Total synthesis of antofine using the net [5+5]-cycloaddition of γ,δ-unsaturated carbene complexes and 2-alkynylphenyl ketones as a key step

Article abstract of DOI:10.1021/jo061053n

A compound containing all of the carbons of the anticancer agent antofine was produced in a single step from the coupling of a γ,δ-unsaturated carbene complex with a 2-alkynylphenyl ketone derivative. Subsequent conversion to antofine was effected in three steps.

Full text of DOI:10.1021/jo061053n

SCHEME 1
Total Synthesis of Antofine Using the Net
[5+5]-Cycloaddition of γ,δ-Unsaturated Carbene
Complexes and 2-Alkynylphenyl Ketones as a
Key Step
Alejandro Camacho-Davila and James W. Herndon*
Department of Chemistry and Biochemistry, New Mexico State
UniVersity, MSC 3C, Las Cruces, New Mexico 88003
jherndon@nmsu.edu
ReceiVed May 23, 2006
SCHEME 2
A compound containing all of the carbons of the anticancer
agent antofine was produced in a single step from the cou-
pling of a γ,δ-unsaturated carbene complex with a 2-alky-
nylphenyl ketone derivative. Subsequent conversion to
antofine was effected in three steps.
Antofine (1, Scheme 1) is a potent anticancer agent1 of the
phenanthroindolizidine alkaloid family of natural products.2 The
anticancer activity of antofine arises through inhibition of protein
and nucleic acid synthesis.3 Antofine also displays other
medicinal properties, including antibiotic and antifungal activ-
ity,4 and antiviral activity.5 Antofine has been the subject of
numerous synthetic investigations. Most of the syntheses of
antofine have involved the preparation of substrates where the
middle aromatic ring of phenanthrene is closed in an early step
of the synthesis. Diverse strategies have been employed for
appending the pyrrolidine ring system. These methods include
strategies based on the alkylation of pyrrole,6 alkylation of
proline,7 use of a pyrrole-substituted Wittig-like reagent,8 and
ortho metalation.9 Other strategies include a pyridine-based
strategy10 and a cyclohexadienone photorearrangement ap-
proach.11 Enantioselective approaches based on alkene meta-
thesis have recently been reported.12
Synthesis of the hydrophenanthrene ring systems through the
coupling of γ,δ-unsaturated carbene complexes with 2-alky-
nylphenylbenzoyl systems in a net [5+5]-cycloaddition process
was recently reported (Scheme 2).13 This complex tandem
process involves carbene-alkyne coupling (A + B to C),
followed by isobenzofuran formation (C to D), followed by
(1) (a) Tanner, U.; Wiegrebe, W. Arch. Pharm. Chem. Life Sci. 1993,
326, 67-72. (b) Stark, D.; Lykkeberg, A. K.; Christensen, J.; Budnik, B.
A.; Abe, F.; Jaroszewski, J. W. J. Nat. Prod. 2002, 65, 1299-1302. (c)
Lee, S. K.; Nam, K. A.; Heo, Y. H. Planta Med. 2003, 69, 21-25. (d)
Lou, H.; Li, X.; Chu, X.; Jiang, Z.; Zhao, Y. Shandong Yike Daxue Xuebao
1995, 33, 158-62. (e) Fang, S.; Zhang, R.; Chen, Y.; Xu, C.; Lu, S. Zhiwu
Xuebao 1989, 31, 934-938.
(2) For a recent review of this class of compounds, see: Li, Z.; Jin, Z.;
Huang, R. Synthesis 2001, 2365-2378.
(8) Lebrun, S.; Couture, A.; Deniau, E.; Grandclaudon, P. Tetrahedron
1999, 55, 2659-2670.
(3) Xi, Z.; Zhang, R.; Yu, Z.; Ouyang, D.; Huang, R. Bioorg. Med. Chem.
Lett. 2005, 15, 2673-2677.
(9) Iwao, M.; Mahalanabis, K. K.; Watanabe, M.; De Silva, S. O.;
Snieckus, V. Tetrahedron 1983, 39, 1955-1962.
(4) (a) Baumgartner, B.; Erdelmeier, C. A. J.; Wright, A. D.; Rali, T.;
Sticher, O. Phytochemistry 1990, 29, 3327-3330. (b) Capo, M.; Saa, J. M.
J. Nat. Prod. 1989, 52, 389-390.
(10) Ciufolini, M. A.; Roschangar, F. J. Am. Chem. Soc. 1996, 118,
12082-12089.
(11) Bhakuni, D. S.; Gupta, P. K. Indian J. Chem., B 1982, 21B, 393-
(5) An, T.; Huang, R.; Yang, Z.; Zhang, D.; Li, G.; Yao, Y.; Gao, J.
Phytochemistry 2001, 58, 1267-1269.
395.
(12) (a) Kim, S.; Lee, J.; Lee, T.; Park, H. G.; Kim, D. Org. Lett. 2003,
5, 2703-2706. (b) Kim, S.; Lee, T.; Lee, E.; Lee, J.; Fan, G.; Lee, S. K.;
Kim, D. J. Org. Chem. 2004, 69, 3144-3149. (c) For synthesis of the
enantiomer, see ref 7b.
(13) For the most recent example, see: Li, R.; Zhang, L.; Camacho-
Davila, A.; Herndon, J. W. Tetrahedron Lett. 2005, 46, 5117-5120.
(6) Govindachari, T. R.; Ragade, I. S.; Viswanathan, N. J. Chem. Soc.
1962, 1357-1360.
(7) (a) Chauncey, B.; Geller, E. Aust. J. Chem. 1970, 23, 2503-2513.
(b) Faber, L.; Weigerebe, W. HelV. Chim. Acta 1973, 56, 2882-2884. (c)
Faber, L.; Wiegerebe, W. HelV. Chim. Acta 1976, 59, 2201-2212.
10.1021/jo061053n CCC: $33.50 © 2006 American Chemical Society
Published on Web 07/25/2006
6682
J. Org. Chem. 2006, 71, 6682-6685

SCHEME 3
SCHEME 4
intramolecular Diels-Alder reaction (D to E), followed by
opening of the oxanorbornene ring to afford G. Use of this
reaction for the synthesis of antofine will require that the
oxanorbornene intermediate undergo an alternative pathway,
dehydration (E to F).14 This pathway was observed in cases
where the Y group is strongly electron donating.15
SCHEME 5
In this manuscript, the use of the process in Scheme 2 as the
key step in a short total synthesis of antofine will be described.
The retrosynthetic analysis for antofine is depicted in Scheme
1. Coupling of ketone 3 and butenylcarbene complex 4 will
afford dihydrophenanthrene 2a, assuming that the presence of
the electron-rich aromatic ring will result in dehydration
(conversion of E to F in Scheme 2). The cyclization precursor
3 should be readily prepared from commercially available
6-bromo-3,4-dimethoxybenzaldehyde 5 through alkynylation
and introduction of the pyrrolidine group. Successful transfor-
mation of initial adduct 2a to secondary amine-phenanthrene
2b, the penultimate synthetic intermediate in several of the
previous total syntheses, would constitute a formal total
synthesis of antofine. Simple condensation of amine 2b with
formaldehyde leads to antofine in high yield.7b,8,12
The alkyne-ketone-carbamate 3 was rapidly assembled
through the sequence of reactions in Scheme 3. Commer-
cially available 6-bromoveratraldehyde (5)16 was transformed
to the ketone 7 via Sonogashira coupling17 followed by
conversion of the alkyne-aldehyde 6 to the corresponding
methyl ketone. For the introduction of the pyrrolidine unit, only
the silyl enol ether-acyliminium ion coupling depicted in
Scheme 3 was successful. The synthesis of the carbomethoxy-
protected adduct 3a was more efficient than the attempted
synthesis of Boc-protected analogue 3b. Methods using proline-
derived sources of the heterocyclic ring were also attempted
but were unsuccessful (Scheme 4). Coupling of carbonyl anion
equivalents derived from aldehyde 6 (e.g., 9a and 9b) and
Boc-protected homoproline iodide or tosylate (10) was unsuc-
cessful despite demonstrated success with similar compounds.18
Methods based on the addition of the organolithium reagent
derived from 12 to Weinreb amide 13a or the corresponding
nitrile 13b were similarly unsuccessful.19
The key step of the reaction (coupling of 3a and 4, Scheme
5) led to dihydrophenanthrene 14 in 66% yield. As expected,
the presence of the electron-rich aromatic ring in 3a resulted in
only a dehydration product, dihydrophenanthrene 14. The
trimethylsilyl substituent was also removed in the key carbene
complex coupling step, possibly during chromatographic puri-
fication.20 Dehydrogenation to form phenanthrene 15 was
effected by treatment with palladium on carbon in refluxing
xylene. The reaction employing DDQ afforded a complex
mixture of products. Formal total synthesis of antofine required
only the removal of the carbomethoxy protecting group from
15 to afford 2b (see Scheme 1). This proved to be very
challenging, and no conditions were found where this process
(14) This pathway is frequently observed for isobenzofuran-alkene
Diels-Alder adducts. Friedrichsen, W. AdV. Heterocycl. Chem. 1999, 73,
1-96.
(19) Nucleophilic addition of hydride to the nitrile has been reported;
however, we are not aware of successful addition of aryllithium reagents.
Toujas, J. L.; Jost, E.; Vaultier, M. Bull. Soc. Chim. Fr. 1997, 134, 713-
717.
(20) This is anticipated to be a facile process because of the strain
associated with forcing a bulky trimethylsilyl group into the bay region
and because protonation of the silicon-bearing carbon of compound 2
(Scheme 1) would afford a carbocation stabilized by an R-methoxy group
and a â-silyl group. For a recent reference to alkoxyvinylsilanes, see:
Barluenga, J.; Alvarez-Garcia, L. J.; Romanelli, G. P.; Gonzalez, J. M.
Tetrahedron Lett. 1997, 38, 6763-6766. These authors did not note any
unusual acid sensitivity for these compounds.
(15) Ghorai, B. K.; Herndon, J. W. Organometallics 2003, 22, 3951-
3957.
(16) It is far more economical to brominate veratraldehyde for large-
scale reactions (Br2 in acetic acid). Charlton, J. L.; Alauddin, M. M. J.
Org. Chem. 1986, 51, 3490-3493.
(17) This is a known compound. Martin, N.; Altable, M.; Filippone, S.;
Martin-Domenech, A.; Poater, A.; Sola, M. Chem.-Eur. J. 2005, 11, 2716-
2729.
(18) Jones, K.; Woo, K. C. Tetrahedron 1991, 47, 7179-7184.
J. Org. Chem, Vol. 71, No. 17, 2006 6683

1H NMR (CDCl3): δ 0.25 (s, 9H), 1.70-1.93 (m, 3H), 2.20 (m,
1H), 3.20-3.57 (m, 4H), 3.60 (s, 3H), 3.94 (s, 6H), 4.39 (m, 1H),
6.97 (s, 1H), 7.42 (nearly coalescing singlets,27 1H). 13C NMR
(CDCl3): δ -0.4, 22.9*, 23.7*, 31.0*, 31.9*, 46.3*, 46.7, 52.1,
54.1*, 54.7*, 56.0, 56.2, 104.7, 111.3*, 115.4*, 116.2, 133.7, 142.4,
149.4, 151.5*, 155.3*, 198.7. IR (neat): 2147 (m), 1698 (s), 1668
(m), 1592 (m), 1512 (m) cm-1. MS (EI): 403 (M, 33), 388 (23),
372 (12), 344 (18), 330 (73), 301 (12), 261 (64), 149 (54), 128
(95), 69 (100). HRMS: Calcd for C21H29NO5Si, 403.1815; found,
403.1804. *These peaks are broadened due to dynamic processes.
Synthesis of Dihydrophenanthrene 14 through Coupling of
Alkyne-Ketone 3a with Carbene Complex 4. To a refluxing
solution of alkyne-ketone 3a (568 mg, 1.40 mmol) in dioxane (15
mL) was added over a 1 h period a solution of carbene complex
425 (450 mg, 1.55 mmol) in dioxane (15 mL). After the addition
was complete, the reflux was continued for 24 h and the mixture
was then allowed to cool to room temperature. The dioxane was
removed on a rotary evaporator, and the residue was suspended in
hexanes-ethyl acetate and filtered through Celite. The filtrate was
concentrated under reduced pressure, and the residue was purified
by flash chromatography on silica gel using 1:1 hexane/ethyl acetate
as eluent. Dihydrophenanthrene 14 was isolated as a white powder
could be accomplished in respectable yield. An alternative
synthetic route was explored that employs the conversion of
carbamate 15 to the lactam 16 through the Bischler-Napieralski
reaction. Classical conditions using phosphorus oxychloride
failed in this reaction; however, the use of triflic anhydride
allowed for successful ring closure.21 The resulting amide was
utilized in previous total syntheses of antofine,9,11 and treatment
with lithium aluminum hydride completed the total synthesis.
In summary, a short total synthesis that affords racemic22
antofine in 23% overall yield in seven steps23 from commercially
available chemicals has been presented. The key step in this
reaction is the carbene complex coupling in Scheme 5, which
incorporates all of the carbons of antofine in a net [5+5]-
cycloaddition process.
Experimental Section24
Preparation of Ketone 3a. To a solution of ketone 725 (2.00 g,
7.23 mmol) in dichloromethane (25 mL) was added triethylamine
(2 mL). This mixture was cooled to 0 °C in an ice-water bath, and
a solution of trimethylsilyl trifluoromethanesulfonate (1.7 mL, 8.8
mmol) dissolved in dichloromethane (5 mL) was added dropwise
over 5 min. The mixture was stirred at 0-5 °C for 30 min and at
room temperature for 1 h. The reaction mixture was then diluted
with hexanes (75 mL) and poured into saturated aqueous sodium
bicarbonate solution (25 mL). The organic phase was washed with
water (4 × 25 mL), dried over sodium sulfate, and concentrated
on a rotary evaporator to afford 2.49 g (98%) of silyl enol ether 17
as an oil, which was used in the next step without purification. 1H
NMR (CDCl3): δ 0.20 (s, 9H), 0.22 (s, 9H), 3.86 (s, 6H), 4.65 (d,
1H, J ) 1.3 Hz), 5.15 (d, 1H, J ) 1.3 Hz), 6.93 (s, 1H), 7.01 (s,
1H). 13C NMR (CDCl3): δ -0.2, 0.0, 55.7, 55.8, 96.2, 96.7, 105.1,
110.5, 112.5, 115.7, 134.3, 147.9, 149.0, 153.6. IR (neat): 2147
(m), 1600 (s) cm-1. To a solution of 8a26 (1.095 g, 6.88 mmol)
1
(380 mg, 66%). Mp 173-175 °C. H NMR (CDCl3): δ 1.60-
2.00 (m, 4H), 2.47 (t, 2H, J ) 4.0 Hz), 2.57 (m, 1H), 2.97 (t, 2H,
J ) 4.0 Hz), 3.34 (m, 1H), 3.45-3.75 (m, 2H), 3.76 (s, 1.5 H, one
rotamer of R2NCOOCH3), 3.85 (s, 3H), 4.03 (s, 6H), 4.15 (s, 1.5H,
one rotamer of R2NCOOCH3), 4.27 (m, 1H), 6.09 (s, 1H), 6.93 (s,
1H), 7.28 (s, 1H), 7.97 (s, 1H). 13C NMR (CDCl3): δ 23.5, 27.7,
29.1, 29.5, 37.9, 46.6, 52.1, 54.8, 55.8, 56.5, 57.9, 91.9, 102.1,
104.9, 125.0, 126.8, 127.6, 128.1, 130.1, 149.0, 155.7, 160.8. IR
(neat): 1693 (s), 1636 (m) cm-1. MS (EI): 411 (M, 40), 283 (100),
268 (3), 128 (77). HRMS: Calcd for C24H29NO5, 411.2036; found,
411.2045.
Formation of Phenanthrene 15 through Dehydrogenation of
14. Compound 14 (0.841 g, 2.00 mmol) was dissolved in xylenes
(15 mL), and 10% palladium on carbon (0.5 g) was added. The
mixture was refluxed for 48 h. Concentration under reduced pressure
and purification by flash column chromatography eluting with
hexane/ethyl acetate 3:2 on silica gel afforded 0.67 g (80%) of
1
phenanthrene 15. Mp 177-180 °C. H NMR (CDCl3): δ 1.83-
2.05 (m, 2H), 2.16 (m, 1H), 2.43 (m, 1H), 2.91 (dd, 1H, J ) 15.7,
13.5 Hz), 3.54 (dd, 1H, J ) 15.8, 4.5 Hz), 3.80-4.00 (m, 4H),
4.01 (s, 3H), 4.06 (s, 6H), 4.12 (s, 3H), 7.26 (dd, 1H, J ) 9.3, 2.4
Hz), 7.32 (s, 1H), 7.85 (d, 1H, J ) 2.4 Hz), 7.90 (s, 1H), 9.30 (d,
1H, J ) 9.3 Hz). 13C NMR (CDCl3): δ 14.1, 23.4, 28.9, 38.2,
46.5, 52.0, 55.4, 55.9, 56.6, 57.4, 60.2, 103.6, 103.8, 106.3, 115.2,
124.5, 125.9, 127.0, 129.4, 130.5, 148.7, 149.8, 155.6, 157.9, 170.9.
IR (neat): 1693 cm-1. MS (EI): 409 (M+, 15), 281 (8), 128 (100).
HRMS: Calcd for C24H27NO5, 409.1892; found, 409.1889.
Formation of Amide 16 through Bischler-Napieralski Cy-
clization of 15. To a cooled (0 °C) solution of 15 (200 mg, 0.48
mmol) in dichloromethane (25 mL) containing (dimethylamino)-
pyridine (176 mg, 2.40 mmol, 3 equiv) was added over 15 min a
solution of trifluoromethanesulfonic anhydride (677 mg, 2.4 mmol,
3 equiv) in dichloromethane (10 mL). The solution was stirred for
16 h while the ice-water bath was kept in place but without any
further addition of ice. The reaction was diluted with dichlo-
romethane (20 mL) and washed successively with saturated aqueous
sodium carbonate solution (10 mL), 20% aqueous acetic acid (10
mL), and again with saturated aqueous sodium carbonate solution.
This solution was dried over sodium sulfate and concentrated under
reduced pressure. Purification by flash column chromatography on
silica gel eluting with 100% ethyl acetate gave 108 mg (60%) of
and silyl enol ether 17 (2.76 g, 7.91 mmol, 1.15 equiv) in
dichloromethane (25 mL) at -78 °C and under an argon atmosphere
was added trimethylsilyl trifluoromethanesulfonate (0.50 mL, 2.6
mmol). The resulting purple solution was stirred for 1 h at -78
°C, and saturated aqueous sodium bicarbonate solution (25 mL)
was added. The reaction mixture was allowed to warm to room
temperature, then extracted with dichloromethane (3 × 15 mL) and
dried over sodium sulfate. After concentration, the residue was
purified by flash chromatography on silica gel, eluting with
hexanes-ethyl acetate (3:2). A yellowish syrup identified as
alkyne-ketone-carbamate 3a (3.17 g, 100% yield) was obtained.
(21) Banwell, M. G.; Bisset, B. D.; Busato, S.; Cowden, C. J.; Hockless,
D. C. R.; Holman, J. W.; Read, R. W.; Wu, A. W. Chem. Commun. 1995,
2551-2553.
(22) Because of the failed reactions noted in Scheme 4, use of the
synthesis in this manuscript for preparation of optically pure antofine must
await advances in the enantioselective addition of acetophenone-derived
silyl enol ethers to acyliminium ions. A report of moderately enantioselective
additions (53% ee in best case) to acyliminium ions recently appeared.
Onomura, O.; Ikeda, T.; Matsumura, Y. Heterocycles 2005, 66, 81-86.
(23) The overall synthesis is six steps from known compound 6, seven
steps from commercially available 6-bromoveratraldehyde (5), and eight
steps from cheap 3,4-dimethoxybenzaldehyde.
(27) This assignment was based on the different appearance of the 60,
200, and 400 MHz proton NMR spectra. The two peaks at δ 7.42 appear
as a single peak in the 60 MHz NMR and as two broad singlets in the 200
MHz spectrum. This broadening is likely due to rotation about the aryl-
CO C-C bond.
(24) For a general experimental, see the Supporting Information.
(25) See Supporting Information.
(26) Boto, A.; Hernandez, R.; Suarez, E. J. Org. Chem. 2000, 65, 4930-
4937.
6684 J. Org. Chem., Vol. 71, No. 17, 2006

compound 16. Mp 265-270 °C (color is significantly darker at
mp). 1H NMR (CDCl3): δ 1.80-2.00 (m, 2H), 2.13 (m, 1H), 2.40
(m, 1H), 2.80 (dd, 1H, J ) 15.0, 13.0 Hz), 3.44 (dd, 1H, J ) 15.0,
4.0 Hz), 3.78-3.85 (m, 3H), 3.99 (s, 3H), 4.01 (s, 3H), 4.09 (s,
3H), 7.22 (s, 1H), 7.23 (dd, 1H, J ) 9.1, 2.5 Hz), 7.79 (d, 1H, J )
2.5 Hz), 7.82 (s, 1H), 9.28 (d, 1H, J ) 9.1 Hz). 13CNMR (CDCl3):
δ 23.5, 32.5, 33.8, 45.2, 55.1, 55.4, 55.86, 55.94, 56.1, 103.8,
104.1, 104.9, 115.2, 123.5, 124.3, 126.7, 129.6, 130.9, 132.5, 149.5,
150.1, 157.7, 164.2. IR(neat): 1631 cm-1. LRMS: 377 (M+, 100),
308 (60), 280 (42), 265 (10), 222 (8), 188 (15). HRMS Calcd for
C23H23NO4, 377.1618; found, 377.1627.28
Antofine (1). To a solution of amide 16 (20 mg, 0.052 mmol)
in THF (10 mL) was added lithium aluminum hydride (100 mg,
2.63 mmol), and the mixture was heated to reflux under an argon
atmosphere for 3 h. The reaction was cooled to room temperature,
and 3 N aqueous sodium hydroxide solution (5 mL) was added.
The mixture was saturated with solid sodium sulfate, and the solids
were extracted with THF (5 × 5 mL). Concentration under reduced
pressure and purification of the residue by preparative TLC on silica
gel eluting with dichloromethane/ methanol (10:1) afforded racemic
antofine as a yellowish solid.29 1H NMR (CDCl3): δ 1.79 (m, 1H),
1.93 (m, 1H), 2.02 (m, 1H), 2.26 (m, 1H), 2.50 (m, 1H), 2.92 (dd,
IH, J ) 13.8, 10.2 Hz), 3.34 (dd, 1H, J ) 13.8, 1.8 Hz), 3.48 (br
t, 1H, J ) 6.5 Hz), 3.63 (m, 1H), 3.72 (d, 1H, J ) 14.7 Hz), 4.01
(s, 3H), 4.06 (s, 3H), 4.11 (s, 3H), 4.70 (d, 1H, J ) 14.7 Hz), 7.20
(dd, 1H, J ) 9.0, 2.5 Hz), 7.30 (s, 1H), 7.81 (d, 1H, J ) 9.0 Hz),
7.86 (d, 1H, J ) 2.5 Hz), 7.91 (s, 1H). 13C NMR (CDCl3): δ 21.6,
31.2, 33.5, 53.7, 55.0, 55.4, 55.9, 56.0, 60.2, 103.9, 104.1, 104.8,
114.8, 123.5, 124.1, 124.2, 125.5, 126.5, 127.0, 130.2, 148.3, 149.3,
157.4. The spectral data were in agreement with those previously
reported for this compound.12a
Acknowledgment. This work was supported by the SCORE
program of NIH. The Universidad Autonoma de Chihuahua and
PROMEP (Mexico) provided a fellowship to A.C.D. High-
resolution mass spectra were acquired at the University of
California at Riverside or through the Nebraska Mass Spec
Facility.
Supporting Information Available: Synthetic procedures for
compounds 4-7. Photocopies of proton and C-13 NMR spectra
for compounds 1, 3a, 7, and 14-17. This material is available free
of charge via the Internet at http://pubs.acs.org.
JO061053N
(28) The spectral data are not in agreement with that reported for this
compound in ref 9. A closer examination reveals that the spectral data
reported for compound 16 in ref 9 are actually the spectral data for antofine.
(29) The yield for this process is reported as quantitative (see ref 9).
Our yield was 75%; however, we only attempted this reaction once.
J. Org. Chem, Vol. 71, No. 17, 2006 6685