Organic Letters
Letter
10′ to H-14′ rather than H-14′ to H-5, which is only possible
with the revised structure 7. Further key HMBC and COSY
correlations (Figure 2) confirm the 6−6−9−4 ring system of 7.
The NOESY spectrum of 7 in CDCl3 was again difficult to
interpret because of the overlapped region around the key H-5
resonance. However, the NOESY spectrum of 7 in C6D6
showed correlations between H-9′ (δH = 2.45) and Me-14
Caryophyllene can adopt four conformations in solution,
termed αα, αβ, βα, and ββ according to the relative disposition
of the exocyclic alkene and the vinyl methyl substituent. NMR
studies have shown that caryophyllene exists as a 48:28:24 αα/
βα/ββ mixture.12 While conformers αα and βα are known to
be in rapid exchange, their exchange with the ββ conformer is
relatively slow, with an experimentally determined barrier of 68
kJ mol−1. In line with previous studies on the biomimetic
synthesis of guajadial and psidial A,8 we propose that littordials
B and C (2 and 3) and littordials E and F (7 and 8) arise from
cycloaddition between o-QMs 15/16 and the βα conformation
of 9 while littordial A (1) is formed from the ββ conformation
(Figure 4).
(δH = 0.76) and between H-5 (δH = 2.13) and H-1 (δH
=
2.04). Littordial E (7) therefore has the same 6−6−9−4 ring
system and relative configuration as guajadial (12).
Our analysis of the 2D NMR spectra for synthetic 3 agreed
with the original structure elucidation for littordial C,4 and this
was confirmed by single-crystal X-ray crystallography (Figure
Figure 3. X-ray crystal structure of littordial C (for clarity, only H
atoms at stereocenters are shown).
The use of butanal as the aldehyde component in the
cascade reaction with 9 and 10 allowed the synthesis of the
revised structure of littordial F (8) in 3% yield alongside
littordials A (1) and B (2) as an inseparable mixture (5:1 ratio
in favor of 2) in a combined yield of 42% (Scheme 1). Since
the structure of 1 was previously established by single-crystal
X-ray crystallography4 and the structure of 2 is also not in
doubt because of its close similarity to 3, we focused our
1
attention solely on the characterization of 8. The H and 13C
NMR spectra of 8 are almost identical to the corresponding
spectra of 7, except for two additional methylene groups in the
n-pentyl side chain appended to C-9′ of 7 compared with the
n-propyl substituent of 8. Thus, we propose that the revised
littordial F structure 8 has the same 6−6−9−4 ring system and
the same relative configuration as 7 and 12. This assignment is
also supported by 2D NMR studies (see the Supporting
Information for full details). The formation of the originally
assigned littordial F structure 6 under our reaction conditions
is mechanistically implausible. Although the proposed littordial
F structure 6 was supported by predicted 13C NMR data
calculated using DFT methods, the conformational flexibility
of this compound makes such a calculation potentially
misleading.5
The absolute configuration of isolated littordial A (1) was
initially assigned by single-crystal X-ray diffraction analysis with
Ga Kα radiation, which was supported by a comparison
between experimental and calculated CD spectra for all of the
littordials.4,5 This configuration arises from the more common
(−) enantiomer of caryophyllene. As expected, the CD spectra
of synthetic littordials C, E, and F (also derived from
(−)-caryophyllene) matched those of the isolated natural
products. However, our measured optical rotations differed
significantly from those of the reported natural products, which
could be due to the presence of minor diastereomeric
impurities in either the synthetic or natural samples.11
Figure 4. Plausible transition states leading to littordials A, B, and F.
Computational modeling was used to assess the possible
cycloaddition pathways leading to littordials A, B, an F and an
additional diastereomer analogous to 14. Calculations were
performed using DFT with the M06-2X functional and the 6-
31+G(d) basis set for geometry optimizations and frequency
analysis, with single-point energies calculated at the 6-
311+(d,p) level.
Given that caryophyllene has four possible conformers and
that o-QM 16 may be formed as either the E or Z isomer and
taking into account the endo/exo stereoselectivity, there are 16
possible reaction pathways, of which those involving the
nonpopulated αβ conformer of caryophyllene are discounted.
o-QMs E-16 and Z-16 are calculated to have essentially the
same stability (ΔG = −1.2 kJ mol−1), and any E/Z selectivity
in the formation 16 is unknown, so both isomers must be
considered as viable reactants. The lowest-energy TSs leading
to littordials A, B, and F are illustrated in Figure 4. This
analysis reveals that in general a concerted Diels−Alder
cycloaddition between 9 and 16 is energetically feasible
under the reaction conditions. The Diels−Alder transition
C
Org. Lett. XXXX, XXX, XXX−XXX