Atropo-enantioselective synthesis of the natural bicoumarin (+)-yisokotanin A via a configurationally stable biaryl lactone

Article abstract of DOI:10.1002/1099-0690(200203)2002:6<1096::AID-EJOC1096>3.0.CO;2-Z

The atropo-enantioselective total synthesis of the axially chiral bicoumarin (+)-isokotanin A (1) is described. Key steps were the formation of a configurationally stable seven-membered biaryl lactone and its kinetic resolution by atroposelective ring cleavage. The previous assignment of the absolute configuration of (+)-isokotanin A (1) (and its synthetic precursors) was confirmed by quantum chemical CD calculations. Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002.

Full text of DOI:10.1002/1099-0690(200203)2002:6<1096::AID-EJOC1096>3.0.CO;2-Z

FULL PAPER
Atropo-Enantioselective Synthesis of the Natural Bicoumarin (؉)-Isokotanin A
via a Configurationally Stable Biaryl Lactone[‡]
Gerhard Bringmann,*[a] Jürgen Hinrichs,[a] Petra Henschel,[a] Jürgen Kraus,[a] Karl Peters,[b]
and Eva-Maria Peters[b]
Keywords: Biaryls / Axial chirality / Kinetic resolution / Natural products / Total synthesis
The atropo-enantioselective total synthesis of the axially
configuration of (+)-isokotanin A (1) (and its synthetic pre-
chiral bicoumarin (+)-isokotanin A (1) is described. Key steps
cursors) was confirmed by quantum chemical CD calcula-
were the formation of a configurationally stable seven-mem- tions.
bered biaryl lactone and its kinetic resolution by atroposelec-
tive ring cleavage. The previous assignment of the absolute
(
Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
2002)
Introduction
Because of their interesting bioactivities and unusual, ap-
parently polyketide-derived,[1] structures, including the axial
chirality, the isokotanins are rewarding target molecules for
stereoselective total synthesis. In the course of the first, and
so far only, reported enantioselective preparation of (ϩ)-
isokotanin A (1), the naturally occurring atropisomer was
assigned the (M) configuration.[3] As isokotanin A (1) con-
tains two C1 substituents in its o- and oЈ-positions next to
the biaryl axis, it seemed profitable to construct this com-
pound in enantiomerically pure form by our ‘‘lactone
method’’,[4] using a recently introduced variant, the kinetic
resolution of configurationally stable biaryl lactones.[5] This
total synthesis of (ϩ)-isokotanin A (1) is described here.
The isokotanins A (1), B (2), and C (3, see Figure 1) are
axially chiral bicoumarins from the sclerotia of the fungus
Aspergillus alliaceus, first isolated in 1994 by Gloer et al.[1]
All of these natural biaryls show antifeedant activity
against the fungivorous beetle Carpophilus hemipterus.[1]
A
short time after its discovery, isokotanin B (2), now named
O-demethyl-6,6Ј-bisiderin, was also isolated from the
ascosttromata of the fungus Petromyces alliaceus by Uda-
gawa’s group.[2] O-Methylation of this bicoumarin gave 6,6Ј-
bisiderin (i.e., isokotanin A, 1).[2]
Results and Discussion
Our synthetic plan was to construct 1 by way of a bi-
phenyl derivative (M)-4 (Scheme 1), which should be avail-
able in enantiomerically pure form by kinetic resolution of
the Ϫ presumably configurationally stable Ϫ lactone 5. It
was planned to prepare this bridged biaryl from the bromo
ester 6 as a single monoaryl precursor. The (P) enantiomer
of 4,4Ј-bisorcinol (biaryl 4 in its deprotected, tetraphenolic
form) had already been prepared earlier, via a six-mem-
bered lactone, although with only moderate atropo-diaster-
eoselectivity in the ring-cleavage step (up to 48% de).[6]
Later, (M)-4 was used as an intermediate in Lin’s synthesis
of (ϩ)-isokotanin A (1).[3]
Similar biaryl lactones with two naphthalene compon-
ents, as in 7 (Figure 2), had previously been prepared in a
five-step reaction sequence starting from their monomeric
bromonaphthalene precursors.[7] Being configurationally
stable at the biaryl axis, they can be reductively transformed
Figure 1. Isokotanins A (1), B (2), and C (3), axially chiral bicou-
marins with insect antifeedant activity
[‡]
Novel Concepts in Directed Biaryl Synthesis, 97. Ϫ Part 96: S.
Tasler, H. Endress, G. Bringmann, Synthesis 2001, 1993Ϫ2002.
[a]
Institut für Organische Chemie, Universität Würzburg,
Am Hubland, 97074 Würzburg, Germany
Fax: (internat.) ϩ 49-(0)931/888-4755
E-mail: bringman@chemie.uni-wuerzburg.de
[b]
Max-Planck-Institut für Festkörperforschung,
Heisenbergstraße 1, 70506 Stuttgart, Germany
1096
WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/0306Ϫ1096 $ 17.50ϩ.50/0 Eur. J. Org. Chem. 2002, 1096Ϫ1106

Atropo-Enantioselective Synthesis of the Natural Bicoumarin (ϩ)-Isokotanin A
FULL PAPER
Scheme 1. Retrosynthetic analysis of isokotanin A (1)
Scheme 2. (a) Cu, DMF, 89%; (b) LiAlH4, THF, 97%; (c) MnO2,
CH2Cl2, 92%; (d) KOH, EtOH, 94%; (e) DCC, DMAP, CH2Cl2,
72%
with high efficiency (with a relative rate constant, krel, of
50) into the corresponding atropisomerically pure diol by
kinetic resolution, with oxazaborolidine-activated borane as
a chiral H-nucleophile.[5] Similar reductions with the corres-
ponding biaryldicarbaldehyde or the respective hydroxy al-
dehyde, in contrast, were found to be essentially nonstereo-
selective (krel ഠ 1).[8]
more coplanar, helical arrangement as observed for similar
six-membered lactones.[4] Whilst, for the latter compounds,
the bridge drastically lowers the atropisomerization barrier,
this is not the case for seven-membered lactones such as 5.
We therefore anticipated that 5 should, like 7,[5] be config-
urationally stable at room temperature. This was confirmed
by HPLC experiments on a chiral phase, which produced
two well-separated peaks (for detailed conditions, see Exp.
Sect.).
Figure 2. The configurationally stable seven-membered biaryl lac-
tone 7
For the synthesis of the key lactone 5 required here,
bromo ester 6 was prepared from 3,5-dimethoxybenzoic
acid by esterification[9] and bromination according to a pro-
cedure reported by Danishefsky.[10] Ullmann coupling of 6
gave the corresponding racemic diester 8[11,12] (Scheme 2),
reduction of which with LiAlH4 provided diol 9,[12] which
was then oxidized to the dialdehyde 10. This known[13] com-
pound can thus now be prepared in a substantially im-
proved overall yield (79%) and without use of toxic reagents
such as HgO or benzene, as employed formerly.[13] Under
Cannizzaro conditions, 10 underwent an intramolecular
disproportionation to the hydroxy acid 11. The final ring
closure was achieved with DCC and DMAP under modi-
Figure 3. Crystal structures of compounds 5 and 9
As previously noted for the model lactone 7,[5] atropo-
fied[14] Steglich conditions[15] to give the seven-membered enantioselective reduction of racemic 5 with oxazaborolid-
lactone 5 in an overall yield of 39% (7 steps).
ine-activated borane[17] resulted in an efficient kinetic res-
Compounds 9 and 5 were obtained in crystalline forms olution, with a high relative rate constant[18] (krel) of 43
suitable for X-ray structure analysis.[16] As can be seen in (Scheme 3, analytical scale). The product alcohol (M)-9 was
Figure 3, the two phenyl rings adopt nearly orthogonal po- obtained with up to 96% ee at the beginning of the reaction,
sitions, not only in the ‘‘open’’ biaryl 9, but also in lactone whilst the less reactive (P) enantiomer of the starting mat-
5. The ester bridge, as part of the seven-membered lactone, erial 5 was enriched to optical purity (for the assignment of
is too long to force the two parts of the molecule into a the absolute configurations, vide infra).
Eur. J. Org. Chem. 2002, 1096Ϫ1106
1097

G. Bringmann, J. Hinrichs, P. Henschel, J. Kraus, K. Peters, E.-M. Peters
FULL PAPER
Scheme 3. Kinetic resolution of biaryl lactone 5 (analytical scale)
Figure 4. Determination of the atropisomerization barrier of lac-
tone 5: thermal racemization of (P)-5 in toluene (A), determination
With enantiomerically pure lactone (P)-5 available, its
atropisomerization barrier was determined by thermal race-
mization experiments in toluene. As depicted (A in Fig-
ure 4), the isomerization is still very slow at 60 °C, while
complete racemization was achieved after 1 h at 100 °C. Al-
though it precludes the possibility of recycling the more
slowly reacting substrate atropisomer during the ring-open-
of the rate constants k (B), and Eyring plot for ∆H϶ and ∆S϶ (C)
Table 1. Experimental activation parameters for the atropisomeriz-
ation [(M)-5 Ǟǟ (P)-5]
ing reaction in situ, in a dynamic kinetic resolution, this T [°C]
k [sϪ1
]
t1/2 [min] ∆G϶ [kJ/mol]
option allows a similarly efficient ‘‘off-line’’ recycling of the
Ϫ still precious Ϫ material with the undesired configura-
60
80
100
4.71·10Ϫ5
2.44·10Ϫ4
1.75·10Ϫ3
245.2
47.3
6.6
109.4
111.4
111.7
tion. The activation parameters for the atropisomerization
[(M)-5 Ǟǟ (P)-5] are summarized in Table 1. As in the case
of the calculation of the isomerization barriers of helically ∆H϶ ϭ 90.1 kJ/mol calcd. (AM1):
∆H϶ ϭ 83.2 kJ/mol
twisted six-membered biaryl lactones,[19] the semiempirical
AM1 parameterization should similarly give reliable results
for the seven-membered lactone 5. Indeed, the experimental
data were in good agreement with the results obtained by
∆S϶ ϭ Ϫ58.9 J/K·mol
quantum chemical calculations (∆H϶
ϭ 83.2 kJ/mol,
AM1).
merization barrier for the lactone 5 is in agreement with
For the thermal atropisomerization of the biaryl 13, also decreased activation parameters for lactones versus cyclic
seven-membered but ether-bridged, (see Figure 5, below), ethers in the field of six-membered bridged biaryls[4,21]
Ϫ
an activation enthalpy ∆H϶ of 120.2 kJ/mol (in cyclohexa- but is still not low enough to permit unhindered rotation at
none) had been determined earlier.[12,20] The lower iso- the biaryl axis at room temperature.
1098
Eur. J. Org. Chem. 2002, 1096Ϫ1106

Atropo-Enantioselective Synthesis of the Natural Bicoumarin (ϩ)-Isokotanin A
FULL PAPER
Figure 5. The ether-bridged biaryls (M)-13 and (P)-13, obtained as
by-products
On a preparative scale, the oxazaborolidine-mediated
kinetic resolution of 5 (krel ϭ 27) was stopped at 56% con-
version (see Scheme 4). After chromatographic separation,
the optical purity of the product alcohol (M)-9 was in-
creased from 75% to 95% ee by a single crystallization step,
while the remaining lactone (P)-5 was enantiomerically al-
most pure (96% ee) straightaway (Scheme 4). LiAlH4 reduc-
tion of (P)-5 gave the diol (P)-9 without significant loss of
enantiomeric purity. On recrystallization, (P)-9 was ob-
tained in stereochemically homogeneous form (Ͼ 99% ee).
Both enantiomers of 9 were separately transformed into
the respective tetramethyl ethers 4 (Scheme 4) by hydroxy/
bromo exchange and subsequent LiAlH4 reduction, as pre-
viously used for other biaryl systems.[22] As by-products, the
known,[12] stereochemically almost pure, seven-membered
biaryl ethers (M)-13 and (P)-13 (Figure 5) were obtained in
Ͻ 20% yield, presumably due to base-catalyzed ring closure
during or after the first hydroxy/halogen exchange on diol 9.
Crystals of (M)-4 were of suitable quality for X-ray struc-
ture analysis.[23] As expected, the two phenyl rings adopt a
nearly orthogonal position, due to the space needed by the
four ortho substituents (Figure 6).
Finally, by treatment of (M)-4 and (P)-4 with boron trib-
romide (Scheme 4, vide supra), (M)-4,4Ј-bisorcinol [(M)-14] (P)-5; (b) LiAlH4, THF, 76%; (c) recrystallization from ethyl acet-
and (P)-4,4Ј-bisorcinol [(P)-14] were obtained. By this strat-
Scheme 4. (a) (S)-12, BH3·THF, THF, 46% of (M)-9 and 43% of
ate/cyclohexane; (d) PPh3, (CBrCl2)2,[53] CH2Cl2; (e) LiAlH4, Et2O,
83% for (M)-4 (two steps), 81% for (P)-4 (two steps); (f) BBr3,
egy, both enantiomers of 14 were prepared in good overall
yields and (nearly) enantiopure from one kinetic resolution
experiment. As illustrated for related cases,[5] recycling of
the undesired atropisomer [e.g., of the (P) products if only
(M)-configured material is needed] is in principle possible
through thermal racemization of the respective ‘‘wrong’’
lactone enantiomer [here (P)-5], but also by oxidative con-
version of (P)-9 back to lactone (P)-5 by the route outlined
in Scheme 2, its subsequent racemization, and renewed
resolution. If, vice versa, only (P)-configured products are
required, the same strategies still apply, now with the (R)
enantiomer of oxazaborolidine 12.
CH2Cl2, 96% for (M)-14, 84% for (P)-14
The assignment of the absolute configurations of (M)-14
and (P)-14 and their precursors was achieved by compar-
ison of their optical rotations with literature data.[3,6,24,25]
An additional confirmation of these previous[26] assign-
Figure 6. Crystal structure of biaryl (M)-4
ments was provided by quantum chemical CD calcula- thesized previously, but as yet without assignment of its ab-
tions.[27] For this purpose, the configurationally stable bi- solute configuration.[12] As clearly shown in Figure 7, the
aryl ether (ϩ)-13 was chosen, since it had already been syn- position and intensity of the maxima calculated for (P)-13
Eur. J. Org. Chem. 2002, 1096Ϫ1106
1099

G. Bringmann, J. Hinrichs, P. Henschel, J. Kraus, K. Peters, E.-M. Peters
FULL PAPER
are in good agreement with the CD spectrum experiment-
ally obtained for (ϩ)-13, whereas the CD curve calculated
for (M)-13 is nearly the mirror image of the experimental
one. The biaryl ether (ϩ)-13 can thus unambiguously be
assigned the (P) configuration.
Scheme 5. (a) (CH3CO)2O, TiCl4, CH2Cl2, 94%; (b) TiCl4, benzene,
82%; (c) ClCO2Me, pyridine, 55% of (M)-16 and 10% of (M)-17
however, was considerably higher than those reported dur-
ing the initial isolation of (M)-1 (223Ϫ226 °C)[1] and its first
total synthesis (240Ϫ242 °C)[3] Ϫ but was in good agree-
ment with the respective value of ‘‘6,6Ј-bisiderin’’ [ϭ (M)-
1, vide supra] (285Ϫ290 °C).[2] The low yield for (ϩ)-isokot-
anin A (1) Ϫ only 26% over the last two steps Ϫ can be
explained by the formation of the by-products (M)-19 and
(M)-20, resulting from the loss of one or two carbonate
groups prior to the ring closure of (M)-16, with subsequent
methylation of all hydroxy functions.
Figure 7. Determination of the absolute configuration of (ϩ)-13
by comparison of the experimental CD spectrum with the spectra
quantum chemically calculated for (P)- and (M)-13
These investigations not only permitted the first attribu-
tion of the absolute configuration of compound 13, but
more importantly also established the axial configurations
for all of the biphenyls of this series. Furthermore, they con-
firmed the assignments already made for diol 9,[3] 4,4Ј-bi-
sorcinol (14),[3,6,24,25] and finally for (ϩ)-isokotanin A (1,
vide infra).[3]
The final steps towards the target molecule (M)-1 were
accomplished by use of Lin’s procedure,[3,28] according to a
sequence already well known for the preparation of dimeric
coumarins.[29,30] The tetramethyl ether (M)-4 was acylated
[29]
in the presence of TiCl4
in almost quantitative yield
(Scheme 5). In the next step, the methyl ethers para to the
biaryl axis were cleaved selectively to give the phenol (M)-
15 in 82% yield. Treatment of this material with methyl
chloroformate gave the carbonate (M)-16, along with 10%
of the (previously[3] undescribed) monoacylated product
(M)-17.
Scheme 6. (a) KOtBu, tBuOH; (b) NaH, Me2SO4, HMPA, 26% of
(M)-1, 19% of (M)-19, and 13% of (M)-20 (two steps)
According to the CD spectrum of (M)-1 (Figure 8), Uda-
In the presence of KOtBu as a base, (M)-16 cyclized to gawa’s 6,6Ј-bisiderin,[2] which had been obtained by partial
give the bicoumarin (M)-18 (Scheme 6). This very polar synthesis from O-demethyl-6,6Ј-bisiderin (ϭ isokotanin B,
compound contained some impurities, but could not be 2), can now also be assigned to have the (M) configuration.
purified by column chromatography, due to its very low Unfortunately, Udagawa et al. did not give the optical rota-
solubility in common organic solvents. Therefore, the crude tion of 6,6Ј-bisiderin, and the CD spectrum was only meas-
(M)-18 obtained was directly transformed into (ϩ)-isokot- ured from 225 to 325 nm (the reported maxima are marked
anin A (1) by use of the reagent combination NaH, in Figure 8). The good agreement in this region, however,
Me2SO4, and HMPA.[3,28,31] The spectroscopic data and the is sufficient for unambiguous assignment of the absolute
optical rotation of (M)-1 were in full accordance both with configuration of 6,6Ј-bisiderin (and the natural O-demethyl-
those of the natural product[1] and with those of material 6,6Ј-bisiderin). Gloer et al. showed (ϩ)-isokotanin B (2) to
synthesized earlier.[2,3] The melting point of 286Ϫ289 °C, have the same axial configuration as (ϩ)-isokotanin A (1).[1]
1100
Eur. J. Org. Chem. 2002, 1096Ϫ1106

Atropo-Enantioselective Synthesis of the Natural Bicoumarin (ϩ)-Isokotanin A
FULL PAPER
Therefore, isokotanin B (2) isolated from A. alliaceus is also CD Calculations
(M)-configured.
The wavefunctions required for the calculation of the ro-
tational strengths for the electronic transitions from the
ground state to excited states of compound 13 were ob-
tained by CNDO/S-CI calculations[37] with a CI expansion
including 576 singly occupied configurations and the
ground state determinant. These calculations were carried
out with LinuX Pentium II workstations by use of the
BDZDO/MCDSPD[38] program package. All single CD
spectra thus obtained were added up by Boltzmann statist-
ics using appropriate heats of formation, to give the calcu-
lated overall CD spectrum for the biaryl ether 13. For better
visualization, the rotational strengths were transformed into
∆ε values and superimposed with a Gaussian band shape
function. The assignment of responsible transitions of the
relevant calculated CD bands at 217 and 274 nm was per-
formed by comparison of the experimental UV spectrum
with the calculated one. For further details concerning
quantum chemical calculation of chiroptical properties, see
the corresponding section in ref.[39]
Figure 8. CD spectrum of synthetic isokotanin A (1) and compar-
ison with literature data (marked as small squares) for ‘‘6,6Ј-
bisiderin"[2]
Conclusion
In summary, an efficient atropo-enantioselective total
synthesis of the bicoumarin (ϩ)-isokotanin A (1) has been
accomplished. In contrast to the first synthesis of the title
compound, in which a Meyers-type homocoupling[32] of a
bromooxazoline was utilized in the coupling step,[3] the
common intermediate (M)-4 has now been prepared by kin-
etic resolution of the configurationally stable seven-mem-
bered biaryl lactone 5, with the option of recycling the
‘‘wrong’’ atropisomer by thermal racemization. The
quantum chemical CD calculation of 13 confirmed the pre-
vious assignment of the absolute configuration of (ϩ)-iso-
kotanin A (1) and its synthetic precursors.
Experimental Section
General Remarks: Melting points were determined with a Reichert-
Jung Thermovar hot-stage apparatus and are uncorrected. IR spec-
tra were recorded with a PerkinϪElmer 1420 spectrometer and are
reported in wavenumbers (cmϪ1). Optical rotations were measured
with a PerkinϪElmer 241 MC polarimeter. CD spectra were taken
with a Jasco J-715 spectropolarimeter, with ethanol (Uvasol ,
Merck) as the solvent. 1H and 13C NMR spectra were recorded
with Bruker AC 250 (250 and 63 MHz), AMX 400 (400 and
101 MHz), and DMX 600 (600 and 151 MHz) machines. Chemical
shifts δ are reported in ppm, and coupling constants J in Hertz
(Hz). As the internal reference, the residual solvent signals [1H:
δ(CDCl3) ϭ 7.26; δ([D6]DMSO) ϭ 2.50; 13C: δ(CDCl3) ϭ 77.0;
δ([D6]DMSO) ϭ 39.4] were used. EI mass spectra (70 eV) were
measured with Finnigan MAT 90 and MAT 8200 mass spectro-
meters; the relative intensities are given in brackets. Microanalyses
were performed by the microanalytical laboratory of the Inorganic
Institute of the University of Würzburg, Germany (Leco CHNS-
932). THF was freshly distilled from potassium, DMF and DMSO
from CaH2, and CH2Cl2 from P2O5. Et2O and toluene were dis-
tilled from sodium wire. Methanol and ethanol were heated under
reflux in the presence of magnesium; the drying agent during the
distillation of tert-butyl alcohol was KOtBu. Acetone, acetonitrile,
2-propanol, n-hexane, cyclohexane, petroleum ether (PE, 50Ϫ70
°C), and ethyl acetate were used after distillation. Pyridine was dis-
tilled from NaH. All air- or moisture-sensitive reactions were car-
ried out with dry glassware under nitrogen or argon. Reaction solv-
ents were removed in vacuo. Column chromatography was per-
formed on 63Ϫ200 µm silica gel (Merck); for preparative TLC,
silica gel aluminum foils (60 F254, 20 ϫ 20 cm, Merck) were used.
Oxazaborolidine (S)-12 (1  in toluene, which was removed in va-
cuo prior to use) and BH3·THF (1  in THF) were obtained from
Aldrich. HPLC analyses were carried out with a combination of a
Waters 510 HPLC pump, a 20-µL injection loop, and Chiralcel OF
and OD-H columns (0.46 ϫ 25 cm, Daicel Chem. Ind.) with UV
detection at 254 nm (biaryl ether 13: 280 nm). The atropisomers
of lactone 5 were separated on the OD-H column with hexane/2-
propanol (70:30; 0.8 mL/min) as the eluent; the retention times for
Computational Section
Conformational Analyses
Conformational analyses of the biaryl lactone 5 and the
biaryl ether 13 were performed on Silicon Graphics OCT-
ANE R10000 workstations by AM1[33] parameterization as
implemented in the VAMP6.5 program package,[34] starting
from preoptimized geometries generated by the TRIPOS[35]
force field.
Helimerization Barrier
The potential surface was calculated through variation of
two particular dihedral angles that describe the atropiso-
merization process of 5. The corresponding transition struc-
tures were optimized by initial calculation of the molecular
forces, followed by a transition state search by the
NS01A[36] algorithm as implemented in VAMP6.5. Force
calculations were subsequently applied to characterize
transition structures by computation of their normal vibra-
tions. Correspondence to their local minima was deter-
mined by IRC calculations.
Eur. J. Org. Chem. 2002, 1096Ϫ1106
1101

G. Bringmann, J. Hinrichs, P. Henschel, J. Kraus, K. Peters, E.-M. Peters
FULL PAPER
(M)-5 and (P)-5 were 11 min and 16 min, respectively. For diol 9,
the OF column was used with hexane/2-propanol (50:50; 1.0 mL/
min) as the eluent, with retention times of 13 and 19 min for (P)-
9 and (M)-9, respectively. For the chromatographic separation of
was recrystallized from ethyl acetate/cyclohexane, affording crystals
suitable for X-ray structure analysis. M.p. 174Ϫ176 °C (ref.[12]
Ϫ1
˜
174Ϫ175.5 °C). IR (KBr): ν ϭ 3360 cm (br. s, OH), 3070 (m,
Ar-H), 2970, 2930, 2910, 2860, 2815 (w, w, w, w, m, CϪH), 1585,
(M)- and (P)-13, the OD-H column was used with hexane/2-pro- 1565 (s, s, CϭC), 1305, 1145, 830 (s, s, s). 1H NMR (250 MHz,
panol (70:30; 0.5 mL/min); (M)-13: 15 min, (P)-13: 21 min. The CDCl3): δ ϭ 3.68 (s, 6 H, 6- and 6Ј-OCH3), 3.86 (s, 6 H, 4- and
samples of bisorcinol 14 were analyzed on the OF stationary phase
(hexane/2-propanol, 80:20; 0.5 mL/min), with signals for (P)-14
after 44 min and for (M)-14 after 51 min.
4Ј-OCH3), 4.18 and 4.24 (AB system, 2J ϭ 11.6 Hz, 4 H, CH2OH),
4
4
6.52 (d, J ϭ 2.4 Hz, 2 H, Ar-H), 6.70 (d, J ϭ 2.4 Hz, 2 H, Ar-
H). 13C NMR (63 MHz, CDCl3): δ ϭ 55.3, 55.9 (OCH3), 63.7
(CH2OH), 98.7, 105.3, 116.3, 142.0, 158.1, 160.4 (Ar-C). MS
(70 eV): m/z (%) ϭ 334 (85) [Mϩ], 317 (18) [335 Ϫ H2O], 316 (100)
[Mϩ Ϫ H2O], 301 (15) [316 Ϫ CH3], 285 (9) [316 Ϫ CH3O], 273 (70)
[316 Ϫ C2H3O], 168 (84) [C9H12O3ϩ]. C18H22O6 (334.37): calcd. C
64.66, H 6.63; found C 64.35, H 6.73.
Methyl 2-Bromo-3,5-dimethoxybenzoate (6): A solution of 3,5-di-
methoxybenzoic acid (10.0 g, 54.9 mmol) in methanol (150 mL)
was heated under reflux in the presence of conc. H2SO4 (292 µL,
5.49 mmol) for 24 h. The solvent was removed, water (150 mL) was
added, and the mixture was extracted with Et2O (4 ϫ 50 mL). The
combined organic extracts were dried (Na2SO4), and the solvent
was evaporated to give methyl 3,5-dimethoxybenzoate[9] (10.3 g,
52.5 mmol, 96%). The purity of this product, which solidified on
rac-4,4Ј,6,6Ј-Tetramethoxy-1,1Ј-biphenyl-2,2Ј-dicarbaldehyde (10):
Diol 9 (3.22 g, 9.63 mmol) and MnO2 (8.37 g, 96.3 mmol) were ad-
ded to CH2Cl2 (300 mL) and heated under reflux for 12 h. After
standing (m.p. 39Ϫ41 °C, ref.[9] 41 °C), was sufficient for the next filtration and evaporation of the solvent, the crude product was
step. According to a known procedure,[10] methyl 3,5-dimethoxyb-
enzoate (10.0 g, 51.0 mmol) in acetonitrile (400 mL) was treated
with N-bromosuccinimide (10.9 g, 61.2 mmol) at 0 °C and stirred
obtained as a colorless solid. This material was crystallized from
2-propanol to give pure dialdehyde 10[42] (2.94 g, 8.90 mmol, 92%).
Ϫ1
M.p. 164Ϫ165 °C (ref.[13] 159Ϫ160 °C). IR (KBr): ν ϭ 3045 cm
˜
at room temperature for 24 h. After removal of the solvent, the (w, Ar-H), 2980, 2945, 2920, 2825 (w, w, w, m, CϪH), 1670 (s, Cϭ
1
residue was purified by column chromatography (PE/Et2O, 2:1), O), 1580 (s, CϭC), 1310, 1145, 830 (s, s, m). H NMR (250 MHz,
yielding the desired compound 6[10] (10.6 g, 38.4 mmol, 75%) after CDCl3): δ ϭ 3.70 (s, 6 H, 6- and 6Ј-OCH3), 3.91 (s, 6 H, 4- and
4
4
recrystallization from Et2O/PE (m.p. 56.5Ϫ58.5 °C, ref.[10] 57Ϫ59
4Ј-OCH3), 6.76 (d, J ϭ 2.4 Hz, 2 H, Ar-H), 7.16 (d, J ϭ 2.4 Hz,
°C). A chromatographically slower fraction contained methyl 2,6- 2 H, Ar-H), 9.63 (s, 2 H, CHO). 13C NMR (63 MHz, CDCl3): δ ϭ
dibromo-3,5-dimethoxybenzoate[40] (2.51 g, 7.09 mmol, 14%), m.p. 55.7, 56.0 (OCH3), 101.4, 104.6, 118.7, 136.7, 158.6, 160.9 (Ar-C),
148Ϫ149 °C (ref.[40] 149Ϫ150 °C).
191.7 (CϭO). MS (70 eV): m/z (%) ϭ 330 (100) [Mϩ], 315 (7) [Mϩ
Ϫ CH3], 302 (15) [Mϩ Ϫ CO], 301 (34) [Mϩ Ϫ CHO], 299 (14)
[Mϩ Ϫ CH3O], 286 (23) [Mϩ Ϫ C2H4O], 271 (49) [286 Ϫ CH3].
C18H18O6 (330.34): calcd. C 65.45, H 5.49; found C 65.47, H 5.47.
Dimethyl
rac-4,4Ј,6,6Ј-Tetramethoxy-1,1Ј-biphenyl-2,2Ј-dicarb-
oxylate (8): Under dry argon, bromo ester 6 (3.83 g, 13.9 mmol)
and activated copper[41] (9.56 g, 150 mmol) in dry DMF (12 mL)
were heated to 165 °C for 24 h. After the mixture had cooled to rac-2Ј-Hydroxymethyl-4,4Ј,6,6Ј-tetramethoxy-1,1Ј-biphenyl-2-
room temperature, the copper was filtered off and washed thor- carboxylic Acid (11): Dialdehyde 10 (2.87 g, 8.69 mmol) was dis-
oughly with CH2Cl2. The solvents were evaporated, and the residue solved in ethanol (200 mL), treated with KOH (8.10 g, 144 mmol),
was subjected to column chromatography (PE/Et2O, 2:1), to yield
and heated to reflux for 1.5 h. The solvent was removed in vacuo,
water (50 mL) was added to the oily residue, and the pH value was
colorless crystals of 8[42] (2.41 g, 6.17 mmol, 89%) after recrystal-
lization from CH2Cl2/PE. M.p. 162 °C (ref.[11] 160Ϫ161 °C, ref.[12] carefully adjusted to 1Ϫ2 with conc. HCl. This mixture was then
Ϫ1
˜
161Ϫ163 °C). IR (KBr): ν ϭ 3070 cm (m, Ar-H), 2970, 2925, extracted with CH2Cl2 (500 mL altogether), and the combined or-
2815 (m, s, m, CϪH), 1705 (s, CϭO), 1580 (s, CϭC), 1440, 1325, ganic fractions were washed with water (2 ϫ 100 mL). During this
1195, 1055, 840 (s, s, s, s, m). 1H NMR (250 MHz, CDCl3): δ ϭ workup procedure, the hydroxy acid 11 precipitated in the form of
3.60 (s, 6 H, CO2CH3), 3.66 (s, 6 H, 6- and 6Ј-OCH3), 3.87 (s, 6 a very pure colorless powder, which was collected by filtration
4
H, 4- and 4Ј-OCH3), 6.65 (d, J ϭ 2.4 Hz, 2 H, 5- and 5Ј-H), 7.08
(1.71 g). Concentration of the organic solution yielded a second
(d, J ϭ 2.4 Hz, 2 H, 3- and 3Ј-H). 13C NMR (63 MHz, CDCl3): batch of the desired product; the combined yield was 2.86 g
δ ϭ 51.8 (CO2CH3), 55.4, 56.0 (4-, 4Ј-, 6- and 6Ј-OCH3), 102.4, (8.21 mmol, 94%). M.p. 210Ϫ211 °C. IR (KBr): ν ϭ 3420 cm
4
Ϫ1
˜
105.1, 120.1, 132.0, 158.1, 159.3 (Ar-C), 167.6 (CϭO). MS (70 eV): (br. s, OH), 3000 (w, Ar-H), 2940, 2835 (m, m, CϪH), 1685 (s, Cϭ
m/z (%) ϭ 390 (100) [Mϩ], 375 (1) [Mϩ Ϫ CH3], 359 (6) [Mϩ
CH3O], 329 (13) [359 Ϫ CH2O], 300 (21) [329 Ϫ CHO], 299 (16)
Ϫ
O), 1585 (s, CϭC), 1445, 1315, 1150, 1070, 825 (s, s, s, s, m). 1H
NMR (250 MHz, [D6]DMSO): δ ϭ 3.54 (s, 3 H, 4-OCH3), 3.62 (s,
[329 Ϫ CH2O], 209 (66). C20H22O8 (390.39): calcd. C 61.53, H 5.68; 3 H, 6-OCH3), 3.78 (s, 3 H, 4Ј-OCH3), 3.83 (s, 3 H, 6Ј-OCH3), 4.04
4
found C 61.63, H 5.70. In addition, a small quantity of methyl 3,5-
dimethoxybenzoate[9] (180 mg, 917 µmol, 7%), the hydro-dehalo-
genation product of ester 6, was obtained.
(br. s, 2 H, CH2), 6.40 (d, 4J ϭ 2.4 Hz, 1 H, 5-H), 6.69 (d, J ϭ
4
4
2.4 Hz, 1 H, 3-H), 6.75 (d, J ϭ 2.4 Hz, 1 H, 5Ј-H), 6.87 (d, J ϭ
2.8 Hz, 1 H, 3Ј-H). 13C NMR (63 MHz, [D6]DMSO): δ ϭ 54.8 (4Ј-
OCH3), 55.2 (6Ј-OCH3), 55.3 (4-OCH3), 55.7 (6-OCH3), 60.5
(CH2), 96.2 (C-5), 101.2 (C-5Ј), 102.0 (C-3), 105.1 (C-3Ј), 114.9,
117.0, 134.7, 142.7, 157.2, 157.9, 159.0, 159.2 (Ar-C), 168.3 (Cϭ
O). MS (70 eV): m/z (%) ϭ 348 (29) [Mϩ], 330 (90) [Mϩ Ϫ H2O],
315 (5) [330 Ϫ CH3], 301 (17) [330 Ϫ CHO], 299 (9) [330 Ϫ CH3O],
286 (15) [301 Ϫ CH3], 271 (36) [299 Ϫ CO], 182 (100) [C9H10O4ϩ],
166 (46) [C9H10O3ϩ]. C18H20O7 (348.35): calcd. C 62.06, H 5.79;
found C 61.77, H 5.54.
rac-2,2Ј-Dihydroxymethyl-4,4Ј,6,6Ј-tetramethoxy-1,1Ј-biphenyl (9):
A solution of diester 8 (3.98 g, 10.2 mmol) in THF (150 mL) was
slowly added to a suspension of LiAlH4 (582 mg, 15.3 mmol) in
THF (50 mL). After stirring for 2 h at room temperature, the mix-
ture was hydrolyzed by careful addition of water (25 mL) and acidi-
fied with 2  HCl. The organic solvent was evaporated, and the
aqueous residue was diluted with water (100 mL) and extracted
with CH2Cl2. The combined organic extracts were washed with
water and dried (Na2SO4). After evaporation of the solvent, the rac-1,3,9,11-Tetramethoxydibenzo[c,e]oxepin-5(7H)-one (5): A solu-
crude product was crystallized from CH2Cl2 to yield pure 9[42] tion of DMAP (192 mg, 1.57 mmol) in CH2Cl2 (250 mL) was satur-
(3.30 g, 9.87 mmol, 97%) as colorless crystals. An analytical sample ated with gaseous HCl (5 min) and then added to a mixture of
1102
Eur. J. Org. Chem. 2002, 1096Ϫ1106

Atropo-Enantioselective Synthesis of the Natural Bicoumarin (ϩ)-Isokotanin A
FULL PAPER
hydroxy acid 11 (1.04 g, 2.99 mmol) and DCC (975 mg, CH2Cl2). CD (EtOH): ∆ε194 ϭ ϩ25.6, ∆ε213 ϭ Ϫ16.3, ∆ε232
4.73 mmol). The resulting slurry was heated to reflux for 4 h under ϩ29.7, ∆ε261 ϭ Ϫ15.3.
argon. After this had cooled to room temperature, excess reagent
ϭ
Determination of the Atropisomerization Barrier of Lactone 5 by
Thermal Racemization Experiments: A solution of enantiomerically
almost pure (92% ee) lactone (P)-5 (2.00 mg, 6.05 µmol) in toluene
(2 mL) was heated with stirring at the respective temperature under
argon. At defined times, samples (200 µL) were taken, concen-
trated, and stored at Ϫ20 °C until HPLC examination. The racem-
ization data thus obtained are summarized (see A in Figure 4 and
the section Results and Discussion); from these data, the activation
parameters for the atropisomerization [(M)-5 Ǟǟ (P)-5] were calcu-
lated.[44]
was destroyed by addition of glacial HOAc (ca. 250 µL). The clear
solution obtained by filtration was washed with water (100 mL)
and brine (100 mL), dried (Na2SO4), and concentrated. Column
chromatography (PE/Et2O, 1:1 Ǟ 1:10) gave crude lactone 5, which
could be further purified by a second chromatographic step (cyclo-
hexane/ethyl acetate, 4:1 Ǟ 2:1) and crystallization from Et2O/PE
(colorless cubes, suitable for X-ray structure analysis, 707 mg,
2.14 mmol, 72%). M.p. 124Ϫ126 °C. IR (KBr): ν ϭ 3060 cmϪ1 (w,
˜
Ar-H), 2975, 2920, 2820 (m, m, m, CϪH), 1700 (s, CϭO), 1585,
1550 (s, m, CϭC), 1450, 1325, 1150, 1045, 830 (s, s, s, s, m). 1H
NMR (250 MHz, CDCl3): δ ϭ 3.78 (s, 3 H, 1-OCH3 or 3-OCH3),
3.82 (s, 3 H, 9-OCH3), 3.85 (s, 3 H, 11-OCH3), 3.88 (s, 3 H, 1-
(M)-2,2Ј,4,4Ј-Tetramethoxy-6,6Ј-dimethyl-1,1Ј-biphenyl
[(M)-4]:
1,2-Dibromotetrachloroethane [(CBrCl2)2, 98.3 mg, 302 µmol] and
PPh3 (79.2 mg, 302 µmol) were added to a solution of (M)-9 (95%
ee, 36.2 mg, 108 µmol) in CH2Cl2 (3 mL). After stirring for 4 h at
room temperature, the mixture was cooled to 0 °C, Et2O (1 mL)
and LiAlH4 (17.2 mg, 453 µmol) were added, and stirring was con-
tinued at room temperature for 5 h. Water (4 mL) and 2  HCl
(4 mL) were used to stop the reaction. The mixture was then ex-
tracted with CH2Cl2, dried (Na2SO4), concentrated, and purified
by column chromatography (PE/Et2O, 3:1 Ǟ 1:1). The two color-
less products were crystallized from Et2O/PE, providing crystals
suitable for an X-ray structure analysis in the case of the main
product (M)-4. The spectroscopic data of (M)-4 (27.1 mg, 89.6
µmol, 83%) were in accordance with those previously obtained for
racemic material.[28,29,45] (M)-4: M.p. 127 °C. [α]2D3 ϭ ϩ33.9 (c ϭ
0.83, CHCl3). CD (EtOH): ∆ε208 ϭ ϩ12.5, ∆ε230 ϭ ϩ5.7, ∆ε243 ϭ
Ϫ4.5, ∆ε285 ϭ ϩ0.8. The biaryl ether (M)-13 (92% ee, 5.40 mg,
17.1 µmol, 16%) was obtained as a by-product, whose spectroscopic
data were identical with literature values.[12,46] (M)-13: M.p.
140Ϫ141 °C (ref.[12] 137.5Ϫ139 °C). [α]2D3 ϭ Ϫ7.5 (c ϭ 0.17, acet-
one) {ref.[12] [α]2D0 ϭ Ϫ16.0 (c ϭ 2.12, acetone)}. CD (EtOH):
∆ε194 ϭ Ϫ18.9, ∆ε204 ϭ Ϫ8.3, ∆ε217 ϭ Ϫ31.0, ∆ε242 ϭ ϩ13.9,
∆ε256 ϭ ϩ12.7, ∆ε274 ϭ ϩ2.8, ∆ε291 ϭ ϩ4.7; calculated for (M)-
13: ∆ε192 ϭ Ϫ9.1, ∆ε200 ϭ Ϫ7.9, ∆ε206 ϭ Ϫ5.8, ∆ε217 ϭ Ϫ32.2,
∆ε274 ϭ ϩ7.4.
2
OCH3 or 3-OCH3), 4.81 and 4.99 (AB system, J ϭ 11.7 Hz, 2 H,
CH2), 6.57 (d, 4J ϭ 2.1 Hz, 1 H, 2-H or 4-H), 6.59 (d, 4J ϭ 2.4 Hz,
4
4
1 H, 2-H or 4-H), 6.71 (d, J ϭ 2.4 Hz, 1 H, 10-H), 6.96 (d, J ϭ
2.4 Hz, 1 H, 8-H). 13C NMR (63 MHz, CDCl3): δ ϭ 55.5 (11-
OCH3), 55.7, 55.8 (1-OCH3 and 3-OCH3), 56.1 (9-OCH3), 70.0
(CH2), 99.8 (C-2 or C-4), 102.9 (C-10), 104.5 (C-2 or C-4), 105.1
(C-8), 116.3, 116.8, 132.9, 137.7, 157.8, 159.1, 160.0, 160.5 (Ar-C),
170.0 (CϭO). MS (70 eV): m/z (%) ϭ 330 (100) [Mϩ], 315 (5) [Mϩ
Ϫ CH3], 301 (7) [Mϩ Ϫ CHO], 299 (6) [Mϩ Ϫ CH3O], 287 (9) [315
Ϫ CO], 271 (26) [299 Ϫ CO]. C18H18O6 (330.34): calcd. C 65.45,
H 5.49; found C 65.05, H 5.52.
Kinetic Resolution of rac-1,3,9,11-Tetramethoxydibenzo[c,e]oxepin-
5(7H)-one (5) (Analytical Scale): BH3·THF (1  in THF, 129 µL,
129 µmol) was added at 0 °C under argon to a solution of oxaza-
borolidine (S)-12 (26.8 mg, 96.6 µmol) in THF (1.5 mL). After stir-
ring for 30 min at room temperature, the solution was added drop-
wise over 5 min to a solution of racemic lactone 5 (10.6 mg, 32.1
µmol) in THF (1.5 mL) at Ϫ20 °C. Stirring was continued at this
temperature. At intervals, 100-µL portions of the reaction mixture
were quenched with 2  HCl and extracted with diethyl ether. The
organic solutions were purified by TLC (cyclohexane/ethyl acetate,
1:5) to yield samples of lactone 5 and diol 9, which were analyzed
by HPLC.[18]
(M)-2,2Ј,4,4Ј-Tetrahydroxy-6,6Ј-dimethyl-1,1Ј-biphenyl [(M)-14]: At
0 °C, a solution of biaryl (M)-4 (26.5 mg, 87.6 µmol) in CH2Cl2
(5 mL) was treated with a BBr3 solution (1.0  in CH2Cl2, 420 µL,
420 µmol). After 5 h of stirring at room temperature, more BBr3
(105 µL, 105 µmol) was added. Complete conversion was reached
after 20 h, and excessive reagent was destroyed with methanol
(1 mL). After removal of the solvent, the crude product was puri-
fied by two consecutive chromatographic steps (each with cyclohex-
ane/ethyl acetate, 1:5), providing (M)-14 (94% ee, 20.7 mg, 84.1
µmol, 96%) as a colorless oil. The spectroscopic data of (M)-14
(M)-2,2Ј-Dihydroxymethyl-4,4Ј,6,6Ј-tetramethoxy-1,1Ј-biphenyl
[(M)-9] and (P)-1,3,9,11-Tetramethoxydibenzo[c,e]oxepin-5(7H)-one
[(P)-5]: The oxazaborolidine/borane system used for the kinetic res-
olution of 5 on a preparative scale was prepared from (S)-12
(503 mg, 1.82 mmol) and BH3·THF (2.42 mL, 2.42 mmol) in THF
(15 mL) as outlined above, cooled to Ϫ20 °C, and added dropwise
over 10 min to a solution of racemic lactone 5 (200 mg, 605 µmol)
in THF (15 mL) at Ϫ20 °C under argon. The course of the reaction
was monitored by HPLC. After 2 h of stirring at Ϫ20 °C (56%
conversion, krel ϭ 27), the reaction mixture was carefully hy-
drolyzed with water (8 mL) and 2  HCl (6 mL), and extracted
with CH2Cl2. The combined organic extracts were dried (Na2SO4),
the solvent was evaporated, and the residue was subjected to col-
umn chromatography (cyclohexane/ethyl acetate, 1:5). Lactone (P)-
5 (86.3 mg, 261 µmol, 43%) was obtained as a colorless oil with
96% ee, while the alcohol (M)-9 (93.2 mg, 279 µmol, 46%) had an
agreed with those reported for the racemic compound.[47] [α]2D3
ϭ
ϩ35.8 (c ϭ 1.01, ethanol) {ref.[48,49] [α]2D5 ϭ ϩ36.2 (c ϭ 0.50, eth-
anol), ref.[24] [α]2D5 ϭ ϩ39.4 (c ϭ 0.50, ethanol), ref.[25] [α]2D0 ϭ ϩ40.0
(c ϭ 3.00, ethanol), ref.[3] [α]2D4 ϭ ϩ38.7 (c ϭ 0.90, ethanol)}. CD
(EtOH): ∆ε195 ϭ Ϫ2.9, ∆ε209 ϭ ϩ9.5, ∆ε229 ϭ ϩ1.7, ∆ε246 ϭ Ϫ1.3,
∆ε270 ϭ Ϫ1.0, ∆ε288 ϭ ϩ1.6.
ee of 75% and was recrystallized from ethyl acetate/cyclohexane (P)-2,2Ј-Dihydroxymethyl-4,4Ј,6,6Ј-tetramethoxy-1,1Ј-biphenyl [(P)-
to give colorless crystals (50% ee, 36.3 mg, 109 µmol, 18%) and 9]: A solution of lactone (P)-5 (96% ee, 74.8 mg, 226 µmol) in THF
enantiomerically enriched mother liquor (95% ee, 56.9 mg, 170 (8 mL) was treated with LiAlH4 (17.2 mg, 453 µmol) at 0 °C. After
µmol, 28%). (M)-9 (95% ee): M.p. 141Ϫ142 °C (ref.[12] 144Ϫ145 stirring for 4 h at room temperature, the reaction was stopped by
°C, ref.[3] 144Ϫ145 °C). [α]2D3 ϭ ϩ58.6 (c ϭ 1.03, acetone) {ref.[12,43] addition of water and the mixture was acidified with 2  HCl. Ex-
[α]1D8 ϭ ϩ54.9 (c ϭ 0.51, acetone), ref.[3] [α]2D4 ϭ ϩ62.5 (c ϭ 1.10,
traction with CH2Cl2, drying of the organic phase (Na2SO4), and
column chromatography (cyclohexane/ethyl acetate, 1:5) of the res-
idue obtained by evaporation of the solvent gave (P)-9 (95% ee,
CHCl3)}. CD (EtOH): ∆ε208 ϭ ϩ59.5, ∆ε228 ϭ ϩ11.8, ∆ε246
ϭ
Ϫ9.7, ∆ε288 ϭ ϩ4.1. (P)-5 (96% ee): [α]2D3 ϭ Ϫ8.4 (c ϭ 0.94,
Eur. J. Org. Chem. 2002, 1096Ϫ1106
1103

G. Bringmann, J. Hinrichs, P. Henschel, J. Kraus, K. Peters, E.-M. Peters
FULL PAPER
57.7 mg, 173 µmol, 76%). Crystallization from ethyl acetate/cyclo- (Na2SO4) and concentrated. From the brownish residue, the pure
hexane provided enantiomerically pure diol (P)-9 (99.9% ee,
product (M)-15[3] was obtained by column chromatography (PE/
42.0 mg, 126 µmol, 56%), while the mother liquor (15.2 mg, 45.5 ethyl acetate, 1:2) in the form of a colorless oil (22.9 mg, 63.9 µmol,
µmol, 20%) had 85% ee. (P)-9 (99.9% ee): M.p. 143.5Ϫ144 °C
82%). [α]2D3 ϭ ϩ35.2 (c ϭ 0.76, CHCl3) {ref.[3] [α]2D4 ϭ ϩ29.3 (c ϭ
(ref.[12] 143.5Ϫ145 °C). [α]2D3 ϭ Ϫ60.4 (c ϭ 0.97, acetone) {ref.[12,50] 0.50, CHCl3)}. CD (EtOH): ∆ε211 ϭ ϩ4.3, ∆ε241 ϭ Ϫ1.5, ∆ε267 ϭ
[α]2D4 ϭ Ϫ55.0 (c ϭ 1.04, acetone), ref.[3] [α]2D4 ϭ Ϫ62.8 (c ϭ 0.97,
ϩ1.0, ∆ε310 ϭ Ϫ0.6.
CHCl3)}. CD (EtOH): ∆ε209 ϭ Ϫ69.2, ∆ε227 ϭ Ϫ13.8, ∆ε246
ϩ10.4, ∆ε287 ϭ Ϫ5.0.
ϭ
(M)-3,3Ј-Diacetyl-6,6Ј-dimethoxy-4,4Ј-bis(methoxycarbonyloxy)-
2,2Ј-dimethyl-1,1Ј-biphenyl [(M)-16]: According to the literat-
(P)-2,2Ј,4,4Ј-Tetramethoxy-6,6Ј-dimethyl-1,1Ј-biphenyl
(CBrCl2)2 (115 mg, 353 µmol) and PPh3 (92.6 mg, 353 µmol) were dropwise at 0 °C to a solution of (M)-15 (21.8 mg, 60.8 µmol) in
[(P)-4]: ure,[3,28] methyl chloroformate (94.0 µL, 1.22 mmol) was added
added to a solution of (P)-9 (99.9% ee, 42.0 mg, 126 µmol) in
dry pyridine (3 mL). After this mixture had been stirred for 25 h
CH2Cl2 (4 mL). After 4 h of stirring at room temperature, the mix-
at 60 °C, 2  HCl (15 mL) was added at room temperature. The
ture was cooled to 0 °C, diluted with Et2O (1 mL), and treated combined organic phases from the extraction with CH2Cl2 were
with LiAlH4 (20.1 mg, 530 µmol). Stirring was continued at room
temperature (6 h), after which water (6 mL) and 2  HCl (6 mL)
were added. Extraction with CH2Cl2, drying of the combined or-
washed with 2  HCl and water and dried with Na2SO4. The solv-
ent was removed, and the residue was chromatographed (cyclohex-
ane/ethyl acetate, 1:2) to give, besides the desired compound (M)-
ganic phases with Na2SO4, and evaporation of the solvents gave 16 (16.0 mg, 33.7 µmol, 55%), whose spectroscopic data were in
a crude product, from which pure (P)-4 was obtained by column
chromatography (PE/Et2O, 3:1 Ǟ 1:1) and crystallization from of
accordance with values for the racemic material,[28] a small quantity
(M)-3,3Ј-diacetyl-4-hydroxy-6,6Ј-dimethoxy-2,2Ј-dimethyl-4Ј-
Et2O/PE. The spectroscopic data of (P)-4 (colorless crystals, methylcarbonate-1,1Ј-biphenyl [(M)-17] (2.51 mg, 6.03 µmol, 10%)
30.9 mg, 102 µmol, 81%) were in accordance with those previously
as a by-product. Both compounds were oils. (M)-17: [α]2D3 ϭ ϩ33.8
obtained for racemic material.[28,29,45] (P)-4: M.p. 125Ϫ126 °C. [α] (c ϭ 0.32, CHCl3). CD (EtOH): ∆ε200 ϭ ϩ15.1, ∆ε220 ϭ Ϫ2.5,
23
ϭ Ϫ35.1 (c ϭ 0.75, CHCl3). CD (EtOH): ∆ε208 ϭ Ϫ15.6,
∆ε230 ϭ ϩ1.2, ∆ε241 ϭ Ϫ2.5, ∆ε264 ϭ ϩ2.5, ∆ε280 ϭ Ϫ0.5, ∆ε297 ϭ
D
Ϫ1
˜
ϩ0.8, ∆ε315 ϭ Ϫ0.4. IR (KBr): ν ϭ 3410 cm (br. m, OH), 3070
∆ε228 ϭ Ϫ9.5, ∆ε243 ϭ ϩ6.9, ∆ε285 ϭ Ϫ3.0. The biaryl ether (P)-
13[12,46] (96% ee, 7.50 mg, 23.7 µmol, 19%) was obtained as a by- (w, Ar-H), 2985, 2950, 2930, 2825 (w, m, w, w, CϪH), 1755, 1680,
product. (P)-13: [α]2D3 ϭ ϩ8.5 (c ϭ 0.38, acetone) {ref.[12] [α]2D0
ϩ15.5 (c ϭ 2.03, acetone)}. CD (EtOH): ∆ε194 ϭ ϩ20.1, ∆ε204
ϩ8.0, ∆ε217 ϭ ϩ32.1, ∆ε241 ϭ Ϫ15.0, ∆ε256 ϭ Ϫ13.2, ∆ε274
ϭ
ϭ
ϭ
1670 (s, m, m, CϭO), 1580 (s, CϭC), 1245, 1195 (s, s). H NMR
(400 MHz, CDCl3): δ ϭ 1.88 (s, 3 H, 2Ј-CH3), 2.15 (s, 3 H, 2-CH3),
2.51 (s, 3 H, 3Ј-COCH3), 2.63 (s, 3 H, 3-COCH3), 3.71 (s, 6 H, 6-
1
Ϫ2.8, ∆ε293 ϭ Ϫ4.9; calculated for (M)-13: ∆ε192 ϭ Ϫ9.1, ∆ε200 ϭ OCH3 and 6Ј-OCH3), 3.93 (s, 3 H, 4Ј-OCO2CH3), 6.41 (s, 1 H, 5-
Ϫ7.9, ∆ε206 ϭ Ϫ5.8, ∆ε217 ϭ Ϫ32.2, ∆ε274 ϭ ϩ7.4.
H), 6.72 (s, 1 H, 5Ј-H). 13C NMR (151 MHz, CDCl3): δ ϭ 16.7 (2-
CH3), 20.4 (2Ј-CH3), 32.4, 33.2 (3-COCH3 and 3Ј-COCH3), 55.7,
55.8, 56.0 (6-OCH3, 6Ј-OCH3 and 4Ј-OCO2CH3), 97.9 (C-5), 102.6
(C-5Ј), 116.3, 118.3, 124.5, 127.4, 136.4, 139.9, 147.8 (Ar-C), 153.8
(4Ј-OCO2CH3), 158.4, 162.3, 165.5 (Ar-C), 202.8, 204.7 (3-COCH3
and 3Ј-COCH3). MS (70 eV): m/z (%) ϭ 416 (5) [Mϩ], 401 (6) [Mϩ
Ϫ CH3], 388 (9) [Mϩ Ϫ CO], 373 (4) [Mϩ Ϫ C2H3O], 357 (5) [388
Ϫ CH3O], 344 (41) [373 Ϫ CHO], 343 (5) [373 Ϫ CH2O], 330 (24)
[373 Ϫ C2H3O], 329 (100) [388 Ϫ C2H3O2], 315 (9) [330 Ϫ CH3].
C22H24O8 (416.43): calcd. 416.1471; found 416.1472 (HRMS).
(P)-2,2Ј,4,4Ј-Tetrahydroxy-6,6Ј-dimethyl-1,1Ј-biphenyl [(P)-14]: At 0
°C, BBr3 (neat, 60.0 µL, 635 µmol) was added to a solution of
biaryl (P)-4 (24.0 mg, 79.4 µmol) in CH2Cl2 (4 mL). After 2 h at
room temperature, the reaction mixture was quenched with meth-
anol (1 mL), concentrated, and purified by column chromato-
graphy (cyclohexane/ethyl acetate, 1:5) to yield a slightly yellow oil,
from which the desired product (P)-14 was obtained as a colorless
solid by crystallization from ethyl acetate/cyclohexane (Ͼ 99% ee,
17.2 mg, 69.8 µmol, 88%). Compound (P)-14 gave spectroscopic
data identical with literature values for the racemic compound.[47] (M)-4,4Ј-Dihydroxy-7,7Ј-dimethoxy-5,5Ј-dimethyl-6,6Ј-bicoumarin
M.p. 129Ϫ131 °C (ref.[25] Ͻ 130 °C).[51] [α]2D3 ϭ Ϫ35.9 (c ϭ 0.97,
ethanol) {ref.[48,49] [α]2D5 ϭ Ϫ36.7 (c ϭ 0.50, ethanol), ref.[25] [α]2D0
Ϫ41.3 (c ϭ 3.00, ethanol), ref.[6] [α]2D5 ϭ Ϫ38.6 (c ϭ 0.28, ethanol)}.
[(M)-18]: According to Lin’s procedure,[3,28] KOtBu (30.3 mg, 270
µmol) and tert-butyl alcohol (4 mL) were added to (M)-16
(16.0 mg, 33.7 µmol). This mixture was stirred at exactly 60 °C for
ϭ
CD (EtOH): ∆ε194 ϭ ϩ4.0, ∆ε209 ϭ Ϫ11.9, ∆ε229 ϭ Ϫ1.6, ∆ε242 ϭ 2 h. After this had cooled to room temperature, water (10 mL) was
ϩ0.9, ∆ε272 ϭ ϩ0.7, ∆ε287 ϭ Ϫ1.7.
added, the pH value was adjusted to 6 with 2  HCl, and the
solution was extracted with CH2Cl2. The organic extracts were
dried (Na2SO4) and concentrated. The crude product was spectro-
scopically identical to rac-18.[28] It still contained impurities, but
could not be purified chromatographically, due to its high polarity
and very low solubility in common organic solvents. Therefore,
(M)-18 was directly used in the next step.
(M)-3,3Ј-Diacetyl-4,4Ј-dihydroxy-6,6Ј-dimethoxy-2,2Ј-dimethyl-
1,1Ј-biphenyl [(M)-15]: According to literature precedence,[3,28,29]
tetramethyl ether (M)-4 (95% ee, 25.1 mg, 83.0 µmol) in CH2Cl2
(8 mL) was stirred with Ac2O (24.3 µL, 258 µmol) and TiCl4 (77.5
µL, 706 µmol) at room temperature for 1 h. After addition of water
(10 mL), the mixture was extracted with CH2Cl2 (4 ϫ 10 mL). The
combined organic phases were dried (Na2SO4) and concentrated, (M)-4,4Ј,7,7Ј-Tetramethoxy-5,5Ј-dimethyl-6,6Ј-bicoumarin (1): By
and the crude product was purified by column filtration (PE/ethyl the known procedure,[3,28] crude bicoumarin (M)-18 was dissolved
acetate, 1:2). Evaporation of the solvent gave (M)-3,3Ј-diacetyl-
in HMPA (2 mL) and treated with NaH (55Ϫ65%, in mineral oil,
4,4Ј,6,6Ј-tetramethoxy-2,2Ј-dimethyl-1,1Ј-biphenyl (30.2 mg, 78.1 3.23 mg, 80.8 µmol). After this had been stirred at room temper-
µmol, 94%) as a colorless solid, spectroscopically identical with ra-
ature for 15 min, Me2SO4 (9.60 µL, 101 µmol) was added and stir-
cemic material obtained earlier.[28,29] This product was immediately ring was continued at the same temperature for a further 90 min.
[3,28]
dissolved in benzene (8 mL), treated with TiCl4
(68.8 µL, 627 Water (5 mL) was added, and the mixture was acidified with 2 
HCl and extracted with ethyl acetate. The organic phases were
room temperature, water (10 mL) was added carefully. The organic combined and washed several times with brine, and then dried with
µmol), and heated under reflux for 75 min. After this had cooled to
phase was separated, the aqueous component was extracted with
CH2Cl2 (5 ϫ 10 mL), and the combined organic extracts were dried
Na2SO4. The solvent was removed, and the residue was purified in
two portions by preparative TLC (cyclohexane/ethyl acetate, 1:2).
1104
Eur. J. Org. Chem. 2002, 1096Ϫ1106

Atropo-Enantioselective Synthesis of the Natural Bicoumarin (ϩ)-Isokotanin A
FULL PAPER
[7]
G. Bringmann, T. Hartung, O. Kröcher, K.-P. Gulden, J.
Lange, H. Burzlaff, Tetrahedron 1994, 50, 2831Ϫ2840.
G. Bringmann, J. Hinrichs, unpublished results.
F. Roblot, R. Hocquemiller, A. Cave, Bull. Soc. Chim. Fr. 1990,
127, 258Ϫ267.
J. N. Haseltine, M. P. Cabal, N. B. Mantlo, N. Iwasawa, D. S.
Yamashita, R. S. Coleman, S. J. Danishefsky, G. K. Schulte, J.
Am. Chem. Soc. 1991, 113, 3850Ϫ3866.
The products were recovered from the silica gel with ethyl acetate.
The desired final product, (ϩ)-isokotanin A (1), was obtained as a
colorless solid, which upon recrystallization from ethyl acetate
formed colorless crystals (3.90 mg, 8.90 µmol, 26% over two steps),
spectroscopically fully identical with the data reported for the isol-
ated natural product[1] and for material already synthesized.[2,3]
(M)-1: M.p. 286Ϫ289 °C [ref.[1] 223Ϫ226 °C (decomp.), ref.[3]
[8]
[9]
´
[10]
240Ϫ242 °C (decomp.), ref.[2] 285Ϫ290 °C (decomp.)]. [α]2D3
ϭ
[11] [11a] C. Wünsche, A. Sachs, A. Einwiller, W. Mayer, Tetrahedron
ϩ21.6 (c ϭ 0.26, CHCl3) {ref.[1] [α]2D3 ϭ ϩ21.4 (c ϭ 0.22, CHCl3),
[11b]
1968, 24, 3407Ϫ3416.
Taylor, Aust. J. Chem. 1970, 23, 147Ϫ183.
J. M. Insole, J. Chem. Res. (S) 1990, 378Ϫ379.
Insole, J. Chem. Res. (M) 1990, 2831Ϫ2867.
F. Dallacker, H. Leidig, Chem. Ber. 1979, 112, 2672Ϫ2679.
D. D. Ridley, E. Ritchie, W. C.
ref.[3] [α]2D4 ϭ ϩ22.4 (c ϭ 0.30, CHCl3)}. CD (EtOH): ∆ε204
ϭ
ϩ151.5, ∆ε215 ϭ Ϫ182.0, ∆ε226 ϭ ϩ57.6, ∆ε249 ϭ Ϫ3.9, ∆ε297
ϭ
[12] [12a]
[12b]
J. M.
1
Ϫ4.7, ∆ε327 ϭ ϩ11.0. H NMR (400 MHz, CDCl3): δ ϭ 2.23 (s, 6
H, 5- and 5Ј-CH3), 3.72 (s, 6 H, 7- and 7Ј-OCH3), 3.94 (s, 6 H, 4-
and 4Ј-OCH3), 5.59 (s, 2 H, 3- and 3Ј-H), 6.78 (s, 2 H, 8- and 8Ј-
H).[52] 13C NMR (151 MHz, CDCl3): δ ϭ 18.7 (5- and 5Ј-CH3),
55.97, 56.00 (4-, 4Ј-, 7- and 7Ј-OCH3), 87.9 (C-3 and C-3Ј), 97.4
(C-8 and C-8Ј), 108.1 (C-4a and C-4aЈ), 123.4 (C-6 and C-6Ј), 137.2
(C-5 and C-5Ј), 156.3 (C-8a and C-8aЈ), 160.1 (C-7 and C-7Ј), 163.0
(CϭO), 170.1 (C-4 and C-4Ј).[52] In addition, (M)-6-(3Ј-acetyl-
4Ј,6Ј-dimethoxy-2Ј-methylphenyl)-4,7-dimethoxy-5-methylcoumarin
[(M)-19] (colorless crystals from ethyl acetate, 2.69 mg, 6.52 µmol,
19% over two steps) and (M)-3,3Ј-diacetyl-4,4Ј,6,6Ј-tetramethoxy-
2,2Ј-dimethyl-1,1Ј-biphenyl [(M)-20, compare preparation of com-
pound (M)-15] (colorless oil, 1.72 mg, 4.45 µmol, 13% over two
[13]
[14] [14a]
E. P. Boden, G. E. Keck, J. Org. Chem. 1985, 50,
2394Ϫ2395. [14b] K. S. Feldman, S. M. Ensel, R. D. Minard, J.
Am. Chem. Soc. 1994, 116, 1742Ϫ1745.
[15] [15a]
B. Neises, W. Steglich, Angew. Chem. 1978, 90, 556Ϫ557;
Angew. Chem. Int. Ed. Engl. 1978, 17, 522Ϫ524. [15b] B. Neises,
W. Steglich, Org. Synth., Coll. Vol. VII 1990, 93Ϫ95.
[16]
In the crystals of 5 and 9, both enantiomeric forms are found;
for reasons of clarity, only the respective (M)-configured atrop-
isomers are shown. Crystal data for 5: C18H18O6, triclinic,
¯
space group P1; unit cell parameters: a ϭ 702.83(6), b ϭ
895.83(6), c ϭ 1283.36(8) pm; α ϭ 78.685(6), β ϭ 83.700(5),
γ ϭ 83.460(5)°; V ϭ 783.99(8)·106 pm3. Crystal data for 9:
C18H22O6, monoclinic, space group P21/c; unit cell parameters:
a ϭ 835.44(8), b ϭ 2521.5(2), c ϭ 862.99(8) pm; β ϭ
99.864(7)°; V ϭ 1716.4(3)·106 pm3. CCDC-171441 (5), -171442
(9) contain the supplementary crystallographic data for
this paper. These data can be obtained free of charge at
steps) were isolated as by-products. (M)-19: M.p. 198Ϫ199 °C. [α]
23
ϭ ϩ16.2 (c ϭ 0.36, CHCl3). CD (EtOH): ∆ε205 ϭ ϩ18.1,
D
∆ε217 ϭ Ϫ17.2, ∆ε229 ϭ ϩ7.3, ∆ε253 ϭ Ϫ1.4, ∆ε305 ϭ ϩ1.0. IR
(KBr): ν˜ ϭ 3065 cmϪ1 (w, Ar-H), 2975, 2905, 2825 (w, m, m,
CϪH), 1705, 1675 (s, s, CϭO), 1570 (s, CϭC), 1355, 1195, 1080 (s,
s, m). 1H NMR (400 MHz, CDCl3): δ ϭ 1.78 (s, 3 H, 2Ј-CH3), 2.25
(s, 3 H, 5-CH3), 2.52 (s, 3 H, 3Ј-COCH3), 3.71, 3.73 (s, s, je 3 H,
6Ј-OCH3 and 7-OCH3), 3.89 (s, 3 H, 4Ј-OCH3), 3.93 (s, 3 H, 4-
OCH3), 5.57 (s, 1 H, 3-H), 6.42 (s, 1 H, 5Ј-H), 6.76 (s, 1 H, 8-H).
13C NMR (151 MHz, CDCl3): δ ϭ 16.4 (2Ј-CH3), 18.8 (5-CH3),
32.7 (3Ј-COCH3), 55.6, 55.8, 55.9, 56.0 (4-, 4Ј-, 6Ј- and 7-OCH3),
87.8 (C-3), 92.8 (C-5Ј), 97.3 (C-8), 108.0 (C-4a), 118.2 (C-3Ј), 123.7
(C-6), 124.5 (C-1Ј), 136.0 (C-2Ј), 137.4 (C-5), 156.1 (C-8a), 157.2
(C-4Ј), 158.4 (C-6Ј), 160.4 (C-7), 163.2 (C-2), 170.2 (C-4), 205.4 (3Ј-
COCH3). MS (70 eV): m/z (%) ϭ 412 (33) [Mϩ], 398 (32) [413 Ϫ
CH3], 397 (100) [Mϩ Ϫ CH3], 382 (2) [Mϩ Ϫ CH2O], 381 (2) [Mϩ
Ϫ CH3O], 369 (2) [Mϩ Ϫ C2H3O], 367 (8) [397 Ϫ CH2O].
C23H24O7 (412.44): calcd. 412.1522; found 412.1523 (HRMS).
www.ccdc.cam.ac.uk/conts/retrieving.html
or from the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK [Fax: (internat.) ϩ 44-1223/336-033;
E-mail: deposit@ccdc.cam.ac.uk].
[17] [17a]
E. J. Corey, R. K. Bakshi, S. Shibata, J. Am. Chem. Soc.
[17b]
1987, 109, 5551Ϫ5553.
E. J. Corey, R. K. Bakshi, S. Shib-
ata, C.-P. Chen, V. K. Singh, J. Am. Chem. Soc. 1987, 109,
[17c]
7925Ϫ7926.
E. J. Corey, C. J. Helal, Angew. Chem. 1998,
110, 2092Ϫ2118; Angew. Chem. Int. Ed. 1998, 37, 1986Ϫ2012.
[18]
The conversion c and the relative rate constant krel were calcu-
lated according to: [18a] C.-S. Chen, Y. Fujimoto, G. Girdaukas,
[18b]
C. J. Sih, J. Am. Chem. Soc. 1982, 104, 7294Ϫ7299.
Kagan, J. C. Fiaud, Top. Stereochem. 1988, 18, 249Ϫ330.
H. B.
[19] [19a]
G. Bringmann, H. Busse, U. Dauer, S. Güssregen, M.
[19b]
Stahl, Tetrahedron 1995, 51, 3149Ϫ3158.
G. Bringmann,
D. Vitt, J. Kraus, M. Breuning, Tetrahedron 1998, 54,
10691Ϫ10698.
Acknowledgments
[20]
Compound 13 had been prepared in three steps from the cor-
responding enantiomerically pure dicarboxylic acid, but with-
out assignment of the absolute configuration (see ref.[12]).
G. Bringmann, J. R. Jansen, Heterocycles 1989, 28,
137Ϫ142.
This work was supported by the Deutsche Forschungsgemeinschaft
(SFB 347 ‘‘Selektive Reaktionen Metall-aktivierter Moleküle’’) and
by the Fonds der Chemischen Industrie. J. H. thanks the Freistaat
Bayern for a generous fellowship. Skillful experimental assistance
[21] [21a]
[21b]
G. Bringmann, T. Hartung, L. Göbel, O.
Schupp, K. Peters, H. G. von Schnering, Liebigs Ann. Chem.
ˇ
by Jan Valusek is gratefully acknowledged.
[21c]
1992, 769Ϫ775.
G. Bringmann, M. Heubes, M. Breuning,
L. Göbel, M. Ochse, B. Schöner, O. Schupp, J. Org. Chem.
2000, 65, 722Ϫ728.
[22] [22a]
[1]
G. Bringmann, T. Pabst, S. Busemann, K. Peters, E.-M.
J. A. Laakso, E. D. Narske, J. B. Gloer, D. T. Wicklow, P. F.
Dowd, J. Nat. Prod. 1994, 57, 128Ϫ133.
K. Nozawa, S. Nakajima, K.-I. Kawai, S.-I. Udagawa, M. Mi-
[22b]
Peters, Tetrahedron 1998, 54, 1425Ϫ1438.
G. Bringmann,
[2]
T. Pabst, P. Henschel, J. Kraus, K. Peters, E.-M. Peters, D. S.
Rycroft, J. D. Connolly, J. Am. Chem. Soc. 2000, 122,
yaji, Phytochemistry 1994, 35, 1049Ϫ1051.
[22c]
[3]
9127Ϫ9133.
G. Bringmann, J. Hinrichs, T. Pabst, P. Hen-
G.-Q. Lin, M. Zhong, Tetrahedron Lett. 1996, 37, 3015Ϫ3018.
[4]
schel, K. Peters, E.-M. Peters, Synthesis 2001, 155Ϫ167.
Crystal data for (M)-4: C18H22O4, trigonal, space group P3221;
unit cell parameters: a ϭ 875.92(3), c ϭ 1918.3(1) pm; V ϭ
1274.6(1)·106 pm3. CCDC-171443 contains the supplementary
crystallographic data for this paper. These data can be obtained
free of charge at www.ccdc.cam.ac.uk/conts/retrieving.html or
from the Cambridge Crystallographic Data Centre, 12 Union
G. Bringmann, M. Breuning, S. Tasler, Synthesis 1999,
[23]
525Ϫ558.
[5] [5a]
G. Bringmann, J. Hinrichs, Tetrahedron: Asymmetry 1997,
[5b]
8, 4121Ϫ4126.
G. Bringmann, J. Hinrichs, J. Kraus, A.
Wuzik, T. Schulz, J. Org. Chem. 2000, 65, 2517Ϫ2527.
G. Bringmann, R. Walter, C. L. J. Ewers, Synlett 1991,
581Ϫ583.
[6]
Eur. J. Org. Chem. 2002, 1096Ϫ1106
1105

G. Bringmann, J. Hinrichs, P. Henschel, J. Kraus, K. Peters, E.-M. Peters
FULL PAPER
Road, Cambridge CB2 1EZ, UK [Fax: (internat.) ϩ 44-1223/
336-033; E-mail: deposit@ccdc.cam.ac.uk].
Boulder, USA; modified by J. Fleischhauer, W. Schleker, B.
Kramer; ported to LinuX by K.-P. Gulden.
G. Bringmann, J. Kraus, U. Schmitt, C. Puder, A. Zeeck, Eur.
J. Org. Chem. 2000, 2729Ϫ2734.
[24]
[25]
[26]
[39]
[40]
[41]
[42]
[43]
H. Musso, W. Steckelberg, Chem. Ber. 1968, 101, 1510Ϫ1518.
H. Heß, H. Musso, Liebigs Ann. Chem. 1979, 431Ϫ437.
The assignment of the absolute configuration for the two
atropisomers of 14 is based on a stereochemical correlation of
an axially chiral ketone with centrochiral compounds. Starting
from 6,6Ј-dinitro-1,1Ј-biphenyl-2,2Ј-dicarboxylic acid, a series
of rotationally hindered biphenyls were assigned, correlated
J. R. Cannon, T. M. Cresp, B. W. Metcalf, M. V. Sargent, G.
Vinciguerra, J. A. Elix, J. Chem. Soc. (C) 1971, 3495Ϫ3504.
R. C. Fuson, E. A. Cleveland, Org. Synth., Coll. Vol. III
1955, 339Ϫ340.
Compounds 8,[11,12] 9,[12] and 10[13] had already been prepared
before, but without complete characterization data.
The absolute configuration of (ϩ)-9 was not determined in
ref.[12]
either by synthesis or by comparison of their CD or ORD spec-
[26a]
tra:
P. Newman, P. Rutkin, K. Mislow, J. Am. Chem. Soc.
1958, 80, 465Ϫ473. [26b] F. A. McGinn, A. K. Lazarus, M. Sie-
[44] [44a]
gel, J. E. Ricci, K. Mislow, J. Am. Chem. Soc. 1958, 80,
A. A. Frost, R. G. Pearson, Kinetik und Mechanismen
[26c]
476Ϫ480.
K. Mislow, Angew. Chem. 1958, 70, 683Ϫ689.
homogener chemischer Reaktionen, Verlag Chemie, Weinheim,
[26d] H. Musso, W. Steckelberg, Justus Liebigs Ann. Chem. 1966,
1964, pp. 173Ϫ174. [44b]L. Ernst, Chem. Unserer Zeit 1983, 17,
693, 187Ϫ196; see also ref.[24]
21Ϫ30.
P. W. Atkins, Physikalische Chemie, VCH,
[44c]
[27]
[44d]
G. Bringmann, S. Busemann in Natural Product Analysis (Eds.:
P. Schreier, M. Herderich, H. U. Humpf, W. Schwab), Vieweg,
Wiesbaden, 1998, p. 195Ϫ212.
G.-Q. Lin, M. Zhong, Huaxue Xuebao (Acta Chim. Sin.) 1997,
55, 97Ϫ101 [Chem. Abstr. 1997, 126, 211942s].
Weinheim, 1990, pp. 766Ϫ770.
G. Wedler, Lehrbuch der
Physikalischen Chemie, VCH, Weinheim, 1987, pp. 783Ϫ790.
[44e] H. Friebolin, Basic One- and Two-Dimensional NMR Spec-
troscopy, 3rd ed., Wiley-VCH, 1998, pp. 308Ϫ311.
Biaryls (M)-4 and (P)-4 had already been synthesized in enanti-
omerically pure form,[28,29] but without provision of the optical
rotations and melting points for these compounds.
The absolute configuration of (Ϫ)-13 and (ϩ)-13 had not previ-
ously been established.[12]
R. K. Haynes, H. Heß, H. Musso, Chem. Ber. 1974, 107,
3733Ϫ3748.
W. Steckelberg, M. Bloch, H. Musso, Chem. Ber. 1968, 101,
1519Ϫ1521.
[28]
[29]
[45]
G. Büchi, D. H. Klaubert, R. C. Shank, S. M. Weinreb, G. N.
Wogan, J. Org. Chem. 1971, 36, 1143Ϫ1147.
[30] [30a]
[46]
[47]
[48]
[49]
[50]
[51]
M. A. Rizzacasa, M. V. Sargent, J. Chem. Soc., Perkin
Trans. 1 1988, 2425Ϫ2428. [30b] G.-Q. Lin, M. Zhong, Tetrahed-
ron: Asymmetry 1997, 8, 1369Ϫ1372.
[31]
[32]
[33]
E. Suzuki, B. Katsuragawa, S. Inoue, Synthesis 1978, 144Ϫ146.
T. D. Nelson, A. I. Meyers, J. Org. Chem. 1994, 59, 2655Ϫ2658.
M. J. S. Dewar, E. G. Zoebisch, E. Healy, J. J. P. Steward, J.
Am. Chem. Soc. 1985, 107, 3902Ϫ3909.
G. Rauhut, J. Chandrasekhar, A. Alex, B. Beck, W. Sauer, T.
Clark, VAMP 6.5, available from Oxford Molecular Ltd., The
Medawar Centre, Oxford Science Park, Sandford-on-Thames,
Oxford, OX4 4GA, England.
The absolute configurations of (ϩ)-14 and (Ϫ)-14 were not as-
signed in these early studies.[48]
[34]
No assignment of the absolute configuration of (Ϫ)-9 was
made in ref.[12]
Because of the starting thermal racemization, the exact melting
point of enantiomerically pure dimeric orcinol (P)-14 is diffi-
cult to determine; compare ref.[24]
[35]
[36]
SYBYL: Tripos Associates, 1699 St. Hanley Road, Suite 303,
St. Louis, MO, 63144.
[52]
[53]
M. J. D. Powell, Nonlinear Optimization, Academic Press, New
York, 1982.
The assignment of the NMR signals was performed as in
refs.[1,2]
[37]
[38]
´
J. Del Bene, H. H. Jaffe, J. Chem. Phys. 1968, 58, 1807Ϫ1813.
G. Bringmann, S. Schneider, Synthesis 1983, 139Ϫ141.
Received September 27, 2001
J. W. Downing, Program Packet BDZDO/MCDSPD, Depart-
ment of Chemistry and Biochemistry, University of Colorado,
[O01464]
1106
Eur. J. Org. Chem. 2002, 1096Ϫ1106