Food Chemistry 139 (2013) 332–338

Contents lists available at SciVerse ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Synthesis of amino acid conjugates of tetrahydrocurcumin and evaluationof their antibacterial and anti-mutagenic properties

J.R. Manjunatha a, B.K. Bettadaiah a, P.S. Negi b, P. Srinivas a,⇑a Department of Plantation Products, Spices and Flavour Technology, CSIR-Central Food Technological Research Institute, Mysore 570 020, Indiab Department of Fruits and Vegetable Technology, CSIR-Central Food Technological Research Institute, Mysore 570 020, India

a r t i c l e i n f o a b s t r a c t

Article history:Received 31 August 2012Received in revised form 20 December 2012Accepted 29 January 2013Available online 6 February 2013

Keywords:TetrahydrocurcuminAmino acidsConjugatesSynthesisAntimutagenicityAntibacterial activity

0308-8146/$ - see front matter � 2013 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.foodchem.2013.01.081

⇑ Corresponding author. Address: Department of PlFlavour Technology, CSIR-Central Food Technologicuvamba Mansion, Mysore 570 020, India. Tel.: +912517233.

E-mail address: psrinivas@cftri.res.in (P. Srinivas).

Tetrahydrocurcumin (THC), the hydrogenated and stable form of curcumin, exhibits physiological andpharmacological activities similar to curcumin. A protocol has been developed for the synthesis of novelconjugates of THC with alanine (2a), isoleucine (2b), proline (2c), valine (2d), phenylalanine (2e), glycine(2f) and leucine (2g) in high yields (43–82%). All the derivatives of THC exhibited more potent anti-micro-bial activity than THC against Bacillus cereus, Staphylococcus aureus, Escherichia coli and Yersinia enterocol-itica. The MIC values of the derivatives were 24–37% of those for THC in case of both Gram-positive andGram-negative bacteria. Derivatives 2g and 2d exhibited maximum anti-mutagenicity against Salmonellatyphimurium TA 98 and TA 1538, respectively at a low concentration of 313 lg/plate, with comparableactivity for THC evident only at 3750 lg/plate. These results clearly demonstrated that the conjugationof THC at the phenolic position with amino acids led to significant improvement of its in vitro biologicalattributes.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Curcumin is the major colour constituent of turmeric (Curcumalonga, Zingiberaceae) used in Indian cuisine mainly for its colouringand flavoring attributes. Curcumin is highly valued for its medici-nal and nutraceutical attributes (Ravindran, Babu, & Sivaraman,2007). Although curcumin has promising therapeutic potential(Corson & Crews, 2007), various studies have highlighted its insta-bility under physiological conditions (Anand, Kunnumakkara,Newman, & Aggarwal, 2007). It is soluble in organic solvents likeacetone and ethyl acetate but insoluble in water at acidic or neutralpH. At pH 7.2, curcumin is sparingly soluble in aqueous media(�20 lg per ml) but degrades rapidly with the half-life of 30 min.The stability further decreases with increase in pH (Wang et al.,1997). Tetrahydrocurcumin (THC, 1) is the major and stable metab-olite of curcumin. Its glucuronidation constitutes the reportedpathway of metabolism of curcumin (Pan, Huang, & Lin, 1999).Glucuronides of curcumin and THC can serve as bio-availableforms of curcumin in vivo. It exhibits physiological and pharmaco-logical activities similar to curcumin and, in some systems, doesshow higher antioxidant activity than curcumin. THC is reported

ll rights reserved.

antation Products, Spices andal Research Institute, Chel-821 2512352; fax: +91 821

to be more stable than curcumin in buffer solutions of physiologi-cal pH of 7.2 and also at basic pH, as well as in plasma.

THC is derived from curcumin by selective reduction of olefinicbonds alpha to the carbonyl group in a diferuloyl backbone. THC isan effective natural antioxidant (Osawa, Sugiyama, Inayoshi, &Kawakishi 1995; Sugiyama, Kawakishi, & Osawa, 1996) and anti-inflammatory compound (Rao, Basu, & Siddiqui 1982), employedin many formulations in the field of cosmetics and nutrition. THCoffers protection to the skin against damage due to ultraviolet radi-ation and other factors which are essential for sun protection prep-arations and cosmetics for aged skin. Tetrahydrocurcuminoidshave superoxide scavenging ability and inhibitors of fat oxidationand offers protection against rancidity of several fat components.THC produces the protective effect to cells against oxidative stressby scavenging free radicals and reactive oxygen species (Nakam-ura, Ohto, Murakami, Osawa, & Ohigashi, 1998). THC is a potentantioxidant under the conditions where the radical initiators areproduced in the polar water medium (Khopde et al., 2000). Admin-istration of THC to streptozotocin (STZ)-nicotinamide-induced dia-betic male Wistar rats significantly improves specific insulinbinding to the receptors, with the receptor numbers and affinitybinding reaching near-normal levels resulting in a significantincrease in plasma insulin with the effect of THC being moreprominent than that of curcumin (Murugan, Pari, & Rao, 2008).THC-inhibits lipoxygenase (LOX) enzyme implicated in inflamma-tory conditions by preventing the activation of LOX-1 (Sneharani,Sridevi, Srinivas, & Rao, 2011). It is also effective in inhibiting

J.R. Manjunatha et al. / Food Chemistry 139 (2013) 332–338 333

cyclooxygenase-2 and phospholipase A2 (Hong et al., 2004). THCinhibits cancer (HT1080) cell migration and invasion by down reg-ulation of extracellular matrix (ECM) degradation enzymes and theinhibition of cell adhesion to ECM proteins (Yodkeeree, Garbisa, &Limtrakul, 2008).

Recent attempts at preparing sugar and amino acid conjugatesof curcumin led to the preparation of a number of water-solublecurcumin derivatives, which exhibited potent antioxidant, antimi-crobial and antimutagenic properties comparable to and in severalcases superior than curcumin (Kumar, Narain, Tripathi, & Misra,2001; Parvathy, Negi, & Srinivas, 2009, 2010; Parvathy & Srinivas,2008). The authors demonstrated that curcumin–b-diglucosideprevented oligomer formation and inhibited fibril formation of a-synuclein, whose aggregation is centrally implicated in Parkinson’sdisease (Bharathi, Parvathy, Srinivas, Indi, & Rao, 2012). Curcuminis unstable and is metabolised to THC and other reduced formsin vivo. Hence, the bioavailability of curcumin is limited. Tetrahy-drocurcumin (THC) is a stable metabolite of curcumin in physio-logical systems and more lipophilic than curcumin. Also, inseveral studies THC is shown to exhibit a wide array of bioactiveproperties more potent than curcumin. Hence, it was envisagedthat it would be of great interest to synthesise new amino acidderivatives of THC and explore their bioactive attributes. In thepresent study, we attempted the syntheses of selected novel aminoacid conjugates of THC and evaluated their in vitro antimicrobialand antimutagenicity potential.

2. Materials and methods

2.1. Apparatus and materials

All the solvents and reagents used for the synthesis were of ana-lytical grade. Curcumin (�95% purity) was procured from Ms. Spi-cex Chemicals Pvt. Ltd., Mysore, India. Palladium on bariumsulphate catalyst, t-Boc amino acids, 4-dimethylaminopyridineand N,N0-dicyclohexylcarbodiimide were purchased from Sigma Al-drich Chemical Co. (St. Louis, MO, USA). Dry 1,4-dioxane, trifluoro-acetic acid, chloroform and dichloromethane were purchased fromMerck Specialty Chemicals, Mumbai, India. 1H and 13C NMR spectrafor the compounds were recorded on a 500 MHz NMR spectrome-ter (Bruker Avance, Reinstetten, Germany), using CD3OD solvent.Coupling constants (J values) are given in Hz. High resolution massspectral analyses of the compounds were carried out using MS(Waters Q-Tof Ultima, Manchester, UK) in the ESI positive mode.Thin-layer chromatographic (TLC) analysis was done on silica gel60 F254 (Merck, Germany) coated on an alumina sheet and 3%methanol in chloroform as the developing solvent. The productswere purified by triturating using ethyl acetate and hexane mix-ture. All the chemicals and Petri plates used for microbial studieswere procured from Hi Media Ltd., Mumbai, India.

2.2. General method for synthesis of curcumin–t-Boc amino acidconjugates

A solution of curcumin (2 g, 5.43 mmol) in dioxane (25 ml) wastaken in a round-bottomed flask. t-Boc amino acid (a–g,13.58 mmol) was added gradually to ensure its complete solubility.4-Dimethylaminopyridine (5.43 mmol) was added to reaction mix-ture and stirred for 5 min. After this N,N0-dicyclohexylcarbodiimide(13.58 mmol) was added and set for stirring until the completionof reaction (�7–12 h) under nitrogen atmosphere. The progressof the reaction was monitored by TLC analysis. After the comple-tion of reaction dicyclohexylurea was filtered off. The filtrate wasmixed with chloroform (25 ml) and subjected to aqueous workup.The organic layer was separated, passed through anhydrous

Na2SO4 and concentrated. Further, the product was purified bytriturating using ethyl acetate and petroleum ether (60–80 �C)mixture to afford t-Boc protected amino acid derivatives of curcu-min in 75–95% yield.

2.3. General method for deprotection of t-Boc group in curcumin–t-Bocamino acid conjugates

Each of the curcumin–amino acid t-Boc conjugates (�3 g) wasdissolved in CH2Cl2 (20 ml). To this solution, trifluoroacetic acid(TFA, 4 ml) in CH2Cl2 (20 ml) was added slowly under ice cold con-dition. The reaction mixture was allowed to stir at 25 �C till thecompletion of the reaction (3–5.5 h). It was followed by neutralisa-tion of TFA by gradual addition of solid K2CO3. The product getsseparated from the mixture and settled at the bottom of the flask.It was filtered, washed with water, dried and then kept in a vac-uum desiccator over KOH pellets for 12 h. The pure products wereobtained in the 60–90% yield. The structure of curcumin–aminoacid conjugates (1a–1g) was confirmed by 1D and 2D-NMR andmass spectral analyses.

2.4. General method for the synthesis of THC–amino acid conjugates(2a–2g)

The solution of curcumin–amino acid conjugates (1 g, 1a–1g) inmethanol (15 ml) was taken in hydrogenation flask and chargedwith Pd/BaSO4 (10 mg, 10% w/w). The mixture was agitated at30 psi H2 pressure till completion of the reaction (0.5–1.5 h), as in-ferred by disappearance of golden yellow colour. Additionally, itwas confirmed by TLC analysis. After completion of the reaction,the catalyst was filtered off and the filtrate was concentrated underreduced pressure to afford pure products (2a–2g) in quantitativeyields. The products were characterised by 1D and 2D-NMR andhigh resolution mass spectral analyses. The physical and spectro-scopic data of the novel THC–alanine conjugate (2a) are presentedbelow with numbering of the carbon atoms as depicted in Table 1.The physical and spectroscopic data of other conjugates (2b–2g)are provided in the Supplementary information.

(2a): [(Z)-5-hydroxy-1,7-bis(4-O-L-alaninoyl-3-methoxyphenyl)hept-4-en-3-one]; off-white solid; m. p. 74–76 �C; 1H NMR(500 MHz, CD3OD): d 1.70 (d, J = 7.2 Hz, 6H, H-17 & H-170), 2.55–2.62 (m, 2H, H-6). 2.72–2.83 (m, 4H, H-1 & H-2), 2.84–2.89 (m,2H, H-7), 3.76 (s, 6H, H-14 & H-140), 4.35 (q, J = 7.3 Hz, 2H, H-16& H-160), 5.56 (s, 1H, H-4), 6.72–6.81 (m, 2H, H-13 & H-130),6.88–6.94 (m, 2H, H-9 & H-90), 6.95–7.01 (m, 2H, H-12 & H-120).13C NMR (125 MHz, CD3OD): d 16.52 (C-17 & C-170), 30.11 (C-1),32.33 (C-7), 40.59 (C-6), 45.70 (C-2), 49.89 (C-16 & C-160), 56.47(C-14 & C-140), 100.86 (C-4), 113.99 (C-9 & C-90), 121.54 (C-13 &C-130), 123.12 (C-12 & C-120), 138.61 (C-11), 138.70 (C-110),142.14 (C-8), 142.31 (C-80), 151.92 (C-10 & C-100), 169.65 (C-15 &C-150), 194.15 (C-5), 206.01 (C-3). Mass: Calculated for formulaC27H34N2O8: 514.2315; Found: [M++1] = 515.4629.

2.5. Antibacterial studies of tetrahydrocurcumin–amino acidconjugates

The antibacterial activity of THC–amino acid conjugates wastested essentially by the method described by Negi, Jayaprakasha,Rao, and Sakariah (1999). Bacillus cereus (F 4810, Public Health Lab-oratory, London, UK), Staphylococcus aureus (FRI 722, Public HealthLaboratory, The Netherlands), Escherichia coli (MTCC 108, MicrobialType Culture Collection, Institute of Microbial Technology, Chandi-garh, India) and Yersinia enterocolitica (MTCC 851, Microbial TypeCulture Collection, Institute of Microbial Technology, Chandigarh,India) were sub-cultured in BHI broth and incubated for 24 h at37 �C. After incubation cells were harvested by centrifugation

Table 1Amino acid conjugates of tetrahydrocurcumin.

Entrya Product Time in h Yield (%)b

2a O OHO

O

O

O

OO

NH2 NH2

1

2

3

4

5

6

7

8

9

10

1112

13

14

1516

178′

9′

10′

11′12′

13′

14′

15′16′

17′

O OHO

O

O

O

OO

NH2 NH2

1

2

3

4

5

6

7

8

9

10

1112

13

14

1516

178′

9′

10′

11′12′

13′

14′

15′16′

17′

(i) 11 85(ii) 05 80(iii) 01 >95

2b O OHO

O

O

O

OO

NH2 NH2

(i) 12 90(ii) 4.5 85(iii) 1.5 >95

2c O OHO

O

O

O

OOHN

HN

(i) 08 75(ii) 05 71(iii) 0.5 >95

2d O OHO

O

O

O

OO

NH2 NH2

(i) 10 92(ii) 04 90(iii) 01 >95

2e O OHO

O

O

O

OO

NH2 NH2

(i) 07 95(ii) 03 90(iii) 0.5 >95

2f O OHO

O

O

O

OOH2N NH2

(i) 12 75(ii) 5.5 60(iii) 0.5 >95

2g O OHO

O

O

O

OO

NH2 NH2

(i) 12 90(ii) 5 86(iii) 1.5 >95

a Conjugates of THC with alanine (2a), isoleucine (2b), proline (2c), valine (2d), phenylalanine (2e), glycine (2f) and leucine (2g).b Isolated yields of products of steps (i), (ii) and (iii) as specified in Fig. 1.

334 J.R. Manjunatha et al. / Food Chemistry 139 (2013) 332–338

(5000 rpm, 10 min) and serially diluted in saline solution (0.85%NaCl, w/v) for use in antibacterial assay. Test samples were pre-pared by dissolving the THC and its amino acid conjugates inDMSO. To flasks containing 20 ml of melted warm agar, differentconcentrations of test material (equivalent amount of DMSO incontrol) and 100 ll (about 103 cfu/ml) of the test organism wereadded and the contents poured onto sterilised Petri plates afterthorough mixing. The plates were incubated at 37 �C for 24 h.The colonies developed after incubation were counted and theinhibitory effect calculated using the following formula:

% inhibition ¼ ð1� T=CÞ � 100

where T is cfu/ml of test sample and C is cfu/ml of control.The minimum inhibitory concentration (MIC) was determined

as the lowest concentration of the compound inhibiting the com-plete growth of bacterium being tested. The growth inhibitionstudies for each compound were done in duplicate and the exper-iments were repeated three times.

2.6. Antimutagenicity tetrahydrocurcumin–amino acid conjugates byAmes test

The antimutagenicity of THC–amino acid conjugates was stud-ied using the tester strains of Salmonella typhimurium (TA 98 and

TA 1538, Microbial Type Culture Collection, Institute of MicrobialTechnology, Chandigarh, India) through the standard plate incor-poration test as described by Maron and Ames (1983). In the anti-mutagenicity studies on THC–amino acid conjugates, the inhibitionof mutagenicity of the sodium azide by the test samples was calcu-lated by determining the number of His+ revertants in the plate.The test samples (313, 625, 1250, 2500, 3750, 5000, 6250 and7500 lg) were assayed by plating with molten soft agar (0.6%,2 ml) containing 0.5 mM of histidine/biotin (0.2 ml) and 0.1 ml of10 h old culture of either S. typhimurium TA 98 or S. typhimuriumTA 1538 on minimal glucose agar plates. Sodium azide was usedas a diagnostic mutagen (1.5 lg/plate) in positive control andplates without sodium azide and without test samples were con-sidered as negative controls. His+ revertants were counted afterincubation of the plates at 37 �C for 48 h. Each sample was assayedusing duplicate plates and the data were presented as mean ± SD ofthree independent assays. The mutagenicity of sodium azide in theabsence of test samples was defined as 100% or 0% inhibition. Thecalculation of percentage inhibition was done according to the fol-lowing formula

% Inhibition ¼ ½1� T=M� � 100

where T is number of revertants per plate in presence of mutagenand test sample, and M is number of revertants per plate in positive

J.R. Manjunatha et al. / Food Chemistry 139 (2013) 332–338 335

control. The number of spontaneous revertants (negative controls)was subtracted from the numerator and the denominator. The anti-mutagenic effect was considered weak when the inhibitory effectwas less than 25%, medium when the inhibitory effect was 25–40% and strong when the inhibitory effect was more than 40% (Ik-ken et al., 1999).

2.7. Statistical analysis

All the experiments were repeated three times and the datawere calculated as mean ± SD. The data of all the assays were ana-lysed by one-way ANOVA. Duncan’s multiple range test (DMRT)was used to make the comparisons among various treatments.

3. Results and discussion

3.1. Synthesis of amino acid conjugates of THC

The reaction of THC (1, Fig. 1) with t-Boc amino acids was inves-tigated as per the protocol for synthesis of conjugates of curcuminwith t-Boc amino acids reported earlier (Parvathy et al., 2010). Itwas observed that the reactions were slower and the yields ofthe products were relatively low. More importantly, the efforts tode-protect the t-Boc group from the THC–t-Boc amino acid conju-gates were not successful as the reaction led to the cleavage of theester group resulting in the formation of THC. Hence, an alternativeroute had to be delineated for the synthesis of amino acids conju-gates of tetrahydrocurcumin (THC). In the present study, a path-way has been envisaged to THC conjugates via the selectivereduction of the double bonds alpha to carbonyl groups in the cur-cumin moiety in the curcumin–amino acid conjugates (Fig. 1).While, such an approach was successful in obtaining THC from cur-cumin, the reduction in the presence of a labile ester group in cur-cumin–amino acid conjugates has never been tried earlier.Interestingly, the selective reduction of double bonds in the pres-ence of carbonyl moiety occurred smoothly when the reaction ofcurcumin–amino acid conjugates with hydrogen was carried outunder pressure in presence of palladium on barium sulphate cata-lyst. Accordingly, conjugates of curcumin with alanine, isoleucine,

O OHO

HO

O

OH

HO

OR

NHBoc2+

a,1a, 2a: R = -CH3 ; b, 1b, 2b: R = -CH(CH3)C2H5 ;c, 1c, 2c: R = -(CH2)3- as part of pyrrolidine ring;d, 1d, 2d: R = -CH(CH3)2 ; e, 1e, 2e: R = -CH2(C6H5);f, 1f, 2f: R = -H; g, 1g, 2g: R = -CH2CH(CH3)2

a - gCurcumin t-Boc amino acids

i

Reagents and condi tion: (i) DCC, DMAP, Dioxane, RT, 7-12 h,(ii i) H2, Pd/BaSO4, 30 p

O OHO

HO

O

OH

Tetrahydrocurcumin

1

H2 Pd/BaSO4

Fig. 1. Synthesis of THC–a

proline, valine, phenylalanine, glycine and leucine were preparedin 68%, 77%, 53%, 83%, 86%, 45% and 77% yields, respectively. Thecorresponding THC conjugates (2a–2g) were obtained from thereduction of respective curcumin–amino acid conjugates in highyields (43–82%, Table 1). The compounds were characterised bytheir 1D and 2D 1H and 13C NMR and high resolution mass spectraldata. The distinguishing feature of all the 1H NMR spectra in THCconjugates of the various amino acids was the presence of themethylene protons in the hepta-3,5-dione portion of the moleculeas multiplet for protons attached to C-1, C-2, C-6 and C-7 in the2.10–2.93 ppm range. This clearly indicated the reduction of thedouble bond of curcumin–amino acid conjugate. In few casesmethylene protons of C-2 and C-6 registered as triplets, while pro-tons of C-1 and C-7 collectively appeared as multiplet integratingto 4 protons. In case of the THC–alanine conjugate (2a), the 13CNMR spectra exhibited the signals at 30.11, 32.33, 40.59 and45.70 corresponding to methylene carbons. Signals related tovinylic carbon (C-5) and carbonyl (C-3) appeared at 194.15 and206.01 respectively. The carbon peaks at 16.52, 49.89 and 169.65corresponds to methyl (C-17, 170), methine (C-16, 160) and car-bonyl (C-15, 150) of alanine. The magnetically equivalent carbonsof aromatic and amino acid moieties appeared together indicatingthe symmetric nature of the molecule. However, the quaternarycarbon signals of C-8 & C-80 and C-11 & C-110 registered small dif-ferences in their chemical shift values. The structure of the com-pound was further ascertained by 2D NMR spectral correlationsand high resolution mass spectral studies.

These structural features along with signals for other carbonsand protons of the aliphatic and aromatic moieties of the respec-tive amino acid and aryl groups were clearly discernable in thespectra for conjugates of THC with other amino acids. These newcompounds were fully characterised by the concurrence of the cal-culated masses with those obtained by high resolution mass spec-tral analyses.

3.2. Antibacterial properties of THC–amino acid conjugate

THC and its amino acid conjugates (2a–2g) were tested for theirantimicrobial potential against two Gram-positive (B. cereus and S.

O OHO

O

O

O

O O

NH-Boc

R

NH-Boc

R

O OHO

O

O

O

O O

NH2

R

NH2

R

2a - 2g

1a - 1g

O OHO

O

O

O

O O

NH2

R

NH2

R

ii

iii

Iner t, 75-95%; (ii) 10% TFA in DCM, 0oC, 3-5.5 h, 60-90%;si, 0.5-1.5 h, > 95%

mino acid conjugates.

Table 2MIC of THC–amino acid conjugates against Gram-positive and Gram-negativebacteria.

Compounds MIC (lMol)

B. cereus S. aureus E. coli Y. enterocolitica

1 1066 1329 1723 21142a 340 437 583 7772b 334 459 543 5852c 309 353 485 6182d 263 482 526 6572e 337 487 600 7502f 257 360 514 6172g 334 376 501 585

336 J.R. Manjunatha et al. / Food Chemistry 139 (2013) 332–338

aureus) and two Gram-negative bacteria (E. coli and Y. enterocoliti-ca). The minimum inhibitory concentration (MIC) values of THCand its amino acid conjugates are presented in Table 2. All thecompounds showed appreciably higher activity than THC. TheTHC–glycine and THC–valine conjugates (2f and 2d) were the mostpotent against B. cereus as they exhibited the lowest MIC values of257 and 263 lmols, respectively. Other compounds showed MICvalues in the range of 309–340 lmol against B. cereus, which wasmuch lower than that of THC (1066 lmol). The compound 2f, alongwith 2c and 2g also showed higher antibacterial activity (353–376 lmol) against S. aureus, which was much lower than that ofTHC (1329 lmol). In case of Gram-negative bacteria, E. coli and Y.enterocolitica, MIC values for all the compounds were in the rangeof 485–583 and 585–777 lmol, respectively. Although these MICvalues were higher than the values against the Gram-positive bac-

0

10

20

30

40

50

60

70

80

90

100

% In

hibi

tion

a

bbc cc

e

a

d

b

d

c

a a a a a

200 mg/l 300 mg/l

0

10

20

30

40

50

60

70

80

90

100

% In

hibi

tion

f

f

b

c

d

d

c

cb

a

d

e

aa

e

e

100 mg/l 150 mg/l

A

C

Fig. 2. Antibacterial activity of tetrahydrocurcumin and tetrahydrocurcumin–amino acidand (D) Yersinia enterocolitica at two different concentrations. 1 2a 2b 2c 2d 2not significantly (p < 0.05) different.

teria, they were still much lower than the corresponding values forTHC (1723 and 2114 lmol, respectively).

Fig. 2(A–D) depicts the growth inhibition of bacteria by THC (1)and its amino acid conjugates. In case of B. cereus (Fig. 2A), 2fshowed highest inhibition and 1 showed least inhibition of growthat both concentrations (100 and 150 mg/l). The growth inhibitionpattern by all the compounds was similar (2f > 2d > 2a > 2c >2b > 2e > 2g > 1) at both concentrations, however, statistically(p < 0.05) similar inhibition of growth was shown by 2a and 2c,2b and 2e at 100 mg/l; and 2e and 2g at 150 mg/l. As 150 mg/lwas higher than the MIC values for 2d and 2f, both showed 100%inhibition of growth at this concentration.

Compound 2f also showed highest inhibition of growth of S.aureus (Fig. 2B), however, minimum inhibition was shown by 2g(which was statistically [p < 0.05] lower than 1) at both concentra-tions. At 100 mg/l, 2f and 2c showed highest but statistically(p < 0.05) similar inhibition of growth followed by 2a, which wasfollowed by 1, 2b, 2d and 2e (all showed statistically (p < 0.05) sim-ilar inhibition of growth) and least inhibition was shown by 2g,whereas the trend for growth inhibition was 2f > 2c > 2d > 2a =2b = 2e > 1 > 2g at 150 mg/l. For E. coli (Fig. 2C) and Y. enterocolitica(Fig. 2D), highest inhibition of growth was shown by 2f and mini-mum inhibition was shown by 1 at 200 and 300 mg/l, however,growth inhibition shown by 1 against Y. enterocolitica was statisti-cally (p < 0.05) similar to 2d and 2g at 200 mg/l (Fig. 2D). In case ofE. coli, 2a showed statistically (p < 0.05) similar inhibition as 2f at200 mg/l. At 300 mg/l, 2a, 2c, 2d, 2f and 2g showed 100% inhibitionof growth (Fig. 2C). The trend of growth inhibition shown by vari-ous compounds at two different concentrations was in concurrencewith their MIC values (Table 2).

0

10

20

30

40

50

60

70

80

90

100

% In

hibi

tion

a

bc

d

c

d

c

dd

bb b

cc

a

b

200 mg/l 300 mg/l

0

10

20

30

40

50

60

70

80

90

100

% In

hibi

tion

a

b

c

d

c cc

a

ede

f

d

b

c

d

a

100 mg/l 150 mg/l

B

D

conjugates against (A) Bacillus cereus, (B) Staphylococcus aureus, (C) Escherichia colie 2f 2g. Values at each concentration followed by same letter in each strain are

0

20

40

60

80

100

313 625 1250 2500 3750 5000 6250 7500 313 625 1250 2500 3750 5000 6250 7500

Inhi

bitio

n (%

)

Concentration (µg / plate)

abb

bc

c

a

d

bc

c

a

b

c

a

b

a a a a(a)

0

20

40

60

80

100

Concentration (µg / plate)

Inhi

bitio

n (%

)

a

d

bc

e

ab

e

abc

d

a

d

b

c

a

b

a

b

a

b

a a

0

20

40

60

80

100

Concentration (µg / plate)

Inhi

bitio

n (%

)

e

c

a a a a

b

b

bd

d

d

e

c

0

20

40

60

80

100

313 625 1250 2500 3750 5000313 625 1250 2500 3750 5000

Concentration (µg / plate)

Inhi

bitio

n (%

)

c

a

abb

a

ab

b

a

a a a a(b)

Fig. 3. Inhibitory effect of tetrahydrocurcumin and tetrahydrocurcumin–amino acid conjugates against the sodium azide induced mutagenicity in (a) Salmonella typhimuriumTA 98 and (b) Salmonella typhimurium TA 1538. (2a) (2b) (2c) (2d) (2e) (2f) (2g) THC (1). Values at each concentrationfollowed by same letter in each strain are not significantly (p < 0.05) different.

J.R. Manjunatha et al. / Food Chemistry 139 (2013) 332–338 337

Earlier workers reported that certain curcumin bio-conjugatescontaining esters and peptides showed enhanced antifungal andantibacterial activities, which was attributed to better cellularuptake, increased cellular concentration and better receptorbinding (Kapoor, Narain, & Misra, 2007). We also observed thatTHC–amino acid conjugates had antimicrobial activities muchhigher than THC against all the bacteria tested in the presentstudy. Apparently, the amino acid portion of the derivatives ren-ders the THC conjugates hydrophilic which assists in the cellularuptake of the covalently bound THC by the bacterial cells. Thehigher antimicrobial activities of THC conjugates could thus beattributed to their higher concentration per se in the bacterialcell or THC that could be liberated in vivo from breakage ofthe ester bond. While comparing the activity of individual com-pounds against bacteria, it was observed that all the compoundsshowed higher MIC values against Gram-negative bacteria thanGram-positive ones. A similar trend of higher resistance toGram-negative bacteria than Gram-positive bacteria has beenobserved earlier (Nostro, Germano, D’Angelo, Marino & Cannatel-li, 2000). This could be due to their differences in cell structureas the Gram-positive bacteria contain an outer peptidoglycanlayer, which is not a good permeability barrier (Scherrer andGerhardt, 1971). In case of Gram-negative bacteria, the outerphospholipidic membrane is impermeable to lipophilic solutes,and porins present in membrane act as a selective barrier tothe hydrophilic solutes making them comparatively more resis-tant to antibacterial compounds (Nikaido and Vaara, 1985).

3.3. Antimutagenicity studies of THC–amino acid conjugates

THC–amino acid conjugates showed significantly (p < 0.05)higher antimutagenic activity against sodium azide induced muta-tion in both the tester strains (S. typhimurium TA 98 and TA 1538)than THC (1) at all the tried concentrations (313–7500 lg/plate,Fig. 3). The highest antimutagenic activity was shown by 2gagainst TA 98 strain, whereas 2d was most effective againstTA1538 strain. THC showed strong antimutagenic activity at3750 and 1250 lg/plate concentration in TA 98 and TA 1538respectively, whereas similar activity was observed for 2c and 2dat 313 lg/plate concentration in both the strains. The antimuta-genic activity of THC and its conjugates was concentration depen-dent, and significant (p < 0.05) variation among conjugates wasobserved (up to 2500 lg/plate in TA 98 and up to 1250 lg/platein TA-1538).

Several natural compounds act as potent antimutagenic agents(Jayaprakasha, Negi, & Jena, 2006; Jayaprakasha, Negi, Jena, & Rao,2007; Negi, Jayaprakasha, & Jena, 2010). Curcumin (Goud, Polasa, &Krishnaswamy, 1993) and its amino acid conjugates (Parvathyet al., 2010) are also reported to be good antimutagens. In the pres-ent study, it was observed that the conjugation of amino acids in-creases the antimutagenic efficacy of THC, as its conjugation withamino acids may facilitate its transport across the membrane bar-rier. This would improve its intracellular uptake in bacteria as ob-served with the curcumin derivatives (Kumar, Dubey, Tripathi,Fujii, & Misra, 2000; Mishra, Narain, Mishra, & Misra, 2005).

338 J.R. Manjunatha et al. / Food Chemistry 139 (2013) 332–338

4. Conclusion

In conclusion, a facile and efficient synthetic procedure hasbeen outlined for the synthesis of several novel amino acid conju-gates of tetrahydrocurcumin (THC). The results of their in vitro bio-logical evaluation established that these amino acid conjugates ofTHC exhibited significant antibacterial and antimutagenic proper-ties. The effects revealed that the conjugation of the phenolic groupto amino acid moiety via an ester linkage improved the antibacte-rial and antioxidant attributes of THC. The molecules may have po-tential food and pharmacological applications.

Acknowledgements

JRM thanks the Indian Council of Medical Research, New Delhi,India for the scholarship. The authors are grateful to Dr. G. Venk-ateswara Rao, Director, CFTRI and Dr. M. C. Varadaraj, Head, Hu-man Resources Development, CFTRI for their support andconstant encouragement for the work. The help of Mr. PadmereMukund Lakshman in the MS studies of the compounds is grate-fully acknowledged.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.foodchem.2013.01.081.

References

Anand, P., Kunnumakkara, A. B., Newman, R. A., & Aggarwal, B. B. (2007).Bioavailability of curcumin: Problems and promises. Molecular Pharmaceutics,4, 807–818.

Bharathi, S. G., Parvathy, K. S., Srinivas, P., Indi, S. S., & Rao, K. S. (2012). Curcumin–glucoside, a novel synthetic derivative of curcumin, inhibits a-synucleinoligomer formation: Relevance to Parkinson’s disease. Current PharmaceuticalDesign, 18(1), 76–84.

Corson, T. W., & Crews, C. M. (2007). Molecular understanding and modernapplication of traditional medicines: Triumphs and trials. Cell, 130, 769–774.

Goud, V. K., Polasa, K., & Krishnaswamy, K. (1993). Effect of turmeric on xenobioticmetabolizing enzymes. Plant Food and Human Nutrition, 44, 87–92.

Hong, J., Bose, M., Ju, J., Ryu, J., Chen, X., Sang, S., et al. (2004). Modulation ofarachidonic metabolism by curcumin and related biketone derivatives: Effectson cytosolic phospholipase A2, cyclooxygenases and 5-lipoxygenase.Carcinogenesis, 25, 1671–1679.

Ikken, Y., Morales, P., Martinez, A., Marin, M. L., Haza, A. I., & Cambero, M. I. (1999).Antimutagenic effect of fruit and vegetable ethanolic extracts against N-nitrosamines evaluated by the Ames test. Journal of Agricultural and FoodChemistry, 47, 3257–3264.

Jayaprakasha, G. K., Negi, P. S., & Jena, B. S. (2006). Antioxidative and antimutagenicactivities of the extracts from the rinds of Garcinia pedunculata. Innovative FoodScience and Emerging Technologies, 7, 246–250.

Jayaprakasha, G. K., Negi, P. S., Jena, B. S., & Rao, L. J. M. (2007). Antioxidant andantimutagenic activities of Cinnamomum zeylanicum fruit extracts. Journal ofFood Composition and Analysis, 20, 330–336.

Kapoor, N., Narain, U., & Misra, K. (2007). Bio-active conjugates of curcumin havingester, peptide, thiol and disulfide links. Journal of Scientific & Industrial Research,66, 647–650.

Khopde, S. M., Priyadarshini, K. I., Guha, S. N., Satav, J. G., Venkatesan, P., & Rao, M. N.A. (2000). Inhibition of radiation-induced lipid peroxidation bytetrahydrocurcumin: Possible mechanism by pulse radiolysis. BioscienceBiotechnology, and Biochemistry, 64, 503–509.

Kumar, S., Dubey, K. K., Tripathi, S., Fujii, M., & Misra, K. (2000). Design and synthesisof curcumin-bioconjugates to improve systemic delivery. Nucleic AcidsSymposium Series, 44, 75–76.

Kumar, S., Narain, U., Tripathi, S., & Misra, K. (2001). Synthesis of curcuminbioconjugates and study of their antibacterial activities against b-lactamaseproducing microorganisms. Bioconjugate Chemistry, 12, 464–469.

Maron, D. M., & Ames, B. N. (1983). Revised methods for the Salmonellamutagenicity test. Mutation Research, 113, 173–815.

Mishra, S., Narain, U., Mishra, R., & Misra, K. (2005). Design, development andsynthesis of mixed bioconjugates of piperic acid–glycine, curcumin–glycine/alanine and curcumin–glycine–piperic acid and their antibacterial andantifungal properties. Bioorganic and Medicinal Chemistry, 13, 1477–1486.

Murugan, P., Pari, L., & Rao, C. A. (2008). Effect of tetrahydrocurcumin on insulinreceptor status in type 2 diabetic rats: Studies on insulin binding toerythrocytes. Journal of Bioscience, 33(1), 63–72.

Nakamura, Y., Ohto, Y., Murakami, A., Osawa, T., & Ohigashi, H. (1998). Inhibitoryeffects of curcumin and tetrahydrocurcuminoids on the tumor promoter-induced reactive oxygen species generation in leukocytes, in vitro and in vivo.Japan Journal of Cancer Research, 89(4), 361–370.

Negi, P. S., Jayaprakasha, G. K., & Jena, B. S. (2010). Evaluation of antioxidant andantimutagenic activities of the extracts from the fruit rinds of Garcinia cowa.International Journal of Food Properties, 13, 1256–1265.

Negi, P. S., Jayaprakasha, G. K., Rao, L. J. M., & Sakariah, K. K. (1999). Antibacterialactivity of turmeric oil: A by-product from curcumin manufacture. Journal ofAgricultural and Food Chemistry, 47, 4297–4300.

Nikaido, H., & Vaara, M. (1985). Molecular basis of bacteria outer membranepermeability. Microbiological Reviews, 49, 1–32.

Nostro, A., Germano, M. P., D’Angelo, V., Marino, A., & Cannatelli, M. A. (2000).Extraction methods and bio-autography for evaluation of medicinal plantantimicrobial activity. Letters in Applied Microbiology, 30, 379–384.

Osawa, T., Sugiyama, Y., Inayoshi, M., & Kawakishi, S. (1995). Antioxidative activityof tetrahydrocurcuminoids. Bioscience, Biotechnology, and Biochemistry, 59(9),1609–1612.

Pan, M. H., Huang, T. M., & Lin, J. K. (1999). Biotransformation of curcumin throughreduction and glucuronidation in mice. Drug Metabolism and Disposition, 27(1),486–494.

Parvathy, K. S., Negi, P. S., & Srinivas, P. (2009). Antioxidant, antimutagenic andantibacterial activities of curcumin–b-diglucoside. Food Chemistry, 115,265–271.

Parvathy, K. S., Negi, P. S., & Srinivas, P. (2010). Curcumin–amino acid conjugates:Synthesis, antioxidant and antimutagenic attributes. Food Chemistry, 120,523–530.

Parvathy, K. S., & Srinivas, P. (2008). Ultrasound-assisted reaction of 2,3,4,6-tetra-O-acetyl-a-D-glucopyranosyl bromide with potassium salt of curcumin under PTCconditions. Ultrasonics Sonochemistry, 15, 571–577.

Rao, T. S., Basu, N., & Siddiqui, H. H. (1982). Anti-inflammatory activity of curcuminanalogues. Indian Journal of Medicinal Research, 75, 574–578.

Ravindran, P., Babu, K. N., & Sivaraman, K. (2007). Turmeric: The genus Curcuma. BocaRaton, Fla.: CRC Press, Taylor & Francis, p. 1–369.

Scherrer, R., & Gerhardt, P. (1971). Molecular sieving by the Bacillus megateriumcell wall and protoplast. Journal of Bacteriology, 107, 718–735.

Sneharani, A. H., Singh, Sridevi Annapurna, Srinivas, P., & Appu Rao, A. G. (2011).Inhibition of lipoxygenase-1 by tetrahydrocurcumin. European Food ResearchTechnology, 233, 561–568.

Sugiyama, Y., Kawakishi, S., & Osawa, T. (1996). Involvement of the beta-diketonemoiety in the antioxidative mechanism of tetrahydrocurcumin. BiochemistryPharmacology, 52(4), 519–525.

Wang, Y. J., Pan, M. H., Cheng, A. L., Lin, L. I., Ho, Y. S., Hsieh, C. Y., et al. (1997).Stability of curcumin in buffer solutions and characterization of its degradationproducts. Journal of Pharmaceutical and Biomedical Analysis, 15, 1867–1876.

Yodkeeree, S., Garbisa, S., & Limtrakul, P. (2008). Tetrahydrocurcumin inhibitsHT1080 cell migration and invasion via downregulation of MMPs and uPA1.Acta Pharmacologica Sinica, 29(7), 853–860.