An important aspect of drug distribution involves a set of proteins positioned within the endothelial cells of the CNS vasculature [198]. These proteins form the essence of what is commonly referred to as the Blood Brain Barrier (BBB). The proteins of the BBB involve structural proteins that line the intercellular walls between endothelial cells and form physical barriers (tight junctions) to prevent paracellular diffusion of chemicals. The BBB also includes a set of proteins that enable transport of molecules into the endothelial cell from the lumen of the capillary for trafficking intracellularly and exit on the brain side. Conversely, the BBB includes a complementary set of proteins that enable transport of molecules to enter the endothelial cell from the brain for trafficking intracellularly to exit to the capillary lumen. Both forms are described as transcellular transport. Finally, the BBB also contains a specific set of proteins (ABC transporters) that essentially capture passively diffusing molecules as they enter the endothelial cell and efflux to the lumen of the capillary. These proteins are known as BBB efflux proteins. It has been appreciated for some time that a number of opioids are substrates for one specific member of this class of transporter, P-glycoprotein (P-gp) [199]. Evidence in support of this includes enhanced brain uptake of morphine, methadone, and fentanyl in P-gp knock-out mice following intra-arterial perfusion [200], or microdialysis [201]. It is known that P-gp both contributes to preventing entry of opioids circulating in the blood to the brain as well as efflux opioids circulating in the CNS extracellular space to the blood [202]. Additionally, another member of the ABC BBB transporter family, the multidrug resistance protein (MRP), has recently been shown to similarly reduce the BBB transport of morphine from the systemic circulation to the CNS and to efflux morphine from the brain to the blood [203]. Specifically, it has been observed that antisense treatment to downregulate the expression of either P-gp [202] or MRP [203] or use of their respective knock-out mice resulted in enhanced analgesia induced by systemically administered morphine. This enhancement occurs presumably by enabling increased CNS levels of morphine. Interestingly, it has also be demonstrated that when opioids are delivered centrally, a component of the drug is efflux by BBB transport proteins to the periphery where activation of peripheral opioid receptors contributes to the overall analgesic effect synergistically with central opioid receptors [202]. Specifically, when p-glycoprotein [202] or MRP transporters [203] are reduced either by antisense or knock-out technology, the analgesic dose–response curves of intracerebroventricularly (i.c.v.) delivered morphine are shifted rightward, indicating diminished potency of the overall effect of i.c.v. morphine, presumably due to reduced efflux of morphine from the brain to the systemic circulation and, consequently, reduced activation of peripheral opioid receptors. These data are important observations in support the concept that P-gp and MRP also contribute to the efflux of opioids from the brain to the systemic circulation.

Analgesia of morphine [202, 204], fentanyl [204], and methadone [204] has been reduced in either P-gp KO mice or in the presence of P-gp inhibitors. It is, however, unclear what the therapeutic impact of P-gp influenced alterations in opioid concentrations may be. There is increasing evidence to suggest that P-gp levels are altered following a variety of environmental or physiological processes. Chronic administration of opioid has been reported to upregulate P-gp levels in a variety of tissues [205]; this would suggest that CNS concentrations under such conditions would be lowered and this may contribute to development of apparent opioid tolerance [206]. It would stand to reason that alterations in expression and/or function P-glycoprotein could arise under conditions of chronic pain that could influence the analgesic responsiveness of P-gp substrates, such as opioids. In fact, a study of peripheral inflammatory hyperalgesia using the standard intraplantar carrageenan model has demonstrated 40 % elevation in P-gp expression with a corresponding reduction in the antinociceptive efficacy of morphine [207]. Further, more detailed analysis has revealed that carrageenan-induced hypersensitivity results in protein trafficking alterations in P-gp at the level of the bilipid membrane [208]. Therefore, these studies provide the proof-of-concept that the condition of chronic pain can influence protein expression and function of P-gp. Peripheral inflammation is one model of chronic pain. More work is needed to understand the status of P-gp under a wide variety of pain conditions of diverse origin. There is clinical evidence to suggest that genetic variability in the ABC1 gene contributes to variability in patient responsiveness to opioids. Specifically it has been shown that methadone [209] or morphine [210] dosing for addiction and pain treatment respectively was associated with genetic variants of ABCB1, the gene that encodes P-gp. It seems that the interaction of the opioids with the P-gp and other transport proteins is highly complex involving both substrate interaction with P-gp and/or modulation of the expression of the P-gp system and/or patient genotype for the P-gp gene [205]. It would be greatly beneficial, therefore, to consider the status and impact of P-gp under the varied conditions of chronic pain and opioid self-administration models that are currently used separately and increasingly combined to further our understanding as to how subjects with established chronic pain respond to chronic opioid treatment.

When the condition of chronic pain is introduced into the complexity of the system, there are a number of points in the pharmacokinetic profile of each compound that may influence the reinforcing effects of a particular opioid. These aspects have been featured in this chapter. Although opioid receptor agonists are thought to converge on a common target, their pathways to the target, activities at the target, and retreat and distributions away from the target and out of the organism can be distinct. In experimental design and analysis of data, these considerations will be valuable to investigators intending to initiate studies of opioid self-administration in models of established chronic pain.