Farnesoid X receptor is essential for the survival of
renal medullary collecting duct cells under
hypertonic stress
Sujuan Xua,1, Shizheng Huangb,1, Zhilin Luana,1, Tingyue Chena, Yuanyi Weia, Miaomiao Xinga, Yaqing Lia, Chunxiu Dua,
Bing Wanga, Feng Zhenga, Nanping Wanga,c, Youfei Guana,c,2, Jan-Åke Gustafssond,e,2, and Xiaoyan Zhanga,c,2
a
Advanced Institute for Medical Sciences, Dalian Medical University, Dalian, 116044 Liaoning, China; bRenal Electrolyte and Hypertension Division,
Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104; cDepartment of Physiology and Pathophysiology, School of Basic Medical
Sciences, Dalian Medical University, Dalian, 116044 Liaoning, China; dCenter for Nuclear Receptors and Cell Signaling, Department of Biology and
Biochemistry, University of Houston, Houston, TX 77204; and eCenter for Biosciences, Department of Biosciences and Nutrition, Karolinska Institutet,
Novum, 14186 Stockholm, Sweden
Contributed by Jan-Åke Gustafsson, April 10, 2018 (sent for review March 6, 2018; reviewed by Chuanming Hao and Baoxue Yang)
Hypertonicity in renal medulla is critical for the kidney to produce
concentrated urine. Renal medullary cells have to survive high
medullary osmolarity during antidiuresis. Previous study reported
that farnesoid X receptor (FXR), a nuclear receptor transcription
factor activated by endogenous bile acids, increases urine concen-
trating ability by up-regulating aquaporin 2 expression in medul-
lary collecting duct cells (MCDs). However, whether FXR is also
involved in the maintenance of cell survival of MCDs under
dehydration condition and hypertonic stress remains largely un-
known. In the present study, we demonstrate that 24-hours water
restriction selectively up-regulated renal medullary expression of
FXR with little MCD apoptosis in wild-type mice. In contrast, water
deprivation caused a massive apoptosis of MCDs in both global
FXR gene-deficient mice and collecting duct-specific FXR knockout
mice. In vitro studies showed that hypertonicity significantly
increased FXR and tonicity response enhancer binding protein
(TonEBP) expression in mIMCD3 cell line and primary cultured
MCDs. Activation and overexpression of FXR markedly increased
cell viability and decreased cell apoptosis under hyperosmotic
conditions. In addition, FXR can increase gene expression and
nuclear translocation of TonEBP. We conclude that FXR protects
MCDs from hypertonicity-induced cell injury very likely via in-
creasing TonEBP expression and nuclear translocation. This study
provides insights into the molecular mechanism by which FXR en-
hances urine concentration via maintaining cell viability of MCDs
under hyperosmotic condition.
bile acid receptor | NFAT5 | hypertonicity | osmoprotection | cell viability
W hole-body water homeostasis is critical for human health
and depends on the balance between water intake stimu-
lated by thirst and water excretion controlled mainly by the
kidneys. Each day, an adult human produces ∼1.5 L of urine
despite 180 L of fluid filtered through glomerular basement
membrane. Approximately, 90% of glomerular filtrate is con-
stitutively reabsorbed in the proximal tubules and descending
loop of Henle. The remaining filtered water is reabsorbed in
renal medullary collecting duct cells (MCDs), where the water
transport is tightly regulated by arginine vasopressin (AVP), a
circulating hormone also known as antidiuretic hormone
(ADH) (1).
Under dehydration condition, AVP is produced by the hypo-
thalamus and released from the neurohypophysis into the blood
in response to increased plasma osmolarity or reduced blood
volume. It promotes water reabsorption in MCDs by increasing
gene expression and apical membrane targeting of aquaporin 2
(AQP2) mainly through the AVP-V2R-PKA pathway (2). How-
ever, AVP also increases sodium chloride (NaCl) reabsorption in
the thick ascending limb (TAL) and distal convolute tubule and
urea permeability in the inner MCDs to establish a hyperosmotic
5600–5605 | PNAS | May 22, 2018 | vol. 115 | no. 21
state in renal medulla, which is essential for the generation of a
concentrated urine (3).
Renal medulla is a unique tissue in which residing cells in-
cluding MCDs are exposed to the harsh hypertonic and hypoxic
environment and have to survive significant rises in NaCl and
urea concentrations during antidiuresis. Multiple osmoprotective
mechanisms have been reported to be important in maintaining
the survival of renal medullary cells under hypertonic stress (4–
7). Among them, it is generally believed that the transcription
factor tonicity-responsive enhancer binding protein (TonEBP)
and its target osmoprotective genes including aldose reductase
(AR) and heat shock protein 70 (HSP70) represent the most
important protective mechanism.
Farnesoid X receptor (FXR), a member of the nuclear re-
ceptor superfamily, is a transcription factor activated by endog-
enous bile acids. In addition to the liver and small intestine
where FXR plays an important role in bile acid, glucose, and
lipid metabolism, the kidney also exhibits highly abundant FXR
expression, especially in renal collecting duct cells (8). It has
been previously reported that activation of FXR increases urine
concentration by up-regulating the expression of AQP2, one of
its direct target genes, in MCDs (9). However, whether FXR is
Significance
As a ligand-activated transcription factor, farnesoid X receptor
(FXR) is abundantly expressed in the liver and small intestine,
where it plays an important role in the maintenance of bile
acid, lipid, and glucose homeostasis. Increasing evidence is
emerging that FXR may be a critical regulatory factor in renal
physiology and pathophysiology. The present study demon-
strates an essential role of FXR in promoting cell survival of
medullary collecting duct cells under hypertonic stress and
identifies FXR-induced TonEBP expression and activation as an
underlying mechanism. This work provides an insight into the
role of FXR in renal urine concentration.
Author contributions: Y.G., J.-Å.G., and X.Z. designed research; S.X., S.H., Z.L., T.C., Y.W.,
M.X., Y.L., C.D., and B.W. performed research; N.W., Y.G., and J.-Å.G. contributed new
reagents/analytic tools; F.Z. and X.Z. analyzed data; and Y.G., J.-Å.G., and X.Z. wrote
the paper.
Reviewers: C.H., Fudan University; and B.Y., Peking University.
The authors declare no conflict of interest.
Published under the PNAS license.
1
S.X., S.H., and Z.L. contributed equally to this work.
2
To whom correspondence may be addressed. Email: guanyf@dmu.edu.cn, jgustafs@
central.uh.edu, or zhangxy@dmu.edu.cn.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
1073/pnas.1803945115/-/DCSupplemental.
Published online May 8, 2018.
www.pnas.org/cgi/doi/10.1073/pnas.1803945115
also involved in the maintenance of cell survival of MCDs under
hypertonic stress remains largely unknown. The present studies
aimed to investigate whether FXR plays an important role in
regulating cell survival of MCDs under hyperosmotic stress
in vitro and in vivo. The underlying mechanism by which FXR
affects cell viability of MCDs exposed to hypertonic condition
was also examined.
Results
Water Restriction Selectively Up-Regulated FXR Expression in Mouse
Renal Medulla. Wild-type (WT) C57BL/6 mice were subjected to
water deprivation (WD) for 24 h. Compared with the mice with
free access to water, water-deprived mice exhibited ∼10-fold and
∼2-fold increase in FXR expression in renal medulla at mRNA and
protein level, respectively (Fig. 1). In contrast to renal medulla,
cortical expression of FXR at both mRNA and protein level
remained unaltered (Fig. 1). Because renal medullas of dehydrated
mice are hypertonic, these results suggest FXR may be regulated by
hypertonicity and plays an important role in urine concentration.
Hypertonicity Induced FXR Expression in Cultured MCDs. Since FXR
was mainly expressed in MCDs (9), we first determined whether
hyperosmolarity can induce FXR expression. mIMCD3 cell line
Fig. 1. Water restriction selectively induced FXR expression in renal me-
dullas of C57BL/6 mice. Male C57BL/6 mice were water-deprived for 24 h. (A)
FXR mRNA expression was selectively induced in renal medulla. FXR mRNA
levels were analyzed by qRT-PCR, *P < 0.05 vs. control (n = 4). (B) FXR protein
expression was significantly increased in renal medulla as assessed by
Western blot. Lamin A/C was used as protein loading control. (C) Statistical
analysis of Western blot shown in B, **P < 0.01 vs. control (n = 4).
Xu et al.
and primary cultured mouse MCDs were exposed to hyperto-
nicity. As shown in Fig. 2, FXR was constitutively and func-
tionally expressed in mIMCD3 cells (Fig. 2 A and B), where its
expression was significantly induced by hypertonicity at mRNA
(Fig. 2C) and protein levels (Fig. 2D). Similarly, in primary
cultured mouse MCD cells, FXR was constitutively expressed (SI
Appendix, Fig. S1) and increased hypertonicity significantly up-
regulated FXR expression at mRNA (SI Appendix, Fig. S2A) and
protein (SI Appendix, Fig. S2 B and C) levels. Compared with the
levels of FXR at 800 mOsm, FXR protein expression at
900 mOsm was markedly reduced, which might be due to re-
duced cell viability of MCD cells treated with 900 mOsm (SI
Appendix, Fig. S2B).
FXR Activation and Overexpression Protected mIMCD3 Cells from
Hypertonicity-Induced Cell Death. To investigate the role of FXR
in cell survival under hypertonic stress, mIMCD3 cells were
treated with an endogenous agonist (CDCA) and a synthetic
agonist (GW4064) of FXR or infected with an adenovirus
expressing FXRα2 (9). Cells were treated with either isotonic
(300 mOsm) or hyperosmotic medium (600 mOsm), and cell
viability was determined by the MTT assay. FXR activation by
both CDCA and GW4064 significantly reduced cell death in-
duced by hypertonic stress, with little effect on cell viability un-
der the isotonic condition (Fig. 3A). Similarly, overexpression of
FXR with Ad-FXRα2 markedly promoted cell survival under
hyperosmotic challenge (Fig. 3B). Flow cytometry analysis was
further performed to examine hypertonicity-induced apoptosis of
mIMCD3 cells. FXR activation by CDCA and GW4064 signifi-
cantly reduced hypertonicity (600 mOsm)-induced cell apoptosis
(Fig. 3C). Consistently, CDCA and GW4064 treatment blocked
hypertonicity-induced cleavage of procaspase-3 (Fig. 3D). These
PHYSIOLOGY
findings suggest that FXR may promote cell survival of MCDs
under dehydration state possibly by blocking caspase-3 activation-
associated apoptosis.
FXR Gene Deficiency Potentiated MCD Cell Apoptosis in Dehydrated
Mice in Vivo. To address whether FXR promotes the survival of
MCDs under dehydrated condition, WT, global FXR gene-
deficient mice (FXR−/−), and collecting duct-specific FXR gene
knockout mice (CD-FXR−/−) were generated and were deprived
of water for 24 h. As shown in Fig. 4 A and B, the TUNEL assay
demonstrated that water restriction induced a small amount
apoptosis of MCDs in WT mice but resulted in a massive apo-
ptosis in FXR−/− mice. FXR−/− mice exhibited much more ap-
optotic cells than WT mice after 24-h water restriction. The
majority of the apoptotic cells were the MCDs (Fig. 4 A and B).
To further confirm the role of collecting duct FXR, collecting
duct-specific FXR gene knockout mice (CD-FXR−/−) were
generated by crossing FXR flox/flox mice with AQP2-Cre mice
as previously reported (10–13) and were subjected to 24-h WD.
Compared with WT mice, CD-FXR−/− mice also displayed a signif-
icant increase in apoptotic cells of medullary collecting ducts after
water restriction (SI Appendix, Fig. S3). Together, these findings
clearly demonstrate that FXR can protect MCDs from hypertonicity-
induced injury and promote their survival under hypertonic stress.
FXR Gene Knockout Mice Exhibited Attenuated Renal TonEBP Expression
After 24-h Water Restriction. Tonicity-responsive enhancer binding
protein (TonEBP) is a transcription factor promoting cellular ac-
cumulation of organic osmolytes in the hypertonic renal medulla by
stimulating expression of its target genes including aldose reductase
(AR) and heat shock protein 70 (HSP70) (7). To elucidate the
underlying mechanism by which FXR protects MCDs from hyper-
tonic stress, renal expression of TonEBP and its target genes AR
and HSP70 was determined. As expected, 24-h dehydration
markedly increased renal TonEBP expression in WT mice at
both mRNA (Fig. 4C) and protein level (Fig. 4D). However,
PNAS | May 22, 2018 | vol. 115 | no. 21 | 5601
 Fig. 2. Hypertonicity-induced FXR expression in
mouse MCDs. (A) Constitutive expression of FXR in
mouse inner MCDs (mIMCD3). Cultured mIMCD3 cells
were transfected with an EGFP-AQP2 expression vec-
tor and then exposed to hypertonic stress (600 mOsm)
for 6 h. Note that FXR was mainly expressed in the
nucleus (red) and AQP2 was predominantly located in
cell membrane (green). (Scale bar, 25 uM.) (B) Lucif-
erase reporter assay demonstrating that hypertonicity
at 600 mOsm significantly induced FXR transcrip-
tion activity. ***P < 0.001 vs. 300 mOsm (n = 12).
(C) Induction of FXR mRNA expression by hyper-
tonicity in a dose-dependent manner. Cultured
mIMCD3 cells were incubated with hypertonic so-
lutions for 6 h. *P < 0.05 vs. 300 mOsm (n = 4). (D)
Western blot assay showing that hypertonicity
significantly induced FXR protein expression in
mIMCD3 cells. Cells were exposed to various hy-
pertonic stress (400, 500, and 600 mOsm) for 6 h.

in FXR − /− mice, 24-h water restriction just slightly induced
TonEBP mRNA and protein expression (Fig. 4 C and D).
These results suggest that FXR may exert osmoprotective
effect on MCDs via increasing TonEBP expression and
transcriptional activity in renal medulla.
FXR Induced TonEBP Expression and Nuclear Translocation in Cultured
MCDs. To confirm the effect of FXR on TonEBP expression and
activity, primary cultured mouse MCDs were treated with two
FXR agonists or infected with Ad-FXRα2. As shown in Fig. 5 A
and B, CDCA and GW4064 treatment markedly up-regulated

Fig. 3. FXR protected mIMCD3 cells from hypertonic stress-induced cell apoptosis. The mIMCD3 cells were subjected to hypertonicity (600 mOsm) for 6 h in
the presence or absence of FXR selective agonist GW4064 and CDCA. (A) MTT cell viability analysis showing that hypertonicity-induced cell death was sig-
nificantly attenuated by GW4064 and CDCA treatment. **P < 0.01 vs. 300 mOsm+DMSO, ##P < 0.01 vs. 600 mOsm+DMSO (n = 10). (B) Overexpression of FXR-
ameliorated mIMCD3 cell death under hypertonic stress. Cells were infected with Ad-FXRα2 or Ad-VP16. After 36 h of infection, the cells were challenged with
hypertonic stress (600 mOsm) for 6 h. Cell viability was determined by the MTT assay. ***P < 0.001 vs. 300 mOsm+Ad-VP16, ###P < 0.001 vs. 600 mOsm+Ad-
VP16 (n = 6). (C) FXR activation blocked hypertonicity-induced apoptosis of mIMCD3 cells. The mIMCD3 were treated with isotonic (300 mOsm) or hypertonic
medium (600 mOsm) for 6 h, and apoptotic cells were analyzed by the flow cytometry. Treatment of cells with GW4064 (2.5 μM) and CDCA (50 μM) markedly
reduced hypertonicity-induced cell apoptosis. ***P < 0.001 vs. 300 mOsm+DMSO, and #P < 0.05 and ##P < 0.01 vs. 600 mOsm+DMSO (n = 3). (D) FXR ac-
tivation with GW4064 and CDCA markedly reduced hypertonicity-induced activation of the caspase-3 apoptotic cascade. Note that protein expression of the
apoptotic marker, cleaved caspase-3, was markedly attenuated by GW4064 and CDCA. The results are representative of three independent experiments.

5602 | www.pnas.org/cgi/doi/10.1073/pnas.1803945115 Xu et al.

Fig. 4. FXR deficiency accelerated MCD cell apoptosis and attenuated TonEBP expression in dehydrated mice. C57BL/6 mice were water-deprived for 24 h.
Apoptotic cells were identified as dark brown staining by the TUNEL assay. Mice with free access to water were used as control. (A) TUNEL assay dem-
onstrating increased apoptotic cells were observed in the kidneys of FXR−/− mice with 24-h water restriction. Immunohistochemical staining of AQP2 (light
blue) was used to indicate renal-collecting ducts. Twenty-four-hour WD barely induced apoptosis of renal medullary cells in WT mice (FXR+/+) but resulted
in a massive apoptosis in FXR−/− mice (arrows). (Scale bar, 10 uM.) (B) Quantitative analysis of apoptotic cells in renal medullas of FXR+/+ and FXR−/− mice.
*P < 0.05 vs. FXR+/+ control mice, ##P < 0.01 vs. FXR+/+ with WD, &&&P < 0.001 vs. FXR−/− control mice (n = 3). (C and D) WD failed to induce TonEBP
expression at mRNA (C) and protein level (D) in the kidneys of FXR−/− mice. FXR +/+ and FXR−/− mice were water-deprived for 24 h. TonEBP mRNA and
protein expression was measured by real-time PCR and immunoblot, respectively. *P < 0.05 vs. FXR+/+ control mice, #P < 0.05 vs. FXR−/− with WD (n = 6).
CD, collecting duct.

TonEBP mRNA and protein expression. FXR activation also tissues with active cholesterol metabolism including the liver,

induced protein expression of FXR and TonEBP target genes
including AR and Hsp70 (Fig. 5B). Similarly, overexpression of
FXRα2 via an adenovirus-based approach increased TonEBP
expression at both mRNA and protein levels (Fig. 5 C and D).
Expression of FXR as well as AR and Hsp70 was also signifi-
cantly induced (Fig. 5D). These findings demonstrate that FXR
can promote TonEBP expression at both mRNA and protein level.
It is clear that transcriptional activity of TonEBP depends on
its nuclear translocation (14, 15). To further test whether FXR
can increase TonEBP transcriptional activity, primary MCDs
were treated with the FXR agonist CDCA and GW4064 or
overexpressed with FXRα2. Both FXR activation and over-
expression markedly promoted TonEBP nuclear translocation
(Fig. 5 E and F). Importantly, in primary cultured MCDs from
FXR−/− mice, Ad-FXRα2 infection not only rescued FXR ex-
pression, but also caused a significant translocation of TonEBP
from the cytoplasm to the nucleus (SI Appendix, Fig. S4). To-
gether, these findings demonstrate that FXR indeed increases
TonEBP activity by enhancing its nuclear translocation.
Discussion
This study demonstrates a critical role of FXR in the survival of
MCDs under dehydration condition. Twenty-four-hour water
restriction selectively up-regulates renal medullary expression of
FXR with little MCD apoptosis in WT mice. In contrast, WD
causes a massive apoptosis of MCDs in both global FXR gene-
deficient mice and collecting duct-specific FXR knockout mice.
In cultured MCDs, activation and overexpression of FXR
markedly promote cell viability and increase TonEBP expression
and activity under hyperosmotic stress. FXR may represent a
factor protecting the MCDs from hypertonic stress-induced in-
jury via increasing TonEBP activity, thereby maintaining the ability
of renal medulla to concentrate urine under hydration status.
FXR belongs to the superfamily of nuclear receptor tran-
scription factors and can be activated by many naturally occurred
endogenous bile acids (16). It is abundantly expressed in the
PHYSIOLOGY
intestine, and adrenal gland, where FXR plays an important role
in bile acid biosynthesis and transport, cholesterol metabolism,
and glucocorticoid production (17–19). It has been previously
reported that FXR is also highly expressed in the kidney and is
critical for maintaining renal function (9). Disruption of FXR
gene causes a significant defect in urine concentrating ability,
resulting in a polyuria phenotype in mice. In addition, FXR
dysfunction is associated with a worsened renal injury in diabetic
mice (20). The present study further demonstrates that FXR
gene deficiency significantly increases cell death of MCDs under
hypertonic conditions, while FXR activation and overexpression
markedly reduces hypertonicity-induced apoptosis of MCDs.
These findings provide evidence that FXR is a key factor in pro-
moting cell survival of MCDs in hypertonic renal medulla, thereby
maintaining renal urine concentrating ability during antidiuresis.
TonEBP, also called nuclear factor of activated T-cell 5
(NFAT5), is a transcription factor critical for osmoadaptive re-
sponse of renal medullary cells to hyperosmolar stress (14). It
exerts an osmoprotective effect on renal medullary cells mainly
via up-regulating gene transcription of many of its target genes
including AR, HSP70, and aquaporins (AQPs) (15). Genetically
modified animals with deficient TonEBP activity in the kidney
suffer from severe medullary atrophy in association with cell
death, demonstrating that TonEBP is essential for the survival of
the renal medullary cells (7). Although it is well accepted that
TonEBP is a key transcription factor in osmoprotection, mech-
anisms involved in TonEBP expression and activation remain
largely unknown. In the present study, we found that renal
TonEBP expression was significantly reduced in FXR−/− mice
with WD, while FXR activation markedly up-regulated expres-
sion of TonEBP and its target genes. These findings raise a
possibility that FXR may be a transcription factor important for
driving TonEBP expression. However, the underlying mecha-
nism warrants further investigation.
The present study also provides clear evidence that FXR can
promote nuclear translocation of TonEBP protein in MCDs. It is

Xu et al. PNAS | May 22, 2018 | vol. 115 | no. 21 | 5603

Fig. 5. FXR up-regulated TonEBP expression and nuclear translocation in primary cultured mouse IMCD cells. Primary MCD cells were treated with GW4064
(2.5 μM) and CDCA (50 μM) for 6 h or infected with Ad-FXRα2. (A) Real-time PCR analysis demonstrating mRNA expression of TonEBP in primary MCD cells
isolated from WT mice. **P < 0.01, ***P < 0.001 vs. DMSO (n = 4). (B) Western blot assay showing that treatment of MCDs with GW4064 and CDCA markedly
induced protein expression of FXR, TonEBP, and osmoprotective genes HSP70 and AR. (C) Quantitative PCR analysis showing that overexpression of FXR
significantly up-regulated TonEBP mRNA levels in MCDs. *P < 0.05 vs. Ad-VP16 (n = 8). (D) Immunoblot assay demonstrating overexpression of FXR markedly
induced FXR, TonEBP, HSP70, and AR expression. (E) FXR activation promoted nuclear translocation of TonEBP. Primary MCDs were treated with GW4064 and
CDCA for 6 h. Confocal immunofluorescence was performed to detect TonEBP (red) nuclear translocation. The nucleus was visualized by staining with DAPI
(blue). (Scale bar, 5 uM.) (F) FXR overexpression with Ad-FXRα2 markedly facilitated nuclear translocation of TonEBP. Cells infected with a control virus (Ad-
VP16) had little effect on TonEBP nuclear localization. (Scale bar, 5 uM.)

well known that nuclear import of TonEBP is essential for its
transcriptional activity. Multiple mechanisms have been reported
to contribute to the nucleocytoplasmic shuttling of TonEBP
protein, which requires the presence of the nuclear export signal
and nuclear localization signal at its N terminus (21). The sig-
naling cascades of Fyn, a shrinkage-activated tyrosine kinase, and
p38, a subgroup of the mitogen-activated kinases, have been
reported to be major signaling pathways for hypertonic activation
of TonEBP (22). In contrast, transcriptional coactivator with
PDZ-binding motif (TAZ) can suppress TonEBP activity
through tyrosine phosphorylation (23). We report here that FXR
activation and overexpression can dramatically diminish cyto-
plasmic TonEBP levels and increase its nuclear localization,
providing a mechanism for TonEBP transactivation. Currently,
the mechanism by which FXR drives nuclear import of TonEBP
is not clear. One possibility worth addressing is that FXR may
form a transcriptional complex with TonEBP, which increases
both TonEBP nuclear translocation and its transactivation.
Given a ubiquitous distribution pattern of both FXR and
TonEBP, they may act in concert in regulating many other bi-
ological processes than those involved in the response to hy-
pertonicity in renal medulla.
Similar to TonEBP, FXR may also represent a hypertonicity-
responsive gene. Twenty-four-hour water restriction selectively up-
regulated FXR expression in renal medulla at both mRNA and
protein levels. Similarly, hypertonic stress significantly induced
FXR mRNA and protein levels and its transcriptional activity in
cultured MCDs. These findings demonstrate that FXR gene is
transcriptionally regulated in response to hypertonicity. However,
it is currently unclear whether there is a hypertonicity-responsive
element in the promoter region of FXR gene. It is also uncertain
whether FXR is under the transcriptional regulation of TonEBP.
Addressing these important issues may significantly advance our
knowledge in understanding the mechanism by which FXR is regu-
lated and the importance of FXR in renal physiology.
In summary, the present studies demonstrate that dehydration
and hypertonicity increase FXR expression and its transcrip-
tional activity in the MCDs. FXR gene deficiency significantly
potentiates MCD death in mice with water restriction. FXR
activation and overexpression markedly promote cell survival of
cultured MCDs. FXR-driven TonEBP expression and nuclear
translocation may contribute to its osmoprotective role in
renal medulla.

5604 | www.pnas.org/cgi/doi/10.1073/pnas.1803945115 Xu et al.

Materials and Methods
Male C57BL/6 WT mice (8-10 wk old) were purchased from the Experimental
Animal Center at Peking University Health Science Center. FXR knockout
mice were purchased from the Jackson Laboratory. Renal collecting duct-
specific FXR gene knockout mice were generated by crossing FXRflox/flox
mice with an aquaporin 2 gene promoter driven-Cre transgenic mouse and
validated as previously described (10–13). Eight- to 10-week-old male ani-
mals used in the study were housed on a daily 12-h light/black cycle under
controlled temperature and humidity in the animal facility of Peking Uni-
versity Health Science Center. All animals had free access to food. Mice were
allowed free access to water or subjected to water restriction for 24 h. After
water restriction, mice were killed and perfused with cold PBS. The left
kidneys were fixed and paraffin-embedded, while the right kidneys were
used to dissect renal cortex and medulla. The use of animals and the study
1. Knepper MA, Kwon TH, Nielsen S (2015) Molecular physiology of water balance. N
Engl J Med 372:1349–1358.
2. Li Y, Wei Y, Zheng F, Guan Y, Zhang X (2017) Prostaglandin E2 in the regulation of
water transport in renal collecting ducts. Int J Mol Sci 18:E2539.
3. Bachmann S, Mutig K (2017) Regulation of renal Na-(K)-Cl cotransporters by vaso-
pressin. Pflugers Arch 469:889–897.
4. Hao CM, et al. (2000) Dehydration activates an NF-kappaB-driven, COX2-dependent
survival mechanism in renal medullary interstitial cells. J Clin Invest 106:973–982.
5. Han Q, et al. (2011) AMPK potentiates hypertonicity-induced apoptosis by suppress-
ing NFκB/COX-2 in medullary interstitial cells. J Am Soc Nephrol 22:1897–1911.
6. Hao CM, Redha R, Morrow J, Breyer MD (2002) Peroxisome proliferator-activated
receptor delta activation promotes cell survival following hypertonic stress. J Biol
Chem 277:21341–21345.
7. Lee SD, et al. (2011) TonEBP stimulates multiple cellular pathways for adaptation to
hypertonic stress: Organic osmolyte-dependent and -independent pathways. Am J
Physiol Renal Physiol 300:F707–F715.
8. Zhang XY, Wang B, Guan YF (2016) Nuclear receptor regulation of aquaporin-2 in the
kidney. Int J Mol Sci 17:E1105.
9. Zhang X, et al. (2014) Farnesoid X receptor (FXR) gene deficiency impairs urine con-
centration in mice. Proc Natl Acad Sci USA 111:2277–2282.
10. Nelson RD, et al. (1998) Expression of an AQP2 Cre recombinase transgene in kidney
and male reproductive system of transgenic mice. Am J Physiol 275:C216–C226.
11. Sinal CJ, et al. (2000) Targeted disruption of the nuclear receptor FXR/BAR impairs bile
acid and lipid homeostasis. Cell 102:731–744.
12. Guan Y, et al. (2005) Thiazolidinediones expand body fluid volume through PPAR-
gamma stimulation of ENaC-mediated renal salt absorption. Nat Med 11:861–866.
Xu et al.
protocols were reviewed and approved by the Animal Care and Use Review
Committee of Dalian Medical University.
Chemicals and reagents, methods of cell viability assay, real-time PCR,
Western blot, immunostaining, flow cytometry, primary culture of mouse
IMCD cells, and luciferase assay, TUNEL assay, immunofluorescence
staining, and statistical analyses are described in SI Appendix, Materials
and Methods.
ACKNOWLEDGMENTS. We thank Dr. Frank Gonzalez for providing FXRflox/
flox mice. This work was supported by the National Natural Science
Foundation of China Grants 81570636, 81722010 (to X.Z.), 81390351, and
91639201 (to Y.G.); Dalian High-level Talent Innovation Support Pro-
gram 2016RD13; and Peacock Plan of Shenzhen city Grant KQTD20140630100746562.
J.Å.G. was supported by Robert A. Welch Foundation Grant E-0004.
13. Gao M, et al. (2015) Disruption of prostaglandin E2 receptor EP4 impairs urinary
concentration via decreasing aquaporin 2 in renal collecting ducts. Proc Natl Acad Sci
USA 112:8397–8402.
14. Ferraris JD, Burg MB (2006) Tonicity-dependent regulation of osmoprotective genes
in mammalian cells. Contrib Nephrol 152:125–141.
15. Burg MB, Ferraris JD, Dmitrieva NI (2007) Cellular response to hyperosmotic stresses.
Physiol Rev 87:1441–1474.
16. Shapiro H, Kolodziejczyk AA, Halstuch D, Elinav E (2018) Bile acids in glucose me-
tabolism in health and disease. J Exp Med 215:383–396.
17. Inagaki T, et al. (2005) Fibroblast growth factor 15 functions as an enterohepatic
signal to regulate bile acid homeostasis. Cell Metab 2:217–225.
18. Hoekstra M, et al. (2012) FXR agonist GW4064 increases plasma glucocorticoid levels
in C57BL/6 mice. Mol Cell Endocrinol 362:69–75.
19. Houten SM, Volle DH, Cummins CL, Mangelsdorf DJ, Auwerx J (2007) In vivo imaging
of farnesoid X receptor activity reveals the ileum as the primary bile acid signaling
tissue. Mol Endocrinol 21:1312–1323.
20. Herman-Edelstein M, Weinstein T, Levi M (2018) Bile acid receptors and the kidney.
Curr Opin Nephrol Hypertens 27:56–62.
21. Cheung CY, Ko BC (2013) NFAT5 in cellular adaptation to hypertonic stress–Regula-
tions and functional significance. J Mol Signal 8:5.
22. Ko BC, et al. (2002) Fyn and p38 signaling are both required for maximal hypertonic
PHYSIOLOGY
activation of the osmotic response element-binding protein/tonicity-responsive en-
hancer-binding protein (OREBP/TonEBP). J Biol Chem 277:46085–46092.
23. Jang EJ, et al. (2012) TAZ suppresses NFAT5 activity through tyrosine phosphorylation.
Mol Cell Biol 32:4925–4932.
PNAS | May 22, 2018 | vol. 115 | no. 21 | 5605