TM residues within the lipophilic environment of the cell membrane are key in ligand recognition and/or signal transduction and are expected to be oriented toward a relatively hydrophilic central cavity (Surratt et al. 1994). Structural motives important for ligand binding and subtype speciﬁcity have been iden- tiﬁed for the opioid receptors from experimental mutagenesis studies and computer modeling. A current view of how GPCRs achieve the transition be- tween resting and active conformations to convey external signals across the
cell membrane is that ligands bind with hydrophilic and aromatic residues within the helical core. This triggers outward movements of TM helices 3, 6, and 7 to promote the formation of the active receptor state that results in G protein coupling and signal transduction, as it was shown for the structurally related muscarinic receptor (Hulme et al. 1999). Charged and polar amino acid residues are useful starting points in identifying important residues for ago- nist and antagonist binding within the binding pocket. For opioid receptors, histidine, asparagine, and tyrosine residues within TM 3, 6, and 7 are critical for receptor activation (Mansour et al. 1997). Random mutagenesis is another tool to identify important structure–function relationships. Such a strategy
with the entire δ-receptor revealed 30 point mutations leading to a constitutive activity of the receptor (Decaillot et al. 2003). Mutagenesis studies are useful tools to identify potential domains important for ligand binding. However, they
do not allow the deﬁnitive identiﬁcation of binding sites, because the mutation of single amino acids could affect the amino acid side-chain interaction and subsequently the secondary and/or tertiary structure of the entire receptor.
The C-terminal portion of the opioid receptor determines coupling to second messenger molecules and is important for receptor trafﬁcking. μ-receptor point mutations of any of the Ser/Thr within the C-terminal tail result in signiﬁcant reductions in the rate of receptor internalization in HEK cell lines (Koch et al. 1998, 2001). This was conﬁrmed using CHO-K1 cells expressing μ-receptors and a mutant form, in which all Ser and Thr residues from the third cytoplasmic loop and C-terminal were changed to alanine (Capeyrou et al. 1997; Wang 2000).
Opioid receptors are prototypical Gi /Go-coupled receptors. Opioid signals are efﬁciently blocked by pertussis toxin that adenosine diphosphate (ADP)-
Fig. 2a Opioid ligands induce a conformational change at the receptor which allows coupling of G proteins to the receptor. The heterotrimeric G protein dissociates into active Gα and Gβγ subunits (a) which can subsequently inhibit adenylyl cyclase (b), decrease the conductance of voltage-gated Ca2+ channels, open rectifying K+ channels (c), or activate the PLC/PKC pathway (d) that modulates Ca2+ channel activity in the plasma membrane (e). b Opioid receptor desensitization and trafﬁcking is activated by G protein-coupled receptor kinases (GRK). After arrestin binding, the receptor is in a desensitized state at the plasma membrane (a). Arrestin-bound receptors can then be internalized via a clathrin-dependent pathway, and either be recycled to the cell surface (b) or degraded in lysosomes (c)
ribosylates and inactivates the α-subunits of Gi/Go proteins. After opioid ago- nists bind to the receptor, dissociation of the trimeric G protein complex into Gα- and Gβγ-subunits can subsequently lead to the inhibition of cyclic 3 5 adenylyl cyclase (cAMP) and/or to direct interaction with K+, Ca2+, and other ion channels in the membrane (Fig. 2a). Ion-channel regulation by opioids
is mainly mediated via direct G protein βγ-subunits (Herlitze et al. 1996). All three opioid receptors couple to various N, T-type and P/Q-type Ca2+ channels, suppress Ca2+ inﬂux and subsequently the excitation and/or neurotransmitter release in many neuronal systems. A prominent example is the inhibition of (pronociceptive) substance P release from primary afferent sensory neurons in the spinal cord and from their peripheral terminals (Kondo et al. 2005).
At the postsynaptic membrane, opioid receptors mediate hyperpolarization by activating K+ channels, thereby preventing excitation or propagation of action potentials. Four families of potassium channels play important roles in antinociception, including voltage-gated K+ channels (Kv), inward recti-
ﬁer K+ channels (Kir), calcium-activated K+ channels (KCa), and two-pore K+ channels (K2P) (Ocana et al. 2004). In the central nervous system (CNS), opioid-induced antinociception is associated with activation of voltage-gated
K+ channels of the Kir channel family (Ocana et al. 1995). Some opioids also induce opening of calcium-activated K+ channels (Stretton et al. 1992).
Apart from Ca2+ and K+ channels, opioid receptors may regulate the
functions of other ion channels. For example, opioids suppress tetrodotoxin- resistant sodium-selective and nonselective cation currents, which are mainly expressed in nociceptors (Gold and Levine 1996; Ingram and Williams 1994). In addition, a number of studies indicate that opioids can directly modulate N - methyl-d-aspartate (NMDA) receptors at presynaptic and postsynaptic sites within the CNS (Mao 1999). NMDA receptors can be blocked at the same
site where Mg2+ and ketamine interfere with the ion channel. In addition, we recently showed that activation of opioid receptors modulates the transient
receptor potential vanilloid type 1 (TRPV1), a member of the ligand gated ion channels (J. Endres-Becker, P.A. Heppenstall, S.A. Mousa, A. Oksche, M.
Schäfer, C. Stein, C. Zöllner, submitted for publication). TRPV1 is involved in thermosensation and nociception and is mainly expressed in peripheral sensory neurons. It was also suggested that opioid receptors regulate the phos- pholipase Cβ (PLCβ) pathway via Gβγ-subunits (Chan et al. 1995) or Gq proteins (Rubovitch et al. 2003). PLC activation mobilizes phosphokinase C (PKC) that opens calcium channels in the plasma membrane. The entry of Ca2+ into the cell stimulates calcium-activated adenylyl cyclases to produce cAMP. However,
the physiological relevance of such a potential bidirectional regulation of in- tracellular cAMP by opioid receptors is not completely solved at the moment and most data indicate that opioids mainly inhibit cAMP production.
Mitogen-activated protein kinase (MAPK), also known as extracellular signal-regulated kinase (ERK), is regarded as a major pathway for growth factor signaling from the cell surface to the nucleus (Seger and Krebs 1995). MAPK activation has been noted in response to agonist stimulation of many GPCRs, including all opioid receptor types (Belcheva et al. 1998; Ignatova et al.
1999; Schulz et al. 2004a). A suggested mechanism includes the activation of PLC, generating diacylglycerol that binds to PKCε, leading to its phosphory- lation. PKCε can then signal to matrix metalloproteinases, which can cleave
membrane-anchored epidermal growth factor receptor (EGF)-type ligands, thereby initiating EGF receptor transactivation and ultimately activation of the MAPK phosphorylation cascade (Belcheva et al. 2005; Pierce et al. 2001). So far, little is known about the functional signiﬁcance of this phenomenon. Besides an impact of opioids on neural development, the MAPK pathway might be involved in homologous desensitization of the μ-receptor (Polakiewicz et al.
1998; Schmidt et al. 2000). Future studies might investigate the role of the MAPK
cascade in opioid tolerance and dependence.
On the cellular level long-term opioid treatment can result in the eventual loss of opioid receptor-activated function (i.e., desensitization). Three gen- eral mechanisms are associated with desensitization of GPCRs: (1) receptor phosphorylation, (2) receptor internalization and/or sequestration, and (3) receptor downregulation (i.e., a reduced total number of receptors). Opioid receptors are substrates for second messenger kinases (i.e., PKC) and for members of the GPCR kinases (GRK). Opioid receptor phosphorylation by these kinases increases the afﬁnity to arrestin molecules. Arrestin-receptor complexes sterically prevent coupling between receptor and G proteins and promote internalization via clathrin-dependent pathways (Fig. 2b; Law et al.
2000). Agonist-induced internalization of the receptor via the endocytic path- way has been thought to contribute directly to tolerance by decreasing the number of opioid receptors on the cell surface. However, more recent stud- ies have shown that morphine fails to promote endocytosis of opioid re- ceptors in cultured cells (Eisinger et al. 2002) and native neurons (Stern- ini et al. 1996), although it is highly efﬁcient in inducing tolerance in vivo (Sim et al. 1996). Moreover, endocytosis and recycling of the opioid recep- tor was shown to dramatically decrease opioid tolerance and withdrawal (Koch et al. 2005). These ﬁndings suggest that desensitization and receptor internalization might be a protective mechanism and prevent the develop- ment of tolerance. Drugs that do not cause receptor internalization, such as morphine, may have higher propensities to induce tolerance. In contrast to full agonists such as [D-Ala(2)-MePhe(4)-Gly-ol]enkephalin (DAMGO), mor- phine stimulated a selective phosphorylation of the carboxy-terminal residue
375, indicating that morphine-desensitized receptors remained at the plasma membrane in a Serin-375-phosphorylated state for prolonged periods (Schulz et al. 2004b). Recycling of opioid receptors to the plasma membrane pro-
motes rapid resensitization of signal transduction, whereas targeting to lyso- somes leads to proteolytic downregulation. A protein that binds preferen-
tially to the cytoplasmic tail of opioid receptors was identiﬁed and named as GPCR-associated sorting protein (GASP) (Whistler et al. 2002). It was sug- gested that GASPs modulate lysosomal sorting and functional downregulation of GPCRs.
Tolerance in vivo describes the phenomenon that the magnitude of a given opioid effect decreases with repeated administration of the same opioid dose, or that increasing doses of an opioid are needed to produce the same effect. All opioid effects (e.g., analgesia, respiratory depression, sedation, consti- pation) can be subject to tolerance development. It is evident that opiate- induced adaptations occur at multiple levels in the nervous and other or- gan systems, beginning with regulation of opioid receptors themselves and extending to complex networks including learned behavior and genetic fac- tors (von Zastrow 2004). The importance of opioid receptor endocytosis for chronic adaptation of the intact nervous system is not yet understood. Indi- rect evidence that arrestin-dependent μ-receptor desensitization contributes to morphine tolerance in vivo comes from studies in arrestin-3 and GRK3 knockout mice. Functional deletion of the arrestin-3 gene resulted in remark- able potentiation and prolongation of the analgesic effect of morphine (Bohn et al. 2000; Bohn et al. 2002) but had no effect on the acute antinociceptive potency of etorphine, fentanyl, or methadone (Bohn et al. 2004). However, to date the majority of studies on opioid tolerance have been performed in the absence of painful tissue injury. This may explain some of the discrep- ancies between experimental (Smith et al. 2003) and clinical studies (Stein et al. 1996). Tolerance is not ubiquitously observed in clinical routine and is in many cases explained by increasing nociceptive stimulation with progressing disease (e.g., cancer pain; Zech et al. 1995). Animal models of pathological situations have also produced evidence for a reversal of tolerance to mor- phine, e.g., during intestinal inﬂammation (Pol and Puig 1997). These ﬁndings indicate differences in the development of opioid tolerance (and possibly in receptor desensitization and recycling) under pathological situations. Future studies are necessary on the development of opioid receptor tolerance in the presence of postoperative pain, arthritis, or other types of inﬂammatory and chronic pain.
Several investigators have focused on the concept that tolerance can be counteracted by NMDA receptor antagonists (Mao 1999; Price et al. 2000). NMDA receptors are a subclass of excitatory amino acid receptors that, once activated, facilitate calcium inﬂux into neurons. Although it has been shown NMDA receptor antagonists, e.g., MK801, inhibit tolerance to the analgesic effects of repeated morphine administration (Elliott et al. 1995; Price et al.
2000), the underlying mechanisms have not been fully elucidated.