Structural determination of molecular stereochemistry using VCD spectroscopy and a conformational code: Absolute configuration and solution conformation of a chiral liquid pesticide, (R)-(+)-malathion

Authors


  • Contribution to the special thematic project “Advances in Chiroptical Methods”

Abstract

The absolute configuration and solution conformation of (R)-(+)-malathion were determined by using vibrational circular dichroism spectroscopy and a fragment-conformational search with a recently published conformational code. The determination of molecular stereochemistry was carried out without a conformational search using molecular mechanics calculations. Density functional theory calculations of the fragments of (R)-malathion, ethyl propionate, (R)-ethyl 2-(methylthio)propanoate, (R)-diethyl 2-(methylthio)succinate, and O,O,S-trimethyl phosphorodithioate were carried out, and the principal conformational features of the fragments were profiled. This fragment-conformational search reduces the time needed for the selection of the predominant conformations for (R)-malathion and significantly improves the accuracy of the determination of absolute configuration. Chirality 21:E172–E180, 2009. © 2009 Wiley-Liss, Inc.

INTRODUCTION

Malathion is one of the most widely used pesticides for suppression of harmful insects, such as mosquito, and has a flexible liquid-state conformation at room temperature.1, 2 On the other hand, pesticide residue tests for the determination of maximum residue levels (MRLs) in food have been addressed for trade and human health purposes.3, 4 Enantiomers of chiral pesticides work differently in biological systems, and the analytical method to determine the optical purity, stereoselective bioactivity, and environmental behavior of chiral pesticides has also been developed.5 Several chiral pesticides including malathion are liquid at room temperature, and therefore, the absolute configuration of these pesticides cannot be determined by the X-ray crystallography in the usual manner.

We have developed an analytical method for the molecular stereochemistry, absolute configuration, and solution conformation, of chiral compounds by using vibrational circular dichroism (VCD) spectroscopy6–8 and a recently published conformational code.9, 10 A chiral compound in a liquid or solution state at room temperature usually has a flexibilble structure and large numbers of conformational isomers. A detailed understanding of the exact conformations in the liquid state or the solution state is needed for the determination of absolute configuration.7, 10 For the conformational search of flexible compounds, molecular mechanics (MM) calculations have been usually used.11 However, especially for conformational searches of flexible compounds with many degrees of freedom, there is a possibility that several population-rich conformations might be missing.12 In such cases, it is necessary to demonstrate conclusively that the range of considered conformations is sufficiently complete for the determination of absolute configuration.

For this purpose, we have developed a conformational code for the exhaustive analysis of conformers of all kinds of molecules.10 At the time that we started our VCD conformational studies, Polavarapu and coworkers reported density functional theory (DFT) calculations of VCD for various flexible compounds such as (+)-enflurane,13 (−)-tert-butylphenylphosphine oxide,14 (−)-2-butanol,15 (+)-n-butyl tert-butyl sulfoxide,16 (+)-2,5-dimethylthiolane,17 (−)-2,3-butanediol,18tert-butylphenylphosphinothioic acid,19 3-(2-methylbutyl)thiophene,20 1,1,1,3,3-pentafluoro-2-(fluoromethoxy)-3-methoxypropane,21 bromochlorofluoromethane,22tert-butyl-1-(2-methylnaphthyl)phosphine oxide,23 and (+)-1-bromo-2-methylbutane.24 As a result, we decided to carry out VCD theoretical calculations for additional large chiral compounds such as (+)-2-hexanol, (+)-2-heptanol, (+)-2-octanol, (+)-2-nonanol, and (+)-2-decanol.7 For the conformational description of (+)-2-octanol, the conformational terms, T (trans), G+ (+gauche), G (−gauche), are inconvenient for the comparison of each conformation to sort them in the table, and instead the displacement of numbers from the terms (T = 1, G+ = 2, and G = 3) has been used for the conformational analysis of the flexible chiral compounds.7 Furthermore, the definition of angle locations and the fine classification for conformational elements from the theoretical calculations of baccatin III9 and (S)-ibuprofen10 have been used, and subsequently, we have proposed a revised conformational code for the exhaustive analysis of conformers of all kinds of chemical compounds for what we call conformome analysis (Fig. 1).10

Figure 1.

Classification of dihedral angles as conformational elements in the revised conformational code.10

On the other hand, we tried the conformational search using MM calculations for various entire molecules such as (+)-2-heptanol. However, the reliability of the relative energy for each conformer for the MM calculations was so ambiguous as to select the conformations for the prediction of VCD with additional DFT calculations. For flexible large molecules, such as natural products, the absolute configurations have been often determined by the model fragments of the entire molecules.25, 26 Therefore, we have addressed the conformational search using DFT calculations for the fragments of entire molecules in combination with the conformational code.

In this article, we describe the absolute configuration and solution conformation of (R)-malathion using VCD spectroscopy and the conformational code without a full conformational search using MM calculations. Instead, DFT calculations were carried out for the fragments of (R)-malathion, namely ethyl propionate, (R)-ethyl 2-(methylthio)propanoate, (R)-diethyl 2-(methylthio)succinate, and O,O,S-trimethyl phosphorodithioate. We also show that this fragment-conformational search improves the accuracy and efficiency of the determination absolute configuration of malathion.

MATERIALS AND METHODS

Materials

The enantiomers of malathion were separated by chiral HPLC (CHIRALCEL OD), using hexane/2-propanol as an eluent. (+)-Malathion ([α]20D +82.1°, c = 0.12, CHCl3) was obtained as the first fraction. The optical rotation of (+)-malathion corresponded closely to the reference value ([α]24D +79.7°, c = 1.25, CHCl3).1

Measurements

The optical rotation was recorded on a Horiba polarimeter SEPA-300. The infrared and VCD spectra were recorded on a commercial Fourier transform VCD spectrometer ChiralIR from BioTools. The VCD spectra of the solution state were recorded with 4–5 h data collection time at 4 cm−1 resolution. The CCl4 solutions were placed in a 72 μm path length cell with BaF2 windows. The VCD spectra of the solution state were corrected by subtraction using a solvent VCD baseline.

Calculations

All geometry optimizations, conformer searches, vibrational frequencies, infrared absorbance intensities, and VCD intensities for (R)-malathion, ethyl propionate, (R)-ethyl 2-(methylthio)propanoate, (R)-diethyl 2-(methylthio)succinate, and O,O,S-trimethyl phosphorodithioate were calculated by using the Gaussian 03 program27 on a Pentium 4 (2.8 GHz) PC. DFT with B3LYP functional and 6-31G(d) basis set was used for the calculations. The theoretical absorbance and VCD spectra were simulated with Lorentzian band shapes and 6 cm−1 full width at half-height. The ab initio frequencies were scaled by 0.97, and the thermal corrections to Gibbs free energies were scaled with 0.9989.

RESULTS AND DISCUSSION

(R)-Malathion [desc-AB-C(tmpd-DEF)-GHI] can be considered as the combination of the fragments, (R)-diethyl 2-(methylthio)succinate [desc-AB-C(MeS)-GHI] and O,O,S-trimethyl phosphorodithioate (tmpd-DEF) (Fig. 2). (R)-Diethyl 2-(methylthio)succinate [desc-AB-C(MeS)-GHI] is further considered to contain the building block, (R)-ethyl 2-(methylthio)propanoate [etpa-AB-C(MeS)], and is finally resolved into ethyl propionate (etpa-AB). The prefix letters are selected to avoid the general chemical abbreviations such as Pr (propyl) and are shown as the italic face in the compound name.

Figure 2.

Chemical structures and representations of the conformational codes for (R)-malathion [desc-AB-C(tmpd-DEF)-GHI], O,O,S-trimethyl phosphorodithioate (tmpd-DEF), (R)-diethyl 2-(methylthio)succinate [desc-AB-C(MeS)-GHI], (R)-ethyl 2-(methylthio)propanoate [etpa-AB-C(MeS)], and ethyl propionate (etpa-AB).

First, the conformational analysis of ethyl propionate (etpa-AB) was carried out using DFT calculations.27 The conformational profile of the predicted stability for etpa-AB is shown in Table 1. In Figure 3, the dihedral angles at angle locations for etpa-1β1α, etpa-3β1α, and etpa-1β2β are indicated for clarity. We have already reported that the population of one conformation is extremely large for the acetoxy group of baccatin III.9 Although the other possible conformations for etpa-AB exist, these conformations in relation to the dihedral angles at angle location of CO[BOND]O bond and with the symmetry relationship to etpa-2α2α and etpa-2α3β are neglected.

Figure 3.

Dihedral angles at angle locations of (a) etpa-1β1α, (b) etpa-3β1α, and (c) etpa-1β2β.

Table 1. Relative Gibbs free energiesa and populations for conformations of ethyl propionate (etpa-AB)
ConformationsElectronic energy (hartree)Thermal correction to Gibbs free energy (hartree)ΔG (kcal mol−1)Population
  • a

    B3LYP/6–31G*.

  • b

    The arrow indicates the conformational change during the optimization process.

etpa-1β1α−347.02213710.11345700.46
etpa-3β1α−347.02194640.1142320.610.16
etpa-1β2β−347.02069240.113931.200.06
etpa-2α2α−347.02047670.1139921.380.04
etpa-2α3β−347.02048320.1141061.440.04
etpa-2α1β−347.02194680.1141780.570.17
etpa-1α3α−347.02069290.1139311.200.06
etpa-1β4  betpa-1β2β
etpa-41α  etpa-3β1α

In this fragment-conformational search, only the stable conformational distribution patterns are selected for the second stage of the conformational search. For example, in the case of the conformational search for (R)-ethyl 2-(methylthio)propanoate [etpa-AB-C(MeS)], the conformational distribution pattern etpa-1βB was selected (Table 2). In the same way, for the conformational search of (R)-diethyl 2-(methylthio)succinate [desc-AB-C(MeS)-GHI], the conformational distribution patterns, etpa-1β6α-3β(MeS), etpa-1α5β-2α(MeS), etpa-1β3β-2α(MeS) as the angle locations A, B, and C and etpa-1β1α as the angle locations H and I were selected (Table 3). Also, several combinations of the conformational distribution patterns were added to verify and raise the accuracy of the choice of the conformational distribution patterns (Table 3).

Table 2. Relative Gibbs free energiesa and populations for conformations of (R)-ethyl 2-(methylthio)propanoate [etpa-AB-C(MeS)]
ConformationsElectronic energy (hartree)Thermal correction to Gibbs free energy (hartree)ΔG (kcal mol−1)Population
  • a

    B3LYP/6–31G*.

  • b

    The arrow indicates the conformational change during the optimization process.

etpa-1β6α-3β(MeS)−784.5191720.13890900.34
etpa-1α5β-2α(MeS)−784.5187230.1385620.060.31
etpa-1β3β-2α(MeS)−784.51825810.138370.240.23
etpa-1α2α-3β(MeS)−784.51650230.1385321.440.03
etpa-1β2α-5α(MeS)−784.51534670.1375191.530.03
etpa-1β1α-2β(MeS)−784.51598820.1382241.570.02
etpa-1α6α-1β(MeS)−784.51570230.137941.570.02
etpa-1β1α-1β(MeS)−784.51458880.1376182.070.01
etpa-1β1α-3β(MeS)betpa-1β6α-3β(MeS)
Table 3. Relative Gibbs free energiesa and populations for conformations of (R)-diethyl 2-(methylthio)succinate [desc-AB-C(MeS)-GHI]
ConformationsElectronic energy (hartree)Thermal correction to Gibbs free energy (hartree)ΔG (kcal mol−1)Population
  • a

    B3LYP/6–31G*.

  • b

    The arrow indicates the conformational change during the optimization process.

desc-13β-3β(MeS)-1β1α1β−1051.7108080.20097800.38
desc-1β3β-2α(MeS)-1β1β1β−1051.7103430.2007330.140.3
desc-1α5β-2α(MeS)-1β1β1α−1051.7098590.2007830.470.17
desc-1α6α-3β(MeS)-2α6β1α−1051.7088110.2007571.110.06
desc-1β3β-5β(MeS)-2α1α1α−1051.7089830.2010031.160.05
desc-1α5β-2α(MeS)-2α1α1α−1051.7083160.2006241.340.04
desc-1α3α-1α(MeS)-5β1α1β−1051.7054180.2004363.040
desc-1α3β-5β(MeS)-3β1α1α−1051.7057680.2013923.420
desc-1β3α-3α(MeS)-3β1α1α−1051.704190.2005113.860
desc-1α4β-1β(MeS)-3β5α1β−1051.7041840.2009914.170
desc-1β5β-2α(MeS)-3α6β1β−1051.7044990.2014084.230
desc-1β3β-1(MeS)-1β1β1βbdesc-1β3β-2α(MeS)-1β1β1β

In applying the stable conformational distribution patterns to the next stage of the conformational search, the prior bonds for the determination of dihedral angle possibly change depending on the relative positional relationship near the connection part of fragments. In that case, the transformation of the conformational elements in accord with the definition is necessary.

The conformational analysis of O,O,S-trimethyl phosphorodithioate (tmpd-DEF) was carried out in the same way (Table 4). Finally, the conformational search of entire (R)-malathion [desc-AB-C(tmpd-DEF)-GHI] was carried out with the predominant distribution patterns (Table 5). The conformational elements in (R)-malathion [desc-AB-C(tmpd-DEF)-GHI] were slightly changed in comparison with those of (R)-diethyl 2-(methylthio)succinate [desc-AB-C(MeS)-GHI] and O,O,S-trimethyl phosphorodithioate (tmpd-DEF) by the steric factor of the fragments. The order of the stability for the fragments, desc-13β-3β(MeS)-1β1α1β, desc-1β3β-2α(MeS)-1β1β1β, and desc-1α5β-2α(MeS)-1β1β1α was changed in the fragments of (R)-malathion. The order of the stability for the fragments, tmpd-3α3α3α, tmpd-2β2β2β, tmpd-1α2β3α, tmpd-1β2β2β, and tmpd-1α3α3α was also changed. However, the population-rich conformations of (R)-malathion were necessarily reflected by the stable species of the fragments. For example, desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β can be considered as the combination of fragments desc-1β3β-2α(MeS)-1β1β1β and tmpd-1β2β2β.

Table 4. Relative Gibbs free energiesa and populations for conformations of O,O,S-trimethyl phosphorodithioate (tmpd-DEF)
ConformationsElectronic energy (hartree)Thermal correction to Gibbs free energy (hartree)ΔG (kcal mol−1)Population
  • a

    B3LYP/6–31G*.

  • b

    The arrow indicates the conformational change during the optimization process.

tmpd-3α3α3α−1407.9710340.09041700.22
tmpd-2β2β2β−1407.9710340.09041700.22
tmpd-1α2β3α−1407.9703960.0899580.110.18
tmpd-1β2β2β−1407.9706480.0904110.240.14
tmpd-1α3α3α−1407.9706440.090490.290.13
tmpd-2β4β1β−1407.9678810.0890911.150.03
tmpd-3α1α4α−1407.9678810.0890921.150.03
tmpd-2β2β1β−1407.9695160.0909321.280.03
tmpd-3α1α3α−1407.9695150.0909781.30.02
tmpd-3α3α1β−1407.9664640.0900522.640
tmpd-2β1α2β−1407.9664650.090092.660
tmpd-1β32βbtmpd-1β2β2β
tmpd-22β3αtmpd-4α2β3α
tmpd-2β32βtmpd-2β2β2β
tmpd-23α3αtmpd-3α3α3α
tmpd-32β1βtmpd-2β2β1β
tmpd-32β2βtmpd-2β2β2β
Table 5. Relative Gibbs free energiesa and populations for conformations of (R)-malathion [desc-AB-C(tmpd-DEF)-GHI]
ConformationsElectronic energy (hartree)Thermal correction to Gibbs free energy (hartree)ΔG (kcal mol−1)Population
  • a

    B3LYP/6–31G*.

  • b

    The arrow indicates the conformational change during the optimization process.

desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β−1981.6663240.24308900.25
desc-1β6α-5α(tmpd-4α2β2β)-1β1α1α−1981.6642250.2419420.60.09
desc-1β6α-1β(tmpd-3α3α3α)-1β1α1β−1981.6647410.2424830.610.09
desc-1α5β-5α(tmpd-1α2β3α)-1β1α1α−1981.6648840.2428670.760.07
desc-1β3β-5α(tmpd-4α2β3α)-1β1β1β−1981.6648670.242930.810.06
desc-1α5β-5α(tmpd-1α2β2β)-1β1α1β−1981.6655750.2436510.820.06
desc-1α3β-5β(tmpd-2β2β2β)-1β1β1β−1981.6647040.242820.850.06
desc-1β3β-5α(tmpd-1α2β3α)-1β1β1α−1981.6655680.2437650.90.06
desc-1α3β-5α(tmpd-1α3α3α)-1β1β1β−1981.6640810.2424521.010.05
desc-1α5β-5β(tmpd-2β2β3α)-1β1α1α−1981.6646890.2430791.020.05
desc-1α3α-5β(tmpd-4β2β2α)-2α1α1β−1981.6630140.2422741.570.02
desc-1β2α-5α(tmpd-2β2β2β)-1α1α1α−1981.6618530.2411441.590.02
desc-1α2α-1β(tmpd-4β3α3α)-1α1α1α−1981.6625970.2419511.630.02
desc-1α3α-5β(tmpd-4β2β2α)-2α1α1β−1981.6630150.2424921.70.01
desc-1β6α-2β(tmpd-1β2β2α)-1β1α1β−1981.6636880.2433021.790.01
desc-1β6α-5β(tmpd-1β3α3α)-2α6β1α−1981.6636940.2433291.80.01
desc-1β6α-5β(tmpd-1β2β3α)-2α1α1β−1981.6637640.2436611.960.01
desc-1β3β-3β(tmpd-1β2β2α)-1β1β1β−1981.6633320.2434862.130.01
desc-1α6α-2β(tmpd-1β2β3α)-1β1α1β−1981.663460.243762.220.01
desc-1α3β-5β(tmpd-4β3α3α)-2α1α1β−1981.6610740.2414012.240.01
desc-1β3α-5β(tmpd-4β2β3α)-2α1α1β−1981.6629580.2433992.310.01
desc-1β3β-3α(tmpd-1α3β3α)-1β1β1β−1981.6634330.2438922.320.01
desc-1β3α-3β(tmpd-2β2β2β)-1β1β1β−1981.6614810.2419732.340
desc-1β2α-5β(tmpd-1β3β3α)-1α1α1β−1981.6616020.2422582.440
desc-1β2α-5α(tmpd-1α3α3α)-1α1α1β−1981.6615820.2423982.540
desc-1β3β-5α(tmpd-1α2β2α)-2α6β1β−1981.6620750.2429332.570
desc-1α3β-6α(tmpd-1β2β3α)-1β1β1β−1981.6628710.2437532.580
desc-1β3β-3β(tmpd-3β4α2β)-1β1β1β−1981.6616920.2428332.750
desc-1β3β-3β(tmpd-3α4β3α)-1β1o1α−1981.6620810.2434282.870
desc-1β3β-2α(tmpd-6α4α2β)-1β1β1β−1981.6612230.2435733.50
desc-1β2α-2α(tmpd-6α4α2β)-1β1β1β−1981.6604660.2428823.550
desc-1β2α-2α(tmpd-3β4β3α)-1β1α1β−1981.6587150.241753.940
desc-1β3α-3α(tmpd-4α2β3α)-1β1β1β−1981.6593760.2444575.220
desc-1β3β-5β(tmpd-1α2β2β)-1β1β1βbdesc-1β3β-5α(tmpd-1α2β2β)-1β1β1β
desc-1β6α-5β(tmpd-3α3α3α)-1β1α1βdesc-1β6α-1β(tmpd-3α3α3α)-1β1α1β
desc-1β6α-1β(tmpd-3α3α2)-1β1α1βdesc-1β6α-1β(tmpd-3α3α3α)-1β1α1β
desc-1α5β-5β(tmpd-1α2β3α)-1β1α1αdesc-1α5β-5α(tmpd-1α2β3α)-1β1α1α
desc-1β3β-5β(tmpd-4α2β3α)-1β1β1βdesc-1β3β-5α(tmpd-4α2β3α)-1β1β1β
desc-1α5β-5β(tmpd-1α2β2β)-1β1α1βdesc-1α5β-5α(tmpd-1α2β2β)-1β1α1β
desc-1α3β-5β(tmpd-1α2β3α)-1β1β1αdesc-1α3β-5α(tmpd-1α2β3α)-1β1β1α
desc-1α5β-5β(tmpd-2β2β2)-1β1α1αdesc-1α5β-5β(tmpd-2β2β3α)-1β1α1α
desc-1α5β-5α(tmpd-2β2β3α)-1β1α1αdesc-1α5β-5β(tmpd-2β2β3α)-1β1α1α
desc-1α3β-5β(tmpd-1α3α3α)-1β1β1βdesc-1α3β-5α(tmpd-1α3α3α)-1β1β1β
desc-1α2α-1β(tmpd-4β3α2)-1α1α1αdesc-1α2α-1β(tmpd-4β3α3α)-1α1α1α
desc-1α2α-5β(tmpd-4β3α3α)-1α1α1αdesc-1α2α-1β(tmpd-4β3α3α)-1α1α1α
desc-1α5β-5β(tmpd-2β2β2)-1β1α1αdesc-1α5β-5β(tmpd-2β2β3α)-1β1α1α
desc-1α3α-5β(tmpd-22β2α)-2α1α1βdesc-1α3α-5β(tmpd-4β2β2α)-2α1α1β
desc-1α3α-5β(tmpd-4β32α)-2α1α1βdesc-1α3α-5β(tmpd-4β2β2α)-2α1α1β
desc-1β3α-5β(tmpd-22β3α)-2α1α1βdesc-1β3α-5β(tmpd-4β2β3α)-2α1α1β
desc-1β3β-5β(tmpd-1α2β2α)-2α6β1βdesc-1β3β-5α(tmpd-1α2β2α)-2α6β1β
desc-1β3β-3β(tmpd-3β32β)-1β1β1βdesc-1β3β-3β(tmpd-3β4α2β)-1β1β1β
desc-1β3β-2α(tmpd-6α32β)-1β1β1βdesc-1β3β-2α(tmpd-6α4α2β)-1β1β1β
desc-1β2α-2α(tmpd-6α32β)-1β1β1βdesc-1β2α-2α(tmpd-6α4α2β)-1β1β1β

The eight conformations within 1 kcal mol−1 of ΔG accounted for greater than 75% of the calculated population distribution (Table 5), and the predicted most stable conformation was desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β (Fig. 4). The measured VCD and IR spectra of (+)-malathion (CCl4, 0.11 M, BaF2, 72 μm path length) were found to be fairly in good agreement with the population weighted VCD and IR spectra of the eight energetically preferred conformations for (R)-malathion (Fig. 5). The absolute configuration of (+)-malathion can be assigned as (R)-malathion from the VCD analysis and corresponds to the reference assignment.1

Figure 4.

Optimized-geometry of desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β (B3LYP/6-31G*).

Figure 5.

Comparison of the measured VCD (Δϵ) and IR (ϵ) spectra of (+)-malathion (thin line: CCl4, 0.11 M, BaF2, 72 μm path length) with the predicted (population weighted) spectra of eight energetically preferred conformations for (R)-malathion (bold line: B3LYP/6-31G*).

The mean absolute deviation (MAD) of B3LYP/6-31G(d)//B3LYP/6-31G(d) was reported as 7.9 kcal mol−1.28 Furthermore, the Gibbs free energies of each conformation in a vacuum state and in a solution state are quite different, and the comparison of the calculated ΔG with the experimental solution data is extremely difficult. However, the accumulation of substantial conformational data well reproduces the observed VCD spectra of flexible molecules from the trial of chiral alkyl alcohols,7 thalidomide,8 baccatin III,9 (S)-ibuprofen,10 cholesterol derivatives, (–)-cis-permethrin, and pravastatin (unpublished results available at http://staff.aist.go.jp/izumi.h/index-e/index-e.html). If only single (most populated) conformer is stable within 1 kcal mol−1 of ΔG and is reasonably selected, it seems to be possible to compare the experimental VCD spectrum with the predicted VCD spectrum from the single conformer.

Figure 6 indicates the dihedral angles at angle locations of desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β. The description of the conformations using the conformational code is especially useful to verify the accuracy of the conformational search and the determination of the absolute configuration in comparison with the predicted VCD spectrum of each conformation. Figure 7 shows the difference of the predicted VCD band shapes which correspond to the difference of conformations at angle locations D, E, and F, whereas Figure 8 indicates the difference of the predicted VCD band shapes which correspond to the difference of conformations at angle locations B and C. In Figures 7 and 8, desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β is the most stable conformation from the DFT calculations. The predicted VCD band shape did not change much from the conformational variations of tmpd-DEF (Fig. 7) but was largely influenced by the conformational variations of desc-1βB-C(tmpd-1α2β2β)-1β1β1β (Fig. 8). For the large VCD bands of desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β at 1250 cm−1 to 1150 cm−1, the rotation around the single bond of angle location C resulted in only minor changes in the calculated VCD band shape of the conformation. On the other hand, the rotation around the single bonds at angle locations B and G significantly changed the VCD band shape of desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β at 1250 cm−1 to 1150 cm−1. The VCD band shapes of (R)-malathion are largely determined by the relative positional relationship of the C[BOND]O single bonds in the two [BOND]COOC2H5 groups. The conclusion suggests that the methodology of the fragment-conformational search used in this study reasonably elucidates the population-rich conformations of (R)-malathion in the solution state.

Figure 6.

Dihedral angles at angle locations of desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β. Hydrogen atoms are omitted for clarity.

Figure 7.

Comparison of the measured VCD (Δϵ) and IR (ϵ) spectra of (+)-malathion (thin line: CCl4, 0.11 M, BaF2, 72 μm path length) with the predicted spectra of conformations desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β (thin line: B3LYP/6-31G*), desc-1β6α-1β(tmpd-3α3α3α)-1β1α1β (broken line: B3LYP/6-31G*), and desc-1β3β-5α(tmpd-4α2β3α)-1β1β1β (bold line: B3LYP/6-31G*) for (R)-malathion.

Figure 8.

Comparison of the measured VCD (Δϵ) and IR (ϵ) spectra of (+)-malathion (thin line: CCl4, 0.11 M, BaF2, 72 μm path length) with the predicted spectra of conformations desc-1β3β-5α(tmpd-1α2β2β)-1β1β1β (thin line: B3LYP/6-31G*), desc-1β6α-2β(tmpd-1β2β2α)-1β1α1β (broken line: B3LYP/6-31G*), desc-1α5β-5α(tmpd-1α2β2β)-1β1α1β (bold line: B3LYP/6-31G*), and desc-1β3β-3β(tmpd-1β2β2α)-1β1β1β (dotted line: B3LYP/6-31G*) for (R)-malathion.

A full MM calculation can search and find nearly all conformations in a short time with modern programs. However, a selection of conformations for the prediction of VCD is helpful because each additional DFT calculation requires exponentially a large computation time depending on the number of atoms. The actual calculation time for this approach was 2 mo, whereas the estimated time for the calculation of all conformations (39 = 19,683) of (R)-malathion would be about 20 yr. This methodology of the fragment-conformational search is useful to reduce the overall computation time and improve the accuracy of the determination of the important conformations in a relatively short time.

A Monte Carlo-type calculation is essential for a conformational analysis of huge molecules such as proteins today. In the future, the combination technique of the Monte Carlo-type calculation and the DFT calculation would be developed. This approach with the conformational code may contribute toward raising the efficiency of such calculation technique.

(R)-Malathion does not have groups that contribute to the interaction among the fragments such as hydrogen-bonds. The methodology of the fragment-conformational search is especially effective in such cases. In some cases, however, it is expected that two different fragments containing hydrogen-bonding substituents may interact with each other. In such cases, a thorough conformational search of the entire combination of the fragments is also necessary. However, though the way to divide the molecule into fragments is extremely important as well as the retro-synthesis method, the conformational distribution patterns of the fragments are helpful for the conformational search of the entire molecule. Especially, in the case of bioactive compounds, the information of each conformational distribution pattern of the fragments is very useful to understand the bioactivity in detail. Application of the fragment-conformational search with the conformational code to peptide-type compounds is now in progress.

CONCLUSIONS

In this study, a new fragment-conformational search technique with a recently published conformational code for the structural determination of (R)-(+)-malathion using VCD spectroscopy was reported. The use of a fragment-conformational search with the conformational code highlights the essential conformational features of the molecular fragments and affords the clue on the conformational information in a solution state. This methodology allows the selection time of the predominant conformations of large bioactive compounds to be shortened and the accuracy of the determination of absolute configuration is significantly improved. Most large molecules consist of the chemical fragments, and each fragment has characteristic conformational distribution patterns. A database of the conformational distribution patterns of numerous common fragments and an associated conformome analysis would be useful for the design of bioactive compounds in the future.

Ancillary