Biochemical and Molecular Action of Nutrients

Garlic and Garlic-Derived Compounds Inhibit HumanSqualene Monooxygenase1

Nisha Gupta and Todd D. Porter2

Division of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky,Lexington, KY 40536-0082

ABSTRACT Although extracts of garlic inhibit cholesterol biosynthesis in cultured hepatocytes, the inhibitorycomponents of garlic and the site or sites of inhibition in the cholesterol biosynthetic pathway have not beenestablished. To elucidate potential mechanisms of inhibition, we examined the effect of fresh garlic extract and 16water- or lipid-soluble compounds derived from garlic on purified recombinant human squalene monooxygenase.Squalene monooxygenase catalyzes the second and likely rate-limiting step in the downstream pathway forcholesterol biosynthesis. A 50% inhibitory concentration (IC50) of squalene epoxidation was achieved with 1 g/L offresh garlic extract; of the 16 garlic compounds tested, only selenocystine (IC50 5 65 mmol/L), S-allylcysteine (IC50

5 110 mmol/L), alliin (IC50 5 120 mmol/L), diallyl trisulfide (IC50 5 195 mmol/L), and diallyl disulfide (IC50 5 400mmol/L) substantially inhibited the enzyme. Kinetic analysis showed that the inhibition by garlic and by thesecompounds was slow and irreversible, suggestive of covalent binding to the enzyme; the ability of thiol-containingcompounds such as glutathione and 2,3-dimercaptopropanol to prevent and reverse the inhibition indicated thatthe garlic compounds were reacting with sulfhydryl groups on the protein. Dithiols were better reversal agents thanmonothiols, further suggesting that these inhibitors bind to the proposed vicinal sulfhydryls present on this enzyme.These results indicate that squalene monooxygenase may be one of the target enzymes through which garlicinhibits cholesterol biosynthesis. J. Nutr. 131: 1662–1667, 2001.

KEY WORDS: ● cholesterol biosynthesis ● garlic ● selenium ● selenocystine ● squalene epoxidase

Garlic has been reputed to possess medicinal propertiessince ancient times. More recently, attention has been focusedon garlic’s ability to decrease blood cholesterol levels, as dem-onstrated in clinical trials in humans (1,2), through meta-analyses of clinical studies (3–5) as well as in experimentalstudies with animals (6–8) and cultured hepatocytes (9–11).Efforts have been made to isolate and identify the activechemical components of garlic; S-allylcysteine (SAC),3 diallyldisulfide (DADS), and allicin and its derivatives (ajoene) allhave been shown to contribute to its hypocholesterolemicactions (11–13). However, the mechanism underlying theinhibitory action of these garlic compounds has not beenestablished. Several studies have reported that feeding garlic-supplemented diets to animals decreased the activity of severalcholesterolgenic enzymes, including 3-hydroxy-3-methylglu-taryl (HMG)-CoA reductase (8,14–15) and acetyl-CoA syn-thetase (12), and it is possible that additional enzymes in thecholesterol synthesis pathway are also inhibited.

Squalene monooxygenase (EC 1.14.99.7, earlier called

squalene epoxidase) is a 64-kDa FAD-containing enzyme boundto the endoplasmic reticulum of eukaryotic cells. The enzymecatalyzes the first oxidative step in cholesterol biosynthesis, theepoxidation of squalene across a C5C double bond to yield2,3-oxidosqualene, a reaction more typical of cytochrome P450-type chemistry. Like the cytochromes P450, this flavoproteinmonooxygenase also is dependent upon NADPH-cytochromeP450 reductase (CPR) for reducing equivalents. Squalene mono-oxygenase evidently plays an important role in the overall regu-lation of cholesterol biosynthesis in that the addition of choles-terol to cells in culture lowers squalene monooxygenase mRNAlevels, suppresses squalene monooxygenase activity and results inthe accumulation of squalene (16).

Garlic can contain high levels of tellurium and seleniumcompounds (17–20); the possibility that these compoundscontribute to the overall block in cholesterol synthesis byinhibiting squalene monooxygenase has been proposed (21)but not yet examined. Indeed, inhibition of this enzyme bytellurium was shown to block cholesterol synthesis inSchwann cells and led to a peripheral demyelinating neurop-athy (22). In previous work from our laboratory, tellurium,selenium and arsenic compounds were shown to inhibit hu-man squalene monooxygenase by binding to vicinal cysteinesulfhydryls on this enzyme (23,24). It is therefore plausiblethat tellurium and selenium compounds in garlic may inhibitsqualene monooxygenase and thereby contribute to the hypo-cholesterolemic actions of this plant.

1 Supported by a grant from the Ohio Valley Affiliate of the American HeartAssociation.

2 To whom correspondence should be addressed.E-mail: tporter@pop.uky.edu.

3 Abbreviations used: CPR, cytochrome P450 reductase; DADS, diallyl disul-fide; DATS, diallyl trisulfide; DMP, 2,3-dimercaptopropanol; GSH, glutathione;HMG, 3-hydroxy-3-methylglutaryl; IC50, 50% inhibitory concentration; SAC,S-allylcysteine; SC, selenocystine.

0022-3166/01 $3.00 © 2001 American Society for Nutritional Sciences.Manuscript received 7 December 2000. Initial review completed 11 January 2001. Revision accepted 6 March 2001.

1662

by on October 19, 2010

jn.nutrition.orgD

ownloaded from

To address this possibility, fresh garlic extract and 16 garlic-derived compounds (Fig. 1) were tested for their ability toinhibit purified human squalene monooxygenase. This workidentifies five chemicals found in garlic as inhibitors ofsqualene monooxygenase and suggests that these compoundsbind to vicinal cysteine sulfhydryls on the enzyme. Althoughthere are several reports that selenium-containing compoundsfrom garlic are effective in the prevention of mammary andother types of cancer (18,19,25), this is the first report to showthat an organoselenium compound, selenocystine, will inhibitan enzyme involved in cholesterol synthesis.

MATERIALS AND METHODS

Materials. Garlic bulbs were purchased from a local grocery.Alliin, S-methylcysteine, S-ethylcysteine, S-propylcysteine and SACwere graciously provided by Wakunaga of America (Mission Viejo,CA). DADS and diallyl sulfide were purchased from Fluka Chemika

(St. Louis, MO); dipropyl disulfide, dipropyl sulfide, allyl mercaptan,allyl methyl sulfide and selenomethionine were purchased from AcrosOrganics (Geel, Belgium); methallyl sulfide was purchased from Al-drich (Milwaukee, WI); diallyl trisulfide (DATS) was purchased fromICN Biomedicals; glutathione (GSH) was purchased from BoehringerMannheim (Indianapolis, IN), GmbH; Se-(methyl)selenocysteine,L-selenocystine (SC), nicotinamide adenine dinucleotide phosphate,reduced form (b-NADPH), FAD, 2,3-dimercaptopropanol (DMP)and cytochrome c were purchased from Sigma Chemical, St. Louis,MO. Radiolabeled squalene (2.6 3 10 Bq) was synthesized by theChemical Synthesis Facility, Department of Medicinal Chemistry,University of Utah. Precoated silica TLC plates were obtained fromWhatman (Clifton, NJ). All garlic derivatives were stored at 4°C andstocks of the water-soluble compounds were made fresh every timebefore use. Lipid-soluble garlic derivatives were dissolved in dimethylsulfoxide to a final concentration of 200 mmol/L.

Preparation of garlic extracts. Garlic cloves were peeled andhomogenized with a small amount of quartz sand in 20 mmol/LTris-HCl buffer (pH 7.4), and the debris was removed by centrifuga-tion at 21,000 3 g for 20 min at 25°C. The clear supernatant wasdivided into aliquots and stored at 280°C.

Purification of squalene monooxygenase. Human recombinantsqualene monooxygenase expressed from the pTYB4 vector was pu-rified according to the protocol described by New England Biolabs(Beverly, MA) for expression of proteins with the IMPACT T7system (Intein-chitin binding domain fusion proteins), as previouslydescribed (23). Protein was quantified with the Coomassie Plus Pro-tein Assay Reagent Kit from Pierce (Rockford, IL) and was thenstored at 280°C until use.

Squalene monooxygenase assays. Squalene epoxidation assayswere carried out as described previously (23). Standard reactionmixtures (200 mL) containing 20 mmol/L Tris-HCl (pH 7.4), 0.1%Triton X-100, 30 mmol/L FAD, 28 pmol CPR, 40 mmol/L 14C-squalene and 58 pmol squalene monooxygenase were preincubatedwith or without inhibitor for 30 min at 37°C. The reactions werethen started by the addition of 1 mmol/L NADPH and incubated foranother 30 min at 37°C. Product formation was linear for 40–60 minat 37°C, and yielded 1.1 nmol/L of 2,3-oxidosqualene/30-min reac-tion. Reactions were stopped by extraction into methylene chlorideand then fractionated on TLC plates with 5% ethyl acetate inhexane. The developed plates were quantified by electronic autora-diography (Packard Instant Imager; Wilmington, DE).

Cytochrome P450 reductase assays. Recombinant rat CPR waspurified by affinity chromatography as described (26) and quantifiedspectrally using a millimolar extinction coefficient of 21.4 at 456 nm.Activity was determined with cytochrome c in 100 mmol/L potassiumphosphate buffer (pH 7.7) with 50 mmol/L cytochrome c and 5 pmolof reductase; reactions were initiated by the addition of 50 mmol/LNADPH and followed at 550 nm for 1 min at room temperature.Preincubations were carried out with garlic extract or garlic com-pounds for 30 min at 37°C.

Statistical analysis. Statistical analyses were carried out by one-way ANOVA with Dunnett’s or Bonferrroni’s multiple comparisonspost-hoc tests, as indicated in the appropriate figure legends. Differ-ences were considered significant at P , 0.05. All graphs and statis-tical comparisons were prepared on Prism 3.0 (GraphPad Software,San Diego, CA) by using values in duplicate or greater with SEM, asshown by error bars where appropriate.

RESULTS

Fresh garlic extract and 16 different water- or lipid-solublecompounds present in garlic (Fig. 1) were tested for theirability to inhibit human squalene monooxygenase. A 50%inhibitory concentration (IC50) was achieved using 1 g/Lgarlic extract and activity was completely inhibited at 150 g/L(Fig. 2A). Of the eight water-soluble compounds, SC, SACand alliin were the most potent inhibitors, with IC50 values of65, 110 and 120 mmol/L, respectively (Fig. 2B). Inhibition bySC was nearly complete at 500 mmol/L, whereas inhibition bySAC and alliin was still incomplete. S-methylcysteine, S-

FIGURE 1 The 16 water- or lipid-soluble compounds from garlictested in this study for inhibition of squalene monooxygenase. Thosecompounds that were found to inhibit the enzyme are underlined.

INHIBITION OF SQUALENE MONOOXYGENASE BY GARLIC 1663

by on October 19, 2010

jn.nutrition.orgD

ownloaded from

ethylcysteine, S-propylcysteine, selenomethionine and Se-(methyl)selenocysteine were not inhibitory at concentrationsup to 1 mmol/L (data not shown). Of the eight lipid-solublecompounds tested, only DATS and DADS showed significantinhibitory potency, with IC50 values of 195 and 400 mmol/L,respectively; at DADS concentrations as high as 1 mmol/L,squalene monooxygenase maintained ;25% of maximal ac-tivity (data not shown). The six other lipid-soluble com-pounds, diallyl sulfide, methallyl sulfide, dipropyl disulfide,dipropyl sulfide, allyl mercaptan and allyl methyl sulfide werenot inhibitory at concentrations up to 1 mmol/L (data notshown).

NADPH-CPR is the obligate electron transfer partner forsqualene monooxygenase. To determine whether inhibition ofCPR by garlic and its derivatives contributes to the inhibitionof squalene epoxidation, reductase activity was determined inthe presence of garlic extract and four of the inhibitory com-pounds identified above. A 5 g/L concentration of garlic ex-tract inhibited .90% of squalene epoxidation and 70% ofcytochrome c reduction by CPR (Fig. 3). Similarly, a 100mmol/L concentration of each of the four garlic compoundsyielded similar levels of inhibition of squalene epoxidation(Fig. 2) and cytochrome c reduction (Fig. 3). However, be-cause the reductase concentration exceeds saturation in thesqualene monooxygenase reactions, the effect of a 70% de-crease in reductase concentration, comparable to the extent ofinhibition of cytochrome c reduction by garlic extract, wasdetermined. A 70% decrease in reductase concentration re-duced squalene monooxygenation by only 10–15%, indicatingthat electron transfer from the reductase is not rate limiting in

the oxygenation of squalene, a finding consistent with earlierstudies (23). Thus, it is unlikely that the inhibition of CPR bygarlic or its components in this in vitro assay contributessignificantly to the decrease in squalene monooxygenase ac-tivity.

To determine whether the garlic compounds were interfer-ing with the binding of squalene or FAD, the extent ofinhibition by a fixed concentration of each garlic compoundwas determined in the presence of several concentrations ofsqualene or FAD. Doubling the concentration of squalene to80 mmol/L decreased the extent of inhibition by the garliccompounds by 20–60%, whereas changing the FAD concen-tration had no effect on inhibition by these agents (data notshown). The protection afforded by squalene suggests thatthese compounds bind in the substrate-binding pocket of theenzyme and supports the conclusion that inhibition of CPRdoes not contribute significantly to the inhibition of squalenemonooxygenation.

Garlic extract, SC, SAC, alliin and DADS were slow,time-dependent inhibitors with t1/2 values (half-life forinactivation) of 13, 22, 36, 41 and 91 min, respectively(Fig. 4). These results suggest that these inhibitors bindirreversibly to squalene monooxygenase. To address thispossibility, enzyme-inhibitor dilution experiments were car-ried out as described previously (24). In these experiments,enzyme was preincubated with either garlic extract or one ofits inhibitory components for 30 min and then diluted12-fold before the addition of 1 mmol/L NADPH to startthe reaction. No recovery of activity was seen after dilutionof garlic or the four garlic derivative-containing incuba-tions, indicating that dilution did not release these com-pounds from the enzyme and that these inhibitors bindirreversibly to squalene monooxygenase (Fig. 5).

Thiol-containing compounds were tested for their ability toprotect against and reverse the inhibition of squalene mono-oxygenase by garlic and by four of the inhibitory compounds.GSH, a monothiol, or DMP, a dithiol, was added before orafter the 30-min preincubation. Addition of either thiol re-agent to the preincubations yielded significant protection frominhibition by garlic and the garlic derivatives, indicating thatthe garlic components were reacting with sulfhydryls onsqualene monooxygenase (Fig. 6A). Inhibition by garlic and

FIGURE 3 Effect of garlic and garlic-derived compounds on cy-tochrome c reduction by cytochrome P450 reductase (CPR). Standardreductase assay components were preincubated for 30 min at 37°Cwith 5 g/L of garlic extract or 100 mmol/L concentrations of the fourgarlic compounds, S-allylcysteine (SAC), alliin, selenocystine (SC) anddiallyl disulfide (DADS), followed by the addition of NADPH to start thereaction. The 30% CPR bar shows the effect of a 70% decrease in CPRconcentration on squalene monooxygenase activity. Data are ex-pressed as means 6 SEM, n 5 4 independent experiments. Inhibition bygarlic extract and by the garlic compounds was significantly greaterthan that produced by a 70% reduction in CPR (P , 0.05, one-wayANOVA with Dunnett’s multiple comparison test).

FIGURE 2 Effect of garlic and its derivatives on squalene mono-oxygenase activity. Standard squalene monooxygenase assay compo-nents were preincubated for 30 min at 37°C with increasing concen-trations of (A) aqueous garlic extract as a final concentration of its freshweight; (B) garlic-derived compounds, S-allylcysteine (SAC), alliin, di-allyl disulfide (DADS), selenocystine (SC) or diallyl trisulfide (DATS).Reactions were started by the addition of NADPH (1 mmol/L finalconcentration) and incubated for 30 min. The dashed line indicatesactivity in the absence of inhibitor (1.1 nmol/L 2,3-oxidosqualene/30min reaction). Data are expressed as means 6 SEM, n 5 4 independentexperiments.

GUPTA AND PORTER1664

by on October 19, 2010

jn.nutrition.orgD

ownloaded from

the garlic compounds could also be partially reversed by thethiol reagents (Fig. 6B). DMP was a significantly more effec-tive reversal agent than GSH, an effect most noticeable withDADS and SC. This difference between the dithiol and themonothiol was significant for three of the five substancestested, i.e., alliin, DADS and SC. Similar results were obtainedwith dithiothreitol and b-mercaptoethanol, confirming thatdithiols were more effective reversal agents (data not shown).These results are similar to those obtained previously in studieson the inhibition of squalene monooxygenase by tellurite (24)and are consistent with the garlic compounds reacting withvicinal cysteine sufhydryls on squalene monooxygenase.

DISCUSSION

Experimental data from animal studies (7,14), clinical trials(1,2), meta-analyses (3–5) and cell culture studies (9–11)

have supported the postulate that garlic extracts and garlic-derived compounds can lower cholesterol levels. Althoughseveral studies have indicated that one site of inhibition in thecholesterol biosynthetic pathway is HMG-CoA-reductase(14,15), additional downstream enzymes have also been sug-gested as targets for garlic and its components (12). In thiswork, we report the inhibition of purified recombinant humansqualene monooxygenase, a downstream enzyme in the cho-lesterol biosynthetic pathway, by an aqueous garlic extract andby four organosulfur compounds and one organoselenium com-pound found in garlic. All four of the organosulfur compounds,SAC, alliin, DATS and DADS, contain an S-allyl moiety.Structurally analogous compounds that lack the allyl group,such as S-propylcysteine and dipropyl disulfide, were not in-hibitory, suggesting that the allyl group enhances the reactiv-ity of these compounds. However, an S-allyl group is notsufficient for reactivity because diallyl sulfide and methallyl

FIGURE 4 Slow, time-dependent inhibition of squalene monoox-ygenase by garlic and garlic-derived compounds. Assay mixtures con-taining NADPH but without squalene monooxygenase were preincu-bated for 30 min at 37°C in the presence of 5 g/L of garlic extract or 100mmol/L of S-allylcysteine (SAC), alliin, selenocystine (SC) or diallyldisulfide (DADS). The reactions were started by the addition ofsqualene monooxygenase and incubated for the indicated times at37°C. Data are expressed as means 6 SEM , n 5 3 independentexperiments. Half-life (t1/2) values of inactivation for garlic, SC, SAC,alliin and DADS were 13, 22, 36, 41 and 91 min, respectively.

FIGURE 5 Irreversible inhibition of human squalene monooxygen-ase by garlic and its derivatives, S-allylcysteine (SAC), alliin, selenocys-tine (SC) and diallyl disulfide (DADS). Assay mixtures were preincu-bated for 30 min at 37°C in a volume of 17 mL containing either freshgarlic extract or garlic derivative. After preincubation, the volume wasincreased to 200 mL and inhibitor concentration was held constant at 5g/L of garlic extract or 0.1 mmol/L garlic derivative (constant concen-tration), or diluted from 60 to 5 g/L of garlic extract, or from 1.2 to 0.1mmol/L garlic derivative (diluted concentration). The concentration ofall other reaction components was held constant and NADPH wasadded to start the reaction. Reactions were carried out for 30 min at37°C. Data are expressed as means 6 SEM, n 5 3 independent exper-iments.

FIGURE 6 Effect of sulfhydryls on inhibition of squalene mono-oxygenase by garlic and its derivatives. (A) Protection experiments:assay mixtures were preincubated at 37°C with 5 g/L of garlic extract,65 mmol/L selenocystine (SC), 110 mmol/L S-allylcysteine (SAC), 120mmol/L alliin or 400 mmol/L diallyl disulfide (DADS) in the absence orpresence of 1 mmol/L of either 2,3-dimercaptopropanol (DMP) or glu-tathione (GSH). After 30 min, the reactions were started by the additionof NADPH and incubated for an additional 30 min. Both thiol treatmentsafforded significant protection from the garlic compounds (P , 0.05,one-way ANOVA with Dunnett’s multiple comparisons test). (B) Rever-sal experiments: after a 30-min preincubation with garlic extract orderivative, 1 mmol/L DMP or GSH was added along with NADPH tostart the reactions. Asterisks indicate the points at which reversal byDMP was significantly greater than that produced by GSH (P , 0.05,one-way ANOVA with Bonferroni’s multiple comparison test). No sig-nificant reversal by either thiol compound was obtained with the garlicextract, and GSH did not reverse the inhibition by DADS or SC. Theformation of 2,3-oxidosqualene in the presence of DMP but no inhibitorwas 1.76 nmol/L over 30 min; in the presence of GSH and no inhibitor,it was 1.38 nmol/L. Data are expressed as means 6 SEM, n 5 3independent experiments.

INHIBITION OF SQUALENE MONOOXYGENASE BY GARLIC 1665

by on October 19, 2010

jn.nutrition.orgD

ownloaded from

sulfide were not inhibitory; both contain an S-allyl group, but,in contrast to DADS, these compounds contain only a singlesulfur atom. It should be noted that alliin and SAC, both ofwhich contain an S-allyl group, also contain only a singlesulfur atom; evidently the peptide moiety of these compoundspromotes their reactivity. Although the thiol protection ex-periments indicate that these compounds react with sulfhy-dryls on squalene monooxygenase, the reactive groups onthese inhibitors and the products of the reaction remain to bedetermined.

Of the four inhibitory organosulfur compounds, three havebeen shown to be potent inhibitors of cholesterol synthesis incultured hepatocytes (SAC, DATS and DADS), whereas al-liin was ineffective (11,27). Similarly, Liu and Yeh (11) foundthat S-ethylcysteine and S-propylcysteine were equipotentwith SAC in decreasing cholesterol synthesis in hepatocytes,but these compounds are not inhibitors of squalene monoox-ygenase. Thus, there is not an exact correspondence betweeninhibitors of cholesterol synthesis in hepatocytes and inhibi-tors of squalene monooxygenase, and it may be that thesecompounds can act at different steps in the cholesterol bio-synthetic pathway, as has been suggested (27).

It is difficult to compare our studies with isolated enzyme toin vivo studies on the reduction in blood cholesterol by garlicconsumption because the amount of garlic consumed and itsform (raw, cooked or as a garlic pill supplement) differ greatlyamong studies. Most importantly, the concentration of garliccompounds in the blood, and, more importantly, in the liver,is not known. Study participants typically consume 1–3 g ofgarlic per day (5,28,29), which presumably yields reasonablelevels of organosulfur compounds in the portal bloodstreamand contributes partially or wholly to the established ability ofgarlic to decrease blood cholesterol levels. Because these or-ganosulfur compounds are irreversible inhibitors of squalenemonooxygenase, it is conceivable, but not yet shown, thateven relatively low levels of these compounds in the bloodmight be sufficient to inhibit squalene monooxygenase activityand thereby decrease cholesterol synthesis.

Studies in vitro with hepatocyte culture are more amenableto comparison. Inhibition of cholesterol synthesis in culturedprimary rat hepatocytes by SAC was evident at 100 mmol/L(11), a concentration that yielded 50% inhibition of squalenemonooxygenase in the present study. Similarly, DATS andDADS were effective inhibitors of cholesterol synthesis atconcentrations of 100–200 mmol/L (10), similar to their IC50values for inhibition of squalene monooxygenase, althoughhigher concentrations were associated with significant cyto-toxicity (11). Thus, inhibitory concentrations of several ofthese garlic compounds with isolated hepatocytes correspondto concentrations that inhibit purified squalene monooxygen-ase. Whether inhibition of squalene monooxygenase contrib-utes to the inhibition of cholesterol synthesis by garlic in cellculture and in vivo remains to be determined.

Garlic has been reported to contain high levels of seleniumand tellurium (17,18), especially if grown in soils containingthese elements, and it has been proposed that tellurium- andselenium-containing compounds present in garlic contributeto the block in cholesterol synthesis by inhibiting squalenemonooxygenase (21). In previous work from our laboratory, itwas shown that micromolar concentrations of selenite andselenium dioxide inhibit purified human squalene monooxy-genase (23). Here, we showed that SC, an organoseleniumcompound present in garlic (19), inhibits squalene monooxy-genase with an IC50 value of 65 mmol/L. Se-(methyl)seleno-cysteine, selenomethionine or cystine were not inhibitory,indicating that the diseleno-bond was necessary for inhibition.

The present work implicates the involvement of cysteinesulfhydryls in the inhibition of squalene monooxygenase bygarlic and garlic-derived compounds. In previous work fromthis laboratory, evidence was presented that the inhibition ofsqualene monooxygenase by tellurium compounds results frombinding to critical sulfhydryl groups on the enzyme (24).Similarly, N-ethylmaleimide, a sulfhydryl-specific reagent, wasshown to inhibit squalene monooxygenase in rat liver micro-somes (22). Thus, squalene monooxygenase is highly sensitiveto inhibition by chemicals that react with sulfhydryls. Theinhibition of squalene monooxygenase by garlic and garlic-derived compounds could be prevented by preincubation witheither GSH or DMP, indicating that these inhibitors reactwith enzyme thiol groups. Addition of GSH or DMP after thepreincubation also partially reversed the inhibition, furthersupporting the thesis that cysteine sulfhydryls on squalenemonooxygenase are the target of binding by these compounds.The present work further suggests that there are differencesbetween the ability of monothiols and dithiols to reverse theinhibition of squalene monooxygenase by garlic compounds.DMP and dithiothreitol (dithiols) were more effective thanGSH or b-mercaptoethanol (monothiols) in reversing theinhibition, suggesting that the garlic compounds were reactingwith vicinal cysteines on the enzyme. The presence of vicinalcysteines was proposed to explain the unusual sensitivity ofsqualene monooxygenase to tellurium compounds (24). Itshould be noted that heating the garlic extract or the fourgarlic derivatives in a boiling water bath for 2 min completelyinactivated these compounds and prevented the inhibition,indicating that they are relatively labile. Several other studieshave noted that raw garlic was significantly more effectivethan cooked garlic in blocking cyclooxygenase activity in theprevention of thrombosis (30,31).

Although we have identified five garlic-derived chemicalsthat can inhibit squalene monooxygenase, we do not knowwhether these are the only inhibitory chemicals in garlicextract, or whether they mediate the inhibition of squalenemonooxygenase by garlic extract. Garlic contains a variety oforganosulfur and organoselenium compounds, with SAC andalliin as the most prominent. SAC appears to be a precursor toalliin; when garlic is crushed, alliinase is released to cleavealliin to yield sulfenic acid (CH25CHCH2SOH), which con-denses to form allicin. Allicin is also highly unstable, andrapidly degrades to a variety of organosulfur compounds, in-cluding DADS, DATS and ajoene. These compounds give riseto allyl mercaptan and, secondarily, allyl methyl sulfide; allylmercaptan has been suggested to be the ultimate active chem-ical species derived from garlic consumption (10). Althoughallyl mercaptan is not inhibitory to squalene monooxygenase,several of the precursors and intermediates leading to thiscompound are inhibitory; most notably, SAC is one of themore effective inhibitors found in this study. SAC is consid-ered to be a principal active ingredient in commercial garlicpreparations; it is stable and readily absorbed from the gastro-intestinal tract (32).

The present work demonstrates that the inhibition ofsqualene monooxygenase by garlic is slow, irreversible andlikely to involve binding to sulfhydryls in the squalene bindingsite of the enzyme. Garlic extract, as well as four of thegarlic-derived components, inhibited squalene epoxidation ina time- and concentration-dependent manner; on the basis ofthe protection and reversal by monothiols and dithiols, it islikely that these inhibitors react with the proposed vicinalcysteine sulfhydryls on squalene monooxygenase. The inhibi-tion by the organoselenium compound, SC, is of particularinterest because overconsumption of selenium by pigs and

GUPTA AND PORTER1666

by on October 19, 2010

jn.nutrition.orgD

ownloaded from

cattle is associated with a variety of pathologies suggestive ofdisruption of lipid metabolism in neural tissue (33). The effectof garlic and its components on the intracellular flux of cho-lesterol intermediates in vivo is the subject of current inves-tigations.

ACKNOWLEDGMENT

We thank Brian Laden for advice and assistance.

LITERATURE CITED

1. Adler, A. J. & Holub, B. J. (1997) Effect of garlic and fish-oil supple-mentation on serum lipid and lipoprotein concentrations in hypercholesterolemicmen. Am. J. Nutr. 65: 445–450.

2. Bordia, A., Verma, S. K. & Srivastava, K. C. (1998) Effect of garlic(Allium sativum) on blood lipids, blood sugar, fibrinogen and fibrinolytic activity inpatients with coronary artery disease. Prostaglandins Leukot. Essent. Fatty Acids58: 257–263.

3. Warshafsky, S., Kamer, R. S. & Sivak, S. L. (1993) Effect of garlic ontotal serum cholesterol. A meta-analysis. Ann. Intern. Med. 119: 599–605.

4. Silagy, C. & Neil, A. (1994) Garlic as a lipid lowering agent—a meta-analysis. J. R. Coll. Physicians Lond. 28: 39–45.

5. Stevinson, C., Pittler, M. H. & Ernst, E. (2000) Garlic for treatinghypercholesterolemia. Ann. Intern. Med. 133: 420–429.

6. Aouadi, R., Aouidet, A., Elkadhi, A., Ben Rayana, M. C., Jaafoura, H.,Tritar, B. & Nagati, K. (2000) Effect of fresh garlic (Allium sativum) on lipidmetabolism in male rats. Nutr. Res. 20: 273–280.

7. Chi, M. S., Koh, E. T. & Stewart, T. J. (1982) Effect of garlic on lipidmetabolism in rats fed cholesterol or lard. J. Nutr. 112: 241–248.

8. Qureshi, A. A., Din, Z. Z., Abuirmeileh, N., Burger, W. C., Ahmad, Y. &Elson, C. E. (1983) Suppression of avian hepatic lipid metabolism by solventextracts of garlic: impact on serum lipids. J. Nutr. 113: 1746–1755.

9. Gebhardt, R. (1993) Multiple inhibitory effects of garlic extracts oncholesterol biosynthesis in hepatocytes. Lipids 28: 613–619.

10. Cho, B.H.S. & Xu, S. (2000) Effects of allyl mercaptan and variousallium-derived compounds on cholesterol synthesis and secretion in Hep-G2cells. Comp. Biochem. Physiol. C 126: 195–201.

11. Liu, L. & Yeh, Y.-Y. (2000) Inhibition of cholesterol biosynthesis byorganosulfur compounds derived from garlic. Lipids 35: 197–203.

12. Focke, M., Feld, A. & Lichtenthaler, K. (1990) Allicin, a naturally oc-curring antibiotic from garlic, specifically inhibits acetyl-CoA synthetase. FEBSLett. 261: 106–108.

13. Sendl, A., Schliack, M., Loser, R., Stanislaus, F. & Wagner, H. (1992)Inhibition of cholesterol synthesis in vitro by extracts and isolated compoundsprepared from garlic and wild garlic. Atherosclerosis 94: 79–85.

14. Quereshi, A. A., Abuirmeileh, N., Din, Z. Z., Elson, C. E. & Burger, W. C.(1983) Inhibition of cholesterol and fatty acid biosynthesis in liver enzymes andchicken hepatocytes by polar fractions of garlic. Lipids 18: 343–348.

15. Qureshi, A. A., Crenshaw, T. D., Abuirmeileh, N., Peterson, D. M. & Elson,

C. E. (1987) Influence of minor plant constituents on porcine hepatic lipidmetabolism. Impact on serum lipids. Atherosclerosis 64: 109–115.

16. Hidaka, Y., Satoh, T. & Kamei, T. (1990) Regulation of squaleneepoxidase in HepG2 cells. J. Lipid Res. 31: 2087–2094.

17. Schroeder, H. A., Buckman, J. & Balassa, J. J. (1967) Abnormal traceelements in man: tellurium. J. Chronic Dis. 20: 147–161.

18. Ip, C., Lisk, D. J. & Stoewsand, G. S. (1992) Mammary cancer pre-vention by regular garlic and selenium-enriched garlic. Nutr. Cancer 17: 279–286.

19. Ip, C., Birringer, M., Block, E., Kotrebai, M., Tyson, J. F., Uden, P. C. &Lisk, D. J. (2000) Comparative speciation influences comparative activity ofselenium-enriched garlic and yeast in mammary cancer prevention. J. Agric. FoodChem. 48: 2062–2070.

20. McSheehy, S., Yang, W., Pannier, F., Szpunar, J., Lobinski, R., Auger, J.& Potin-Gautier, M. (2000) Speciation analysis of selenium in garlic by two-dimensional high-performance liquid chromatography with parallel inductivelycoupled plasma mass spectrometric and electrospray tandem mass spectromet-ric detection. Anal. Chim. Acta 421: 147–153.

21. Larner, A. J. (1995) How does garlic exerts its hypocholesterolaemicaction? The tellurium hypothesis. Med. Hypotheses 44: 295–297.

22. Wagner, M., Toews, A. D. & Morell, P. (1995) Tellurite specificallyaffects squalene epoxidase: investigations examining the mechanism of tellu-rium-induced neuropathy. J. Neurochem. 64: 2169–2176.

23. Laden, B. P., Tang, Y. & Porter, T. D. (2000) Cloning, heterologousexpression, and enzymological characterization of human squalene monooxy-genase. Arch. Biochem. Biophys. 374: 381–388.

24. Laden, B. P. & Porter, T. D. (2001) Inhibition of squalene monooxy-genase by tellurium compounds: evidence of interaction with vicinal sulfhydryls.J. Lipid Res. 42: 235–240.

25. Ip, C. & Lisk, D. J. (1995) Efficacy of cancer prevention by highselenium-garlic is primarily dependent on the action of selenium. Carcinogenesis16: 2649–2652.

26. Shen, A. L., Porter, T. D., Wilson, T. E. & Kasper, C. B. (1989) Struc-tural analysis of the FMN binding domain of NADPH-cytochrome P-450 oxi-doreductase by site-directed mutagenesis. J. Biol. Chem. 264: 7584–7589.

27. Gebhardt, R. & Beck, H. (1996) Differential inhibitory effects of garlic-derived organosulfur compounds on cholesterol biosynthesis in primary rat he-patocyte cultures. Lipids 31: 1269–1276.

28. Ali, M. & Thomson, M. (1995) Consumption of garlic clove a day couldbe beneficial in preventing thrombosis. Prostaglandins Leukot. Essent. FattyAcids 53: 211–212.

29. Vorberg, G. & Schneider, B. (1990) Therapy with garlic: results of aplacebo-controlled, double-blind study. Br. J. Clin. Pract. Sym. Suppl. 69: 7–11.

30. Ali, M. (1995) Mechanism by which garlic (Allium sativum) inhibitscyclooxygenase activity. Effect of raw versus boiled garlic extract on the synthe-sis of prostanoids. Prostaglandins Leukot. Essent. Fatty Acids 53: 397–400.

31. Ali, M., Bordia, T. & Mustafa, T. (1999) Effect of raw versus boiledaqueous extract of garlic and onion on platelet aggregation. ProstaglandinsLeukot. Essent. Fatty Acids 60: 43–47.

32. Nagae, S., Ushijima, M., Hatono, S., Imai, J., Kasuga, S., Matsuura, H.,Itakura, Y. & Higashi, Y. (1994) Pharmacokinetics of the garlic compoundS-allylcysteine. Planta Med. 60: 214–217.

33. Davidson-York, D., Galey, G. D., Blanchard, P. & Gardner, I. A. (1999)Selenium elimination in pigs after an outbreak of selenium toxicosis. J. Vet. Diagn.Investig. 11: 352–357.

INHIBITION OF SQUALENE MONOOXYGENASE BY GARLIC 1667

by on October 19, 2010

jn.nutrition.orgD

ownloaded from