Metathesis reactions of Δ22-steroids

Article abstract of DOI:10.1016/j.tetlet.2009.03.191

Metathesis reactions of Δ²²-steroids are studied. The cross metathesis reactions of model Δ²²-steroids with excess of simple alkenes are sluggish or do not occur at all. In contrast, derivatives of both trans- and cis-Δ²²-

Full text of DOI:10.1016/j.tetlet.2009.03.191

Tetrahedron Letters 50 (2009) 2904–2907
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Metathesis reactions of
D
22-steroids
*
Dorota Czajkowska, Jacek W. Morzycki
Institute of Chemistry, University of Białystok, Piłsudskiego 11/4, 15-443 Białystok, Poland
a r t i c l e i n f o
a b s t r a c t
Metathesis reactions of
D D
22-steroids are studied. The cross metathesis reactions of model 22-steroids
Article history:
Received 15 December 2008
Revised 12 March 2009
Accepted 27 March 2009
Available online 2 April 2009
with excess of simple alkenes are sluggish or do not occur at all. In contrast, derivatives of both trans-
and cis-
D
22-cholesterol undergo ring closing metathesis reactions but the former reacts faster. However,
the side chain double bond in stigmasterol and ergosterol is too crowded for metathesis reactions pro-
moted by currently available catalysts.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Metathesis
RCM
Stigmasterol
Steroids
Steroids containing a C22–C23 double bond in the side chain,
such as stigmasterol (1a) or ergosterol (1b), have been frequently
used for the synthesis of various medicinally important com-
pounds, for example, vitamin D derivatives hydroxylated on the
side chain.1 The key intermediates in these syntheses are C22-alde-
hydes that are available by ozonolysis of the C22–C23 double
bonds in the protected steroid precursors. Various olefination
methods exist to reconstruct the double bond. However, the C22-
aldehydes are very sensitive to bases and may epimerize at C20
during the olefination step (Scheme 1).
There is no such danger if an olefin metathesis step is applied
instead. In addition, a cross metathesis approach might afford the
desired products in the most direct way. However, a preliminary
study showed that the steroid C22–C23 double bond is poorly
accessible to metathesis catalysts. Modern ruthenium and molyb-
denum carbene complexes are known to promote various chal-
Thus i-steroidal alcohols were obtained and subjected to ester-
ification with different-sized -alkenyl monoterephthalates in the
presence of DCC/DMAP. The stigmasterol derived esters were
x
subjected to metathesis reactions promoted by various second
21
22
CHO
20
ergosterol or
stigmasterol
side chain
modified steroids
Scheme 1.
R
R
lenging
metathesis
reactions
including
formation
of
tetrasubstituted double bonds,2 but they do not catalyze reactions
of stigmasterol or ergosterol. These steroids contain a double bond
on the side chain sterically hindered at both allylic positions and
do not react even with simple terminal olefins (e.g., ethylene).
We have recently shown3 that occasionally, metathesis reac-
tions that do not proceed intermolecularly may be carried out suc-
cessfully via ring closing metathesis (RCM). Therefore a RCM
approach to steroids with unsaturated side chains was attempted
(Scheme 2). The idea was to transfer an alkylidene group from
the remote 6b position.
1
HO
O
HO
O
HO
c: R =
d: R =
e: R =
f: R =
a: R =
b: R =
; Δ7
* Corresponding author. Tel.: +48 85 7457585; fax: +48 85 7457581.
E-mail address: morzycki@uwb.edu.pl (J.W. Morzycki).
Scheme 2.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.03.191

D. Czajkowska, J. W. Morzycki / Tetrahedron Letters 50 (2009) 2904–2907
2905
generation catalysts. Unfortunately, the RCM did not occur and the
side chain double bond remained intact. Only self-metathesis was
observed leading to the corresponding dimers.
To check if the above approach is conceptually correct the mod-
el olefin 2f lacking steric hindrance in the side chain was prepared
from the corresponding aldehyde. However, even in this case, no
RCM reaction was observed and only dimeric products were
formed (Scheme 3).
Inspection of the Dreiding models and molecular modeling
using the MM+ force-field (HyperChem from HyperCube) sug-
gested that introduction of a short spacer between the oxygen
atom at C6 and the ester group would diminish the steric energy
of the desired macrocycle. Therefore, a series of i-steroidal deriva-
tives were prepared by solvolysis of sterol p-tosylates with ethyl-
ene glycol (instead of hydrolysis) under buffered conditions
(Scheme 4).
R
R
The i-steroidal hydroxyethyl ethers thus obtained were esteri-
fied with 3-butenyl monoterephthalate using the DCC/DMAP
method. Thorough computer-assisted analysis showed that a four
atom alkene is the most suitable for RCM reactions and therefore
3-butenyl esters were used in further studies (Scheme 5).
Unfortunately, stigmasterol and ergosterol derivatives (5a and
5b) did not afford the RCM products. Various first and second gen-
eration metathesis catalysts were tested but only products of self-
metathesis (dimers) were formed accompanied by trace amounts
of isomerization products. In the case of reactions with the Schrock
molybdenum catalyst, the starting sterols were recovered. Cyclore-
version was due to the slightly acidic character of the Schrock com-
plex or products of its decomposition. Reactions of 4-methylpent-
3-enyl esters (instead of 3-butenyl) also proved unsuccessful. Since
the RCM reactions of 5a did not work, the intramolecular enyne
reaction was attempted.4 The analogous 3-butynyl ester 6a was
prepared and subjected to metathesis with the second generation
Grubbs’ catalyst but the reaction also failed to afford the desired
macrocyclic product.
catalyst
PhH or PhCH3,
80 ºC, 24-72 h
O(CH2) CH=CH2
n
O(CH2) CH=
n
2
O
O
O
O
O
O
3
a: R =
or f: R =
2
Scheme 3.
R
R
COOH
O
1. p-TsCl, Py
O
2. (CH2OH)2,
DCC, DMAP
HO
4
AcOK or KHCO3
1
O
OH
R
On the other hand, the less hindered esters 5c–f yielded the
same RCM product 7 as an inseparable mixture of cis and trans iso-
mers in addition to the corresponding dimers (in the case of ester
5f, compound 7 was the only product). The recovered starting ester
was usually contaminated by small amounts (less than 5%) of its
isomers.
5
6a
O
O
O
O
O
The RCM reaction of the model compound 5f promoted by the
second generation Hoveyda catalyst in toluene at 80 °C proceeded
smoothly and was almost complete within 15 minutes (Table 1).
O
O
O
O
O
Also trans-D
22-cholesterol derivative 5c yielded the desired macro-
cyclic product 7 but the reaction required a much longer time
(24 h). The cis isomer 5d appeared to be significantly less reac-
tive—after 48 h only 19% of 7 was isolated in addition to 50% of di-
mer 8d. Surprisingly, a similar result was obtained for the cis-
configured model compound 5e.
c: R =
e: R =
f: R =
a: R =
b: R =
; Δ7
d: R =
Scheme 4.
O
H2/Pd
R
O
9
O
O
O
RCM
7
PhCH3, 80 ºC
R
5
O
O
O(CH2)2CH=
O
O
2
O
O
O
O
O
8
c: R =
d: R =
e: R =
f: R =
a: R =
b: R =
; Δ7
Scheme 5.

2906
D. Czajkowska, J. W. Morzycki / Tetrahedron Letters 50 (2009) 2904–2907
Table 1
Reactions of compounds 5 in toluene at 80 °C catalyzed by the Hoveyda second generation complex
Side chain structure
Amount of catalyst (mol %)
Reaction time
Recovered 5 (isolated yield, %)
Product 7 (isolated yield, %)
Dimer 8 (isolated yield, %)
18
24 h
10
64
10
c
25
20
20
48 h
11
18
5
19
30
84
50
48
0
d
48 h
e
f
15 min
Table 2
The structure of a macrocyclic compound 7 was proved by char-
acteristic upfield shifts of the 18-methyl group resonances to d 0.00
and d 0.17 for the trans and cis isomers, respectively, due to the
anisotropic effect of the nearby aromatic ring. A 1.7:1 ratio of iso-
mers was obtained with the trans (J22,23 = 15.3 Hz) isomer prevail-
ing over the cis (J22,23 = 10.8 Hz) isomer. Hydrogenation of the
mixture yielded quantitatively the saturated product 9, showing
the 18-methyl group resonance at d = ꢀ0.10. Finally, cyclorever-
sion of 9 was carried out under routine conditions (p-TsOH, diox-
ane–water 3:1 v/v mixture) to afford compound 10 (Scheme 6).
Reactions of compounds 11 with 3-butenyl stearate (2 equiv) in toluene at 80 °C, 48 h,
Hoveyda second generation catalyst (20 mol %), c = 9–12 mmol/L
Side chain structure
Yield (%)
12
c
13
44
43
d
In general, the RCM reactions of
easily than the analogous cross metatheses (Scheme 7). The reac-
tion of trans-
22-i-cholesteryl methyl ether 11c with 2 equiv of
D
22-steroids proceeded more
e
f
D
3-butenyl stearate, carried out at higher concentration (9–
12 mmol/L), yielded only 12% of product 12, while ring closing
metathesis of 5c gave 7 in 64% yield (Table 2). The much less
crowded alkene 11f afforded a 43% yield of 12 within 48 h. The
RCM reaction of analogous alkene 5f was almost complete in
15 min and afforded 7 in 84% yield. Surprisingly, the RCM approach
proved not to be advantageous for cis isomers. The RCM reaction of
the cis alkene 5e was not better than the CM reaction of 11e with
the same side chain. The alkenes with stigmasterol (11a) or ergos-
terol (11b) side chains did not react at all.
In conclusion, it should be noted that there is still need for new
metathesis catalysts that are less sterically demanding. The cur-
O
O
O
p-TsOH
dioxane:H2O (3:1)
40 ºC
O
HO
O
HO
O
O
O
O
9
10
Scheme 6.
R
O
16
O
O
catalyst
+
O
PhCH3, 80 ºC
16
12
11
OCH3
OCH3
c: R =
d: R =
e: R =
f: R =
a: R =
b: R =
; Δ7
Scheme 7.

D. Czajkowska, J. W. Morzycki / Tetrahedron Letters 50 (2009) 2904–2907
2907
rently available catalysts are able to react with 1,1-disubstituted
References and notes
olefins yielding tetrasubstituted products.5 However, 1,10-disubsti-
tuted olefins branched at both allylic positions are still challenging
substrates. This study reveals that the RCM approach can be supe-
rior to the CM process but only, if the appropriate stereoisomer of
the starting olefin is used.
1. (a) Chapelon, A.-S.; Moraléda, D.; Rodriguez, R.; Ollivier, C.; Santelli, M.
Tetrahedron 2007, 63, 11511–11616; (b) D’herde, J. N.; De Clercq, P. J.
Molecules 2006, 11, 655–660; (c) Jiang, B.; Shi, H.-P.; Xu, M.; Wang, W.-J.;
Zhou, W.-S. Tetrahedron 2008, 64, 9738–9744; (d) Morzycki, J. W.; Schnoes, H.
K.; DeLuca, H. F. J. Org. Chem. 1984, 49, 2148–2151; (e) D’Ambra, T. E.; Javitt, N.
B.; Lacy, J.; Srinivasan, P.; Warchol, T. Steroids 2000, 65, 401–407.
2. (a) Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413–4450; (b) Schuster, M.;
Blechert, S. Angew. Chem. Int. Ed. 1997, 36, 2036–2056; (c) Fürstner, A. Angew.
Chem. Int. Ed. 2000, 39, 3012–3043; (d) Trnka, T.; Grubbs, R. H. Acc. Chem. Res.
2001, 34, 18–29; (e) Trost, B. M.; Frederiksen, M. U.; Rudd, M. T. Angew. Chem.
Int. Ed. 2005, 44, 6630–6666; (f) Nakamura, I.; Yamamoto, Y. Chem. Rev. 2004,
104, 2127–2198; (g) Deiters, A.; Martin, S. F. Chem. Rev. 2004, 104, 2199–2238.
3. (a) Czajkowska, D.; Morzycki, J. W. Tetrahedron Lett. 2007, 48, 2851–2855; (b)
Wojtkielewicz, A.; Morzycki, J. W. Steroids 2007, 72, 552–558.
Acknowledgment
This work was supported by the Polish State Committee for Sci-
entific Research (Grant Nos. N N204 240334 and PBZ-KBN-126/
T09/2004).
4. (a) Diver, S. T. J. Mol. Catal. A: Chem. 2006, 254, 29–42; (b) Diver, S. T.; Giessert, A.
J. Chem. Rev. 2004, 104, 1317–1382; (c) Mori, M. J. Mol. Catal. A: Chem. 2004, 213,
73–79.
Supplementary data
5. (a) Flynn, A. B.; Ogilvie, W. W. Chem. Rev. 2007, 107, 4698–4745; (b) Dragutan,
V.; Dragutan, I.; Verpoort, F. Platinum Metals Rev. 2005, 49, 33–40; (c) Schrodi, Y.;
Pederson, R. I. Aldrichim. Acta 2007, 40, 45–52.
Supplementary data (experimental procedures as well as 1H
and 13C NMR spectra for all new compounds) associated with this
article can be found, in the online version, at doi:10.1016/
j.tetlet.2009.03.191.