Hepatic transmethylation reactions in micropigs with alcoholic liver disease



Alcoholic liver disease is associated with abnormal hepatic methionine metabolism, including increased levels of homocysteine and S-adenosylhomocysteine (SAH) and reduced levels of S-adenosylmethionine (SAM) and glutathione (GSH). The concept that abnormal methionine metabolism is involved in the pathogenesis of alcoholic liver disease was strengthened by our previous findings in a micropig model where combining dietary folate deficiency with chronic ethanol feeding produced maximal changes in these metabolites together with early onset of microscopic steatohepatitis and an eightfold increase in plasma aspartate aminotransferase. The goal of the present study was to determine potential mechanisms for abnormal levels of these methionine metabolites by analyzing the transcripts and activities of transmethylation enzymes in the livers of the same micropigs. Ethanol feeding or folate deficiency, separately or in combination, decreased transcript levels of methylenetetrahydrofolate reductase (MTHFR), methionine adenosyltransferase (MAT1A), glycine-N-methyltransferase (GNMT) and S-adenosylhomocysteine hydrolase (SAHH). Ethanol feeding alone reduced the activities of methionine synthase (MS) and MATIII and increased the activity of GNMT. Each diet, separately or in combination, decreased the activities of MTHFR and SAHH. In conclusion, the observed abnormal levels of methionine metabolites in this animal model of accelerated alcoholic liver injury can be ascribed to specific effects of ethanol with or without folate deficiency on the expressions and activities of hepatic enzymes that regulate transmethylation reactions. These novel effects on transmethylation reactions may be implicated in the pathogenesis of alcoholic liver disease. (HEPATOLOGY 2004;39:1303–1310.)

Abnormal hepatic methionine metabolism can result from chronic exposure to ethanol or from folate deficiency. Hyperhomocysteinemia has been described in chronic alcoholics,1, 2 and decreased liver S-adenosylmethionine (SAM) levels were found in experimental animals and cultured hepatocytes exposed to ethanol.3–6 Folate deficiency also results in abnormal methionine metabolism, because 5-methyltetrahydrofolate (5-MTHF) is substrate for the methionine synthase (MS) reaction that converts homocysteine to methionine, the substrate for SAM (Fig. 1). Furthermore, folate deficiency is frequently associated with alcoholic liver disease, in part, because chronic exposure to ethanol reduces the intestinal transport of folic acid and increases folate excretion in the urine.7–9

Figure 1.

Transmethylation reactions in the liver. Arrows indicate the directions of change in methionine metabolites in response to the diets as previously described.15 Methylenetetrahydrofolate reductase (MTHFR) converts 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate (5-MTHF) in a flavin adenine dinucleotide (FAD)-dependent reaction that is normally down-regulated by S-adenosylmethionine (SAM). 5-MTHF is substrate with cofactor cobalamin (Cob) for methionine synthase (MS) that generates methionine from homocysteine. Demethylated Cob I is oxidized to Cob II, which is reduced to the methyl donor Cob III by FADdependent methionine synthase reductase (MSR) with SAM as substrate. In the alternate salvage pathway, betaine is substrate for betaine homocysteine methyltransferase (BHMT) in the conversion of homocysteine to methionine. Methionine adenosyltransferase (MAT1A) adds adenosine triphosphate (ATP) to methionine for generation of SAM, the methyl donor for multiple reactions. Glycine N-methyltransferase (GNMT) converts SAM to S-adenosylhomocysteine (SAH) and is normally down-regulated by 5-MTHF. SAH is substrate for S-adenosyl homocysteine hydrolase (SAHH) in the generation of homocysteine, but it is reversible and may increase SAH during homocysteine accumulation.

Previous data support the concept that changes in methionine metabolism are involved in the development of alcoholic liver disease. Studies in rats and micropigs have demonstrated MS activity is reduced during chronic ethanol exposure.10–13 Rats that were fed ethanol by gastric tube demonstrated decreased methionine adenosyltransferase MAT1-III activity with decreased SAM and SAM/S-adenosylhomocysteine (SAH) ratio in association with increased DNA strand breaks.5 A MAT1A knockout mouse model developed hepatic steatosis with reduced liver SAM and glutathione (GSH) and enhanced expression of several proinflammatory genes.14 The administration of SAM to chronic ethanol-fed baboons attenuated reductions in SAM and GSH as well as the histopathology of early alcoholic liver disease.3

Recently, we explored the hypothesis that the combination of folate deficiency with chronic ethanol feeding would accelerate the development of alcoholic liver injury by enhancing abnormal methionine metabolism.15 This hypothesis is based on the integral role of folate in methionine metabolism (Fig. 1) and prior evidence for the effects of ethanol on promoting folate deficiency and altering methionine metabolism.3–7, 16 We developed a micropig model for alcoholic liver disease based on 14 weeks of feeding diets that either contained ethanol or were deficient in folate, or both factors in combination.15 Compared with the control group fed a folate-sufficient diet without ethanol, each folate-deficient diet reduced liver folate by one half, and the combined diet resulted in a fourfold elevation in plasma homocysteine. Ethanol feeding alone increased liver homocysteine levels by one half and reduced liver SAM by one third and the GSH/oxidized glutathione (GSSG) ratio by one half. The combined diet reduced mean liver methionine levels by two thirds and increased liver SAH twofold; the urinary excretion of 8-hydroxy-2'-deoxyguanosine (oxo[8]dG), an index of DNA oxidation, by twofold; DNA strand breaks by one third; and the mean level of plasma aspartate aminotransferase by eightfold. Only the micropigs that were fed the combined diet developed a histopathology of alcoholic liver injury with steatosis, hepatocellular necrosis, and focal inflammation, whereas liver histology remained normal in the other three groups. This demonstration of abnormal histopathology after just 14 weeks of the combined diet contrasts with a 1-year requirement to attain a similar histopathology in micropigs fed ethanol with a folate-sufficient diet,17 and strengthens the concept that abnormal levels of one or more methionine metabolites participate in the pathogenesis of alcoholic liver disease.

The goal of the present study was to determine the bases for the observed changes in methionine metabolites by measuring the transcripts and activities of enzymes that regulate transmethylation reactions in liver tissues from the same experimental micropigs.


SAM, S-adenosylmethionine; 5-MTHF, 5-methyltetrahydrofolate; MS, methionine synthase; MAT, methionine adenosyltransferase; SAH, S-adenosylhomocysteine; GSH, glutathione; GSSG, oxidized glutathione; FDE, folate deficient with ethanol; cDNA, complementary DNA; MTHFR, methylenetetrahydrofolate reductase; oxo(8)dG, 8-hydroxy-2'-deoxyguanosine; GNMT, glycine N-methyltransferase; SAHH, S-adenosyl homocysteine hydrolase; PCR, polymerase chain reaction; PMSF, phenylmethyl sulfonyl fluoride; FAD, flavin adenine dinucleotide; NADPH, nicotinamide adenine dinucleotide phosphate; Cob, cobalamin.

Materials and Methods

Animals and Diets

As previously described,15 24 juvenile Yucatan micropigs were obtained from Sinclair Farms (Columbia, MO) and were pair-fed diets that provided 90 kcal/kg body weight with 15% of kcal as vitamin-free casein, 30% of kcal as corn oil, and 55% of kcal as carbohydrate (control) or diets in which carbohydrate was reduced to 15% and ethanol was substituted to provide 40% of kcal. Folic acid was absent or was added to diets at 14.5 μg/kg body weight, and each diet was supplemented with a vitamin and mineral mix that included levels that were in excess of pig requirements for choline at 60.3 mg and methionine at 675 mg/kg body weight18 (Dyets, Inc., Bethlehem, PA). Thus there were four feeding groups of six animals each: folate sufficient control, folate deficient only, folate sufficient with ethanol, and folate deficient with ethanol (FDE). The micropigs were housed in facilities approved by the National Institutes of Health and were cared for following standards and procedures outlined in the National Academy of Sciences Guide for the Care and Use of Laboratory Animals (http://bob.nap.edu/readingroom/books/labrats/). All procedures were reviewed and approved by the Animal Welfare Committee of the University of California Davis. After 14 weeks of feeding, liver tissues were surgically removed from all animals under anesthesia, freeze-clamped in liquid nitrogen, and frozen at −70°C for further analysis.


5-[14C]-MTHF and S-adenosyl-L-[methyl-3H]-methionine were purchased from Amersham Phamacia Biotech (Piscataway, NJ). [Methyl-3H]-L-methionine was purchased from Moraveck Biochemical, Inc. (Brea, CA). Cyto-Scin scintillation cocktail was purchased from Fisher Scientific (Houston, TX). AG 50W-X4 resin and poly-prep chromatography columns were purchased from BioRad Laboratories Life Science Group (Richmond, CA). The TOPO TA-cloning kit was purchased from Invitrogen (Carlsbad, CA). The QIAprep Spin Miniprep Kit was obtained from QIAGEN, Inc. (Valencia, CA). All other chemicals and reagents were obtained from Sigma Chemical Co. (St. Louis, MO) and various commercial sources.

Isolation, Purification, and Cloning of Nucleic Acids

Total RNA was isolated from liver tissue by acid guanidinium thiocyanate-phenol chloroform extraction19 using Trizol reagent (Life Technologies, Rockville, MD). The quality of the RNA extracts was analyzed by gel electrophoresis in 1.0% agarose gel. First-strand complementary DNA (cDNA) was synthesized from 3 μg of total RNA by the SuperScript Preamplification System after DNase treatment and spectrophotometry reading at 260 nm according to the manufacturer (Life Technologies). Initially, we designed pig-specific sense and antisense primers to amplify 400- to 800-bp fragments within the open reading frame of each gene. Primers for methylenetetrahydrofolate reductase (MTHFR), MS, glycine-N-methyltransferase (GNMT), and 18S transcripts were obtained from pig-specific sequences, and primers for MAT1A and S-adenosyl homocysteine hydrolase (SAHH) were obtained from highly homologous regions of different species in Genbank and the multiple sequence alignment program Clustal W version 1.8 (Table 1).20 A fragment of the open reading frame of each gene was obtained by polymerase chain reaction (PCR) using Taq polymerase, and each amplified cDNA fragment was ligated to the TA-pCR-2.1 TOPO vector. Transformation into chemically competent Escherichia coli cells (Life Technologies) was followed by DNA isolation from transformants by the mini-preparation procedure described by QIAGEN, Inc. Following sequence confirmation, each cDNA fragment was resolved in 1.0% agarose gel, and specific bands were purified by the QIAEX II gel extraction procedure (QIAGEN, Inc.).

Table 1. Primers Used for Pig cDNA Constructs
GeneSenseAntisenseFragment (bp)
  • *

    From published sequences of pig.

  • From homologous regions of different species.


Quantitation of Transcripts by Real-Time PCR

Pig-specific primers were designed to amplify 51 bp from the cDNA clone of each gene by Hot Start real-time PCR in the GeneAmp 5700 sequence detection system (Applied Biosystems) and are shown in Table 2. PCR products were detected by the increase in fluorescence caused by the binding of SYBR Green to double-stranded DNA. Transcripts of each gene were quantified using standard curves obtained with each purified cDNA fragment. The data are expressed as a ratio with 18S ribosomal RNA simultaneously amplified in separate PCR reactions.21

Table 2. Primers Designed From Authentic Pig cDNA Sequence for Real-Time PCR Quantitation
GeneSenseAntisenseFragment (bp)

Enzyme Assays

The activities of transmethylation enzymes were measured in supernatant or cytosol fractions from each liver specimen. Each preparation contained the protease inhibitors aprotinin, leupeptin, benzamidine, and phenylmethyl sulfonyl fluoride (PMSF).


Frozen liver tissue was homogenized in 10 volumes of 0.1 M potassium phosphate buffer (pH 6.3). The liver homogenate was centrifuged at 105,000g for 1 hour at 4°C and 500 μg protein of the cytosol fraction were incubated for 50 minutes at 37°C with 50 μL 1.0 M phosphate buffer (pH 6.3), 40 μL 0.025 M menadione bisulfate, 4 μL 0.05 M ascorbic acid, 10 μL 0.81 mM flavin adenine dinucleotide (FAD), 4 μL 0.1 M ethylenediaminetetraacetic acid (pH 6.3), and 12 μL 5-[14C]-MTHF (1618 cpm/nmol) in a total volume of 280 μL. The reaction was stopped by the addition of 250 μL 0.6 M sodium acetate, 150 μL 0.4 M dimedone, and 100 μL 0.1 M formaldehyde. MTHFR activity was quantified by measuring the inverse enzyme reaction of radio-labeled product [14C]-5-MTHF to [14C]-N5, N10-methylenetetrahydrofolate using menadione as the electron acceptor.22 Oxidized [14C]-N5, N10-methylenetetrahydrofolate was dissociated into [14C]-formaldehyde and tetrahydrofolate, the [14C]-formaldehyde was coupled with dimedone and extracted into toluene, and the radioactivity was determined in a scintillation counter.23


As previously described,15 liver tissue was homogenized in five volumes of 0.028 M sodium phosphate buffer (pH = 7.0) followed by centrifugation at 40,000g for 30 minutes. Approximately 4.0 mg of supernatant protein were incubated with 25 μmol potassium phosphate buffer (pH = 7.5), 100 μmol sodium chloride, 100 μmol 2-mercaptoethanol, 1.25 μmol L-homocysteine, 50 nmol S-adenosylmethionine, 37.5 nmol 5-MTHF, 7.5 μL 5-[14C]-MTHF (1,233 cmp/nmol), 15 nmol cyanocobalamin, and 50 nmol FADH2 reduced with platinum oxide and hydrogen gas in a total volume of 0.5 mL. After 40 minutes of incubation at 37°C, samples were boiled for 3 minutes and chilled in ice for 5 minutes, followed by the addition of 500 μL of methionine-coated Norit-charcoal. After 10 minutes of incubation at 25°C, samples were centrifuged at 16,000g and 0.5 mL counted in a liquid scintillation counter.24


The synthesis of [3H]-SAM by adenosine triphosphate activation of [methyl-3H]-L-methionine was measured according to previously described methods.25, 26 Liver tissue was homogenized in eight volumes of 10 mM Tris HCl (pH 7.5) containing 0.3 M sucrose, 0.1% beta-mercaptoethanol, and 0.1% of the protease inhibitors benzamidine and PMSF. The homogenate was centrifuged at 30,000g for 60 minutes and 70 μL of supernatant normalized to approximately 800 μg of protein were incubated with 130 μL of a reaction mixture containing 80 mM TES pH 7.4, 50 mM KCl, 40 mM MgCl2, 10 mM dithiothreotol (DTT) 0.5 mM ethylenediaminetetraacetic acid, 5 mM adenosine triphosphate, and 5 mM hot/cold [methyl-3H]-L-methionine. The reaction was incubated for 40 minutes at 37°C, stopped by adding 2 mL cold distilled water, and immediately applied to a column containing 1 mL of AG 50W-X4 cation exchange resin (BioRad) previously activated for 4 hours with ammonium hydroxide and equilibrated with distilled water to pH 7.0. After sample application, the column was washed with 20 mL of distilled water and the [3H]-SAM generated was eluted with 4 mL of 3.0 N ammonium hydroxide. The eluate was neutralized with 1 mL of acetic acid and counted in a liquid scintillation counter (Beckman model LS 3801, Fullerton, CA).


Activity was determined in liver homogenates by using a charcoal assay to adsorb unreacted S-adenosyl-L-[methyl-3H]-methionine.27 Briefly, approximately 1,200 μg whole homogenate protein were incubated with 10 μL 1.0 M Tris HCl buffer (pH 7.5), 20 μL of 10 mM glycine, 20 μL of 1 mM S-adenosyl methionine, and 10 μL of S-adenosyl-L-[methyl-3H]-methionine (10,226 dpm/nmol) in a total volume of 100 μL. After 15 minutes of incubation at 25°C, the reaction was terminated by adding 50 μL of trichloroacetic acid followed by 350 μL of activated charcoal (38 mg/mL in 0.1 M acetic acid). Samples were incubated for 20 minutes at 25°C and centrifuged for 3 minutes at 14,000g, and 200 μL of the supernatant fraction were counted in a liquid scintillation counter.


Activity was measured in the forward hydrolysis direction.28 Briefly, 0.2 g of liver tissue were homogenized in 4 mL of 50 mM potassium phosphate buffer (pH 7.0). The homogenate was centrifuged at 30,000g for 60 minutes at 4°C and 60 μL of supernatant normalized to approximately 800 μg protein were incubated with 860 μL of reaction mixture containing 50 mM potassium phosphate buffer (pH 7.0), 1.2 mM MTT (3′[4,5-dimethylthiazol-2-yL]-2,5, diphenyl-tetrazolium bromide), 0.16% gelatin, 1.1 U adenosine deaminase, 0.45 U nucleoside phosphorylase, and 0.1 U xanthine oxidase. Following 5 minutes of incubation at 37°C, the SAHH reaction was started by the addition of 80 μL of 1.0 mM SAH to a total assay volume of 1 mL. SAHH-specific activity was related to the production of formazan by chart-recording the change in absorbance at 728 nm at 37°C in a Gilford model 260 spectrophotometer (Nova Biotech, El Cahon, CA).

Statistical Analyses

All group values are expressed as mean ± SEM. Significant differences among the groups were determined by two-way ANOVA. If the interaction term for the effects of ethanol feeding and folate deficiency was significant, subgroup analyses were performed by one-way ANOVA with Tukey's test for post hoc comparisons. Relationships among variables were determined by linear regression analysis using SPSS 10.0 for windows (SPSS, Inc., Chicago, IL).


Table 3 illustrates changes in mean ± SEM levels of transmethylation enzyme transcripts expressed as ratios with 18S. There was no effect of the feeding protocols on the expression of MS transcripts. Three of the transcript expressions in the FDE group were modified by interactions of the two feeding variables. Subsequent analyses showed the reducing effects of ethanol feeding and folate deficiency on MTHFR and SAHH transcripts in the FDE group were similar but nonadditive, and the effect of ethanol to reduce the expression of MAT1A was counteracted by the effect of folate deficiency. The transcript expressions of GNMT in the FDE group were affected independently and additively by ethanol feeding and folate deficiency.

Table 3. Transcripts of Transmethylation Enzymes
GeneRatio MultipleTranscript Level/I8S
  • NOTE: Values are expressed as a ratio with the transcript level of 18S.

  • *

    Effects of folate deficiency; P < .03, P < .05, P < .05, P < .02.

  • Effects of ethanol; P < .03, P < .02, P < .02, P < .03.

  • Interaction; P < .05, P < .005, P < .05.

  • Abbreviations: FS, folate sufficient control; FD, folate deficient only; FSE, folate sufficient with ethanol; FDE, folate deficient with ethanol.

MTHFR(10−5)0.32 ± 0.050.17 ± 0.03*0.16 ± 0.040.18 ± 0.03*
MS(10−4)0.81 ± 0.110.68 ± 0.080.64 ± 0.050.84 ± 0.14
MATIA(10−3)0.74 ± 0.160.34 ± 0.06*0.27 ± 0.060.48 ± 0.06*
GNMT(10−3)0.72 ± 0.160.35 ± 0.07*0.43 ± 0.090.24 ± 0.04*
SAHH(10−4)0.38 ± 0.050.24 ± 0.02*0.19 ± 0.050.21 ± 0.03*

Table 4 shows changes in mean ± SEM values for transmethylation enzyme activities in each feeding group. Ethanol feeding acted independently to decrease MS and MATIII activities and to increase GNMT activity. MTHFR activity was decreased independently by ethanol feeding and folate deficiency in group FDE, while the two factors acted independently and additively to decrease SAHH activity.

Table 4. Activities of Transmethylation Enzymes
Enzyme (Specific Activity)Enzyme Activity
  • NOTE: Values are expressed as pmol/mg protein/min unless indicated otherwise.

  • *

    Effects of folate deficiency; P < .005, P < .02.

  • Effects of ethanol; P < .0001, P < .0001, P < .001, P < .0001; P < .0001.

  • Abbreviations: FS, folate sufficient control; FD, folate deficient only; FSE, folate sufficient with ethanol; FDE, folate deficient with ethanol.

MTHFR (pmol/mg/min)22.6 ± 0.9916.8 ± 1.34*12.1 ± 0.9011.5 ± 0.56*
MS (pmol/mg/min)8.2 ± 0.448.9 ± 0.605.3 ± 0.536.1 ± 0.31
MATIII (pmol/mg/min)34.6 ± 1.4231.9 ± 2.8226.4 ± 1.5226.1 ± 1.19
GNMT (pmol/mg/min)2.4 ± 0.663.5 ± 0.537.9 ± 0.656.4 ± 1.00
SAHH (mU/mg/min)18.7 ± 0.9215.3 ± 1.33*11.4 ± 2.457.6 ± 1.04*

Certain changes in transmethylation enzyme activities correlated with our previous findings of changes in metabolite levels,15 according to linear regressions using measurements from all 24 animals. MTHFR activities were correlated negatively with homocysteine levels (r = −0.655, P < .001) and the SAM/SAH ratio (r = −0.757, P < .0001) and positively with GSH levels (r = 0.703, P < .001). MATIII activity was correlated positively with the SAM/SAH ratio (r = 0.532, P < .01). GNMT activity was correlated negatively with SAM (r = −0.515, P < .01) and the SAM/SAH ratio (r = −0.550, P < .01). SAHH activity was correlated negatively with both SAH (r = −0.467, P < .02) and homocysteine (r = −0.649, P < .001) and positively with the SAM/SAH ratio (r = 0.617, P < .001).


In previous studies, we developed the micropig model of alcoholic liver disease and demonstrated that 1 year of ethanol feeding produced the histological features of inflammation, steatonecrosis, centrilobular fibrosis, cirrhosis, and apoptosis in castrated males.13, 17, 29 Findings from our recent and much shorter 14-week study of uncastrated males strengthens the concept that abnormalities of certain methionine metabolites contribute to the pathogenesis of alcoholic liver disease. To summarize, micropigs in the FDE group demonstrated the greatest increase in SAH and decrease in the SAM/SAH ratio, the lowest GSH/GSSG ratio, and the greatest increase in plasma aspartate aminotransferase, together with the histopathology of alcoholic liver injury.15 Finding a significant correlation between lowered levels of SAM and GSH in that study provided a plausible link between abnormal methionine metabolism and antioxidant defense, because SAM is an essential activator of the transsulfuration pathway for GSH production (Fig. 1). The findings that the DNA strand breaks and the urinary excretion of oxo(8)dG were each maximally increased in pigs fed the FDE diet links abnormal methionine metabolism to DNA damage and increased susceptibility to apoptosis.

The present study shows that the single and combined effects of ethanol feeding and folate deficiency on methionine metabolites and alcoholic liver injury can be ascribed, in part, to specific effects on the expressions and activities of transmethylation enzymes. The previous finding of increased homocysteine in plasma and livers of the experimental animals15 is consistent with present findings of decreased regulatory activities of MTHFR, MS, and reversible SAHH (Fig. 1; Tables 3 and 4). Decreased SAM in the ethanol groups is consistent with reduced activity of MATIII and with increased activity of GNMT. Increased SAH in both folate-deficient and ethanol groups can be explained by the combination of increased GNMT activity and reduced activity of SAHH.

The observed suppression of transcript levels of four of the five transmethylation enzymes (Table 3) may be ascribed to the DNA oxidant effects of ethanol that were exacerbated by folate deficiency, as shown previously by enhanced oxo(8)dG excretion in the FDE group.15 Because SAM is the major methyl donor and SAH opposes its effects on DNA methylation,30 hypomethylation also may have played a part in altered transcription of the transmethylation genes in our study. In addition to its DNA effects, ethanol-induced oxidant stress can modify proteins in general by enhancing the irreversible covalent binding of acetaldehyde to lysine residues.31, 32

The present finding of decreased MTHFR activity in response to both ethanol feeding and folate deficiency may constitute another explanation for low serum folate levels in chronic alcoholics.16 However, we cannot determine whether decreased MTHFR activity resulted in selectively decreased end product 5-MTHF in the experimental groups, because folate analogs were not measured in their livers. In this regard, tetrahydrofolate, not 5-MTHF, is the major constituent of plasma folate in the pig,33 whereas the distribution of pig liver folates is not known. The active site of MTHFR incorporates FAD, which interacts with nicotinamide adenine dinucleotide phosphate (NADPH) to provide reducing equivalents for the synthesis of 5-MTHF.34, 35 The observed decreased GSH/GSSG ratio15 is consistent with increased activity of glutathione reductase.36 Because glutathione reductase is also a flavoprotein,37 its presumed increased activity would predictably decrease the availability of FAD and NADPH for the MTHFR reaction. Furthermore, SAM is required to stabilize the binding of FAD to the active site of MTHFR.38 Therefore, decreased MTHFR activity could be ascribed to conformational changes in the presence of reduced SAM, FAD, and NADPH levels.

The observed decrease in MS activity could be explained in part by potential oxidant effects of ethanol feeding on the conformation of this protein in relation to its dependence on cobalamin (Cob) and the interactive MS reductase reaction (Fig. 1). During transfer of the methyl group of 5-MTHF via Cob III to homocysteine, demethylated Cob I is oxidized to Cob II, which is reduced back to the active Cob III state by MS reductase with SAM as substrate.39 Therefore, decreased SAM that results from ethanol feeding could reduce the activity of MS reductase. In addition, others have demonstrated irreversible inhibition of purified rat liver MS activity by acetaldehyde, suggesting another protein modification due to covalent binding to this metabolite.40

Liver MAT is the product of MAT1A that encodes a dimer (MATIII) and a tetramer (MATI) that catalyze the synthesis of SAM from methionine by the addition of an adenosyl group from adenosine triphosphate.41 The present finding of reduced MATIII activity in the ethanol-fed micropigs (Table 4) is consistent with findings of others in the ethanol-fed rat model.5 On the other hand, lack of correlation of product SAM levels with MATIII activity emphasizes that SAM is regulated both by its synthesis by MAT and its subsequent metabolism by GNMT (Fig. 1). Inactivation of MATIII has been ascribed to covalent modification of the cysteine 121 residue in the setting of oxidant stress and is specifically related to a decrease in the GSH/GSSG ratio.42, 43

GNMT constitutes approximately 1% of all mammalian liver protein44 and plays a pivotal role in regulation of the SAM/SAH ratio by catalyzing the conversion of SAM to SAH by donation of a methyl group to glycine for production of N-methylglycine.27 In the present experiment, the role of GNMT in regulation of the conversion of SAM to SAH was reflected by negative correlations of GNMT activity with both SAM and the SAM/SAH ratio. Although SAH is a potent inhibitor of most methyl transfer reactions,30 GNMT is relatively resistant to inhibition because its Ki for SAH is very high (35 umol/L) compared with most other methyltransferases, such as that of DNA methyltransferase (1.4 umol/L).45 Because GNMT transcripts were decreased independently and additively by folate deficiency and ethanol, the finding of increased GNMT activity in the presence of ethanol feeding must represent a posttranscriptional modification of the enzyme. A likely explanation for increased GNMT activity in the ethanol-fed pigs is based on its known inhibition by avid binding to 5-MTHF.46 We can predict that hepatic 5-MTHF levels were decreased in the ethanol-fed pigs, because they are regulated by the activity of MTHFR, which was decreased in the same animals (Table 4). The predictable low levels of 5-MTHF would account for release of inhibition of GNMT and its increased activity in the ethanol-fed animals.

SAHH cleaves SAH to adenosine and homocysteine, but this reaction is readily reversible by the same enzyme28 (Fig. 1). The finding that both ethanol feeding and folate deficiency inhibited SAHH activity is novel and is consistent with the negative correlation of SAHH activity with both substrate SAH and product homocysteine. Under normal physiological conditions, the rapid metabolism of SAHH end products homocysteine and adenosine drives the SAHH reaction in the hydrolysis direction. On the other hand, the accumulation of homocysteine and adenosine promotes the reversibility of SAHH toward the synthesis of SAH, because the affinity and maximal velocity of the reaction are more favorable in this reverse direction.28, 47 Under these circumstances, experimental alteration in the SAHH protein with reduction of its activity would be expected to promote the accumulation of SAH by a block in the hydrolysis direction and homocysteine by a block in the reverse direction.

In conclusion, the present observations on the effects of ethanol feeding with or without folate deficiency on the transcripts and activities of hepatic transmethylation enzymes are consistent with and explain our previous findings in the same experimental animals of abnormal levels of methionine metabolites that were associated with accelerated liver injury.15 The significance of the findings from both studies can be understood in the context of recent discoveries of the role of homocysteine in the induction of fatty liver and apoptosis through the endoplasmic reticulum stress pathway.48, 49 GRP78, a marker of endoplasmic reticulum stress, and SREBP, which enhances lipid synthesis, were up-regulated in a genetic mouse model of hyperhomocysteinemia that developed fatty liver.48 Others found that fatty liver in this model was associated with increased hepatocellular apoptosis by terminal deoxynucleotidyl transferase-mediated nick end labeling (TUNEL) assay and increased caspase 12 levels.49 There is no direct evidence that SAH is more or less effective than homocysteine as a trigger of the endoplasmic reticulum stress pathway. In this context, the present findings on ethanol-induced changes in transmethylation reactions that resulted in increased levels of homocysteine and SAH assume importance for further understanding of the pathogenesis of alcoholic liver disease.


We are indebted to Jan Peerson for expert statistical consultation.