Capricious selectivity in electrophilic deuteration of methylenedioxy substituted aromatic compounds

Article abstract of DOI:10.1021/jo5019427

Ring deuteration via the S_EAr mechanism, which is usually problem-free, is found to be troublesome with methylenedioxy substituent aromatics. We report a case where the deuteration not only partially fails at one of the ortho positions but also

Full text of DOI:10.1021/jo5019427

Note
pubs.acs.org/joc
Capricious Selectivity in Electrophilic Deuteration of Methylenedioxy
Substituted Aromatic Compounds
Monika Pohjoispaa,† Raul Mera-Adasme,‡ Dage Sundholm,*,† Sami Heikkinen,† Tapio Hase,†
́
̈
̈
and Kristiina Wahala*,†
̈
̈
̈
†Department of Chemistry, University of Helsinki, P.O. Box 55, FIN-00014 Helsinki, Finland
‡Department of Chemistry, Ludwig Maximilian University, Munich, Germany
S
* Supporting Information
ABSTRACT: Ring deuteration via the SEAr mechanism,
which is usually problem-free, is found to be troublesome
with methylenedioxy substituent aromatics. We report a case
where the deuteration not only partially fails at one of the
ortho positions but also is completely prevented by a
conformation dependent effect at the other o-position. Such
selectivity discrepancies are important due to the widespread
occurrence of methylenedioxy substituted natural products. Density functional theory calculations were used to elucidate the
exchange reaction mechanism in 1,2-dialkoxybenzenes.
he methylenedioxy group is a common aromatic ring
Tsubstituent of numerous biologically active compounds
that is present for example in the plant lignans podophyllotoxin
and cubebin, which are hot lead compounds for the discovery
and development of novel anticancer and antimicrobial agents.1
In studying the biological activity, use is often made of stable
isotope labeled analogs (very often containing several
deuteriums). The first efforts to deuterate an aromatic
compound took place in 1934, when benzene was heated in
3% heavy water in the presence of a nickel catalyst.2 In the
same year, Ingold et al. introduced deuterium atoms into the
aromatic nucleus by means of ordinary electrophilic reagents,
such as deuterated aqueous sulfuric acid without using any
heterogeneous catalysts.3 They also proposed that the
mechanism of the exchange reaction is an ordinary electrophilic
aromatic substitution and proved it by comparing the
efficiencies of certain acidic and basic deuterating agents and
by studying how certain aromatic substituents influence the
reaction.4
The electrophilic aromatic H/D exchange reactions catalyzed
by acids and occasionally by bases are utilized in the deuterium
labeling of various polyphenolic compounds.5,6 In our studies,
we found that 35% DCl/D2O and the use of the ionic liquid 1-
butyl-3-methylimidazolium chloride [bmim]Cl as a cosolvent
under microwave irradiation constitute an expedient deutera-
tion method for naturally occurring polyphenols, such as
lignans.7 For lignans, the aromatic substituents are usually
ortho−para directing and activating hydroxy and/or methoxy
groups.8 Employing these methods, stable isotopically pure
polydeuterated lignans 1−6 were synthesized via proton
exchange at the activated positions (Figure 1).
Figure 1. Deuterium labeled lignans: lignanolactones 1−3 and
tetrahydrofuran lignans 4−6.
lignane, 7, Scheme 1) and brassilignan (3,3′,4,4′-tetra-
methoxy-9,9′-epoxylignane, 8), which differ only slightly in
the aromatic substituents. In the deuteration reaction, all the
aromatic protons of brassilignan were exchanged in over 95%
isotopic purity as expected (8 → 5, Scheme 1), whereas with
dehydroxycubebin (7) the degree of deuteration (7 → 9,
Scheme 1) remained incomplete despite recycled labeling
attempts. According to electron ionization mass spectrometry
(EI-MS) measurements, a mixture of di-, tri-, and tetradeu-
terated products was formed from 7 at best. Quantitative 13C
NMR studies demonstrated that the 6- and 6′-sites of 7 were
fully deuterated, while only one-third of the 2- and 2′-sites (R2
= H/D = 2:1; see 9 in Scheme 1) carried deuteriums and
remarkably the positions 5 and 5′ remained intact.
The reaction was repeated several times with fresh
deuteration reagents by recycling the isolated crude product
9, but the level of deuteration did not improve. Our earlier
However, diverging regioselectivities in the H/D exchange
reaction were observed for two closely related analogs, namely
dehydroxycubebin (3,3′,4,4′-bismethylenedioxy-9,9′-epoxy-
Received: August 21, 2014
Published: September 30, 2014
© 2014 American Chemical Society
10636
dx.doi.org/10.1021/jo5019427 | J. Org. Chem. 2014, 79, 10636−10640

The Journal of Organic Chemistry
Note
The 9,9′-epoxylignanes 7 and 8 can adopt three main
conformations, the extended 12 and two π-stacked conforma-
tions 13 and 14 (depicted for 7 in Figure 2). The calculations
Scheme 1. Deuteration of the Plant Lignans
Dehydroxycubebin 7 and Brassilignan 8
a
a
The treatment of 8 gives a hexadeuterated analog 5 in over 95%
isotopic purity.
experiments have shown that certain aromatic sites, e.g.
structures having two meta substituted hydroxy or methoxy
groups, are so highly activated that the deuterium atoms are to
some extent exchanged back to hydrogens during workup.6 For
such compounds a selective dedeuteration of the labile
deuterium labels is performed to obtain isotopically stable
products. However, the deuterium labels at C-2 and C-2′ in 9
are not labile, as they could not be replaced using the
dedeuteration procedure,6 i.e. by treating with methanolic HCl.
Our initial hypothesis was that the strained methylenedioxy
ring may somewhat deform the planar aromatic ring and thus
prevent the 5- and 2-positions from being deuterated.
Computational studies and additional experiments with
model compounds were performed to test the hypothesis and
to gain a deeper understanding of the observed discrepancy in
the reactivity of dehydroxycubebin 7.
Figure 2. Optimized structures for the three different conformations of
dehydroxycubebin 7: the extended 12 and the two π-stacked
conformations 13 and 14.
imply that the stacked sandwich conformations 13 and 14 are
preferred under the reaction conditions (Figure 2 and Table 1).
Table 1. Energy Difference (in kcal/mol) between the
Stacked and Extended Conformations for 7 and 8
molecule (functional)
E(13) − E(12)
E(14) − E(12)
7 (B3LYP)
7 (PWPB95)
8 (B3LYP)
8 (PWPB95)
−2.0
−3.3
−3.2
−4.8
−1.2
−2.6
−0.6
−1.4
Deuteration experiments with model compounds gave
similar but not identical results as obtained with the lignans.
In 1,3-benzodioxole (1,2-methylenedioxybenzene 10, Scheme
2), the para/meta positions to the methylenedioxy substituent
The energies obtained in the PWPB95 calculations are
expected to be more accurate because the molecules are
treated at a higher level of theory. The sandwich conformation
with the substituents oriented in opposite directions (13,
Figure 2) is energetically slightly lower than the one having the
aryl rings stacked with identical orientation (14, Figure 2),
probably due to a more favorable multipole interaction between
the oxygen containing substituents of 13.
Scheme 2. Deuteration of Model Compounds 1,3-
Benzodioxole 10 and Veratrole 11
The H/D exchange reaction is an ordinary electrophilic
aromatic substitution taking place in two steps: first the
electrophile attacks giving rise to a positively charged resonance
stabilized intermediate (an arenium ion, also called a Wheland
intermediate or sigma complex), and the leaving group departs
in the second step (Scheme 3). Simultaneous attack-and-
departure mechanisms are not known.13 The degree of
deuterium incorporation is expected to be governed by
were fully deuterated, while only ca. 30% of the ortho positions
were deuterated. With veratrole (1,2-dimethoxybenzene 11), all
aromatic positions were successfully deuterated. Acid catalyzed
detritiation experiments9 with aryl tritiated 1,3-benzodioxoles,
and deuteration experiments10 with 4-hydroxy-1,3-benzodiox-
ole (sesamol), have shown that the para ring positions are
always more reactive than the ortho position. This has been
rationalized by suggesting a deformation of the aromatic ring
(the Mills−Nixon effect) and by the quasi-aromatic nature of
the heterocyclic ring of 10.11 The deviant behavior of 10 has
recently been reported in other types of reactions12 as well.
The model compound 10 thus serves to demonstrate that
diminished reactivity at the ortho sites is characteristic of
methylenedioxy substituted systems. However, it is not clear
why just one ortho site in each aryl ring of 7 is H/D exchanged
at ca. 30% yield and the other two ortho sites remain
untouched, whereas in 10 both ortho sites participate fully in
the exchange.
a
Scheme 3. Mechanism of H/D Exchange
a
R denotes the rest of the lignan molecule.
10637
dx.doi.org/10.1021/jo5019427 | J. Org. Chem. 2014, 79, 10636−10640

The Journal of Organic Chemistry
Note
equilibrium factors of the participant D and H species present
in D2O and at the reactive aryl sites.
One of them is conformation independent, possibly due to a
proposed quasi-aromatic nature of the five-membered methyl-
enedioxy substituent ring11 restricting the complete deuteration
at the ortho positions, which is observed also for sesamol10 and
the model compound 10. The other reason seems to be steric
interaction, which is conformation dependent and blocks even a
partial deuteration at C-5 (C-5′) of 7.
In the computational studies, the two reaction steps, namely
the addition of a proton (deuterium) from one face of the ring
plane and abstraction of a proton from the other face were
studied separately. The reaction from both faces are illustrated
in Figure 3, where the face of the ring plane exposed to the
solvent is denoted “up”, while the other is “down”.
EXPERIMENTAL SECTION
■
General Experimental. Lignanolactones were synthesized by the
tandem Michael addition−alkylation method followed by Raney nickel
desulfurization and debenzylation.15 To obtain lignanofurans the
lignanolactones were further reduced to dibenzylbutanediols with
15
LiAlH4 and then cyclized to the corresponding lignanofurans either
with concentrated HCl under microwave heating16 or during the
deuteration reaction as described below. The ionic liquid [bmim]Cl
was synthesized following the published procedure.17 DCl solution (35
wt % in D2O, 99 atom % D) was commercially available from Sigma-
Figure 3. Proton addition from the “up” and “down” faces of structure
13 is illustrated for dehydroxycubebin (7).
1
Aldrich. H NMR and 13C NMR (500 or 300 MHz and 150 or 75
MHz, respectively) spectra were recorded in CDCl3 or (CD3)2CO.
Chemical shifts are given in δ in ppm and J values in Hz, setting the
scale relative to the solvent signals.18 Mass spectra (EI, 70 eV
ionization energy, or ESI TOF) were acquired.
The calculated differences in the activation energies suggest
that the deuterium is added from the “up” face, where the steric
pressure is weaker (left of Figure 3), and the proton leaves from
the “down” face. A transition state with the water molecules
approaching from the “down” face is not feasible for 7. The
calculated energies suggest that the deuteration reaction can
proceed normally from the exposed outer “up” face of the
molecule, but the proton cannot leave C-5 (see numbering in
Scheme 1) as the less flexible 7 does not allow water molecules
to approach from the interring “down” face. Thus, the proton
departure is impaired and the second step of the electrophilic
aromatic substitution reaction cannot occur. Instead, the
reaction is reversed and the deuteron is liberated back to the
solution. Conversely, both reaction steps are possible for 8,
whose methoxy groups can freely rotate and do not obstruct
the approach of the water molecules on the “down” face.
The geometric data show that the interring distance is
roughly the same for 7 and 8 in both the neutral and
protonated stage (3.5 and 3.3 Å, respectively). The interring
distance is always somewhat smaller for the protonated
structure, probably due to cation−π interactions between the
protonated ring and the other aryl ring.14 The short distance
between the aryl rings may be one reason behind the difficulties
of solvating the “down” proton. The standard deviation of the
distances of 0.2 Å is also larger for 8, indicating a greater
coplanarity between the rings of 7 with a standard deviation of
0.1 Å. The increased coplanarity is most likely due to
interactions with the more rigid methylenedioxy substituent
of 7 as compared to the methoxy groups in 8.
The structural analysis combined with the deuteration results
indicate that the aryl rings cannot rotate freely in either of the
molecules. Thus, there are two atropisomers (13 and 14) for
both compounds, and the molecules do not switch between the
two conformers under the reaction conditions. The small
energy difference of 0.7 kcal/mol between the two closed
conformers of 7 suggests that the two conformers are present
under the experimental conditions, which is the reason for the
incomplete deuteration at C-2 and C-2′. The computational
studies show that the second step of the reaction is sterically
prevented for only one of the conformers, whereas for the other
conformer the C-2 and C-2′ sites are more exposed to the
solvent.
Deuteration Procedure. Lignanolactone or lignanodiol (20−90
mg/0.055−0.302 mmol), or the model compounds veratrol or
methylenedioxybenzene, and [bmim]Cl (8 mol equiv) were dried
under high vacuum (1−0.5 mbar) after which they were placed in a
dry 5 mL pressure-proof reaction vial. 35% DCl/D2O (70−90 mol
equiv of DCl) was added, and the vial was sealed. Microwave power of
80 W was used to reach 40 °C. The reaction was held at this
temperature with stirring and continuous compressed air cooling (2
bar) for 30 min. After the mixture was allowed to cool to rt, 0.5−1.0
mL of D2O was added and the mixture was stirred for 10 min and
extracted with EtOAc. The organic phase was washed with water and
brine, dried over anhydrous Na2SO4, filtered, and concentrated in a
rotary evaporator, and the product was dried under high vacuum. If
necessary, the deuteration procedure was repeated until the isotopic
purity was more than 90%. The crude product was purified by column
chromatography using EtOAc and CH2Cl2 as eluent, and the
deuterated lignan (74−84%) was obtained as a white waxlike solid
or colorless oil.
Characterization of Products. The products were characterized
1
by H NMR, 13C NMR, LRMS, and HRMS and comparison with the
unlabeled analogues. The sites of deuteration were determined by
comparing the 1H NMR and 13C NMR spectra with the spectra of the
unlabeled compound assigned with COSY, HSQC, and HMBC. The
1H NMR spectra of deuterium labeled compounds are the same as
those for unlabeled compounds, with the exception that signals from
the deuterated aromatic sites are lacking and the coupling patterns of
the protons attached to the adjacent carbons are simplified
accordingly. In 13C NMR, a carbon atom attached to one deuterium
produces a low intensity triplet since the spin of deuterium is 1.
Deuteration percentages of carbons C-2 and C-6 were determined
by performing line shape fitting (Lorentzian lines) to measure areas for
signals of both deuterated and nondeuterated carbons from the
quantitative 13C-spectrum using PERCH software (PERCH Solutions
Ltd., Kuopio, Finland). The quantitative 13C-spectrum was recorded
using the following: inverse-gated 1H-decoupling sequence, 45°
excitation pulse flip angle, 0.4 s acquisition time, 200 s relaxation
delay, and 484 repetitions. Carbon C-6 appears to be fully deuterated
(clear 1:1:1 triplet), whereas the signal of C-2 consists of a singlet
(nondeuterated C-2) and partially overlapping 1:1:1 triplet (deu-
terated C-2). Location C-5 remained virtually unaffected (Figure S1 in
Supporting Information). The deuteration percentage in each location
was evaluated from signal areas of deuterated and nondeuterated
carbons (Table S1 in Supporting Information). This resulted in 100%
and 29% deuteration in locations C-6 and C-2, respectively. The result
for C-2 approximately translates into a 2:1 ratio between non-
deuterated and deuterated C-2.
To conclude, two factors are suggested to be responsible for
the deviant deuteration behavior preventing the complete
deuteration of dehydroxycubebin 7 at C-2 (2′) and C-5 (5′).
10638
dx.doi.org/10.1021/jo5019427 | J. Org. Chem. 2014, 79, 10636−10640

The Journal of Organic Chemistry
Note
The isotopic purities of deuterated lignans were determined from
the ion clusters in the molecular ion region in EI or ESI mass spectra
by comparison with those of undeuterated compounds.
The spectroscopic data are given as Supporting Information.
[2,4,6,2′,4′,6′-2H6]-3,3′-Dihydroxylignano-9,9′-lactone (d6-
Enterolactone, 1). White solid (75 mg, 82% yield). The product
was characterized by comparison with the nondeuterated compound.19
1H NMR (300 MHz, (CD3)2CO) δ 2.47−2.72 (4H, m, H-7′, 8, 8′),
3.53 (2H, dd, J = 6.2, 8.9, H-9a, 9′a), 3.84 (3H, s, OCH3), 3.85 (3H, s,
OCH3), 3.91 (2H, dd, J = 6.8, 8.9, H-9b, 9′b). 13C NMR (75 MHz,
CDCl3) δ 39.1 (C-7, 7′), 46.7 (C-8, 8′), 55.9 (OCH3), 56.0 (OCH3),
73.4 (C-9, 9′), 111.0 (t, C-5, 5′)D, 111.8 (t, C-2, 2′)D, 120.3 (C-6,
6′)D, 133.0 (C-1, 1′), 147.5 (C-4, 4′), 149.0 (C-3, 3′). Proportions of
isotopologues in EIMS spectra were 3% 2H5 and 100% 2H6. EIMS (70
eV) m/z: M+ 378 (93%), 180 (4), 155 (73), 154 (100), 140 (10), 124
(15). HRMS (EI) m/z: [M+] calcd for C22H22D6O5 378.2313, found
378.2308.
2.88 (1H, dd, J = 6.6, 13.8, H-7a), 2.97 (1H, dd, J = 5.7, 13.8, H-7b),
3.88 (1H, t, J = 8.7, H-9′a), 4.04 (1H, dd, J = 6.9, 8.7, H-9′b), 7.09
(1H, s, H-5′), 7.13 (1H, s, H-5), 8.22 (2H, s, 3-OH, 3′-OH). 13C
NMR (75 MHz, (CD3)2CO) δ 35.2 (C-7), 38.6 (C-7′), 42.1 (C-8′),
46.8 (C-8), 71.5 (C-9′), 114.1 (t, C-4′)D, 114.2 (t, C-4)D, 116.2 (t, C-
2′)D, 116.8 (t, C-2)D, 120.3 (t, C-6′)D, 121.2 (t, C-6)D, 130.1 (C-5),
130.2 (C-5′), 140.7 (C-1), 141.1 (C-1′), 158.3 (C-3, 3′), 178.8 (C-9).
[2,4,6,2′,4′,6′-2H6]-3,3′-Dihydroxy-9,9′-epoxylignan (d6-Entero-
furan, 6). White solid (14 mg, 74% yield). The product was
characterized by comparison with the nondeuterated compound.22 1H
NMR (300 MHz, (CD3)2CO) δ 2.13−2.26 (2H, m, H-8, 8′), 2.50
(2H, dd, J = 8.7, 13.5, H-7a, 7′a), 2.67 (2H, dd, J = 5.7, 13.5, H-7b,
7′b), 3.43 (2H, dd, J = 6.6, 8.7, H-9a, 9′a), 3.79 (2H, dd, J = 7.2, 8.7,
H-9b, 9′b), 7.08 (2H, s, H-5, 5′), 8.13 (2H, s, 3-OH, 3′-OH). 13C
NMR (75 MHz, (CD3)2CO) δ 39.7 (C-7, 7′), 47.5 (C-8, 8′), 73.7 (C-
9, 9′), 113.6 (t, C-4, 4′)D, 116.2 (t, C-2, 2′)D, 120.4 (t, C-6, 6′), 130.0
(C-5, 5′), 143.1 (C-1, 1′), 158.3 (C-3, 3′). Proportions of
isotopologues in ESI-TOF spectra were 4% 2H5 and 100% 2H6.
HRMS (EI) m/z: [M+] calcd for C18H14D6O3 290.1789; found
290.1793.
2
Proportions of isotopologues in EIMS spectrum were 15% H5 and
2
100% H6. HRMS (EI) m/z: [M+] calcd for C18H12D6O4 304.1582;
found 302.1567.
[2,5,6,2′,5′,6′-2H6]-4,4′-Dihydroxy-3,3′-dimethoxylignano-9,9′-
lactone (d6-Matairesinol, 2). Colorless oil (19 mg, 95% yield). The
product was characterized by comparison with the nondeuterated
compound.19,20 1H NMR (300 MHz, CDCl3) δ 2.42−2.65 (4H, m, H-
7′, 8, 8′), 2.84−2.98 (2H, m, H-7), 3.81 (3H, s, OCH3′), 3.81 (3H, s,
OCH3) 3.88 (1H, dd, J = 6.9, 9.0, H-9′a), 4.15 (1H, dd, J = 6.9, 9.0, H-
9′b), 5.5 (br, 4- and 4′-OH). 13C NMR (75 MHz, CDCl3) δ 34.47 (C-
7), 38.20 (C-7′), 40.99 (C-8′), 46.60 (C-8), 55.81 (OCH3), 55.86
(OCH3), 71.39 (C-9′), 110.69 (t, C-2′)D, 111.24 (t, C-2)D, 113.74 (t,
C-5′)D, 114.07 (t, C-5)D, 120.95 (t, C-6′)D, 121.70 (t, C-6)D,129.36
(C-1′), 129.60 (C-1), 144.37 (C-4′), 144.50 (C-4), 146.59 (C-3′),
146.70 (C-3), 178.94 (C-9). Proportions of isotopologues in EIMS
[6,6′-2H2]-(3,4),(3′,4′)-Bismethylenedioxy-9,9′-epoxylignan (d2-
Dehydroxycubebin, 9). Pale yellow solid (18 mg, 75% yield). The
product was characterized by comparison with the nondeuterated
compound 7.23,24 1H NMR (500 MHz, (CD3)2CO) δ 2.13−2.21 (2H,
m, H-8, 8′), 2.51 (2H, dd, J = 8.0, 14.0, H-7a, 7′a), 2.60 (2H, dd, J =
6.5, 14.0, H-7b, 7′b), 3.42 (2H, dd, J = 6.0, 8.5, H-9a, 9′a), 3.80 (2H,
dd, J = 6.5, 8.5, H-9b, 9′b), 5.94 (4H, s, O−CH2−O, O−CH2−O′),
6.65 (2H, s, H-2, 2′), 6.72 (2H, s, H-5, 5′). 13C NMR (75 MHz,
(CD3)2CO) δ 39.6 (C-7, 7′), 47.4 (C-8, 8′), 73.6 (C-9, 9′), 101.7 (C-
10, 10′), 108.6 (C-5, 5′), 109.8 (C-2, 2′), 122.1 (t, C-6, 6′)D, 135.4 (C-
1, 1′), 146.8 (C-4, 4′), 148.6 (C-3, 3′). Proportions of isotopologues in
2
2
spectra were 9% H5 and 100% H6. EIMS (70 eV) m/z: M+ 364
(55%), 224 (6), 167 (16), 140 (100), 125 (10). HRMS (EI) m/z:
[M+] calcd for C20H16D6O6 364.1793; found 364.1780.
2
2
2
[2′,4′,6′-2H3]-3′-Hydroxylignano-9,9′-lactone (3). Pale yellow
viscous oil (21 mg, 78% yield). The product was characterized by
comparison with the nondeuterated compound.6 1H NMR (500 MHz,
CDCl3) δ 2.43 (1H, dd, J = 9.0, 13.0 Hz, H-7′a), 2.47−2.54 (1H, m,
H-8′), 2.59 (1H, dd, J = 5.0, 11.0 Hz, H-7′b), 2.59−2.63 (1H, m, H-8),
2.94 (1H, dd, J = 7.0, 14.0, H-7a), 3.09 (1H, dd, J = 5.0, 14.0, H-7b),
3.84 (1H, dd, J = 8.0, 9.0 Hz, H-9′a), 4.08 (1H, dd, J = 7.5, 9.0 Hz, H-
9′b), 5.42 (1H, s, OH), 7.11 (1H, s, H-5′), 7.17 (2H, d, J = 7.5 Hz, H-
2, 6), 7.23 (1H, t, J = 7.5, H-4), 7.30 (2H, t, J = 7.2, H-3, 5) 13C NMR
(75 MHz, CDCl3) δ 35.3 (C-7), 38.4 (C-7′), 41.3 (C-8′), 46.7 (C-8),
71.5 (C-9′), 113.7 (t, C-4′)D, 115.4 (t, C-2′)D, 120.8 (t, C-6′)D, 127.1
(C-4), 129.0 (C-3, 5), 129.5 (C-2, 6), 129.9 (C-5′), 137.9 (C-1), 139.7
(C-1′), 156.1 (C-3′), 179.1 (C-9). Proportions of isotopologues in
EIMS were 100% H2, 68% H3, and 43% H4. When the natural
abundance of 13C is taken into account, the ratio of species d2:d3:d4 is
about 6:3:2. EIMS (70 eV) m/z: 345 (9%), 344 (18), 343 (28), [2H2]-
M+ 342 (41), 206 (4), 163 (5), 162 (4), 138 (53), 137 (100), 136
(76). HRMS (EI) m/z: [M+] calcd for C20H18D2O5 342.1436; found
342.1452.
COMPUTATIONAL DETAILS
■
The initial molecular structures were constructed manually and
preoptimized at the semiempirical level25,26 as described below. The
structures were then optimized at the density functional theory (DFT)
level using the Becke−Perdew (BP86) functional27,28 and split-valence
basis sets augmented with polarization functions29 employing the
resolution of the identity approximation to speed up the
calculations.30−32 The Cartesian coordinates of the optimized
molecular structures of 12, 13, 14 are given as Supporting Information.
The relative energies were calculated at the DFT level using the
B3LYP33−36 functional in combination with triple-ζ polarization basis
sets.37 The energies were also assessed using the PWPB95 double-
hybrid functional in combination with quadruple-ζ polarization basis
sets.37 The D3 dispersion correction was used in all calculations.38 The
conductor-like screening model with a dielectric constant of 80 was
used to model solvent effects.39,40 The chain of spheres algorithm was
used to speed up the hybrid and double-hybrid DFT calculations.41
The calculations were performed with the ORCA software.42,43
Reaction Path Calculations. Solvent molecules were included to
allow for a more accurate description of the mechanism of the
deuteration reaction. For each system, four water molecules were
explicitly considered to hydrate the proton in addition to the dielectric
continuum. For each reaction path, a proton was initially placed 2.2 Å
farther away from the carbon atom where the reaction occurs than in
the corresponding protonated optimized geometry. The four water
molecules were manually added close to the proton to avoid clashes
among them and with the organic molecule. The structure was
preoptimized at the semiempirical PM6-D3H4 level of theory25,26
while keeping the proton−carbon distance fixed. The obtained
conformation was taken as the starting point for the next step, in
2
2
EIMS spectra were 7% H2 and 100% H3. EIMS (70 eV) m/z: M+
285 (100%), 175 (16), 137 (52), 136 (56), 111 (87), 110 (88), 91
(89). HRMS (EI) m/z: [M+] calcd for C18H15D3O3 285.1444; found
285.1443.
[2,5,6,2′,5′,6′-2H6]-4,4′-Dihydroxy-3,3′-dimethoxy-9,9′-epoxyli-
gnan (d6-Anhydrosecoisolariciresinol, 4). White solid (16 mg, 84%
yield). The product was characterized by comparison with the
nondeuterated compound.21 1H NMR (500 MHz, (CD3)2CO) δ
2.15−2.20 (2H, m, H-8, 8′), 2.49 (2H, dd, J = 8.3, 13.5, H-7a, 7′a),
2.60 (2H, dd, J = 6.3, 13.5, H-7b, 7′b), 3.43 (2H, dd, J = 6.0, 8.5, H-9a,
9′a), 3.79 (2H, dd, J = 6.0, 8.5, H-9b, 9′b), 3.81 (6H, s, OCH3,
OCH3′), 7.26 (2H, s, OH, OH′). 13C NMR (75 MHz, (CD3)2CO) δ
39.6 (C-7, 7′), 47.7 (C-8, 8′), 56.3 (3-OCH3, 3′-OCH3), 73.8 (C-9,
9′), 112.8 (t, C-2, 2′)D, 115.3 (t, C-5, 5′)D, 121.7 (t, C-6, 6′)D, 133.0
(C-1, 1′), 145.8 (C-4, 4′), 148.3 (C-3, 3′). Proportions of
2
2
isotopologues in EIMS spectra were 9% H5 and 100% H6. EIMS
(70 eV) m/z: M+ 350 (72%), 141 (62), 140 (100), 126 (9), 125 (10).
HRMS (EI) m/z: [M+] calcd for C20H18D6O5 350.2000; found
350.1996.
[2,5,6,2′,5′,6′-2H6]-3,3′,4,4′-Tetramethoxy-9,9′-epoxylignan (d6-
Brassilignan, 5). White solid (17 mg, 77% yield). The product was
characterized by comparison with the nondeuterated compound 8.22
1H NMR (300 MHz, CDCl3) δ 2.13−2.25 (2H, m, H-8, 8′), 2.53 (2H,
dd, J = 8.1, 13.8, H-7a, 7′a), 2.64 (2H, dd, J = 6.0, 13.8, H-7b, 7′b),
10639
dx.doi.org/10.1021/jo5019427 | J. Org. Chem. 2014, 79, 10636−10640

The Journal of Organic Chemistry
Note
which the proton was moved 0.1 Å closer to carbon. The procedure
was repeated until a structure with a carbon−proton distance equal to
the equilibrium geometry for the protonated molecules was reached.
After the procedure was completed, all the semiempirical structures
were refined by geometry optimizations at the BP86/def2-SVP level of
theory with a fixed proton−carbon distance. Single-point B3LYP/def2-
TZVP energies were calculated using the BP86-optimized geometries.
(13) Smith, M. B.; March, J. March’s advanced organic chemistry:
reactions, mechanisms, and structure, 5th ed.; Wiley: 2001; p 675.
(14) Ma, J. C.; Dougherty, D. A. Chem. Rev. 1997, 97, 1303.
(15) Raffaelli, B.; Leppala, E.; Chappuis, C.; Wahala, K. Environ.
̈
̈
̈
̈
̈
Chem. Lett. 2006, 4, 1.
(16) Pohjoispaa, M.; Wahala, K. Molecules 2013, 18, 13124.
̈
̈
̈
̈
̈
(17) Dupont, J.; Consorti, C. S.; Suarez, P. A. Z.; de Souza, R. F. Org.
Synth. 2002, 79, 236; Coll. Vol. 2004, 10, 184.
(18) Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J. Org. Chem. 1997,
62, 7512.
ASSOCIATED CONTENT
* Supporting Information
■
S
(19) Eklund, P.; Lindholm, A.; Mikkola, J.-P.; Smeds, A.; Lehtila, R.;
̈
Spectroscopic data for deuterated compounds (1−6, 9) and the
Cartesian coordinates of 12−14. This material is available free
of charge via the Internet at http://pubs.acs.org.
Sjoholm, R. Org. Lett. 2003, 5, 491.
̈
(20) Fischer, J.; Reynolds, A. J.; Sharp, L. A.; Sherburn, M. S. Org.
Lett. 2004, 6, 1345.
(21) Pan, H.; Lundgren, L. N. Phytochem. 1995, 39, 1423.
(22) Ward, R. S.; Hughes, D. D. Tetrahedron 2001, 57, 2057.
(23) Rehnberg, N.; Magnusson, G. J. Org. Chem. 1990, 55, 4340.
(24) Gaboury, J. A.; Sibi, M. P. J. Org. Chem. 1993, 58, 2173.
(25) Rezac, J.; Hobza, P. J. Chem. Theory Comput. 2012, 8, 141.
́
(26) Stewart, J. P. The mopac2012 program. http://openmopac.net/
MOPAC2012.html.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: kristiina.wahala@helsinki.fi.
*E-mail: dage.sundholm@helsinki.fi.
Notes
■
The authors declare no competing financial interest.
(27) Becke, A. D. Phys. Rev. A 1988, 38, 3098.
(28) Perdew, J. P. Phys. Rev. B 1986, 33, 8822.
(29) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297.
ACKNOWLEDGMENTS
■
̈
(30) Eichkorn, K.; Treutler, O.; Ohm, H.; Haser, M.; Ahlrichs, R.
Chem. Phys. Lett. 1995, 242, 652.
̈
This research has been supported by the Academy of Finland
(AF) through its Computational Science Research Programme
(LASTU) and the AF-DAAD (Deutscher Akademischer
Austauschdienst) Mobility Project 1257329. We acknowledge
the Finnish Center for Scientific Computing (CSC) for
computational resources, Magnus Ehrnrooth Foundation, and
University of Helsinki for financial support. We also thank
Gudrun Silvennoinen and Eduardo de Pedro Beato for
technical assistance.
(31) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102, 346.
(32) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theoret.
Chim. Acta 1997, 97, 119.
(33) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(34) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200.
(35) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(36) Stevens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623.
(37) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571.
̈
(38) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104.
REFERENCES
■
(1) (a) Yousefzadi, M.; Sharifi, M.; Behmanesh, M.; Moyano, E.;
Bonfill, M.; Cusido, R. M.; Palazon, J. Eng. Life Sci. 2010, 281.
(b) Cunha, W. R.; e Silva, M. L. A.; Veneziani, R. C. S.; Ambrosio, S.
́
R.; Bastos, J. K. In Phytochemicals − A Global Perspective of Their Role
in Nutrition and Health; Rao, V., Ed.; InTech: Rijeka, Croatia, 2012; p
213. (c) Niwa, A. M.; Marcarini, J. C.; Sartori, D.; Maistro, E. L.;
Mantovani, M. S. J. Food Comp. Anal. 2013, 30, 1. (d) Carvalho, M. T.
M.; Rezende, K. C. S.; Evora, P. R. B.; Bastos, J. K.; Cunha, W. R.; e
Silva, M. L. A.; Celotto, A. C. Phytother. Res. 2013, 27, 1784.
(2) Horiuti, J.; Polanyi, M. Nature 1934, 377.
(39) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 2 1993,
̈
̈
799.
(40) Sinnecker, S.; Rajendran, A.; Klamt, A.; Diedenhofen, M.;
Neese, F. J. Phys. Chem. A 2006, 110, 2235.
(41) Neese, F.; Wennmohs, F.; Hansen, A.; Becker, U. Chem. Phys.
2009, 356, 98.
(42) Neese, F. 2004, ORCA, an ab initio, density functional and
semiempirical program package, version 2.9.
(43) Neese, F. WIREs Comput. Mol. Sci. 2012, 2, 73.
(3) Ingold, C. K.; Raisin, C. G.; Wilson, C. L. Nature 1934, 734.
(4) (a) Ingold, C. K.; Raisin, C. G.; Wilson, C. L. J. Chem. Soc. 1936,
915. (b) Ingold, C. K.; Clifford, G. R.; Wilson, C. L. J. Chem. Soc. 1936,
1637. (c) Best, A. P.; Wilson, C. L. J. Chem. Soc. 1938, 28.
(5) Wahala, K.; Makela, T.; Backstrom, R.; Brunow, G.; Hase, T. J.
̈
̈
̈
̈
̈
̈
̈
Chem. Soc., Perkin Trans. 1 1986, 95.
(6) (a) Leppala, E.; Pohjoispaa, M.; Koskimies, J.; Wahala, K. J. Label.
̈
̈
̈
̈
̈
̈
̈
Compd. Radiopharm. 2008, 51, 407. (b) Wahala, K.; Rasku, S.
̈
̈
̈
Tetrahedron Lett. 1997, 38, 7287.
(7) (a) Hakala, U.; Wahala, K. J. Org. Chem. 2007, 72, 5817.
̈
̈
̈
(b) Pohjoispaa, M.; Hakala, U.; Silvennoinen, G.; Wahala, K. J. Label.
̈
̈
̈
̈
̈
Compd. Radiopharm. 2010, 53, 429.
(8) Ayres, D. C.; Loike, J. D. Lignans. Chemical, biological and clinical
properties; Cambridge University Press: Cambridge, 1990; p 16.
(9) Czernohorsky, J. H.; Richards, K. E.; Wright, G. J. Aust. J. Chem.
1972, 25, 1459.
(10) Hill, R. K.; Vaidya, N. A.; Morton, G. H. J. Label. Compd.
Radiopharm. 1982, 19, 1265.
(11) Dauksa
̌
s, V. K.; Purvaneckas, G. V.; Udrenaite, E. B.; Gineityte,
̇ ̇ ̇
V. L.; Barauskaite, A. V. Heterocycles 1981, 15, 1395.
̇
(12) Vanchura, B. A.; Preshlock, S. M.; Roosen, P. C.; Kallepalli, V.
A.; Staples, R. J.; Maleczka, R. E., Jr.; Singleton, D. A.; Smith, M. R., III.
Chem. Commun. 2010, 46, 7724.
10640
dx.doi.org/10.1021/jo5019427 | J. Org. Chem. 2014, 79, 10636−10640