A proteomics study of chemically induced cirrhosis in rat liver revealed the mechanism of thioacelamide hepatotoxicity 4

Cluster 1 Energy Metabolism Cluster 2 Iron and Heme Related Cluster 3 Carbonyl Metabolizing Cluster 4 Glutathione Related Long chain acyl-CoA dehydrogenase ↓ β-oxidation Ferritin l

Chapter 4 Discussion

4.1 General discussion

Oral administration and intraperitoneal injection of TAA are both established methods in the generation of fibrosis and cirrhosis models of animals, particularly in the rats We adopted the latter, which is achieved by injecting 200-300 mg/kg of TAA to the

rats for a span of 2 to 3 months (Chiijiwa et al., 1994; Zhao et al., 2002) TAA was

chosen for our studies mainly because TAA-induced nodules in rats are more prominent than that of those induced by other chemicals and the histology of the TAA model was

more akin to human liver cirrhosis (Li et al., 2002)

TAA is a thiono-sulfur containing compound that is both necrogenic (Landon et

al , 1986) and carcinogenic (Kizer et al., 1985) to the liver In 1984, TAA toxicity was

first associated with thioacetamide S-oxide (TSO), which is derived from the biotransformation of TAA by the microsomal FAD-monooxygense (FMO) system

(Chieli et al., 1984) Recently, certain isoforms of cytochrome P450 such as CYP2E1 and CYP2B (Lee et al., 2003) were reported as also being involved in the metabolic

activation of TAA to its toxic metabolites

During the biotransformation of TAA, both flavin-containing monooxygenase (FMO) and cytochrome P450 reduce dioxygen to superoxide anion, which is in turn,

catalyzed (Ekström et al., 1989) to form hydrogen peroxide, an important reactive

oxygen species (ROS) that is held responsible for cellular oxidative stress Therefore, it

(Sanz et al., 2002), an increase in malondialdehyde (MDA) (So et al., 2002) and the disappearance of tetraploid hepatocytes (Díez-Fernández et al., 1993) Nevertheless, the

detailed biochemical mechanisms underlying this hepatotoxic process of TAA remain largely unknown

In our attempt to unravel the pathogenesis of TAA-induced liver cirrhosis in rat liver and also to identify potential biomarkers for early detection of liver fibrosis, our 2-DE-MS approach had successfully identified 59 protein spots which displayed differential expression upon TAA administration, some of which could be correlated with the chronicity of TAA administration Subsequently, these proteins were categorized according to their common functions and properties As a result, we obtained four functional clusters as shown in Table 4-1 We attempted to deduce the basis of TAA-induced toxicity related to fibrosis These deductions were then collaborated and supported by evidences available in the literature The inter-relationship of the four clusters of proteins is presented as Fig 4-11

Cluster 1

Energy Metabolism

Cluster 2 Iron and Heme Related

Cluster 3 Carbonyl Metabolizing

Cluster 4 Glutathione Related

Long chain acyl-CoA

dehydrogenase (↓) β-oxidation Ferritin light chain (↓) iron storage Aldose reductase (↑)

detoxify aldehydes

GPx selenium dependent (↓)

anti-oxidant, binding to H2O2

glutathione transferase Pi (↑)

S-anti-oxidant, binding to lipid and DNA hydroxyperoxides

Enoyl CoA hydratase

(5 isoforms) (↓)

β-oxidation

& branched chain amino acid breakdown

dehydrogenase (↓) detoxification

Malate dehyrogenase

4.2 Cluster 1: energy metabolism

Proteins categorized as the first cluster are involved in fatty acid β-oxidation, the Krebs cycle, breakdown of branched chain amino acids such as isoleucine and valine as well as breakdown of methionine

4.2.1 Fatty acid catabolism

As shown in Table 3-1 in Chapter 3, long chain acyl-CoA dehydrogenase (LCAD) was down-regulated at least two folds in the treated rat liver samples from the third week of TAA treatment Fatty acids are broken down through the β-oxidation of

fatty acyl-CoA (LCAD), which is a flavoenzyme responsible for the formation of a trans

- α, β double bond in fatty acyl-CoA through dehydrogenation, the first committed step of fatty acid breakdown This reaction involves the removal of a proton at Cα of fatty acyl-CoA and transfer of a hydride ion equivalent from Cβ to FAD prosthetic group of LCAD, forming FADH2

Subsequently, electron transferring flavoprotein (ETF) transfers an electron pair from the FADH2 prosthetic group to the flavo-iron-sulfur protein ETF: ubiquinone oxidoreductase and is followed by reactions in the electron transport chain in the

Following the dehydrogenation of the fatty acyl-CoA to form an enoyl-CoA, enoyl-CoA hydratase catalyzes the stereo-specific addition of H2O to its substrate’s

trans-α, β double bond to form 3-L-Hydroxyacyl-CoA, a secondary alcohol Coincidentally, the second enzyme in the β-oxidation pathway, short chain enoyl-CoA hydratase (3 isoforms) was down-regulated A detailed description of this pathway and the differentially expressed proteins are illustrated in Fig 4-1

β-Ketoacyl-CoA

CoA dehydrogenase NAD +

3-L-hydroxyacyl-NADH + H +

Fatty acyl-CoA (2 C shorter) + Acetyl-CoA

or Fatty acyl-CoA (2 C shorter) + Succinyl-CoA (from β-

oxidation of odd chain fatty acids derived from diet

β-Ketoacyl-CoA thiolase

LCAD FAD

FADH 2

ETF ETF ox

Electron transport chain

4.2.1.1 Cluster 1: Hypothesis derived from observations

It has been well documented in the literature that the first step of fatty acyl-CoA breakdown involving acyl-CoA dehydrogenases and electron transfer flavoproteins (ETF) are crucial for survival This has been exemplified in the deficiency of acyl-CoA dehydrogenases in sudden infant death syndrome (SIDS) and Jamaican vomiting sickness, both of which can lead to death SIDS involves the Lys 304 → Glu mutation in medium chain acyl-CoA dehydrogenases (MCAD) while the second involves consumption of the poisonous ackee fruit that contain hypoglycin A, which was converted to methylenecyclopropylacetyl-CoA (MCPA-CoA), a mechanism inhibitor of acyl-CoA dehydrogenases

Our results show that enzymes involved in fatty acid metabolism are regulated Decrease in β-oxidation might lower ATP production It was also suspected that a decreased β-oxidation may lead to accumulation of fatty acids in the affected livers

down-(Grimbert et al., 1993) (Fig 4.2)

TAA Administration

Decrease in β-Oxidation of Lipid?

Accumulation of Fatty Acids?

Figure 4-2 Suppression of fatty acid β-oxidation by TAA, as implicated by the 2DE-MS

data, may result in the accumulation of fatty acids in the liver, causing fatty change or steatohepatitis

4.2.1.2 Supporting evidence from scientific literature

This hypothesis turned out to be supported by published results Firstly, a paper

by Müller and Dargel (Müller et al., 1984) showed rat livers chronically treated with

TAA exhibited a decrease in fatty acid oxidation They also showed that the structural and functional alterations of mitochondria from chronically treated rats remained unaffected for at least two weeks after cessation of TAA application Besides, several

early reports have also recorded fatty change in liver as a result of TAA toxicity (Smith et

al , 1968; Vorbrodt et al., 1966) For instance, Zimmermann (Zimmerman, 1976) reported that TAA is moderately steatogenic while Fernandez (Fernandez et al., 1997)

found that the nodular cirrhosis of TAA-treated rat livers developed increased collagen content and lipid accumulation

This hypothesis and its accompanying verifications are illustrated in detail as Fig 4-3 Collectively, our results are in concert with the biochemical and toxicological observations pertaining to TAA-administered livers These results indicate that the down-regulation of enzymes involved in β-oxidation might be the underlying cause for the observed TAA-induced decrease in β-oxidation and development of steatohepatitis

TAA

Administration

Decrease in Oxidation

β-Fatty changes, steatosis

Lipid Accumulation & Cirrhosis

II

I

III

Figure 4-3 I.) Rat livers chronically treated with TAA was shown to have decreased fatty acid oxidation

(Muller and Dargel, 1984) II.) Fatty change or steatosis in TAA-treated livers had been observed and

documented since the 1960’s (Smith et al., 1968; Vorbrodt et al., 1966; Zimmerman, 1976) III.) Nodular

cirrhosis of TAA treated rat livers were observed to develop increased collagen content and lipid accumulation

(Fernandez et al., 1997)

4.2.2 Metabolism of methionine

Impairment of methionine metabolism was first reported (Kinsell et al., 1947) in

human liver cirrhosis over 50 years ago and later in several experimental models of liver

injury (Lieber et al., 1990) Our results showed that two methyltransferases: glycine

N-methyltransferase (GNMT) and thioether N-methyltransferase were down-regulated in the TAA-treated liver tissues Both enzymes are methyltransfereases that catalyze the second step of methionine degradation (Fig 4-4) They are responsible for the conversion of S-adenosylmethionine (SAMe) to S-adenosylhomocysteine (SAH) and thus play an

important role in regulating their ratios (Fu et al., 1996) GNMT is also a major folate binding protein (Raha et al., 1994) Therefore it is able to induce changes in tissue folate

status which would result in chromosome breakage or abnormal DNA methylation (Duthie, 1999)

A close inspection of literature reveals that GNMT mRNA had been reported to

be down-regulated in HCC, hepatitis C virus (HCV)-induced and alcoholic livers (Avila

et al , 2000; Chen et al., 1998) In fact, down-regulation of methionine adenosyl transferase mRNA was also observed in cirrhotic models of rats (Avila et al., 2000)

However, this was not observed in our 2-D gels

As shown in Fig 4-4, the methionine breakdown pathway is important for

the production of succinyl-CoA, which is a high energy intermediate in the

tricarboxylic acid (TCA) cycle and an important precursor for heme biosynthesis

in bone marrow and liver

Methionine

S-Adenosylmethionine

(SAM)

Methionine adenosyl transferase

Methyl acceptor Methylated acceptor

Methylase such as:

i.) Glycine N-methyl transferase ii.)Thioether methyl transferase

β-Cystathionine lyase

α-α-keto acid

H 2 O

NH 3

4.2.3 Branched-chain amino acid catabolism

Mitochondrial short-chain enoyl-CoA hydratase, which is found to be regulated in our 2-D gels, catalyzes the transfer of water to a fatty acid, can also play a major role in the metabolic breakdown of branched chain amino acids such as valine and isoleucine The metabolic products of valine and isoleucine are further broken down to propionyl-CoA, which is also a product of methionine breakdown and odd-chain fatty acid degradation It is then converted to succinyl-CoA by a series of reactions involving the participation of biotin and co-enzyme B12

Acetyl-CoA acetyl transferase

H 2 O

β-hydroxyisobutyryl-CoA hydrolase

β-hydroxyisobutyrate dehydrogenase

H 2 O CoASH

NAD + NADH

Short chain enoyl-CoA hydratase

4.2.4 Summary for Cluster 1

In summary, cluster 1 consists of proteins that are involved in the degradation of lipid and amino acids and thus, it is important in energy metabolism Interestingly, all these proteins are down-regulated Besides, it is interesting to note that these three affected pathways are involved in the formation of succinyl-CoA, a high energy intermediate in the Krebs cycle

Met breakdown

i.) GNMT (↓)

ii.) Thioether methyltrasferase

Valine and Isoleucine breakdown

i.) Enoyl-CoA hydratase (↓)

4.3 Cluster 2: iron and heme metabolism

4.3.1 Hypothesis I: succinyl-CoA, heme and cytochrome P450

Succinyl-CoA is an important precursor for the biosynthesis of heme, which, in the liver, is important for the biosynthesis of liver hemoproteins, especially cytochrome P450 Therefore, we suspected that the down-regulation of the 3 primary metabolic pathways in Cluster 1 will probably lead to the depletion of succinyl-CoA

Succinyl-CoA and glycine are precursors for heme biosynthesis In the first committed step of heme biosynthesis, both precursors are catalyzed by δ-aminolaevulinic acid (ALA) synthase to form ALA This first step is also the rate-limiting step for heme biosynthesis Besides, it was shown that the rate of hepatic synthesis of ALA is

modulated by the supply of succinyl-CoA (Bonkowsky et al., 1977), as illustrated in Fig

4-8 As a result, we propose that the depletion of succinyl-CoA would contribute to the decrease in heme biosynthesis The drop in hepatic heme concentration will probably lead to a decrease in the expression level of cytochrome P450 because heme is a positive

regulator of cytochrome P450 gene transcription (Bhat et al., 1988; Satyabhama et al.,

1986)

However, a close inspection of our list of differentially expressed proteins failed

be differentially expressed These proteins include hemoproteins such as catalase and sulfite oxidase and heme binding proteins such as albumin, haptoglobin and hemopexin With the exception of hemopexin that was up-regulated, all of them were down-regulated

Catalase Sulfite Oxidase

Heme binding proteins

Liver hemoproteins

Figure 4-7 We hypothesized that the depletion of succinyl-CoA would result in

the lowering of heme biosynthesis This, in turn, results in the lowering of cytochrome P450 a major hemoprotein in the liver The table on the right, meanwhile, shows a list of iron-related proteins that are down-regulated with TAA treatment, with the exception of hemopexin

4.3.2 Supporting evidence

To support our hypothesis (Fig 4-8), we found that in vivo TAA administration

to male rat had been demonstrated to inhibit δ-aminolevulinic acid (ALA) synthetase

(Matsuura et al., 1983) In the same paper, Matsuura and co-workers also described that

TAA administration resulted in the decrease of cytochrome P450 concentration and drug metabolizing enzyme activities Though not absolutely conclusive, our proteomics data does suggest that the depletion of succinyl-CoA, as implicated in the section 4.2, could be one underlying factor for the inhibition of ALA-synthetase and the decrease in cytochrome P450 Furthermore, TAA was also shown to be an inhibitor of heme

synthesis (Satyabhama et al., 1988 and Bhat G J et al., 1986) and it blocks the

phenobarbitone-mediated increase in the transcription of cytochrome P450 mRNAs in the

liver (Satyabhama et al., 1988) probably due to decrease in heme, which serves as a positive regulator (Bhat G J et al., 1986) of cytochrome P450 gene transcription

Our hypothesis and the supporting publications are collectively presented as Fig 4-8 Taken together, we showed that the proteomics data that we obtained can provide logical and plausible answers to explain the effects of TAA on heme and cytochrome P450 biosynthesis as reported in other studies