Nephrol Dial Transplant (2004) 19: 2472–2479
doi:10.1093/ndt/gfh452
Original Article
Reduced renal Naþ–Kþ–Cl- co-transporter activity and inhibited
NKCC2 mRNA expression by Leptospira shermani:
from bed-side to bench
Mai-Szu Wu1, Chih-Wei Yang1, Ming-Jen Pan2, Chiz-Tzung Chang1 and Yung-Chih Chen1
1
Division of Nephrology, Chang Gung Memorial Hospital, Keelung and 2Graduate Institute of Veterinary Medicine,
National Taiwan University, Taipei, Taiwan
Abstract
Background. Renal involvement is common in leptospirosis. Interstitial nephritis with interstitial oedema
and mononuclear cellular infiltration are the usual
findings. Clinically, non-oligouric acute renal failure,
hypokalaemia and sodium wasting appear frequently
in leptospirosis. The outer membrane protein from
leptospira has been thought to be responsible for the
disorder. However, the exact mechanisms of renal
involvement are still unclear.
Methods. To address these questions, we first performed detailed in vivo clearance tests in three patients
with leptospirosis (Leptospira shermani) and hypokalaemia to define the tubular lesion. These tests
indicated a defective Naþ–Kþ–Cl- co-transporter and
a poor response to furosemide infusion. We performed
further in vitro studies in a model of medullary thick
ascending limb (mTAL) cultured cells derived from
normal mouse.
Results. Outer membrane protein extract from
L.shermani (0.3 mg/ml) inhibited Naþ–Kþ–Cl- co-transporter activity in mTAL cells (control, 10.15±0.52;
L.shermani, 6.47±0.47 nmol/min/mg protein). The
basolateral Naþ–Kþ ATPase remained intact. Reverse
transcription–polymerase chain reaction (RT–PCR)
further revealed that the outer membrane protein
extract from L.shermani downregulated Naþ–Kþ–Clco-transporter (mNKCC2) mRNA expression. These
changes were dose dependent and could be reversed by
the antibody against outer membrane protein extract
from L.shermani. Experiments with a less pathogenic
strain of leptospira (L.bratislava) and Escherichia
coli did not affect Naþ–Kþ–Cl- co-transporter
activity.
Conclusions. We conclude that L.shermani leptospirosis downregulates mNKCC2 mRNA expression and
Correspondence and offprint requests to: Mai-Szu Wu, Division of
Nephrology, Chang Gung Memorial Hospital, 222, Mai-Chin
Road, Keelung, Taiwan. Email: [email protected]
inhibits Naþ–Kþ–Cl- co-transporter activity in TAL
cells. These alterations may explain the observed
electrolyte disorders in leptospirosis.
Keywords: kidney; leptospirosis; Naþ–Kþ ATPase;
Naþ–Kþ–Cl- co-transporter; thick ascending limb
Introduction
Leptospirosis, a zoonosis caused by the spirochete
leptospira, has 240 serotypes or serova. Leptospirosis
affects most mammals in the world, and it can spread
from animal reservoirs to humans. Carrier animals
may shed virulent leptospira into urine. The pathogen
may persist for months or years. Humans are usually
the final host [1]. Leptospirosis is a self-limited disease
in 85–90% of cases. However, 5–10% of infections can
cause renal tubular damage, microvascular injury,
acute renal failure and interstitial nephritis [2].
Among 24 serogroups of pathogenic leptospires,
Leptospira santarosai serovar shermani (L.shermani) is
the most frequently encountered serovar in both
humans and animals in Taiwan [3]. The protein
fraction extracted from the outer membrane of
leptospira is thought to be responsible for the tubular
damage in leptospira infection.
Leptospirosis is an important but usually ignored
cause of acute renal failure in Taiwan [2]. Renal
involvement in L.shermani infection appears to be a
special form of acute renal failure characterized by a
higher frequency of polyuria and the presence of
hypokalaemia with an elevated urinary fractional
excretion of potassium [4]. Previous indirect evidence
examining the relative integrity of the distal nephron in
guinea pigs suggested that L.interrogans causes a
dysfunctional proximal tubular lesion [5]. It is not
clear whether L.shermani infection stimulates the same
Nephrol Dial Transplant Vol. 19 No. 10 ß ERA–EDTA 2004; all rights reserved
Leptospirosis and Naþ–Kþ–Cl- co-transporter
2473
pathophysiological mechanisms in human. We evaluated three patients with L.shermani infection, one
patient with L.bratislava infection and one patient
with Escherichia coli infection. They all presented with
non-oliguric acute renal failure. During the recovery
phase, hypokalaemia and metabolic alkalosis developed in leptospirosis patients. A renal potassium
wasting resulted in hypokalaemia. The characteristic
clinical manifestations suggested a specific tubulointerstitial lesion caused by L.shermani. We hypothesized
that L.shermani exerted its effects on a specific renal
segment, which created the unique clinical manifestation. To address this possibility, we first defined the
involved renal segments by using classic renal clearance tests. We also performed in vitro experiments
to investigate the underlying cellular mechanisms
involved in renal epithelial transport. In the second
part of the study, we investigated the effect of the outer
membrane protein extract from L.shermani on renal
epithelial transport mechanisms using a model of renal
epithelial cells derived from the normal mouse [6]. A
better understanding of these mechanisms will enhance
our knowledge about infection-related renal tubulointerstitial disease.
Materials and methods
Case reports
Case 1. A 44-year-old male hunter was admitted to the
nephrology ward due to fever, myalgia and a non-oliguric
acute renal failure [serum creatinine (Cr) 3.5 mg/dl;
309 mmol/l]. He had experienced a dog bite while hunting.
A microscopic agglutination test (MAT) confirmed infection
with L.shermani. The fever, myalgia and azotaemia subsided
after penicillin treatment. Metabolic alkalosis (pH 7.55,
PCO2 42.5 mmHg, HCO–3 37 mEq/l) with hypokalaemia
(serum, 2.2 mEq/l; urine, 35 mEq/l) developed during the
recovery phase (Cr 1.4 mg/dl; 124 mmol/l). Several clearance
tests were performed to define the specific lesion along the
nephron (Table 1).
Case 2. A 66-year-old farmer came to our ward reporting 1
week of malaise, anorexia and fever. Hyperbilirubinaemia
and azotaemia (Cr 5.3 mg/dl; 469 mmol/l) were found upon
admission. MAT confirmed L.shermani infection. Hypokalaemia (serum 2.5 mEq/l; urine, 42 mEq/l) due to renal
potassium wasting developed after recovery of renal function (Cr 1.3 mg/dl; 115 mmol/l). Several clearance tests were
performed to define the location of the tubular lesion
(Table 1).
Case 3. A 46-year-old male farmer came to our ward
reporting 2 weeks of fever, jaundice and anorexia. Hyperbilirubinaemia and azotaemia (Cr 4.4 mg/dl; 389 mmol/l) were
found upon admission. Leptospira shermani infection was
confirmed by MAT. Hypokalaemia (serum, 2.6 mEq/l; urine,
38 mEq/l) due to renal potassium wasting developed after
recovery of renal function (Cr 1.6 mg/dl; 141 mmol/l). Several
clearance tests were performed to define the location of the
tubular lesion (Table 1).
Case 4. A 77-year-old male patient came to our ward after 1
week of fever, abdominal pain and disturbances of consciousness. The patient developed renal failure (Cr 8.3 mg/dl;
734 mmol/l) and hepatic failure. MAT showed increased titres
of 1:400 to L.bratislava. Leptospira DNA was detected by
polymerase chain reaction (PCR) in urine and blood samples. Renal function and hyperbilirubinaemia (Cr 1.4 mg/dl;
124 mmol/l) recovered almost completely after 1 month of
penicillin therapy. Hypokalaemia (serum 2.6 mEq/l; urine,
31 mEq/l) developed during the recovery phase. Several
clearance tests were performed to define the tubular lesions
(Table 1).
Case 5. A 46-year-old female visited our emergency department complaining of intermittent fever and dyspnoea.
Abnormal renal function (Cr 5.5 mg/dl; 486 mmol/l) was
noted upon arrival. Blood culture revealed E.coli infection.
Leptospira was not found by serology or by PCR. Acute
renal failure due to E.coli sepsis was diagnosed. Renal
function (Cr 1.3 mg/dl; 115 mmol/l) improved after antibiotic
treatment, and renal clearance tests were performed to
evaluate tubular function during recovery from acute tubular
necrosis.
Diagnosis of leptospirosis
In patients with suspected leptospirosis, diagnosis was made
by 4-fold changes of MAT titre in paired sera between the
acute and convalescence phase, a single titre in MAT >1:400
or the demonstration of leptospira DNA in urine or in blood
samples by PCR.
Clearance tests
Several in vivo clearance tests were used to define the tubular
lesion as previously described [7]. Briefly, a bicarbonate test
was used to test proximal tubular bicarbonate reabsorption
and collecting tubular proton pump function. A furosemide
Table 1. Clearance tests in patients with L.shermani, L.bratislava and E.coli infection
Bicarbonate loading
Furosemide
FeHCO–3
Cosm
U–B PCO2
Patient
1
2
3
4
5
4.0%
1.2%
2.1%
20.9%
2.0%
27
29.2
34.2
30.5
29.5
Thiazide
CCl
Cosm
CCl
Before
After
Before
After
Before
After
Before
After
4.1%
6.2%
4.4%
4.5%
3.8%
4.8%
6.4%
4.7%
13.2%
10.7%
3.3%
6.5%
4.5%
4.7%
3.2%
3.5%
6.7%
4.4%
16.6%
13.5%
4.53%
6.1%
4.6%
4.3%
4.1%
12.1%
13.5%
11.5%
14.1%
10.2%
2.9%
6.3%
5.1%
3.9%
3.6%
17.9%
13.3%
13.5%
14.5%
12.5%
2474
test was performed to assess furosemide-sensitive sodium
absorption along the thick ascending limb (TAL). A thiazide
test was performed to evaluate distal tubular sodium
reabsorption.
Cultured cells
The experiments were carried out on subcultured cells
derived from isolated medullary TALs (mTALs) and cortical
TALs (cTALs) microdissected out from the kidney of
1-month-old C57BL/6 mice as previously described [6].
Subcultured mTAL cells were routinely grown in a modified
culture medium [Dulbecco’s modified Eagle’s medium
(DMEM): Ham’s F12, 1:1 (v/v); 60 nM sodium selenate;
5 mg/ml transferin; 2 mM glutamine; 5 mg/ml insulin; 50 nM
dexamethasone; 1 nM triiodothyronine; 10 ng/ml epidermal
growth factor; 2% fetal calf serum; 20 mM HEPES, pH 7.4]
at 37 C in a 5% CO2–95% air atmosphere. All experiments
were performed between the 6th and 15th passages on sets
of confluent cells grown on Petri dishes.
RNA extraction and reverse transcription–PCR
(RT–PCR)
RNA was extracted from isolated mTAL and cTAL segments
microdissected from an adult mouse kidney and from
confluent mTAL and cTAL cultured cells by the method of
Chomczynski and Sacchi as previously described [8]. Total
RNA concentration was treated with RNase-free DNase I
(Boehringer Mannheim, Germany) at 37 C for 30 min and
RNA concentration was evaluated by spectrophotometry.
RNA (100 mg) was reverse transcribed with avian myeloblastosis virus reverse transcriptase (Boehringer Mannheim)
at 42 C for 60 min. A 150 ng aliquot of cDNA and nonreverse-transcribed RNA were amplified for 32 cycles in
100 ml total volume containing 50 mM KCl, 20 mM Tris–HCl
pH 8.4, 10 mM dNTP, 1.5 mM MgCl2, 1 U of Taq polymerase and 10 pmol of inducible nitric oxide sythetase (iNOS)
primers. The thermal cycling program was as follows: 94 C
for 1 min, 60 C for 1 min and 72 C for 3 min. The two
primers from the NKCC2 gene [8] were as follows: antisense
strand 50 -CTT GGC TTC GGT TTT AGA TGA CCC G-30 ,
sense strand 50 -GCA ATG CTG GCA TTT AGA CCC TCC
G-30 . Amplification products (406 bp) were run on a 4%
agarose gel with ethidium bromide and photographed. As a
control, the identity of the amplified products of b-actin
(460 bp) was processed at the same time.
Quantitation of PCR products:
competitive PCR assay
Competitive RT–PCR was performed for the measurement
of NKCC2 and -actin mRNA. The test template for all
PCRs was an aliquot of cDNA collected from mTAL cells.
To quantitate the tested cDNAs, various amounts of mutant
cDNA templates were added to compete with test cDNA
on an equimolar basis, as previously described. For
NKCC2 and -actin, deletion cDNA mutant templates were
developed to create 105 and 103 bp deletions in the middle
of the molecules, resulting in mutant cDNAs of 301 and
357 bp, respectively. Following agarose gel electrophoresis,
amplification bands stained by ethidium bromide were
quantitated from the film negative by scanning densitometry.
M.-S. Wu et al.
The ratio of mutant to wild-type band density was calculated
for each lane and plotted as a function of the amount of
initial mutant template added to the reaction. The amount of
cDNA was derived from linear regression analysis with
duplicate or triplicate assays. The mean values for assays
were expressed as a percentage change from the control.
Preparation of outer membrane protein
extract of leptospira
Two commonly encountered pathogenic leptospira serovars,
L.shermani and L.bratislava, and a non-pathogenic L.biflexa
serovar patoc were obtained from ATCC (MD) and cultured
in Johnson–Harris bovine serum albumin–Tween-80 medium
(Difco, MI). Leptospires were counted using dark-field
microscopy as previously described [9]. Protein extract from
the outer membrane of leptospires was obtained as previously described [10]. Briefly, leptospira cells were cultured
for 5–7 days at 28 C until a cell density of 108/ml was
obtained. A 200 ml aliquot of cultured cells was centrifuged
at 15 000 g for 30 min and resuspended in 8 ml of 1 M NaCl.
The bacteria were observed under dark-field microscopy until
conversion to spherical form, were centrifuged again at
15 000 g for 30 min and were resuspended in 8 ml of distilled
water. Treatment of this cell suspension with 0.04% sodium
dodecyl sulfate (SDS) solubilized the outer membrane. The
cells were removed by centrifugation, and the supernatant
was filtered through a 0.45 mm membrane filter and
lyophilized. Protein concentration was measured by the
Bradford method (Protein Assay, Bio-Rad Laboratories,
CA). Each extraction procedure yielded 0.5 mg of protein
per serovar.
Antiserum for L.shermani
Antiserum against L.shermani was prepared in New Zealand
white rabbits by immunizing outer membrane protein as
previously described [11]. A 400 mg aliquot of detergentsoluble protein was mixed with Freund’s complete adjuvant,
and injected intradermally into various sites along the back
of the rabbit. A booster of the same dose was given 2 weeks
later. The animals were bled 2 weeks after the last
inoculation. The titre of the antiserum was determined by
an MAT which revealed a titre of 1:10 000.
Transport studies
The potassium-transporting capacities of the cells were
studied by radiolabelled rubidium (86Rbþ), used as the
tracer for potassium influx and efflux studies. Confluent cells
in 24-well trays were washed with 1 ml of modified
phosphate-buffered saline (PBS, in mM: NaCl 136; KCl 5;
Na2HPO4 1; glucose 5; HEPES 10; pH 7.4) pre-warmed to
37 C. The medium was removed and rubidium influx was
initiated by adding 1 ml of PBS containing 2 mCi/ml 86Rbþ.
The reaction was stopped by removing the solution and
performing three rinses with 1 ml of ice-cold 100 mM MgCl2
solution. Cells were solubilized with 0.5 ml of 1 M NaOH and
radioactivity was determined by scintillation counting. The
Kþ influx was measured by incubating cells without (total
influx) or with 0.5 mM ouabain or ouabain plus 104 M
furosemide in order to determine the ouabain-sensitive influx
(Os) mediated by the Naþ–Kþ ATPase pumps, and the
Leptospirosis and Naþ–Kþ–Cl- co-transporter
ouabain-resistant furosemide-sensitive influx
mediated by Naþ–Kþ–Cl- co-transport [12].
2475
(Or-Fs)
that a TAL defect is specific for L.shermani infection.
Neither L.bratislava nor E.coli infection show the same
tubular dysfunction.
Cell viability
Cell viability was estimated by a tetrazolium-based 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay
(MTT) [13] to determine the non-specific cytotoxicity of
the outer membrane protein extract of L.shermani and
L.bratislava to mTAL cells. Cells were seeded in 96-well
plates (Corning Co., Corning, NY). After culturing for
3 days, cells were exposed to various concentrations (0.1–
1 mg/ml) of outer membrane protein extract. Following a 48 h
incubation, 40 ml of 5 mg/ml 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide was added to each well. After
2 h at 37 C, the cells were lysed by adding 100 ml of 20%
(w/v) SDS and 50% (v/v) N,N-dimethylformamide (pH 4.7)
and incubated overnight at 37 C. The absorbance at 570 nm
was measured for each well using a Dynex microplate reader.
The reported cell viability was the percentage of viable
cells in comparison with the control wells. Duplicate measurements were made for each test on at least two separate
occasions.
Characteristics of cultured TAL cells
We examined whether L.shermani specifically inhibits
Naþ–Kþ–Cl- co-transporter activity in TAL cells
and explored the cellular mechanisms of this effect.
We performed in vitro studies to explore these possibilities. We microdissected the mTAL from normal
mice and put the segments into culture (Figure 1A).
Confluent mouse mTAL cells grown on collagencoated Petri dishes formed a monolayer of cuboid
shaped cells and formed small domes (Figure 1A) [6].
Cultured cTAL cells were also cultivated using similar
methods. RT–PCR was used to detect NKCC2, the
kidney-specific Naþ–Kþ–Cl- co-transporter mRNA.
Statistical analysis
Results are expressed as means±SD from (n) experiments
performed in duplicate or triplicate. Significant differences
between groups were analysed by Student’s t-test and oneway ANOVA. The statistical analysis was performed using
the StatviewTM program (Macintosh). A P-value <0.05 was
considered significant.
Results
Clearance test
The patients with L.shermani infection presented with
acute renal failure, hyperbilirubinaemia and fever
during the acute phase. Both renal and hepatic failure
recovered gradually after institution of penicillin
therapy. However, metabolic alkalosis and hypokalaemia developed during the recovery phase. The hypokalaemia was secondary to renal potassium wasting.
Several clearance tests were performed to define the
tubular lesion. A defect in TAL sodium reabsorption
was suspected because of the poor response to
furosemide infusion. The chloride clearance (CCl)
and osmolality clearance (Cosm) remained unchanged
after furosemide infusion (Table 1). Patient 4 was
infected by another species of leptospira, L.bratislava.
He presented evidence of renal tubular acidosis with
defective bicarbonate reabsorption in proximal tubules
since fractional excretion of bicarbonate was as high as
20.9% after bicarbonate infusion. A normal response
to furosemide and thiazide demonstrated that the
distal nephron remained intact. Patient 5 recovered
from the E.coli sepsis with acute renal failure,
hyperbilirubinaemia and fever. The patient responded
normally to bicarbonate loading, and to furosemide
and thiazide infusion (Table 1). These results suggest
Fig. 1. Properties of cultured mouse mTAL cells. (A) Left panel: a
microdissected mTAL segment. Right panel: confluent TAL cells
grown on Petri dishes formed layers of cuboid-shaped cells and
formed small domes (magnification 250). (B) Illustration of an
ethidium bromide-stained 4% agarose gel showing the amplified
products of the expected size (406 bp) obtained with the NKCC2
primers in microdissected mTAL (lane 1) and cultured TAL cells
(7th passage) (lane 2). As controls, no band was detected using nonreverse-transcribed RNA from cultured cells (lane 3) or by omitting
cDNA (lane 4). The expression of -actin was also shown as an
internal control. Molecular weight standards (M) were the 1 kb
ladder from Gibco-BRL. (C) Time course of 86Rbþ influx performed
on cultured mTAL cells grown on Petri dishes and incubated without
() or with ouabain (), or with ouabain plus furosemide (m). Each
point is the mean from five separate experiments performed in
duplicate.
2476
M.-S. Wu et al.
Substantial amounts of NKCC2 transcripts were
detected in both microdissected mTAL and cultured
mTAL cells (Figure 1B). cTAL segment and cultured
cTAL cells also expressed substantial amounts of
NKCC2 (data not shown). A single 406 bp band of
the expected size was amplified in both microdissected mTALs and cultured cells. As a negative
control, samples lacking amplified products were
obtained by using non-reversed-transcribed RNA from
cultured TAL cells or by omitting cDNA (Figure 1B).
The different components of 86Rbþ influx measured
in the absence or presence of ouabain and furosemide allowed us to distinguish between the ouabainsensitive (Os) component of 86Rbþ influx mediated by
the Naþ–Kþ ATPase pumps and the ouabain-resistant
furosemide-sensitive (Or-Fs) component of 86Rbþ
influx reflecting Naþ–Kþ–Cl- co-transport activity
(Figure 1C). In all cases, the influx of 86Rbþ increased
linearly with time for up to 10 min. These results
indicate that cultured TAL cells is an appropriate
model to test the effect of L.shermani outer membrane
extract on Naþ–Kþ–Cl- co-transport.
The effect of L.shermani outer membrane
extract in Rbþ transport
We analysed the effects of membrane extract from
L.shermani, L.bratislava and E.coli on the uptakes
of Rbþ mediated by the Naþ–Kþ ATPase pump and
the Naþ–Kþ–Cl- co-transporter. Pre-incubation with
outer membrane extract (0.3 mg/ml) from L.shermani
for 48 h decreased the Or-Fs component of 86Rbþ
influx, which represents Naþ–Kþ–Cl- co-transport in
mTAL cells (control, 10.15±0.52; L.shermani, 6.47±
0.47 nmol/min/mg protein, P<0.001) (Figure 2A). The
outer membrane protein extract from L.bratislava
(10.21±0.57 nmol/min/mg protein) and E.coli (9.80±
0.48 nmol/min/mg protein) did not affect influx
(Figure 2A). Similarly, the outer membrane extracts
did not alter the Os component of 86Rbþ influx, which
represented Naþ–Kþ ATPase (Figure 2B).
Competitive RT–PCR
mRNAs for mNKCC2 and -actin were measured by
competitive RT–PCR, and the results were expressed
as percentage change from controls. Significant
changes in mRNA levels were found in mNKCC2
but not in control -actin mRNA. At 48 h after
adding L.shermani outer membrane protein to mTAL
cells at 0.1, 0.2 and 0.3 mg/ml, there were 97.7, 98.5
and 98.8% decreases in mNKCC2 mRNA, respectively, compared with controls (P < 0.01) (Figure 3).
Changes in these levels were compared by optical
density obtained by scanning densitometry of the
PCR product at the exponential phase. However, the
outer membrane extract at levels up to 0.3 mg/ml
did not alter the expression of NKCC2 in cTAL cells
(data not shown).
Fig. 2. Effect of membrane extracts from L.shermani, L.bratislava
and E.coli on 86Rbþ influx. The (A) ouabain-resistant furosemidesensitive (Or-Fs) and the (B) ouabain-sensitive (Os) component
of 86Rbþ influx (nmol/min/mg protein) were measured on sets of
confluent mTAL cells grown on Petri dishes and incubated without
(control) or with 0.3 mg/ml membrane extract from L.shermani (LS),
0.3 mg/ml membrane protein extract from L.bratislava (LB) or
0.3 mg/ml membrane protein extract from E.coli (E.coli). Values are
the means±SD of 10 separate experiments performed in duplicate.
***P<0.001 vs control values.
Effect of L.shermani antiserum on mNKCC2 mRNA
expression and Naþ–Kþ–Cl- co-transporter activity
To clarify the specificity of the decreased mRNA
levels, a rabbit antiserum raised against the outer
membrane protein of L.shermani was incubated with
the outer membrane protein extract from L.shermani
before adding it to the cell culture medium. After 48 h
incubation, the inhibitory effect of L.shermani membrane extract on mNKCC2 mRNA expression was
partially abolished by the antiserum (Figure 4). Adding
antiserum alone slightly decreased mNKCC2 mRNA
expression (Figure 4). These results suggest that the
inhibitory effect was due at least in part to a direct
action of outer membrane protein extract.
Besides preventing changes in mRNA, antiserum to
L.shermani also partially abolished decreases in Naþ–
Kþ–Cl- co-transporter activity (control, 10.94±0.87;
L.shermani, 6.83±0.47; antiserum þ L.shermani, 9.12±
0.79 nmol/min/mg protein) (Figure 5). These findings
Leptospirosis and Naþ–Kþ–Cl- co-transporter
Fig. 3. Representative competitive RT–PCR results of mNKCC2
and -actin by L.shermani outer membrane proteins. The outer
membrane protein induced 97.7, 98.5 and 98.8% decreases in
mNKCC2 mRNA at doses of 0.1, 0.2 and 0.3 mg/ml, respectively,
compared with controls. M ¼ marker; control (lane 1); LS 0.1 (lane
2) ¼ L.shermani 0.1 mg/ml; LS 0.2 (lane 3) ¼ L.shermani 0.2 mg/ml; LS
0.3 (lane 4) ¼ L.shermani 0.3 mg/ml.
provided further evidence that the inhibitory effect on
Naþ–Kþ–Cl- co-transporter activity was mediated
mostly, or perhaps entirely, by the L.shermani outer
membrane protein extract.
Cell viability
To ensure that the changes induced by L.shermani
were not related to cell damage, we used cell viability
as an index of cell injury [13]. Viability was measured
in cultured mTAL cells incubated without or with
various concentrations (0.1–0.3 mg/ml) of L.shermani
membrane extract for 48 h. The percentage of viable
cells was not significantly different between L. shermani membrane extract-treated and untreated cells
(0.1 mg/ml, 98±4%; 0.2 mg/ml, 99±5%; 0.3 mg/ml,
101±8% of viable cells vs untreated cells n ¼ 5).
These results indicated that cultured mTAL cells
remained viable after 48 h incubation with L.shermani,
suggesting that the observed decreases in mNKCC2
expression were not related to cell damage.
Discussion
Pathogenic Leptospira outer membrane contains lipopolysaccharide, glycolipid and lipoproteins that determine virulence and are the main targets for immunity.
2477
Fig. 4. Effect of L.shermani outer membrane protein extract and
antiserum on mNKCC2 expression in mTAL cells. (A) Ethidium
bromide-stained 4% agarose gel of RT–PCR showing the amounts
of amplified products of mNKCC2 compared with that of -actin,
used as control, in cultured mTAL cells incubated without (lane 1),
or with 0.1 mg/ml (lane 2), 0.2 mg/ml (lane 3), 0.3 mg/ml (lane 4),
antiserum (lane 5) and 0.3 mg/ml outer membrane protein extract
plus antiserum for the outer membrane protein extract (lane 6).
Molecular weight standards (M) were the 1 kb ladder from GibcoBRL. (B) The bars represent the amplified products ratio for
mNKCC2 and -actin over untreated in L.shermani outer membranetreated cells. Control ¼ untreated cells; LS 0.1 ¼ L.shermani 0.1 mg/
ml; LS 0.2 ¼ L.shermani 0.2 mg/ml; LS 0.3 ¼ L.shermani 0.3 mg/ml;
antiserum ¼ antiserum against L.shermani outer membrane protein
extract; antiserum þ LS 0.3 ¼ antiserum against L.shermani outer
membrane protein extract plus L.shermani 0.3 mg/ml.
Leptospira endotoxin derived from the outer membrane from virulent strains of leptospira may provide a
mechanism that explains the pathogenesis of leptospirosis and has been the focus of recent leptospiral
research [9]. The leptospira endotoxin differs from
that of Gram-negative bacteria, which lacks 2-keto-3deoxyoctonic acid (KDO), an authentic chemical
component of endotoxin. KDO induces less pyrogenecity and less lethality when administered to
animals, whereas it may induce necrosis of mammalian
cells [14]. In contrast, the outer membrane of avirulent
leptospira does not contain endotoxin-like components. Peptidoglycan, a protein extracted from the
outer membrane endotoxin of L.interrogans, activates
macrophages [14] and induces the release of tumour
necrosis factor-a from human monocytes [15].
2478
M.-S. Wu et al.
-
Fig. 5. Effect of outer membrane protein extract from L.shermani,
and antiserum against L.shermani outer membrane extract on 86Rbþ
influx. The ouabain-resistant furosemide-sensitive (Or-Fs) component of 86Rbþ influx (nmol/min/mg protein) was measured on sets of
confluent mTAL cells grown on Petri dishes and incubated without
(control), with 0.3 mg/ml membrane extract from L.shermani (LS 0.3),
with antiserum against L.shermani outer membrane protein extract
(antiserum), or with antiserum against L.shermani outer membrane
protein extract plus L.shermani 0.3 mg/ml (antiserum þ LS). Values
are the means ± SD of 10 separate experiments performed in
duplicate. ***P < 0.001 vs control values.
The clinical clearance tests in the present study
suggested that there were defective responses to
furosemide in L.shermani leptospirosis patients. These
results indicated that the tubular lesion is mainly
located in TAL. The renal tubular target of L.shermani
leptospirosis may be the Naþ–Kþ–Cl- co-transporter,
which is uniquely found in TAL cells. Because of
this finding, we further evaluated the effect of
L.shermani outer membrane extract on the Naþ–Kþ–
Cl- co-transporter. We found that the L.shermani outer
membrane protein extract reduced Naþ–Kþ–Cl- cotransporter activity and mRNA expression.
Neutralizing antibody reversed the effect of outer
membrane extract on Naþ–Kþ–Cl- co-transporter.
These results suggest that the outer membrane protein
extract exerts a specific effect on TAL cells. These
delicate transport processes are well regulated by many
intra-renal hormones, and keep our internal milieu in
homeostasis. The inhibitory effect of the L.shermani
outer membrane on Naþ–Kþ–Cl- co-transporter may
provide pathogenic clues regarding the frequent
appearance of renal potassium wasting [16] and
concentration defects [17] in patients with leptospirosis. The inhibition of the Naþ–Kþ–Cl- co-transporter
results in more distal sodium delivery and subsequently
induces renal potassium loss. A reduction in Naþ–Kþ–
Cl co-transport activity also compromises medullary
hypertonicity and may be responsible for the lack of
vasopressin responsiveness.
Ionic transport additionally acts as a cellular signal
in many cellular functions [18]. Cell volume regulation,
apoptosis, cytokine release and many other cellular
functions are regulated by the ionic transport process.
Thus, alterations in ionic transport may affect many
cellular functions and lead to acute or chronic renal
dysfunction [19]. Our recent studies examining leptospirosis effects on cytokine and chemokine expression
further supported this view by indicating that altered
transport activity may be the first event in acute and
chronic tubulointerstitial renal disease during leptospirosis [9]. Together, these findings point to a role for
infection in renal disease.
Importantly in the present study, the inhibitory
effect was limited to the medullary segment but not the
cortical region. Recent studies indicated that mouse
TAL expresses a total of six NKCC2 isoforms
generated by the combinatorial association of three 50
exon cassettes (A, B and F) with two alternative 30 ends
[20]. The two 30 ends predict C-terminal cytoplasmic
domains of 129 amino acids (the C4 C-terminus)
and 457 amino acids (the C9 terminus). The three
C9 isoforms (A9/F9/B9) all express Naþ–Kþ–2Clco-transport activity, whereas C4 isoforms (A4/F4/
B4) are non-functional in Xenopus oocytes. Activation
or inhibition of protein kinase A does not affect the
activity of the F9 isoforms [20]. This finding suggests
that the outer membrane protein extract may have
a specific isoform of the NKCC2 as a target. The
affected isoform was limited to the medullary segment.
Our previous studies also indicated that the affected
NKCC2 was cAMP-dependent, potassium-dependent
and bumetamide-sensitive [6]. The characteristics of
NKCC2 suggested that it may be isoform A, which
appeared mainly in the apical membrane of mTAL
cells. These results indirectly suggest that specific isoforms were affected by the outer membrane protein
extract. Further experiments will be necessary to
specify the affected isoforms.
The current study provides a good example of the
fact that clinical findings and laboratory investigations
can be linked. Potential mechanisms can be verified
in the laboratory, and laboratory findings can be
applied towards clinical problems. Using this method,
further investigations and clinical observations can be
developed.
In conclusion, the outer membrane protein extract
from L.shermani is able to reduce the expression of
mNKCC2 mRNA and Naþ–Kþ–Cl- co-transporter
activity. This effect may represent the initial event in
tubulointerstitial injury in L.shermani leptospirosis
renal injury.
Acknowledgements. The authors would like to thank Yi-Ching
Ko, Chung-Tseng Huang and Hsiau-Mai Yu for their excellent and
dedicated technical support. This study was supported by grants
from the National Science Council, Taiwan.
Conflict of interest statement. None declared.
Leptospirosis and Naþ–Kþ–Cl- co-transporter
References
1. Farr RW. Leptospirosis. Clin Infect Dis 1995; 21: 1–6; quiz 7–8
2. Yang CW, Pan MJ, Wu MS et al. Leptospirosis: an ignored
cause of acute renal failure in Taiwan. Am J Kidney Dis 1997;
30: 840–845
3. Pan MJ, Yang CW, Li CL. Leptospirosis. Infect Dis Rep 1996;
12: 389–394
4. Lin CL, Wu MS, Yang CW, Huang CC. Leptospirosis
associated with hypokalaemia and thick ascending limb
dysfunction. Nephrol Dial Transplant 1999; 14: 193–195
5. Ballard SA, Williamson M, Adler B, Vinh T, Faine S. Interactions of virulent and avirulent leptospires with primary
cultures of renal epithelial cells. J Med Microbiol 1986; 21: 59–67
6. Wu MS, Bens M, Cluzeaud F, Vandewalle A. Role of F-actin
in the activation of Na(þ)–K(þ)–Cl cotransport by forskolin
and vasopressin in mouse kidney cultured thick ascending limb
cells. J Membr Biol 1994; 142: 323–336
7. Wu MS, Hong JJ, Lin JL, Yang CW, Chien HC. Multiple tubular
dysfunction induced by mixed Chinese herbal medicines containing cadmium. Nephrol Dial Transplant 1996; 11: 867–870
8. Igarashi P, Whyte DA, Li K, Nagami GT. Cloning and
kidney cell-specific activity of the promoter of the murine
renal Na–K–C1 cotransporter gene. J Biol Chem 1996; 271:
9666–9674
9. Yang CW, Wu MS, Pan MJ et al. The leptospira outer
membrane protein LipL32 induces tubulointerstitial nephritismediated gene expression in mouse proximal tubule cells. J Am
Soc Nephrol 2002; 13: 2037–2045
10. Bey RF, Johnson RC. Immunogenicity and humoral and cellmediated immune responses to leptospiral whole cell, outer
envelope, and protoplasmic cylinder vaccines in hamsters and
dogs. Am J Vet Res 1982; 43: 835–840
2479
11. Brown JA, LeFebvre RB, Pan MJ. Protein and antigen profiles
of prevalent serovars of Leptospira interrogans. Infect Immun
1991; 59: 1772–1777
12. Wu MS, Yang CW, Bens M et al. Cyclosporine stimulates
Naþ–Kþ–Cl cotransport activity in cultured mouse medullary thick ascending limb cells. Kidney Int 2000; 58: 1652–1663
13. Hansen MB, Nielsen SE, Berg K. Re-examination and further
development of a precise and rapid dye method for measuring
cell growth/cell kill. J Immunol Methods 1989; 119: 203–210
14. Isogai E, Hirose K, Kimura K et al. Role of platelet-activatingfactor (PAF) on cellular responses after stimulation with
leptospire lipopolysaccharide. Microbiol Immunol 1997; 41:
271–275
15. Cinco M, Vecile E, Murgia R, Dobrina P, Dobrina A.
Leptospira interrogans peptidoglycans induce the release of
tumor necrosis factor alpha from human monocytes. FEMS
Microbiol Lett 1996; 138: 211–214
16. Seguro AC, Lomar AV, Rocha AS. Acute renal failure of
leptospirosis: nonoliguric and hypokalemic forms. Nephron
1990; 55: 146–151
17. Abdulkader RC, Seguro AC, Malheiro PS, Burdmann EA,
Marcondes M. Peculiar electrolytic and hormonal abnormalities in acute renal failure due to leptospirosis. Am J Trop Med
Hyg 1996; 54: 1–6
18. O’Neill WC. Physiological significance of volume-regulatory
transporters. Am J Physiol 1999; 276: C995–C1011
19. Warntges S, Grone HJ, Capasso G, Lang F. Cell volume
regulatory mechanisms in progression of renal disease. J
Nephrol 2001; 14: 319–326
20. Plata C, Mount DB, Rubio V, Hebert SC, Gamba G. Isoforms
of the Na–K–2Cl cotransporter in murine TAL II. Functional
characterization and activation by cAMP. Am J Physiol 1999;
276: F359–F366
Received for publication: 17.11.03
Accepted in revised form: 23.4.04