Chiral recognition by CD-sensitive dimeric zinc porphyrin host. 1. Chiroptical protocol for absolute configurational assignments of monoalcohols and primary monoamines

Article abstract of DOI:10.1021/ja010249w

A general microscale protocol for the determination of absolute configurations of primary amino groups or secondary hydroxyl groups linked to a single stereogenic center is described. The chiral substrates are linked to the achiral trifunctional bidentate carrier molecule (3-aminopropylamino)acetic acid (1, H₂NCH₂CH₂CH₂NHCH₂COOH) and the resultant conjugates are then complexed with dimeric zinc porphyrin host 2 giving rise to 1:1 host/guest sandwiched complexes. These complexes exhibit exciton-coupled bisignate CD spectra due to stereodifferentiation leading to preferred porphyrin helicity. Since the chiral sense of twist between the two porphyrins in the complex is dictated by the stereogenic center of the substrate, the sign of the couplet determines the absolute configuration at this center. The twist of the porphyrin tweezer in the complex can be predicted from the relative steric sizes of the groups flanking the stereogenic center, such that the bulkier group protrudes from the complex sandwich. In certain α-hydroxy esters and α-amino esters, electronic factors and hydrogen bonding govern the preferred conformation of the complex, and hence the CD spectra.

Full text of DOI:10.1021/ja010249w

5962
J. Am. Chem. Soc. 2001, 123, 5962-5973
Chiral Recognition by CD-Sensitive Dimeric Zinc Porphyrin Host. 1.
Chiroptical Protocol for Absolute Configurational Assignments of
Monoalcohols and Primary Monoamines
Tibor Kurta´n, Nasri Nesnas, Yuan-Qiang Li, Xuefei Huang, Koji Nakanishi,* and
Nina Berova*
Contribution from the Department of Chemistry, Columbia UniVersity, New York, New York 10027
ReceiVed January 29, 2001
Abstract: A general microscale protocol for the determination of absolute configurations of primary amino
groups or secondary hydroxyl groups linked to a single stereogenic center is described. The chiral substrates
are linked to the achiral trifunctional bidentate carrier molecule (3-aminopropylamino)acetic acid (1, H2NCH2-
CH2CH2NHCH2COOH) and the resultant conjugates are then complexed with dimeric zinc porphyrin host 2
giving rise to 1:1 host/guest sandwiched complexes. These complexes exhibit exciton-coupled bisignate CD
spectra due to stereodifferentiation leading to preferred porphyrin helicity. Since the chiral sense of twist
between the two porphyrins in the complex is dictated by the stereogenic center of the substrate, the sign of
the couplet determines the absolute configuration at this center. The twist of the porphyrin tweezer in the
complex can be predicted from the relative steric sizes of the groups flanking the stereogenic center, such that
the bulkier group protrudes from the complex sandwich. In certain R-hydroxy esters and R-amino esters,
electronic factors and hydrogen bonding govern the preferred conformation of the complex, and hence the CD
spectra.
Introduction
The configurational assignment of substrates containing
primary amino groups or secondary hydroxyl groups linked to
a single stereogenic center is a challenging task. Since many of
these substrates exhibit important biological activities and often
are available only in limited amounts, determination of their
absolute configurations is of academic as well as of practical
importance. In the following, we have developed a chemical/
chiroptical method that allows the microscale determination of
absolute configuration of such substrates. Namely, the substrate
is derivatized with carrier 1 (Figure 1) to give an ester or amide.
As outlined in Figure 2, treatment of these esters or amides
with a CD sensitive dimeric zinc porphyrin host, referred to as
tweezer 2,1-3 yields complex 3, which exhibits intense exciton-
coupled CD reflecting their absolute configurations.
Figure 1. The three aliphatic diamino carrier molecules for derivati-
zation of chiral monoamines and monoalcohols.
The assignment of absolute configurations of primary mono-
amines and secondary monoalcohols by chiroptical spectroscopy
is mostly based on empirical and semiempirical CD sector and
helicity rules, and thus in contrast to methods based on exciton-
coupled CD, the Cotton effects (CE) are smaller and milligram
quantities of sample are often required.4 Benzene sector rules
* To whom correspondence should be addressed. E-mail:
kn5@columbia.edu and ndb1@columbia.edu.
Figure 2. Complexation of a chiral conjugate of carrier 1 with zinc
porphyrin tweezer 2; the broken line in structure 2 depicts the orientation
of the effective electric transition moment. Upon treatment of tweezer
2 with the conjugate, the primary amino group of the carrier moiety
coordinates with the zinc in P-1, and this is followed by coordination
of the sec-amino group to the zinc in P-2 to yield tweezer complex 3;
the helicity between the two porphyrin rings is dictated by the absolute
configuration of the substrate and affords an exciton-coupled CD (see
Figure 5).
(1) Huang, X.; Rickman, B. H.; Borhan, B.; Berova, N.; Nakanishi, K.
J. Am. Chem. Soc. 1998, 120, 6185-6186.
(2) Huang, X.; Borhan, B.; Berova, N.; Nakanishi, K. J. Indian Chem.
Soc. 1998, 75, 725-728.
(3) Huang, X.; Borhan, B.; Rickman, B. H.; Nakanishi, K.; Berova, N.
Chem. Eur. J. 2000, 6, 216-224.
(4) (a) Snatzke, G.; Kajtar, M.; Snatzke, F. In Optical Rotatory Dispersion
and Circular Dichroism; Ciardelli, F., Salvadori, P., Eds.; Heyden: London,
1973; pp 148-172. (b) Smith, H. E.; Fontana, L. P. J. Org. Chem. 1991,
56, 432-435. (c) Schoenfelder, W.; Snatzke, G. Isr. J. Chem. 1980, 20,
142-149. (d) Harada, N.; Ohashi, M.; Nakanishi, K. J. Am. Chem. Soc.
1968, 90, 7349-7351. (e) Stanchev, S.; Rakovska, R.; Berova, N.; Snatzke,
G. Tetrahedron: Asymmetry 1995, 6, 183-198.
have been forwarded for benzyl carbinols and aryl carbinamines5
while the induced CE of chiral monoamines were used in their
10.1021/ja010249w CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/02/2001

Configuration of Monoalcohols and Primary Monoamines
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 5963
acid-base complexes with benzoylbenzoic acid6 and poly((4-
carboxyphenyl)acetylene).7 The stereochemistry of secondary
monoalcohols has been investigated by induced CD of their
complexes with Rh2(OCOCF3)4,8 copper hexafluoroacetyl-
acetonate,9 and the optical rotation and CD of their 2,4-
dinitrobenzenesulfenyl derivatives.10
if both derivatives of the chiral amine or alcohol have to be
prepared and measured.
In the following we describe a microscale method for
determining the absolute configurations of secondary mono-
alcohols and primary monoamines in which the hydroxyl and
amino groups are linked to the stereogenic center. It is an
extension of the conventional exciton chirality method. Since
the presence of two or more chirally oriented chromophores is
a prerequisite for the exciton chirality method,17,18 it cannot be
applied directly to compounds in which the hydroxyl or amino
group is the only site available for the introduction of a
chromophore. Application to such compounds requires either a
chromophore already present in the substrate, such as R-aryl-
substituted alcohols19 or allylic alcohols,20 or employment of a
host molecule as a sensitive bichromophoric CD reporter
group.3,21
In recent years porphyrins and zinc porphyrins have attracted
widespread attention as reporter groups with multifaceted
properties for structural studies by CD spectroscopy.22 The
chromophore 5-(carboxyphenyl)-10,15,20-triphenylporphyrin22
has an intense red-shifted Soret band at ca. 414 nm and
propensity to undergo intramolecular π-π stacking in a
stereocontrolled manner. This renders it a versatile and powerful
reporter group in configurational assignments of various sub-
strates with a single stereogenic center, e.g. acyclic diamines
and amino alcohols,23 diols,24,25 and R-hydroxy acids.26 In
another protocol, two zinc tetraphenylporphyrin residues were
linked by a pentanediol spacer.1 The resulting zinc porphyrin
tweezer 2 (Figure 2), an achiral CD reporter “receptor”, was
capable of binding various chiral acyclic R,ω-diamines through
zinc-amine coordination that resulted in formation of 1:1
macrocyclic host/guest complexes.1 The stereoselective com-
plexation gave rise to one clearly preferred porphyrin helicity,
represented by an exciton-coupled CD in the porphyrin spectral
region with signs controlled by the absolute configuration of
the bound diamine. It was subsequently found that this zinc
porphyrin tweezer 2 also forms complexes with conjugate 7a
(Scheme 1) prepared from primary monoamines 4a and carrier
5 (the N-Boc protected form of the trifunctional bidentate carrier3
molecule 6). In such cases, the preferred porphyrin helicity of
the resultant complex 8a was also dictated by the relative steric
size of substituents attached to the stereogenic center. The
stereodifferentiation between the large and medium groups, R1
The most widely used method for configurational assignments
of both secondary monoalcohols11,12 and monoamines13 is the
Mosher NMR method and its modified versions which are based
on the ring current effect of the introduced aryl moiety. The
original protocol requires derivatization of the chiral alcohol
or amine with both (R)- and (S)-enantiomers of a chiral aromatic
acid, such as R-methoxy-R-(trifluoromethyl)phenylacetic acid
(MTPA), R-methoxyphenylacetic acid (MPA), or other auxiliary
reagent, and measurement of the ∆δS,R values for the protons
flanking the alcoholic and amino functions. The Mosher NMR
method has also found some of its most sensitive applications
in using 19F- and 13C-NMR spectroscopy.11 Recently, this
method has undergone considerable progress in terms of
introducing new auxiliary reagents14 and novel procedures15 that
require the preparation of only one of the two chiral esters or
amides. Despite such improvements there are cases12,16 where
the Mosher method cannot be applied with certainty due to either
the lack of protons on one side of the molecule or the small
∆δS,R values that approach the limits of experimental error. The
required amount of sample is also a restriction in Mosher’s
method since milligram quantities are usually needed, especially
(5) (a) Smith, H. E.; Neergaard, J. R. J. Am. Chem. Soc. 1997, 119, 116-
124. (b) Smith, H. E. In Circular Dichroism, Principles and Applications,
2nd ed.; Berova, N., Nakanishi, K., Woody, R. W., Eds.; Wiley-VCH: New
York, 2000; pp 397-430.
(6) (a) Takenaka, S.; Miyauchi, Y.; Tokura, N. Tetrahedron Lett. 1976,
3811-3814. (b) Takenaka, S.; Ako, M.; Kotani, T.; Matsubara, A.; Tokura,
N. J. Chem. Soc., Perkin Trans. 2 1978, 95-99. (c) Takenaka, S.; Kondo,
K.; Tokura, N. J. Chem. Soc., Perkin Trans. 2 1975, 1520-1524.
(7) (a) Yashima, E.; Matsushima, T.; Okamoto, Y. J. Am. Chem. Soc.
1997, 119, 6345-6359. (b) Yashima, E.; Matsushima, T.; Okamoto, Y. J.
Am. Chem. Soc. 1995, 117, 11596-11597.
(8) (a) Gerards, M.; Snatzke, G. Tetrahedron: Asymmetry 1990, 1, 221-
236. (b) Frelek, J.; Szczepek, W. J. Tetrahedron: Asymmetry 1999, 10,
1507-1520.
(9) Dillon, J.; Nakanishi, K. J. Am. Chem. Soc. 1974, 96, 4055-4057.
(10) Craine, L. E.; Bicknell, L. K.; Mailloux, R. C.; Mark, J. P.; Mitchell,
S. A.; Vicente, S. R.; Jasinski, J. P.; Woudenberg, R. C. J. Org. Chem.
1993, 58, 1251-1258.
(11) (a) Trost, B. M.; Belletire, J. L.; Godleski, S.; McDougal, P. G.;
Balkovec, J. M.; Baldwin, J. J.; Christy, M. E.; Ponticello, G. S.; Varga, S.
L.; Springer, J. P. J. Org. Chem. 1986, 51, 2370-2374. (b) Dale, J. A.;
Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512-519. (c) Pehk, T.; Lippmaa,
E.; Lopp, M.; Paju, A.; Borer, B. C.; Taylor, R. J. K. Tetrahedron:
Asymmetry 1993, 4, 1527-1532. (d) Latypov, S. K.; Seco, J. M.; Quinoa,
E.; Riguera, R. J. Org. Chem. 1996, 61, 8569-8577. (e) Dale, J. A.; Dull,
D. L.; Mosher, H. S. J. Org. Chem. 1969, 34, 2543-2549
(12) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092-4096.
(13) (a) Trost, B. M.; Bunt, R. C.; Pulley, S. R. J. Org. Chem. 1994, 59,
4202-4205. (b) Seco, J. M.; Quinoa, E.; Riguera, R. J. Org. Chem. 1999,
64, 4, 4669-4675. (c) Seco, J. M.; Latypov, S. K.; Quinoa, E.; Riguera, R.
J. Org. Chem. 1997, 62, 7569-7574. (d) Latypov, S. K.; Seco, J. M.;
Quinoa, E.; Riguera, R. J. Org. Chem. 1995, 60, 1538-1545.
(14) (a) Ferreiro, M. J.; Latypov, S. K.; Quinoa, E.; Riguera, R.
Tetrahedron: Asymmetry 1996, 7, 2195-2198. (b) Parve, O.; Aidnik, M.;
Lille, U.; Martin, I.; Vallikivi, I.; Vares, L.; Pehk, T. Tetrahedron:
Asymmetry 1998, 9, 885-896. (c) Harada, N.; Watanabe, M.; Kuwahara,
S.; Sugio, A.; Kasai, Y.; Ichikawa, A. Tetrahedron: Asymmetry 2000, 11,
2843. (d) Fukushi, Y.; Yajima, C.; Mizutani, J. Tetrahedron Lett. 1994,
35, 599-602.
(17) Berova, N.; Nakanishi, K. In Circular Dichroism, Principles and
Applications, 2nd ed.; Berova, N., Nakanishi, K., Woody, R. W., Eds.;
Wiley-VCH: New York, 2000; pp 337-382.
(18) Harada, N.; Nakanishi, K. In Circular Dichroic Spectroscopy Exciton
Coupling in Organic Stereochemistry; University Science Books: Mill
Valley, CA, 1983.
(19) (a) Adam, W.; Lukacs, Z.; Viebach, K.; Humpf, H.-U.; Saha-
Moeller, C. R.; Schreier, P. J. Org. Chem. 2000, 65, 186-190. (b) Pini,
D.; Petri, A.; Rosini, C.; Salvadori, P.; Giorgi, R.; Di Bugno, C.; Turbanti,
L.; Marchetti, F. Tetrahedron 1994, 50, 205-216.
(20) (a) Harada, N.; Iwabuchi, J.; Yokota, Y.; Uda, H.; Nakanishi, K. J.
Am. Chem. Soc. 1981, 103, 5590-5591. (b) Johnson, R. A.; Krueger, W.
C.; Nidy, E. G.; Pschigoda, L. M.; Garry, M. J. J. Org. Chem. 1980, 45,
1528-1532. (c) Gonnella, N. C.; Nakanishi, K.; Martin, V. S.; Sharpless,
K. B. J. Am. Chem. Soc. 1982, 104, 3775-3776. (d) Humpf, H.-U.; Berova,
N.; Nakanishi, K.; Jarstfer, M. B.; Poulter, C. D. J. Org. Chem. 1995, 60,
3539-3542.
(21) Kato, N. J. Am. Chem. Soc. 1990, 112, 254-257.
(22) Huang, X.; Nakanishi, K.; Berova, N. Chirality 2000, 12, 237-
255.
(15) (a) Lopez, B.; Quinoa, E.; Riguera, R. J. Am. Chem. Soc. 1999,
121, 9724-9725. (b) Seco, J. M.; Quinoa, E.; Riguera, R. Tetrahedron
1999, 55, 569-584. (c) Latypov, S. K.; Seco, J. M.; Quinoa, E.; Riguera,
R. J. Am. Chem. Soc. 1998, 120, 877-882.
(16) (a) Seco, J. M.; Quinoa, E.; Riguera, R. Tetrahedron: Asymmetry
2000, 11, 2781-2791. (b) Earle, M. A.; Hultin, P. G. Tetrahedron Lett.
2000, 41, 7855-7858.
(23) Matile, S.; Berova, N.; Nakanishi, K. Enantiomer 1996, 1, 1-12.
(24) Jiang, H.; Huang, X.; Nakanishi, K.; Berova, N. Tetrahedron Lett.
1999, 40, 7645-7649.
(25) Matile, S.; Berova, N.; Nakanishi, K.; Fleischhauer, J.; Woody, R.
W. J. Am. Chem. Soc. 1996, 118, 5198-5206.
(26) Rickman, B. H.; Matile, S.; Nakanishi, K.; Berova, N. Tetrahedron
1998, 54, 5041-5064.

5964 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
Kurta´n et al.
Scheme 1. Formation of Complex 8 from Tweezer 2 and
Conjugates 7 Prepared from Chiral Monoamines 4a, Mono
Alcohols 4b, and Carrier 6 (the Boc protected carrier 5 is
employed in the synthesis)
Figure 3. UV-VIS and CD spectra of conjugate 10-23 complexed
with porphyrin tweezer 2 in methylcyclohexane at room temperature
(conjugate 10-23 is the definition for the ester formed between carrier
10 and chiral alcohol substrate 23; see Table 1).
the intramolecular hydrogen bond between the amino group and
the carbonyl oxygen was also considered to make its conforma-
tion less flexible. The complexes between tweezer 2 and pyridine
oxide alcohol conjugates 11 indeed gave exciton-coupled CD
spectra in nonpolar solvents but their signs could not be
correlated unambiguously with the absolute configuration of the
alcohols. The complexes from tweezer 2 and conjugates 12 also
exhibited distinctive CD spectra in methylcyclohexane and
hexane; however, usually three Cotton effects were observed
(see Figure 3). Such multiple Cotton effects are most likely due
to overlapping CD couplets of various conformers. In addition,
their intensities were relatively small and did not increase when
the substrate substituents at the stereogenic center were made
bulkier. The use of carriers 9 and 10 was thus abandoned.
Most likely, in the alcohol conjugates 7b (Scheme 1), 11 and
12 (Scheme 2), the aromatic moieties restrict the conformational
freedom of the complexing nitrogens. This suggests that when
complexation occurs between the carrier nitrogens and the zinc
porphyrin, conformation of the amide and ester moieties might
be different before and after binding so that the complexes do
not reflect the subtle differences stemming from the substituents
surrounding the stereogenic center. On the other hand, in the
case of conjugates formed between monoamines 4a and carrier
6 (Scheme 1), the rigid amide bond gave rise to a major
conformer, so that signs of the CD couplets could be correlated
with the absolute configurations.3 Obviously this is not the case
with conjugates formed between alcohols and the three carriers
6, 9, and 10. The more flexible ester bonds do not impart
sufficient conformational rigidity; hence overlapping CD cou-
plets arising from multiple conformers render the interpretations
difficult or impossible.
Scheme 2. Unsuited Carriers and Their Alcohol Conjugates
(see Supporting Information for synthesis)
and R2, respectively, led to a preferred chiral sense of twist
between the two interacting porphyrins that gave rise to a
positive or negative exciton couplet reflecting the relative steric
bulk. This procedure is applicable to both acyclic and cyclic
aromatic amines, aliphatic amines, and amino esters and can
be performed at the low microgram level.3 However, the method
cannot distinguish between hydrogen and deuterium since the
stereoselectivity is based on relative steric size and not relative
mass.
Extension of this method to secondary monoalcohols 4b
would be intriguing and desirable, since it could become a
general microscale method where its application would not
require any preexisting CD chromophore in the hydroxyl-
containing substrate. The conjugates of secondary monoalcohols
4b with this carrier molecule 6 were first studied to check the
applicability. The UV-VIS spectra of zinc porphyrin tweezer
complexes with the monoalcohol conjugates 7b were almost
identical with those prepared from monoamine conjugates 7a
(Scheme 1). However, none of these complexes 8b gave
distinctive CD couplets in the solvents tested, most likely due
to the lack of conformational rigidity of esters compared to
amides. This led to the design of two further carriers 9 and 10
(Scheme 2). In the pyridine oxide conjugates 11, the increased
electronic repulsion between the negatively charged oxygen of
the pyridine oxide and the carbonyl oxygen was supposed to
reduce the conformational flexibility of the alcohol conjugates.
Similarly, in the case of the p-phenylenediamine conjugate 12,
Results and Discussion
Selection of Carrier 1 (Figure 1). The search for a different
type of carrier molecule has highlighted some important
attributes to be considered. First, it should be an achiral molecule
containing two amino nitrogens with sufficient Lewis basicity
to bind to the tweezer zinc atoms and an additional carboxyl
group to couple to alcohols or amines. Second, when linked to
the substrate (monoalcohol or amine) the carrier molecule should
lead to a bidentate conjugate where the stereogenic center is in
close proximity to the nitrogen of the secondary amine, which
upon coordination to zinc adopts a chirality controlled by the

Configuration of Monoalcohols and Primary Monoamines
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 5965
Scheme 3. Synthesis of diBoc Protected Carrier Molecule
17a
Figure 4. CD spectra of conjugates 1-25, 1a-25, and 1b-25 prepared
from (+)-isomenthol 25 and carriers 1, 1a, and 1b (Figure 1) with
varying chain lengths, in methylcyclohexane, respectively; carrier 1
gives the highest amplitude.
a Methods A and B: Derivatization of chiral alcohols/amines with
(3-aminopropylamino)acetic acid.
no protecting groups to yield the product as the free amine,
ready for complexation with the porphyrin tweezer. The chiral
alcohol or amine 4 is first reacted with bromoacetic acid in the
presence of EDC/DMAP, and then the resultant bromoester 20
is treated with an excess of 1,3-diaminopropane (13) to afford
the free amine of conjugate 19.
While method A gives conjugates in excellent yields and is
applicable to most chiral alcohols, it is not suited for some
benzylic alcohols where deprotection by TFA can also cleave
the ester bond. Method B is thus preferred for benzylic
monoalcohols. In the case of chiral amines, both methods can
be used to yield amine conjugates in quantitative yields. Thirty
equivalents of the resulting monoalcohol or monoamine con-
jugate was added to ca. 1 µM solution of porphyrin tweezer 2
and UV/CD were measured in dichloromethane and methylcy-
clohexane. It was feasible to form the conjugate with as little
as 50 µg of chiral substrate followed by the use of a pipet
column for purification prior to UV/CD measurements (see
Experimental Section).
Helicity of the Bisporphyrin Complex. Figure 5 depicts
conjugate 1-30 and its CD spectrum with porphyrin tweezer 2
in methylcyclohexane. In this conjugate, the methine hydrogen
of the ester is syn coplanar with the carbonyl oxygen (see also
Figure 2, complex 3)17,28 and this in turn defines the orientation
of the naphthyl and methyl groups. It was found that the addition
of conjugate 1-30 to achiral porphyrin tweezer 2 leads to the
formation of a 1:1 host-guest tweezer complex (see the
following paper) via coordination of amine nitrogens in 1-30
to the zinc atoms of porphyrins P-1 and P-2 in tweezer 2.29
The zinc in porphyrin P-1 coordinates with the terminal primary
amino group. The P-2, however, approaches the secondary
amino group from the side of the less bulky group (M) (see
also Figure 2) giving rise to a stable configuration of the
nitrogen, which is governed by the chirality of the stereogenic
center in the substrate. This stereoselection leads to the most
stable conformer I (Figure 5) in which the “less bulky group”
(M) at the stereogenic center is sandwiched between P-1 and
P-2 while the “bulkier group” (L) is pointing out and away from
the P-2 porphyrin ring (Figure 2). In this preferred conformation
absolute configuration of the stereogenic center in the substrate.
It was also expected that if the carrier molecule contains a
flexible diaminoalkyl chain, this would allow the conjugate to
be sandwiched between the host tweezer without noticeable
conformational distortion in the substrate. Moreover, an optimal
fit between the pentanediol bridge and the two porphyrin rings
in tweezer 2 should yield intense CD amplitudes.27 The CD
amplitude is taken as the difference in ∆ꢀ of the split extrema;
the sign is defined positive when the first Cotton effect (longer
wavelength) is positive and vice versa (see Table 1).
Such considerations led to the design of (3-aminopropyl-
amino)acetic acid (1), (2-aminoethylamino)acetic acid (1a), and
(4-aminobutylamino)acetic acid (1b) carrier molecules (Figure
1). The conjugates of (1S,2R,5R)-(+)-isomenthol (25, Table 1)
with these three carriers, 1-25, 1a-25, and 1b-25, were prepared
and their CD spectra were measured after complexation to
tweezer 2 (Figure 4). While all three complexes gave desired
bisignate curves, that of conjugate 1-25 provided the highest
amplitude (+159). The addition or removal of a methylene
group in 1, as in 1a-25 and 1b-25, resulted in smaller CD
amplitudes, +36 and +38, respectively, demonstrating that the
trimethylene chain in 1 represents the optimal distance between
the two nitrogens complexing with the zinc atoms of tweezer
2.
Conjugate Synthesis and Complexation with Tweezer 2.
The formation of a bidentate conjugate between carrier 1 and a
chiral monoalcohol or monoamine prior to complexation with
the zinc porphyrin tweezer 2 was performed by two independent
routes (Scheme 3). Method A involved the use of the diBoc
protected carrier molecule 17, which was synthesized from 1,3-
diaminopropane (13) in four steps, via intermediates 14, 15,
and 16 in 85% overall yield. Chiral secondary monoalcohols
or primary monoamines 4 were then directly linked to the diBoc
protected carrier molecule 17 in the presence of 1-(3-dimethyl-
aminopropyl)-3-ethylcarbodiimide (EDC) and 4-(dimethyl-
amino)pyridine (DMAP). This was followed by removal of the
Boc protecting groups with 20% TFA in dichloromethane to
afford the TFA salt of conjugate 19. The latter was converted
to its free amine with Na2CO3 before complexation to tweezer
2. In method B, conjugate 19 was prepared in two steps with
(28) Jeffrey, G. A.; Sundaralingam, M. In AdVances in Carbohydrate
Chemistry and Biochemistry; Tipson, R. S., Horton, D., Eds.; Academic
Press: Orlando, FL, 1985; Vol. 43, pp 204-272.
(29) In Tweezer 2, P-1 and P-2 are indistinguishable; however, in the
complex of 2 and conjugate 1-30, P-1 refers to the porphyrin close to the
terminal amine, and P-2 refers to the porphyrin close the secondary amine.
(27) (a) Hayashi, T.; Nonoguchi, M.; Arya, T.; Ogoshi, H. Tetrahedron
Lett. 1997, 38, 1603-1606. (b) Schneider, H.-J.; Wang, M. J. Org. Chem.
1994, 59, 7464-7472. (c) Crossley, M. J.; Mackay, L. G.; Try, A. C. J.
Chem. Soc., Chem. Commun. 1995, 1925-1927.

5966 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
Kurta´n et al.
Table 1. Structures and Schematic Representations of Chiral Substrates (alcohols and amines) and CD Data of Their Conjugates with Carrier
1 after Complexation with Tweezer 2a
a The predicted CD couplet is based on relative steric sizes or conformational A values of substituents. For synthetic Methods A and B see
Scheme 3. b Enantiomer of (S)-30 exhibited an enantiomeric CD spectrum.

Configuration of Monoalcohols and Primary Monoamines
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 5967
Figure 6. CD titration curves of porphyrin tweezer 2 with 0.5-300
equiv of conjugate 1-26 in hexane.
the P-2 porphyrin of tweezer 2 coordinates to the secondary
amine nitrogen from the direction of the M or the L group
(Figures 2 and 5). The preferred conformer determines the sign
of the CD couplet and hence the observed CD exciton chirality
reveals the absolute configuration of the substrate.
Figure 5. Complex formation between conjugate 1-30 and tweezer 2
leads to two conceivable conformations with opposite sense of twists.
The predominant conformation I, the one in which the L (larger) group
is protruding away from the P-1/P-2 sandwich, gives rise to a typical
CD couplet representing the sense of twist between the two porphyrins
and hence the absolute configuration at the stereogenic center; the
dashed curve in the UV represents tweezer 2 prior to complexation.
Spectra were measured in methylcyclohexane.
The assignment of M and L to substituents at the stereogenic
center is thus critical for predicting the sign of the CD couplet.
The sign and intensity of the CD couplet depend on the degree
of stereoselection between the conformers with opposing
porphyrin helicity (Figure 5). While steric considerations play
an important role in conformational analysis, it should be
emphasized that factors other than steric bulk, such as electronic
interactions, hydrogen bonding, and rotational degrees of
freedom, cannot be ignored. The conformer that predominates
is a result of these interrelated factors that often sway in opposite
directions. We therefore performed a systematic study of the
factors that may be encountered and their effects on the
predominant conformation aided by Molecular Modeling cal-
culations.30 To clarify the concept and applicability of the
method, the chiral substrates investigated were classified into
two general categories: namely, L/M assignments based on the
relative steric bulk of substituents derived from A values (Table
1) and L/M assignments based on the overriding electronic
factors studied by molecular modeling (Table 2).
A. L/M Based on Relative Steric Bulk. In chiral alcohols
21 to 37 and chiral amines 38 to 43 presented in Table 1, the
hydroxyl and amino functions are attached to stereogenic centers
carrying aliphatic and/or aromatic substituents. In such sub-
strates, the L and M substituents can be distinguished on the
basis of conformational energy differences for the cyclohexane
model, or so-called A values (kcal/mol) defined by Winstein
and Holness,31 and later described by Eliel32 and Bushweller,33
and more recently by Lightner.34 Even in cases where the
difference between the two R groups is subtle, such as in alcohol
22 where R groups are methyl (A value of 1.74) and ethyl (A
value of 1.79), the tweezer 2 complex of conjugate 1-22 reflects
I, supported by NMR ROESY and anisotropic porphyrin ring
current effects on M and L groups (following paper), porphyrins
P-1 and P-2 adopt a predicted positive helicity resulting from
the clockwise arrangement of their effective electric transition
moments which run alongside the C-5-C-15 axis25 of the zinc
tetraphenylporphyrin (Figure 2). Thus the overriding positive
twist between the transition moments in conformer I leads to a
positive CD couplet, which is in full agreement with the
observed positive CD couplet with ACD ) +170.
While measurements of CD spectra of conjugate 1-30 after
complexation in dichloromethane, hexane, chloroform, aceto-
nitrile, and toluene resulted in varying amplitudes, they all gave
the same CD sign, i.e., a positive exciton couplet. Since hexane,
methylcyclohexane, and dichloromethane gave the highest
amplitudes, all other tweezer complexes were measured in these
solvents. The CD data were generally recorded in 1 µM solutions
of tweezer 2 with 30 equiv of the conjugates. A titration of
tweezer 2 with 0.5 to 300 equiv of conjugate 1-26 is shown in
Figure 6. A substantial increase in the CD amplitude is observed
up to 5 equiv of the conjugate, the slight increase being
continued up to 30 equiv (Figure 6); half of the maximal
amplitude was attained even with 0.5 equiv of conjugate 1-26.
Above 40 equiv, the CD amplitude gradually decreases most
likely due to the formation of a 1:2 host-guest complex in
which the chiral twisted conformation of the tweezer is
disrupted.3
Analysis of Representative Chiral Substrates. The CD data
for monoalcohols and monoamines with previously established
absolute configurations, in both methylcyclohexane and dichlo-
romethane, are given in Tables 1 and 2. The schematic
representations depict the substituents attached to the stereogenic
center that leads to the prediction of the CD couplet sign. If the
hydroxyl or amino group is placed in the rear, as shown, the
clockwise arrangement of L, M, and hydrogen results in a
positive CD couplet, and vice versa. The absolute sense of twist
between the porphyrin electric transition moments and hence
the sign of the exciton-coupled CD is dependent on whether
(30) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.;
Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J. Comput.
Chem. 1990, 11, 440-467.
(31) Winstein, S.; Holness, N. J. J. Am. Chem. Soc. 1955, 77, 5562-
5578.
(32) Eliel, E. L.; Wilen, S. H. In Stereochemistry of Organic Compounds;
Wiley: New York, 1994; pp 695-697.
(33) Bushweller, C. H. In Conformational BehaVior of Six-Membered
Rings. Analysis, Dynamics, and Stereoelectronic Effects; Juaristi, E., Ed.;
VCH Publishers: New York, 1995; pp 25-58.
(34) Boiadjiev, S. E.; Lightner, D. A. J. Am. Chem. Soc. 2000, 122,
11328-11339.

5968 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
Kurta´n et al.
Table 2. Structures and Schematic Representations of Chiral Substrates (alcohols and amines) and CD Data of Their Conjugates with Carrier
1 after Complexation with Tweezer 2a
a The predicted CD couplet is based on electronic factors, i.e., mainly H-bonding and lone pair electron repulsion, which exerts an apparent large
and medium effect on the substituent. For synthetic Methods A and B see Scheme 3.
this difference in accordance with A values. Replacement of
the ethyl with an isopropyl, as in 23, or a pentyl group, as in
21, consistently resulted in an enhanced stereodifferentiation
and hence larger CD amplitudes of the same sign. In 1-24 the
sign of the CD couplet is opposite in agreement with the M
and L sequence.
the conformational energy A values. The CD amplitudes in 31,
32, and 34 also reflect the bulk of the aromatic group. Alcohol
36 is, in a sense, comparable to alcohol 34 with a carbomethoxyl
in place of a methyl and hence results in an identical sign with
comparable amplitude. However, the introduction of an ad-
ditional R-methyl group, as in alcohol 37, results in two groups
of comparable steric size and the tweezer complex gives
opposite signs in methylcyclohexane and dichloromethane; thus
alcohol 37 clearly represents a borderline case.
The tweezer 2 complexes of the conjugates prepared from
amines 38 to 43 with carrier 1 gave exciton-coupled CD with
identical signs and comparable amplitudes to the ones measured
as conjugates of carrier 6 (Scheme 1).3 Amines 38 to 40 are
analogous to alcohols 21 to 26 in which the substituents are
simply aliphatic groups, and hence follow the same trend in M
and L assignments. The opposite configuration of amines 38
and 39 is reflected in their opposite CD signs, while the
amplitudes reflect the size difference between isopropyl and
cyclohexyl groups. The CD couplet sign of menthylamine 40
is the same as its alcohol analogue 26, and the amplitudes are
also comparable. Amine 41 is comparable to its alcohol analogue
32, possessing an identical CD sign but a larger amplitude, most
likely due to the increased rigidity of the amide in its conjugate
compared to the ester in the alcohol conjugate of 32. In amine
43 the p-chlorophenyl group is larger than the benzofuran ring.
Thus for aliphatic/benzylic alcohols and amines, the steric
bulk of the M and L groups, predictable based on cyclohexane
Monoalcohols with multiple stereogenic centers (25-29) also
showed CD signs consistent with the absolute configuration of
the carbon bearing the hydroxyl group. In diastereomers
isomenthol 25 and menthol 26 with CD couplets of opposite
signs, the substantially smaller amplitude of isomenthol 25
derives from the axial orientation of the C-5 methyl group which
adds to the bulk of the M side, resulting in decreased
stereorecognition. The conjugates 1-26 and 1-27 gave intense
CD couplets where the porphyrin host had to differentiate
between the isopropyl (A value of 2.21) vs hydrogen, in 1-26,
and the phenyl (A value of 2.8) vs hydrogen, in 1-27, flanking
the secondary OH group. In alcohol 28, a saturated methylene
at the R position and an acetoxyl group at the â position are
larger than the double bond. In alcohol 29, a quaternary carbon
is assigned L while the methylene is assigned M.
Due to the red-shifted Soret band of tweezer 2 at ca. 420
nm, this method is also applicable to substrates which inherently
contain a chromophore, e.g. a phenyl or a naphthyl, that absorb
below 300 nm, thus interference of such shorter wavelength
transitions with the porphyrin is excluded. In benzylic alcohols
(30-37) the aromatic rings are assigned L in accordance with

Configuration of Monoalcohols and Primary Monoamines
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 5969
Figure 7. (a) The complex formed from tweezer 2 and conjugate 1-44. The stick model represents the optimal conformation (obtained from an
MM2 Monte Carlo conformational search) of partial conjugate 1-44 lacking the H2NCH2CH2- tail before complexation with tweezer 2; the coplanarity
of the phenyl and carbonyl groups leads to the M* assignment for the phenyl ring. (b) The complex formed from tweezer 2 and conjugate (R)-1-50.
The stick model represents the optimal conformation of partial conjugate 1-50 (without the H2NCH2CH2- tail) before complexation with tweezer
2; due to the π-H-bonding between the amide N-H hydrogen and the aromatic ring, the phenyl adopts a conformation perpendicular to the carbonyl
group, thus making it L*.
conformational A values,32,33 appears to be the main factor in
determining their recognition and hence the predicted CD
couplet.
carrier’s carbonyl and the benzylic methine proton, which in
other cases is usually syn-coplanar, is found to be +51°. The
other conformers obtained from the conformational search that
predominantly gave structures with phenyl protruding from the
side were at least 0.63 kcal/mol higher in energy. This
corresponds to a stereodifferentiation of the preferred conformer
by a factor of 3:1 (data not shown). Therefore in 44 and 45 the
methylenes flanking the stereogenic center have a larger
apparent size, hence L*, compared to the phenyl moiety which
lies more or less parallel to P-2. The introduction of a methylene
group into the aliphatic ring in 44, i.e. 45, results in a larger
CD amplitude as the aliphatic ring size is further increased.
In the amine analogues of 44 and 45, i.e., 50 and 51, the
presence of N-H, a hydrogen bond donor, favors a conforma-
tion where the substituents adopt an orientation leading to
different M*/L* assignments. An MM2 Monte Carlo confor-
mational search of the partial conjugate 1-50 (lacking CH2CH2-
NH2) revealed that in its most stable conformer, the N-H proton
of the amide is partially overlapping with the aromatic ring,
probably due to π-hydrogen bonding (Figure 7b). This type of
hydrogen bonding has been observed in other similar cases35
where the hydrogen bond donor is also two atoms away from
the aromatic moiety. The optimal projection angle between the
carrier’s carbonyl and the benzylic methine proton was calcu-
lated to be +29° (see Figure 7b). This angle difference (+51°
in analogue 44), along with the difference in the conformation
of the cyclopentene ring, renders the phenyl group to protrude
to the side almost perpendicular to the carrier’s carbonyl; this
is contrary to analogue 44 (Figure 7a). Thus the P-2 porphyrin
B. L/M Based on Electronic Factors. The chiral alcohols
44 to 49 and chiral amines 50 to 54 presented in Table 2
represent cases in which electronic factors override the effect
of relative steric bulk. Unlike alcohols, the conjugates of amines
contain an amide NH capable of H-bonding, and when such
bonds are present, this could lead to opposite L/M assignments
for alcohols and amines as shown in Table 2. The L and M
assignments in these cases do not directly reflect the steric bulk,
i.e., cannot be derived from conformational A values, but instead
refer to substituents that behave as if they are larger (or smaller)
due their electronic effect on the overall conformation. Therefore
we employ the use of an asterisk as in M* and L*, where M*
refers to the group that is pointing toward P-2, while L* refers
to the group that is pointing away from P-2, due to factors other
than steric bulk.
In contrast to alcohol 32 and amine 41, in bicyclic benzylic
alcohols 44/45 and amines 50/51, the A values of substituents
do not reflect their actual steric bulk because rotation of the
phenyl moiety is restricted. The overall steric demand of such
substituents was clarified through molecular modeling of the
conjugate prior to complexation. An MM2 Monte Carlo
conformational search on MacroModel30 7.0 was performed with
a “partial” conjugate 1-44 lacking the flexible CH2CH2NH2
chain (Figure 7). The truncated conjugate was used to reduce
the number of potential conformers to those of interest, mainly
focusing on the orientation of the carrier’s carbonyl with respect
to the methine proton. The most stable conformation of partial
conjugate 1-44 is shown in Figure 7a. It is clear that the phenyl
ring is almost in the plane of the carrier carbonyl and hence
has less steric demand on P-2. In this case, the ester group adopts
a conformation in which the projection angle between the
(35) (a) Crisma, M.; Formaggio, F.; Valle, G.; Toniolo, C.; Saviano, M.;
Iacovino, R.; Zaccaro, L.; Benedetti, E. Biopolymers 1997, 42, 1-6. (b)
Jimenez, A. I.; Cativiela, C.; Gomez-Catalan, J.; Perez, J. J.; Aubry, A.;
Paris, M.; Marraud, M. J. Am. Chem. Soc. 2000, 122, 5811-5821. (c)
Adams, H.; Harris, K. D. M.; Hembury, G. A.; Hunter, C. A.; Livingstone,
D.; McCabe, J. F. Chem. Commun. (Cambridge) 1996, 2531-2532.

5970 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
Kurta´n et al.
Figure 8. (a) The complex formed from tweezer 2 and conjugate 1-46. The stick model represents the optimal conformation (obtained from an
AMBER Monte Carlo conformational search) of partial conjugate 1-46 lacking the H2NCH2CH2CH2NH- tail before complexation with tweezer 2;
the protruding methyl ester group is assigned L*. (b) The complex formed from tweezer 2 and conjugate 1-52. The stick model represents the
optimal conformation (obtained from an AMBER Monte Carlo conformational search) of partial conjugate 1-52 lacking the H2NCH2CH2CH2NH-
tail before complexation with tweezer 2; due to the hydrogen bonding between the amide N-H and the ester carbonyl, the ester becomes the M*
group in this case.
ring avoids approaching from the phenyl side, hence the L*
assignment. This is also supported by the fact that an additional
methylene group in 51 increases the steric demand of M*, hence
diminishing the stereodifferentiation, thus resulting in a smaller
CD amplitude. It should be noted that this change from the five-
membered-ring 50 to the six-membered-ring 51 causes, as
expected, an opposite trend in the CD amplitude as compared
to the alcohol analogues 44 and 45. The conformer in which
the N-H proton is not interacting with the aromatic ring was
found to be 3.1 kcal/mol higher in energy (data not shown).
The remaining chiral alcohols 46 to 49 and amines 52 and
54 (except for 53) all contain one aliphatic substituent and one
ester group that can behave as an H-bond acceptor. Surprisingly,
experimental results show that in alcohols it is the ester function
that is the L* group, whereas in amines the ester function is
the M* group (Table 2). Amine 53 has a methoxyl group that
is also an H-bond acceptor and hence is treated as an ester group.
The effect of the H-bond in the amine cases was studied by
molecular modeling to evaluate its effective relative steric
demand (Figure 8). The alcohols (which are esters in their
conjugates with carrier 1) lacking such H-bonding properties
were also modeled for comparison purposes. The truncated
conjugate of 1-46 (lacking the NHCH2CH2CH2NH2 chain) was
modeled via a Monte Carlo conformational search in the solvent
CHCl3 using the AMBER force field. This resulted in a preferred
conformation that would lead to positive porphyrin helicity upon
complex formation with tweezer 2. Namely in the conformer
shown in Figure 8a, the aliphatic methyl group is in the plane
of the carrier carbonyl and with a carbonyl-methine projection
angle of +46°. This leaves the ester group protruding from the
side (hence L*). This conformational result was also supported
by X-ray crystallographic data of similar esters.36 Alternate
conformers (not shown) that would have resulted in negative
porphyrin helicity were at least 2.4 kcal/mol higher in energy.
Another contributing electronic factor that needs to be consid-
ered is the potential electronic repulsion that may take place
between the partially negative ester carbonyl oxygen and the
π-electrons of the P-2 porphyrin ring. This favors the orientation
in which the ester/lactone groups are pointing away from the
P-2 porphyrin ring. In all four cases 46 to 49, the P-2 porphyrin
ring avoids approach from the side of the ester. Therefore in
alcohol cases 46 and 47, where one of the substituents is an
ester and the other one is a methyl or isopropyl, the former
will have the effect of the “larger” group pointing away from
P-2, and hence is given the assignment L*. Lactones 48 and 49
also follow the same trend, with P-2 avoiding complexation from
the side of the ester group.
In the amine substrates 52 to 54, the amide N-H proton
engages in H-bonding to the methoxyl oxygen in 53 and to the
ester carbonyl oxygen in 52 and 54, as shown for 52 in Figure
8b. This type of H-bonding has been well studied and
encountered in X-ray crystallographic37 data of similar com-
pounds. This was also confirmed by molecular modeling
calculations using the AMBER force field and the Monte Carlo
conformational search tool in solvent CHCl3. In the most stable
conformer of partial conjugate 1-52 (lacking the NHCH2CH2-
CH2NH2), the carbonyl of the ester group, the amide N-H
proton, and the carbonyl of the amide are all coplanar due to
the hydrogen bond between N-H and the ester carbonyl (Figure
8b). The projection angle between the carrier amide carbonyl
and methine hydrogen was -43°. This orients the ester group
(or the methoxyl in 53) into the plane of the amide carbonyl
which ultimately makes the aliphatic chain protrude to the side
(hence L*). The alternate conformer in which the ester carbonyl
(36) (a) Vazquez, M. T.; Pujol, M. D.; Solans, X. Acta Crystallogr., Sect.
C: Cryst. Struct. Commun. 1998, C54, 428-430. (b) Laarhoven, W. H.;
Prinsen, W. J. C.; Behm, H.; Bosman, W. P.; Beurskens, P. T. J. Crystallogr.
Spectrosc. Res. 1989, 19, 215-222.
(37) (a) Vrielink, A.; Obel-Jorgensen, A.; Codding, P. W. Acta Crys-
tallogr., Sect. C: Cryst. Struct. Commun. 1996, C52, 1300-1302. (b)
Padiyar, G. S.; Seshadri, T. P. Acta Crystallogr., Sect. C: Cryst. Struct.
Commun. 1996, C52, 1693-1695.

Configuration of Monoalcohols and Primary Monoamines
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 5971
Table 3. Structures and Schematic Representations of Some Chiral Natural Products and CD Data of Their Conjugates with Carrier 1 after
Complexation with Tweezer 2
Figure 9. (a) Conventional representation of ginkgolide A 58 conjugate with carrier 1. (b) The same structure presented in Figure 9a oriented in
a fashion to overlap with the stereo stick models in part c and to divide the cage molecule into two halves along the 10-oxygen function and 10-H.
The larger and smaller halves are shown in red (L) and black (M), respectively. (c) Stereoview of the optimal conformation of partial conjugate
1-58 (lacking the H2NCH2CH2- tail) before complexation with tweezer 2 obtained from an AMBER Monte Carlo conformational search, emphasizing
the orientation of the conjugate carbonyl relative to the ginkgolide cage.
and the N-H are not H-bonded and thus with the ester group
protruding was 1.24 kcal/mol higher in energy. Moreover, since
the N-H proton diminishes the electron density on the carbonyl
via hydrogen bonding, the electronic repulsion between the ester
carbonyl and P-2, observed in the conjugates of alcohols 46 to
49, may not play a dominant role in amino compounds 52 to
54.
It should be noted that the ester-containing substrates (Table
2) were included in the study mainly for probing the applicability
of the method. However, since these substrates in fact contain
two functional groups, their absolute configuration can be readily
determined by the conventional exciton chirality method; i.e.,
after hydrolysis of the ester and lactone groups, the R-hydroxy
or R-amino acids can be derivatized to the corresponding
porphyrins derivatives26 giving rise to exciton-coupled CD.
C. Application Extended to Complex Natural Products.
The configurational assignment protocol has been applied to
more complex compounds of biological relevance, 55 to 58,
Table 3. While in all previous cases the sign of the CD couplet
was dictated by the “size” of the groups next to the stereogenic
center, i.e., the R-position, this is not the case with cholestanols
55 and 56. In these steroids, the alcohol function is flanked by
two methylene groups at C-2 and C-4, but as C-1 is also a
methylene while C-5 is at the ring juncture, the C-5 side
becomes the larger group, L. Therefore in the case of 56 with
the hydroxyl pointing away from the viewer, the counterclock-

5972 J. Am. Chem. Soc., Vol. 123, No. 25, 2001
Kurta´n et al.
wise relationship among L, M, and H predicts a negative
exciton-coupled CD. In steroid 55, however, the opposite
3-hydroxyl configuration gives the opposite sign for the exciton-
coupled CD. In the caged ginkgolide molecules, dianhydro
ginkgolide C 10-monoacetate38 57 and ginkgolide A 58,
assignments of L and M to the moieties flanking the 7- and
10-hydroxyls, respectively, require a more thorough examination
of the three-dimensional structure.
Experimental Section
Zinc porphyrin tweezer 2 is commercially available from TCI
(Japan).39
Preparation of Conjugates 19. Method A: via diBoc Reagent
17. To a solution of secondary alcohol/primary monoamine (15 µmol)
and 17 (18.5 µmol) in anhydrous CH2Cl2 (5 mL) were added EDC
(20.4 µmol) and DMAP (0.2 equiv) with stirring for 1 h at 0 °C and
overnight at room temperature. The mixture was then washed with
NaHCO3 and brine and dried over Na2SO4. Subsequent purification
by flash chromatography afforded the diBoc protected conjugate 18.
Deprotection was carried out by dissolving 18 in CH2Cl2 (2 mL)
followed by the addition of TFA (0.4 mL). After being stirred at room
temperature for 2 h, the solution was evaporated and dried in vacuo to
give conjugate 19 as its TFA salt. This protocol was successful with
microscale amounts of alcohol as low as 0.35 µmol which is equivalent
to 55 µg in the case of (S)-(-)-isomenthol 25.
Method B: via Bromoacetate 20. To a solution of secondary
alcohol/primary monoamine (10 µmol) and bromoacetic acid (2 equiv)
in CH2Cl2 (5 mL)were added EDC (2 equiv) and DMAP (0.2 equiv)
with stirring at 0 °C for 1 h and then stirring overnight at room
temperature. The mixture was then washed with NaHCO3 and brine
and dried over Na2SO4. Subsequent purification by flash chromatog-
raphy afforded bromoacetate derivative 20. To a solution of 1,3-
diaminopropane 13 (10 equiv) and diisopropylethylamine (DIPEA) (0.1
equiv) in THF (3 mL) was added a solution of 20 (1 equiv) in THF (1
mL) dropwise and with stirring for 1 h at 0 °C and overnight at room
temperature. After evaporation of the solvent, the remaining residue
was dissolved in CH2Cl2 (15 mL) and washed three times with brine
until 13 could no longer be detected by TLC. The organic layer was
dried over Na2SO4 and evaporated to yield conjugate 19 as the free
diamine.
In ginkgolide 57 (Table 3), an imaginary divider along the
7-hydroxyl will show that the right-hand side, i.e., the tert-butyl
side, is smaller in size (hence M) than the diene side (hence L).
With ginkgolide A 58 (Figure 9), the case is much less obvious
since the 10-OH is somehow buried in the cage of the molecule,
which is more concave in nature than 57 due to the absence of
the two double bonds. A Monte Carlo conformational search,
using an AMBER force field, of the ginkgolide partial conjugate
1-58 (lacking CH2CH2NH2) shows that lactone C lies perpen-
dicular to the rest of the cage and that the lactone side (also
containing the rest of the molecule shown in red in Figure 9a,b)
is much larger than the quaternary carbon bearing the tert-butyl
group (see also the stereoview in Figure 9c) and hence the
lactone side is assigned L. This assignment is dependent on the
projection angle between the ester carbonyl and the methine
proton 10-H, calculated to be -27° in the most stable conformer.
Therefore in conjugate 1-58, the L assignment reflects not only
the steric size of the directly attached lactone C but also the
entire moiety shown in red in Figure 9b. Ginkgolides are rigid
molecules and hence molecular modeling calculations give only
a small number of possible conformers. All the other conformers
obtained from the search that were within 2.4 kcal/mol showed
similar projection angles ranging from -27° to -35°, and led
to the same L and M assignment and hence a negative exciton-
coupled CD of its tweezer 2 complex. These L and M
assignments in ginkgolide conjugates 1-57 and 1-58 also
corroborate those of conjugates 1-46 to 1-49.
General Procedure for Milligram Scale Preparation of Host-
Guest Complexes for Measurement of CD. Preparation of Tweezer
2 Complex of 1-35. In a typical experiment, a 1 µM tweezer 2 solution
was prepared by the addition of a 10 µL aliquot of tweezer 2 (0.1 mM
in anhydrous CH2Cl2) to 1 mL of CH2Cl2. The exact concentration of
the diluted tweezer 2 solution was determined by UV from the known
ꢀ value of the Soret band in CH2Cl2 (ꢀ ) 890 000 L mol-1 cm-1). The
free amine solution of conjugate 1-35 (1.8 mg, 3.54 µmol) was prepared
from its TFA salt after the addition of 0.5 mL of MeOH followed by
solid Na2CO3 (10 mg). The solvent (MeOH) was then dried under a
stream of argon followed by placement under high vacuum (0.2 Torr)
for 20 min. Anhydrous CH2Cl2 (1 mL) was then added to yield the
free amine solution of conjugate 1-35 (3.54 mM). An aliquot of 10 µL
of the latter solution (ca. 30 equivalents) was added to the prepared
porphyrin tweezer 2 solution to afford tweezer 2/conjugate 1-35 host-
guest complex. The UV/VIS and CD spectra were recorded at 25 °C
and corrected for background. The CD spectra were recorded on
JASCO-720 and JASCO-810 spectropolarimeter driven by a JASCO
V500/FP-750 analysis program for Windows. The CD spectra were
measured in millidegrees and normalized into ∆ꢀmax [L mol-1 cm-1]/λ
[nm] units. In the UV spectra, the red shift of the tweezer Soret band
indicated that complexation took place.
Conclusion
A microscale procedure based on the CD exciton chirality
method has been developed for the determination of absolute
configurations of monoalcohols and primary monoamines.
Conventional exciton chirality methods cannot be applied to
such substrates due to the lack of required two sites for
chromophoric derivatization. This new protocol takes advantage
of host/guest complexation between chiral substrate and bichro-
mophoric zinc porphyrin host as a CD reporter group. The
method is applicable to alcohols and amines carrying aliphatic
groups and/or aromatic groups. The substrate is derivatized with
carrier 1, the resultant conjugate complexed with tweezer 2,
and the CD spectrum is measured. A clockwise relationship of
L, M, and H groups of the substrate denotes a positive CD
couplet, and vice versa; the L (large) and M (medium)
assignments are based on the conformational A values. Although
there are other CD methods available for the configurational
assignments of R-hydroxy and R-amino esters, an attempt has
been made to extend this method to this class of substrates as
well. This has led to further clarification of additional electronic
factors that govern the L and M assignments. Furthermore, NMR
analysis and molecular modeling (see succeeding paper) can
lead to independent L/M assignments of substituents which is
critical for interpreting the CD data and determining absolute
configurations, thus extending the protocol to more complicated
cases.
Acknowledgment. This research was supported by NIH
grants GM 34509 and AI 10187. We acknowledge the assistance
of Cara Cesario in syntheses and CD measurements. We also
extend our thanks to Professors Claudio Villani, Universita` di
Chieti ‘G. D’Annunzio’, and Maurizio Botta, Universita` di
Siena, for providing samples 48 and 49, and Professor Shu`
Kobayashi, University of Tokyo, for providing samples 36 and
37. T. Kurta´n is indebted to the Fulbright Foundation and Y.-
Q. Li to the Suntory Institute for Bioorganic Research for
(39) The synthesis of the zinc porphyrin tweezer 2 was first accomplished
in our laboratories (see ref 1) and is now commercially available from TCI,
Japan. Address: Tokyo Kasei Kogyo Co., Ltd., 3-1-13 Nihonbashi-Honcho,
Chuo-Ku, Tokyo 103-0023, Japan. (TCI America, 9211 North Harborgate
Street, Portland, OR 97203.)
(38) Maruyama, M.; Terahara, A.; Itagaki, Y.; Nakanishi, K. Tetrahedron
Lett. 1967, 4, 303-308.

Configuration of Monoalcohols and Primary Monoamines
J. Am. Chem. Soc., Vol. 123, No. 25, 2001 5973
financial support. We thank Professor Ronald Breslow for the
use of the O2 Silicon Graphics computer.
58, bromoacetates of 25 to 27, 29 to 31, 33, 34, 44, 45, and
bromoacetamides of 39 to 41, 51; synthetic schemes for
conjugates 11 and 12, and CD data of their tweezer 2 complexes;
and CD measurements of the tweezer 2 complex of conjugate
1-30 in different solvents (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
Supporting Information Available: Materials and general
procedures, syntheses of 14 to 17, carrier 9, and conjugates 11
and 12; 1H NMR and MS data of conjugates 1-21 to 1-58, 10-
23, 1a-25, 1b-25, diBoc conjugates 17-21 to 17-24, 17-28, 17-
32, 17-35 to 17-39, 17-42, 17-43, 17-46 to 17-49, 17-52 to 17-
JA010249W