how to search the world wide web using yahoo! search

Nephrol Dial Transplant (2011) 26: 2127–2137doi: 10.1093/ndt/gfq687Advance Access publication 25 November 2010

Protective effects of tubular liver-type fatty acid-binding proteinagainst glomerular damage in murine IgA nephropathy

Nan Zuo1,2, Yusuke Suzuki1, Takeshi Sugaya1,3, Ken Osaki1, Yasuhiko Kanaguchi1, LiNing Wang2

and Yasuhiko Tomino1

1Division of Nephrology, Department of Internal Medicine, Juntendo University Faculty of Medicine, Tokyo, Japan, 2Division ofNephrology, The First Affiliated Hospital of China Medical University, Shenyang, China and 3CMIC Company Limited, Tokyo, Japan

Correspondence and offprint requests to: Dr. Yasuhiko Tomino, E-mail: [email protected]

AbstractBackground. Liver-type fatty acid-binding protein (L-FABP) in proximal tubules was reported to have renopro-tective roles in experimental tubulointerstitial diseases viaits anti-oxidative properties. Since tubuloglomerular cross-talk was recently discussed in the progression of renal dis-eases, to investigate whether tubular L-FABP may have animpact on the progression of glomerular damage, we in-duced IgA nephropathy (IgAN) in mice (Tg) transgeni-cally tubular overexpressing human L-FABP (hL-FABP).Methods. We reconstituted IgAN by bone marrow trans-plantation (BMT) from IgAN-prone mice into Tg andwild-type (WT) mice. Renal damage was evaluated at 6and 12 weeks after BMT. During in vitro experiments, me-sangial cells (MC) were stimulated by aggragated IgA(AIgA), and their supernatants (AIgA-MC medium) werecollected. Stable cell line of mouse proximal tubular cell(mProx) transfected with or without hL-FABP genewas cul-tured with the AIgA-MC medium.Results. Although mesangial IgA deposition and serumIgA level were not different between WT (WT/ddY) andTg (Tg/ddY) recipients, WT/ddY mice showed a signifi-cantly higher urinary albumin level and mesangial matrixexpansion with a significantly higher glomerular damagescore. Furthermore, CD68+ macrophage infiltration wasalso significantly attenuated in Tg/ddYmice. Up-regulationof renal hL-FABP was associated with significant suppres-sion of renal heme oxygenase-1 (HO-1) expression and ac-cumulation of 4-hydroxy-2-nonenal (4-HNE) and MCP-1expression in Tg/ddY mice. In vitro experiments showedthat AIgA-MC medium and recombinant TNF-α signifi-cantly up-regulated hL-FABPexpression, whichwas partial-ly blocked by anti-TNF-α antibody, and major mediators ofoxidative stress (HO-1 and 4-HNE) and inflammation(MCP-1). Importantly, such up-regulation of the mediatorsin mProx with hL-FABP was significantly suppressed muchmore than that in mProx.Conclusions. Tubular L-FABP activated by MC-originhumoral factors may lessen progression of glomerular dam-

age at early stages of IgAN by reducing oxidative stress andinflammatory mediators.

Keywords: IgA nephropathy; liver-type fatty acid-binding protein (L-FABP); oxidative stress; tubuloglomerular cross-talk

Introduction

Liver-type fatty acid-binding protein (L-FABP) is a 14.4-kDa protein expressed in human proximal tubules andknown to facilitate the cellular uptake, transport and me-tabolism of fatty acids and regulate the expression of genesinvolved in lipid metabolism [1,2]. Moreover, L-FABP canbind long-chain fatty acid oxidation products with highaffinity and may be a crucial cellular antioxidant molecule[3]. Because L-FABP is not expressed in the kidneys ofrodents, human L-FABP (hL-FABP) chromosomal trans-genic (Tg) mice were generated [4]. Several experimentalmodels using the hL-FABP Tg mice with renal diseases,such as protein overload [4], unilateral ureteral obstruction[5] and diabetic nephropathy [6], showed that tubularoverexpression of hL-FABP plays renoprotective rolesvia its anti-oxidative properties in such tubulointerstitialdiseases. However, the pathophysiological role of L-FABPin glomerular damage of glomerulonephritis remains to bedetermined.

IgA nephropathy (IgAN) is defined as glomeruloneph-ritis with predominant deposition of IgA in the glomerularmesangium [7], and frequently progresses to end-stagerenal failure [8]. Although IgAN is the most common pri-mary glomerular disease worldwide, no specific therapy isavailable to date, and the exact etiological mechanismremains elusive. Recent studies on the pathogenic mechan-isms of IgAN have stressed the role of intra-renal oxidativestress [9]. Treatments with antioxidants such as fish-oilextracts are reported to be beneficial in both experimental[10] and human IgAN [9]. Furthermore, Kobori et al. [11]recently showed that immunoreactivities of intra-renal

© The Author 2010. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.For Permissions, please e-mail: [email protected]

Dow

nloaded from https://academ

ic.oup.com/ndt/article-abstract/26/7/2127/1892815 by guest on 05 April 2019

heme oxygenase-1 (HO-1) and 4-hydroxy-2-nonenal (4-HNE), markers of oxidative stress, were significantly high-er in IgAN patients than those in the control group. Inaddition, they also provided direct evidence for the roleof activated reactive oxygen species (ROS) in the earlystages of experimental IgAN through interaction with theactivated renin–angiotensin system (RAS) [12,13].

The ddY mice are known to develop spontaneous IgAN[14]. Since they are not inbred, the incidence of IgANamong these mice is highly variable. Recently, we foundthat ddY mice could be classified into three groups: early-onset, late-onset and quiescent groups [15]. We reportedthat inbreeding of early-onset ddYmice for >20 generationsproduced a stable IgAN-onset ddY mice model [16]. Pre-vious reports including our studies indicated that bone mar-row cells (BMCs) may contribute to the pathogenesis ofIgAN [17–19]. Indeed, we experimentally proved that IgANcan be replicated by BM transplantation (BMT) with theIgAN-onset ddY mice as donors in all healthy recipients[20]. In addition, this model can provide the same stage ofIgAN. The present study aimed to assess the influence oftubular L-FABP on glomerular damage without influenceof severe tubular damage, this is important for examiningthe pathophysiological process from early phase of IgAN.Recently, Ohashi et al. [13] demonstrated that the inhibitionof tubular oxidative stress in an experimental IgAN modelcould improve glomerular damage, indicating the possibilityof tubuloglomerular cross-talk in IgAN.

In the present study, we induced IgAN in hL-FABP TgandWT recipient mice by BMT from IgAN-onset ddYmiceto investigate the pathophysiological impact of tubularL-FABP on the progression of glomerular damage in IgAN.

Materials and methods

IgAN model

Details of the generation of hL-FABP Tg mice have been reported previ-ously [4]. Tg mice were backcrossed for more than six generations ontoC57BL/6 mice to obtain homozygous mutant mice on an inbred back-ground. Only female mice were used in this study. Our inbred IgAN-onsetddY mice [15] were used as donors for BMT. Eleven- to 14-week-old fe-male Tg mice (Tg; n = 19; body weight, 22.8 ± 0.4 g; mean ± SD) and WTlittermates on a C57BL/6 background (WT; n = 19; body weight, 22.1 ±0.3 g) were used in this study. BMTwas performed in WT and Tg mice asdescribed previously by our group [20]. Both kinds of transplanted mice(WT/ddY and Tg/ddY mice) were sacrificed at 6 (n = 6 in each group)and 12 weeks (n = 10 in each group) after BMT. Normal WT (n = 3) andTg (n = 3) micewere used as controls.EveryexperimentalprotocolwithmicewasapprovedbytheEthicsReviewCommitteeforAnimalExperimentationofJuntendoUniversitySchool ofMedicine.

Serum and urinary analyses

Serum IgA concentrations were measured by Mouse IgA ELISAQuantita-tionKit (BETHYL, Tokyo, Japan). Urinary albumin from samples collectedover 24 h using metabolic cages was measured with an enzyme-linkedimmunosorbent assay kit (Albuwell, Exocell, Philadelphia, PA, USA).Urinary hL-FABP was measured by the sandwich ELISA kit for hL-FABP(CMIC Co., Tokyo) [4–6]. Intra-assay reproducibility was determinedusing the same sample eight times, with a coefficient of variation within10%. This assay system does not detect L-FABP of rodents, particularlythose derived fromWT mice. Urinary hL-FABP was expressed as the ratioof the urinary L-FABP in nanograms to urinary creatinine in micrograms.

Histological analyses

Kidney sections (3-μm thickness) were fixed in 4% paraformaldehyde andstained with periodic acid–Schiff reagent to assess histological changes bylight microscopy. The extent of glomerular damage (glomerular patho-logical score) was evaluated using a previously described semiquantitativescoring system with modification [21]. Briefly, we examined >30glomeruli per animal (n = 5 in each group) and scored each aspect asfollows: (i) matrix expansion: 0, absent; 1, <25% of glomeruli affected;2, 25–50% glomeruli affected; and 3, >50% of glomeruli affected; (ii)the adhesion of the capillary tuft to the Bowman’s capsule: 0, no; and 1,yes; (iii) tuft numbers of mesangial cell (MC) proliferation determinedby MC >3 in one tuft: 0, absent; 1, one tuft; 2, two tufts; and 3, >3 tufts;(iv) capillary collapse: 0, absent; 1, <25% of glomeruli affected; 2, 25–50%glomeruli affected; and 3, >50% of glomeruli affected. The expansion ofmesangial matrix was estimated by the percentage of intra-glomerular areaoccupied by mesangial matrix in 20 glomeruli of each mouse (n = 6 in eachgroup) with Image J (version 1.43q, NIH), and the average was used foranalysis.

Glomerular infiltration of macrophages was evaluated by immunostain-ing in frozen section (3-μm thick) using biotinylated rat anti-mouse CD68monoclonal antibody (Serotec). We examined >30 glomeruli in each ani-mal (n = 3 in each group). The paraffin sections were incubated withmonoclonal mouse anti-hL-FABP antibody (a gift from CMIC Co.) [4]and subsequent HRP-labeled anti-mouse antibody (Dako Cytomation,Carpentaria, CA) before using DAB kit (Dako Cytomation). The frozensections were also used for immunofluorescence analyses with phyco-erythrin (PE)-conjugated goat anti-mouse IgA antibody (BD Biosciences,Pharmingen, San Diego, CA, USA).

Real-time quantitative PCR analysis

Total RNA was isolated by TRIZOL reagent (Life Technologies Inc.,Carlsbad, CA, USA) according to the instructions provided by the manu-facturer. One-microgram aliquots of total RNA were reverse-transcribedusing Random Decamers (Ambion, Austin, TX, USA) and reverse tran-scriptase, M-MLV (Invitrogen Life Technologies, Carlsbad, CA). The pro-ducts were subjected to real-time PCR using Applied Biosystems 7500Real Time PCR System with SYBR Green Master Mix (Applied Biosys-tems, Tokyo, Japan) and specific primers (Table 1). The gene expressionof the target sequence was normalized to that of GAPDH.

Preparation of isolated glomerulus

The glomeruli were isolated by standard mechanical sieving technique, aspreviously described [22]. Briefly, kidneys were decapsulated, and sinusfat was excised. Isolated cortices were minced to a paste-like consistencywith a surgical blade. The tissue paste was then forced through a series ofmesh filters with 200-, 100- and 70-μM openings. Retained material onthe last sieve was rinsed off and transferred to PBS at 4°C. Glomerularfractions were collected by centrifugation (1500 rpm for 5 min) andwashed with PBS containing protease inhibitors.

Western blotting

Protein extracts were prepared from kidney tissues, isolated glomeruli (fortype IV collagen and fibronectin) or cells with lysis buffer–PBS containing1% Nonidet P40 plus Complete Mini 1 tablet/100 mL (Roche, Mannheim,Germany). The protein samples (20 μg) were separated on 7.5–12.5%sodium dodecyl sulfate–polyacrylamide gels. Monoclonal mouse anti-hL-FABP (a gift from CMIC Co.), [4] polyclonal rabbit anti-HO-1

Table 1. Primer sequences for quantitative real-time PCR assay

Gene name Primers

hL-FABP Forward: 5’ AAA TCG TGC AGA ATG GGA AG 3’Reverse: 5’ TCT CCC CTG TCA TTG TCT CC 3’

MCP-1 Forward: 5’ TCC CAA TGA GTA GGC TGG AG 3’Reverse: 5’ CCT CTC TCT TGA GCT TGG TGA 3’

GAPDH Forward: 5’CAA GGA CTG CGA GAG AAG GT 3’Reverse: 5’ CCT GGT GTG GGT CTT CAG AT 3’

2128 N. Zuo et al.

Dow



(StressGen Biotechnology, Victoria, BC, Canada), monoclonal mouseanti-4-HNE (Japan Institute for the Control of Aging, Shizuoka, Japan),polyclonal goat anti-fibronectin (Santa Cruz Biotechnology), polyclonalrabbit anti-collagen IV (Abcam Inc., Cambridge, MA, USA) and mouseanti-GAPDH (Abcam Inc., Cambridge, MA, USA) antibodies were usedas primary antibodies. Blots were visualized by enhanced chemilumines-cence (ECL system; Amersham Int., Buckinghamshire, UK).

Cell culture

Murine MCs were obtained from primary cell culture of C57BL/6 mice,as described previously [23]. The cells were cultured with RPMI 1640(Sigma, St. Louis, MO, USA) supplemented with 20% FCS (Life Tech-nologies), 100 U/mL penicillin G, 100 μg/mL streptomycin and 5 μg/mLinsulin–transferrin–sodium selenite media supplement (Sigma) in a 5%CO2 environment at 37°C. MCs were identified by immunostaining withdesmin, negative staining for von Willebrand factor and cytokeratin, andmorphological observation. The cells survived after incubation in amedium in which L-valine was replaced with D-valine, excluding contam-ination with fibroblasts. MCs at passage 3 were used in the experiments.

mProx, a stable C57BL/6 mouse proximal tubular cell line, was ob-tained as described previously [24]. Because mProx does not expressL-FABP, hL-FABP including the promoter region was transfected intomProx cells by FuGene 6 (Roche), and the ‘resultant’ mProx stably ex-pressing hL-FABP was named mProx-L [25]. Both mProx and mProx-Lwere maintained in Dulbecco’s modified Eagle’s medium (DMEM)(Sigma) supplemented with 10% FCS under 5% CO2 environment at37°C. In all experiments, the cultured cells grown to confluence weregrowth-arrested with a DMEM medium containing 0.5% FCS for 16 hbefore the commencement of experiments.

Preparation of spent (condition) medium

Mouse IgA (Cappel, MP Biomedicals, Solon, OH, USA) were heat-aggregated by incubation at 63°C for 150 min to obtain aggregated IgA(AIgA), as described previously [26]. Growth-arrested MCs were culturedin 60-mm tissue culture dishes (IWAKI) (1 × 106 cells/dish) with RPMI1640 medium containing 0.5% FCS and different concentrations of AIgA(final concentrationwas 10, 50 and 250 μg/mL) for 48 h. The spentmedium(AIgA-MC medium) after culture was collected and kept frozen at −80°Cuntil analyses. Spent medium from MCs cultured without the addition ofAIgA preparation was used as control medium. AIgA-MC medium wasdiluted by 10-fold with DMEM containing 0.5% FCS to stimulate mProxand mProx-L.

Based on preliminary data (data not shown), growth-arrested mProxand mProx-L were cultured in 60-mm collagen IV-coated tissue culturedishes (IWAKI) (1 × 106 cells per well) and stimulated with recombinantTNF-α (R&D Systems, Minneapolis, MN, USA) (final concentration: 2–250 ng/L) for 24 h or 10-fold diluted AIgA-MC medium for either 12 (forhL-FABP mRNA expression) or 24 h (for MCP-1 mRNA expression orhL-FABP and 4-HNE protein expression) with or without pre-treatmentby neutralizing anti-TNF-α antibody (R&D Systems, Minneapolis, MN,USA) (final concentration: 0.1–5 μg/ml) 1 h before.

Statistical analyses

All data were expressed as mean ± SD. Differences between groups wereexamined for statistical significance using the one-way analysis of vari-ance (ANOVA). A P-value <0.05 was considered statistically significant.

Results

Glomerular damage in Tg mice with replicated IgAN wassignificantly milder than that in WT mice

Although both mesangial IgA deposition (Figure 1A) andserum elevation of IgA levels (Figure 1B) were similar be-tween WT/ddYand Tg/ddY mice after BMT, urinary albu-min level in Tg/ddY mice even at 6 weeks was significantly

lower than that in WT/ddY mice (P < 0.05). Thereafter,urinary albumin level in WT/ddY mice sharply rose to82.8 ± 65.6 mg/dL at 12 weeks, but that of Tg/ddY micewas still significantly suppressed (8.2 ± 2.2 mg/dL, P <0.005) (Figure 2A).

Next, we evaluated glomerular histology (Figure 2B).Glomerular pathological score regarding the extracellularmatrix, adhesion, mesangial cell proliferation and capillarycollapsing showed significant attenuation of glomerulardamage in Tg mice (P < 0.0005) (Figure 2C). Since the ex-tent of the mesangial matrix expansion was strikingly dif-ferent between both groups, we examined the expansion ofmesangial matrix and found that WT/ddY mice showedsignificant mesangial matrix expansion at 12 weeks (P <0.005), which is significantly suppressed in Tg/ddY mice(P < 0.01) (Figure 2D). Furthermore, glomerular protein ex-pression of type IV collagen (P < 0.0005) and FN (P < 0.05)in Tg/ddY mice was also significantly suppressed comparedwith that of WT/ddY mice at 12 weeks (Figure 2E). More-over, glomerular infiltration of CD68+ macrophages in Tg/ddYmicewas significantly milder than that inWT/ddYmiceat 12 weeks (P < 0.05) (Figure 2F).

Tubular hL-FABP was up-regulated in Tg mice with IgAN

Immunohistochemical analysis showed that hL-FABP ex-pression in control Tg mice was spread profusely throughthe cytoplasm of the proximal tubules. In Tg/ddY mice,

WT/ddY

Tg/ddYWT/ddY

Tg/ddY

WT/ddY Tg/ddY

0

100

200

300

400

500

Seru

m I

gA(m

g/dL

)

Time after BMT

0w 6w 12w

***

**

****

6w

12w

A

B

Fig. 1. Mesangial IgA deposition and serum level of IgA after bonemarrow transplantation (BMT). Similar reconstitution of glomerularIgA deposition (original magnification ×400) (A) and serum level ofIgA (B) were observed. Data are mean ± SD (*P < 0.05, **P < 0.01,***P < 0.005).

Renoprotective effects of L-FABP in murine IgAN 2129

Dow



immunostaining for hL-FABP showed enhanced nuclearlocalization of the proximal tubules (Figure 3A) and waslocalized more in nuclei of tubular cells at 12 weeks thanthat at 6 weeks. In addition, real-time PCR and westernblotting analyses revealed that gene (P < 0.001) and pro-tein (P < 0.01) expression of hL-FABP were significantlyincreased in the Tg/ddY kidney at 6 weeks after BMT(Figure 3B and C). Consistently, urinary hL-FABP levelwas also increased significantly at 6 weeks (P < 0.05)(Figure 3D).

Renal oxidative stress and inflammation were attenuatedin Tg/ddY mice

HO-1 is considered to be one of the most sensitive indica-tors of cellular oxidative stress [27], and 4-HNE is a majoraldehydic product of lipid peroxidation [28]. Both HO-1(Figure 4A) and 4-HNE-modified protein (Figure 4B)were not detected in the control WT and Tg kidneys. Al-though HO-1 overexpression and accumulation of 4-HNEcan be observed at 6 and at 12 weeks, respectively, theirrenal expression levels in Tg/ddY were significantly lower

than those in WT/ddY (P < 0.05). Renal MCP-1 mRNAexpression was up-regulated at 12 weeks in WT/ddY mice(P < 0.005), which was significantly higher than that inTg/ddY mice (P < 0.01) (Figure 4C).

AIgA-MC medium enhanced the expression of hL-FABP incultured mProx-L, partly via TNF-α

Next, we approached the underlying mechanism of reno-protection in Tg/ddY mice. We firstly analyzed the de-tailed mechanism of tubular up-regulation of hL-FABPin this disease. We incubated mProx-L with dilutedAIgA-MC media or AIgA alone with corresponding con-centration (final concentration of AIgA was 1–25 μg/mL).AIgA-MC media induced a significant up-regulation ofhL-FABP mRNA and protein expression compared withcontrol medium and was significantly more effective thanAIgA itself (Figure 5A and B). The pre-incubation of neu-tralizing anti-TNF-α antibody with mProx-L in AIgA-MCmedium can significantly suppress the up-regulation ofhL-FABP at 1 μg/mL (P < 0.05) and 5 μg/mL (P < 0.01)(Figure 5C). In contrast, recombinant TNF-α itself in-

WT/ddY Tg/ddY

0

30

60

90

120

150

180

A

****

Time after BMT

*

0w 6w 12w

*

Uri

nary

alb

umin

(m

g/dL

) WT/ddY

Tg/ddY

0w 6w 12w

B

0

1

2

3

4

5

6

Glo

mer

ular

pat

holo

gica

l sc

ore

C****

30

35

40

45

50

55

60

0w 6w 12w

*****

Mes

angi

al m

atri

x fr

acti

on(%

)

D

WT/ddY 12w Tg/ddY 12w

Fig. 2. Evaluation of glomerular injury in WT and Tg mice with IgAN. (A). Urinary albumin levels were attenuated significantly in Tg/ddY. (B).Representative photomicrographs of periodic acid–Schiff (PAS) staining (original magnification ×400). (C). The glomerular pathological score wassignificantly lower in Tg/ddY mice at 12 weeks (n = 5). (D). Tg/ddY mice significantly improved expansion of mesangial matrix at 12 weeks (n = 6).(E). Western blotting with protein extracts from isolated glomeruli. Glomerular deposition of type IV collagen and FN in Tg/ddY mice was attenuated at12 weeks (n = 3). (F). The number of CD68+ macrophages in glomeruli was also lower in Tg/ddY mice (n = 3) (original magnification ×400). Data aremean ± SD (*P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.0005).

2130 N. Zuo et al.

Dow



duced a significant up-regulation of hL-FABP protein ex-pression at 50 and 250 ng/L (P < 0.005) (Figure 5D),suggesting a role of MC-origin TNF-α in tubular L-FABP expression.

AIgA-MC medium-induced expression levels of 4-HNEand MCP-1 in mProx-L were significantly lower thanthose of mProx

To further examine the renoprotective mechanism ofL-FABP, we simultaneously stimulated mProx and mProx-Lwith AIgA-MC medium (final concentration of AIgAwas 25 μg/mL). The stimulation induced a significantup-regulation of 4-HNE-modified protein in mProx (P <0.05), which was signif icantly higher than that inmProx-L (P < 0.05) (Figure 5E), and also significantly in-

creased MCP-1 mRNA expression in both mProx (P <0.001) and mProx-L (P < 0.001). However, the expressionlevel of MCP-1 in mProx was ~4.7-fold higher (P < 0.001)than that of mProx-L (Figure 5F).

Discussion

In this study, we induced IgAN in hL-FABP Tg and WTrecipients by BMT from IgAN-onset ddY donors. Althoughmesangial IgA deposition and serum IgA levels were simi-larly reconstituted, glomerular damage in Tg/ddY micewas strongly protected. Moreover, renal oxidative stressand inflammation were significantly suppressed in theTg/ddY kidneys with up-regulation of renal hL-FABP, sug-gesting that renal hL-FABP may reduce oxidative stressand inflammation through inhibition of the accumulation

WT/ddY Tg/ddY WT/ddY Tg/ddY

FN

GAPDH

6w 12w

WT/ddY Tg/ddY

6w

WT/ddY Tg/ddY

12w

Col E

F

IV

GAPDH

0

0.2

0.4

0.6

0.8

1

1.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Rel

ativ

e pr

otei

n ex

pres

sion

Col

IV

/GA

PD

HR

elat

ive

prot

ein

expr

essi

onF

N/G

AP

DH

****

*

0

1

2

3

4

5

6

num

bers

of

CD

68+

mac

roph

ages

in g

lom

erul

i

*

WT/ddT 12w Tg/ddY 12w

Fig. 2. Continued.


Dow



of lipid peroxidation products in the present IgAN model.We also provided experimental evidence that up-regulationof hL-FABP in the proximal tubules can be induced byhumoral factors released from the glomerular mesangialcells after IgA deposition, which may be partially due toTNF-α. Overexpression of hL-FABP can suppress tubularexpression of some major mediators of oxidative stress andinflammation by such humoral factors and thus may con-tribute to the renoprotection in Tg/ddY mice.

The pathophysiological roles of renal L-FABP in kidneydiseases have not yet been fully clarified. Recent animalexperiments have confirmed the renoprotective role ofL-FABP in tubulointerstitial disease models [4–6], in whichoverexpression of tubular L-FABP attenuated the tubuloin-terstitial damage. Although renal damage in the presentmodel was mainly located in the glomeruli, tubular expres-sion of hL-FABP was also significantly up-regulated in Tg/ddY mice, together with a rise of urinary hL-FABP level,suggesting that certain stimuli may act on proximal tubularcells in this glomerular disease and induce up-regulation ofhL-FABP expression and subsequent renoprotection.

In IgAN, mesangial IgA deposition is one of the initialand critical events that subsequently lead to glomerulardamage with a progressive clinical course. It had been de-monstrated that proximal tubular epithelial cells (PTEC) donot express known specific IgA receptors [29], and IgA israrely deposited in tubulointerstitium in IgAN [30]. IgA

binding activates mesangial cells to produce inflammatorycytokines and chemokines, including TNF-α [31,32]. Nor-mally, the glomerular barrier is impermeable to proteins,but an increase in the glomerular barrier pore size allowingthe leakage of proteins into the tubular lumen has been ob-served in various glomerular diseases [33]. Recently, Chanet al. established a novel in vitro model to provide experi-mental evidence of glomerulotubular communication inIgAN: ‘spent medium’ from human mesangial cells incu-bated with IgA from IgAN patients induced proliferationand activation of PTEC [29]. Their finding suggests thathumoral factors released from the glomerular mesangialcells maintained a glomerulotubular cross-talk in the devel-opment of IgAN. Based on the aforementioned findings, wehypothesized that up-regulation of tubular L-FABP in ourIgAN model may also be due to this kind of glomerulotub-ular cross-talk. To confirm our hypothesis, we followed theaforementioned experimental model to stimulate mProx-Lby AIgA-MC media or directly by AIgA, and found thatAIgA-MC media can induce a significant up-regulation ofhL-FABP gene expression and protein synthesis comparedwith AIgA alone. This finding suggests that certain humoralmediators released fromMC after mesangial IgA depositioninduced the up-regulation of L-FABP in proximal tubularcells. To confirm whether TNF-α released from mesangialcells after IgA deposition may activate renal tubular cells[29], we incubated mProx-L with neutralizing anti-TNF-α

A control Tg

Tg/ddY 12w WT/ddY 12w

Tg/ddY 6w

0

0.4

0.8

1.2

1.6

2

2.4

control Tg

6w 12w

Tg/ddY

****

0

0.5

1

1.5

2

6w 12 w

Tg/ddY

hL-FABP

GAPDH

0

20

40

60

80

100

0w 6w 12w

***

Time after BMTcontrol

Tg

Rel

ativ

e ge

ne e

xpre

ssio

nhL

-FA

BP

/GA

PD

H

Rel

ativ

e pr

otei

n ex

pres

sion

hL-F

AB

P/G

AP

DH

Uri

nary

hL

-FA

BP

(ng/

mg

crea

tini

ne)

B

C D

Fig. 3. Expression of hL-FABP in Tg/ddY kidney. (A). Representative immunostaining of hL-FABP in the kidneys (original magnification ×200).Renal hL-FABP mRNA (B) and protein (C) expression. (D). Urinary hL-FABP level expressed as a ratio to urinary creatinine (μg/g creatinine). Data aremean ± SD (*P < 0.05, **P < 0.01, ***P < 0.001).

2132 N. Zuo et al.

Dow



antibody in AIgA-MCmedium or recombinant TNF-α, andfound that TNF-α may be a potent mediator of hL-FABPsynthesis. These findings are in conflict with the results ofa previous study by Memon et al. [34]. They reported thatrecombinant TNF-α decreased L-FABPmRNA levels in theliver of Syrian hamsters. Dubé et al. [35] also reported thatTNF-α did not modify the cellular levels of L-FABP inCaco-2 cells. However, these two studies investigated dif-ferent organs and cell lines from those used in our study,indicating that the expression of L-FABP may be regulatedin an organ-specific manner. Further studies are needed toclarify this point.However, it is possible that IgAmay directlyactivate tubular L-FABP in vivo. Since the nephritogenic IgAmay be in polymeric or immune complex (IC), leaked poly-meric or IC IgA in urine may be absorbed by specific or non-

specific receptors such as megalin and thus activates tubularcells. Although present experiments with AIgA alone did notenhance L-FABP in vitro, we have to think carefully aboutthis possibility in IgAN.

L-FABP can prevent the peroxidation of intracellularfatty acids by promoting fatty acid metabolism and byregulating gene expression involved in lipid metabolism[1,2]. Wang et al. [36] demonstrated in vitro that L-FABPmay reduce oxidative stress in hypoxia–reoxygenation.Recent renal disease models of hL-FABP Tg mice alsosuggested that acting as an antioxidant, endogenous renalL-FABP can effectively improve the course of tubulointer-stitial injury [4–6]. In our model, renal HO-1, 4-HNE andMCP-1 mRNA expression were up-regulated at 6 and12 weeks after BMT in WT/ddY mice, respectively. All ofthem were significantly suppressed in Tg/ddY mice. Theseresults indicate that oxidative stress and inflammation inmurine IgAN may be attenuated by tubular L-FABP. InTg/ddY mice, glomerular damage evaluated by mesangialmatrix expansion, MC proliferation, adhesion and capillarycollapsing were significantly attenuated compared withthat of WT/ddY mice. Comparably, Tg/ddY mice main-tained the baseline value of urinary albumin level duringthe experimental period, while WT/ddY mice showed sig-nificant albuminuria until 12 weeks. Moreover, glomerularinfiltration of CD68+ macrophages was also significantlysuppressed in Tg/ddY mice. These findings point to allevi-ation of glomerular damage in Tg/ddY mice. A series ofprevious studies stressed the pathogenic role of oxidativestress in the development of IgAN and showed that treat-ment with antioxidants could improve the development ofIgAN [9,11,37–41]. For example, Chen et al. [37] reportedthat IgA-containing IC simulated the production of oxygen-free radicals byMC in situ. Kashem et al. [38] demonstratedhigh superoxide production by polymorphonuclear leuco-cytes isolated from patients with IgAN, and that suchproduction correlated with the level of proteinuria. Des-camps-Latscha et al. [39] found that high plasma levels ofadvanced oxidation protein products (AOPP) closely re-flected a progressive form of IgAN. Recently, Kobori andco-workers [11] also found augmentation of intra-renal oxi-dative stress at early stages of IgAN, and that the activatedintra-renal ROS played some roles in the development ofIgAN by interacting with activated RAS. Trachtman et al.[40] showed that dietary treatment with antioxidantagent–vitamin E attenuated renal functional and structuralchanges in the experimental IgAN model. Kuemmerle et al.[41] also reported that administration of alpha tocopherolameliorated renal injury in a rat experimental model ofIgAN. Treatment with antioxidants such as fish-oil extractsin human IgAN was also reported to produce encouragingresults [9]. Up-regulation of tubular L-FABP and suppres-sion of oxidative stress of kidney were also observed inour IgAN model with subsequent amelioration of glomeru-lar injury. Taking the aforementioned data into conside-ration, the present study suggests that, as an endogenousantioxidant, tubular L-FABP can improve the progressionof glomerular damage in IgAN.

Ohashi et al. [13] demonstrated the association betweenrenal damage and changes in ROS and RAS in an experi-mental IgAN model. Both ROS and RAS were activated

0

1

2

3

4

5

B

WT/ddY6w 12w

Tg/ddY6w 12w

4-HNE

GAPDH

*

WT/ddY 12w Tg/ddY 12w

WT/ddY Tg/ddY

controlWT

controlTg

Pro

duct

ion

of 4

-HN

E4-

HN

E/G

AP

DH

AHO-1

GAPDHcontrol

WT WT/ddYcontrol

Tg6w 12w

Tg/ddY6w 12w

0

4

8

12

16

20 *

WT/ddY 6w Tg/ddY 6wRel

ativ

e pr

otei

n ex

pres

sion

HO

-1/G

AP

DH

0

1

2

3

4

5

6

0w 6w 12w

**

Rel

ativ

e ge

ne e

xpre

ssio

n M

CP

-1/G

AP

DH

C ***

Fig. 4. Intra-renal oxidative stress and inflammation in WT/ddY and Tg/ddY mice. Renal HO-1 protein expression at 6 weeks (A), production of4-HNE-modified protein at 12 weeks (B) and MCP-1 gene expression at12 weeks (C) were significantly suppressed in Tg/ddY mice. Data aremean ± SD (*P < 0.05, **P < 0.01, ***P < 0.005).


Dow



AIgA-MC mediumA B

C D

E

F

AIgA

0

0.5

1

1.5***

********

Rel

ativ

e ge

ne e

xpre

ssio

n

hL-F

AB

P/G

AP

DH

0 0 1 1 5 5 25 25AIgA (μg/ml)

0

1

2

3

4

hL-FABP

GAPDH

*******

****

******

****

Rel

ativ

e pr

otei

n ex

pres

sion

hL-F

AB

P/G

AP

DH

0 0 1 1 5 5 25 25AIgA (μg/ml)

0

0.5

1

1.5

2

2.5

control 2ng/L 10ng/L 50ng/L 250ng/L

hL-FABP

GAPDH

Rel

ativ

e pr

otei

n ex

pres

sion

hL-F

AB

P/G

AP

DH

******

Recombinant TNF-α

0

2

4

6

8

10

12

Rel

ativ

e ge

ne e

xpre

ssio

n

MC

P-1

/GA

PD

H

****

0

0.2

0.4

0.6

0.8

1

Pro

duct

ion

of 4

-HN

E m

odif

ied

prot

ein,

4-H

NE

/GA

PD

H

AIgA-MC medium (-) (+) (-) (+)

mProx mProx-L

4-HNE

GAPDH

* *

********

AIgA-MC medium (-) (+) (-) (+)

hL-FABP

GAPDH

0

1

2

3

4

5

****

**

Rel

ativ

e pr

otei

n ex

pres

sion

hL-F

AB

P/G

AP

DH

control medium

AIgA-MC medium

TNFa Ab (μg/ml)

(-)(-)

0

(+)(-)

0

(-)(+)

0

(-)(+)

0.1

(-)(+)

1

(-)(+)

5

Fig. 5. In vitro study. mRNA (A) and protein (B) expression of hL-FABP in mProx-L were up-regulated significantly by AIgA-MC medium (n = 4).(C). Up-regulation of hL-FABP in mProx-L induced by IgA-MC medium was partially blocked by neutralizing anti-TNF-α antibody (n = 4). (D).Recombinant TNF-α induced up-regulation of L-FABP (n = 3). Production of 4-HNE-modified protein (n = 4) (E) and MCP-1 mRNA expression level(n = 4) (F) in mProx-L were significantly lower than those in mProx after stimulation of AIgA-MC medium (final concentration of AIgA is 25 μg/mL).Data are mean ± SD (*P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001).

2134 N. Zuo et al.

Dow



in mice with IgAN, and administration of tempol (oxygenradical scavenger) significantly suppressed the immunoac-tivity of HO-1 and Ang II in the proximal and distal tubulesand finally resulted in amelioration of renal damage, indi-cating that inhibition of oxidative stress in tubules couldimprove glomerular damage. It also implied the possibilityof tubuloglomerular cross-talk in IgAN. Tang et al. [42] alsoprovided compelling experimental evidence in support ofthe existence of a tubuloglomerular ‘cross-talk’mechanismin the proteinuric state. After exposure to transferrin, PTECcontinued to secrete PDGF, which contributed at least inpart to the proliferation of MC through autocrine and/orparacrine mechanisms. Recently, we found that tubularenhancement of anti-oxidative action by L-FABP alsoattenuated the acute glomerular damage in acute glome-rulonephritis models (anti-GBM antibody-induced glome-rulonephritis) (our unpublished data). This protective

effect was linked to attenuation of polymorphonuclearcell (PMN) flux in acute phase which may be due tohemodynamic change in glomeruli by tubular activationof L-FABP. The present experiment also showed signifi-cant attenuation of glomerular CD68+ macrophage infil-tration in Tg mice with decrease of MCP-1 expressionand oxidative stress. In addition, MCP-1 production byAIgA-MC medium in mProx-L was down-regulated sig-nificantly less than that in mProx in association withdown-regulation of 4-HNE. Oxidative stress is known tomodulate MCP-1 gene expression, which is, at least inpart, transcriptionally regulated via activation of nuclearfactor-kappaB (NF-κB) [43,44]. Although it is possiblethat a similar mechanism by hemodynamic alterationmay exist, the present data suggests that anti-oxidative ac-tion by tubular L-FABP may reduce glomerular macro-phage infiltration, at least in part, via anti-inflammatoryaction on MCP-1, providing protection against glomerulardamage. Further studies are imperative to clarify more de-tailed mechanisms. Based on the results of the presentstudy, we propose a hypothetical schema that definesthe mechanisms discussed in the present study (Figure 6).

Although L-FABP is known to play a protective role intubulointerstitial injury [4–6], we could not observe obvioustubulointerstitial damage in our IgAN model during therelatively short experimental process, which was still atthe early stage up to 12weeks after BMTandmainly showedglomerular damage. Therefore, the present model can avoidindirect influence of severe tubulointerstitial fibrosis onglomerular damage and clearly demonstrate the protectiveaction of tubular L-FABP against glomerular damage.

In the present model, severe glomerular injury was mostprominent at 12 weeks, while mRNA expression of hL-FABP and urinary level of hL-FABP at 6 weeks were high-er than those at 12 weeks. This discrepancy may relate tothe difference of L-FABP localization at 6 and 12 weeks,as demonstrated in Figure 3A that hL-FABP was localizedmore in nuclei of tubular cells of Tg mice at 12 weeks.L-FABP shuttles FFA to the nucleus, where L-FABP directlyinteracts with PPARα known as an anti-inflammatorymediator [45], suggesting that this shift of localization maychange the efficacy of L-FABP anti-inflammatory actionvia PPARα transcriptional activity [46].In conclusion, overexpression of hL-FABP in proximal

tubules protected against the progression of glomerulardamage of murine IgAN in association with attenuationof oxidative stress. These results suggest that suppressionof oxidative stress in the proximal tubules was crucial inimproving the progression of IgAN, particularly at earlystages. Because the genomic DNA of hL-FABP, includingits promoter region, was microinjected into fertilized eggsto generate the Tg mice, transcription of the hL-FABP genein the Tg mice may be regulated in a manner similar to thatin humans. It is possible that L-FABP in human IgAN hassimilar dynamics to that in our experimental IgAN model,as a self-protective system. In fact, Nakamura et al. [47]reported that urinary L-FABP was significantly higher inIgAN patients compared to both normal subjects and thosediagnosed with thin basement membrane nephropathy.Thus, this study seems to provide some promising data onthe improvement of progression of IgAN by agents that can

A

B

Oxidative

stress

Humoral

factors

L-FABP

TNF- or ?

Glomerular

damageMesangial cell

Inflammatory

mediators

(MCP-1 or ?)

?

Glomerulus

PTC

Oxidative

stress

Humoral

factors

Glomerular

damageMesangial cell

Inflammatory

mediators

(MCP-1 or ?)

?

Glomerulus

PTC

IgA

IgA

Fig. 6. Schematic diagram of the renoprotective effect of L-FABP inmurine IgAN. (A). Mechanism of progression of IgAN in WT/ddYmice: certain humoral factors from glomeruli after mesangial depositionof IgA induce oxidative stress in the proximal tubular cells (PTC), whichsubsequently up-regulate the gene expression of certain inflammatorymediators such as MCP-1. Oxidative stress and/or up-regulation ofinflammatory mediators are transmitted to the glomerular compartmentvia an as yet undefined mechanism that results in aggravation ofglomerular damage. (B). The renoprotective actions of L-FABP in Tg/ddY mice: TNF-α and/or other mediators released from glomeruli aftermesangial IgA deposition induce up-regulation of L-FABP in PTC. Thelatter suppresses oxidative stress and subsequently down-regulates theexpression of inflammatory mediators, which alleviates glomerulardamage.


Dow



up-regulate tubular expression of L-FABP such as PPARαagonists [48] and help design of new therapies for IgAN.

Acknowledgements. We thank Ms Shibata for her excellent technicalsupport and Dr Tadahiro Kajiyama for management of early-onset ddYmice.

Conflict of interest statement. None declared.

References

1. Zimmerman AW, Veerkamp JH. New insights into the structure andfunction of fatty acid-binding proteins. Cell Mol Life Sci 2002; 59:1096–1116

2. Antonenkov VD, Sormunen RT, Ohlmeier S et al. Localization of aportion of the liver isoform of fatty-acid-binding protein (L-FABP) toperoxisomes. Biochem J 2006; 394: 475–484

3. Ek-Von Mentzer BA, Zhang F, Hamilton JA. Binding of 13-HODEand 15-HETE to phospholipid bilayers, albumin, and intracellularfatty acid binding proteins. Implications for transmembrane and intra-cellular transport and for protection from lipid peroxidation. J BiolChem 2001; 276: 15575–15580

4. Kamijo A, Sugaya T, Hikawa A et al. Urinary excretion of fatty acidbinding protein reflects stress overload on the proximal tubules. Am JPathol 2004; 165: 1243–1255

5. Kamijo-Ikemori A, Sugaya T, Obama A et al. Liver type fatty acidbinding protein attenuates renal injury induced by unilateral ureteralobstruction. Am J Pathol 2006; 169: 1107–1117

6. Kamijo-Ikemori A, Sugaya T, Sekizuka A et al. Amelioration of dia-betic tubulointerstitial damage in liver-type fatty acid binding proteintransgenic mice. Nephrol Dial Transplant 2009; 24: 788–800

7. Barratt J, Feehally J. IgA nephropathy. J Am Soc Nephrol 2005; 16:2088–2097

8. Floege J, Feehaly J. IgA nephropathy: recent developments. J Am SocNephrol 2000; 11: 2395–2403

9. Donadio JV, Grande JP. IgA nephropathy. N Engl J Med 2002; 347:738–748

10. Grande JP, Walker HJ, Holub BJ et al. Suppressive effects of fish oilon mesangial cell proliferation in vitro and in vivo. Kidney Int 2000;57: 1027–1040

11. Kobori H, Katsurada A, Ozawa Y et al. Enhanced intrarenal oxidativestress and angiotensinogen in IgA nephropathy patients. BiochemBiophys Res Commun 2007; 358: 156–163

12. Ohashi N, Katsurada A, Miyata K et al. Activation of reactive oxygenspecies and the rennin-angiotensin system in IgA nephropathy modelmice. Clin Exp Pharmacol Physiol 2009; 36: 509–515

13. Ohashi N, Katsurada A, Miyata K et al. Role of activated intrarenalreactive oxygen species and renin-angiotensin system in IgA nephro-pathy model mice. Clin Exp Pharmacol Physiol 2009; 36: 750–755

14. Imai H, Nakamoto Y, Asakura K et al. Spontaneous glomerular IgAdeposition in ddY mice: an animal model of IgA nephritis. Kidney Int1985; 27: 756–761

15. Suzuki H, Suzuki Y, Yamanaka T et al. Genome-wide scan in a novelIgA nephropathy model Identifies a susceptibility locus on murinechromosome 10, in a region syntenic to human IGAN1 on chromo-some 6q22-23. J Am Soc Nephrol 2005; 16: 1289–1299

16. Suzuki H, Suzuki Y, Narita I et al. Toll-like receptor 9 affects severityof IgA nephropathy. J Am Soc Nephrol 2008; 19: 2384–2395

17. Barratt J, Feehally J, Smith AC. Pathogenesis of IgA nephropathy.Semin Nephrol 2004; 24: 197–217

18. van den Wall Bake AW, Daha MR, Evers-Schouten J et al. SerumIgA and the production of IgA by peripheral blood and bone marrowlymphocytes in patients with primary IgA nephropathy: evidence forthe bone marrow as the source of mesangial IgA. Am J Kidney Dis1988; 12: 410–414

19. Harper SJ, AllenAC, Pringle JH et al. Increased dimeric IgAproducingB cells in the bone marrow in IgA nephropathy determined by in situhybridisation for J chain mRNA. J Clin Pathol 1996; 49: 38–42

20. Suzuki H, Suzuki Y, Aizawa M et al. Th1 polarization in murine IgAnephropathy directed by bone marrow-derived cells. Kidney Int 2007;72: 319–327

21. Parlakpinar H, Acet A, Gul M et al. Protective effects of melatoninon renal failure in pinealectomized rats. Int J Urol 2007; 14: 743–748

22. Fong JS, Drummond KN. Method for preparation of glomeruli formetabolic studies. J Lab Clin Med 1968; 71: 1034–1039

23. Hisada Y, Sakurai H, Sugaya T. Cell to cell interaction betweenmesangial cells and macrophages induces the expression of monocytechemoattractant protein-1 through nuclear factor-κB activation.Biochem Biophys Res Commun 2000; 269: 309–316

24. Okada H, Kikuta T, Inoue T et al. Dexamethasone induces connectivetissue growth factor expression in renal tubular epithelial cells in amouse strain-specific manner. Am J Pathol 2006; 168: 737–747

25. Yamamoto T, Noiri E, Ono Y et al. Renal L-type fatty acid–bind-ing protein in acute ischemic injury. J Am Soc Nephrol 2007; 18:2894–2902

26. Gómez-Guerrero C, López-Armada MJ, González E et al. SolubleIgA and IgG aggregates are catabolized by cultured rat mesangialcells and induce production of TNF-alpha and IL-6, and proliferation.J Immunol 1994; 153: 5247–5255

27. Applegate LA, Luscher P, Tyrrell RM. Induction of heme oxygenase:a general response to oxidant stress in cultured mammalian cells.Cancer Res 1991; 51: 974–978

28. Ji C, Kozak KR, Marnett LJ. IkappaB kinase, a molecular targetfor inhibition by hydroxy-2-nonenal. J Biol Chem 2001; 276:18223–18228

29. Chan LY, Leung JC, Tsang AW et al. Activation of tubular epithelialcells by mesangial-derived TNF-a: glomerulotubular communicationin IgA nephropathy. Kidney Int 2005; 67: 602–612

30. Frascà GM, Vangelista A, Biagini G et al. Immunological tubulo-interstitial deposits in IgA nephropathy. Kidney Int 1982; 22: 184–191

31. van den Dobbelsteen ME, van der Woude FJ, Schroeijers WE et al.Binding of dimeric and polymeric IgA to rat renal mesangial cellsenhances the release of interleukin 6. Kidney Int 1994; 46: 512–519

32. Mené P, Simonson MS, Dunn MJ. Physiology of the mesangial cell.Physiol Rev 1989; 69: 1347–1424

33. D’Amico G, Bazzi C. Pathophysiology of proteinuria. Kidney Int2003; 63: 809–825

34. Memon RA, Bass NM, Moser AH et al. Down-regulation of liver andheart specific fatty acid binding proteins by endotoxin and cytokinesin vivo. Biochim Biophys Acta 1999; 1440: 118–126

35. Dubé N, Delvin E, Yotov W et al. Modulation of intestinal and liverfatty acid-binding proteins in Caco-2 cells by lipids, hormones andcytokines. J Cell Biochem 2001; 81: 613–620

36. Wang G, Gong Y, Anderson J et al. Antioxidative function of L-FABPin L-FABP stably transfected Chang liver cells. Hepatology 2005; 42:871–879

37. Chen A, Chen WP, Sheu LF et al. Pathogenesis of IgA nephropathy:In vitro activation of human mesangial cells by IgA immune complexleads to cytokine secretion. J Pathol 1994; 173: 119–126

38. Kashem A, Endoh M, Nomoto Y et al. Fcα R expression on poly-morphonuclear leukocyte and superoxide generation in IgA nephro-pathy. Kidney Int 1994; 45: 868–875

39. Descamps-Latscha B, Witko-Sarsat V, Nguyen-Khoa T et al. Earlyprediction of IgA nephropathy progression: proteinuria and AOPPare strong prognostic markers. Kidney Int 2004; 66: 1606–1612

40. Trachtman H, Chan JC, Chan W et al. Vitamin E ameliorates renalinjury in an experimental model of immunoglobulin A nephropathy.Pediatr Res 1996; 40: 620–626

41. Kuemmerle NB, Krieg RJ Jr, Chan W et al. Influence of alpha toco-pherol over the time course of experimental IgA nephropathy. PediatrNephrol 1999; 13: 108–112

42. Tang S, Leung JC, Tsang AW et al. Transferrin up-regulates chemo-kine synthesis by human proximal tubular epithelial cells: implicationon mechanism of tubuloglomerular communication in glomerulo-pathic proteinuria. Kidney Int 2002; 61: 1655–1665

43. Habibzadegah-Tari P, Byer K, Khan SR. Oxalate induced expressionof monocyte chemoattractant protein-1 (MCP-1) in HK-2 cells in-volves reactive oxygen species. Urol Res 2005; 33: 440–447

2136 N. Zuo et al.

Dow



44. Sung FL, Zhu TY, Au-Yeung KK et al. Enhanced MCP-1 expressionduring ischemia/reperfusion injury is mediated by oxidative stressand NF-kappaB. Kidney Int 2002; 62: 1160–1170

45. Duan SZ, Usher MG, Mortensen RM et al. PPARs: the vasculature,inflammation and hypertension. Curr Opin Nephrol Hypertens 2009;18: 128–133

46. Hostetler HA, Balanarasimha M, Huang H et al. Glucose regulatesfatty acid binding protein interaction with lipids and PPAR{alpha}.J Lipid Res 2010; 51(11): 3103–3116

47. Nakamura T, Sugaya T, Ebihara I et al. Urinary liver-type fatty acid-binding protein: discrimination between IgA nephropathy and thinbasement membrane nephropathy. Am J Nephrol 2005; 25: 447–450

48. Negishi K, Noiri E, Maeda R et al. Renal L-type fatty acid -bindingprotein mediates the bezafibrate reduction of cisplatin-induced acutekidney injury. Kidney Int 2008; 73: 1374–1384

Received for publication: 15.2.10; Accepted in revised form: 18.10.10

Nephrol Dial Transplant (2011) 26: 2137–2144doi: 10.1093/ndt/gfq675Advance Access publication 13 December 2010

Effect of chronic kidney disease on the expression of thiamin and folicacid transporters

Farhan J. Bukhari1, Hamid Moradi2, Pavan Gollapudi2, Hyun Ju Kim2, Nosratola D. Vaziri2,3

and Hamid M. Said1,2,3

1Department of Medical Research, VA Medical Center, 5901 East 7th St., Long Beach, CA 90822, USA, 2Department of Medicineand 3Department of Physiology/Biophysics, University of California School of Medicine, Irvine, CA 92697, USA

Correspondence and offprint requests to: Hamid M. Said; E-mail: [email protected]

AbstractBackground. Chronic kidney disease (CKD) is associatedwith significant cardiovascular, neurological and metaboliccomplications. Thiamin and folate are essential for growth,development and normal cellular function, and their uptakeis mediated by regulated transport systems. While plasmafolate and thiamin levels are generally normal in patientswith CKD, they commonly exhibit features resembling vita-min deficiency states. Earlier studies have documentedimpaired intestinal absorption of several B vitamins inexperimental CKD. In this study, we explored the effectof CKD on expression of folate and thiamin transporters inthe key organs and tissues.Methods. Sprague-Dawley rats were randomized to under-go 5/6 nephrectomy or sham operation and observed for12 weeks. Plasma folate and thiamin concentrations andgene expression of folate (RFC, PCFT) and thiamin trans-porters (THTR-1 and THTR-2) were determined in theliver, brain, heart and intestinal tissues using real-timePCR. Hepatic protein abundance of these transporterswas determined using western blot analysis.Results. Plasma folate and thiamin levels were similar be-tween the CKD and the control groups. However, expres-sions of both folate (RFC and PCFT) and thiamin(THTR-1, THTR-2) transporters were markedly reducedin the small intestine, heart, liver and brain of the CKD an-imals. Liver protein abundance of folate and thiamin trans-porters was significantly reduced in the CKD animals when

compared with the sham-operated controls. Furthermore,we found a significant reduction in mitochondrial folateand thiamin transporters in the CKD animals.Conclusions. CKD results in marked down-regulation inthe expression of folate and thiamin transporters in theintestine, heart, liver and brain. These events can leadto reduced intestinal absorption and impaired cellularhomeostasis of these essential micronutrients despite theirnormal plasma levels.

Keywords: anemia; cardiovascular disease; malnutrition; neuropathy;uremia

Introduction

Chronic kidney disease (CKD) has emerged as a majorpublic health problem worldwide [1]. CKD is associatedwith an increased risk of cardiovascular disease, neuro-logic disorders, malnutrition and progression to end-stagerenal disease. Vascular and neurologic complications inparticular remain an important source of morbidity andmortality in this population [2,3].

The water-soluble vitamins are a group of structurallyand functionally unrelated compounds that share the com-mon feature of being essential for normal cellular function,growth and development. Folate and thiamin (vitamin B1)

Chronic kidney disease and vitamin transporter expression 2137

© The Author 2010. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.For Permissions, please e-mail: [email protected]

Dow



how to search the world wide web using yahoo! search

Documents