α2Adrenoceptors in Pain Modulation : Which Subtype Should Be Targeted to Produce Analgesia?

Sir Ivan Magill Professor of Anaestheticsm.maze@ic.ac.ukSenior Lecturer Magill Department of Anaesthetics Chelsea and Westminster Campus Imperial College of Science, Technology, andMedicineUniversity of LondonLondon, United Kingdomm.fujinaga@ic.ac.ukThe α2-adrenoceptor agonists may become an attractive alternative to the currently used analgesics because they are devoid of respiratory depressant effect and addictive liability. However, some of their pharmacologic properties, namely sedation and hypotension, are hindering the clinical use of subtype nonselective α2-adrenoceptor agonists for pain management in the ambulatory setting. This has provoked two parallel lines of investigation to elucidate which function each of the three subtypes (i.e. , α2A, α2B, and α2C) mediates and to define the distribution of the subtypes. The study by Ongjoco et al. 1in this issue of ANESTHESIOLOGY sheds further light on this issue. The rationale for this approach is to identify a single receptor subtype target that mediates the desired analgesic effect and is incapable of producing the undesirable pharmacologic properties. Armed with this information, chemists can then direct a “target-based” drug-design program to synthesize a novel, subtype-selective, α2-adrenoceptor agonist. The genes for each of these receptor subtypes have been cloned, making it possible to investigate their transcription, expression, and their role in physiologic and pathologic settings. The mRNA transcription patterns of these receptor subtypes can be elucidated qualitatively (e.g. , by in situ hybridization) and quantitatively (e.g. , by northern assay, ribonuclease protection assay or reverse-transcription polymerase chain reaction [RT-PCR] assay). Furthermore, recombinant DNA methodology has made it possible to address the role of the α2-adrenoceptor subtypes by creating genetically modified reagents with dysfunctional (“transgenic”) or deficient (“knockout”) subtypes. Resolution of the protein sequence from the genome has facilitated the production of antibodies directed against the cognate proteins, which facilitates immunologic-based investigations (e.g. , qualitative immunohistochemistry assay and quantitative western assay).Early pharmacologic studies showed that the activation of α2Aadrenoceptor mediated the analgesic effect of α2-adrenoceptor agonists. A prazosin-sensitive α2-adrenoceptor subtype was shown to inhibit neurotransmitter release from spinal cord preparations, suggesting a role for the α2B- or α2C-adrenoceptor subtypes in antinociception. 2In support of this finding, ST-91, an α2Csubtype–preferring agonist, was shown to induce antinociception in rats when it was administered intrathecally, and this effect was blocked by prazosin. 3More sophisticated analysis of the subtypes involved could not been accomplished by classic pharmacologic approaches because of a lack of subtype-selective agonists or antagonists. The development of molecular biologic techniques has provided more subtype-specific approaches. Studies using D79N mice, in which the gene for α2A-adrenoceptor subtype has been rendered dysfunctional, established a role for α2Asubtype in antinociception, at least in the development of thermal analgesia. 4–6Yet the D79N mice exhibited some residual α2-adrenoceptor agonist–induced effects on spinal analgesia in the substance P–induced pain test. 6Studies using other transgenic and knockout models have shown that the α2Asubtype is responsible for the hypotensive and bradycardic actions of α2-adrenoceptor agonists, whereas the α2Bsubtype is responsible for its vasoconstriction properties and the α2Csubtype for hypothermic action and modulation of dopaminergic activity. 7–9A recent study showed that intrathecally administered α2Cantisense oligodeoxynucleotides decreased mechanical antinociception induced by clonidine and other antinociceptive agents in the rat, 10suggesting that the α2Csubtype is also involved in antinociception. However, studies of genetically modified reagents may also provide misleading information. In the case of the knockout models, these animals not only lack the protein of interest, but the organisms have had an opportunity to adapt to the deficiency during their development. These putative adaptive responses to the knockout remain largely undocumented. Therefore, in the situation in which the null mutation yields no loss of the investigated function, one cannot yet conclude that the molecule is not necessary for that function. 11The mechanism whereby antisense technology interferes with expression of the protein of interest in fully developed animals (“knockdown”) remains poorly understood. In fact, the administration of sense oligonucleotides have also resulted in altered behavioral function, 12which underscores the necessity for demonstrating that expression of the cognate protein (e.g. , by western assay) has been reduced before ascribing a functional role to that protein.Previously, detailed descriptions of the distribution of α2-adrenoceptor subtypes had not been possible by conventional approaches using radiolabeled ligand binding assays because of the lack of subtype selective ligands. Recently, immunohistochemical analysis of specific protein expression and in situ hybridization assay of gene transcription have been used to overcome this problem. Studies using rats and mice have shown that α2Ais the predominant receptor subtype in the brain, which is widely distributed throughout the brain. In the rat spinal cord, α2Aand α2Csubtypes have been identified by in situ hybridization; α2Aare diffusely distributed, whereas the α2Cis located mainly in the cells of the dorsal root ganglia. 13,14In contrast, Smith et al. , 15using in situ hybridization, previously demonstrated that α2Aand α2Bsubtypes predominate in human spinal cord, with the α2Csubtype only sparsely represented. This discrepancy between rodents and humans instigated this current study regarding α2-adrenoceptor subtype distribution in human dorsal root ganglia. Because of the small size of dorsal root ganglia, they implemented a more sensitive quantitative RT-PCR assay. Data from these studies showed that mRNAs for α2Band α2Csubtypes were almost equally transcribed in the dorsal root ganglia, but that the α2AmRNA transcript was relatively deficient in this location. This distribution differs from those found in rodents and that present in the human spinal cord. There are possible methodologic reasons to explain these discrepancies. Quantitative RT-PCR assay provides less precise quantitation than that obtained by either ribonuclease protection assay or northern assay. An additional concern relates to degradation of the tissue mRNA over time because the material was harvested from autopsies performed from 2.5 to 20 h after death. Are we to assume that the mRNAs of three receptor subtypes degrade to the same degree over time? Notwithstanding the possible technical concerns, RT-PCR assay does not provide qualitative information about its localization on different cell types in the dorsal root ganglia; such information necessitates immunohistochemical and in situ hybridization assays. For examples, it is possible that, even though there is a relative deficiency of α2AmRNA, the transcripts may all be present on important nociceptive neurons. Furthermore, there is not necessarily a 1:1 stoichiometric relation between the amount of transcript (mRNA) and the amount of protein subsequently translated and functionally expressed in the plasma membrane.Although no one technique is likely to be the panacea for all our questions about something as complex as the molecular basis for behavior, the approach by Ongjoco et al. 1exploits the profound advances that have been made in the neurosciences in the last decade. These data challenge the unproven assumption that data in rodents can be extrapolated to humans. Still, these molecular genetic techniques, and especially the information that they provide, have reduced the mechanistic questions, which have plagued our clinical discipline, into tractable segments.