Biotechnol. Appl. Biochem. (2008) 51, 73–78 (Printed in Great Britain) doi:10.1042/BA20070255 73

The antibacterial efficacy of an aceraceous plant [Shantungmaple (Acer truncatum Bunge)] may be related to inhibition ofbacterial β-oxoacyl-acyl carrier protein reductase (FabG)

Feng Zhang*, Shi-Yun Luo†, Yan-Bin Ye‡, Wen-Hua Zhao*, Xu-Guang Sun†,Zhi-Qun Wang†, Ran Li†, Ying-Hui Sun§, Wei-Xi Tian§ and Ying-Xia Zhang*¶1

*Department of Chemical Biology, School of Chemical Biology and Pharmaceutical Sciences, Capital Medical University, Beijing100069, People’s Republic of China, †Department of Ocular Microbiology, Beijing Institute of Ophthalmology, Beijing Tong RenHospital, Capital Medical University, Beijing 100005, People’s Republic of China, ‡Experimental Centre, School of ChemicalBiology and Pharmaceutical Sciences, Capital Medical University, Beijing 100069, People’s Republic of China, §Departmentof Biology, Graduate School of Chinese Academy of Sciences, Beijing 100049, People’s Republic of China, and ¶Municipal KeyLaboratory of Beijing for Regulation of Liver Protection and Regeneration, Capital Medical University, Beijing 100069, People’sRepublic of China

Polyphenols, including flavonoids, are the major com-ponents of the extracts from aceraceous plants. Theypossess remarkable antibacterial and antitumouractivity. Our study focused on whether the inhibitionof the bacterial type II fatty acid synthesis system is themechanism for the antibacterial effect of the relatedplant polyphenols. Extracts obtained from the fallenleaves of the Shantung maple (Acer truncatum Bunge)using different solvents, and the related pure compoundPGG (1,2,3,4,6-penta-O-galloyl-β-D-glucose), potentlyinhibited the FabG (β-oxoacyl-ACP reductase) stepsin the fatty-acid-elongation cycle with the IC50 valuesbetween 0.9 and 7.2 μg/ml. An ethyl acetate extractappeared to inhibit FabG reductase in a mixed mannerwith NADPH, as did PGG with NADPH, demonstratingthat they interfered with the binding of the cofactor tothe enzyme. Gram-positive and Gram-negative bac-teria and some fungi were used to evaluate the anti-bacterial abilities of different extract samples. Theexperiments showed that a higher polyphenol contentof the extracts led to a more potent inhibitory capacityagainst FabG, thus enhancing the antibacterial efficacy.

Introduction

Fatty acid biosynthesis, the first stage in membrane lipid bio-genesis, is catalysed in most bacteria by a series of small, sol-uble proteins that are separate enzymes catalysing sequentialreactions. This arrangement is termed the ‘FAS II’ (type IIfatty acid synthase) system and contrasts noticeably with thetype I FAS system of eukaryotes, which comprises a dimerof a single large multifunctional polypeptide. Fatty acids areessential for bacterial growth and viability. Thus the bacterialpathway offers several unique sites for selective inhibition

by chemotherapeutic agents. The well known antimicrobialagent triclosan has been revealed to be FabI, the enoyl-ACP(acyl carrier protein) reductase of the FAS II inhibitor [1].

As bacterial resistance to existing antibiotics hasincreased at an alarming rate, posing a serious threat toglobal public health, discovery of new antibacterial agentsis becoming an urgent medical need. Recently, FabH [β-oxoacyl-ACP synthase III; also known as KAS III (3-ketoacyl-ACP synthase)] has formed the focus of an antibacterialtarget, arousing the interest of many researchers. FabHis essential for the initiation of fatty acid synthesis [2]and catalyses a regulated step thought to play a key rolein determining the amount of fatty acids produced bythe pathway [3,4]. These properties have led FabH to betargeted for drug development by several groups. The mostextensively evaluated case is the development of compoundsthat act against FabH [5]. Co-crystal structures of inhibitorswith FabH aided in improving the design [6,7]. In the bacterialpathway, FabG (β-oxoacyl-ACP reductase) catalyses theNADPH-dependent reduction of the β-oxoacyl-ACPintermediate to β-hydroxyacyl-ACP in the elongation cyclesof the FAS II system. FabG is highly conserved and widely ex-pressed, and only a single isoform is known in bacteria. Thesefeatures suggest that drugs targeting this enzyme would havebroad-spectrum antibacterial activity. Despite the growing

Key words: antibacterial activity, enzyme inhibition, β-oxoacyl-acyl carrierprotein reductase (FabG), Shantung maple (Acer truncatum Bunge), type IIfatty acid synthase.

Abbreviations used: ACP, acyl carrier protein; cfu, colony-forming units;EGCG, (−)-epigallocatechin gallate; ECG, (−)-epicatechin gallate; FabG,β-oxoacyl-ACP reductase; FabH, β-oxoacyl-ACP synthase III; FabI,enoyl-ACP reductase; FAS, fatty acid synthase; MIC, minimum inhibitoryconcentration; PGG, 1,2,3,4,6-penta-O-galloyl-β-D-glucose.

1 To whom correspondence should be addressed (emailzhangyue@ccmu.edu.cn).

C© 2008 Portland Press Ltd

74 F. Zhang and others

collection of antibacterial agents that inhibit the FAS IIsystem, there are a very limited number of FabG inhibitors.

The Shantung maple (Acer truncatum Bunge) is a localspecies of the genus Acer in China and is a prominent speciesin the hardwood forests of the North of the country. Itis horticulturally important and widely planted for thebrilliant autumn colours of its leaves. It has been used as aChinese folk medicine for the treatment of coronary-arterycirrhosis, cerebrovascular diseases and angina pectoris[8]. We recently reported that the extracts from Acertruncatum are significant inhibitors of FAS I with potentantitumour activity [9]. In view of the 67% homology ofoxoacyl reduction domain of FAS I with bacterial FabG, weinvestigated the inhibitory effect of the extracts on FabGprior to an antibacterial and antifungal test.

Materials and methods

MaterialsAcetyl-CoA, malonyl-CoA and NADPH were purchasedfrom Sigma–Aldrich. All other reagents were local productswith purity of analytical grade.

Preparation and assay of FabGThe engineered strain producing FabG was generouslyprovided by Dr Charles Rock (Infectious Diseases, St. JudeChildren’s Research Hospital, Memphis, TN, U.S.A.). FabGwas prepared as described previously [10,11]. Briefly, N-terminally histidine-tagged FabG was expressed in Escheri-chia coli strain BL21(DE3) and purified by nickel-chelationaffinity chromatography. The preparation was homogeneouson the basis of PAGE in the presence or absence of SDS. Theprotein was stored in 50% (v/v) glycerol at −20 ◦C. FabGactivity was determined using a Shimadzu UV2550 double-wavelength/double-beam spectrophotometer at 37 ◦C bymeasuring the change in absorbance at 340 nm for 1.5 min.The reduction reaction mixture contained 100 mM phos-phate buffer, pH 7.5, 40 mM ethyl acetoacetate, 50 mMNADPH and 10–20 μg of FabG in a total volume of 2.0 ml.The reaction was initiated by the addition of enzyme.

Inhibition studiesReversible inhibition was measured by adding the inhibitor(extract) to the reaction system before the enzyme initiatedthe reaction. Inhibitors were dissolved in 50% (v/v) ethanoland then added to the reaction mixture described above.Reaction mixtures containing 50% ethyl alcohol solventalone were used as controls. It was found that the finalconcentration of ethyl alcohol (0.5%) did not interferewith FabG and FAS I activity. The half-inhibition concen-tration, IC50, was obtained from the percentage inhibition–concentration curves for the inhibitors. Inhibition of FabG

by different concentrations of inhibitor solution was mea-sured and repeated three times.

Time-dependent inhibition was determined by takingaliquots to measure the residual activity at the indicatedtime intervals after the enzyme solution has been mixedwith the inhibitor dissolved in buffer, pH 7.5, containingethanol. The final concentration of ethanol was 7.5% (v/v).This solvent without any inhibitor was used as the control.In these experiments, ethanol at 7.5% (v/v) did not affectFabG activity over several hours.

Preparation of the extracts from fallen leavesof A. truncatumThe fallen leaves of A. truncatum were collected from Baiwangmountain of Beijing and were identified by Professor ChenYu-ting of the Beijing University of Traditional Chinese Medi-cine, Beijing, China. After appropriate treatment, the cleanedand dried leaves were cut into small pieces and mixedwith 50% (v/v) ethanol at a ratio of 1:20 (w/v), followedby magnetic stirring for 24 h at room temperature (25 ◦C)to afford a dark-red concentrated solution on removal ofethanol under reduced pressure, which was then mixed withan equal volume of light petroleum (boiling range 30–60 ◦C)to extract and remove lipids. The extraction was repeatedthree times to give a residual solution free of lipid, which wasthen extracted in turn with an equal volume of ethyl acetatethree times. The pooled ethyl acetate extract was evapor-ated under reduced pressure to yield a brown residue whichwas further dried in a vacuum at <50 ◦C. For comparison,50% ethanol extraction in the ratio of 1:20 (w/v) was alsoperformed without further treatment with light petroleumand afforded a dried substance after removal of solvent.

Determination of MIC (minimal inhibitoryconcentration)The MICs of the test extracts against different bacterialstrains were determined by the doubling-dilution method.ethyl acetate, ethyl alcohol extracts and isolated PGG(1,2,3,4,6-penta-O-galloyl-β-D-glucose) from fallen leaves ofA. truncatum as inhibitors were dissolved in 1% DMSO to afinal concentration of 2 mg/ml. Each kind of Gram-positiveand Gram-negative bacterial strain was inoculated on to ablood-agar plate (for fungi, an inoculated potato dextrose/agar plate was used). The plate was incubated at 37 ◦C for24 h (for fungi, at 28 ◦C for 72 h), followed by a modulatingbacterial (or fungal) suspension with aseptic normalsaline (0.9% NaCl). The concentration of suspension wasMcFarland standard no. 0.5 as measured using a turbidimeter.Then 0.2 ml of the above suspension was added to 40 mlof normal saline and the final concentration of bacterial (orfungi) suspension was 106 cfu (colony-forming units)/ml. A100 μl volume of bacterial (or fungal) suspension (106 cfu)was used to inoculate each well of a 96-well plate containing

C© 2008 Portland Press Ltd

Antibacterial efficacy of aceraceous plant 75

Figure 1 Inhibition of FabG activity by extracts and PGG

The inhibitory effects of extracts and PGG on β-oxoacyl-ACP reductioncatalysed by E. coli FabG are shown. �, PGG; �, ethyl acetate extract; �,ethanol extract.

100 μl of Mueller–Hinton broth and preset concentrationsof inhibitors. The plate was incubated at 37 ◦C for 24 h (forfungi, at 28 ◦C for 48 h) and, meanwhile, the control test wasdone. The microbial growth was determined by observingturbidity in the well plate. The lowest concentration thatlooked transparent was recorded as the minimum inhibitoryconcentration.

Results

Inhibition of FabG by extractsWe have isolated PGG (for structure, see Figure 6 below),one of the components of the extract, by high-speed coun-tercurrent chromatography and have elucidated its struc-ture by analysis of its NMR and mass spectra [12]. Inhibitionby PGG was studied for comparison with that shown bythe whole extract.

The IC50 (the inhibitor concentration inhibiting 50% ofthe activity of FAS) value for the ethyl acetate extract fromA. truncatum indicated its fast-binding reversible inhibitorycapacity. The IC50 values were obtained from the percentageinhibition–concentration curves for the inhibitor (Figure 1).For the ethyl acetate extract, the ethanol extract and PGG,the IC50 values were 2.7, 7.2 and 0.9 μg/ml respectivelywith an 8-fold discrepancy between the highest and lowestvalues. The inhibition patterns were studied in the presenceof three fixed extract concentrations and NADPH as thevariable substrate (Figures 2 and 3). The respective kineticparameters were calculated by secondary plots on the basisof Figures 2 and 3 and are listed in Table 1. It can be seenthat PGG is a mixed-type inhibitor of FabG with respect toNADPH (Figure 2), but the ethyl acetate extract exhibitedan ‘abnormal’ inhibition pattern for NADPH, that is, the

Figure 2 Lineweaver–Burk plot for the inhibition of FabG by PGG

Double-reciprocal plots of 1/v versus 1/[NADPH] at different concentrationsof PGG (�, 0; �, 0.25 μg/ml; �, 0.50 μg/ml; and �, 0.75 μg/ml) are shown.

Figure 3 Lineweaver–Burk plot for the inhibition of FabG by the ethylacetate extract

Double-reciprocal plot of 1/v versus 1/[NADPH] at different concentrationsof extract (�,0; �, 1.25 μg/ml; �, 1.87 μg/ml; and �, 2.5 μg/ml) are shown.

straight lines in the double-reciprocal plot intersect inthe first quadrant (Figure 3).

A time course of inhibition of FabG by ethyl acetateextract (Figure 4) showed no obvious slow-binding inactiv-ation. The remaining activity of FabG fluctuated overthe assay time, though a loss of most of the activity wasobserved during an initial period of several minutes. Figure 5shows the kinetic course of FabG inactivation by PGG; aspeedy irreversible loss of activity was observed.

Antibacterial resultsThe MICs of more than ten bacterial strains, and somefungi too, are listed in Table 1. From Table 1 it is easyto conclude that the antibacterial activity of ethyl acetateextract is stronger than that of the ethyl alcohol extract.This conclusion seems to hold even, in one case, for a

C© 2008 Portland Press Ltd

76 F. Zhang and others

Table 1 IC50, Ki (inhibition constant) and MIC values for extracts and compounds from A. truncatum for β-oxoacyl-ACP reduction of E. coli FabG and (b)antibacterial assay (MIC values)

The species listed in (a) are: aStaphylococcus aureus A.T.C.C. 25923; bStaphylococcus epidermidis; cE. coli A.T.C.C. 25922; dPseudomonas aeruginosa ATCC 27853.Data for EGCG, taxifolin and quercetin are taken from [16]. The species listed in (b) are as follows. Bacteria: 1, Oerskovia xanthineolytica; 2, Ps. alcaligenes; 3,Branhamella catarrhalis; 4, Acinetobacter baumannii; Fungi: 5, Candida krusei; 6, Penicillium spp.; 7, Alternaria alternate; 8, Fusarium spp.

(a)

MIC (mg/ml)

Extract or compound IC50 (μg/ml) Ki (μg/ml) S. a.a S. e.b E. colic Ps. a.d

ExtractsEthyl acetate 2.7 0.25 0.125 0.25 0.25Ethanol 7.2 0.5 0.25 0.5 > 0.5

CompoundsPGG 0.90 0.21 0.25 0.06 0.25 0.125EGCG 2.3 4.5 0.23Taxifolin 6.1 >0.06Quercetin 6.1 0.006

(b)

MIC (mg/ml)

Extract Species . . . 1 2 3 4 5 6 7 8

Ethyl acetate 0.25 0.25 0.125 0.25 0.50 1.0 0.25 0.50Ethanol 0.25 0.50 0.50 0.50 0.50 1.0 0.50 0.50

Figure 4 Kinetic time course of inhibition of FabG by the ethyl acetateextract

Remaining activity at different times in the presence of 75 μg/ml ethyl acetateextract is shown. The FabG solution (10.7 μM) was mixed with extract(150 μg/ml) in a 0.1 M K2HPO4/KH2PO4 buffer, pH 7.5. The final concentrationof FabG was 5.35 μM before aliquots were taken to assay the remaining activity.

fungus (Alternaria alternate). However, Table 1(a) shows thatPGG possesses more potent antibacterial efficacy than doeither the ethyl acetate and ethyl alcohol extracts withrespect to some strains.

Discussion

FabG has a monomeric molecular mass of 25.5 kDa andexists as a tetramer in solution [13]. FabG catalyses the

Figure 5 Kinetic time course of inhibition of FabG by PGG

Remaining activity at different times in the presence of 30 μg/ml PGG isshown. The FabG solution (21 μM) was mixed with PGG (60 μg/ml) in0.1 M K2HPO4/KH2PO4 buffer, pH 7.5, and the final concentration of FabGwas 10.5 μM before aliquots were taken to assay the remaining activity.

keto reduction by a compulsory ordered mechanism withthe NADPH cofactor as the leading substrate and NADP+

as the last product to leave. An allosteric interaction usuallyoccurs when the oligoprotein contains more than two sub-units, as shown in Figure 3. It is most probable that the com-ponents of extract bind to the allosteric sites of FabG,causing an ‘abnormal’ inhibition pattern. TLC detectionrevealed that the extract comprises flavonoids and poly-phenols that may be the allosteric effectors.

FabG is ubiquitously expressed in all bacteria, is highlyconserved across species and is the only known isoenzymethat catalyses the essential keto-reduction step in the elong-ation cycle [14]. Thus FabG represents an ideal focus for the

C© 2008 Portland Press Ltd

Antibacterial efficacy of aceraceous plant 77

Figure 6 Chemical structures of EGCG, PGG, taxifolin and quercetin

development of new antibiotics. However, FabI is, like FabG,a member of the short-chain dehydrogenase/reductasesuperfamily [15]. The component proteins of this family bindnicotinamide nucleotide cofactors such as NADPH/NADH+

using a similar protein fold. Zhang and Rock [16] indicatedthat EGCG [(−)-epigallocatechin gallate; for structure,see Figure 6] inhibited the binding of cofactors such asNADPH/NADH+ in both FabG and FabI enzymes. Theirfinding leads to the hypothesis that the extract may also in-hibit FabI, notwithstanding that this was not assayed. EGCGexhibited a mixed-type inhibition for NADPH in FabG re-ductase, meaning that it binds both to the free enzyme andto the enzyme–substrate complex, whereas PGG inhibitsFabG mostly by occupying the NADPH-binding site. Theauthors suggested that some natural-products polyphenolcompounds might have potential antibacterial activity.

We used Gram-positive and -negative bacteria toevaluate the antibacterial abilities of different samples. Asshown in Table 1, in comparison with either quercetin ortaxifolin (for structures, see Figure 6), the antibacterialactivity (MIC) of PGG against E. coli is weaker, but its inhib-itory capacity against FabG is much greater than that of theformer. A comparison of PGG and EGCG with regard totheir respective IC50 and Ki values demonstrated that PGG isthe stronger inhibitor of FabG, but their antibacterial activ-ities against E. coli are almost equal.

However, another set of MICs for Staphylococcusepidermidis and Pseudomonas aeruginosa seemed to showthat the notion of FabG as a primary antibacterial targetis reasonable. It can be seen from Figure 4 that the ethylacetate extract has no slow-binding inhibitory activitytowards FabG, meaning its limited inhibition is based onreversible binding to the enzyme. PGG, however, exhibiteda potent activity towards FabG (Figure 5), inasmuch as inthe initial 15 min, approx. 70% of enzyme activity was lost.

Accordingly, the MICs of PGG for the above two bacteria aremuch less than those for the ethyl acetate extract (Table 1).

As is well known, fatty acid biosynthesis in fungi oreukaryotes is performed by a type-I FAS that is a dimer ofa single large multifunctional polypeptide. We have recentlyreported [9] that an ethyl acetate extract from aceraceousplant (A. truncatum) effectively inhibited type-I FAS withan IC50 of 0.7 μg/ml for the overall reaction and one of16 μg/ml for the β-oxoacyl reduction reaction (its domainshowing 67% homology with bacterial FabG) and exhibitedin the present study a remarkable antifungal effect (Table 1).

Li et al. [17] reported that EGCG exhibited a typicaltime-dependent inhibition of FabG, which possibly resultedfrom the EGCG-induced aggregation of FabG. It was obser-ved that FabG inactivation and aggregation occurrednearly simultaneously, with a lag time that decreased withincreasing EGCG concentration. These results suggest thatsome chemical reactions required for aggregation and inac-tivation occurred during the lag time. Both PGG and EGCGare polyphenols, and a similar reaction most likely occurredbetween PGG and FabG, as shown in Figure 5. However,plant polyphenols have multiple cellular targets, and FabGmight be one of them. On the basis of the present findings,we postulate that the antibacterial activity of the aceraceousplant extract in bacteria is, to some extent at least, dueprimarily to the inhibition of FabG.

Acknowledgments

We thank (through W.-X.T.) Dr Charles Rock forgenerously providing the engineered strain of E. coliproducing FabG.

References

1 Hong, T. T. and Schweizer, H. P. (1999) J. Bacteriol. 181,5489–5497

2 Lai, C. Y. and Cronan, J. E. (2003) J. Biol. Chem. 278,51494–51503

3 Heath, R. J. and Rock, C. O. (1996) J. Biol. Chem. 271,1833–1836

4 Heath, R. J. and Rock, C. O. (1996) J. Biol. Chem. 271,10996–11000

5 Khandekar, S. S., Gentry, D. R., Van Aller, G. S., Warren, P., Xiang,H., Silverman, C., Doyle, M. L., Chambers, P. A., Konstantinidis,A. K., Brandt, M. et al. (2001) J. Biol. Chem. 276, 30024–30030

6 Daines, R. A., Pendrak, I., Sham, K., Van Aller, G. S.,Konstantinidis, A. K., Lonsdale, J. T., Janson, C. A., Qiu, X.,Brandt, M., Khandeker, S. S. et al. (2003) J. Med. Chem. 46,5–8

7 He, X., Reeve, A. M., Desai, U. R., Kellogg, G. E. and Reynolds,K. A. (2004) Antimicrob. Agents Chemother. 48, 3093–3102

C© 2008 Portland Press Ltd

78 F. Zhang and others

8 Shi, W. Sh. and Lei, X. (2003) Chinese Trad. Pat. Med. 25,948–950

9 Zhao W. H., Zhang, J. F., Wang, Z., Zhang, Y. X. and Tian,W. X. (2006) J. Enz. Inhib. Med. Chem. 21, 589–596

10 Heath, R. J. and Rock, C. O. (1995) J. Biol. Chem. 270,26538–26542

11 Heath, R. J., Rubin, J. R., Holland, D. R., Zhang, E., Snow, M. E.and Rock, C. O. (1999) J. Biol. Chem. 274, 11110–11114

12 Zhao, W. H., Gao, Ch. Ch., Ma, X. F., Bai, X. Y. and Zhang, Y.X. (2007) J. Chromatogr. B 850, 523–527

13 Sheldon, P. S., Kekwick, R. G., Smith, C. G., Sidebottom, C. andSlabas, A. R. (1992) Biochim. Biophys. Acta 1120, 151–159

14 Lai, C. Y. and Cronan, J. E. (2004) J. Bacteriol. 186, 1869–187815 Bergler, H, Wallner, P., Ebeling, A., Leitinger, B., Fuchsbichler,

S., Aschauer, H., Kollenz, G., Hogenauer, G. and Turnowsky, F.(1994) J. Biol. Chem. 269, 5493–5496

16 Zhang, Y. M. and Rock, C. O. (2004) J. Biol. Chem. 279,30994–31001

17 Li, B. H., Zhang, R., Du,Y. T., Sun,Y. H. and Tian, W. X. (2006)Biochem. Cell Biol. 84, 755–762

Received 23 November 2007/17 January 2008; accepted 21 January 2008Published as Immediate Publication 21 January 2008, doi:10.1042/BA20070255

C© 2008 Portland Press Ltd