http://informahealthcare.com/phbISSN 1388-0209 print/ISSN 1744-5116 online

Editor-in-Chief: John M. PezzutoPharm Biol, Early Online: 1–8

! 2014 Informa Healthcare USA, Inc. DOI: 10.3109/13880209.2014.960944

ORIGINAL ARTICLE

A novel animal model of metabolic syndrome with non-alcoholic fattyliver disease and skin inflammation

Nagaraj M. Kulkarni1,2, Mallikarjun S. Jaji1, Pranesha Shetty1, Yeshwant V. Kurhe1, Shilpee Chaudhary1, G. Vijaykant1,J. Raghul1, Santosh L. Vishwakarma1, B. Navin Rajesh1, Jeyamurugan Mookkan1, Uma Maheswari Krishnan2,and Shridhar Narayanan2

1Department of Biology, Drug Discovery Research, Orchid Chemicals and Pharmaceuticals Ltd., Chennai, Tamil Nadu, India and 2Centre for

Nanotechnology and Advanced Biomaterials (CeNTAB), School of Chemical and Biotechnology, Sastra University, Thanjavur, Tamil Nadu, India

Abstract

Context: Metabolic syndrome and non-alcoholic fatty liver disease (NAFLD) are the emergingco-morbidities of skin inflammation. Occurrence of skin inflammation such as psoriasis issubstantially higher in NAFLD patients than normal. Currently, there are no animal models tostudy the interaction between these co-morbidities.Objective: The present study seeks to develop a simple mouse model of NAFLD-enhanced skininflammation and to study the effect of NAFLD on different parameters of skin inflammation.Materials and method: Metabolic syndrome and NAFLD were induced in C57BL/6 mice byfeeding high-fat diet (HFD, 60% kcal) and high fructose liquid (HFL, 40% kcal) in drinking water.Skin inflammation was induced by repeated application of oxazolone (1% sensitization andrepeated 0.5% challenge) in both normal and NAFLD mice and various parameters of skininflammation and NAFLD were measured.Results: HFD and HFL diet induced obesity, hyperglycemia, hyperinsulinemia, and histologicalfeatures of NAFLD in mice. Oxazolone challenge significantly increased ear thickness, earweight, MPO activity, NF-kB activity, and histological features of skin inflammation in NAFLDmice as compared with normal mice. Overall, induction of oxazolone-induced skin inflamma-tion was more prominent in NAFLD mice than normal mice. Hence, HFD and HFL diet followedby topical oxazolone application develops metabolic syndrome, NAFLD, and enhanced skininflammation in mice.Discussion and conclusion: This simple model can be utilized to evaluate a therapeutic strategyfor the treatment of metabolic syndrome and NAFLD with skin inflammation and also tounderstand the nexus between these co-morbidities.

Keywords

Animal model, metabolic syndrome,non-alcoholic fatty liver disease (NAFLD),nuclear factor-kappa B, oxazolone, skininflammation

History

Received 16 July 2014Revised 27 August 2014Accepted 28 August 2014Published online 27 November 2014

Introduction

Non-alcoholic fatty liver disease (NAFLD) refers to a

histological spectrum of liver damage from simple steatosis

to steatohepatitis in those without a history of excessive

alcohol consumption (520 g/d) (Angulo et al., 2002). NAFLD

is considered as the manifestation of metabolic syndrome in

the liver. The metabolic syndrome is a condition characterized

by a cluster of alterations including insulin resistance, obesity,

dyslipidemia, hypertension, a pro-inflammatory, and pro-

thrombotic state (Almeda-Valdes et al., 2009). The develop-

ment of NAFLD is strongly associated with metabolic

syndrome as reflected by the fact that approximately 90%

of the patients with NAFLD have more than one feature of

metabolic syndrome and about 33% have three or more criteria

(Marchesini et al., 2003). Increasing severity of NAFLD

represents worsening inflammatory and insulin-resistant state

with poorer metabolic outcome (Bhatia et al., 2012).

There are a number of observations in the literature linking

NAFLD and metabolic syndrome to systemic inflammation

(Bhatia et al., 2012; Hamminga et al., 2006). The liver is a

key metabolic organ and plays important role in the regulation

of systemic inflammation. Liver disease like NAFLD

increases various inflammatory mediators like TNF-a, IL-6,

C-reactive protein, glucose, and plasminogen activator

inhibitor-1 (Bhatia et al., 2012). Obesity and diabetes are

associated with chronic low-grade inflammatory state through

release of pro-inflammatory cytokines including TNF-a, and

IL-6 (Hamminga et al., 2006). In rodents, a high-fat diet

(HFD) results in NAFLD and up-regulation of NF-kB

activity, which leads to hepatic production of various pro-

inflammatory cytokines and activation of Kupffer cells and

macrophages (Bhatia et al., 2012; Cia et al., 2005).

Interestingly, several recent studies have found prevalence

of NAFLD and metabolic syndrome in skin inflammation

like psoriasis (Gisondi et al., 2009; Matsumoto et al., 2004).

Correspondence: Dr. Shridhar Narayanan, Vice President and Head,Infection iScience, AstraZeneca India Pvt. Ltd., Off Bellary Road,Hebbal, Bangalore 560024, Karnataka, India. Tel: +91 9611598805.E-mail: shridhar.narayanan@gmail.com

Phar

mac

eutic

al B

iolo

gy D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y L

aure

ntia

n U

nive

rsity

on

12/0

8/14

For

pers

onal

use

onl

y.

Gisondi et al. (2009) reported that occurrence of NAFLD is

substantially more in psoriasis patients than in controls (47%

versus 28%, p50.001). Several medications used to treat

skin inflammation are known to cause liver damage, which

can affect the outcome of drugs used to treat metabolic

syndrome, especially NAFLD. Drugs like methotrexate

cause NAFLD-like histological changes in liver and hence

may worsen the pre-existing NAFLD in patients with skin

inflammation (Wenk et al., 2010). Moreover, no specific

therapy is approved for the treatment of NAFLD.

Animal models of human disease are necessary to allow

the study of and understand the efficacy and safety of drugs at

real-life level. The ideal disease model should include all the

characteristics of the disease along with the co-morbidities.

HFD with high-fructose liquid (HFL) in drinking water has

been shown to induce NAFLD in mice (Tetri et al., 2008).

Chronic oxazolone-induced experimental skin inflammation

is being used as a preliminary model for screening of novel

drugs for skin inflammation like dermatitis and psoriasis in

drug discovery (Peterson, 2006; Yeom et al., 2012). The aim

of the present study was to develop a simple mouse model of

NAFLD-enhanced skin inflammation and to study the effect

of NAFLD on different parameters of skin inflammation.

Materials and methods

Animals

Male C57BL/6 mice of 4–5 weeks age and 13–18 g body

weight were procured from the National Institute of Nutrition,

Hyderabad, India. All mice were maintained in 12 h light/dark

cycle with free access to standard laboratory chow diet and

water ad libitum in controlled environment (23 ± 2 �C). Mice

were handled according to the guidelines of experimental

animal care issued by the committee for the purpose of

control and supervision of experiments on animals (CPCSEA)

and the experimental protocol was approved by the institu-

tional animal ethical committee (IAEC) (Protocol no. 04/

IAEC-01/PCP/2011).

Materials

Normal chow diet was procured from Nutri Lab� Rodent

(Tetragon Chemie Pvt. Ltd., Bangalore, India) and 60 kcal%

HFD procured from Open Source diets (Cat# D12492, New

Brunswik, NJ). Fructose was procured from Sisco Research

Laboratories (Mumbai, India). Oxazolone was purchased

from Sigma (St Louis, MO). All other reagents were of the

highest commercially available grade.

Methodology

Male C57BL/6 mice were divided into two groups. The first

group (n¼ 22) was fed with normal chow feed and the second

group (n¼ 29) fed with 60 kcal% HFD and HFL 40 % for

60 d. On the 45th day, two animals from each group were

sacrificed; liver histopathology examination was carried out

to confirm the induction of NAFLD. At the end of induction

period (day 60), the remaining animals were grouped as

follows. Mice fed with normal chow diet (n¼ 20) were further

divided into three groups; normal untreated control (n¼ 8),

normal + vehicle (n¼ 6), and normal + oxazolone (n¼ 6).

Mice fed with HFD and HFL (n¼ 27) were further divided

into three groups (n¼ 9) as untreated NAFLD control,

NAFLD + vehicle, and NAFLD + oxazolone. Basal ear thick-

ness was measured and animals were sensitized with 1%

oxazolone or vehicle (acetone:olive oil, 4:1) on both ears

(20 ml each). Seven days later, the animals were challenged

with 0.5% oxazolone or vehicle for 2 weeks as shown in

Figure 1(a). Ear thickness was measured with a digital

micrometer (Mitutoyo, Kanagawa, Japan) before and 24 h post

each challenge. On 15th day, animals were kept for overnight

fasting and, on 16th day, blood was collected from retro-

orbital sinus under light isoflurane anesthesia and plasma

separated by centrifugation at 6000 rpm for 10 min for

biochemical analysis. Following blood collection, animals

were sacrificed by cervical dislocation and liver, ear, and fat

pad were collected, weighed, and frozen (�80 �C) for further

analysis. Part of liver and ear tissues were kept in 10%

formalin for histopathology examination.

Biochemical estimations in plasma

Plasma glucose, triglyceride (TG), cholesterol (TC), total

bilirubin, total protein, alanine aminotransaminase (ALT),

aspartate aminotransferase (AST), and alkaline phosphatase

(ALP) were measured from plasma by random access clinical

chemistry analyzer Erba XL 300 using commercially avail-

able ERBA diagnostics kit (OpenSource Diets, New

Brunswik, NJ). Plasma insulin was estimated by using

ELISA kits following manufacturer’s instructions (Millipore,

Billerica, MA).

Cytokines in serum and ear homogenates

Briefly, the treated mouse ear tissue extracts were prepared by

tissue lysis with T-PER buffer (Pierce, Rockford, IL) supple-

mented with 1� Protease inhibitor cocktail (Calbiochem,

San Diego, CA) and 1� phosphatase inhibitor cocktail

(Calbiochem, San Diego, CA) using a tissue homogenizer.

Homogenates were then centrifuged at 14 000 rpm for 30 min.

IL-6 and TNF-a were measured in the supernatant (tissue

homogenate) or serum using ELISA kits following manufac-

turer’s instructions (GE Healthcare, Little Chalfont, UK).

Myeloperoxidase (MPO) activity in plasma and earhomogenates

MPO activity was measured in plasma and ear homogenates

(homogenized as described above for assessment of cyto-

kines) (De Vry et al., 2005). Briefly, 50 ml of plasma or

supernatant from ear homogenate was added to 200 ml of an

assay reaction mixture containing 0.5% hexadecyltrimethylm-

monium bromide (in 50 mM potassium phosphate, pH 6.4),

0.165 mg/ml o-dianisidine hydrochloride and 0.0015% H2O2.

Absorbance was measured at 460 nm.

Liver triglycerides

The liver lipids were extracted using a modified Folch

extraction protocol (Folch et al., 1957). Briefly, approxi-

mately 100 mg of liver tissue was homogenized with

methanol (1 ml) then the homogenate was centrifuged at

4000 rpm for 5 min and supernatant transferred into separate

2 N.M. Kulkarni et al. Pharm Biol, Early Online: 1–8

Phar

mac

eutic

al B

iolo

gy D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y L

aure

ntia

n U

nive

rsity

on

12/0

8/14

For

pers

onal

use

onl

y.

tubes (15 ml). Then the pellet was re-suspended in chloro-

form:methanol (2:1) solution for 2 min. The homogenate was

then centrifuged and supernatant separated and mixed with

first supernatant. About 0.1 M potassium chloride was added

to supernatant and mixed well by vortexing. After centrifu-

gation of the mixture, bottom phase (organic phase) was

removed to a new tube (2 ml). The samples were evaporated

in Turbovap LV evaporator (GE Healthcare, Little Chalfont,

UK). Residue was reconstituted in 400 ml of mixture of

N-butyl alcohol:Triton X-100:methanol (3:1:1) and mixed

properly by vortexing. The samples were used for the

triglyceride estimation using commercially available kit

(Erba Diagnostics).

Histopathology

Liver and ear tissue samples were fixed in 10% formalin and

embedded in paraffin. Sections measuring 3–5 mm in thick-

ness were cut and stained with hematoxylin and eosin (H&E).

Liver sections were examined using a light microscope

(NIKON, ECLIPSE-E200, Tokyo, Japan) and graded by the

in-house method based on Hubscher (2006) and Brunt and

Tiniakos (2002). It included scoring of microvacuolation,

namely 1 (mild), 2 (moderate), and ballooning degeneration

as 1 (mild), 2 (moderate), 3 (marked), and 4 (severe).

Ear tissues were analyzed quantitatively for epidermal

thickness and inflammatory infiltrate by an investigator

blinded to treatment as described previously (Boehncke

et al., 2001). Briefly, maximal epidermal thickness was

measured from the tip of the rete ridges to the border of the

viable dorsal or ventral epidermis. The mean value of six such

rete ridges was measured using the ocular micrometer of the

microscope at 400� each division measuring 2.45mm. The

inflammatory grading was done based on the presence of

dermal inflammation without Munro’s abscess (score 0.5),

mild inflammation with Munro’s abscess (score 1), moderate

inflammation with Munro’s abscess (score 2), and marked

inflammation with Munro’s abscess (score 3).

Western blot analysis

Briefly, the treated mice ear tissue extracts were prepared by

tissue lysis as described earlier for the cytokine estimation.

The protein content was determined using a bicinchoninic acid

assay (BCA� protein assay kit, Pierce, Appleton, WI) and

samples were prepared in SDS, bromophenol blue, and

100 mg/mL DTT, and boiled for 5 min at 99 �C. The 50 mg

of protein was resolved on a 10% SDS-PAGE and blotted

on to a nitrocellulose membrane and the blots were blocked

for 1 h at room temperature with 5% non-fat milk protein in

(a)

(b)(c)

0 1 2 3 4 5 60.0

0.1

0.2

0.3

0.4

0.5

0.6 Normal+vehicle

Normal+oxazolone

NAFLD+vehicle

NAFLD+oxazolone***

******

***

#####

###

###

$$$$$$

$$$

Oxazolone challenge

Ear

Thi

ckne

ss (

mm

)

0.0

2.5

5.0

7.5

10.0

12.5

15.0 **

##

##

Normal + Vehicle Normal +OxazoloneNAFLD + Vehicle NAFLD+Oxazolone

Ear

wei

ght

(mg)

(e)(d)

0

100

200

Normal+vehicle Normal+oxazolone

NAFLD+vehicle NAFLD+oxazolone

$$**#

Cha

nge

in M

PO

acti

vity

(pl

asm

a)(P

erce

nt r

elat

ive

to n

orm

al+

vehi

cle)

0

100

200

300

400

Normal+vehicle Normal+oxazoloneNAFLD+vehicle NAFLD+oxazolone

**$$

##

Cha

nge

in M

PO

acti

vity

(Ea

r)(P

erce

nt r

elat

ive

to n

orm

al+

vehi

cle)

Day -60 Day 0 Day 1 Day 7 Day 9 Day 11 Day 13 Day 15 Day 16

Dietary Manipulation Grouping Oxazolone 1 % Sensitization Oxazolone 0.5 % Challenge End of study

Figure 1. (a) Experimental design. (b–c) Effect of oxazolone on ear edema and ear weight in normal and NAFLD mice. (d and e) Effect of oxazoloneon change in MPO activity in plasma and ear homogenate in normal and NAFLD mice. Change in MPO activity is expressed as percent relative to thenormal + vehicle group (n¼ 6–9, $$p50.01, $$$p50.001 as compared with normal + vehicle, **p50.01, ***p50.001 as compared withNAFLD + vehicle, and #p50.05, ##p50.01, and ###p50.001 as compared with normal + oxazolone, one-way ANOVA followed by the Newman–Keuls multiple comparison test).

DOI: 10.3109/13880209.2014.960944 Novel animal model of NAFLD and skin inflammation 3

Phar

mac

eutic

al B

iolo

gy D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y L

aure

ntia

n U

nive

rsity

on

12/0

8/14

For

pers

onal

use

onl

y.

Tris-buffered saline (TBS) containing 0.1% Tween 20, then

blots were probed with antibodies against anti-rabbit pNF-kB

(Cell Signaling Technology, Danvers, MA) and anti-rabbit

GAPDH (Cell Signaling Technology, Danvers, MA) over-

night at 4 �C. Horseradish peroxidase-conjugated secondary

antibody was subsequently incubated with blot for 1 h room

temperature. The ECL (Enhanced chemiluminescence; Pierce,

Appleton, WI) was used to visualize signal (Li et al., 2012).

Statistical analysis

All values are expressed as mean ± standard error of the mean

(SEM). The Graphs were generated using Graph-Pad Prism�

(Version 4, GraphPad Software, La Jolla, CA). Statistical

analysis was performed by Student ‘t’ test or one-way

ANOVA followed by the Newman–Keuls multiple compari-

son test or two-way ANOVA followed by the Bonferroni post

test as applicable. For histopathology, non-parametric analysis

was done by the Mann–Whitney U test or the Kruskal Wallis

test followed by Dunn’s multiple comparison test as applic-

able. Results were considered statistically significant at

p50.05.

Results

There was no significant difference observed between normal

control versus normal + vehicle and NAFLD control versus

NAFLD + vehicle, indicating vehicle did not affect any of

the parameters in both normal and NAFLD mice. Hence, the

data for normal control and NAFLD control are not shown in

the results.

Induction of obesity and NAFLD with dietarymanipulation

Induction of obesity and NAFLD in mice was achieved by

feeding HFD and HFL diet to C57BL/6 mice. Feeding mice

with HFD and HFL significantly increased body weight

(p50.01), epididymal (p50.01), and inguinal fat (p50.01)

as compared with animals on the normal chow diet (Table 1).

On 45th day of dietary manipulation, two animals were

sacrificed and induction of NAFLD in HFD- and HFL-

fed mice was confirmed by the histopathological examin-

ation for the presence of ballooning degenerations and

microvacoulations (data not shown). Oxazolone or vehicle

application did not affect the body weight and fat pad in both

normal and HFD- and HFL-fed mice.

Induction of hyperglycemia and hyperinsulinemiawith dietary manipulation-

Hyperglycemia and hyperinsulinemia are important criteria of

metabolic disorder. Mice fed with HFD and HFL diet showed

a significant increase in fasting plasma glucose (p50.01) and

insulin (p50.01) as compared to mice on normal chow diet

(Table 1). Mice fed with HFD and HFL caused elevation in

plasma ALT levels indicating the induction of NAFLD in this

model. Dietary manipulation did not cause any significant

difference in other biochemical parameters tested compared

to normal chow-fed mice (Table 1). Oxazolone application in

both normal and NAFLD mice did not affect any of the

biochemical parameters as compared with their respective

controls (Table 1).

Induction of ear edema in normal and NAFLD mice

Ear edema was induced by topical application of oxazolone

to both ears of normal and NAFLD mice as shown in

Figure 1(b–e). Ear thickness was measured before and

24 h post each oxazolone challenge. Ear thickness

increased significantly from the second challenge onwards

in NAFLD + oxazolone mice as compared with

NAFLD + vehicle mice (p50.001) and remained till the end

of the experiment (p50.001) (Figure 1b). Whereas oxazolone

application to the normal mice showed significant increase in

ear thickness from third challenge onwards and remained till

the end of the experiment (p50.001). Ear thickness in

NAFLD + oxazolone mice was significantly higher than

normal + oxazolone mice (p50.001). This increase in ear

thickness was statistically significant from the second chal-

lenge onwards till the end of the experiment, indicating

increased inflammatory response in NAFLD mice. At the end

of the study, animals were sacrificed and ear weight was

recorded. Oxazolone application significantly increased ear

weight in both normal and NAFLD mice (p50.01)

(Figure 1c). Ear weight of NAFLD + oxazolone mice was

significantly higher than normal + oxazolone mice (p50.01).

Table 1. Effect of oxazolone on body weight, fat pad and biochemical parameters in normal and NAFLD mice.

Normal + vehicle Normal + oxazolone NAFLD + vehicle NAFLD + oxazolone

Body weight (g) 24.33 ± 1.33 25.05 ± 1.16 33.87 ± 1.29$$ 32.83 ± 1.07$$Epididymal fat (g) 0.52 ± 0.08 0.42 ± 0.03 1.45 ± 0.16$$ 1.40 ± 0.15$$Inguinal fat (g) 0.40 ± 0.07 0.30 ± 0.04 1.23 ± 0.19$$ 1.12 ± 0.13$$Glucose (mg/dl) 160.05 ± 10.44 158.15 ± 3.99 311.71 ± 15.31$$ 272.17 ± 13.68$$Insulin (ng/ml) 0.350 ± 0.30 0.323 ± 0.40 1.599 ± 0.31$$ 1.710 ± 0.34$$Triglyceride (mg/dl) 103.5 ± 13.22 76.83 ± 6.45 87.89 ± 2.32 81.44 ± 5.24Total cholesterol (mg/dl) 111.67 ± 6.76 112.17 ± 2.06 125.33 ± 5.00 115.56 ± 5.84ALT (IU/l) 39.65 ± 3.41 48.7 ± 4.32 59.33 ± 12.29 64.49 ± 14.28AST (IU/l) 100.03 ± 10.78 82.1 ± 7.75 92.63 ± 10.20 80.98 ± 6.03ALP (IU/l) 228.17 ± 29.41 227.67 ± 9.30 216.33 ± 6.92 204.56 ± 13.57Total bilirubin (mg/dl) 0.43 ± 0.14 0.27 ± 0.04 0.54 ± 0.08 0.36 ± 0.05Total protein (g/dl) 5.75 ± 0.11 5.80 ± 0.09 5.58 ± 0.07 5.65 ± 0.07TNF-a (pg/ml) 12.58 ± 3.16 13.92 ± 3.19 26.17 ± 3.77$$ 29.56 ± 4.57$$IL-6 (pg/ml) 15.36 ± 2.89 19.12 ± 5.11 38.22 ± 3.69$$ 45.15 ± 6.58$$

Values are expressed as mean ± SEM (n¼ 6–9).$$p50.01 as compared with normal + vehicle, one-way ANOVA followed by the Newman–Keuls multiple comparison test.

4 N.M. Kulkarni et al. Pharm Biol, Early Online: 1–8

Phar

mac

eutic

al B

iolo

gy D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y L

aure

ntia

n U

nive

rsity

on

12/0

8/14

For

pers

onal

use

onl

y.

Effect of oxazolone on MPO and cytokine levels innormal and NAFLD mice

MPO activity, an indicator of polymorphonuclear leukocyte

influx, was measured in both plasma and ear homogenate.

Oxazolone application significantly increased MPO activity

in plasma and ear homogenate in both normal and NAFLD

mice (p50.01) as compared with their respective control

(Figure 1d). In accordance with ear thickness and ear weight,

MPO activity in plasma and ear homogenate was signifi-

cantly higher in the NAFLD + oxazolone group than the

normal + oxazolone group (p50.05, p50.01) (Figure 1d

and e). TNF-a and IL-6 levels were measured in serum and

ear homogenate. Serum TNF-a, and IL-6 levels were

significantly elevated in NAFLD mice as compared

with normal mice (Table 1). In ear homogenate, there was

no significant difference observed in TNF-a levels in

both normal and NAFLD mice (data not shown). While

IL-6 levels significantly increased in normal + oxazolone

(285.48 ± 17.61 pg/ml versus 558.10 ± 53.88 pg/ml) and

NAFLD + oxazolone groups (260.71 ± 18.60 pg/ml versus

504.52 ± 54.47 pg/ml) as compared with their respective

controls, no significant difference was observed between

normal + oxazolone and NAFLD + oxazolone groups.

Effect of dietary manipulation on liver triglyceridelevels and liver histopathology

Liver triglyceride levels were significantly higher in NAFLD

mice than normal mice (p50.01) indicating accumulation of

lipid in the liver with HFD and HFL diet (Figure 2a).

Microvacuolations and ballooning degenerations were scored

by a pathologist blinded to treatment. Liver sections of HFD-

and HFL-fed mice showed significant induction of ballooning

degenerations and microvacuolations (Figure 2b and d).

Animals fed with normal chow diet did not show any

pathology of NAFLD (Figure 2c). Application of oxazolone

on ears did not affect the liver histopathology in both normal

and NAFLD mice.

Effect of oxazolone on ear histopathology in normaland NAFLD mice

Application of oxazolone to ears showed psoriasis and

dermatitis like histopathology in both normal and NAFLD

mice (Figure 3). Munro’s microabscesses were present in

both normal + oxazolone and NAFLD + oxazolone groups

(Figure 3d). Epidermal thickness, as measured using an ocular

micrometer by an investigator blinded to treatment, was

significantly higher in both normal + oxazolone and

NAFLD + oxazolone groups as compared with their respect-

ive controls (p50.01) (Figure 3e). Epidermal thickness

in the NAFLD + oxazolone group was greater than the

normal + oxazolone group. Dermal inflammatory score was

significantly more in the NAFLD + oxazolone group than

the normal + oxazolone group (Figure 3f). Inflammation

was not observed in un-induced normal and NAFLD mice

(Figure 3a). Overall, induction of skin histopathology

resembling psoriasis and dermatitis was more prominent in

the NAFLD + oxazolone group than the normal + oxazolone

group (Figure 3b and c).

Effect of oxazolone on NF-kB expression in normaland NAFLD mice

NF-kB is a key regulatory element in a variety of immune and

inflammatory pathways and hence is a crucial mediator

involved in the pathogenesis of various skin inflammatory

(a)(b)

0

10

20

30

40

Normal NAFLD

**

Liver

TG

(mg/

g o

f tiss

ue)

-0.50.00.51.01.52.02.53.03.54.04.5

Normal

NAFLD***

***

Balloniong degenerations Microvacuolations

Balloning

degenerations/

Microvacuolations

(c) (d)

Figure 2. (a) Elevation in liver triglyceride (TG) levels by dietary manipulation with HFD and HFL diet. (b) Ballooning degenerations andmicrovacoulations score in liver histopathology indicating induction of NAFLD by HFD and HFL diet. (c) Liver histopathology section of normalchow-fed mice. (d) Liver histopathology section of HFD and HFL-fed mice, showing hepatic steatosis (solid arrow) and ballooning degenerations(dotted arrow) (n¼ 6–9, **p50.01, ***p50.001 as compared with normal, the Mann–Whitney U test).

DOI: 10.3109/13880209.2014.960944 Novel animal model of NAFLD and skin inflammation 5

Phar

mac

eutic

al B

iolo

gy D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y L

aure

ntia

n U

nive

rsity

on

12/0

8/14

For

pers

onal

use

onl

y.

conditions. Chronic oxazolone treatment in mice strongly

stimulated the phosphorylation of NF-kB in both normal and

NAFLD mice. Phosphorylation of NF-kB was significantly

higher in NAFLD mice than in normal mice treated with

oxazolone (p50.01). However, there was no significant

difference observed between normal and NAFLD mice

treated with vehicle (Figure 4).

Discussion

This is the first attempt to evaluate the effect of NAFLD on

skin inflammation and eventually to develop an animal

model. Recent literature indicates that the association among

skin inflammation, metabolic syndrome, and NAFLD is

growing in patient population (Gerdes et al., 2010; Gottlieb

et al., 2008; Saraceno et al., 2008; Wenk et al., 2010).

However, the exact pathogenic mechanism between these co-

morbidities is still unclear. Moreover, several medications

used to treat skin inflammation are also known to induce liver

diseases. Lack of proper animal model has hampered the

understanding of the nexus between such co-morbidities and

also evaluation of potential pharmacological interventions.

Hence, the animal model developed in the present study will

enable us to study the interaction among NAFLD, metabolic

syndrome, and skin inflammation and also to evaluate

different therapeutic strategies for the treatment of such

co-morbidities.

In the present study, we induced metabolic syndrome and

NAFLD first by combination of HFD and HFL in drinking

water to C57BL/6 mice for 2 months. Combination of HFD

and HFL is associated with the development of obesity,

increased de novo lipogenesis in the liver, hepatic steatosis,

impaired glucose tolerance, insulin resistance, and hyperten-

sion (Tetri et al., 2008). Mice fed with HFD and HFL

experienced rapid gain in body weight and increased fat pad

weight as compared with normal chow-fed mice.

Histopathological examination of liver clearly demonstrated

that the HFD- and HFL-fed mice showed ballooning degen-

erations and microvacuolations indicating the development of

NAFLD. The induction of NAFLD was further confirmed by

increased liver triglyceride levels and plasma level of ALT.

NAFLD mice also showed significant elevation of plasma

(a)

(c)

(e)

(b)

(d)

(f)

0

10

20

30

40

Normal + vehicle Normal + OxazoloneNAFLD + vehicle NAFLD + Oxazolone

**

$$

Epi

dermal

thickness

(�m)

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Normal+oxazoloneNAFLD+oxazolone

#

$

Normal+vehicleNAFLD+vehicle

***

Dermal

Inflammatory score

Figure 3. Effect of oxazolone on ear histopathology in normal and NAFLD mice. Mice ears were excised 24 h after oxazolone challenge and stainedwith hematoxylin and eosin. Mice applied with vehicle showed normal histology (a). Challenging with oxazolone caused significant increase inepidermal thickness (solid arrow) and inflammatory infiltration (dotted arrow) in both normal (b) and NAFLD (c) mice (more pronounced in theNAFLD mice). Ear histopathology also showed the presence of Munro’s microabscess (solid arrow) (d). Epidermal thickness (e) and dermalinflammatory score (f) were higher in the NAFLD + oxazolone group as compared with normal + the oxazolone group (n¼ 6–9, $p50.05, $$p50.01as compared with normal + vehicle, **p50.01 as compared with NAFLD + vehicle, and #p50.05, as compared with normal + oxazolone, theKruskal–Wallis test followed by Dunn’s multiple comparison test).

6 N.M. Kulkarni et al. Pharm Biol, Early Online: 1–8

Phar

mac

eutic

al B

iolo

gy D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y L

aure

ntia

n U

nive

rsity

on

12/0

8/14

For

pers

onal

use

onl

y.

glucose and insulin levels along with the elevation of

circulating cytokines like TNF-a, and IL-6 in serum than

normal mice. These data suggest that the current animal

model displayed the most important criteria of metabolic

syndrome like obesity and insulin resistance along with

NAFLD.

For pharmacological screening of new drugs for skin

inflammation, acute and chronic oxazolone-induced skin

inflammation is being used as the preliminary screening

model in drug discovery for years. This model is particularly

useful for the evaluation of broad acting anti-inflammatory

drugs (Peterson, 2006). Oxazolone application to ears causes

histopathological changes similar to that of psoriasis and

atopic dermatitis. Moreover, standard drugs like hydrocorti-

sone, phosphodiesterase inhibitors, and dexamethasone show

significant inhibitory activity in this model (Peterson, 2006).

In the present study, application of oxazolone to the ears of

normal C57BL/6 mice caused significant increase in ear

thickness, ear weight, MPO activity, and NF-kB expression.

Histopathology of ear displayed epidermal thickness and

infiltration of inflammatory cells along with the presence of

Munro’s microabscess, a characteristic histopathological

feature of psoriasis (Boehncke et al., 2001). A similar

application of oxazolone to NAFLD mice improved the

induction of disease compared with that of normal mice. Ear

thickness, ear weight, MPO activity, and NF-kB activity were

significantly higher in NAFLD mice than that of normal mice.

Incidence of microabscess, epidermal thickness, and dermal

inflammatory score were also higher in NAFLD mice than

normal mice. The data from this study clearly show that the

presence of co-morbidities like obesity, insulin resistance, and

NAFLD increased the severity of oxazolone-induced skin

inflammation.

These results are in accordance with clinical reports

where the incidence of skin inflammation was found higher

in patients with metabolic disorder and NAFLD (Gisondi

et al., 2009; Saraceno et al., 2008; Wenk et al., 2010).

Elevated levels of pro-inflammatory cytokines, especially

TNF-a in adipose tissue, are an important feature of obesity

and contribute significantly to insulin resistance and devel-

opment of NAFLD (Baker et al., 2011; Wenk et al., 2010).

Accumulation of free fatty acids within hepatocytes causes

hepatic TNF-a production and further increasing the

inflammatory condition in metabolic syndrome. Adipocyte

and hepatic TNF-a may act on skin to worsen skin

inflammation by promoting keratinocyte proliferation,

increased inflammation, and the upregulation of vascular

adhesion molecules and transcription factor NF-kB (Bhatia

et al., 2012; Wenk et al., 2010). In the present study, serum

levels of TNF-a and IL-6 were significantly higher in

NAFLD mice indicating the presence of chronic low-grade

systemic inflammation. This increased systemic inflamma-

tion may be responsible for enhanced skin inflammation in

NAFLD mice.

Notably, several studies have indicated that the increase in

the activation of factor NF-kB has an important role in the

development and maintenance of cutaneous inflammatory

diseases, including psoriasis, contact dermatitis, and atopic

dermatitis (Bell et al., 2003). In the skin, the transcription of

cytokines, such as IL-1, IL-6, and TNF-a, and many of the

effectors of cytokine action, such as vascular cell adhesion

molecule-1, and cyclo-oxygenase-2, are regulated by NF-kB

(De Vry et al., 2005). In this study, oxazolone treatment

increased NF-kB activation in ear tissue of both normal and

NAFLD mice, with pronounced increase in NAFLD mice

treated with oxazolone than normal mice. These data clearly

shows that NAFLD-enhanced oxazolone-induced skin inflam-

mation may be mediated by increased NF-kB activation.

Further, MPO activity, an indicator of neutrophil accumula-

tion and a characteristic feature of cutaneous inflammation,

was significantly higher in plasma and ear tissues of NAFLD

mice than normal mice, indicating the increased severity of

inflammation in NAFLD mice.

Most of the drugs used for the treatment of skin

inflammation are known to cause liver damage and there

are no formal guidelines regarding systemic treatment of skin

inflammation in NAFLD patients. Methotrexate induced

liver damage resembles that of NAFLD and may worsen

the condition if used in such patient population (Wenk et al.,

2010). The present animal model can be used to study the

effect of anti-inflammatory drugs in the presence of both

metabolic syndrome and NAFLD. In our preliminary studies,

orally administered methotrexate caused severe mortality at

1 mg/kg in this model and worsened NAFLD, whereas it was

well tolerated in normal mice (unpublished data). This finding

encourages us to further utilize this model to study the effect

of such drugs alone and in combination with drugs used in the

management of metabolic syndrome on NAFLD and skin

inflammation. Hence, pharmacotherapy of skin inflammation

and NAFLD in the presence of co-morbidities can be

established using this model.

(a)

(b)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

Normal+vehicle NAFLD+vehicle

Normal+oxazoloneNAFLD+oxazolone

$$$

***##

Relative Density

Normal +oxazolone

Normal +Vehicale

NF–kB p65 kDa

GAPDH 3– kDa

NAFLD +Vehicale

NAFLD +oxazolone

Figure 4. Effect of oxazolone on NF-kB activity in normal andNAFLD mice. (a) Mice ears were excised 24 h after last oxazolonechallenge and analyzed for phosphorylation levels of NF-kB.(b) GAPDH was used as a control and relative density after vehicle oroxazolone treatment was expressed from two independent experiments(n¼ 4, $$$p50.001 as compared with normal + vehicle, ***p50.001 ascompared with NAFLD + vehicle, and ###p50.001, as compared withnormal + oxazolone, one-way ANOVA followed by the Newman–Keulsmultiple comparison test).

DOI: 10.3109/13880209.2014.960944 Novel animal model of NAFLD and skin inflammation 7

Phar

mac

eutic

al B

iolo

gy D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y L

aure

ntia

n U

nive

rsity

on

12/0

8/14

For

pers

onal

use

onl

y.

In conclusion, the present study is an attempt to develop an

animal model of metabolic syndrome and NAFLD-enhanced

skin inflammation. Dietary manipulation followed by topical

oxazolone application increases systemic inflammation and

NF-kB activation in skin resulting in metabolic syndrome and

NAFLD-enhanced skin inflammation in mice. This simple

animal model can be utilized to evaluate therapeutic strategy

for the treatment of metabolic syndrome and NAFLD with

skin inflammation like psoriasis or dermatitis and also to

understand the nexus between these co-morbidities.

Declaration of interest

The authors report no conflicts of interest. The authors alone

are responsible for the content and writing of this article.

References

Almeda-Valdes P, Cuevas-Ramos D, Anguilar-Salinas CA. (2009).Metabolic syndrome and non-alcoholic fatty liver disease. AnnHepatol 8:S18–14.

Angulo P. (2002). Nonalcoholic fatty liver disease. N Engl J Med 346:1221–31.

Baker RG, Hayden MS, Ghosh S. (2011). NF-kB, inflammation andmetabolic disease. Cell Metab 13:11–22.

Bell S, Degitz K, Quirling M, et al. (2003). Involvement of NF-kappaBsignalling in skin physiology and disease. Cell Signal 15:1–7.

Bhatia LS, Curzen NP, Calder PC, Byrne CD. (2012). Non-alcoholicfatty liver disease: A new and important cardiovascular risk factor?Eur Heart J 33:1190–200.

Boehncke W, Hardt-Weinelt K, Nilsson H, et al. (2001). Anatagnoisticeffects of the Staphylococcal Enterotoxin A mutant, SEA on psoriasisin SCID-hu xenogeneic transplantation model. J Invest Dermatol 116:596–601.

Brunt EM, Tiniakos DG. (2002). Pathology of steatohepatitis. Best PractRes Clin Gastroenterol 16:691–707.

Cia D, Yuan M, Frantz DF, et al. (2005). Local and systemic insulinresistance resulting from hepatic activation of IKK-beta and NF-kappaB. Nat Med 11:183–90.

De Vry CG, Valdez M, Lazarov M, et al. (2005). Topical application of anovel immunomodulatory peptide, RDP58, reduces skin inflammation

in the phorbol ester-induced dermatitis model. J Invest Dermatol 125:473–81.

Folch J, Lees M, Stanley GHS. (1957). A simple method for the isolationand purification of total lipids from animal tissues. J Biol Chem 226:497–509.

Gerdes S, Rostami-Yazdi M, Mrowietz U. (2010). Adipokines andpsoriasis. Exp Dermatol 20:81–7.

Gisondi P, Targher G, Zoppini G, Girolomoni G. (2009). Non-alcoholicfatty liver disease in patients with chronic plaque psoriasis. J Hepatol51:758–64.

Gottlieb AB, Chao C, Dann F. (2008). Psoriasis comorbidities.J Dermatol Treat 19:5–21.

Hamminga EA, Van Der Lely AJ, Neumann HAM, Thio HB. (2006).Chronic inflammation in psoriasis and obesity: Implications fortherapy. Med Hypotheses 67:768–73.

Hubscher SG. (2006). Histological assessment of non-alcoholic fattyliver disease. Histopathology 49:450–65.

Li M, Zhong X, He Z, et al. (2012). Effect of erythromycin on cigarette-induced histone deacetylase protein expression and nuclear factor-kBactivity in human macrophages in vitro. Int Immunopharmacol 12:643–50.

Marchesini G, Bugianesi E, Forlani G, et al. (2003). Nonalcoholic fattyliver, steatohepatitis, and the metabolic syndrome. Hepatology 37:917–23.

Matsumoto T, Suziki N, Watanabe H, et al. (2004). Nonalcoholicsteatohepatitis associated with psoriasis vulgaris. J Gastroenterol 39:1102–5.

Peterson TK. (2006). In vivo pharmacological disease models forpsoriasis and atopic dermatitis in drug discovery. Basic ClinPharmacol Toxicol 99:104–15.

Saraceno R, Ruzzetti M, De Martino MU, et al. (2008). Does metabolicsyndrome influence psoriasis? Eur Rev Med Pharmacol Sci 12:339–41

Tetri LH, Basaranoglu M, Brunt EM, et al. (2008). Severe NAFLD withhepatic necroinflammatory changes in mice fed trans fats and a high-fructose corn syrup equivalent. Am J Physiol Gastrointest LiverPhysiol 295:G987–95.

Wenk KS, Arrington KC, Ehrlich A. (2010). Psoriasis and non-alcoholic fatty liver disease. J Eur Acad Dermatol Venereol 25:383–91.

Yeom M, Kim S, Lee B, et al. (2012). Oral administration ofglucosylceramide ameliorates inflammatory dry-skin condition inchronic oxazolone-induced irritant contact dermatitis in the mouseear. J Dermatol Sci 67:101–10.

8 N.M. Kulkarni et al. Pharm Biol, Early Online: 1–8

Phar

mac

eutic

al B

iolo

gy D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y L

aure

ntia

n U

nive

rsity

on

12/0

8/14

For

pers

onal

use

onl

y.