Content uploaded by Martina Kocan
Author content
All content in this area was uploaded by Martina Kocan on Oct 14, 2015
Content may be subject to copyright.
ORIGINAL PAPER
Orthosteric, Allosteric and Biased Signalling at the Relaxin-3
Receptor RXFP3
Martina Kocan
1
•
Sheng Yu Ang
1
•
Roger J. Sum mers
1
Received: 9 July 2015 / Revised: 12 August 2015 / Accepted: 13 August 2015
Ó Springer Science+Business Media New York 2015
Abstract Relaxin-3 is a neuropeptide that has roles in
stress, memory and appetite regulation. The peptide acts on
its cognate receptor RXFP3 to induce coupling to inhibi-
tory G proteins to inhibit adenylyl cyclase and activate
MAP-kinases such as ERK1/2, p38MAPK and JNK. Other
relaxin family peptides can activate the receptor to produce
alternative patterns of signalling and there is an allosteric
modulator 135PAM1 that displays probe-selectivity. There
are now a variety of selective peptide agonists and antag-
onists that will assist in the determination of the physio-
logical roles of the relaxin-RXFP3 system and its potential
as a drug target.
Keywords Relaxin-3 RXFP3 Biased signalling
Allosteric modulator
Relaxin-3 is the most recently described relaxin family
peptide, categorised by the presence of the characteristic
RxxxRxxI/V relaxin-binding motif in the B-chain. Unlike
other relaxins, the relaxin-3 sequence is well-conserved
across species [1–3] and is believed to be the ancestral
peptide of the family [1]. In mammals it is primarily a
neuropeptide [4] that is involved in stress, memory and
appetite regulation [5–14].
The Relaxin-3 Receptor-RXFP3
The receptor for relaxin-3, RXFP3, is encoded by an
intronless gene and is a family A, G protein coupled
receptor (GPCR) [15, 16]. Unlike RXFP1 and RXFP2,
RXFP3 has a very small N-terminal domain and a distinct
signalling profile. RXFP3 was originally termed SALPR
or somatostatin and angiotensin-like receptor [17], for its
resemblance to class A small peptide GPCRs before
recognition that its cognate ligand was relaxin-3 [16]. The
receptor and ligand expression profile suggests that
relaxin-3 is a neuropeptide [4, 18]. In rat brain, relaxin-3
is expressed primarily in the nucleus incertus, in the
pontine raphe nucleus, the anterior, lateral, and ventro-
lateral periaqueductal gray and in an area dorsal to the
lateral substantia nigra [7]. High levels of RXFP3 mRNA
are found in the olfactory bulb, paraventricular and
supraoptic nuclei, preoptic and posterior areas of the
hypothalamus, hippocampus, septum, and amygdala with
lower levels in cortex, peraqueductal grey, nucleus
incertus, and areas of brain stem [15]. No peripheral tis-
sues express high levels of RXFP3. In contrast to RXFP1
and RXFP2 that require the complete two-chain relaxin
family peptide for activation, the relaxin-3 B-chain alone
can bind to and activate RXFP3, although with lower
potency than the two-chain peptide [19, 20]. Thus binding
and activation of RXFP3 differs from that observed for
relaxin/RXFP1.
Structural Features of RXFP3: N-terminal
and Extracellular Loop Regions
Comparative sequence analysis of mammalian RXFP3
sequences, molecular modelling and mutagenesis were
used to map the relaxin-3 binding site [21] and
Special Issue: In Honor of Philip Beart.
& Roger J. Summers
Roger.Summers@monash.edu
1
Drug Discovery Biology, Monash Institute of Pharmaceutical
Sciences, Monash University, 399 Royal Parade, Parkville,
Melbourne, VIC 3052, Australia
123
Neurochem Res
DOI 10.1007/s11064-015-1701-3
demonstrated that removal of the first 33 amino acids of the
RXFP3 N-terminus had little effect. This concentrated
attention on the glutamic and aspartic acid residues in the
extracellular loops as potential partners for arginine resi-
dues in relaxin-3. E141 and D145 in ECL1 and E244 in
ECL2 were identified as essential for relaxin-3 binding
allowing the docking of the NMR solution structure of
relaxin-3 [22] into a homology model of RXFP3 based on
the CXCR4 peptide antagonist crystal structure [23]. R12
and R16, on the B-chain helical segment of relaxin-3, are
believed to interact with E244 and D145 on RXFP3
whereas R26 forms a salt bridge with E141 [21]. This
model allows for the relaxin-3 C-terminus to insert into a
‘‘GPCR binding pocket’’ within the TM domains. Thus
relaxin-3 residues in the core b-helix are involved in
RXFP3 binding and the C-terminal RW residues drive
activation.
Structural Features of RXFP3: Transmembrane
Spanning Regions
Chimeras of RXFP3 and the closely related RXFP4 have
been used to investigate the role of the transmembrane
(TM) spanning regions in responses to relaxin-3 and the
RXFP4 cognate ligand insulin-like peptide 5 (INSL5)
[24]. At the wild-type RXFP3 receptor, relaxin-3 (pIC
50
9.3) was *200 fold more potent than INSL5 (pIC
50
7.0)
at competing for
125
I R3/I5 binding whereas at the
RXFP4 receptor both ligands displayed similar potency
(pIC
50
’s 8.8 and 8.7) [25]. A chimera of RXFP3 con-
taining TM3 and TM5 of RXFP4 displayed increased
binding affinity for INSL5 but did not activate GTPcS
binding demonstrating that these regions are not sufficient
for INSL5 activation of RXFP3. A similar binding affinity
and GTPcS binding for both relaxin-3 and INSL5 was
achieved in a chimera of RXFP3 containing TM2, TM3,
TM5 and ECL2 of RXFP4 compared with wild-type
RXFP4, suggesting that all of these regions influenced
binding and functional responses to INSL5 but not
relaxin-3 [25]. Interestingly none of the chimeras in
which RXFP3 regions were replaced by corresponding
regions from RXFP4 or those where RXFP4 regions were
replaced by corresponding regions from RXFP3, dis-
played major changes in potency for relaxin-3 either in
binding or functional assays. Thus relaxin-3 is able to
bind to and activate with similar potency a variety of
chimeras showing a wide degree of structural diversity,
whereas the requirements for INSL5 were much more
specific [25]. The mode of activation of RXFP3 by
relaxin-3 is also unlike the interaction between relaxin
family peptides and RXFP1 and RXFP2, involving pri-
marily an interaction with ECL2.
Ligands that Act at RXFP3
Orthosteric Ligands: Relaxin-3 and Analogues
Human relaxin-3 has a tertiary structure that resembles
insulin and other relaxin family peptides [26]. Replacement
of the relaxin-3 A-chain with the corresponding region
from INSL5 (R3/I5) does not influence RXFP3 binding or
activation, but increases specificity by reducing binding to
and activation of RXFP1 [27]. A similar outcome can be
achieved by truncation of the relaxin-3 A-chain [20]orby
deletion of the A-chain disulphide bond [28]. The analogue
with a B23-27 deletion and arginine addition at B23 is a
high-affinity, RXFP3-selective, competitive antagonist
(analogue 3).
Important residues that are essential for relaxin-3 bind-
ing to RXFP3 are R8, R12, R16, I5 and F20 [19]. The
B-chain residues R26 and W27 are also required for acti-
vation [19]. Removal of B-chain C-terminal residues to
C22 and addition of an arginine residue at the N-terminus,
when combined with the A-chain of INSL5, produces a
high affinity RXFP3-selective antagonist R3(BD23-27)R/
I5 [19] (see below). Modification of the B-chain of
R3(BD23-27)R/I5 with the cysteine residues mutated to
serine [H3(B1-22R)] produces a single chain antagonist
[29].
Biased Agonists Acting at RXFP3
Ligand-directed signalling bias is an increasingly recog-
nized feature of GPCRs that is characterized by stabilisa-
tion of particular receptor confirmations by ligands to cause
selective activation of downstream signal transduction
pathways [30–32]. Although it was suggested [16] that
only relaxin-3 or its B-chain could bind to and activate
RXFP3, more recent studies have shown that several
relaxin peptides interact with RXFP3 to activate distinct
signalling profiles through different, although sometimes
overlapping pathways (Fig. 1). In binding and adenylyl
cyclase inhibition assays, relaxin-3 appeared to be selective
for RXFP3, with no receptor activation by human relaxin
(relaxin) or insulin-like peptide 3 (INSL3) [16]. However,
activation by other relaxin peptides was not examined over
a wider range of signal transduction pathways, and the
sensitivity of inhibitory cAMP assays may also be influ-
enced by the degree of activation of adenylyl cyclase by
forskolin and by the time of stimulation.
Allosteric Modulators
3-[3,5-Bis(trifluoromethyl)phenyl]-1-(3,4-dichlorobenzyl)-
1-[2-(5-methoxy-1H-indol-3-yl)ethyl]urea (135PAM1) is a
Neurochem Res
123
positive allosteric modulator (PAM) that only displays
activity with C-terminal amidated relaxin-3 or R3/I5 (probe
selectivity
1
)[33] (Fig. 1). Binding studies with HEK293-
RXFP3 membranes demonstrated that 135PAM1 (B1 lM)
does not directly compete for
125
I R3/I5(amide) binding,
but at higher concentrations behaves as a PAM. R3/
I5(amide) competes for
125
I R3/I5(amide) binding in a
conventional manner [33]. In HEK293-RXFP3 cells co-
expressing the promiscuous G protein Gaq
i5
, 135PAM1
shifted Ca
2?
concentration–response relationships in a
limiting manner (i.e. the concentration–response curves
shift only so far with increasing 135PAM1 concentrations)
to relaxin-3(amide) or R3/I5(amide) but not the free-acid
(OH) peptides. Similar results were obtained for inhibition
Fig. 1 Signal transduction pathways activated by RXFP3 in response
to orthosteric, biased and allosteric agonists. Orthosteric ligands such
as the cognate ligand relaxin-3 interact with RXFP3 to cause coupling
of the receptor with Ga
oB
,Ga
i2
,Ga
oA
and Ga
i3
proteins to inhibit
adenylyl cyclase. Relaxin-3 also induces phosphorylation of MAP
kinases, activates reporter gene transcription and RXFP3/b-arrestin
interactions. Biased agonists such as relaxin activate RXFP3 to
promote coupling to Ga
oB
and Ga
i2
but not Ga
oA
,Ga
i3
or b-arrestins.
Relaxin inhibits forskolin-stimulated cAMP accumulation and acti-
vates AP-1 but not NF-jB. Allosteric modulation is observed at
RXFP3 in the presence of the positive allosteric modulator 135PAM1
that sensitises responses to relaxin-3 amide but not the endogenously
produced peptide. Transactivation of the EGFR is observed following
activation of RXFP3 in tissues where they coexist
1
Probe selectivity—the allosteric modulation by 135PAM1 only
occurs with amidated relaxin analogues and not with the native
peptides.
Neurochem Res
123
of CRE reporter gene responses. Although this is the only
published example to date of allosteric activity at RXFP3,
it does identify an allosteric site on RXFP3 that can be
modulated by small molecules.
RXFP3 Antagonists
R3(BD23-27)R/I5 has been useful for defining the physi-
ological functions of RXFP3. Generically, chimeric pep-
tides with the INSL5 A-chain and relaxin-3 B-chain lack
activity at RXFP1 and become antagonists when the
relaxin-3 B-chain is truncated [19, 34]. During recombi-
nant production of R3(BD23-27)/I5 an extra arginine
remained at the N-terminus due to incomplete peptidase
‘‘trimming of the RR from the RRRR furin cleavage
domain’’, and the modified R3(BD23-27)R/I5 proved to be
a potent antagonist at RXFP3 in both rat and human in vitro
systems. The analogue that lacks R23 is a weak agonist
with lower affinity at RXFP3 [34]. In rats, administration
of R3(BD23-27)R/I5 i.c.v. blocks increased food intake in
response to the RXFP3 agonist R3/I5. Recently it has been
shown that R3(BD23-27)R/I5 has a complex antagonist
profile and blocks some but not all pathways activated by
RXFP3. It also has weak biased agonist properties (see
below).
The identification of the important role of R23 in
R3(BD23-27)R/I5 suggested that these properties might be
retained in a B-chain only analogue. The resulting H3(B1-
22R) is a high affinity antagonist of RXFP3 that is simpler
to produce than the two chain peptides [29]. The peptide
blocks increases in feeding produced by i.c.v. injection of
R3/I5 in rats [29] and has identified a role for the
endogenous relaxin-3 system in addiction [12] as well as in
motivated food seeking and consumption in mice [13].
The studies of structure–activity relationships for the
relaxin peptides show that unlike RXFP1 and RXFP2,
RXFP3 can be activated by peptides comprising only the
B-chain. In addition agonists and antagonists have been
developed with improved selectivity profiles. Both relaxin
and some RXFP3 antagonists display biased agonist
properties at RXFP3 and although the physiological
importance of this is unclear, biased agonists could repre-
sent an approach to novel therapies. The utility of allosteric
modulation in vivo is currently limited by probe selectivity.
Relaxin Family Peptide Receptor 3 (RXFP3)
Signalling
Canonical Signalling Pathways
RXFP3 is coupled to inhibitory Ga
i/o
proteins since stim-
ulation by relaxin-3 causes PTX-sensitive inhibition of
forskolin-stimulated cAMP accumulation [16]. This sig-
nalling pattern is also associated with PI3K- and PKC-
dependent phosphorylation of ERK1/2 and other MAPKs
[35, 36] (Fig. 1). Both relaxin-3 or its B-chain inhibit
adenylyl cyclase activity [16], but more recent studies
suggest that relaxin also interacts with RXFP3 to activate a
unique signalling pattern (see below) [35].
Like many other Ga
i/o
-coupled GPCRs, activation of
RXFP3 results in a rapid and transient (peak response
2-5 min) increase in ERK1/2 phosphorylation in CHO-K1
and HEK293 cells stably expressing RXFP3 [35, 36]
(Fig. 1). ERK1/2 activation also occurs with relaxin-3
B-chain dimer and relaxin albeit with lower potency and
efficacy [36] but not with porcine relaxin or INSL3 [35]
(Fig. 1). PTX pre-treatment inhibits relaxin-3 stimulated
ERK1/2 phosphorylation in CHO-RXFP3 cells by *90 %,
and completely abolishes the response in HEK-RXFP3 or
SN56 cells [36]. At least two pathways downstream of Ga
i/
o
contribute to RXFP3-mediated ERK1/2 phosphorylation
in CHO-K1, HEK293 and SN56 cells, since the MAPK
response is blocked *50 % by the PI3K inhibitors
LY294002 or Wortmannin, and the remainder by general
and isoform-selective PKC inhibitors [36] (Fig. 2). Since
ERK1/2 phosphorylation appears to be involved in the
central feeding responses in rats [37–39], this may be
physiologically relevant.
Reporter gene assays have also been used to identify
signal transduction mechanisms activated by RXFP3. Since
several MAPK signalling pathways (p38 MAPK, [40];
JNK, [41]; and ERK1/2, [42, 43] converge on activator
protein 1 (AP-1) elements to increase gene transcription,
the use of AP-1-linked reporter gene assays with selective
inhibitors can provide useful information on signalling in
different cellular backgrounds. Inhibition of Ga
i/o
proteins
by PTX blocks RXFP3 mediated AP-1 reporter activation
in SN56 cells but not in CHO-RXFP3 and HEK-RXFP3
cells [35] suggesting that AP-1 reporter gene activation
was downstream of Ga
i/o
in the mouse-derived cell line but
not in the other two cell types [35]. AP-1 reporter gene
activation in CHO-RXFP3 and SN56 cells in response to
relaxin-3 is completely blocked by the p38 MAPK inhi-
bitor (RWJ67657), whereas MEK (PD98059) or JNK
inhibition (SP600125) only partially blocks the response.
In HEK-RXFP3 cells, JNK inhibition completely blocks
relaxin-3-stimulated AP-1 reporter activation, whereas p38
MAPK or MEK inhibition was only partially effective [35].
This suggests that the hierarchy of the different MAPK
signalling pathways varies with the cell background
(Fig. 2). In vivo studies show that MAPK signalling is
activated in forced swim tests in rats, with marked
increases in pMEK1/2, pERK1/2 and pJNK1/2/3 [38]. This
is associated with an increase in relaxin-3 mRNA in the
nucleus incertus (NI) [5], but more direct links between
Neurochem Res
123
relaxin-3, RXFP3, MAPK phosphorylation and stress
responses remain to be demonstrated.
In the NFjB reporter gene assay, in CHO-K1 and
HEK293 cells transiently expressing RXFP3, and in SN56
cells, stimulation of RXFP3 by relaxin-3 increased acti-
vation [35]. This response was blocked by PTX pre-treat-
ment [35] suggesting that it occurs entirely downstream of
Ga
i/o
(Fig. 2). However, the physiological significance of
this pathway remains to be determined.
Ligand-Directed Signalling Bias at RXFP3
Ligand-directed signalling bias has been described for
relaxin [44] and for the RXFP3 antagonist R3(BD23-27)R/
I5 [45]. The first study that showed biased signalling pro-
duced by relaxin at RXFP3 came from metabolic responses
recorded by microphysiometry [44] where relaxin pro-
duced a small change in the extracellular acidification rate
in CHO-RXFP3 cells. Later studies in CHO-K1, HEK293
and SN56 cells, showed that relaxin, porcine relaxin and
INSL3 all weakly inhibited forskolin-stimulated cAMP
accumulation [35]. The ability of INSL3 to activate RXFP3
was species-specific for the human but not the mouse
receptor [35]. Although previous studies reported no
inhibition of cAMP accumulation by either porcine relaxin
or INSL3 [16] the sensitivity of inhibitory cAMP assays is
highly dependent on both the degree of activation of
adenylyl cyclase by forskolin and the time of stimulation
and the differences observed are probably due to different
experimental paradigms.
Relaxin and porcine relaxin also caused AP-1 reporter
gene activation [35] in CHO-K1 and HEK293 cells
expressing RXFP3 and in SN56 cells with an order of
potency relaxin [relaxin-3 [porcine relaxin. Some of the
AP-1 activation appeared to be independent of Ga
i/o
,as
PTX failed to completely block porcine relaxin-stimulated
activation in CHO-RXFP3 cells. Thus, both relaxin-3 and
porcine relaxin can activate AP-1 by a Ga
i/o
-independent
mechanism, suggesting ligand-directed signalling bias [35].
However, in SN56 cells, all of the AP-1 reporter gene
responses observed after stimulation of RXFP3 were
blocked by PTX, suggesting differences in the pathways
involved in mediating the downstream response [35].
In CHO-K1 cells expressing RXFP3, AP-1 responses to
relaxin are strongly inhibited by the p38 MAPK inhibitor
(RWJ67657) or the JNK inhibitor (SP600125), but only
weakly by the MEK inhibitor (PD98059), suggesting that
p38 MAPK and JNK are the major MAPKs involved
Fig. 2 Summary of the effects of pharmacological inhibitors on signal transduction pathways activated by RXFP3 in response to orthosteric or
biased agonists. Pharmacological inhibitors are shown in red and affected pathways in grey (Color figure online)
Neurochem Res
123
(Fig. 2). However, in HEK293 cells expressing RXFP3,
AP-1 responses to relaxin were decreased by p38 MAPK or
MEK inhibition but not by JNK inhibition, suggesting that
p38 MAPK and ERK predominated in these cells. In SN56
cells, p38 MAPK, JNK or MEK inhibition were equally
effective in blocking the response to relaxin, suggesting
that all three kinases were equally important [35]. Thus
species and cell type will affect the signalling pattern fol-
lowing RXFP3 activation. Specific assays of pERK1/2, p38
MAPK and pJNK activity in response to relaxin family
peptides has confirmed these findings [45] although the
Ga
i/o
-independent pathway still remains to be identified for
relaxin-3 and porcine relaxin. In CHO-RXFP3 cells,
relaxin activates p38 MAPK and ERK1/2 with lower effi-
cacy than relaxin-3 but the two peptides have similar
efficacy for JNK1/2/3 phosphorylation. All of the MAPK
responses to relaxin and relaxin-3 involved PTX-sensitive
G proteins [35, 36, 45] (Fig. 2).
Signalling bias was also seen with the RXFP3 antago-
nist, R3(BD23-27)R/I5, which blocked relaxin-3 AP-1
reporter gene activation but not relaxin AP-1 activation or
relaxin-3 NFjB activation [45]. R3(BD23-27)R/I5 also
activated the serum response element (SRE) reporter but
did not block either relaxin or relaxin-3 SRE activation.
While R3(BD23-27)R/I5 blocked relaxin-3-stimulated
p38MAPK and ERK1/2 phosphorylation, it was a weak
partial agonist for these signalling pathways. p38MAPK
activation in response to R3(BD23-27)R/I5 was G protein-
independent. Bioluminescence resonance energy transfer
(BRET) technology was used to determine interactions
between RXFP3 and G proteins and confirmed that relaxin-
3-activated RXFP3 interacts with Ga
i2
,Ga
i3
,Ga
oA
and
Ga
oB
whereas relaxin or R3(BD23-27)R/I5 can only cause
interactions with Ga
i2
or Ga
oB
(Fig. 3). Relaxin-3, but not
R3(BD23-27)R/I5 or relaxin, promoted RXFP3/b-arrestin
interactions and these were blocked by R3(BD23-27)R/I5
[45]. This represents strong evidence for ligand directed
signalling bias at RXFP3. While signalling bias in response
to relaxin in particular is easily observed in recombinant
systems it is not known if this is physiologically important
or if it can be observed in vivo. Certainly relaxin does not
appear to be expressed in regions that also express RXFP3,
although receptors located in the brainstem region may be
exposed to circulating relaxin. The physiological signifi-
cance or the viability of biased signalling as a potential
target for pharmacological intervention remains to be
established.
Allosteric Modulation of RXFP3
Since the RXFP3 allosteric modulator 135PAM1 (Fig. 1)
has only been described in receptor binding studies, and in
recombinant systems expressing the promiscuous chimeric
G protein Gaq
I5
or in CRE reporter gene assays [33], it is
not known what pattern of signalling is observed in sys-
tems that naturally express RXFP3. In addition, since
135PAM1 has poor solubility and displays probe selec-
tivity with the C-terminal amides that are not naturally
occurring, it has limited use experimentally [33].
Transactivation of Epidermal Growth Factor
Receptors
Epidermal growth factor receptors (EGFR) are expressed
endogenously in some cells [46] and can influence RXFP3
signalling [36] (Fig. 1). In RXFP3 expressing HEK cells that
endogenously express EGFR, but not in CHO cells where
they are absent, the EGFR inhibitor tyrphostin (AG1478)
inhibited relaxin-3 ERK1/2 activation by *40 % suggesting
that EGFR are transactivated by RXFP3 [36] (Fig. 2). The
inhibitor of platelet-derived growth factor (PDGF) receptor
autophosphorylation AG370, had no effect on ERK1/2
responses to relaxin-3 in either HEK or CHO cells expressing
RXFP3 suggesting that transactivation of the PDGF receptor
was not involved [36]. Both AG1478 and AG370 on their
own have no effect on ERK1/2 signalling (van der Wes-
thuizen unpublished). In addition, in SN56 cells, that natu-
rally express RXFP3, tyrphostin AG1478 inhibited *62 %
of the ERK1/2 response [36] suggesting that transactivation
of tyrosine kinase receptors could play an important role in
responses to relaxin-3 in some systems. It is possible that in
some cells, transactivation of EGFR plays a role in ERK1/2
responses and hence in RXFP3 mediated responses in the
CNS, however the physiological significance of transacti-
vation of EGFR by RXFP3 is still unclear.
RXFP3 Coupling to Signalling Proteins
G proteins Coupling to RXFP3 When Stimulated
by Human Relaxin-3
The inhibition of forskolin-stimulated cAMP accumulation
in CHO-K1 or HEK293 cells stably expressing RXFP3 or
in mouse SN56 cells [16, 36] by relaxin-3 is completely
prevented by PTX pre-treatment (Fig. 2) which also
inhibited extracellular acidification rate [44] in the
cytosensor microphysiometer and ERK1/2 phosphorylation
responses to relaxin-3 [47]. Although these studies suggest
that RXFP3 is a Ga
i/o
coupled receptor, identification of
the specific G proteins involved was also performed in
RXFP3 expressing cells transiently transfected with PTX
insensitive (C351I mutation) variants of proteins, that were
subsequently treated with PTX (to remove the influence of
endogenous Ga
i/o
proteins) and then examined for recovery
of function. Ga
i2
was the major G protein involved in the
Neurochem Res
123
inhibition of cAMP accumulation in CHO, whereas in
HEK293 cells, Ga
i3
,Ga
oB
and Ga
oA
were all involved
[47]. More recent BRET studies confirm ligand-induced
interactions between RXFP3-Rluc8 and G proteins (Gc2-
Venus) in CHO cells where treatment with relaxin-3
caused activation of Ga
i2
,Ga
oA
and Ga
oB
, but also
revealed an interaction between RXFP3 and Ga
i3
[45]
(Fig. 3). Thus RXFP3 has the potential to couple to Ga
i2
,
Ga
i3
,Ga
oA
and Ga
oB
following activation by relaxin-3, but
the receptor preferentially couples to discrete subsets of
these G proteins dependent on cell type.
G protein Coupling to RXFP3 Following
Stimulation with the Biased Ligands Relaxin
and R3(BD23-27)R/I5
Not only did the biased ligands relaxin and R3(BD23-27)R/
I5 produce different signalling patterns but both showed a
distinct pattern of G protein coupling when examined by
BRET [45]. Whereas relaxin-3 promoted coupling to Ga
oA
,
Ga
oB
,Ga
i2
and Ga
i3
the biased ligands could only promote
weak coupling of RXFP3 to Ga
i2
or Ga
oB
[45] (Fig. 3).
Interactions Between RXFP3 and b-Arrestins
Relaxin-3 but not relaxin or the RXFP3 antagonist
R3(BD23-27)R/I5 promoted interactions between RXFP3
and b-arrestins, as measured by real-time kinetic BRET
between RXFP3-Rluc8 and a b-arrestin fusion protein (b-
arrestin 1-Venus or b-arrestin 2-Venus) (Fig. 4). R3(BD23-
27)R/I5 blocked relaxin-3 stimulated recruitment of b-ar-
restin 1 and b-arrestin 2 to RXFP3 [45]. However, treat-
ment with PTX to block Ga
i/o
proteins only partially
blocked b-arrestin recruitment, indicating that RXFP3
recruits b-arrestins both by G-protein-dependent and -in-
dependent pathways (Fig. 4). To determine whether b-
Fig. 3 Detection of RXFP3-G
protein interactions following
treatment with the cognate
agonist relaxin-3 and the biased
ligands relaxin or
R3(BD23–27)R/I5.
Examination by BRET of
interactions between RXFP3-
Rluc8, Gc2-Venus, Gb1 and
one of the Ga subunits (Ga
i2
,
Ga
i3
,Ga
oA
,Ga
oB
,Gas, Gaq)
co-transfected into Flp-In CHO
cells showed that relaxin-3
(1 lM) induced interactions
between RXFP3 and Ga
i2
,Ga
i3
,
Ga
oA
or Ga
oB
(a–d) whereas
relaxin (1 lM) or
R3(BD23–27)R/I5 (1 lM)
induced interactions only
between RXFP3 and Ga
i2
or
Ga
oB
proteins and produced a
smaller signal compared with
relaxin-3 (a, d). BRET
interaction only occurs when
receptor and G-protein subunits
are within 10 nm (f)
Neurochem Res
123
arrestin interaction was required for ERK1/2 activation
downstream of RXFP3, cells were transfected with con-
structs expressing dominant negative (V53D) or wild type
(WT) b-arrestin 1. There was no inhibitory effect of the
dominant negative b-arrestin 1 on ERK1/2 activation
suggesting that, unlike several other GPCRs [48, 49], the
RXFP3/b-arrestin 1 interaction contributes little to ERK1/2
signalling [45].
RXFP3 was internalised following 10 min of stimula-
tion with relaxin-3 but not by relaxin, porcine relaxin or
INSL3 as assessed by radioligand internalisation assays
(70–90 % of receptors internalised) or confocal micro-
scopy [35] (Fig. 4). Taken together, these studies indicate
that only treatment with the cognate ligand relaxin-3 causes
RXFP3 to undergo classical b-arrestin-dependent inter-
nalisation. This represents another example of signalling
bias between relaxin and relaxin-3. The detailed mecha-
nisms involved in RXFP3 internalisation, phosphorylation
and recycling/degradation in response to relaxin-3 all
remain to be determined.
Conclusions
There has been significant progress in the last few years in
understanding the biological roles of the relaxin-3-RXFP3
system in stress [6], memory and appetite regulation [14,
18, 19], although the signalling pathways associated with
these functions are less clear. The development of tool
compounds that are selective agonists and antagonists
based on the relaxin-3 peptide structure has been successful
and there are now also potent single chain peptide
Fig. 4 RXFP3 receptors interact with b-arrestins and undergo
internalisation following stimulation with the cognate ligand
relaxin-3 but not with the biased ligands relaxin or R3(BD23–27)R/
I5. The microscopy shows HEK293 cells transfected with GFP2-
labelled RXFP3 receptors. Unstimulated cells or cells incubated with
vehicle show RXFP3 receptors localised to the plasma membrane (a,
b). Exposure of cells to relaxin-3 but not relaxin causes internalisation
of RXFP3 receptors (c, d). Scale bar 10 lM. Examination of
interactions between RXFP3-Rluc8 and the b-arrestin fusion proteins
(b-arrestin 1-Venus or b-arrestin 2-Venus) co-transfected into Flp-In
CHO cells using BRET showed that that interactions occurred
between RXFP3 and both b-arrestins following treatment with
relaxin-3 (1 lM) but not relaxin (1 lM) or R3(BD23–27)R/I5
(1 lM). The interactions were partially blocked by pre-treatment
with PTX suggesting that RXFP3/b-arrestin interactions utilise both
Gi/o-dependent and G protein-independent mechanisms (e, f). BRET
interaction only occurs when receptor and b-arrestins are within
10 nm (g)
Neurochem Res
123
antagonists. This has allowed progress to be made in
identifying the signalling pathways activated by RXFP3
and how the receptor is regulated. However it is likely that
further progress with drugs that target RXFP3 will require
the development of small molecule ligands. To date this
has resulted in the identification of a PAM that displays
probe selectivity but does indicate the presence of an
allosteric site on the receptor that could potentially be
exploited.
Acknowledgments RJS is supported by National Health and
Medical Research Council of Australia Program Grant 1055134.
References
1. Wilkinson TN, Speed TP, Tregear GW, Bathgate RA (2005)
Evolution of the relaxin-like peptide family. BMC Evol Biol 5:14
2. Yegorov S, Good-Avila SV, Parry L, Wilson BC (2009) Relaxin
family genes in humans and teleosts. Ann N Y Acad Sci 1160:42–44
3. Wilkinson TN, Speed TP, Tregear GW, Bathgate RA (2005)
Coevolution of the relaxin-like peptides and their receptors. Ann
N Y Acad Sci 1041:534–539
4. Bathgate RA, Samuel CS, Burazin TC, Layfield S, Claasz AA,
Reytomas IG, Dawson NF, Zhao C, Bond CP, Summers RJ, Parry
LJ, Wade JD, Tregear GW (2002) Human relaxin gene 3 (H3)
and the equivalent mouse relaxin (M3) gene. Novel members of
the relaxin peptide family. J Biol Chem 277:1148–1157
5. Tanaka M, Iijima N, Miyamoto Y, Fukusumi S, Itoh Y, Ozawa H,
Ibata Y (2005) Neurons expressing relaxin 3/INSL 7 in the
nucleus incertus respond to stress. Eur J Neurosci 21:1659–1670
6. Banerjee A, Shen PJ, Ma S, Bathgate RA, Gundlach AL (2010)
Swim stress excitation of nucleus incertus and rapid induction of
relaxin-3 expression via CRF1 activation. Neuropharmacology
58:145–155
7. Ma S, Bonaventure P, Ferraro T, Shen PJ, Burazin TC, Bathgate
RA, Liu C, Tregear GW, Sutton SW, Gundlach AL (2007)
Relaxin-3 in GABA projection neurons of nucleus incertus sug-
gests widespread influence on forebrain circuits via G-protein-
coupled receptor-135 in the rat. Neuroscience 144:165–190
8. McGowan BM, Stanley SA, Smith KL, White NE, Connolly
MM, Thompson EL, Gardiner JV, Murphy KG, Ghatei MA,
Bloom SR (2005) Central relaxin-3 administration causes
hyperphagia in male Wistar rats. Endocrinology 146:3295–3300
9. Ganella DE, Callander GE, Ma S, Bye CR, Gundlach AL,
Bathgate RA (2013) Modulation of feeding by chronic rAAV
expression of a relaxin-3 peptide agonist in rat hypothalamus.
Gene Ther 20:703–716
10. Ganella DE, Ma S, Gundlach AL (2013) Relaxin-3/RXFP3 sig-
naling and neuroendocrine function—a perspective on extrinsic
hypothalamic control. Front Endocrinol 4:128
11. Ryan PJ, Buchler E, Shabanpoor F, Hossain MA, Wade JD,
Lawrence AJ, Gundlach AL (2013) Central relaxin-3 receptor
(RXFP3) activation decreases anxiety-and depressive-like beha-
viours in the rat. Behav Brain Res 244:142–151
12. Ryan PJ, Kastman HE, Krstew EV, Rosengren KJ, Hossain MA,
Churilov L, Wade JD, Gundlach AL, Lawrence AJ (2013)
Relaxin-3/RXFP3 system regulates alcohol-seeking. Proc Natl
Acad Sci USA 110:20789–20794
13. Smith CM, Chua BE, Zhang C, Walker AW, Haidar M, Hawkes
D, Shabanpoor F, Hossain MA, Wade JD, Rosengren KJ,
Gundlach AL (2014) Central injection of relaxin-3 receptor
(RXFP3) antagonist peptides reduces motivated food seeking and
consumption in C57BL/6 J mice. Behav Brain Res 268:117–126
14. Halls ML, Bathgate RA, Sutton SW, Dschietzig TB, Summers RJ
(2015) International Union of Basic and Clinical Pharmacology.
XCV. Recent advances in the understanding of the pharmacology
and biological roles of relaxin family peptide receptors 1-4, the
receptors for relaxin family peptides. Pharmacol Rev 67:389–440
15. Liu C, Chen J, Sutton S, Roland B, Kuei C, Farmer N, Sillard R,
Lovenberg TW (2003) Identification of relaxin-3/INSL7 as a
ligand for GPCR142. J Biol Chem 278:50765–50770
16. Liu C, Eriste E, Sutton S, Chen J, Roland B, Kuei C, Farmer N,
Jornvall H, Sillard R, Lovenberg TW (2003) Identification of
relaxin-3/INSL7 as an endogenous ligand for the orphan G-pro-
tein-coupled receptor GPCR135. J Biol Chem 278:50754–50764
17. Matsumoto M, Kamohara M, Sugimoto T, Hidaka K, Takasaki J,
Saito T, Okada M, Yamaguchi T, Furuichi K (2000) The novel
G-protein coupled receptor SALPR shares sequence similarity
with somatostatin and angiotensin receptors. Gene 248:183–189
18. Bathgate RA, Halls ML, van der Westhuizen ET, Callander GE,
Kocan M, Summers RJ (2013) Relaxin family peptides and their
receptors. Physiol Rev 93:405–480
19. Kuei C, Sutton S, Bonaventure P, Pudiak C, Shelton J, Zhu J,
Nepomuceno D, Wu J, Chen J, Kamme F, Seierstad M, Hack
MD, Bathgate RA, Hossain MA, Wade JD, Atack J, Lovenberg
TW, Liu C (2007) R3(BDelta23 27)R/I5 chimeric peptide, a
selective antagonist for GPCR135 and GPCR142 over relaxin
receptor LGR7: in vitro and in vivo characterization. J Biol Chem
282:25425–25435
20. Hossain MA, Rosengren KJ, Haugaard-Jonsson LM, Zhang S,
Layfield S, Ferraro T, Daly NL, Tregear GW, Wade JD, Bathgate
RA (2008) The A-chain of human relaxin family peptides has
distinct roles in the binding and activation of the different relaxin
family peptide receptors. J Biol Chem 283:17287–17297
21. Bathgate RA, Oh MH, Ling WJ, Kaas Q, Hossain MA, Gooley
PR, Rosengren KJ (2013) Elucidation of relaxin-3 binding
interactions in the extracellular loops of RXFP3. Front Endocri-
nol 4:13
22. Rosengren KJ, Zhang S, Lin F, Daly NL, Scott DJ, Hughes RA,
Bathgate RA, Craik DJ, Wade JD (2006) Solution structure and
characterization of the LGR8 receptor binding surface of insulin-
like peptide 3. J Biol Chem 281:28287–28295
23. Wu B, Chien EY, Mol CD, Fenalti G, Liu W, Katritch V,
Abagyan R, Brooun A, Wells P, Bi FC, Hamel DJ, Kuhn P,
Handel TM, Cherezov V, Stevens RC (2010) Structures of the
CXCR4 chemokine GPCR with small-molecule and cyclic pep-
tide antagonists. Science 330:1066–1071
24. Grosse J, Heffron H, Burling K, Akhter Hossain M, Habib AM,
Rogers GJ, Richards P, Larder R, Rimmington D, Adriaenssens
AA, Parton L, Powell J, Binda M, Colledge WH, Doran J,
Toyoda Y, Wade JD, Aparicio S, Carlton MB, Coll AP, Reimann
F, O’Rahilly S, Gribble FM (2014) Insulin-like peptide 5 is an
orexigenic gastrointestinal hormone. Proc Natl Acad Sci USA
111:11133–11138
25. Zhu J, Kuei C, Sutton S, Kamme F, Yu J, Bonaventure P, Atack J,
Lovenberg TW, Liu C (2008) Identification of the domains in
RXFP4 (GPCR142) responsible for the high affinity binding and
agonistic activity of INSL5 at RXFP4 compared to RXFP3
(GPCR135). Eur J Pharmacol 590:43–52
26. Rosengren KJ, Lin F, Bathgate RA, Tregear GW, Daly NL, Wade
JD, Craik DJ (2006) Solution structure and novel insights into the
determinants of the receptor specificity of human relaxin-3. J Biol
Chem 281:5845–5851
27. Liu C, Chen J, Kuei C, Sutton S, Nepomuceno D, Bonaventure P,
Lovenberg TW (2005) Relaxin-3/insulin-like peptide 5 chimeric
peptide, a selective ligand for G protein-coupled receptor
Neurochem Res
123
(GPCR)135 and GPCR142 over leucine-rich repeat-containing G
protein-coupled receptor 7. Mol Pharmacol 67:231–240
28. Shabanpoor F, Akhter Hossain M, Ryan PJ, Belgi A, Layfield S,
Kocan M, Zhang S, Samuel CS, Gundlach AL, Bathgate RA,
Separovic F, Wade JD (2012) Minimization of human relaxin-3
leading to high-affinity analogues with increased selectivity for
relaxin-family peptide 3 receptor (RXFP3) over RXFP1. J Med
Chem 55:1671–1681
29. Haugaard-Kedstrom LM, Shabanpoor F, Hossain MA, Clark RJ,
Ryan PJ, Craik DJ, Gundlach AL, Wade JD, Bathgate RA,
Rosengren KJ (2011) Design, synthesis, and characterization of a
single-chain peptide antagonist for the relaxin-3 receptor RXFP3.
J Am Chem Soc 133:4965–4974
30. Baker JG, Hill SJ (2007) Multiple GPCR conformations and
signalling pathways: implications for antagonist affinity esti-
mates. Trends Pharmacol Sci 28:374–381
31. Kenakin T, Miller LJ (2010) Seven transmembrane receptors as
shapeshifting proteins: the impact of allosteric modulation and
functional selectivity on new drug discovery. Pharmacol Rev
62:265–304
32. Evans BA, Sato M, Sarwar M, Hutchinson DS, Summers RJ
(2010) Ligand-directed signalling at beta-adrenoceptors. Br J
Pharmacol 159:1022–1038
33. Alvarez-Jaimes L, Sutton SW, Nepomuceno D, Motley ST, Cik
M, Stocking E, Shoblock J, Bonaventure P (2012) In vitro
pharmacological characterization of RXFP3 allosterism: an
example of probe dependency. PLoS One 7:e30792
34. Hossain MA, Bathgate RA, Rosengren KJ, Shabanpoor F, Zhang
S, Lin F, Tregear GW, Wade JD (2009) The structural and
functional role of the B-chain C-terminal arginine in the relaxin-3
peptide antagonist, R3(BDelta23-27)R/I5. Chem Biol Drug Des
73:46–52
35. van der Westhuizen ET, Christopoulos A, Sexton PM, Wade JD,
Summers RJ (2010) H2 relaxin is a biased ligand relative to H3
relaxin at the relaxin family peptide receptor 3 (RXFP3). Mol
Pharmacol 77:759–772
36. van der Westhuizen ET, Werry TD, Sexton PM, Summers RJ
(2007) The Relaxin Family Peptide Receptor 3 (RXFP3) activates
ERK1/2 through a PKC dependent mechanism. Mol Pharmacol
71:1618–1629
37. Morikawa Y, Ueyama E, Senba E (2004) Fasting-induced acti-
vation of mitogen-activated protein kinases (ERK/p38) in the
mouse hypothalamus. J Neuroendocrinol 16:105–112
38. Shen CP, Tsimberg Y, Salvadore C, Meller E (2004) Activation
of Erk and JNK MAPK pathways by acute swim stress in rat
brain regions. BMC Neurosci 5:36
39. Sasaguri K, Kikuchi M, Hori N, Yuyama N, Onozuka M, Sato S
(2005) Suppression of stress immobilization-induced phospho-
rylation of ERK 1/2 by biting in the rat hypothalamic paraven-
tricular nucleus. Neurosci Lett 383:160–164
40. Roux PP, Blenis J (2004) ERK and p38 MAPK-activated protein
kinases: a family of protein kinases with diverse biological
functions. Microbiol Mol Biol Rev 68:320–344
41. Davis RJ (2000) Signal transduction by the JNK group of MAP
kinases. Cell 103:239–252
42. Price MA, Cruzalegui FH, Treisman R (1996) The p38 and ERK
MAP kinase pathways cooperate to activate ternary complex
factors and c-fos transcription in response to UV light. EMBO J
15:6552–6563
43. Whitmarsh AJ, Davis RJ (1996) Transcription factor AP-1 reg-
ulation by mitogen-activated protein kinase signal transduction
pathways. J Mol Med 74:589–607
44. Van Der Westhuizen E, Sexton PM, Bathgate RA, Summers RJ
(2005) Responses of GPCR135 to human gene 3 (H3) relaxin in
CHO-K1 cells determined by microphysiometry. Ann N Y Acad
Sci 1041:332–337
45. Kocan M, Sarwar M, Hossain MA, Wade JD, Summers RJ (2014)
Signalling profiles of H3 relaxin, H2 relaxin and R3(BDelta23-
27)R/I5 acting at the relaxin family peptide receptor 3 (RXFP3).
Br J Pharmacol 171:2827–2841
46. Shi L, Shi SQ, Saade GR, Chwalisz K, Garfield RE (2000)
Studies of cervical ripening in pregnant rats: effects of various
treatments. Mol Hum Reprod 6:382–389
47. Van der Westhuizen ET (2008) Molecular characterisation of
human and mouse relaxin-3 receptors (RXFP3) in recombinant
and endogenously expressing cell lines. PhD Thesis, Monash
University, Melbourne
48. Luttrell LM, Gesty-Palmer D (2010) Beyond desensitization:
physiological relevance of arrestin-dependent signaling. Phar-
macol Rev 62:305–330
49. Cottrell GS, Padilla BE, Amadesi S, Poole DP, Murphy JE, Hardt
M, Roosterman D, Steinhoff M, Bunnett NW (2009) Endosomal
endothelin-converting enzyme-1: a regulator of beta-arrestin-de-
pendent ERK signaling. J Biol Chem 284:22411–22425
Neurochem Res
123
