cyclase gene deletion reduced morphine-induced tolerance and dependence, but not acute analgesia (655). Glycogen synthase kinase 3beta and cyclin-dependent kinase 5 inhibitors abolished morphine-induced tolerance, but not analgesia (880). NRM administration of GABA-A receptor antagonists augmented the development of morphine tolerance, whereas NRM GABA-A agonists attenuated morphine tolerance development. The GABA-A receptor-mediated IPSC was significantly increased in morphine-tolerant animals, an effect blocked by PKA inhibition (709). Whereas co treatment with the 5HT1A agonist, 8-OH-DPAT in the dorsal, but not median raphe nucleus delayed morphine analgesic tolerance, the 5HT2 antagonist, ketanserin failed to exert effects in the same sites (829). The CCK-2 antagonist, LY225910 administered into the NAC blocked morphine-induced tolerance, but failed to affect the acute antinociceptive effect of morphine (1245).

Neuropeptide SF facilitated spinal morphine analgesia in normal and morphine tolerant animals (519). Supraspinal agmatine prevented the development of supraspinal morphine analgesic tolerance (592). The tricyclic antidepressant amitryptaline prevented morphine tolerance during its development and following a subsequent morphine challenge, and up regulated the spinal glutamate transporters GLAST and GLT-1 expression (1099). The NMDA antagonists MK-801 and CPP attenuated morphine tolerance in male, but not intact or ovariectomized female mice. MK-801 also facilitated morphine-induced hyperalgesia in male, but not female mice (147). There was an additive effect of dextromethorphan on the inhibitory effect of anti-NT4 on morphine tolerance (455). The glycine site-specific NMDA antagonist, (+)-HA966 reversed morphine tolerance and enhanced morphine antinociception in morphine-tolerant rats (5). Opioid tolerance was attenuated by AS targeting the regulator of PKC and Ca2+/calmodulin-dependent protein kinase II (1108) as well as by the antagonist, KN93, but not KN92. KN93 also prevented morphine dependence (1109). Inactivation of p38MAPK in spinal microglia inhibits morphine analgesic tolerance (254). Nicotine and morphine produced analgesic cross-tolerance that was blocked by the Ca2+ channel antagonists, nimodipine, diltiazem and flunarizine, but not verapamil (94).

The endothelin receptor antagonist, BMS182874 reinstated morphine-induced GTP stimulation in tolerant animals to acute levels, and reduced the EC50 value in tolerant cells to that like acute morphine (87). The induction of morphine hyperallodynia in infraorbital nerve-injured rats was blocked by the 5HT1A agonist, F13640 before or after allodynia induction; F13640 also blocked morphine CPP and naloxone-induced conditioned place aversion (230). Administration of combined calcium-magnesium gels, but not calcium alone attenuated the development of morphine tolerance, morphine dependence and morphine withdrawal (933).

5a-v. Other forms of opioid tolerance 

Chronic morphine administration resulted in tolerance to delta opioid receptor-mediated antinociception induced by Delt (922). Intrathecal administration of (−)-oxymorphone, but not (+)-oxymorphone up regulated spinal DYN, produced thermal and tactile hypersensitivity, and produced analgesic tolerance (386).

5b. Animal models in dependence and withdrawal responses

This section will be divided into the following sub-sections: (i) cellular effects on morphine dependence and withdrawal, (ii) organismic effects on morphine dependence and withdrawal, (iii) opioid effects on morphine dependence and withdrawal, (iv) peptide-transmitter effects on morphine dependence and withdrawal, and (v) other forms of opioid dependence and withdrawal.

5b-i. Cellular effects on morphine dependence and withdrawal responses 

Morphine dependence was associated with changes in cytoplasmic and mitochondrial enzymes including proteins belonging to GTPase and GST super families, ATPase, asparaginase or proteasome subunit p27 families (93). Genetic variants of the P-glycoprotein gene Abcb1b modulate opioid-induced hyperalgesia, tolerance and dependence (661). Inhibition of adenylyl cyclase II activity following chronic opioid agonist exposure and withdrawal is modulated by phosphorylation (995). Both adenosine and naloxone increased regional cerebral blood flow in morphine-dependent rats accompanied by elevated blood pressure and heart rate (572). Altered periodicity and expression of circadian clock gene, mPer1, occurred in mouse brain and kidney under morphine dependence and withdrawal (1193). Heroin produced oxidative stress and exogenous antioxidant-alleviated withdrawal responses (1246). MOR and orexin/hypocretin mRNA levels in the LH and striatum are enhanced by morphine withdrawal (1323). Opiate withdrawal induces dynamic expressions of AMPA receptors and its regulatory molecule CaMKIIalpha in hippocampal synapses (1320). Although acute and withdrawal-induced morphine and nicotine increased central and peripheral concentrations of pregnenolone, progesterone and allopregnanolone that was abolished by either adrenalectomy or gonadectomy, a challenge dose of morphine or nicotine 24 h after withdrawal failed to induce effects (233).

Agmatine decreased the calcium signal in morphine-dependent CHO cells by activation of IRAS, a candidate for imidazoline I1 receptor (1236). The non-competitive NMDA antagonist reversed the decreases in postsynaptic density proteins in the hippocampus of rat offspring of morphine-addicted mothers (1268). Naloxone-precipitated morphine withdrawal up regulated MOR labeling in NAC, C/P, medial-basal thalamus, basolateral and basomedial amygdala, and VTA in prepubertal male, but not female mice with the GABA B agonist, baclofen re-establishing MOR levels in the first three brain areas in males (281). Naloxone-precipitated, but not spontaneous withdrawal in morphine-dependent rats increased MAPK1/2 phosphorylation in cortex and striatum in a manner similar to acute sufentanil, morphine and SNC-80, but not U50488H; morphine tolerance did not change MAPK ½ activity (42). Consumption of low naltrexone doses in the drinking water of morphine-dependent rats decreased withdrawal signs and increased MOR mRNA expression in the NTS, but not the LC, VTA, frontal cortex, striatum or amygdala (1145).

Pregnant females treated with morphine pellets had offspring with higher morphine-induced and endothelin-1-induced G-protein stimulation (927). Acute and chronic morphine treatment respectively stimulated and decreased Na+, K+-ATPase activity with the latter effect augmented by naloxone precipitated withdrawal; these effects were inversely correlated with cAMP accumulation, and were modulated by PKA inhibitors and protein phosphatase inhibitors (1239). There is a lack of cAMP-specific phosphodiesterase 4 activation during naloxone-precipitated morphine withdrawal in rats (585). Chronic morphine increased NR1 NMDA mRNA in the central amygdala, and NR1 and NR2B NMDA protein in the NAC (55).

Persistent and reversible morphine withdrawal-induced morphological reductions in spine density were observed in the NAC shell, but not core (280). Enhanced Fos expression was observed in glutamic acid decarboxylase immunoreactive neurons of the mouse PAG during opioid withdrawal (433). Galanin attenuates CREB phosphorylation induced by chronic morphine and naloxone challenge in Cath.a cells and primary striatal cultures (457).

5b-ii. Organismic effects on morphine dependence and withdrawal responses 

A requirement for protein synthesis was proposed in the modeling of the onset of drug dependence (536). By adding a transient early phase and a delay of the buildup of protein, a modified ‘Joyce model’ of opioid dependence and withdrawal shows excellent fit of data (936). Genetic differences were noted among 12 strains of inbred mice in the ability of CFA to lower the ED50 of morphine analgesia, enhance morphine tolerance and alter physical dependence signs precipitated by naloxone (660). Global opiate withdrawal scores were evident 56 h later in female relative to male mice, but males display earlier signs of opiate withdrawal using paw tremors and wet dog shakes (876). Maternal deprivation specifically enhanced vulnerability to opiate dependence in rats (1150). Brief early handling increased morphine dependence in adult rats without differences in pre-pro-Enk or MOR levels in the limbic system (1151). Heroin self-administration respectively lowered and elevated ICSS thresholds in naïve and dependent rats. Naloxone increased heroin consumption and reversed heroin-lowering of ICSS thresholds in naïve rats, and increased heroin consumption and elevated ICSS thresholds further in dependent rats (568). Conditioned withdrawal is readily established to discrete cues associated with naloxone-precipitated withdrawal from acute, infrequent opioid exposure (30). Morphine withdrawal for 18 h and 4 days reduced subsequent morphine analgesia, but the latter response was restored 20 days after withdrawal; this effect was augmented by stress applied after 18 hours. RU38486, a glucocorticoid receptor antagonist increased morphine analgesia four days after withdrawal (294). Naloxone-precipitated withdrawal from morphine produced potentiated startle responses which on the first day were blocked by inactivation of the basolateral or central nuclei of the amygdala or BNST, but not NAC, and which on the second day were blocked by inactivation of the NAC, but not the basolateral or central nuclei of the amygdala or BNST (450).

Although the major human metabolite, 6beta-naltrexol, was as potent as naltrexone in precipitating withdrawal in opiate-dependent monkeys, it was far less potent than naltrexone in MOR binding, shifting alfentanil-induced analgesia to the right, and blocking opiate-induced respiratory depression and itching/scratching responses (598).

5b-iii. Opioid effects on morphine dependence and withdrawal responses 

Novel depots of buprenorphine suspended in sesame oil have long-acting ameliorating effects for the management of physical dependence to morphine in mice (684).

5b-iv. Peptide-transmitter effects on morphine dependence and withdrawal responses 

Whereas over expression of CREBGFP in the LC aggravated morphine withdrawal behaviors and sensitized the cAMP-signaling pathway, dnCREBGFP in the LC attenuated morphine withdrawal behaviors and hyperpolarized LC neurons (442). Mice lacking calcium-calmodulin-dependent protein kinase IV displayed less morphine analgesic tolerance, but no changes in acute morphine analgesia or physical dependence (600). Spinal modulation of CGRP by endocannabinoids participates in the development of opioid physical dependence (1131). ATP-dependent K+ channel blockade decreased nicotine-induced inhibition of withdrawal in morphine-dependent rats (1299). Morphine withdrawal responses were enhanced by inhibition of spinal phosphoinositide 3-kinase (1266). The D1 receptor agonist, SKF82958 was rewarding in morphine-dependent rats, and blocked naloxone-induced conditioned place aversions and somatic signs of opioid withdrawal while increasing P-Glu-R1, but not P-CREB in the NAC. Naloxone reduced SKF 82958-mediated P-GluR1 induction in morphine-dependent rats (183). Yohimbine pretreatment potentiated the efficiency of clonidine to decrease naloxone-precipitated opioid withdrawal signs (1073). Scopolamine attenuated such naloxone-precipitated withdrawal signs as jumping, writhing, weight loss, genital grooming, teeth chattering, ptosis, diarrhea and irritability (1241). Lack of neuropeptide Y attenuates the somatic signs of opiate withdrawal (864). RF9, a neuropeptide FF receptor antagonist, prevented opioid-induced tolerance associated with hyperalgesia (1039).

Cross-talk between NO and ERK1/2 signaling pathways in the spinal cord mediates naloxone-precipitated withdrawal in morphine-dependent rats (165). Conantoxins and variants derived from cone snail venom inhibit naloxone-induced withdrawal jumping in morphine-dependent mice (1209). The aqueous and methanolic, but not chloroform extracts of rhizome and aerial parts of Valeriana officinalis L reduced naloxone-induced jumping in morphine-dependent mice (1017). Withdrawal responses from morphine, DAMGO and U50488H as well as the induced Ach response are reduced by isoquinoline alkaloid derivatives from A. mexicana and A. constricta as well as papaverine (166).

5b-v. Other forms of opioid dependence and withdrawal responses 

Elevations in ICSS reward thresholds and somatic withdrawal signs are noted following spontaneous and naloxone-precipitated withdrawal from fentanyl (145). Caffeine paired with naloxone produced a quasi-morphine withdrawal syndrome in wild-type, but not adenosine A(2A)R(−/−) KO mice, although both groups had similar levels of striatal mu opioid receptors (98). MOR and KOR agonists antagonize icilin-induced wet-dog shaking in rats (1215).

6. Learning and Memory

Learning and memory effects of endogenous opioid peptides, their receptors, their agonists and their antagonists, as well as genetically altered animals continue to be studied extensively. Recent developments will be reviewed for animal models in CPP (Section 6a), conditioned aversion paradigms (Section 6b), drug discrimination and spatial learning (Section 6c), as well as memory and amnesia (Section 6d). 

6a. Opiates and conditioned place preferences (CPP)

The following sections examine opioid CPP, non-opioid effects upon opioid CPP, and opioid effects upon non-opioid CPP respectively.

6a-i. Opioid CPP 

CPP can be induced by both morphine and its metabolite, M6G in mice (1165). Rats that displayed high levels of seeking for novelty displayed greater morphine-induced CPP and increased morphine self-consumption (894). Prenatal morphine exposure during the time of opiate receptor appearance does not alter adult morphine-induced CPP or self-administration (952). Restraint, tail-pinch and social defeat in an agonistic encounter each produce the reinstatement of morphine-induced CPP that had been extinguished (950). Involvement of endogenous ligands for mu-, delta- and kappa-opioid receptors was observed in modulating morphine-induced CPP expression in rats (663). Morphine-induced CPP was increased by combinations of methyl-6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate and restraint stress in out bred Sprague-Dawley and inbred Fischer 344 rats, yet this treatment decreased morphine-induced CPP in inbred Lewis rats (412). Naloxone, but neither proglumide nor MK-801, altered effects of morphine pre exposure on morphine-induced taste aversions to saccharin (355). Artificial rearing produced a greater morphine-induced CPP as adults than maternal rearing (694). There were differential changes in CREB in rat hippocampus, prefrontal cortex and NAC during the three phases of morphine-induced CPP in rats (1322).

6a-ii. Non-opioid effects on Opioid CPP 

Inhibition of the ERK pathway or protein synthesis during re-exposure to morphine or cocaine erases previously learned place preferences (1143). Administration of protein synthesis blockers such as anisomycin or cycloheximide into the hippocampus, amygdala, or NAC, but not the VTA after a conditioning session blocked a morphine-induced CPP which did not return after further conditioning (777). However, another study (1272) failed to observe any effect of the protein synthesis inhibitor, anisomycin in the basolateral amygdala following retrieval to impair expression of morphine CPP. Intrahippocampal inhibition of protein kinase A attenuates morphine-induced conditioned place preference (1018).

Inhibition of the cyclooxygenase pathway attenuated morphine-induced conditioned place preference in mice (398). D1/D2 dopamine receptor antagonists and D3 dopamine receptor agonists block foot shock stress-induced enhancement of morphine conditioned place preference (257). Mice lacking dopamine beta-hydroxylase fail to show morphine-induced CPP or hyper locomotion with viral restoration of dopamine beta-hydroxylase in the NTS, but not LC restoring morphine-induced CPP; the enzyme in both sites partially restores hyper locomotion (865). CCK B receptors in the NAC are necessary for the expression of morphine conditioned place preference by acting through D2 dopamine receptors (783). Ultra-low doses of nicotine reduce the expression of morphine-induced conditioned place preference in mice that was reversed by hexamethonium (1013). Theophylline inhibits tolerance and sensitization induced by morphine in a conditioned place preference paradigm (981). The too few mutant zebra fish displays food-induced, but not morphine-induced CPP (630). Administration of l-threo-3,4-dihydroxyphenylserine or carbidopa restored morphine-induced CPP in DBH KO mice that was then blocked by NK1 receptor antagonism (517). Corticosterone suppression can affect in the absence of an inescapable stressor block a morphine-induced CPP and induced changes in NAC DA, indicating that corticosterone is involved in the expression, but not the induction of stress-induced sensitization of morphine CPP (277).

Microinjection of D1 (SCH39166) or D2 (L-sulpiride) antagonists into the NAC shell, and to a lesser degree, the NAC core impaired the acquisition, but not the expression of a single-trial morphine-induced CPP (342). Repeated administration of apomorphine, SKF38393, quinpirole and sulpiride into the NAC enhanced morphine-induced CPP, whereas repeated NAC SCH23390 reduced morphine-induced CPP (1293). Ifenprodil, a NR2B subunit of the NMDA receptor antagonist, blocked CPP induced by morphine, but not natural rewards or social interaction. CPP induced by morphine, but not natural rewards or social interaction augments NR2B NMDA expression in the NAC and hippocampus (711). Intra-ventral pallidal glutamate antagonists block expression of morphine-induced CPP (258). The glycine site/NMDA receptor antagonist, MRZ2/576, reduced the acquisition and expression of CPP and locomotor activity induced by morphine in mice (1330). Ventricular administration of a neuropeptide FF agonist blocked the acquisition of a morphine CPP (724). Agmatine enhanced CPP induced by low morphine doses, an effect augmented by L-arginine, a NO precursor and blocked by L-NAME and aminoguanidine, NOS inhibitors (574); this potentiation was also modulated by alpha2-adrenoceptors (1098). An inhibitory effect of paeonol on morphine-induced locomotor sensitization and conditioned place preference was observed in mice (328). Mice that either lacked the pre-pro-orexin gene or received the orexin antagonist, SB334867A into the VTA failed to display morphine CPP or locomotor sensitization while reducing morphine-induced increases in DA levels (819). Prenatal cocaine exposure blocked morphine CPP as well as increased threat, avoidance and fleeing during a social encounter after isolation (326). Post-training and post-reactivation administration of amphetamine enhances morphine CPP (103).

6a-iii. Opioid effects on non-opioid CPP 

The kappa and partial mu opioid agonist, nalbuphine blocked morphine-induced CPP and increased NAC DA metabolites, but failed to affect morphine-induced locomotor sensitization (1110). Glycyl-glutamine, an inhibitory dipeptidyl synthesized from BEND(1–31) inhibited the acquisition and expression a nicotine CPP, and blocked acquisition of a conditioned place aversion induced by chronic nicotine paired with mecamylamine, but not one paired with the kappa agonist, U50488H (404).

6b. Opiates and conditioned aversion paradigms

Morphine administration 20 h prior to naloxone produced greater conditioned place aversion than naloxone alone and better than morphine-6-beta-naloxol pairing, implicating constitutively active mu receptors in this process (1032). The glutamate release inhibitor riluzole attenuated the formation of conditioned place aversion induced by naloxone in rats undergoing a single morphine exposure (525).

6c. Opiates and drug discrimination and spatial learning

The VTA and PAG appear more critical than the parabrachial nucleus in mediating antagonist-induced disruptions of the discriminative stimulus effects of systemic morphine for food reinforcement in both sexes of rats (613). Olfactory repeated discrimination reversal in rats was impaired by morphine but only at doses that affected performance of the well-learned performance discrimination (380). Discriminative-stimulus effects of methamphetamine and morphine are attenuated by cAMP-related compounds (1259). The NMDA antagonist, LY235959 produced additive and supra-additive effects on schedule-controlled responding when paired with butorphanol and nalbuphine, but only additive or sub-additive effects when paired with morphine or buprenorphine. LY235959 potentiated analgesia induced by all opioid agonists (349). Chlordiazepoxide and dizocilpine, but not morphine, selectively impair acquisition under a novel repeated-acquisition and performance task in rats (913). The NMDA antagonists, ketamine, phencyclidine and MK-801 generalized to the discriminative stimulus effects of U50488H, but not TRK-820 in rats. (800). Old mice lacking liver-derived IGF-I had impairments in the acquisition of spatial memory that were accompanied by increased hippocampal Enk and DYN immunoreactivity but lower mRNA levels of the opioid peptides in the hippocampus (1094).

6d. Opiates and memory

Mu and kappa, but not delta opioid receptor agonists in the NAC regulate attentional learning in the blocking paradigm (504). Odor paired with tactile stroking in rat neonates produced odor preferences, and the acquisition and consolidation of this response was blocked by naltrexone (974). Odor-shock pairings in neonatal animals produce an odor preference that activates c-Fos in the granule cell layer of the olfactory bulb and the anterior piriform cortex. Post-training naltrexone treatment in the amygdala turned the odor preference into an aversion, blocked Fos expression in the piriform cortex, and activated Fos expression in the central nucleus of the amygdala (973). Beta-casomorphin-5, a mu-opioid agonist from bovine milk ameliorated scopolamine-induced impairments in spontaneous alternation behavior and passive avoidance in a mu- and mu1-antagonist sensitive manner (984). Morphine-induced state-dependent memory of passive avoidance was blocked by centrally-acting muscarinic and nicotinic antagonists, but not peripheral nicotinic antagonists. Centrally-acting, but not peripherally-acting anticholinesterase drugs enhanced morphine-induced retrieval (510). Morphine-induced amnesia in passive avoidance was decreased by L-arginine, whereas L-NAME and low morphine doses produced amnesia. Apomorphine inhibited morphine-induced amnesia, and the L-NAME-induced inhibition of morphine amnesia was blocked by DA D1 or D2 receptor antagonists (949).

Morphine-induced deficits in passive avoidance learning were ameliorated in part by treadmill running exercise (14). Opioid receptor antagonism in the PAG prevented blocking of prediction errors during Pavlovian fear conditioning (762). Central, but not peripheral opioid receptor antagonism alleviated retrieval of infant fear memories in developing rats (1202). Big DYN increased step-through latencies on a passive avoidance task, increased locomotor activity and increased anxiolytic behaviors on the open field and elevated plus maze with all effects blocked by MK-801. In contrast, DYN A and DYN B increased memory through a NBNI-dependent mechanism, and failed to affect locomotor or anxiolytic responses (621).

Mecamylamine-induced reductions in passive avoidance responses were blocked by DYN A (1–13) and DYN A (2–13) with the latter effect insensitive to NBNI antagonism. Mecamylamine-induced extracellular Ach decreases were also abolished by DYN A(2–13) in a kappa antagonist-insensitive manner (474). Ventricular administration of D1 and D2 agonists as well as D2 antagonists on a passive avoidance test day reversed morphine’s pre-test amnestic effects, whereas central D1 antagonism prevented the restoration of passive avoidance responses by morphine administration on the test day (1294). Central L-glutamate and MK-801 administered prior to a passive avoidance test respectively reversed and enhanced amnesia induced by pretraining morphine, and respectively increased and decreased morphine state-dependent learning (1296). Morphine state-dependent learning sensitization was decreased by pretreatment with L-NAME and L-arginine (1292). The CB1 agonist, WIN55,212–2, mimicked the ability of pre-test morphine to restore passive avoidance learning, and enhanced morphine’s ability as well; the CB1 antagonist AM-251 blocked morphine’s and WIN55,212–2’s memory-restorative effect (1297). Histamine administration prior to passive avoidance training decreased learning, whereas pre-test histamine reversed this amnestic effect. Mice sensitized with morphine or apomorphine exhibit reversals of histamine’s pre-training amnestic effects with morphine’s effects blocked by naloxone, SCH23390 or sulpiride (1298). Nicotine reversed morphine-induced amnesia and morphine-induced state-dependent learning, effects blocked by atropine (1300).

7. Eating and Drinking

This section will review ingestive effects as functions of opioid agonists (Section 7a), opioid antagonists (Section 7b), and the interaction of POMC-derived peptides (Section 7c). 

7a. Opioid agonists and ingestive behavior

Whereas DAMGO in the NAC increased consumption of two types of flavored food pellets, intake of the preferred pellet in a choice test was respectively enhanced and reduced by DAMGO and naltrexone in the NAC; systemic naltrexone had no effect (1230). DAMGO-induced increases in high-fat feeding were blocked by naltrexone and muscarinic antagonism, but not by antagonists of DA, glutamate or nicotinic receptors (1218). Administration of morphine, but not DAMGO, Dyn or U50488H into the rostral lateral hypothalamus induced feeding, whereas naltrexone into the same site elicited c-Fos activation (652).

Whereas bicuculline administered into the ventral pallidum increased saccharin, but not quinine or water in water-deprived rats, pallidal muscimol and SCH23390 reduced this response. Pallidal DAMGO initially suppressed, but then stimulated saccharin intake in water-deprived rats (1029). DAMGO increased fat preference and increased food intake in Osborne-Mendel fat-preferring rats, but only switched preference in S5B/PI fat-resistant rats. Osborne-Mendel rats also had increased MOR and MOR mRNA in the arcuate nucleus (65). Morphine time-dependently prevented acetic acid-induced writhing and the suppression of palatable food intake; haloperidol inhibited the former, but not latter response (1069). Mice with deletion of the DA transporter and thus elevated levels of DA enhances their tendency to work for a food reward without affects on Pavlovian or operant learning for that reward; increased DYN in the C/P and in the core, but not shell of the NAC were observed (156). Chronic infusion of OFQ/N increases food and ethanol intake in alcohol-preferring rats (224). Whereas OFQ/N prolongs feeding induced by food deprivation by decreasing activity of AMSH neurons involved in feeding termination, AMSH at doses capable of reducing deprivation-induced feeding, failed to affect OFQ/N-induced feeding (109). Novel ORL-1 agonists, OS-500 and OS-462 produced greater degrees of feeding than OFQ/N following ventricular, but not systemic administration, an effect blocked by the ORL-1 antagonist, UFP-101, but not NC-797 (308).

The ability of the CRF2 receptor agonists, urocortin II and III to inhibit food intake was unaffected by OFQ/N (338). OFQ/N-induced feeding was respectively stimulated by muscimol and inhibited by bicuculline in cockerels (1100). BEND KO mice displayed increased systolic blood pressure, increased MPOA c-Fos activity and higher levels of urinary epinephrine secretion following a high-sodium diet relative to heterozygotes and controls (155). Preoperative epidural morphine together with postoperative transdermal fentanyl restored normal behavior and weight gain in pigs receiving abdominal surgery (715).

7b. Opioid antagonists and ingestive behavior

Increased body weight was observed in MOR KO mice, an effect accompanied by increased hypothalamic NPY mRNA (445). A diaryl ether derivative, (6-(4-{[(3-methylbutyl)amino]methyl}phenoxy)- nicotinamide, capable of displacing MOR, KOR and DOR agonists, suppressed food intake and weight gain in obese rats and mice, but failed to do so in MOR KO mice (1306). Naloxone decreased palatable food intake and increased latency to feed in sated, but not food-restricted rats without affecting food-anticipatory activity (63). Administration of BFNA or CTAP into the dorsal caudomedial shell of the NAC significantly reduced sucrose intake, but did not alter overall chow intake or body weight in rabbits, and also produced specific losses of MOR coupling to their G-proteins (1197). Chronic prevention of MOR G-protein coupling by BFNA in the pontine parabrachial nucleus persistently decreased consumption of standard but not palatable food (1198).

CPPs induced by exposure to either high-sugar (Fruit Loops) or high-fat (Cheetos) snack foods were each dose-dependently suppressed by naltrexone (516). Food-induced behavioral sensitization that conditioned activity was cross-sensitized with cocaine and morphine, and blocked by naltrexone and noncompetitive AMPA antagonism (648). Naloxone failed to reduce sucrose intake in mice with Enk deleted, with BEND and Enk deleted, and with DYN deleted, but showed normal inhibitory responses in mice with only BEND deleted (458). Naltrexone reduced sucrose intake, particularly in a group of rats exposed to a consummatory contrast paradigm and only for the relatively more valuable sucrose solution (1097). Naloxone reduced the ability of orexin and NPY to increase saccharin intake (374). Two studies by the same laboratory showed that central naloxone blocked the ability of central administration of interleukin-1beta to suppress water and salt intake in fluid-deprived and sodium-depleted rats induced by hypernatremia and hypovolemia; interleukin-1beta failed to affect saccharin intake (269, 708).

7c. POMC-derived peptides and ingestion

Anticipation of ingestion of a corn oil emulsion increased hypothalamic POMC and orexin mRNA expression for up to 30 min of presentation and until ingestion (789). Mice lacking 11 beta-hydroxysteroid dehydrogenase type 1 show lower levels of arcuate CART and MC-4 receptors, but higher levels of melanin-concentrating hormone, and when on a high-fat diet display an up-regulation of AGRP concomitant with high-fat-induced hyperphagia (274). Starvation in Xenopus laevis produced significant reductions of immunopositive Menk, CART, Fos and urocortin-1 in the hypothalamic magnocellular nucleus accompanied by an increase in CRF (159).

Electroacupuncture produced greater decreases in weight loss and serum leptin levels, coupled with greater increases in serum BEND than diet restriction in female obese subjects (154).

8. Alcohol and Drugs of Abuse

The interaction between opiates and other drugs of abuse, particularly alcohol, continues to be a vigorous area of investigation. This section is organized into a consideration of how the opioid system works in the general area of drugs of abuse (Section 8a), in opiate self-administration (Section 8b) and in interactions with ethanol (Section 8c), THC (Section 8d), stimulants such as cocaine and amphetamine (Section 8e) and other abused drug classes (Section 8f). 

8a. Opiates and drugs of abuse: reviews

New challenges and opportunities are reviewed in managing substance abuse in Malaysia (751).

8b. Opiates and self-administration studies

8b-i. Animal studies 

Disrupting reconsolidation of conditioned withdrawal memories in the basolateral amygdala reduces suppression of heroin seeking in rats (463). A novel bivalent morphine/heroin vaccine prevented relapse to heroin addiction in rodents (34). Unlimited access to heroin increased daily self-administration across the circadian cycle with decreased levels of food intake occurring across the circadian cycle with daytime intake increasing meal frequency and nocturnal intake displaying smaller and briefer meals; this reduced body weight gain (196). Conditioned heroin-seeking, but not sucrose-seeking behaviors increased early gene expression of ania-3, MKP-1, c-fos and Nr4a3 in the medial frontal cortex, and of ania-3 in the orbitofrontal cortex and NAC (609). Destruction of dopamine neurons in the rostral linear nucleus and periaqueductal gray blocked the rewarding and sensitizing properties of heroin (351).

Activation of group II metabotropic glutamate receptors in the NAC shell attenuates context-induced relapse to heroin seeking (117). Adenosine A2a blockade prevented the synergy between MOR and cannabinoid CB1 receptors and eliminated heroin-seeking behavior in addicted rats (1269). CRF antagonism, but not adrenalectomy blocked acute food-deprivation-induced reinstatement of heroin seeking in rats (1012). Nondependent monkeys increased choice of heroin over food as the heroin dose increased with chronic naloxone and buprenorphine, but not methadone blocking this response. Heroin-dependent monkeys in withdrawal also showed increased heroin choice over food with methadone more effective than buprenorphine in blocking this latter effect (832). Whereas heroin and DAMGO reduced VTA GABA firing rats and internal capsule stimulation post-spike discharges, acquisition of heroin self-administration behavior increased the firing rate of VTA GABA neurons (1068). Prenatal heroin exposure disrupted the development of cholinergic-induced translocation and activation of PKC isoforms (493). Expression of NR1/NR2B N-methyl-D-aspartate receptors enhances heroin toxicity in HEK293 cells (292). 18-methoxycoronaridine acted in the medial habenula and interpedunuclar nucleus to decrease morphine self-administration in rats (403). Cardiovascular responses failed to change after naloxone administration in propofol-sedated piglets during opioid overdose (125).

Although a CS previously paired with naltrexone-precipitated withdrawal suppressed heroin seeking in extinction, it elevated responding if rats had prior experience of heroin taking in the presence of the withdrawal CS (464).

8b-ii. Human studies 

The prescribing of naloxone to actively-injecting heroin users in Chicago has been associated with a reversal in heroin-induced deaths between 2000 and 2003 (749). Diamorphine (heroin) prescriptions were a long-term commitment in the United Kingdom and were not associated with serious drug, health or social problems (773). The use of “take home naltrexone” for homeless drug users was evaluated for awareness and risk perception (1232). An intranasal diamorphine spray was effective as an alternative to injectable diamorphine for maintenance treatment (784). The incidence of heroin use in Zurich, Switzerland rose between 1975 and 1990, and then declined between 1990 and 2002 (850). There was an association between the availability of heroin and methadone and fatal poisoning in England and Wales between 1993 and 2004 (797).

The greatest increases and then declines in methadone deaths happened on Saturdays (798). Heroin-using drivers displayed impairments that correlated with morphine and M6G blood levels (48). Massive decreases in Canadian heroin supply coincided with an Australian heroin shortage (1227). Malay subjects displayed a higher prevalence of current injection drug use, needle sharing and HIV infection than Chinese subjects (184). Methadone-induced overdose deaths were associated with testing positive for a tricyclic anti-depressant or benzodiazepine (179). Methadone maintenance patients are cross-tolerant to morphine analgesia at very high plasma morphine concentrations (44). Methadone-stabilized pregnant patients were safely transferred to buprenorphine using an immediate release morphine transition (532). S(+)-ketamine attenuated the increase in EEG activity and amplitude height of sensory-evoked potentials during rapid opioid detoxification (362). Nonketotic hyperglycemia coma occurred in toddlers after unintentional methadone ingestion (1122). Effective and ineffective methadone dosages overlap considerably for individual opioid-dependent patients (1130). Central morphine concentrations are on average twice as high as peripheral morphine concentrations in heroin overdose victims (244). Although most overdoses occur within 5 min after buying a drug and administering it alone, knowledge of dangers of mixing benzodiazepines with heroin actually increases the likelihood of such behaviors (284).

Opioid-dependent individuals who had used heroin, had injected drugs, had witnessed a drug overdose or who had a history of one or more accidental overdoses were significantly more willing to treat with naloxone a companion who had overdosed (624). Opiate overdoses in Rhode Island were more likely in males under 54 in a private residence on weekends around 9 PM (770). Hydrocodone-related fatalities in Ontario typically occurred in the presence of another drug (1179). A strict correlation was not observed for the relationship between plasma cortisol levels, withdrawal symptoms and craving in abstinent and treated heroin addicts (825). Salvia divinorum, which contains the psychotropic, diterpene and the KOR agonist, salvinorin A is mostly smoked to achieve a short-lived but intense psychedelic experience in recreational users (408).

Post-mortum analyses of heroin abusers indicated that subjects with MOR A118G polymorphisms displayed exaggerated down-regulation of prepro-Enk, but also prepro-DYN especially in the NAC shell (300). The ability of levomethadone to reduce pupil size and miosis was attenuated in healthy human volunteers possessing the 118A>G single-nucleotide polymorphism of the MOR gene (697). However, a meta-analysis failed to find a relationship for a risk of substance dependence with the Asn40Asp (A118G) single nucleotide polymorphism of the human MOR (36). Sex, body weight, benzodiazepine use and creatinine clearance are not factors in prescription of heroin for heroin dependency (969). Subjective heroin effects were rated more positively in heroin inhaling than in injecting patients (970); mast cell mediator tryptase concentrations occurred in the latter, but not former treatment (971).

Heroin addicts display increased polysialic neural cell adhesion molecule expression in the hippocampus (1203). 6-acetylmorphine assays are more accurate than morphine assays in detecting heroin use in urine samples (71). Analyses for papaverine metabolites were more sensitive for detecting illicit heroin use than analysis of 6-monacetylmorphine (884). DRI and CEDIA, but not REMEDi systems are best for preliminary tests of opiates in human urine (487). Pericardial fluid can be used in forensic assays of heroin and cocaine use (237). Straws used to inhaled heroin vapors contained heroin, its main metabolite, 6-acetylmorphine, caffeine and morphine (595); this was confirmed in an in vitro simulation of ‘chasing the dragon’ (596).

The greater incidence of anesthesiologists to fentanyl abuse might be explained by second-hand exposure of fentanyl in air around the mouth of patients during surgery (406). Detoxified opioid-dependent individuals showed greater improvement on all assessment measures following implantation of naltrexone pellets relative to levomethadone treatment (420). Implants of 1.8–3.6 g of naltrexone in opioid dependent individuals maintained plasma levels of naltrexone above 1 nl/ml for 4 to 6 months respectively (1172). Sustained-release naltrexone treatment dose-dependently increased retention in treatment and increased the percentage of negative urine samples in opioid dependent individuals (231). Depot naltrexone was used in lieu of incarceration for coerced treatment for addicted offenders (730), and antagonized the reinforcing, subjective and physiological effects of heroin (1084). Judicially mandated naltrexone by opioid-dependent criminal offenders was analyzed (112, 855), and provision of naloxone to injection drug users as an overdose prevention strategy in New York City proved efficacious (376, 1231). Intranasal naloxone should be used earlier in a tiered-response emergency medical service unit (76). A minority of patients showed recurrent opioid toxicity within 12 h after pre-hospital care of presumed heroin overdose (124) with survival success greater when cardiac arrest is witnessed by emergency workers (123). Slow tapering from methadone maintenance in a program encouraging indefinite maintenance was generally ineffective in opioid-dependent individuals (160). Buprenorphine maintenance was more effective than tramadol detoxification in increasing outpatient treatment participation in opioid-dependent subjects (158). Naltrexone and buprenorphine was more effective than naltrexone alone in opioid dependent individuals in terms of retention (947) as well as lower positive urine samples for morphine and cocaine metabolites (397).

Buprenorphine/naloxone doses greater than 8/2 mg provide minimal incremental value in terms of withdrawal suppression (238). Integrated buprenorphine and naloxone were effective for HIV clinical care in opioid-dependent individuals (1082), and its efficacy was maintained as a function of counseling sessions (347). Introduction of buprenorphine for treatment of substance abuse was robustly predicted by a facility’s prior use of naltrexone and medically-offered withdrawal programs (603). Earlier adoption of buprenorphine was found in private rather than public substance abuse treatment centers (597). Buprenorphine-based outpatient treatment was far more cost-effective than conventional outpatient and inpatient as well as anesthesia-induced detoxification procedures in Australian opiate-dependent individuals (1014), but adverse effects were observed in children after unintentional buprenorphine exposure (391). Slow-release buprenorphine or methadone was better than slow-release morphine in treating heroin addicts in terms of quality of life, symptoms and additional consumption (400). Interactions between buprenorphine and nonnucleoside reverse-transcriptase inhibitor and protease inhibitor antiretrovirals were evaluated in opioid-dependent, buprenorphine/naloxone maintained HIV-negative volunteers (752, 753). Oral slow release morphine over 6 months was safe and efficacious for maintenance treatment in heroin addicts (1148). Buprenorphine tablets were more effective than solutions in producing drug steady state in opiate-dependent individuals (232).

Analgesia and sedation in the presence of a naltrexone implant presented a novel pharmacological challenge for emergency care (854). Naloxone and naltrexone, but not tramadol increased measures of opioid withdrawal in opioid-dependent patients (170). Although most patients in a Veterans Administration study successfully detoxified from opioid use following naltrexone treatment, a one-year follow-up revealed that very few remained engaged and stabilized in abstinence-oriented outpatient treatment (266). Predictors of retention in naltrexone maintenance for opioid dependence included methadone use and higher average bags per day of heroin (1083). Low-dose treatment with naltrexone may help reducing the manifestation of opioid withdrawal in dependent populations (719). Further, although behavioral naltrexone therapy improved retention in treatment for heroin-dependent patients, it generally achieved abstinence from opioids after three months, but only in 22% after six months (853). Naltrexone with fluoxetine prevented relapse to heroin addiction in Russian women (615). Anesthetic opioid detoxification did not differ from clonidine detoxification treatment when combined with psychosocial support in opioid dependent patients (336). Opioid demand in heroin-dependent individuals not in treatment is a function of drug supply, unit price and cocaine use (417).

In two psychonautic studies, concomitant administration of naltrexone with codeine obviated opioid withdrawal syndrome in a single subject (870). Generic controlled release oxycodone approval was not followed by an increase in misuse or abuse (54). A patient genotyped as CYP2D6 PM could not metabolize oxycodone, but had better success with hydrocodone (1089). An automated DOI kit for urinary oxycodone was more reliable that a point-of-care assay (436).

8c. Opiates and ethanol

8c-i. Animal behavioral models 

Enhanced morphine-induced ethanol drinking occurs in alcohol-preferring alko rats sensitized to morphine (859). Spontaneous opioid-mediated and withdrawal-associated CRF1-mediated ethanol drinking was observed in Sardinian alcohol-preferring rats (978). The Naples low-excitability rat line displayed greater alcohol intake and preferences as well as greater sensitivity to quinine and naltrexone than Naples high-excitability rats (892). The greater preferences for ethanol noted in C57BL/6J and hybrid C57BL/6J x CD-1 mice relative to CD-1 mice could not be linked to mesolimbic DA transporter neurochemistry and/or Enk levels, but rather with D2 receptor expression (1033). Whereas mu, but not delta opioid antagonism blocked ethanol-induced sensitization, this effect was not attributable to putative mu1 or mu3 opioid receptors (881). Both MOR and KOR antagonism blocked the reinforcing effects of ethanol on an artificial nipple administered to neonatal rats (846). GABAB receptor stimulation attenuates the locomotor effects of morphine in mice bred for extreme sensitivity to the stimulant effects of ethanol (476). KOR antagonism blocks Nac Da concentrations during operant ethanol, but not sucrose self-administration (298). Low and high doses of buprenorphine respectively increased and decreased ethanol consumption with the latter effect more effectively eliminated by the ORL-1 antagonist, UFP-101 relative to naltrexone (223). OFQ/N agonists reduce ethanol intake while increasing food intake in alcohol-preferring rats in a naloxone-sensitive manner (307). OFQ/N blocked ethanol-induced increases of GABA release, and decreased presynaptic GABA transmission in the central amygdala (962).

Although Roman high avoidance strains consume more alcohol than Roman low avoidance strains, the latter show enhanced responsiveness on the hole board test following systemic ethanol. Whereas Roman high avoidance rats have higher Pro-DYN in the NAC shell and Pro-Enk in the cingulate, Roman low avoidance rats have higher Enk gene transcription in the dorsal striatum (425). Female, but not male mice lacking prepro-DYN show lower preferences and consumption of alcohol, but not changes in ethanol-induced loss of righting reflexes, ethanol withdrawal, ethanol-induced CPP or conditioned taste aversions to ethanol (106). Estradiol valerate increased alcohol intake, decreased body weight and food intake, and reduced the number, but not content of arcuate BEND neurons (537). Losers in a social defeat test consumed more ethanol than winners, and thereby displayed lower anxiety levels. U50488H stimulated ethanol intake more in losers than winners, but only increased approach behaviors in losing mice not consuming ethanol (619). Antagonism of endogenous KOR systems potentiated the increased responsiveness of mesoaccumbal Da neurons to ethanol (1290). Naltrexone pretreatment blocked reinstatement behaviors on an operant ethanol lever induced by ethanol-associated context cues (149). Naltrexone reduced ethanol intake, but not craving behaviors in heavy- and light-drinking mice (329). Naltrexone was more effective in blocking the acquisition of alcohol intake in periadolescent relative to adult alcohol-preferring rats whereas greater tolerance to repeated naltrexone dosing was noted in adult relative to periadolescent rats (979). Both naltrexone and acamprosate reduced alcohol and water intake in a schedule-induced polydipsia task, and reduced head entries for food following the highest antagonist doses (324). Naloxone blocked analgesia induced by alcohol and nicotine alone as well as its synergistic combination (162).

8c-ii. Ethanol-induced changes in opioid systems 

Whereas acute ethanol respectively decreased and increased Pro-Enk mRNA expression in the VTA and prefrontal cortex, chronic ethanol increased Pro-Enk mRNA in the core and shell of the NAC (765). Chronic ethanol decreased brain interstitial levels of Menk using in vivo microdialysis (1141). Intubated ethanol perinatally from Days 1–22 increased hypothalamic Menk levels in males and females, NAC Menk levels in females, and lowered Menk levels in the central nucleus of the amygdala in male and female animals (704). Alcohol administration increased extracellular levels of DYN A(1–8) in the rat NAC (727). Alcohol, nicotine and cocaine evoked release of endogenous morphine from the M.edulis pedal ganglia (1326). Chronic naltrexone prevented acute ethanol-induced increases in BEND plasma levels in Warsaw low-ethanol preferring, but not high-ethanol preferring rat strains (1289).

Ethanol induces apoptotic death of developing fetal rat BEND neurons through suppression of cAMP production and activation of transforming growth factor-beta 1-linked apoptotic signaling (191). Ethanol suppression of hypothalamic POMC levels and splenic natural killer cell cytolytic activity is associated with a reduction in the expression of proinflammatory cytokines, but not anti-inflammatory cytokines in neuroendocrine and immune cells (189). Prenatal ethanol exposure altered the expression of period genes governing the circadian function of BEND neurons in the hypothalamic suprachiasmatic nucleus (190). Chronic ethanol inhibits natural killer cell function by suppressing the influences of BEND, CRF and the autonomic nervous system signals to the spleen (122). Rats exposed to ethanol through breast-feeding displayed decreased BEND in thymic cells whereas pre-natally ethanol-exposed rats displayed decreased 5HT content in immune cells (250).

8c-iii. Human studies 

There are associations between the promoter and 3′ end of Pro-DYN as well as intron 2 of the human KOR with alcohol dependence in human individuals (1251). Acamprosate and naltrexone prevented decreases in ACTH and cortisol during alcohol abstinence, and this was associated with a reduced risk of relapse (575). Acamprosate increased BEND plasma concentrations in patients with high, but not low alcohol preference (576). Naltrexone was found to be superior to acamprosate in relapse prevention of alcoholism among individuals with low levels of clinical depression and alcohol dependence (803). Alcohol-dependent and PTSD subjects displayed significantly higher stress ratings and decreased ACTH responses on the cold pressor task response (129). Targeted naltrexone was better than daily naltrexone in affecting average daily drinking (466). Naltrexone’s reduction in blood alcohol level following acute ethanol intoxication accounted for the changes in subjective and behavioral responses to alcohol but not the reduction in HR (902). Naltrexone appeared effective in the treatment of alcohol dependence particularly with respect to heavy drinking (907). Cognitive behavioral therapy when combined with both naltrexone and acamprosate produced the greatest improvement in outcome measures in alcohol dependent individuals relative to each treatment alone (35, 339; but see (289)). Naltrexone produced strong positive associations between the number of positive social celebratory events and drinking although such events occurred on a minority of days (37). Although naltrexone failed to alter the Positive and Negative Syndrome Scale in alcoholic patients, it further augmented the dose-dependent effects of low, but not high doses of ketamine (616). G-hydroxybutyrate was more effective than naltrexone or disulfiram in reducing craving and altered biological markers of alcohol abuse (826). Although there was no clear advantage of naltrexone, disulfiram or their combination in the treatment of alcohol abuse in patients with psychotic spectrum disorders (903), they were effective in patients suffering from PTSD (904). Asp40 carriers acted as genetic moderators of naltrexone’s effects on alcohol cue reactivity (757). Both the A118G polymorphism in the MOR gene and a polymorphism, G1510A, in the acetaldehyde dehydrogenase 2 gene were associated with alcohol dependence in a Japanese population (845). Intervention with morphine at the level of the neuroendocrine-immune axis decreased post-operative pneumonia rate in long-term alcoholics (1060). Nonalcoholic subjects with a family history of alcoholism exhibit lower plasma ACTH and BEND levels following stress and CRF stimulation conditions (1026).

8d. Opiates and THC

CB1 knockout mice display significant increases in striatal preproEnk and preproDYN as well as D4 DA receptor gene expression (396). Both morphine and the CB1 agonist HU210 inhibited Ach and GABA release in the NAC core that was blocked respectively by naloxone and the CB1 antagonist, SR141716 A. Co treatment of the two antagonists reversed their respective blockade of the agonist-induced inhibitory effects (999). MOR and CB1 receptors display reciprocal inhibition of receptor signaling and neuritogenesis (955), thereby failing to form hetero-oliogomers (215). Correspondingly, the acute and chronic suppression by the CB1 cannabinoid receptor inverse agonist AM251 on food intake and body weight was observed in both wild type and MOR KO mice (195). The CB-1 antagonist, SR 141716A blocked the ability of morphine to significantly reduce ventral pallidal GABA efflux, and blocked heroin, but not cocaine self-administration following administration into the NAC, but not the pallidum (157). Morphine in the PVN inhibited in a naloxone-sensitive manner the penile erections induced by PVN administration of the CB1 antagonist SR 141716A as well as the concomitant increases in PVN glutamic acid and NO (1076). Tolerance occurred to the reinforcing effects of morphine in delta9-tetrahydrocannibinol-treated mice (515). Cannabinoid type 2 agonists induce transcription of the MOR gene in Jurkat T cells (113). The acquisition of self-administration of the CB1 agonist WIN 55,212–2 was enhanced by both NBNI administered prior to the session in wild-type mice and in pro-DYN KO mice (766). Prenatal cannabis exposure was associated with increased MOR in the amygdala, reduced KOR mRNA in the mediodorsal thalamus and reduced Pro-Enk expression in the C/P. Prenatal alcohol exposure decreased KOR mRNA in the amygdala, claustrum, putamen and insula cortex (1191).

Cannabidiol, a major constituent of cannabis, is an allosteric modulator at mu- and delta-opioid receptors (556). Whereas systemic naloxone and naltrexone prevented THC-induced release of Ach from prefrontal cortex and hippocampus, VTA administration of naloxonazine and the D1 DA receptor antagonist, SCH39,166 produced the same pattern of effects (912).

8e. Opiates and stimulants

8e-i. Animal behavioral studies 

An AS to the MOR coding sequence 16–32 attenuated cocaine-induced behavioral sensitization and reward (495). Both contingent and non-contingent administration of cocaine increased Pro-DYN, but not Pro-Enk mRNA levels in the C/P, but not in the NAC or central nucleus of the amygdala (1331). Acute and chronic administration of the selective delta opioid agonist SNC80 decreased both cocaine- and food-maintained responding in rhesus monkeys (288). Context-dependent extinction occurs for both cocaine- and morphine-induced floor preferences (878). Like repeated forced swim stress, prior activation of kappa opioid receptors with U50488H potentiates cocaine place preference conditioning (758). A combination of cocaine and heroin self-administered at a maximum of three infusions per hour was preferred to either drug alone, and continued during all hours of the light:dark schedule; increased fixed ratio schedules led to extinction (733). Discrete-trials heroin self-administration produced sensitization to the reinforcing effects of cocaine in rats (1199). Administration of estradiol valerate to 8-week old female rats destroyed arcuate BEND neurons and attenuated the acquisition of cocaine self-administration both with and without estrogen replacement therapy (975).

The NK-1 receptor antagonist, GR82334 failed to affect cocaine self-administration or hyper locomotion, but attenuated morphine-induced locomotor activity while increasing heroin self-administration (914). Chronic buprenorphine enhanced acute cocaine-, but not acute heroin-induced locomotor increases early in buprenorphine treatment (1056). A combination of morphine and buprenorphine was more effective than U50488H in reducing self-injurious behavior induced by methamphetamine in mice (799). Sub-additive withdrawal occurs from cocaine and kappa opioid agonist combinations in Planaria (935).

8e-ii. Anatomical, molecular and neurochemical studies 

MOR KO mice display reduced psycho stimulant effects on Da dynamics in the NAC (745). Cocaine-induced locomotor activity and Fos expression in enkephalinergic D2-type neurons in the NAC are sensitized for six months after repeated cocaine administration outside the home cage (478). Prenatal administration of morphine alone or in combination with cocaine significantly increased mu opioid receptor binding and MOR mRNA on post-natal days 1 and 7, but not 14 in the offspring, effects blocked by the D2 receptor antagonist, sulpiride (88). Prenatal cocaine and morphine respectively increase and decrease brain cyclin-dependent kinase (Cdk5) in rat pups (89). The mu antagonist, clocinnamox blocked cocaine- and methamphetamine-induced preDYN mRNA expression only in the rostral patch compartment of the dorsal striatum, attenuated zif/268 mRNA throughout the dorsal striatum, but did not alter regional psycho stimulant-induced Fos expression (479). DA D2 receptor antagonism in the arcuate nucleus attenuated cocaine-induced NAC BEND increases, and produced extinction behavior in cocaine self-administering rats (295).

DYN KO mice showed decreased NAC DA levels, but no changes in DA uptake under baseline and cocaine conditions, an effect enhanced by kappa agonists and blocked by antagonists. DYN KO mice showed less cocaine-induced locomotor activity, and NBNI failed to alter this response (185). In repeated cocaine-treated animals, CREB responses to DYN are mediated in opposite directions by drugs affecting the D1 and D3 DA receptors (1308). Mice lacking glucocorticoid receptors display decreased DYN and SP as well as D1 and D2 receptors in the dorsal striatum, but not NAC under baseline, but not following cocaine behavioral sensitization. In contrast, Enk mRNA levels are altered during cocaine use in glucocorticoid receptor KO mice (507).

Increased striatal c-fos and SP, but not Enk gene expressions were noted in animals given access to running wheels and cocaine than either manipulation alone, an effect that peaked after two exposures (1220).

Methamphetamine-induced increases in striatal dopamine transmission were potentiated by morphine treatment (897). Amphetamine-induced increases in activity as well as striatal Pre-Enk and Pre-DYN mRNA and gene expression were blocked by SL327 and U0126, inhibitors of MAPK and ERK (1027). Chronic methamphetamine down-regulated MOR after 8 days, and after drug cessation, MOR levels returned to normal on Day 11 and then were up-regulated on Day 21 (210). L-methamphetamine and selective MAO inhibitors decrease morphine-reinforced and non-reinforced behavior in rats (459). Acute and chronic amphetamine treatment decreased concanavalin A-lymphocyte proliferation together with increased Menk in the NAC, prefrontal cortex, spleen, thymus and splenic macrophages (43). The rewarding effects of morphine and methamphetamine were respectively aggravated and enhanced by administration of an astrocyte-conditioned medium into the NAC and cingulated cortex respectively. The glial modulator propentofylline suppressed both morphine’s and methamphetamine’s rewarding effects (818).

Ecstasy, 3,4-methylenedioxy-N-methylamphetamine, increased pro-DYN mRNA in the C/P following acute and chronic treatment, the prefrontal cortex following acute treatment, the NAC and hypothalamus following chronic treatment, but decreased gene expression in the VTA following acute and chronic treatment. Accordingly, DYN A levels were increased in the VTA following chronic Ecstasy treatment, and decreased in the NAC, prefrontal cortex and hypothalamus after acute Ecstasy treatment (282).

8e-iii. Human studies 

Comparisons between heroin and heroin-cocaine polyabusers were made in terms of incidence of psychopathology (61). There was an association of DOR gene polymorphisms in methamphetamine dependence and psychosis (602). A 3- or 4-repeat allele of a 68-bp element in the promoter region of the Pro-DYN gene was found significantly more frequently in patients with methamphetamine dependence than in controls (849). Linkages between an IVS2+G691C single-nucleotide polymorphism and methamphetamine patients and transient psychosis were found (498). The Cocaine- and Heroin-Craving Questionnaires appear to predict dropout rates and in-treatment drug use by clients relative to visual analog scores (462).

8f. Opiates and other drug abuse classes

Three-marker haplotypes of the MOR gene were significant for smoking initiation, but were marginal for nicotine dependence (1309). Morphine increased locomotor activity to a greater degree in nicotine-withdrawn as compared to control mice, and effect accompanied by increased Da and 5HT in the striatum (1162). Beta2 nicotinic Ach receptor KO mice fail to display intra-VTA nicotine self-administration, but have normal morphine self-administration and normal nicotine withdrawal behaviors (86). Reinstatement of nicotine-induced place preferences was produced by nicotine or morphine, effects blocked by the calcium channel antagonists, nimodipine and flunarizine (91). Nicotine induced c-fos within CRF and Enk, but not DYN cells within the PVN (698).

Norlaudanosoline and nicotine increase endogenous invertebrate ganglionic morphine (1328). Chronic swim stress increased nicotine’s ability to increase plasma ACTH and BEND, but not corticosterone, whereas stress itself increased plasma corticosterone, but not ACTH or BEND (703). There was an association of the MOR A118G variant with reductions in the relative reinforcing value of nicotine in females, but not males (944). Naltrexone was efficacious in smoking cessation especially in females (587). A high dose of oral naltrexone appeared to augment the efficacy of the nicotine patch in producing smoking cessation after 4–6 weeks (866). Naltrexone and cognitive behavioral treatment based on the community reinforcement approach decreased craving in abstinent smokers (972). However, nicotine self-administration in rats was blocked by mecamylamine and hexamethonium, but not by naloxone (273).

9. Sexual Activity and Hormones, Pregnancy, Development and Endocrinology

This section will examine developments in the last year relating the endogenous opioid system to sexual activity (Section 9a), pregnancy (Section 9b), development (Section 9c), and general endocrinology (Section 9d). 

9a. Sexual activity and hormones

Fentanyl and naloxone respectively decreased and increased sexually motivated song in male European starlings, a pattern opposite to Da (1001). MOR, DOR and KOR were expressed in human spermatozoa with MOR agonists and delta antagonists reducing motility (9). MOR is located on the acrosomal region and the neck region of human sperm (19). Plasma testosterone was respectively decreased and increased 4 and 24 h after morphine, fentanyl and buprenorphine, whereas plasma estradiol was decreased both 4 and 24 h after morphine, tramadol and buprenorphine in male rats (176). Insulin-like growth factor and growth factor-associated signal transduction pathways inhibited estradiol and progesterone facilitation of female reproductive behaviors as well as the behavioral effects of a DOR agonist (327). DYN A and B as well as Lenk-Arg(6) were more predominant in NAC, C/P and SN during estrus than during diestrus or proestrus, effects associated with cyclic fluctuations in the enzymatic cleavage of DYN (966). Estrogen in female rats suppressed 3H-DAMGO binding in rat cortical membranes in a manner similar to that of CCK-8 and a CCK-8 agonist, JMV-180 with the latter effects blocked by the CCK1 receptor antagonist, JMV-179 (931).

Estrogen without progesterone increased ORL-1 mRNA expression in the anteroventral periventricular hypothalamus, MPOA and VMH and increased OFQ/N mRNA levels in the caudal part of the posterodorsal medial amygdala; estrogen and progesterone increased MPOA OFQ/N mRNA in the female rat (1040). LSN2120310, a selective estrogen receptor modulator for both ERalpha and ERbeta treats hot flushes in a morphine-dependent rat model (1178). Different progestins differentially modulate estradiol-induced BEND synthesis and release (915). The reproduction phase-related expression of BEND-like immunoreactivity in the nucleus lateralis tuberis of the female Indian major carp correlates with the number of leutinizing hormone cells and ovary during spawning (985). OFQ/N reduced tuberoinfindibular DA neurons just preceding its stimulation of the prolactin secretory response, effects blocked by ORL-1 antagonism that in turn inhibited the suckling-induced prolactin response (207). Naloxone acts as an antagonist of estrogen receptor activity in MCF-7 cells (334). DORs are found in a subpopulation of GnRH nerve terminals, primarily in the external layer of the ME. DPDPE-induced inhibition of cAMP accumulation and Go down regulation in a GnRH-secreting cell line is blocked by pertussis toxin (910). Estradiol valerate blocked ovariectomy-induced reductions in BEND and allopregnanolone, whereas progesterone increased plasma, hypothalamic and intermediate pituitary BEND. In contrast, medroxyprogesterone increased BEND only in hippocampus and intermediate pituitary (81). Oral tibolone maximally increased BEND in frontal lobe, hypothalamus and neurointermediate pituitary, also increased BEND in parietal lobe, anterior pituitary and plasma, but failed to alter hippocampal BEND in ovariectomized rats in parallel to allopregnanolone increases (392). Sexual dimorphism was observed in the organization of glutaminergic cells in the rat hypothalamic infindibular area for neurokinin B, but not for DYN (225). There was no role for endogenous opioid peptides in the reproductive suppression in subordinate female highveld mole-rats (1124).

Decreased BEND is found in serum and seminal plasma in infertile men (313). Exercise failed to elevate Pro-Enk peptide F during either follicular or luteal phases of the menstrual cycle in women (611). Both fluoxetine and clomipramine prevent premature ejaculation induced by naltrexone in opioid detoxification treatment (3). Normalization of hyperinsulinemia by chronic opioid receptor blockade occurred in hyperandrogenemic women (434).

9b. Pregnancy

Fetal morphine metabolism and clearance are constant during late gestation (387). Naloxone increased serum prolactin, but not the DOPAC/DA ratio in anti-progesterone-treated, but not control pregnant rats, and failed to affect mifepristone-induced decreases in TH immunoreactivity in the arcuate and periventricular hypothalamic nuclei (1052). Pro-Enk A mRNA in the luminal and glandular epithelium fell and GABA A receptor subunit in the luminal epithelium and stromal cells fell during endometrial receptivity (932). Interleukin-1beta increased SON Oxy secretion and electrical activity in virgin, but not pregnant rats, whereas naloxone enhanced interleukin-1beta-induced SON Oxy secretion and electrical activity in pregnant, but not virgin rats (146). Caesarean section is associated with lower maternal concentrations of epinephrine, NE, ACTH, cortisol, prolactin and BEND compared with other modes of delivery (1171). Elevations in heat pain, but not cold pressor pain were not before and after parturition in pregnant women relative to non-pregnant women (172).

9c. Development

DAMGO and U69593, mu and kappa opioids, induced the differentiation of embryonic stem cells to neural progenitors (578). NMDA lesions placed in the rostral lateral PAG prevented morphine-induced inhibition of maternal behavior. Morphine-sensitized dams would choose to hunt insects rather than engage in maternal behavior, and lesions of the rostral lateral PAG restored the maternal response (1081). DYN induced a kappa receptor-sensitive immediate unconditioned increase in neonatal responsiveness to a surrogate nipple in newborn rats, but did not alter water intake through an intra-oral cannula. Pairing Dyn and suckling on a nipple increased responsiveness 1 day later, indicating conditioning (905). MOR and ORL-1 receptor binding are significantly and respectively increased in the NAC and VTA of two-day old rat pups relative to their dams, effects attributable to increased G-protein coupling which in turn showed greater effects in female relative to male pups in the NAC (483). Neonatally-handled male mice display greater body weights, increased plasma ACTH and corticosterone and decreased hypothalamic ACTH and CRF as adults, effects blocked by neonatal administration of AS directed against POMC (379). Neonatal treatment with naloxone increased the population of Sertoli cells and sperm production in adult rats (262). Facial expressions of pain, high activity levels, poor response to routine care and poor ventilator synchrony were associated with placebo versus morphine therapy, and could be used as markers for persistent pain in preterm infants (126).

9d. Endocrinology

Oxytocin increased ACTH, BEND, LH and prolactin secretion of cycling porcine pituitary cells, with oxytocin and CRF producing additive increases in pituitary BEND (608). The endocrine disruptor, bisphenol-A administered prenatally and postnatally led to an enhancement of the DA-dependent rewarding effect induced by morphine (785). DAMGO reversed the inhibitory effects of PGE1 upon Na,K-ATPase activity in SH SY5Y neuroblastoma cells; DAMGO increased this activity itself through a cyclosporine- and nifedipine-dependent Ca2+ channel system (1229).

10. Mental Illness and Mood

This section summarizes the few studies examining opioid involvement in mental illness (Section 10a) and mood (Section 10b). 

10a. Mental Illness

Manipulations of Dyn, CREB, BDNF, MCH or Clock proteins in the NAC and VTA in rodents produce unique behavioral phenotypes directly relevant to depression (834). The kappa agonist, salvinorin A, induced two animal models of depression, namely increased immobility on the forced swim test and increased thresholds for ICSS without altering locomotor activity in an open field. These effects were accompanied by decreased DA, but not 5HT concentrations in the NAC (169). Delta agonists, including DPDPE and Delt II, decreased immobility in the forced swim test and increased frontal cortical BDNF mRNA expression in a NTI-sensitive manner, suggestive of anti-depressant actions (1127). The DOR agonist, (+)BW373U86 decreased immobility and increased BDNF mRNA in the frontal cortex, effects blocked by NTI, but not naltrexone or NBNI. Lenk and Menk produced similar patterns of effects, but up regulated BDNF mRNA in the hippocampus through DOR- and MOR-sensitive mechanisms. In contrast, BEND, endomorphin-1 and endomorphin-2 increased BDNF mRNA expression in frontal cortex, hippocampus and amygdala in a naltrexone-sensitive manner without affecting immobility (1304). The enkephalinase inhibitor, RB101 produced antidepressant and increased locomotor effects without inducing seizures, convulsions of alterations in BDNF mRNA expression (543). Whereas ORL-1 agonists produce anxiolytic-type effects on the elevated plus maze, light:dark aversion, operant conflict, startle, ultrasonic vocalizations and hole board tests, either ORL-1 antagonists and OFQ/N KO animals display anti-depressant effects on the forced swimming and tail suspension tests (388). The ORL-1 agonist, Ro64-6198, produced selective inhibition of marble burying in mice without affecting locomotor activity, thereby exhibiting anxiolytic and antidepressant actions (841).

Sustained sadness in depressed women was associated with decreased MOR binding potential in the left inferior temporal cortex, and in the anterior cingulate in those who did not respond to anti-depressants (567). Adherence to naltrexone in opioid-dependent patients produced less depression symptoms (267). Naltrexone reduced alcohol use and craving as well as symptoms measured by the Hamilton Rating Scale for Depression and Young Mania Rating Scale in patients with bipolar disorder and alcohol dependence (137). Nalmefene significantly improved scores on the Yale-Brown Obsessive-Compulsive Scale modified for Pathological Gambling markedly improving 59% of the subjects (413). Naltrexone rendered one-session exposure less effective for phobia treatment (610). Long-term use of high-dose oral naltrexone is safe using hepatic transaminase profiles in otherwise healthy patients with impulse-control disorders who restrict their intake of over-the-counter analgesics (584).

10b. Mood

People with high neuroticism displayed a graded cortisol, but not ACTH response to naloxone, whereas those with low neuroticism displayed a cortisol response that plateau (718). Reductions by naltrexone were observed on repressive coping and disclosure of emotional material (1278). Volunteers displayed less alertness, increased sedation and increased effort to perform a driving test following oxycodone/paracetamol relative to the NSAID, bromfenac (1157). Midgestational women who were battered showed higher levels of anxiety and depression than non-battered pregnant women, and plasma ACTH and BEND levels showed a significant linear relationship in battered, but not non-battered women (1105).

11. Seizures and Neurological Disorders

11a. Seizures

Alfentanil enhanced the amplitude and number of hippocampal CA1 population spikes in control animals, but reduced in such spikes in pilocarpine-treated epileptic rats because of an increase in eliptiform population spikes; these effects were naloxone-reversible (986). Morphine enhanced pilocarpine-induced seizures and status epilepticus (360). The non-peptide delta agonists, SNC80 and (+)BW373U86 produced bilateral ictal and paroxysmal spike and discharges and brief changes in EEG recordings that were subject to quick tolerance and sensitivity to compounds used to treat absence seizures (542). A ketogenic diet decreased Pro-Enk gene expression induced by KA of the granular cells of the hippocampus as well as diminishing KA-induced AP-1 DNA-binding activity, Fos and Jun expression, and the phophorylated form of the three types of JNKs (848). EEG and convulsant effects were noted following the delta opioid agonist SNC80 in one of four rhesus monkeys (260). Encephalitis induced by neurotrophic Borna disease virus induced seizures and DYN loss through hippocampal dentate cell loss. Kappa agonists prevented the seizure activity that was associated with an absence of DYN in dentate gyrus granule cells and an up regulation of Enk in CA1 interneurons (1054). Kappa agonists and antagonists respectively decreased and increased handling-induced convulsions in ethanol withdrawal-seizure prone, but not ethanol withdrawal-seizure resistant mice (69). HSV-1 infection caused loss of hippocampal DYN A reactivity, and the kappa agonist, U50488H blocked ictal activity (1055). Single nucleotide polymorphisms in the MOR subunit gene were not associated with idiopathic generalized epilepsy (66).

11b. Neurological disorders

Estradiol benzoate, but not tamoxifen was capable of blocking the methamphetamine-induced increase in striatal PPE mRNA levels as well as prevent loss of striatal dopamine transporter binding (263). In Parkinson Disease-related research, MPTP-treated monkeys displayed striatal DA denervation and dyskinesia accompanied by an increase in Pro-Enk-A mRNA levels in the rostral and lateral putamen and rostral caudate, effects blocked by pretreatment with the NR1A/2B NMDA antagonist, CI-1041, but not levodopa (802). Primates with Parkinsonian features with or without L-DOPA-induced dyskinesia displayed normal phenotypes of striatal medium spiny neurons, but showed axonal collateralization of striatofugal cells containing Dyn that project to the pars interna and externa of the globus pallidus (811). Naltrexone is one of eight nondopaminergic drugs in which the MPTP-lesioned promate correctly predicted phase II efficacy (356). Mice lacking preproOFQ/N gene expression display less loss of TH neurons in the SN and CP of MPTP-treated mice, but failed to show changes in methamphetamine-induced losses of DA in the CP (138). Aphakia mice that possess a naturally-occurring Pitx3 deficiency, display levo-dopa-sensitive locomotor deficits and such neuronal losses as DA transporter binding and DA receptor expression, Enk, DYN and neurotensin in a manner similar to adult animals receiving neurotoxin administration (1146). Hemiparkinsonian Nur77 KO mice showed exacerbated rotational responses to L-DOPA, and failed to show up-regulation of striatal Enk mRNA levels; DYN levels were unchanged (1071). SP induced toxic effects on SN DA neurons in a microglia-dependent manner in neuron-glial cultures, an effect reversed by DYN (107). The G-allele of the A118G single nucleotide coding region polymorphism of the MOR gene as well as a history of never smoking were independently associated with increased risk of earlier onset of levodopa-induced dyskinesia in Parkinson’s disease (1074).

R6/2 transgenic mice that display symptoms of Huntington’s disease display decreases in NR2A NMDA receptors in proportionally more striatal cells that contain Enk (20). Huntingtin inclusions do not down-regulate Pro-Enk genes in the R6/2 Huntington’s disease mouse (980). Endomorphin-2 protects against the beta-amyloid aggregate creator, Abeta1–42 in vitro and in vivo (1095). Perinatal 6-OHDA lesions decreased striatal DA and SP and increased striatal Menk, effects reversed by chronic perinatal administration of D1 (SKF-38393), but not D2 (quinpirole) agonists (1044). An adenosine A2A antagonist, CSC, but not U50488H produced a reversion of decreases in the duration of levo-dopa-induced rotations in 6-OHDA-treated rats that was accompanied by an increase in levo-dopa-induced preproDYN mRNA in the lesioned ventromedial striatum (120). Animals receiving traumatic brain injury displayed poor motor function for up to five days post-trauma when treated with morphine (1066). Oral exposure to manganese increased manganese levels and neuron injury in striatum and GP with apoptotic cells containing NOS, Enk and ChAT; these effects reduced locomotor activity and striatal DA content (690).

Decoy peptides that bind DYN noncovalently prevent NMDA receptor-mediated neurotoxicity in the spinal cord as well as DYN-induced paralysis and allodynia (1228). Mice bearing the G93A SOD1 mutation that leads to clinical symptoms of familial amyotrophic lateral sclerosis display prolonged survival with daily administration of morphine and taineptine (216). Intrathecal mu and delta, but not kappa agonists dose-dependently and opioid receptor-selectively induced spastic paraparesis in animals exposed to spinal ischemia (545). Intrathecal nicorandil and small-dose morphine can induce spastic paraparesis after a noninjurious interval of spinal cord ischemia in the rat (366). There was a transient loss of motor-evoked responses associated with caudal injection of morphine in a patient with spondylolisthesis undergoing spinal fusion (401). Unilateral labyrinthectomy causing vestibular behavioral deficits was accompanied by pre-pro-Enk up regulation and increased Fos activity in the medial vestibular nucleus. Blockade of this up regulation with naloxone or antisense probes increased the behavioral deficits (590). DYN anti-sera applied soon after spinal cord ischemia attenuated NOS up-regulation, and reduced both spinal edema and cell injury (1022). Endomorphin-1 and to a lesser extent endomorphin-2 inhibited Cu2+ and AAPH, a water-soluble initiator in inducing oxidation of low density lipoprotein, showing protection against free radical-induced neurodegenerative disorders (670). Naloxone failed to alter a postoperative coma in a patient with complete basilar syndrome after anterior cervical disectomy (1134). Neuroprotection by endogenous and exogenous PACAP following stroke was mediated in part by Enk (201).

Gastrointestinal, Renal and Hepatic Functions

14a. Gastric Function

Naloxone or vagotomy does not influence centrally octreotide-induced inhibition of gastric acid secretion in rats. (383). Morphine and opium dependence reduced the severity of postoperative adhesions following abdominal gastric surgery (573).

14b. Intestinal Function

Morphine increased GI transit in horses over a 6 h period decreasing fecal weight and moisture as well as defecation frequency (115). The peripherally-acting opioid antagonist, N-methylnaltrexone reversed these effects (116). Morphine tolerance and dependence increased c-Kit expression in duodenum and ascending colon in a CTOP-sensitive manner (60). Morphine and WIN55212-2 each reduced GI transit, but CB1 antagonism or CB1 KO mice failed to affect morphine-induced inhibition. Similarly, neither naloxone nor naltrexone altered CB1 agonist-induced inhibition of GI transit (168). Morphine delivered by pellets produced far greater Salmonella infection in Peyer’s patches, mesenteric lymph nodes and spleen than morphine delivered by minipump, an effect attributable to greater inhibition of GI transit. Whereas BEND relaxed rabbit jejunum contractility, naloxone respectively increased and decreased contractility at low and high concentrations respectively (239). Chemical coding of Type I myenteric neurons with different axonal projection patterns in the porcine ileum revealed 78% immunoreacted to Enk (540). Delt II, but not DPDPE or U50488H administered by minipump induced moderate Salmonella infection in the spleen, but did not affect GI transit (341). Both salvia divinorum and its active ingredient, Salvinorin A inhibited electrically-induced contractions of the guinea pig ileum, effects reduced by naloxone and NBNI, but not CTOP or NTI (167). Kappa and delta, but not mu opioid agonists enhanced the inhibition of plasma extravasation during chronic relative to acute intestinal inflammation; these effects were reversed by receptor-selective antagonists and NOS inhibitors. Chronic inflammation increased KOR and DOR protein levels in the whole jejunum and mucosa (524). Pro-Enk A mRNA expression is widely expressed in human GI tract, esophagus, pancreas and gallbladder (794). An inherent acceleratory effect of insulin on small intestinal transit could be observed in naloxone-induced inhibition of this response (889). Chronic desipramine treatment produced subsentivity to kappa agonist-induced inhibition of the peristaltic reflex, but supersensitivity to mu agonist-induced inhibition of propulsion velocity. Chronic desipramine treatment decreased both MOR and KOR in the myenteric plexus and colon of guinea pigs (164). The selective mu opioid receptor antagonist, alvimopan, improves delayed GI transit of postoperative ileus in rats (371). Long pulses of intestinal electrical stimulation inhibited canine intestinal motility in a naloxone-insensitive and NO-insensitive manner, but that was reversed by Ach and NE antagonists (686). Enteral naloxone was effective in treating opioid-induced constipation in a pediatric intensive care unit (1123). Low-dose naltrexone was effective for the treatment of irritable bowel syndrome (554).

14c. Nausea and Emesis

Ondansetron, but not droperidol was an effective prophylaxis of nausea and vomiting after intrathecal morphine in women undergoing caesarean section (891). Loperamide reduced kinetosis-induced nausea, and reduced the increases in ACTH and antidiuretic hormone induced by kinetosis in human volunteers (872).

14d. Glucose Function

DYN A(1–17) increased glucagon release from pancreatic islets with pancreatic beta cells treated with high glucose eliminating DREAM interactions with the Pro-DYN promoter downstream regulatory element, and thereby increasing Pro-DYN promoter activity. DREAM-KO beta cells show increased Pro-DYN activity as well (509). Metformin decreases plasma glucose in STZ-induced diabetic rats through an increase in adrenal BEND secretion to stimulate mu opioid receptors that leads to an increase in GLUT-4 gene expression and a decrease of hepatic PEPCK gene expression (205). Metaformin also decreased fasting plasma glucose in healthy humans that was accompanied by an increase in plasma BEND and a decrease in serum cholesterol (873). The ginsenoside, Rh2 lowered plasma glucose and increased plasma BEND in a naloxone-sensitive manner in control, but not MOR KO STZ rats (625). Exocrine pancreatic secretion is inhibited by both OFQ/N and DOR agonists (672). The plasma glucose lowering action of Hei-Shug-Pian, the fire-processed product of the root of Aconitum was naloxone-reversible in STZ-induced diabetic rats (675). Myricetin lowered plasma glucose and increased plasma BEND in STZ-induced diabetic mice, effects eliminated by bilateral adrenalectomy, opioid receptor antagonists and by deletion of the MOR gene (682).

14e. Hepatic and Pancreatic Function

The analgesic effects of cholestasis induced by bile duct ligation were reduced by peripheral and direct intraplantar injections of naloxone and naloxone methiodide (833). Delta opioid agonists diminished the growth of the biliary tree in the development of cholestasis (736). Menk regulated oxidant-antioxidant status in the liver of CBA mice (1053). Chronic morphine produced pro-oxidant effects in hepatic glutathione concentrations and its synthesis pathway in a naltrexone-sensitive manner (887). Naltrexone and NOS inhibitors corrected the blunted chronotopic and inotropic responses to beta-adrenergic stimulation in cirrhotic rats (305). Naltrexone reduced development of hepatic fibrosis and cirrhosis, MMP2 activity, and decreased the number of hepatic stellate cells in bile duct ligated rats. Activation of hepatic stellate cells increased expressed delta-1 receptors and increased TIMP-1 expression following delta-1 and delta-2 agonists respectively (306). Monocytes and granulocytes from bile duct-ligated rats increased the percentage of opioid receptor labeling, but a decrease in opioid receptor expression on leukocytes due to cholestasis (793).

There is a low utility of plasma OFQ/N in patients with hepatocellular carcinoma (1058). Cholestatic patients with pruritus found limited relief in visual analogue scales for pruritus following naltrexone with virtually half showing withdrawal symptoms associated with the antagonist (721). Pro-Enk hypermethylation was enhanced in pure pancreatic juice compared with p53 mutation in the diagnosis of pancreatic carcinoma (858). A patient with pancreatic cancer with metastases to the liver has survived after four years of treatment with intravenous alpha-lipoic acid and naltrexone (80). Rifampin-mediated induction of oxycodone hepatic metabolism resulted in negative urine opioid screening (637). Naltrexone administration in alcoholic patients did not induce hepatic abnormalities and indeed alanine aminotrasferase and aspartate aminotransferase declined across the 12-week treatment (1270).

14f. Renal Function

PVN OFQ/N decreased RSNA, increased urine flow rate, decreased urinary sodium and potassium excretion, and increased free water clearance, effects that were blocked by renal denervation, [Arg8]VP and ORL-1 receptor antagonism. The ORL-1 antagonist, UFP-101 in the PVN increased RSNA and decreased urine flow (614). OFQ/N increased cystomanometric bladder capacity and voiding and decreased maximum bladder pressure and urine leakage in controlling neurogenic detrusor overactivity in patients (631). Naloxone abolished the recurrent inhibition of the bladder C fiber reflex in the cat (750). Renal failure and elevated creatinine kinase in a heroin addict was treated by fasciotomy and hyberbaric oxygen therapy (4).

Cardiovascular Responses

15a. Heart rate

Morphine stimulated vascular endothelial growth factor-like signaling in mouse retinal endothelial cells (187). U50488H increased HR, but decreased both MAP and renal blood flow in lambs aged 1 and 6 weeks (929). KOR is most abundant in the ventricular and atrial myocardium of the heart; DOR, but not MOR is found more in the atria than ventricles. Ligation of a renal artery increased Pro-Enk levels in ventricular myocardium, whereas isoprenaline administration decreased it (1210). Whereas the vagotonic influence of a Menk analogue was not dependent upon a sympatholytic influence, kappa stimulation with U50488H produced a sustained sympatholytic effect that was not easily reversed by NBNI (64). Mu receptor agonism with DAMGO and antagonism with BFNA respectively increased and decreased HR and systolic blood pressure in control rats, but not rats maintained on a high-fat diet despite the latter’s baseline increases in systolic blood pressure (470). Activation of DOR through the opening of KATP channels reduces the severity of post resuscitation myocardial dysfunction (332). The monosialosyl ganglioside GM-1 reduces the vagolytic efficacy of delta2-opioid receptor stimulation (265). Repeated delta-1 opioid receptor stimulation reduces delta2-opioid receptor responses in the sinoarterial node (275). Bradycardia induced by D-Ala2, Leu5, Arg6-enkephalin (dalargin) is associated with activation of peripheral KOR (738). Negative inotropic and chronotropic effects of DOR antagonists are mediated through non-opioid receptors (739). Naloxone-precipitated morphine withdrawal induced PKC inhibitor-sensitive increases in Fos in right and left ventricles, and induced PKC inhibitor-insensitive increases in TH and NE in the heart (22). Baroreceptor activation by pressure in the carotid sinus or by electrical stimulation produced depressor and bradycardic responses that were attenuated by NTS administration of endomorphin-2, but not naloxone (1161). The opioid dipeptide, kytorphin inhibited isoprenaline-induced increases in twitch tension of cardiac muscle without affecting twitch tension itself; this effect was blocked by a kytorphin antagonist and naloxone (657). NO release from lobster heart was enhanced by morphine and decreased by naloxone and the NOS inhibitor, L-NAME (173). However, L-NAME, but not naltrexone exaggerated the hemodynamic response to clonidine in bile duct-ligated rats (1112). Naloxone reduced the immediate lethal response to severe shock in a canine model with a combination of hypertonic-hyperoncotic solutions (385). Administration of naloxone and butorphanol combined with exposure to electric foot shock increased heart and respiratory rate in the lizard, Uromastix hardwickii and the rooster, gallus domesticus (940). Buprenorphine and naloxone in combination with antiretroviral drugs increased the QT interval in HIV-negative and opioid-dependent subjects (56). Abrupt withdrawal from oxycontin decreased the left ventricular ejection fraction and produced new regional wall motion abnormalities (959). Increases in serum BEND by programmed exercise training were correlated with improvement of clinical symptoms and quality of life in female mitral valve prolapse patients (496).

15b. Cardioprotection and ischemic preconditioning

Chronic morphine was more effective than acute morphine in restoring diastolic and contractile function during cardioprotection; G(i) inhibition reduced affected both acute and chronic morphine effects. Whereas a G(s) inhibitor or PKA inhibitor blocked only chronic effects, a PKC inhibitor blocked only acute effects (888). Morphine attenuated hemorrhagic shock-induced hyper permeability (182). Morphine and naloxone respectively decreased and increased the rate of hemorrhage-induced hypovolemia in sheep (363). The JAK/STAT pathway is essential for opioid-induced cardioprotection with JAK2 acting as a mediator of STAT3, Akt and GSK-beta (418). Morphine-induced neuroprotection of hippocampal neurons following oxygen-glucose deprivation was partially blocked by single and combined treatment with chelerythrine, a PKC blocker, epsilonv(1–2), a nPKCepsilon antagonist and MK-801 (333). Opioid preconditioning with Tan-67 induces opioid receptor-dependent neuroprotection against ischemia in rats (1319). Opioid receptor-independent protection of ischemic rat hepatocytes was observed following morphine (581). Central and peripheral opioid receptor antagonists block the cardioprotective effects of fentanyl (649).

Myocardial ischemic tolerance occurs in the newborn rat and involves opioid receptors and mitochondrial K+ channels (807). Menk-Arg6-Phe7 produced cardioprotection in the ischemic preconditioning model (283). Activation of ERK and suppression of calcineurin interact in cardioprotection produced by DOR activation (500). The delta agonist, SNC-121 blunted the ischemic-induced increases in cardiac myocyte death produced by mineral oil layering to reduce gas exchange (882). Hypoxic preconditioning reduces glutamate-induced neuronal injury by increasing cortical DADL binding density without affecting DOR mRNA in a NTI-sensitive manner. In contrast, prolonged hypoxia causes severe neuronal injury and decreases DADL binding and DOR mRNA levels (1307). ARD-353, a nonpeptide delta agonist decreased infarct size following coronary artery occlusion that was blocked by a delta-1 antagonist (1201). Kappa agonists significantly reduced infarct size and plasma lactate dehydrogenase level by ischemia-reperfusion in a NBNI-sensitive manner (1312). Kappa-opioid receptor antagonism improved recovery from myocardial stunning in chronically instrumented dogs undergoing left anterior descending artery ischemia (419). Morphine and fentanyl respectively enhance and retard recovery of ventricular function after cardiopulmonary bypass (808).

15c. Blood pressure

Morphine exposure increased MAP and HR, and spontaneous morphine withdrawal produced a prolonged 72 h increase in MAP as measured by telemetry (761). BEND acts in limbic circuits as a depressor in regulation of arterial pressure as it relates to the development of essential hypertension (617). Blockade of endothelin-1 release contributes to the anti-angiogenic effect of POMC over expression in endothelial cells (628). Endomorphins restore the endothelium-dependent relaxation of the rabbit aorta rings exposed to high D-glucose through a NO-c-GMP pathway (683). Endomorphin2-ol was more potent than endomorphin2, but endomorphin1-ol was less potent than endomorphin1 in reducing systolic arterial pressure and heart rate with the former measure reversed by pretreatment with naloxone, atropine, L-NAME and bilateral vagotomy (1281). Morphine, meperidine, fentanyl and remifentanil all produced a concentration-dependent vasorelaxation of human artery rings that was insensitive to naloxone (431). Vasorelaxation induced by U50488H in the pulmonary artery was blocked by the K(V) channel blocker 4-AP, but not glibenclamide or TEA (1088). Arterial partial pressure of oxygen was significantly in horses receiving morphine during upper respiratory tract surgery (699). Stimulation of sensory neuropeptide release by OFQ/N leads to hyperaemia in acutely inflamed rat knees (1303). Remifentanil, sufentanil and fentanyl each induced a dose-dependent vasodepressor response in the cat pulmonary vascular bed that was attenuated by naloxone and diphenhydramine, but not glibenclamide (562,564,565). In contrast, meperidine induced a dose-dependent vasodilator response in the cat pulmonary vascular bed that was attenuated by naloxone and diphenhydramine, but not glibenclamide (563). The dextro-, but not levo-isomer of tramadol produces a concentration-dependent and naloxone-reversible relaxation of rodent aorta precontracted with phenylephrine (939). Tramadol stereoselectively attenuated endothelium-dependent relaxation in isolated rat aorta induced by Ach (1030). Spinal estrogen attenuates the exercise pressor reflex without changing gene expression of opiate receptors in the DRG (998). Hemorrhage induces c-Fos in arcuate POMC neurons and increases POMC mRNA in the MBH. Caudal arcuate administration of lidocaine inhibited hemorrhagic hypotension and bradycardia without affecting MAP or HR (405). Methylnaltrexone inhibits opiate and VEGF-induced angiogenesis through receptor transactivation (1043). Transcutaneous magnetic stimulation of the Jianshi-Neiguan acupoints decreased the reflex pressor response, effects blocked by median nerve denervation or by pretreatment with general, kappa, delta, but not mu antagonists (1324). Naloxone blocked the ability of SP(1–7) to inhibit SP-induced vasodilation in a rodent model of inflammation of the hindpaw (1217).

Exaggerated opioid analgesia is observed in persons at enhanced risk for hypertension (754). Combined spinal-epidural anesthesia with intrathecal bupivacaine produced analgesia with lower incidence of maternal hypotension for cesarean delivery (1114). Low, but not high levels of trait anger in chronic low back pain sufferers were associated by naloxone-induced impairments in post-pain blood pressure (143). Formation clearance of morphine to M3G is reduced during the first 10 days in neonates undergoing venoarterial extra corporeal membrane oxygenation (901). Remifentanil attenuated hemodynamic changes after induction and tracheal intubation during cesarean delivery (566).

Respiration and Thermoregulation

DAMGO applied to the rostral ventrolateral medulla of newborn rats depressed fourth cervical ventral root inspiratory activity with facial nerve activity continuing to synchronize with pre-inspiratory bursts both in normal pups and pups receiving transverse sectioning between the pre-Botzinger complex and the parafacial respiratory group (867). DAMGO also depressed confocally imaged respiratory center neurons in on-line-calibrated newborn rat brainstem slices (977). Fentanyl-induced modulation of inspiration and expiration in juvenile rats was dissociated with caudal transactions at the level of the facila nerve blocked the latter response, but not the former response (511). Recurrent hypoxia in rats during development increases subsequent respiratory sensitivity to fentanyl (805). Morphine increases the transmesothelial resistance of the sheep parietal pleura in an in vitro preparation (1170). Morphine decreases gill ciliary activity through coupling to NO release in a bivalve mollusk (722). Optimal assessment of anesthesia associated with regular respiration, loss of blink, papillary and pedal withdrawal reflexes was obtained with intra-nasal doses of fentanyl, droperidol and medetomidine in mice (632). Both naloxone and peripherally-acting naloxone methiodide reversed the respiratory depression and analgesia induced by morphine, methadone and heroin, effects that did not appear to be sex-dependent in mice (651). Naloxone and epinephrine were equally effective for cardiopulmonary resuscitation in a rat asphyxia model (193, 194). Naloxone reversed buprenorphine-induced respiratory depression (1147). Naloxone did not increase the minimum alveolar anesthetic concentration of sevoflurane in mice (664). Lenk decreased the burst frequency of the membrane potential of pre-Botzinger complex respiratory center neurons (503). OFQ/N reduced respiratory frequency in an apnea model of isolated brainstem-spinal cord newborn rat preparation, effects reversed by an ORL-1 antagonist, an activator of adenylyl cyclase and a phosphodiesterase-4 (but not 3 or 5) inhibitor (976). OFQ/N inhibited acid-invoked cough in guinea pigs by blocking acid-induced increases in Ca2+ in vagal jugular ganglionic neurons by inhibiting the sustained component of acid-induced inward current (640). Two analogues of OFQ/N, but not nocistatin reversed the inhibition by OFQ/N of electric field stimulation-induced excitatory nonadrenergic-non cholinergic and cholinergic constriction of the guinea pig isolated bronchus (1267). Endomorphin-1 and endomorphin-2 inhibited the response to electrical field stimulation in rat isolated bronchus at low, but not higher frequencies, and also inhibited cholinergic constriction in a naloxone-reversible manner, but had no effect upon the contractile response to exogenous Ach (1280); these effects were present in diabetic rats, but the endomorphin response was less marked (1282). 

The acute agonal period during heroin overdose death is characterized by variations in respiratory distress (482). Heroin produced contact allergy and respiratory and mucosal complaints (475). Maximal buprenorphine-induced decreases in respiratory rate and pupil diameter did not vary across doses in experienced opiate users, nor did subjective effects change (226). High doses of buprenorphine cause naloxone-insensitive respiratory depression, particularly in conjunction with benzodiazepine use (763). Whereas fentanyl produced dose-dependent decreases in respiratory rate in volunteers undergoing anesthesia, buprenorphine caused respiratory depression that leveled at 50% of baseline rates (255). Midazolam and morphine can produce harmful effects on cerebral oxygenation and hemodynamics in ventilated premature infants (1144). In neonates receiving central line placement for ventilation, combinations of morphine and tetracine provided superior analgesia with the former producing respiratory depression and the latter producing erythema (1096). Codeine was no more effective than placebo in an objective measurement of cough in chronic obstructive pulmonary disease (1046). There was no significant difference in the time to extubation after use of remifentanil and sufentanil in combination with propofol as anesthesia in adults undergoing nonemergency intracranial surgery (285).

16b. Thermoregulation

DAMGO administered into the NRM blocked noxious stimulation-evoked suppression of PGE2-induced BAT temperature increases (821). SNC80-induced hypothermia was attenuated by central L-NAME administration as well as by combined peripheral L-NAME and 7-NI administration (942). SNC80-induced produced hypothermia in a NTI-sensitive manner, an effect blocked by the 5HT1A antagonist, WAY100635, and enhanced by non-hypothermic doses of fluoxetine (943). Hydromorphone produced post anesthetic hyperthermia in cats (843). The ability of the GABA B agonist, baclofen in the pre-optic/anterior hypothalamus to decrease tonic activity and increase temperature sensitivity through membrane hyper polarization and decrease of input resistances was inhibited by the mu-opioid agonist, PL-017 (1252). Bile duct-ligated rats displayed hypothermia that was reversed by opioid antagonism and NOS inhibition (791). Naloxone restored heat stress-induced declines in GABA and glycine and prevented heat stress-induced increases in glutamate and aspartate (1019). The SSRI fluoxetine abated naloxone-induced increases in tail-skin temperature in ovariectomized rats (740). The 5HT2A antagonist mirtazapine increased tail temperatures following naloxone administration in morphine-dependent rats (886). Post-burn local hyperthermia in reducing burn injury was blocked by naloxone (1011). Intrathecal morphine intensified the intra-operative hypothermia induced by bupivacaine spinal anaesthesia for caesarean section (491).

Immunological Responses

Intrathecal chronic administration at the maximum tolerated doses of morphine hydromorphone, L-methadone, and naltrexone produced granulomas (21). POMC gene transfer, and particularly AMSH, reduced anchorage-dependent growth of melanoma cells in mice by reducing foci formation in lung attenuating their migratory and adhesive capabilities (678). Morphine decreased metalloproteinase activity through the NO-NOS system, implicating it in the treatment of fibrosarcoma (1016). Morphine stimulated CCL2 production by human neurons (964). Chronic, but not acute morphine induces apoptosis in neurons and astrocytes accompanied by up-regulation of FasL, Fas, Bad and active fragments of caspases-8 and –3 (317). Morphine promoted Jurkat cell apoptosis through pro-apoptotic FADD/P53 and anti-apoptotic PI3K/Akt/NF-kappaB pathways (1273). Morphine enhanced apoptosis of cultured human colonic cells in a NOS inhibitor-sensitive fashion, and produced a breach in the host defense barrier in a NOS inhibitor-insensitive fashion (361). Non-apoptotic cell death is induced by morphinone in human promyelocytic leukemia HL-60 cells (1104). Reduction of the brain apoptosis-related proteins, Fas, FasL, Bad and Bax occurred following naltrexone in mice (987). Morphine-induced suppression of natural killer cell activity was blocked by the NPY Y1 antagonist, BIBP3226 (992). Morphine reciprocally regulates IL-10 and IL-12 production by monocyte-derived human dendritic cells and enhances T cell activation (772). Morphine, DAMGO, DPDPE and U-69593 all produced neuroprotective effects against low-temperature-induced cell death in cultured hamster hippocampal neurons (1106). TPA increases expression and regulation of MOR in TPA-differentiated HL-60 promyelocytic leukemia cells (75). Interferon-gamma down-regulates transcription of the MOR gene in neuronal and immune cells (612). Morphine modulates the monocyte-macrophage conversion phase in an opiate antagonist-sensitive manner (456), and naloxone inhibits macrophage activation and atherosclerosis formation in mice (688). Acute morphine and morphine withdrawal inhibit phagocytosis through pertussis toxin-sensitive and –insensitive mechanisms respectively with the latter, but not former manipulation increasing cAMP levels (705). Over expression and blockade of cyclin-dependent kinase 5 respectively augmented and diminished DAMGO-induced neuroprotection from serum deprivation (1195). DAMGO induced up-regulation of annexin V apoptosis-inducing gene expression in PC12 cells stably transfected with MOR (1291). DAMGO inhibited ERK-1 and Akt signaling pathways activated by the chemokine receptor CXCL12 and abolished its neuroprotective effects during NMDA-induced neurotoxicity, effects reversed by general and mu antagonists (883). SNC, but not mu or kappa agonists promoted neural differentiation from multipotent stem cells through the activation of Trk-dependent tyrosine kinase (814). There is a requirement for Id1 in BEND opioid-induced oligodendrogensis in cultured adult rat hippocampal progenitors (899). Insulin decreased histamine and 5HT, but not BEND content in the thymus of in vitro preparations (251). The nucleus of rat peritoneal mast cells contained serotonin and histamine, but not BEND, insulin or triiodothyronine (252). EDAC fixation increases histamine, but not BEND, levels in such cells relative to paraformaldehyde (253). 

Loperamide reversed the inhibition of insulin-stimulated 2DG uptake by tumor necrosis factor alpha in myoblast C2C12 cells (601). Chimeric DNA vaccine reverses morphine-induced immunosuppression and tumorigenesis (206). Morphine dependence diminished virus evolution in SHIV/SIV-infected rhesus macaques with an inverse relation between virus evolution and onset of clinical symptoms (1121). Chronic morphine exposure causes pronounced virus replication in the cerebral compartment and accelerated onset of AIDS in SIV/SHIV-infected Indian rhesus macaques (620). HIV-1 gp120 up-regulation of MOR occurred in TPA-differentiated HL-60 cells (74). Menk maintained the viability of SIV-infected cells through suppression of the expression of caspase-3 (1248). Delta opioid agonists attenuate HIV protein TAT(1–72)-induced oxidative stress in SK-N-SH cells (1176). The increased vulnerability of ApoE4 neurons to HIV proteins and opiates were provided protection by diosgenin and L-deprenyl (1137). HIV-1 Tat and opiate-induced changes in astrocytes promote chemotaxis of microglia through the expression of MCP-1 and alternative chemokines (312). CCR2 mediates the increases in glial activation by exposure to HIV-1 Tat and opiates (311). There was a correlation between SIV Tat evolution and AIDS progression in cerebrospinal fluid of morphine-dependent as compared to control macaques infected with SIV and SHIV (847). Naltrexone inhibits alcohol-mediated enhancement of HIV infection of T lymphocytes (1189). Further, an in vitro model of morphine withdrawal manifests the enhancing effect of HIV infection of human T lymphocytes through the induction of substance P (1190). Acupuncture reduced leukocyte migration into the mouse air pouch in a peripheral, but not spinal naloxone-reversible manner, but did not produce an anti-inflammatory effect (580).

Down-regulation of the opioid growth factor receptor is associated with progression of squamous cell carcinoma of the head and neck (760), yet the opioid growth factor receptor is unaltered with the progression of human pancreatic and colon cancers (1286). Regulation of corneal repair was effectuated by particle-mediated gene transfer of opioid growth factor receptor complementary DNA (1287) with corneal safety of topically applied naltrexone validated (1285). Adaptation of the homeostatic ocular surface epithelium occurred following chronic treatment with naltrexone (1288). Pro-ENK KO mice developed less severe clinical signs of experimental autoimmune encephalomyelitis than wild-type mice that was accompanied by a reduction in MOG(35–55)-specific IFNgamma-producing cells (1212). Tumor-cell-targeted Menk analogues containing unnatural amino acids display in vitro antitumor activity (481). DADL stimulates Akt-dependent phosphorylation of c-jun in T cells (1010). DPDPE triggers monocyte adhesion even in pertussis toxin-treated cells, indicating involvement of G proteins other than Gi. This adhesion requires integrins, PI3Kgamma activation, and involves Src kinases, a guanine nucleotide exchange factor and a small GTPase (893). Chemotaxis of human and rat leukocytes is observed following the delta-selective non-peptidergic opioid SNC 80 (868). Deletion of the DOR in mice increases skin differentiation and delays wound healing (95). Kappa opioids induce a reversible inhibition of CFU-GM from CD133(+) cord blood cells (286). OFQ/N and ORL-1 KO respectively stimulate and reduce TNFalpha and IFNgamma transcripts in the spleen when challenged with staphylococcal enterotoxin A (407). Lipopolysaccharide induces ProEnk transcription in hypophysiotropic neurons of the rat periventricular hypothalamic nucleus suggesting a neuroendocrine role for Enk during immune stress (321). Interleukin-1 beta contributed to the up regulation of KOR mRNA in the DRG in response to peripheral inflammation (926). Lipopolysaccharide-induced interleukin-6 production in a mouse monocyte cell line is reduced by the kappa agonist, U50488H in a NBNI-sensitive manner (879). Lipopolysaccharide induced PC1/3 and PC2 substrate pro-Enk in the marginal zone of rat spleens (629). The expression of Pro-DYN gene is down-regulated by activation with lipopolysaccharide in U-937 macrophage cells (1086).

Source and Footnotes, as well as some capitals i dont copy: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2693002/?tool=pubmed