Indoxyl sulphate inhibits osteoclast differentiation and function.
ABSTRACT Patients with chronic kidney disease (CKD) develop various bone abnormalities characterized by impaired bone remodelling. Recent data suggest that accumulation of the uraemic toxin indoxyl sulphate (IS) may be one of the factors involved in bone abnormalities in CKD patients. Indeed, it was recently reported that IS induces skeletal resistance to parathyroid hormone in cultured osteoblastic cells. However, it is not yet known whether IS also affects osteoclast cells.
In the present study, we assessed the direct effect of IS at uraemic concentrations and in the presence (to reach the 3 mM concentration) or absence of added inorganic phosphate (Pi) on osteoclast (OCL) differentiation and bone-resorbing activity in two well-established cellular models of monocyte/macrophage (peripheral blood mononuclear cells and the RAW 264.7 cell line).
We found that IS inhibits both OCL differentiation and bone-resorbing activity in a dose-dependent manner and that these effects were enhanced in the presence of Pi at 3mM concentration. IS induced a gradual inhibition of JNK, Akt, p38, ERK1/2 phosphorylation and AP-1 DNA-binding activity. The effects of IS on OCL differentiation and AP-1 were prevented by probenecid, a competitive inhibitor of organic anion transporters, suggesting that IS's effects occur subsequently to its intake.
Our findings strongly suggest that IS not only inhibits osteoblast function but also has an inhibitory effect on OCL function and thus could affect bone remodelling in CKD patients.
- SourceAvailable from: Eric Devine[Show abstract] [Hide abstract]
ABSTRACT: Protein binding prevents uremic toxins from removal by conventional extracorporeal therapies leading to accumulation in maintenance dialysis patients. Weakening of the protein binding may enhance the dialytic elimination of these toxins. In ultrafiltration and equilibrium dialysis experiments, different measures to modify the plasma binding affinity and capacity were tested: (i), increasing the sodium chloride (NaCl) concentration to achieve a higher ionic strength; (ii), increasing the temperature; and (iii), dilution. The effects on the dissociation constant KD and the protein bound fraction of the prototypical uremic toxin indoxyl sulfate (IS) in plasma of healthy and uremic individuals were studied. Binding of IS corresponded to one site binding in normal plasma. KD increased linearly with the NaCl concentration between 0.15 (KD = 13.2 ± 3.7 µM) and 0.75 M (KD = 56.2 ± 2.0 µM). Plasma dilution further reduced the protein bound toxin fraction by lowering the protein binding capacity of the plasma. Higher temperatures also decreased the protein bound fraction of IS in human plasma. Increasing the NaCl concentration was effective to weaken the binding of IS also in uremic plasma: the protein bound fraction decreased from 89% ± 3% to 81% ± 3% at 0.15 and 0.75 M NaCl, respectively. Dilution and increasing the ionic strength and temperature enhance the free fraction of IS allowing better removal of the substance during dialysis. Applied during clinical dialysis, this may have beneficial effects on the long-term outcome of maintenance dialysis patients.Toxins 01/2014; 6(2):416-30. · 2.13 Impact Factor
- Jornal Brasileiro de Nefrologia 01/2014;
- [Show abstract] [Hide abstract]
ABSTRACT: Many small solutes excreted by the kidney are bound to plasma proteins, chiefly albumin, in the circulation. The combination of protein binding and tubular secretion allows the kidney to reduce the free, unbound concentrations of such solutes to lower levels than could be obtained by tubular secretion alone. Protein-bound solutes accumulate in the plasma when the kidneys fail, and the free, unbound levels of these solutes increase more than their total plasma levels owing to competition for binding sites on plasma proteins. Given the efficiency by which the kidney can clear protein-bound solutes, it is tempting to speculate that some compounds in this class are important uremic toxins. Studies to date have focused largely on two specific protein-bound solutes: indoxyl sulfate and p-cresyl sulfate. The largest body of evidence suggests that both of these compounds contribute to cardiovascular disease, and that indoxyl sulfate contributes to the progression of chronic kidney disease. Other protein-bound solutes have been investigated to a much lesser extent, and could in the future prove to be even more important uremic toxins.Seminars in Nephrology 03/2014; 34(2):106-117. · 2.83 Impact Factor
Nephrol Dial Transplant (2012) 27: 2176–2181
Advance Access publication 1 December 2011
Indoxyl sulphate inhibits osteoclast differentiation and function
Anaı ¨s Mozar1,2, Loı ¨c Louvet1,2, Corinne Godin1,2, Romuald Mentaverri1,2, Michel Brazier1,2,
Saı ¨d Kamel1,2and Ziad A. Massy1,2,3
1INSERM ERI-12 (EA4292), Amiens, France,2University of Picardie Jules Verne, Amiens, France and3Divisions of Clinical
Pharmacology and Nephrology, Amiens University Hospital, Amiens, France
Correspondence and offprint requests to: Ziad A. Massy; E-mail: firstname.lastname@example.org
Background. Patients with chronic kidney disease (CKD)
develop various bone abnormalities characterized by im-
paired bone remodelling. Recent data suggest that accumu-
lation of the uraemic toxin indoxyl sulphate (IS) may be
one of the factors involved in bone abnormalities in CKD
patients. Indeed, it was recently reported that IS induces
skeletal resistance to parathyroid hormone in cultured os-
teoblastic cells. However, it is not yet known whether IS
also affects osteoclast cells.
Methods. In the present study, we assessed the direct effect
of IS at uraemic concentrations and in the presence (to
reach the 3 mM concentration) or absence of added inor-
ganic phosphate (Pi) on osteoclast (OCL) differentiation
and bone-resorbing activity in two well-established cellular
models of monocyte/macrophage (peripheral blood mono-
nuclear cells and the RAW 264.7 cell line).
Results. We found that IS inhibits both OCL differentiation
and bone-resorbing activity in a dose-dependent manner
and that these effects were enhanced in the presence of Pi
at 3mM concentration. IS induced a gradual inhibition of
JNK, Akt, p38, ERK1/2 phosphorylation and AP-1 DNA-
binding activity. The effects of IS on OCL differentiation
and AP-1 were prevented by probenecid, a competitive
inhibitor of organic anion transporters, suggesting that
IS’s effects occur subsequently to its intake.
Conclusion. Our findings strongly suggest that IS not only
inhibits osteoblast function but also has an inhibitory effect
on OCL function and thus could affect bone remodelling in
Keywords: adynamic bone disease; chronic kidney disease; indoxyl
The uraemic toxin indoxyl sulphate (IS) is an organic anion
belonging to the family of protein-bound retention solutes.
It is synthesized in the liver from indole, which itself is
produced by the intestinal flora from tryptophan . It
has been demonstrated that IS modulates the activity of
many cell types and induces the production of reactive
oxygen species. IS has a suppressive effect on woundrepair
in human umbilical vein endothelial cells . The accumu-
lation of IS in renal proximal cells promotes cytotoxic ef-
fects and thus accelerates the progression of kidney failure
. Shimoishi et al.  showed that IS’s induction of sys-
temic vascular oxidative stress in rats could be prevented
by treatment with the oral absorbent AST-120 (Kreme-
zin?), which absorbs indole in the intestines. In another
in vivo study in a rat model of kidney failure with low bone
turnover, Iwasaki et al.  showed that administration of
AST-120 suppressed the IS-induced osteoblast cytotoxicity
and the decrease in bone formation.
Recently, Fukagawa et al. demonstrated in a primary
mouse calvariae osteoblast cell culture that addition of IS
suppresses parathyroid hormone (PTH)-stimulated intra-
cellular cAMP production, decreases PTH receptor ex-
pression and induces oxidative stress in these cells .
They suggested that IS accumulation in the blood of
CKD patients is at least one of the factors that induces
skeletal resistance to PTH implicated in the outcome of
adynamic bone disease (ABD) observed in patients with
end-stage renal disease . ABD is characterized by qui-
escent osteoblasts and osteoclasts (OCLs) and markedly
reduced bone turnover activity. Since 1979, uraemic tox-
ins have been proposed to play a role in low bone turnover
disease such as ABD . As mentioned above, recent data
suggest that accumulation of IS in the blood may be one
of the uraemic factors involved in ABD via its effects on
osteoblastic cells . Limited data suggest that uraemic
toxins may also play a role in OCL differentiation and
bone resorption. We recently observed in vitro that the
uraemic toxin, inorganic phosphate (Pi), directly inhibits
both osteoclastic differentiation and bone-resorbing activ-
ity induced by receptor activator of nuclear factor kappa
B ligand (RANKL) and macrophage colony stimulating
factor (M-CSF) . To date, the direct effects of IS on
multinucleated OCL differentiation and function have not
yet been evaluated. Hence, we decided in the present work
to assess the effects of uraemic concentrations of IS (as
defined by the EUTox work group) on OCL differentiation
and bone-resorbing activity.
? The Author 2011. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please e-mail: email@example.com
at New York University on November 28, 2012
Materials and methods
All chemicals were purchased from Sigma unless otherwise stated.
Uraemic toxin preparation
For our experiments, we referred to the list of uraemic toxins provided by
EUTox and thus worked with IS at a normal concentration (ISn: 0.6 mg/L,
i.e. 2 lM), a uraemic concentration (ISu: 53 mg/L, i.e. 211 lM) and the
maximum concentration found in uraemic patients (ISm: 236 mg/L, i.e.
940 lM) . All experiments in the presence of Pi were performed with
3 mM of NaH2PO4salt, taking into account the 1.1 mM present in the
alpha modified eagle’s medium (aMEM) medium supplemented with 10%
foetal calf serum (FCS).
Peripheral blood mononuclear cells and RAW 264.7 cell culture and
Peripheral blood mononuclear cells.
and added factors were replaced twice a week; cells were cultured for
(25 ng/mL) and human M-CSF (30 ng/mL) (R&D Systems). In order to
evaluate bone resorption activity, the cells were seeded on bovine bone
in . After each medium change, we pooled and stored supernatants from
each experimental conditionin orderto quantifytypeI collagenC-terminal
telopeptide (CTX) degradation products. In all experiments, we confirmed
that there was no differentiation in the absence of RANKL.
Peripheral blood mononuclear
RAW 264.7 cells.
(ATCC) routine culture was previously described in . For differentia-
tion into OCL-like cells, RAW 264.7 cells were gently scraped and seeded
in 96-well plates at a density of 103cells per well in aMEM differentiation
medium supplemented with 10% FCS and 30 ng/mL of recombinant mur-
ine RANKL (R&D Systems). The medium was replaced once and cells
were cultured for 5 days. In all experiments, we confirmed that there was
no differentiation in the absence of RANKL.
To evaluate the effects of IS on RANKL intracellular signalling, RAW
264.7 cells were seeded for 24 h at a density of 2 3 106cells per well in
aMEM. RANKL (100 ng/mL) was added for 10 min before preparation of
the total cell extracts and 30 min before preparation of the nuclear extracts
(50 ng/mL). IS was added 2 min before RANKL addition in order to reach
the indicated concentrations.
To block the effect of IS on differentiation, we used a range of concen-
trations of the organic anion transporter (OAT) inhibitor probenecid (from 1
to 100 lM). Cells were treated with probenecid for the last 2 days of
differentiation only because longer exposure was cytotoxic (data not
Correctness of total and ionized calcium concentrations in the super-
natants collected from the various experimental conditions was checked
with a RXL Dimension Dade-Behring auto-analyser and a Ciba-Corning
634 Ca11/pH analyser, respectively.
RAW 264.7 mouse monocyte/macrophage cells
Tartrate-resistant acid phosphatase staining
Osteoclasts differentiated from RAW cells or PBMCs were stained for
tartrate-resistant acid phosphatase (TRAP) using a commercially available
kit (leukocyte acid phosphatase staining kit 387A, Sigma) according to the
manufacturer’s protocol. Multinucleated (more than three nuclei) TRAP-
positive cells were counted as OCLs under microscopic examination.
Cell viability and specific controls
We used the water-soluble tetrazolium salt WST-1 (Roche Applied Sci-
ence) to measure cell proliferation and viability (as already described
in ). Briefly, after overnight starvation (in 0.5% FCS medium), RAW
264.7 cells seeded in 96-well plates at 5 3 103and 1 3 103cells per well
were cultured for 2 and 5 days, respectively, under different conditions.
We checked that the effects of IS were not due to the potassium ions in
the IS salt used for these experiments since KCl at 0.9 mM (the maximum
concentration used with IS) inhibited neither the differentiation of PBMC
progenitors into OCLs nor bone resorption after 2 weeks of culture (data
not shown). Likewise, we checked that the effects of IS were not due to
the presence of sulfate; we did not observe any effect of sodium sulphate
(at the various concentrations used for IS) on RAW 264.7 progenitor
differentiation (data not shown).
Since IS is primarily a protein-bound molecule, the free serum concen-
tration corresponds to a small fraction of the total serum concentration. To
test the role of protein-binding effects, we performed experiments with
physiological concentrations (40 g/L) of human serum albumin (HSA) in
murine monocyte RAW 264.7 model. Unfortunately, HSA did affect our
model by itself inducing the differentiation of RAW 264.7 cells into OCL-
like cells (data not shown). These phenomena prevented us from gaining a
clear picture of protein-binding effects.
Bone resorption activity measurement
Morphological assessment: pit area measurement.
resorption lacunae (pits) were visualized by staining bone slices with
haematoxylin red and a toluidine blue solution containing 1% sodium
borate. The percentage of resorbed bone surface area was quantified using
After OCL removal,
Quantitative assessment: measurement of type I collagen fragment
and fragments of type I collagen are released into the supernatant. We used
an enzyme-linked immunosorbent assay (ELISA) (CrossLaps?; Osteomedi-
cal) for the quantitative determination of degradation products of type I
collagen C-terminal telopeptides (CTX) in PBMC culture supernatants. The
level of CTX was assessed according to the manufacturer’s protocol.
Immunoblot assay of kinase phosphorylation
After stimulation, cells were washed once with phosphate-buffered saline
and lysed with 50 lL lysis buffer (25 nM Tris–HCl, pH 7.6, 150 nM NaCl,
10% glycerol, 1% Nonidet P-40, 2 mM Na3VO4and protease inhibitors)
(for 20 min and at 4?C) and protein-containing supernatants were collected.
Equal amounts of proteins (40 lg) were resolved by sodium dodecyl
sulphate–polyacrylamide gel electrophoresis, transferred to a nitrocellulose
membrane and immunoblotted with specific antibodies against phospho-
JNK, -Akt, -ERK1/2 and -p38 (Cell Signalling). Secondary antibodies were
obtained from Santa Cruz. Equal loading was confirmed using a b-actin
antibody. Immunoblots were visualized byanenhanced chemiluminescence
detection kit (Amersham). Protein expression levels were quantified using a
densitometer (Genegenius; Ozyme). The data were represented as the ratio
of phosphorylated proteins to that of b-actin. The value under control con-
ditions (i.e. in the absence of RANKL) was considered to be 1.
Determination of the DNA-binding activity of transcription factors
using the TransAM? ELISA
NF-jB and AP-1 DNA-binding activity assays were performed using the
Trans-AM? ELISA-based kits from Active Motif. Nuclear extracts were
prepared using the Active Motif kit according to the manufacturer’s proto-
col. Equal quantities of protein (10 lg) for each sample were then incubated
in 96-well plates coated with an oligonucleotide containing the NF-jB or
AP-1 consensus binding site. Activated transcription factors of the extracts
specifically bounded to the respective immobilized oligonucleotide were
detected using a primary antibody directed against p65 (NF-jB kit) or
c-Jun (AP-1 kit) and a secondary antibody conjugated to horseradish per-
oxidase. Developing solution was then added for 10 min at room temper-
ature and opticaldensities (450–655 nm) were determined with a microplate
reader (Molecular Devices). Results are expressed as a percentage of the
control condition (Ct, i.e. complete medium without RANKL).
All results are expressed as mean ? SEM. Statistical significance was
determined in an analysis of variance and a P-value <0.05 was considered
to be significant.
IS inhibits the RANKL-induced differentiation of RAW
264.7 cells into osteoclasts
As shown in Figure 1, the RANKL-induced differentiation
of RAW 264.7 cells into OCLs was dose dependently and
IS inhibits OCL differentiation2177
at New York University on November 28, 2012
significantly inhibited when IS was added to culture media
at ISu and ISm concentrations, when compared with the Ct
and normal concentration of IS (ISn). When 3 mM Pi was
added to ISm, we observed a marked inhibition of OCL
differentiation when compared with Pi alone. In order to
exclude the possibility that IS’s inhibitory effect was due to
a change in cell viability, we assessed the effect of IS on the
viability of these cells over 2 and 5 days (using WST-1
reagent). We did not see any change in RAW 264.7 pre-
cursor cell viability in the presence of IS, regardless of the
highest concentration used or the presence or absence of
3 mM Pi (data not shown).
IS reduces the RANKL-induced differentiation of PBMCs
IS’s effects on OCL differentiation were confirmed with
PBMC cell progenitors. IS significantly and dose depend-
ently inhibited the differentiation into active OCL cells
normally induced by RANKL and M-CSF alone (Ct) in
this cell model (Figure 2A and B). These effects were sig-
nificantly enhanced in the presence of 3 mM Pi.
IS reduces the bone-resorbing activity of PBMC-derived
After counting, human OCL cells were washed out of bone
slices. Bone resorption appeared as characteristic pits on
the surface of the bone slices. By evaluating the bone re-
sorption surface area under the various conditions, we
found that ISu and ISm concentrations significantly and
dose dependently decreased the bone-resorbing activity
of OCL cells compared with the Ct and ISn conditions
(Figure 3A). We found that effects of ISu and ISm were
enhanced in the presence of 3 mM Pi. The IS’s effects on
the bone-resorbing activity of OCL were also assessed by
measuring the amount of CTX in the supernatants. We
observed a dose-dependent decrease in the amounts of
CTX released under ISu and ISm conditions compared with
the medium-only and ISn conditions (Figure 3B). We did
not observe significantly greater inhibition when Pi was
added to ISu, relative to the effects of ISu alone. This
may be due to the high observed variability of CTX release
from one experiment to another.
OATs are directly involved in the effects of IS on OCL
In order to explore the molecular mechanism through
which IS might inhibit the differentiation of RAW 264.7
cells into OCLs, we assessed the role of OATs. By using
different concentrations of the OAT inhibitor probenecid
(1, 10 and 100 lM), we were able to block the inhibitory
effect exerted by ISu (Figure 4) and ISm (data not shown)
on OCL-like differentiation in a dose-dependent manner.
Cells cultured in the presence of uraemic or maximal con-
centrations of IS and 100 lM probenecid displayed a sim-
ilar differentiation rate to that seen in the Ct, whereas
probenecid alone did not modulate OCL differentiation.
These results suggest that IS exerts its effects after OAT-
mediated entry into cells.
Fig. 1. IS inhibits RANKL-induced differentiation of RAW 264.7 cells
into OCLs. RAW 264.7 cells were cultured for 5 days in the presence
of RANKL (30 ng/mL, Ct1) and the indicated concentrations of IS [IS
at normal concentration (ISn, 0.6 mg/L), uraemic concentration (ISu,
53 mg/L) and maximum concentration (ISm, 236 mg/L)] in the presence
or absence of 3 mM Pi. TRAP1 multinucleated osteoclasts (RAW-Diff
OCLs) were assessed in each well. Data are represented as a percentage
of Ct1 and represent the mean 6 SEM of four independent experiments.
**P < 0.01, ***P < 0.001 versus Ct1. $$P < 0.01 compared with cells
cultured in the presence of 3 mM Pi.
Fig. 2. IS inhibits RANKL-induced differentiation of PBMCs into OCLs.
To induce differentiation, PBMCs were cultured for 14 days on bovine
bone slices in the presence of RANKL (25 ng/mL)and M-CSF (30 ng/mL)
(Ct1) and the indicated concentration of IS in the presence or absence of
3 mM Pi. (A) The number of multinucleated TRAP1 cells was assessed in
each well. (B) Pictures of TRAP1 OCL cells cultured on bone slices in the
presence of the indicated concentrations in IS alone (ISn, ISu, ISm; see the
legend to Figure 1 for details). The results are quoted as a percentage of
Ct1 and represent the mean 6 SEM of four independent experiments.
***P < 0.001 versus Ct1. $$P < 0.01, $$$P < 0.001 compared with cells
cultured in the presence of 3 mM Pi. @P < 0.01 and @@P < 0.001 IS
alone compared with IS 1 3 mM Pi.
2178 A. Mozar et al.
at New York University on November 28, 2012
IS inhibits the activation of RANKL protein kinases
The activation states of ERK1/2, p38, JNK and Akt were
determined by western blotting with specific antibodies
against the phosphorylated forms of these enzymes. Con-
cerning the effect of IS itself, weak activation of ERK1/2,
JNK, p38 and Akt was observed at the ISm concentration.
As expected, the addition of RANKL activated all four
kinases. Under these experimental conditions, we showed
that IS inhibited ERK1/2, JNK, p38 and Akt phosphoryla-
tions in a dose-dependent manner (Figure 5).
IS inhibits RANKL-induced DNA-binding activities of
NF-jB and AP-1
We next evaluated the effect of IS on the DNA-binding
activities of NF-jB and AP-1 transcriptional factors by
using the ELISA TransAM?method. We observed that
IS alone did not exert any effect on the DNA-binding ac-
tivities of these two transcriptional factors in RAW 264.7
cells. As previously reported, incubation with RANKL in-
creased the DNA-binding activities of both transcriptional
factors. We then observed that increasing uraemic concen-
trations of IS led to a significant dose-dependent inhibition
of AP-1-binding activity (Figure 6A). Significant inhibition
of NF-jB DNA-binding activity was observed only at the
ISm concentration (Figure 6B). Probenecid at 500 lM (but
not 100 lM) was able to prevent the inhibitory effect ex-
erted by ISm on AP-1-binding activity (P < 0.001 for ISm
alone versus ISm 1 probenecid, in the presence of
Using two well-established cellular models of OCL differ-
entiation, we demonstrated that the addition of IS at urae-
mic and maximum concentrations dose dependently
inhibited not only OCL differentiation from human and
mouse monocyte/macrophage models but also bone-
resorbing activity. Interestingly, these effects were signifi-
cantly enhanced by the presence of 3 mM Pi. Molecular
investigations showed that IS dose dependently inhibits the
activities of protein kinases Akt, JNK, p38 and ERK1/2 and
the DNA-binding activity of AP-1—all key proteins in-
volved in RANKL-induced OCL differentiation. We were
able to implicate the OATs in these effects by using pro-
benecid, a pharmacological inhibitor of these transporters
suggesting that IS might need to be taken up by monocyte/
macrophage precursors to inhibit differentiation.
Our data emphasize the fact that several uraemic toxins
could affect OCL differentiation and function. In fact, we
had already observed in vitro that the uraemic toxin Pi at
uraemic concentration directly inhibited both OCL differ-
entiation and bone-resorbing activity induced by RANKL
and M-CSF . In the present study, we observed that
the uraemic toxin IS has comparable effects on OCL differ-
entiation and function. Moreover, we demonstrated that the
Fig. 3. IS inhibits RANKL-induced bone resorption. (A) After removing
OCL cells, the bone slices were stained with haematoxylin red and tolui-
dine blue to visualize the resorption lacunae (pits). For each condition, we
assessed the total pit surface area. (B) Quantitative assay of C-terminal of
type I collagen telopeptide (CTX) degradation products in PBMC culture
supernatants. The results are quoted as a percentage of Ct1 and represent
the mean 6 SEM of four independent experiments. **P < 0.01, ***P <
0.001 compared with bone resorption when PBMCs were cultured in Ct1.
$P < 0.05, $$P < 0.01 and $$$P < 0.001 compared with cells cultured in
the presence of 3 mM Pi. @P < 0.01 and @@P < 0.001 IS alone com-
pared with IS 1 3 mM Pi.
Fig. 4. The OAT inhibitor probenecid blocks IS’s inhibition of RAW
264.7 cell differentiation. RAW 264.7 cells were cultured for 5 days in
the presence of RANKL (30 ng/mL) and 0, 1, 10 and 100 lM probenecid
in the presence or absence of IS. The number of TRAP1 multinucleated
osteoclasts derived from RAW 264.7 cells were counted in each well. The
results are quoted as a percentage of Ct1 and represent the mean 6 SEM
of two independent experiments. ***P < 0.001 versus Ct1. $$P < 0.01
and $$$P < 0.001 compared with cells cultured in the presence of ISu.
IS inhibits OCL differentiation 2179
at New York University on November 28, 2012
effects of these two uraemic toxins (i.e. IS and Pi) were
additive. Other uraemic toxins may have opposite effects.
Recently, it has been shown that the uraemic toxin beta(2)-
microglobulin stimulates osteoclastogenesis via up-regulation
of tumour necrosis factor-alpha and interleukin-1 expres-
sion . However, it seems that the uraemic milieu has
rather inhibitory effects on both OCL differentiation and
bone-resorbing activity. Indeed, when we compared the
impact of pooled uraemic serum from CKD haemodialysis
patients with that of healthy control serum (10% of healthy
control serum or FCS), we observed that PBMC differen-
tiation into OCLs and bone-resorbing activity were greatly
decreased . A note of caution should be added since the
final concentrations in the present experimental conditions
by far exceed free IS concentrations in vivo. Unfortunately,
HSA did affect our model by itself inducing the differen-
tiation of RAW 264.7 cells into OCL-like cells. These
phenomena prevented us from gaining a clear picture of
protein-binding effects. However, one could advocate that
the situation may be different between in vitro and in vivo.
Indeed, the short-term exposure in vitro to high IS concen-
trations in the present study may mimic the long-term ex-
posure in patients to lower IS concentrations.
Bone remodelling depends not only on osteoblast activ-
ity but also on the OCLs’ bone-resorbing activity. Hence,
disturbance of osteoblast and/or OCL activities can induce
bone metabolism disorders. Along with CKD progression,
the kidney’s functions gradually become unable to main-
tain systemic mineral homeostasis, resulting in the various
abnormalities of bone and vascular physiology observed in
CKD-MBD (CKD associated with mineral and bone dis-
orders) . The various bone disorders observed in CKD
patients are probably depending not only on the serum
concentration in PTH but also on the concentrations of
various uraemic toxins. It has been suggested that when
compared with healthy patients, dialysed CKD patients
require two to three times more PTH to maintain a normal
bone turnover rate probably to counterbalance direct effects
of uraemic toxins on bone in these patients . It is note-
worthy that in a rat model of CKD with low bone turnover,
Iwasaki et al.  showed that serum IS rises while bone
turnover decreases with the impairment in renal function.
Oral administration of indole-absorbing Kremezin? pre-
vents both the progression of renal failure and the decrease
in bone formation . Accumulation of IS in CKD patients
could be at least one of the factors that induce skeletal
resistance to PTH by its deleterious effects on osteoblasts
and OCLs. Recently, Goto et al.  reported that IS cor-
related negatively with two serum markers of bone forma-
tion (alkaline phosphatase and bone-specific alkaline
phosphatase) independently of intact PTH but not with a
serum marker of bone resorption (TRAP 5b). However,
serum markers may not be ideal to test bone turnover and
bone histomorphometric data are needed. Hence, by affect-
ing both osteoblast and OCL activities, IS may promote
low bone turnover diseases outcomes (such as ABD) ob-
served in CKD patients.
Our data confirm and extend previous works showing
that IS acts as a bone toxin by affecting both osteoblast and
OCL activities. More work is required to understand how
IS enters the cell to exert its effects and to determine
whether OATs play a role in entry. The relevance of these
findings must, however, be confirmed by further studies in
a clinical setting.
Fig. 5. Effect of IS on RANKL-induced phosphorylation of the signalling kinases ERK, p38, JNK and Akt. The cells were pre-incubated for 24 h in the
differentiation medium aMEM supplemented with 10% FCS. RANKL 100 ng/mL was added for 10 min before the preparation of cellular extracts. The
IS was added just before the RANKL, in order to reach the indicated concentrations. The phosphorylation state of the kinases was determined by
immunoblot analysis. The results are representative of three independent experiments. Protein expression was quantified and data are expressed as the
phosphorylated protein/b-actin ratio. The Ct condition in the absence of RANKL was considered as 1.
2180 A. Mozar et al.
at New York University on November 28, 2012
Acknowledgements. The authors thank the ‘Laboratoire de Biochimie du
Centre Hospitalier Universitaire d’Amiens Nord’ and the ‘Laboratoire de
Biologie Endocrinienne et Osseuse du Centre Hospitalier Universitaire
d’Amiens Sud’ for their valuable technical help with this work.
Authors’ contributions: Anaı ¨s Mozar, Loı ¨c Louvet and Ziad Massy
participated in the conception and design of the study and acquisition of
the data; they analysed and interpreted the data and drafted the first and
final version of the paper. Saı ¨d Kamel, Michel Brazier and Romuald
Mentaverri participated in the conception and design of the study and
interpretation of the data and revised the manuscript for content. Corinne
Godin participated in the acquisition and interpretation of data and revised
the manuscript for content.
Funding. Conseil Re ´gional de Picardie.
Conflict of interest statement. A.M., L.L., C.G., R.M., M.B., S.K., Z.A.M.
authors declare no conflict of interest.
1. Niwa T, Takeda N, Tatematsu A et al. Accumulation of indoxyl
sulfate, an inhibitor of drug-binding, in uremic serum as demonstrated
by internal-surface reversed-phase liquid chromatography. Clin Chem
1988; 34: 2264–2267
2. Dou L, Jourde-Chiche N, Faure V et al. The uremic solute indoxyl
sulfate induces oxidative stress in endothelial cells. J Thromb Hae-
most 2007; 5: 1302–1308
3. Motojima M, Hosokawa A, Yamato H et al. Uremic toxins of or-
ganic anions up-regulate PAI-1 expression by induction of NF-
kappaB and free radical in proximal tubular cells. Kidney Int
2003; 63: 1671–1680
4. Shimoishi K, Anraku M, Kitamura K et al. An oral adsorbent, AST-
120 protects against the progression of oxidative stress by reducing
the accumulation of indoxyl sulfate in the systemic circulation in renal
failure. Pharm Res 2007; 24: 1283–1289
5. Iwasaki Y, Yamato H, Nii-Kono T et al. Administration of oral char-
coal adsorbent (AST-120) suppresses low-turnover bone progression
in uraemic rats. Nephrol Dial Transplant 2006; 21: 2768–2774
6. Nii-Kono T, Iwasaki Y, Uchida M et al. Indoxyl sulfate induces
skeletal resistance to parathyroid hormone in cultured osteoblastic
cells. Kidney Int 2007; 71: 738–743
7. Somerville PJ, Kaye M. Evidence that resistance to the calcemic
action of parathyroid hormone in rats with acute uremia is caused
by phosphate retention. Kidney Int 1979; 16: 552–560
8. Mozar A, Haren N, Chasseraud M et al. High extracellular inorganic
like cells. J Cell Physiol 2008; 215: 47–54
9. Vanholder R, De Smet R, Glorieux G et al. Review on uremic toxins:
classification, concentration, and interindividual variability. Kidney
Int 2003; 63: 1934–1943
10. Menaa C, Esser E, Sprague SM. Beta2-microglobulin stimulates os-
teoclast formation. Kidney Int 2008; 73: 1275–1281
11. Chillon JM, Mozar A, Six I et al. Pathophysiological mechanisms and
consequences of cardiovascular calcifications: role of cardiovascular
and bone remodeling. Ann Pharm Fr 2009; 67: 234–240
12. Moe S, Drueke T, Cunningham J et al. Definition, evaluation, and
classification of renal osteodystrophy: a position statement from Kid-
neyDisease:ImprovingGlobalOutcomes (KDIGO).KidneyInt 2006;
13. Stubbs JR, Liu S, Tang W et al. Role of hyperphosphatemia and 1,25-
dihydroxyvitamin D in vascular calcification and mortality in fibroblas-
tic growth factor 23 null mice. J Am Soc Nephrol 2007; 18: 2116–2124
14. Goto S, Fujii H, Hamada Y et al. Association between indoxyl sulfate
and skeletal resistance in hemodialysis patients. Ther Apher Dial
2010; 14: 417–423
Received for publication: 12.7.11; Accepted in revised form: 4.10.11
Fig. 6. The effect of IS on the RANKL-induced DNA-binding activity of
the transcriptional factors NF-jB and AP-1. RAW 264.7 cells were pre-
incubated for 24 h in aMEM differentiation medium supplemented with
10% FCS. RANKL 50 ng/mL was added for 30 min before the preparation
of nuclear extracts. The IS was added just before the RANKL, in order to
reach the indicated concentration. The DNA-binding activity was deter-
mined with NF-jB-p65 and AP-1-cjun TransAM? ELISA kits from Ac-
tive Motif. The results (means 6 SEM) are quoted as a percentage of the
Ct experiment in the absence of RANKL. (A) Effects of IS on NF-jB
DNA-binding activity in the presence or absence of 50 ng/mL of murine
RANKL. (B) Effects of IS on AP-1 DNA-binding activity in the presence
or absence of 50 ng/mL murine RANKL. Data are represented as a per-
centage of the Ct condition in the absence of RANKL and represent the
mean 6 SEM of three independent experiments 6 SEM. *P < 0.05,
**P < 0.01 and ***P < 0.001 when compared with Ct condition with
IS inhibits OCL differentiation2181
at New York University on November 28, 2012