news and views

Neuropharmacology increase selectivity for the J.L receptor. Further­ more, the long-lasting analgesic activity of Another opiate for the masses? endomorphin-1 and -2 could reflect pro­ tection from exoproteolytic degradation David Julius conferred by the carboxy-terminal amidation. By analogy with other peptide hormones he active ingredients of natural toxins and neurotransmitters8, including the endor­ Tand herbal remedies have helped to Endomorph ins phins, one would expect the endomorphins unlock the secrets of biological Dynorphins (3-endorphin Enkephalins to be excised from a larger polyprotein pre­ processes ranging from cell division to cursor. Cloning the gene for this hypothetical synaptic transmission. A notable example is pro-endomorphin would serve at least three opium- an extract of the poppy plant that purposes. First, it would confirm that the has been used for centuries to induce eu­ endomorphins are indeed endogenous phoria and relieve pain. The active compo­ neuropeptides, derived by specific proteolyt­ nent of opium is morphine, and its synthetic ic cleavage events. Second, sequence analysis congener is heroin 1'2• The analgesia, reward­ of the gene would clarify any genetic relation­ seeking behaviour and physical dependence ship between endomorphins and the pep­ elicited by rnorphine2'3 are mainly mediated Figure I Known agonists for the K, J.1 and o opioid tides that led to their isolation, including Tyr­ through the J.L opioid receptor and, on page receptors. Zadina et al.' have identified two MIF-1 and Tyr-W-MIF-1, and other biologi­ 499 of this issue, Zadina and colleagues4 naturally occurring endorphins (termed cally active peptides might be discovered in report the isolation of two new J.L-selective endomorphin-1 and -2) which bind with high the process. Third, the endomorphin gene peptide ligands from mammalian brain. specificity and high affinity to the J.1 receptor. would provide sensitive and specific molec­ These peptides (called endomorphin-1 and Relative affinities of various agonists for different ular probes for determining where these -2) rival morphine in both their potency as opioid-receptor subtypes are depicted by arrows. peptides are expressed in the nervous system. agonists and their analgesic activity. If they If the endomorphins are the endogenous are, indeed, natural ligands for the J.L recep­ peptide from mammalian brain extracts7, ligands for J.L receptors, they might be found tor, then they should provide new molecular Tyr-W-MIF-1 (Tyr-Pro-Trp-Gly-NH2) . Tyr­ in pain-modulating, reward-seeking or tools for dissecting the endogenous opiate W-MIF-1 is a more potent opiate agonist emetic centres of the nervous system (for pathways in the brain and spinal cord. than Tyr-MIF-1 , and it has a respectable example, the spinal-cord dorsal horn, peri­ The quest to determine how the opiate selectivity for J.L receptors over oor K recep­ aqueductal grey, area postrema, striatum, alkaloids mediate euphoric, analgesic and tors (200-300-fold). But its affinity for J.L amygdala and nucleus accumbensf None­ addictive states has helped to shape a central receptors (based on competitive binding theless, questions about the peptide distribu­ tenet of modern neuropharmacology: assays with the J.L-selective endorphin ana­ tion can perhaps be addressed with the namely, that toxins and drugs usurp sites at logue 3H-DAMGO) is modest (K; 70 nM). specific antibody that was used by Zadina and which endogenous factors normally act to To identify a peptide with greater affinity colleagues4 to purify the endomorphins. transduce biological signals. Two discoveries for the J.L receptor, Zadina et al. used a solid­ Could the endomorphins provide relief affirmed this hypothesis. First, radio labelled phase synthesis technique known as poly­ from pain without eliciting the negative opiate alkaloids were shown to bind to ethylene pin technology to generate 20 Tyr­ symptoms that are associated with mor­ specific, high-affinity sites in the brain, W-MIF-1 derivatives, representing all of the phine, such as respiratory depression, corresponding to three distinct guanine­ possible amino-acid substitutions at the nausea, tolerance and addiction? Although nucleotide-binding (G) protein-coupled fourth position of the peptide. Phe4 - the pharmacological and genetic studies3 indi­ opioid-receptor subtypes, termed K, J.L and 85 peptide with phenylalanine at the fourth site cate that this is unlikely, the answer may be (Fig. 1). Second, endogenous morphine-like (Tyr-Pro-Trp-Phe-NH2)- turned out to 'yes' if the endomorphins are more J.L-selec­ substances were identified in the brain. have the desired characteristics, including tive than morphine, or if the activation of J.L These so-called endorphins are peptides that sub-nanomolar affinity for J.L receptors, and receptors by structurally distinct agonists activate opioid-receptor subtypes, with very high selectivity of binding to J.L versus o has differential effects on signalling. We next characteristic rank-order potencies. But, (>4,000-fold) or K (> 15,000-fold) sites. When need to examine interactions between the until now, no single endorphin had been it was injected into the brain or spinal column endomorphins and cell lines that express shown to bind with both high selectivity and of mice, synthetic Phe4 proved to be as potent cloned opioid receptors. Finally, does this high affinity to the J.L receptor, begging the as morphine in producing a long-lasting work signal the beginning or the end of the question of whether the true 'endogenous analgesic response. Moreover, the response search for endogenous opiate peptides? The morphine' had, in fact, been discovered. was reversed by administration of broad­ modest binding selectivity of other known The route that was taken by Zadina et al. 4 spectrum or J.L-selective opiate antagonists. endorphins to the three opioid-receptor to uncover the new endomorphins is inter­ Next, the authors tookadaringleap offaith subtypes indicates that peptides with com­ esting in itself, because it involved an un­ -they predicted that the nervous system parable selectivity for the o and K receptors usual interplay between pharmacology and might express its own version of the Phe4 pep­ are out there waiting to be discovered. 0 combinatorial chemistry. The story begins tide. Using an antibodythatwas specific for the David Julius is in the Departmen t of Cellular and with the isolation of a tetrapeptide called Phe4 peptide, they isolated immunologically Molecular Pharmacology, University of Californ ia

Tyr-MIF-1 (Tyr-Pro-Leu-Gly-NH2) from crossreactive material from bovine brain at San Francisco, San Francisco, California 94143- bovine and human brain6• This peptide acts extracts. These extracts indeed contained 0450, USA. as both an agonist and an antagonist in phys­ endogenous Phe4 peptide (endomorphin-1), l. Snyder, S. Drugs and the Brain (Freeman, New York, 1986). iological assays for binding to the J.L receptor, along with the closely related species Tyr-Pro­ 2. fields, H. Pain (McGraw- Hill, New York t987). 3. Matthes, H. eta!. Nat11re 383, 819-823 ( 1996). which measure the opiate-dependent inhibi­ Phe-Phe-NH2 (endomorphin-2). Because 4. Zadina, J. E., Hac kler, L., Ge, L -J. & Kastin, A. J. Nature 386, tion of electrically induced contractions in other endorphins contain the amino-terminal 499-502 ( 1997). the guinea-pig ileum. An antibody was signature sequence Tyr-Gly-Gly-Phe, it may 5. Ki effer, B. Cell. Mol. Neurobiol. 15, 615-635 ( 1995 ). 6. Horvath , A. & Kastin , A. f. Bioi. Chem. 264, 2 175-2 179 ( 1989). generated to Tyr-MIF-1, then a radio­ be that a stable structure imposed by the pro­ 7. Erchegyi, )., Kasli n, A . & Zadina, ). Peptides 13, 623-63 1 ( 1992). immunoassay was used to isolate a related line residue at the second position helps to 8. Herbert, E. & Uh ler, M. Cel/30, 1-2 (1982).

442 NATU RE IVOL386 j3 APRIL 1997