Introduction
Methuselah (Mth) is a G protein–coupled receptor (GPCR) involved in aging. Drosophila melanogaster mth mutants have an extension of life span and increased resistance to stress1. The development of pharmacological modulators of the receptor would allow for manipulation of its function in order to gain information about the receptor's physiological role and activity. In this issue, Ja et al. use mRNA display selection to identify high-affinity peptide inhibitors for Mth2. The peptides identified are potent antagonists of the receptor and compete with a peptide mimicking the natural agonist ligand for binding to the receptor. When expressed in vivo in flies, the new Mth peptide antagonists induced an extension of life span.
The D. melanogaster mth gene encodes a seven-transmembrane protein with a large extracellular domain similar in size to that of the peptide hormone receptors belonging to the B-family of GPCRs, which are stimulated by moderately large peptide hormones. The mth gene was first discovered in a screen for mutations that increase the life span of D. melanogaster. Heterozygous mutant flies lived 
35% longer than the parental strain and were more resistant to stresses, including toxins (such as dietary paraquat) and starvation. Homozygous mutant flies showed pre-adult lethality, which suggests that the gene may also have an important developmental role1. D. melanogaster mth mutants were also found to have a 50% reduction in neurotransmitter release from presynaptic motor neurons at larval neuron-muscular junctions, and a decrease in both synaptic area and the density of docked and clustered vesicles3.
Two peptides, SunA and SunB, which are products of the stunted gene, had previously been identified as the natural ligands for Mth. sun mutant flies, similar to mth mutants, have an increased life span and increased resistance to stress4. The stunted gene encodes for the e subunit of F1F0 ATP synthase, though no difference in ATP levels was found between sun mutant flies and wild-type flies4. The function of the e subunit is unclear. It is, however, known that F1F0 ATP synthase is found on the plasma membrane, and in mitochondrial membranes5. Therefore it is possible that the peptide ligands reach the extracellular space when the e subunit is released from F1F0 ATP synthase on the plasma membrane. Or, alternatively, it could be released from F1F0 ATP synthase on the mitochondrial membrane and then secreted from the cell (Fig. 1).
Figure 1: The 
subunit of F1F0 ATP synthase, Sun, is an agonist of Methuselah.
![Figure 1 : The |[epsiv]| subunit of F1F0 ATP synthase, Sun, is an agonist of Methuselah.](/nchembio/journal/v3/n7/thumbs/nchembio0707-371-F1.jpg)
Sun (red circle) could get to the extracellular space by secretion from the mitochondrial ATP synthase, or it could be found on the cell surface. The peptide antagonists identified by Ja et al. compete with Sun for binding to Mth. Both flies expressing a peptide antagonist and sun mutant flies have enhanced longevity.
Full size image (24 KB)In the current report, the authors set out to identify high-affinity peptide ligands that bind to the N-terminal ectodomain of Mth based on the logic that other class B GPCR ectodomains maintain recognition of their cognate ligands independently of the transmembrane components of the receptor2. The crystal structure of the N-terminal ectodomain of Mth reveals a folded structure with three primarily 
-structure–containing domains meeting to form a shallow groove6. Using mRNA display selection they identified several peptide antagonists that had high affinity for the ectodomain of Mth in vitro. These peptides competed with the Sun peptide, reducing its binding by 80%, which suggests that the agonist and antagonists bind to Mth at the same site of the ectodomain. By electron density mapping, the binding site was determined to be near the C terminus of the ectodomain, which suggests that the peptides bind at an interface between the Mth ectodomain and possibly the extracellular loops.
Ja et al.2 followed up on this idea with the hypothesis that extracellular loop 2 of Mth, which contains a partial RWR motif (also contained within the peptide antagonists), may be the site of interaction with the ectodomain. However, a synthetic peptide mimicking this area did not compete with an antagonist peptide; thus, it remains unclear which extracellular loop of Mth is involved in ligand binding. This may be explained by the fact that purified ectodomains of some class B receptors only show low-affinity ligand binding. Thus, it is likely that the ectodomain itself is not fully accountable for ligand binding. Both the ectodomain and the first extracellular loop of another class B peptide hormone receptor, the glucagon receptor, are important for agonist binding and potency7. In the case of rhodopsin, a member of the class A family of GPCRs and the only GPCR for which a crystal structure (of the inactive form) has been solved, it is thought that extracellular loop 2 may contact the chromophore, and it has been suggested that this loop may regulate the stability of the active state of the receptor, in which the all-trans chromophore acts8. The crystal structures of the ectodomain of the group 2 subtype 3 metabotropic glutamate receptor (a class C GPCR) with agonists bound suggest that only the N-terminal portion of the ectodomain is involved in agonist binding9. It seems, therefore, that different classes of GPCRs may bind to and be activated by their ligands in different manners.
The current study provides insight into how Mth is activated by its ligand by identifying the binding area of an antagonist and the ectodomain of the receptor. The results suggest that both the ectodomain and one or more extracellular loops may be involved in GPCR activation. Technical difficulties in crystallization of membrane proteins present a large obstacle in the path toward understanding the mechanisms of activation of GPCRs. However, this study, in using just the ectodomain of Mth, provides both insight into its activation and a method that could be extended to other class B GPCRs while we wait on the elusive structures of more full-length and activated GPCRs.
As with mth and sun mutations, expression of the Mth peptide inhibitors in vivo produced flies with an extended life span. Mutations of some residues simultaneously abolished the binding to Mth and the effect on life span, which suggests that the peptides could function as antagonists in vivo. This study further confirms the role of Mth in aging, and it also confirms that the stunted gene product is indeed the ligand of Mth. These antagonists will prove useful for further examining the role of Mth in aging and synapse development. Given that the ligand (Sun) doesn't appear to have a role in ATP synthesis, it seems unlikely that Mth signaling exerts its aging effects through general control of energy production. A well-studied insulin-like growth factor signaling pathway is known to control aging in Caenorhabditis elegans, D. melanogaster and mice10, though no Mth homologs have been identified in mammals so far. It would be interesting to see whether Mth signaling is also involved in this pathway, or whether the Mth-mediated control of longevity is a separate pathway.
