Mechanism of activation of the relaxin family peptide receptors RXFP1 and RXFP2
AuthorBruell, Shoni Denise Rega
AffiliationBiochemistry and Molecular Biology
Document TypePhD thesis
Access StatusThis item is embargoed and will be available on 2021-12-18. This item is currently available to University of Melbourne staff and students only, login required.
© 2019 Shoni Denise Rega Bruell
The relaxin family peptide receptors RXFP1 and RXFP2 are the cognate receptors for the peptide hormones relaxin and INSL3 respectively, best known for their roles in reproductive biology. Being GPCRs, these receptors are members of the largest family of membrane-bound receptors known in the human genome, but they are unique within this family due to the existence of a single low-density lipoprotein type A (LDLa) module on the N-terminus of their large ectodomain. The LDLa module is imperative to normal receptor signalling and is hypothesised to be a tethered ligand, interacting with the receptor transmembrane domain (TMD) to bring about an active conformation. This module is connected to the leucine-rich repeats that make up the remainder of their extracellular domain by a stretch of amino acids 32 long in RXFP1 and 25 in RXFP2. These linking residues have been termed the linker, and a series of accidental and intentional discoveries led to the notion that the linker plays an important role within the activation mechanism of both RXFP1 and RXFP2 in response to their peptide ligands. The work presented within this thesis delves into the details of this activity, exploring the various regions of the linker, as well as the LDLa modules, with the use of mutagenesis and the construction of chimeras of the two receptors. The mutants and chimeras were systematically tested in stably or transiently transfected HEK293T cells in a series of ligand binding, activity (focussing on their ability to prompt an accumulation of intracellular cAMP) and cell surface/total expression assays. By comparing the behaviour of the mutant and chimeric receptors to that of their wild-type counterparts we have been able to paint a detailed picture of the binding and activation mechanisms of RXFP1 and RXFP2 in response to either active peptide ligand, adding our findings to a growing understanding of their activity quickly emerging from the lab and the field at large. Complementing the data generated from cell-based assays are nuclear magnetic resonance (NMR) studies performed concurrently using recombinantly expressed and purified RXFP1 and RXFP2 LDLa-linker proteins in titrations with relaxin as well as the receptor extracellular loops, as expressed on a soluble scaffold based on thermostabilized protein GB1. While the NMR experiments were carried out by a collaborating student, the RXFP2 LDLa-linker and soluble scaffold proteins were designed and characterized for use in NMR during this PhD project, and the details are outlined herein. In Chapter 2 the LDLa modules of the two receptors were swapped, such that RXFP1 contained the LDLa module from RXFP2 and vice versa. We found that while ligand-induced activity was weakened in the chimeric receptors, they were able to produce a robust signal, and for RXFP2 (but not RXFP1) the signal was slightly closer to wild-type levels upon subsequent swapping of the TMDs, such that they would match the non-native LDLa module . The result contradicted previous findings in which RXFP1 with the LDLa module from RXFP2 was shown to be incapable of signalling upon relaxin stimulation. We rationalized this discrepancy as being due to the differing design in the cloning of the chimeric constructs. While our versions swapped only the LDLa modules themselves, the previous versions had also swapped a large portion of the neighbouring linker residues. This alerted us to a possible function being carried out by the linker and guided our future work. We proceeded to mutate residues of the linker to alanine, carrying out an extensive scan of the RXFP1 linker that is presented in Chapter 3. We discovered that the initial residues formed something of a motif with the sequence GDNNGW, and when mutated – the residues Asp2, Gly5 and Trp6 in particular – the receptors lost their ability to stimulate cAMP production and bind relaxin effectively. This information along with NMR data led to the conclusion that this motif made up an activation region that was involved in the tethered ligand activation mechanism of RXFP1. Of note, we observed that the fold-difference in affinity for relaxin exhibited by the mutants of the activation domain was not commensurate with the enormous weakening in potency in relaxin signalling assays shown by the same mutants. This contrasted with mutants of the central part of the linker, in particular Phe54 and Tyr58, which when mutated individually or together displayed similar fold-differences from wild-type in relaxin potency and affinity. Coupled with compelling NMR data we concluded from this evidence that the central portion of the linker constituted a relaxin binding site that had hitherto never been described . We pursued a similar investigation focussing on the linker of RXFP2 in Chapter 4. The supposed activation region is largely conserved in the similar receptor, with the initial sequence of the linker GDTSGW. Indeed, mutation to alanine of these residues and the Phe found three residues along showed that both relaxin and INSL3 activation was dependent on equivalent residues to those identified in the RXFP1 mutagenesis campaign. Similarly, relaxin binding was severely compromised in the mutant receptors, while INSL3 binding was not. This result mirrored prior knowledge coming from mutagenesis and A-chain truncations of the relaxin and INSL3 peptides and highlighted a differing mode of action on RXFP2 in response to the two similar agonists. We postulated that the activation region of RXFP2 consisted of the linker residues from the GDxxGW motif found in both receptors, but they were assisted or accompanied by the actions of the N-terminal residues of the INSL3 A-chain . We further investigated the relaxin binding site from the RXFP1 linker by creating another two chimeras, in which the implicated residues were inserted into an equivalent position in the RXFP2 linker. The linker of RXFP2 is shorter than that of RXFP1 and hence a relaxin binding site had not been identified at a similar position, but by including the residues we supposed to be contributing to this binding site we were able to increase the potency and affinity for relaxin of the chimera to more closely resemble the behaviour of RXFP1 with its native ligand. In addition, the dissociation kinetics of the interaction, measured using a NanoBRET assay, resembled more closely the case of RXFP1 than RXFP2. This information helped us to confirm the existence of the relaxin binding site and highlight major differences between the two receptors with a focus on the linker . The work presented in this thesis gives a deep and detailed look at an integral part of the binding and activation mechanisms of RXFP1 and RXFP2 in response to the peptide hormones relaxin and INSL3, paving the way for their use in a therapeutic setting. The resolution of the role that the LDLa module plays has altered the prevailing view about receptor activation in a number of aspects. Firstly, the complex binding mode of both relaxin and INSL3 has been defined more thoroughly, and secondly, we now know to focus on the linker when defining TM interactions. It therefore highlights the utility of using a combination of approaches – cell-based assays partnered with mutagenesis and chimeras alongside protein expression and NMR – to reach valid conclusions about molecular systems.
KeywordsGPCR; Relaxin; INSL3; Mutagenesis; Molecular and cell biology; Structural biology; Relaxin family peptide receptor (RXFP)
- Click on "Export Reference in RIS Format" and choose "open with... Endnote".
- Click on "Export Reference in RIS Format". Login to Refworks, go to References => Import References