Ia

YT

ARA

KAIKKLLMP

1

dctiaoYopahsmit

0h

Experimental and Toxicologic Pathology 65 (2013) 817– 823

Contents lists available at SciVerse ScienceDirect

Experimental and Toxicologic Pathology

jo ur nal homepa ge: www.elsev ier .de /e tp

mmunohistochemistry of LAMP-2 and adipophilin for phospholipidosis in livernd kidney in ketoconazole-treated mice

oshiji Asaoka, Yuko Togashi, Naoko Imura, Takafumi Sai, Tomoya Miyoshi, Yohei Miyamoto ∗

oxicology and Pharmacokinetics Laboratories, Pharmaceutical Research Laboratories, Toray Industries, Inc., 6-10-1, Tebiro, Kamakura, Kanagawa 248-8555, Japan

a r t i c l e i n f o

rticle history:eceived 20 August 2012ccepted 19 November 2012

eywords:dipophilin

mmunohistochemistryetoconazoleidneyAMP-2iverice

hospholipidosis

a b s t r a c t

Drug-induced phospholipidosis is an abnormal accumulation of phospholipids in the lysosomes followingrepeated administration of cationic amphiphilic drugs. Phospholipidosis is detected histopathologicallyas cytoplasmic vacuolation; however, it is difficult to distinguish from lipid accumulation since their mor-phological features are similar. In this study, we investigated the usefulness of immunohistochemistryfor lysosome-associated membrane protein-2 (LAMP-2) and adipophilin, a membrane protein of cyto-solic non-lysosomal lipid droplets, in the liver and kidneys of mice orally administered ketoconazole,an inducer of hepatic phospholipidosis. In 7-week-old mice administered ketoconazole (300 mg/kg/day)for 7 days, cytoplasmic vacuolation was histopathologically observed in centrilobular hepatocytes andproximal tubular epithelial cells under the fasted condition. The cytoplasmic vacuolation consisted offoamy vacuoles, which were revealed to be phospholipidosis-characteristic lamellar bodies by elec-tron microscopy. Furthermore, lipid-like vacuoles were observed in the perilobular hepatocytes, andrevealed to be lipid droplets by electron microscopy. In immunohistochemistry, the foamy vacuoles and

lipid-like vacuoles were positive for LAMP-2 and adipophilin, respectively. These results indicate thatimmunohistochemistry for LAMP-2 and adipophilin could distinguish between phospholipidosis andlipid accumulation. Additionally, it could detect ketoconazole-induced phospholipidosis in the glycogen-rich livers of non-fasted mice. In conclusion, ketoconazole induced phospholipidosis in not only the liverbut also the kidneys, and immunohistochemistry for LAMP-2 and adipophilin could be useful for thepathological evaluation of drug-induced phospholipidosis in mice.

C

. Introduction

Drug-induced phospholipidosis is a phospholipid storageisorder characterized by the formation of phospholipid-drugomplexes in lysosomes, which appear as lamellar bodies underhe transmission electron microscope. The first case of drug-nduced phospholipidosis in human occurred after treatment withn antianginal drug, 4,4′-diethylaminoethoxyhexestrol which isne of cationic amphiphilic drugs (CADs) (Chatman et al., 2009;amamoto et al., 1971), and then, the phospholipidosis was alsobserved in rats (Yamamoto et al., 1976). CADs, including antide-ressants, and antianginal, antimalarial and cholesterol-loweringgents, have side chains with a charged hydrophilic ring andydrophobic regions, and are known to induce phospholipido-is (Halliwell, 1997; Reasor and Kacew, 2001). Although the

olecular mechanism underlying CAD-induced phospholipidosis

s still not fully understood, it is considered to be associated withhe formation of an indigestible phospholipid-drug complex, the

∗ Corresponding author. Tel.: +81 467 329665; fax: +81 467 329768.E-mail address: Youhei Miyamoto@nts.toray.co.jp (Y. Miyamoto).

940-2993/$ – see front matter. Crown Copyright © 2012 Published by Elsevier GmbH. Attp://dx.doi.org/10.1016/j.etp.2012.11.008

rown Copyright © 2012 Published by Elsevier GmbH. All rights reserved.

suppression of phospholipase activity, and an imbalance in phos-pholipid turnover (Halliwell, 1997; Reasor and Kacew, 2001). Inaddition, some CADs have adverse effects such as acute pneu-monitis or hepatitis in clinical settings (Guigui et al., 1988; Lewiset al., 1984; Okumura et al., 1983). The relationship betweendrug-induced phospholipidosis and clinical side effects remainsunclear; however, phospholipidosis would be a significant concernfor drug development, and evaluations in non-clinical studies arevery important to reduce the risk in human.

Phospholipidosis is observed as cytoplasmic vacuolation onlight microscopy, but difficult to distinguish from lipid (non-phospholipid) accumulation since their morphological features aresimilar. Transmission electron microscopy is utilized to achievea definitive diagnosis of phospholipidosis, but the throughput isvery low. Lysosome-associated membrane protein-2 (LAMP-2) isdistributed ubiquitously to cells and involved in the direct uptakeof cytosolic proteins in lysosomes and participates in the endo-cytic pathway (Cuervo and Dice, 1996; Lichter-Konecki et al., 1999;

Tanaka et al., 2000). Adipophilin is distributed to the membraneof lipid droplets in restricted cells such as lactating mammaryepithelial cells, adrenal cortex cells and steatosis hepatocytes,and functions in the packaging, transportation and deposition of

ll rights reserved.

8 oxico

cfwdcbe

irmurd(u

hpap

2

2

YbhYo(nKIdaard&

2

amf(iT

2

eto

2

r5(

18 Y. Asaoka et al. / Experimental and T

ytosolic lipid droplets (Heid et al., 1998). Immunohistochemistryor LAMP-2 and adipophilin to evaluate hepatic phospholipidosisas reported in rats (Obert et al., 2007). Furthermore, in silico pre-ictions based on physicochemical properties, in vitro assays usingultured cells, and biomarkers for detecting phospholipidosis haveeen investigated (Fischer et al., 2012; Hirode et al., 2008; Monteitht al., 2006; Tomizawa et al., 2006).

Mice are widely used in non-clinical research such as toxic-ty and pharmacological studies and as experimental animals inesearch and development for new drug candidates. Geneticallyodified mice such as the p53+/− and rasH2 strain’s are often

sed for studying carcinogenicity. In addition, mice have drawnenewed attention as a suitable model for the early stages of drugevelopment because the compound requirements are much lowerKramer et al., 2010). Therefore, it is considered that the easy eval-ation for drug-induced phospholipidosis in mice is valuable.

In this study, we investigated the usefulness of immuno-istochemistry for LAMP-2 and adipophilin using formalin-fixedaraffin sections for the detection of phospholipidosis in mice orallydministered ketoconazole, a well-known inducer of hepatic phos-holipidosis (Pakuts et al., 1990; Whitehouse et al., 1994).

. Materials and methods

.1. Animal experiments

Male Crlj:CD1(ICR) mice (Charles River Laboratories, Japan Inc.,okohama, Japan) were housed in a plastic cage with softwood chipedding under controlled conditions (12 h light/dark cycle, 40–60%umidity at 19–25 ◦C), and fed on a standard diet (CRF-1, Orientaleast Co. Ltd., Tokyo, Japan) and tap water ad libitum. Seven-week-ld mice were divided into 4 groups and administered ketoconazole300 mg/kg/day; Tokyo Chemical Industry Ltd., Tokyo, Japan) orot with/without fasting for approximately 15 h before autopsy.etoconazole in a 0.25% tragacanth gum (Wako Pure Chemical

ndustries Ltd., Tokyo, Japan) solution was administered orally onceaily for 7 days. A day after the final administration, the mice werenesthetized with pentobarbital sodium, bled from the postcava,nd euthanized by exsanguination. Then, the liver and kidneys wereemoved, weighed and fixed. The animal experiments were con-ucted according to Guidelines for Animal Experiments, Research

Development Division, Toray Industries Inc.

.2. Blood chemistry

The blood samples were centrifuged at 1870 × g for 15 mint 4 ◦C, and the resulting plasma was stored below −60 ◦C untileasurements were made. Plasma levels of aspartate aminotrans-

erase (AST), alanine aminotransferase (ALT), alkaline phosphataseALP), leucine aminopeptidase (LAP), urea nitrogen (UN) and creat-nine were determined using a 7070 analyzer (HitachiKoki Co. Ltd.,okyo, Japan).

.3. Histopathology

Livers and kidneys were fixed in 10% neutral buffered formalin,mbedded in paraffin, and cut into 4 �m thick sections. These sec-ions were stained with hematoxylin and eosin (H.E.) and observedn light microscopy.

.4. Single immunostainings for LAMP-2 and adipophilin

The sections were deparaffinized and treated with epitopeetrieval solution (ImmunoSaver, Nissin EM Ltd., Tokyo, Japan) for

min at 98 ◦C for antigen retrieval using a microwave processorMI-77, Azumaya Corp., Tokyo, Japan). Then, the sections were

logic Pathology 65 (2013) 817– 823

incubated with a peroxidase-blocking solution and a blocking solu-tion for non-specific protein. The sections were incubated withthe primary antibody; either a rat monoclonal antibody againstmouse LAMP-2 (Santa Cruz BioTech. Inc., Burlingame, CA, USA) at1:1000 or a guinea pig polyclonal antibody against adipophilin (Pro-gen Biotechnik Co., Heidelberg, German) at 1:1000, for 60 min atroom temperature. The sections were incubated with the secondaryantibodies and a VECTASTAIN Elite ABC kit (Vector LaboratoriesInc., Burlingame, CA, USA); subsequently, the target antigen wasdetected with an ImmPACTTM DAB peroxidase substrate kit (VectorLaboratories Inc.).

2.5. Double immunostaining for LAMP-2 and adipophilin

LAMP-2, the first target antigen, was stained brown in the samemanner as for single immunostaining. For double immunostaining,the sections were incubated with avidin and biotin blocking kitreagent (Vector Laboratories Inc.) to block the avidin and biotinactivities related with the first target. Then, the sections wereincubated with the blocking solution for non-specific protein and,the primary antibody, the guinea pig polyclonal antibody againstadipophilin at 1:1000, for 60 min at room temperature. The sectionswere incubated with the secondary antibody and VECTASTAIN EliteABC kit (Vector Laboratories Inc.). Finally, adipophilin, the secondtarget antigen, was detected with an ImmPACTTM VIP peroxidasesubstrate kit (Vector Laboratories Inc.) as a purple chromogen.

2.6. Electron microscopy

Livers and kidneys were fixed in 2.5% glutaraldehyde, post-fixedin 1% osmium tetroxide, and embedded with epoxy resin. Ultra-thin sections were stained with 2% uranyl acetate and lead citrate,and observed under a transmission electron microscope (H7000,Hitachi High-Technologies Co. Ltd., Tokyo, Japan).

2.7. Statistical analysis

The significance of differences between the vehicle-treated andketoconazole-treated mice were analyzed with Student’s t-test orWelch’s t-test.

3. Results

3.1. Blood chemistry and organ weights

The plasma levels of AST, ALT, ALP and LAP in the ketoconazole-treated mice under fasted and non-fasted conditions increasedcompared to those in the vehicle-treated control mice (Table 1).The plasma levels of UN and creatinine in the ketoconazole-treatedmice increased relatively little, though the change was significantfor creatinine. Liver weight increased approximately 1.5 fold in theketoconazole-treated mice compared to the vehicle-treated mice.Kidney weight did not change (Table 2).

3.2. Histopathology of the liver

In the control mice, variably sized, round, clear, membrane-bound vacuoles were observed in the perilobular hepatocytes ofthe fasted animals (Fig. 1A). These vacuoles are herein referred toas lipid-like vacuoles. Conversely, irregular non-membrane-boundvacuoles were observed in the hepatocytes of the non-fasted ani-mals (Fig. 2A). These vacuoles are referred to as glycogen-like

vacuoles.

In the ketoconazole-treated mice, cytoplasmic vacuolation inthe centrilobular hepatocyte was observed in addition to lipid-like vacuoles in the peripheral hepatocytes in the fasted animals.

Y. Asaoka et al. / Experimental and Toxicologic Pathology 65 (2013) 817– 823 819

Table 1Blood chemistry in the ketoconazole-treated mice under fasted and non-fasted conditions.

Condition Fasted Non-fasted

Test article Vehicle Ketoconazole Vehicle KetoconazoleThe number of animals 4 4 5 6

AST (U/L) 42 ± 12 119 ± 29** 35 ± 2 155 ± 57**

ALT (U/L) 19 ± 6 92 ± 47 21 ± 4 121 ± 47**

ALP (U/L) 310 ± 99 907 ± 795 240 ± 57 535 ± 236 *

LAP (U/L) 46 ± 4 88 ± 22* 49 ± 7 92 ± 10**

UN (mg/dL) 26.6 ± 4.5 36.8 ± 2.3 17.4 ± 3 26.2 ± 5Creatinine (mg/dL) 0.15 ± 0.06 0.16 ± 0.03 0.08 ± 0.02 0.12 ± 0.03*

Data represent as the means ± SD.* P < 0.05, significantly different from the vehicle control mice.

** P < 0.01, significantly different from the vehicle control mice.

Table 2Organ weights in the ketoconazole-treated mice under fasted and non-fasted conditions.

Condition Fasted Non-fasted

Test article Vehicle Ketoconazole Vehicle KetoconazoleThe number of animals 4 4 5 6

Liver (g) 1.24 ± 0.12 1.98 ± 0.13 ** 2.08 ± 0.21 3.12 ± 0.32 **

Kidney (mg) 416 ± 59 446 ± 30 542 ± 53 559 ± 32

D

Tmrlha

tkd

FahH

ata represent as the means ± SD.* P < 0.05, significantly different from the vehicle control mice.** P < 0.01, significantly different from the vehicle control mice.

he cytoplasmic vacuolation consisted of round to irregular, clearembrane-bound vacuoles (Fig. 1D). These vacuoles are herein

eferred to as foam-like vacuoles. In the non-fasted animals, lipid-ike vacuoles and foamy vacuoles were observed in the hepatocytesaving glycogen-like vacuoles (Fig. 2D). Foamy Kupffer cells werelso detected.

Electron microscopy revealed variably sized lipid droplets in

he hepatocytes in the fasted control mice (Fig. 3A). In the fastedetoconazole-treated mice, variably sized lamellar bodies and lipidroplets were observed in the hepatocytes (Fig. 3B).

ig. 1. Histopathology and immunohistochemistry of livers in fasted mice. In the vehicle-nd are immunohistochemically positive for adipophilin and negative for LAMP-2 (B and Cepatocytes, and the lipid-like vacuoles are seen in the perilobular hepatocytes (D). Foam.E. staining, B and E: LAMP-2 immunostaining, C and F: adipophilin immunostaining. CV

3.3. Immunohistochemistry of the liver

Among the control mice, lipid-like vacuoles in the hepatocytesof the fasted animals were positive for adipophilin and negative forLAMP-2 (Fig. 1B and C), while glycogen-like vacuoles in the hepa-tocytes of the non-fasted animals were negative for both LAMP-2and adipophilin (Fig. 2B and C). Furthermore, granules positive for

LAMP-2 were observed along cytoplasmic membranes of the hepa-tocytes in the non-fasted mice. Small vacuoles in the Ito cells werepositive for adipophilin and negative for LAMP-2.

treated control mice, lipid-like vacuoles are seen in the perilobular hepatocytes (A),). In the ketoconazole-treated mice, foam-like vacuoles are seen in the centrilobulary vacuoles are positive for LAMP-2 and negative for adipophilin (E and F). A and D:: central vein, P: portal triad. A bar shows 50 �m.

820 Y. Asaoka et al. / Experimental and Toxicologic Pathology 65 (2013) 817– 823

Fig. 2. Histopathology and immunohistochemistry of livers in non-fasted mice. In the control mice, glycogen-like vacuoles are seen in the hepatocytes (A), and are immuno-histochemically negative for both LAMP-2 and adipophilin (B and C). Granules positive for LAMP-2 are seen along cytoplasmic membranes of the hepatocytes (B). In thek the ha tivelyi

cLh2p

tIckir

3

bckpm

Ft

etoconazole-treated mice, foam-like vacuoles and lipid-like vacuoles are seen innd lipid-like vacuoles are positive for LAMP-2 and adipophilin (E and F), respecmmunostaining. CV: central vein, P: portal triad. Bars show 50 �m.

In the ketoconazole-treated mice under fasted and non-fastedonditions, foamy vacuoles in the hepatocytes tested positive forAMP-2 and negative for adipophilin, and lipid-like vacuoles in theepatocytes were positive for adipophilin and negative for LAMP-

(Figs. 1E and F, 2E and F). In addition, foamy Kupffer cells wereositive for LAMP-2 and negative for adipophilin.

In the double immunostaining for LAMP-2 and adipophilin,he satiability was similar to that in each single immunostaining.n the fasted control mice, lipid-like vacuoles in the hepato-ytes were positive for adipophilin (Fig. 4A and B). In the fastedetoconazole-treated mice, foamy vacuoles and lipid-like vacuolesn the hepatocytes tested positive for LAMP-2 and adipophilin,espectively (Fig. 4C and D).

.4. Histopathology of the kidneys

In the control mice, small lipid-like vacuoles were observed atasal sites of the proximal tubular epithelial cells under the fasted

ondition, but not non-fasted condition (Fig. 5A and 6A). In theetoconazole-treated mice, foamy vacuoles were observed in theroximal tubular epithelial cells in both fasted and non-fasted ani-als (Figs. 5D and 6D).

ig. 3. Electron microscopy of livers in fasted mice. In the control mice, numerous variablyreated mice, numerous variably sized lamellar bodies (*) and lipid droplets are seen in th

epatocytes having glycogen-like vacuoles (D), and foamy vacuoles (arrowheads). A and D: H.E. staining, B and E: LAMP-2 immunostaining, C and F: adipophilin

Electron microscopy revealed relatively large lysosomes in theproximal tubular epithelial cells in the fasted control mice (Fig. 7A).In the fasted ketoconazole-treated mice, variably sized lamellarbodies were observed in proximal tubular epithelial cells (Fig. 7B).

3.5. Immunohistochemistry of the kidneys

In the control experiment, lipid-like vacuoles in the proximaltubular epithelial cells tested positive for adipophilin and negativefor LAMP-2 in the fasted mice (Fig. 5B and C). Granules positivefor LAMP-2 were observed at apical sites of the proximal tubularepithelial cells in the non-fasted animals (Fig. 6B and C).

In the ketoconazole-treated mice, foamy vacuoles in theproximal tubular epithelial cells under fasted and non-fasted con-ditions were positive for LAMP-2 and negative for adipophilin(Figs. 5E and F, 6E and F).

4. Discussion

Ketoconazole, an antifungal agent, induces hepatic phospholipi-dosis in mice and rats (Hirode et al., 2008; Whitehouse et al., 1994).Ketoconazole is converted to a number of metabolites by hepatic

sized lipid droplets (arrowheads) are seen in the hepatocytes (A). In ketoconazole-e hepatocytes (B). Bars show 5 �m.

Y. Asaoka et al. / Experimental and Toxicologic Pathology 65 (2013) 817– 823 821

Fig. 4. Immunohistochemistry for LAMP-2 and adipophilin by double staining in livers of fasted mice. In the control mice (A and B), lipid-like vacuoles in the hepatocytesw zole-tL e by ai show

mipvllmtt

FcsC

ere stained purple by adipophilin immunostaining (arrows) (B). In the ketoconaAMP-2 immunostaining (arrowheads), and lipid-like vacuoles were stained purplmmunostaining for LAMP-2 and adipophilin. CV: central vein, P: portal triad. Bars

icrosomes, and a major metabolite, de-N-acetyl ketoconazole,n mice, is considered to be responsible for ketoconazole-inducedhospholipidosis (Whitehouse et al., 1994). In this study, foamyacuoles appeared in hepatocytes and proximal tubular epithe-ial cells in ketoconazole-treated mice under fasted conditions, and

ipid-like vacuoles were also observed in the hepatocytes. Electron

icroscopy revealed the foamy vacuoles and lipid-like vacuoleso be lamellar bodies and lipid droplets, respectively, indicatinghat ketoconazole induced phospholipidosis in not only the liver

ig. 5. Histopathology and immunohistochemistry of the kidneys in fasted mice. In the conells (A), and are immunohistochemically positive for adipophilin and negative for LAMPites of the proximal tubular epithelial cells (D), and are positive for LAMP-2 and negative

and F: adipophilin immunostaining. Bars show 50 �m.

reated mice (C and D), foamy vacuoles in the hepatocytes were stained brown bydipophilin immunostaining (arrows) (D). A and C: H.E. staining, B and D: double-50 �m.

but also the kidneys in mice. Ketoconazole was reported to causephospholipidosis in the liver, lung and spleen (Whitehouse et al.,1994; Pakuts et al., 1990), but little is known about its effects onthe kidneys in humans or animals. Thus, our findings about renalphospholipidosis are the first of their kind in ketoconazole-treated

mice.

Since phospholipidosis resembles lipid accumulation on lightmicroscopy, electron microscopy is generally utilized to achieve adefinitive diagnosis, though the throughput is very low. Recently,

trol mice, lipid-like vacuoles are seen at basal sites of the proximal tubular epithelial-2 (B and C). In the ketoconazole-treated mice, foamy vacuoles are seen at apical

for adipophilin (E and F). A and D: H.E. staining, B and E: LAMP-2 immunostaining,

822 Y. Asaoka et al. / Experimental and Toxicologic Pathology 65 (2013) 817– 823

F ce. In

p les areL P-2 im

imftllm2hrtidi

aail

usuf

Fc

ig. 6. Histopathology and immunohistochemistry of the kidneys in non-fasted miroximal tubular epithelial cells (B). In the ketoconazole-treated mice, foamy vacuoAMP-2 and negative for adipophilin (E and F). A and D: H.E. staining, B and E: LAM

t was reported that immunohistochemistry for LAMP-2 as aarker for lysosome membranes and adipophilin as a marker

or lipid droplets was useful for evaluating phospholipidosis inhe livers of rats (Obert et al., 2007). In the fasted control mice,ipid-like vacuoles in the hepatocytes and proximal tubular epithe-ial cells, which were revealed to be lipid droplets by electron

icroscopy, tested positive for adipophilin and negative for LAMP-. In the fasted ketoconazole-treated mice, foamy vacuoles in theepatocytes and proximal tubular epithelial cells, which wereevealed to be lamellar bodies by electron microscopy, were posi-ive for LAMP-2 and negative for adipophilin. Given these results,mmunohistochemistry for LAMP-2 and adipophilin is useful foristinguishing between phospholipidosis and lipid accumulation

n the liver and kidneys in mice.Furthermore, double immunostaining for LAMP-2 and

dipophilin also could distinguish phospholipidosis from lipidccumulation in the liver in the ketoconazole-treated mice. Thismmunostaining revealed the distribution of phospholipidosisesion and lipid droplets on the same slide.

It is difficult to detect phospholipidosis in a glycogen-rich liversing light microscopy, because its histopathological features are

imilar to those of lipid accumulation. However, glycogen-like vac-oles in the non-fasted ketoconazole-treated mice were negativeor both LAMP-2 and adipophilin. Thus, the immunohistochemistry

ig. 7. Electron microscopy of the kidneys in fasted mice. In the control mice, a small numbells (A). In the ketoconazole-treated mice, a large number of variably sized lamellar bod

the control mice (A–C), granules positive for LAMP-2 are seen at apical sites of the seen at apical sites of the proximal tubular epithelial cells (D), and are positive formunostaining, C and F: adipophilin immunostaining. Bars show 50 �m.

was able to distinguish between phospholipidosis and lipid accu-mulation in the glycogen-rich liver. Additionally, granules positivefor LAMP-2 were observed in the hepatocytes and proximal tubularepithelial cells in the non-fasted control mice, and considered to benormal lysosomes.

Sudan III and oil red O are used for staining fatty tissues, andacid hematin staining is used to distinguish between phospho-lipids and non-phospholipids. These traditional stains need frozensections untreated with alcohol and xylene, whereas immunohis-tochemistry for LAMP-2 and adipophilin can to be performed withformalin-fixed paraffin sections.

The relationship between drug-induced phospholipidosis andtoxicity remains unclear. However, phospholipidosis might beassociated with gentamicin-induced nephrotoxicity, since the lat-ter was suppressed by inhibition of phospholipid-drug complexformation in proximal tubular epithelial cells (Kishore et al., 1992;Samadian et al., 1993). Additionally, phospholipidosis is consideredone of the causes of the severe liver injury induced by ketocona-zole treatment in humans (Lewis et al., 1984; Okumura et al.,1983). In this study, ketoconazole increased levels of hepatotoxicmarkers, ALT, AST, ALP and LAP, and liver weight in mice, and

these changes might be related with phospholipidosis. Further-more, CYP2A, CYP3A and CYP4A were reported to be induced inmouse liver by ketoconazole (Whitehouse et al., 1994; Sapone et al.,

er of relatively large lysosomes (arrows) are seen in the proximal tubular epithelialies (*) are seen in the proximal tubular epithelial cells (B). Bars show 5 �m.

oxico

2wabake

oisd

A

a

R

C

C

C

F

G

H

H

H

K

Y. Asaoka et al. / Experimental and T

003; Casley et al., 2007), which might be associated with the livereight gain in our study. Conventional nephrotoxic markers, UN

nd creatinine, did not change in the ketoconazole-treated mice,ut it would be useful to measure sensitive markers such as Kim-1nd �-GST for a better understanding of the relationship betweenetoconazole-induced phospholipidosis and nephrotoxicity (Swaint al., 2011, 2012).

In conclusion, ketoconazole induced phospholipidosis in notnly the liver but also the kidneys in mice, and immunohistochem-stry for LAMP-2 and adipophilin using formalin-fixed paraffinections could be useful for evaluating drug-induced phospholipi-osis in the mouse liver and kidneys.

cknowledgments

We are grateful to Kohei Ueda and Yasushi Miyauchi for theirssistance with the electron microscopic evaluation.

eferences

asley WL, Ogrodowczyk C, Larocque L, Jaentschke B, LeBlanc-Westwood C, Men-zies JA, et al. Cytotoxic doses of ketoconazole affect expression of a subsetof hepatic genes. Journal of Toxicology and Environmental Health Part: A2007;70:1946–55.

hatman LA, Morton D, Johnson TO, Anway SD. A strategy for risk management ofdrug-induced phospholipidosis. Toxicologic Pathology 2009;37:997–1005.

uervo AM, Dice JF. A receptor for the selective uptake and degradation of proteinsby lysosomes. Science 1996;273:501–3.

ischer H, Atzpodien EA, Csato M, Doessegger L, Lenz B, Schmitt G, et al. In silico assayfor assessing phospholipidosis potential of small druglike molecules: training,validation, and refinement using several data sets. Journal of Medical Chemistry2012;55:126–39.

uigui B, Perrot S, Berry JP, Fleury-Feith J, Martin N, Metreau JM, et al. Amiodarone-induced hepatic phospholipidosis: a morphological alteration independent ofpseudoalcoholic liver disease. Hepatology 1988;8:1063–8.

alliwell WH. Cationic amphiphilic drug-induced phospholipidosis. ToxicologicPathology 1997;25:53–60.

eid HW, Moll R, Schwetlick I, Rackwitz HR, Keenan TW. Adipophilin is a specificmarker of lipid accumulation in diverse cell types and diseases. Cell and TissueResearch 1998;294:309–21.

irode M, Ono A, Miyagishima T, Nagao T, Ohno Y, Urushidani T. Gene expres-sion profiling in rat liver treated with compounds inducing phospholipidosis.Toxicology and Applied Pharmacology 2008;229:290–9.

ishore BK, Ibrahim S, Lambricht P, Laurent G, Maldague P, Tulkens PM. Comparativeassessment of poly-l-aspartic and poly-l-glutamic acids as protectants againstgentamicin-induced renal lysosomal phospholipidosis, phospholipiduria andcell proliferation in rats. Journal of Pharmacology and Experimental Therapeu-tics 1992;262:424–32.

logic Pathology 65 (2013) 817– 823 823

Kramer JA, O’Neill E, Phillips ME, Bruce D, Smith T, Albright MM, et al. Early toxicologysignal generation in the mouse. Toxicologic Pathology 2010;38:452–71.

Lewis JH, Zimmerman HJ, Benson GD, Ishak KG. Hepatic injury associatedwith ketoconazole therapy, analysis of 33 cases. Gastroenterology 1984;86:503–13.

Lichter-Konecki U, Moter SE, Krawisz BR, Schlotter M, Hipke C, Konecki DS. Expres-sion patterns of murine lysosome-associated membrane protein 2 (Lamp-2)transcripts during morphogenesis. Differentiation 1999;65:43–58.

Monteith DK, Morgan RE, Halstead B. In vitro assays and biomarkers for drug-induced phospholipidosis. Expert Opinion on Drug Metabolism and Toxicology2006;2:687–96.

Obert LA, Sobocinski GP, Bobrowski WF, Metz AL, Rolsma MD, Altrogge DM, et al. Animmunohistochemical approach to differentiate hepatic lipidosis from hepaticphospholipidosis in rats. Toxicologic Pathology 2007;35:728–34.

Okumura H, Aramaki T, Satomura K, Iizuka K, Ohta M, Katsuta Y, et al. Severe hepatitisduring ketoconazole therapy. Gastroenterologia Japonica 1983;18:142–7.

Pakuts AP, Parks RJ, Paul CJ, Bujaki SJ, Mueller RW. Ketoconazole-inducedhepatic lysosomal phospholipidosis: the effect of concurrent barbiturate treat-ment. Research Communications in Chemical Pathology and Pharmacology1990;67:55–62.

Reasor MJ, Kacew S. Drug-induced phospholipidosis: are there functional conse-quences? Experimental Biology and Medicine (Maywood) 2001;226:825–30.

Samadian T, Dehpour AR, Amini S, Nouhnejad P. Inhibition of gentamicin-induced nephrotoxicity by lithium in rat. Histology and Histopathology 1993;8:139–47.

Sapone A, Pozzetti L, Canistro D, Broccoli M, Bronzetti G, Potenza G, et al. Inductionand suppression of cytochrome P450 isoenzymes and generation of oxygen radi-cals by procymidone in liver, kidney and lung of CD1 mice. Mutation Research2003;527:67–80.

Swain A, Turton J, Scudamore C, Maguire D, Pereira I, Freitas S, et al. Nephrotoxi-city of hexachloro-1,3-butadiene in the male Hanover Wistar rat; correlation ofminimal histopathological changes with biomarkers of renal injury. Journal ofApplied Toxicology 2012;32:417–28.

Swain A, Turton J, Scudamore C, Pereira I, Viswanathan N, Smyth R, et al. Urinarybiomarkers in hexachloro-1,3-butadiene-induced acute kidney injury in thefemale Hanover Wistar rat; correlation of �-glutathione S-transferase, albuminand kidney injury molecule-1 with histopathology and gene expression. Journalof Applied Toxicology 2011;31:366–77.

Tanaka Y, Guhde G, Suter A, Eskelinen EL, Hartmann D, Lüllmann-Rauch R, et al.Accumulation of autophagic vacuoles and cardiomyopathy in LAMP-2-deficientmice. Nature 2000;406:902–6.

Tomizawa K, Sugano K, Yamada H, Horii I. Physicochemical and cell-based approachfor early screening of phospholipidosis-inducing potential. The Journal of Toxi-cological Sciences 2006;31:315–24.

Whitehouse L, Menzies A, Mueller R, Pontefract R. Ketoconazole-induced hepaticphospholipidosis in the mouse and its association with de-N-acetyl ketocona-zole. Toxicology 1994;94:81–95.

Yamamoto A, Adachi S, Kitani T, Shinji Y, Seki K. Drug-induced lipidosis in humancases and in animal experiments, accumulation of an acidic glycerophospho-

lipid. The Journal of Biochemistry 1971;69:613–5.

Yamamoto A, Adachi S, Matsuzawa Y, Kitani T, Hiraoka A, Seki K. Studies on drug-induced lipidosis: VII. Effects of bis-beta-diethyl-aminoethylether of hexestrol,chloroquine, homochlorocyclizine, prenylamine, and diazacholesterol on thelipid composition of rat liver and kidney. Lipids 1976;11:616–22.