Al

SPa

b

a

ARRA

KOUINA

1

doapfini

tdobcima

1d

Process Biochemistry 45 (2010) 1406–1414

Contents lists available at ScienceDirect

Process Biochemistry

journa l homepage: www.e lsev ier .com/ locate /procbio

staxanthin restricts weight gain, promotes insulin sensitivity and curtails fattyiver disease in mice fed a obesity-promoting diet

aravanan Bhuvaneswaria, Elumalai Arunkumara,eriyasamy Viswanathanb, Carani Venkatraman Anuradhaa,∗

Department of Biochemistry and Biotechnology, Annamalai University, Annamalai Nagar, Chidambaram 608 002, Tamil Nadu, IndiaDepartment of Pathology, Rajah Muthiah Medical College and Hospital, Annamalai University, Annamalai Nagar 608 002, Tamil Nadu, India

r t i c l e i n f o

rticle history:eceived 18 January 2010eceived in revised form 30 April 2010ccepted 15 May 2010

eywords:besitynbalanced diet

a b s t r a c t

The study evaluated the effects of astaxanthin (ASX) in mice rendered obese by feeding an unbalanceddiet. Adult male mice of body weight 25–35 g were fed either normal chow or a high fat-high fructosediet (HFFD). Fifteen days later, mice in each group were divided in to two and treated with either ASX(6 mg/kg b.w.) in olive oil or olive oil alone. The mice were killed at the end of 60 days. Insulin sensi-tivity, markers of liver injury, inflammation and nitro-oxidative stress, antioxidants and cytochrome P4502E1 (CYP2E1) activity were assayed. Liver structural integrity was also assessed by histology withhemotoxylin and eosin and Masson’s trichrome stains. HFFD-fed mice registered significant increase in

nsulin resistanceitro-oxidative stressstaxanthin

body weight and liver weight and displayed hyperglycemia, hyperinsulinemia and insulin resistance andelevated plasma aminotransferases. Lipid deposition, oxidative damage, defective antioxidant systemand upregulated transforming growth factor-� (TGF-�1) expression were observed in HFFD-fed mice.ASX supplementation promoted insulin sensitivity and prevented liver injury by decreasing CYP2E1,myeloperoxidase, and nitro-oxidative stress and by improving the antioxidant status in them. Lipid depo-

�1 exnefic

sition and increased TGF-new data showing the be

. Introduction

Obesity, an exaggeration of normal adiposity, is a global epi-emic characterized by excess fat in adipose tissue and a spectrumf metabolic derangements like dyslipidemia, glucose intolerancend insulin resistance (IR) [1]. Ectopic fat storage occurs in obesity,articularly in the liver leading to a condition termed nonalcoholicatty liver disease (NAFLD) characterized by varying degrees of livernjury that progresses from steatosis to steatoheptitis, fibrosis andecrosis. Due to its prominent association with IR/obesity, NAFLD

s regarded as the hepatic manifestation of metabolic syndrome [2].Fat accumulation triggers free radical production and sensitizes

he liver to additional necro-inflammatory insults, thus promotingisease progression [3]. Evidence for the close association betweenbesity and IR with low-grade inflammatory responses in liver haseen consistently documented [4]. Production of proinflammatory

ytokines and oxidative stress in the liver activate the residentmmune cells of the liver like Kupffer cells and stellate cells, which

ediate liver injury [5]. One such cytokine secreted by the hep-tic stellate cells is transforming growth factor-�1 (TGF-�1) that

∗ Corresponding author. Tel.: +91 0 4144 239141; fax: +91 0 4144 238080.E-mail address: cvaradha@hotmail.com (C.V. Anuradha).

359-5113/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.oi:10.1016/j.procbio.2010.05.016

pression induced by HFFD were also abolished by ASX. This study providesial effects of ASX in obese mice.

© 2010 Elsevier Ltd. All rights reserved.

has been implicated in fibrogenesis. TGF-�1, plays a central rolein early stages of fibrogenesis by promoting the accumulation ofa number of extracellular matrix proteins, specifically collagen byinhibiting its degradation and activating its biosynthesis [6].

Experimental studies have reported that animals fed a high fatplus high fructose diet (HFFD) for more than 2 months developweight gain, dyslipidemia, moderate hyperglycemia, oxidativestress and IR [7,8]. Besides, consumption of a calorie-rich dietresults in lipid accumulation, excess production of inflammatorycytokines, and macrophage infiltration that favour the progressionof liver disease [9]. Consequently, in this study, we adopted theHFFD-fed mice as a model of fatty liver disease.

Natural compounds have been proposed as potential therapeu-tic agents in the prevention and/or treatment of number of diseases.Being aware of the increasing prevalence of obesity/IR, it is imper-ative to test and identify the possible therapeutic role of bioactivecompounds that could be added to our diet. Astaxanthin (3,3′-dihydroxydiketo � carotene, ASX) is one such compound of interestused in this study. ASX is a red carotenoid pigment produced by

marine-algal species like Haemotococcus pluvialis [10], Chlorellazofingiensis [11] and Scenedesmus obliqus [12]. It is ubiquitous in themarine environment and is responsible for the pinkish-red hue ofthe flesh of sea animals that ingest the marine algae. ASX is achiev-ing commercial success in cosmetic industry and aquaculture and

Bioch

iAaapwiseasagdttl

2

2

UL

2

uAlcalom

ia3Tr

w2w(o(w

Fm(

ad1tT

S. Bhuvaneswari et al. / Process

s widely used as a nutraceutical by athletes and sportsmen [13].SX is also known for its pharmacological activities that includenticancer, anti-ulcerative, anti-inflammatory and antigenotoxicctivities [14]. There has been considerable interest on the disease-reventing effects of ASX. This is revealed by mounting evidence,hich show that addition of ASX reduces oxidative stress and

nflammation in cardiovascular diseases [15], restores blood pres-ure in spontaneously hypertensive (SHR) rats [16], increases thendurance capacity in mice against exercise-induced fatigue [17]nd improves memory in BALB/c mice [18]. Previous studies havehown its anti-obesity effects in mice fed high fat diet [19] andntidiabetic effects in db/db mice [20,21]. Although ASX has under-one such vast research by many scientists, its effects on fatty liverisease in obese mice have not yet been investigated. This study washerefore designed to investigate the effects of ASX on insulin sensi-ivity, hepatic function, inflammatory markers, TGF-�1 expression,ipid profile and oxidative and nitrosative stress in HFFD-fed mice.

. Materials and methods

.1. Chemicals and kits

Astaxanthin (ASX) was purchased from Sigma–Aldrich Pvt. Ltd., St. Louis, MO,SA. The solvents and chemicals of analytical grade were purchased from Himediaaboratories Pvt. Ltd., Mumbai, India.

.2. Animals

Adult male mus musculus albino mice of Swiss strain weighing 25–35 g weresed for the study. The animals were procured from and maintained in the Centralnimal House, Department of Experimental Medicine, Rajah Muthiah Medical Col-

ege and Hospital, Annamalai Nagar. They were individually housed under hygieniconditions (22–24 ◦C) in polypropylene cages under 12 h light/12 h dark cycle. Thenimals used in the present study were cared according to the principles and guide-ines of the Institutional Animal Ethical Committee. After acclimatization for a periodf 1 week, the animals were divided at random into four groups consisting of sixice each. The following groups were maintained for a total period of 60 days.

Group 1: CON—Animals received control diet and were orally administered oliveoil (0.3 ml/kg b.w./day) from day 16 till the end of the experimental period.Groups 2: HFFD—Animals received HFFD and were orally administered olive oil(0.3 ml/kg b.w./day) from day 16 till the end of the experimental period.Group 3: HFFD + ASX—Animals received HFFD and were orally administered ASXat a dose (6 mg/kg b.w./day) in olive oil from day 16 till the end of the experimentalperiod.Group 4: CON + ASX—Animals received control diet and were orally administeredASX in olive oil at a dose (6 mg/kg b.w./day) from day 16 till the end of the experi-mental period.

Food and water were provided ad libitum to animals. The dosage of ASX usedn this study is based on a previous report which tested three doses of ASX (1.2, 6,nd 30 mg/kg/day) for anti-obesity effects, among which ASX was effective at 6 and0 mg/kg [19]. Hence the minimum dosage (6 mg/kg) was selected for our study.he body weight was measured every 4th day and food intake of the animals wereecorded every 24 h. Food was replaced daily.

The high fat-high fructose diet (HFFD) of the following composition (g/100 g)as prepared in the laboratory: 45.0 fructose, 10.0 ground nut oil, 10.0 beef tallow,

2.5 casein, 0.3 dl-methionine, 1.2 vitamin mixture, 5.5 mineral mixture and 5.5heat bran. HFFD contained 45% (w/w) fructose (39% of calories), 20% (w/w) fat

10% beef tallow; 10% groundnut oil; 40% of calories), and 22.5% (w/w) casein (21%f calories). The standard laboratory chow consisted of 60% (w/w) starch, 22.08%w/w) protein and 4.38% (w/w) fat. The normal chow diet provided 382.61 cal/100 ghile HFFD provided 471.25 cal/100 g.

The development of IR was assessed on the 16th day of the experimental period.or this, fasting blood samples were collected from control and experimental ani-als for the measurement of glucose and insulin. Homeostatic model assessment

HOMA) values were computed for each group.

At the end of the 60th day, mice were fasted overnight before being

naesthetized with ketamine hydrochloride (30 mg/kg, i.m.) and then killed byecapitation. Blood was collected and plasma separated by centrifugation (1500 × g,5 min). Liver tissue was removed immediately and transferred to containers con-aining ice-cold saline. A 10% liver homogenate was prepared in ice-cold 0.1 Mris–hydrochloride (Tris–HCl) buffer, pH 7.4.

emistry 45 (2010) 1406–1414 1407

2.3. Glucose, insulin and insulin sensitivity indices

Plasma glucose and insulin were measured using kits obtained from Agappediagnostic Pvt. Ltd., Kerala and Accubind microwells, Monobind Inc., CA, USA, respec-tively. IR was assessed by computing homeostatic model assessment (HOMA) [22]and quantitative insulin sensitivity check index (QUICKI) [23]. The formulae usedare given below:

HOMA = Insulin(�U/ml) × glucose(mM)22.5

QUICKI = 1log(glucose mg/dl) + log(insulin �U/ml)

2.4. Aminotransferases, CYP2E1 and myeloperoxidase (MPO) activities

Alanine aminotransferase (ALT) and aspartate transaminase (AST) in plasmawere measured using kits procured from Agappe Diagnostic Pvt. Ltd. Kerala. OneInternational Unit is defined as �mol of pyruvate formed per minute under exper-imental conditions. CYP2E1 activity was determined by the method of Wu et al.by monitoring the oxidation of p-nitrophenol to p-nitrocatechol in the presence ofNADPH and oxygen at 546 nm in a spectrophotometer [24]. Hepatic MPO activitywas measured according to the method of Bradley et al. [25] with some modifi-cations. The liver tissue was homogenized in 50 mM phosphate buffer containing0.5% hexadecyltrimethyl-ammonium bromide, pH 6.0 (Sigma–Aldrich, St Louis,MO) and centrifuged at 2000 × g for 10 min at 4 ◦C. The supernatant (100 �l) wasadded to 1.9 ml of 10 mM phosphate buffer pH 6.0 and 1 ml of 1.5 M o-dianisidinehydrochloride (Sigma–Aldrich) containing 0.0005% (w/v) hydrogen peroxide. Thechanges in absorbance were recorded at 450 nm on a spectrophotometer for 5 minat 1 min intervals. Optical density units were converted into units of concentrationusing the molar absorptivity co-efficient for oxidized to o-dianisidine hydrochloride(10,062 M−1 cm−1).

2.5. Nitro-oxidative stress markers

The oxidative stress markers such as thiobarbituric acid reactive substances(TBARS) [26], lipid hydroperoxides (LHP) [27] and protein carbonyl [28] were quan-tified. Total nitrite as an index of nitric oxide [29] and nitrosothiol content [30]were assayed. The activities of enzymatic (superoxide dismutase (SOD), catalase(CAT), glutathione peroxidase (GPX) and glutathione-S-transferase (GST)) and non-enzymatic (vitamins C and E and reduced glutathione (GSH)) in the hemolysate andliver were measured by methods outlined elsewhere [31].

2.6. Plasma and hepatic lipid profile

Lipids were extracted from the liver tissue according to the method of Folch etal. [32]. The content of triglycerides (TG), cholesterol and free fatty acids (FFA) weremeasured in plasma and liver by standard methods reported elsewhere [31].

2.7. Western blotting analysis of TGF-ˇ1 in liver

Proteins were extracted from the frozen liver. Tissue was homogenized in an ice-cold extraction buffer containing 20 mM Tris–HCl at PH 7.4, 150 mM NaCl (sodiumchloride), 1 mM EDTA (ethylene diamine dichloroacetic acid), 0.5% Triton X-100,0.1% SDS (sodium dodecyl sulphate), 1 mM PMSF (phenyl methosulphonyl fluorideand 10 �l of protease inhibitor cocktail) and centrifuged (10,000 × g, 15 min at 4 ◦C).After measuring the protein content by the method of Lowry et al. [33], samples(50 �g protein) were separated by 12% sodium dodecyl sulphate-polyacyrlamidegel electrophoresis (SDS-PAGE) and transferred on to a nitrocellulose membrane at80 mA for 1 h and 15 min. Membrane was then blocked overnight in a blocking buffer[3% bovine serum albumin in Tris-buffered saline pH 7.5 containing 0.1% tween-20(TBST)]. Then, the membrane was incubated with TGF-�1 antibody (rabbit poly-clonal anti-TGF-�1, Santa Cruz Biotechnology (1:1000 dilution)) and agitated gentlyat 4 ◦C overnight. The membrane was washed with TBST for 5 min, and then incu-bated with anti-rabbit IgG conjugated with horseradish peroxidase (1:1500 dilution)for 2 h at room temperature. The protein bands were detected using diaminobenzi-dene (DAB) substrate. The bands were then scanned and the intensity was measuredusing densitometry software (AlphaEaseFC Software, Alphatech innotech Co., Johan-nesburg, South Africa). �-Actin was used as the internal control (1:1000 dilution).

2.8. Histology and immunohistochemistry

Four animals from each group were anesthetized using light ether, perfused with10% formalin and the liver tissue was separated and placed in buffered formalin. Theywere later sectioned using a microtome (4–5 �m thickness), dehydrated in graded

alcohol and embedded in paraffin. Sections were stained with hemotoxylin andeosin (H and E) or Masson’s trichrome stain (MT). For immunohistochemistry, 4 �mparaffin sections were deparaffinized with xylene and rehydrated with graded con-centrations of isopropyl alcohol. Separated sections were incubated with peroxideblocking reagent for 10 min, rinsed with phosphate buffer and again incubated withpower block solution for 10 min. Non-specific binding was minimized by leaving the

1 Biochemistry 45 (2010) 1406–1414

so(wtcwc

2

ov

3

3

mHm(ida44batlpoci

3

H

Fig. 1. Body weight changes of animals during the experimental period.

TL

Vw

TL

Vw

408 S. Bhuvaneswari et al. / Process

ections in 3% BSA in phosphate buffered saline for 30 min. Sections were incubatedvernight with anti-TGF-�1 (1:250 dilution) antibody or with anti-DNP antibody1:500 dilution). The sections were rinsed well with phosphate buffer and incubatedith superenhancer reagent for 30 min. After rinsing with phosphate buffer, incuba-

ion was done with supersensitive polymer-horseradish peroxidase immunohisto-hemistry detection system (Biogenex laboratories, San Ramon, CA, USA). Sectionsere washed with buffer and incubated with DAB solution for 5 min. Sections were

ounterstained with hematoxylin and observed under the light microscope.

.9. Statistical analysis

Values are expressed as means ± SD. Data within the groups were analyzed byne–way analysis of variance followed by Duncan’s multiple range test (DMRT). Aalue of P < 0.05 was considered statistically significant.

. Results

.1. Weight gain, liver weight and liver index.

Fig. 1 shows the body weights of animals during the experi-ental period. The final body weights (g) were CON, 31.3 ± 0.7;FFD, 36.7 ± 0.7; HFFD + ASX, 33.0 ± 0.8; CON + ASX, 30.9 ± 1.0. Theean final body weight of HFFD-fed mice was significantly higher

17%) than control, while the value was significantly lower (11%)n HFFD + ASX mice than the mice fed HFFD alone. No significantifference in final body weight was observed between controlnd CON + ASX groups. The food intake (g/day/mice) were CON,.4 ± 0.31; HFFD, 4.3 ± 0.35; HFFD + ASX, 4.3 ± 0.29 and CON + ASX,.5 ± 0.33. There was no significant difference in food intakeetween different groups of animals. Table 1 shows the liver weightnd liver index (LI; liver weight/body weight × 100) of experimen-al animals at the end of the study period. Both liver weight andiver index were significantly higher in HFFD-fed mice when com-ared to control. However, increased liver weight and liver indexbserved in HFFD-fed mice were significantly brought down to nearontrol levels when ASX was administered. The increase in liverndex suggests hepatomegaly in HFFD-fed mice.

.2. Glucose, insulin and insulin sensitivity indices

Plasma glucose and insulin levels were significantly higher inFFD-fed mice than in the control group (P < 0.01) on the 16th and

able 1iver weight and liver index (liver weight/body weight × 100) of experimental animals.

Parameters CON HFFD

Liver weight (g) 1.26 ± 0.06 1.76 ±Liver index (g) 4.02 ± 0.21 4.80 ±alues are means ± SD of 6 mice from each group. CON, control mice; HFFD, high fat-higith astaxanthin (6 mg/kg b.w./day); CON + ASX, control mice treated with astaxanthin (6a Significant as compared to CON (P < 0.05; ANOVA followed by DMRT).b Significant as compared to HFFD (P < 0.05; ANOVA followed by DMRT).

able 2evels of fasting glucose and insulin in plasma and HOMA and QUICKI values of experime

PARAMETERS CON HFFD

15th dayGlucose (mg/dl) 80.11 ± 4.12 115.81 ±Insulin (�U/ml) 16.54 ± 1.23 25.33 ±HOMA 3.23 ± 0.18 7.24 ±QUICKI 0.31 ± 0.002 0.26 ±

60th dayGlucose (mg/dl) 83.91 ± 4.52 134.81 ±Insulin (�U/ml) 17.42 ± 1.46 30.54 ±HOMA 3.60 ± 0.25 10.13 ±QUICKI 0.32 ± 0.002 0.27 ±

alues are means ± SD of 6 mice from each group. CON, control mice; HFFD, high fat-higith astaxanthin (6 mg/kg b.w./day); CON + ASX, control mice treated with astaxanthin (6a Significant as compared to CON (P < 0.05; ANOVA followed by DMRT).b Significant as compared to HFFD (P < 0.05; ANOVA followed by DMRT).

CON, control; HFFD, high fructose-high fat diet; HFFD + ASX, high fructose-high fat diet + astaxanthin (6 mg/kg b.w./day); CON + ASX, control + astaxanthin(6 mg/kg b.w./day). Values are mean ± SD of six experiments. *Significant when com-pared to control; #significant when compared to HFFD.

60th days (Table 2). The addition of ASX from the 16th day tillthe end of the study was able to return glucose and insulin lev-els to those obtained in control mice. We performed the insulinsensitivity index HOMA and QUICKI, which also showed signifi-cant alterations in HFFD-fed mice indicating the development ofIR on the 16th day itself. However, no statistical differences wereobserved in HOMA and QUICKI values among CON, CON + ASX andHFFD + ASX.

3.3. Aminotransferases, MPO and CYP2E1 activities

Table 3 shows the plasma levels of AST and ALT and hepatic

CYP2E1 and MPO activities observed in experimental animals. Theactivities of all these enzymes were significantly higher in HFFDalone-fed group (P < 0.05) whereas the activities were near-normalin ASX-supplemented groups even when HFFD feeding was contin-ued.

HFFD + ASX CON + ASX

0.13a 1.39 ± 0.08b 1.27 ± 0.070.36a 4.21 ± 0.27b 4.11 ± 0.25

h fructose diet fed mice; HFFD + ASX, high fat-high fructose diet fed mice treatedmg/kg b.w./day).

ntal animals at the 15th day and at the end of the study period.

HFFD + ASX CON + ASX

7.61a – –1.94a – –0.60a – –0.001a – –

8.98a 98.77 ± 4.74b 82.66 ± 4.742.63a 22.24 ± 1.76b 18.67 ± 0.960.90a 5.40 ± 0.36b 3.81 ± 0.350.001a 0.30 ± 0.002b 0.31 ± 0.003

h fructose diet fed mice; HFFD + ASX, high fat-high fructose diet fed mice treatedmg/kg b.w./day).

S. Bhuvaneswari et al. / Process Biochemistry 45 (2010) 1406–1414 1409

Table 3Activities of aspartatate aminotransferase (AST), alanine aminotransferase (ALT), CYP2E1 and MPO in experimental animals.

Parameters CON HFFD HFFD + ASX CON + ASX

AST (IU/l) 47.61 ± 2.42 93.28 ± 6.73a 66.62 ± 4.45b 48.62 ± 2.61ALT (IU/l) 51.20 ± 2.81 99.32 ± 7.62a 68.96 ± 3.98b 52.11 ± 3.43CYP2E1(Pmol min−1 mg protein−1) 3.77 ± 0.20 8.29 ± 0.57a 4.84 ± 0.26b 3.58 ± 0.18MPO (�mol min−1 mg protein−1) 5.43 ± 0.29 8.93 ± 0.61a 6.09 ± 0.36b 5.41 ± 0.32

Values are means ± SD of 6 mice from each group. CON, control mice; HFFD, high fat-high fructose diet fed mice; HFFD + ASX, high fat-high fructose diet fed mice treatedwith astaxanthin (6 mg/kg b.w./day); CON + ASX, control mice treated with astaxanthin (6 mg/kg b.w./day).

a Significant as compared to CON (P < 0.05; ANOVA followed by DMRT).b Significant as compared to HFFD (P < 0.05; ANOVA followed by DMRT).

Table 4Concentrations of cholesterol, triglycerides and free fatty acids in plasma (mmol/l)and liver (mg/g tissue) of experimental animals.

Parameters CON HFFD HFFD + ASX CON + ASX

CholesterolPlasma 2.50 ± 0.15 3.57 ± 0.21 2.80 ± 0.19 2.58 ± 0.14Liver 3.20 ± 0.17 5.83 ± 0.41a 3.98 ± 0.27b 3.07 ± 0.16

TriglyceridesPlasma 0.46 ± 0.03 0.71 ± 0.05a 0.50 ± 0.04b 0.53 ± 0.03Liver 5.48 ± 0.32 8.02 ± 0.56a 6.40 ± 0.39b 5.20 ± 0.33

Free fatty acidsPlasma 2.45 ± 0.15 3.69 ± 0.24a 2.89 ± 0.24b 2.53 ± 0.16Liver 6.47 ± 0.32 9.81 ± 0.88a 7.57 ± 0.39b 6.33 ± 0.34

Values are means ± SD of 6 mice from each group. CON, control mice; HFFD, highfat-high fructose diet fed mice; HFFD + ASX, high fat-high fructose diet fed micetreated with astaxanthin (6 mg/kg b.w./day); CON + ASX, control mice treated withastaxanthin (6 mg/kg b.w./day).

3

iFawCfcisflsg

Table 6Levels of nitrite and nitrosothiols in liver (�mol/mg protein) of experimentalanimals.

Parameters CON HFFD HFFD + ASX CON + ASX

Nitrite 13.97 ± 0.74 6.88 ± 0.49a 12.71 ± 0.79b 13.57 ± 0.73Nitrosothiol 31.95 ± 1.83 58.76 ± 4.12a 38.97 ± 2.35b 31.18 ± 1.60

Values are means ± SD of 6 mice from each group. CON, control mice; HFFD, highfat-high fructose diet fed mice; HFFD + ASX, high fat-high fructose diet fed micetreated with astaxanthin (6 mg/kg b.w./day); CON + ASX, control mice treated withastaxanthin (6 mg/kg b.w./day).

a Significant as compared to CON (P < 0.05; ANOVA followed by DMRT).

TL

Vw

a Significant as compared to CON (P < 0.05; ANOVA followed by DMRT).b Significant as compared to HFFD (P < 0.05; ANOVA followed by DMRT).

.4. Plasma and liver lipids

Dyslipidemia was noted in HFFD-fed mice. This included 54%ncrease in TG, 42% increase in cholesterol and 50% increase inFA compared to control. Similarly lipid analysis in the liverlso revealed excess storage. The total lipid levels (mg/g tissue)ere CON, 42.64 ± 2.4; HFFD, 51.70 ± 4.3; HFFD + ASX, 44.83 ± 2.8;ON + ASX, 41.76 ± 2.6. The total lipid content in the liver of HFFD-

ed mice was increased 1.2-fold when compared with mice fedontrol diet whereas in ASX + HFFD, the value did not differ signif-cantly from control. Levels of TG, total cholesterol and FFA were

ignificantly higher by 46%, 82% and 52%, respectively in HFFD-ed mice. ASX treatment of HFFD-fed mice restored plasma andiver lipid levels to near-normal values (Table 4). There were noignificant differences in lipid levels between CON and CON + ASXroups.

able 5evels of thiobarbituric acid reactive substances (TBARS), lipidhydroperoxides (LHP), and

Parameters CON H

PlasmaTBARS (�mol/dl) 0.92 ± 0.05 2LHP (nmol/dl) 0.90 ± 0.04 2PC (�mol/dl) 1.78 ± 0.14 2

LiverTBARS (nmol/mg protein) 1.26 ± 0.07 2LHP (nmol/mg protein) 1.36 ± 0.07 2PC (nmol/mg protein) 1.15 ± 0.06 2

alues are means ± SD of 6 mice from each group. CON, control mice; HFFD, high fat-higith astaxanthin (6 mg/kg b.w./day); CON + ASX, control mice treated with astaxanthin (6a Significant as compared to CON (P < 0.05; ANOVA followed by DMRT).b Significant as compared to HFFD (P < 0.05; ANOVA followed by DMRT).

b Significant as compared to HFFD (P < 0.05; ANOVA followed by DMRT).

3.5. Oxidative stress markers

The levels of oxidative stress markers such as LHP, TBARSand PC were measured to study the extent of oxidative dam-age and are presented in Table 5. The levels were significantlyincreased in HFFD group. ASX treatment of HFFD-fed mice con-trolled the levels of lipid peroxidation products and proteinoxidation products. The levels of nitrosative stress markers suchas total nitrite and nitrosothiol are tabulated in Table 6. Signifi-cantly higher levels of nitrosothiol (P < 0.05) and lower levels ofnitrite (P < 0.05) were observed in HFFD-fed mice. ASX admin-istration brought back the levels of nitrosothiol and nitrite tonear-normal.

The activities of enzymatic and non-enzymatic antioxidants inthe hemolysate and liver of animals in each group are shown inTables 7 and 8, respectively. The activities of SOD, CAT, GPX and GSTin HFFD-fed group were significantly decreased in circulation (by62%, 36%, 43% and 33%, respectively) and in the liver (by 26%, 50%,23% and 36%, respectively) as compared to control mice. The con-

centration of non-enzymatic antioxidants followed a similar trendboth in plasma and liver. Co-treatment with ASX prevented thedecline in antioxidant system in circulation and liver.

protein carbonyl (PC) in plasma and liver of experimental animals.

FFD HFFD + ASX CON + ASX

.19 ± 0.19a 1.33 ± 0.08b 0.96 ± 0.06

.16 ± 0.17a 1.22 ± 0.08b 0.92 ± 0.05

.25 ± 0.19a 1.84 ± 0.13b 1.69 ± 0.14

.22 ± 0.16a 1.61 ± 0.10b 1.28 ± 0.08

.72 ± 0.21a 1.75 ± 0.10b 1.39 ± 0.08

.66 ± 0.20a 1.75 ± 0.11b 1.18 ± 0.07

h fructose diet fed mice; HFFD + ASX, high fat-high fructose diet fed mice treatedmg/kg b.w./day).

1410 S. Bhuvaneswari et al. / Process Biochemistry 45 (2010) 1406–1414

Table 7Activities of enzymatic antioxidants in hemolysate and liver of experimental animals.

Parameters CON HFFD HFFD + ASX CON + ASX

HemolysateSOD (U/mg Hb) 3.98 ± 0.20 1.51 ± 0.13a 3.50 ± 0.20b 3.81 ± 0.30CAT (�mol of H2O2 consumed min−1 mgHb−1) 39.90 ± 2.02 25.31 ± 1.89a 35.51 ± 2.20b 38.42 ± 2.26GPX (�mol of GSH consumed min−1 mgHb−1) 8.10 ± 0.43 4.60 ± 0.37a 7.50 ± 0.46b 8.06 ± 0.55GST (�mol of CDNB conjugate consumed min−1 mgHb−1) 5.55 ± 0.30 2.71 ± 0.18a 4.78 ± 0.26b 5.35 ± 0.29

LiverSOD (U/mg protein) 4.94 ± 0.27 3.64 ± 0.28a 4.37 ± 0.29b 4.79 ± 0.26CAT (�mol of H2O2 consumed min−1 mg protein−1) 48.72 ± 2.75 24.44 ± 2.13a 41.44 ± 2.50b 47.43 ± 2.45GPX (�mol of GSH consumed min−1 mg protein−1) 6.01 ± 0.34 4.58 ± 0.38a 5.39 ± 0.35b 5.90 ± 0.34GST (�mol of CDNB conjugate formed min−1 mg protein−1) 5.39 ± 0.31 3.44 ± 0.29a 4.80 ± 0.27b 5.29 ± 0.34

Values are means ± SD of 6 mice from each group. CON, control mice; HFFD, high fat-high fructose diet fed mice; HFFD + ASX, high fat-high fructose diet fed mice treatedwith astaxanthin (6 mg/kg b.w./day); CON + ASX, control mice treated with astaxanthin (6 mg/kg b.w./day).

a Significant as compared to CON (P < 0.05; ANOVA followed by DMRT).b Significant as compared to HFFD (P < 0.05; ANOVA followed by DMRT).

Table 8Levels of non-enzymatic antioxidants in plasma and liver of experimental animals.

Parameters CON HFFD HFFD + ASX CON + ASX

PlasmaGSH (mg/dl) 18.64 ± 0.95 10.48 ± 0.83a 16.55 ± 1.00b 18.38 ± 1.07Vitamin C (mg/dl) 1.54 ± 0.09 0.96 ± 0.08a 1.33 ± 0.07b 1.45 ± 0.07Vitamin E (mg/dl) 1.33 ± 0.07 0.87 ± 0.06a 1.18 ± 0.08b 1.27 ± 0.07

LiverGSH (mg/100 g tissue) 86.58 ± 4.77 64.98 ± 5.73a 76.81 ± 4.73b 83.28 ± 4.82Vitamin C (�g/mg protein) 2.23 ± 0.12 1.15 ± 0.10a 1.98 ± 0.11b 2.19 ± 0.13Vitamin E (�g/mg protein) 4.13 ± 0.23 2.56 ± 0.20a 3.62 ± 0.24 b 3.98 ± 0.25

V fat-higw thin (6

3

ltis

FiiLqntc

alues are means ± SD of 6 mice from each group. CON, control mice; HFFD, highith astaxanthin (6 mg/kg b.w./day); CON + ASX, control mice treated with astaxana Significant as compared to CON (P < 0.05; ANOVA followed by DMRT).b Significant as compared to HFFD (P < 0.05; ANOVA followed by DMRT).

.6. Blotting analysis of TGF-ˇ1

Fig. 2(a) represents the protein expression of TGF-�1 in

iver while Fig. 2(b) represents the results of the densitome-ry analysis. Increased expression of TGF-�1 protein, observedn HFFD-fed mice, was reduced significantly when ASX wasupplemented. On the other hand, the protein expression was

ig. 2. Expression of transforming growth factor �1 (TGF-�1) in liver of exper-mental animals. (a) Expression of TGF-�1 in liver of experimental animals bymmunoblotting. Lane 1, CON; Lane 2, HFFD; Lane 3, HFFD + ASX (6 mg/kg b.w./day);ane 4, CON + ASX (6 mg/kg b.w./day). (b) Results of the densitometry analysis. Foruantitative assessment, the band intensity was measured in a densitometer andormalised with �-actin values. The changes relative to control are expressed inhe bar diagram. Values are means of 5 determinations. *Significant compared toontrol; #significant compared to HFFD.

h fructose diet fed mice; HFFD + ASX, high fat-high fructose diet fed mice treatedmg/kg b.w./day).

similar in ASX-untreated control mice and ASX-treated controlmice.

3.7. Histological changes

Figs. 3 and 4 show the histological analysis of liver by H and Eand MT stain, respectively. H and E staining revealed necrosis andmild inflammatory cell infiltration (active hepatitis) in HFFD-fedmice (Fig. 3B). Furthermore, liver of HFFD-mice shows microvesic-ular and macrovesicular fatty vacuoles. On the other hand, liverfrom mice administered ASX during HFFD feeding shows reduc-tion in inflammatory cell infiltration, fat accumulation and necrosis.MT staining of liver section from HFFD-fed mice revealed areasof necrosis involving all the zones and also deposition of colla-gen (Fig. 4B) compared to control (Fig. 4A). These abnormalitiesassociated with HFFD feeding were attenuated by ASX treatment.ASX-administered control mice showed normal collagen distribu-tion in liver as that of control (Fig. 4A and D).

3.8. Immunohistochemistry

Figs. 5 and 6 are the photomicrographs showing the localizationof DNP–protein adduct and TGF-�1 protein, respectively in liver.The intensity of DNP and TGF-�1 staining was more pronouncedin HFFD-fed mice than in control. It is noteworthy that there was amarked reduction in DNP–protein adduct and TGF-�1 immunore-activity in the liver of HFFD-fed mice treated with ASX. There wasno apparent difference in the immunoreactivity between CON andCON + ASX.

4. Discussion

The major findings of the study are the following: (1) ASXmarkedly decreased the weight gain and limited the hepatomegaly

S. Bhuvaneswari et al. / Process Biochemistry 45 (2010) 1406–1414 1411

F . (A)a ctiond ogy of

isl

pibsgp

Ftc

ig. 3. Photomicrograph of liver sections stained with hematoxylin and eosin 40×rrows show fat vacuoles of microvesicular and macrovesicular types. (C) Liver seeposition; microvesicular type of fatty changes is minimal. (D) Normal liver histol

nduced by the high calorie diet; whole body insulin sensitivity wasubstantially improved and (2) ASX ameliorated oxidative insult,iver injury and inflammation imposed by the high calorie diet.

The rise in bodyweight upon the consumption of HFFD has beenreviously reported [8] and is expected from the increased energy

ntake from the high calorie diet. The rise in liver weight mighte due to excess lipid/collagen accumulation. Confirming earliertudies [7,8], HFFD-fed mice developed significant elevation in bothlucose and insulin levels indicating the development of IR. Theresence of IR is further indicated by higher values of HOMA-IR and

ig. 4. Photomicrograph of liver sections stained with Masson trichrome 40×. (A) Normhe increased deposition of collagen around the focal areas. (C) Liver section of HFFD + Aollagen around central vein. (D) Normal liver mice treated with ASX (6 mg/kg b.w./day) s

Normal liver histology of control mice. (B) Liver section from HFFD-fed mice. Thefrom HFFD + ASX (6 mg/kg b.w./day) mice. The arrows represent reduction in lipidcontrol + ASX mice (6 mg/kg b.w./day).

lower values of QUICKI. Animals fed HFFD showed hepatic injuryresulting in the release of cellular enzymes such as AST and ALT into circulation.

Accumulation of lipids, increases in cholesterol, TG and FFA inliver of HFFD-fed mice were observed. Histological evaluation of the

liver also showed fatty changes of micro- and macrovesicular typesconsistent with biochemical measurements. Diet rich in fructoseinduces the lipogenic pathways by activating the gene expressionof sterol retinal binding protein-1, fatty acid synthase and acetylCoA carboxylase and by decreasing the activities of enzymes of fatty

al liver histology of control mice. (B) Liver section of HFFD-fed mice. Arrows showSX (6 mg/kg b.w./day). The arrows represent marked decrease in the deposition ofhows normal deposition of collagen around central vein.

1412 S. Bhuvaneswari et al. / Process Biochemistry 45 (2010) 1406–1414

F rimeni nd ASa

aTtc

io

Fms

ig. 5. Immunohistochemical localization of DNP–protein adducts in liver of expes reduced in ASX (6 mg/kg b.w./day) treated HFFD-fed mice (C). Both control (A) adducts.

cid oxidation resulting in VLDL-TG secretion causing a rise in liverG [7]. Furthermore, excess delivery of FFA from the insulin resis-

ant adipocytes to the liver could result in esterification of TG thatontributes to the development of liver steatosis.

Excess storage of lipids induces a state of lipotoxicity thats related to the formation of fatty acyl COA, diacylglycerol andther potentially toxic lipid metabolites which activate deleteri-

ig. 6. Representative photomicrograph for the accumulation of TGF-�1 protein in liver ofice (B) and this expression is markedly reduced in ASX (6 mg/kg b.w./day) treated HFFD

how minimal expression for this protein.

tal animals. DNP-immunostaining is increased in liver of HFFD-fed mice (B) andX (6 mg/kg b.w./day) treated control mice (D) show negligible expression of DNP

ous stress-sensitive pathways that may suppress the insulin signalcascade [34]. In addition, cholesterol or saturated fatty acid accu-

mulation can elicit stress in the endoplasmic reticulum and triggerthe unfolded protein response, an event that down regulates insulinsignaling [35].

The reversion of dyslipidemia and prevention of lipid accu-mulation in liver by ASX suggest that it might act to shift lipid

animals from each group. TGF-�1 immunostaining is increased in liver of HFFD-fed-fed mice (C). Both control (A) and ASX (6 mg/kg b.w./day) treated control (D) mice

Bioch

mso

ftobd[iir

oa[ostaft

gssacEa

cklorGnt

teaFsTmdi

dComcCiosmaie

i

S. Bhuvaneswari et al. / Process

etabolism from synthesis and storage to oxidation. Further, theuppression of lipotoxicity by ASX could mediate the improvementf insulin sensitivity by deactivating the stress-sensitive pathways.

Reducing energy intake and increasing energy expenditure areundamental to the treatment of obesity. It is important to note thathe food intake was similar in all the groups but HFFD-mice (withr without ASX) received more calories than control mice. ASX haseen shown to promote fatty acid metabolism in skeletal muscleuring exercise via carnitine palmitoyltransferase (CPT) I activation36]. Ikeuchi et al. [17,19] also noted that ASX does not affect foodntake but lowers body weight in obese mice possibly by increas-ng energy expenditure. This explains the mechanism of weighteduction and increased energy expenditure by ASX in HFFD-mice.

Oxidative lipid and protein modification in HFFD-fed mice couldccur through the generation of free radicals from glucose by mech-nisms like autooxidation, protein glycation and polyol pathway37]. The formation of S-nitrosothiol by the reaction between thi-ls especially GSH and peroxynitrite, a highly reactive nitrogenpecies causes depletion of GSH and promotes oxidative stress. Fur-her, carbonylation and nitrosylation of proteins alter the structurend function of cellular and membrane proteins and impair cellunction [38,39]. Piconi et al. [40] reported that protein modifica-ion/damage could reduce insulin activity.

Liver cytotoxicity and oxidative burden inflicted by HFFD is sug-estive of inefficient antioxidant system. The antioxidant enzymesuch as CAT and GPX are themselves susceptible to reactive oxygenpecies (ROS) attack [41]. The decreased SOD activity in HFFD-fednimals may be attributed to inactivation of the enzyme due to gly-ation of Lys122 and Lys128 residues [42]. Decreased GSH, vitaminsand C could be due to increased utilization to combat free radicalsnd/or decreased regeneration from their oxidized forms.

ASX is recognized to be a free radical scavenger as well as ahain breaking antioxidant [43]. The presence of hydroxyl andeto moieties on each ionone ring of ASX confer unique featuresike higher antioxidant activity and more polar configuration thanther carotenoids [44]. Transmembrane orientation of ASX is likelyesponsible for its antioxidant properties and its biological effects.oto et al. [45] reported that polar ring of ASX can scavenge the ROSear the membrane surface, while the polyene chain would inhibithe radical chain reaction into the membrane.

Glucose-induced ROS production results in the activation ofranscription factor NF-KB (nuclear factor kappa B) and subsequentxpression/production of cytokines which are documented to playn important role in downregulation of insulin signal cascade [46].urther, insulin receptor gene expression requires efficient tran-cription of certain factors [47] which are modulated by GSH levels.hus ASX, by its antioxidant activity and by replenishing GSH, pro-otes insulin sensitivity and alleviates oxidative stress and liver

amage. ASX has also been shown to prevent mitochondrial ROS-nduced modification of CPT I via its antioxidant effect [36].

CYP2E1, a member of cytochrome P450 enzymes, is a prooxi-ant enzyme and is the major microsomal source of superoxide.YP2E1 upregulation could be responsible for the developmentf liver injury by its ability to induce ROS production in HFFD-ice [48]. Since CYP2E1 is normally suppressed by insulin, it is

onceivable that reduced insulin action would result in increasedYP2E1 activities. In this context it is important to look at the stud-

es by Schattenberg et al. [49], which indicate the potential rolef C-Jun-N terminal kinase-dependent downregulation of insulinignaling cascade by CYP2E1 overexpression in a dietary mouseodel of steatohepatitis. ASX administration decreased CYP2E1

ctivity to near-normal, which is a strong evidence for the antiox-dant role of ASX and gives a possible clue to its insulin sensitivityffects.

MPO originates from activated neutrophils and monocytes ands an indicator for inflammatory cell infiltration and tissue injury.

emistry 45 (2010) 1406–1414 1413

Increased MPO activity suggests neutrophil migration and activa-tion and is likely to account for the increased parenchymal injuryand hepatitis in HFFD-fed mice. On the contrary, ASX producedinhibition of MPO activity that reflect the reduction in migrationof neutrophils and other inflammatory cells in the liver, therebyalleviating liver damage.

TGF-�1 is a profibrogenic cytokine secreted by the activatedstellate cells of the liver. Elevated glucose, ROS and lipid peroxi-dation products increase TGF-�1 causing cell death, inflammationand fibrosis [50]. Thus, the overexpression of TGF-�1 in HFFD-fed mice may arise from ROS generation and glucotoxicity whichis prevented by ASX treatment. There is a report showing thatASX protects mesangial cells in culture from glucose-induced ROSproduction, the activation of transcription factors and cytokineexpression [46]. Thus, we interpret that ASX acts by suppressingthe expression of TGF-�1, a stimulator of fibrogenic process. Reduc-tion in collagen in liver section of HFFD + ASX animals is affirmativewith its antifibrogenic role. Although, no studies have identifiedthe antifibrogenic effects of ASX, the anti-inflammatory action hasbeen increasingly recognized. For instance, ASX but not vitamin Ewas able found to reduce swelling of rat paw in an inflammatorymodel [51].

One limitation of the study is that the bioavailability of ASX hasnot been tested, but this has been reported by several investigators[21,52]. Rufer et al. [52] studied ASX concentration and distributionof isomeric forms in an human intervention study and observed thatASX is absorbed into plasma without any appreciable metabolictransformation. Animal studies also observed high bioavailabilityof ASX. ASX is shown to be metabolized in the liver and accumulatesin tissues (liver and kidney) of mice [21].

Human and animal studies have shown that western diet whichis high in fat and sugar is associated with obesity and other relatedmetabolic diseases [53]. The composition of HFFD mimics that of atypical unhealthy western diet [7]. We used HFFD to induce obesityin mice which serve as a good model to investigate the possible pro-tective role of ASX. ASX exerts anti-hyperglycemic effects in db/dbmice model as well [20,21]. ASX has been shown to preserve �-cell function, reduce glucose toxicity, prevent renal cell damageand delay the progression of diabetic nephropathy in db/db mice[21].

The present observations reflect that the high calorie diet, richin fat and fructose could be a driving force for the disease pro-cess in terms of oxidative stress, inflammation and fibrosis. Withthe addition of ASX, it is possible to prevent an early liver injuryand activation of the fibrogenic process. ASX is of interest to manyrecent researchers because of the anti-inflammatory and anti-atherogenic effects, which stem from its antioxidant role. Besidesantioxidant function, other effects specifically receptor activationor interaction with endogenous bioactive factors by ASX must alsobe considered in future. ASX is one of the ingredients in a numberof health products. In this vein, clinical studies have revealed thathealthy adults can consume 6 mg per day from H. pluvialis algalextract [54].

In summary, we have used a dietary model of fatty liver diseaseassociated with IR, which mimics the human NAFLD, to show thatASX restricts weight gain, promotes insulin sensitivity and curtailsliver disease. These changes were verified by examining the reduc-tion in CYP2E1 activity, oxidative stress, TGF-�1 expression andMPO activity. These abilities of ASX make it a potential drug forpreventing liver disease arising from IR/obesity.

Conflict of interest

The authors declare that there are no conflicts of interest.

1 Bioch

A

D

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

414 S. Bhuvaneswari et al. / Process

cknowledgment

The authors thank the University Grants Commission, Newelhi, India for providing financial support.

eferences

[1] Richard D. Energy expenditure: a critical determinant of energy balance withkey hypothalamic controls. Minerva Endocrinol 2007;32:173–83.

[2] Tessari P, Coracina A, Cosma A, Tiengo A. Hepatic lipid metabolism and non-alcoholic fatty liver disease. Nutr Metab Cardiovasc Dis 2009;19:291–302.

[3] Matteoni CA, Younossi ZM, Gramlich T, Boparai N, Liu YC, Mc-Cullough AJ. Non-alcoholic fatty liver disease: a spectrum of clinical and pathological severity.Gastroenterology 1999;116:1413–9.

[4] Wellen KE, Hotamisligil GS. Inflammation, stress, and diabetes. J Clin Invest2005;115:1111–9.

[5] Cheng KS, Tang HL, Chou FT, Chou JW, Hsu CH, Yu CJ, et al. Cytokine evalua-tion in liver cirrhosis and hepatocellular carcinoma. Hepatogastroenterology2009;56:1105–10.

[6] Breitkopf K, Haas S, Wiercinska E, Singer MV, Dooley S. Anti-TGF-betastrategies for the treatment of chronic liver disease. Alcohol Clin Exp Res2005;29:121S–31S.

[7] Aragno M, Tomasinelli CE, Vercellinatto I, Catalano MG, Collino M, Fantozzi R,et al. SREBP-1c in NAFDL induced by Western-type high-fat diet plus fructosein rats. Free Radic Biol Med 2009;47:1067–74.

[8] Messier C, Whately K, Liang J, Du L, Pussiant D. The effects of high-fat, highfructose and combination diet on learning, weight, and glucose regulation inC57BL/6 mice. Behav Brain Res 2007;178:139–45.

[9] Wei Y, Wang D, Topczewski F, Pagliassotti MJ. Fructose-mediated stress sig-naling in the liver: implications for hepatic insulin resistance. J Nutr Biochem2007;18:1–9.

10] Lorenz RT, Cysewski GR. Commercial potential for Haematococcus microalgaeas a natural source of astaxanthin. Trends Biotechnol 2000;18:160–7.

11] Ip P-F, Chen F. Employment of reactive oxygen species to enhance astaxan-thin formation in Chlorella zofingiensis in heterotrophic culture. Proc Biochem2005;40:3491–6.

12] Qin S, Liu G-X, Hu Z-Y. The accumulation and metabolism of astaxanthin inScenedesmus obliquus (Chlorophyceae). Proc Biochem 2008;43:795–802.

13] Malmsten C. Dietary supplementation with astaxanthin-rich algal mealimproves muscle endurance—a double blind study on male students. Gustavs-berg, Sweden: Karolinska Institute; 1998.

14] Guerin M, Huntley ME, Olaizola M. Haematococcus astaxanthin: applicationsfor human health and nutrition. Trends Biotechnol 2003;21:210–6.

15] Pashkow FJ, Watumull DG, Campbell CL. Astaxanthin: a novel potential treat-ment for oxidative stress and inflammation in cardiovascular disease. Am JCardiol 2008;101:58D–68D.

16] Hussein G, Nakamura M, Zhao Q, Iguchi T, Goto H, Sankawa U, et al. Antihy-pertensive and neuroprotective effects of astaxanthin in experimental animals.Biol Pharm Bull 2005;28:47–52.

17] Ikeuchi M, Koyama T, Takahashi J, Yazawa K. Effects of astaxanthin supplemen-tation on exercise-induced fatigue in mice. Biol Pharm Bull 2006;29:2106–10.

18] Zhang X, Pan L, Wei X, Gao H, Liu J. Impact of astaxanthin-enriched algal powderof Haematococcus pluvialis on memory improvement in BALB/c mice. EnvironGeochem Health 2007;29:483–9.

19] Ikeuchi M, Koyama T, Takahashi J, Yazawa K. Effects of astaxanthin in obesemice fed a high-fat diet. Biosci Biotechnol Biochem 2007;71:893–9.

20] Uchiyama K, Naito Y, Hasegawa G, Nakamura N, Takahashi J, Yoshikaw T. Astax-anthin protects �-cells against glucose toxicity in diabetic db/db mice. RedoxReport 2002;7(5):290–3.

21] Naito Y, Uchiyama K, Aoi W, Hasegawa G, Nakamura N, Yoshida N, et al. Preven-tion of diabetic nephropathy by treatment with astaxanthin in diabetic db/dbmice. BioFactors 2004;20:49–59.

22] Mathews DR, Hosker JP, Rudenkl AS, Naylor BA, Treacher DF, Turner RC.Homeostasis model assessment: insulin resistance and �-cell function fromfasting plasma glucose and insulin concentrations in man. Diabetologia1985;28:412–9.

23] Katz A, Nambi SS, Mather K, Baron AD, Follmann DA, Sullivan G, et al. Quanti-tative insulin sensitivity check index: a simple, accurate method for assessinginsulin sensitivity in humans. J Clin Endocrinol Metab 2000;85:2402–10.

24] Wu DF, Clejan L, Potter B, Cederbaum AI. Rapid decrease of cytochromeP-450IIE1 in primary hepatocyte culture and its maintenance by added 4-methylpyrazole. Hepatology 1990;12:1379–89.

25] Bradley PP, Priebat DA, Christensen RD, Rothstein G. Measurement of cuta-neous inflammation: estimation of neutrophil content with an enzyme marker.J Invest Dermatol 1982;78:206–9.

[

[

emistry 45 (2010) 1406–1414

26] Nieuhas WG, Samuelson S. Formation of malondialdehyde from phospho-lipids arachidonate during microsomal lipid peroxidation. Eur J Biochem1968;6:126–30.

27] Jiang ZY, Hunt JV, Wolf SP. Detection of lipid hydroperoxides using the FOXmethod. Anal Biochem 1992;202:384–9.

28] Levine RL, Garland D, Oliver CN, Amici A, Climent I, Lenz AG, et al. Determi-nation of carbonyl content in oxidatively modified proteins. Methods Enzymol1990;186:464–78.

29] Rock E, Astier C, Lab C, Malpuech C. Magnesium deficiency in rats induces a risein plasma nitric oxide. Magnes Res 1995;8:237–42.

30] Cook JA, Kim SY, Teague D. Convenient colorimetric and fluorometric assaysfor S-Nitrosothiols. Anal Biochem 1996;238:150–8.

31] AnithaNandhini AT, Balakrishnan SD, Anuradha CV. Taurine modulates antiox-idant potential and controls lipid peroxidation in the aorta of high fructose-fedrats. J Biochem Mol Biol Biophys 2002;6:129–33.

32] Folch J, Lees M, SloaneStanley GH. A simple method for the isolation and purifi-cation of total lipides from animal tissues. J Biol Chem 1957;226:497–509.

33] Lowry OH, Rosebrough N, Farr AL, Randall RJ. Protein measurement with theFolin phenol reagent. J Biol Chem 1951;193:265–75.

34] Unger RH. Lipotoxic diseases. Annu Rev Med 2002;53:319–36.35] Beltroy EP, Liu B, Dietschy JM, Turley SD. Lysosomal unesterified cholesterol

content correlates with liver cell death in murine Niemann-Pick type C disease.J Lipid Res 2007;48:869–81.

36] Aoi W, Naito Y, Takanami Y, Ishii T, Kawai Y, Akagiri S, et al. Astaxanthinimproves muscle lipid metabolism in exercise via inhibitory effect of oxidativeCPT I modification. Biochem Biophys Res Commun 2008;366:892–7.

37] Wolff SP, Dean RT. Glucose autoxidation and protein modification. The potentialrole of ‘autoxidative glycosylation’ in diabetes. Biochem J 1987;245:243–50.

38] Evans JL, Goldfine ID, Maddux BA, Grodsky GM. Oxidative stress and stress-activated signaling pathways. A unifying hypothesis of type 2 diabetes. EndocrRev 2009;23:599–622.

39] Grune T, Blasig IE, Sitte N, Roloff B, Haseloff R, Davies KJ. Peroxynitrite increasesthe degradation of aconitase and other cellular proteins by proteosome. J BiolChem 1998;273:10857–62.

40] Piconi L, Quagliaro L, Ceriello A. Oxidative stress in diabetes. Clin Chem Lab Med2003;41:1144–9.

41] Datta K, Sinha S, Chattopadhyay P. Reactive oxygen species in health and dis-eases. Natl Med J India 2000;13:304–10.

42] Arai K, Maguchi S, Fujii S, Ishibashi H, Oikawa K, Taniguchi N. Glyca-tion and inactivation of human Cu–Zn-superoxide dismutase. J Biol Chem1987;262:16969–72.

43] Miki W. Biological functions and activities of animal carotenoids. J Agric FoodChem 1991;63:141–6.

44] Kobayashi M. In vivo antioxidant role of astaxanthin under oxidativestress in the green algae Haematococcus pluvialis. Appl Microbiol Biotechnol2000;54:550–5.

45] Goto S, Kogure K, Abe K, Kimata Y, Kimtahama K, Yamashita E, et al. Effi-cient radical trapping at the surface and iniside the phospholipid membrane isresponsible for highly potent antiperoxidative activity of the carotenoid astax-anthin. Biochem Biophys Acta 2001;1512:251–8.

46] Manabe E, Handa O, Naito Y, Mizushima K, Akagiri S, Adachi S, et al. Astaxanthinprotects mesengial cells from hyperglycemia-induced oxidative signaling. J CellBiochem 2008;103:1925–37.

47] Araki E, Murakami T, Shirotoni T. A cluster of four SP1 binding sites requiredfor efficient expression of the human insulin receptor gene. J Biol Chem1991;6:3944–8.

48] Weltman MD, Farrell GC, Liddle C. Increased hepatocyte CYP2E1 expression in arat nutritional model of hepatic steatosis with inflammation. Gastroenterology1996;111:1645–53.

49] Schattenberg JM, Wang Y, Singh R, Rigoli RM, Czaja MJ. Hepatocyte CYP2E1overexpression and steatohepatitis lead to impaired hepatic insulin signaling.J Biol Chem 2005;280:9887–94.

50] Chen S, Jim B, Ziyadeh FN. Diabetic nephropathy and transforming growthfactor-beta: transforming our view of glomerulosclerosis mand fibrosis build-up. Semin Nephrol 2003;23:532–43.

51] Kurashige M, Okimasu E, Inoue M, Utsumi K. Inhibition of oxidative injuryof biological membranes by astaxanthin. Physiol Chem Phys Med NMR1990;22:27–38.

52] Rufer CE, Moeseneder J, Briviba K, Rechkemmer G, Bub A. bioavailability ofastaxanthin stereoisomers from wild (Oncorhynchus spp.) and aquacultured(Salmo salar) salmon in healthy men: a randomized, double blind study. Br J

Nutr 2008;99(5):1048–54.

53] Dirlewanger W, Schneiter P, JequierbE, Tappy L. Effects of fructose onhepatic glucose metabolism in humans. Am J Physiol Endocrinol Metab2000;279:907–11.

54] Fassett RG, Coombes JS. Astaxanthin, oxidative stress, inflammation and car-diovascular disease. Future Cardiol 2009;5:333–42.