Nephrol Dial Transplant (2016) 31: 401–412

doi: 10.1093/ndt/gfv353
Advance Access publication 20 October 2015

Gut Lactobacillus protects against the progression of renal damage

by modulating the gut environment in rats

Ayumi Yoshifuji, Shu Wakino, Junichiro Irie, Takaya Tajima, Kazuhiro Hasegawa, Takeshi Kanda,

Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
Hirobumi Tokuyama, Koichi Hayashi and Hiroshi Itoh
Department of Internal Medicine, School of Medicine, Keio University, Tokyo, Japan

Correspondence and offprint requests to: Shu Wakino; E-mail: shuwakino@z8.keio.jp

Keywords: chronic kidney disease, gut microbiota, protein

A B S T R AC T bound uremic retention solutes, tight junction, Toll-like
receptor
Background. The role of gut microbiota in the progression of
chronic kidney disease (CKD) has not been fully elucidated.

ORIGINAL ARTICLE
Methods. Renal failure was induced in 6-week-old spontan-
eously hypertensive rats by 5/6 nephrectomy (Nx). We analyzed INTRODUCTION
the gut microbiota population to identify the relevant species
Chronic kidney disease (CKD) is a worldwide health problem
potentially involved in inducing renal damage. Human colon
that can lead to cardiovascular disease and end-stage kidney
Caco-2 cells were used to delineate the mechanism involved
disease that requires renal replacement therapy. Although its
in the molecular changes in the gut of Nx rats.
pathogenesis and pathophysiology have not been fully eluci-
Results. Nx rats showed an increase in Bacteroides (Bact) and a
dated, the progression of CKD is accompanied by the accumu-
decrease in Lactobacillus (Lact) species compared with sham-
lation of uremic substances that have deleterious effects. For
operated rats. Lact, but not Bact, populations were significantly
example, the protein-bound uremic solutes indoxyl sulfate
associated with urinary protein excretion. Treatment of Nx rats
(IS) and p-cresyl sulfate (PCS), which are normally excreted
with 1 × 1010 CFU/kg/day Lact ameliorated increased urinary
into the urine by the kidneys, accumulate in patients with
protein excretion and higher serum levels of the uremic toxins,
CKD. These substances have been shown to cause damage to
indoxyl sulfate and p-cresyl sulfate, and serum urea nitrogen le-
renal tubular cells or endothelial cells, as well as to other cells,
vels. Lact also attenuated systemic inflammation in Nx rats, as
by increasing cellular oxidative stress [1]. In an in vivo study
evaluated by serum lipopolysaccharide, interleukin-6 and C-
where these substances were administered exogenously, they in-
reactive protein levels. Histologically, renal sclerosis in Nx rats
duced damage to renal or cardiovascular tissues by enhancing
was restored by Lact treatment. A reduction in the expression of
oxidative stress, inflammatory responses or fibrotic changes [2–
tight junction proteins and the Toll-like receptor 2 (TLR2), a
4]. A recent study also demonstrated that another uremic sol-
putative Lact receptor, in the colons of Nx rats were mitigated
ute, indole acetic acid (IAA), caused endothelial cell damage.
by Lact. Treatment of Caco-2 cells with indole downregulated
In addition, IAA serum levels correlated with cardiovascular
tight junction protein expression, which was abolished by ex-
events in CKD patients, suggesting that it may be a cause of car-
posure to Lact. The effects of Lact were reversed by treatment
diovascular complications in CKD [5]. Since the precursors of
with OxPAPC, a TLR inhibitor. Similarly, the increase in the
these substances, indoles or cresols, are produced by bacteria in
permeability of the Caco-2 cell monolayer was reversed by
the intestine from the amino acid tryptophan or tyrosine, re-
the administration of Lact. Lact upregulated TLR2 expression
spectively, and are subsequently absorbed from the colon, the
in Caco-2 cells. Lact also attenuated the increase in serum in-
changes in the influx of these uremic solutes from the intestine
doxyl sulfate and urea levels and urinary protein excretion in
may affect renal function [6]. This kind of interrelationship
Nx rats even in the pseudogerm-free environment.
between the gut and kidney is called ‘the intestinal-renal syn-
Conclusions. Lact supplementation mitigated the systemic in-
drome’[7], and the modulation of this axis, including regulation
flammation and proteinuria associated with renal failure, sug-
of the colon environment or colon barrier, may be a plausible
gesting that in the gut microbiota, Lact plays a protective role
strategy to slow the progression of CKD. A decreased expres-
against the progression of CKD.
sion of tight junction proteins in the colon has been observed

© The Author 2015. Published by Oxford University Press 401

on behalf of ERA-EDTA. All rights reserved.
 in CKD, and stabilization of the intestinal barrier by changing
the colon environment, using a charcoal absorbent for indole,
was found to improve the associated endotoxemia, oxidative
stress and inflammatory response [8].
Another factor that affects the intestinal environment is
changes in the intestinal microbiota, which have been shown to
correlate with kidney diseases [9, 10]. Manipulation of micro-
biota, such as with the use of probiotics, reduced blood urea ni-
trogen (BUN) levels and enhanced longevity in CKD rats [11, 12].
Changes in the distribution of intestinal microbiota and reduc-
tions in uremic toxins caused by treatment with probiotics have
been observed in hemodialysis patients [13], although the clinical
relevance of these alterations has not been determined [14].
This study assessed the role of the gut environment in CKD
by using 5/6 nephrectomized (Nx) spontaneous hypertensive
rats (SHRs), in which we previously observed severe renal dam-
age and inflammation [15]. We found that a decrease in the
population of Lactobacillus (Lact) in Nx rats, along with disrup-
tion of the intestinal barrier, contributed to the progression of
CKD. We also showed that Lact supplementation ameliorated
renal damage and intestinal changes through activation of a
Toll-like receptor (TLR) that recognizes Lact. These findings
provide compelling evidence for the clinical use of Lact in
CKD patients to reverse the inflammatory response and protect
against the progression of renal damage.
ORIGINAL ARTICLE
M AT E R I A L S A N D M E T H O D S
Animal experiments
Renal failure was induced in 6-week-old male SHRs (Charles
River, Wilmington, MA, USA) via the Nx model, as described
previously [15]. In the first experiment, rats were randomly
assigned to two experimental groups, sham-operated SHR
(Sham, n = 8) and Nx rats (n = 8). In the second experiment,
rats were randomly assigned to three experimental groups:
sham-operated SHR (Sham, n = 8), Nx (n = 8) and Nx rats
treated with 1 × 1010 CFU/kg/day Lactobacillus acidophilus
NT (Lact) (Nx+Lact, n = 8). Lact was kindly provided by
Nitto Pharmaceutical Industries (Kyoto, Japan). Twelve weeks
after nephrectomy, body weight and systolic blood pressure
(SBP) were measured by the tail-cuff method (KN-210, Nat-
sume, Tokyo, Japan). Twenty-four-hour urinary protein excre-
tion and serum concentrations of BUN, serum creatinine and IS
were measured as described previously [15, 16]. Serum PCS and
IAA levels were measured by high-performance liquid chroma-
tography [17]. Serum lipopolysaccharide (LPS) levels were
measured by turbidimetric-kinetic assay using a toxinometer
MT6500 (Wako Pure Chemical Industries, Osaka). Serum
Table 1. The primers used for bacterial gene analysis
Bacteria Forward
Lactobacilli AGCAGTAGGGAATCTTCCA
Bacteroides GAAGGTCCCCCACATTG
Prevotella CACCAAGGCGACGATCA
Universal primers TCCTACGGGAGGCAGCAGT
402
levels of interleukin-6 (IL-6) and C-reactive protein (CRP)
were measured using commercial ELISA kits (Abcam Japan,
Tokyo, Japan). At sacrifice, tissue samples of the kidney and as-
cending colon were obtained, snap frozen and used for subse-
quent experiments. All experiments were performed in
accordance with the animal experiment guidelines of the Keio
University School of Medicine.
Morphological examination
Kidneys were fixed in 10% formaldehyde and embedded in
paraffin blocks. Glomerular sclerosis was assessed by periodic
acid-Schiff (PAS) staining and renal fibrosis was assessed by
Masson-trichrome staining. Glomerulosclerosis was evaluated
Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
by counting sclerotic glomeruli and calculating the glomerulo-
sclerotic index [18, 19]. The proportion of the fibrotic area was
measured using Image-Pro Plus 3.0 (Media Cybernetics, Silver
Spring, MD, USA).
Analysis of gut bacteria
The gut microbiota population was examined as described
previously [20]. Briefly, DNA was extracted from the bead-
treated suspension of fecal samples. Amplification of fecal
16S rDNA was performed using 25 μM of the labeled primers
516f (50 TGCCAGCAGCCGCGGTA 30 ) and 1510r (50 GGTTA
CCTTGTTACGACTT 30 ). For terminal restriction fragment
length polymorphism (T-RFLP) analysis of the microbiota,
the resulting amplicons were digested with 2U of Bsl I (Cell Sig-
naling, Danvers, MA, USA) for 1 h and the fragments were se-
parated using an automated DNA sequencer (ABI PRISM
3130xl DNA Sequencer, Applied Biosystems, Carlsbad, CA,
USA). The lengths and peak areas of the resulting electrophero-
gram were determined using GeneMapper software (Applied
Biosystems). The relative abundances of operational taxonomic
units (OTUs) were calculated by dividing the peak area of each
fragment by the sum of all peak areas. The T-RFLP patterns of
each sample were compared using the dissimilarity index [21].
The amounts of bacteria in the fecal samples of each species
were quantified by real-time quantitative PCR using the 7500
Fast Real-time PCR System (Applied Biosystems), as previously
reported [22]. The primers for each species are shown in Table 1
[23]. Results with the specific Lactobacillus, Bacteroides and
Prevotella primers were normalized relative to those for the uni-
versal primers.
Cell culture
Caco-2 sigmoid colon cancer cells (HTB-37; ATCC, LOT
60143947) were grown in non-fibrin-coated flasks in Dulbecco’s
modified eagle’s medium (DMEM, Invitrogen, Auckland, NZ),
containing glucose, L-glutamine, NaHCO3 and pyridoxine
Reverse
CACCGCTACACATGGAG
CAATCGGAGTTCTTCGTG
GGATAACGCCTGGACCT
GACTACCAGGGTATCTAATCCTGTT
A. Yoshifuji et al.
(Sigma, St Louis, MO, USA) at 37°C in 5% CO2. To evaluate the
effects of uremic toxins, Caco-2 cells were treated with indole
(Sigma), the precursor of IS, at various doses, as well as with
2.0 × 109 CFU/plate Lact in antibiotic-free DMEM media con-
taining 0.4% FBS (Invitrogen). The concentration of indole in
the intestinal juice is estimated to be in the range of 0.1–100 μM
because the ‘human’ fecal concentration of indole reportedly
ranges from 250 to 1100 μM [24, 25] and indole is estimated to
be highly concentrated in the intestinal juice at least by 10- to
100-fold because of the adsorption of water [26]. Furthermore,
intestinal cells were covered with the mucous layer [26]. Based
on this evidence, we tested the effects of indole in the preliminary
experiments and found that 10 μM of indole was toxic to Caco-2
cells. We further examined the effects of indole on the expression
of tight junction proteins at the concentrations ranging from 0.1
to 1 μM and observed that indole downregulated ZO-1 and
occludin between the concentration of 0.1 and 1 nM, although
claudin-1 was not altered even at 1 μM. Hence the appropriate
concentration used in the experiment was determined to be be-
tween 0.1 and 1 nM. To investigate the role of the Toll-like recep-
tor 2 (TLR2), Caco-2 cells were pretreated with a nonselective
TLR blocker, oxidized 1-palmitoyl-2-arachidonyl-snglycero-3-
phosphorylcholine (OxPAPC, InvivoGen, San Diego, CA, USA),
30 min prior to the administration of Lact or indole. The cells
were harvested 24 h after the treatment and the expressions of
tight junction proteins, including occludin, ZO-1 and claudin-1,
were examined by immunoblotting.
Permeability assay
Permeability was estimated by measuring the paracellular
transport of FITC-labeled 4-kDa dextran (FD4) as described
previously [27]. Briefly, sterilized FD4 was added into the
apical well at 25 mg/mL with or without indole, Lact or a com-
bination of both for 6 h. The translocation of FD4 across the
cells was determined by sampling the basolateral solution.
FD4 fluorescence at 485/535 nm was measured using a Perkin
Elmer (Waltham, MS, USA) multimode plate reader (ARVO
X3). All permeability experiments were conducted in triplicate
for five sets of cell preparations. FD4 permeability was ex-
pressed as the apparent permeability coefficient (Papp), calcu-
lated as described previously [27].
Immunoblotting
Ascending colon tissues and Caco-2 cells were lysed and so-
nicated in a lysis buffer and centrifuged at 15 000 g for 15 min.
Aliquots of supernatants were electrophoresed on SDS-PAGE
gels and transferred to nitrocellulose membranes, which were
then incubated with primary antibodies against occludin, ZO-1
and claudin-1 (Invitrogen). Following incubation with a second-
ary antibody (HRP-linked anti-rabbit IgG; GE Healthcare,
Table 2. The primers used for real-time PCR
Gene Forward
TLR2 (rat) GTACGCAGTGAGTGGTGCAAGT
GAPDH (rat) GTTACCAGGGCTGCCTCTC
TLR2 (human) GCCAGCAGGTTCAGGATGTC
β-actin (human) GTCAGGTCACTATCGGC
Effects of Lactobacillus on CKD
Backhamshire, UK), immunoreactive bands were detected
using an ECL detection kit (Amersham Biosciences, Uppsala,
Sweden).
Real-time reverse transcriptase polymerase chain reaction
(RT-PCR)
Total RNA was extracted from the ascending colon tissue
and Caco-2 cells using TRIzol reagent (Invitrogen). Equal
amounts (1 μg) of total RNA from each sample were converted
to cDNA using the Prime Script RT reagent kit with gDNA Era-
ser (Takara, Ohtsu, Japan). Real-time RT-PCR was performed
using an ABI Step One Plus sequence detector (Applied Biosys-
tems). Primers used are shown in Table 2. Levels of mRNA were
Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
normalized to those of β-actin.
Pseudogerm-free model experiment
A pseudogerm-free condition was created by administering
a broad-spectrum antibiotic cocktail consisting of ampicillin,
neomycin sulfate, vancomycin and metronidazole as described
previously [28]. Rats were randomly assigned to the following
five groups: (1) sham-operated SHR (Sham, n = 6), (2) Nx
SHR (Nx, n = 6), (3) Sham plus antibiotics (Sham+Anti, n = 6),
(4) Nx plus antibiotics (Nx+Anti, n = 6) and (5) Nx+Anti trea-
ted with Lact (Nx+Anti+Lact, n = 6). Eight weeks after neph-
rectomy, 24-h urinary protein excretion levels and SBP were
ORIGINAL ARTICLE
measured and the rats were sacrificed. The concentration of
BUN and serum concentrations of creatinine and IS were mea-
sured as described [15].
Statistical analysis
Data are expressed as mean ± SEM and analyzed by one-way
analysis of variance, followed by the Bonferroni’s post hoc test.
A P-value <0.05 was considered statistically significant.
R E S U LT S
Decreased Lact population in nephrectomized rats
The microbiota between Nx SHRs, a model of severe ne-
phrosclerosis [15] and Sham-operated SHRs were compared.
Body weight was significantly lower (286.0 ± 38.2 g versus
366.8 ± 5.4 g, P < 0.05) and SBP was significantly higher
(Figure 1A) in Nx compared with Sham rats. In addition,
serum creatinine, IS and BUN levels, as well as urinary protein
excretion, were higher in Nx than in Sham rats (Figure 1B–E),
confirming that these rats were an appropriate model of severe
renal dysfunction. T-RFLP evaluation of intestinal microbiota
in feces from the terminal ileum showed that Bacteroides
(Bact) and Prevotella species were more abundant, whereas
Lact species were less abundant in the Nx rats compared with
Reverse
GGCCGCGTCATTGTTCTC
GGGTTTCCCGTTGATGACC
AGGCATCCCGCTCACTGTAA
CATGGATGCCACAGGATTCC
403
 Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
ORIGINAL ARTICLE

F I G U R E 1 : Effects of renal impairment on biochemical parameters and microbiota populations in rats. Spontaneously hypertensive rats
underwent 5/6 nephrectomy (Nx) or sham nephrectomy (Sham). Measurements were taken of (A) SBP, (B) serum creatinine concentrations, (C)
BUN, (D) urinary protein excretion and (E) serum IS concentrations. (F) Relative abundance of microbiota based on the average number of each
subfamily measured by T-RFLP analysis. (G) Populations of Bacteroides species (upper panel) and Lactobacillus (lower panel) in the colons of
Sham and Nx groups were examined by RT-PCR. (H) Relationship between urinary protein excretion levels and Bacteroides (upper panel) and
Lactobacillus (lower panel) populations. *P < 0.05, **P < 0.01 versus Sham (n = 8 per group).

the Sham rats (Figure 1F). Confirmatory real-time PCR using
species-specific primers showed that Bact species were higher
and Lact lower in number in Nx compared with Sham rats
(Figure 1G), whereas Prevotella species were similar. The Lact
population was significantly associated with urinary protein ex-
cretion levels (Figure 1H; R 2 = 0.425, P < 0.01), whereas the
Bact population was not.
Effects of L. acidophilus on rats with CKD
Since decreased Lact in the intestine was associated with
urinary protein excretion, we investigated the effects of Lact
supplementation in our rat model of CKD by treating Nx
rats with 1 × 1010 CFU/kg/day Lact for 12 weeks (Nx+Lact).
RT-PCR using DNA extracted from fecal samples revealed

404 A. Yoshifuji et al.

that the population of Lact was higher in Nx+Lact than in Nx day). Similarly, body weight was lower in Nx (296.4 ± 21.2 g)
rats (Figure 2A). Daily chow intake was lower in Nx than in Sham (361.4 ± 9.4 g) rats, although this reduction im-
(24.3 ± 4.3 g/day) than in Sham (36.0 ± 2.7 g/day) rats, and proved in Nx+Lact rats (322.0 ± 24.9 g). Systolic blood pressure
this decrease was not improved by Lact treatment (28.2 ± 3.1 g/ was similar in Nx and Nx+Lact rats (Figure 2B). Although the

Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
ORIGINAL ARTICLE

F I G U R E 2 : Effects of Lactobacillus on nephrectomized rats. Spontaneously hypertensive rats underwent sham nephrectomy (Sham) or 5/6
nephrectomy (Nx) or Nx followed by Lactobacillus treatment (Nx+Lact), as described. (A) Numbers of Lactobacillus species in the colons, as
determined by RT-PCR. Measurements were made of (B) SBP, (C) BUN, (D) serum creatinine level, (E) daily urinary protein excretion, (F) serum
indoxyl sulfate (IS), (G) serum p-cresyl sulfate (PCS), (H) serum indole acetic acid (IAA), (I) serum lipopolysaccharide (LPS), (J) serum
interleukin-6 (IL-6) and (K) serum C-reactive protein (CRP) levels. The evaluation of (L) glomerular sclerosis by PAS staining and (M) interstitial
fibrosis by Masson-trichrome staining are shown. Lower panels represent the quantification of glomerular sclerosis and interstitial fibrosis. Scale
bar, 100 μm. *P < 0.05, **P < 0.01 versus Sham; #P < 0.05 versus Nx (n = 8 per group).

Effects of Lactobacillus on CKD 405

 increase in BUN levels caused by Nx improved following Lact
treatment (Figure 2C), serum creatinine levels were not altered
(Figure 2D), suggesting that supplementation with Lact was not
sufficient for the improvement of renal dysfunction in Nx SHRs.
The Lact-induced reduction of BUN may have been caused by
decreased urea production in the intestine or decreased urea
entry from the intestine. However, the increased urinary protein
excretion in Nx rats was mitigated in Nx+Lact rats (Figure 2E),
indicating that Lact had a renoprotective effect. Similarly, serum
concentrations of IS (Figure 2F), PCS (Figure 2G), LPS (Fig-
ure 2I), IL-6 (Figure 2J) and CRP (Figure 2K), which were higher
in Nx than in Sham rats, were mitigated in Nx+Lact rats. How-
ever, the increase in serum IAA levels (Figure 2H) in Nx rats was
not altered in Nx+Lact rats. Nx rats showed a significantly high-
er degree of glomerulosclerosis, as evaluated by the glomerulo-
sclerotic index (Figure 2L) [18, 19] and increased tissue fibrotic
area on Masson-trichrome staining (Figure 2M) than Sham rats.
Lact treatment improved the glomerulosclerotic but not the
fibrotic index.
Molecular changes in the intestinal barrier induced
by Lact
Disruption of the intestinal barrier has been shown to con-
tribute to increased levels of serum cytokines during renal fail-
ure, as well as to the progression of renal impairment [8]. Colon
ORIGINAL ARTICLE
epithelial tight junctions are involved in maintaining the intes-
tinal barrier, wherein the proteins occludin, ZO-1 and
claudin-1 are found to be key components of intestinal tight
junctions. The expression of all three proteins was lower in
the Nx model compared with Sham-operated rats. Although
Lact supplementation restored the expression of occludin and
ZO-1, it had no effect on claudin-1 expression (Figure 3A). It
is possible that the concentrations of uremic toxins or their pre-
cursors affect the population of microbiota or the expression of
tight junction proteins [29]. Although the fecal concentrations
of indole, p-cresol and phenol increased in Nx, Lact treatment
had no effect on any of these compounds (Figure 3B), indicat-
ing that the treatment with Lact had little effect on the produc-
tion or degradation of these substances in the intestine. The
expression of tight junction proteins has been reported to be
regulated by the activation of pattern recognition receptors,
TLRs [30, 31]. It has also been reported that Lact is recognized
by TLR2 and activates TLR2 in various cells [32, 33]. We found
that the expression of TLR2 was lower in the Nx than in
the Sham group. This decrease was reversed in the Nx+
Lact group (Figure 3C). These results suggested that the restor-
ation of intestinal tight junction protein expression and the
concomitant increase in TLR2 expression by Lact leads to the
suppression of increased intestinal permeability and to the re-
duction in systemic IS levels as well as in LPS levels, thus
mitigating renal damage and systemic inflammation in Nx
rats [34, 35].
Indole- and Lact-modified tight junction molecules
in Caco-2 cells
The mechanisms by which Nx and Lact induce molecular
changes in the intestinal barrier were investigated using the
Caco-2 colonic cell line. Indole, which is produced in the
406
intestine by some microorganisms, dose-dependently downre-
gulated the expression of occludin and ZO-1, while having no
effect on claudin-1 expression (Figure 4A). These indole-
induced reductions in occludin and ZO-1 were restored by
the coadministration of Lact to Caco-2 cells. These effects of
Lact were inhibited by the nonselective TLR blocker OxPAPC
(Figure 4B). Although indole did not alter the expression of
TLR2, it was increased by Lact treatment (Figure 4C). These
data implied that, in Nx rats, the expression of tight junction
proteins was disrupted by the reduction in the Lact population
through the reduced activation of TLR2. Finally, the permeabil-
ity of the cell layer was assayed using FD-4 dye as described in
the Materials and Methods section. Indole, at a concentration of
Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
1.0 nM, increased the permeability of the Caco-2 cell layer, and
this increase was inhibited by treatment with Lact (Figure 4D).
This result implied that the increased intestinal permeability in
Nx rats was improved by Lact treatment. It is deduced that Lact
contributes to maintenance of the intestinal environment.
Lact-mitigated proteinuria in rats with CKD under
pseudogerm-free conditions
The probiotic effects of Lact on proteinuria in Nx rats were
confirmed using pseudogerm-free conditions, where the effects
of microbiota other than Lact are minimized. Treatment of
Sham and Nx rats with an antibiotic cocktail consisting of
ampicillin, neomycin sulfate, vancomycin and metronidazole
yielded a pseudogerm-free condition in these animals, whereas
some of these rats were supplemented with Lact, as confirmed
by Gram staining of the feces of each experimental group
(Figure 5A). Treatment of Sham and Nx rats with antibiotics
had no effect on the levels of expression of serum creatinine,
BUN, or IS, or on urinary protein excretion (Figure 5B–E).
Treatment of Nx rats with Lact had no effect on serum creatin-
ine levels (Figure 5B) but attenuated the increases in BUN and
IS levels and urinary protein excretion even under pseudogerm-
free conditions (Figure 5C–E). These findings confirmed the
probiotic effects of Lact.
DISCUSSION
It has been strongly suggested that the gut microbiome is an im-
portant factor in the maintenance of a healthy condition [36,
37]. By using the recently developed T-RFLP technique fol-
lowed by real-time PCR, this study analyzed the species present
in the gut microbiota and found that the Lact population was
reduced under CKD conditions. The number of Lact was
found to be associated with urinary protein excretion levels,
and supplementation with Lact mitigated proteinuria in rats
with CKD. Furthermore, Lact helped to improve proteinuria
by restoring the expression of intestinal barrier proteins and re-
ducing systemic inflammation, even under pseudogerm-free
conditions. Lact had a direct effect on intestinal cells by revers-
ing the indole-induced downregulation of tight junction pro-
teins through activation of the TLR2 pathway. Thus this study
identified a specific species of microbiota that may be protective
against renal damage and revealed the detailed mechanisms
underlying these effects.
A. Yoshifuji et al.
 Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
ORIGINAL ARTICLE
F I G U R E 3 : Intestinal conditions in nephrectomized rats. Rats underwent sham nephrectomy (Sham), 5/6 nephrectomy (Nx) or Nx plus Lact
treatment (Nx+Lact), as described in the legend to Figure 2. (A) Expression levels of the tight junction proteins occludin (left panel), ZO-1 (middle
panel) and claudin-1 (right panel) in the colon. Each upper subpanel shows an immunoblot of a tight junction protein and each lower subpanel
shows the densitometric analysis of expression. (B) Fecal concentrations of indole (upper panel), p-cresol (middle panel) and phenol (lower panel).
(C) Levels of expressions of TLR2 in the colon, as shown by real-time RT-PCR. *P < 0.05, **P < 0.01 versus Sham; #P < 0.05 versus Nx (n = 8 per
group).

Colon microbiota are altered under uremic conditions.
For example, the populations of certain microbes, such as
Clostridiaceae, Enterobacteriacea and Verrucomicrobiacea,
which produce indole or p-cresol, are higher, whereas the po-
pulations of microbial families with butyrate-producing en-
zymes, including Lact and Prevotellae, are lower in subjects
with uremia compared with healthy subjects [38]. It was
reported that in CKD patients, probiotics with the combin-
ation of L. acidophilus, Bifidobacterium longum and Strepto-
coccus thermophiles produced a significant reduction of BUN
and enhanced well-being [39]. Although several other studies
also revealed that probiotics or prebiotics in CKD patients

Effects of Lactobacillus on CKD 407

 Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
ORIGINAL ARTICLE

F I G U R E 4 : Expression of intestinal barrier proteins in Caco-2 cells. (A) Caco-2 cells were incubated with various concentrations of indole and the
levels of expression of the tight junction proteins occludin (left panel), ZO-1 (middle panel) and claudin-1 (right panel) were assessed by im-
munoblotting. Each upper subpanel shows an immunoblot and each lower subpanel shows the densitometric analysis of that blot. *P < 0.05 versus
unstimulated cells (n = 5). (B) Effects of Lactobacillus (Lact) and the nonselective TLR blocker OxPAPC (Ox) on the expression of occludin (upper
panel), ZO-1 (middle panel) and β-actin (lower panel) in indole-treated (1.0 nmol/L) Caco-2 cells. Each upper subpanel shows an immunoblot
and each lower subpanel shows the densitometric analysis of that blot. – denotes no treatment. *P < 0.05 versus unstimulated cells, #P < 0.05 versus
indole-stimulated cells, ¶P < 0.05 versus indole-stimulated cells treated with Lact (n = 5). (C) Effects of Lact and indole on the expression of TRL2
mRNA in indole-treated Caco-2 cells. *P < 0.05 versus control (n = 5). (D) Effects of Lact and indole on the permeability of the Caco-2 cell
monolayer. *P < 0.05 versus unstimulated cells, #P < 0.05 versus indole-stimulated cells (n = 5).

reduce BUN levels or serum IS levels, some papers have to develop more efficient therapeutic strategies or to
reported no beneficial effects of probiotics, prompting a elucidate more precise mechanisms for dysbiosis in CKD
controversy regarding probiotic therapy [6, 14, 40]. In this [6, 14]. Our data provide relevant information for these fu-
sense, it has been proposed that future studies are necessary ture studies.

408 A. Yoshifuji et al.

 Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
ORIGINAL ARTICLE
F I G U R E 5 : The effects of Lactobacillus on nephrectomized rats in a pseudogerm-free environment. (A) Gram staining of stool samples of SHRs
(Sham), 5/6 nephrectomized SHRs (Nx), Sham rats treated with antibiotics (Sham+Anti), Nx rats treated with antibiotics (Nx+Anti) and Nx rats
treated with antibiotics and Lactobacillus (Nx+Anti+Lact). Measurements of (B) serum creatinine, (C) BUN (D) and serum IS levels and (E) daily
urinary protein excretion in the five rat groups. *P < 0.05, **P < 0.01 versus Sham+Anti; #P < 0.05 versus Nx+Anti (n = 6 per group).

In our experiment using the pseudogerm-free condition, al-
though the number of Lact should be decreased by antibiotics,
the antibiotic cocktail used in this study had no effects on pro-
teinuria. These data imply that antibiotic therapy also reduces
other microbiota with unfavorable effects on the kidney, such as
Clostridium species or Bact species, which are the microbiota
producing indole. Our analysis showed an increase in the popu-
lation of Bact, and it was reported that ∼20% of Bact species
produce indole [41]. Consistently, fecal concentrations of
indole in the intestine were higher in Nx compared with
Sham rats. Because indole directly downregulates the expres-
sion of tight junction proteins that form a barrier of intestinal
cells, and also because indole is a direct precursor of the uremic
toxin IS, the increase in the population of Bact in the Nx model
could lead to increased intestinal permeability and increased
serum levels of IS as well as LPS. Our findings show that the
gut contains various types of microbiota that may have an effect
on renal damage, either positively or negatively. Both a decrease
in renoprotective Lact and an increase in renal pathogenic Bact
contributed to the progression or maintenance of renal damage
in Nx rats. Collectively, it is suggested that the balance between
these bacteria may have some role in CKD.
In the present study, Lact treatment salvaged the expression
of tight junction proteins in the intestine of Nx rats. Our study
provided the molecular mechanism and evidence for the direct
action of Lact in intestinal cells. In the previous study, the
downregulation of tight junction proteins by infection of
Caco-2 cells with Escherichia coli or Salmonella was reversed
by the addition of Lactobacillus amylophilus via modulation
of the extracellular signal-regulated kinase (ERK) and c-Jun

Effects of Lactobacillus on CKD 409

 NH2-terminal kinase (JNK) pathways [42]. In addition, activa-
tion of the TLR2 pathway has been reported to lead to ERK or
JNK activation [43, 44]. TLRs are mucosal pattern recognition
receptors that recognize microbes, leading to the preservation of
tight junction integrity [30, 31]. Taken together, our findings
suggest that Lact regulates the expression of tight junction pro-
teins through the direct activation of TLR2. Consistent with this
hypothesis, our experiments using Caco-2 cells showed that the
restoration of indole-induced downregulation of ZO-1 and
occludin by Lact was inhibited by pretreatment with the TLR
antagonist OxPAPC. In CKD, reduction in the Lact population
and the subsequent downregulation of TLR2 also contributed
to downregulation of tight junction proteins (Figure 6). Supple-
cytokines, endotoxins, uremic toxins and microorganisms [9,
32]. Tight junction proteins form an effective barrier against
the influx of these noxious substances from the gastrointestinal
lumen to the internal milieu. In a uremic state, intestinal barrier
function is impaired, resulting in increased circulating endo-
toxin levels and enhanced systemic inflammation [48]. This
systemic inflammation contributes to the progression of renal
impairment [10]. In the previous report, Nx rats exhibited low-
grade inflammation, leading to impairment of kidney structure
and function [49]. In this study, Lact restored intestinal tight
junction protein expression, which possibly reduced intestinal
permeability as shown in the experiment described in Fig-
ure 4D. These effects stabilized intestinal barrier function,

mentation with Lact successfully reversed these changes. We
also found that tight junction proteins were downregulated by
the direct action of indole. Indole has been shown to activate
arylhydroxycarbon (AhR) and the pregnane X receptors
(PXRs) expressed in cell nuclei [45, 46]. Our results suggest a
possible cross-talk between these nuclear receptors and TLR
signaling in the intestinal cells (Figure 6) [10, 47].
The intestine plays an important role in regulating systemic
inflammation by serving as a barrier against the entry of various
ORIGINAL ARTICLE
Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
blocked the translocation of LPS and attenuated systemic in-
flammation, as indicated by the reduction in both serum IL-6
and CRP levels. These effects ultimately improved renal sclerot-
ic changes and proteinuria in Nx rats.
Moreover, the increase in uremic solutes IS and PCS has
been reported to induce renal inflammatory responses [4, 50,
51]. Our study implied that in Nx rats, besides an impaired
clearance, the increase in intestinal indole and p-cresol produc-
tion, presumably by an increase in the population of indole-

F I G U R E 6 : Scheme depicting the effects of CKD on microbiota populations and systemic inflammation. Intestinal indole downregulates the
expression of tight junction proteins, presumably through binding to the arylhydroxycarbon receptor (AhR) or PXR. In contrast, Lactobacillus
maintains protein expression through the activation of TLR2. In CKD, indole concentrations increase and the population of Lactobacillus de-
creases, which both result in the downregulation of expression of tight junction proteins. These mechanisms enhance intestinal permeability, thus
allowing various uremic toxins such as indole and p-cresol to enter the systemic circulation, consequently elevating the systemic indoxylsulfate (IS)
and p-cresylsulfate (PCS) levels. Increased permeability also allows bacterial lipopolysaccharide (LPS) to translocate to the systemic circulation,
leading to systemic inflammation along with increases in serum cytokine levels such as IL-6 and CRP. These changes ultimately cause renal cell
damage. Lactobacillus supplementation reverses these conditions and mitigates renal damage.

410 A. Yoshifuji et al.

and p-cresol-producing microbiota, and the disruption of the
intestinal barrier due to the increased levels of the indole itself
or the reduction in the Lact population could be responsible for
the observed increase in serum IS and PCS levels. Treatment of
Nx rats with Lact stabilized barrier function and reduced the
serum levels of IS and PCS, resulting in reductions in renal in-
flammation and in renal sclerosis. Because Lact treatment did
not alter fecal concentrations of indole, as well as p-cresol, or
renal excretory function, Lact-induced reductions in systemic
IS and PCS were considered to be primarily due to blocking
of indole and p-cresol entry from the intestine. These mechan-
isms may explain the renoprotective effects of Lact (Figure 6).
This barrier-stabilizing effect may also explain why Lact re-
duced BUN levels without affecting renal function, as serum
urea levels are also determined by urea production by intestinal
microbiota and by urea entry into the intestine [52]. Several
other metabolic pathways, including uptake via the intestinal
wall, and metabolic conjugation reactions, such as sulfotrans-
feration of indole to IS or that of p-cresol to PCS by the intes-
tinal wall, may be involved in uremic toxin generation [6]. The
effects of Nx and Lact treatment on these pathways remain to be
investigated.
In conclusion, Lact exhibited renal protective effects in CKD
rats by modulating the gut environment and regulating system-
ic inflammation. These effects suggest a novel mechanism by
which renoprotective Lact can improve systemic inflammation
and slow the progression of renal damage.
AC K N O W L E D G E M E N T S
This work was supported by the Scientific Research Fund of the
Ministry of Education, Culture, Sports, Science, and Technol-
ogy of Japan (grant no. 70573286).
C O N F L I C T O F I N T E R E S T S TAT E M E N T
None declared.
REFERENCES
1. Vanholder R, Schepers E, Pletinck A et al. The uremic toxicity of indoxyl
sulfate and p-cresyl sulfate: a systematic review. J Am Soc Nephrol 2014; 25:
1897–1907
2. Muteliefu G, Shimizu H, Enomoto A et al. Indoxyl sulfate promotes vascu-
lar smooth muscle cell senescence with upregulation of p53, p21, and pre-
lamin A through oxidative stress. Am J Physiol Cell Physiol 2012; 303:
C126–C134.
3. Shimizu H, Yisireyili M, Higashiyama Y et al. Indoxyl sulfate upregulates
renal expression of ICAM-1 via production of ROS and activation of
NF-κB and p53 in proximal tubular cells. Life Sci 2013; 92: 143–148.
4. Watanabe H, Miyamoto Y, Honda D et al. p-Cresyl sulfate causes renal
tubular cell damage by inducing oxidative stress by activation of NADPH
oxidase. Kidney Int 2013; 83: 582–592
5. Dou L, Sallée M, Cerini C et al. The cardiovascular effect of the uremic sol-
ute indole-3 acetic acid. J Am Soc Nephrol 2015; 26: 876–887
6. Vanholder R, Glorieux G. The intestine and the kidneys: a bad marriage can
be hazardous. Clin Kidney J 2015; 8: 168–179
Effects of Lactobacillus on CKD
7. Ritz E. Intestinal-renal syndrome: mirage or reality? Blood Purif 2011; 31:
70–76
8. Vaziri ND, Yuan J, Khazaeli M et al. Oral activated charcoal adsorbent
(AST-120) ameliorates chronic kidney disease-induced intestinal epithelial
barrier disruption. Am J Nephrol 2013; 37: 518–525
9. Anders HJ, Andersen K, Stecher B. The intestinal microbiota, a leaky gut,
and abnormal immunity in kidney disease. Kidney Int 2013; 83: 1010–1016
10. Ramezani A, Raj DS. The gut microbiome, kidney disease, and targeted in-
terventions. J Am Soc Nephrol 2014; 25: 657–670
11. Ranganathan N, Patel B, Ranganathan P et al. Probiotic amelioration of
azotemia in 5/6th nephrectomized Sprague-Dawley rats. Sci World J
2005; 5: 652–660
12. Ranganathan N, Patel BG, Ranganathan P et al. In vitro and in vivo assess-
ment of intraintestinal bacteriotherapy in chronic kidney disease. ASAIO J
2006; 52: 70–79
13. Vaziri ND, Wong J, Pahl M et al. Chronic kidney disease alters intestinal
Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
microbial flora. Kidney Int 2013; 83: 308–315
14. Rossi M, Klein K, Johnson DW et al. Pre-, pro-, and synbiotics: do they have
a role in reducing uremic toxins? A systematic review and meta-analysis. Int
J Nephrol 2012; 2012: 631–673
15. Sugano N, Wakino S, Kanda T et al. T-type calcium channel blockade as a
therapeutic strategy against renal injury in rats with subtotal nephrectomy.
Kidney Int 2008; 73: 826–834
16. Saito S, Shimizu H, Yisireyili M et al. Indoxyl sulfate-induced activation of
( pro)renin receptor is involved in expression of TGF-β1 and α-smooth
muscle actin in proximal tubular cells. Endocrinology 2014; 155:
1899–1907
17. Itoh Y, Ezawa A, Kikuchi K et al. Protein-bound uremic toxins in hemodi-
alysis patients measured by liquid chromatography/tandem mass spec-
trometry and their effects on endothelial ROS production. Anal Bioanal
ORIGINAL ARTICLE
Chem 2012; 403: 1841–1850
18. Wu L, Cox A, Roe C et al. Transforming growth factor β1 and renal injury
following subtotal nephrectomy in the rat: role of the renin-angiotensin sys-
tem. Kidney Int 1997; 51: 1553–1567
19. Kelly DJ, Zhang Y, Gow R et al. Tranilast attenuates structural and function-
al aspects of renal injury in the remnant kidney model. J Am Soc Nephrol
2004; 15: 2619–2629
20. Nagashima K, Hisada T, Sato M et al. Application of new primer-enzyme
combinations to terminal restriction fragment length polymorphism profil-
ing of bacterial populations in human feces. Appl Environ Microbiol 2003;
69: 1251–1262
21. Hiraishi A, Iwasaki M, Shinjo H. Terminal restriction pattern analysis of
16S rRNA genes for the characterization of bacterial communities of acti-
vated sludge. J Biosci Bioeng 2000; 90: 148–156
22. Takayama F, Taki K, Niwa T. Bifidobacterium in gastro-resistant seamless
capsule reduces serum levels of indoxyl sulfate in patients on hemodialysis.
Am J Kidney Dis 2003; 41(Suppl): S142–S145
23. Larsen N, Vogensen FK, van den Berg FW et al. Gut microbiota in human
adults with type 2 diabetes differs from non-diabetic adults. PLoS One
2010; 5: e9085
24. Karlin DA, Mastromarino AJ, Jones RD et al. Fecal skatole and indole and
breath methane and hydrogen in patients with large bowel polyps or cancer.
J Cancer Res Clin Oncol 1985; 109: 135–141
25. Zuccato E, Venturi M, Di Leo G et al. Role of bile acids and metabolic ac-
tivity of colonic bacteria in increased risk of colon cancer after cholecystec-
tomy. Dig Dis Sci 1993; 38: 514–519
26. Hall JE. Digestion and absorption in the gastrointestinal tract. In: Hall JE
(ed). Guyton and Hall Textbook of Medical Physiology, 13th edn. Philadel-
phia, PA: Elsevier, 2015, pp. 833–842
27. Vllasaliu D, Casettari L, Fowler R et al. Absorption-promoting effects of
chitosan in airway and intestinal cell lines: a comparative study. Int J
Pharm 2012; 430: 151–160
28. Emmelot CH, van der Waaij D. The dose at which neomycin and poly-
myxin B can be applied for selective decontamination of the digestive
tract in mice. J Hyg (Lond) 1980; 84: 331–340
29. McCall IC, Betanzos A, Weber DA et al. Effects of phenol on barrier
function of a human intestinal epithelial cell line correlate with altered
tight junction protein localization. Toxicol Appl Pharmacol 2009; 241:
61–70
411
 30. Cario E, Gerken G, Podolsky DK. Toll-like receptor 2 controls mucosal in-
flammation by regulating epithelial barrier function. Gastroenterology
2007; 132: 1359–1374
31. Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F et al. Recognition of com-
mensal microflora by toll-like receptors is required for intestinal homeosta-
sis. Cell 2004; 118: 229–241
32. Karczewski J, Troost FJ, Konings I et al. Regulation of human epithelial tight
junction proteins by Lactobacillus plantarum in vivo and protective effects
on the epithelial barrier. Am J Physiol Gastrointest Liver Physiol 2010; 298:
G851–G859
33. Sakai F, Hosoya T, Ono-Ohmachi A et al. Lactobacillus gasseri SBT2055 in-
duces TGF-β expression in dendritic cells and activates TLR2 signal to pro-
duce IgA in the small intestine. PLoS One 2014; 9: e105370
34. Vaziri ND, Goshtasbi N, Yuan J et al. Uremic plasma impairs barrier func-
tion and depletes the tight junction protein constituents of intestinal epithe-
lium. Am J Nephrol 2012; 36: 438–443
35. Vaziri ND, Yuan J, Norris K. Role of urea in intestinal barrier dysfunction
and disruption of epithelial tight junction in CKD. Am J Nephrol 2013; 37:
1–6
36. Sekirov I, Russell SL, Antunes LC et al. Gut microbiota in health and dis-
ease. Physiol Rev 2010; 90: 859–904
37. Cénit MC, Matzaraki V, Tigchelaar EF et al. Rapidly expanding knowledge
on the role of the gut microbiome in health and disease. Biochim Biophys
Acta 2014; 1842: 1981–1992
38. Wong J, Piceno YM, Desantis TZ et al. Expansion of urease and uricase-
contaning, indole- and p-cresol-forming and contraction of short chain
fatty acid-producing intestinal bacteria in ESRD. Am J Nephrol 2014; 39:
230–237
39. Ranganathan N, Ranganathan P, Friedman EA et al. Pilot study of probiotic
dietary supplementation for promoting healthy kidney function in patients
ORIGINAL ARTICLE
with chronic kidney disease. Adv Ther 2010; 27: 634–647
40. Schulman G, Berl T, Beck GJ et al. Randomized placebo-controlled EPPIC
trials of AST-120 in CKD. J Am Soc Nephrol 2015; 26: 1732–1746
41. Jenkins SG, Birk RJ, Zabransky RJ. Differences in susceptibilities of species
of the Bacteroides fragilis group to several beta-lactam antibiotics: indole
production as an indicator of resistance. Antimicrob Agents Chemother
1982; 22: 628–634
412
42. Yu Q, Wang Z, Yang Q. Lactobacillus amylophilus D14 protects tight junc-
tion from enteropathogenic bacteria damage in Caco-2 cells. J Dairy Sci
2012; 95: 5580–5587
43. Ahmad R, Shihab PK, Jasem S et al. FSL-1 induces MMP-9 production
through TLR-2 and NF-κB/AP-1 signaling pathways in monocytic
THP-1 cells. Cell Physiol Biochem 2014; 34: 929–942
44. Xu C, Wang Y, Sun R et al. Modulatory effects of vasoactive intestinal pep-
tide on intestinal mucosal immunity and microbial community of weaned
piglets challenged by an enterotoxigenic Escherichia coli (K88). PLoS One
2014; 9: e104183
45. Schroeder JC, Dinatale BC, Murray IA et al. The uremic toxin 3-indoxyl
sulfate is a potent endogenous agonist for the human aryl hydrocarbon re-
ceptor. Biochemistry 2010; 49: 393–400
46. Heath-Pagliuso S, Rogers WJ, Tullis K et al. Activation of the Ah receptor by
tryptophan and tryptophan metabolites. Biochemistry 1998; 37: 11508–11515
47. Venkatesh M, Mukherjee S, Wang H et al. Symbiotic bacterial metabolites
Downloaded from http://ndt.oxfordjournals.org/ at Serials Dept / Cornell University Medical Library on September 30, 2016
regulate gastrointestinal barrier function via the xenobiotic sensor PXR and
Toll-like receptor 4. Immunity 2014; 41: 296–310
48. Vaziri ND. CKD impairs barrier function and alters microbial flora of the
intestine: a major link to inflammation and uremic toxicity. Curr Opin Ne-
phrol Hypertens 2012; 21: 587–592
49. Kondo M, Tahara A, Hayashi K et al. Renoprotective effects of novel
interleukin-1 receptor-associated kinase 4 inhibitor AS2444697 through
anti-inflammatory action in 5/6 nephrectomized rats. Naunyn Schmiede-
bergs Arch Pharmacol 2014; 387: 909–919
50. Shimizu H, Bolati D, Adijiang A et al. NF-kappaB plays an important role in
indoxyl sulfate-induced cellular senescence, fibrotic gene expression, and
inhibition of proliferation in proximal tubular cells. Am J Physiol Cell Phy-
siol 2011; 301: C1201–C1212
51. Yisireyili M, Shimizu H, Saito S et al. Indoxyl sulfate promotes cardiac fi-
brosis with enhanced oxidative stress in hypertensive rats. Life Sci 2013; 92:
1180–1185
52. Cummings JH. Absorption and secretion by the colon. Gut 1975; 16:
323–329
Received for publication: 4.4.2015; Accepted in revised form: 8.9.2015
A. Yoshifuji et al.