www.elsevier.com/locate/ytaap

Toxicology and Applied Pharm

Implications of oxidative stress and hepatic cytokine (TNF-a and IL-6)

response in the pathogenesis of hepatic collagenesis in chronic

arsenic toxicity

Subhankar Das, Amal Santra, Sarbari Lahiri, D.N. Guha Mazumder*

Institute of Post Graduate Medical Education and Research, Kolkata, India

Received 23 June 2004; accepted 23 August 2004

Available online 30 November 2004

Abstract

Introduction: Noncirrhotic portal fibrosis has been reported to occur in humans due to prolonged intake of arsenic contaminated water.

Further, oxystress and hepatic fibrosis have been demonstrated by us in chronic arsenic induced hepatic damage in murine model. Cytokines

like tumor necrosis factor a (TNF-a) and interleukin 6 (IL-6) are suspected to play a role in hepatic collagenesis. The present study has been

carried out to find out whether increased oxystress and cytokine response are associated with increased accumulation of collagen in the liver

due to prolonged arsenic exposure and these follow a dose–response relationship.

Methods: Male BALB/c mice were given orally 200 Al of water containing arsenic in a dose of 50, 100, and 150 Ag/mouse/day for 6 days

a week (experimental group) or arsenic-free water (b0.01 Ag/l, control group) for 3, 6, 9 and 12 months. Hepatic glutathione (GSH), protein

sulfhydryl (PSH), glutathione peroxidase (GPx), Catalase, lipid peroxidation (LPx), protein carbonyl (PC), interleukin (IL-6), tumor necrosis

factor (TNF-a), arsenic and collagen content in the liver were estimated from sacrificed animals.

Results: Significant increase of lipid peroxidation and protein oxidation in the liver associated with depletion of hepatic thiols (GSH,

PSH), and antioxidant enzymes (GPx, Catalase) occurred in mice due to prolonged arsenic exposure in a dose-dependent manner. Significant

elevation of hepatic collagen occurred at 9 and 12 months in all the groups associated with significant elevation of TNF-a and IL-6.

However, arsenic level in the liver increased progressively from 3 months onwards. There was a positive correlation between the hepatic

arsenic level and collagen content (r = 0.8007), LPx (r = 0.779) and IL-6 (r = 0.7801). Further, there was a significant negative correlation

between GSH and TNF-a (r = �0.5336)) and LPx (r = �0.644).

Conclusion: Increasing dose and duration of arsenic exposure in mice cause progressive increase of oxystress and elevation of cytokines

associated with increasing level of collagen in the liver.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Arsenic; Hepatotoxicity; Proinflammatory cytokines; Oxystress

Arsenic is a known environmental toxin (Abernathy et

al., 1999). Reports of chronic arsenic toxicity due to

drinking of arsenic contaminated water have been reported

from many countries. An important feature in chronic

arsenic toxicity in West Bengal is a form of hepatic fibrosis

0041-008X/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.taap.2004.08.010

* Corresponding author. 37/C, Block B, New Alipur, Kolkata-700053,

India. Fax: +91 033 24751799.

E-mail address: dngm@apexmail.com (D.N. Guha Mazumder).

that causes portal hypertension but does not progress to

cirrhosis (Guha Mazumder et al., 1988, 1998). Franklin et

al. (1950) first reported periportal fibrosis that extended

intralobularly in one of the three cases of arsenic toxicity

due to chronic intake of arsenic in medicinal form.

Prolonged consumption of arsenic such as Fowler’s solution

has been known to cause similar hepatic lesion (Nevens et

al., 1990). Reports by Datta et al. (1979) and Guha

Mazumder et al. (1988) highlighted that non-cirrhotic portal

hypertension does occur in chronic arsenic toxicity due to

acology 204 (2005) 18–26

S. Das et al. / Toxicology and Applied Pharmacology 204 (2005) 18–26 19

prolonged drinking of arsenic contaminated water.

Although the association between arsenic toxicity and

hepatic fibrosis has been firmly established, the mechanism

of this type of liver damage remains unelucidated.

Oxidative damage is thought to underlie several chronic

liver diseases that are associated with fibrosis. It is thought

that activation of reactive oxygen species (ROS) is a

complex process that can result in peroxidative damage of

the major cellular components including amino acids,

carbohydrates, lipids, proteins, and nucleic acids (Sies and

Cadenas, 1985). Evidence of oxidative stress has been

detected in almost all the experimental conditions of arsenic

toxicity (Chang et al., 1991; Chouchane and Elizabeth,

2001; Lee et al., 1989). In our earlier investigation (Santra

et al., 2000), oxidative stress was supposed to be the

mainstay of arsenic induced hepatic fibrosis. In vivo, ROS

react with cellular lipids to produce a number of reactive

aldehydes. Malondialdehyde (MDA) and 4-hydroxynonenal

(HNE) are produced in alcoholic, viral, and cholestatic liver

injury (Kawamura et al., 2000; Niemela et al., 1999; Paradis

et al., 1997; Parola et al., 1996) and are supposed to

stimulate the synthesis of collagen mRNA (Lee et al., 1995;

Parola et al., 1996). Collagen production has also been

reported to be stimulated by MDA in vitro (Chojkier et al.,

1989). It has been shown that peroxidative damage of lipids

plays an important role in free radical-induced excess

collagen deposition in ethanol-intoxicated animals (Poli et

al., 1987). An involvement of lipid peroxidation has also

been proposed to explain the development of liver fibrosis

occurring as a consequence of excess deposition of iron

(Bacon et al., 1983) and copper (Sokol et al., 1990) in

experimental animal model.

Inflammatory cytokine activity is increased in many

forms of experimental and clinical liver injury including

alcoholic liver disease. In animal model, alcoholic liver

injury has been demonstrated to occur due to failure of

antioxidant defense system due to increased oxystress

(Zindenberg et al., 1991). Monocytes and Kuppfer cells

produce cytokines such as TNF-a, IL-6, IL-8 in response

to stimuli such as endotoxin (Pena et al., 1999).

Elevation of serum IL-6 and TNF-a level is always

associated with alcoholic liver disease (Bird et al., 1990;

Hong et al., 2002; McClain and Cohen, 1989), but the

significance of such elevation is not clear. It has been

suggested that TNF-a enhances the production of type I

collagen in fat storing cells (FSC) by a post-transcriptional

mechanism (Weiner et al., 1990). Further, myofibroblasts

are known to produce many cytokines including IL-6

(Greenwel et al., 1991) which inhibit proliferation of rat

FSC (Matsuoka et al., 1989). The current study is carried

out to find out whether hepatic collagenesis increases with

increased arsenic exposure for prolonged period in a

dose-dependent manner and whether cytokines are

involved in collagen deposition in the liver in association

with membrane damage caused by arsenic induced

oxystress.

Materials and methods

Seven- to 8-week-old male BALB/c mice weighing 20–

22 gm were taken for the study and were housed under 12 h

dark and light cycles at laboratory environment and were

given laboratory animal chow and water ad libitum. All

procedures were performed in accordance with the guide-

lines of the Institution’s animal care committee.

Experimental protocol. Solutions of arsenic were prepared

in three different concentrations by dissolving requisite

amount of sodium arsenite (NaAsO2, E.Merck, Germany)

in distilled water in such a way that 200 Al of each solution

contains 50, 100, and 150 Ag of arsenic, respectively. Arsenicsolution (200 Al) in three different concentration (50–150 Ag)was fed to the experimental animals by a feeding needle 6

days a week for 12 months. Mice fed with arsenic-free (b0.01

ppm) water served as control. Batches of animals were

sacrificed at the end of 3, 6, 9, and 12 months. Liver obtained

from the sacrificed animals was cut into pieces and kept in

aliquot under �80 8C until use. The experimental group

which received 50 Ag of As per day was designated as GroupI, while those receiving 100 Ag and 150 Ag of As per day weredesignated as Group II and Group III, respectively.

Estimation of hepatic thiol. A portion of liver tissue was

immediately transferred into a 5% sulfosalicylic acid (SSA)

solution for the assay of hepatic GSH according to the

method of Ball (1966). Briefly, liver tissue was homogen-

ized into ice-cold 5% SSA and centrifuged at 5000 rpm for

10 min. The supernatant was used to react with dithioni-

trobenzoic acid (DTNB) in 0.5 M phosphate buffer pH 6.8

and the absorbance was read at 412 nm. Standard curve was

prepared using known GSH dilutions.

The hepatic protein sulfhydryl (PSH) was measured in

the pellet, which was obtained after precipitating the liver

homogenate with SSA. The pellet was resuspended in 6 M

guanidine pH 6.0, and the sample was read by a

spectrophotometer at 412 and 530 nm before and after 30

min of incubation with 10 mM DTNB solution according to

the method of Grattagliano et al. (1996).

Estimation of protein oxidation. The hepatic protein

oxidation in terms of formation of protein carbonyl (PC)

content was evaluated by incubating liver homogenate with

2% dinitrophenylhydrazine (DNPH) in 2 N HCl. The

proteins were subsequently precipitated with 50% TCA

and were washed three times each with 1:1 (w/v) ethanol/

ethylacetate and 10% TCA. The final precipitate was solved

in 6 M guanidine and the spectrum of the DNPH versus HCl

controls was followed at 350 to 375 nm according to the

method of Levine et al. (1990) using the molar extinction

coefficient of 21.5 nmol/l�1 cm�1 for aliphatic hydrazones.

Estimation of hepatic catalase and glutathione peroxidase

(GSH-Px). A portion of the liver was minced and a 10%

S. Das et al. / Toxicology and Applied Pharmacology 204 (2005) 18–2620

homogenate was prepared in ice-cold potassium phosphate

buffer (10 mM, pH 7.4) containing 0.15 M KCl using a

glass homogenizer fitted with a Teflon pestle. The homo-

genate was centrifuged at 105000 g for 60 min to get

cytosolic fraction. Catalase activity was determined by the

Beers and Sizer assay (1952) in which the disappearance of

H2O2 is followed spectrophotometrically at 240 nm. The

decrease in absorbance for 2–3 min was recorded in

cytosolic fraction with 0.059 M H2O2 in 0.05 M potassium

phosphate buffer pH 7.

The method used to determine GSH-Px activity was

according to the method of Paglia and Valentine (1967). The

cytosol, containing the enzyme source was mixed with 0.05

M sodium phosphate buffer, sodium azide, 0.15 M GSH,

glutathione reductase (GR) and 0.89 mM NADPH. The

reaction was initiated by addition of 0.0022 M H2O2. The

change in absorbance was recorded at 1-min intervals at 340

nm and the specific activity was calculated using an

extinction coefficient of 6.22 cm/Amol for NADPH.

Estimation of hepatic lipid peroxidation (LPx). Hepatic lipid

peroxidation was assayed by determining the production of

thiobarbituric acid-reactive substance (Utley et al., 1967).

Liver homogenate containing 10% trichloroacetic acid

(TCA) and 0.67% thiobarbituric acid (TBA) were heated

for 45 min in a boiling water bath, cooled and centrifuged at

4000 rpm for 10 min. The absorbance of the supernatant

fraction was determined at a wavelength of 535 nm. An

extinction coefficient of 1.56 � 10�1 M�1 cm�1 was used to

calculate the concentration of malondialdedhyde (MDA).

Estimation of tissue collagen content. A portion of the liver

was lyophilized, crashed to powder and defatted under 1:1

(vol/vol) methanol/chloroform solution and pelleted in a

speed vacuum. The defatted dried matter was hydrolyzed in

6(N) HCl for 18–20 h at 110 8C. Aliquots of the

hydrolysates were used to measure hydroxyproline content

spectrophotometrically by reacting with Ehrlic’s reagent

according to the method of Woessner (1961).

Estimation of cytokines. Tissue levels of TNF-a and IL-6

were quantified using an enzyme linked immunosorbent

assay (ELISA) kit. A piece of liver (kept at �80 8C) washomogenized in extraction buffer (Tris 50 mmol/l pH 7.2,

NaCl 150 mmol/l, Triton-X 100 1%) plus 100 Al of proteaseinhibitor cocktail solution (Complete Mini; Roche, Basel,

Switzerland). The homogenate was shaken on ice for 90 min

and then centrifuged at 3000 � g at 4 8C for 15 min

(Rudiger and Clavien, 2002). The supernatants were then

analyzed for TNF-a and IL-6 by using a commercial ELISA

kit (Quantakine mTNF-a, Quantakine mIL-6; R&D Sys-

tems) according to the manufacturer’s guidelines.

Estimation of hepatic arsenic (As) content. The arsenic

content of liver tissue of control and experimental animals

obtained after 3, 6, 9, and 12 months was estimated as

described by Atallah and Kalman (1991). Briefly, liver

tissue was digested with 1 ml of 20% NaOH and 650 mg of

sodium peroxodisulfate. The arsenic content was then

estimated from the digested material with atomic absorption

spectrometer (FIHGAAS, model no. AA 100, Perkin

Elmer).

Estimation of protein. The protein content of the liver

homogenate and the subcellular fractions were determined

by the method of Lowry et al. (1951).

Statistical analysis. Statistical analysis was performed by

using Student’s t test. Data are expressed as mean F SD.

Correlation and linear regression coefficients were calcu-

lated and drawn by using SPSS (VERSION 10.1) and

Microsoft excel 2000. In all instances, P b 0.05 was

considered as the minimum level of significance.

Results

Effect of chronic arsenic exposure on hepatic thiol

Continued feeding of arsenic contaminated water to

animals of all the experimental groups caused significant

elevation of hepatic GSH level at the end of 3 months.

However, the values reached to normal level at 6 months of

arsenic feeding. A significant depletion of hepatic GSH was

found in all the experimental animals at the end of 9 and 12

months (P b 0. 01 and 0.001, Table 1).

There was no change of hepatic protein thiol (PSH) in the

three experimental group of mice at the end of 3 months. A

significant reduction of its value was observed at 6 months

in group II (P b 0.02) and group III (P b 0.01) animals.

Feeding of arsenic for 9 and 12 months led to a significant

progressive decrease (P b 0. 01 and 0.001) of PSH content

in the liver of experimental animals of all the groups.

Effect on hepatic glutathione peroxidase (GSH-Px) and

hepatic catalase activity

Values of hepatic GSH-Px and catalase activity in both

control and experimental animals are given in Table 1.

Feeding of arsenic for 3 months caused significant elevation

of both the enzymes in all three experimental groups of

mice. However, further feeding of As caused significant

reduction in the level of both the enzymes in group II and III

animals. A gradual dose-dependent depletion of these two

enzymes was found in all the experimental animals at 9 and

12 months.

Effect on hepatic lipid peroxidation

Data on lipid peroxidation (LPx) in terms of MDA

formation of whole liver homogenate of control and

experimental animals are presented in Table 1. A significant

Table 1

Results of various components of antioxidant defense system during increasing dose and duration of arsenic (As) feeding in mice

GSH

(Ag/mg tissue)

LPx

(nm MDA/mg pr.)

PSH (nmol/mg pr) PC (nmol/mg pr) GPx (Am NADPH

oxidized/min/mg pr.)

Catalase (Am H2O2

reduced/min/mg pr.)

3 Mon Cont 9.38 F 1.20 1.40 F 0.41 11.27 F 1.15 10.25 F 1.70 8.23 F 0.48 6.72 F 0.21

50 Ag 10.69 F 1.15a 1.57 F 0.50 11.13 F 0.85 11.15 F 2.30 9.19 F 0.67y 7.14 F 0.40y

100 Ag 10.98 F 1.58h 1.83 F 0.86 10.85 F 0.99 11.85 F 2.20 9.45 F 0.80E 7.26 F 0.35E

150 Ag 11.75 F 1.12E 1.87 F 0.68 10.17 F 1.27 12.35 F 1.56 10.24 F 0.94E 7.40 F 0.34E

6 Mon Cont 9.66 F 0.78 1.60 F 0.47 10.79 F 1.45 10.56 F 1.60 8.28 F 0.64 6.75 F 0.42

50 Ag 8.97 F 1.01 2.06 F 0.82 9.56 F 1.30 11.40 F 2.10 7.86 F 0.89 6.42 F 0.68

100 Ag 8.81 F 1.27 2.22 F 0.63a 8.78 F 0.85h 12.30 F 1.90a 7.54 F 0.68a 6.34 F 0.33a

150 Ag 8.46 F 1.78 2.31 F 0.83a 8.06 F 0.76y 13.70 F 2.50y 7.42 F 0.82a 6.24 F 0.46h

9 Mon Cont 9.75 F 0.89 1.40 F 0.33 11.03 F 0.74 11.12 F 2.00 8.20 F 0.29 6.63 F 0.41

50 Ag 7.76 F 1.75y 2.76 F 1.25y 8.81 F 0.96y 18.24 F 3.10E 7.77 F 0.56a 6.29 F 0.25a

100 Ag 7.55 F 1.62y 2.89 F 0.81E 8.27 F 1.23E 19.84 F 2.70E 7.49 F 0.66y 6.18 F 0.32h

150 Ag 6.47 F 0.96E 3.02 F 0.51E 7.86 F 0.91E 21.56 F 3.60E 7.12 F 0.96y 6.08 F 0.30y

12 Mon Cont 9.55 F 0.70 1.48 F 0.32 10.65 F 0.96 11.76 F 1.80 8.03 F 0.38 6.62 F 0.43

50 Ag 7.26 F 0.77E 3.25 F 0.81E 7.76 F 1.11E 24.54 F 2.45E 6.95 F 0.45E 6.15 F 0.26y

100 Ag 6.85 F 1.15E 4.48 F 0.64E 6.73 F 0.78E 34.24 F 4.20E 6.76 F 0.74E 6.08 F 0.23y

150 Ag 6.06 F 0.74E 5.87 F 0.98E 6.01 F 0.82E 40.70 F 3.93E 6.05 F 1.01E 5.84 F 0.37E

Results are mean F SD. a = P b 0.05, h = P b 0.02, y = P b 0.01, E = P b 0.001, sample size (n) = 10 in each group.

S. Das et al. / Toxicology and Applied Pharmacology 204 (2005) 18–26 21

increase (P b 0.05) of hepatic MDA level was seen in

groups II and III animals after 6 months of arsenic feeding.

Further increase of hepatic MDA formation was noted in all

the experimental animals at 9 and 12 months. The increase

was found to be highly significant (P b 0. 01 and 0.001)

and was dose- and duration-dependent.

Glutathione (GSH) vs. lipid peroxidation (LPx)

Fig. 1 shows the correlation of data of hepatic glutathione

and lipid peroxidation of all the control and experimental

animals. An inverse correlation (Fig. 1, r = �0.644, P b

0.001) was found between hepatic GSH and LPx.

Effect on hepatic protein carbonyl (PC)

Data on protein oxidation in the liver in the form of

protein carbonyl (PC) of control and experimental animals

are given in Table 1. A significant increase of hepatic PC

content was seen in group II (12.30 F 1.9 nmol/mg protein,

P b 0.05) and group III (13.70 F 2.5 nmol/mg protein, (P b

0.01) animals at 6 months of arsenic feeding. Further

feeding of arsenic showed a significant (P b 0. 01 and

Fig. 1. Data showing inverse correlation between GSH and LPx in the liver

of control and experimental animals (r = �0.644, P b 0.001).

0.001) elevation of hepatic PC in all the experimental

animals in a dose-dependent manner. The rate of increase

was 64.02% in group I, 78.4% in group II, and 93.88% in

group III at the end of 9 months. A 108–246% increase of

hepatic protein carbonyl content was observed at 12 months

of exposure.

Effect on hepatic collagen content

Fig. 2 shows the hepatic collagen content of both the

control and experimental animals exposed to different doses

of arsenic. Feeding of arsenic for 9 and 12 months caused a

significant (P b 0. 01 and 0.001) and progressive increase

of hepatic collagen level in all the experimental animals in a

dose-dependent manner.

Correlation of hepatic collagen content with lipid

peroxidation (LPx) and protein oxidation

Fig. 3 shows correlation of data of hepatic collagen

content and lipid peroxidation in the liver of all the control

Fig. 2. Concentration of total collagen in liver of control and experimental

animals. Results are mean F SD. a = P b 0.05, E = P b 0.001, sample size

(n) = 10 in each group.

Fig. 3. Data showing positive correlation between hepatic collagen

content and MDA level of control and experimental animals (r = 0.7177,

P b 0.001).

Fig. 5. Hepatic arsenic (As) content of the control and experimental animals

at different months of As feeding. Results are mean F SD. k = P b 0.001,

Sample size (n) = 10 in each group.

S. Das et al. / Toxicology and Applied Pharmacology 204 (2005) 18–2622

and experimental animals. A positive correlation (r =

0.7177, P b 0.001) was found between hepatic MDA level

and collagen content in the study. Similarly, a positive

correlation (Fig. 4, r = 0.8149, P b 0.001) was also found

between the hepatic protein oxidation (in terms of protein

carbonyl formation) and hepatic collagen content.

Effect on hepatic arsenic

Arsenic levels in the liver of the experimental animals

with various dose and duration of arsenic exposure are

shown in Fig. 5. A 2- to 4-fold increase in hepatic arsenic

content was observed at 3 and 6 months of arsenic feeding,

while a 7- to 14 -fold and 15- to 20-fold increase were seen

at 9 and 12 months, respectively. The elevation of hepatic

arsenic level in experimental animals was found to be dose-

and duration-dependent.

Correlation of hepatic arsenic content with lipid

peroxidation and hepatic collagen level

Fig. 6 shows a positive correlation (r = 0.7790, P b

0.001) between hepatic arsenic level and hepatic MDA

value. This implied that increased hepatic arsenic level was

associated with increase of hepatic membrane damage. The

increased hepatic arsenic level was also found to be

associated with increased hepatic collagenesis, as was

evident from a significant positive correlation (r = 0.8007,

Fig. 4. Data showing positive correlation between the levels of protein

oxidation and collagen content in the liver of control and experimental

animals (r = 0.8149, P b 0.001).

P b 0.001 Fig. 7) between hepatic arsenic and hepatic

collagen content.

Effect on hepatic TNF-a and hepatic IL-6

There was no significant difference of values of TNF-a

(Fig. 8) and IL-6 (Fig. 9) in the liver tissue of experimental

animals compared to control following 3 and 6 months of

arsenic exposure. However, there was significant progres-

sive increment of their values compared to control following

9 and 12 months of arsenic feeding and these followed a

dose-dependent relationship.

Glutathione (GSH) vs. tumor necrosis factor (TNF-a)

A significant inverse correlation (r = � 0. 0.5336, P b

0.001 between hepatic GSH and hepatic TNF a content was

found when all the data of TNF-a level in the liver were

correlated with the respective glutathione level (Fig. 10).

Hepatic IL-6 vs. hepatic collagen

A significant positive correlation (r = 0. 0.7801, P b

0.001) between hepatic IL-6 and hepatic collagen content

Fig. 6. Data showing positive correlation between hepatic arsenic level

and lipid peroxidation of control and experimental animals (r = 0.779,

P b 0.001).

Fig. 7. Data showing positive correlation between the concentration of

arsenic and collagen in the liver of control and experimental animals (r =

0.8007, P b 0.001).

Fig. 9. Concentration of IL-6 in the liver of control and experimental

animals. Results are mean F SD. y = P b 0.01, k = P b 0.001, sample size

(n) = 10 in each group.

S. Das et al. / Toxicology and Applied Pharmacology 204 (2005) 18–26 23

was observed in arsenic fed experimental animals (Fig. 11)

when all the respective data were plotted.

Discussion

In the course of metabolism of xenobiotic compounds

causing chronic liver diseases, reactive oxygen species

(ROS) are frequently produced in the liver (Vessey, 2002).

Arsenic has been shown to induce ROS production

(Yamanaka et al., 1989, 1990), which can lead to hepatic

fibrosis when the exposure is chronic (Santra et al., 2000).

Under conditions of oxidative stress, accumulation of large

quantities of free radicals in living system is known to be

detrimental to the viability of cells. In our present study,

significant elevation of hepatic GSH was observed in mice

exposed to As (50–150 Ag) intoxication for a period of 3

months, which may be a form of adaptation on the part of

the system to counteract the oxidative stress. The protective

effect of GSH against the toxic effects of As (III), As (V),

MMA & TMA has been confirmed by Ochi et al. (1994).

Arsenic toxicity is primarily due to the binding of As3+ to

sulfhydryl containing enzymes (Leonard and Lauwerys,

1980; Wang and Rossman, 1996). Reduced glutathione

(GSH) is one of the most abundant cellular thiols and

Fig. 8. Concentration of TNF-a in the liver of control and experimental

animals during different months of arsenic exposure. Results are mean FSD. a = P b 0.05, y = P b 0.01, E = P b 0.001, sample size (n) = 10 in each

group.

potentially a target for arsenic. Numerous toxic or poten-

tially toxic compounds, including some metals, are either

taken up or removed from the cell by GSH-mediated

pathway (Ballatori, 1994). Important protective role of GSH

against arsenic induced oxidative damage has been demon-

strated by Ito et al. (1998). Moreover, in case of inorganic

arsenic, a major mechanism postulated for its detoxification

by mammals and many other organisms involves GSH-

dependent methylation and protein binding (Aposhian,

1997). Furthermore, it has been shown that GSH partici-

pates in the biliary excretion of arsenic (Gyurasics et al.,

1991). However, by 6 month, the level of GSH came down

to base level and was followed by its progressive depletion

at 9 and 12 months. Reduction of GSH level in all the three

experimental groups during such prolonged exposure

appears to be due to its increased utilization compared to

its synthesis.

The liver possesses an antioxidant defense system which

is mediated by several enzymes functioning in a concerted

manner by removing peroxide and superoxide anion

generated within the cell. Glutathione peroxidase (GSH-

Px) and catalase are the major enzymes that remove

hydrogen peroxide generated by superoxide dismutase in

cytosol and mitochondria (Chance et al., 1979) by

oxidizing the GSH to GSSG (Halliwell, 1994). The sharp

rise of GSH-Px and catalase activity at 3 months of arsenic

Fig. 10. Data showing inverse correlation between GSH and TNF-a level in

the liver of control and experimental animals (r = �0.5336, P b 0.001).

Fig. 11. Data showing positive correlation between IL-6 and collagen

content in the liver of control and experimental animals (r = 0.7801,

P b 0.001).

S. Das et al. / Toxicology and Applied Pharmacology 204 (2005) 18–2624

feeding might be due to adaptation on part of the system to

counteract the oxidative stress. Further feeding of arsenic

contaminated water caused a significant progressive

decrease of GSH-Px and catalase activity suggesting failure

of adaptive mechanism towards chronic arsenic insult. The

reductions of those enzymes were found to be dose-

dependent. Similar reduction of these enzymes during

prolonged arsenic exposure has been reported earlier

(Santra et al., 2000).

Free radicals are also implicated to alter the intracellular

thiol homeostasis during the metabolism of various xeno-

biotics (Chai et al., 1994; DiMonte et al., 1984). A

significant loss of hepatic protein sulfhydryl in groups II

and III experimental animals at 6 month was found. Further,

a significant depletion of hepatic PSH content in all the

experimental animals was seen at 9 and 12 months. This

depletion of protein sulfhydryl possibly means its increased

utilization towards GSH synthesis. GSH depletion has been

shown to affect PSH pool (Vendemiale et al., 1998), which

is necessary for the preservation of protein function during

oxidative stress (Guerrieri et al., 1999; Starke et al., 1997).

A positive correlation between depletion of protein sulf-

hydryl groups and loss of cell viability has been observed by

DiMonte et al. (1984). Loss of protein thiols preceded cell

death, and occurred more rapidly in cells with decreased

levels of GSH.

The peroxidation (LPx) of polyunsaturated fatty acids

present in the membrane lipids is considered as an index to

monitor the function of membrane integrity (Baird et al.,

1980). Lipid peroxidation (LPx) can be demonstrated in

association with cellular injury from a wide variety of toxins.

It provides a logical link between free radical initiated

processes and resultant membrane dysfunction. Continued

feeding of arsenic contaminated water for 6 months and

onwards caused decreased intracellular concentration of

GSH and GSH-Px/Catalase level associated with significant

lipid peroxidation. The protective role of GSH against cell

damage is confirmed by the significant inverse correlation

(Fig. 1, r = �0.644, P b 0.001) between hepatic GSH and

MDA level. Oxidative damage and LPx of membrane lipids

have been reported to occur within a wide variety of chronic

liver diseases induced by different metals like iron (Bacon

and Britton, 1990), copper (Sternleib, 1980) in experimental

animals. Protein damage from oxidative stress may occur

either directly or as a result of lipid peroxidation (Spector,

1984). In our present study, exposure of arsenic in all the

experimental animals for 9 and 12 months caused significant

increase in protein carbonyl suggesting failure of antioxidant

defense system to prevent protein oxidation.

Following arsenic exposure, it was found to be deposited

in most organs of the body, for example, skin, hair, thyroid,

squamous epithelium, skeleton, lens of the eye, liver, lungs

and kidney (Lindgren et al., 1982; Marafante and Vahter,

1984, 1987). In our present study, hepatic arsenic concen-

tration was found to be significantly increased in all the

experimental groups compared to control animals at 3

months onwards and the increase was found to be dose- and

duration-dependent (Fig. 5).

Clinical evidences of portal fibrosis with or without portal

hypertension due to prolonged drinking of arsenic contami-

nated water was reported earlier by Datta et al. (1979) and

Guha Mazumder et al. (1988, 1998). Histological evidence

of liver fibrosis was also reported in experimental animal

after 15 months of feeding arsenic contaminated water

collected from an arsenic endemic area in West Bengal.

Evidence of oxystress and membrane damage due to lipid

peroxidation was reported in that study (Santra et al., 2000).

Electron microscopic study of rats fed with higher doses of

arsenic trioxide for 4 months showed evidence of large

bundles of collagen fibers around proliferating bile ducts and

spotty necrosed areas of liver (Ishinishi et al., 1980). A

significant increase in the hepatic collagen and 4-hydrox-

yproline levels were also reported in experimental mice after

feeding arsenic in high doses (120 to 360 ppm, Sarin et al.,

1999). In our present study, a significant increase in hepatic

collagen level was seen in all three experimental groups at 9

and 12 months (Fig. 2). The increased hepatic collagen

content was associated with increased level of hepatic MDA

and hepatic protein carbonyl content as seen by the positive

correlation ship (Fig. 3, r = 0.7177, P b 0.001 and Fig. 4, r =

0.8149, P b 0.001). Peroxidative damage of the major

cellular components including proteins and lipids are

supposed to occur in oxidative stress (Sies and Cadenas,

1985). A significant positive correlation (r = 0.7790, P b

0.001 Fig. 6) between hepatic arsenic level and hepatic MDA

level was also seen in our study. We have also found a

significant positive correlation (r = 0.8007, P b 0.001 Fig. 7)

between hepatic arsenic and hepatic collagen content. It is

possible that oxystress-induced protein oxidation and lipid

peroxidation might have influenced collagen synthesis due

to accumulation of arsenic following prolonged exposure.

Niemela et al. (1995) and Pietorangelo et al. (1994) showed

that lipid peroxidation products in liver parenchymal cells

are needed to initiate the fibrogenic process.

Of the various cytokines involved in collagenesis, TNF-a

and IL-6 are the important ones. A progressive and

significant increased levels of hepatic TNF-a were found

S. Das et al. / Toxicology and Applied Pharmacology 204 (2005) 18–26 25

in all the experimental animals at 9 and 12 months of arsenic

feeding (Fig. 8). The elevation was also found to be

associated with increased level of hepatic collagen during

that time. TNF-a plays an important role in activation of fat-

storing cells (FSC) (Bachem et al., 1993), the predominant

cells involved in the production of type 1 collagen (Krull

and Gressner, 1992). In our present study, an inverse

correlation (r = �0. 5336, P b 0.001 Fig. 10) was observed

between hepatic GSH and hepatic TNF-a. There is some

report that decrease of hepatic GSH could cause withdrawal

of the effect of glycine-gated chloride channel and increases

synthesis of TNF-a (Ikejima et al., 1997).

The present study showed increased level of hepatic IL-6

in all the experimental animals at 9 and 12 months of arsenic

exposure (Fig. 9). A significant positive correlation (r =

0.7801, P b 0.001) between hepatic IL-6 and hepatic

collagen content (Fig. 11) was found in our study—

indicating a possible role of IL-6 in hepatic collagenesis

during arsenic-induced toxicity.

The present study highlights that increased accumulation

of collagen occurs following prolonged arsenic exposure in

a dose-dependent manner and is associated with high

arsenic level in the liver. Cytokines like TNF-a and IL-6

are probably associated with collagenesis and are found to

be associated with protein oxidation and membrane damage

caused by oxystress which has been counteracted initially

by antioxidant defence mechanisms but later failed.

Acknowledgments

The authors acknowledge the help of the Director,

Institute of Post Graduate Medical Education and Research,

Kolkata, India and staff members of the Department of

Gastroenterology in conducting the study.

References

Abernathy, C.O., Liu, Y.P., Longfellow, D., Aposhian, H.V., Beck, B.,

Fowler, B., Goyer, R., Menzer, R., Rossman, T., Thompson, C.,

Waalkes, M., 1999 (July). Arsenic: health effects, mechanisms of

actions, and research issues. Environ. Health Perspect. 107 (7),

593–597.

Aposhian, H.V., 1997. Enzymatic methylation of arsenic species and other

new approaches to arsenic toxicity. Annu. Rev. Pharmacol. Toxicol. 37,

397–419.

Atallah, R., Kalman, D.A., 1991. Online photooxidation for the determi-

nation of organoarsenic compounds by AAS with continuous arsine

generation. Talanta 38, 167–173.

Bachem, M.G., Shell, K.M., Melchior, R., Kroof, J., Gressner, A.M., 1993.

TNF-alpha and TGF-beta1 stimulate fibronectin synthesis and the

transdifferentiation of fat storing cells in the rat liver into myofibroblast.

Virchows Arch. B. Cell Pathol. 63, 123–130.

Bacon, B.R., Britton, R.S., 1990. The pathology of hepatic iron overload: a

free radical mediated process. Hepatology 11, 127–137.

Bacon, B.R., Tavill, A.S., Brittenham, G.M., Park, C.H., Recknagel, R.O.,

1983. Hepatic lipid peroxidation in vivo in rats with chronic iron

overload. J. Clin. Invest. 71 (3), 429–439.

Baird, M.B., Birnbaum, L.S., Sfeier, G.T., 1980. NADPH driven lipid

peroxidation in rat liver nuclei and nuclear membranes. Arch. Biochem.

Biophys. 200, 108–115.

Ball, C.R., 1966. Estimation and identification of thiols in rat spleen after

cysteine or glutathione treatment relevant to protection nitrogen

mustards. Biochem. Pharmacol. 15, 809–816.

Ballatori, N., 1994. Glutathione mercaptides as transport forms of metals.

Adv. Pharmacol. 27, 271–298.

Beers, R.F., Sizer, I.W., 1952. A spectrophotometric method for measuring

the breakdown of hydrogen peroxide by catalase. J. Biol. Chem. 195,

133.

Bird, G.L., Sheron, N., Goka, A.K., Alexander, G.J., Williams, R.S., 1990.

Increased plasma tumour necrosis factor in severe alcoholic hepatitis.

Ann. Intern. Med. 112 (12), 917–920.

Chai, Y.C., Hendrich, S., Thomas, J.A., 1994. Arch. Bichem. Biophys. 310

(1), 264–272.

Chance, B., Sies, H., Boveris, A., 1979. Hydroperoxide metabolism in

mammalian organs. Physiol. Rev. 59, 527–605.

Chang, W.C., Chen, S.H., Wu, H.L., Shi, G.Y., Murota, S.I., Morita, I.,

1991. Cytoprotective effect of reduced glutathione in arsenical induced

endothelial cell injury. Toxicology 69, 101–110.

Chojkier, M., Houghum, K., Brenner, D.A., 1989. Stimulation of collagen

gene expression by ascorbic acid in cultured human fibroblast. A role

for lipid peroxidation. J. Biol. Chem. 264, 16957–16962.

Chouchane, S., Elizabeth, T.S., 2001. In vitro effect of arsenical compounds

on glutathione related enzymes. Chem. Res. Toxicol. 14, 517–522.

Datta, D.V., Mitra, S.K., Chuttani, P.N., Chakraborti, R.N., 1979. Chronic

arsenic oral intoxication as a possible etiological factor in idiopathic

portal hypertension (non-cirrhotic portal fibrosis) in India. Gut 20,

378–384.

DiMonte, D., Ross, D., Bellemo, G., Eklow, L., Orrenius, S., 1984.

Alterations in intracellular thiol homeostasis during the metabolism of

menadione by isolated rat hepatocytes. Arch. Biochem. Biophys. 235

(2), 334–342.

Franklin, M., Bean, W.B., Handin, R.C., 1950. Fowler’s solution as an

etiologic agent in cirrhosis. Am. J. Med. Sci. 219, 589–595.

Grattagliano, I., Vendemiale, G., Sabba, C., Buonamico, P., Altomare, E.,

1996. Oxidation of circulating proteins in alcoholics: role of acetalde-

hyde and xanthine oxidase. J. Hepatol. 25, 28–36.

Greenwel, P., Scwartz, M., Rosas, M., Peyrol, S., et al., 1991.

Characterization of fat storing cell lines derived from normal and

cirrhotic livers. Differences in the productions of IL-6. Lab. Invest. 65,

644–653.

Guerrieri, F., Vendemiale, G., Grattaliagno, I., Cocco, T., Pellecchia, G.,

Altomere, E., 1999. Mitochondrial oxidative alterations following

partial hepatectomy. Free Radical Biol. Med. 26, 34–41.

Guha Mazumder, D.N., Chakraborty, A.K., Ghosh, A., et al., 1988. Chronic

arsenic toxicity from drinking tubewell water in rural West Bengal.

Bull. World Health Organ 66, 499–506.

Guha Mazumder, D.N., Das Gupta, J., Santra, A., Pal, A., Ghosh, A.,

Sarker, S., 1998. Chronic arsenic toxicity in West Bengal—The worst

calamity in the World. Ind. Med. Assoc. 96 (1), 4–7.

Gyurasics, A., Varga, F., Gregus, Z., 1991. Glutathione dependent biliary

excretion of arsenic. Biochem. Pharmacol. 42, 465–468.

Halliwell, B., 1994. Free radicals, antioxidants, and human diseases:

curiosity, cause or consequence? Lancet 344, 721–724.

Hong, F., Kim, W.H., Tian, Z., Jaruga, B., Ishac, E., Shen, X., Gao, B.,

2002. Elevated IL-6 during ethanol consumption acts as a potential

endogenous protective cytokine against ethanol induced apoptosis in

the liver: involvement of induction of Bcl-2 and Bcl-x(L) proteins.

Oncogene 21 (1), 32–43 (Jan. 3).

Ikejima, K., Qu, W., Stachlewitz, R.F., Thurman, R.G., 1997. Kupffer

cell contains a glycine-gated chloride channel. Am. J. Pathol. 272,

G1581–G1586.

Ishinishi, N., Tomita, M., Hisanga, A., 1980. Study on chronic toxicity of

arsenic trioxide in rats with special reference to the liver damage.

Fukuoka Acta Med. 71, 27–34.

S. Das et al. / Toxicology and Applied Pharmacology 204 (2005) 18–2626

Ito, H., Okamoto, K., Kato, K., 1998. Enhancement of expression of stress

proteins by agents that lower the levels of glutathione in cells. Biochim.

Biophys. Acta 29 (1397(2)), 223–230.

Kawamura, K., Kobayashi, Y., Kageyama, E., et al., 2000. Enhanced

hepatic lipid peroxidation in patients with primary biliary cirrhosis. Am.

J. Gastroenterol. 95, 3596.

Krull, N.B., Gressner, A.M., 1992. Differential expression of keratan

sulphate proteoglycans fibromodulin, lumican and aggrecan in normal

and fibrotic rat liver. FEBS Lett. 312 (1), 47–52 (Nov. 2).

Lee, T.C., Wei, M.L., Chang, W.J., Ho, I.C., Lo, J.F., Jan, K.Y., Huang, H.,

1989. Elevation of Glutathione levels and glutathione-transferase

activity in arsenic resistant Chinese hamster ovary cells. In Vitro Cell.

Div. Biol. 25, 442–448.

Lee, K.S., Buck, M., Houglum, K., Chojkier, M., 1995. Activation of HSC

by TGF-alpha and collagen type I is mediated by oxidative stress

through c-myb expression. J. Clin. Invest. 96, 2461–2468.

Leonard, A., Lauwerys, R.R., 1980. Carcinogenecity, teratogenecity and

mutagenecity of arsenic. Mutat. Res 75 (1), 49–62.

Levine, R., Garland, D., Oliver, C.N., et al., 1990. Determination of

carbonyl content in oxidatively modified proteins. Methods Enzymol.

186, 464–478.

Lindgren, A., Vahter, M., Dencker, L., 1982. Autoradiographic studies on

the distribution of arsenic in mice and hamster administered 74As-

arsenite or arsenate. Acta Pharmacol. Toxicol. 51, 253–265.

Lowry, O.H., Rosenberg, N.J., Far, A.L., Randall, J., 1951. Protein

measurement with the folin phenol reagent. J. Biol. Chem. 193, 265.

Marafante, E., Vahter, M., 1984. The effect of methyl transferase inhibition

on the metabolism of (74As) arsenite in mice and rabbits. Chem. Biol.

Interact. 50, 49–57.

Marafante, E., Vahter, M., 1987. Effects of low dietary intake of

methionine, choline or proteins on the biotransformation of arsenite

in the rabbit. Toxicol. Lett. 37, 41–46.

Matsuoka, M., Pham, N.T., Tsukamoto, H., 1989. Differential effects of IL-

1 alpha, TNF-alpha, TGF-beta1 on cell proliferation and collagen

formation by cultured fat storing cells. Liver 9, 71–78.

McClain, C.J., Cohen, D.A., 1989. Increased tumor necrosis factor

production by monocytes in alcoholic hepatitis. Hepatology 9 (3),

349–351.

Nevens, F., Van, S.W., Sciot, R., Desmet, V., Groote, I.D., 1990. Arsenic and

noncirrhotic portal hypertension—A report of eight cases. J. Hepatol. 11,

80–85.

Niemela, O., Parkkila, S., Yla, H.S., Villaneuva, J., Ruetner, B., Halsted,

C.H., 1995. Sequential acetaldehyde production, lipid peroxidation and

fibrogenesis in a micropig model of alcohol induced liver disease.

Hepatology 22, 1208–1214.

Niemela, O., Parkkila, S., Britton, R.S., et al., 1999. Hepatic lipid

peroxidation in hereditary hemochromatosis and alcoholic liver injury.

J. Lab. Clin. Med. 133, 451.

Ochi, T., Kaise, T., Oyabhta, Y., 1994. Glutathione plays different roles

in the induction of the cytotoxic effects of inorganic and organic

arsenic compounds in cultured BALB/c 3T3 cells. Experientia 50 (2),

115–120.

Paglia, D.E., Valentine, W.N., 1967. Studies on the quantitative and

qualitative characterization of erythrocyte glutathione peroxidase.

J. Lab. Clin. Med. 70, 158–169.

Paradis, V., Kollinger, M., Faber, M., Holstege, A., Poynard, T., Bedossa,

P., 1997. In situ detection of lipid peroxidation by-products in chronic

liver diseases. Hepatology 26, 135–142.

Parola, M., Pinzani, M., Casini, A., et al., 1996. Induction of procollagen

type I gene expression and synthesis in human hepatic stellate cells

by 4-hydroxy 2, 3-nonenal and other 4-hydroxy-2, 3-alkenals is

related to their molecular structure. Biochem. Biophys. Res.

Commun. 222, 261.

Pena, L.R., Hill, D.B., McClain, C.J., 1999. Treatment with glutathione

precursor decreases cytokine activity. J. Parenter. Enteral Nutr. 23 (1),

1–6.

Pietorangelo, A., Gauldi, R., Cassalgrandi, G., Geerts, A., De Blesser, P.,

Montosi, G., Ventura, E., 1994. Enhanced hepatic collagen type I

mRNA expression into fat storing cells in a rodent model of

hemochromatosis. Hepatology 19, 714–721.

Poli, G., Albano, E., Dianzani, M.U., 1987. The role of lipid peroxidation in

liver damage. Chem. Phys. Lipids 45, 117–142.

Rudiger, H., Clavien, P.A., 2002. Tumor necrosis factor alpha, but not Fas,

mediates hepatocellular apoptosis in the murine ischemic liver. Gastro-

enterology 122, 202–210.

Santra, A., Maity, A., Das, S., Lahiri, S., Chakraborty, S.K., Guha

Mazumder, D.N., 2000. Hepatic damage caused by chronic arsenic

toxicity in experimental animals. Clin. Toxicol. 38, 395–405.

Sarin, S.K., Sharma, G., Banerjee, S., Sathayat, R., Malhotra, V., 1999.

Hepatic fibrogenesis using chronic arsenic ingestion: studies in a

murine model. Ind. J. Exp. Biol. 37, 147–151.

Sies, H., Cadenas, E., 1985. Oxidative stress: damage to intact cells and

organs. Philos. Trans. R Soc. Lond., Ser. B. Biol. Sci. 17 (311(1152)),

617–631.

Sokol, R.J., Devereaux, M., Mierau, G.W., Hambidge, K.M., Shikes, R.H.,

1990. Oxidant injury to hepatic mitochondrial lipids in rats with dietary

copper overload. Modification by vitamin E deficiency. Gastroenterol-

ogy 99 (4), 1061–1071 (Oct).

Spector, A., 1984. The search for a solution to senile cataracts. Proctor

lecture. Invest. Opthalmol. Visual Sci. 25, 130–146.

Starke, D.W., Chen, Y., Bapna, C.P., Lesnefsky, E.J., Mieyal, J.J., 1997.

Sensitivity of protein sulfhydryl repair enzymes to oxidative stress. Free

Radical Biol. Med. 23, 373–384.

Sternleib, I., 1980. Gastroenterology 78, 1615–1628.

Utley, H.G., Bernheim, F., Hochstein, P., 1967. Effects of sulfhydryl

reagents on peroxidation in microsomes. Arch. Biochem. Biophys.

118, 29.

Vendemiale, G., Grattagliano, I., Signorile, A., Altomere, E., 1998. Ethanol

induced changes of intracellular thiol compartmentation and protein

redox status in the liver: effect of tauroursodeoxycholate. J. Hepatol. 28

(1), 46–53.

Vessey, D.A., 2002. Hepatic metabolism of xenobiotics in humans. In:

Zakim, D., Boyer, T.D. (Eds.), Hepatology—A text book of liver

disease, 4th ed. Saunders.

Wang, Z., Rossman, T.G., 1996. The carcinogenecity of arsenic. In: Chang,

L.W. (Ed.), Toxicology of Metals, pp. 219–227.

Weiner, F.R., Giambrone, M.A., Czaja, MJ., 1990. Ito cell gene expression

and collagen regulation. Hepatology 11, 111–117.

Woessner Jr., J.F., 1961. Arch. Biochem. Biophys. 93, 440–447.

Yamanaka, K., Hasegawa, A., Swamura, R., Okuda, S., 1989. DNA strand

breaks in mammalian tissues induced by methylarsenics. Biol. Trace

Elem. Res. 21, 413–417.

Yamanaka, K., Hoshino, M., Okamoto, M., Sawamura, R., Hasegawa,

A., Okuda, S., 1990. Induction of DNA damage by dimethylar-

sine, a metabolite of inorganic arsenic, is for the major part likely

due to its peroxyl radical. Biochem. Biophys. Res. Commun. 168,

58–64.

Zindenberg, C.S., Olin, K.L., Villarweva, J., 1991. Ethanol induced

changes in hepatic free radical defense mechanisms and fatty acid

composition in the miniature pig. Hepatology 13, 1185.