Content uploaded by Chrishan S Samuel
Author content
All content in this area was uploaded by Chrishan S Samuel on Mar 10, 2014
Content may be subject to copyright.
Relaxin requires the angiotensin II type 2 receptor to
abrogate renal interstitial fibrosis
Bryna S. Man Chow
1,2
, Martina Kocan
3,4
, Sanja Bosnyak
3
, Mohsin Sarwar
3,4
, Belinda Wigg
5
,
Emma S. Jones
4
, Robert E. Widdop
4
, Roger J. Summers
3,4
, Ross A.D. Bathgate
1,2
, Tim D. Hewitson
5,6
and
Chrishan S. Samuel
1,2,4
1
Florey Institute of Neuroscience and Mental Health, Florey Department of Neuroscience and Mental Health, University of Melbourne,
Parkville, VIC, Australia;
2
Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, VIC, Australia;
3
Drug
Discovery Biology Laboratory, Monash Institute of Pharmaceutical Sciences, Parkville, VIC, Australia;
4
Department of Pharmacology,
Monash University, Clayton, VIC, Australia;
5
Department of Nephrology, Royal Melbourne Hospital, Parkville, VIC, Australia and
6
Department of Medicine, University of Melbourne, Royal Melbourne Hospital, Parkville, VIC, Australia
Fibrosis is a hallmark of chronic kidney disease, for which
there is currently no effective cure. The hormone relaxin is
emerging as an effective antifibrotic therapy; however, its
mechanism of action is poorly understood. Recent studies
have shown that relaxin disrupts the profibrotic actions of
transforming growth factor-b1(TGF-b1) by its cognate
receptor, relaxin family peptide receptor 1 (RXFP1),
extracellular signal–regulated kinase phosphorylation, and a
neuronal nitric oxide synthase–dependent pathway to
abrogate Smad2 phosphorylation. Since angiotensin II also
inhibits TGF-b1 activity through its AT2 receptor (AT2R), we
investigated the extent to which relaxin interacts with the
AT2R. The effects of the AT2R antagonist, PD123319, on
relaxin activity were examined in primary rat kidney
myofibroblasts, and in kidney tissue from relaxin-treated
male wild-type and AT2R-knockout mice subjected to
unilateral ureteric obstruction. Relaxin’s antifibrotic actions
were significantly blocked by PD123319 in vitro and in vivo,
or when relaxin was administered to AT2R-knockout mice.
While heterodimer complexes were formed between RXFP1
and AT2Rs independent of ligand binding, relaxin did not
directly bind to AT2Rs but signaled through RXFP1-AT2R
heterodimers to induce its antifibrotic actions. These findings
highlight a hitherto unrecognized interaction that may be
targeted to control fibrosis progression.
Kidney International advance online publication, 15 January 2014;
doi:10.1038/ki.2013.518
KEYWORDS: AT2 receptor; fibrosis; kidney disease; Smad2; relaxin; TGF-b
Characterized by an excessive accumulation of the
extracellular matrix (ECM), primarily collagen, fibrosis is a
universal response to chronic injury and inflammation
in the kidney.1,2 Prolonged exposure to pathological stimuli
and/or profibrotic cytokines causes significant disruption
to the regulatory processes that control the rate at which
the extracellular matrix is synthesized and degraded,
where an imbalance between extracellular matrix synthesis
and degradation results in excessive collagen deposition
at the site of injury.3A failure to resolve this process
causes significant nephron destruction leading to progressive
organ dysfunction and failure, with damage dependent on
the extent of fibrogenesis.1–3 Angiotensin (Ang) II
and transforming growth factor-b1 (TGF-b1) are among
the most potent cytokines that drive this pathological
process.4,5
Despite fibrosis being the final common pathway for all
forms of renal disease and an inevitable feature of end-stage
kidney failure, there are currently no effective treatments to
ameliorate the structural and functional changes that it
causes. Furthermore, the cellular and molecular events that
underlie this process are poorly understood. Thus, the
identification of agents that can alter collagen turnover
and remodeling to prevent or even reduce the fibrosis
that accompanies progressive renal disease is key to both
understanding the cellular and molecular pathways involved
and to developing novel treatment strategies.
The ovarian and cardiovascular hormone, relaxin, has
emerged as a rapid-acting but safe antifibrotic that
ameliorates renal fibrosis in several experimental models,
regardless of etiology.6–9 Although clinical trials have recently
explored its vasodilatory benefits in acute heart failure,10
clinical assessment of the antifibrotic potential of relaxin is
less well developed. Despite end-stage kidney disease being
progressive, it can take a decade to develop, making it
particularly difficult to design, fund, and run trials with hard
end points. To this end, a thorough understanding of the
signal-transduction mechanisms involved in the antifibrotic
http://www.kidney-international.org basic research
&2014 International Society of Nephrology
Correspondence: Chrishan S. Samuel, Department of Pharmacology,
Monash University, Clayton, Victoria 3800, Australia.
E-mail: chrishan.samuel@monash.edu or Tim D. Hewitson, Department of
Nephrology, Royal Melbourne Hospital, Parkville, VIC 3050, Australia.
E-mail: tim.hewitson@mh.org.au
Received 14 May 2013; revised 23 September 2013; accepted 17
October 2013
Kidney International 1
actions of relaxin will significantly facilitate the development
of novel therapeutic targets for intervention and design of
better clinical trials.
Recent studies have demonstrated that human gene-2
(H2) relaxin (the major stored and circulating form of
human relaxin) signals through its cognate G-protein-
coupled receptor relaxin family peptide receptor 1 (RXFP1)
to activate extracellular signal–regulated kinase phosphory-
lation (pERK)1/2 and a neuronal nitric oxide (NO) synthase
(nNOS)-NO-cyclic guanosine monophosphate (cGMP)–
dependent pathway in human11 and rat12,13 renal
myofibroblasts to inhibit TGF-b1 activity, at the level of
Smad2 phosphorylation (pSmad2), an intracellular protein
that promotes the profibrotic actions of TGF-b1.11,14
This in turn inhibits TGF-b1-induced myofibroblast
differentiation and myofibroblast-derived aberrant matrix/
collagen production,11,12 while allowing for an upregulation
of the matrix metalloproteinases (MMP-1/MMP-13, MMP-2,
and MMP-9) that are associated with the breakdown of
existing collagen.13 Furthermore, by suppressing the TGF-b1/
pSmad2 axis that inhibits iNOS activity in myofibroblasts,15
H2 relaxin is able to release iNOS, which through higher
levels of NO specifically contributes to the MMP-promoting
actions of the hormone.13
To further understand how H2 relaxin inhibits the
profibrotic influence of TGF-b1 in various fibroblast culture
models,11–13,16–19 this study sought to find the points at which
H2 relaxin interacts with the well-established Ang II-TGF-b1
system.20 Ang II is a well-known vasoconstrictor that increases
blood pressure, and a potent profibrogenic cytokine.21 These
classical actions of Ang II along with its ability to promote
TGF-b1 activity are mediated through the angiotensin type 1
receptor (AT1R). Conversely, Ang II also negatively regulates
TGF-b1 activity and tissue remodeling by acting at the
angiotensin type 2 receptor (AT2R).22,23 Both AT1Rs and
AT2Rs are expressed in the kidney. Given our previous
findings that H2 relaxin separately inhibits the collagen-
stimulatory actions of Ang II or TGF-b1 in other organs,19 we
aimed to examine the interaction between H2 relaxin and the
AT2R to determine how this influences the profibrotic actions
of TGF-b1. The experiments were performed in primary renal
myofibroblasts in vitro and in an experimental model of
tubulointerstitial renal fibrosis in vivo.
RESULTS
The antifibrotic actions of H2 relaxin are blocked by the AT2R
antagonist PD123319 in vitro
Consistent with our previous findings,12,13 treatment of renal
myofibroblasts with recombinant H2 relaxin (100 ng/ml;
16.8 nmol/l) for 72 h promoted ERK1/2 (p42/p44 mitogen-
activated protein kinase (MAP kinase)) phosphorylation
(pERK1/2), nNOS expression, and nNOS phosphorylation
(pnNOS) by 0.8- to 1-fold (Figure 1a) and levels of collagen-
degrading MMPs (MMP-9, MMP-2, MMP-13) by 0.85- to
1.3-fold (Figure 1c), while inhibiting TGF-b1 expression,
pSmad2, and a-smooth muscle actin (a-SMA) levels (a
marker of myofibroblast differentiation) by 0.5-fold
(Figure 1b; Supplementary Figure S1 online) (all Po0.01
vs. respective values from untreated cells). All these H2
relaxin–induced effects were blocked by the AT2R antagonist
PD123319 (0.1 mmol/l; all Po0.01 vs. H2 relaxin treatment),
whereas PD123319 (0.1 mmol/l) alone did not affect basal
levels of the various parameters measured (Figure 1).
The antifibrotic actions of H2 relaxin are abrogated by the
absence of, or blockade of, AT2Rin vivo
To substantiate the above findings in isolated cells (Figure 1),
in vivo studies examined male AT2R
þ/þ
(wild-type) and
AT2R
/y
(knockout) mice after unilateral ureteric obstruction
(UUO). Their kidneys were assessed at day 2 (when
fibrogenesis can be measured) and day 5 after injury (when
renal fibrosis is well established). Mice were also pretreated or
delayed-treated with H2 relaxin (0.5 mg/kg per day) alone or
in combination with PD123319 (3 mg/kg per day). Total
kidney collagen concentration (Figure 2a), collagen IV
staining (Figure 2b), and collagen I staining (Figure 2c) were
all progressively increased in both AT2R
þ/þ
and AT2R
/y
mice after UUO, but were 0.15- to 1.1-fold greater in AT2R
/y
mice compared with UUO–injured AT2R
þ/þ
mice by day 5
after injury (all Po0.05 vs. respective measurements from
injured AT2R
þ/þ
mice). Immunohistochemically stained
sections of AT2R
/y
mice showed an expanded interstitium,
with increased deposition of collagen IV (Figure 2b) and
collagen I (Figure 2c) by day 5 after injury. Pretreatment and
delayed treatment of AT2R
þ/þ
mice with recombinant H2
relaxin significantly reduced both renal collagen concentration
(Figure 2a) and collagen IV staining (Figure 2b) by 0.33- to
0.4-fold, and further reduced collagen I staining (Figure 2c) to
levels beyond those measured in day 5 (and even day 2) post-
UUO animals (all Po0.05 vs. respective measurements from
day 5 UUO AT2R
þ/þ
mice). These collagen-inhibitory effects
of H2 relaxin treatment were completely lost when it was
administered to AT2R
/y
mice, or when it was coadminis-
tered with PD123319 to AT2R
þ/þ
mice (all Po0.01 vs.
respective values from H2 relaxin–pretreated and delayed-
treated AT2R
þ/þ
mice; Figure 2).
Kidney protein extracts from day 5 UUO controls and
mice that were pretreated with H2 relaxin±PD123319 were
further evaluated for additional targets of relaxin activity.
Compared with AT2R
þ/þ
mice, the kidneys of day 5
UUO–injured AT2R
/y
mice had 0.5- to 0.6-fold lower
pERK1/2, nNOS, pnNOS (Figure 3a), and MMP-13 levels
(Figure 3b), but 1- to 1.3-fold increased TGF-b1, pSmad2,
and a-SMA (Figure 3b). Pretreatment of AT2R
þ/þ
mice with
recombinant H2 relaxin caused a significant elevation in
renal pERK and pnNOS (by 1- to 1.7-fold; Figure 3a) and
MMP-13 levels (by 1.6-fold; Figure 3b), but reduced TGF-b1
and a-SMA expression, as well as pSmad2 phosphorylation
(by 0.5- to 0.6-fold of that in day 5–injured AT2R
þ/þ
mice;
Figure 3b; all Po0.05 vs. respective values from day 5–injured
AT2R
þ/þ
mice). Again, the effects of relaxin were abrogated
when it was administered to AT2R
/y
mice or when it was
basic research BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers
2Kidney International
H2 relaxin (16.8 nmol/l)
Phospho-p44 MAPK (44 kDa)
Phospho-p42 MAPK (42 kDa)
p44 MAPK (44 kDa)
p42 MAPK (42 kDa)
α-Tubulin (55 kDa)
α-Tubulin (55 kDa)
α-Tubulin (55 kDa)
MMP-13 (57 kDa)
L-MMP-2 (72 kDa)
L-MMP-9 (92 kDa)
A-MMP-2
A-MMP-9
TGF-β1 (25 kDa)
TGF-β1
α-SMA (43 kDa)
α-SMA
Smad2 (60 kDa)
pSmad2 (60 kDa)
pSmad2
MMP-9 MMP-2 MMP-13
pnNOS (140 kDa)
pnNOSnNOSpERK1/2
–
–
–
–
–
–
–
––
–++
++
–
––
–++
++
–
––
–++
++
–
––
–++
++
–
––
–++
++
–
––
–++
++
–
–––
––
–
––
––
–
2.5
1.25
1
0.75
0.5
0.25
0
1.25
1
0.75
0.5
0.25
0
1.25
1
0.75
0.5
0.25
0
**
**
** **
**
** **
** **
†† †† ††
††
††
††
†† †† ††
1.5
1
0.5
0
Relative OD
Relative OD
Relative OD
2
2.5
1.5
1
0.5
0
2
2.5
1.5
1
0.5
0
2
2.5
1.5
1
0.5
0
2
2.5
1.5
1
0.5
0
2
2.5
1.5
1
0.5
0
2
–
–– –
––
++
+
+
+
+
++
++
+
+++
+
+
+++
–
–
–
–––
––
+++
+
+
+++
+
+
++
–
–
–
–++
++
+
nNOS (155 kDa)
PD123319 (0.1 μmol/l)
H2 relaxin (16.8 nmol/l)
PD123319 (0.1 μmol/l)
H2 relaxin (16.8 nmol/l)
PD123319 (0.1 μmol/l)
H2 relaxin (16.8 nmol/l)
PD123319 (0.1 μmol/l)
H2 relaxin (16.8 nmol/l)
PD123319 (0.1 μmol/l)
H2 relaxin (16.8 nmol/l)
PD123319 (0.1 μmol/l)
Figure 1 | The antifibrotic actions of human gene-2 (H2) relaxin are blocked by the AT2 receptor (AT2R) antagonist PD123319 in vitro.
Representative western blots of renal (a) phosphorylated (phospho-) p44 and p42 mitogen-activated protein kinase (MAPK) (extracellular
signal–regulated kinase 1/2 phosphorylation (pERK1/2)), total p44 and p42 MAPK (ERK1/2), unphosphorylated neuronal nitric oxide synthase
(nNOS), nNOS phosphorylation (pnNOS), and a-tubulin; (b) transforming growth factor-b1 (TGF-b1), Smad2 phosphorylation (pSmad2),
unphosphorylated Smad2, a-smooth muscle actin (SMA), and a-tubulin; and (c) representative zymographs of latent (L) and active (A) matrix
metalloproteinase-9 and -2 (MMP-9 and -2) levels and representative western blots of MMP-13 and a-tubulin expression from untreated
(control) rat renal myofibroblasts and cells treated with H2 relaxin (16.8 nmol/l) alone, H2 relaxin (16.8 nmol/l), PD123319 (0.1 mmol/l), or
PD123319 (0.1 mmol/l) alone after 72 h in culture. The total (a) p44 and p42 MAPK (ERK1/2), (b) unphosphorylated Smad2, and (a–c)a-tubulin
blots were included to demonstrate the quality and equivalent loading of protein samples. Also shown is the relative mean±s.e.m. Optical
density (OD) levels of (a) pERK1/2 (corrected for total ERK1/2 levels), nNOS, and pnNOS (both corrected for a-tubulin levels); (b) TGF-b1, a-SMA
(both corrected for a-tubulin levels), and pSmad2 (corrected for Smad2 levels); and (c) MMP-9, MMP-2, and MMP-13 (corrected for a-tubulin
levels) from each of the groups studied, as determined by densitometry scanning (from n¼4 experiments conducted in duplicate), to that of
the untreated group, which was expressed as 1 in each case. **Po0.01 vs. untreated cells;
ww
Po0.01 vs. H2 relaxin alone-treated cells.
Kidney International 3
BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers basic research
coadministered with PD123319 to AT2R
þ/þ
mice (all Po0.01
vs. respective values from H2 relaxin–treated AT2R
þ/þ
mice;
Figure 3), confirming that the AT2R was necessary for H2
relaxin to mediate its antifibrotic actions.
Constitutive heterodimers are formed between RXFP1 and
the AT2R
Bioluminescence resonance energy transfer (BRET) saturation
assays carried out in human embryonic kidney (HEK)293
cells to examine the formation of dimers between RXFP1 and
AT2R indicated that constitutive dimerization occurred
between RXFP1 and AT2R. Saturation curves were con-
structed by keeping a constant amount of Rluc8-tagged AT2R
and increasing the amount of Venus-tagged RXFP1. The
BRET curve detecting RXFP1 and AT2R interactions showed
saturation with a maximum BRET ratio of B0.1 (Figure 4a),
providing strong evidence that RXFP1 and AT2Rs formed
heterodimers. In contrast, no dimerization was observed
between RXFP1 and the thyrotropin-releasing hormone
receptor 1, which was used as a negative control (Figure 4a).
The effect of recombinant H2 relaxin and Ang II on the
RXFP1-AT2R dimers was then examined using real-time
kinetic BRET assays. The addition of H2 relaxin (30 nmol/l),
Ang II (1 mmol/l), or a combination of H2 relaxin and Ang II
Preventative (P) Delayed (D)
treatment treatment
†† †† ¶¶ ¶¶
[–D2 – D5] [D2 – D5]
Preventative (P) Delayed (D)
treatment treatment
[–D2 – D5] [D2 – D5] Preventative (P) Delayed (D)
treatment treatment
[–D2 – D5] [D2 – D5]
*
§§
**
##
2
Collagen concentration
(% collagen content /
dry wt kidney tissue)
Total collagen IV
(% fractional area)
Total collagen I
(% fractional area)
1.5
1
0.5
H2 relaxin
PD123319
D2
–
–
–
–
–
––
–
–– ––
+
+
+++ +
+
+
D5
AT2R+/+
AT2R–/y
AT2R+/+
AT2R–/y AT2R+/+
20
15
10
5
0
AT2R–/y
20
15
10
5
0D2 D5 After UUO
H2 relaxin
PD123319
H2 relaxin
PD123319
AT2R+/+ (D5)AT2R+/y (D5)
AT2R+/+ (D5)AT2R+/y (D5)
–
–
–
–
–
–
–
–
+++
+
+++
+–
+ H2 relaxin (P)
+ H2 relaxin (D)
+ H2 relaxin (P)
+ H2 relaxin (P)
+ H2 relaxin (D)
+ H2 relaxin (P)
+ H2 relaxin (D) + H2 relaxin (D)
+ H2 relaxin + PD (D)
+ H2 relaxin + PD (P)
––
–
After UUO
##
*
§§
##
**
§§
††
††
†† ††
**
¶¶
¶¶
**
¶¶
¶¶
D2 D5 After UUO
–
–
–
––
––+++
+
++ +
+– ––––
+ H2 relaxin + PD (D)
+ H2 relaxin + PD (P)
Figure 2 | The collagen-inhibitory actions of human gene-2 (H2) relaxin were abrogated by the absence of, or blockade of, AT2
receptors (AT2Rs) in vivo.Shown is the relative mean±s.e.m. (a) renal collagen concentration (% collagen content/dry weight kidney tissue),
(b) collagen IV staining (as a % of the fractional area), and (c) collagen I staining (as a % of the fractional area) from day 2 (D2)– and day 5
(D5)–injured AT2R
þ/þ
and AT2R
/y
mice; D5-injured AT2R
þ/þ
and AT2R
/y
animals that were pretreated (from 2 days before ( D2) to injury)
or delayed-treated (from D2 after unilateral ureteric obstruction (UUO)) with H2 relaxin (0.5 mg/kg per day); and D5-injured AT2R
þ/þ
mice that
were pretreated or delayed-treated with H2 relaxin (0.5 mg/kg per day) and PD123319 (3mg/kg per day), from n¼5–7 mice per treatment
group. As a reference, uninjured AT2R
þ/þ
mice have renal collagen concentrations of 0.75–0.85%. Also shown are representative images of (b)
collagen IV and (c) collagen I staining from the preventative (P) and delayed (D) treatment groups investigated at D5 after injury. Bar ¼100mm.
##
Po0.01 vs. D2-injured AT2R
þ/þ
mice;
yy
Po0.01 vs. D2-injured AT2R
/y
mice; *Po0.05 vs. D5-injured AT2R
þ/þ
mice;
ww
Po0.01 vs. D5-injured
AT2R
þ/þ
mice pretreated with H2 relaxin alone;
zz
Po0.01 vs. D5-injured AT2R
þ/þ
mice delayed-treated with H2 relaxin alone.
4Kidney International
basic research BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers
AT2R+/+
pERK1/2
**
** **
††
**
††
**
††
**
††
**
††
**
††
**
†† **
††
**
††
**
†† **
††
**
**
**
**
**
**
*
*
††
*
††
**
**
**
**
H2 relaxin
PD123319
3
3
2.5
2
Relative ODRelative OD
1.5
TGF-β1pSmad2
MMP-13
α-SMA
1
0.5
3
2.5
2
1.5
H2 relaxin
PD123319
1
0.5
0
3
2.5
2
1.5
1
0.5
0
0
3
2.5
2
1.5
1
0.5
0
2.5
2
1.5
1
–
–– –
+
–
+
H2 relaxin (0.5 mg/kg per day)
PD123319 (3 mg/kg per day)
+
+–
–
–
–
–
–
–
–
–
–
–
–
–
–
+
–
+
–
+
–
+
–
+
–
+
+
+
+
+
+
+
–
–
–
–
–
+
–
+
+
+
–
–– –
+
–
+
+
+––
–– –
+
–
+
+
+–
0.5
Relative OD
0
3
2.5
2
1.5
1
0.5
0
3
2.5
2
1.5
1
0.5
0
nNOS pnNOS
AT2 R +/+
AT2 R –/γ
H2 relaxin (0.5 mg/kg per day) –
–
–
–
–
–
–
–
+
–
+
–
+
–
+
–
+
+
+
Phospho-p44 MAPK (44 kDa)
Phospho-p42 MAPK (42 kDa)
p44 MAPK (44 kDa)
p42 MAPK (42 kDa)
nNOS (155 kDa)
pnNOS (140 kDa)
α-Tubulin (55 kDa)
+PD123319 (3 mg/kg per day)
AT2 R +/+ AT 2 R +/+
AT2 R –/γAT 2 R –/γ
AT2 R +/+
TGF-β1 (25 kDa)
pSmad2 (60 kDa)
Smad2 (60 kDa)
MMP-13 (57 kDa)
α-SMA (43 kDa)
α-Tubulin (55 kDa)
AT2 R +/+ AT 2 R +/+
AT2 R –/γAT 2 R –/γ
Figure 3 |The antifibrotic actions of human gene-2 (H2) relaxin are abrogated by the absence of, or blockade of, AT2receptors (AT2Rs)
in vivo.Representative western blots of renal (a) phosphorylated (phospho-) p44 and p42 mitogen-activated protein kinase (MAPK) (ERK1/2
phosphorylation (pERK1/2)), total p44 and p42 MAPK (ERK1/2), unphosphorylated neuronal nitric oxide synthase (nNOS), nNOS
phosphorylation (pnNOS), and a-tubulin; (b) transforming growth factor-b1 (TGF-b1), Smad2 phosphorylation (pSmad2), unphosphorylated
Smad2, a-smooth muscle actin (a-SMA), matrix metalloproteinase-13 (MMP-13), and a-tubulin, from day 5–injured AT2R
þ/þ
and AT2R
/y
mice±pre-treatment with H2 relaxin (0.5 mg/kg per day) alone; and day 5–injured AT2R
þ/þ
mice pre-treated with H2 relaxin (0.5 mg/kg per
day) and PD123319 (3 mg/kg per day). The total (a) p44 and p42 MAPK (ERK1/2), (b) unphosphorylated Smad2, and (aand b)a-tubulin blots
were included to demonstrate the quality and equivalent loading of protein samples. Also shown is the relative mean±s.e.m. optical density
(OD) levels of (a) pERK1/2 (corrected for total ERK1/2 levels), nNOS, and pnNOS (both corrected for a-tubulin levels); (b) TGF-b1, a-SMA, MMP-13
(all corrected for a-tubulin levels), and pSmad2 (corrected for Smad2 levels) from each of the groups studied, as determined by densitometry
scanning (from n¼5–6 mice/treatment group), to that of the day 5–injured AT2R
þ/þ
mouse group, which was expressed as 1 in each case.
**Po0.01 vs. day 5–injured AT2R
þ/þ
mice;
ww
Po0.01 vs. day 5–injured AT2R
þ/þ
mice pretreated with H2 relaxin alone.
Kidney International 5
BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers basic research
RXFP1 and AT2R
RXFP1 and TRHR1
0.12
0.10
0.08
0.06
BRET ratio
0.04
0.02
0.00
0.0 0.5
Receptor-Venus / Receptor-Rluc8
1.0 1.5 2.0
–1
0.2
0.1
Ligand or vehicle
addition
H2 relaxin
Ang II
H2 relaxin+Ang II
0.0
Ligand-induced BRET ratio
–0.1
01
Time (min)
AT2R (44 kDa)
α-Tubulin (55 kDa)
AT2R (44 kDa)
1.5
1.0
Relative OD
0.5
0.0
HEK293 + AT2R
Vehicle
Vehicle + PD123319
H2 relaxin + PD123319
0
50
pERK1/2 (% FBS)
100
150
200
0 5 10 15
Time (min)
20 25
H2 relaxin
Vehicle
–11
150
100
50 Candesartan cilexetil
CGP42112
H2 relaxin
0
–50
% Binding
–10 –9
[Ligand]
–8 –7 –6
HEK293
Vehicle + PD123319
H2 relaxin + PD123319
0
50
pERK1/2 (% FBS)
100
150
200
0 5 10 15
Time (min)
20 25
H2 relaxin
Vehicle
Vehicle + PD123319
**
*
**
** H2 relaxin + PD123319
0
50
pERK1/2 (% FBS)
100
150
200
0 5 10 15
Time (min)
20 25
H2 relaxin
Vehicle
HEK293-R HEK293-RXEP1 + AT2R
Vehicle + PD123319
H2 relaxin + PD123319
0
50
pERK1/2 (% FBS)
100
150
200
0 5 10 15
Time (min)
20 25
H2 relaxin
+H2
relaxin
–H2
relaxin
128 kDa
89 kDa
39 kDa
33 kDa
2345
Figure 4 | Human gene-2 (H2) relaxin does not directly interact with the AT2receptor (AT2R). (a) Bioluminescence resonance energy
transfer (BRET) saturation curves and (b) real-time kinetic BRET curves detecting (a) ligand-independent and (b) ligand-induced interactions
between relaxin family peptide receptor 1 (RXFP1) and AT2Rs. Human embryonic kidney (HEK)293 cells were transiently co-transfected with
(aand b) RXFP1-Venus and AT2R-Rluc8 or (a) RXFP1-Venus and thyrotropin-releasing hormone receptor 1 (TRHR1)-Rluc8 (negative control). The
BRET ratios are plotted as a function of the expression ratio of (Receptor-Venus)/(Receptor-Rluc8). (b) The ligand-induced BRET ratios were
detected before and after treatment with agonist (H2 relaxin (30 nmol/l); Ang II (1 mmol/l); H2 relaxin (30 nmol/l) plus angiotensin II (Ang II)
(1 mmol/l) or vehicle (phenol red-free þ10% fetal bovine serum (FBS) þ0.01% bovine serum albumin) and calculated as described in Materials
and Methods. Data shown (aand b) are the mean±s.e.m. of n¼3 experiments performed in triplicate. Competition binding curves for
CGP42112, candesartan cilexetil, and H2 relaxin (c) were also generated from HEK293 cells stably transfected with AT2Rs. Data shown (c) are the
mean±s.e.m. of n¼4 performed in triplicate. Additionally shown are representative western blots of renal AT2R expression (d) from untreated
versus H2 relaxin (16.8 nmol/l)–treated rat renal myofibroblasts (from n¼4 experiments in duplicate). Functional responses of RXFP1-AT2R
dimers in response to H2 relaxin, using the AlphaScreen extracellular signal–regulated kinase 1/2 (ERK1/2) phosphorylation assay—shown are
the effects of (e) vehicle (20 mmol/l sodium acetate buffer, pH 5.0) alone, vehicle þPD123319 (1 mmol/l), H2 relaxin (30 mmol/l) alone, and H2
relaxin (30 mmol/l) þPD123319 (1 mmol/l) in untransfected HEK293 cells, (f) HEK293 cells transfected with AT2Rs, (g) HEK293 cells expressing
RXFP1 (HEK-RXFP1), and (h) HEK-RXFP1 cells transfected with AT2Rs on ERK1/2 phosphorylation (pERK1/2) (p42/p44 mitogen-activated protein
kinase) activity over a 60-min period (of which responses over the first 20 min are shown). All data (e–h) are expressed as the mean±s.e.m. of
n¼4–6 experiments. *Po0.05, **Po0.01 vs. respective measurements from vehicle-treated cells;
w
Po0.05 vs. respective measurement from H2
relaxin–treated cells. OD, optical density.
6Kidney International
basic research BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers
had no effect on existing heterodimers between RXFP1 and
the AT2R (Figure 4b).
H2 relaxin does not directly interact with the AT2R
To determine whether H2 relaxin displayed any direct
binding affinity for the AT2R, competition-binding assays
were carried out with H2 relaxin, an AT2R agonist
CGP42112, and an AT1R antagonist candesartan in
HEK293 cells expressing the AT2R (Figure 4c). Only
CGP42112, but not H2 relaxin or candesartan, was able to
bind to the AT2R (Figure 4c), confirming that H2 relaxin did
not directly bind to the AT2R. Furthermore, western blotting
of AT2Rs endogenously expressed in primary rat renal
myofibroblasts confirmed that H2 relaxin (16.8 nmol/l;
72 h) did not alter AT2R receptor expression (Figure 4d),
despite the antibody used detecting the expected 44-kDa
product and two larger-molecular-weight products.
H2 relaxin mediates the cross-talk through RXFP1-AT2R
dimerization at the level of pERK1/2
To determine the downstream consequences of RXFP1-AT2R
dimerization (Figure 4a) and whether this was possibly linked
to the ability of H2 relaxin to stimulate pERK1/2 signaling
downstream of RXFP1 (Figure 1),12,13 untransfected HEK293
cells or HEK293-expressing RXFP1 (HEK-RXFP1) and
transfected with AT2R were measured for pERK1/2 activity
using the AlphaScreen ERK1/2 phosphorylation assay.24 H2
relaxin (30 nmol/l, a dose previously used to stimulate
intracellular signaling12,25) did not stimulate any pERK1/2
activity over 60 min, when administered to untransfected
cells (Figure 4e) or cells transiently transfected with AT2Rs
(Figure 4f), confirming the findings from the competition-
binding assays (Figure 4c) that H2 relaxin did not directly
bind to the AT2R. In contrast, pERK1/2 activity was
significantly increased over 2–5 min, when H2 relaxin was
administered to HEK-RXFP1 alone (Figure 4g) or to HEK-
RXFP1 cells transfected with the AT2R (Figure 4h) (Po0.05
vs. respective vehicle treatment of cells at both time points),
before returning to baseline within 10 min. The H2
relaxin–induced stimulation of pERK1/2 in HEK-RXFP1 cells
was unaffected by PD123319 (1 mmol/l) (Figure 4g), but was
significantly, although not totally, abrogated when the
AT2R was coexpressed (Figure 4h), the latter being consistent
with transient AT2R expression resulting in coexpression
(of both receptors) in only a proportion of RXFP1-expressing
cells.
DISCUSSION
This study explored the interaction between H2 relaxin-
RXFP112 and AT2R22,23,26–29 activation, which inhibits
AT2R
H2 relaxin
RXFP1
TGF-β1
TGF-β receptor
iNOS
iNOS
NO
Smad2Smad2
ERK1/2
ERK1/2
nNOS
NO
cGMP
Altered gene transcription
P
P
P
P
P
P
Smad2
Smad4Smad3
MMPs
(-1/-13, -2, -9)
Differentiation
(α-SMA)
Collagen deposition
Fibrosis
Nucleus
Intracellular
Extracellular
RXFP1-AT2R
heterodimer
Figure 5 |A schematic illustration of the proposed signal-transduction mechanisms of human gene-2 (H2) relaxin’s antifibrotic actions,
via relaxin family peptide receptor 1-AT2 receptor (RXFP1-AT2R) heterodimers. Previous findings from primary renal myofibroblasts11–13
demonstrated that H2 relaxin signaled through a extracellular signal–regulated kinase 1/2 phosphorylation (pERK1/2) and a neuronal nitric
oxide (NO) synthase (nNOS)-NO-cyclic guanosine monophosphate (cGMP)–dependent pathway to inhibit Smad2 phosphorylation (pSmad2)
and translocation to the nucleus as a means of disrupting transforming growth factor-b1 (TGF-b1) activity. This in turn ameliorated
TGF-b1-mediated myofibroblast differentiation and myofibroblast-induced aberrant collagen deposition (the basis of fibrosis), while allowing
for an upregulation of matrix metalloproteinases (MMPs) associated with collagen degradation. Furthermore, the H2 relaxin–induced
suppression of the TGF-b1, which itself inhibits inducible NOS (iNOS) expression in myofibroblasts, releases iNOS, which through higher levels
of NO specifically contributes to the MMP–promoting actions of the antifibrotic hormone. Our findings from the current study now suggest
that H2 relaxin signals through constitutive RXFP1-AT2R heterodimers and a RXFP1/AT2R-pERK1/2-nNOS-NO-cGMP pathway to inhibit the
TGF-b1/pSmad2 axis and the downstream effects of TGF-b1, allowing it to induce the various actions detailed above (as indicated by the
red arrows). a-SMA, a-smooth muscle actin.
Kidney International 7
BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers basic research
TGF-b1 expression and/or activity while exerting cardio/
renoprotection through an nNOS-NO-cGMP–dependent
pathway. Consistent with our previous findings, we found
that H2 relaxin signals through an RXFP1-pERK1/2-nNOS-
NO-cGMP–dependent pathway to abrogate the TGF-b1/
pSmad2 axis12,13 and inhibit the profibrotic influence of
TGF-b1 on renal myofibroblast differentiation,12 while
regulating collagen-degrading MMPs (MMP-1/-13, MMP-2
and MMP-9).13 Surprisingly, however, this was completely
abolished by the AT2R antagonist PD123319 at all levels.
Furthermore, the antifibrotic actions of relaxin in vivo,
regardless of whether it was administered before or after
UUO, were completely lost when the AT2R was either absent
(in AT2R
/y
mice) or antagonized (with PD123319),
confirming that the AT2R was essential for the antifibrotic
actions of H2 relaxin. However, H2 relaxin did not produce
these effects by a direct action at the AT2R, as the peptide did
not bind to AT2Rs or alter the dynamics of already formed
RXFP1-AT2R complexes. Instead, we demonstrated for the
first time that H2 relaxin signals through constitutive RXFP1-
AT2R heterodimers to induce its downstream effects,11–13
resulting in the inhibition of the TGF-b1/pSmad2 axis and
hence reduced TGF-b1-induced collagen deposition (the
basis of fibrosis) (Figure 5).
These findings were quite striking and may help explain
why H2 relaxin only displays its antifibrotic effects under
pathological conditions, without affecting normal extracel-
lular matrix and fibroblast function under physiological
conditions.6–9,16–19 As AT2Rs are expressed at low levels in
tissues22,23,27 and fibroblasts under physiologically quiescent
states, but are markedly increased in number and activity
under pathological conditions, the increased availability of
AT2Rs in injured/diseased tissues improves the functional
importance of these receptors,22,23,26–29 not only on their own
but also through interactions with other receptors. Hence,
pathological conditions would be more conducive to RXFP1-
AT2R heteromerization taking place and hence for H2 relaxin
to mediate its antifibrotic actions through these RXFP1-AT2R
heteromers. Furthermore, as activation of AT2Rs decreases
AT1R expression30 and antagonizes the effects of AT1R
activation,30,31 our findings suggest that H2 relaxin may
also indirectly affect the Ang II–AT1R–TGF-b1 interaction via
effects on RXFP1-AT2R dimers.
Consistent with our findings here, H2 relaxin reduces Ang
II infusion–induced renal oxidative stress, glomerular sclero-
sis, arterial pressure, and albuminuria in hypertensive rats.32
These protective effects, however, were lost in animals
cotreated with the general NOS inhibitor L-NAME. Along
with previous studies demonstrating that nNOS is widely
distributed in the kidney and shown to be a marker of renal
injury,33 our current observations suggest that the nNOS-NO
pathway is absolutely required for the protective/antifibrotic
actions of H2 relaxin in kidney/renal myofibroblasts, and are
likely mediated via RXFP1-AT2R heterodimers.
The accelerated tubulointerstitial fibrosis (represented
by increased collagen IV, collagen I, and total collagen
concentration) in AT2R
/y
mice subjected to UUO that was
measured in this study is consistent with previous findings,34
confirming that AT2Rs protect from renal fibrosis. These
findings are also consistent with our previous observations
that RXFP1
/y
mice similarly underwent more rapidly
progressive interstitial renal fibrosis after UUO.12 We have
now, however, provided further insights into the mechanisms
by which AT2Rs protected against renal fibrosis progression
at the in vitro and in vivo level, which involved an
upregulation of ERK1/2 phosphorylation, nNOS activity
(nNOS-mediated NO production), and collagen-degrading
MMP levels, and downregulation of TGF-b1/pSmad2,
myofibroblast differentiation, and aberrant collagen levels.
Taken together, these combined findings suggested that H2
relaxin uses the antifibrogenic and antiremodeling effects of
AT2Rs (via RXFP1-AT2R dimers) to mediate its antifibrotic
actions.
Although our previous studies12 confirmed that RXFP1 in
renal myofibroblasts was essential for H2 relaxin’s inhibition
of myofibroblast differentiation and myofibroblast-mediated
collagen deposition, we now demonstrate that constitutive
RXFP1-AT2R heteromer complexes are formed
independently of any ligand (H2 relaxin or Ang II)
binding. These findings resemble other heteromerization
paradigms among G-protein-coupled receptors.35,36 In
addition, we now demonstrate that H2 relaxin appeared to
signal through these constitutive RXFP1-AT2R heterodimers
to induce downstream functional effects, at the level of
pERK1/2 (which had previously been shown to be activated
by relaxin-RXFP112,13 or Ang II-AT2R28 to mediate organ
protection), to inhibit the TGF-b1/Smad2 axis. Our added
finding that pERK1/2 responses were almost completely
blocked by PD123319 under conditions where RXFP1-AT2R
dimers are present also suggested that these RXFP1-AT2R
heteromers may be regulated by AT2R blockade, as a novel
means by which the actions of H2 relaxin may be inhibited.
In conclusion, we have demonstrated a novel mechanism
by which H2 relaxin disrupts the profibrotic effects of
TGF-b1 in activated myofibroblasts. The AT2R is critically
required for H2 relaxin to disrupt the TGF-b1/pSmad2 axis
via a RXFP1/AT2R-pERK1/2-nNOS-NO-cGMP–dependent
pathway, which in turn inhibits the eventual downstream
effects of TGF-b1 on myofibroblast differentiation, aberrant
collagen deposition, and collagen-degrading MMP levels.
This essential requirement is via the formation of AT2R-
RXFP1 constitutive heterodimers. The relative absence of the
AT2R under normal physiological conditions and its
upregulation in models of fibrosis may explain why the
antifibrotic effects of H2 relaxin are only seen under
pathological conditions. These findings further demonstrate
that H2 relaxin acts at multiple levels to disrupt both TGF-
b1 signal transduction and the profibrotic interaction
between Ang II and TGF-b1. Furthermore, we have
identified the AT2R as a novel therapeutic target that may
enhance the antifibrotic potential of H2 relaxin to reduce
fibrosis progression.
8Kidney International
basic research BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers
MATERIALS AND METHODS
Materials
Recombinant H2 relaxin was generously provided by Corthera (San
Mateo, CA; a subsidiary of Novartis International AG, Basel,
Switzerland) and is bioactive in rats12,13,18,19 and mice.12,17,19,37
CGP42112 (AT2R agonist) was obtained from GL Biochem
(Shanghai) (Shanghai, China), whereas candesartan cilexetil (AT1R
antagonist) was obtained from AstraZeneca (So
¨derta
¨lje, Sweden).
Animals
A rat and mouse model of UUO, which mimics the pathology of
human progressive renal disease,38 was used as an experimental
model of primary tubulointerstitial fibrosis. In each case, a single
ureter was ligated under general anesthesia with the contralateral
kidney left intact. Tissue was collected from the obstructed kidneys of
male Sprague–Dawley rats (obtained from the Animal Resource
Centre, Perth, WA, Australia) for the propagation of renal fibroblasts.
Similarly, the role of the renal AT2R was examined in male littermate
AT2R
þ/þ
and AT2R
/y
mice (on an FVB/N background; kindly
provided by Professor Lutz Hein, University of Freiburg, Freiburg,
Germany; and validated previously39), subjected to UUO.
Animals were housed in a controlled environment and main-
tained on a fixed lighting schedule with free access to rodent lab
chow (Barastock Stockfeeds, Pakenham, VIC, Australia) and water.
These experiments were approved by the Florey Institute of
Neuroscience and Mental Health’s and Monash University’s Animal
Ethics Committees, which adhere to the Australian code of practice
for the care and use of laboratory animals for scientific purposes.
Cell culture
Myofibroblasts propagated from fibrotic kidneys40 of Sprague–
Dawley rats, 3 days after UUO were used for these studies, as
described before,12,13,41 as they behave like their counterparts from
fibrotic human kidneys, and respond to H2 relaxin in a similar
manner as TGF-b1-stimulated human renal fibroblasts.11 Cells were
characterized by immunocytochemical staining. As 100% of cells
stained positive for the mesenchymal marker vimentin, and 60–70%
of cells positively stained for a-SMA, it was concluded that
fibroblasts constituted 100% of the cell population used for
experimentation, of which 60–70% were myofibroblasts.
Cultures were maintained in Dulbecco’s modified Eagle’s
medium supplemented with 10% fetal bovine serum (FBS), 2.2%
HEPES, 1% L-glutamine, penicillin (50 U/ml), and streptomycin
(50 mg/ml) (Dulbecco’s modified Eagle’s medium-FBS) at 371C. All
described experiments were independently replicated at least 3–6
separate times in duplicate, with rat renal myofibroblasts used
between passages 13 and 18 (during which time they expressed
RXFP1 and the AT2R).
Evaluating the effects of AT2R blockade on the antifibrotic
effects of H2 relaxin in vitro
To elucidate whether there was any interaction between H2 relaxin
and the AT2R, rat myofibroblasts were seeded into 12-well plates at
an equal density of 1–1.2510
5
cells per well and continuously
treated with recombinant H2 relaxin (100 ng/ml; 16.8 nmol/l)12,13 in
the absence or presence of the specific AT2R antagonist PD123319
(0.1 mmol/l) over 72 h. Untreated cells and cells treated w ith
PD123319 alone (at 0.1 mmol/l) for 72 h were used as appropriate
controls. After 72 h, proteins were extracted from cell layers
with Trizol reagent (Invitrogen, Carlsbad, CA; according to the
manufacturer’s instructions) for further analysis.
Evaluating the effects of AT2R blockade on the antifibrotic
effects of H2 relaxin in vivo
The importance of the AT2R in mediating H2 relaxin’s antifibrotic
actions was further investigated in 6- to 8-week-old male AT2R
þ/þ
and AT2R
/y
mice on an FVB/N background, subjected to the
UUO–induced model of tubulointerstitial renal fibrosis.12,37
Subgroups of AT2R
þ/þ
and AT2R
/y
mice (n¼5–6 per genotype
and treatment group) that were subjected to UUO were either left
untreated until 2 or 5 days postoperatively (injury controls) or were
either pretreated with recombinant H2 relaxin (0.5 mg/kg per day;
via subcutaneously implanted osmotic minipumps; model 1007D;
Alzet, Cupertino, CA) from 2 days before UUO until 5 days after
injury (preventative treatment) or treated with the same dose of H2
relaxin (via 1003D pumps; Alzet) from days 2 to 5 after UUO
(delayed treatment). This dose of H2 relaxin had previously been
used to successfully prevent/reverse fibrosis progression in various
models of renal disease, regardless of etiology,6–9 and was found to
produce B20 ng/ml of circulating relaxin after 5 days of
administration.42 Further subgroups of AT2R
þ/þ
mice were either
pretreated or delayed-treated with H2 relaxin (as above) in
combination with PD123319 (3 mg/kg per day;43 via osmotic
minipumps over the 7- or 3-day treatment periods, respectively).
At 2 and 5 days after UUO, all animals were killed by an overdose
of anesthetic, and their obstructed kidneys were collected for total
protein extraction and various analyses.
Western blotting
Equal amounts of total protein (10–30 mg; from the preventative study
groups) from each sample was electrophoresed on 10.5% acrylamide
gels, as described previously.13 Western blot analyses were then
performed with primary polyclonal antibodies to either pERK1/2
(Thr202/Tyr204; Cell Signaling Technology, Danvers, MA),
unphosphorylated nNOS (BD Biosciences, San Jose, CA), pnNOS
(Ser1417; Pierce Biotechnology, Rockford, IL), TGF-b1 (Santa Cruz
Biotechnology, Santa Cruz, CA), and the AT2R (Santa Cruz
Biotechnology); primary monoclonal antibodies to pSmad2 (Ser465/
467; Cell Signaling Technology), a-SMA (Dako Corporation,
Carpinteria, CA), or MMP-13 (Abcam, Cambridge, MA); and the
appropriate secondary antibodies. Membranes probed with pERK1/2
and pSmad2 were stripped and reprobed with total ERK1/2 and
(unphosphorylated) Smad2, respectively, whereas a-tubulin levels
(Millipore Corporation, Bedford, MA) were additionally assessed to
demonstrate equivalent loading of samples. Blots detected with the
ECL detection kit (Amersham Pharmacia Biotech, Piscataway, NJ)
were quantified by densitometry with a GS710 Calibrated Imaging
Densitometer and Quantity-One software (Bio-Rad Laboratories,
Richmond, CA). The density of each parameter was corrected for a-
tubulin protein levels (or total ERK1/2 and Smad2 levels for pERK1/2
and pSmad2, respectively) and expressed relative to the untreated
control group (in vitro studies) or the day 5–injured AT2R
þ/þ
mouse
group (in vivo studies), defined as 1 in each case.
Gelatin zymography
Conditioned media or tissue extracts from the preventative study
groups were incubated with 5 mmol/l amino-phenyl mercuric
acetate (Sigma-Aldrich) for 6 h at 37 1C, to stimulate the activation
of latent MMPs, before gelatin zymography. Equal volumes of
Kidney International 9
BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers basic research
samples were electrophoresed on gelatin zymographs, as detailed
before.13 Gelatinolytic activity was indicated by clear bands and
assessed by densitometry, as described above.
Hydroxyproline assay
Equivalent tissue portions (containing cortex and medulla) from
AT2R
þ/þ
and AT2R
/y
mice were lyophilized to dry weight,
hydrolyzed in 6 mmol/l hydrochloric acid, and assessed for hydro-
xyproline content, as described previously.12,19,37 Hydroxyproline
values were then multiplied by a factor of 6.94 to extrapolate total
collagen content (as hydroxyproline represents B14.4% of the
amino-acid composition of collagen in most mammalian tissues),44
which was then divided by the dry weight tissue to yield collagen
concentration (% collagen content/dry weight tissue; corrected for
the size of the tissue portion analyzed).
Immunohistochemistry
Renal collagen IV and collagen I were identified from paraffin-
embedded kidney sections from each of the animals studied, using
goat anti-collagen IV (Southern Biotechnology, Birmingham, AL)
and rabbit anti-collagen I (Biodesign International, Saco, ME)
primary antibodies, respectively.37 Binding was visualized with the
avidin–biotin complex (ABC Elite; Vector, Burlingame, CA) and
3,30-diaminobenzidine (Sigma-Aldrich), and was morphometrically
assessed using point-counting.37 Using an eye-piece graticule, a
minimum of 10 fields at 20 original magnification were counted,
with results being expressed as a percentage of points with positive
immunohistochemical staining. Equivalent areas of renal cortex
were assessed in each animal, with glomeruli and large vessels
excluded from analysis.
BRET assays
BRET experiments were performed in HEK293 cells as described
previously.45–48 For BRET saturation assays,48 cells were co-
transfected with a constant amount of Rluc8-tagged AT2R
receptors and increasing amounts of Venus-tagged RXFP1
receptors. The expression levels of Rluc8- and Venus-tagged
constructs for each BRET experiment (n¼3 independent
experiments) were detected by luminescence (LUMIstar; BMG
Labtech, Mornington, VIC, Australia) and fluorescence (Envision;
Perkin-Elmer, Waltham, MA) measurements, respectively. The
actual Receptor-Venus/Receptor-Rluc8 expression ratios were then
plotted.
For real-time kinetic BRET assays,45–47 cells were assayed before
and after treatment with H2 relaxin (30 nmol/l) and/or Ang II
(1 mmol/l) or vehicle (phenol red-free þ10% FBS þ0.01% bovine
serum albumin) for each Receptor-Venus/Receptor-Rluc8 expression
ratio. The ligand-induced BRET signal was calculated by subtracting
the ratio of emission through the ‘acceptor wavelength window’ over
emission through the ‘donor wavelength window’ for a vehicle-
treated cell sample from the same ratio for a second aliquot of the
same cells treated with agonist, as described previously.45–47
Competition-binding assays
Whole-cell competition binding assays were conducted in HEK293
cells stably transfected with the AT2R.49–51 The unlabeled ligands
used were as follows: CGP42112 (AT2R agonist), candesartan (AT1R
antagonist), and H2 relaxin at concentrations ranging from 1pmol/l
to 1 mmol/l, which were prepared in Dulbecco’s modified Eagle’s
medium containing 0.1% bovine serum albumin (Dulbecco’s
modified Eagle’s medium þ0.1% bovine serum albumin)
(DMEM þ0.1% BSA) on the day of the experiment. For each
experiment, each ligand concentration was tested in triplicate. The
ability of each ligand to inhibit specific binding of [
125
I]Sar
1
Ile
8
Ang
II was assessed. Nonlinear regression of the data was performed
using GraphPad Prism 5.03 (GraphPad Software, San Diego, CA).
AlphaScreen pERK1/2 accumulation assays
Parental HEK293 cells and HEK293 cells expressing RXFP1 (HEK-
RXFP1) were transiently transfected with AT2Rs using polyethyle-
nimine at a ratio of 1:6 (DNA:polyethylenimine). Receptor-
mediated ERK1/2 phosphorylation (pERK1/2) was determined
48 h after transfection using the ERK1/2 SureFire kit (TGR
Biosciences, Hindmarsh, Australia) according to the manufacturer’s
instructions.24 Briefly, cells were seeded at 30,000 cells per well and
allowed to grow overnight at 37 1Cin5%CO
2. Before stimulation,
the cells were serum-starved for 6 h. The cells were stimulated with
vehicle (20 mmol/l sodium acetate buffer, pH 5.0) or relaxin
(30 nmol/l) in the absence or presence of PD123319 (1 mmol/l)
over 60 min to determine the time course of pERK1/2. Cells exposed
to PD123319 were pretreated with the AT2R antagonist for 30 min.
Data were normalized to the maximal response elicited by 10% FBS,
determined at 5 min.
Statistical analysis
Results were analyzed by one-way analysis of variance followed by
the Newman–Kuels post hoc test for multiple comparisons between
groups, using GraphPad Prism 5.03 (GraphPad Software). All data
are expressed as the mean±standard error of the mean (s.e.m.),
with a value of Po0.05 regarded as statistically significant.
DISCLOSURE
All the authors declared no competing interests.
ACKNOWLEDGMENTS
We thank Ms Chongxin Zhao for technical support. We are grateful to
Professor Lutz Hein (University of Freiburg, Germany) for providing
the AT2R knockout mice; A/Professor Kevin Pfleger (Western
Australian Institute for Medical Research, Perth, WA, Australia), Drs
Andreas Loening and Sanjiv Gambhir (Stanford University, Palo Alto,
CA), and Dr Atsushi Miyawaki (RIKEN Brain Science Institute, Wakocity,
Japan) for providing cDNA constructs (for the BRET assays); and
Professor Walter Thomas (University of Queensland) for providing the
HEK293 cells stably transfected with AT2Rs. This study was supported
by a University of Melbourne Fee Remission Scholarship to BSMC; an
Australian Research Council Postgraduate Scholarship to MS; National
Health and Medical Research Council of Australia (NHMRC) Project
Grants 436713, 454375, 628634, and APP1045848; NHMRC Senior
Research Fellowships to RADB and CSS; a Monash University Mid-
Career Fellowship to CSS; and by the Victorian Government’s
Operational Infrastructure Support Program.
SUPPLEMENTARY MATERIAL
Figures S1. Full Western blots of TGF-b1, pSmad2, Smad2, a-SMA
and a-tubulin.
Supplementary material is linked to the online version of the paper at
http://www.nature.com/ki
REFERENCES
1. Eddy AA. Progression in chronic kidney disease. Adv Chronic Kidney Dis
2005; 12: 353–365.
2. Wynn T. Cellular and molecular mechanisms of fibrosis. J Pathol 2008;
214: 199–210.
10 Kidney International
basic research BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers
3. Eddy AA. Molecular insights into renal interstitial fibrosis. J Am Soc
Nephrol 1996; 7: 2495–2508.
4. Mezzano SA, Ruiz-Ortega M, Egido J. Angiotensin II and renal fibrosis.
Hypertension 2001; 38: 635–638.
5. Stahl PJ, Felsen D. Transforming growth factor-b, basement membrane,
and epithelial–mesenchymal transdifferentiation. Am J Pathol 2001; 159:
1187–1192.
6. Bathgate RAD, Hsueh AJ, Sherwood OD. Physiology and molecular
biology of the relaxin peptide family. In: Neill JD (eds). Knobil and Neill’s
Physiology of Reproduction, 3rd edn Raven Press: New York, NY, 2006,
pp 679–770.
7. Samuel CS, Hewitson TD, Unemori EN et al. Drugs of the future: the
hormone relaxin. Cell Mol Life Sci 2007; 64: 1539–1557.
8. Samuel CS, Hewitson TD. Relaxin and the progression of kidney disease.
Curr Opin Nephrol Hypertens 2009; 18: 9–14.
9. Bennett RG. Relaxin and its role in the development and treatment of
fibrosis. Transl Res 2009; 154:1–6.
10. Teerlink JR, Cotter G, Davison BA et al. Serelaxin, recombinant human
relaxin-2, for treatment of acute heart failure (RELAX-AHF): a randomised,
placebo-controlled trial. Lancet 2013; 381: 29–39.
11. Heeg MH, Koziolek MJ, Vasko R et al. The antifibrotic effects of relaxin in
human renal fibroblasts are mediated in part by inhibition of the Smad2
pathway. Kidney Int 2005; 68: 96–109.
12. Mookerjee I, Hewitson TD, Halls ML et al. Relaxin inhibits renal
myofibroblast differentiation via RXFP1, the nitric oxide pathway, and
Smad2. FASEB J 2009; 23: 1219–1229.
13. Chow BS, Chew EG, Zhao C et al. Relaxin signals through a RXFP1-pERK-
nNOS-NO-cGMP-dependent pathway to up-regulate matrix
metalloproteinases: the additional involvement of iNOS. PLoS One 2012;
7: e42714.
14. Li JH, Zhu HJ, Huang XR et al. Smad7 inhibits fibrotic effect of TGF-beta
on renal tubular epithelial cells by blocking Smad2 activation. J Am Soc
Nephrol 2002; 13: 1464–1472.
15. Zhang HY, Phan SH. Inhibition of myofibroblast apoptosis by
transforming growth factor beta(1). Am J Respir Cell Mol Biol 1999; 21:
658–665.
16. Unemori EN, Amento EP. Relaxin modulates synthesis and secretion of
procollagenase and collagen by human dermal fibroblasts. J Biol Chem
1990; 265: 106981–110685.
17. Unemori EN, Pickford LB, Salles AL et al. Relaxin induces an extracellular
matrix-degrading phenotype in human lung fibroblasts in vitro and
inhibits lung fibrosis in a murine model in vivo.J Clin Invest 1996; 98:
2739–2745.
18. Bennett RG, Kharbanda KK, Tuma DJ. Inhibition of markers of hepatic
stellate cell activation by the hormone relaxin. Biochem Pharmacol 2003;
66: 867–874.
19. Samuel CS, Unemori EN, Mookerjee I et al. Relaxin modulates cardiac
fibroblast proliferation, differentiation, and collagen production and
reverses cardiac fibrosis in vivo.Endocrinology 2004; 145: 4125–4133.
20. Border WA, Noble NA. Interactions of transforming growth factor-beta
and angiotensin II in renal fibrosis. Hypertension 1998; 31: 181–188.
21. Ru
¨ster C, Wolf G. Angiotensin II as a morphogenic cytokine stimulating
renal fibrogenesis. J Am Soc Nephrol 2011; 22: 1189–1199.
22. Jones ES, Vinh A, McCarthy CA et al. AT2 receptors: functional relevance
in cardiovascular disease. Pharm Ther 2008; 120: 292–316.
23. Carey RM. Cardiovascular and renal regulation by the angiotensin type 2
receptor. Hypertension 2005; 45: 840–844.
24. Van der Westhuizen ET, Werry TD, Sexton PM et al. The relaxin family
peptide receptor 3 activates extracellular signal-regulated kinase 1/2
through a protein kinase C-dependent mechanism. Mol Pharmacol 2007;
71: 1618–1629.
25. Halls ML, Bathgate RA, Summers RJ. Relaxin family peptide receptors
RXFP1 and RXFP2 modulate cAMP signaling by distinct mechanisms. Mol
Pharmacol 2006; 70: 214–226.
26. Siragy HM, Carey RM. The subtype 2 (AT2) angiotensin receptor mediates
renal production of nitric oxide in conscious rats. J Clin Invest 1997; 100:
264–269.
27. Matsubara H. Pathophysiological role of angiotensin II type 2 receptor in
cardiovascular and renal diseases. Circ Res 1998; 83: 1182–1191.
28. Nie W, Yan H, Li S et al. Angiotensin-(1–7) enhances angiotensin II induced
phosphorylation of ERK1/2 in mouse bone marrow-derived dendritic
cells. Mol Immunol 2009; 46: 335–361.
29. Savoia C, Ebrahimian T, He Y et al. Angiotensin II/AT2 receptor-induced
vasodilation in stroke-prone spontaneously hypertensive rats involves
nitric oxide and cGMP-dependent protein kinase. J Hypertens 2006; 24:
2417–2422.
30. Yang J, Chen C, Ren H et al. Angiotensin II AT(2) receptor decreases AT(1)
receptor expression and function via nitric oxide/cGMP/Sp1 in renal
proximal tubule cells from Wistar–Kyoto rats. J Hypertens 2012; 30:
1176–1184.
31. Nakajima M, Hutchinson HG, Fujinaga M et al. The angiotensin II type 2
(AT2) receptor antagonizes the growth effects of the AT1 receptor: gain-
of-function study using gene transfer. Proc Natl Acad Sci USA 1995; 92:
10663–10667.
32. Sasser JM, Molnar M, Baylis C. Relaxin ameliorates hypertension and
increases nitric oxide metabolite excretion in angiotensin II but not
N
o
-nitro-L-arginine methyl ester hypertensive rats. Hypertension 2011; 58:
197–204.
33. Szabo AJ, Wagner L, Erdely A et al. Renal neuronal nitric oxide synthase
protein expression as a marker of renal injury. Kidney Int 2003; 64:
1765–1771.
34. Ma J, Nishimura H, Fogo A et al. Accelerated fibrosis and collagen
deposition develop in the renal interstitium of angiotensin type 2
receptor null mutant mice during ureteral obstruction. Kidney Int 1998;
53: 937–944.
35. Svendsen AM, Zalesko A, Knig J et al. Negative cooperativity in H2
relaxin binding to a dimeric relaxin family peptide receptor 1. Mol Cell
Endocrinol 2008; 296: 10–17.
36. Siddiquee K, Hampton J, McAnally D et al. The apelin receptor inhibits the
angiotensin II type 1 receptor via allosteric trans-inhibition. Br J
Pharmacol 2013; 168: 1104–1117.
37. Hewitson TD, Ho WY, Samuel CS. Anti-fibrotic properties of relaxin: in vivo
mechanism of action in experimental renal tubulointerstitial fibrosis.
Endocrinology 2010; 151: 4938–4948.
38. Diamond JR, Ricardo SD, Klahr S. Mechanisms of interstitial fibrosis in
obstructive nephropathy. Semin Nephrol 1998; 18: 594–602.
39. Hein L, Barsh GS, Pratt RE et al. Behavioural and cardiovascular effects of
disrupting the angiotensin II type-2 receptor in mice. Nature 1995; 377:
744–777.
40. Rodemann HP, Muller GA. Abnormal growth and clonal proliferation of
fibroblasts derived from kidneys with interstitial fibrosis. Proc Soc Exp Biol
Med 1990; 195: 57–63.
41. Masterson R, Hewitson TD, Kelynack K et al. Relaxin down-regulates renal
fibroblast function and promotes matrix remodelling in vitro.Nephrol Dial
Transplant 2004; 19: 544–552.
42. Samuel CS, Zhao C, Bathgate RA et al. Relaxin deficiency in mice is
associated with an age-related progression of pulmonary fibrosis. FASEB J
2003; 17: 121–123.
43. Gemici B, Tan R, Ongu
¨tGet al. Expressions of inducible nitric oxide
synthase and cyclooxygenase-2 in gastric ischemia–reperfusion: role of
angiotensin II. J Surg Res 2010; 161: 126–133.
44. Gallop PM, Paz MA. Posttranslational protein modifications, with special
attention to collagen and elastin. Physiol Rev 1975; 55: 418–487.
45. Kocan M, See HB, Seeber RM et al. Demonstration of improvements to
the bioluminescence resonance energy transfer (BRET) technology for the
monitoring of G protein-coupled receptors in live cells. J Biomol Screen
2008; 13: 888–898.
46. Kocan M, See HB, Sampaio NG et al. Agonist-independent interactions
between beta-arrestins and mutant vasopressin type II receptors
associated with nephrogenic syndrome of inappropriate antidiuresis. Mol
Endocrinol 2009; 23: 559–571.
47. Kocan M, Dalrymple MB, Seeber RM et al. Enhanced BRET technology for
the monitoring of agonist-induced and agonist-independent interactions
between GPCRs and beta-arrestins. Front Endocrinol (Lausanne) 2010; 1:12.
48. Mercier JF, Salahpour A, Angers S et al. Quantitative assessment of beta 1-
and beta 2-adrenergic receptor homo- and heterodimerization by
bioluminescence resonance energy transfer. J Biol Chem 2002; 277:
44925–44931.
49. D’amore A, Black MJ, Thomas WG. The angiotensin II type 2 receptor
causes constitutive growth of cardiomyocytes and does not antagonize
angiotensin II type 1 receptor-mediated hypertrophy. Hypertension 2005;
46: 1347–1354.
50. Thomas WG, Motel TJ, Kule CE et al. Phosphorylation of the angiotensin II
(AT(1A)) receptor carboxyl terminus: a role in receptor endocytosis. Mol
Endocrinol 1998; 12: 1513–1524.
51. Bosnyak S, Jones ES, Christopoulos A et al. Relative affinity of angiotensin
peptides and novel ligands at AT(1) and AT(2) receptors. Clin Sci 2011;
121: 297–303.
Kidney International 11
BSM Chow et al.: Relaxin signals through RXFP1-AT2R dimers basic research
