Intramolecular diyl trapping reactions en route to the bicyclo [3.2.1] framework; an approach to aphidicolin

Article abstract of DOI:10.1016/j.tetlet.2009.06.083

We describe an application of the intramolecular diyl trapping cycloaddition reaction to the assembly of the bicyclo [3.2.1] framework, and utilize the outcome to complete a formal total synthesis of aphidicolin.

Full text of DOI:10.1016/j.tetlet.2009.06.083

Tetrahedron Letters 50 (2009) 4994–4997
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Intramolecular diyl trapping reactions en route to the bicyclo [3.2.1]
framework; an approach to aphidicolin
*
Wei Zhong, R. Daniel Little
Department of Chemistry and Biochemistry, University of California, Santa Barbara, CA 93106-9510, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We describe an application of the intramolecular diyl trapping cycloaddition reaction to the assembly of
the bicyclo [3.2.1] framework, and utilize the outcome to complete a formal total synthesis of aphidicolin.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 11 May 2009
Revised 1 June 2009
Accepted 16 June 2009
Available online 21 June 2009
Keywords:
TMM diyl
Trimethylenemethane
Aphidicolin
Intramolecular diyl trapping
Cycloaddition
The trimethylenemethane diyl (TMM) made its first appearances
in the literature in 1948 and 1950.1 Longuet-Higgins prediction that:
‘alternate hydrocarbons with no classical structure of the K-type
should be paramagnetic in the ground state’,2 provided an apropos
description of TMM. Years later, Dowd reported the first synthesis
and ESR spectral characterization of TMM.3 A report from his group
appearing in 1976 established unambiguously that TMM does pos-
sess a triplet ground state, as predicted.4 Berson et al. initiated a de-
tailed and systematic investigation of the reactivity of
cyclopentadiyls of the TMM variety, with their first reports appear-
ing in 1971.5 We entered the field with a publication describing
the use of an intermolecular cycloaddition that we refer to as a ‘diyl
trapping reaction’, to construct the linearly fused tricyclopentanoid
framework that is common to a host of bioactive natural products.6
Through the efforts of many, the reactivity pattern for TMM diyls has
emerged.7 Thus, in addition to both intermolecular and intramolec-
ular cycloaddition, the diyls can engage in atom transfer processes,8
cut DNA,9 and undergo several interesting fragmentation–recombi-
nation processes.10
and have formulated the following simple guidelines: (a) linearly
fused cycloadducts arise in those instances where an electron-
withdrawing group is appended to either of the sp2-hybridized car-
bons of the diylophile (note structure 2), (b) in contrast, the posi-
tioning of an alkyl group on the internal carbon of the diylophile
leads preferentially to the bridged framework, and (c) the larger
the group (e.g., R2 = CH3 vs CH(OR)2 vs C(OR)2CH3), the greater
the preference.7
R1
R1
CH3
R2
N
N
1
2
CH3
A striking observation made by Carroll and Little in 198111 was
that the intramolecular cycloaddition of the diyl derived from dia-
zene 1 led to the formation of an approximately 1:1 mixture of the
linearly fused and bridged cycloadducts, 3 and 4, respectively. Prior
to that time, only the linear pathway had been observed. Eventu-
ally, we learned how to generate either skeletal type by design,
H3C
4, bridged
3, linearly fused
The ability to gain access to either framework intrigued us suf-
ficiently to wonder whether TMM diyl chemistry could be applied
to the synthesis of a natural product wherein the bicyclo [3.2.1]
subunit that is found in the bridged adducts constituted an
* Corresponding author. Tel.: +1 805 893 3693; fax: +1 805 893 4120.
E-mail address: little@chem.ucsb.edu (R.D. Little).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.06.083

W. Zhong, R. D. Little / Tetrahedron Letters 50 (2009) 4994–4997
4995
important structural feature of the target. To explore this notion,
we set our sights upon the anti-tumor agent aphidicolin (5).12 It
has long been the focus of practitioners of total synthesis, and their
efforts have led to notable achievements.13 We selected diazene 6
as the diyl precursor of interest. If our guidelines are correct then
the presence of the large hetero-alkyl substituent appended to
the diylophile will ensure the preferential formation of the bridged
framework. Once formed, our plan calls for an oxidative cleavage of
chloroethoxycarbonyl) azo dicarboxylate set the stage for conver-
sion of the biscarbamate unit to the diazene linkage. This was
most efficiently accomplished electrochemically using a controlled
potential electrolysis at a potential of À1.65 V (vs Ag/AgCl).16 Oxi-
dation of the resulting hydrazine using cupric chloride delivered
diazene 6. This protocol proved to be superior to the use of sapon-
ification followed by a ferricyanide oxidation and further high-
lights the utility of electrochemistry.17
the C–C
p
-bond in 7 and the selective conversion of one of the two
With diazene 6 in hand, the stage was set to convert it to the
TMM diyl and therefore, to test the utility of our guidelines. The
intramolecular diyl trapping reaction was initiated simply by
refluxing a 2 mM solution of diazene 6 dissolved in acetonitrile.18
Once TLC analysis showed that the starting material had disap-
peared, the reaction mixture was cooled to room temperature
and the solvent was removed. We were pleased to discover the for-
mation of the bridged cycloadduct as the major product, thereby
adding validity to the predictive power of our guidelines. Chro-
matographic separation of the ꢀ10:1 mixture of bridged to linearly
fused cycloadducts led to the isolation of the desired bridged
framework 7a,b in 80–83% yields.19 In addition to possessing the
primary structural features of the C,D ring system of the natural
product, one finds embedded within 7a,b the carbons destined to
become carbon numbers 5-10, as well as the C-10 angular methyl
(natural product numbering).
ketals to the methyl ketone found in structure 8. We focus upon
structure 8 since it has previously been converted to the natural
product,13m and because it presents a challenging [3.2.1] frame-
work upon which to apply the diyl chemistry and test our guide-
lines. We describe our results herein.
OH
OH
H
O
O
8
O
10
H
6
HO
H
OH
5, aphidicolin
OH
O
8
O
O
O
O
O
O
CH3CN
reflux
O
O
O
N
N
8
N
N
O
O
O
O
80-83%
H
O
O
7a,b
6
7
6
While the yield was high, we were disappointed by the forma-
tion of diastereomers at C-8. This outcome is likely to be a conse-
quence of the stepwise nature of the cycloaddition arising from
the triplet diyl. Undoubtedly, the first bond is formed via a 6-endo
trig cyclization of diyl 12 leading to 13. Sigma bond formation at
either the top or the bottom face of the resulting allylic framework
in 13, accounts for the stereochemical outcome.
Diazene 6 was assembled via a sequence that began with the si-
lyl ether 9 of glycidol (Scheme 1). Ring opening of the epoxide was
accomplished using the dianion of 3-methyl-3-buten-2-ol. Subse-
quent Doering–Parikh oxidation followed by installation of the ke-
tal units and desilylation provided an efficient route to structure
10.14 A Swern oxidation followed by reaction with sodium cyclo-
pentadienide delivered fulvene 11.15 Diimide reduction of the
Diels–Alder adduct formed via the reaction of 11 with bis(2,2,2-tri-
O
O
O
O
O
O
O
O
O
O
TBDPSO
1 - 4
O
O
HO
O
12
13
9
10
5, 6
O
Having established the utility of the intramolecular diyl trap-
ping reaction for the assembly of the [3.2.1] skeleton, we pro-
ceeded to convert the cycloadducts to structure 8, an aphidicolin
convergent point.13m Toward that end, ozonolytic cleavage of the
O
O
O
O
OO
O
O
7, 8
p-bond in 7a,b, followed by a reductive workup, selective protec-
N
N
tion of the primary hydroxyl group, and oxidation using PCC/Celite
afforded a 1:1 mixture of diastereomeric ketones 14a,b. Once the
desired epimer was identified,20 the unwanted form 14b was con-
verted to a 1:3 mixture of 14a to 14b upon treatment with DBN
and pyridine; the isomers were then separated, and the desired
form was carried forward. Clearly the need to effect epimerization
is a shortcoming of our pathway21 (Scheme 2).
6
11
Scheme 1. Reagents and conditions: (1) n-BuLi, Et2O, TMEDA; 3-methyl-3-buten-2-
ol, Et2O, then add epoxide 9 in THF (91–93%); (2) SO3–pyridine, DMSO, TEA, DCM
(88–95%); 3. HOCH2CH2OH, p-TsOH, PhCH3 (87–90%); (4) 1 M TBAF, THF (80–83%);
(5) Swern; and then (6) 60% NaH, CpH, THF (68–79%, two steps); (7)
(Cl3CCH2O2CN@)2, DCM then (KO2CN@)2, AcOH (80–85%); (8) À1.65 V (vs Ag/AgCl,
Hg/Mg), DMF; CuCl2 (90%).
To complete the sequence, we elected to reduce the ketone and
convert the resulting alcohol to a xanthate ester prior to affecting a

4996
W. Zhong, R. D. Little / Tetrahedron Letters 50 (2009) 4994–4997
6. Little, R. D.; Bukhari, A.; Venegas, M. G. Tetrahedron Lett. 1979, 20, 305–308.
O
7. (a) Little, R. D. Chem. Rev. 1996, 96, 93–114; (b) Allan, A.; Carroll, G. L.; Little, R.
D. Eur. J. Org. Chem. 1998, 1, 1–12.
O
O
O
8. (a) Carroll, G. L.; Kim Allan, A.; Schwaebe, M. K.; Little, R. D. Org. Lett. 2000, 2,
2531–2534; (b) Maiti, A.; Gerken, J. B.; Masjedizadeh, M. R.; Mimieux, Y. S.;
Little, R. D. J. Org. Chem. 2004, 69, 8574–8582; (c) Billera, C. F.; Little, R. D. J. Am.
Chem. Soc. 1994, 116, 5487–5488.
1 - 3
O
O
O
O
H
H
O
9. (a) Bregant, T. M.; Groppe, J.; Little, R. D. J. Am. Chem. Soc. 1994, 116, 3535–
3636; (b) Spielmann, H. P.; Fagan, P. A.; Bregant, T. M.; Little, R. D.; Wemmer, D.
E. Nucleic Acids Res. 1995, 23, 1576–1583.
10. (a) Mikesell, P. J.; Little, R. D. Tetrahedron Lett. 2001, 42, 4095–4097; (b) Carroll,
G. L.; Harrison, R.; Gerken, J.; Little, R. D. Tetrahedron Lett. 2003, 44, 2109–2112.
11. (a) Little, R. D.; Carroll, G. L. Tetrahedron Lett. 1981, 22, 4389–4392; (b) Little, R.
D.; Carroll, G. L. J. Am. Chem. Soc. 1984, 105, 928–932.
14a, α-isomer
14b, β-isomer
4
7a,b
OTBDPS
Scheme 2. Reagents and conditions: (1) (a) O3, MeOH/DCM, À78 °C; (b) NaBH4,
MeOH/H2O, reflux; (2) TBDPSCl, TEA, DCM, (70–75% two steps); (3) PCC/Celite, DCM
(>98%); (4) DBN, pyr CH2Cl2, reflux (1:3 14a to 14b after 6 h), then combine
materials and recycle.
12. Brundret, K. M.; Dalziel, W.; Hesp, B.; Jarvis, J. A. J.; Neidle, S. J. Chem. Soc., Chem.
Commun. 1972, 1027–1028.
13. (a) Trost, B. M.; Nishimura, Y.; Yamamoto, K. J. Am. Chem. Soc. 1979, 101, 1328–
1330; (b) Corey, E. J.; Tius, M. A.; Das, J. J. Am. Chem. Soc 1980, 102, 1742–1744;
(c) McMurry, J. E.; Andrus, A.; Ksander, G. M.; Musser, J. H.; Johnson, M. A.
Tetrahedron 1981, 37, 319–327; (d) Ireland, R. E.; Godfrey, J. D.; Thaisrivongs, S.
J. Am. Chem. Soc. 1981, 103, 2446–2448; (e) Van Tamelen, E. E.; Zawacky, S. R.;
Russell, R. K.; Carlson, J. G. J. Am. Chem. Soc. 1983, 105, 142–143; (f) Bettolo, R.;
Tagliatesta, P.; Lupi, A.; Bravetti, D. Helv. Chim. Acta 1983, 66, 1922–1928; (g)
Ireland, R. E.; Dow, W. C.; Godfrey, J. D.; Thaisrivongs, S. J. Org. Chem. 1984, 49,
1001–1013; (h) Tanis, S. P.; Chuang, Y. H.; Head, D. B. Tetrahedron Lett. 1985, 26,
6147–6150; (i) Holton, R. A.; Kennedy, R. M.; Kim, H. B.; Krafft, M. E. J. Am.
Chem. Soc. 1987, 109, 1597–1600; (j) Lupi, A.; Patamia, M.; Marini Bettolo, R.
Helv. Chim. Acta 1988, 71, 872–875; (k) Rizzo, C. J.; Smith, A. B., III Tetrahedron
Lett. 1988, 29, 2793–2796; (l) Rizzo, C. J.; Smith, A. B., III J. Chem. Soc., Perkin
O
O
O
O
O
1 - 5
O
O
H
O
H
7
8
14a
OTBDPS
OH
Trans.
1 1991, 969–979; (m) Toyota, M.; Nishikawa, Y.; Fukumoto, K.
Scheme 3. Reagents and conditions: (1) NaBH4, MeOH (99%); (2) (a) NaH; CS2; MeI,
THF (91%); (3) (n-Bu)3SnH, AIBN, PhH (91%); (4) CSA, acetone (92%); (5) TBAF, THF
(89%).
Tetrahedron 1994, 50, 11153–11166; (n) Toyota, M.; Nishikawa, Y.;
Fukumoto, K. Tetrahedron Lett. 1994, 35, 6495–6498; (o) Tanaka, T.; Okuda,
O.; Murakami, K.; Yoshino, H.; Mikamiyama, H.; Kanda, A.; Kim, S.-W.; Iwata, C.
Chem. Pharm. Bull. 1995, 43, 1407–1411; (p) Bilodeau, F.; Dube, L.;
Deslongchamps, P. Tetrahedron 2003, 59, 2781–2791; (q) Justicia, J.; Oltra, J.
E.; Cuerva, J. M. J. Org. Chem. 2005, 70, 8265–8272.
radical initiated cleavage of the C–O bond.22 Subsequent conver-
sion of the bridgehead ketal to a carbonyl unit was readily achieved
using catalytic quantities of CSA in wet acetone. That it reacts in
preference to the ketal appended to the 3-carbon bridge is
undoubtedly a consequence of the strain release that accompanies
the conversion from sp3 to sp2 hybridization with the attendant re-
moval of the crowded vicinal quaternary centers. A routine fluo-
ride-induced removal of the silyl group led efficiently to the
desired structure 8. Verification that we had reached the conver-
gent point in our synthetic sequence was achieved by comparing
our spectral data with those reported in the literature13m,23
(Scheme 3).
In conclusion, the results reported herein clearly indicate that
the intramolecular diyl trapping reaction can indeed be used to
assemble the bicyclo [3.2.1] framework that is common to bioac-
tive natural products. The results substantiate the principle that
this framework will be produced when the internal carbon of the
diylophile is appended with a large alkyl group. Judicious selection
of this group allows functional group manipulation once the cyclo-
addition has been completed.
14. Parikh, J. P.; Doering, W. E. J. Am. Chem. Soc. 1967, 89, 5505–5507.
15. Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651–1660.
16. (a) Little, R. D.; Carroll, G. L. J. Org. Chem. 1979, 44, 4720; (b) Schwaebe, M. K.;
Little, R. D. Electrochim. Acta 1997, 42, 2201–2203.
17. (a) Sperry, J. B.; Wright, D. L. Chem. Soc. Rev. 2006, 35, 605–621; (b) Moeller, K.
D. Tetrahedron 2000, 56, 9527–9554.
18. The low concentration is used to assure that triplet dimerization is not
competitive with cycloaddition.
19. The ratio of bridged to linearly fused cycloadducts was determined by 1H NMR.
20. The desired epimer was identified by comparing our data with those of Toyota,
Nishkawa, and Fukuoto for structure 8. See Ref. 13m.
21. We suspect, but have not proven, that through additional experimentation a
substantially more effective epimerization could be identified. Since the total
synthesis was not our major objective, we opted not to explore the options.
22. Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin Trans. 1, 1975, 1574–
1585. In the present instance: To a solution of NaH (8 mg, 0.2 mmol, 60% in
mineral oil) in THF (0.5 mL) at 0 °C was added alcohol (45 mg, 0.079 mmol) in
THF (0.5 mL) dropwise. The reaction mixture was allowed to warm to room
temperature and stirred for 2 h. Carbon disulfide (15
lL, 0.25 mmol) was added
and stirred for 30 min. MeI (15 L, 0.24 mmol) was added rapidly to the
l
resulting yellow solution and stirred for 30 min. The reaction mixture was
poured into H2O (1 mL), and extracted with DCM (3 Â 2 mL). The combined
extracts were dried over Na2SO4, filtered, and then concentrated to dryness in
vacuo. The residue was purified by column chromatography over silica gel
(pentane/ether, 3:1, v/v) to afford xanthate (47 mg, 0.072 mmol, 91%) as a
white foam. To a solution of xanthate (47 mg, 0.072 mmol) in dry benzene
(1 mL) that was bubbled with argon for 30 min were added tri-n-butyltin
hydride (31 lL, 0.12 mmol) and AIBN (2 mg, 0.012 mmol). The resulting
Acknowledgments
solution was refluxed for 2 h and one more portion of tri-n-butyltin hydride
and AIBN were added and refluxed for 2 more hours, and then cooled to room
temperature. The solution was concentrated to dryness in vacuo and quenched
with satd. NaHCO3 (2 mL), extracted with ether (3 Â 2 mL), dried over Na2SO4,
filtered, and then concentrated to dryness in vacuo. The residue was purified
by column chromatography over silica gel (pentane/ether, 3:1, v/v) to afford
the deoxygenated compound (36 mg, 0.065 mmol, 91%) as a colorless oil.
23. The spectral data for structure 8: 1H NMR (CDCl3, 400 MHz): d 4.00–3.83 (m,
4H), 3.65 (t, 1H, J = 6.4), 2.38–2.30 (m, 1H), 2.22–2.17 (m, 1H), 2.15 (s, 3H),
2.14–2.10 (m, 1H), 2.03–1.99 (m, 1H), 1.89–1.76 (m, 3H), 1.68–1.59 (m, 4H),
1.57–1.44 (m, 2H), 1.40–1.26 (m, 3H). 13C NMR (CDCl3, 100 MHz): d 212.9,
110.9, 64.6, 64.1, 62.7, 57.1, 43.2, 42.6, 40.4, 32.5, 31.8, 29.4, 26.5, 26.4, 26.1.
ESI-MS: found 291.1555, C15H24O4Na+ Calculated 291.1572. For structure 10,
1H NMR (CDCl3, 200 MHz): d 5.18 (d, 1H, J = 1.0), 4.81 (d, 1H, J = 1.4), 3.96–
3.70 (m, 8H), 3.46 (s, 2H, 2.57 (br, 1H), 2.11–2.03 (m, 2H), 1.84–1.75 (m, 2H),
1.40 (s, 3H). 13C NMR (CDCl3, 50 MHz): d 148.5, 109.9, 109.8, 109.2, 65.4, 65.0,
64.1, 33.3, 24.1, 23.8. For fulvene 11, 1H NMR (CDCl3, 200 MHz): d 6.85–6.80
(m, 1H), 6.58–6.53 (m, 1H), 6.47–6.43 (m, 1H), 6.22 (s, 1H), 6.17–6.13 (m,
1H), 5.24 (d, 1 H, J = 1.2), 4.86 (d, 1 H, J = 1.4), 4.02–3.77 (m, 8 H), 2.27–2.17
(m, 2 H), 2.09–1.99 (m, 2H), 1.48 (s, 3H). 13C NMR (CDCl3, 50 MHz): d 148.6,
145.4, 139.1, 134.7, 131.6, 126.3, 121.1, 110.7, 109.3, 64.6, 64.3, 37.8, 24.2.
For diazene 6, 1H NMR (CDCl3, 200 MHz): d 5.79 (s, 1H), 5.24 (d, 1H, J = 1.0),
The Supported Activity is sponsored by an educational donation
provided by Amgen. The authors are grateful for their support.
R.D.L. is grateful to the former students who set the stage for gen-
erating the guidelines referred to in the text, particularly Drs. Car-
roll, Dannecker-Doerig, Masjedizadeh, Moeller, and Ott.
References and notes
1. (a) Coulson, C. A. J. Chim. Phys. 1948, 45, 243–248; (b) Longuet-Higgins, H. C. J.
Chem. Phys. 1950, 18, 265–274.
2. Longuet-Higgins refers to Kekuké structures as ‘K-structures’. See Ref. 1b.
3. Dowd, P. J. Am. Chem. Soc. 1966, 88, 2587–2589.
4. Baseman, R. J.; Pratt, D. W.; Chow, M.; Dowd, P. J. Am. Chem. Soc. 1976, 98,
5726–5727.
5. (a) Berson, J. A.; Bushby, R. J.; McBride, J. M.; Tremelling, M. J. Am. Chem. Soc.
1971, 93, 1544–1546; See also: (b) Berson, J. A. Non-Kekulé Molecules as
Reactive Intermediates. In Reactive Intermediates; Moss, R. A., Platz, M. S., Jones,
M., Jr., Eds.; Wiley-Interscience: New York, 2004. Chapter 5, pp 165–203.

W. Zhong, R. D. Little / Tetrahedron Letters 50 (2009) 4994–4997
4997
5.16 (s, 1H), 5.11 (s, 1H), 4.83 (dd, 1 H, J = 1.4, 2.8), 3.99–3.76 (m, 8H), 2.14–
2.05 (m, 2H), 1.87–1.78 (m, 2H), 1.72–1.63 (m, 4H), 1.46 (s, 3H), 1.26–1.11
(m, 4H), 0.94 (d, 1H, J = 6.8). 13C NMR (CDCl3, 50 MHz): d 148.5, 147.3, 118.1,
109.9, 109.2, 108.6, 72.8, 64.5, 64.4, 64.2, 37.4, 24.2, 21.1, 20.8. For structures
7a,b, 1H NMR (CDCl3, 200 MHz): d feature peak 5.68 (s, 1H), 5.37 (s, 1H),
3.92–3.76 (m, 20H), 3.10 (m, 1H), 2.60–2.33 (m, 8H), 2.22–2.18 (m, 1H), 2.12–
2.05 (m, 3H), 2.00–1.94 (m, 4H), 1.89–1.80 (m, 3H), 1.74–1.64 (m, 7H), 1.59–
1.37 (m, 4H), 1.27 (s, 3H), 1.21 (s, 3H). For structure 14a, 1H NMR (CDCl3,
400 MHz): d 7.70–7.66 (m, 4H), 7.44–7.35 (m, 6H), 4.03–3.82 (m, 8H), 3.76–
3.73 (dd, 1H, J = 6.4), 3.66–3.60 (m, 1H), 2.47 (d, 1H, J = 6.4), 2.33 (d, 1H,
J = 8.4), 2.20–2.03 (m, 2H), 1.91–1.64 (m, 6H), 1.58–1.42 (m, 2H), 1.30 (s, 3H),
1.05 (s, 9H). 13C NMR (CDCl3, 100 MHz): d 218.4, 135.6, 135.5, 134.0, 129.4,
127.5, 111.9, 107.9, 64.9, 64.5, 64.4, 64.1, 54.3, 52.6, 50.6, 33.9, 32.8, 31.6,
26.8, 26.7, 25.1, 23.9, 19.9, 19.2.