Olefin Metathesis: Remote Substituents Governing the Stereoselectivity of 11-Membered-Ring Formation

Article abstract of DOI:10.1021/ol036156w

(Matrix presented) Stereospecific RCM reaction of 7 (R¹ = Me) results in the formation of (Z)-11-membered macrocycle 9c. Cascade RCM/ROM/RCM transformation of 7 (R¹ = H) affords the thermodynamically more stable spirocycle product 12.

Full text of DOI:10.1021/ol036156w

ORGANIC
LETTERS
2004
Vol. 6, No. 2
205-208
Olefin Metathesis: Remote Substituents
Governing the Stereoselectivity of
11-Membered-Ring Formation
Georgios Vassilikogiannakis,* Ioannis Margaros, and Maria Tofi
Department of Chemistry, UniVersity of Crete, 71409 Iraklion, Crete, Greece
Vasil@chemistry.uoc.gr
Received November 4, 2003
ABSTRACT
1
Stereospecific RCM reaction of 7 (R ) Me) results in the formation of (Z)-11-membered macrocycle 9c. Cascade RCM/ROM/RCM transformation
1
of 7 (R ) H) affords the thermodynamically more stable spirocycle product 12.
Recent decades have seen a burgeoning of interest in the
olefin metathesis reaction and the intensity of research
activity involved has grown exponentially as the power of
this reaction became ever more apparent.1 Much work has
focused on unveiling a plethora of olefin metathesis catalysts
with improvements in their reactivity profiles, and their
sensitivity to air and water, constantly being attained by the
judicious tailoring of the ligands employed and by careful
selection of an appropriate transition metal to effect the
requisite metallocarbene-mediated reactions. Elegant syn-
thetic applications using the knowledge so-garnered have
begun to emerge in the literature and such examples are
becoming more and more sophisticated as time progresses.
The popularity of olefin metathesis as a synthetic tool can
disguise the fact that each novel example investigated still
has much to teach us about the dynamics of this complex
sequence of reactions which eventually leads to the formation
of a new olefin upon the union of two precursor olefins.
These latter features were never better illustrated than by
the recent publication of the first total synthesis of coleoph-
omones B and C.2
In this instance two extremely strained and rigid 11-
membered rings were formed by employing the second-
generation Grubbs’ catalyst (I)3 in two independent ring-
closing metathesis (RCM) reactions. The surprise elicited
by the outcome of this investigation can be said to originate
not only from the successful accomplishment of a challenging
and relatively unprecedented macrocyclization metathesis
employing an isoprenyl coupling partner, but also in the fact
that this reaction occurred with complete stereoselectivity.
Furthermore, the stereoselectivity observed was reversed in
the two substrates examined, in one case the E double bond
was formed exclusively and in the other the Z isomer was
the sole product.2 The proliferation of medium-sized rings
among the molecules we seek to synthesize combined with
the obvious advantages of RCM as a tool for accomplishing
their construction together mean that comprehending the
source of this specificity is pivotal for future reference. It is
also of broad relevance to ask how general the specificity
for formation of a single isomer is when 11-membered rings4
are targeted, since larger ring sizes are frequently associated
with mixed geometries.5
(1) For selected recent reviews in olefin metathesis see: (a) Grubbs, R.
H.; Chang, S. Tetrahedron 1998, 54, 4413. (b) Furstner, A. Angew. Chem.,
Int. Ed. 2000, 39, 3012. (c) Connon, S. J.; Blechert, S. Angew. Chem., Int.
Ed. 2003, 42, 1900. (d) Schrock, R. R.; Hoveyda, A. H. Angew. Chem.,
Int. Ed. 2003, 42, 4592.
(2) Nicolaou, K. C.; Vassilikogiannakis, G.; Montagnon, T. Angew.
Chem., Int. Ed. 2002, 41, 3276.
(3) For the preparation of the catalyst see: Scholl, M.; Ding, S.; Lee, C.
W.; Grubbs, R. H. Org. Lett. 1999, 1, 953.
10.1021/ol036156w CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/23/2003

To instigate the probing of such issues, compounds 6a
and 7a (R1 ) R2 ) H) were synthesized by using the methods
developed by the Nicolaou group (Scheme 1).2 Our plan was
more general explorations would be pursued in search of a
set of general rules for selectivity in the formation of 11-
membered-ring olefins by metathesis, beginning with sub-
strates where the bridging carbocycle was absent (C9-C11).
When 6a was treated with Grubbs’ catalyst (I, 0.1 equiv)
in refluxing dichloromethane, macrocycle 8a was isolated
as the sole product in 81% yield (Scheme 2). The newly
Scheme 1. Preparation of the Olefin Metathesis Precursorsa
Scheme 2. Stereoselective RCM of Monoprenylated
Substratesa
a Reagents and conditions: (a) cat. I (0.1 equiv), CH2Cl2, 40
°C, 1 h, 81% of 8a; 14 h, 68% of 8b; (b) cat. I (0.1 equiv), CH2Cl2,
40 °C, 30 min, 83% of 9a; 17 h, 71% of 9b. Mes ) mesityl; Cy
) cyclohexyl.
formed macrocyclic double bond was confirmed as having
the predicted Z-stereochemical arrangement by extensive nOe
studies. Surprisingly, however, when 7a (E configuration at
∆7,8)6 was likewise treated with catalyst I under the same
conditions, the Z-macrocyclic double bond olefin isomer
(∆14,15) 9a was once again obtained as the sole product in
83% yield. Since this stereochemical arrangement for ∆14,15
(assigned by nOe studies, Scheme 2) was not as predicted,
further support for the geometry of the newly formed
macrocyclic double bond of 9a was desirable. Upon hy-
drolysis of the vinylogous esters of 8a and 9a, facilitated by
K2CO3, tricarbonyl compounds, identical in every respect,
were obtained. To explain the discrepancy between this and
previously reported results for trisubstituted 11-membered
macrocyclic double bond formation,2 it was first assumed
a Reagents and conditions: (a) K2CO3 (2.0 equiv), 3-bromo-2-
methylpropene or allyl bromide (1.5 equiv), acetone, 56 °C, 2 h,
90-93%; (b) Et2AlCN (1.2 equiv), toluene, 0 to 25 °C, 1.5 h, 79-
85%; (c) PCC (4.0 equiv), CH2Cl2, 40 °C, 7 h, 64-69%; (d)
concentrated H2SO4 (cat.), MeOH, 65 °C, 12 h, 85%; (e) LiHMDS
(1.1 equiv), THF, -78 °C, 1 h, then prenyl-Br (1.2 equiv), -78 to
0 °C, 3 h, 85%; (f) LDA (1.1 equiv), THF, slow addition of a
solution of starting material in THF/HMPA (7/1), -78 °C, 1 h;
then prenyl-Br (2.0 equiv), -78 to 20 °C, 12 h, 85%; (g) 1.0 M
HCl/THF (10/1), 25 °C, 14 h, 95%; (h) Et3N (2.0 equiv), THF, 25
°C, 12 h, 92-95%; (i) Et3N (2.0 equiv), 4-DMAP (1.0 equiv), THF,
25 °C, 3 d, 81-85%; (j) excess CH2N2, Et2O, 0 °C, 1 h, 32-45%
of 6a-d plus 30-40% of 7a-d.
to validate these compounds as models initially by ensuring
they followed the stereochemical precedent.2 Subsequently,
(5) Prunet, J. Angew. Chem., Int. Ed. 2003, 42, 2826.
(4) For examples of the synthesis of 11-membered rings by using olefin
metathesis see: (a) El Sukkari, H.; Gesson, J. P.; Renoux, B. Tetrahedron
Lett. 1998, 39, 4043. (b) Winkler, J. D.; Holland, J. M.; Kasparec, J.;
Axelsen, P. H. Tetrahedron 1999, 55, 8199. (c) Hoye, T. R.; Promo, M. A.
Tetrahedron Lett. 1999, 40, 1429. (d) Arisawa, M.; Kato, C.; Kaneko, H.;
Nishida, A.; Nakagawa, M. J. Chem. Soc., Perkin Trans. 1 2000, 1873. (e)
Fu¨rstner, A.; Radkowski, K.; Wirtz, C.; Goddard, R.; Lehmann, C. W.;
Mynott, R. J. Am. Chem. Soc. 2002, 124, 7061.
(6) E- and Z-configured isomers at ∆7,8 of 7a,b were separated by flash
column chromatography. Z isomers proved inactive to catalyst I, which is
consistent with previous observation in a similar substrate.2
(7) The Z geometry of the double bond (∆14,15) was unambiguously
confirmed by extensive nOe studies.
(8) The two geometrical isomers (∆7,8) of 7d were independently treated
with Grubbs’ catalyst after their careful chromatographic separation. The
reaction conditions and the yields were similar for both.
206
Org. Lett., Vol. 6, No. 2, 2004

that the Z-macrocycle represents the thermodynamic product.
Since bisubstituted macrocyclic olefins within strained ring
systems are known to readily undergo ring-opening meta-
thesis (ROM) reactions and, thus, can equilibrate to the
thermodynamically most stable product, it was also reason-
able to assume that 9a represented just such a thermodynamic
product. To confirm this explanation, and despite our initial
intention to test general 11-membered macrocycles, we were
forced to move closer toward the coleophomone skeleton
by adding an extra methyl group (R1 ) Me) to the previously
examined compounds.
Scheme 3. Olefin Metathesis of Bisprenylated Substrates 6c
and 6d: Mechanistic Rationale of the Cascade RCM/ROM/
RCM Transformation of 6d into the Spirocycle 11a
When regioisomers 6b and 7b6 (for their preparation see
Scheme 1) were independently treated with Grubbs’ catalyst
(I, 0.1 equiv) in refluxing dichloromethane, 8b and 9b were
correspondingly formed, as the sole products, and isolated
in good yields (68% and 71%, respectively, Scheme 2). The
newly formed macrocyclic double bond was, in each case,
confirmed7 as having the Z-stereochemical arrangement.
Once again and in contrast to the previous report,2 the
regiochemistry of the methyl enol ether used to protect the
tricarbonyl moiety was, therefore, not influential in changing
the geometry of the macrocyclic double bond in the products
obtained from these RCM reactions.
Having examined the stereoselectivity of the RCM reaction
of monoprenylated 6a,b and 7a,b we proceeded to examine
the bisprenylated congeners 6c and 6d in an effort to realize
which structural alteration was responsible for the observed
discordance. Not surprisingly, treatment of 6c with catalyst
I under the established conditions afforded exclusively the
Z-macrocycle7 (8c, 88%, Scheme 3). On the other hand,
application of the same conditions for 15 min to substrate
6d resulted in the formation of an inseparable mixture of
macrocycle 8d and spirocycle 11 (8d:11 1:1.1, 95%).
Increasing the reaction time further to 3 h led to the complete
conversion of 8d to the more thermodynamically stable
product, spirocycle 11 (Scheme 3). These results strongly
suggest that trisubstituted macrocyclic bonds of this type (e.g.
as in 8c) are formed in a nonreversible reaction, presumably
because the newly formed trisubstituted double bond is too
sterically encumbered to easily react further with the large
metathesis catalyst, while in the case of a bisubstituted
macrocyclic double bond (e.g. 8d) ring-opening metathesis
(ROM) is facile such that the product distribution eventually
funnels through to the thermodynamic product, spirocycle
11. This RCM/ROM/RCM cascade transformation (6d f
11) is illustrated in Scheme 3.
a Reagents and conditions: for R1 ) Me (a) cat. I (0.1 equiv),
CH2Cl2, 40 °C, 1.5 h, 88% of 8c; for R1 ) H (a) cat. I (0.1 equiv),
CH2Cl2, 40 °C, 45% of 8d plus 50% of 11 after 15 min; and 87%
of spectroscopically pure 11 after 3 h.
mation, however, was once again the exclusive formation
of the Z-macrocyclic double bond in 9c (∆14,15).7
Scheme 4. Olefin Metathesis of Bisprenylated Substrates 7c
and 7da
When a 1:1 mixture of geometrical isomers of 7c was
subjected to the action of catalyst I, a 3:1 mixture of E/Z
isomers (∆7,8) of 9c was isolated (Scheme 4). This transfor-
mation increased the ratio of geometrical isomers in favor
of the E isomer. This observation may be attributed to a
closer proximity between the reacting olefins in the E
substrate and/or to the formation of a more strained macro-
cycle when starting from the Z isomer. The sluggish
participation in the RCM reaction and the resultant low yield
of macrocycle from the Z isomer (∆7,8) of 7c is responsible
for the moderate overall yield obtained from RCM of the 7c
mixture (63%). The most important feature of this transfor-
a Reagents and conditions: (a) cat. I (0.1 equiv), CH2Cl2, 40
°C, 3 h, 63% of 9c; (b) cat. I (0.1 equiv), CH2Cl2, 40 °C, 20 min,
81% of 12.
Org. Lett., Vol. 6, No. 2, 2004
207

Reaction of the two geometrical isomers (∆7,8) of 7d8 with
catalyst I resulted in the rapid formation of spirocycle 12.
Only trace amounts of the more polar macrocycle 9d were
isolated following the application of the reaction conditions
(20 min at 40 °C, Scheme 4).
Scheme 5. Possible Mechanistic Explanation for the Complete
Inversion of the Stereoselectivity of RCM Based on the
Substitution at C1
In conclusion, the reaction of substrates 6a-c (where
tricarbonyl protection places the methyl enol ether within
the pendant six-membered ring) follows the precedent set
for RCM reactions of this type by Nicolaou’s synthesis of
coleophomone C,2 wherein the more thermodynamically
stable Z-macrocyclic olefin is formed exclusively. To our
surprise, and in stark contrast to the coleophomone B
precedent, however, RCM of substrates 7a-c (where tricar-
bonyl protection situates the methyl enol ether at the benzylic
position) also results in the stereospecific formation of the
Z macrocycle. To explain this anomaly and draw a general
conclusion the structural differences between our compounds
(7a-d) and the intermediates used for the construction of
coleophomone macrocycles must be considered. The only
remaining difference, after the extensive and thorough
modifications described above, is the substitution at the
aromatic position (C1). When C1 bears a large bulky
substituent (R3) rotation around the C6-C7 bond is severely
restricted. A large R3 group also, therefore, forces the
aromatic ring to stay out of conjugation and perpendicular
to the plane of the C7dC8 bond. This arrangement favors
(based on molecular model studies) the formation of the trans
metalocyclobutane intermediate (see TSb, Scheme 5), lead-
ing to the (E)-macrocycle. On the other hand, rotation around
the C6-C7 bond is much easier when C1 is not substituted
(R3 ) H) as in compounds 7a-d, thus affording the
opportunity for this system to adopt a conjugated conforma-
tion in which the dihedral angle between the aromatic ring
and the C7dC8 double bond is closer to 0. In this instance,
the formation of the cis metalocyclobutane between the two
coupling partners, far away from the main skeleton of the
molecule (see TSa, Scheme 5), is favored. On the basis of
the previous discussion, the reason for the formation of the
Z macrocycles from 6a-c now becomes apparent, since
rotation around both C6-C7 and C7-C8 bonds is permitted
with these examples.
truth highly dependent on the individual molecular environ-
ment. Even a subtle difference in substitution far away from
the reaction center is capable of resulting in complete
reversion of the configuration at the newly formed double
bond. However, once this assessment has been made empiri-
cally RCM is a capable and efficient tool for making even
highly strained macrocycles stereoselectively. Furthermore,
the RCM reaction is not reversible when the olefin formed
is trisubstituted, even in very strained macrocycles, but is
readily reversible when said olefin is bisubstituted. It would
appear that the mystery lies not with the metathesis reaction,
which is eminently predictable, but with the conformational
character of these substrates and macrocycles individually.
Acknowledgment. This work was supported financially
by the Greek Ministry of Education (B′ EPEAEK Graduate
Program).
Medium-sized macrocycles (8-13 membered) exhibiting
an intrinsic olefin are a common feature in natural products
with widespread medicinal, agricultural, and perfumery
applications. Despite our initial design endeavoring to seek
out rules for the geometrical selectivity of RCM reaction in
this type of substrates, the general, and pertinent, conclusion
suggested by current work is that the absolute prediction of
which of the two possible geometric arrangements for the
olefins in these macrocycles is preferred upon RCM is in
Supporting Information Available: 1H and 13C NMR
spectra of the olefin metathesis precursors and products;
atropisomerization of macrocycles 8b, 8c, 9b, and 9c results
in a broadening of some 1H NMR signals and makes
acquiring satisfactory 13C NMR spectra impossible. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL036156W
208
Org. Lett., Vol. 6, No. 2, 2004