In Silico Kinetics and Mechanism of Interaction of cis-2-Butene-1,4-dial with 2′-DeoxycytidineLiudmyla K. Sviatenko,†,‡ Leonid Gorb,†,§,∥ Dmytro Hovorun,§,∥ and Jerzy Leszczynski*,†

†Interdisciplinary Center for Nanotoxicity, Department of Chemistry and Biochemistry, Jackson State University, 1325 J.R. LynchStreet, P.O. Box 17910, Jackson, Mississippi 39217-0510, United States‡Kirovohrad State Pedagogical University, Kirovohrad 25006, Ukraine§Institute of Molecular Biology and Genetics, NAS of Ukraine, 150 Zabolotny Str., Kyiv 03143, Ukraine∥Key Laboratory, Institute of Molecular Biology and Genetics, NAS of Ukraine, 150 Zabolotny Str., Kyiv 03143, Ukraine

*S Supporting Information

ABSTRACT: Newly proposed approach involving computa-tional analysis of multistep chemical reactions has beensuccessfully applied to study the interaction between 2′-deoxycytidine and cis-2-butene-1,4-dial, a metabolite of furan.The new method comprises a combination of few steps. Theyinclude the prediction of the reaction mechanism, calculation ofGibbs free energies for the reaction pathway, and conversion of barrier energies to rate constants. On the basis of the results ofprevious steps, corresponding kinetic equations are constructed and solved. Such a procedure allows one to indicate the definiteconcentration of reaction species (reactants, intermediates, and products) at any moment in time. Obtained results show that 2′-deoxycytidine reacts with cis-2-butene-1,4-dial to form primary products, which are represented by four polycyclic diastereomers.These primary products further transform to more stable secondary product by dehydration, which is catalyzed by acid. Theobtained data demonstrate that cis-2-butene-1,4-dial plays a key role in furan-induced carcinogenesis.

■ INTRODUCTION

cis-2-Butene-1,4-dial (BDA) represents a microsomal metabo-lite of furan,1 an industrially important chemical found incanned or jarred foods that undergo heat treatment and also incigarette smoke. Furan has been classified as reasonablyanticipated to be a human carcinogen based on sufficientevidence of carcinogenicity from studies in experimentalanimals.2,3 There are also experimental studies that indicatereactions of BDA with biomolecules suggesting its involvementin carcinogenesis induced by furan.4−13 Experimental data4−7,14

partially include the kinetics of primary BDA-adduct formationfor DNA bases, obtained by HPLC analysis. These works alsoproposed the pathway of the interaction between BDA andDNA. The present article aimed to extend the experimentalfindings by computationally designed step by step modelingand thermodynamic and kinetic characterization of those steps.Recently, we have developed the computational approach

suitable to predict the rates of decay and appearance inmultistep chemical and biological reactions.15 The suggestedmethodology consists of two steps. The first one generates amultistep Gibbs free-energy reaction profile for the trans-formations of the reagents to products based on reliable DFTlevel of theory. The second one converts the activation Gibbsfree energies to the set of rate constants, constructs thecorresponding kinetic equations, and solves them. Theoutcome of this technique is computationally simulatedkinetics, which allows one to predict the concentration of anyreagent, product, and intermediate at any period of time. The

proposed procedure extends data on the experimentallysuggested mechanism by the addition of results obtainedfrom the computationally predicted steps. This allows us toapply the suggested computational technique to study thedetails of the interaction between DNA and BDA.The first DNA site we studied was an adenine, which was

modeled by the nucleoside (dAdo). It has been found that theinteraction of dAdo with BDA results in primary dAdo-BDAadducts. This process consists of five-steps.15 BDA attaches bycarbonyl carbon to exocyclic nitrogen atom N6 of dAdo. Then,the next step involves an attack of endocyclic nitrogen atom N1

on the CC double bond. This step leads to cyclization.During the following phase, two protons transfer, and the finalattack of the hydroxyl group on the carbonyl carbon results inthe formation of primary products. The latter transforms to asecondary product by a four-step pathway which involvesopening of the furan cycle and the dehydration catalyzed byacid.The present work concentrates on the interaction of BDA

with 2′-deoxycytidine (dCyd). UV absorbance, NMR, and massspectral analysis4,5 allowed one to identify the primary reactionproducts, which are oxadiazabicyclo(3.3.0)octaimine adducts.According to experimental findings, the reaction mechanisminvolves an initial attack of the exocyclic nitrogen atom N4 ofdCyd on the carbonyl carbon of BDA (Scheme 1).5 Further,

Received: February 8, 2014Published: May 19, 2014

Article

pubs.acs.org/crt

© 2014 American Chemical Society 981 dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989

two cyclization processes occur. The first one involves theattachment of endocyclic nitrogen atom N3 to the CCdouble bond. The second one is a result of the attack of thehydroxyl group on carbonyl carbon. Consequently, based onthe obtained data a general mechanism of interaction with BDAfor dAdo and dCyd has been suggested.There are eight possible diastereomers for the dCyd-BDA

adduct. However, four structures with a trans configurationabout the C10 and C11 ring junction did not form due to ringstrain.4 The existence of four structures with a cis conformationwas confirmed by NMR spectral data.4 The interconversion ofthe diastereomers may be explained by equilibrium hydrolysisof the hemiacetal. The half-life of the primary dCyd-BDAproducts at physiological pH is 275 h.4 These products mayinduce point mutations. There is no experimental data relatedto products of dCyd-BDA primary adduct decomposition.However, there is evidence of the rearrangement of structurallysimilar 4-oxo-nonenal-dCyd primary adducts to the more stablesecondary etheno-2′-deoxycytidine adduct by dehydration.16

This study is aimed at investigating the mechanism andkinetics of the process of interaction between dCyd and BDAwith an accuracy close to that of experimentally observed dataand to compare the results with previously reported data fordAdo-BDA adduct formation.15

■ COMPUTATIONAL PROCEDURES

All calculations were carried out using the Gaussian 09 programpackage.17 All molecular geometries (reagents, intermediates,transition states, and products) were fully optimized at theM05-2X/6-311+G(d) level of theory18 and characterized asminima (no imaginary frequencies) or as transition states (oneimaginary frequency) by calculations of analytic harmonicvibrational frequencies. A water molecule was explicitlyinvolved in the reaction process as a catalyst. To take intoaccount an aqueous medium, the single-point computations forM05-2X/6-311+G(d) gas-phase optimized geometries werecarried out using PCM method. Single point calculations werechosen because M05-2X gas-phase optimization and PCM/M05-2X optimization in the solvent resulted in similargeometry for several tested reactive species and also due tothe fact that gas-phase optimization needs less computationalresources than does the solvent. In the case of the acid-catalyzed mechanism, a more detailed account of solvation isneeded; therefore, full optimization at the PCM/M05-2X/6-311+G(d) level was performed. IRC for key structures wascalculated to confirm that transition states connect to both thereactants and products. Structures of all transition states areincluded in the Supporting Information, represented by theirgeometries in Cartesian coordinates.

Time dependent (TD) density functional theory at thePCM/M05-2X/6-311+G(d) level was used to simulate theUV−vis spectra of dCyd and dCyd-BDA adducts. The outputhas been visualized with the GausView 5 program, which is astandard tool to visualize the results obtained by the Gaussian09 set of programs. The same level of theory was applied tocompute the NMR spectra of products using the GIAOmethod.Reaction kinetics was obtained by solving the system of

differential eqs 1 with a Mathcad 15 program.

∑ ∑ ∑= −

≠ ≠

nt

k n n kddi

i

j

j i

ji j il

l i

il

( ) ( ) (1)

∑ ∑ ∑= −

≠ ≠

nt

k n n n k nddi

i

j

j m i

ji j m il

l o i

il o

( , ) ( , )

The rate constants for uni- and bimolecular reactions werecomputed according to eqs 2 and 3.

= · · −Δ −⧧k

k Th

e (s )G RTuni

/ 1T

(2)

= · · · · ·−Δ − −⧧⎜ ⎟⎛⎝

⎞⎠k

k Th

ec1

(L mol s )G RTbi

/ 1 1T

(3)

where k is the Boltzmann constant, T the temperature, h thePlanck constant, ΔGT

≠ the Gibbs free energy, R the universal gasconstant, and c the transformation coefficient equal to 1 mol·L−1.The calculation of kinetics was performed for experimental

conditions (temperature is 310 K, and initial concentrations ofdCyd and BDA are equal to 5 mmol·L−1 and 25 mmol·L−1,respectively).

■ RESULTS AND DISCUSSIONThe simulated UV−vis and NMR spectra of dCyd and dCyd-BDA products were compared with experimental data toconfirm the suggested geometrical structures for adducts. Thecalculated 13C and 1H chemical shifts for primary dCyd-BDAadducts in DMSO showed a good agreement with experimentalones (see Tables S1 and S2 of Supporting Information).4 Thepredicted UV−vis spectra for primary dCyd-BDA adducts inwater match the experimental data4,5 (see Figure S1 ofSupporting Information). Absorption maxima at 252 nm(exp. 278 nm) was found for all primary adducts. SecondarydCyd-BDA adduct appears in a UV−vis spectrum at λmax 254nm with higher absorbance as compared to that of the primary

Scheme 1. Mechanism of Interaction of dCyd with BDA

Chemical Research in Toxicology Article

dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989982

Figure 1. PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) computer generated initial pathways of the dCyd reaction with BDA and thecorresponding Gibbs free energy diagram.

Figure 2. PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) computer generated pathways of INT1 and INT2 transformations and correspondingGibbs free energy diagrams.

Figure 3. PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) computer generated pathways of INT3-INT6 transformations and correspondingGibbs free energy diagrams.

Chemical Research in Toxicology Article

dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989983

adducts. Thus, the simulated UV−vis and NMR spectraconfirm the geometrical structure of dCyd-BDA adducts.Reaction Mechanism of Primary dCyd-BDA Adducts

Formation. BDA initially attaches by carbonyl carbon to theexocyclic N4 atom of dCyd. This process is accompanied by aproton transfer between nitrogen and oxygen atoms and resultsin the formation of two stereoisomeric intermediates INT1 and

INT2 with similar stability (Figure 1). The Gibbs free energyvalues for this step are close to the ones reported earlier fordAdo. The formation of the C−N bond and breakage of theC−O π bond are found to occur before the proton transferbetween heteroatoms.Further, intermediates INT1 and INT2 undergo cyclization,

similar to the case of dAdo. A five-membered cycle appears as a

Figure 4. PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) computer generated pathways of INT7 and INT8 transformations and correspondingGibbs free energy diagrams.

Chemical Research in Toxicology Article

dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989984

result of the intermolecular attack of an endocyclic N3 atomonto the ethene carbon atom. The ring closing for INT1 yieldstwo stereomeric intermediates, INT3 and INT4 (Figure 2).Transformation of INT2 leads to intermediates INT5 andINT6. The predicted Gibbs free activation energy for thesepathways is less than 20 kcal/mol.These zwitterionic intermediates INT3-INT6 are unstable

and quickly transform to the more stable cycle-open formsINT7 and INT8 by proton transfer from the oxygen atom tocarbon atom (Figure 3). The Gibbs free energy activationbarriers for these stages are below 14 kcal/mol. This step variesfrom the previously studied dAdo mechanism. In the case ofdAdo, zwitterionic intermediates do not lose their five-membered ring during proton transfer. In addition, this stepis endothermic for dAdo but exothermic for dCyd.Since the intermediates INT7 and INT8 are cycle-open

forms, they have two carbonyl groups which can be attacked bythe nitrogen atom of imine group. This results in the formationof five- and six-membered rings (Figure 4). While in the case ofdAdo, the similar intermediates maintain a five-membered ring.As a result, the intermediates INT7 and INT8 transform to fourisomeric intermediates (INT9, INT10, INT12, and INT13)with five-membered rings and to two isomeric intermediates(INT11 and INT14) with six-membered rings. All thesecyclization processes occur rapidly and lead to stableintermediates.Further cyclization in intermediates INT9 and INT12 takes

place through the attachment of the hydroxyl oxygen tocarbonyl carbon with a corresponding proton transfer. Thesecyclization processes pass through Gibbs free energy activationbarriers of 20−22 kcal/mol and lead to the formation ofproducts PR15 and PR16, and products PR17 and PR18(Figure 4). These four products are diastereomers with similarstability. They have different orientations of the hydroxyl groupand five-membered furan ring. The cyclization in INT10 andINT13 intermediates is not possible due to the steric separationof hydroxyl and carbonyl groups. The departure of the watermolecule is also unlikely to occur because this pathway requireshigh activation Gibbs free energies of 48−51 kcal/mol (Figure4).By comparing data for dAdo and dCyd, it is easy to see that

primary product formation occurs in a similar manner for bothnucleosides. The free activation energy barriers in most caseshave the same values. In addition, the unfeasibility of secondary

adduct formation by dehydration of INT10 and INT13intermediates is also revealed for dAdo.On the basis of the obtained results, one suggests that the

mechanism of interaction between dCyd and BDA, which leadsto the formation of primary products (PR15-PR18), consists offive steps. The first step is the rate limiting one. Figure 5displays the Gibbs free energy profile of the reaction that startsfrom the reactants and terminates at the primary products.To acquire the kinetic picture of the whole reaction process,

the values of Gibbs free activation energy displayed in Figures1−4 were converted to the set of rate constants (Table S3 asSupporting Information). On the basis of these rate constants,the kinetic eqs (Scheme S1, Supporting Information) whichdescribe the chemical processes were obtained. This allows oneto develop kinetics for reactants, intermediates, and products.A comparison between experimental and simulated kinetics

suggests that the calculated rate of chemical reactions shown inFigures 1−4 is considerably reduced compared to theexperimentally found ones. In our opinion, the main sourceof the discrepancy is the influence of water bulk as it waspostulated by us earlier.15 This effect may be partially correctedby applying a model with a larger, explicit hydration shell (seeFigure S2 in Supporting Information). However, calculationswith a larger explicit hydration shell need a large amount ofcomputation time. For the present article, only one or twoexplicit water molecules were used, assuring correct trends forthe studied phenomena. Application of a single scaling factorresulted in a 35% decrease of all calculated activation energiesshown in Figures 1−4. This provides an excellent agreementbetween the scaled computationally predicted kinetics and theexperimental one (see Figure 6) both for the decay of reagent(dCyd) and accumulation of products (PR15-PR18). It shouldbe noted that also in the case of dAdo the 33% reduction of allpredicted activation energies has been applied to simulateexperimental data.The computed full set of the rate constants allows us to

analyze the appearance and decay of each reaction species ofthe multistep chemical reaction and use it to predict theproducts ratio. As can be seen from Figure S3 (SupportingInformation), the major products are PR16 and PR18.

Reaction Mechanism for the Transformation ofPrimary dCyd-BDA Adducts into a Secondary one. Theproducts PR15-PR18 are stable. As we already mentionedabove, there is literature evidence that at 37 °C, the half-lives of

Figure 5. Relative Gibbs free energy profiles (kcal/mol) for the most favorable pathways of the primary dCyd-BDA adduct formation calculated atthe PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) level.

Chemical Research in Toxicology Article

dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989985

primary dCyd-BDA adducts are 96 h at pH 2.8 and 275 h at pH7.4.4 However, there is no information about products of theirtransformation. As mentioned above, a rearrangement of the 4-oxo-nonenal-dCyd adduct, which is structurally similar todCyd-BDA adducts, occurs through a dehydration reaction andleads to the etheno-2′-deoxycytidine adduct.16 We also decidedto study this reaction path because primary dAdo-BDAproducts transform to a secondary one by dehydration.Dehydration of the primary dCyd adducts proceeds throughfuran ring-opening and a further loss of a water molecule. Wefound that this pathway has to proceed through the highactivation barrier, which is more than 40 kcal/mol (Figure 7).Thus, dehydration is unfavorable to occur.

According to experimental evidence, an acid decreases half-lives for primary adducts.4 In addition, an acid-catalyzedmechanism for the dehydration of 4-oxo-nonenal-dCyd adductwas proposed.16 Furthermore, primary dAdo-BDA productsrearrangement is facilitated by acid. Therefore, we modeled themechanism of transformation of primary dCyd-BDA adductsinto a secondary one using an acid as a catalyst. An acetic acidmolecule was chosen for this purpose. The protonation of theopen form of the primary products INT9 and INT12 takesplace onto nitrogen atoms and results in the formation ofpositively charged intermediates INT21 and INT22 (Figure 8).Further water molecule abstraction leads to intermediatesINT23 and INT24, which transform into secondary productPR20 through a deprotonation reaction.We concluded that dehydration of primary dCyd-BDA

adducts catalyzed by acid is more favorable as compared withthe water-catalyzed mechanism. This is in accordance withexperimental observations.4 The primary → secondary dAdo-BDA adduct transformation follows the same pathway as thedCyd-BDA one. The difference is in the free activation energyvalues for the dehydration step, which is about 3 kcal/molhigher in the case of dCyd.As earlier, the calculated relative Gibbs free energies of

activation barriers were converted to rate constants which wereinvolved in a system of kinetic equations (Scheme S2 and TableS4 of Supporting Information). The half-life for the trans-formation of primary dCyd-BDA adducts to a secondary one isexperimentally found to be 96 h.4 Applying a single scalingfactor to reduce the Gibbs free energy barriers by 42%, we

Figure 6. Kinetic plot of the primary dCyd-BDA adducts (PR15-PR18) formation reaction calculated at 37 °C at the PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) level.

Figure 7. PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) computer generated pathways of primary → secondary dCyd-BDA adducttransformation and corresponding Gibbs free energy diagram.

Chemical Research in Toxicology Article

dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989986

found that the calculated half-time (90 h) quite well matchesthe experimental one. In the case of dAdo, the correspondingreduction of the computational data amount to 44%. Thecomputationally predicted kinetic plots for the transformationof primary dCyd-BDA adducts into the secondary product aredisplayed in Figure 9 and Figures S4 and S5 (as SupportingInformation).The comparison of the results described above with the ones

presented in our earlier work15 allows us to conclude that dAdoand dCyd react with BDA in the same manner. It is not

surprising because both of these compounds have a similarreactive site moiety. The zwitterionic intermediates are morestable in the case of dAdo because the purine ring system betterstabilizes charge delocalization than the pyrimidine one. As canbe seen from the atomic charges and bond lengths (Figure S6as Supporting Information) in the case of the dAdo-BDAintermediate, the formed C2−N bond is shorter and stronger,the negative charge of the carbon atom C3 is more delocalizedthrough the moiety C3C4O, the C2−C3 and C4−O bonds arelonger, and the C3−C4 bond is shorter than that in case of

Figure 8. PCM/M05-2X/6-311+G(d) computer generated pathways of acid-catalyzed primary → secondary dCyd-BDA adduct transformation andcorresponding Gibbs free energy diagram.

Chemical Research in Toxicology Article

dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989987

dCyd-BDA intermediate. The formation of unstable zwitterionsin the case of dCyd follows by cycle-open intermediates. Theprimary dCyd adducts are slightly more stable, compared to thereactant, than in the case of dAdo reactions.The main difference between processes considered for these

two nucleosides is the stability of primary products, which isgreater in the case of dCyd. This is the reason for fasterformation of primary products and their slower transformationinto the secondary one for dCyd, as compared with dAdo. Theacid-catalyzed mechanism passes through the step ofcarbocation formation. In the case of dAdo, the carbocationand the transition state that leads to this intermediate are morestabilized because of the conjugation system which is larger indAdo, as compared to that in dCyd.

■ CONCLUSIONSThe mechanism and kinetics of interaction between dCyd andBDA were investigated using computational methods. Thestructure of the proposed primary adducts was supported bythe simulated UV−vis and NMR spectra. It was found thatinteraction of dCyd with BDA passes through a five stepprocess and results in primary adducts formation. This processinvolves the initial attaching of an exocyclic nitrogen atom N4of dCyd to a carbonyl carbon of BDA. Further, two consequentcyclizations take place. The first cycle forms owing to the attackof endocyclic nitrogen atom N3 on the double CC bond, andthe second one is a result of the addition of hydroxyl to thecarbonyl group. The transformation of the primary products toa secondary one consists of four steps. They include thedehydration of the ring-opening form of adducts, which iscatalyzed by acid. The calculated results confirm that BDAforms a stable adduct with dCyd under physiologicalconditions. This suggests that furan through its metabolitemay cause mutagenesis. A comparison of thermodynamics andkinetics of dAdo and dCyd reactions with BDA allows one toconclude that these nucleosides have a similar mechanism ofinteraction with BDA.

■ ASSOCIATED CONTENT*S Supporting InformationPCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) calculatedUV−vis spectra for dCyd and primary and secondary dCydadducts of BDA; PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) calculated 13C NMR spectra for primary dCydadducts of BDA in DMSO and experimental data; PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) calculated 1H NMR

spectra for primary dCyd adducts of BDA in DMSO andexperimental data; system of differential equations for thekinetic model of primary adduct formation; PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) calculated Gibbs free en-ergies (kcal/mol) and rate constants (s−1, L·mol−1·s−1) for thekinetic model of primary adduct formation; PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) level calculated geometriesand relative Gibbs free energies for transition state TS1(Reactants → INT1); the system of differential equations forthe kinetic model of primary → secondary adduct trans-formation; the PCM/M05-2X/6-311+G(d) calculated Gibbsfree energies (kcal/mol) and rate constants (s−1) for the kineticmodel of primary → secondary adduct transformation; plots ofthe concentration versus time for dCyd and primary adductscalculated at 37 °C at the PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) level; plots of the concentration versus timefor primary and secondary adducts calculated at 37 °C at thePCM/M05-2X/6-311+G(d) level; the plots of the concen-tration versus time for intermediates calculated at 37 °C at thePCM/M05-2X/6-311+G(d) level; the PCM/M05-2X/6-311+G(d)//M05-2X/6-311+G(d) level calculated structuresof dAdo-BDA and dCyd-BDA intermediates with selectedatomic charges and bond lengths; and the geometries oftransition states, intermediates, and products in Cartesiancoordinates. This material is available free of charge via theInternet at http://pubs.acs.org.

■ AUTHOR INFORMATION

Corresponding Author*Phone: 601-979-3482. Fax: 601-979-7823. E-mail: jerzy@icnanotox.org.

FundingThis research was funded by HRD-0833178 from the NSFCREST program. Computation time was provided by theExtreme Science and Engineering Discovery Environment(XSEDE) by National Science Foundation Grant NumberOCI-1053575 and XSEDE award allocation NumberDMR110088, and by the Mississippi Center for SupercomputerResearch.

NotesThe authors declare no competing financial interest.

■ ABBREVIATIONS

BDA, cis-2-butene-1,4-dial; dAdo, 2′-deoxyadenine; dCyd, 2′-deoxycytidine; DFT, density functional theory; DMSO,dimethyl sulfoxide; DNA, deoxyribonucleic acid; GIAO,gauge-including atomic orbital; HPLC, high performance liquidchromatography; INT, intermediate; NMR, nuclear magneticresonance; PCM, polarizable continuum model; PR, product;UV−vis, ultraviolet−visible spectroscopy

■ REFERENCES(1) Chen, L.-J., Hecht, S. S., and Peterson, L. (1995) Identification ofcis-2-butene-1,4-dial as a microsomal metabolite of furan. Chem. Res.Toxicol. 8, 903−906.(2) (2011) Furan, in Report on Carcinogens, 12th ed., pp 205−207,U.S. Department of Health and Human Services, Public HealthService, National Toxicology Program, Washington, DC.(3) (2008) Exploratory Data on Furan in Food: Individual FoodProducts, U.S. Food and Drug Administration, Silver Spring, MD.http://www.fda.gov.

Figure 9. Kinetic plot for the primary → secondary dCyd-BDA adducttransformation reaction calculated at 37 °C at the PCM/M05-2X/6-311+G(d) level.

Chemical Research in Toxicology Article

dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989988

(4) Gingipalli, L., and Dedon, P. C. (2001) Reaction of cis- and trans-2-butene-1,4-dial with 2′-deoxycytidine to form stable oxadiazabicy-clooctaimine adducts. J. Am. Chem. Soc. 123, 2664−2665.(5) Byrns, M. C., Predecki, D. P., and Peterson, L. A. (2002)Characterization of nucleoside adducts of cis-2-butene-1,4-dial, areactive metabolite of furan. Chem. Res. Toxicol. 15, 373−379.(6) Byrns, M. C., Vu, C. C., and Peterson, L. A. (2004) Theformation of substituted 1,N6-etheno-2′-deoxyadenosine and 1,N2-etheno-2′-deoxyguanosine adducts by cis-2-butene-1,4-dial, a reactivemetabolite of furan. Chem. Res. Toxicol. 17, 1607−1613.(7) Byrns, M. C., Vu, C. C., Neidigh, J. W., Abad, J.-L., Jones, R. A.,and Peterson, L. A. (2006) Detection of DNA adducts derived fromthe reactive metabolite of furan, cis-2-butene-1,4-dial. Chem. Res.Toxicol. 19, 414−420.(8) Peterson, L. A. (2013) Reactive metabolites in the biotransfor-mation of molecules containing a furan ring. Chem. Res. Toxicol. 26, 6−25.(9) Phillips, M. B., Sullivan, M. M., Villalta, P. W., and Peterson, L. A.(2014) Covalent modification of cytochrome c by reactive metabolitesof furan. Chem. Res. Toxicol. 27, 129−135.(10) Peterson, L. A., Phillips, M. B., Lu, D., and Sullivan, M. M.(2011) Polyamines are traps for reactive intermediates in furanmetabolism. Chem. Res. Toxicol. 24, 1924−1936.(11) Lu, D., and Peterson, L. A. (2010) Identification of furanmetabolites derived from cysteine−cis-2-butene-1,4-dial−lysine cross-links. Chem. Res. Toxicol. 23, 142−151.(12) Marinari, U. M., Ferro, M., Sciaba, L., Finollo, R., Bassi, A. M.,and Brambilla, G. (1984) DNA-damaging activity of biotic andxenobiotic aldehydes in Chinese hamster ovary cells. Cell Biochem.Funct. 2, 243−248.(13) Peterson, L. A., Naruko, K. C., and Predecki, D. (2000) Areactive metabolite of furan, cis-2-butene-1,4-dial, is mutagenic in theames assay. Chem. Res. Toxicol. 13, 531−534.(14) Vu, C. C., and Peterson, L. A. (2005) Synthesis of a 2′-deoxyguanosine adduct of cis-2-butene-1,4-dial, a reactive metaboliteof furan. Chem. Res. Toxicol. 18, 1012−1017.(15) Sviatenko, L., Gorb, L., Hovorun, D., and Leszczynski, J. (2012)Interaction of 2′-deoxyadenosine with cis-2-butene-1,4-dial: computa-tional approach to analysis of multistep chemical reactions. J. Phys.Chem. A 116, 2333−2342.(16) Pollack, M., Oe, T., Lee, S. H., Silva Elipe, M. V., Arison, B. H.,and Blair, I. A. (2003) Characterization of 2′-deoxycytidine adductsderived from 4-oxo-2-nonenal, a novel lipid peroxidation product.Chem. Res. Toxicol. 16, 893−900.(17) Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E.,Robb, M. A., Cheeseman, J. R., Scalmani, G., Barone, V., Mennucci, B.,Petersson, G. A., Nakatsuji, H., Caricato, M., Li, X., Hratchian, H. P.,Izmaylov, A. F., Bloino, J., Zheng, G.,Sonnenberg, J. L., Hada, M.,Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima,T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Montgomery, J. A., Jr.,Peralta, J. E., Ogliaro, F., Bearpark, M., Heyd, J. J., Brothers, E., Kudin,K. N., Staroverov, V. N., Kobayashi, R., Normand, J., Raghavachari, K.,Rendell, A., Burant, J. C., Iyengar, S. S., Tomasi, J., Cossi, M., Rega, N.,Millam, J. M., Klene, M., Knox, J. E., Cross, J. B., Bakken, V., Adamo,C., Jaramillo, J., Gomperts, R., Stratmann, R. E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J. W., Martin, R. L., Morokuma,K., Zakrzewski, V. G., Voth, G. A., Salvador, P., Dannenberg, J. J.,Dapprich, S., Daniels, A. D., Farkas, O., Foresman, J. B., Ortiz, J. V.,Cioslowski, J., and Fox, D. J. (2009) Gaussian 09, revision A.01,Gaussian Inc., Wallingford, CT.(18) Zhao, Y., and Truhlar, D. G. (2008) Density functionals withbroad applicability in chemistry. Acc. Chem. Res. 41, 157−167.

Chemical Research in Toxicology Article

dx.doi.org/10.1021/tx5000427 | Chem. Res. Toxicol. 2014, 27, 981−989989