Received: 30 May 2017 Revised: 6 July 2017 Accepted: 9 July 2017

DOI: 10.1002/chir.22741

REGULAR ARTICLE

Absolute configuration of the ocimene monoterpenoids

from Artemisia absinthium

Luis F. Julio1 | Eleuterio Burgueño‐Tapia2 | Carmen E. Díaz3 | Nury Pérez‐Hernández4 |

Azucena González‐Coloma1 | Pedro Joseph‐Nathan5

1
Instituto de Ciencias Agrarias, Consejo
Superior de Investigaciones Científicas,
Abstract
Madrid, Spain The absolute configuration (AC) of the naturally occurring ocimenes
2
Departamento de Química Orgánica, (−)‐(3S,5Z)‐2,6‐dimethyl‐2,3‐epoxyocta‐5,7‐diene (1) and (−)‐(3S,5Z)‐2,6‐
Escuela Nacional de Ciencias Biológicas,
dimethylocta‐5,7‐dien‐2,3‐diol (2), isolated from the essential oils of domesti-
Instituto Politécnico Nacional, Mexico
City, Mexico cated specimens of Artemisia absinthium, followed by vibrational circular
3
Instituto de Productos Naturales y dichroism (VCD) studies of 1, as well as from the acetonide 3 and the
Agrobiología, Consejo Superior de monoacetate 4, both derived from 2, since secondary alcohols are not the best
Investigaciones Científicas, La Laguna,
Tenerife, Spain
functional groups to be present during VCD studies in solution due to intermo-
4
Escuela Nacional de Medicina y lecular associations. The AC follows from comparison of experimental and
Homeopatía, Instituto Politécnico calculated VCD spectra that were obtained by Density Functional Theory
Nacional, Mexico City, Mexico computation at the B3LYP/DGDZVP level of theory. Careful nuclear magnetic
5
Departamento de Química, Centro de
resonance (NMR) measurements were compared with literature values, provid-
Investigación y de Estudios Avanzados del
Instituto Politécnico Nacional, Mexico ing for the first time systematic 1H and 13C chemical shift data. Regarding
City, Mexico homonuclear 1H coupling constants, after performing a few irradiation experi-
ments that showed the presence of several small long‐range interactions, the
Correspondence
Pedro Joseph‐Nathan, Departamento de complete set of coupling constants for 3, which is representative of the four
Química, Centro de Investigación y de studied molecules, was determined by iterations using the PERCH software.
Estudios Avanzados del Instituto
Politécnico Nacional, Apartado 14‐740,
This procedure even allowed assigning the pro‐R and pro‐S methyl group sig-
Mexico City, 07000 Mexico. nals of the two gem‐dimethyl groups present in 3.
Email: pjoseph@nathan.cinvestav.mx
KEYWORDS
Funding information absolute configuration, monoterpenes, ocimenes, very‐high‐resolution nuclear magnetic resonance,
MINECO/FEDER, Spain, Grant/Award
vibrational circular dichroism
Number: CTQ2015‐64049‐C3‐1‐R; predoc-
toral fellowship to LFJ, Grant/Award
Number: JAE‐CSIC

1 | INTRODUCTION (−)‐cis‐chrysanthenol, linalool, and chrysanthenyl ace-

The perennial Asteraceae Artemisia absinthium L,
popularly known as wormwood or ajenjo in Spain, has
been studied widely. Thujone free populations of this
species have been domesticated for cultivation, thus
providing a chemically stable new variety.1-4 The hydro‐
distilled semi‐industrial vapor pressure essential oils,
characterized by the presence of (−)‐(Z)‐epoxyocimene,
tate,1,4 show antifungal,4 leishmanicidal, and trypanocidal
activities,2,5 as well as toxic effects against Trichinella
spiralis.6 The extraction procedure also generates a water
residue or hydrolate as a by‐product, which shows nema-
ticidal activity7 against Meloidogyne javanica. The major
active component of the hydrolate was characterized
as (−)‐(Z)‐2,6‐dimethylocta‐5,7‐diene‐2,3‐diol,7 which
together with (−)‐(Z)‐epoxyocimene, isolated from the

Chirality. 2017;1–10. wileyonlinelibrary.com/journal/chir © 2017 Wiley Periodicals, Inc. 1

2 JULIO
volatile fraction, are two ocimene monoterpenoids
(Scheme 1) whose absolute configuration (AC) determi-
nation is described in this article.
2 | MATERIALS AND METHODS
2.1 | General
Optical rotations were measured on Perkin‐Elmer
(Boston, MA) 141 or 341 polarimeters. Routine 1H and
13
C nuclear magnetic resonance (NMR) measurements
were performed on a Bruker (Karlsruhe, Germany)
AMX2 500 spectrometer, while very‐high resolution 1H
NMR measurements were performed on a Varian (Palo
Alto, CA) Mercury 300 spectrometer. All NMR studies
were carried out using deuterochloroform solutions,
and chemical shifts were referred to internal
tetramethylsilane (TMS). Infrared (IR) and vibrational
circular dichroism (VCD) spectra were measured using
a BioTools (Jupiter, FL) dual PEM ChiralIR FT‐VCD
spectrophotometer. Mass spectra (MS) were determined
on a Waters (Milford, MA) Micromass Autospec spec-
trometer operated at 70 eV, while high‐resolution elec-
tron impact mass spectra (HRESIMS) data were
obtained using a Waters Micromass LCT Premier
spectrometer.
2.2 | Plant material
A domesticated population of A. absinthium var. Candial
was used in this work. The seedlings were planted8 in
an experimental field at Ejea de los Caballeros, Zaragoza,
Spain, in 2008. A total of 1080 flowering plants, harvested
in August 2016 from all the blocks with a combined
weight of 829 kg, were submitted to semi‐industrial vapor
pressure extraction.
SCHEME 1 Formulas of 1–4
ET AL.
2.3 | Extraction and fractionation
The semi‐industrial vapor pressure extraction was carried
out in a 3000 L stainless steel industrial distillation facility
(http://www.cita‐aragon.es). The 829 kg sample of vegetal
material was steam‐distilled for 1 h to provide 2.3 L of
essential oil and 5 L of hydrolate.
A sample of 6 g of the essential oil was column
chromatographed on 150 g of silica gel (35−70 μm) with
increasing polarity mixtures of hexanes‐AcOEt (0−5%)
to provide 1.48 g of a fraction rich in 1. A 100‐mg portion
of this fraction was further purified by flash column chro-
matography on a prepacked cartridge (ExtraBond Flash
OT SI 20 g 70 mL 26.8 x 154 mm, Scharlau, Barcelona,
Spain) eluting with increasing polarity mixtures of hex-
anes‐AcOEt (0−5%) at a flow rate of 18 mL/min to afford
75 mg of a highly enriched sample of 1.
The hydrolate was filtered and solid phase‐extracted by
stirring in the presence of 280 g of activated charcoal for 1 h.
After filtering, the black solid was oven‐dried with an air
flow at 50 °C for 24 h. The adsorbed organic material was
extracted in a Soxhlet with AcOEt for 4 h. The resulting
organic extract was filtered and evaporated under reduced
pressure to give 6.5 g (2.3%) of an oily residue. A 6 g portion
thereof was chromatographed on a 150 g Si‐gel column
eluting with mixtures of CH2Cl2:AcOEt of increasing polar-
ity (1:0 to 7:3) to afford 2262 mg (37.7%) of 2.
2.4 | Compounds
2.4.1 | (−)‐(3S,5Z)‐2,6‐Dimethyl‐2,3‐
epoxyocta‐5,7‐diene (1)
Oil. [α]589 −10.3, [α]578 −11.1, [α]546 −12.3, [α]436 −17.8 (c
1.15, CHCl3). 1H NMR δ 6.73 (1H, ddd, J = 17.3, 10.7,
1.0 Hz, H‐7), 5.41 (1H, t, J = 7.7 Hz, H‐5), 5.24 (1H, d,
J = 17.0 Hz, H‐8 cis), 5.13 (1H, dt, J = 11.0, 1.5 Hz, H‐8
trans), 2.74 (1H, t, J = 6.3 Hz, H‐3), 2.48 (1H, m, H‐4),
2.32 (1H, m, H‐4′), 1.84 (3H, m, H‐10), 1.30 (3H, s, H‐1
or H‐9), 1.29 (3H, s, H‐1 or H‐9). 13C NMR δ 134.5 (C‐6),
133.3 (C‐7), 124.9 (C‐5), 114.4 (C‐8), 63.6 (C‐3), 58.3 (C‐
2), 27.4 (C‐4), 24.8 (C‐1 or C‐9), 19.8 (C‐10), 18.7 (C‐1 or
C‐9). EIMS m/z (rel. Int.): 152 [M]+ (0.4), 93 (13), 91
(12), 81 (50), 79 (100), 77 (29), 71 (9), 53 (15), 43 (16), 41
(22), 39 (19). HREIMS: m/z 152.1200 [M]+, calcd for
C10H16O, 152.1201.
2.4.2 | (−)‐(3S,5Z)‐2,6‐Dimethylocta‐5,7‐
dien‐2,3‐diol (2)
Oil, [α]589 −50.2 (c 0.21, CHCl3). 1H NMR δ 6.74 (1H, ddd,
J = 17.3, 11.0, 0.8 Hz, H‐7), 5.48 (1H, t, J = 7.7 Hz, H‐5),
5.25 (1H, d, J = 17.3 Hz, H‐8 cis), 5.13 (1H, dt, J = 10.7,
JULIO ET AL. 3

1.6 Hz, H‐8 trans), 3.42 (1H, m, H‐3), 2.33 (2H, m, H‐4),
1.86 (3H, m, H‐10), 1.23 (3 H, s, H‐1 or H‐9), 1.19 (3H, s,
H‐1 or H‐9). 13C NMR δ 135.3 (C‐6), 133.3 (C‐7), 126.9
(C‐5), 114.6 (C‐8), 77.9 (C‐3), 72.7 (C‐2), 29.8 (C‐4), 26.5
(C‐1 or C‐9), 23.6 (C‐1 or C‐9), 19.9 (C‐10). EIMS m/z
(rel int): 170 [M]+ (1), 82 (57), 81 (51), 79 (35), 71 (30),
68 (20), 67 (54), 59 (100), 55 (21), 43 (59), 41 (36).
HREIMS: m/z 170.1304 [M]+, calcd for C10H18O2,
170.1307.
2.4.3 | (−)‐(5S)‐2,2,4,4‐Tetramethyl‐5‐[(2Z)‐
3‐methylpenta‐2,4‐dien‐1‐yl]‐1,3‐dioxolane
(3)
A solution containing 250 mg of 2 in 25 mL of anhydrous
acetone was stirred in the presence of 1 g of anhydrous
CuSO4 for 10 h. The reaction mixture was filtered, the
solvent was removed under reduced pressure, and the res-
idue was purified by flash chromatography eluting with
CH2Cl2 to afford 166.1 mg of 3 as oil. [α]589 −14.1, [α]578
−15.6, [α]546 −17.8, [α]436 −31.6, [α]365 −54.3 (c 1.00,
CHCl3).1H NMR see Table 1. 13C NMR δ 134.1 (C‐6),
133.3 (C‐7), 125.9 (C‐5), 114.4 (C‐8), 106.6 (acetonide C),
83.0 (C‐3), 80.2 (C‐2), 28.5 (C‐13), 27.6 (C‐4), 26.9 (C‐11),
26.3 (C‐9), 22.9 (C‐1), 19.8 (C‐10). EIMS m/z (rel int):
210 [M]+ (0,2), 195 (9), 129 (52), 81 (47), 79 (38), 77 (10),
71 (32), 59 (100), 53 (9), 43 (38), 41 (17). HREIMS: m/z
210.1617 [M]+, calcd for C13H22O2, 210.1620.
2.4.4 | (+)‐(3S,5Z)‐2,6‐Dimethylocta‐5,7‐
dien‐2‐ol‐3‐yl acetate (4)
A solution containing 100 mg of 2 in 0.1 mL of anhydrous
acetic acid and 5 mg of NaOH was heated in a domestic
Corbero microwave oven9 operated at 480 W for 5 min
twice. The solution was diluted with 2 mL of distilled
water and extracted twice with AcOEt. The organic layers
were combined and the solvents were removed under
reduced pressure. The residue was purified by flash chro-
matography eluting with a mixture of hexanes‐AcOEt
(1:5) to afford 78 mg of oil. [α]589 +58.3, [α]578 +61.0,
[α]546 +70.4, [α]436 + 132.8, [α]365 + 242.3 (c 0.98,
CHCl3).1H NMR δ 6.74 (1H, ddd, J = 17.3, 10.7, 0.6 Hz,
H‐7), 5.34 (1H, t, J = 7.6 Hz, H‐5), 5.23 (1H, d,
J = 17.0 Hz, H‐8 cis), 5.12 (1H, m, H‐8 trans), 4.83 (1H,
dd, J = 9.6, 3.3 Hz, H‐3), 2.55 (1H, m, H‐4), 2.44 (1H, m,
H‐4′), 2.03 (3H, s, acetate Me), 1.80 (3H, m, H‐10, 1.22
(6 H, s, H‐1 and H‐9). 13C NMR δ 170.8 (carbonyl), 134.7
(C‐6), 133.4 (C‐7), 125.8 (C‐5), 114.3 (C‐8), 79.2 (C‐3),
72.3 (C‐2), 27.7 (C‐4), 26.7 (C‐1 or C‐9), 25.2 (C‐1 or C‐
9), 20.9 (acetate Me), 19.8 (C‐10), EIMS m/z (rel int): 212
[M]+ (not observed), 152 [M ‐ AcO]+ (17), 137 (13), 109
(59), 94 (16), 93 (16), 81 (35), 79 (43), 67 (11), 59 (30), 43
(100). HRESI‐TOFMS: m/z 235.1306 [M + Na]+, calcd
for C12H20O3Na 235.1310.
2.5 | VCD measurements
Samples of 14.1 mg of an 80% mixture of 1 and achiral
contaminants, of 8.4 mg of 3, and of 9.6 mg of 4 were in
each case dissolved in 150 μL of 100% atom‐D CDCl3
and placed in spectrophotometer cells having BaF2
windows and a pathlength of 0.1 mm for data acquisition
at a resolution of 4 cm−1 over 6 h. A baseline correction
was performed by subtracting the spectrum of the solvent
acquired under identical instrument conditions. The
sample0 s stability was monitored in either case by
300 MHz 1H NMR analysis immediately before and after
the VCD measurements.

TABLE 1 Accurate 1H NMR parameters of acetonide 3 derived from very‐high‐resolution measurements and PERCH iterations
1
H δ Multiplicity J
4
1 1.126 dq J1,3 = 0.13, 4J1,9 = 0.33
4
3 3.750 qqdd J1,3 = 0.13, 4J3,9 = 0.12, 3J3,4R = 5.55, 3J3,4S = 8.08
4R 2.379 dddddq 3
J3,4R = 5.55, 2J4R,4S = −15.15, 3J4R,5 = 8.29, 6J4R,8cis = 0.34, 6J4R,8trans = 0.34, 5J4R,10 = 1.03
4S 2.444 dddddq 3
J3,4S = 8.08, 2J4R,4S = −15.15, 3J4S,5 = 6.32, 6J4S,8cis = 0.43, 6J4S,8trans = 0.56, 5J4S,10 = 1.51
3
5 5.423 dddddq J4R,5 = 8.29, 3J4S,5 = 6.32, 4J5,7 = 0.93, 5J5,8cis = 0.80, 5J5,8trans = 1.72, 4J5,10 = 1.33
4
7 6.751 ddd J5,7 = 0.93, 3J7,8cis = 17.29, 3J7,8trans = 10.84
6
8cis 5.250 ddddd J4R,8cis = 0.34, 6J4S,8cis = 0.43, 5J5,8cis = 0.80, 3J7,8cis = 17.29, 2J8cis,8trans = 1.52
6
8trans 5.141 ddddd J4R,8trans = 0.34, 6J4S,8trans = 0.56, 5J5,8trans = 1.72, 3J7,8trans = 10.84, 2J8cis,8trans = 1.52
4
9 1.250 qd J1,9 = 0.33, 4J3,9 = 0.12
5
10 1.848 ddd J4R,10 = 1.03, 5J4S,10 = 1.51, 4J5,10 = 1.33
4
11 1.344 q J11,13 = 0.67
4
13 1.430 q J11,13 = 0.67
4 JULIO ET AL.

2.6 | Computational methods and in 20 conformers for 4 in a 2.95 kcal/mol energy

gap, the next conformer being 3.22 kcal/mol above the
The molecular models of 1, 3, and 4, were constructed in
global minimum. Further conformational optimizations
the Spartan 08 (Wavefunction, Irvine, CA) suite, and the
were performed using the Gaussian 03 (Gaussian,
initial conformational search for these three molecules
Wallingford, CT) software to provide the thermochemical
were carried out using the Monte Carlo protocol and
parameters summarized in Table 2. For calculating vibra-
molecular mechanics force field (MMFF94) calculations
tional normal modes and rotational strengths, using DFT
in energy windows of 10 kcal/mol. In the case of epoxide
at the B3LYP/DGDZVP level of theory, the value of ΔG
1 this provided 10 conformers in a 5.46 kcal/mol energy
was estimated for each conformer and used as a criterion
gap, for acetonide 3 it provided 19 conformers in a
for weighting the IR and VCD spectra for 1, 3, and 4.
7.96 kcal/mol gap, and for monoacetate 4 it provided
Finally, four, four, and 10 conformers, for 1, 3, and 4,
100 conformers in a 9.99 kcal/mol energy gap. Single‐
respectively, were found within a 2 kcal/mol range. The
point calculations of all conformers were done using Den-
band‐shapes were calculated with Lorentzian functions
sity Functional Theory (DFT) with the B3LYP functional
and a bandwidth of 6 cm−1. Calculated and experimental
and the DGDZVP basis set implemented in the same soft-
spectra were compared using the CompareVOA
ware package. This resulted in four conformers for 1 in a
(BioTools) software,10 and the pertinent parameters are
0.85 kcal/mol energy range, the next conformer being
shown in Table 3. The final graphical comparison of
3.36 kcal/mol above the global minimum; in five con-
experimental and calculated IR and VCD spectra shown
formers for 3 in a 2.23 kcal/mol, interval, the next con-
in Figures 2–4 were prepared using four, four, and 10
former being 3.29 kcal/mol over the global minimum;
conformers for 1, 3, and 4, respectively.

TABLE 2 Thermochemical parameters of ocimenes (S)‐1, (S)‐3, and (S)‐4

ΔEMMFFa %b ΔE6‐31G(d)c %b ΔEDGDZVPd %b ΔGDGDZVPe %f

(S)‐1a 0.56 14.6 0.28 24.2 0.01 36.5 0.00 44.8
(S)‐1b 0.02 36.1 0.00 39.1 0.00 36.7 0.17 33.6
(S)‐1c 0.00 37.2 0.22 26.9 0.41 18.2 0.70 13.7
(S)‐1d 1.13 5.5 0.85 9.4 0.86 8.6 1.03 7.9
(S)‐3a 0.00 63.5 0.31 28.0 0.00 47.5 0.00 50.8
(S)‐3b 1.39 6.1 0.75 13.2 0.67 15.4 0.32 29.8
(S)‐3c 1.46 5.4 0.92 10.0 0.80 12.2 0.89 11.2
(S)‐3d 1.31 6.9 0.00 47.2 0.41 23.8 1.15 7.3
(S)‐3e 2.46 1.0 2.23 1.1 2.21 1.1 2.39 0.9
(S)‐4a 1.44 2.8 1.89 1.4 0.97 5.1 0.00 21.8
(S)‐4b 0.61 11.1 0.19 24.1 0.00 26.3 0.09 18.7
(S)‐4c 2.40 0.5 1.24 4.1 0.95 5.3 0.18 16.0
(S)‐4d 0.35 17.2 0.74 9.6 0.45 12.3 0.36 11.9
(S)‐4e 2.76 0.3 2.36 0.6 1.47 2.2 0.52 9.1
(S)‐4f 0.58 11.8 0.00 33.4 0.22 18.2 0.57 8.3
(S)‐4 g 1.93 1.2 1.09 5.3 0.88 5.9 0.82 5.4
(S)‐4 h 2.90 0.2 2.90 0.2 2.30 0.5 1.03 3.8
(S)‐4i 3.86 0.0 2.03 1.1 1.90 1.1 1.31 2.4
(S)‐4j 1.11 4.8 1.14 4.8 1.13 3.9 1.39 2.1
(S)‐4 k 0.00 31.3 0.96 6.7 0.77 7.18 2.33 0.4
a
Relative to 1c (24.58 kcal/mol), 3a (45.62 kcal/mol), 4 k (32.62 kcal/mol).
b
Calculated according to ΔE ≅ − RT ln K.
c
Relative to 1b (−292327.45 kcal/mol), 3d (−413560.00 kcal/mol), 4f (−436101.52 kcal/mol).
d
Relative to 1b (−292360.10 kcal/mol), 3a (−413607.66 kcal/mol), 4b (− 436158.29 kcal/mol).
e
Relative to 1a (−292336.91 kcal/mol), 3a (−413430.82 kcal/mol), 4a (−435998.34 kcal/mol).
f
Calculated according to ΔG = −RT ln K.
JULIO ET AL.
TABLE 3 Confidence level for the IR and VCD spectra compari-
son of 1, 3, and 4
Compound anHa SIRb SEc S−Ed ESIe
(S)‐1 0.981 72.0 70.2 4.9 65.2
(S)‐3 0.976 94.8 81.3 17.7 63.6
(S)‐4 0.985 96.0 80.1 9.6 70.5
a
Anharmonicity factor.
b
IR spectral similarity in percentage.
c
VCD spectral similarity for the correct enantiomer in percentage.
d
VCD spectral similarity for the opposite enantiomer in percentage.
e
Enantiomer similarity index, calculated as the SE − S−E difference.
f
Confidence level for the absolute configuration determination in percentage.
3 | RESULTS AND DISCUSSION
The method of choice to determine the AC of the ocimene
monoterpenoids found in Artemisia absinthium is VCD,
which has been used extensively for this group of small
terpenoids.11,12 Although we had no direct AC determina-
tion experience with this type of compound, it turned out
from the beginning of this study that, due to their size,
these molecules are quite volatile and must therefore be
treated accordingly, in particular when recovering
samples during the process of going from the initial 1H
NMR spectrum to the VCD measurement, and then back
to the 1H NMR determination, a precautionary routine we
always use to be sure the studied compound are stable
during the long‐time vibrational optical activity
measurements.
It is well known that primary and secondary hydroxy
groups contained in organic molecules are prone to show
relevant intermolecular associations that render, in
relatively poor VCD data, comparisons between experi-
mental and calculated spectra.13,14 Therefore, although
dimethyloctadienediol 2 is a constituent of A. absinthium,
it was not subjected to VCD studies. Instead, both its
derived acetonide 3 and monoacetate 4 were studied by
this technique, together with naturally occurring epoxide
1. Both 3 and 4 were obtained by simple procedures
outlined in the Materials and Methods, although it should
be noted that acetylation of (−)‐2 provides monoacetate
(+)‐4, showing that AC assignments based on optical rota-
tion analogy can lead to errors, in line with our observa-
tions on angular 3′‐acyloxypyranocoumarins,15 where
the replacement of an ester for another ester in several
cases changes the sign of the optical rotation. It should
further be noted that preparation of acetonide 3 and
monoacetate 4 from diol 2 occurs by replacement of
hydroxy groups hydrogen atoms for substituents without
touching either bond of the sole stereogenic center of 2,
which means the AC of the natural product is retained
in derivatives 3 and 4.
5
Before VCD measurements, it was mandatory to accu-
rately establish the identity of the studied ocimenes, so
Cf they were subjected to NMR, MS, and polarimetric stud-
100
ies; the pertinent data are summarized in the Materials
and Methods. Of relevance are certain long‐range 1H cou-
100
pling constants that became evident in the spectra of these
100 compounds, and therefore, as a representative molecule
of the series, we undertook a very detailed 1H NMR study
of acetonide 3, since this molecule also contains two gem‐
dimethyl arrangements that deserve individual 1H and
13
C NMR assignments.
A very‐high‐resolution 300 MHz 1H NMR spectrum of
3 was obtained from an ultrasound argon‐bubbled CDCl3
degassed solution for which we reached a 0.13 Hz magnet
homogeneity after careful manual shimming under
temperature regulation at 27 °C. This revealed that the
acetonide methyl groups are quartets (4Jq = 0.67 Hz)
due to mutual spin coupling, and posed the question
why such an interaction was not directly evident for the
gem‐dimethyl signals at C‐2. Since these two methyl sig-
nals, owing to C‐1 and C‐9, were not very sharp singlets,
it was assumed that in addition to mutual coupling, they
should have an additional long‐range coupling, the logical
candidate for which should be H‐3 at δ 3.75. Effectively,
irradiation of the H‐3 signal removed a long‐range cou-
pling, thus allowing observation of the H‐1 and H‐9 quar-
tets. In turn, irradiation of the H‐10 methyl signal at δ 1.85
changed various multiplicities in the spectrum, and there-
fore to completely define the overall spin system it was
decided to evaluate it using the PERCH v. 2011.1 soft-
ware16 (PERCH Solutions, Kuopio, Finland) as we have
done for other terpenoids.17,18
The methodology used for the PERCH evaluation of
the spin system of 3 is similar to that recently described
for hydroxy‐bis‐farinosin.19 Thus, on the one hand, the
300 MHz 1H NMR FID data were imported into the prep-
aration module of the PERCH shell, followed by Fourier
transformation, phase and baseline correction, peak pick-
ing, and integration. On the other hand, the most stable
molecular conformer of 3 was calculated in the molecular
modeling portion of the PERCH software and used to
obtain an initial predicted spectrum. For this purpose
the chemical shifts of a manually assigned spectrum,
which are in agreement with those having partial assign-
ments,20,21 as well as several directly observed coupling
constant values, were manually fed into the table of the
graphical spectral parameter editor of the PERCH soft-
ware. The resulting set of values was then submitted to
the iterator module of the software, using the total‐line‐
shape‐fitting method, which allowed obtaining an initial
calculated spectrum. For the distinction of the
diastereotopic hydrogen atoms at the C‐4 methylene,
the assignments were interchanged to show which
6 JULIO
total‐line‐shape fitting analysis gave the lowest root mean
square (RMS). A similar procedure was used to assign the
pro‐R and pro‐S methyl groups of the two gem‐dimethyl
groups present in 3. In each interchange of signals the
optimization of the spectral parameters was carried out
using the total‐line‐shape fitting analysis, obtaining, at
the end of this process, an excellent agreement between
the experimental and the calculated spectra, with a final
RMS error of 0.054%. Full 1H NMR data of 3 are summa-
rized in Table 1, while the experimental and simulated 1H
NMR spectra are compared in Figure 1. Line narrowing
using mathematical algorithms, like weighting functions
for resolution enhancement, was avoided since it becomes
critical to determine at which point a given weighted
NMR spectrum starts to look unreal. It should also be
noted that although PERCH‐calculated spectra provide
chemical shift values with six significant figures after the
decimal point, and because the experimental 300 MHz
1
H NMR spectrum was acquired with a magnet homoge-
neity of 0.13 Hz, the chemical shifts in Table 1 are given
with three digits after the decimal point, which consti-
tutes a proper description, as is usual.22-25 The table also
reports PERCH‐derived J values with two figures after
the decimal point, which the software outputs with four.
The 4R and 4S distinction follows from the coupling
FIGURE 1 Comparison of the PERCH calculated (top) and the experimental (bottom) 300 MHz 1H NMR spectra of acetonide 3. The
labeled (*) peak is due to moisture
ET AL.
constant values, while in the case of the diastereotopic
methyl groups, inversion of the signal assignment of a
given pair slightly increases the RMS value of the fit.
Once all hydrogen atoms were securely assigned, the
13
C NMR data summarized in the Materials and Methods
followed from a routine gHSQC experiment.
The conjugated double bonds system present in these
ocimenes has been drawn either as a folded4,26 s‐cis or
an extended20,21,27 s‐trans diene, and therefore, for the
DFT calculation needed for the VCD studies, it is of rele-
vance to know if both geometries coexist, in which case
the relative abundance of them also has to be estimated,
or if one atom arrangement highly predominates over
the other one. Inspection of a 500 MHz 1H NMR NOESY
plot of 3 reveals the extended s‐trans geometry from the
interactions of H‐4 at δ 2.41 and H‐7 at δ 6.76, while there
is no evidence for an interaction between H‐7 at δ 6.76
and H‐10 at δ 1.85. In turn, the Z configuration of the
C‐5/C‐6 double bond is confirmed by the NOE interaction
of H‐4 at δ 2.41 and H‐7 at δ 6.76. From there it follows
that the extended s‐trans arrangement predominates over
the s‐cis arrangement. Cogent confirmation of the exclu-
sive existence of only one arrangement follows from
molecular mechanics calculations, which reveal a
7 kcal/mol difference in favor of the C‐6/C‐7 s‐trans
JULIO ET AL.
FIGURE 2 Comparison of the experimental IR (b) and VCD (d)
spectra of (−)‐1 with the DFT B3LYP/DGDZVP calculated IR (a)
and VCD (c) spectra for (S)‐1
arrangement as compared to the s‐cis arrangement. A
similar 7 kcal/mol difference was observed during the
MM calculations of 1 and 4, and although MM energy is
not sufficient to access final conformational distributions,
experience dictates that the order of contributing con-
formers can only vary in less than the initial 3 kcal/mol
interval.
Once the structure and geometry of the studied
ocimenes were secured, we turned our attention to the
calculation procedures of the VCD spectra that, in order
to determine the absolute configuration, have to be
contrasted with the experimental spectra. For this pur-
pose we arbitrarily selected the 3S absolute configuration,
for which the molecular models of 1, 3, and 4 were con-
structed in the Spartan 08 suite and subjected to an initial
7
FIGURE 3 Comparison of the experimental IR (b) and VCD (d)
spectra of (−)‐3 with the DFT B3LYP/DGDZVP calculated IR (a)
and VCD (c) spectra for (S)‐3
Monte Carlo conformational search, using molecular
mechanics force field (MMFF94). The simplest molecule
to calculate is epoxide 1, which has only 84 electrons
and is the smallest molecule we ever calculated over the
last 11 years. The conformational freedom of 1 is
restricted by the 2,3‐epoxide and by the two double bonds,
leaving only rotation freedom at the C‐3/C‐4 and C‐4/C‐5
bonds because C‐6/C‐7 is trans, which resulted in a total
of 10 conformers in a 5.46 kcal/mol energy gap. Single‐
point calculations of these 10 conformers were done using
DFT at the B3LYP/DGDZVP level of theory, as imple-
mented in the Spartan 08 software package, which pro-
vided only four conformers in the lowest 0.85 kcal/mol
gap. Further conformational optimizations for these four
conformers were done using the Gaussian 03 software,
which provided the thermochemical parameters that are
8 JULIO
FIGURE 4 Comparison of the experimental IR (b) and VCD (d)
spectra of (+)‐4 with the DFT B3LYP/DGDZVP calculated IR (a)
and VCD (c) spectra for (S)‐4
given in Table 2. Finally, calculation of the vibrational
normal modes and associated rotational strengths was
also done with the Gaussian 03 software using DFT at
the same B3LYP/DGDZVP level of theory. The obtained
ΔG values for each conformer were used as the criteria
for obtaining the Boltzmann distribution weighted IR
and VCD spectra of 1.
A similar procedure was undertaken in the case of
acetonide 3, which contains 116 electrons. The Monte
Carlo search provided 19 conformers in a 7.69 kcal/mol
range, which were reduced to five conformers in a
2.23 kcal/mol energy gap after single‐point optimization.
The Gaussian optimization located the four contributing
conformers in a narrow 0.80 kcal/mol gap, and finally cal-
culations of the vibrational normal modes and associated
ET AL.
FIGURE 5 The minimum energy conformers of acetonide 3 and
of monoacetate 4
rotational strengths placed these four conformers in a
1.15 kcal/mol energy gap. Again, the four conformers
were Boltzmann‐weighted to obtain the calculated IR
and VCD spectra of 3.
In the case of 4, the initial Monte Carlo search gave
100 conformers in a 9.99 kcal/mol gap, which single‐point
calculations reduced to 20 conformers in a 2.95 kcal/mol
gap, to leave at the end 10 conformers (Table 2) in a
1.4 kcal/mol energy interval, in which both 1 and 3 show
only four conformers. This result deserves two brief com-
ments, which are that for 4 some conformational freedom
is also associated with the hydroxy and acetoxy groups,
and, more important, it should be observed in Table 2 that
the eleventh conformer of the molecular mechanics force
field calculation turned out to be the most stable con-
former after the complete energy calculation procedure.
The band‐shapes of the so‐far computed IR and VCD
spectra of 1, 3, and 4 were calculated with Lorentzian
functions and a bandwidth of 6 cm−1. Calculated and
experimental spectra were contrasted using the
CompareVOA10 software, which provided the parameters
summarized in Table 3, thus revealing that the absolute
configuration of the C‐3 stereogenic center is S in all
cases. Of relevance is to note that for 1 the similarity of
JULIO ET AL.
the IR spectra is relatively low (72%), while for 3 and 4, as
frequently happens, it is far above 90%. This is in line with
the purity of the studied sample of 1, which contained
some 20% of nonchiral contaminants. The final graphical
comparison of experimental and calculated IR and VCD
spectra shown in Figures 2−4 were prepared using four,
four, and 10 conformers for 1, 3, and 4, respectively.
Finally, it should be noted that the absolute configura-
tion assignment of the ocimenes is based on visual inspec-
tion of the phases of the bands shown in Figures 2−4 in
combination with the data given in Table 3. In the present
study we are dealing with small monoterpenes possessing
a sole stereogenoic center, and there is no reasonable need
to do further band analysis. The two main conformers of
acetonide 3 and the four lower energy conformers of
monoacetate 4 are depicted in Figure 5.
4 | CONCLUSION
The AC of naturally occurring (−)‐(3S,5Z)‐2,6‐dimethyl‐
2,3‐epoxyocta‐5,7‐diene (1) followed after comparison of
the experimental and DFT‐calculated VCD spectra
obtained at the B3LYP/DGDZVP level of theory. In the
case of the also naturally occurring (−)‐(3S,5Z)‐2,6‐
dimethylocta‐5,7‐dien‐2,3‐diol (2), the AC followed from
the study of the derived (−)‐(5S)‐2,2,4,4‐tetramethyl‐5‐
[(2Z)‐3‐methylpenta‐2,4‐dien‐1‐yl]‐1,3‐dioxolane (3) and
(+)‐(3S,5Z)‐2,6‐dimethylocta‐5,7‐dien‐2‐ol‐3‐yl acetate
(4), since secondary alcohols are not good candidates for
VCD studies due to intermolecular associations in solu-
tion that unnecessarily complicate the calculations. The
results show that the simple acetylation of a secondary
alcohol changes the sign of the optical rotation,
demonstrating once again that AC determinations based
on the analogy of optical activity can lead to wrong
results. We also completely assigned the 1H NMR spec-
trum of 3, including the determination of several long‐
range coupling constants between atoms separated by 4,
5, or 6 bonds, as well as the assignment of the individual
signals owing to the pro‐R and pro‐S methyl groups of
the gem‐dimethyl groups of both the monoterpenoid and
the acetonide portions of the molecule. With the complete
1
H NMR assignment in hand, the 13C NMR signal assign-
ments became trivial from a gHSQC measurement.
A C K N O WL E D G M E N T S
This work was supported by grant CTQ2015‐64049‐C3‐1‐
R (MINECO/FEDER, Spain), and JAE‐CSIC (predoctoral
fellowship to LFJ). We thank Ing. Jesús Burillo (CITA)
for the pilot plant extraction.
9
ORCID
Pedro Joseph‐Nathan http://orcid.org/0000-0003-3347-
3990
RE FER EN CES
1. Martín L, Julio LF, Burillo J, Sanz J, Mainar AM, González‐
Coloma A. Comparative chemistry and insect antifeedant action
of traditional (Clevenger and Soxhlet) and supercritical extracts
(CO2) of two cultivated wormwood (Artemisia absinthium L.)
populations. Ind Crop Prod. 2011;34:1615‐1621.
2. González‐Coloma A, Bailen M, Díaz CE, Fraga BM, Martínez‐
Díaz R, Zuñiga GE, Contreras RA, Cabrera R, Burillo J. Major
components of Spanish cultivated Artemisia absinthium popula-
tions: Antifeedant, antiparasitic, and antioxidant effects. Ind
Crop Prod. 2012;37:401‐407.
3. Bailen M, Julio LF, Díaz CE, Sanz J, Martínez‐Díaz RA, Cabrera
R, Burillo J, Sanz J, González‐Coloma A. Chemical composition
and biological effects of essential oils from Artemisia
absisnthium L. cultivated under different environmental condi-
tions. Ind Crop Prod. 2013;49:102‐107.
4. Julio LF, Burillo J, Giménez C, Cabrera R, Díaz CE, González‐
Coloma A. Chemical and biocidal characterization of two
cultivated Artemisia absinthium populations with different
domestication levels. Ind Crop Prod. 2015;76:787‐792.
5. Martínez‐Díaz RA, Ibáñez‐Escribano A, Burillo J, De las Heras
L, Del Prado G, Agulló‐Ortuño MT, Julio LF, González‐
Coloma A. Trypanocidal, trichomonacidal and cytotoxic
components of cultivated Artemisia absinthium Linnaeus
(Asteraceae) essential oil. Mem Inst Oswaldo Cruz. 2015;110:
93‐699.
6. García‐Rodríguez JJ, Andrés MF, Ibañez‐Escribano A, Julio LF,
Burillo J, Bolás‐Fernández F, González‐Coloma A. Selective
nematocidal effects of essential oils from two cultivated Artemisia
absinthium populations. Z Naturforsch. 2015;70:275‐280.
7. Julio LF, González‐Coloma A, Burillo J, Díaz CE, Andrés MF.
Nematicidal activity of the hydrolate byproduct from the semi
industrial vapor pressure extraction of domesticated Artemisia
absinthium against Meloidogyne javanica. Crop Prot.
2017;94:33‐37.
8. Burillo J. Cultivo experimental de ajenjo Artemisia absinthium L.
como potencial insecticida de origen natural. In: Burillo J,
González‐Coloma A, eds. Insecticidas y Repelentes de Origen
Natural, Centro de Investigación y Tecnología Agroalimentaria.
Zaragoza, Spain; 2009:19‐30.
9. Patnam R, Chang FR, Kuo RY, Pan WB, Wu YC. Microwave
enhanced acetylation of alcohols. J Chem Res S. 2002;6:301‐302.
10. Debie E, De Gussem E, Dukor RK, Herrebout W, Nafie LA,
Bultinck P. A confidence level algorithm for the determination
of absolute configuration using vibrational circular dichroism
or Raman optical activity. ChemPhysChem Spec Iss Jacobus van0 t
Hoff. 2011;12:1542‐1549.
11. Joseph‐Nathan P, Gordillo‐Román B. Vibrational circular
dichroism absolute configuration determination of natural prod-
ucts. In: Kinghorn AD, Falk H, Kobayashi J, eds. Progress in the
10
Chemistry of Organic Natural Products. Vol.100 Switzerland:
Springer International Publishing; 2015:311‐451.
12. Burgueño‐Tapia E, Joseph‐Nathan P. Vibrational circular
dichroism: Recent advances for the assignment of the absolute
configuration of natural products. Nat Prod Commun.
2015;10:1785‐1795. Reprinted 2017;12:641–651
13. Burgueño‐Tapia E, Joseph‐Nathan P. Absolute configuration of
eremophilanoids by vibrational circular dichroism. Phytochemis-
try. 2008;69:2251‐2256.
14. Burgueño‐Tapia E, Sánchez‐Castellanos M, Joseph‐Nathan P.
Optimization of the considered number of conformers for
the absolute configuration determination of catechin and
epicatechin peracetates by VCD. Nat Prod Comun. 2017;12:
683‐686.
15. Buendía‐Trujillo AI, Torres‐Valencia JM, Joseph‐Nathan P,
Burgueño‐Tapia E. The absolute configuration of angular 3′‐
acyloxypyranocoumarins by vibrational circular dichroism
exciton chirality. Tetrahedron Asymmetry. 2014;25:1418‐1423.
16. Laatikainen R, Tiainen M, Korhonen S‐P, Niemitz M. Comput-
erized analysis of high‐resolution solution‐state spectra. In:
Harris RK, Wasylishen RE, eds. Encyclopedia of Magnetic
Resonance. Chichester, UK: Wiley; 2012:677.
17. Ortega AR, Sánchez‐Castellanos M, Pérez‐Hernández N, Robles‐
Zepeda RE, Joseph‐Nathan P, Quijano L. Relative stereochemis-
try and absolute configuration of farinosin, an eudesmanolide
from Encelia farinosa. Chirality. 2016;28:415‐419.
18. Pérez‐Hernández N, Gordillo‐Román B, Arrieta‐Baez D, Cerda‐
García‐Rojas CM, Joseph‐Nathan P. Complete 1H NMR assign-
ments of cedranolides. Magn Reson Chem. 2017;55:169‐176.
19. Ortega AR, Ortiz‐Pastrana N, Quijano L, Becera‐Martínez E,
Olmedo‐Aguirre JO, Joseph‐Nathan P. Structure and absolute
configuration of hydroxy‐bis‐dihydrofarinosin from Encelia
farinosa. Magn Reson Chem. 2017;55:530‐539.
20. Tsankova E, Bohlmann F. A monoterpene from Aster bakeranus.
Phytochemistry. 1983;22:1285‐1286.
21. Moore CJ, Possner S, Hayes P, Paddon‐Jones GC, Kitching W.
An asymmetric dihydroxylation route to (3R,5E)‐2,6‐dimethyl‐
JULIO ET AL.
2,3‐epoxyocta‐5,7‐diene: The major volatile component from
male fruit‐spotting bugs. J Org Chem. 1999;64:9742‐9744.
22. Becerra‐Martínez E, Ramírez‐Gualito KE, Pérez‐Hernández N,
Joseph‐Nathan P. Total 1H NMR assignment of 3β‐
acetoxypregna‐5,16‐dien‐20‐one. Steroids. 2015;104:208‐213.
23. Molina‐Salinas GM, Rivas‐Galindo VM, Said‐Fernández S,
Lankin DC, Muñoz MA, Joseph‐Nathan P, Pauli GF, Waksman
N. Stereochemical analysis of leubethanol, and anti‐TB‐active
serrulatane from Leucophyllum frutescens. J Nat Prod.
2011;74:1842‐1850.
24. Villanueva‐Cañongo C, Pérez‐Hernández N, Hernández‐Carlos
B, Cedillo‐Portugal E, Joseph‐Nathan P, Burgueño‐Tapia E.
Complete 1H NMR assignments of pyrrolizidine alkaloids and
a new eudesmanoid from Senecio polypodioides. Magn Reson
Chem. 2014;52:251‐257.
25. Muñoz MA, Pérez‐Hernández N, Pertino MW, Schmeda‐
Hirschmann G, Joseph‐Nathan P. Absolute configuration and
1
H NMR characterization of rosmaridiphenol diacetate. J Nat
Prod. 2012;75:779‐783.
26. Anisimov AV, Phan LC, Tarakanova AV, Lebedev MY,
Berentsveig VV. Epoxidation of (Z)‐β‐ocimene under phase‐
transfer‐catalytic conditions. J Org Chem USSR (Engl Transl).
1992;28:1105‐1108.
27. Agrebi A, Agnaniet H, Bikanga R, Makani T, Anguilé JJ, Lebibi
J, Casablanca H, Morére A, Menut C. Essential oil of
Plectranthus tenuicaulis for flavor and fragance: Synthesis of
derivatives from natural and synthetic 6,7‐epoxyocimenes.
Flavour Fragr J. 2012;27:188‐195.
How to cite this article: Julio LF, Burgueño‐
Tapia E, Díaz CE, Pérez‐Hernández N, González‐
Coloma A, Joseph‐Nathan P. Absolute
configuration of the ocimene monoterpenoids from
Artemisia absinthium. Chirality. 2017;1‐10. https://
doi.org/10.1002/chir.22741