T. K. Chakraborty et al. / Tetrahedron Letters 51 (2010) 4425–4428
4427
minor Z-isomer 26 could be separated easily at this stage through
silica gel column chromatography. The E geometry of 20 was
confirmed from NOE experiment. 1H NOE study showed that irra-
diation of the olefinic singlet signal at d 5.59 caused no enhance-
ment of the proton signal at d 2.18 corresponding to allylic
methyl. Similarly on irradiation of allylic methyl singlet signal at
d 2.18, there was no enhancement of proton signal at d 5.59 corre-
sponding to olefinic proton. On the other hand, irradiation of the
olefinic singlet signal at d 5.6 enhanced the peak of allylic CH2
quatrate at d 2.1 to prove the E geometry of 20. To carry out the
crucial radical-mediated cyclization reaction, the epoxy alcohol
20 was treated with Cp2TiCl1 (generated in situ using Cp2TiCl2,
ZnCl2, and activated Zn powder) in THF at ꢁ20 °C, then allowed
to come to room temperature and stirred for another 12 h. This
metal-mediated radical cyclization gave 2719 in 82% yield.
The relative stereochemistry of the newly generated centre was
confirmed by NOESY experiment. NOESY spectrum of 27 showed a
strong dipolar coupling between C8–H and 10-CH3 indicating that
they are in the same side of the cyclic ring. Similarly NOE cross
peaks between C8–H/C11–H, 10-CH3/C11–H proved that C11–H also
to be in the same side of the bicyclic ring. Stereochemistry of C6–H
center was fixed from the NOE between C6–H/C9–H (Fig. 2).
The Ti(III)-mediated ring closure reaction of 20 was expected to
provide the desired compound 19 in the same way as 3 was trans-
formed into 4 carrying the CH2OH and CH2CO2Et substituents in cis
orientations in the newly formed five-membered ring. But the for-
mation of the unwanted isomer 2719, contrary to our expectation,
has now given us a challenge to find a way out to access the right
intermediate 19 in order to complete the total synthesis of
penifulvins.
12
13
EtO2C
H
4
H
5
3
9
2
H
8
7
6
1
H
11
HO
10
CH3
HO
H
H
Figure 2. Important NOEs that were used to establish the relative stereochemist-
ries in 27.
hand to build chiral quaternary centers using Ti(III)-mediated ring
opening of 2,3-epoxy alcohols could be tested for its practical
applicability in assembling the core bicylic ring system of these
molecules.
Retrosynthetic analysis of one of the penifulvins, penifulvin A
(12) is illustrated in Scheme 4. We envisaged that compound 12
could have been derived from intermediates 17–19 via functional
groups manipulations. The bicylic core of 19 could be obtained
from epoxy alcohol 20 via Cp2TiCl-mediated radical cyclization.
We commenced our synthesis fromcommerciallyavailable start-
ing material cyclopentanone 21 as shown in Scheme 5. The first step
involved a crucial Michael addition reaction13 in which cyclopenta-
none 21 was first converted to the corresponding TMS enol ether
derivative using LDA and TMSCl in THF at ꢁ78 °C. The TMS-enol
ether intermediate, without purification and characterization, was
mixed with mesityl oxide in CH2Cl2 and the mixture was added
dropwise to a TiCl4 solution in CH2Cl2 at ꢁ78 °C to give 22 in 56%
yield. Next our objective was to protect selectively one of the
carbonyl groups of 22. We achieved it by treating 22 with 1,3-prop-
anedithiobis(trimethylsilane) and ZnI2 in diethyl ether at room tem-
perature.14 A variety of reagents and conditions, such as (CH2OH)2,
PTSA, benzene; CH2(CH2OH)2, PTSA, benzene; (CH2OTMS)2, TMSOTf,
CH2Cl2; CH2(CH2SH)2, BF3ꢂEt2O, and CH2Cl2 were used but none of
them were found to be effective except the above-mentioned
condition.
Acknowledgments
The authors wish to thank DST, New Delhi for the Ramanna Fel-
lowship (SR/S1/RFOC-06/2006; T.K.C.) and CSIR, New Delhi for re-
search fellowships (A.K.C. and R.S.).
References and notes
The resulting mono-keto compound was then subjected to one
carbon olefination. Several reagents and conditions such as Ph3P+-
CH3Iꢁ, n-BuLi, THF; Ph3P+CH3Iꢁ, KOtBu, diethyl ether; Cp2TiMe2,
and toluene were used but none of them gave the desired product.
Finally the transformation was successfully carried out using Mg
and TiCl4 in CH2Cl2–THF solvent mixture at 0 °C to afford 23 in 62%
1. (a) Chakraborty, T. K.; Dutta, S. J. Chem. Soc., Perkin Trans. 1 1997, 1257–1259;
(b) Chakraborty, T. K.; Das, S. Tetrahedron Lett. 2002, 43, 2313–2315.
2. Chakraborty, T. K.; Samanta, R.; Das, S. J. Org. Chem. 2006, 71, 3321–3324.
3. Chakraborty, T. K.; Samanta, R.; Ravikumar, K. Tetrahedron Lett. 2007, 48, 6389–
6392.
4. Chakraborty, T. K.; Samanta, R.; Roy, S.; Sridhar, B. Tetrahedron Lett. 2009, 50,
3306–3310.
5. Shim, S. H.; Gloer, J. B.; Wicklow, D. T. J. Nat. Prod. 2006, 69, 1601–1605.
6. Shim, S. H.; Swenson, D. C.; Gloer, J. B.; Dowd, P. F.; Wicklow, D. T. Org. Lett.
2006, 8, 1225–1228.
7. (a) Horner, L.; Hoffman, H.; Wippel, H. G.; Klahre, G. Chem. Ber. 1959, 92, 2499–
2505; (b) Wadsworth, W. S., Jr.; Emmons, W. D. J. Am. Chem. Soc. 1961, 83,
1733–1738; Cf.: (c) Wadsworth, W. S., Jr. In Organic Reactions; Wiley: New
York, 1977; Vol. 25. pp 73–254.
8. Gao, Y.; Klunder, J. M.; Hanson, R. M.; Ko, S. Y.; Masamune, H.; Sharpless, K. B. J.
Am. Chem. Soc. 1987, 109, 5765–5780.
t
yield.15 Allylic oxidation of 23 using SeO2 and TBHP in BuOH at
50 °C gave the desired allylic alcohol in 39% yield.16 Extensive
decomposition of the starting material was observed when other
reagents and conditions such as SeO2, H2O2, dioxane; SeO2, H2O2,
tBuOH; and SeO2, AcOH–H2O were used. The relative stereochemis-
try of the newly generated center carrying the –OH group was
established latter by NOESY spectroscopy. Next, to deprotect the
dithiane moiety, the allylic alcohol intermediate was treated with
AgNO3 in EtOH at 50 °C to give 24 in 90% yield.17
Our next objective was to carry out two carbon olefination reac-
tion. A variety of reagents such as (MeO)2P(O)CO2Me, (CF3CH2O)2-
P(O)CO2Me, and Ph3PCH2CO2Me were used but Mayer–Schuster
rearrangement18 was found to be effective. First the carbonyl com-
pound 24 was treated with lithiated ethyl ethynyl ether in THF at
ꢁ78 °C followed by Mayer–Schuster rearrangement, using anhy-
drous AuCl3 in CH2Cl2–EtOH solvent system at room temperature
to afford 25, as an inseparable mixture of E and Z olefins (3.4:1),
in 61% yield in two steps. Then this inseparable mixture of com-
pounds was treated with m-CPBA in CH2Cl2 at 0 °C to give the E
olefinic epoxy compound 20 (single diasteromer) in 65% yield.
The stereochemistry of the epoxy ring was not determined as its
opening with Ti(III) was expected to lead to a radical center. The
9. (a) Green, M. L. H.; Lucas, C. R. J. Chem. Soc., Dalton Trans. 1972, 1000–1003; (b)
Nugent, W. A.; RajanBabu, T. V. J. Am. Chem. Soc. 1988, 110, 8561–8562; For
reviews on Cp2TiCl see: (c) Barrero, A. F.; Quílez del Moral, J.; Sánchez, E. M.;
Arteaga, J. F. Eur. J. Org. Chem. 2006, 1627–1641; (d) Daasbjerg, K.; Svith, H.;
Grimme, S.; Gerenkamp, M.; Mück-Lichtenfeld, C.; Gansäuer, A.; Barchuk, A.
Top. Curr. Chem. 2006, 263, 39–69; (e) Gansäuer, A.; Rinker, B. Tetrahedron 2002,
58, 7017–7026; (f) Gansäuer, A.; Bluhm, H. Chem. Rev. 2000, 100, 2771–2788.
10. Spectral data of compound 11: Rf = 0.6 (SiO2, 25% EtOAc in petroleum ether);
½
a 2D8
ꢃ
+15.3 (c 0.063, CHCl3); 1H NMR (CDCl3, 400 MHz): d 1.1 (Me9), 1.32 and
1.44 (acetonide methyls), 1.36 (m, H40), 1.43 (m, H7), 1.75 (m, H4), 2.02 (m,
3
3
H50), 2.12 (m, H5), 2.47 (d, J2–2 = 14.7 Hz, H2), 2.79 (d, J2–2 = 14.7 Hz, H20),
0
0
3
3
3
3.64 (s, 3H), 3.80 (dd, J8–8 = 12.5 Hz, J7–8 = 3.2 Hz, H80), 4.06 (dd, J8–
0
0
= 12.5 Hz, J7–8 = 5.4 Hz, H8), 4.43 (dt, J = 1.4, 6.1 Hz, H6); 13C NMR (CDCl3,
3
80
100 MHz): d 173.3, 97.2, 72.4, 57.9, 51.0, 48.3, 47.8, 41.0, 36.7, 31.1, 30.1, 28.7,
26.2; IR (KBr): 2983, 2859, 1727, 1427, 1178 cmꢁ1; MS (ESI): m/z 265 (20)
[M+Na]+.
11. (a) Gaich, T.; Mulzer, J. J. Am. Chem. Soc. 2009, 131, 452–453; (b) Gaich, T.;
Mulzer, J. Org. Lett. 2009, 12, 272–275.
12. Cozzi, P. G.; Hilgraf, R.; Zimmermann, N. Eur. J. Org. Chem. 2007, 5969–5994.