Absolute configuration of tropane alkaloids from Schizanthus species by vibrational circular dichroism Matías Reina a , Eleuterio Burgueño-Tapia b , María A. Bucio c , Pedro Joseph-Nathan c, * a Instituto de Productos Naturales y Agrobiología (IPNA), CSIC, La Laguna, Tenerife, Spain b Departamento de Química Orgánica, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, Prolongación de Carpio y Plan de Ayala, Col. Santo Tomás, México, D.F., 11340, Mexico c Departamento de Química, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, Apartado 14-740, México, D.F., 07000, Mexico article info Article history: Received 8 December 2009 Received in revised form 1 February 2010 Available online 26 February 2010 Keywords: Schizanthus grahamii S. pinnatus Solanaceae Absolute configuration Vibrational circular dichroism 3a,6b-Tropanediol alkaloids Schizanthus species abstract The absolute configuration (AC) of 6b-hydroxy-3a-senecioyloxytropane (1), 3a-hydroxy-6b-tigloyloxy- tropane (2), 3a-hydroxy-6b-senecioyloxytropane (3), and 3a-hydroxy-6b-angeloyloxytropane (4) was assigned as (1R,3R,5S,6R) using density functional theory (DFT) calculations at the B3LYP/DGDZVP level of theory in combination with experimental vibrational circular dichroism (VCD) measurements and comparison with the spectra of similar tropanes. The AC of 1 followed from a sample isolated from Schi- zanthus grahamii, while those of the mixture of 2 and 3, isolated from the same source, were determined by comparing the VCD measurement to a weighted calculation of the individual VCD spectra according to a 69:31 ratio of 2:3 determined by 1 H NMR signal integration. In turn, Schizanthus pinnatus provided a 7:3 mixture of 1:4 whose AC was determined using the experimental VCD absorptions in the 1150–950 cm 1 spectral region which were compared with those observed for 1–3 and with those described for other 3a,6b-tropanediol derivatives. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Tropane alkaloids, a class of compounds possessing the 8-aza- bicyclo[3.2.1]octane skeleton, mainly occur in the Solanaceae, Ery- throxylaceae, and Convolvulaceae plant families (Griffin and Lin, 2000). The plants containing these alkaloids have been used in tra- ditional medicine for centuries and some of these alkaloids have important pharmacological activities as anesthetics, against car- diovascular disorders, renal failure and antidote to snakebites (Lounasmaa and Tamminen, 1993; Fodor and Dharanipragada, 1994; Gringauz, 1997; Silva et al., 2001; Muñoz et al., 2006 and therein). Although Index Kewensis lists 72 samples for the genus Schizanthus (Solanaceae), after evaluating synonyms it seems to have 31 species indigenous to the southern part of the Andian chain of mountains, mainly in Chile (Griffin and Lin, 2000) with one representative, Schizanthus grahamii Gill., whose dispersal area reaches Argentina (Coccuci, 1989). The chemical study of these plants has attracted much attention because they contain a wide range of tropane-derived alkaloids, such as hydroxytropanes esterified with different isomeric residues like angelic, tiglic, or senecioic acids, or with cinnamic, mesaconic, and itaconic acids (San Martín et al., 1980, 1987; Gambaro et al., 1983; De la Fuente et al., 1988; Muñoz et al., 1991) found in alkaloids with a single tropane skeleton, while belladonnine and truxilline (Moore et al., 1996; Dembitsky, 2008) contain two tropane moieties due to their dimeric nature. One alkaloid, grahamine, possesses a complex, trimeric structure containing three 3,7-dihydroxytropane ring systems (Hartmann et al., 1990). Since tropane alkaloids appear in their host plants as complex mixtures, considerable efforts have been devoted recently using different analytical techniques for the separation and identification of these molecules. These include capillary electrophoresis with UV detection and coupling to MS with electrospray ionization (ESI) (Humam et al., 2005), very fast GC and capillary LC–MS (Bieri et al., 2006a,b) and LC–UV detection–multiple mass spectrometry and LC–NMR (Zanolari et al., 2003). Most tropane alkaloids in the Schizanthus genus are derived from 3a,6b-tropanediol (5) with var- ious acyl residues, a situation that generates difficulty to ascertain the correct enantiomer (1R,3R,5S,6R or 1S,3S,5R,6S) since only a few molecules have known absolute configuration (AC) (Muñoz et al., 2010). Recent determinations include the AC assignment of both diasteroisomers of 6b-hydroxyhyoscyamine (Muñoz et al., 2006), and of semisynthetic ()-(1S,3S,5R,6S)-3a,6b-diacetyloxy- tropane (Muñoz et al., 2010) using vibrational circular dichroism (VCD) in combination with density functional theory (DFT) calcula- tions, as well as the study of (1R,3R,5S,6R)-trans-3-hydrox- ysenecioyloxy-6-senecioyloxytropane which was ascertained 0031-9422/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.phytochem.2010.02.004 * Corresponding author. Tel.: +52 55 5747 7112; fax: +52 55 5747 7137. E-mail address: [email protected](P. Joseph-Nathan). Phytochemistry 71 (2010) 810–815 Contents lists available at ScienceDirect Phytochemistry journal homepage: www.elsevier.com/locate/phytochem
Absolute configuration of tropane alkaloids from Schizanthus speciesby vibrational circular dichroism
Matías Reina a, Eleuterio Burgueño-Tapia b, María A. Bucio c, Pedro Joseph-Nathan c,*
a Instituto de Productos Naturales y Agrobiología (IPNA), CSIC, La Laguna, Tenerife, Spainb Departamento de Química Orgánica, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, Prolongación de Carpio y Plan de Ayala,Col. Santo Tomás, México, D.F., 11340, Mexicoc Departamento de Química, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, Apartado 14-740, México, D.F., 07000, Mexico
a r t i c l e i n f o a b s t r a c t
Article history:Received 8 December 2009Received in revised form 1 February 2010Available online 26 February 2010
Keywords:Schizanthus grahamiiS. pinnatusSolanaceaeAbsolute configurationVibrational circular dichroism3a,6b-Tropanediol alkaloidsSchizanthus species
0031-9422/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.phytochem.2010.02.004
The absolute configuration (AC) of 6b-hydroxy-3a-senecioyloxytropane (1), 3a-hydroxy-6b-tigloyloxy-tropane (2), 3a-hydroxy-6b-senecioyloxytropane (3), and 3a-hydroxy-6b-angeloyloxytropane (4) wasassigned as (1R,3R,5S,6R) using density functional theory (DFT) calculations at the B3LYP/DGDZVP levelof theory in combination with experimental vibrational circular dichroism (VCD) measurements andcomparison with the spectra of similar tropanes. The AC of 1 followed from a sample isolated from Schi-zanthus grahamii, while those of the mixture of 2 and 3, isolated from the same source, were determinedby comparing the VCD measurement to a weighted calculation of the individual VCD spectra according toa 69:31 ratio of 2:3 determined by 1H NMR signal integration. In turn, Schizanthus pinnatus provided a 7:3mixture of 1:4 whose AC was determined using the experimental VCD absorptions in the 1150–950 cm�1
spectral region which were compared with those observed for 1–3 and with those described for other3a,6b-tropanediol derivatives.
� 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Tropane alkaloids, a class of compounds possessing the 8-aza-bicyclo[3.2.1]octane skeleton, mainly occur in the Solanaceae, Ery-throxylaceae, and Convolvulaceae plant families (Griffin and Lin,2000). The plants containing these alkaloids have been used in tra-ditional medicine for centuries and some of these alkaloids haveimportant pharmacological activities as anesthetics, against car-diovascular disorders, renal failure and antidote to snakebites(Lounasmaa and Tamminen, 1993; Fodor and Dharanipragada,1994; Gringauz, 1997; Silva et al., 2001; Muñoz et al., 2006 andtherein). Although Index Kewensis lists 72 samples for the genusSchizanthus (Solanaceae), after evaluating synonyms it seems tohave 31 species indigenous to the southern part of the Andianchain of mountains, mainly in Chile (Griffin and Lin, 2000) withone representative, Schizanthus grahamii Gill., whose dispersal areareaches Argentina (Coccuci, 1989). The chemical study of theseplants has attracted much attention because they contain a widerange of tropane-derived alkaloids, such as hydroxytropanesesterified with different isomeric residues like angelic, tiglic, orsenecioic acids, or with cinnamic, mesaconic, and itaconic acids(San Martín et al., 1980, 1987; Gambaro et al., 1983; De la Fuente
ll rights reserved.
x: +52 55 5747 7137.Joseph-Nathan).
et al., 1988; Muñoz et al., 1991) found in alkaloids with a singletropane skeleton, while belladonnine and truxilline (Moore et al.,1996; Dembitsky, 2008) contain two tropane moieties due to theirdimeric nature. One alkaloid, grahamine, possesses a complex,trimeric structure containing three 3,7-dihydroxytropane ringsystems (Hartmann et al., 1990).
Since tropane alkaloids appear in their host plants as complexmixtures, considerable efforts have been devoted recently usingdifferent analytical techniques for the separation and identificationof these molecules. These include capillary electrophoresis with UVdetection and coupling to MS with electrospray ionization (ESI)(Humam et al., 2005), very fast GC and capillary LC–MS (Bieriet al., 2006a,b) and LC–UV detection–multiple mass spectrometryand LC–NMR (Zanolari et al., 2003). Most tropane alkaloids in theSchizanthus genus are derived from 3a,6b-tropanediol (5) with var-ious acyl residues, a situation that generates difficulty to ascertainthe correct enantiomer (1R,3R,5S,6R or 1S,3S,5R,6S) since only afew molecules have known absolute configuration (AC) (Muñozet al., 2010). Recent determinations include the AC assignment ofboth diasteroisomers of 6b-hydroxyhyoscyamine (Muñoz et al.,2006), and of semisynthetic (�)-(1S,3S,5R,6S)-3a,6b-diacetyloxy-tropane (Muñoz et al., 2010) using vibrational circular dichroism(VCD) in combination with density functional theory (DFT) calcula-tions, as well as the study of (1R,3R,5S,6R)-trans-3-hydrox-ysenecioyloxy-6-senecioyloxytropane which was ascertained
a Conformers 1a and 1b have axial N-Me while 1c–1j have equatorial N-Me
M. Reina et al. / Phytochemistry 71 (2010) 810–815 811
(Humam et al., 2008) using electronic circular dichroism, also incombination with DFT calculations.
In this paper, the AC of 3a,6b-tropanediol monoesters 1–4(Fig. 1) determined by applying the VCD methodology are reported.The AC determination of 1 follows from a pure sample isolatedfrom S. grahamii, while in the case of 2 and 3 it was made fromVCD spectral measurements of a 2:3 mixture isolated from thesame species, and the calculation of the individual VCD spectraof both molecules which were weighted according to the 69:31(2:3) ratio determined from 1H NMR signal integration. The ACassignment of 4 was made after comparison of absorptions in the1150–950 cm�1 VCD spectral region of a 7:3 mixture of 1:4 iso-lated from Schizanthus pinnatus Ruiz & Pav. with those observedfor 1–3 and from the VCD literature data (Muñoz et al., 2010).
orientation.b Relative to 1a with E6-31G(d) = �494752.48 kcal/mol.C From Spartan’04 energies according to DG = �RT ln K.d Relative to 1a with EDGDZVP = �494813.22 kcal/mol.E From Gaussian 03W energies according to DG = �RT ln K.
2. Results and discussion
Conformational analysis of (1R,3R,5S,6R)-6b-hydroxy-3a-sene-cioyloxytropane (1) was initiated following the Monte Carlo proto-col (Chang et al., 1989) at the MMFF94 level of theory (Halgren,1996a,b,c,d; Halgren and Nachbar, 1996) as implemented in thePC Spartan’04 program, considering axial or equatorial orientationsfor the N-Me group which gave 16 and 22 minimum energy con-formers, respectively. These structures were optimized using den-sity functional theory (DFT) (Perdew, 1986) at the B3LYP/6-31G(d)level. The two different sets of conformers were then mixed, pro-viding 10 conformers, when using a 2.5 kcal/mol cutoff, fromwhich only two have the axial N-Me group, while eight have theequatorial N-Me group, accounting for 72.8% and 27.2% of the totalconformational population, respectively (Table 1). These conform-ers were re-optimized, and the IR and VCD frequencies calculatedusing the B3LYP functional and the DGDZVP basis set (Burgueño-Tapia and Joseph-Nathan, 2008). The use of this B3LYP/DGDZVPcombination of functional and basis set requires less computingtime than the 6-31G(d) basis set, while producing very similar re-sults, as is evident in published data (Cerda-García-Rojas et al.,2007, 2008). This situation seems to be associated with the factthat DGauss basis sets, such as DGDZVP, are optimized for DFTmethods (Godbout et al., 1992), and are demonstrated as an accu-rate and efficient approach for the study of molecular geometries,vibrational properties, and reaction energies (Andzelm and Wim-mer, 1992). Computer time could become crucial when studyinglarger molecules, as was done for stypotriol triacetate (Muñoz
N
OR
R1O
11234
Sen HHHH
TglSenAng
O O
O
Sen = Tgl =
Ang =
5
R R
H H
Fig. 1. Molecular structure of 3a,6b-tropanediols monoesteres 1–4 and diol 5.
et al., 2009), a C33H46O7 compound with 300 electrons which re-quired almost 1100 h of computer time when using the B3LYP/DGDZVP level of theory.
After B3LYP/DGDZVP optimization of 1, eight relevant conform-ers were observed accounting for 98.5% of the conformational map(Fig. 2). Two of these conformers, with an axial N-Me orientation,show hydrogen bonding between the C6 hydroxyl group and thenitrogen atom and contribute to 72.0% of the conformational pop-ulation. The carbonyl and double bond atoms of the senecioyl esterresidue in both conformers are coplanar, differing only in the C2–C3–O–C10 torsion angle values, which are of �156.6 and �82.5 de-grees for 1a and 1b (Fig. 2), respectively. The calculated VCD spec-tra of these eight conformers of 1 were combined in a singleweighted plot (Fig. 3) according to the Boltzmann conformationalpopulation derived from the relative free energy values of theiroptimized structures and compared with the experimental VCDspectrum showing a good concordance as can be seen in Fig. 3.
The AC assignment of alkaloids 2 and 3 follows from a 69:31mixture of 2:3. For this purpose, the conformational analysis ofeach alkaloid, using the same methodology as for 1, was under-taken. After B3LYP/DGDZVP reoptimization, 12 conformers wereselected for (1R,3R,5S,6R)-2 in a 2.5 kcal/mol cutoff contributingwith 96.4% of the total conformational population (Table 2), fromwhich eight conformers correspond to the equatorial (82.4%) N-Me group orientation and four to the axial (14.0%) N-Me group ori-entation. The VCD curve (trace c, Fig. 4) was generated according tothe Boltzmann conformational population derived from theDGDZVP free energy values. By analogy, conformational analysisand VCD calculations were made for (1R,3R,5S,6R)-3 yielding ninelow energy conformers corresponding to 99.4% of the total confor-mational population (Table 3) from which five with an equatorialN-Me group account for 85.8% and four with the axial N-Me groupaccount for 13.6%. The corresponding VCD spectrum of(1R,3R,5S,6R)-3 (trace d, Fig. 4), shows similarity to the spectrumof (1R,3R,5S,6R)-2 (trace c, Fig. 4), since the ester groups are achiraland have no relevant contributions to the VCD spectrum. The finalcalculated VCD curve for the 2:3 mixture (trace b, Fig. 4) was ob-tained by a combination of the individual VCD spectra in a singleweighted plot according to the 69:31 ratio of 2:3 obtained fromthe 1H NMR signal integration analysis and is compared (trace a,Fig. 4) to the experimental VCD spectrum, thereby establishingthe absolute configuration of both molecules.
In a previous paper (Muñoz et al., 2010), two specific distinctiveregions in the VCD spectra of 3a,6b-tropanediol derivatives werehighlighted, one between 1300 and 1200 cm�1 that drasticallychanges with the tropane moiety conformation, and the otherone between 1150 and 950 cm�1 showing little variation with dif-
Fig. 2. DFT B3LYP/DGDZVP minimum energy conformers of (1R,3R,5S,6R)-1.
b
a
-30
-20
-10
0
10
20
30
950115013501550
-30
-20
-10
0
10
20
30
Δε x
10
3
Wavenumbers(cm-1)
Δε x
10
3
Fig. 3. (a) Experimental and (b) DFT B3LYP/DGDZVP Bolztmann weighted VCDspectra of (1R,3R,5S,6R)-1.
Table 2B3LYP calculated relative energies (kcal/mol) and conformational population (%) forthe most stable conformers of (1R,3R,5S,6R)-2.
a Conformers 2e, 2h, 2k and 2l have axial N-Me while 2a–d, 2f, 2g, 2i and 2j haveequatorial N-Me orientation.
b Relative to 2a with E6-31G(d) = �494751.34 kcal/mol.C From Spartan’04 energies according to DG = �RT ln K.d Relative to 2a with EDGDZVP = �494811.87 kcal/mol.E From Gaussian 03W energies according to DG = �RT ln K.
812 M. Reina et al. / Phytochemistry 71 (2010) 810–815
fering conformations, but being very sensitive to the absolute con-figuration. In the cases of (1S,3S,5R,6S)-3a,6b-diacetyloxytropane(Muñoz et al., 2010) and (1S,3S,5R,6S)-6b-hydroxyhyoscyamine(Muñoz et al., 2006) this last region shows, on going from higherto lower wavenumbers, two absorptions with negative intensityand two with positive intensity collapsed into a broad band, whilethe alkaloid (1R,3R,5S,6R)-6b-hydroxyhyoscyamine (Muñoz et al.,2006) has similar absorption bands, but with opposite sign.
The weighted VCD spectra (Fig. 4), calculated for the most sta-ble conformations of 2 and 3 show absorptions at 1065, 1038,1025, and 1001 cm�1 with De +8.8, +10.1, +17.7, and�10.0, respec-tively. These absorptions were observed in the experimental VCDspectrum at 1080, 1053, 1039, and 1012 cm�1 with De +9.4,+13.9, +28.5, and �8.4, respectively. In addition to direct VCD com-parison, it is clear that in 3,6-tropanediol derivatives, observation
ca
db
-20
-10
0
10
20
30
-20
-10
0
10
20
30
-20
-10
0
10
20
30
40
950115013501550-20
-10
0
10
20
30
950115013501550
Wavenumbers (cm-1) Wavenumbers (cm-1)
Δε x
103
Δε x
103
Fig. 4. (a) Experimental VCD spectrum of a 69:31 mixture of 2:3, (b) DFT//B3LYP/DGDZVP spectrum of a Boltzmann weighted 69:31 ratio of (1R,3R,5S,6R)-2 and (1R,3R,5S,6R)-3, (c) DFT//B3LYP/DGDZVP spectra of (1R,3R,5S,6R)-2 and of (d) (1R,3R,5S,6R)-3.
Table 3B3LYP calculated relative energies (kcal/mol) and conformational population (%) forthe most stable conformers of (1R,3R,5S,6R)-3.
a Conformers 3e–h have axial N-Me while 3a–d and 3i have equatorial N-Meorientation.
b Relative to 3a with E6-31G(d) = �494752.46 kcal/mol.C From Spartan’04 energies according to DG = �RT ln K.d Relative to 3a with EDGDZVP = �494813.23 kcal/mol.E From Gaussian 03W energies according to DG = �RT ln K.
M. Reina et al. / Phytochemistry 71 (2010) 810–815 813
of the 1150–950 cm�1 VCD spectral region is critical to assign theAC.
Taking advantage of this observation, the AC of 6b-angeloyloxy-3a-hydroxytropane (4) contained in a mixture with 6b-hydroxy-3a-senecioyloxytropane (1) was determined as (1R,3R,5S,6R). Theexperimental VCD spectrum obtained from the 1:4 (7:3) mixture(Fig. 5), shows absorptions at 1060, 1047, 1038, and 1022 cm�1
with absorptivities of +26.0, +21.4, +31.8, and �35.5, respectively,in excellent agreement with those observed in the experimentalVCD spectra of 1 1061 (+24.9), 1049 (+19.5), 1037 (+28.3), and1020 (�20.5) cm�1, and of the 2:3 mixture (Table 4).
3. Concluding remarks
In summary, tropane alkaloids 1–4 isolated from two Schizan-thus species possess the (1R,3R,5S,6R) absolute configuration asevidenced by VCD. The most stable conformations for(1R,3R,5S,6R)-1 show the N-Me group in an axial orientation withrespect to the six-membered ring due to hydrogen bonding ofthe C6 hydroxyl group and the nitrogen atom. In contrast, the moststable conformations for 2 and 3, in which the C6 substituent is anester, correspond to those with the N-Me group in an equatorial ori-entation. Observed and calculated absorptions between 1150 and950 cm�1 in the VCD spectra of 1–3 were in concordance withthose described for 3a,6b-tropanediol derivatives having the(1R,3R,5S,6R) AC, confirming that these absorptions and theirintensities are fundamental for determining the AC of 3a,6b-tro-panediol derivatives. Thus, in the case of 4, the AC determinationfollows from comparison of specific VCD bands thereby avoidingthe need of costly DFT calculations. This seems to be the first casein which spectral analogy permits the AC determination of a natu-ral product in a similar fashion as has been done for decades usingoptical rotatory dispersion and electronic circular dichroism mea-surements performed in the ultraviolet and visible regions.
4. Experimental
4.1. General
The 1H NMR spectra were recorded from 99.8% atom-D CDCl3
solutions containing TMS as the internal standard at 300 MHz on
-40
-30
-20
-10
0
10
20
30
40
-30
-20
-10
0
10
20
30
-20
-10
0
10
20
30
950115013501550
Wavenumbers (cm-1)
Δε x
103
Δε x
103
Δε x
103
a
b
c
Fig. 5. Experimental VCD spectral comparison of (a) (1R,3R,5S,6R)-1 and(1R,3R,5S,6R)-4 (7:3) mixture, (b) (1R,3R,5S,6R)-1 and (c) (1R,3R,5S,6R)-2 and(1R,3R,5S,6R)-3 (69:31) mixture.
814 M. Reina et al. / Phytochemistry 71 (2010) 810–815
a Varian Mercury spectrometer. Alumina (Merck, Art. 1077) wasused for column chromatography (CC) purifications.
Table 4Wavenumbers (m in cm�1) and intensities (De � 103) in the 1150–950 cm�1 region of the
a Measured from a 69:31 mixture of 2:3.b Measured from a 7:3 mixture of 1:4.
4.2. IR and VCD measurements
These measurements were performed using a BioTools-BOMEMChiralIR FT–VCD spectrophotometer equipped with dual photo-elastic modulation. All samples were dissolved in 100% atom-DCDCl3 (150 lL) and placed in a BaF2 cell with a pathlength of100 lm. Data were acquired at a resolution of 4 cm�1 using 1(7.8 mg), the mixture of 2 and 3 (7.1 mg), and the mixture of 1and 4 (7.7 mg). Four data blocks, 1 h each, were acquired andadded. The identity, stability, and composition of each tested sam-ple were verified by 1H NMR measurement immediately prior andafter VCD measurements.
4.3. Tropane alkaloids
Extracts derived from air-dried and finely powdered leaves of S.grahamii and S. pinnatus, as well as chromatographic fractionsthereof, were available in our laboratory from previous studies(San Martín et al., 1987; De la Fuente et al., 1988). Purification ofeach of the alkaloid samples was performed by neutral aluminaCC, eluting with EtOAc and an EtOAc:MeOH (97:3) mixture to give6b-hydroxy-3a-senecioyloxytropane (1) (8.2 mg), and a mixture of1 and 6b-angeloyloxy-3a-hydroxytropane (4) (7:3, 0.591 g) from S.grahamii, as well as a 69:31 mixture of 3a-hydroxy-6b-tigloyloxy-tropane (2) and 3a-hydroxy-6b-senecioyloxytropane (3) (32.0 mg)from S. pinnatus. The composition of the two studied mixtures oftropane alkaloids follows from careful 1H NMR signal integrationof the ester vinyl protons setting an observation pulse width of45� and a 10 s pulse delay to allow complete magnetization recov-ery. They are estimated to be accurate within 1%.
4.4. DFT calculations
Conformational searches were started using a Monte Carloguided protocol considering an initial energy cutoff of 10 kcal/mol above the global minimum value. The searches were con-ducted independently considering the N-methyl group axial orequatorial. All structures were optimized at the B3LYP/6-31G(d) le-vel of theory, and the different sets of conformers were mixed.Only the conformations found in a 2.5 kcal/mol energy range werefurther optimized using the B3LYP hybrid functional and theDGDZVP basis set, and then IR and VCD frequencies were calcu-lated. The final computed VCD spectra of the 2:3 mixture was gen-erated by weighting the individual VCD spectra according to the69:31 (2:3) ratio obtained by 1H NMR signal integration. Confor-mational searches were made using the Spartan004 software pack-age from Wave-Function (Irvine, CA, USA), while geometryreoptimizations and vibrational spectra were calculated using theGaussian 03 W software package from Gaussian, Inc. (Wallingford,CT, USA). The frequencies were scaled using an anharmonicity fac-tor of 0.97, estimated from the IR absorption bands, and plotted asLorenzian bands with half-widths of 6 cm�1. For optimizations andIR and VCD calculations, typically around 12 h of computational
VCD spectra of 1, of a 2 and 3 mixture, and of a 1 and 4 mixture.
M. Reina et al. / Phytochemistry 71 (2010) 810–815 815
time per conformer were required when using a Pentium D desk-top personal computer (PC) with 2 GB RAM operated at 3 GHz.
Acknowledgements
This work was partially supported by the Ministerio de Educa-ción y Ciencia, Spain (CTQ2006-15597-C02-01PPQ), SIP-IPN-Mex-ico, CONACYT-Mexico and CYTED.
References
Andzelm, J., Wimmer, E., 1992. Density functional Gaussian-type-orbital approachto molecular geometries, vibrations, and reaction energies. J. Chem. Phys. 96,1280–1303.
Bieri, S., Muñoz, O., Veuthey, J.L., Christen, P., 2006a. Analysis of isomeric tropanealkaloids from Schizanthus grahamii by very fast gas chromatography. J. Sep. Sci.29, 96–102.
Bieri, S., Varesio, E., Muñoz, O., Veuthey, J.-L., Christen, P., 2006b. Use of porousgraphitic carbon column for the separation of natural isomeric tropanealkaloids by capillary LC and mass spectrometry. J. Pharm. Biomed. Anal. 40,545–551.
Burgueño-Tapia, E., Joseph-Nathan, P., 2008. Absolute configuration oferemophilanoids by vibrational circular dichroism. Phytochemistry 69, 2251–2256.
Cerda-García-Rojas, C.M., Catalán, C.A.N., Muro, A.C., Joseph-Nathan, P., 2008.Vibrational circular dichroism of africanane and lippifoliane sesquiterpenesfrom Lippia integrifolia. J. Nat. Prod. 71, 967–971.
Chang, G., Guida, W.C., Still, W.C., 1989. An internal-coordinate Monte Carlo methodfor searching conformational space. J. Am. Chem. Soc. 111, 4379–4386.
Coccuci, A.A., 1989. El mecanismo floral de Schizanthus. Kurtziana 20, 113–132.De la Fuente, G., Reina, M., Muñoz, O., San Martín, A., Girault, P.J., 1988. Tropane
alkaloids from Schizanthus pinnatus. Heterocycles 27, 1887–1897.Dembitsky, V.M., 2008. Bioactive cyclobutane-containing alkaloids. J. Nat. Med. 62,
1–33.Fodor, G., Dharanipragada, R., 1994. Tropane alkaloids. Nat. Prod. Rep. 12, 443–450.Gambaro, V., Labbé, C., Castillo, M., 1983. Angeloil, tigloyl and senecioyloxytropane
alkaloids from Schizanthus hookerii. Phytochemistry 22, 1838–1839.Godbout, N., Salahub, D.R., Andzelm, J., Wimmer, E., 1992. Optimization of
Gaussian-type basis set for local spin density functional calculations. Part I.Boron through neon optimization techniques and validation. Can. J. Chem. 70,560–571.
Griffin, W.J., Lin, G.D., 2000. Chemotaxonomy and geographical distribution oftropane alkaloids. Phytochemistry 53, 623–637.
Gringauz, A., 1997. Introduction to Medicinal Chemistry-How Drugs Act and Why.VCH, New York. 721p.
Halgren, T.A., 1996a. Merck molecular force field. I. Basis, form, scope,parameterization, and performance of MMFF94. J. Comp. Chem. 17, 616–641.
Halgren, T.A., 1996b. Merck molecular force field. II. MMFF94 van der Waals andelectrostatic parameters for intermolecular interactions. J. Comp. Chem. 17,587–615.
Halgren, T.A., 1996c. Merck molecular force field. III. Molecular geometries andvibrational frequencies for MMFF94. J. Comp. Chem. 17, 553–586.
Halgren, T.A., 1996d. Merck molecular force field. V. Extension of MMFF94 usingexperimental data, additional computational data, and empirical rules. J. Comp.Chem. 17, 490–519.
Halgren, T.A., Nachbar, R.B., 1996. Merck molecular force field. IV. Conformationalenergies and geometries for MMFF94. J. Comp. Chem. 17, 520–552.
Hartmann, R., San-Martin, A., Muñoz, O., Breitmaier, E., 1990. Grahamine, anunusual tropane alkaloid from Schizanthus grahamii. Angew. Chem. Int. Ed. Engl.29, 385–386.
Humam, M., Bieri, S., Geiser, L., Muñoz, O., Veuthey, J.-L., Christen, P., 2005.Separation of four isomeric tropane alkaloids from Schizanthus grahamii by non-aqueous capillary electrophoresis. Phytochem. Anal. 16, 349–356.
Humam, M., Christen, P., Muñoz, O., Hostettmann, K., Jeannerat, D., 2008. Absoluteconfiguration of tropane alkaloids bearing two a,b-unsaturated ester functionsusing electronic CD spectroscopy: application to (R,R)-trans-3-hydroxysenecioyloxy-6-senecioloxytropane. Chirality 20, 20–25.
Lounasmaa, M., Tamminen, T., 1993. The tropane alkaloids. In: Cordell, G.A. (Ed.),The Alkaloids, vol. 44. Academic Press, San Diego, pp. 1–114.
Moore, J.M., Casale, J.F., Cooper, D.A., 1996. Comparative determination of totalisomeric truxillines in illicit, refined, South American cocaine hydrochlorideusing capillary gas chromatography-electron capture detection. J. Chromatogr.A 756, 193–201.
Muñoz, O., Hartmann, R., Breitmaier, E., 1991. Schizanthine X, a new alkaloid fromSchizanthus grahamii. J. Nat. Prod. 54, 1094–1096.
Muñoz, M.A., Muñoz, O., Joseph-Nathan, P., 2006. Absolute configuration of naturaldiastereoisomers of 6b-hydroxyhyoscyamine by vibrational circular dichroism.J. Nat. Prod. 69, 1335–1340.
Muñoz, M.A., Areche, C., San-Martín, A., Robirosa, J., Joseph-Nathan, P., 2009. VCDdetermination of the absolute configuration of stypotriol. Nat. Prod. Comm. 4,1037–1040.
Muñoz, M.A., Muñoz, O., Joseph-Nathan, P., 2010. Absolute configurationdetermination and conformational analysis of (�)-(3S, 6S)-3a,6b-diacetoxytropane using vibrational circular dichroism and DFT techniques.Chirality 22, 234–241.
Perdew, J.P., 1986. Density-functional approximation for the correlation energy ofthe inhomogeneous electron gas. Phys. Rev. B 33, 8822–8824.
San Martín, A., Rovirosa, J., Gambaro, V., Castillo, M., 1980. Tropane alkaloids fromSchizanthus hookeri. Phytochemistry 19, 2007–2008.
San Martín, A., Labbé, C., Muñoz, O., Castillo, M., Reina, M., De la Fuente, G.,González, A., 1987. Tropane alkaloids from Schizanthus grahamii.Phytochemistry 26, 819–822.
Silva, G.L., Cui, B., Chávez, D., You, M., Chai, H.-B., Rasoanaivo, P., Lynn, S.M., ONeill,M.J., Lewis, J.A., Besterman, J.-M., Monks, A., Farnsworth, N.R., Cordell, G.A.,Pezzuto, J.M., Kinghorn, A.D., 2001. Modulation of the multidrug-resistancephenotype by new tropane alkaloid aromatic esters from Erytroxylum pervillei. J.Nat. Prod. 64, 1514–1520.
Zanolari, B., Wolfender, J.-L., Guilet, D., Marston, A., Queiroz, E.F., Paulo, M.Q.,Hostettmann, K., 2003. On-line identification of tropane alkaloids fromErythroxylum vacciniifolium by liquid chromatography–UV detection–multiplemass spectrometry and liquid chromatography–nuclear magnetic resonancespectrometry. J. Chromatogr. A 1020, 75–89.
