COMMENTARY Neuropeptide receptor positive allosteric modulation in epilepsy: Galanin modulation revealed Daniel Hoyer1 Neuropsychiatry/Neuroscience Research, WSJ-386/745, Novartis Institutes for Biomedical Research Basel, CH 4002 Basel, Switzerland he therapeutic potential of neu- Galanin and Epilepsy neuropeptide receptor agonists are largely ropeptides has long been recog- There is ample evidence that galanin, like peptide analogs. Lu et al. describe an LMW T nized; some successes have been NPY or somatostatin, plays a role in epi- PAM that is reasonably potent at and se- registered, for example, in the lepsy, because when applied to the brain/ lective for a neuropeptide receptor. There tachykinin, endothelin, somatostatin, an- hippocampus these neuropeptides or are a few examples of allosteric modulators giotensin, calcitonin gene-related peptide their analogs have antiepileptic properties at corticotropin-releasing factor, CGRP, (CGRP), orexin, or glucagon-like peptide (2–4). Further, knockout or knockdown of GLP1, or ghrelin receptors, primarily 1 (GLP1) fields (1), primarily in peripheral their receptors result in increased sensi- antagonists (6). When it comes to agonists, applications. However, given the nature tivity to seizures, as reported for NPY2, almost no neuropeptide receptor PAMs of receptor–ligand interactions at the sst2/sst4 receptors (depending on species), are described. More generally, LMW, cen- orthosteric site of G protein-coupled re- and GalR1/GalR2. GalR1 and GalR2 are trally acting, bioavailable, and metaboli- ceptors (GPCRs), which seem to involve expressed in the hippocampus, whereas cally stable agonists acting at neuropeptide multiple pharmacophores that are hard to GalR3 is not. It is known that galanin, receptors are rare. L-692,429 is a ghrelin display on low-molecular-weight (LMW) receptor PAM, one of the very few peptide ligands, it is not surprising that the number Lu et al. describe an LMW receptor PAMs to be described (7). Addi- of such ligands, especially agonists, is very tionally, LMW ago-allosteric compounds limited, particularly considering brain- PAM that is reasonably acting on GLP1 receptors exist (8). Ghrelin penetrant ones. This is not because large and GLP1 are not commonly accepted as pharmaceutical companies have neglected potent at and selective neuropeptides, although this may change as modern high throughput screening tech- central effects are being investigated. Al- fi nology to nd such drug candidates. Many for a neuropeptide losteric modulation of GPCRs offers great synthetic neuropeptide analogs exist, but opportunities, especially for receptors for most of them are peptides, with inherent receptor. which the chemical space is limited (e.g., limitations in biovailability, plasma half- glutamate, GABAB, calcium sensing, and life/proteolytic degradation, brain pene- now neuropeptides); it is also expected that NPY, or somatostatin are released dur- tration, and cost of goods. More recently, orthosteric and allosteric sites are very dif- alternate approaches in the GPCR field ing epileptogenesis, and one may argue ferent (although ago-allosteric ligands emerged with negative or positive alloste- that such a release may contribute to may act on both), and it seems that allo- ric modulators (PAM), acting as antago- seizures; however, this is not so, because steric sites with a receptor subfamily are nists or agonists, but only in the presence mice constitutively overexpressing the rather divergent, a good basis for selectivity. of the endogenous agonist. Galanin, which neuropeptides or subjected to lentivirus- An allosteric modulator is expected to be acts on three GPCRs (GalR1–3), has long or recombinant adeno-associated virus- largely devoid of (side) effects (if selective been proposed to play a role in seizures; mediated overexpression are also less however, with few exceptions, the tools susceptible to seizures (2, 5). Lu et al. (2) enough) because it is inoperant in the ab- described are still peptides or peptidomi- now make the point that increased galanin sence of the endogenous ligand (6, 9). In metics. Now, Lu et al. (2) report in PNAS release plays a protective role, although not addition, it is not expected to induce de- on the discovery of CYM2503, an LMW, enough galanin is released to provide full sensitization, an issue with some neuro- potent, selective, and in vivo active GalR2 protection. However, the GalR2 PAM peptides. Negative allosteric modulators PAM, effective in several modalities (which per se has no activating effect), by (antagonists) are in clinical development linked to epilepsy in rodents. modulating endogenous peptide induces (10), and a few are on the market (9); it re- This is a major contribution to the a large degree of protection in different mains to be seen whether neuropeptide re- galanin and GPCR fields in that a PAM models of seizure and epileptogenesis. ceptor PAMs will be as successful, but the of GalR2 is described: CYM2503 does not Their article also suggests that targeting data presented here are very promising. Fi- interact with the orthosteric site of GlaR2 receptor may suffice, although pre- nally, whether a GalR2 PAM has peripheral GalR2 and is devoid of effect on GalR1 vious work also suggested a role for GalR1. effects is to be investigated. Interestingly, binding or function. CYM2503 is effective It remains to be seen whether a com- Fallarino et al. (11) report on the effects of in vivo in various models of mouse or bined effect at GalR1 and GalR2 may im- an MgluR4 PAM in animal models of mul- rat seizures and epileptogenesis and is as prove the outcome, although this may be tiple sclerosis [experimental autoimmune fi active as a reference antiepileptic, making dif cult to achieve; it is assumed that allo- encephalomyelitis (EAE)]; that article also at least two points. (i) GalR2 PAM steric sites among GPCR subfamily mem- concludes that endogenous glutamate is re- seems to be a valid principle in the treat- bers are highly different. In other words, to leased as a protective mechanism in rodent ment of seizures/epileptogenesis. (ii) The achieve a PAM for both GalR1 and GalR2 EAE models and that an mGluR4 PAM, just study validates the fact that galanin, seems rather unrealistic. which like other neuropeptides [somato- statin, neuropeptide Y (NPY)] is released Allosteric Modulators of Peptide Author contributions: D.H. wrote the paper. after seizure induction and can indeed Receptors The author declares no conflict of interest. be modulated to provide protection The present work highlights another major See companion article on page 15229. against seizures. advantage of the PAM approach, because 1E-mail: [email protected]. www.pnas.org/cgi/doi/10.1073/pnas.1010365107 PNAS | August 24, 2010 | vol. 107 | no. 34 | 14943–14944 Downloaded by guest on September 28, 2021 like the GalR2 PAM by Lu et al. (2), is able brain-penetrant galanin receptor agonist galanin peptide analogs (e.g., NAX5055) that to confer additional protection. (12–17). Numerous attempts in that direction have antiepileptic effects when applied sys- This represents the near-culmination of a were made: galmic and galnon are two LMW temically have since been described; how- long search. Tamas Bartfai received galanin nonpeptide GalR agonists; however, their ever, NAX5055 has a relatively short half-life fi from Viktor Mutt in late 1986 and since af nity for GalRs and a number of other in vivo and, being a peptide, will lack bio- then has published more than 140 papers on GPCRs is in the high micromolar range, thus galanin, galanin receptors, and their func- a more selective compound was needed. In- availability (18); furthermore, cost of goods is tional and pathophysiological roles, with terestingly, although galnon is unselective, it not to be neglected with peptides. Thus, emphasis on epilepsy, pain, and other CNS is a good tool for screening modulators or CYM2503 seems to be very close to the goal disorders, the ultimate aim being to discover antagonists at recombinant cell lines, set by Bartfai and his collaborators, and an LMW, systemically active, selective and galnon being more stable than galanin. A few many others, some time ago. 1. Hökfelt T, Bartfai T, Bloom F (2003) Neuropeptides: Op- retagogues act both as ghrelin receptor agonist and as 13. Fisone G, et al. (1989) N-terminal galanin-(1-16) frag- portunities for drug discovery. Lancet Neurol 2:463–472. positive or negative allosteric modulators of ghrelin ment is an agonist at the hippocampal galanin recep- 2. Lu X, et al. (2010) GalR2 positive allosteric modulator signaling. Mol Endocrinol 19:2400–2411. tor. Proc Natl Acad Sci USA 86:9588–9591. exhibits anticonvulsant effects in animal models. Proc 8. Knudsen LB, et al. (2007) Small-molecule agonists for 14. Bartfai T, et al. (1991) M-15: High-affinity chimeric pep- Natl Acad Sci USA 107:15229–15234. the glucagon-like peptide 1 receptor. Proc Natl Acad tide that blocks the neuronal actions of galanin in the 3. Vezzani A, et al. (1994) Enhanced neuropeptide Y re- Sci USA 104:937–942. hippocampus, locus coeruleus, and spinal cord. Proc lease in the hippocampus is associated with chronic 9. Conn PJ, Christopoulos A, Lindsley CW (2009) Alloste- Natl Acad Sci USA 88:10961–10965. seizure susceptibility in kainic acid treated rats. Brain ric modulators of GPCRs: A novel approach for the 15. Bartfai T, et al. (2004) Galmic, a nonpeptide galanin recep- Res 660:138–143. treatment of CNS disorders. Nat Rev Drug Discov 8: tor agonist, affects behaviors in seizure, pain, and forced- 4. Vezzani A, Hoyer D (1999) Brain somatostatin: A can- 41–54. swim tests. Proc Natl Acad Sci USA 101:10470–10475. didate inhibitory role in seizures and epileptogenesis. 10. Berry-Kravis E, et al. (2009) A pilot open label, single 16. Crawley JN, Robinson JK, Langel U, Bartfai T (1993) Eur J Neurosci 11:3767–3776. dose trial of fenobam in adults with fragile X syn- Galanin receptor antagonists M40 and C7 block galanin- 5.