Absolute Configuration of (â)-Centratherin, a Sesquiterpenoid Lactone

Article pubs.acs.org/jnp

Absolute Configuration of (−)-Centratherin, a Sesquiterpenoid Lactone, Defined by Means of Chiroptical Spectroscopy Fernando M. S. Junior,†,‡ Cody L. Covington,‡ Ana Carolina F. de Albuquerque,† Jonathas F. R. Lobo,† Ricardo M. Borges,† Mauro B. de Amorim,† and Prasad L. Polavarapu*,‡ †

Instituto de Pesquisas de Produtos Naturais, Universidade Federal do Rio de Janeiro, 21941-902, Rio de Janeiro, RJ, Brazil Department of Chemistry, Vanderbilt University, Nashville, Tennessee 37235, United States

‡

S Supporting Information *

ABSTRACT: (−)-Centratherin is a bioactive sesquiterpenoid lactone, whose absolute configuration (AC) was not established, but has been proposed based on those of germacrane precursors. To verify this proposal, the experimental electronic circular dichroism (ECD), electronic dissymmetry factor (EDF), optical rotatory dispersion (ORD), vibrational circular dichroism (VCD), and vibrational dissymmetry factor (VDF) spectra of (−)-centratherin have been analyzed with the corresponding density functional theoretical predictions. These analyses suggest the AC of naturally occurring (−)-centratherin to be (6R,7R,8S,10R,2′Z).

S

centratherine using reliable stereochemical methods is important. The most widely practiced methods to determine the AC are X-ray diffraction using the Bijvoet method10 or chiroptical spectroscopy.11 Although X-ray analysis has been a powerful tool for AC assignment, most often high-quality crystals cannot be obtained and natural products usually do not contain heavy atoms, which are important features for unambiguously assigning the AC by X-ray analysis.12,13 Chiroptical spectroscopy methods based on electronic transitions, namely, electronic circular dichroism (ECD) and optical rotatory dispersion (ORD), have been around for several decades. Two other chiroptical spectroscopy methods based on vibrational transitions, namely, vibrational circular dichroism (VCD) and vibrational Raman optical activity (VROA), are relatively new methodologies and offer convenient approaches for deducing the ACs of chiral molecules. A combination of these different chiroptical methods provides a reliable approach toward stereostructural analysis.14,15 The interpretation of these data using empirical or semiempirical methods does not normally yield reliable results. Therefore, the current trend16 relies on the evaluation of the

esquiterpenoid lactones (SLs) constitute one of the largest and most widely distributed groups of cytotoxic and antitumor compounds of plant origin. The relationship between chemical structure and cytotoxic activity is based on the presence of an α-methylene-γ-lactone moiety.1 Centratherin, 1 (Scheme 1), is an SL belonging to the furanoheliangolide group, whose complex skeleton possesses, besides the α-methylene-γ-lactone moiety necessary for cytotoxic activity, an angelic ester, a primary allylic alcohol, and a 3-(2H)-furanone moiety.2 Centratherin also possesses antimicrobial,3 anti-inflammatory,4 and trypanocidal activities.5 Owing to the presence of a complex skeleton, varying structural assignments for centratherin were found in the earlier literature.6,7 In 1982, Le Quesle and co-workers, and Herz and Goedken, determined8,9 the relative configuration of centratherin (also referred to as lychnophorolide A) based on, among other experimental data, X-ray diffraction data of lychnophorolide A8 and a related compound, goyazensolide,9 2 (Scheme 1). However, its absolute configuration (AC) was not defined, but instead, the structure “was drawn to show the configuration which is in accord with the germacrane precursors from which lychnophorolide A may be regarded as having arisen”.8 Since the biological activities of many compounds come from a specific AC, verification of the proposed configuration for © XXXX American Chemical Society and American Society of Pharmacognosy

Received: June 20, 2015

A

DOI: 10.1021/acs.jnatprod.5b00546 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

In order to demonstrate that the NMR chemical shift calculation protocol is able to accurately distinguish the centratherin structure from that of a diastereomer, we also compared the experimental data obtained for the isolated compound with the calculated data for a known diastereomer, budlein A (3) (Scheme 1).29 This comparison is presented in Table S2 of the Supporting Information, along with the MAD and RMSD values. The comparison between experimental data for centratherin (δexp) and the calculated data for budlein A (δscal) yields larger MAD and RMSD values (Table S2, Supporting Information). Therefore, these data allow us to distinguish the two diastereomers and provide credence to the relative configuration assigned for the isolated compound (Scheme 1). For further verification, we additionally employed Smith and Goodman’s CP3 parameter method,30 which is based on comparing the differences in calculated chemical shifts of diastereomers with differences in experimental chemical shifts. This CP3 parameter method has been established31 to provide more reliable assignments than the statistical parameters, such as MAD and RMSD, listed in Tables S1 and S2 of the Supporting Information. The CP3 parameter, calculated by the Web applet provided by the authors,30 generated values of 0.22 and −0.57 for, respectively, correct and incorrect assignments between calculated and experimental data, which in conjunction with Bayes’ theorem has led, with 100% certainty, to the assignment of the experimental data of centratherin (1) and budlein A (3) to their proposed structures (Scheme 1). Therefore, the isolated compound can be confirmed as centratherin. Next, the analysis was focused on the determination of the AC of 1 by means of chiroptical spectroscopy. ECD Analysis. The experimental electronic absorption (EA) (Figure 1A) and ECD spectra (Figure 1B) in the 187−400 nm region measured in acetonitrile show two EA and six ECD bands. In the ECD spectrum there are three positive Cotton effects (CEs), at ∼300, 250, and 210 nm, and three negative CEs, at ∼270, 230, and 190 nm. According to previous studies, sesquiterpenoid lactones with a trans-fused α-methylene-γlactone moiety show negative CEs at 220 and 260 nm in their ECD spectra, with an α-orientation for H-7, implying a 7R AC, which is in accordance with the corresponding predicted spectrum.32−35 Accordingly the negative CEs observed at ∼230 and 270 nm may be considered to arise from the π−π* and n−π* transitions of the lactone moiety, respectively. The experimental EA and ECD spectra are compared to those calculated for the (6R,7R,8S,10R,2′Z) structure in Figure 1A and B. The agreement between experimental and calculated EA spectra is good, as the two resolved EA bands in the 200− 280 nm range of the experimental spectrum have corresponding bands in the predicted spectrum. The agreement between the experimental and predicted ECD spectra may also be considered to be good, as the six CEs in the experimental spectrum have corresponding features in the predicted spectrum. However, such visual comparison depends on specific bands being correlated and may inadvertently introduce user bias. Since predicted transition wavelengths often deviate from the experimental band positions, a scale factor is often necessary to match the predicted and experimental band positions. This is somewhat equivalent to translating the calculated spectrum along the x-axis. When alternating positive and negative CEs are present in the ECD spectrum, one can arrive at opposite conclusions depending on the direction and extent of translation. This can be clearly seen in the similarity

Scheme 1

experimental spectra with reference to those predicted using reliable quantum chemical methods. The popularity of this approach is reflected in the determination of ACs of several chiral natural products by means of chiroptical spectroscopy methods over the past several years.17−24 In this article the first analysis of the stereostructure of centratherin by means of chiroptical spectroscopy is discussed. The experimental ECD, ORD, and VCD spectra for solution samples of centratherin are analyzed in terms of the corresponding quantum chemical predictions to determine the AC.

■

RESULTS AND DISCUSSION The isolated compound was determined as centratherin based on, among other facts, 1D and 2D NMR studies and by comparison of its 1H and 13C NMR spectra with published data.25,26 As incorrect assignments for the structure of centratherin have been published,6,7 we also made comparisons of the calculated 13C NMR chemical shifts for centratherin with those obtained experimentally for the isolated compound (Table S1, Supporting Information). The 13C NMR shielding constants (σ) were calculated by a gauge including atomic orbitals-hybrid density functional theory (GIAO-HDFT) calculation procedure27 at the GIAO-mPW1PW91/6-31G(d)//mPW1PW91/6-31G(d) level of theory in the gas phase, using the Gaussian 09 software package. 13C NMR chemical shifts were calculated as δcalc = σref − σ, where σref is the shielding constant of TMS predicted at the same level of theory. The resulting 13C NMR chemical shifts were then scaled by the expression δscal = 1.05δcalc − 1.22, obtained from linear correlation between calculated and experimental chemical shifts of a pool of adequately chosen compounds.28 The calculated 13C NMR chemical shifts show good agreement with experimental data. The smaller magnitudes of the mean absolute (MAD) and root-mean-square (RMSD) deviations permitted confirmation of the isolated compound as centratherin. B



Article

the assignment of opposite AC to (−)-centratherin cannot be ruled out. Therefore, even though the visual comparison of experimental and calculated ECD spectra may suggest a good agreement, the similarity overlap plot shown for ECD (SimECD in Figure 2) suggests that ECD spectroscopic analysis alone can be ambiguous. To overcome such ambiguities arising from individually comparing the experimental ECD and EA spectra with the corresponding predicted spectra, it was suggested36 that the comparison of the electronic dissymmetry factor (EDF) (i.e., ratio of CD to absorption) spectra provides a robust approach for determining the AC. The experimental EDF spectrum is compared with the corresponding predicted spectrum for the (6R,7R,8S,10R,2′Z) isomer in Figure 1C. Even though there are six well-defined CEs, there is only one with a large dissymmetry factor at ∼330 nm. This large EDF band originates from small absorption, yet with significant CD associated with it. This positive CE with large EDF is satisfactorily reproduced in the predicted spectrum. The similarity overlap plot (Figure 2) for EDF spectra eliminates the ambiguity mentioned earlier, because there is only one positive maximum of 0.87 in the SimEDF plot (at a scale factor of 1.09). Thus, the analysis of EDF clearly suggests favoring the assignment of the (6R,7R,8S,10R,2′Z) configuration to (−)-centratherin. Thus, the comparison of ECD spectra can now be reconciled by considering the scale factor of 1.03 and SimECD value of 0.78 to be appropriate. ORD Analysis. The comparison of the experimental ORD curve with that calculated for the (6R,7R,8S,10R,2′Z) isomer is shown in Figure 3. The numerical specific rotation (SR) values

Figure 1. Comparison of experimental EA (A, bottom panel) and ECD (B, middle panel) and EDF (C, top panel) spectra of (−)-centratherin with those predicted for the (6R,7R,8S,10R,2′Z) structure. The calculated spectra are presented without x-axis scaling.

plot shown in Figure 2. A SimECD value of 0.78 is obtained at a wavelength scale factor of 1.03, which indicates the feasibility of assignment of the (6R,7R,8S,10R,2′Z) configuration to (−)-centratherin. The presence of a significant negative SimECD value (−0.64) at a scale factor of 0.93 suggests that

Figure 3. Comparison of experimental ORD of (−)-centratherin (black) with that predicted for the (6R,7R,8S,10R,2′Z) diastereomer (red).

are tabulated in the Supporting Information. The experimental ORD curve of (−)-centratherin shows negative values at longer wavelengths and change of sign at 546 nm and increasing positive values at shorter wavelengths. The calculated ORD curve showed the same pattern as that for the experimental ORD curve. However, as the predicted ECD and EDF spectra needed to be red-shifted by approximately 40−50 nm in comparison with the corresponding experimental spectra, this shift is also expected to be applicable for the calculated ORD curve. For this reason, the larger magnitude for SR at 365 nm in the experimental data can be reproduced only when the calculations are performed at less than 365 nm. Except for this difference, the trends seen in the experimental and predicted ORD values are considered to be in agreement, thereby inferring the (6R,7R,8S,10R,2′Z) AC assignment for (−)-centratherin.

Figure 2. Similarity overlap plot showing the similarity overlap of experimental spectra with corresponding predicted spectra for EA, ECD, and EDF as a function of wavelength scale factor for the (6R,7R,8S,10R,2′Z) structure. The predicted ECD spectrum with constituent rotational strengths of individual transitions represented as vertical bars is presented in the Supporting Information. C



Article

VCD Analysis. The experimental vibrational absorption (VA) and VCD spectra of (−)-centratherin are compared to the corresponding predicted spectra for the (6R,7R,8S,10R,2′Z) isomer in Figure 4A and B. To codisplay the experimental and

Figure 5. Similarity overlap plot showing the similarity overlap of experimental spectra with corresponding predicted spectra for VA, VCD, and VDF as a function of wavenumber scale factor for the (6R,7R,8S,10R,2′Z) structure.

mismatch between experimental and calculated VCD intensities). Nevertheless, there are not many VCD bands with dissymmetry factors greater than 4 × 10−5 in the predicted spectrum, and most of the VDF comparison comes from a single band at ∼1320 cm−1. Bisignate VCD Couplet in the Carbonyl Stretching Region. The experimental VCD spectrum exhibits a positive bisignate couplet at ∼1700 cm−1 in the carbonyl region (see expanded VA and VCD spectra in Figure 6). Such couplets are

Figure 4. Comparison of experimental VA (A, bottom panel) and VCD (B, middle panel) and VDF (C, top panel) spectra of (−)-centratherin with those predicted for the (6R,7R,8S,10R,2′Z) structure. The calculated spectra are presented without x-axis scaling.

calculated VCD spectra on the same y-axis scale, the experimental VCD intensities were scaled by 0.5. Visual comparison indicates excellent agreement between experimental and calculated spectra, for VA as well as VCD. As mentioned in the context of ECD spectra, a visual comparison may inadvertently introduce user bias, and therefore, it is necessary to evaluate the similarity overlap plots, which are shown in Figure 5. The maximum SimVA is 0.8 and that for SimVCD is 0.5. Since the SimVCD plot yields only positive values across the reasonable frequency scale factor range, the possibility for the opposite absolute configuration is eliminated. The experimental vibrational dissymmetry factor (VDF) spectrum is compared with the corresponding predicted spectrum for the (6R,7R,8S,10R,2′Z) isomer in Figure 4C. As mentioned earlier, the experimental VCD intensities had to be multiplied by 0.5 to display the experimental VCD spectrum on the same y-axis scale as the predicted VCD spectrum. For this reason, although the SimVDF has a maximum value of 0.62, SimVCD_NN has a maximum value of only 0.2 (due to a

Figure 6. Experimental VA (bottom) and VCD (top) spectra of (−)-centratherin in the 1800−1550 cm−1 region, emphasizing the VCD couplet in the carbonyl stretching region. The inset identifies the three CO groups (A, B, and C) and the neighboring coupled CC groups (A′, B′, and C′).

often attributed to the VCD exciton coupling (EC) mechanism.37 However, centratherin is one example where application of the EC model cannot be applied or understood without performing quantum chemical calculations. The complex skeleton of centratherin has three CO groups (A, B, and C in the inset of Figure 6), a doubly conjugated carbonyl, a conjugated lactone, and a conjugated angelic ester, D



Article

as solvent. The chemical shifts are expressed relative to TMS. The CP3 parameter was calculated from the equation30 CP3 = [(∑i(δiA,exp − δiB,exp)n(δiA,calc − δiB,calc)]/∑i(δiA,exp − δiB,exp)2, where n = 3 when (δiA,calc − δiB,calc)/(δiA,exp − δiB,exp) > 1 and n = 1 otherwise. The chemical shifts δA and δB represent those for the corresponding atoms in the two diastereomers and the summation index i runs over all atoms. Optical Rotation. The optical rotation data at six different wavelengths, namely, 633, 589, 546, 436 405, and 365 nm, were measured in CH3CN, at a concentration of 2.76 mg/mL, using an Autopol IV polarimeter (Supporting Information). Electronic Circular Dichroism. ECD spectra in the ∼187−400 nm region were measured in CH3CN, at a concentration of 2.32 mg/ mL, using a Jasco J-720 spectrometer using 1 and 0.1 mm quartz cells. The reported Δε values are expressed in units of L mol−1 cm−1. Vibrational Circular Dichroism. VCD spectra in the 1800−1100 cm−1 region were measured in CH3CN, at concentration of 23 mg/mL (and 9 mg/mL for the carbonyl region), using a ChiralIR spectrometer (BioTools Inc.) and a 200 μm path length SL3 cell equipped with BaF2 windows. The reported Δε values are expressed in units of L mol−1 cm−1. Computational Details. The structure of centratherin, with the proposed relative configuration,8,9 (6R,7R,8S,10R,2′Z), was constructed using the molecular editor and visualizer program Avogadro v.1.1.0,39 and a conformational analysis was carried out with the MMFF94s force field using the Conflex software.40 These geometries, within a 25 kcal/mol energy range, were reoptimized using the semiempirical PM6 method available in GAUSSIAN09.41 The resulting geometries, within a 5 kcal/mol energy range, were reoptimized using DFT at the B3LYP/6-31G(d) level of theory. The conformers, within a 3 kcal/mol energy range, had their geometries reoptimized at the B3LYP/Aug-cc-pVDZ level of theory. These reoptimized structures identified 16 low-energy conformers (see Supporting Information for optimized Cartesian coordinates of these conformers), and all of these structures were used for ORD, ECD, and VCD calculations. The long-range corrected CAM-B3LYP level was used for ORD and ECD, while the B3LYP level was used for VCD, all of them with Dunning’s aug-cc-pVDZ correlation consistent basis set with augmented functions42 with the GAUSSIAN09 series of programs. The polarizable continuum model (PCM)43 was used to take the influence of solvent (CH3CN) into consideration when using DFT methods. The calculated spectra were obtained by weighting the individual spectra of conformers, according to Boltzmann populations determined using Gibbs energies. The relative energies and Boltzmann weighting for the 16 lowest energy conformations are presented in the Supporting Information. The structures of the four lowest energy conformers and their overlaid structures are also displayed in the Supporting Information. The main differences in these four structures are in the relative orientations of the 4-hydroxymethyl group. The vibrational frequencies, electronic transition wavelengths, rotational strengths, dipole strengths, oscillator strengths, and SRs for the four lowest energy conformers are also presented, along with their EA, ECD, VA, and VCD spectra, in the Supporting Information. Similarity Analysis. Similarity between experimental and predicted spectra can be evaluated using different measures, namely, carbo-similarity index (SI), enantiomeric similarity index (ESI),44 and Tanimoto coefficient.45 All these similarity indices, providing similarity overlap, are evaluated by our in-house-written program CDSpecTech,46 and they all lead to the same conclusions. For this reason, we will not present all similarity indices and, instead, use the index denoted as “Sim”, in analogy with the notation of Shen et al.,45 which is written for absorption and CD spectra as

with similar intergroup distances (5 Å A−B, 3.8 Å B−C, 6.3 Å A−C), but different dihedral (positive and negative) angles among them. As a result, the correct determination of which CO groups are involved in generating the observed positive bisignate couplet is not possible, unless the vibrational modes obtained from reliable quantum calculations are analyzed. The harmonic vibrational analysis at the B3LYP/aug-ccpVDZ/PCM level of theory predicts that the vibrations responsible for the positive bisignate couplet in the CO stretching region originate predominantly from groups A and B. The dihedral angle between groups A and B is positive. If it were assumed that the vibrations of groups B and C were responsible, due to their similarity, then the dihedral angle would be negative (which should give a negative couplet) and the assigned AC would be incorrect. In the presence of multiple CO groups, one would not know, a priori, which groups may participate in exciton coupling and which groups will not, and therefore the utility of the EC model is subjective. Moreover in the present case, the CO stretching motions are strongly coupled to the neighboring stretching motions of CC groups (labeled A′, B′, and C′ in the inset to Figure 6), as inferred from interaction force constants and atomic displacements in individual normal modes. Such interactions are not built into the EC model, and therefore, indiscriminate applications of the EC model can lead38 to either right conclusions for wrong reasons or incorrect conclusions altogether. In summary, starting from the reported relative configuration of (−)-centratherin, its absolute configuration is determined by the combined application of different chiroptical spectroscopic properties, namely, ECD, EDF, ORD, VCD, and VDF. The prediction of these chiroptical spectroscopic properties matched the corresponding experimental data quite well. The quantification of agreement between experimental and calculated spectra is assessed through similarity overlap analysis. All these analyses confirmed the AC of naturally occurring (−)-centratherin as (6R,7R,8S,10R,2′Z).

■

EXPERIMENTAL SECTION

Plant Material. The leaves of Eremanthus crotonoides (DC.) Sch. Bip. (Asteraceae) were collected in Restinga de Jurubatiba National Park, Rio de Janeiro, Brazil. Botanical identification was done by the botanist Dr. Marcelo Guerra Santos, Universidade do Estado do Rio de Janeiro (Brazil), and the voucher specimen (M. Guerra Santos 2150) is registered at the Herbarium of the Faculdade de Formaçaõ de Professores, Universidade Estadual do Rio de Janeiro, Brazil. Extraction and Purification. The dried and powdered leaves (1.8 kg) of E. crotonoides were extracted with 98% EtOH (5 L) at room temperature for 7 days with regular agitation. The EtOH extract was evaporated to give a residue (EC-E, 145.3 g), which was suspended in H2O and sequentially extracted with n-hexane, CH2Cl2, EtOAc, and nBuOH to give n-hexane (EC-H, 58.3 g), CH2Cl2 (EC-D, 36.5 g), EtOAc (EC-EA, 12.9 g), and n-BuOH (EC-B, 27.6 g) fractions, respectively. All solvents were spectroscopic grade and were purchased from Tedia (Fairfield, OH, USA). Fraction EC-D (6.0 g) was submitted to flash chromatography using Isolera Biotage and a cartridge with 10 g of normal-phase silica gel (Snap 10 g Biotage) at 12 mL min−1 to give 14 fractions. Centratherin was isolated from fraction 10 at 300 mL volume at 98:2 (v/v) EtOAc− n-hexane. Mobile phases used were n-hexane and EtOAc with an increasing gradient of the stronger solvent. 1H and 13C NMR analysis confirms the identity of the purified compound. NMR Analysis. The NMR spectra were recorded on a Varian VNMRSYS-500 (500 MHz) (Varian Inc.; Palo Alto, CA, USA) using 5 mm sample tubes and CDCl3 (99.9% D, CIL, Tewksbury, MA, USA)

Sim ABS =

SimCD =

I fg I ff + Igg − I fg

(1)

I fg I ff + Igg − |I fg |

(2)

where E


Journal of Natural Products I fg =

■

∫ f (x:σ) g(x) dx

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.5b00546. Energies, Boltzmann populations, and Cartesian coordinates of 16 lowest energy conformers, vibrational frequencies, electronic transition wavelengths, rotational strengths, oscillator strengths, dipole strengths, and ORD for the four lowest energy conformers, and graphical view of the lowest energy conformer (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone: (615)322-2836. Fax: (615)322-4936. E-mail: Prasad.L. [email protected]. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Picman, A. K. Biochem. Syst. Ecol. 1986, 14, 255−281. (2) Brown, D. S.; Paquette, L. A. Heterocycles 1992, 34, 807−834. (3) Giesbrecht, A. M.; Davino, S. C.; Nassis, C. Z.; Young, M. C. M.; Lopes, J. L. C.; Rodrigues, D. C.; Vichnewski, W.; Nasi, A. M. T. T.; Dias, D. A.; Lopes, J. N. C. Quim. Nova 1990, 13, 312−314. (4) Rungeler, P.; Castro, V.; Mora, G.; Goren, N.; Vichnewski, W.; Pahl, H. L.; Merfort, I.; Schmidt, T. J. Bioorg. Med. Chem. 1999, 7, 2343−2352. (5) Grael, C. F.; Vichnewski, W.; Souza, G. E.; Lopes, J. L.; Albuquerque, S.; Cunha, W. R. Phytother. Res. 2000, 14, 203−206. (6) Ohno, N.; Mccormick, S.; Mabry, T. J. Phytochemistry 1979, 18, 681−682. (7) Bohlmann, F.; Zdero, C.; Robinson, H.; King, R. M. Phytochemistry 1980, 19, 2381−2385. (8) Le Quesne, P. W.; Menachery, M. D.; Pastore, M. P.; Kelley, C. J.; Brennan, T. F.; Onan, K. D.; Raffaul, R. F. J. Org. Chem. 1982, 47, 1519−1521. (9) Herz, W.; Goedken, V. L. J. Org. Chem. 1982, 47, 2798−2800. (10) Harada, N. Chirality 2008, 20, 691−723. (11) Berova, N.; Polavarapu, P. L.; Nakanishi, K.; Woody, R. Comprehensive Chiroptical Spectroscopy; John Wiley & Sons, 2012. (12) Flack, H. D.; Bernardinelli, G. Chirality 2008, 20, 681−690. (13) Eliel, L. E.; Wilen, H. S. Stereochemistry of Organic Compounds; Wiley-Interscience: New York, 1994. (14) Polavarapu, P. L. Chirality 2008, 20, 664−672. (15) Nicu, V. P.; Mándi, A.; Kurtán, T.; Polavarapu, P. L. Chirality 2014, 26, 525−531. (16) Polavarapu, P. L. Chem. Rec. 2007, 7, 125−136. (17) Derewacz, D. K.; Mcnees, C. R.; Scalmani, G.; Covington, C. L.; Shanmugam, G.; Marnett, L. J.; Polavarapu, P. L.; Bachmann, B. O. J. Nat. Prod. 2014, 77, 1759−1763. (18) Junior, F. M. S.; Covington, C. L.; De Amorim, M. B.; Velozo, L. S. M.; Kaplan, M. A. C.; Polavarapu, P. L. J. Nat. Prod. 2014, 77, 1881−1886. (19) Li, X.; Ferreira, D.; Ding, Y. Curr. Org. Chem. 2010, 14, 1678− 1697. (20) Giuseppe, M.; Santoro, E.; Andolfi, A.; Cimmino, A.; Troselj, P.; Petrovic, A.; Superchi, S.; Evidente, A.; Berova, N. J. Nat. Prod. 2013, 76, 588−599. (21) Sánchez-Castellanos, M.; Bucio, M.; Hernández-Barragán, A.; Joseph-Nathan, P.; Cuevas, G.; Quijano, L. Chirality 2015, 27, 247− 252. (22) Xie, G.; Tian, J.; Kövér, K.; Mándi, A.; Kurtán, T. Chirality 2014, 26, 574−579. (23) Shi, X.; Zhang, A.; Pescitelli, G.; Gao, J. Chirality 2012, 24, 386− 390. (24) Batista, J. M.; Batista, A. N.; Rinaldo, D.; Vilegas, W.; Ambrósio, D. L.; Cicarelli, R. M.; Bolzani, V.; Kato, M.; Nafie, L. A.; López, S. N.; Furlan, M. J. Nat. Prod. 2011, 74, 1154−1160. (25) Vichnewski, W.; Semir, J.; Leitão Filho, H. F.; Nakashima, C.; Lunardello, M.; Gutierrez, A.; Herz, W. Rev. Latinoam. Quim. 1990, 21, 28−29. (26) Soares, A. C. F.; Silva, A. N.; Matos, P. M.; Silva, E. H.; Heleno, V. C.; Lopez, N. P.; Lopes, J. L.; Sazz, D. C. Quim. Nova 2012, 35, 2205−2209. (27) Costa, F. L. P.; De Albuquerque, A. C. F.; Dos Santos, F. M.; De Amorim, M. B. J. Phys. Org. Chem. 2010, 23, 972−977. (28) Costa, F. L. P.; De Albuquerque, A. C. F.; Borges, R. M.; Dos Santos, F. M., Jr.; De Amorim, M. B. J. Comput. Theor. Nanosci. 2014, 11, 219−225. (29) Calsteren, M. R. V.; Jankowski, C. K.; Reyes-Chilpa, R.; Jiménez-Estrada, M.; Campos, M. G.; Zarazua-Lozada, A.; Oropeza, M.; Lesage, D. Can. J. Chem. 2008, 86, 1077−1084. (30) Smith, S. G.; Goodman, J. M. J. Org. Chem. 2009, 74, 4597− 4607. (31) Bagno, A.; Saielli, G. WIREs Comput. Mol. Sci. 2015, 5, 228− 240.

and f(x:σ) represents the calculated spectrum as a parametric function of the scale factor σ, and g(x) is the experimental spectrum. In the case of electronic and vibrational absorption spectra, SimABS in eq 1 will be relabeled as SimEA and SimVA, respectively. In the case of electronic and vibrational CD spectra, SimCD in eq 2 will be relabeled as SimECD and SimVCD, respectively. In the case of dissymmetry factor spectra (ratio of CD to absorption), SimCD will be relabeled as SimEDF and SimVDF, respectively, for electronic and vibrational spectra. The maximum overlap for absorption spectra is +1; that for CD spectra is ±1. SimCD = +1 indicates perfect agreement between experimental and predicted spectra for the AC used in the calculations. SimCD = −1 indicates perfect agreement between experimental and predicted spectra for the AC opposite to that used in the calculations. The quantification of the agreement between experimental and simulated spectra is obtained through the aforementioned similarity analysis by numerically evaluating the overlap among experimental and predicted spectra (see eqs 1 and 2) for EA, VA, ECD, and VCD individually and also for the EDF and VDF spectra. For evaluating the similarity, normalized spectra were used for comparisons of the VA, VCD, EA, and ECD spectra. The DF spectra were compared in both normalized and non-normalized form, with the non-normalized ratings given the suffix “_NN”. As described in detail previously,47 the experimental DF should be devoid of errors in concentration and path length, and therefore the differences between normalized and nonnormalized DF spectra reflect the level of mismatch between experimental and predicted spectral intensities. VDF spectra were generated with a baseline tolerance of 40 L mol−1 cm−1 and robustness criterion of 40 ppm. EDF spectra were generated with a baseline tolerance of 10 L mol−1 cm−1and without imposing the robustness criterion. The similarity overlap plots display the similarity index as a function of scale factor for electronic transition wavelengths in EA, ECD, and EDF spectra and for vibrational frequencies in VA, VCD, and VDF spectra. These plots were generated for VA, VCD, VDF, EA, ECD, and EDF using the program CDSpecTech.46

■

Article

ACKNOWLEDGMENTS

We wish to thank CNPq, CAPES, and NSF (CHE-0804301) for financial support. This work was conducted in part using the resources of the Advanced Computing Center for Research and Education (ACCRE) at Vanderbilt University, Nashville, TN, USA. F



Article

(32) Waddell, T. G.; Stocklin, W.; Geissman, T. A. Tetrahedron Lett. 1969, 10, 1313−1316. (33) Stocklin, W.; Waddell, T. G.; Geissman, T. A. Tetrahedron 1970, 26, 2397−2409. (34) Ren, Y.; et al. Tetrahedron 2012, 68, 2671−2678. (35) Youssef, D.; Frahm, A. W. Phytochemistry 1996, 41, 1107−1111. (36) Covington, C. L.; Polavarapu, P. L. J. Phys. Chem. A 2013, 117, 3377−3386. (37) Taniguchi, T.; Monde, K. J. Am. Chem. Soc. 2012, 134, 3695− 3698. (38) Covington, C. L.; Nicu, V. P.; Polavarapu, P. L. J. Phys. Chem. A 2015, 119, 10589. (39) Hanwell, M. D.; Curtis, D. E.; Lonie, D. C.; Vandermeersch, T.; Zurek, E.; Hutchison, G. R. J. Cheminf. 2012, 4, 17. (40) Conflex Program; CONFLEX Corporation: Tokyo, Japan, www. conflex.us. (41) 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, N. J.; 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, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.01; Gaussian, Inc.: Wallingford, CT, USA, 2009. (42) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007−1023. (43) Mennucci, B.; Cappelli, C.; Cammi, R.; Tomasi, J. Chirality 2011, 23, 717−729. (44) Debie, E.; De Gussem, E.; Dukor, R. K.; Herrebout, W.; Nafie, L. A.; Bultinck, P. ChemPhysChem 2011, 12, 1542−1549. (45) Shen, J.; Zhu, C.; Reiling, S.; Vaz, R. Spectrochim. Acta, Part A 2010, 76, 418−422. (46) Covington, C. L.; Polavarapu, P. L. CDSpecTech: Computer programs for calculating similarity measures of comparison between experimental and calculated dissymmetry factors and circular intensity differentials, https://sites.google.com/site/cdspectech1/, 2014. (47) Polavarapu, P. L.; Covington, C. L. Chirality 2014, 26, 539−552.

G


Absolute Configuration of (â)-Centratherin, a Sesquiterpenoid Lactone

Recommend Documents

Absolute Configuration of (â)-Centratherin, a Sesquiterpenoid Lactone