Optical rotation computation, total synthesis, and stereochemistry assignment of the marine natural product Pitiamide A

Article abstract of DOI:10.1021/ja9945313

We report the joint application of ab initio computations and total synthesis to assign the absolute configuration of a new natural product. The expected specific rotations of the (7S,10R)- and (7R,10R)-isomers of pitiamide A in a CHCl₃ solvent continuum model were determined as +8 and -39, respectively, by CADPAC calculations of the electric-dipole-magnetic-dipole polarizability tensor. Total syntheses of these two stereoisomers of the marine metabolite were achieved by a convergent strategy that utilized Evans' oxazolidinone alkylation, a novel water-accelerated modification of Negishi's zirconocene-catalyzed asymmetric carbometalation as well as an unusual segment condensation via Mitsunobu alkylation of a nosyl-activated amide. The experimental optical rotation measurements confirmed the results of the computational optical rotation predictions. On the basis of NMR comparisons, the configuration of pitiamide A was assigned as (7R,10R). These studies highlight the considerable structural significance of [α]_D data, but, because the optical rotation of the natural product was different from either synthetic diastereomer, our work serves also as an illustration of potential problems with obtaining accurate experimental [α]_D data for natural samples.

Full text of DOI:10.1021/ja9945313

4608
J. Am. Chem. Soc. 2000, 122, 4608-4617
Optical Rotation Computation, Total Synthesis, and Stereochemistry
Assignment of the Marine Natural Product Pitiamide A
Seth Ribe, Rama K. Kondru, David N. Beratan,* and Peter Wipf*
Contribution from the Department of Chemistry, UniVersity of Pittsburgh, Pittsburgh, PennsylVania 15260
ReceiVed December 30, 1999
Abstract: We report the joint application of ab initio computations and total synthesis to assign the absolute
configuration of a new natural product. The expected specific rotations of the (7S,10R)- and (7R,10R)-isomers
of pitiamide A in a CHCl3 solvent continuum model were determined as +8 and -39, respectively, by CADPAC
calculations of the electric-dipole-magnetic-dipole polarizability tensor. Total syntheses of these two
stereoisomers of the marine metabolite were achieved by a convergent strategy that utilized Evans’ oxazolidinone
alkylation, a novel water-accelerated modification of Negishi’s zirconocene-catalyzed asymmetric carbometa-
lation as well as an unusual segment condensation via Mitsunobu alkylation of a nosyl-activated amide. The
experimental optical rotation measurements confirmed the results of the computational optical rotation
predictions. On the basis of NMR comparisons, the configuration of pitiamide A was assigned as (7R,10R).
These studies highlight the considerable structural significance of [R]D data, but, because the optical rotation
of the natural product was different from either synthetic diastereomer, our work serves also as an illustration
of potential problems with obtaining accurate experimental [R]D data for natural samples.
The chlorinated lipid pitiamide A (1) was isolated in 1997
from an extract of a mixed assemblage of Lyngbya majuscula
and Microcoleus sp. growing on intact colonies of the hard coral
Porites cylindrica on Guam.1 Pitiamide may act as a feeding
deterrent to both vertebrate and invertebrate herbivore species,
but the determination of its biological profile has been limited
by the lack of availability of sufficient quantities of the natural
product.
Figure 1. Structure of pitiamide A as assigned by Paul and co-
workers.1
computed rotations are Boltzmann-weighted averages of values
obtained for individual fragments and can be directly compared
to experimental specific rotations. Our earlier theoretical analysis
of hennoxazole A represented an a posteriori assignment of
stereochemistry. We decided to further apply this methodology
for the computational prediction of the absolute configuration
of a natural product, i.e., an a priori assignment of the relative
and absolute stereochemistry of pitiamide A.
The structure of this unusual vinyl chloride-containing marine
metabolite (Figure 1) was determined by 2D NMR spectral
analysis and mass spectroscopy, and its [R]D was found to be
-10.3 (c 3.0, CHCl3, 27 °C).2 The absolute and relative
configurations at C(7) and C(10) were not assigned. Pitiamide
A is a temperature- and light-sensitive oil, and due to the
flexibility of most of its backbone chain, a stereochemical
assignment based entirely on NMR spectroscopic methods has
not been realized. Elucidation of its relative and absolute
stereochemistry by circular dichroism (CD) is also not promis-
ing, in particular since the stereocenter at C(10) is not part of
a strongly asymmetrically perturbed chromophore absorbing at
λ >180 nm (vide infra). Accordingly, calculation of the expected
molar rotation of pitiamide A diastereomers and enantiomers
represents the most attractive solution to this structural problem,
but is most challenging due to the unusually high level of
flexibility in the lipid backbone.3
Optical Rotation Computations
In accordance with principles developed in our earlier work,3,4
we dissected pitiamide A into two chiroptically independent
fragments I and II (Figure 2). This structural simplification is
necessary because current coupled Hartree-Fock methods are
limited to 255 basis functions, a number that correlates at the
6-31G* level approximately with a limit of 14 carbon or
heteroatoms in a given molecule. We selected fragments I and
II because they provide increments5-7 for the contribution of
asymmetric carbons C(7) and C(10), respectively, to the molar
rotation of pitiamide A. In some cases, an extended π-system
can make considerable contributions to the rotation angle,4
Recently, we devised a Monte Carlo/ab initio approach to
assign the configuration of the complex natural product hen-
noxazole A from computation of the electric-dipole magnetic-
dipole polarizability tensor of suitable fragment molecules.3 The
* To whom correspondence should be addressed.
(1) Nagle, D. G.; Park, P. U.; Paul, V. J. Tetrahedron Lett. 1997, 38,
6969.
(2) Possibly because of instabilities in the polarimeter, the actual
measurements fluctuated from -15.3 to -8.7. Prof. V. Paul, personal
communication.
(4) Kondru, R. K.; Wipf, P.; Beratan, D. N. J. Phys. Chem. 1999, 103,
6603.
(5) (a) van’t Hoff, J. H. Arch. Neerl. Sci. Exactes Nat. 1874, 9, 455. (b)
van’t Hoff, J. H. Die Lagerung der Atome im Raume; Vieweg: Braunsch-
weig, Germany, 1894; p 95.
(6) Kondru, R. K.; Lim, S.; Wipf, P.; Beratan, D. N. Chirality 1997, 9,
469.
(7) Wipf, P.; Lim, S. Chimia 1996, 50, 157.
(3) Kondru, R. K.; Wipf, P.; Beratan, D. N. J. Am. Chem. Soc. 1998,
120, 2204.
10.1021/ja9945313 CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/29/2000

Marine Natural Product Pitiamide A
J. Am. Chem. Soc., Vol. 122, No. 19, 2000 4609
Raman optical activity and vibrational CD is based on ground
state linear response calculations.3,4,37-40 Applications of this
theoretical strategy toward quantitatively reliable quantum
chemical computations of molar rotation angles are now
beginning to have a large impact on the field.36-40 The
expression for the optical rotational angle, φ in radians, is41
φ ) 4πNâω2(n2+2)/3c2
(1)
where
â ) -ω-1(G′xx + G′yy + G′zz)/3
(2)
Figure 2. Fragments (I-III) for pitiamide A molar rotation angle
computations.
and G′ii are the diagonal elements of the electric-dipole-
magnetic-dipole polarizability tensor.37 N is the number of
molecules per unit volume, n is the refractive index of the
medium, and c is the speed of light. The specific rotation angle
and therefore we included the diene π-system along with the
chlorine atom in fragment I. In general, stereocenters more than
three atoms removed from each other have negligible mixed-
term contributions to the rotation of the plane of polarized light,
as long as the rotation of the chain can occur freely.6 To test
this approximation, we selected a fragment III that contains
both stereocenters at C(7) and C(10).
(measured at the sodium D-line), in units of deg [dm (g/mL)]-1
is
,
[R]D ) 1.343 × 10-4 âνj2(n2 + 2)/3MW
(3)
The selection of amide-containing fragments in our force-
field calculations was complicated by the intrinsic trend of the
molecular mechanics parametrization to overvalue intramolecu-
lar hydrogen-bonded conformations and favor an eight-
membered H-bridge between the C(8) carbonyl group and the
amide NH. As a molecular dynamics analysis of pitiamide A
showed, populated states at room temperature can be signifi-
cantly different from the minimum energy conformers.8 Exten-
sive (10 ns) stochastic dynamics analysis9 of pitiamide A at
300 K using a CHCl3-solvent continuum model10 indicated that
for both diastereomeric configurations, any hydrogen-bonded
conformations were unlikely to account for more than 5-10%
of the molecular ensemble.11 Accordingly, our fragmentation
strategy was judged to be adequate, and we omitted the C(12)-
amide substituent in any subsequent fragment computations. The
effect of this structural simplification as well as the omission
of the C(17)-C(18) double bond on the optical rotation of
pitiamide should be minimal, because both groups are quite
remote from the nearest stereocenter at C(10). In fact, subsequent
NMR and CD analyses of synthetic derivatives validated this
approach (vide infra). Although basing the ab initio calculations
on force-field optimized fragment geometries is clearly less than
ideal, extensive prior tests have demonstrated the viability of
this approach.3,4,6,12
where â is in units of (bohr)4, MW is the molar mass in g/mole,
and νj is the frequency of the sodium D-line in cm-1 37
From
.
[R]D, the molar rotation is defined as [M]D ) [R]DMW/100. The
diagonal elements of the electric-dipole magnetic-dipole polar-
izability tensor, G′ii,42 are calculated using CADPAC43 or
DALTON.44 The CADPAC methodology employed here is most
reliable far from electronic resonance. Note that the G′ii
elements are computed relatively well despite the fact that linear-
(13) Walden, P. Z. Phys. Chem. 1894, 15, 196.
(14) Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of Organic
Compounds; Wiley: New York, 1994.
(15) Nakanishi, K.; Berova, N.; Woody, R. W. Circular Dichroism
Principles and Applications; VCH Publishers Inc.: New York, 1994.
(16) Kauzmann, W.; Eyring, H. J. Chem. Phys. 1941, 9, 41.
(17) Freudenberg, K.; Todd, J.; Seidler, R. Ann. Chem. 1933, 501, 199.
(18) Brewster, J. H. J. Am. Chem. Soc. 1959, 81, 5475.
(19) Brewster, J. H. Tetrahedron 1961, 13, 106.
(20) Moscowitz, A. AdV. Chem. Phys. 1962, 4, 67.
(21) Applequist, J. J. Chem. Phys. 1973, 58, 4251.
(22) Tinoco, I., Jr.; Woody, R. W. J. Chem. Phys. 1964, 40, 160.
(23) Sathyanarayana, B. K.; Stevens, E. S. J. Org. Chem. 1987, 52, 3170.
(24) Gould, R. R.; Hoffmann, R. J. Am. Chem. Soc. 1970, 92, 1813.
(25) Pao, Y. H.; Santry, D. P. J. Am. Chem. Soc. 1966, 88, 4157.
(26) Kuhn, W. Stereochemie; Freudenberg, K., Ed.; Deuticke: Leipzig,
1933; Chapter 8, p 394.
(27) Born, M. Proc. R. Soc. London A 1935, 150, 84.
(28) Kirkwood, J. G. J. Chem. Phys. 1937, 5, 479.
(29) Nolte, H. J.; Buss, V. Tetrahedron 1975, 31, 719.
(30) Maestro, M.; Moccia, R.; Taddei, G. Theor. Chim. Acta 1967, 8,
80.
(31) Eyring, H.; Walter, J.; Kimball, G. E. Quantum Chemistry; Wiley:
New York, 1944.
(32) Rauk, A.; Barriel, J. M. Chem. Phys. 1977, 25, 409.
(33) Bak, K. L.; Hansen, A. E.; Ruud, K.; Helgaker, T.; Olsen, J.;
Jørgensen, P. Theor. Chim. Acta. 1995, 90, 441.
For the interpretation of chiroptical data, many empiri-
cal,3-6,13-19 semiempirical,20-25 classical,26,27 and quantum
mechanical6,28-35 models have been developed. Direct applica-
tion of the Rosenfeld equation, the most fundamental quantum
mechanical expression for chiroptical effects, is not yet feasible
because it includes a sum over all molecular excited states. An
alternative approach employed recently with great success for
(34) Hansen, A. E.; Bouman, T. D. J. Am. Chem. Soc. 1985, 107, 4828.
(35) Snatzke, G. In Chirality: from Weak Bosons to the Alpha-Helix;
Janoschek, R., Ed.; Springer-Verlag: Berlin, 1991; pp 59-85.
(36) Costante, J.; Hecht, L.; Polavarapu, P. L.; Collet, A.; Barron, L. D.
Angew. Chem., Int. Ed. Engl. 1997, 36, 885.
(8) See also: McDonald, D. Q.; Still, W. C. Tetrahedron Lett. 1992,
33, 7747.
(9) Van Gunsteren, W. F.; Berendsen, H. J. C. Mol. Simul. 1988, 1, 173.
(10) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T. J. Am.
Chem. Soc. 1990, 112, 6127.
(37) Polavarapu, P. L. Mol. Phys. 1997, 91, 551.
(11) Calculations of the optical rotation of hydrogen-bonded conforma-
tions of C(5)-C(14) segments of pitiamide A provided absolute values that
were in the same order of magnitude as open-chain conformations. These
control studies addressed the potential pitfall that small percentages of
intramolecular hydrogen-bonded conformations could significantly alter the
overall optical rotation of pitiamide A and thus introduce a major source
of error in our computational strategy.
(38) Kondru, R. K.; Wipf, P.; Beratan, D. N. Science 1998, 282, 2247.
(39) Polavarapu, P. L. Tetrahedron: Asymmetry 1997, 8, 3397.
(40) (a) Polavarapu, P. L.; Chakraborty, D. K. J. Am. Chem. Soc. 1998,
120, 6160. (b) Polavarapu, P. L.; Zhao, C. J. Am. Chem. Soc. 1999, 121,
246.
(41) Rosenfeld, L. Z. Physik 1928, 52, 161.
(42) Amos, R. D. Chem. Phys. Lett. 1982, 87, 23.
(12) Kondru, R. K.; Chen, C. H.-T.; Curran, D. P.; Beratan, D. N.; Wipf,
P. Tetrahedron: Asymmetry 1999, 10, 4143.
(43) Amos, R. D.; Rice, J. E. The Cambridge Analytic DeriVatiVe
Package; issue 6.3, Cambridge, U.K., 1997.

4610 J. Am. Chem. Soc., Vol. 122, No. 19, 2000
Ribe et al.
Table 1. Computed Molar Rotation Angles for Fragments I-III,
and Pitiamide A (1) Based on Fragment Analysis
[M]D (for indicated stereoisomer)
fragment I
fragment II
fragment III
pitiamide A
+111 (7S)
+66 (10S)
+123 (7S,10S)
+150 (7S,10S)
+17 (7S,10R)
+31 (7S,10R)
The values shown for fragments I-III are averages of at least two
independent calculations that differ in their results by <10%. The value
for pitiamide is the average of fragments (I + II) and III. The
configurations of fragments I-III and pitiamide are indicated next to
the computed molar rotation angles.
response methods are not guaranteed to reproduce the molecular
electronic absorption spectra.
Conformation has a major influence on specific rotation
angles.3,4,45 Unique low-energy geometries for our fragments
were obtained from Monte Carlo conformational searches using
the Macromodel46 program with the MM2* force field. This
force field has been found to produce geometries and relative
energies of high reliability.47 After building the required structure
of a specified configuration, energy minimization was performed
using MM2* with the PRCG algorithm and a CHCl3 continuum
solvent model.46 Low-energy structures were chosen from a
Monte Carlo sampling of 5000 conformations (fragments I and
II) or 2000 conformations (fragment III); each new conforma-
tion was energy minimized using a 3000 step energy minimiza-
tion iteration method forcing all newly found structures to be
fully relaxed, and duplicates were automatically rejected. All
conformations generated within ∼(2-3) kT of the lowest energy
structure (those thermally accessible in solution) were used in
the optical rotation angle computations.
Figure 3. Retrosynthetic analysis of pitiamide A.
that our calculations used a CHCl3 solvation model, and are
thus in agreement with the solvent environment used for the
specific rotation determination of the natural product.
Total Syntheses of (7S,10R)- and (7R,10R)-Pitiamide A
The 1-chlorodiene moiety of pitiamide A is a highly unusual
structural feature among marine natural products,1,49 and we
were interested in devising an efficient synthetic strategy for
the preparation of this moiety as well as for the catalytic
asymmetric introduction of the â-carbonyl stereocenter at C(10).
In addition, total synthesis of pitiamide A stereoisomers offered
the opportunity to test the predictive power of our computational
methodology for optical rotation calculations. Accordingly, we
embarked on a total synthesis of the marine lipid guided by the
retrosynthetic analysis shown in Figure 3. Formation of the
C(8)-C(9) bond in 1 was envisioned to result from the addition
of an organometallic reagent 3 to Weinreb amide 2. A novel
Mitsunobu chain extension50 with sulfonimide 4 and deprotec-
tion of the o-nitrophenylsulfonyl group51 should complete the
highly convergent segment assembly.52 Hydroxamate 2 can be
obtained from the Evans-oxazolidinone-derived53 imide 5, which
also serves to install the R-carbonyl stereocenter at C(7). Inspired
by the elegant studies of Negishi and Kondakov,54 we planned
to use an asymmetric zirconocene-catalyzed carbometalation of
monosubstituted alkene 6 toward the key organometallic
intermediate 3. The γ,δ-unsaturated carboxylate 4 could be
obtained by Johnson ortho ester Claisen rearrangement55 from
allylic alcohol 7. Finally, a palladium-mediated coupling56
A total number of 180 conformations of fragments I-III were
subjected to calculations of the molar rotation angles in the static
field approximation implemented in CADPAC with a 6-31G*
basis set.48 The results are summarized in Table 1. The
composite values for fragments I and II (a simple additivity of
molar rotation angles was assumed based on van’t Hoff’s
principle5) were combined with values for fragment III and
averaged to provide predictions for the molar rotations of
pitiamide diastereomers and enantiomers. On the basis of this
computational analysis, the syn [(7S,10S) or (7R,10R)] and anti
[(7S,10R) or (7R,10S)] stereoisomers of pitiamide A were
assigned molar rotations [M]D of +150, -150, +31, and -31,
respectively. In units of [R]D, (7S,10S)-, (7R,10R)-, (7S,10R)-,
and (7R,10S)-stereoisomers of pitiamide A are thus predicted
to have rotations of +39, -39, +8, and -8 deg [dm (g/mL)]-1
,
respectively. Accordingly, the (7R,10S)-configuration provides
the closest agreement in value and sign with the experimental
[R]D of -10.3 reported by Paul et al.1 It is important to stress
(44) Helgaker, T.; Jensen, H. J. A.; Olsen, J.; Ruud, K.; Anderson, A.
T.; Bak, K. L.; Bakken, V.; Christiansen, O.; Dahle, P.; Dalskov, E. K.;
Enevoldsen, T.; Fernandez, B.; Heiberg, H.; Hettema, H.; Jonsson, D.;
Kirpekar, S.; Kobayashi, R.; Koch, H.; Mikkelsen, K. V.; Norman, P.;
Packer, P.; Saue, T.; Taylor, P. R.; Vahtras, O.; DALTON, an electronic
structure program, Release 1.0, 1997.
(49) Mynderse, J. S.; Moore, R. E. Phytochem. 1978, 17, 1325.
(50) (a) Hughes, D. L. Org. Prep. Proc. Int. 1996, 28, 127. (b) Hughes,
D. L. Org. React. 1992, 42, 335. (c) Mitsunobu, O. Synthesis 1981, 1.
(51) Fukuyama, T.; Jow, C.-K.; Cheung, M. Tetrahedron Lett. 1995, 36,
6373.
(45) (a) Kauzmann, W.; Eyring, H. J. Chem. Phys. 1941, 9, 41. (b) Arndt,
E. R.; Stevens, E. S. Biopolym. 1994, 34, 1527. (c) Arndt, E. R.; Stevens,
E. S. Biopolymers 1996, 38, 567. (d) Johnson, W. C.; Tinoco, I. J. Am.
Chem. Soc. 1972, 94, 4389. (e) Schellman, J. A. J. Chem. Phys. 1966, 44,
55.
(46) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.;
Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J. Comput. Chem.
1990, 11, 440.
(52) For N-methylation of a nosylated amide, see: Katagiri, N.; Take-
bayashi, M.; Kokufuda, H.; Kaneko, C.; Kanehira, K.; Torihara, M. J. Org.
Chem. 1997, 62, 1580.
(53) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982,
104, 1737.
(54) (a) Kondakov, D. Y.; Negishi, E.-I. J. Am. Chem. Soc. 1995, 117,
10771. (b) Kondakov, D. Y.; Negishi, E.-I. J. Am. Chem. Soc. 1996, 118,
1577.
(47) Pettersson, I.; Liljefors, T. In ReViews in Conputational Chemistry;
Lipkowitz, K. B., Boyd, D. B., Eds.; VCH: New York, 1996; Vol. 9, p
167.
(48) Calculations were performed on a 600 MHz dec-alpha workstation
and required between 4 and 12 days of CPU time for each fragment.
(55) (a) Wipf, P. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Paquette, L. A., Eds.; Pergamon: Oxford, 1991; Vol. 5, pp
827-874. (b) Johnson, W. S.; Werthemann, L.; Bartlett, W. R.; Brocksom,
T. J.; Li, T.-T.; Faulkner, D. J.; Petersen, M. R. J. Am. Chem. Soc. 1970,
92, 741.

Marine Natural Product Pitiamide A
J. Am. Chem. Soc., Vol. 122, No. 19, 2000 4611
Scheme 1
Scheme 3
Scheme 4
Scheme 2
Attachment of the (S)- or, alternatively, the (R)-benzylox-
azolidinone53 via the acyl chloride followed by deprotonation
with sodium hexamethyldisilazane (NaHMDSA) and alkylation
at -78 °C with MeI provided (S)-559 in 72% yield over the
between alkenyl zirconocene 9, obtained by hydrozirconation57
of the corresponding alkyne, and the readily available 1-iodo-
2-chloroethene (8) should allow access to the chlorodiene
moiety. Although some tactical and experimental modifications
became necessary, this retrosynthetic plan ultimately proved
successful for the preparation of the desired two diastereomers
of pitiamide A.
1
three-step sequence and in 56% based on 10 (Scheme 3). H
NMR analysis of the crude reaction mixture revealed a 92%
diastereoselectivity for the methylation reaction. It is important
to note that the minor diastereomer can readily be removed
chromatographically, and essentially stereochemically pure (S)-5
was used for further conversions. The reaction sequence shown
in Scheme 3 was repeated with the (R)-benzyloxazolidinone
substrate, and (R)-5 was obtained in 63% yield and 91%
enantiomeric excess (ee) from 13.
For the synthesis of the C(1)-C(8) segment 5, tert-butyldim-
ethylsilyl-protected butynol 10 was hydrozirconated with Cp2-
ZrHCl57 (Scheme 1). In situ transmetalation to ZnCl2 followed
by the addition of vinyl iodide 8 58 and 5-8 mol % of Pd-
(PPh3)4 catalyst56 cleanly provided the cross coupling product
11 in high yield exclusively in the desired all-(E) double bond
configuration. The tert-butyldimethylsilyl (TBDMS)-ether was
cleaved with tetrabutylammonium fluoride (TBAF) in tetrahy-
drofuran (THF) at room temperature to give the somewhat
unstable alcohol 12. Mesylation with methanesulfonyl chloride
and 2,6-lutidine followed by nucleophilic displacement with the
anion of diethyl malonate provided the diester intermediate
which was saponified without further purification. After neu-
tralization with 1 N HCl and bulb to bulb distillation, dienyl
acid 13 was isolated in nearly quantitative yield based on alcohol
12. Although 13 is relatively unstable, it could be obtained in
very high purity after only one distillation, and no further
purification was necessary. Alternatively, carboxylic acid 13
could also be obtained much more directly by hydrozircona-
tion-cross coupling of alkynoate 14 (Scheme 2) followed by
saponification. However, chromatographic separation of the
alkene side product derived from 14 and the desired 13 was
extremely tedious, especially on large scale, and therefore this
route was abandoned in favor of the longer but similarly high
yielding sequence shown in Scheme 1.
For the preparation of the C(9)-C(12) segment, we initially
planned to use an unprotected unsaturated alcohol as a substrate
for the zirconocene-catalyzed asymmetric methylalumination,
as reported for 5-hexen-1-ol by Negishi and Kondakov.54
However, we found that 3-buten-1-ol was not a suitable
substrate; it remained inert to the reaction conditions. Further-
more, protection of 3-butene-1-ol either as the benzyl ether or
as the triisopropylsilyl ether still did not yield any of the desired
methylated product, even after the reaction temperature was
raised to 40 °C. In earlier work, we were able to demonstrate
the use of water as a highly effective additive for the zirconocene
dichloride-catalyzed carbometalation of alkynes,60 and we
decided to attempt similarly increasing the reaction rate of the
Negishi-Kondakov process by addition of water. Indeed,
treatment of a mixture of trimethylaluminum and 5-8 mol %
of Erker’s chiral catalyst 1761 in CH2Cl2 with 1 equiv of water
led to a vastly more reactive methylating agent (Scheme 4).
Methylalumination of alkene 6 proceeded in 12 h at -20 °C
and provided, after quenching of the trialkyl alane intermediate
with air, the desired (10R)-alcohol 18 in 85% yield and 80%
ee.62,63 The absolute configuration of 18 was assigned by
(59) The (S)-stereochemistry was assigned based on well-established
literature precedence; see ref 53 and Wipf, P.; Kim, Y.; Fritch, P. C. J.
Org. Chem. 1993, 58, 7195. Comparison of the molar rotation angle of
(S)-5, +23.5, to that of the independently assigned structurally closely related
compound 11, +25.9, in the latter paper provides further proof for our
assignment of the (S)-configuration of the R-methylation product.
(60) Wipf, P.; Lim, S. Angew. Chem., Int. Ed. Engl. 1993, 32, 1068.
(61) Erker, G.; Aulbach, M.; Knickmeier, M.; Wingbermu¨hle, D.; Kru¨ger,
C.; Nolte, M.; Werner, S. J. Am. Chem. Soc. 1993, 1215, 4590.
(56) (a) Negishi, E.-I.; Okukado, N.; King, A. O.; Van Horn, D. E.;
Spiegel, B. I. J. Am. Chem. Soc. 1978, 100, 2254. (b) Panek, J. S.; Hu, T.
J. Org. Chem. 1997, 62, 4912. (c) Panek, J. S.; Hu, T. J. Org. Chem. 1997,
62, 4914.
(57) (a) Wipf, P.; Jahn, H. Tetrahedron 1996, 52, 12853. (b) Schwartz,
J.; Labinger, J. A. Angew. Chem., Int. Ed. Engl. 1976, 15, 333.
(58) Negishi, E.; Okukado, N.; Lovich, S. F.; Luo, F.-T. J. Org. Chem.
1984, 49, 2629.
(62) The %ee was assigned by chiral HPLC with a Chiracel OD column.

4612 J. Am. Chem. Soc., Vol. 122, No. 19, 2000
Ribe et al.
Scheme 5
Scheme 7
Table 2. Optical Rotation Values for (7S,10R)-1 and (7R,10R)-1
as a Function of Wavelength, Concentration, Temperature, and
Solvent
Scheme 6
conditions
(nm, c [g/100 mL], solvent, T [°C])
[R] for
(7S,10R)-1
[R] for
(7R,10R)-1
589, 0.5-0.7, CHCl3, 22
578, 0.4-0.7, CHCl3, 22
546, 0.4-0.7, CHCl3, 22
436, 0.4-0.7, CHCl3, 22
365, 0.4-0.7, CHCl3, 22
589, 0.05-0.06, CHCl3, 22
589, 0.5, 95% EtOH, 22
589, 0.5, CHCl3, 5
11.6
12.1
13.4
24.2
45.8
12.2
13.4
10.2
-30.3
-31.7
-36.2
-69.3
-131.2
-26.3
-24.2
-37.2
conditions with 4. Neighboring group participation of the ketone
functionality in 24 caused significant amounts of pyran side
products. Circumventing this problem by a protective group
strategy was straightforward but added a few steps to the total
synthesis. After reduction of the ketone with NaBH4, the
secondary alcohol was protected as the ethoxyethyl ether.
Desilylation now proceeded uneventfully, and Mitsunobu
displacement of the primary hydroxyl group led to imide 25 in
76% yield from 23. To the best of our knowledge, this is the
first application of a nosyl-protected amide in a segment
coupling.52 The desired (7S,10R)-diastereomer of pitiamide A
was readily obtained after deprotection of the nosyl group with
thiophenol and potassium carbonate in dimethylformamide
(DMF),51 removal the ethoxy ethyl ether with pyridinium
p-toluenesulfonate (PPTs) in MeOH, and oxidation with Dess-
Martin periodinane.66 Calculated for the longest linear sequence
of 18 steps originating from 10, the overall yield in this synthesis
is a very satisfactory 18%. Analogous methodology was used
for the preparation of the (7R,10R)-diastereomer of pitiamide
A (Scheme 7). Highlights of both synthetic routes are the
preparation of the chlorodiene segment by a hydrozirconation-
transmetalation sequence and the use of a water-accelerated
zirconocene-catalyzed methylalumination for the enantioselec-
tive preparation of the linchpin segment 20.
desilylation and comparison of the [R]D of diol 19 to a literature
reference.64 Alcohol 18 was readily converted to iodide 20 with
triphenylphosphine and iodine in a mixture of acetonitrile/ether.
The carbon skeleton of the third pitiamide segment, 4, was
constructed in a straightforward manner by Claisen rearrange-
ment of alcohol 7 (Scheme 5).55 Conversion of the methyl ester
into the primary amide 22 with trimethylaluminum and am-
monium chloride according to the useful protocol of Weinreb
et al.65 followed by N-nosylation provided imide 4 in 73%
overall yield from 7. The amide nitrogen was activated with
the nosyl51 group to facilitate the planned Mitsunobu displace-
ment.50,52
With all three synthetic segments in hand, we proceeded to
the segment condensation stage. Conversion of the oxazolidi-
none (S)-5 to the N,O-dimethylhydroxamate, followed by
addition of the in situ prepared organolithium species derived
from 4 provided ketone 24 in good yield (Scheme 6). Unfor-
tunately, silyl ether cleavage of 24 led to an unstable intermedi-
ate that could not be alkylated under the standard Mitsunobu
Optical Rotation, CD, IR, and NMR Temperature Shift
Coefficient Analysis of (7S,10R)- and (7R,10R)-Pitiamide
A
With multimilligram quantities of synthetic stereoisomers of
pitiamide A in hand, we determined the wavelength-, concentra-
tion-, temperature-, and solvent-dependence of the optical
rotation for each isomer. The results are summarized in Table
2. It is important to note that a minor amount (ca. 10%) of the
(10S)-isomer was present as an impurity in either sample,
because the enantiomeric excess of the methylalumination
reaction was 80%.
(63) The water effect in the asymmetric methylalumination of alkenes
is a generally useful modification that not only provided considerable rate
acceleration but in some cases also increased enantioselectivities: Wipf,
P.; Ribe, S. To be submitted for publication.
(64) On the basis of the [R]D of diol 19, the %ee of the conversion of 6
to 18 proceeded in 80-82%, in good agreement with the HPLC study.
Veith, H. J.; Collas, M.; Zimmer, R. Liebigs Ann. Org. Bioorg. Chem. 1997,
391.
As expected, the values of optical rotation measurements for
both isomers increase steadily as the probing wavelength is
(65) Levin, J. I.; Turos, E.; Weinreb, S. M. Synth. Commun. 1982, 12,
989.
(66) (a) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277.
(b) Ireland, R. E.; Liu, L. J. Org. Chem. 1993, 58, 2899.

Marine Natural Product Pitiamide A
J. Am. Chem. Soc., Vol. 122, No. 19, 2000 4613
Figure 5. Graphical display of amide proton temperature shifts (∆δ/
∆T) for (7S,10R)-1 and (7R,10R)-1 in DMSO-d6. The slope is ∆δ/
∆T.
Figure 4. Circular dichroism spectra of pitiamide A isomers obtained
from five averaged scans between 210 and 300 nm.
decreased. At 365 nm, (7S,10R)-1 exhibits a specific rotation
of +45.8, whereas (7R,10R)-1 measures -131.2. More impor-
tantly, there is no significant concentration dependence of [R]D
for either of the two isomers: Reducing the concentration 10-
fold from 0.6 to 0.06 g/100 mL led to a change of [R]D from
+11.6 to +12.2 for (7S,10R)-1 and from -30.3 to -26.3 for
(7R,10R)-1. Although these fluctuations are close to the
experimental error of [R]D determination, the solvent effect is
more noticeable. It is well known that solvent generally alters
the magnitude and in rare cases even the sign of optical
rotations.14,67 Switching from CHCl3 to EtOH increased the [R]D
for (7S,10R)-1 and (7R,10R)-1 to +13.4 and -24.2, respec-
tively.68 Note that no dramatic changes in sign were observed
for either pitiamide A diastereomer. Finally, both compounds
demonstrated a moderate temperature sensitivity of the [R]D.
Cooling the CHCl3 solutions to 5 °C led to an [R]D of +10.2
for (7S,10R)-1 and -37.2 for (7R,10R)-1. These variations can
be explained by changes in conformational equilibria.69
Figure 6. N-H stretch region FT-IR absorptions for (7S,10R)-1 and
(7R,10R)-1 in CH2Cl2 at 0.001 M concentration at room temperature,
after subtraction of the spectrum of pure CH2Cl2.
In summary, although there were noticeable shifts in specific
rotation values as a consequence of solvent and temperature
changes, concentration dependence did not appear to be an issue
with pitiamide A, and none of the variations were significant
enough to cause any concern about the relevance of our
theoretical calculations.
Because the possibility of intramolecular hydrogen bonding
in pitiamide A was an issue that complicated our computational
analysis to some degree, we were interested in examining the
actual spectroscopic evidence for any intra- or intermolecular
association in solution. Amide proton temperature coefficients
(∆δ/∆T) detected by 1H NMR can provide strong evidence for
intramolecular hydrogen bonding.72 However, the large ∆δ/∆T
(-7.4 to -7.5 ppb/K in CDCl3 and -5.8 to -6.5 ppb/K in
dimethyl sulfoxide (DMSO)-d6) measured for both (7S,10R)-1
and (7R,10R)-1 are in accord with a “random coil” or a relatively
flexible backbone structure (Figure 5). This interpretation is also
supported by the relatively complex CD spectra of these
compounds. Analysis of the IR spectra leads to similar conclu-
sions (Figure 6). In the N-H stretch region, bands at >3435
cm-1 are expected to arise exclusively from solvent-exposed
amide.73 As shown in Figure 6, both synthetic pitiamide A
stereoisomers show bands at 3445 cm-1 in CH2Cl2 at 0.001 M
In many cases, CD can be used for the assignment of the
absolute configuration of organic compounds.15 In the absence
of strong, coupled chromophores or intrinsically chiral chro-
mophores,70 however, CD spectra tend to be less informative,
and in flexible molecules such as pitiamide A a stereochemical
assignment based on CD effects would be speculative at best.
Nonetheless, we determined the CD spectra for both (7S,10R)-1
and (7R,10R)-1 in cyclohexane (Figure 4). Although a relative
assignment of the configuration of the natural compound by
CD might be possible with these two reference spectra in hand,
the complexity of the CD bands illustrates the rather problematic
nature of any predictive interpretation.71
(67) Archelas, A.; Furstoss, R. J. Org. Chem. 1999, 64, 6112.
(68) However, adding small amounts (5%) of MeOH to the CHCl3
solution did not significantly alter [R]D values.
(69) Horsman, G.; Emeis, C. A. Tetrahedron Lett. 1965, 3037.
(70) (a) Forster, L. S.; Moscowitz, A.; Berger, J. G.; Mislow, K. J. Am.
Chem. Soc. 1962, 84, 4353. (b) Kirk, D. N. Tetrahedron 1986, 42, 777. (c)
Person, R. V.; Monde, K.; Humpf, H.-U.; Berova, N.; Nakanishi, K.
Chirality 1995, 7, 128. (d) Dong, J.-G.; Guo, J.; Akritopoulou-Zanze, I.;
Kawamura, A.; Nakanishi, K.; Berova, N. Chirality 1999, 11, 707.
(71) Unfortunately, we were unable to record the CD spectrum of a
natural sample of pitiamide A. Two samples that were generously provided
to us by Professor Paul decomposed during shipment from Guam to
Pittsburgh.
(72) Smith, J. A.; Pease, L. G. CRC Crit. ReV. Biochem. 1980, 8, 315.
(73) (a) Gardner, R. R.; Liang, G.-B.; Gellman, S. H. J. Am. Chem. Soc.
1999, 121, 1806. (b) Boussard, G.; Marraud, M. Biopolymers 1979, 18,
1297. (c) Maxfield, F. R.; Leach, S. J.; Stimson, E. R.; Powers, S. P.;
Scheraga, H. A. Biopolymers 1980, 18, 2507.

4614 J. Am. Chem. Soc., Vol. 122, No. 19, 2000
Ribe et al.
Table 3. Specific Rotation Values for (7S,10R)-1 and (7R,10R)-1
Based on Ab Initio Computation of the Electric Dipole-Magnetic
Dipole Polarizability Tensor and from Measurements of Synthetic
Compounds
conclude that (either because of an instrument calibration error
or a minor impurity in the natural sample) the specific rotation
of the natural product was changed by 20-30 units.76,77 Another
interpretation, albeit speculative at present, is that natural
pitiamide A was originally isolated as a mixture of diastereomers
or enantiomers. The latter hypothesis is supported by the
results of computations [R] results of total synthesis [R]
wavelength
[nm]
(7S,10R)-1
(7R,10R)-1
(7S,10R)-1
(7R,10R)-1
1
589 ([R]D)
578
546
436
365
+8
+8
-39
-41
+11.6
+12.1
+13.4
+24.2
+45.8
-30.3
-31.7
-36.2
-69.3
-131.2
observation of a set of minor H NMR signals between 2.40
and 2.45 and at 2.30 ppm for the natural sample that are
characteristic for the (7S,10R)-isomer. Although relatively rare,
natural products have previously been shown to consist of
racemic or diastereomeric mixtures.78
+9
-46
+15
+21
-71
-102
concentration,74 and bands at 3320-3410 cm-1 that would arise
from intramolecular hydrogen bonding are absent.
Conclusions
We used ab initio coupled Hartree-Fock calculations in
combination with force-field Monte Carlo conformational
analysis and Boltzmann averaging to predict off-resonance
optical rotations for all four diastero- and enantiomeric con-
figurations of the marine natural product pitiamide A. On the
basis of computations of the electric-dipole-magnetic-dipole
polarizability tensor within the CADPAC program and use of
the diagonal tensor elements to calculate the optical rotatory
parameter â, we predicted specific rotations of approximately
+39, -39, +8, and -8 for the (7S,10S)-, (7R,10R)-, (7S,10R)-,
and (7R,10S)-stereoisomers of pitiamide, respectively. This a
priori analysis proved to be quantitatively accurate by com-
parison with experimental values of (7S,10R)-1 and (7R,10R)-1
subsequently obtained by total synthesis. Highlights of the
natural product total syntheses were the preparation of the
chlorodiene segment of pitiamide A by a hydrozirconation-
transmetalation sequence and the use of a novel water-
accelerated zirconocene-catalyzed methylalumination for the
enantioselective preparation of the linchpin segment 20 as well
as a segment coupling by N-alkylation of a nosylated amide.
The dependence of the optical rotation values on solvent,
wavelength, temperature, and concentration was extensively
explored, and again calculation was found to be an accurate
tool for prediction of the ORD spectrum down to relatively short
(436 nm) wavelengths. One of the major approximations of the
computational work, the neglect of a possible intramolecular
hydrogen bonding interaction between the ketone carbonyl and
the amide NH in pitiamide A, was validated by solution NMR
and IR studies of the synthetic material. The CD curves for
(7S,10R)-1 and (7R,10R)-1 were measured and confirmed that
this powerful chiroptical technique is nonetheless unsuitable for
a stereochemical assignment of the highly flexible lipid pitiamide
Comparison of Optical Rotation Data from
Computational Analysis to Experimental Values
Total synthesis provided us with (7S,10R)- and (7R,10R)-
stereoisomers of pitiamide A, and therefore the results of the
computational optical rotation analysis could be directly com-
pared to experimental values for both compounds. As Table 3
shows, the calculation provided a priori extremely accurate
predictions of optical rotation values. It is important to note
that the synthetic material is not 100% diastereomerically pure;
minor amounts (<10%) of the (10S)-stereoisomer slightly alter
the optical rotation in addition to the normal experimental
variation.75 Within the limits of the static-field approximation,
the computations provide a quantitatively valid determination
of the ORD spectra as close to the UV as 436 nm. This is a
consequence of the absence of a strongly absorbing chro-
mophore with maxima >300 nm in compound 1. It is clear from
the results listed in Table 3 that an ab initio computational
assignment of the relative and absolute stereochemistry of
pitiamide A stereoisomers is not only qualitatively feasible but
also quantitatively accurate.
Assignment of the Stereochemistry of Natural Pitiamide
A
On the basis of the experimentally reported specific rotation
of pitiamide A, [R]D -10.3 (c 3.0, CHCl3),1 and the chiroptical
data obtained both from our computational and synthetic studies,
the configuration of the natural product should be assigned as
(7R,10S). However, much to our surprise, a careful comparison
of NMR data between the natural and the (7S,10R)-isomer
revealed subtle differences in the 1H NMR shifts that compel a
different assignment (Table 4 and Figure 7). Because NMR data
are much less susceptible to impurities, comparison of 1H NMR
spectra clearly indicates a (7R*,10R*) relatiVe stereochemistry
for the natural product. If one further assumes that the sign of
the optical rotation of the natural product is correct as
measured,1,2 the absolute configuration of natural pitiamide A
is therefore assigned as (7R,10R). All other physical data,
including all 13C NMR resonances, between the two synthetic
compounds and the natural product were identical. We can only
1
A. Finally, H NMR studies in combination with the sign of
the specific rotation value reported for pitiamide A led to a
definitive assignment of the configuration of the natural product
as (7R,10R)-1. Our work highlights the considerable structural
significance of [R]D data and the accuracy of current ab initio-
based computational chiroptical analysis, but also the potential
pitfalls with obtaining meaningful optical activity measurements
for natural product isolates.
Experimental Section
(74) Variable concentration NMR studies in CHCl3 were used to establish
that no intermolecular aggregation occurred at 0.001 M in CHCl3.
(75) Optical rotation values recorded in Table 3 represent the average
of three independent measurements. The experimental variations in these
three determinations were less than 10% of the average values.
(76) Although the vast majority of total syntheses provides optical
rotations in good agreement with data reported for the isolated natural
product, it is not uncommon that differences arise. See, for example: (a)
Edmondson, S.; Danishefsky, S. J.; Sepp-Lorenzino, L.; Rosen, N. J. Am.
Chem. Soc. 1999, 121, 2147. (b) Hanessian, S.; Cantin, L.-D.; Andreotti,
D. J. Org. Chem. 1999, 64, 4893. (c) McCauley, J. A.; Nagasawa, K.;
Lander, P. A.; Mischke, S. G.; Semones, M. A.; Kishi, Y. J. Am. Chem.
Soc. 1998, 120, 7647. (d) Wipf, P.; Xu, W. J. Org. Chem. 1996, 61, 6556.
(2R)-4-(tert-Butyl-diphenylsilanyloxy)-2-methyl-butan-1-ol (18).
To a solution of 0.314 g (4.36 mmol) of trimethylaluminum in 2 mL
of CH2Cl2 was added a solution of 0.0368 g (0.0551 mmol) of 17 in 4
mL of CH2Cl2. The resulting yellow reaction mixture was chilled to
(77) The effect of impurities on optical rotation is particularly problematic
for small rotation values: Baldwin, J. E.; Hackler, R. E.; Scott, R. M. Chem.
Commun. 1969, 1415.
(78) For recent examples, see: (a) Beuerle, T.; Engelhard, S.; Bicchi,
C.; Schwab, W. J. Nat. Prod. 1999, 62, 35. (b) Van Wagoner, R. M.; Jompa,
J.; Tahir, A.; Ireland, C. M. J. Nat. Prod. 1999, 62, 794.

Marine Natural Product Pitiamide A
J. Am. Chem. Soc., Vol. 122, No. 19, 2000 4615
Table 4. 1H NMR Shifts and Assignments for Natural Pitiamide A1 and Synthetic (7R,10R)-1 and (7S,10R)-1
position
natural pitiamide A
(7R,10R)-1
(7S,10R)-1
1
2
3
4
6.10 (d, 13.0 Hz)
6.40 (dd, 13.3, 10.8 Hz)
5.97 (dd, 15.0, 10.5 Hz)
5.64 (dt, 15.0, 7.0)
2.08-2.00 (m)
6.11 (d, 13.2 Hz)
6.41 (dd, 13.1, 10.9 Hz)
5.98 (dd, 15.1, 10.9 Hz)
5.65 (dt, 15.2, 7.0 Hz)
2.09-2.02 (m)
6.11 (d, 13.2 Hz)
6.41 (dd, 13.1, 10.9 Hz)
5.98 (dd, 15.1, 10.8 Hz)
5.65 (dt, 15.2, 7.0 Hz)
2.11-2.03 (m)
5
6a
6b
7
1.89-1.72 (m)
1.81-1.74 (m)
1.81-1.75 (m)
1.45-1.31 (m)
2.50-2.46 (m)
1.47-1.32 (m)
2.52-2.46 (m)
1.46-1.33 (m)
2.52-2.44 (m)
8
9
2.37 (dd, 6.8, 4.3 Hz)
2.10-2.00 (m)
2.39 (dd, 6.5, 4.3 Hz)
2.09-2.02 (m)
2.52-2.29 (m)
10
11
12a
12b
13
14
15
16
17
18
19
20
21
22
23
2.11-2.03 (m)
1.45-1.31 (m)
1.47-1.32 (m)
1.46-1.33 (m)
3.29 (dt, 20.0, 6.5 Hz)
3.18 (dt, 19.3, 6.5 Hz)
5.72 (bm)
3.30 (dt, 20.1, 6.6 Hz)
3.20 (dt, 19.3, 6.5 Hz)
5.73 (bm)
3.30 (dt, 20.1, 6.6 Hz)
3.19 (dt, 19.2, 6.5 Hz)
5.72 (bm)
2.23 (t, 7.3 Hz)
2.24 (t, 7.2 Hz)
2.24 (t, 7.2 Hz)
2.40-2.29 (m)
5.41 (dt, 15.3, 6.3 Hz)
5.49 (dt, 15.3, 6.4 Hz)
1.96 (q, 6.9 Hz)
1.46-1.33 (m)
0.88 (t, 7.4 Hz)
0.93 (d, 6.7 Hz)
1.07 (d, 7.0 Hz)
2.32 (q, 6.5 Hz)
5.40 (dt, 15.0, 6.3 Hz)
5.48 (dt, 15.5, 6.5 Hz)
1.95 (q, 7.0 Hz)
1.45-1.31 (m)
0.87 (t, 7.3 Hz)
0.91 (d, 6.5 Hz)
1.06 (d, 7.0 Hz)
2.33 (q, 6.5 Hz)
5.41 (dt, 15.3, 6.3 Hz)
5.49 (dt, 15.3, 6.5 Hz)
1.96 (q, 7.0 Hz)
1.47-1.32 (m)
0.88 (t, 7.4 Hz)
0.92 (d, 6.7 Hz)
1.07 (d, 7.0 Hz)
Figure 7. 1H NMR shifts for natural pitiamide A1 and synthetic (7S,10R)-1 and (7R,10R)-1.
-50 °C and then 18 µL (1 mmol) of water was added. The solution
was warmed slowly to room temperature to effect complete consump-
tion of water which was accompanied by a color change from a light
yellow to a dark orange/red color. Note: The addition of H2O to
trimethylaluminum is exothermic and cooling the mixture below -20
°C allows for a controlled reaction. After warming to room temperature,
the red/orange solution was cooled to -20 °C and then treated with
0.3350 g (1.081 mmol) of alkene 6. A change in color back to light
yellow was observed upon addition of alkene. The reaction mixture
was allowed to stir for 12 h before air was vigorously bubbled through
the solution until all volatiles were evaporated. The resulting slurry
was washed with 2 N NaOH and extracted with CH2Cl2. The organic
layer was washed with brine, dried (MgSO4), filtered, concentrated,
and chromatographed on SiO2 (EtOAc/hexane, 1:4) to yield 0.3266 g
(0.9550 mmol, 85%) of (R)-18 in 80% ee as a colorless oil. The
enantiomeric excess was determined by chiral HPLC on a Chiralcel
OD column (1% i-PrOH/hexanes): [R]D +5.6 (c 1.0, CHCl3); IR (neat)
3354, 3071, 3050, 2954, 2930, 2858, 1472, 1428, 1390, 1361, 1112,
1041, 997, 823 cm-1; 1H NMR δ 7.71-7.68 (m, 4 H), 7.48-7.38 (m,
6 H), 3.81-3.68 (m, 2 H), 3.57-3.46 (m, 2 H), 2.48 (bs, 1 H), 1.90-
1.84 (m, 1 H), 1.68-1.59 (m, 1 H), 1.56-1.47 (m, 1 H), 1.07 (s, 9 H),
0.91 (d, 3 H, J ) 6.7 Hz); 13C NMR δ 135.8, 133.6, 129.9, 127.9,
68.5, 62.7, 37.0, 34.1, 27.0, 19.3, 17.4; MS (EI) m/z (rel intensity) 285
([M-t-Bu]+, 76), 267 (22), 229 (37), 199 (100), 181 (9), 139 (12), 69
(20); HRMS (EI) Calcd for C17H21O2Si (M-C4H9) 285.1311, found
285.1317.
N-Oct-4-enoyl-(2-nitrobenzene)sulfonamide (4). A suspension of
0.524 g (13.1 mmol) of NaH (60% dispersed in mineral oil) in 30 mL
of THF was treated portionwise at 0 °C with 0.731 g (5.18 mmol) of
amide 22. After the evolution of H2 had subsided, the mixture was
warmed to room temperature and then 1.169 g (7.62 mmol) of 2-nitro-
benzenesulfonyl chloride was added. The reaction mixture was kept at
room temperature overnight before it was carefully quenched first with
H2O and then 1 N HCl. The solution was extracted with EtOAc, and
the organic layer was washed with brine, dried (MgSO4), filtered, and
chromatographed on SiO2 (EtOAc/hexanes, 1:3) to afford 1.435 g (4.403
mmol, 85%) of 4 as a light yellow solid: mp 73-75 °C; IR (neat)
3276, 2959, 2928, 2855, 1732, 1699, 1545, 1446, 1362, 1181 cm-1
;
1H NMR δ 8.72 (bs, 1 H), 8.45-8.41 (m, 1 H), 7.87-7.78 (m, 3 H),
5.48 (dt, 1 H, J ) 15.3, 6.5 Hz), 5.34 (dt, 1 H, J ) 15.4, 6.3 Hz), 2.44
(t, 2 H, J ) 6.9 Hz), 2.31 (q, 2 H, J ) 6.8 Hz), 1.91 (q, 2 H, J ) 7.0
Hz), 1.31 (s, 2 H, J ) 7.4 Hz), 0.84 (t, 3 H, J ) 7.4 Hz); 13C NMR δ
170.7, 148.3, 135.1, 134.1, 133.2, 132.8, 131.9, 127.2, 125.0, 36.7,
34.7, 27.3, 22.6, 13.8; MS (EI) m/z (rel intensity) 326 (M+, 17), 297
(26), 244 (38), 227 (47), 186 (58), 140 (31), 123 (31), 96 (48), 81
(53), 67 (54), 55 (100); HRMS (EI) Calcd for C14H18N2O5S 326.0936,
found 326.0922.
(3R,6S)-1-(tert-Butyldiphenylsilanyloxy)-12-chloro-3,6-dimethyl-
dodeca-9,11-dien-5-one (24). To a degassed solution of 0.104 g (0.230
mmol) of alkyl iodide 20 and 3 mL of Et2O was added 0.35 mL (0.51
mmol) of a 1.45 M solution of t-BuLi at -78 °C in hexanes. The
reaction mixture was kept at -78 °C for 15 min before warming to
room temperature. The solution was then immediately cooled to -78

4616 J. Am. Chem. Soc., Vol. 122, No. 19, 2000
Ribe et al.
°C and a solution of 0.0327 g of 23 (0.141 mmol) in 1 mL of Et2O
was added. The reaction mixture was kept at -78 °C for 1 h before
quenching by addition of a saturated solution of NH4Cl. After extraction
with Et2O, the organic layer was washed with brine, dried (MgSO4),
filtered, concentrated, and chromatographed on SiO2 (EtOAc/hexanes,
1:4) to yield 0.0621 g (0.123 mmol, 87%) of 24 as a colorless oil:
[R]D +13.9 (c 1.6, CHCl3); IR (neat) 3070, 2929, 2857, 1711, 1461,
(3R,6S)-N-{[12-Chloro-5-(1-ethoxy)-ethoxy]-3,6-dimethyl-dodeca-
9,11-dienyl}-2-nitro-(N-oct-4-enoyl)benzenesulfonamide ((R,S)-29).
A solution of alcohol (R,S)-28 (0.055g, 0.11 mmol) in 2 mL of freshly
distilled ethyl vinyl ether was cooled to 0 °C and a few crystals of
pyridinium p-toluenesulfonate were added. After 1 h, the mixture was
diluted with Et2O and washed with 1 N NaOH. The organic layer was
extracted with brine, dried (MgSO4), filtered, and concentrated. A
solution of the crude product in 2 mL of THF was treated with 20 µL
(0.2 mmol) of a 1 M solution of TBAF in THF. After 4 h the reaction
mixture was washed with a saturated NH4Cl solution and extracted
with Et2O. The organic layer was washed with brine, dried (MgSO4),
filtered, and concentrated. To a solution of the crude alcohol, 0.0358
g (0.136 mmol) of PPh3, and 0.0557 g (0.171 mmol) of the nosyl-
protected amide 4 in 1.5 mL of THF was added 20 µL (0.13 mmol) of
diethyl azodicarboxylate at room temperature. After 3 h, the reaction
mixture was concentrated under a fast stream of N2 and chromato-
graphed on SiO2 (EtOAc/hexanes, 1:4) to yield 0.063 g (0.098 mmol,
89% from (R,S)-28 of the nosyl-protected amide (R,S)-29 as light
yellow oil that was used without further purification.
(3R,6R)-N-{[12-Chloro-5-(1-ethoxy)-ethoxy]-3,6-dimethyl-dodeca-
9,11-dienyl}-2-nitro-(N-oct-4-enoyl)benzenesulfonamide ((R,R)-29).
Using an analogous procedure as for the preparation of (R,S)-29, 0.0530
g (0.107 mmol) of (R,R)-28 in 2 mL of freshly distilled ethyl vinyl
ether and a few crystals of pyridinium p-toluenesulfonate afforded the
ethoxy ethyl ether. The crude product was dissolved in 2 mL of THF
and treated with 20 µL (0.20 mmol) of a 1 M solution of TBAF. A
solution of the crude desilylated alcohol in 1.5 mL of THF was treated
with 0.0363 g (0.138 mmol) of triphenyl phosphine, 0.0543 g (0.166
mmol) of 4, and 20 µL (0.13 mmol) of diethyl azodicarboxylate to
afford after chromatography on SiO2 (EtOAc/hexanes, 1:4) 0.0534 g
(0.0833 mmol, 89%) of (R,R)-29 as a light yellow oil.
1
1428, 1362, 1112, 978, 703 cm-1; H NMR δ 7.68-7.66 (m, 4 H),
7.46-7.36 (m, 6 H), 6.40 (dd, 1 H, J ) 13.0, 10.8 Hz), 6.08 (d, 1 H,
J ) 13.2 Hz), 5.96 (dd, 1 H, J ) 15.2, 10.7 Hz), 5.64 (dt, 1 H, J )
15.1, 7.0 Hz), 3.69 (t, 2 H, J ) 6.6 Hz), 2.49-2.40 (m, 2 H), 2.28-
2.19 (m, 2 H), 2.05 (q, 2 H, J ) 7.4 Hz), 1.79-1.72 (m, 1 H), 1.59-
1.23 (m, 3 H), 1.06-1.04 (m, 12 H), 0.86 (d, 3 H, J ) 6.1 Hz); 13C
NMR δ 214.0, 135.8, 135.1, 134.2, 133.8, 129.8, 127.9, 127.0, 119.1,
62.2, 49.0, 46.0, 39.6, 32.2, 30.5, 27.1, 26.2, 20.3, 19.4, 16.5; MS (EI)
m/z (rel intensity) 439 ([M-t-Bu]+, 46), 219 (6), 199 (100), 169 (12),
139 (14), 126 (17), 113 (18), 101 (20), 79 (29), 57 (35); HRMS (EI)
Calcd for C26H32O2SiCl (M-C4H9) 439.1860, found 439.1853.
(3R,6R)-1-(tert-Butyldiphenylsilanyloxy)-12-chloro-3,6-dimethyl-
dodeca-9,11-dien-5-one ((R,R)-24). Analogous to the preparation of
24, 0.156 g (0.345 mmol) of alkyl iodide 20, 0.64 mL (0.77 mmol) of
a 1.2 M solution of t-BuLi in hexanes, 5 mL of Et2O, and 0.048 g
(0.21 mmol) of (R)-23 afforded 0.083 g (0.17 mmol, 81%) of (R,R)-
24 as a colorless oil: [R]D -10.8 (c 0.48, CHCl3); IR (neat) 3070,
1
2930, 2857, 1711, 1472, 1461, 1428, 1377, 1112, 979, 703 cm-1; H
NMR δ 7.68-7.66 (m, 4 H), 7.46-7.36 (m, 6 H), 6.40 (dd, 1 H, J )
13.1, 10.7 Hz), 6.08 (d, 1 H, J ) 13.2 Hz), 5.96 (dd, 1 H, J ) 15.2,
10.7 Hz), 5.64 (dt, 1 H, J ) 15.2, 6.9 Hz), 3.69 (t, 2 H, J ) 6.6 Hz),
2.51-2.19 (m, 4 H), 2.03 (q, 2 H, J ) 7.3 Hz), 1.80-1.72 (m, 1 H),
1.61-1.54 (m, 1 H), 1.49-1.32 (m, 2 H), 1.06-1.04 (m, 12 H), 0.85
(d, 3 H, J ) 6.3 Hz); 13C NMR δ 214.0, 135.8, 135.2, 134.1, 133.8,
129.8, 127.9, 126.9, 119.1, 62.2, 49.0, 46.0, 39.6, 32.0, 30.5, 27.1, 26.2,
20.2, 19.4, 16.6; MS (EI) m/z (rel intensity) 439 ([M-t-Bu]+, 68), 355
(6), 263 (21), 247 (6), 225 (12), 199 (100), 181 (16), 135 (15), 101
(17), 79 (21); HRMS (EI) Calcd for C26H32O2SiCl (M-C4H9) 439.1860,
found 439.1874.
(7S,10R)-Pitiamide A ((7S,10R)-1). To a solution of 0.024 g (0.037
mmol) of (S,R)-29 and 0.014 g (0.099 mmol) of K2CO3 in 2 mL of
DMF was added 12 µL (0.12 mmol) of thiophenol. After 2.5 h, the
reaction mixture was diluted with Et2O and passed through a pad of
basic alumina. The filtrate was washed with saturated NaHCO3 and
brine. The organic layer was dried (MgSO4) and chromatographed on
SiO2 (EtOAc/hexanes, 1:3) to yield 0.0160 g of the deprotected amide
as a colorless oil. This intermediate was immediately dissolved in 1
mL of MeOH and a few crystals of pyridinium p-toluenesulfonate were
added. After 2 h, the reaction mixture was quenched with a solution of
concentrated NaHCO3 and extracted with Et2O. The organic layer was
dried (MgSO4), filtered, and concentrated to yield a colorless oil that
was dissolved in 2 mL of CH2Cl2 and treated with 0.0220 g (0.0512
mmol) of Dess-Martin periodinane. The reaction mixture was stirred
for 3 h and then quenched with saturated NaHCO3 solution. The mixture
was extracted with Et2O, dried (MgSO4), and chromatographed on SiO2
(EtOAc/hexanes, 2:3) to afford 0.0083 g (0.022 mmol, 59%) of (7S,-
10R)-1 as a colorless oil: [R]D +11.6 (c 0.49, CHCl3); IR (neat) 3303,
(3R,6S)-1-(tert-Butyldiphenylsilanyloxy)-12-chloro-3,6-dimethyl-
dodeca-9,11-dien-5-ol ((R,S)-28). To an ice-cold solution of 0.0645 g
(0.129 mmol) of 24 in 1 mL of EtOH was added 0.0162 g (0.428 mmol)
of NaBH4. The solution was warmed to room temperature and after 3
h the mixture was carefully quenched by dropwise addition of 1 N
HCl. The mixture was extracted with Et2O, washed with brine, dried
(MgSO4), filtered, and concentrated. The residue was chromatographed
on SiO2 (EtOAc/hexanes, 1:2) to yield 0.055 g (0.11 mmol, 86%) of
(R,S)-28 as a colorless oil: IR (neat) 3436, 3070, 2930, 1583, 1472,
1
1462, 1428, 1112, 977, 823, 702 cm-1; H NMR δ 7.69-7.67 (m, 4
H), 7.46-7.34 (m, 6 H), 6.46-6.38 (m, 1 H), 6.08 (d, 1 H, J ) 13.2
Hz), 6.03-5.94 (m, 1 H), 5.75-5.64 (m, 1 H), 3.76-3.64 (m, 2 H),
3.60-3.50 (m, 1 H), 2.23-1.96 (m, 2 H), 1.82-1.11 (m, 9 H), 1.06
(s, 9 H), 0.91-0.86 (m, 6 H); 13C NMR δ 136.2, 135.8, 134.2, 134.0,
129.8, 127.8, 126.4, 118.6, 73.7, 72.8, 62.5, 62.2, 41.9, 41.4, 40.5, 38.9,
38.7, 38.6, 32.7, 31.2, 30.6, 27.1, 26.8, 26.7, 21.2, 19.7, 19.4, 15.4,
13.9; MS (EI) m/z (rel intensity) 498 (M+, 14), 441 (18), 423 (21),
309 (56), 297 (63), 269 (65), 253 (22), 199 (100), 183 (21), 78 (51);
HRMS (EI) Calcd for C30H43O2SiCl 498.2721, found 498.2721.
(3R,6R)-1-(tert-Butyldiphenylsilanyloxy)-12-chloro-3,6-dimethyl-
dodeca-9,11-dien-5-ol ((R,R)-28). Using an analogous procedure as
for the preparation of (R,S)-28, 0.0530 g (0.107 mmol) of (R,R)-24
and 0.014 g (0.37 mmol) of NaBH4 in 1 mL of Et2O afforded 0.0467
g (0.0937 mmol, 88%) of (R,R)-28 as a colorless oil: IR (neat) 3433,
2959, 2927, 2867, 1708, 1644, 1549, 1457, 1376, 1261, 977, 817 cm-1
;
1H NMR (500 MHz) δ 6.41 (dd, 1 H, J ) 13.1, 10.9 Hz), 6.11 (d, 1
H, J ) 13.2 Hz), 5.98 (dd, 1 H, J ) 15.1, 10.8 Hz), 5.87-5.67 (m, 1
H), 5.65 (dt, 1 H, J ) 15.2, 7.0 Hz), 5.49 (dt, 1 H, J ) 15.3, 6.4 Hz),
5.41 (dt, 1 H, J ) 15.3, 6.3 Hz), 3.30 (dt, 1 H, J ) 20.1, 6.6 Hz), 3.19
(dt, 1 H, J ) 19.2, 6.5 Hz), 2.52-2.44 (m, 2 H), 2.37-2.29 (m, 3 H),
2.24 (t, 2 H, J ) 7.2 Hz), 2.11-2.03 (m, 3 H), 1.96 (q, 2 H, J ) 6.9
Hz), 1.81-1.75 (m, 1 H), 1.46-1.33 (m, 5 H), 1.07 (d, 3 H, J ) 7.0
Hz), 0.93 (d, 3 H, J ) 6.7 Hz), 0.88 (t, 3 H, J ) 7.4 Hz); 13C NMR
(125 MHz) δ 214.3, 172.7, 134.9, 133.6, 131.7, 128.6, 126.9, 119.1,
48.6, 45.9, 37.3, 36.8, 36.4, 34.7, 32.0, 30.4, 28.7, 25.9, 22.6, 20.2,
16.5, 13.7; MS (EI) m/z (rel intensity) 381 (M+, 38), 346 (8), 267 (46),
210 (44), 185 (31), 168 (15), 155 (23), 126 (22), 114 (72), 87 (29), 79
(32), 67 (29), 55 (100); HRMS (EI) Calcd for C22H36NO2Cl 381.2435,
found 381.2448.
1
3070, 2930, 1583, 1472, 1462, 1428, 1111, 977, 822, 702 cm-1; H
NMR δ 7.70-7.67 (m, 4 H), 7.46-7.36 (m, 6 H), 6.42 (dd, 1 H, J )
13.0, 10.8 Hz), 6.09 (d, 1 H, J ) 13.1 Hz), 6.00 (dd, 1 H, J ) 15.1,
10.8 Hz), 5.71 (dt, 1 H, J ) 14.7, 7.3 Hz), 3.79-3.51 (m, 3 H), 2.19-
1.16 (m, 11 H), 1.06 (s, 9 H), 0.90-0.85 (m, 6 H); 13C NMR δ 136.2,
135.8, 134.2, 134.0, 129.8, 127.8, 126.4, 118.7, 73.6, 72.7, 62.4, 62.3,
42.2, 41.0, 40.7, 39.3, 39.0, 37.8, 32.8, 31.6, 30.6, 27.1, 26.8, 26.6,
20.9, 19.6, 19.4, 15.3, 13.5; MS (EI) m/z (rel intensity) 498 (M+, 31),
441 (22), 423 (17), 309 (45), 297 (16), 269 (22), 242 (30), 225 (55),
199 (100), 189 (73), 149 (30), 101 (61); HRMS (EI) Calcd for C30H43O2-
SiCl 498.2721, found 498.2728.
(7R,10R)-Pitiamide A ((7R,10R)-1). Using an analogous sequence
as for the preparation of (7S,10R)-1, a solution of 0.0400 g (0.062
mmol) of (R,R)-29 in 3 mL of DMF, 0.0180 g (0.130 mmol) of K2-
CO3, and 20 µL (0.2 mmol) of thiophenol afforded 0.0260 g (0.0571
mmol) of the deprotected amide as a colorless oil. To this oil was added
1 mL of MeOH followed by the addition of a few crystals of PPTs to
afford the deprotected alcohol as a colorless oil that was dissolved in

Marine Natural Product Pitiamide A
J. Am. Chem. Soc., Vol. 122, No. 19, 2000 4617
2 mL of CH2Cl2 and treated with 0.0430 g (0.100 mmol) of Dess-
Martin periodinane. Chromatography on SiO2 (EtOAc/hexanes, 2:3)
afforded 0.0150 g (0.039 mmol, 63%) of (7R,10R)-1 as a colorless
oil: [R]D -30.3 (c 0.71, CHCl3); IR (neat) 3299, 2959, 2928, 2873,
(18), 126 (27), 114 (100), 101 (17), 79 (30), 65 (29); HRMS (EI) Calcd
for C22H36NO2Cl 381.2435, found 381.2439.
Acknowledgment. We thank the PRF (33532-AC), NSF
(CHE-9727657 and CHE-9453461), and NIH (GM-55433) for
support of this research. We would also like to thank Professor
Valerie Paul (Guam) for spectra of natural pitiamide A and
helpful discussions.
1
1709, 1644, 1549, 1457, 1376, 1262, 977, 818 cm-1; H NMR (500
MHz) δ 6.41 (dd, 1 H, J ) 13.1, 10.9 Hz), 6.11 (d, 1 H, J ) 13.2 Hz),
5.98 (dd, 1 H, J ) 15.1, 10.9 Hz), 5.75-5.78 (m, 1 H), 5.65 (dt, 1 H,
J ) 15.2, 7.0 Hz), 5.49 (dt, 1 H, J ) 15.3, 6.5 Hz), 5.41 (dt, 1 H, J )
15.3, 6.3 Hz), 3.30 (dt, 1 H, J ) 20.1, 6.6 Hz), 3.20 (dt, 1 H, J ) 19.3,
6.5 Hz), 2.52-2.46 (m, 1 H), 2.39 (dd, 2 H, J ) 6.5, 4.3 Hz), 2.33 (q,
2 H, J ) 6.5 Hz), 2.24 (t, 2 H, J ) 7.2 Hz), 2.09-2.02 (m, 3 H), 1.96
(q, 2 H, J ) 7.0 Hz), 1.81-1.74 (m, 1 H), 1.47-1.32 (m, 5 H), 1.07
(d, 3 H, J ) 7.0 Hz), 0.92 (d, 3 H, J ) 6.7 Hz), 0.88 (t, 3 H, J ) 7.4
Hz); 13C NMR (125 MHz) δ 214.3, 172.8, 135.0, 133.7, 131.9, 128.7,
127.0, 119.2, 48.6, 46.1, 37.5, 37.0, 36.6, 34.8, 32.1, 30.5, 28.9, 26.1,
22.8, 20.3, 16.6, 13.9; MS (EI) m/z (rel intensity) 381 (M+, 24), 280
(20), 267 (73), 210 (51), 196 (43), 185 (19), 168 (18), 155 (21), 143
Supporting Information Available: Complete experimental
data for 11, 12, 13, (S)-5, (R)-5, 19, 20, 21, 22, and 23; 1H and
13C NMR spectra for all synthetic intermediates and pitiamide
A stereoisomers. 1H and 13C NMR spectra for natural pitiamide
A (PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA9945313