Diazonamide synthesis studies: Use of negishi coupling to fashion diazonamide-related biaryls with defined axial chirality

Article abstract of DOI:10.1021/ol026694t

(matrix presented) The syntheses of a bis indole and an indole salicylate with the required axial chirality for diazonamide A are reported. Atropselectivity in these biaryl systems is enforced by an sp³ stereogenic center in a lactone tether in

Full text of DOI:10.1021/ol026694t

ORGANIC
LETTERS
2002
Vol. 4, No. 20
3525-3528
Diazonamide Synthesis Studies: Use of
Negishi Coupling to Fashion
Diazonamide-Related Biaryls with
Defined Axial Chirality
Ken S. Feldman,* Kyle J. Eastman, and Guillaume Lessene
Department of Chemistry, The PennsylVania State UniVersity,
UniVersity Park, PennsylVania 16802
ksf@chem.psu.edu
Received August 7, 2002
ABSTRACT
The syntheses of a bis indole and an indole salicylate with the required axial chirality for diazonamide A are reported. Atropselectivity in these
biaryl systems is enforced by an sp3 stereogenic center in a lactone tether in both cases.
The control of biaryl atropselectivity remains an enduring
challenge in the synthesis of structurally and functionally
complex natural products biosynthesized via oxidative
coupling of component aryl residues.1 One prominent target
in this context is the marine principle diazonamide A,2,3 an
ascidian isolate notable for both its high potency against
several cancer cell lines and its architectural novelty. A truly
remarkable chapter in the diazonamide story was added
recently by Harran and co-workers,3b who showed convinc-
ingly that the natural product possesses the structure 2 and
not the previously assigned 1.
A longstanding interest in developing methods for dia-
stereoselective biaryl synthesis4 drew us to the diazonamide
problem, and initially a synthesis route was conceived that
relied on two key strategic elements: (1) late-stage biomi-
metic (?) transannular oxidative cyclization of a macrolactam
precursor 3 to set the stereochemistry at both C(10) and the
C(24)/C(26) bond and (2) establishment of the pivotal
C(16)-C(18) axial chirality via the conformational prefer-
ences dictated by an sp3 stereogenic center at C(27), 4,
Scheme 1. In light of Harran’s report, the specifics of this
approach must be modified. Nevertheless, molecular me-
chanics (MM) calculations suggest that the juxtaposition of
C(10) and C(30) in a now bis indole analogue of 3 (3.51 Å
in the low-energy conformer 3, P ) P1 ) H)5 might still be
used to advantage in a synthesis route to 2, again provided
that the central C(16)-C(18) stereochemical relationship can
be established correctly and maintained throughout the effort.
(1) Lessene, G.; Feldman, K. S. In Modern Arene Chemistry; Astruc,
D., Ed.; Wiley-VCH: New York, 2002.
(2) Lindquist, N.; Fenical, W.; Van Duyne, G. D.; Clardy, J. J. Am. Chem.
Soc. 1991, 113, 2303-2304.
(3) Leading references to the work of several groups can be found in:
(a) Li, J.; Jeong, S.; Esser, L.; Harran, P. G. Angew. Chem., Int. Ed. 2001,
40, 4765-4770. (b) Li, J.; Burgett, A. W. G.; Esser, L.; Amezeva, C.;
Harran, P. G. Angew. Chem., Int. Ed. 2001, 40, 4770-4773. (c) Nicolaou,
K. C.; Huang, X.; Guiseppone, N.; Bheema Rao, P.; Bella, M.; Reddy, M.
V.; Snyder, S. A. Angew. Chem., Int. Ed. 2001, 40, 4705-4709. (d) Bagley,
M. C.; Moody, C. J.; Pepper, A. G. Tetrahedron Lett. 2000, 41, 6901-
6904. (e) Magnus, P.; Lescop, C. Tetrahedron Lett. 2001, 42, 7193-7196.
(f) Vedejs, E.; Zajac, M. A. Org. Lett. 2001, 3, 2451-2454. (g) Hang, H.
C.; Drotleff, E.; Elliott, G. I.; Ritsema, T. A.; Konopelski, J. P. Synthesis
1999, 398-400. (h) Wipf, P.; Methot, J.-L. Org. Lett. 2001, 3, 1261-
1264. (i) Boto, A.; Ling, M.; Meek, G.; Pattenden, G. Tetrahedron Lett.
1998, 39, 8167-8170. (j) Fuerst, D. E.; Stoltz, B. M.; Wood, J. L. Org.
Lett. 2000, 2, 3521-3523.
(4) Feldman, K. S.; Sahasrabudhe K.; Quideau, S.; Hunter K. L.; Lawlor,
M. D. In Plant Polyphenols 2: Chemistry and Biology; Gross, G. G.,
Hemingway, R. W., Yoshida, T., Eds.; Plenum Publishing Corporation:
New York, 2000, pp 101-125.
10.1021/ol026694t CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/30/2002

Scheme 2. A shorter (two-step) synthesis of 7 from indole
itself (3-formylation via POCl3/DMF, 4-iodination with
Tl(TFA)3/CuI/KI) has been described,8 but inconsistent yields
Scheme 1
Scheme 2
upon scale-up and concerns about the toxicity of thallium
salts made the longer route a more attractive option. The
zincate precursor 11 can be synthesized from o-iodophenol
(9) through initial o-formylation under Skattebøl’s conditions9
followed by protection and oxidation steps. The key coupling
reaction between 8 and 11 followed the procedure of
Knochel10 as modified by Fu.11 Initial magnesiation of 11
was sensitive to reaction temperature, and optimum yields
were achieved at -30 °C without any detectable decomposi-
tion of the arylmagnesium product. Transmetalation to zinc
during slow warming from -78 °C to room-temperature
preceded addition of the indole iodide 8 and the palladium
catalyst. TLC analysis indicated that coupling proceeded
rapidly at room temperature without any substantial decom-
position of components. Some exploration of different
palladium/ligand recipes with 11 (or its PMB ether analogue)
(see Scheme 2) revealed that the Pd(t-Bu3P)2-based protocol
provided the most consistently high yields of biaryl product
13 at gram-scale reaction.
The synthesis of both biaryl 4 and biaryl 5 are described
herein. In both cases, a Negishi coupling6 is used to link
C(16) with C(18) in good yield, an approach distinct from
that pursued by other workers in this area. Subsequent
elaboration of the tether bearing a stereogenic center provides
single diastereomers of the target biaryls, in accordance with
the MM-generated expectations that a proton rather than an
NHBOC will project toward the peri-positioned aryl residue
in 4, whereas in 5, a proton rather than an NBOC2 will project
toward the adjacent indole C(2) hydrogen.
The synthesis of biaryl lactone 4 commences with N-car-
bobenzyloxy-4-iodo-indole-3-carboxaldehyde (8), which could
be prepared conveniently from gramine (6) in five steps,7
(5) A 10 000-step conformational search of 3 (P ) P1 ) H) using the
directed Monte Carlo algorithm of Macromodel 6.5 (cf.: 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) led to a family of four low-energy conformers corresponding to the
four possible atropisomers about the indicated biaryl linkages. Two of these
species were of equivalently low energy, 3 as shown and its rotomer about
the C(24)/C(26) bond. However, only 3 among these conformers brings
C(10) and C(30) within bonding distance, 3.51 Å. The enthalpies of rotation
about the C(24)/C(25) and C(16)/C(18) bonds were estimated by the
difference between the “global minimum” conformer 3 and planar versions
of 3 where the relevant biaryl dihedral was locked at either 0° or 180°, The
four values obtained (two rotational directions per bond) are shown with 3.
Only rotation about the C(24)/C(26) bond in one direction is predicted by
this analysis to be facile at room temperature (∆Hq ≈ 14 kcal/mol).
(6) Negishi, E.-i. In Metal-Catalyzed Cross-Coupling Reactions; Dieder-
ich, F., Stang, P. J., Eds.; Wiley-VCH: New York, 1998; pp 1-47.
(7) Iwao, M. Heterocycles 1993, 36, 39-32.
Continuing elaboration of the coupled product 13 required
introduction of tryptophan functionality at C(3) on the indole
(8) Somei, M.; Yamada, F.; Kunimoto, M.; Kaneo, C. Heterocycles 1984,
22, 797-801.
(9) Hofsløkken, N. U.; Skattebøl, L. Acta Chem. Scand. 1999, 53, 258-
262.
(10) (a) Rottla¨nder, M.; Palmer, N.; Knochel, P. Synlett 1996, 573-
575. (b) Boudier, A.; Bromm, L. O.; Lotz, M.; Knochel, P. Angew. Chem.,
Int. Ed. 2000, 39, 4414-4435.
(11) Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 2719-2724.
3526
Org. Lett., Vol. 4, No. 20, 2002

of 16 was obtained.15 Further studies to determine the
enantiomeric excess were deferred in light of the reorienting
results of Harran. However, acquisition of 4 as essentially a
single diastereomer serves as a proof-of-concept for the thesis
that the C(16)-C(18) bond in a putative diazonamide
precursor can be set under the influence of a proximal sp3
stereogenic center.
Scheme 3
The extension of this strategy to the bis indole 5 relevant
to the revised diazonamide A structure is shown in Scheme
4. 4,7′-Bis indoles have been identified as minor components
Scheme 4
unit, Scheme 3. Emmons-Horner extension of the aldehyde
in 13 with phosphonate 14b, itself available from the known
acid 14a,12 provided predominantly the (Z)-alkene-containing
product 15 (10:1 Z/E) in good yield. The incorporation of
the p-methoxybenzyl group in 15 was in response to the
unsatisfactory performance in subsequent steps of the more
obvious choice, a benzyl ester. The use of Schmidt’s
conditions (DBU in CH2Cl2)13 was critical for success of
this Wittig transformation, as other bases (t-BuOK, NaH,
tetramethylguanidine) either failed to deliver acceptable
yields of 15 or produced E/Z mixtures of product. Asym-
metric hydrogenation of 15 mediated by (S,S)-DuPHOS
ligated rhodium provided a ∼1.4:1 mixture of diastereomeric
4-aryl tryptophan derivatives in almost quantitative yield.
Palladium-catalyzed hydrogenolysis of the benzyl function-
alities in this intermediate furnished the phenoxy acid
derivative 16, which was cyclized under Keck’s modification
of Steglich esterification conditions to provide the desired
lactone 4 in excellent yield.14 The fact that a nearly unbiased
mixture of diastereomeric hydroxy acids 16 was converted
into a single diastereomer of lactone 4 is consistent with a
scenario wherein the biaryl linkage in 16 is freely rotating
under the reaction conditions and only one diastereomer
cyclizes rapidly to a lactone. DNOE measurements (cf. 4)
secure the relative stereochemical assignment. The absolute
stereochemistry has not been rigorously established, but an
earlier description of the DuPHOS-promoted asymmetric
hydrogenation of a dehydrotryptophan derivative suggests
that the (S)-configuration at C(27) (diazonamide numbering)
of complex mixtures derived from shotgun oligomerization
of 5,6-dihydroxyindole, but deliberate preparations of this
biaryl unit have yet to be described. Thus, application of
the Negishi protocol successful with 8 and 11 to 4,7′-bis
indole formation remained an open question. In particular,
the presumably increased steric demand of the N-salicamide
unit in 19, absent in the 8/11 coupling, posed a concern. In
the event, the known 7-iodoindole (17)16 was converted into
the coupling partner 19 without incident, and exposure of
the derived zinc reagent to 4-iodoindole 20 under palladium
catalysis provided the desired 4,7′-bis indole product 21 in
(12) Travins, J. M.; Etzkorn, F. A. J. Org. Chem. 1997, 62, 8387-8393.
(13) Schmidt, U.; Griesser, H.; Leitenberger, V.; Lieberknecht, A.;
Mangold, R.; Meyer, R.; Riedl, B. Synthesis 1992, 487-490.
(14) (a) Bolden, E. P.; Keck, G. E. J. Org. Chem. 1985, 50, 2394-
2395. (b) Neises, B.; Steglich, W. Angew. Chem., Int. Ed. Engl. 1978, 17,
522-524.
(15) Wang, W.; Xiong, C.; Yang, J.; Hruby, V. J. Tetrahedron Lett. 2001,
42, 7717-7719.
(16) Somei, M.; Saida, Y.; Funamoto, T.; Ohta, T. Chem. Pharm. Bull.
1987, 35, 3146.
Org. Lett., Vol. 4, No. 20, 2002
3527

excellent yield. In this Negishi coupling, optimum yields
attended use of P(2-furyl)3 as the stabilizing ligand, in
contrast to the P(t-Bu)3 alternative favored with the salicylate-
derived zinc reagent 12. Similar Emmons-Horner homolo-
gation of the hindered aldehyde in 21 with phosphonate 22,17
followed by rhodium(I)-mediated asymmetric hydrogenation
of the derived alkene in 23 with (R,R)-DuPHOS as a chiral
director, provided the expected tryptophan derivative. Ni-
trogen protection of the NHBOC unit within this species with
a second BOC group followed by hydrogenolysis of the
benzyl groups afforded the free phenoxy acid 24. The use
of the (R,R)-DuPHOS reagent in this instance was prompted
by the requirement for the (R)-configuration at C(27) to lock
the biaryl linkage in the desired (R)-axial chirality. As with
16, the hydrogenation product existed as a ∼1.5:1 mixture
of diastereomers.
The phenoxy acid 24 was cyclized by treatment with
Keck’s modified Steglich esterification conditions to furnish
the lactone 5 in good yield. The gross structure and
stereochemistry of the lactone 5 was ascertained by single-
crystal X-ray analysis (see Supporting Information for
details).18 As with 16, the presumably equilibrating mixture
of diastereomers 24 apparently cyclized through only the one
isomer featuring a proton (rather than the NBOC2 group)
projected toward the adjacent indole’s C(2) hydrogen. The
second BOC group on the tryptophan nitrogen provided the
necessary steric bulk to enforce formation of a single
atropisomeric lactone. Earlier studies demonstrated that when
this nitrogen bore only a single BOC group, the derived
lactone was formed as a 1:1 mixture of isomers. Chiral HPLC
analysis of 5 using an ss-Welk-01 column (95:5 hexane/2-
propanol as the eluent) provided an indirect assay of the
enantioselectivity of the DuPHOS-mediated hydrogenation.
By this protocol, the lactone 5 exists as a 75:25 ratio of
enantiomers.
Acknowledgment. Financial support from the National
Institutes of Health (GM35727) is gratefully acknowledged.
Supporting Information Available: Spectral data (1H
NMR, 13C NMR, IR, MS), combustion analysis for select
1
compounds, and copies of H NMR and 13C NMR spectra
for 4, 5, 8, 10, 11, 13, 14b, 15, 16, 19, 21, 23, and 24, HPLC
traces of lactone 5 derived from both (R,R)- and (S,S)-
DuPHOS-mediated hydrogenation of 23, a thermal ellipsoid
representation from the X-ray analysis of 5, and the ac-
companying CIF. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL026694T
(17) Horenstein, B. A.; Nakanishi, K. J. Am. Chem. Soc. 1989, 111, 6242.
(18) Lactone 5 was crystallized from ethyl acetate and pentane. The
difference Fourier map shows electron density at and around the special
position (0.50, 0.638, 0.25). However, neither ethyl acetate nor pentane
could be successfully fitted into the density present due to the crystal-
lographic symmetry involved. This feature, in addition to the high value
for Rint of 0.09 (after absorption correction), limits the crystallographic
R-factor to 0.18 for this structure. Nevertheless, the relative stereochemical
assignment of interest here is unambiguous.
3528
Org. Lett., Vol. 4, No. 20, 2002