The ectodomain of the relaxin GPCR receptor, RXFP1, comprises an N-terminal LDLa module, essential for activation (1), tethered to a leucine-rich repeat (LRR) domain by a 32-residue linker. Activation is proposed to proceed by relaxin binding with strong affinity to the ectodomain and then, through an unknown process, enable the LDLa module to bind and activate the transmembrane domain. The binding site of relaxin is thought to be the LRR domain, but we have found mutations within a conserved region of the linker immediately C-terminal to the LDLa module (GDNNGW, residues 41-46) significantly weaken relaxin affinity, suggesting an additional binding site. Using NMR spectroscopy and titrations of 15N-labelled LDLa-linker with relaxin or sub-stoichiometric concentrations of a paramagnetic (Mn2+) labelled relaxin we have elucidated a discrete relaxin-binding site (residues 46-63) on the linker (2). Additional NMR experiments show residues 49-52 of the linker that are most sensitive to relaxin have a weak propensity for helix which on relaxin titration stabilizes. Mutations within GDNNGW show loss of relaxin-binding and an increase in structural disorder around this helical region of the linker, suggesting an indirect role in binding relaxin. Cell-based signaling and binding assays of receptor mutants confirm that the linker binds relaxin and is essential for receptor activation. Further, we engineered a soluble protein scaffold (3) to include the transmembrane domain exoloop-2 and show by NMR titrations that the LDLa-linker interacts specifically with this loop. We therefore propose for binding of relaxin, two sites are required: one on the LRR domain, the other on the linker. Mechanistically, the LDLa-linker may weakly interact with the transmembrane exoloops; this is intensified by relaxin binding which also stabilizes a helical conformation within the linker. Stabilization of the linker helix positions or orients the LDLa module to interact and activate the transmembrane domain of RXFP1.