lable at ScienceDirect

European Journal of Medicinal Chemistry 92 (2015) 693e699

Contents lists avai

European Journal of Medicinal Chemistry

journal homepage: http: / /www.elsevier .com/locate/ejmech

Short communication

Monocarbonyl analogs of curcumin inhibit growth of antibioticsensitive and resistant strains of Mycobacterium tuberculosis

Patrick R. Baldwin a, Analise Z. Reeves b, c, Kimberly R. Powell d, Ruth J. Napier c,Alyson I. Swimm d, Aiming Sun a, Kyle Giesler a, Bettina Bommarius d,Thomas M. Shinnick b, James P. Snyder a, Dennis C. Liotta a, Daniel Kalman d, *

a Department of Chemistry, Emory University, Atlanta GA 30322, USAb Division of Tuberculosis Elimination, Centers for Disease Control and Prevention, Atlanta GA 30333, USAc Microbiology and Molecular Genetics Graduate Program, Emory University School of Medicine, Atlanta GA 30322, USAd Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta GA 30322, USA

a r t i c l e i n f o

Article history:Received 11 November 2014Received in revised form5 January 2015Accepted 10 January 2015Available online 10 January 2015

Keywords:CurcuminCurcumin analogsTuberculosisMycobacteria

* Corresponding author. Emory University SchooStreet, Whitehead Research Building #144, Atlanta, G

E-mail address: dkalman@emory.edu (D. Kalman)

http://dx.doi.org/10.1016/j.ejmech.2015.01.0200223-5234/© 2015 Published by Elsevier Masson SAS

a b s t r a c t

Tuberculosis (TB) is a major public health concern worldwide with over 2 billion people currentlyinfected. The rise of strains of Mycobacterium tuberculosis (Mtb) that are resistant to some or all first andsecond line antibiotics, including multidrug-resistant (MDR), extensively drug resistant (XDR) and totallydrug resistant (TDR) strains, is of particular concern and new anti-TB drugs are urgently needed. Cur-cumin, a natural product used in traditional medicine in India, exhibits anti-microbial activity that in-cludes Mtb, however it is relatively unstable and suffers from poor bioavailability. To improve activityand bioavailability, mono-carbonyl analogs of curcumin were synthesized and screened for their capacityto inhibit the growth of Mtb and the related Mycobacterium marinum (Mm). Using disk diffusion andliquid culture assays, we found several analogs that inhibit in vitro growth of Mm and Mtb, includingrifampicin-resistant strains. Structure activity analysis of the analogs indicated that Michael acceptorproperties are critical for inhibitory activity. However, no synergistic effects were evident between themonocarbonyl analogs and rifampicin on inhibiting growth. Together, these data provide a structuralbasis for the development of analogs of curcumin with pronounced anti-mycobacterial activity andprovide a roadmap to develop additional structural analogs that exhibit more favorable interactions withother anti-TB drugs.

© 2015 Published by Elsevier Masson SAS.

1. Introduction

TB is a major public health concern, with over 2 billion peoplecurrently infected, 8.6 million new cases per year, and more than1.3 million deaths per year. The current drug regimen combinationfor TB consists of isoniazid, rifampicin, ethambutol, and pyr-azinamide, administered over six months [1,2]. Although thistreatment has a high success rate, the utility of this regimen islimited by compliance issues, which has resulted in the rise ofstrains that are resistant to some or all of the first and second lineantibiotics [3]. These strains, called MDR, XDR, and TDR Mtb, haveworse disease outcome [4]. Recent efforts in TB drug development

l of Medicine, 615 MichaelA 30322, USA..

.

have resulted in the discovery of new therapeutics includingbedaquiline, which retain activity against MDR and XDR strains.However, additional drugs are urgently needed.

Natural products and their plant-derived analogs are often asource of drugs or drug templates with limited toxicity, which hasthe potential to mitigate compliance issues during protractedadministration. One natural product candidate of interest is cur-cumin [1,7-bis(4-hydroxy-3-methoxy phenyl)-1,6-heptadiene-3,5-dione], a phenolic compound originally extracted from the plantCurcuma longa and the primary component of the spice turmeric[5,6] (Fig. 1). For centuries, various Asian cultures have used cur-cumin as a traditional medicine to treat numerous disorders,particularly those associated with the skin and digestive tract.Curcumin has been found to have anti-cancer and anti-inflammatory properties, but the comparative action of mono-carbonyl derivatives demonstrates the superior efficacy of this class

Fig. 1. Monoketone curcumin analogs evaluated for anti-mycobacteria properties.

P.R. Baldwin et al. / European Journal of Medicinal Chemistry 92 (2015) 693e699694

of molecules [7e10]. Moreover, curcumin inhibits the growth ofvarious microorganisms including Escherichia coli, Bacillus subtilis,Helicobacter pylori, and Mtb [11e13]. In addition, curcumin actssynergistically with co-administered antibiotics to suppress growthof Staphylococcus aureus in vitro [14]. While curcumin inhibitsgrowth of various bacterial species, this effect requires a relativelyhigh inhibitory concentration [IC] compared to other antimicrobialagents, which is not achievable in vivo due to limited bioavailability[15] and chemical instability [16e18].

One approach to overcome these limitations and improve theinhibitory activity of curcumin is to develop structural analogs[10,19,20]. Curcumin has three sectors that can be targeted formodification: the b-diketone moiety, the aromatic rings, and theflanking double bonds conjugated to the b-diketone moiety. Weevaluated a series of monocarbonyl analogs (Fig. 1) for their ca-pacity to inhibit growth of the pathogenic mycobacteria Mtb andMm.

2. Experimental procedures

2.1. Chemicals

All compounds were prepared by previously described meth-odology [20,21]. Briefly, the monocarbonyl analogs were generatedby allowing the ketones to react with a variety of aromatic alde-hydes under basic aldol condensation conditions. The structures ofthe final products (>95% pure, Fig. 1) have been characterized bystandard analytical techniques as previously reported: UBS-109[22e25], ECMN-909 [25], ECMN-951 [25], EF-31 [21,23,26], SEF-31 [21,27], EF-24 [20,21,28], U2-289 [29] and U2-260 [30]. Curcu-min (�94% curcuminoid content, �80% curcumin), Rifampicin anddimethyl sulfoxide (DMSO) were purchased from Sigma (St. Louis,MI).

2.2. Bacterial strains, and growth conditions

Mycobacterium tuberculosis strain H37Rv, H37Rv-RifR, and Bei-jing F2, and Mycobacterium marinum strain 1218R were grown inDifco Middlebrook 7H9 broth (Becton, Dickinson, and Company,Sparks, MD) supplemented with BBL Middlebrook ADC Enrichment(Becton, Dickinson, and Company, Sparks, MD) and 0.5% Tween 80(Mtb) or 0.025% Tween 80 (Mm) (7H9-ADC broth). Difco Mid-dlebrook 7H10 agar (Becton, Dickinson, and Company, Sparks, MD),supplemented with 10% oleic acid-albumin-dextrose-catalase(7H10-OADC) was used for Mm. Stocks of Mm were grown at30 �C in 5% CO2 until OD600 of 0.8, centrifuged, the supernatantremoved, and the bacteria resuspended in fresh 7H9 broth. Aliquotswere stored at �80 �C. For some experiments, strains with muta-tions that rendered M. marinum (1218RrifR) or Mtb (H37RvrpoBH526Y) resistant to rifampicin were used [31].

2.3. Disk diffusion assay

Mm was grown in 7H9 broth until it reached an OD600 of 0.35.The culture was diluted to OD600 of 0.04, and 100 mL of the dilutedstock was spread on 7H10 plates. The plates were allowed to air dryfor 10 min. Next, sterilized round filter paper (6 mm diameter BBLdisks; Becton, Dickinson, Sparks, MD) was placed in the center ofeach plate, to which 10 ml of a particular curcumin analog (100 mMin DMSO or sterilized water) was added. Three plates were assessedfor each analog. The plates were then incubated at 30 �C in 5% CO2for 7 days, after which time the extent of the zone lacking bacteriaaround the filter was measured (“zone of inhibition”).

2.4. Determination of 50% inhibitory concentration (IC50) forcurcumin analogs against Mm and Mtb

Mm was grown in 7H9-ADC broth until OD600 of 0.35 and then

P.R. Baldwin et al. / European Journal of Medicinal Chemistry 92 (2015) 693e699 695

diluted to an OD600 of 0.04. Cultures were grown for 72 h in 7H9-ADC broth containing no drug, 1% DMSO (vehicle control), or cur-cumin analogs (UBS-109, EF-24, EF-31, and ECMN-909 in DMSO) inconcentrations of 0.780 mM,1.560 mM, 3.125 mM, 6.250 mM,12.5 mM,25 mM, 50 mM, or 100 mM in duplicate. The percent inhibition wasnormalized to the level achievedwith the vehicle alone. Mtb strainsH37Rv, H37Rv-RifR, and Beijing F2 were grown to mid-log in 7H9-ADC broth and then diluted to an OD600 of 0.05. Cultures weregrown in 7H9-ADC broth containing no drug, 1% DMSO, or curcu-min analogs (UBS-109 and EF-24) in concentrations of 1 mM, 5 mM,10 mM, 20 mM, or 50 mM in triplicate. Growth was determined bymeasuring the OD600 of cultures for 14 days using aspectrophotometer.

2.5. Determination of synergistic effects of UBS-109 and rifampicinon Mm

Cultures of Mm were grown in 7H9-ADC broth containing nodrug, 1% DMSO (vehicle control), rifampicin alone at various con-centrations (4 mg/mL, 2 mg/mL, 1 mg/mL, 0.5 mg/mL, or 0.250 mg/mL),or rifampicin at various concentrations in combination with UBS-109 at various concentrations (6.250 mM,12.5 mM, 25 mM, 50 mM, or100 mM) in duplicate. Growth was determined by measuring the

Fig. 2. Monocarbonyl curcumin analogs inhibit the growth of Mycobacterium marinum (Monocarbonyl analogs (100 mM) were added to Mm containing plate, incubated for 7 daysanalogs to inhibit growth of Mm using a liquid culture assay. Indicated concentrations of anormalized to untreated Mm culture. Data represents the mean ± SEM from three experimculture assay. Data represents the mean ± SEM from three experiments.

OD600 of cultures after 72 h using a spectrophotometer. The datawas normalized to the vehicle control. Synergy was determined bycomparing combination treatments to rifampicin-only and UBS-109-only. Statistical analysis was done using nonparametricANOVA. Values less than or equal to 0.05 were considered statis-tically significant.

3. Results

3.1. Evaluation of monocarbonyl analogs against Mm

A series of monocarbonyl analogs (Fig. 1) were synthesized andscreened for anti-mycobacterial properties using a disk diffusionassay. After 7 days, the diameter of the circular zone surroundingthe disk that remained free of bacterial growth (the “zone of inhi-bition”) was measured for each analog (Fig. 2A). Monocarbonylanalogs displayed a range of effects, with some showing little in-hibition of growth compared to curcumin (e.g. U2-260) whereasothers inhibited growth to substantially greater extent (e.g. UBS-109). By comparison, the vehicle control had no effect. To furthercharacterize the growth inhibitory effect of these analogs, Mmwascultured in liquid media in the presence of various concentrationsof UBS-109 or EF-24 for 72hr and the OD600 measured at various

Mm). (A) Initial screening of inhibitory properties analyzed by disk diffusion assay.at 32 �C, and zone of inhibition measured.; (B) Assessment of the capacity of specificnalogs were added to Mm culture, incubated for 72hr, and the OD600 measured. Dataents.; (C) Assessment of the capacity of ECMN-909 to inhibit Mm growth in a liquid

P.R. Baldwin et al. / European Journal of Medicinal Chemistry 92 (2015) 693e699696

time points (Fig. 2B). The IC50 was determined from these data to be10 mM for UBS-109 and 25 mM for EF-24.

3.2. A Michael acceptor is required for anti-mycobacterial activity

While differences in solubility and diffusion through 7H10 agarmay contribute to the variability in zones of inhibition, the ability ofcompounds with diverse substitution to inhibit Mm is most likelyinfluenced by variation in protein target-ligand interactions, as hasbeen suggested for kinase inhibition [21]. Comparison of thestructural features of the analogs (Fig. 1) as it affects bacterialgrowth indicates that variability of all three domains within themolecules, the aromatic rings, the central ring or the Michaelacceptor, can regulate antimicrobial activity. However, the Michaelacceptor, comprised of two unsaturated C]C bonds flanking thecarbonyl group, is known to be a highly active structural compo-nent of these analogs and is likely to play a key role in their anti-microbial activity [32e35].

To further elaborate the relationship between the Michaelacceptor and antimicrobial activity, we evaluated two additionalanalogs, ECMN-909 and ECMN-951, in which the conjugated un-saturated ketone is altered or eliminated. Both are metabolitesidentified in an in vitro study of UBS-109 in liver S9 fractions fromfive different species [25]. ECMN-909 contains only one doublebond flanking the carbonyl, which reduces the capacity of themolecule to undergo Michael additions. Likewise, ECMN-951 con-tains no flanking double bonds (Fig. 1), and cannot participate in aMichael reaction. In disk diffusion assays, ECMN-909 displayed thesecond largest zone of inhibition amongst all analogs (4.6 ± 0) mmcompared to 5.7 mm for UBS-109; however, no zone of inhibitionwas evident with ECMN-951. Liquid culture assays confirmed theseresults, and indicated that ECMN-909 was somewhat less effectivein inhibiting bacterial growth (IC50 ¼ ~50 mM) compared to UBS-109 (IC50 ¼ 10 mM) (Fig. 2C). Together, these data suggest that theterminal aromatic rings and the central 6-membered ring can

Fig. 3. UBS-109 does not display synergistic interactions when combined rifampicinupon the inhibition growth of Mycobacterium marinum (Mm). The effect of combina-tion of rifampicin and UBS-109 on the growth of Mm.Mmwas cultured in the presenceof high and low concentrations of rifampicin and UBS-109 for 72hr, then the OD600 wasmeasured. Data normalized to untreated Mm culture. Data normalized to untreatedMm culture. Data represents the mean ± SEM from three experiments.

influence the antimicrobial activity of UBS-109, but the Michaelacceptor is required. It also suggests that the dominant and firstreduced metabolite (ECMN-909) is likewise a reasonably effectiveanti-TB agent. We speculate that a similar train of events may applyto the other analogs in Fig. 1.

3.3. Curcumin analogs are not synergistic with rifampicin

To determine whether UBS-109 exhibits a synergistic effectwhen co-administered with a key anti-tuberculosis drug, Mm wascultured in the presence of rifampicin, UBS-109 or a combination ofthe two in various concentrations. As shown in Fig. 3, high con-centrations of rifampicin or UBS-109 inhibited growth of Mm,whereas less inhibition was evident at lower concentration ofeither drug alone (6.25 mM for UBS-109 and 0.3 mM for rifampicin).When delivered in combination at low concentrations, no signifi-cant difference was evident compared to either drug alone.Together, these data suggest that no synergistic interaction occursupon co-administration of UBS-109 and rifampicin.

3.4. Evaluation of curcumin analogs on rifampicin-resistant Mm

To determine whether curcumin analogs can be used to treatdrug-resistant mycobacterial species, we evaluated the ability ofUBS-109 to inhibit the growth of a rifampicin-resistance strain ofMm (Mmrif). Using liquid culture assays, we grew Mmrif in thepresence of DMSO, rifampicin or UBS-109. As a control, rifampicinwas found to have no effect at any concentration on inhibiting thegrowth of Mmrif. As shown in Fig. 4, UBS-109 inhibited growth ofMmrif with an IC50 of ~4 mM.

3.5. Evaluation of curcumin analogs on M. tuberculosis

To validate the effects of UBS-109 on other clinically relevantmycobacteria, we tested the effects of UBS-109 and EF-24 on severalMtb strains using the disk diffusion and liquid assays (Fig. 5) withtwo different Mtb strains, H37Rv or Beijing F2. A representative discdiffusion assay with H37Rv is shown in Fig. 5A, and UBS-109induced a marked zone of inhibition. Data from H37Rv or BeijingF2 grown in the presence of various concentrations of UBS-109 orEF-24 for two weeks and the OD600 measured daily are shown inFig. 5BeG. The data indicate that UBS-109 inhibits the growth of

Fig. 4. UBS-109 inhibits the growth of rifampicin-resistant Mycobacterium marinum(Mm). Liquid culture assay validating UBS-109 inhibits rifampicin-resistant Mm(different concentration of analogs were added to Mm culture, incubated for 72hr, thenthe OD600 was measured). Data normalized to untreated Mm culture. Data representsthe mean ± SEM from three experiments.

Fig. 5. Monoketone curcumin analogs inhibit the growth of Mycobacterium tuberculosis (Mtb). (A) Initial screening of inhibitory properties against Mtb analyzed by disk diffusionassays. 100 mM of curcumin analogs was added to Mtb strain H37Rv containing plate, incubated for 28 days at 37 �C, and zone of inhibition measured; (B) Liquid culture assayvalidating that UBS-109 inhibits H37Rv strain of Mtb. Different concentration of analogs were added to Mtb culture, incubated for 8d, then the OD600 was measured.; (C) Liquidculture assay validating that UBS-109 inhibits Beijing strain of Mtb; (D) Liquid culture assay validating that UBS-109 inhibits rifampicin-resistence strain of H37Rv Mtb; (E) Liquidculture assay validating that EF-24 inhibits H37Rv strain of Mtb; (F) Liquid culture assay validating that EF-24 inhibits Beijing strain of Mtb; (G) Liquid culture assay validating thatEF-24 inhibits rifampicin-resistence strain of H37Rv Mtb.

P.R. Baldwin et al. / European Journal of Medicinal Chemistry 92 (2015) 693e699 697

H37Rv with an IC50 of ~10 mM, and the Beijing strainwith an IC50 of20 mM (Fig. 5B, C). EF-24 also blocks Mtb, though not as effectivelyas UBS-109 (IC50 of ~20 mM for H37Rv and 50 mM for Beijing stains;Fig. 5D, E). Finally, the effect of UBS-109 and EF-24 on therifampcin-resistant Mtb (H37Rv RifR) was evaluated. UBS-109 de-pletes H37Rv RifR Mtb with an IC50 of ~7 mM (Fig. 5D). EF-24 alsoblocks the rifampicin-resistant Mtb strain, though with lower ef-ficacy compared to UBS-109 (IC50 of 20 mM; Fig. 5G). These datasuggest that curcumin analogs are as effective against Mm in cul-ture as they are against Mtb H37Rv, but less effective against MtbBeijing.

4. Discussion

The use of curcumin as a potential therapeutic for TB has beendiscounted because of poor bioavailability. One approach to over-coming this disadvantage is the synthesis and bio-evaluation of

structural variations of curcumin such as members of the mono-carbonyl family depicted in Fig. 1. Analogs developed by our groupand others display a range of pharmacological effects against avariety of diseases [7e10,22,24,26] and may have improvedbioavailability compared to curcumin [22,28,36]. To date, only onemember of this class, EF-24, has been subjected to both bioavail-ability evaluation and pharmacokinetics treatment. The bioavail-ability of the compound falls at 35% and 60% in i.p. and oralexperiments, respectively [28]. The corresponding i.p. and p.o.pharmacokinetics parameters (Tmax, Cmax, t½ and AUC; Table 1S,supporting information) for EF-31 [36] and UBS-109 [28] arequite similar to those for EF-24 suggesting a comparablebioavailability.

In the present work we show that monocarbonyl curcuminanalogs also inhibit growth of pathogenic mycobacteria species(Mtb and Mm). UBS-109 exhibited the best inhibitory effect, asdetermined by its IC50 and its activity against two species of

P.R. Baldwin et al. / European Journal of Medicinal Chemistry 92 (2015) 693e699698

pathogenic mycobacteria. UBS-109, and several other analogsdeliver significantly lower IC50 values compared to curcumin,which has been reported to inhibit growth of Mtb [13]. We haveestablished that all three structural domains of the analogs (Fig. 1)can be tailored to influence anti-mycobacterial activity. Thisobservation carries over to anti-cancer and anti-inflammatory ef-fects by this family of analogs as well [10,19,21e24,26]. First, thepresence of a mono-carbonyl group linking the aromatic rings inthese analogs improved growth inhibition in comparison to cur-cumin, which incorporates keto-enol functionality at the center ofthe structure (Fig. 1). Furthermore, the data suggests that fluoro-substitution of the aromatic ring reduces anti-mycobacterial ac-tivity by ~2-fold compared to UBS-109. Since the drug-target pro-teins of Mtb have yet to be identified, it is not yet clear what type ofprotein-ligand interactions within the bacteria are responsible forthe halogen's modest reduction of antibacterial activity.

Perhaps the most important structural characteristic of theseanalogs is the two unsaturated bonds flanking the carbonyl, whichconstitute a double Michael acceptor. In general, this domain isrequired for biological activity of curcumin analogs. While it isrecognized that these bonds are required formost biological effects,they also represent a potential limitation in the usefulness of suchcompounds as therapeutics. Michael acceptors can render com-pounds unstable and facilitate their degradation. In addition,Michael acceptors can in principle undergo numerous reactionswith proteins, which can contribute to nonspecific effects,including toxicity [32]. This limitation may likewise complicate theuse of these particular curcumin analogs in treating TB patients. Inspite of these observations, the rifamycin family of antibiotics,including Rifampicin (Fig. 3), are unsaturated amides and, thus,Michael acceptors themselves. Two clinically valuable members ofthe series, Rifabutin and Rifalazil, contain a secondmasked Michaelacceptor in their naphthoquininoid rings, a feature pictured by oneof the tautomers of Rifampicin as well. Recent re-evaluation of thepresence of Michael acceptors in potential drugs, some of which arein medical use, suggests that such drawbacks can be overcome[32e35,37]. Our data concur with previous studies by Changtamet al., who showed that the elimination of unsaturated bondswithin curcumin decreased anti-mycobacterial activity [13].

Currently, a major impediment to TB treatment is the rise ofstrains that are resistant to some or all first and second line anti-biotics. Our data indicate that monocarbonyl curcumin analogs areeffective against rifampicin-resistant strains of Mtb and Mm.Currently, drug regimens for TB consist of combinations of drugs,which reduce the likelihood of developing antibiotic resistance. Thedevelopment of new therapeutics that displays synergy with co-administered antibiotics is of obvious benefit. While attractive asa therapeutic against Mtb and Mm, no additive or synergistic effectwas evident when UBS-109 was administered in combination withrifampicin. Nevertheless, the prospect remains that examination ofadditional analogs may permit structural insights that could guidethe design of analogs that do display synergistic effects whencombined with other anti-TB drugs. In this context we anticipatesampling a wider range of classical antibiotics with a broadercollection of monocarbonyl analogs of curcumin in search of com-binations that reduce the possibility of Mtb developing resistanceto both drugs.

The present study has focused on the effects of a number ofmonocarbonyl analogs of curcumin on bacterial growth in vitro.Future work will characterize effects of these analogs on Mtbinfection in vivo. Importantly, this class of compounds has beenimplicated in regulating a myriad of cellular processes inmammalian cells [10]. Thus, the efficacy of these compounds in vivomay depend on whether these pathways impact Mtb infection. Inthis regard, curcumin has also been found to inhibit inflammatory

responses; however agents that disrupt inflammation can haveboth positive and negative effects on Mtb infections [38]. Never-theless, development of a new analog that facilities the clearance ofMtb by targeting both bacterial and host cellular pathways mayrepresent a novel means to treat Mtb strains that are resistant toconventional anti-TB drugs.

Acknowledgments

We thankmembers of the Kalman Laboratory, R. Sonowal and G.Patel, for helpful discussions.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ejmech.2015.01.020.

References

[1] Y. Zhang, W.W. Yew, Mechanisms of drug resistance in Mycobacteriumtuberculosis, Int. J. Tuberc. Lung Dis. 13 (2009) 1320e1330.

[2] K. Duncan, J.C. Sacchettini, Approaches to tuberculosis drug development, in:G.F. Hatfull, W.R. Jacobs (Eds.), Molecular Genetics of Mycobacteria, ASMPress, Washington, DC, 2000, pp. 297e307.

[3] R. Loddenkemper, B. Hauer, Drug-resistant tuberculosis: a worldwideepidemic poses a new challenge, Dtsch. Arztebl Int. 107 (2010) 10e19.

[4] D.J. Sloan, G.R. Davies, S.H. Khoo, Recent advances in tuberculosis: new drugsand treatment regimens, Curr. Respir. Med. Rev. 9 (2013) 200e210.

[5] S. Bisht, A. Maitra, Systemic delivery of curcumin: 21st century solutions foran ancient conundrum, Curr. Drug Discov. Technol. 6 (2009) 192e199.

[6] M. Singh, P. Sasi, V.H. Gupta, G. Rai, D.N. Amarapurkar, P.P. Wangikar, Pro-tective effect of curcumin, silymarin and N-acetylcysteine on antituberculardrug-induced hepatotoxicity assessed in an in vitro model, Hum. Exp. Toxicol.31 (2012) 788e797.

[7] D.K. Agrawal, P.K. Mishra, Curcumin and its analogues: potential anticanceragents, Med. Res. Rev. 30 (2010) 818e860.

[8] C. Zhao, Z. Liu, G. Liang, Promising curcumin-based drug design: mono-carbonyl analogues of curcumin (MACs), Curr. Pharm. Des. 19 (2013)2114e2135.

[9] Y. Zhang, C. Zhao, W. He, Z. Wang, Q. Fang, B. Xiao, Z. Liu, G. Liang, S. Yang,Discovery and evaluation of asymmetrical monocarbonyl analogs of curcuminas anti-inflammatory agents, Drug Des. Dev. Ther. 8 (2014) 373e382.

[10] D. Shetty, Y. Kim, H. Shim, J. Snyder, Eliminating the heart from the curcuminmolecule: monocarbonyl curcumin mimics (MACs), Molecules 20 (2015)249e292.

[11] Bhawana, R.K. Basniwal, H.S. Buttar, V.K. Jain, N. Jain, Curcumin nanoparticles:preparation, characterization, and antimicrobial study, J. Agric. Food Chem. 59(2011) 2056e2061.

[12] R. De, P. Kundu, S. Swarnakar, T. Ramamurthy, A. Chowdhury, G.B. Nair,A.K. Mukhopadhyay, Antimicrobial activity of curcumin against Helicobacterpylori isolates from India and during infections in mice, Antimicrob. AgentsChemother. 53 (2009) 1592e1597.

[13] C. Changtam, P. Hongmanee, A. Suksamrarn, Isoxazole analogs of curcumi-noids with highly potent multidrug-resistant antimycobacterial activity, Eur. J.Med. Chem. 45 (2010) 4446e4457.

[14] K. Moghaddam, M. Iranshahi, M. Yazdi, A. Shaverdi, The combination effect ofcurcumin with different antibiotics against Staphylococcus arueus, Int. J. GreenPharmacol. 3 (2009) 141e143.

[15] P. Anand, A.B. Kunnumakkara, R.A. Newman, B.B. Aggarwal, Bioavailability ofcurcumin: problems and promises, Mol. Pharm. 4 (2007) 807e818.

[16] Y.-J. Wang, M.-H. Pan, A.-L. Cheng, L.-I. Lin, Y.-S. Ho, C.-Y. Hsieh, J.-K. Lin,Stability of curcumin in buffer solutions and characterization of its degrada-tion products, J. Pharm. Biomed. Analysis 15 (1997) 1867e1876.

[17] L. Shen, H.-F. Ji, The pharmacology of curcumin: is it the degradation prod-ucts?, Trends in Molecular Medicine, 18 138e144.

[18] O.N. Gordon, C. Schneider, Vanillin and ferulic acid: not the major degradationproducts of curcumin, Trends in Molecular Medicine, 18 361e363.

[19] C. Mosley, D. Liotta, J. Snyder, Highly active anticancer curcumin analogues,in: B. Aggarwal, Y.-J. Surh, S. Shishodia (Eds.), The Molecular Targets andTherapeutic Uses of Curcumin in Health and Disease, Springer, US, 2007, pp.77e103.

[20] B.K. Adams, E.M. Ferstl, M.C. Davis, M. Herold, S. Kurtkaya, R.F. Camalier,M.G. Hollingshead, G. Kaur, E.A. Sausville, F.R. Rickles, J.P. Snyder, D.C. Liotta,M. Shoji, Synthesis and biological evaluation of novel curcumin analogs asanti-cancer and anti-angiogenesis agents, Bioorg. Med. Chem. 12 (2004)3871e3883.

[21] A. Brown, Q. Shi, T.W. Moore, Y. Yoon, A. Prussia, C. Maddox, D.C. Liotta,H. Shim, J.P. Snyder, Monocarbonyl curcumin analogues: heterocyclic pleio-tropic kinase inhibitors that mediate anticancer properties, J. Med. Chem. 56

P.R. Baldwin et al. / European Journal of Medicinal Chemistry 92 (2015) 693e699 699

(2013) 3456e3466.[22] S. Zhu, T.W. Moore, N. Morii, R.B. Howard, R.F. Arrendale, P. Reddy, T.J. Evers,

H. Zhang, G. Sica, Z.G. Chen, A. Sun, H. Fu, F.R. Khuri, D.M. Shin, J.P. Snyder,M. Shoji, Synthetic curcumin analog UBS109 inhibits the growth of head andneck squamous cell carcinoma xenografts, Curr. Cancer Drug Targets 14(2014) 380e393.

[23] G.P. Nagaraju, S. Zhu, J. Wen, A.B. Farris, V.N. Adsay, R. Diaz, J.P. Snyder,S. Mamoru, B.F. El-Rayes, Novel synthetic curcumin analogues EF31 andUBS109 are potent DNA hypomethylating agents in pancreatic cancer, CancerLett. 341 (2013) 195e203.

[24] M. Yamaguchi, T.W. Moore, A. Sun, J.P. Snyder, M. Shoji, Novel curcuminanalogue UBS109 potently stimulates osteoblastogenesis and suppressesosteoclastogenesis: involvement in Smad activation and NF-[small kappa]Binhibition, Integr. Biol. 4 (2012) 905e913.

[25] T.W. Moore, S. Zhu, R. Randolph, M. Shoji, J.P. Snyder, Liver S9 fraction-derivedmetabolites of curcumin analogue UBS109, ACS Med. Chem. Lett. 5 (2014)288e292.

[26] A. Olivera, T.W. Moore, F. Hu, A.P. Brown, A. Sun, D.C. Liotta, J.P. Snyder,Y. Yoon, H. Shim, A.I. Marcus, A.H. Miller, T.W. Pace, Inhibition of the NF-kappaB signaling pathway by the curcumin analog, 3,5-Bis(2-pyridinylmethylidene)-4-piperidone (EF31): anti-inflammatory and anti-cancer properties, Int. Immunopharmacol. 12 (2012) 368e377.

[27] I. Landais, S. Hiddingh, M. McCarroll, C. Yang, A. Sun, M.S. Turker, J.P. Snyder,M.E. Hoatlin, Monoketone analogs of curcumin, a new class of Fanconi anemiapathway inhibitors, Mol. Cancer 8 (2009), 133e133.

[28] J. Reid, S. Buhrow, J. Gilbert, L. Jia, M. Shoji, J. Snyder, M. Ames, Mousepharmacokinetics and metabolism of the curcumin analog, 4-piperidinone,3,5-bis[(2-fluorophenyl)methylene]-acetate(3E,5E) (EF-24; NSC716993), Cancer Chemother. Pharmacol. 73 (2014) 1137e1146.

[29] A. Sun, M. Shoji, Y.J. Lu, D.C. Liotta, J.P. Snyder, Synthesis of EF24�tripeptidechloromethyl ketone: a novel curcumin-related anticancer drug deliverysystem, J. Med. Chem. 49 (2006) 3153e3158.

[30] J. Wu, Y. Zhang, Y. Cai, J. Wang, B. Weng, Q. Tang, X. Chen, Z. Pan, G. Liang,S. Yang, Discovery and evaluation of piperid-4-one-containing mono-carbonylanalogs of curcumin as anti-inflammatory agents, Bioorg. Med. Chem. 21(2013) 3058e3065.

[31] R.J. Napier, W. Rafi, M. Cheruvu, K.R. Powell, M.A. Zaunbrecher, W. Bornmann,P. Salgame, T.M. Shinnick, D. Kalman, Imatinib-sensitive tyrosine kinasesregulate mycobacterial pathogenesis and represent therapeutic targetsagainst tuberculosis, Cell Host Microbe 10 (2011) 475e485.

[32] M.H. Johansson, Reversible michael additions: covalent inhibitors and pro-drugs, Mini Rev. Med. Chem. 12 (2012) 1330e1344.

[33] C.-U. Lee, T.N. Grossmann, Reversible covalent inhibition of a protein target,Angew. Chem. Int. Ed. 51 (2012) 8699e8700.

[34] M. Hagel, D. Niu, T. St Martin, M.P. Sheets, L. Qiao, H. Bernard, R.M. Karp,Z. Zhu, M.T. Labenski, P. Chaturvedi, M. Nacht, W.F. Westlin, R.C. Petter,J. Singh, Selective irreversible inhibition of a protease by targeting a non-catalytic cysteine, Nat. Chem. Biol. 7 (2011) 22e24.

[35] M.H. Potashman, M.E. Duggan, Covalent modifiers: an orthogonal approach todrug design, J. Med. Chem. 52 (2009) 1231e1246.

[36] S. Zhu, T.W. Moore, X. Lin, N. Morii, A. Mancini, R.B. Howard, D. Culver,R.F. Arrendale, P. Reddy, T.J. Evers, H. Zhang, G. Sica, Z.G. Chen, A. Sun, H. Fu,F.R. Khuri, D.M. Shin, J.P. Snyder, M. Shoji, Synthetic curcumin analog EF31inhibits the growth of head and neck squamous cell carcinoma xenografts,Integr. Biol. 4 (2012) 633e640.

[37] I.M. Serafimova, M.A. Pufall, S. Krishnan, K. Duda, M.S. Cohen, R.L. Maglathlin,J.M. McFarland, R.M. Miller, M. Frodin, J. Taunton, Reversible targeting ofnoncatalytic cysteines with chemically tuned electrophiles, Nat. Chem. Biol. 8(2012) 471e476.

[38] K.D. Mayer-Barber, B.B. Andrade, S.D. Oland, E.P. Amaral, D.L. Barber,J. Gonzales, S.C. Derrick, R. Shi, N.P. Kumar, W. Wei, X. Yuan, G. Zhang, Y. Cai,S. Babu, M. Catalfamo, A.M. Salazar, L.E. Via, C.E. Barry Iii, A. Sher, Host-directed therapy of tuberculosis based on interleukin-1 and type I interferoncrosstalk, Nature 511 (2014) 99e103.