Challenges in the assignment of relative and absolute configurations of complex molecules: Computation can resolve conflicts between theory and experiment

Article abstract of DOI:10.1016/j.tet.2013.10.004

The configuration of (-)-brevianamides was assigned as (2S,13S) based on X-ray structure analysis and hydrolysis experiments. However, our theoretical investigation of its chiroptical properties strongly implied that the correct configuration should be (2R,13R). The reasons for the incorrect earlier assignment are analyzed by calculations of conversion energy barriers among different intermediates, starting materials and final products. This study demonstrates that conflicting theoretical and, experimental results suggest that it is premature to assign the configuration of a natural product.

Full text of DOI:10.1016/j.tet.2013.10.004

Tetrahedron 69 (2013) 10351e10356
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Challenges in the assignment of relative and absolute configurations
of complex molecules: computation can resolve conflicts between
theory and experiment
a
b
a
b
,
c
a,
*
*
Jie Ren , Guo-You Li , Lan Shen , Guo-Lin Zhang , Laurance A. Nafie , Hua-Jie Zhu
a
Chinese Center for Chirality, Key Laboratory of Medicinal Chemistry and Molecular Diagnostics Education Committee of China, Hebei University,
Baoding, 071002 Hebei Province, China
Chengdu Institute of Biology, Chengdu, Sichuan, China
b
c
Department of Chemistry, Syracuse University, Syracuse, NY 13244, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The configuration of (ꢀ)-brevianamides was assigned as (2S,13S) based on X-ray structure analysis and
hydrolysis experiments. However, our theoretical investigation of its chiroptical properties strongly
implied that the correct configuration should be (2R,13R). The reasons for the incorrect earlier assign-
ment are analyzed by calculations of conversion energy barriers among different intermediates, starting
materials and final products. This study demonstrates that conflicting theoretical and, experimental
results suggest that it is premature to assign the configuration of a natural product.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 20 September 2013
Received in revised form 2 October 2013
Accepted 3 October 2013
Available online 17 October 2013
Keywords:
Absolute configuration reassignment
DFT
Chiroptical spectroscopy
Transition state
X-ray
2
1. Introduction
molecules and their structures. The AC of 1 was selected as one of
the examples in this study. After comparing measured to calculated
Absolute configuration (AC) determination for complex chiral
compounds has challenged chemists for a long time. With the de-
velopment of super-computer technology, to use theoretical
methods to assign AC becomes more and more popular in the
world. Especially, by means of modern quantum theory, re-
searchers can now identify many complex chiral compounds’ AC.
This greatly accelerates the progress of different theoretical
methods used in AC assignment in stereochemistry.
chiroptical properties of 1, we concluded that the reported AC was
incorrect and should be re-assigned as (2R,13R)-1. This example of
mis-assignment of AC is the most extreme case uncovered to date
in our theoretical studies. It demonstrates that, given the power of
current computational chemistry using DFT, chemists should now
Recently, a novel bioactive alkaloid, brevianamide M (1), and its
1
oxidation product 2 have been reported (Fig. 1). The relative con-
figuration of 1 was established as (2R,13R) or (2S,13S) by X-ray
crystallography using Mo radiation. This compound can be hydro-
lyzed in acidic aqueous solution at 100 C to afford L-phenylalanine,
O
N
ꢁ
H
which strongly implied that C-13 should have an S-configuration.
O
1
Therefore, the AC of 1 was assigned as (2S,13S). This procedure for
N
13
2
AC assignment is normally definitive and cogent. In our recent
studies of structural properties of chiral molecules, we use density
functional theory (DFT) with different basis sets to examine the
relationship between optical rotation values of the target chiral
NH
H
HO
Brevianamide M, 1
*
Corresponding authors. E-mail address: hjzhu@mail.kib.ac.cn (H.-J. Zhu).
Fig. 1. The structures of 1 and 2.
0
040-4020/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2013.10.004

10352
J. Ren et al. / Tetrahedron 69 (2013) 10351e10356
reexamine AC assignments based in part on indirect chemical evi-
dence, such as the acid hydrolysis mentioned above. The details of
how the new assignment of AC for 1 was systematically reached are
explained below, including how the acid hydrolysis provided mis-
leading evidence.
of (2R,13R)-1 by 1.66 kcal/mol at the B3LYP/6-311þþG(2d,p) level
in TEE. This energy decreased to 0.80 kcal/mol in Gibbs free energy
(GFE), or 1.46 kcal/mol with zero-point energy (ZPE) correction at
the B3LYP/6-311þþG(2d,p) level in the gas phase. Thus, if the
transition state barriers are not high enough to maintain the con-
figuration at C-13, (2R,13R)-1 can convert into the more stable
epimer (2R,13S)-1 with almost 92% conversion yield from (2R,13R)-
1 to (2R,13S)-1 using ZPE energy, or 80% conversion using GFE. Fi-
ꢁ
The experimental optical rotation (OR) of 1 is ꢀ147.7 in ace-
1
tone. However, it was found that (2S,13S)-1 had positive OR values
(
Table 1) both in the gas phase and in chloroform using widely used
3
DFT methods. Four DFT quantum chemistry models were used in
the OR computations. Optimization of different conformations was
nally, (2R,13S)-1 could hydrolyze to afford 80e92% of
L-phenylala-
ꢁ
nine under strong HCl conditions at 100 C.
Table 1
OR values for (2S,13S)-1 from methods 1e4
Method 1a
Method 2b
Method 3c
Method 4d
[
[
a
a
]
]
D
þ510.5
þ209.8
þ532.9
þ222.4
þ519.5
þ217.0
þ600.7
þ354.0
e
Dsp
a
b
c
B3LYP/6-311þþG(2d,p)//B3LYP/6-31G(d).
B3LYP/6-311þþG(2d,p)//B3LYP/6-31þG(d,p).
B3LYP/6-311þþG(2d,p)//B3LYP/6-311þþG(2d,p).
d
e
Δ
Δ
B3LYP/6-311þþG(2d,p)//PCM/B3LYP/6-311þþG(2d,p).
Single point energy at the B3LYP/aug-cc-pVDZ level in the chloroform using
PCM model was used in OR computations.
performed at the B3LYP/6-31G(d) level in the gas phase; the OR
computations were then performed at the B3LYP/6-311þþG(2d,p)
level in the gas phase (methods 1, B3LYP/6-311þþG(2d,p)//B3LYP/
6
-31G(d)). Three additional models were used for OR calculations
Δ
as indicated in Table 1. They were B3LYP/6-311þþG(2d,p)//B3LYP/
-31þG(d,p), method 2, B3LYP/6-311þþG(2d,p)//B3LYP/6-
11þþG(2d,p), method 3, and B3LYP/6-311þþG(2d,p)//PCM/
B3LYP/6-311þþG(2d,p), method 4. The total electronic energy
TEE) was used in OR computations using Boltzmann statistics. All
conformations were used in a single point energy (SPE) calculation
at the B3LYP/aug-cc-pVDZ level in chloroform using the PCM
model. The SPE were then used in OR computations again to afford
Scheme 1. The plausible isomerization from (2R,13R)-1 to (2R,13S)-1.
6
3
We determined experimentally that
converted into
L
-phenylalanine cannot be
ꢁ
D
-phenylalanine in 6 N HCl at 100 C over 12 h since
(
all recovered product showed no loss of OR. Therefore, this ex-
cluded both the possibility of conversion of -phenylalanine to
phenylalanine or vice versa. Accordingly, the obtained -phenylal-
D
L-
L
anine must originate from the hydrolysis procedure instead of an
inter-conversion under HCl catalysis.
[a]Dspe shown in Table 1. The methods used were validated in our
4
recent study. All OR values are summarized in Table 1.
The electronic circular dichroism (ECD) of (2R,13R)-1 was com-
puted at the B3LYP/6-311þþG(2d,p)//B3LYP/6-311þþG(2d,p) level
All OR predictions for (2S,13S)-1 are positive ranging from about
þ210 to þ600. The absolute values of OR are much bigger than the
experimental OR (ꢀ147.7 in acetone), which is typical for DFT-level
calculations. Since the relative configuration was well established
using X-ray, the positive OR values of from þ210 to þ600 suggest
that 1 should have the absolute configuration of (2R,13R). However,
there would be concern that the computation of OR values is based
on the determination of electronic transitions in the inaccessible
far-UV region, which is not easy to calculate with reliability. By
contrast, the relative configuration of the structure of (2R,13R) or
5
to further explore the AC of 1. All conformations were used for
frequency computations to obtain GFE data. Then, GFE magnitudes
were used in the Boltzmann statistics in ECD simulations. As ex-
pected, the computed ECD for (2R,13R)-1 was in good agreement
with the experimental ECD (Fig. 2). The theoretical prediction of
(
2S,13S)-1 was well established by X-ray and its absolute configu-
ration was confirmed by hydrolysis of 1 to afford the -amino acid.
L
Thus, we questioned whether the DFT methods may have given
a wrong prediction in this example.
However, after we carefully examined the structure 1, we found
that the previously assigned absolute configuration was indeed
incorrect. A possible reason for this discrepancy is that the H on C-
1
3 could be enolized under strong acid conditions (6 N HCl aqueous
ꢁ
solution, 100 C for 12 h), and thus epimerized (Scheme 1). Theo-
retically, if the diastereomer (2R,13S)-1 had lower energy than
(
2R,13R)-1, this isomerization conversion could explain the hydro-
lysis result. Thus, relative energetics of (2R,13S)-1 and (2R,13R)-1
were examined. After conformational searches using the MMFF54S
force field, all accessible conformations were used in optimizations
at the B3LYP/6-31G(d) level in the gas phase. The low energy
conformations from 0 to 2.5 kcal/mol were then used for further
optimizations at the B3LYP/6-311þþG(2d,p) level in the gas phase.
It was found that the energy of (2R,13S)-1 is really lower than that
Fig. 2. Comparison of the computed ECD with the experimental CD.

J. Ren et al. / Tetrahedron 69 (2013) 10351e10356
10353
þ
positive and negative Cotton effect matched very well with those in
the experimental ECD. Therefore, both of OR and ECD provide ev-
idence that the previously assigned AC (ꢀ)-(2S,13S)-1 is not correct.
The actual absolute configuration of (ꢀ)-1 must be (2R,13R).
Why did the early experiments provide the wrong predictions?
The major reason must be low barriers of the isomerization from
In the absence of H catalysis (TS-1), a 33.4 kcal/mol barrier was
found at the B3LYP/6-311þþG(2d,p) level in the gas phase using
TEE. The barrier is 32.9 kcal/mol using GFE, or 31.5 kcal/mol using
zero-point energy corrections in the gas phase. Here, all the barriers
are quite high, and it is difficult to get conversion via TS-1. The
energy barrier decreased to 25.1 kcal/mol at the B3LYP/6-
311þþG(2d,p) level in water with the PCM model using SPE data.
(
2R,13R)-1 to (2R,13S)-1. Therefore, conversion barriers were in-
þ
vestigated. All possible transition state (TS) structures were ana-
However, in the presence of another H , e.g., TS-2 and TS-4, the
conversion benefits from a lower TS barrier of 23.1 kcal/mol (TS-4)
6
lyzed using DFT methods similar to those recently used (Fig. 3).
Theoretically, four TS structures were plausible for the formation
of (2R,13S)-1. However, only TS-1, TS-2, and TS-4 were found in the
calculations, TS-3 does not exist in the computations.
in the gas phase in TEE, or 17.9 kcal/mol in SPE in water (Table 2). If
Table 2
The TS activation energy magnitudes for the four different procedures computed at
the B3LYP/6-311þþG(2d,p) level in the gas phase or in water
_
_
Cl
Cl
H+
O
H+
O
a
Gb
Ec
d
D
E
0
D
D
DE
l
H
+H O
H
O
TS-1
TS-2
TS-4
TS-5
31.5
26.8
21.5
25.4
32.9
28.6
23.3
26.7
33.4
28.2
23.1
27.7
25.1
16.5
17.9
21.3
HN
N
HN
H
HO
N
H
N
N
a
HO
Using zero-point energy correction.
Using Gibbs free energy.
Using total electronic energy.
b
c
TS-1
TS-2
d
Using single point energy obtained in water via PCM model.
_
_
Cl
Cl
H+
H+
the reaction involvesTS-2, the barriers in the gas phase are higher
by about 5.1e5.3 kcal/mol than those in TS-4. However, TS-2 de-
creased to 16.5 kcal/mol in water and is about 1.4 kcal/mol lower
than TS-4. Both barriers are low enough for the isomerization. The
H+
O
H
O
H
O
O
H+
HN
N
HN
N
3
D TS structure (TS-4) is illustrated in Fig. 4a. Once the isomeri-
H
H
N
HO
zation completed, the intermediate could convert into (2R,13S)-1
via TS-5 as illustrated in Fig. 4b and c. The procedure involved HCl
again. This energy barrier is only 25.4 kcal/mol in ZPE, 26.7 kcal/mol
in GFE, or 27.7 kcal/mol in TEE. The barrier decreased to 21.3 kcal/
mol in water using the PCM model. The reaction coordinate for the
N
HO
TS-3
TS-4
Fig. 3. The four TS structures used in TS computations.
TS-5
TS-4
HCl+H+
+
(2R,13R)-1
Intermediate
G=0.80
2R,13S)-1
G=
3.3
G=
Complex of
2R,13R)-1
HCl+H+
2
26.7
(
(
+
(d)
Fig. 4. The 3D structures of TS-4, TS-5, and the coordinates for the conversion from (2R,13R)-1 to (2R,13S)-1.

10354
J. Ren et al. / Tetrahedron 69 (2013) 10351e10356
whole procedure is illustrated in Fig. 4c. (2R,13S)-1 finally decom-
posed into -phenylalanine as reported previously.
Indeed, the possibility for 1 to lose a molecule of water to 3
existed. The size of the barrier for dehydration to form 3 via TS-7 is
therefore of immediate interest. If the barrier is lower than that
leading through TS-4, the loss of a water, while reversible, could
determine the reaction pathway. However, if the barrier via TS-7 is
higher than that via TS-4, the loss of a molecule of water to 3would
L
Another feasible route for the AC conversion involves an S
N
2
þ
procedure (Scheme 2). At the first step, H catalysis leads the loss of
a water molecule to form an imine (3), the imine then hydrolyzed
into an amide (4) after cleavage of the C2eN. Under HCl catalysis, 4
Scheme 2. The possible conversion from 1 to L-phenylalanine 7.
can go through TS-6 to give intermediate 6. In this procedure, the
AC of C-13 may change from R to S. After reaction with a molecule of
be inconsequential. The process should go through TS-4 and TS-5.
Therefore, the pathway via TS-7 was investigated at the B3LYP/6-
311þþG(2d,p) level (Fig. 6). The position of a phenyl ring in TSs
was investigated to study the effect of phenyl ring on the barriers.
water, phenylalanine 7 formed. Therefore, to obtain
L-phenylala-
nine, the TS-6 conversion must change the AC of C-13.
The barriers via TS-6 were investigated at the B3LYP/6-
31G(d) level in the gas phase; the effect of the phenyl ring on
the barriers were also carefully considered. Finally, the lowest
barrier TS structures were further computed at the B3LYP/6-
3
11þþG(2d,p) level. However, it was found that the energy
barrier via TS-6 is 64.9 kcal/mol in free energy when the energy
of a complex of HCl and 4 was used as the relative zero-point
reference. On the other hand, the relative energy of the com-
plex of HCl and 4 is higher by 8.0 kcal/mol in GFE than that of
the complex of HCl and 1. Thus, the TS barrier was about
7
2.8 kcal/mol when the energy of the complex of HCl and 1 was
used as the relative zero-point reference. This may be due to the
ring strain in the aziridinium TS structure. This conversion is
therefore quite difficult. The 3D TS structure of TS-6 is illus-
trated below (Fig. 5).
Fig. 6. The plausible transition state for TS-7.
The predicted barrier (33.78 kcal/mol) via TS-7 was higher than
that (23.28 kcal/mol) via TS-4 by a value of 10.5 kcal/mol at the
B3LYP/6-311þþG(2d,p) level in the gas phase. This barrier is also
higher by 7.0 kcal/mol than that via TS-5. Since the barriers in TS-7
were bigger than those via TS-4 and TS-5, respectively, this sug-
gests the favorable route to the final products should proceed via
TS-4 and TS-5. The specific TS barriers involved in TS-6 and TS-7 are
summarized in Table 3.
After careful examination of the possible procedures, the whole
sequence from 1 to 7 is illustrated below (Scheme 3).
In route A, after (2R,13S)-1 formed, it could be hydrolyzed into 8a
þ
ꢀ
or 8b. Under the H O eCl system, either 8a or 8b should be further
3
hydrolyzed into 9 and 10. Compound 9 would not change anymore.
However, intermediate 10, which is stable due to the trans-
orientation, should be hydrolyzed into 7 and 11 including a mole-
cule of NH
formed, it could be hydrolyzed into 4. In the H
termediate 12 formed and itshould befurtherhydrolyzedinto9 and 7.
4
Cl. Another possible way is route B, after (2R,13S)-1
þ
ꢀ
3
O eCl system, in-
Fig. 5. The plausible transition state for TS-6.

J. Ren et al. / Tetrahedron 69 (2013) 10351e10356
10355
Table 3
The corresponding barriers involved in TS-6 and TS-7 at the B3LYP/6-311þþG(2d,p)
level in the gas phase and in water using PCM model
a
Gb
Ec
d
D
E
0
D
D
D
E
l
O
N
H
O
N
H
TS-6
TS-7
65.1
32.0
10.6
6.6
64.9
33.8
10.5
7.1
67.1
32.9
9.9
51.7
29.1
11.2
7.8
O
[O]
N
13
2
O
e
f
N
D
D
(
(
D
D
E)
E)
NH
5.3
NH
a
b
c
Using zero-point energy correction.
Using Gibbs free energy.
Using total electronic energy.
Using single point energy obtained in water via PCM model.
The barrier energy via TS-7 was subtracted from that via TS-4. The positive
OH
O
(
2R,13R)-1
(R)-2
d
e
Scheme 4. The oxidation from 1 to 2.
values exhibit that the barrier via TS-7 is higher than that via TS-4.
The energy via TS-7 was subtracted from that via TS-5. The positive values ex-
hibits that the barrier via TS-7 is higher than that via TS-5.
f
As mentioned above, (2R,13R)-1 can be oxidized into (R)-2. Thus,
conversion of (2R,13R)-1 was performed (Scheme 4), the obtained
(
R)-2 had OR values of ꢀ340. It had almost the same OR values as
the one (ꢀ359) isolated from the broth. It exhibited the early ob-
Route A:
Route B:
Scheme 3. Two plausible routes for the conversions of 1e7.
The experimental conversion of (2R,13R)-1 to (2R,13S)-1 was
tained (ꢀ)-2 has (R) configuration derived from (2R,13R)-1 via ox-
idation. Furthermore, the (R)-(ꢀ)-2 was used for ECD
measurement. Three stable conformations were found and they
were used in ECD computations at the B3LYP/6-311þþG(2d,p)//
B3LYP/6-311þþG(2d,p) level in the gas phase using GFE after
Boltzmann statistics. The recorded ECD agreed with the predicted
ECD well (Fig. 7).
studied next. The starting material did not dissolve in water at room
ꢁ
temperature. Once the temperature was raised to about 65 C, it
dissolved in water quickly and decomposed immediately, and
phenylalanine formed with OR value of ꢀ32. This OR value is al-
most the same as that of pure natural -phenylalanine. This
L-
L
exhibited that the conversion from (2R,13R)-1 to (2R,13S)-1 may be
complete, or at least the conversion may be over 92% as predicted
using ZPE energy.
Furthermore, the three conformations of (R)-2 were used in VCD
computations
at
the
B3LYP/6-311þþG(2d,p)//B3LYP/6-
Compound 2 was isolated from the same pathway as 1. Logically,
311þþG(2d,p) in the gas phase in the gas phase based on
2
is likely to have the same AC at C-13 as 1 based on a common
precursor. For example, (2R,13R)-1 can be oxidized into (R)-2. Thus,
R)-2 was used in the OR computations. Theoretically, the com-
puted OR for (R)-2 was ꢀ619.2 at the B3LYP/6-311þþG(2d,p)//
B3LYP/6-311þþG(2d,p) level in the gas phase. The experimental
OR was ꢀ359 for (R)-2. Thus, the early reported (ꢀ)-2 should be (R)
configuration (Scheme 4).
methods described in a review of the application of VCD to the
7
determination of AC in a wide variety natural product molecules,
(
and a more recent review of the use of VCD and Raman optical
8
activity (ROA) for the determination of AC in chiral molecules.
Solvent-subtracted IR and VCD spectra were used. As shown in
Fig. 8, the predicted IR and VCD spectra agree closely with the ex-
perimental IR and VCD spectra of (ꢀ)-2.

10356
J. Ren et al. / Tetrahedron 69 (2013) 10351e10356
Fig. 9. The X-ray structure for (ꢀ)-1.
significance for both experimental and theoretical chemists. All
sources of evidence need to be considered carefully, and compu-
tational methods should be used, where possible, to confirm con-
clusions based only on chemical reasoning.
Fig. 7. The predicted ECD for (R)-2 and experimental ECD for (ꢀ)-2.
Acknowledgements
H.J.Z. thanks the funds support from Hebei University. P. Wipf in
University of Pittsburg is thanked for his valid suggestions in
mechanism study and editing. The measurement of VCD for com-
pound (ꢀ)-2 was performed in BioTools Company in US. Dr. Dukor,
and Dr. Wang at BioTools are thanked for their assistances in VCD
study. G.-Y. Li has the equal contribution to the first author.
References and notes
1. Li, G.-Y.; Yang, T.; Luo, Y.-G.; Chen, X.-Z.; Fang, D.-M.; Zhang, G.-L. Org. Lett. 2009,
11, 3714e3717.
2. (a) Shen, L.; Zhao, S.-D.; Zhu, H.-J. Chem. J. Chin. Univ. 2011, 32, 2568e2573; (b) Li,
Q.-M.; Ren, J.; Shen, L.; Bai, B.; Liu, X.-C.; Wen, M.-L.; Zhu, H.-J. Tetrahedron 2013,
6
9, 3067e3074.
. (a) Zhao, S. D.; Shen, L.; Luo, D. Q.; Zhu, H. J. Curr. Org. Chem. 2011, 15, 1843e1862;
b) Zhu, H. J. Modern Organic Stereochemistry; Science: Beijing, China, 2009; (c)
3
(
Polavarapu, P. L. Mol. Phys. 1997, 91, 551e554; (d) de Meijere, A.; Khlebnikov, A.
F.; Kozhushkov, S. I.; Kostikov, R. R.; Schreiner, P. R.; Wittkopp, A.; Rinderspacher,
C.; Menzel, H.; Yufit, D. S.; Howard, J. A. K. Chem.dEur. J. 2002, 8, 828e842; (e)
Taniguchi, T.; Martin, C.; Monde, K.; Nakanishi, K.; Berova, N.; Overman, L. J. Nat.
Prod. 2009, 72, 430e432; (f) McCann, D. M.; Stephens, P. J. J. Org. Chem. 2006, 71,
6074e6098; (g) Mennucci, B.; Claps, M.; Evidente, A.; Rosini, C. J. Org. Chem.
2007, 72, 6680e6691; (h) Cheng, M.; Li, Q.; Lin, B.; Sha, Y.; Ren, J.; He, Y.; Wang,
Q.; Hua, H.; Ruud, K. Tetrahedron: Asymmetry 2006, 17, 179e183.
4
5
. Liao, T.-G.; Ren, J.; Fan, H.-F.; Xie, M.-J.; Zhu, H.-J. Tetrahedron: Asymmetry 2008,
1
9, 808e815.
. (a) Stephens, P. J.; Pan, J. J.; Devlin, F. J.; Krohn, K.; Kurtan, T. J. Org. Chem. 2007,
2, 3521e3536; (b) Li, X.-N.; Zhang, Y.; Cai, X.-H.; Feng, T.; Liu, Y.-P.; Li, Y.; Ren, J.;
Fig. 8. Predicted VCD and IR at the B3LYP/6-311þþG(2d,p)//B3LYP/6-311þþG(2d,p) in
the gas phase for (R)-2; comparison of experimental VCD for (ꢀ)-2 and the calculated
7
VCD for (R)-2. IR and VCD spectra of (ꢀ)-2 in CDCl
3
(6 mg/0.3 mL); 100-mm path-length
Zhu, H.-J.; Luo, X.-D. Org. Lett. 2011, 13, 5896e5899; (c) Lu, Z. Y.; Zhu, H. J.; Fu, P.;
Wang, Y.; Zhang, Z. H.; Lin, H. P.; Liu, P. P.; Zhuang, Y. B.; Hong, K.; Zhu, W. M. J.
Nat. Prod. 2010, 73, 911e914; (d) Zhang, Y. L.; Ge, H. M.; Zhao, W.; Dong, H.; Xu,
Q.; Li, S. H.; Li, J.; Zhang, J.; Song, Y. C.; Tan, R. X. Angew. Chem., Int. Ed. 2008, 47,
2
cell with BaF windows; 19 h collection for sample and solvent; instrument optimized
ꢀ
1
at 1400 cm . Solvent-subtracted spectra are compared to the corresponding calcu-
lated spectra.
5823e5826; (e) Giorgio, E.; Parrinello, N.; Caccamese, S.; Rosini, C. Org. Biomol.
Chem. 2004, 2, 3602e3607; (f) Specht, K. M.; Nam, J.; Ho, D. M.; Berova, N.;
Kondru, R. K.; Beratan, D. N.; Wipf, P.; Pittman, R. A., Jr.; Kahne, D. J. Am. Chem.
Soc. 2001, 123, 8961e8966.
6. (a) Xiao, W. L.; Lei, C.; Ren, J.; Liao, T.-G.; Pu, J.-X.; Pittman, C. U.; Lu, Y.; Zheng, Y.-
T.; Zhu, H.-J.; Sun, H.-D. Chem.dEur. J. 2008, 14, 11584e11592; (b) Li, L.-C.; Ren, J.;
Jiang, J.-X.; Zhu, H.-J. Eur. J. Org. Chem. 2007, 1026e1030.
. (a) Nafie, L. A. Nat. Prod. Commun. 2008, 3, 451e466; (b) Nakahashi, A.; Yaguchi,
Y.; Miura, N.; Emura, M.; Monde, K. J. Nat. Prod. 2011, 74, 707e711; (c) Wang, F.;
Polavarapu, P. L.; Lebon, F.; Longhi, G.; Abbate, S.; Catellani, M. J. Phys. Chem. A
Efforts to prepare single crystals of 1 of sufficient quality for AC
determination were successful, and X-ray diffraction data was ob-
tained using a Cu-radiation source. The X-ray experiments con-
firmed that our theoretical predictions are correct (Fig. 9). The
enantiomer (ꢀ)-1 should be (2R,13R). Now, all evidences, both ex-
perimental and computational, agree well for the AC assignment of
7
2002, 106, 12365e12369; (d) Socolsky, C.; Rates, S. M. K.; Stein, A. C.; Asakawa, Y.;
(
ꢀ)-(2R, 13R)-1.
Bardon, A. J. Nat. Prod. 2012, 75, 1007e1017; (e) Taniguchi, T.; Monde, K. J. Am.
Chem. Soc. 2012, 134, 3695e3698; (f) Nicu, V. P.; Baerends, E. J.; Polavarapu, P. L. J.
Phys. Chem. A 2012, 116, 8366e8373.
Experimental results are always the fundamental touchstone for
different theories. However, if the interpretation of the experi-
mental results is not correct, the conclusions reached will be af-
fected, just as in our initial thinking in this study. With this idea in
mind, the example analyzed in this paper, carries a high level of
8
. (a) He, Y.; Wang, B.; Dukor, R. K.; Nafie, L. A. Appl. Spectrosc. 2011, 70, 699e723;
(b) Zuber, G.; Hug, W. J. Phys. Chem. A 2004, 108, 2108e2118; (c) Zou, Y.; Ga-
rayalde, D.; Wang, Q.; Nevado, C.; Goeke, A. Angew. Chem., Int. Ed. 2008, 47,
10110e10113.