Total synthesis of tropoloisoquinolines: Imerubrine, isoimerubrine, and grandirubrine

Article abstract of DOI:10.1021/ja0101072

A unified, ready access to the tropoloisoquinoline alkaloids imerubrine (1), grandirubrine (2), and isoimerubrine (3) is delineated and features sequential application of the intramolecular Diels - Alder reaction of an acetylene-tethered oxazole and the [

Full text of DOI:10.1021/ja0101072

J. Am. Chem. Soc. 2001, 123, 3243-3246
3243
Total Synthesis of Tropoloisoquinolines: Imerubrine, Isoimerubrine,
and Grandirubrine1
Jae Chol Lee and Jin Kun Cha*
Contribution from the Department of Chemistry, The UniVersity of Alabama, Tuscaloosa, Alabama 35487
ReceiVed January 11, 2001
Abstract: A unified, ready access to the tropoloisoquinoline alkaloids imerubrine (1), grandirubrine (2), and
isoimerubrine (3) is delineated and features sequential application of the intramolecular Diels-Alder reaction
of an acetylene-tethered oxazole and the [4 + 3] cycloaddition of an oxyallyl. A regioselective synthesis of 1
was achieved by stereo- and regioselective oxidation of an 8-oxabicyclo[3.2.1]oct-6-en-3-one cycloadduct by
means of the Moriarty method. Such a post-cycloaddition functionalization complements the synthetic utility
of an R-alkoxy-substituted oxyallyl so as to broaden the scope of the oxyallyl [4 + 3] cycloaddition reaction.
Imerubrine (1), isolated from the plants Abuta imene and
Abuta refescens of the Menispermacae family, was shown by
X-ray analysis to possess an unusual tropolone ether structure
(Figure 1).2 Grandirubrine (2), the corresponding free tropolone,
was subsequently isolated from Abuta grandifolia.3 This unique
class of naturally occurring tropoloisoquinoline alkaloids pres-
ently includes isoimerubrine (3), pareirubrines A and B (4 and
5), and pareitropone (6).4 These alkaloids 1-6 are structurally
similar to colchicine (7) and its congeners which are the only
previously known tropoloisoquinoline alkaloids.5 The biosyn-
thetic pathways of their shared tropolone moieties were postu-
lated to involve fused cyclopropane intermediates.6 Such a
biosynthesis might also account for the presence of the more
widespread azafluoranthene alkaloids in the same Abuta plants.2b
Cytotoxic properties were reported for some of these alkaloids,
in particular 6,4 which might well be related to the well-known
antimitotic properties of 7. Both 4 and 5 were known to
preferentially crystallize as the keto tautomers shown in Figure
1.
These alkaloids pose considerable synthetic challenges, which
are in part due to the paucity of general methods for preparing
the fused tropolone ring. To date, only two total syntheses of 1
Figure 1.
and 2 were recorded: the Banwell group utilized an acid-
promoted ring expansion of an appropriate homo-o-benzo-
quinone for the regioselective syntheses.7 Boger’s syntheses
were based on the [4 + 2] cycloaddition reaction of a
cyclopropenone ketal and an R-pyrone.8 Another noteworthy
thread of both of these elegant syntheses is characterized by
the development of synthetic strategies which take advantage
of the close structural relationship between these tropoloiso-
quinoline alkaloids and colchicine (7).9 Herein we detail the
regiocontrolled syntheses of 1-3 by extension of our previously
reported synthesis of (-)-7,10 which constitutes a unified entry
to these structurally related alkaloids.
* Address correspondence to this author. E-mail: jcha@bama.ua.edu.
(1) Part 13 in the series of synthetic studies on [4 + 3] cycloadditions
of oxyallyls. See also: (a) Part 12: Lee, J. C.; Cha, J. K. Tetrahedron
2000, 56, 10175. (b) Part 11: Sung, M. J.; Lee, H. I.; Chong, Y.; Cha, J.
K. Org. Lett. 1999, 1, 2017.
(2) (a) Cava, M. P.; Buck, K. T.; Noguchi, I.; Srinivasan, M.; Rao, M.
G.; DaRocha, A. I. Tetrahedron 1975, 31, 1667. (b) Silverton, J. V.; Kabuto,
C.; Buck, K. T.; Cava, M. P. J. Am. Chem. Soc. 1977, 99, 6708.
(3) Menachery, M. D.; Cava, M. P. Heterocycles 1980, 14, 943.
(4) (a) Morita, H.; Matsumoto, K.; Takeya, K.; Itokawa, H.; Iitaka, Y.
Chem. Pharm. Bull. 1993, 41, 1418. (b) Morita, H.; Matsumoto, K.; Takeya,
K.; Itokawa, H.; Iitaka, Y. Chem. Lett. 1993, 339. (c) Morita, H.; Matsumoto,
K.; Takeya, K.; Itokawa, H. Chem. Pharm. Bull. 1993, 41, 1478. (d) Itokawa,
H.; Matsumoto, K.; Morita, H.; Takeya, K. Heterocycles 1994, 37, 1025.
(e) Morita, H.; Takeya, K.; Itokawa, H. Bioorg. Med. Chem. Lett. 1995, 5,
597.
(5) For general reviews, see: (a) Buck, K. T. Alkaloids (Academic Press)
1984, 23, 301. (b) Wildman, W. C.; Pursey, B. A. Alkaloids (Academic
Press) 1968, 11, 407. (c) Capraro, H.-G.; Brossi, A. Alkaloids (Academic
Press) 1984, 23, 1. (d) Boye´, O.; Brossi, A. Alkaloids (Academic Press)
1992, 41, 125. (e) Le Hello, C. Alkaloids (Academic Press) 2000, 53, 287.
(6) (a) Battersby, A. R.; McDonald, E.; Stachulski, A. V. J. Chem. Soc.,
Perkin Trans. 1 1983, 3053. (b) Sheldrake, P. W.; Suckling, K. E.;
Woodhouse, R. N.; Murtagh, A. J.; Herbert, R. B.; Barker, A. C.; Staunton,
J.; Battersby, A. R. J. Chem. Soc., Perkin Trans. 1 1998, 3003.
(7) (a) Banwell, M. G.; Hamel, E.; Ireland, N. K.; Mackay, M. F.
Heterocycles 1994, 39, 205. (b) Banwell, M. G.; Ireland, N. K. J. Chem.
Soc., Chem. Commun. 1994, 591. (c) Banwell, M. G. Pure Appl. Chem.
1996, 68, 539.
(8) Boger, D. L.; Takahashi, K. J. Am. Chem. Soc. 1995, 117, 12452.
(9) For related syntheses of 7, see: (a) Boger, D. L.; Brotherton, C. E.
J. Am. Chem. Soc. 1986, 108, 6713. (b) Banwell, M. G.; Lambert, J. N.;
Mackay, M. F.; Greenwood, R. J. J. Chem. Soc., Chem. Commun. 1992,
974.
10.1021/ja0101072 CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/17/2001

3244 J. Am. Chem. Soc., Vol. 123, No. 14, 2001
Lee and Cha
Scheme 1
Scheme 3
Scheme 2
Preparation of Furan 8. Our synthesis began with the known
and readily available 5,6,7-trimethoxyisoquinoline (11).12 The
iodide 12 was required for the Sonogashira coupling with
trimethylsilylacetylene.13 Unfortunately, 11 proved to resist
direct iodination, although acid-catalyzed bromination (NBS,
H2SO4 catalyst) had successfully been achieved by Boger.12
Thus, the iodo functionality was installed prior to the isoquino-
line construction to afford 12 uneventfully by a slight modifica-
tion of Boger’s procedure (Scheme 3). Alkylation of N-p-
toluenesulfonylaminoacetaldehyde dimethyl acetal (15) with
3,4,5-trimethoxybenzyl bromide (16), followed by iodination14
of 14, provided the aryl iodide 13 in 90% overall yield.
Subsequent cyclization of 13 was best achieved in two steps to
give 12 (75-80% yield) by treatment with 6 N HCl-dioxane,
followed by base-induced elimination of p-toluenesulfonic acid
(t-BuOK, room temperature). On the other hand, direct acid-
mediated cyclization of 13 (in a 6 N HCl solution) resulted in
12 in only 10-30% yield, while the undesired 4-p-toluene-
sulfonylisoquinoline (structure not shown) was obtained in 60-
70% yield. Acetylene 17 was then prepared in 84% overall yield
by means of the Sonogashira reaction with trimethylsilylacet-
ylene and subsequent removal of the silyl group by n-Bu4NF.
The Reissert intermediate 18 was next prepared (96%) by
employing (Boc)2O-KCN to set the stage for the necessary
introduction of the oxazole moiety to the isoquinoline 17.
DIBAL-H reduction of the cyano group of 18 to the aldehyde
19 proved to be challenging, probably due to steric congestion.
Ultimately, 19 was obtained in 63% yield by employing an
excess (2.5-5.0 equiv) of DIBAL-H at -78 °C in toluene.
Treatment with TosMIC according to the procedure of van
Results and Discussion
Retrosynthetic Analysis. In parallel with the total synthesis
of (-)-7 (Scheme 1),10 consecutive application of the intramo-
lecular Diels-Alder reaction of an acetylene-tethered oxazole,
the [4 + 3] cycloaddition of an oxyallyl, and double elimination
of the resulting cycloadduct seemed to promise easy access to
1-3 and also other members of this family. The regiocontrolled
introduction of the tropolone and tropolone ether subunits should
be readily available by employing an R-alkoxy-substituted
oxyallyl or the Moriarty oxidation of the unsubstituted [4 + 3]
cycloadduct, as previously demonstrated in the preparation of
simpler derivatives (e.g., hinokitiol).11 Although both furans 8
and 9 could undergo the key oxyallyl cycloaddition reaction, 8
presented itself as an ideal advanced intermediate (Scheme 2):
the regioselective introduction of the tropolone functionality was
anticipated to be more facile than with the furan 9; and the
intramolecular Diels-Alder reaction of an acetylene-tethered
oxazole 10 (as 10a or 10b) was also ideally suited for a
convenient construction of the tetracyclic furan 8, but not for
9. Of particular interest was the investigation of the regiochem-
istry of the [4 + 3] cycloaddition of the highly functionalized
furan 8 and an R-alkoxy oxyallyl in view of scant literature
precedents. Finally, the preparation of 10 was in turn expected
to be straightforward by means of the Sonogashira reaction and
subsequent elaboration.
(12) (a) Boger, D. L.; Brotherton, C. E.; Kelley, M. D. Tetrahedron 1981,
37, 3977. (b) Boger, D. L.; Brotherton, C. E. J. Org. Chem. 1984, 49, 4050.
(13) (a) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975,
4467. Cf.: (b) Thorand, S.; Krause, N. J. Org. Chem. 1998, 63, 8551.
(14) Janssen, D. E.; Wilson, C. V. Organic Synthesis; Wiley: New York,
1963; Collect. Vol. IV, p 547.
(10) (a) Lee, J. C.; Jin, S.-j.; Cha, J. K. J. Org. Chem. 1998, 63, 2804.
(b) Reference 1a.
(11) Lee, J. C.; Cho, S. Y.; Cha, J. K. Tetrahedron Lett. 1999, 40, 7675.

Total Synthesis of Tropoloisoquinoline
J. Am. Chem. Soc., Vol. 123, No. 14, 2001 3245
Leusen15 provided oxazole 10b (where P ) Boc) in 64% yield.
The pivotal intramolecular Diels-Alder-retro-Diels-Alder
reactions16 of 10b and concomitant elimination of the Boc group
proceeded smoothly by thermolysis (o-dichlorobenzene, reflux)
to furnish the requisite tetracyclic furan 8 in 85-90% yield.
Scheme 4
Oxyallyl [4 + 3] Cycloaddition and Double Elimination.
The key [4 + 3] cycloaddition reaction of 8 was achieved by
adaptation of Albizati’s procedure involving in situ generation
(with TMSOTf) of the R-methoxy trimethylsiloxyallyl cation
from the trimethylsilyl enol ether 20 of pyruvic aldehyde
dimethyl acetal (Scheme 4).17-19 Not surprisingly, a 1:1 mixture
of the desired cycloadduct 21 (26%) and the regioisomer 22
(29%) were isolated, along with 42% of recovered starting
material.20 The overall yield and the material balance were
optimal when the cycloaddition reaction was allowed to proceed
to ∼50% conversion. Additionally, each cycloadduct proved
to be a single diastereomer. The regiochemistry of these
cycloadducts was unequivocally established from the splitting
pattern (an AB quartet) of the methylene protons at C-11 or
C-9 (imerubrine numbering). The stereochemistry of 21 was
assigned on the basis of the diagnostic vicinal coupling constant
(J ) 5.0 Hz),21 with which the exo proton at C-9 is coupled to
the bridgehead proton at C-8. This stereochemical assignment
was consistent with that of Albizati’s previous examples
involving simple furans17 and can be rationalized by the
“compact” (endo-like) transition state of the W-shaped oxyallyl
cation as depicted in Scheme 4.19 On the other hand, the
stereochemistry of 22 could not be determined due to the
absence of the vicinal coupling constant for the proton at C-11.
It is interesting to recall that the cognate cycloaddition
reaction of the identical R-methoxy trimethylsiloxyallyl cation
in the previous synthesis of (-)-colchicine took place with an
exceptional level of regioselectivity,10 where the regiodirecting
influence of the C-3 aryl substituent of the furan substrate must
outweigh that of the C-2 alkyl moiety. Thus, lack of regiocontrol
in the cycloaddition of 8 can be attributed to the presence of
the two aryl groups at C-2 and C-3 of the furan functionality,
which are anticipated to exert comparable directing power.
Finally, as demonstrated in the previous synthesis of (-)-
colchicine, treatment of 21 with excess amounts of TMSOTf
and Et3N in CH2Cl2 by a slight modification of the Fo¨hlisch
and Mann methods22 gave imerubrine (1) in 76% yield. The
synthetic substance was found to exhibit identical physical and
spectroscopic data to those reported for the natural product, as
well as the synthetic material.2,7,8 In contrast, 22 proved resistant
toward elimination of the ether bridge under identical conditions.
While speculative at this juncture, it is tempting to suggest that
the unexpected failure of 22 to undergo ring opening might be
attributed to the axial orientation of the methoxy group at C-11,
which should impede the requisite enolization, an obligatory
step for double elimination of the oxa bridge.
Regioselective Synthesis of 1. To improve on the nonregio-
selective cycloaddition reaction of 8 and R-methoxy oxyallyl,
a regiocontrolled synthesis of 1 would seem attainable by
elaboration of the unsubstituted [4 + 3] cycloadduct. The
Moriarty oxidation23 of 8-oxabicyclo[3.2.1]oct-6-en-3-one com-
pounds was previously shown to be very sensitive to steric
effects, and the desired regioisomer was thus anticipated to be
the major, if not the sole, product (vide infra).11 Toward this
(15) van Leusen, A. M.; Hoogenboom, B. E.; Siderius, H. Tetrahedron
Lett. 1972, 2369.
(16) An efficient assembly of fused-ring furans from acetylene-tethered
oxazoles has been amply demonstrated by Jacobi and co-workers: Jacobi,
P. A.; Blum, C. A.; DeSimone, R. W.; Udodong, U. E. S. J. Am. Chem.
Soc. 1991, 113, 5384 and references therein.
(17) Murray, D. H.; Albizati, K. F. Tetrahedron Lett. 1990, 31, 4109.
(18) Until recently the preparation of R-heteroatom-substituted oxyallyl
cations has received relatively scant attention: (a) Sasaki, T.; Ishibashi,
Y.; Ohno, M. Tetrahedron Lett. 1982, 23, 1693. (b) Fo¨hlisch, B.; Krimmer,
D.; Gehrlach, E.; Ka¨shammer, D. Chem. Ber. 1988, 121, 1585. (c) Reference
17. (d) Stark, C. B. W.; Eggert, U.; Hoffmann, H. M. R. Angew. Chem.,
Int. Ed. Engl. 1998, 37, 1266. See also: (e) Harmata, M.; Jones, D. E.
Tetrahedron Lett. 1997, 38, 3861. (f) Harmata, M.; Jones, D. E.; Kahraman,
M.; Sharma, U.; Barnes, C. L. Tetrahedron Lett. 1999, 40, 1831. See also:
(g) Harmata, M.; Fletcher, V. R.; Claassen, R. J., II J. Am. Chem. Soc.
1991, 113, 9861. (h) Walters, M. A.; Arcand, H. R.; Lawrie, D. J.
Tetrahedron Lett. 1995, 36, 23. (i) Walters, M. A.; Arcand, H. R. J. Org.
Chem. 1996, 61, 1478. For a recent review on this important area, see: (j)
Harmata, M. Recent Res. DeV. Org. Chem. 1997, 1, 523.
(19) For general reviews of the oxyallyl [4 + 3] cycloadditons, see: (a)
Noyori, R.; Hayakawa, Y. Org. React. 1983, 29, 163. (b) Hoffmann, H. M.
R. Angew. Chem., Int. Ed. Engl. 1984, 23, 1. (c) Mann, J. Tetrahedron
1986, 42, 4611. (d) Rigby, J. H.; Pigge, F. C. Org. React. 1997, 51, 351.
(20) The regiochemistry of the oxyallyl cycloaddition in the synthesis
of colchine (7) is controlled by the C-3 aryl moiety of the furan in preference
to the C-2 alkyl group (Scheme 1). In contrast, the two aryl groups at C-2
and C-3 in the cycloaddition reaction of 8 are expected to exert comparable
directing effects in opposition to each other. In the synthesis of 7, there
was also an unusual directing influence by the amino protecting group.
(21) (a) Hoffmann, H. M. R.; Clemens, K. E.; Smithers, R. H. J. Am.
Chem. Soc. 1972, 94, 3940. (b) Takaya, H.; Makino, S.; Hayakawa, Y.;
Noyori, R. J. Am. Chem. Soc. 1978, 100, 1765.
(22) (a) Fo¨hlisch, B.; Sendelbach, S.; Bauer, H. Liebigs Ann. Chem. 1987,
1. (b) Barbosa, L.-C. A.; Mann, J.; Wilde, P. D. Tetrahedron 1989, 45,
4619. (c) For an excellent review on ring-opening reactions of oxabicylic
compounds, see: Chiu, P.; Lautens, M. Top. Curr. Chem. 1997, 190, 1.
(23) (a) Moriarty, R. M.; Hu, H.; Gupta, S. C. Tetrahedron Lett. 1981,
22, 1283. (b) Moriarty, R. M.; Prakash, O.; Vavilikolanu, P. R.; Vaid, R.
K.; Freeman, W. A. J. Org. Chem. 1989, 54, 4008. (c) Moriarty, R. M.;
Berglund, B. A.; Penmasta, R. Tetrahedron Lett. 1992, 33, 6065.

3246 J. Am. Chem. Soc., Vol. 123, No. 14, 2001
Lee and Cha
Scheme 5
Scheme 6
A convenient synthesis of grandirubrine (2) was directly
available (62%) by ring opening of the R-hydroxy ketone 27,
which was in turn prepared (96%) by acidic hydrolysis of 25
with no evidence of potential tautomerization (Scheme 6).
Finally, as previously reported by Cava, Itokawa, and Boger,3,4d,8
treatment of 2 with TMSCHN2 resulted in a 1:1 mixture of 1
and 3. The synthetic substances 1-3 were shown to exhibit the
identical physical and spectroscopic data to those reported for
the natural products, as well as the synthetic materials.2,7,8
end, the [4 + 3] cycloadduct 24 was first secured in 73% overall
yield by way of the Fo¨hlisch reaction24 (Scheme 5). It was
pleasing that the Moriarty oxidation [i.e., treatment with
(diacetoxyiodo)benzene in methanolic potassium hydroxide] of
ketone 24 indeed gave 25 as the sole regioisomer in 83% yield.
In passing, we note that the stereochemical outcome of the
Moriarty oxidation of 8-oxabicyclo[3.2.1]oct-6-en-3-ones comple-
ments that of the Rubottom-type oxidation of the corresponding
silyl enol ether or of the related direct hydroxylation of the
enolate.25 In any event, 25 was uneventfully converted (92%)
by O-methylation and subsequent acidic hydrolysis to the
R-methoxyketone 21, which was identical in all aspects to that
which was previously prepared by means of an R-methoxy
oxyallyl (Scheme 4). Thus, the sequential application of the
Fo¨hlisch cycloaddition, the Moriarty oxidation, and ring opening
afforded a regioselective synthesis of 1.
Conclusion
The [4 + 3] cycloaddition reaction of an oxyallyl to a suitably
functionalized furan, followed by double elimination of the
resulting oxa bridge, has provided ready access to the tropolo-
isoquinoline alkaloids, imerubrine, grandirubrine, and isoimeru-
brine (1-3), as well as colchicine (7). The requisite furan
substrate was readily prepared by the intramolecular Diels-
Alder reaction of an acetylene-tethered oxazole. Additionally,
stereo- and regioselective oxidation of the 8-oxabicyclo[3.2.1]-
oct-6-en-3-one cycloadducts by means of the Moriarty method
provided a regioselective synthesis of 1 and broadened the scope
of the oxyallyl [4 + 3] cycloaddition.
Acknowledgment. This work is dedicated to the memory
of Professor Arthur G. Schultz. We thank the National Institutes
of Health for generous financial support (GM35956).
Total Synthesis of 2 and 3. Double elimination of 24 under
typical conditions gave granditropone (26), which had previously
been prepared by Boger8 to serve as the key precursor to 1-3.
Supporting Information Available: Experimental proce-
dure and the 1H and 13C NMR spectra of all synthetic
intermediates (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
(24) Sendelbach, S.; Schwetzler-Raschke, R.; Radl, A.; Kaiser, R.; Henle,
G. H.; Korfant, H.; Reiner, S.; Fo¨hlisch, B. J. Org. Chem. 1999, 64, 3398
and references therein.
(25) For related chemistry, see also: (a) Bunn, B. J.; Cox, P. J.; Simpkins,
N. S. Tetrahedron 1993, 49, 207. (b) Gethin, D. M.; Simpkins, N. S.
Tetrahedron 1997, 53, 14417.
JA0101072