In situ nitrosonium ion generation - α-Oximinylation of enol ethers from steroidal spiroketals: Introduction of C23 (R)-OH in cephalostatin intermediates

Article abstract of DOI:10.1139/V06-113

Nitrosonium ion generated in situ from the reaction of t-BuNO₂ and BF3·Et₂ is an effective oximinylating agent for enol ethers derived from steroidal spiroketals. The scope and limitations of this method has been studied. The difficu

Full text of DOI:10.1139/V06-113

1442
In situ nitrosonium ion generation — ␣-
Oximinylation of enol ethers from steroidal
spiroketals: Introduction of C23 (R)-OH in
cephalostatin intermediates1
Seongmin Lee and Philip L. Fuchs
Abstract: Nitrosonium ion generated in situ from the reaction of t-BuNO2 and BF3·Et2 is an effective oximinylating
agent for enol ethers derived from steroidal spiroketals. The scope and limitations of this method has been studied. The
difficult reduction of C23 ketone to C23 (R)-alcohol has now been selectively achieved via L-Selectride® reduction.
Application to cephalostatin intermediates is discussed.
Key words: oximinylation, nitrosonium ion, steroid spiroketal, cephalostatins.
Résumé : L’ion nitrosonium généré in situ par réaction du t-BuNO2 avec le BF3·OEt2 est un agent d’oximinylation ef-
ficace pour les éthers énoliques dérivés de spirocétals stéroïdaux. On a étudié la portée et les limitations de cette mé-
thode. Faisant appel à une réduction à l’aide de L-Selectride®, on a maintenant réalisé la difficile réduction
stéréosélective de la cétone C23 en alcool-(R) C23. On discute de l’application à des intermédiaires de la céphalosta-
tine.
Mots clés : oximinylation, ion nitrosonium, spirocétal stéroïdal, céphalostatines.
[Traduit par la Rédaction] Lee and Fuchs 1447
Introduction
(7). Although it is evident from the NIH 60-cell line
COMPARE studies that cephalostatin 1 and OSW-1 are re-
lated, significant biological differences3 and recent bio-
activity data of C22 deoxy OSW-1 analogs (8) suggest that
they may have a modified mechanism of action. Correlation
of cytotoxicity with energy for E-ring oxacarbenium ion ac-
cess in cephalostatins and ritterazines has been proposed (9).
SAR studies indicate that the hydroxyl groups of both north
and south units of the cephalostatin family play an essential
role in tumor eradication. Hydroxyl groups provide “polarity
match” (10), a requirement for the exhibition of bioactivity,
and affect the heat of formation of the E-ring oxacarbenium
ion,4 thus adjusting bioactivity. When the C17 alcohol, for
instance, in the north unit of cephalostatin 1 was replaced by
a hydrogen, the antitumor activity dropped dramatically.5
The presence of the C23 hydroxyl group also contributes to
the anticancer activity. The recently synthesized C23′-deoxy
cephalostatin 1(11) is about 10 times less potent than
cephalostatin 1.
The 45 members of the cephalostatin and ritterazine fam-
ily, along with analogs, afford the basis for elucidating some
of the structure–activity relationships (SAR) of these potent
cytotoxins (1). All members of the cephalostatin family pos-
sess two steroidal spiroketals connected by a pyrazine. The
most active compounds bear a highly oxygenated north unit
and a substantially less polar south unit. In terms of
bioactivity, cephalostatin 1 (Fig. 1) shows average 1 nmol/L
GI50s in 2 day tests in the NCI 60-line screen and
10–14 mol/L GI50s in 6day tests in the Purdue 6 line mini-
panel (2). Cephalostatin 1, cephalostatin 7, ritterazine M,
and ritterazine K have been synthesized by our group (3),
while we and others (4) have also been active in the synthe-
sis and testing of analogs.
The mechanism of action of the cephalostatins and
ritterazines is currently unknown, although recent data indi-
cates that ritterazine B is an apoptotic agent like OSW-1 (5)
and cephalostatin 7 (6). The role of spiroketals and sugars as
hydrogen bond donors and (or) acceptors has been suggested
Our second generation synthesis of cephalostatin 1 (4)
uses a “redox” strategy(12), requiring a scalable protocol ca-
Received 21 December 2005. Accepted 29 April 2006. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on
17 October 2006.
S.M. Lee and P.L. Fuchs.2 Department of Chemistry, Purdue University, West Lafayette, IN 47097, USA.
1This article is part of a Special Issue dedicated to Dr. Alfred Bader.
2Corresponding author (e-mail: pfuchs@purdue.edu).
3P. Huang. Unpublished results.
4T.G. LaCour and P.L. Fuchs. Unpublished results.
5Compare, for example, ritterazine A (17 -OH, 24 nmol/L) and ritterazine T (17 -H, 590 nmol/L).
Can. J. Chem. 84: 1442–1447 (2006)
doi:10.1139/V06-113
© 2006 NRC Canada

Lee and Fuchs
1443
Scheme 1.
OH
OBz
O
OBz
O
O
O
C23 (R)
2
1
AcO
AcO
OBz
OH
C23 (R)
OH
O
BzO
OBz
OH
C23 (R)
OH
O
O
OH
3
4
AcO
AcO
North unit of cephalosatin 1
that the C14,C15-saturated spiroketal forms the nitroimine in
acceptable yield, our attempts to use this procedure on sub-
strate 1 only afforded the C23 nitroimine 5 in very low
yield. We also investigated SN2 chemistry with equatorial
C23 bromide 8 and iodide 9, prepared from 1 with
phenyltrimethylammonium bromide (PTAB) or iodine
monochloride, respectively. Unfortunately, numerous re-
agents that convert alkyl halides to alkoxy groups, including
superoxide (19), KNO2 (20), O2–Bu3SnH (21), AgBF4–
glyme–H2O (22), AgOAc (23), and CsF–BzOH (24), were
either unreactive or severely destructive.
The known methods for α-oximinylation of ketones were
not applicable to our substrate (25). Thus, we were delighted
that the formation of the desired oxime spiroketal 12 took
place smoothly from the reaction of rockogenin diacetate 11
with t-BuNO2 and BF3·Et2. Optimization of this new proce-
dure is summarized in Table 1. The choice of Lewis acid is
critical for this event, as BF3·Et2 gave the best yields while
FeCl3, TiCl4, SnCl4, ZnBr2, Sc(OTf)3, and TMSOTf failed to
give reasonable conversions, probably because of their de-
creased oxophilic character relative to BF3·OEt2.
Minimization of undesired lactone 13, which comes from
oxime spiroketal 12, requires using excess t-BuONO
(5 equiv.) and catalytic BF3·OEt2 (0.5 equiv.) in acetic acid
at ambient temperature for the minimum time. The use of
less t-BuONO (Table 1, entries 1–5 and 7) and BF3·Et2
(Table 1, entry 11), solvents other than AcOH (Table 1, en-
tries 1, 3, and 4), lower temperature (Table 1, entry 2), or
prolonged reaction time (Table 1, entry 10) results in the for-
mation of significant amounts of lactone 13.
Fig. 1.
OH
C23(R)
O
HO
OH
O
OH
C17
N
N
X
Z
cephalostatin 1
X, Y, Z = H:
O
O
O
X =OH, Y=H, Z = H: cephalostatin 2
X =OH, Y=Me, Z= H: cephalostatin 3
X=H, Y=H, Z = OMe: cephalostatin 19
OH
C23' (R)
Y
pable of efficient introduction of the C23 (R)-alcohol
(Scheme 1). Eighteen of nineteen known cephalostatins pos-
sess the north C23 (R)-alcohol and seventeen of those bear
the south C23 (R)-alcohol (Fig. 1).
Many efforts have been made to introduce the C23 (R)-al-
cohol (13), such as the reduction of the C23 carbonyl group
(14), sulfinylation of a C22,C23 enol ether with TFAA-
activated DMSO followed by allylic rearrangement (15), and
addition of allyl stannane to an aldehyde (16). However,
none of these methods provided the requisite C23 (R)-alco-
hol in workable excess (de < 30%). At present, the only
acceptable method is the dimethyldioxirane (DMDO) oxida-
tion of an enone – vinyl ether to quantitatively achieve
stereospecific cyclization to the prized (R)-alcohol. However,
this method employs 750 mL of DMDO for the production
of 15.7 g of C23 alcohol (17) and also requires the use of
stoichiometric amounts of toxic selenium and stannyl re-
agents to prepare 2. Herein, we describe a scalable pathway
leading to C23 (R)-OH.
The proposed mechanism of oximinylation involves
nitrosonium ion (26) mediated nitrosylation of the enol ether
11a formed via Lewis acid opening of the steroid spiroketal
–
(Scheme 3). First, nitrosonium ion (NO+BF4 ) is generated
in situ from a reaction of t-BuNO2 and BF3·OEt2. Then, the
attack of enol ether 11a on the nitrogen atom of nitrosonium
ion takes place to produce nitroso spiroketal 12a. Finally, the
nitroso spiroketal 12a is rapidly isomerized to give oxime
spiroketal 12. The presence of a NOBF4 (–141.6 ppm,
CD3CN) peak in the 19F NMR supports this mechanism. The
reaction of spiroketal 11 with 3 equiv. of commercial
NOBF4 (Table 1, entry 19) in acetic acid affords the same
Results and discussion
The known methods for α-oxidation of ketones, enols, and
spiroketals (Scheme 2) were initially surveyed to access the
desired C23 ketone 7. We first attempted the synthesis of a
masked C23 ketone, such as nitroimine 5 or oxime 6. Al-
though Suàrez and co-workers (18) recently demonstrated
© 2006 NRC Canada

1444
Can. J. Chem. Vol. 84, 2006
Scheme 2.
O
O
O
O
O
NOH
NNO2
O
O
< 10 %
5
6
7
NaNO2
BF3·OEt2
NOCl
AgBF4
OBz
O
O
X
PTAB
or
O
O
KO2
ICl
OH
X = Br
X = I
8
9
O
1
AcO
AgOAc
O
O
OH
O
2
H
O
O
O
10
7
not desired product
Table 1. Optimization of α-oximinylation.
AcO
O
O
O
O
O
N
O
conditions
OH
AcO
Entry
Lewis acid (equiv.)
t-BuNO2 (equiv.)
Solvent
T (°C)
Time (h)
Yields of 12 and 13 (%)
1
2
3
4
5
BF3·OEt2 (1)
BF3·OEt2 (1)
BF3·OEt2 (1)
BF3·OEt2 (1)
BF3·OEt2 (1)
3
3
3
3
3
CH2Cl2
CH2Cl2
Toluene
MeCN
AcOH
25
0
3
45, 31
NR
3
25
25
25
3
40, 23
35, 48
62, 17
3
0.5
6
7
—
3
2
AcOH
AcOH
25
25
1
1
NR
37, 45
BF3·OEt2 (2)
8
BF3·OEt2 (0.5)
BF3·OEt2 (0.5)
BF3·OEt2 (0.5)
BF3·OEt2 (0.2)
NOBF4 (3)
3
5
5
5
AcOH
AcOH
AcOH
AcOH
AcOH
25
25
25
25
25
1
32, 52
92, 0
9
0.2
2
10
11
12
50, 36
41, 44
82, 0
2
0.2
Note: NR = No reation.
oxime spiroketal 12 in 82% yield after 10 min, thus exhibit-
ing similar reactivity and product distribution pattern to that
observed in the reaction of spiroketal 11 with BF3·Et2 and t-
BuONO (Table 1, entry 16). The Lewis acid promoted sec-
ond-order Beckmann fragmentation (27) of α-oxygenated
oxime spiroketal 12 appears to account for the formation of
lactone 13. The scope and limitation of the oximinylation of
steroid spiroketals was then explored.6 The results in Ta-
ble 2 reveal some interesting features. First, C23 oximinyl-
ation works for only 5/6 spiroketals, not for 5/5 and 6/5
spiro systems, such as 21 and 22. Hecogenin acetate (start-
ing material for 15) containing a C12 ketone, which is diffi-
6 Supplementary data for this article are available on the journal Web site (http://canjchem.nrc.ca) or may be purchased from the Depository
of Unpublished Data, Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0R6, Canada. DUD 5081. For more
information on obtaining material refer to http://cisti-icist.nrc-cnrc.gc.ca/irm/unpub_e.shtml.
© 2006 NRC Canada

Lee and Fuchs
1445
Scheme 3.
t
3
NO+BF4
+
t-Bu OBF2
O N O
-
F
F
B
B
F
F
F
F
LA
H
AcO
O
O
-
NO+BF4
O
O
N
O
4
AcO
AcO
O
4
O
N
O
OH
N
O
O
3
AcO
O
N
BF3
O
O
O
:Nu
O
Table 2. α-Oximinylation of steroid spiroketals.
O
OBz
O
O
Y
O
Y
O
AcO
O
O
O
O
O
Y
Y
O
O
X
AcO
AcO
2
O
OBz
OAc
O
O
Y
O
Y
O
OAc
AcO
AcO
O
O
Y
Y
O
O
AcO
Y
MeO
Y
OBz
O
O
OBz
O
O
O
AcO
AcO
Note: isolated yields. Y = H2 (substrates) or Y = NOH (products).
© 2006 NRC Canada

1446
Can. J. Chem. Vol. 84, 2006
Table 3. Reduction of the C23 ketone.
O
X
OH
O
O
O
O
C23(S)
OH
14
15
C23(R)
AcO
–
–
–
α
14
∆
Entry
Substrate
Hydride
Conditions
Products
Yield (%)
C23 (R):C23 (S)
0 °C, THF
1:20
1:19
1:11
1:4.2
1:1.4
1
2
3
4
5
23
23
23
23
23
LiAl(OtBu)3H
NaBH4
24, 25
24, 25
24, 25
24, 25
24, 25
93
97
85
94
81
25 °C, MeOH
25 °C, THF
–78 °C, 2 h
–78 °C, THF
(R)-CBS/BH3
CeCl3/NABH4
LS-Selectride®
6
7
8
23
23
23
L-Selectride®
K-Selectride®
(S)-CBS/BH3
–78 °C, THF
–78 °C, THF
25 °C, 0.5 h, THF
24, 25
24, 25
24, 25
89
83
91
2.5:1
4.6:1
6.2:1
9
10
7
7
L-Selectride®
–78 °C, 3 h, THF
25 °C, 0.5 h, THF
2, 26
2, 26
87
92
10.3:1
7.3:1
(S)-CBS/BH3
L-Selectride®
11
7
–78 °C, 3 h, PhMe
2, 26
85
5.5:1
cult to enolize, smoothly gave oxime spiroketal 15. The
presence of more readily enolized C3 ketone resulted in low
yield of 14 because of the formation of by-products. It is no-
table that olefin moieties (6 and 22) are intact under the re-
action conditions. Alcohol functionality was converted to
nitrite 16a. Most importantly, employing the new protocol,
we were able to obtain a key intermediate (5/6 spiroketal 6)
in reasonable yield.
C23 ketones 23 and 7 were readily prepared by p-
toluenesulfonic acid catalyzed deoximinylation. We next
turned our attention to establishing the requisite C23 (R)
stereochemistry (Table 3). In the case of C14,C15-saturated
C23 ketone 23, (S)-CBS best effected the desired reduction
to afford C23 (R) and C23 (S) in a reproducible ratio of
6.2:17 (Table 3, entry 8, C23 (R) (78%) and C23 (S) (13%),
isolated yield, 50 mmol scale; previous results (28) show
formation of C23 (R) and C23 (S) in a ratio of 1.7:1 (72%)).
We were pleased that the L-Selectride® reduction of
C14,C15-unsaturated ketone 7 (Table 3, entry 9) delivered
C23 axial alcohol 2 in a highly sterereoselective fashion.
In conclusion, an efficient and economical in situ method
to generate nitrosonium fluoroborate is presented. This
chemistry provides a large-scale pathway for the establish-
ment of the C23 (R) stereochemistry of steroidal sapogenins.
Further research is ongoing regarding the application of this
method for the synthesis of the north unit of cephalostatin 1.
References
1. G.R. Pettit, R. Tan, J.P. Xu, Y. Ichihara, M.D. Williams, and
M.R. Boyd. J. Nat. Prod. 61, 953 (1998), and refs. therein.
2. T.G. LaCour, C. Guo, S. Ma, J. Jeong, M.R. Boyd, and P.L.
Fuchs. Bioorg. Med. Chem. Lett. 9, 2587 (1999), and
refs. therein.
3. (a) T.G. LaCour, C. Guo, M.R. Boyd, and P.L. Fuchs. Org.
Lett. 2, 33 (2000); (b) W. Li, T.G. LaCour, M.R. Boyd, and
P.L. Fuchs. J. Am. Chem. Soc. 124, 4548 (2002), and
refs. therein; (c) S.M. Lee, T.G. LaCour, D.A. Lantrip, and
P.L. Fuchs. Org. Lett. 4, 313 (2002).
4. (a) C.H. Heathcock and S.C. Smith. J. Org. Chem. 59, 6828
(1994), and refs. therein; (b) R. Jautelat, A. Muller-Fahrnow,
and E. Winterfeldt. Chem. Eur. J. 5, 1226 (1999); (c) S. Basler,
A. Brunck, R. Jautelat, and E. Winterfeldt. Helv. Chim. Acta.
83, 1854 (2000), and refs. therein.
5. (a) Y. Mimaki, M. Kuroda, A. Kameyama, Y. Sashida, and T.
Hirano. Bioorg. Med. Chem. Lett. 7, 633 (1997); (b) S. Kubo,
Y. Mimaki, M. Terao, Y. Sashida, T. Nikaido, and T. Ohmoto.
Phytochemistry, 31, 3969 (1992).
6. (a) T. Komiya, N. Fusetani, S. Matsugana, M. Kelley, and K.
Nakagawa. Cancer Chemother. Pharmacol. 51, 202 (2003);
(b) V.M. Dirsch, I.M. Mueller, S.T. Eichhorst, G.R. Pettit, Y.
Kamano, M. Inoue, and A.M. Vollmar. Cancer Res. 63, 8869
(2003).
7. Y. Pan, R.L. Merriaman, L.R. Tanzer, and P.L. Fuchs. Bioorg.
Med. Chem. Lett. 2, 967 (1992).
8. (a) L. Deng, H. Wu, B. Yu, M. Jiang, and J. Wu. Bioorg. Med.
Chem. Lett. 14, 2781 (2004); (b) J. Morzycki, A.
Wihtiekewicz, and S. Wolcznski. Bioorg. Med. Chem. Lett. 14,
3323 (2004).
9. (a) T.G. LaCour, C. Guo, M.R. Boyd, and P.L. Fuchs. Org.
Lett. 2, 33 (2000); (b) C. Guo, T.G. LaCour, and P.L. Fuchs.
Bioorg. Med. Chem. Lett. 9, 419 (1999).
Acknowledgments
This investigation was generously supported by funds pro-
vided by the National Institutes of Health (CA 60548). We
acknowledge Arlene Rothwell and Karl Wood for providing
MS data.
7 Preincubation (~30 min) of (S)-CBS and BH3 Me2 in THF before the addition of 23 is critical to obtain good stereoselectivity.
© 2006 NRC Canada

Lee and Fuchs
1447
10. S. Fukuzawa, S. Matsunaga, and N. Fusetani. J. Org. Chem.
62, 4484 (1997), and refs. therein.
20. R.M. Moriarty and V.K. Singh. Tetrahedron Lett. 34, 8029
(1993).
11. W. Li, T.G. LaCour, M.R. Boyd, and P.L. Fuchs. J. Am. Chem.
21. E. Nakamura, T. Inubushi, S. Aoki, and D. Machii. J. Am.
Soc. 124, 4548 (2002).
Chem. Soc. 113, 8980 (1991).
12. (a) S.M. Lee and P.L. Fuchs. J. Am. Chem. Soc. 124, 13978
(2002); (b) S.M. Lee and P.L. Fuchs. Org. Lett. 4, 313 and 317
(2002).
22. J. Morzycki and A. Jastrzebska. Tetrahedron Lett. 42, 5989
(2001).
23. M. Zarraga and J.D. Martin. Tetrahedron Lett. 31, 1633
(1990).
13. Hydroxylation of steroids — C5: (a) J.T. Dixon, C.W.
Holzapfel, and F.R. Heerden. Synth. Commun. 23, 135 (1993);
(b) P. Bovicelli, A. Gambacorta, and E. Mincione. Tetrahedron
Lett. 33, 7411 (1992); hydroxylation of steroids — C14:
(c) M.D. Kaufman, P.A. Grieco, and D.W. Bougie. J. Am.
Chem. Soc. 115, 11648 (1993); hydroxylation of steroids —
C17: (d) F.A. Davies, L.C. Vischwarkarma, J.M. Billmers, and
J. Finn. J. Org. Chem. 49, 3243 (1984).
24. T. Sato and J. Otera. Synlett, 336 (1995).
25. (a) A.A. Mohammed and G. Nagendrappa. Tetrahedron Lett.
44, 2753 (2003); (b) J. Rasmussen and A. Hassner. J. Org.
Chem. 39, 2558 (1974); (c) Y.H. Kim, Y.J. Park, and K. Kim.
Tetrahedron Lett. 30, 2833 (1989); (d) M.N. Mallya, G.
Nagendrappa, J.S. Prasad, M.A. Sridhar, and M.K. Lokanath.
Tetrahedron Lett. 42, 2565 (2001).
14. (a) C. Betancor, R. Freire, I. Perez-Martin, T. Prange, and E.
Suàrez. Org. Lett. 4, 1295 (2002); (b) D.H. Barton, P.G.
Sammes, M.V. Taylor, and E. Werstiuk. J. Chem. Soc. Chem.
Commun. 1977 (1970).
15. J. Jeong and P.L. Fuchs. J. Am. Chem. Soc. 116, 773 (1994).
16. S. Kim, S.C. Sutton, C. Guo, T.G. LaCour, and P.L. Fuchs. J.
Am. Chem. Soc. 121, 2056 (1999).
17. J.S. Lee and P.L. Fuchs. Org. Lett. 5, 2247 (2003).
18. A.G. Gonzalez, R. Frerire, M.G. Garcia-Estrada, and E.
Suàrez. Tetrahedron, 28, 1289 (1972).
26. (a) G.A. Olah. Aldrichimica Acta, 12, 43 (1979); (b) M.P.
Doyle and W. Wierenga. J. Am. Chem. Soc. 94, 3896 (1972);
(c) R.H. McGirk and E.H. White. J. Am. Chem. Soc. 95, 3854
(1973); (d) G.A. Olah, J. Shih, B.P. Singh, and P.G. Gupta. J.
Org. Chem. 48, 3356 (1983).
27. (a) H. Tso, B.A. Gilbert, and J. Huw. J. Chem. Soc. Chem.
Commun. 669 (1993); (b) P.A. Petukhov and A.V. Tkachev.
Synlett, 580 (1993); (c) D. Miljkovic and J. Pedtrovic. J. Org.
Chem. 38, 3585 (1973); (d) A.F. Ferris, G.S. Johnson, and
F.E. Gould. J. Org. Chem. 25, 496 (1960).
19. E.J. Corey, K.C. Nicolaou, M. Shibasaki, Y. Machida, and C.
28. C. Betancor, R. Freire, I. Perez-Martin, T. Prange, and E.
Shiner. Tetrahedron Lett. 37, 3183 (1975).
Suàrez Org. Lett. 4, 1295 (2002).
© 2006 NRC Canada