In Situ Methylene Capping: A General Strategy for Efficient Stereoretentive Catalytic Olefin Metathesis. the Concept, Methodological Implications, and Applications to Synthesis of Biologically Active Compounds

Article abstract of DOI:10.1021/jacs.7b06552

In situ methylene capping is introduced as a practical and broadly applicable strategy that can expand the scope of catalyst-controlled stereoselective olefin metathesis considerably. By incorporation of commercially available Z-butene together with robust and readily accessible Ru-based dithiolate catalysts developed in these laboratories, a large variety of transformations can be made to proceed with terminal alkenes, without the need for a priori synthesis of a stereochemically defined disubstituted olefin. Reactions thus proceed with significantly higher efficiency and Z selectivity as compared to when other Ru-, Mo-, or W-based complexes are utilized. Cross-metathesis with olefins that contain a carboxylic acid, an aldehyde, an allylic alcohol, an aryl olefin, an α substituent, or amino acid residues was carried out to generate the desired products in 47-88% yield and 90:10 to >98:2 Z:E selectivity. Transformations were equally efficient and stereoselective with a ～70:30 Z-:E-butene mixture, which is a byproduct of crude oil cracking. The in situ methylene capping strategy was used with the same Ru catechothiolate complex (no catalyst modification necessary) to perform ring-closing metathesis reactions, generating 14- to 21-membered ring macrocyclic alkenes in 40-70% yield and 96:4-98:2 Z:E selectivity; here too, reactions were more efficient and Z-selective than when the other catalyst classes are employed. The utility of the approach is highlighted by applications to efficient and stereoselective syntheses of several biologically active molecules. This includes a platelet aggregate inhibitor and two members of the prostaglandin family of compounds by catalytic cross-metathesis reactions, and a strained 14-membered ring stapled peptide by means of macrocyclic ring-closing metathesis. The approach presented herein is likely to have a notable effect on broadening the scope of olefin metathesis, as the stability of methylidene complexes is a generally debilitating issue with all types of catalyst systems. Illustrative examples of kinetically controlled E-selective cross-metathesis and macrocyclic ring-closing reactions, where E-butene serves as the methylene capping agent, are provided.

Full text of DOI:10.1021/jacs.7b06552

Article
pubs.acs.org/JACS
In Situ Methylene Capping: A General Strategy for Efficient
Stereoretentive Catalytic Olefin Metathesis. The Concept,
Methodological Implications, and Applications to Synthesis of
Biologically Active Compounds
Chaofan Xu,† Xiao Shen,† and Amir H. Hoveyda*
Department of Chemistry, Merkert Chemistry Center, Boston College, Chestnut Hill, Massachusetts 02467, United States
S
* Supporting Information
ABSTRACT: In situ methylene capping is introduced as a practical and broadly applicable strategy that can expand the scope of
catalyst-controlled stereoselective olefin metathesis considerably. By incorporation of commercially available Z-butene together
with robust and readily accessible Ru-based dithiolate catalysts developed in these laboratories, a large variety of transformations
can be made to proceed with terminal alkenes, without the need for a priori synthesis of a stereochemically defined disubstituted
olefin. Reactions thus proceed with significantly higher efficiency and Z selectivity as compared to when other Ru-, Mo-, or W-
based complexes are utilized. Cross-metathesis with olefins that contain a carboxylic acid, an aldehyde, an allylic alcohol, an aryl
olefin, an α substituent, or amino acid residues was carried out to generate the desired products in 47−88% yield and 90:10 to
>98:2 Z:E selectivity. Transformations were equally efficient and stereoselective with a ∼70:30 Z-:E-butene mixture, which is a
byproduct of crude oil cracking. The in situ methylene capping strategy was used with the same Ru catechothiolate complex (no
catalyst modification necessary) to perform ring-closing metathesis reactions, generating 14- to 21-membered ring macrocyclic
alkenes in 40−70% yield and 96:4−98:2 Z:E selectivity; here too, reactions were more efficient and Z-selective than when the
other catalyst classes are employed. The utility of the approach is highlighted by applications to efficient and stereoselective
syntheses of several biologically active molecules. This includes a platelet aggregate inhibitor and two members of the
prostaglandin family of compounds by catalytic cross-metathesis reactions, and a strained 14-membered ring stapled peptide by
means of macrocyclic ring-closing metathesis. The approach presented herein is likely to have a notable effect on broadening the
scope of olefin metathesis, as the stability of methylidene complexes is a generally debilitating issue with all types of catalyst
systems. Illustrative examples of kinetically controlled E-selective cross-metathesis and macrocyclic ring-closing reactions, where
E-butene serves as the methylene capping agent, are provided.
selective olefin metathesis.5 Alkyl/oxo catalysts3 have proven to
be severely inefficient in promoting transformations with aryl-
INTRODUCTION
■
A hallmark of Ru-dichloro catalysts, which have been pivotal to
substituted or hindered terminal olefins and/or when an
the emergence of olefin metathesis as a central transformation
in chemistry,1 is their ability to promote reactions of hindered
aldehyde is present, and their activity is entirely inhibited by a
carboxylic acid.6 Ru-based catechothiolate catalysts tolerate
alkenes and in the presence of commonly occurring polar
such functional groups,6a but the derived methylidene species
functional units such as a neighboring alcohol, an aldehyde, or a
decompose too rapidly and reactions involving terminal
alkenes,6a the most accessible substrate class, are low yielding.
Here, we demonstrate that by adding Z-butene to a reaction
mixture that contains an appropriate Ru dithiolate catalyst,
carboxylic acid. Grubbs and co-workers reported in 2011 that
kinetic control of Z-selectivity2 can be achieved by substituting
anionic chlorides with an alkyl and an oxo ligand.3 In 2013, we
disclosed the design of a sterically and electronically distinct set
of Ru-based complexes, containing a monodentate N-
heterocyclic carbene and, more importantly, a bidentate
disulfide4 ligand, which may be used for efficient catalytic Z-
Received: June 23, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
methylidene complex formation can be avoided, and, as a result,
an unprecedented range of terminal alkenes may be converted
to the desired products by efficient and exceptionally Z-
selective cross-metathesis (CM) and macrocyclic ring-closing
metathesis (MRCM) processes. Equally notable is that
stereoselectivity is generally higher when the methylene
capping strategy is adopted. The broad applicability and
general utility of the approach are highlighted by concise (i.e.,
little or no protection/deptrotection sequences) and stereo-
selective syntheses of a platelet aggregate inhibitor,7 prosta-
glandins E2 and F2α,8 as well as a 14-membered ring stapled
peptide that is a precursor to a potent opioid receptor agonist.9
The importance of these difficulties is manifested in the
context of Z-selective olefin metathesis reactions catalyzed by
Ru-based carbenes2 (Scheme 1b). Complexes Ru-1a,b may be
used for homocoupling,3a CM,3b or MRCM11 with mono-
substituted olefins; the derived methylidene complexes are
sufficiently stable for transformations to proceed with
reasonable efficiency. Nonetheless, a small number of instances
notwithstanding,12 reactions are inefficient with olefins
containing an allylic or homoallylic substituent or when an
aldehyde and/or an allylic alcohol is present, and generally no
product is generated when a substrate contains a carboxylic
acid.6a Moreover, there are no reported examples of reactions
with aryl olefins. These are major shortcomings especially
because the principal advantage of Ru (vs Mo or W13)
complexes is that an electrophilic or a Brønsted acid moiety
typically does not inhibit catalyst activity.
Catechothiolate Ru species (e.g., Ru-2a, Scheme 1b) offer
functional group compatibility that is similar to that of the
original dichloride derivatives6 and have been used for Z-
selective ring-opening/cross-metathesis,14 ring-opening meta-
thesis polymerization,15 and CM.6 However, we have shown
that the corresponding methylidene complexes (cf., iii) are too
unstable (partly due to 1,2-shift of the sulfide trans to the NHC
to the carbene carbon),6 and thus reactions with terminal
olefins are typically inefficient (i.e., ≤10% conversion). There
are only a few high yielding examples where one of the
substrates is a terminal alkene; in these cases, either the olefin
partner is much more reactive14 (strained bicyclic alkene) so
that homocoupling of the former can be largely avoided, and/or
H-bonding (between Z-2-butene-1,4-diol and sulfide ligands)
allows for methylidene formation to be minimized.6a There are
therefore no available Z-selective cross-metathesis protocols
that can be used efficiently with a wide range of
monosubstituted olefins that might contain an allylic, aryl,
alkyl, or a hydroxy substituent and/or an aldehyde or a
carboxylic acid moiety.
RESULTS AND DISCUSSION
■
1. General Problem of Methylidene Complex Inter-
mediacy. Monosubstituted alkenes are inexpensive and widely
accessible compounds that can be converted to valuable
derivatives by olefin metathesis.1 A general and key problem
is that reaction of a carbene or alkylidene (i, M = Ru, Mo or W,
respectively; Scheme 1a) with a terminal olefin releases the
Scheme 1. Methylidene Problem and the State-of-the-Art in
Z-Selective Olefin Metathesis with Ru-Based Complexes
2. Concept of in Situ Methylene Capping. In search of a
solution, we envisaged a strategy that may be referred to as in
situ methylene capping. This would entail incorporating a third
alkene component (A, Scheme 2), the CM of which with
monosubstituted olefins (i.e., formation of v) would be faster
than homocoupling or CM of terminal olefins a or b, which
would instead generate an unstable methylidene complex (iii).
Ideally, compound A would be efficiently converted to Ru-
carbene v, which might then react with a terminal alkene to
afford Z-1,2-disubstituted olefin intermediates viii-1 and viii-2
via carbene vii. Removal of excess A (to minimize non-
productive metathesis) and terminal alkene byproduct c (to
avoid methylidene formation) in vacuo would allow for efficient
CM between viii-1 and viii-2, affording the final product and
regenerating the capping agent (vs ethylene). We could
contemplate the above sequence because, unlike alkyl-oxo
species Ru-1a,b, catechothiolate complexes are able to promote
reactions between disubstituted olefins.6
The ideal capping agent must possess several key attributes.
Its substituents (G in A, Scheme 2) are to be sufficiently large
to ensure minimal methylidene formation (i.e., high vi/ix
ratio), and yet it needs to be small enough for A to compete
effectively with a terminal alkene for reaction with Ru-2a so
that complex v can be generated efficiently (vs methylidene iii
from homocoupling and/or cross-metathesis of a and/or b).
The capping agent should allow the transformation to be
exceptionally stereoretentive; its high stereoisomeric purity
homocoupling product via metallacyclobutane ii (one stereo-
isomer shown) and then methylidene species iii, which can
decompose at a rate10 that is fast enough to bring a
transformation to a halt. Further complicating matters, the
ethylene byproduct can convert the relatively stable i to
methylidene iii. Ethylene removal under vacuum can improve
matters to some extent, but methylidene formation is
unavoidable as long as a monosubstituted alkene is present.
B
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 2. Concept of in Situ Methylene Capping in Olefin Metathesis
must be directly reflected in the final product. Because excess A
would have to be removed later, it is imperative that it is readily
accessible, inexpensive, and appropriately volatile.
3. Identifying an Effective Methylene Capping Agent.
We began by investigating the homocoupling reactions of 1a
and Z-1,2-disubstituted alkenes 1b−d (Scheme 3). There was
minimal conversion to 2 when 1a was subjected to 2.0 mol %
Ru-2a (<5% conversion (conv), 100 Torr, 22 °C, 8 h). In
contrast, with methyl-substituted 1b (G = Me), the desired
product (2) was obtained in 74% yield and >98:2 Z:E ratio.
focus on cases where high efficiency and/or stereoselectivity are
possible only with the combination of Ru-2a and Z-3 (Scheme
4). With 5.0 mol % Ru-2a, homoallylic alcohol 4 was isolated in
56% yield and 95:5 Z:E selectivity; there was <20% conversion
without Z-3. With Ru-1a there was <5% conversion when the
derived Z-methyl-substituted alkenes were used as substrates,
and with the corresponding terminal alkenes under the
previously disclosed conditions11b,12 [5.0 mol %, THF (0.5
M), 35 °C, 8 h], 4 was isolated in 49% yield and 68:32 Z:E
selectivity. Unlike other instances to be discussed below
(Scheme 4), where 1.0 mol % Ru-2a and 1 h is sufficient at
the first stage, longer reaction times were needed for the
homoallylic alcohol, perhaps because internal chelation of the
hydroxyl group with the metal center can lower catalyst activity.
Despite the presence of excess carboxylic acid (3.0 equiv),
products that also contain an unprotected indole (5b), a
hydroxy unit (5c), an aldehyde group (5d), or a β-branched
substituent (5e,f) were isolated in 58−74% yield and 97:3 to
>98:2 Z:E selectivity; in all cases, there was <5% conversion
without the capping agent. When 2.0 mmol of the unsaturated
ester substrate was used, ∼0.3 g of 5a could be isolated (87%
conv, 56% yield) in 98:2 Z:E selectivity. para-Ketone-
substituted phenol 6a was prepared efficiently and with high
Z-selectivity; this alkene was isolated in 28% yield and 88:12
Z:E selectivity with Ru-1a (under formerly reported con-
ditions11b,12). Compounds 6b and 6c were secured in 58% and
47% yield and 98:2 and 90:10 Z:E selectivity, respectively; in
the case of 6b, the terminal alkene bearing an allylic methyl
group does not readily undergo homocoupling and was
therefore used directly (without methylene capping). More-
over, when 6.5 mol % Ru-2a was used in the second stage of
the process, 6b was isolated in 82% yield (89% conv vs 58%
yield with 4.0 mol %) and 98:2 Z:E selectivity. Again, with Ru-
1a, reactions between the corresponding terminal alkenes were
much less efficient and stereoselective (6b and 6c in 76:24 and
55:45 Z:E selectivity and 14% and 29% yield, respectively). The
lower efficiency in the formation of allylic alcohol 6c is not due
to adventitious alkene isomerization because none of the
ketone byproduct could be detected. Chelation of the hydroxy
group with the Ru center might reduce catalytic activity, an
effect that is likely less critical with reactions of less substituted,
and thus more reactive, primary allylic alcohols.6
Scheme 3. Preliminary Studies: Identifying an Effective
a
Capping Agent
a
Reactions were carried out under N2 atm. Conversion and Z:E ratios
determined by analysis of 1H or 13C NMR spectra of unpurified
product mixtures ( 2%). Yields for purified products ( 5%).
Experiments were run in duplicate or more. See the Supporting
Information for details.
There appears to be a strict requirement on the size of G:
reactions with a slightly larger ethyl- and n-butyl-substituted 1c
and 1d were far less efficient (26% and <5% conv, respectively).
These findings indicated that Z-butene (Z-3) would be a
suitable capping agent. Indeed, treatment of a 1:5 mixture of 1a
with a THF solution of commercially available Z-3 (unpurified)
to 3.0 mol % Ru-2a afforded 2 in 70% yield and 98:2 Z:E
selectivity. Although when a single batch of 5.0 mol % Ru-2a
was used, the efficiency and stereoselectivity were similar, the
sequential addition procedure allowed for lower total catalyst
loading, and we therefore adopted it from here on.
With monosubstituted aryl olefins, yields were low due to
4. Z-Selective Cross-Metathesis with a Methylene
Capping Agent. We then investigated Z-selective CM with
swift stilbene/methylidene complex formation. Accordingly, Z-
C
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 4. Catalytic Z-Selective Cross-Metathesis with Z-Butene as the Methyelene Capping Agent
a
(a) Reactions were carried out under N2 atm. Conversion and Z:E ratios determined by analysis of 1H or 13C NMR spectra of unpurified product
mixtures ( 2%). Yields for purified products ( 5%). For 7a−c, Z-β-methylstyrene was used (condition B); Z-1-propenylboronic acid pinacol ester
used to prepare 8 (condition C). (b) α-Branched terminal alkene was used directly (without treatment with Z-3). Experiments were run in duplicate
or more. See the Supporting Information for details.
β-methylstyrenes, accessed in one step by catalytic cross-
coupling between a commercially available arylboronic acid
pinacol ester and Z-1-bromopropene, were used to access 7a−c
in 48−64% yield and 92:8−96:4 Z:E selectivity. The present
approach is considerably more efficient and/or stereoselective
than what is feasible with alternative catalysts. For example, 7c
was obtained in 65:35 Z:E selectivity with Ru-1a (5.0 mol %,
THF, 35 °C, 4 h), and CM with aryl olefins and a tungsten
monoaryloxide pyrrolide (MAP) complex2a (more Z-selective
than Mo alkylidenes) afforded the thermodynamically favored
isomer predominantly (up to >98% E; fast postmetathesis
isomerization).
D
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Aldehyde-containing alkenyl-B(pin) 8 (54% yield, 95:5 Z:E)
was synthesized from commercially available Z-1-propenylbor-
onic acid pinacol ester. Z,E-Dienes 9a−c were accessed easily
(Scheme 4). The carboxylic acid in 9a is incompatible with Ru-
1a,b or various Mo-based catalysts. Synthesis of aryl-substituted
9b and 9c is inefficient with Ru-1a,b (≤25% yield),16 and less
stereoselective with a Mo complex due to postmetathesis Z-to-
E interconversion; for example, with a MAP catalyst containing
a pentafluorophenylimido ligand17 (5.0 mol %, 100 Torr, 22
°C, 15 min), 9b was obtained in 72% yield and 83:17 Z:E ratio
(vs 80% yield, 97:3 Z:E with Ru-2a).
Amino acid-derived alkenes 10a−c were obtained in 47−88%
yield and 91:9 to >98:2 Z:E selectivity (Scheme 4). In sharp
contrast, with Ru-1a and monosubstituted olefins derived from
amino acids glycine (see 10a) or methionine (see 10c), there
was complete catalyst inhibition (<10% yield),18 and with a Mo
catalyst,17 Z selectivity was entirely eradicated (e.g., 10b in 32%
yield and 50:50 Z:E with the aforementioned MAP complex vs
47% yield, 91:9 Z:E with Ru-2a).
An additional advantage of the strategy described here is that
by avoiding the highly reactive methylidene complexes, there is
less erosion of kinetic Z-selectivity by postmetathesis isomer-
ization; this is partly why reactions with Ru-1a and terminal
alkenes, regardless of the type of functional group present, are
often less Z-selective. Furthermore, the combination of Ru-2a
and the intermediacy of Z-methyl-substituted alkene inter-
mediates causes Z selectivities to be higher than when terminal
alkenes are directly involved (without the capping agent). As an
example, with Ru-2a, homocoupling of the ester-substituted
terminal alkene used to prepare 6c (Scheme 4) and its Z-
methyl-substituted alkene derivative proceeded with 78:22 Z:E
and 93:7 Z:E selectivity (14% and 79% yield), respectively.
Control experiments19 indicate that lower Z:E ratios in these
latter instances in all likelihood originate from a larger energy
gap between transition states leading to trisubstituted as
compared to disubstituted metallacyclobutanes I and II (R =
Me vs R = H; Scheme 5); greater steric pressure is exerted by
Reaction with E-3 is probably slower because of steric pressure
caused by the metallocyclobutane substituent oriented toward
the sizable NHC ligand.
5. Application of in Situ Methylene Capping to
Preparation of Z-Macrocyclic Alkenes. To probe applic-
ability of the capping strategy to MRCM, we first confirmed
that monosubstituted olefins in a diene substrate can be capped
efficiently and stereoselectively (Scheme 6a). Treatment of a
0.1 molar solution containing diene 11a and 20 equiv of Z-3
with 1.0 mol % Ru-2a delivered 11b in 89% yield and >98:2
Z,Z:Z,E selectivity (22 °C, 1 h); removal of excess Z-3 in vacuo
and addition of 4.0 mol % Ru-2a afforded macrocyclic alkene
12a in 56% yield and 96:4 Z:E selectivity (5.0 mM, 400 Torr,
12 h, 35 °C).
The same conditions with Ru-2a but without the capping
agent afforded 12a in only 11% yield and 59:41 Z:E selectivity
(Scheme 6b). There was <10% conversion with 10 mol % Ru-
1a and 11b. Cyclization of dienes 11c and 11d, with only one
Z-disubstituted olefin, was less stereoselective (87:13 and
86:14, respectively vs 96:4 Z:E for 11b). Moreover, Z:E ratios
for 12a are not time-dependent, which further underscores the
larger energy gap between transition states leading to
trisubstituted as compared to disubstituted metallacyclobutanes
(see Scheme 5).
The lower Z selectivity for 11c,d (Scheme 6b) might be due
to competitive homocoupling by the monosubstituted olefin
moieties. This generates internal alkene stereoisomeric mixtures
that can be directly converted to 12a with lower Z selectivity via
a Ru carbene derived from the Z-methyl-substituted alkene20
(“backbiting”). Thus, because of the capping strategy, which
diminishes the possibility of homocoupling side reactions and
methylidene formation, together with the ability of Ru-2a to
promote RCM of 1,2-disubstituted olefins efficiently, the scope
of Z-selective olefin metathesis can be expanded.
Scheme 5. Origin of Higher Z-Selectivity with 1,2-
Disubstituted Alkenes
6. Z-Selective Macrocyclic Ring-Closing Metathesis
with a Methylene Capping Agent. We synthesized 14- to
21-membered ring structures in 53−70% yield and in 96:4−
98:2 Z:E selectivity (Scheme 7a); acyclic dienes bearing two
terminal alkenes were used as starting materials. These results
compare favorably to reactions with Ru-1a,b, where higher
catalyst loadings (7.5 vs 5.0 mol %) and reaction temperatures
(60 vs 35 °C) as well as longer reaction times (24 vs 13 h) are
required11 and Z selectivity is generally lower.11a For example,
secondary alcohol 12f was generated in 56% yield and 65:35
Z:E ratio with Ru-1a as compared to 53% yield and 96:4 Z:E
selectivity with Ru-2a. These distinctions may be due to
postmetathesis isomerization caused by generation of methyl-
idene complexes in reactions of terminal alkenes with Ru-1a,b
as well as a smaller energy difference between transition states
for stereoisomeric disubstituted metallacyclobutanes (see
Scheme 5 and related discussion). As was noted, there is
minimal reaction between 1,2-disubstituted olefins when Ru-1a
was used, and, as a result, fully substituted metallacycle
intermediates cannot be accessed through this catalyst system.
the monodentate N-heterocyclic carbene (NHC) on the two
substituents in II when R is a methyl group (vs one Cα group
in a disubstituted metallacyclobutane when R is an H atom).
An important attribute of the approach is that with a
stereoisomeric mixture of E- and Z-butene (E- and Z-3),
obtained directly from cracking of crude oil and less expensive
than the corresponding pure forms of either isomer, can be
used to synthesize the desired products with similar efficiency
and stereoselectivity. The examples in eq 1 are representative.
E
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 6. Preliminary Studies for Macrocyclic Ring-Closing Metathesis
a
Reactions were carried out under N2 atm. Conversion (disappearance of the disubstituted alkene intermediate) and Z:E ratios determined by
analysis of 1H or 13C NMR spectra of unpurified product mixtures ( 2%). Yields for purified products ( 5%). Experiments were run in duplicate or
more. See the Supporting Information for details.
In a limited number of examples, Ru-1b,11b which contains a
larger NHC ligand, stereoselectivity was improved but at the
cost of diminished efficiency (e.g., 12f in 95:5 Z:E, 45% yield;
same conditions as Ru-1a).
higher stereochemical purity as compared to the formerly
reported route (42% overall yield, 96:4 Z:E selectivity in three
steps vs 30% overall yield, 80:20 Z:E in seven steps7).
Treatment of 16 with lithium hydroxide has been shown to
deliver 17.7
The case of carboxylic acid 12j (Scheme 7b) is notable
because Mo or W complexes are inapplicable,21 and there was
<5% conversion when Ru-1a was used [reported conditions,
1,2-dichloroethane (5.0 mM), 60 °C, 400 Torr, 12 h]. When
MRCM of 11a was performed in the presence of an equivalent
of 13, a highly electrophilic keto-aldehyde 12a was obtained
without significant change in yield or stereoselectivity.
7. Applications to Synthesis of Biologically Active
Compounds. The approach outlined above facilitates the
synthesis of a variety of biologically active molecules. As
highlighted by the examples discussed and illustrated below, the
robustness of dithiolate Ru catalysts and their tolerance of key
polar functional units, along with the various advantages
inherent in adopting the methylene capping strategy, combine
to make concise and highly stereoselective multistep sequences
possible.
a. Stereoselective Synthesis of a Platelet Aggregate
Inhibitor. Preparation of 17 (Scheme 8) demonstrates utility
and allows for exploration of the limits of the approach. With
Ru-1a, CM between 14 and methyl ester 15 afforded 16 in 35%
yield and 71:29 Z:E selectivity, and with Ru-2a, reaction
between 20 and carboxylic acid 1a was inefficient (e.g., 25%
conv at 20 mol % loading at 22 °C, 32 h; >98:2 Z:E). This is
probably because loss of catechothiolate ligand through
reaction with Brønsted acid,4b although sufficiently slow for
relatively efficient transformations with more reactive olefin
substrates, becomes too competitive when aryl olefins are
involved. With methyl ester 15, product 16 was isolated in 58%
yield and 96:4 Z:E ratio. The methylene-capping approach thus
furnished access to 16 with better efficiency and considerably
b. Stereoselective Synthesis of Prostaglandins without
Resorting to Protecting Groups. Prostaglandins E2 and F2α
(Scheme 9) belong to a family of potent eicosanoid lipid
mediators8 that play key roles in a range of physiological
disorders; several members are “block-buster” drugs.22 The
corresponding CM processes further challenge the limits of the
state-of-the-art due to the presence of two alcohol units and a
carboxylic acid moiety as well as the possibility of internal
chelation between the Ru-carbene with the nearby carbonyl
oxygen. Still, treatment of 21a and unsaturated carboxylic acid
1a with Z-3 and 2.0 and 1.0 mol % Ru-2a, respectively,
followed by mixing of the resulting mixtures (no isolation/
purification) and addition of Ru-2a (15 mol %) under reduced
pressure afforded prostaglandin E2 in 51% yield and >98:2 Z:E
selectivity (40% yield and >98:2 Z:E at 10 mol % loading).
Prostoglandin F2α was synthesized in 59% yield and >98:2
Z:E selectivity (Scheme 9). Separate treatment of 21a and 21b
(without 1a) with the capping agent was necessary so that low
catalyst loading would suffice (2.0 mol %). The solutions were
then mixed and treated with additional amounts of Ru-2a for
CM. The amount of Ru dithiolate complex needed to access
these multifunctional products is higher than before. Never-
theless, the present approach is still efficient overall as
stereoselectivity is exceptionally high, there is no need for
protection, and subsequent deprotection/hydrolysis to unmask
the hydroxy or the carboxylic acid moieties, which are
functional groups, was demonstrated to be central to the
exhibited agonist activity.23 By varying the identity of the
reaction partner (e.g., 1a), desirable analogues can be
F
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 7. Z-Selective Macrocyclic Ring-Closing Metathesis with Z-Butene as the Methylene Capping Agent
a
Reactions were carried out under N2 atm. Conversion (disappearance of the disubstituted alkene intermediate) and Z:E ratios determined by
analysis of 1H or 13C NMR spectra of unpurified product mixtures ( 2%). Yields for purified products ( 5%). Experiments were run in duplicate or
more. See the Supporting Information for details.
a
Scheme 8. Application to Stereoselective Synthesis of a Platelet Aggregate Inhibitor
a
1
Reactions were carried out under N2 atm. Conversion and Z:E ratios determined by analysis of H NMR spectra of unpurified product mixtures
( 2%). Yields for purified products ( 5%). Experiments were run in duplicate or more. See the Supporting Information for details.
synthesized directly from the same core molecule (e.g., 21),
accessible in high diastereo- and enantiomeric purity by
reported procedures.24
c. Stereoselective Synthesis of a Relatively Strained
Stapled Peptide. Hydrocarbon-stapled peptides9,26 are α-
helical ligands that have been used for gaining a greater
understanding of protein−protein associations and are leading
candidates for therapeutic regulation of intercellular inter-
actions.25 Reliable access to stapled peptides with stereochemi-
cally defined olefin linkages makes possible the examination of
conformational preferences on biological activity. Z-Selective
MRCM of peptidic dienes is feasible with Ru-1a,b18 but only
G
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 9. Z-Selective Cross-Metathesis with a Capping Agent: Application to Stereoselective Synthesis of Prostaglandins
a
1
Reactions were carried out under N2 atm. Conversion and Z:E ratios determined by analysis of H NMR spectra of unpurified product mixtures
( 2%). Yields for purified products ( 5%). Experiments were run in duplicate or more. See the Supporting Information for details.
Scheme 10. Z-Selective Macrocyclic RCM with a Capping
Agent: Stereoselective Synthesis of a Stapled Peptide
Scheme 11. Ru Catechothiolate in Paraffin Pellet:
Stereoselective Synthesis of a Stapled Peptide
a
a
Reactions were carried out under N2 atm. Conversion (disappearance
of the disubstituted alkene intermediate) and Z:E ratios determined by
1
analysis of H NMR spectra of unpurified product mixtures ( 2%).
Yield for purified 23 ( 5%). Experiments were run in duplicate or
more. See the Supporting Information for details.
peptide 23. In contrast, subjection of 22 to 2.0 mol % Ru-2a
and 3 (20 equiv; 22 °C, 12 h), followed by removal of the
capping agent in vacuo and addition of 10 mol % Ru-2a, led to
the formation of 23 in 72% yield and >98:2 Z:E selectivity.
Macrocyclic peptide 23 has been converted to a potent δ and μ
opioid receptor agonist after a pair of routine transformations.26
As noted previously, Ru catechothiolate complexes are
robust, as manifested by the fact that they retain their activity
in the presence of various polar functional units. Still, what
makes their use even more convenient is that the derived
after incorporation of extended hydrocarbon chains. This
measure is presumably to ensure placement of Ru carbenes at
an appropriately safe distance from a polar functionality, which
could lead to reaction inhibition (cf., CM reactions leading to
10a and 10c, Scheme 4).
Attempts to carry out MRCM of diene 22 (Scheme 10) with
Ru-1a did not yield any of the 14-membered ring macrocyclic
H
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
paraffin pellets,28 which may be sliced with a knife into smaller
fragments for multiple applications, may be utilized for further
ease of operation. The transformation illustrated in Scheme 11
is a case in point.
similar to Z-alkene products, a wide range of desirable linear
and macrocyclic E-olefins, many of which cannot be accessed
by the available protocols,29,30 can now be easily accessed.
There are reactions that can be promoted only by high-
oxidation-state Mo-based catalysts but require 1,2-disubstituted
olefin substrates.17 The in situ capping strategy should allow
these important catalytic processes to be performed with more
easily accessible terminal alkenes as well as deliver products
with higher efficiency and stereoisomeric purity.
Applications of the strategies outlined above to other
unresolved problems in stereoselective catalytic olefin meta-
thesis are in progress in these laboratories and will be disclosed
in due course.
CONCLUSIONS
■
The in situ methylene capping approach offers the possibility of
carrying out exceptionally Z-selective cross-metathesis and
MRCM with substrates that contain naturally ubiquitous
functional groups. This advance was made possible because
methylidene complex intermediacy can be bypassed by the
methylene capping strategy and also as a result of several key
attributes of ruthenium catechothiolate catalysts, the ability to
promote efficient and highly Z-selective reactions with 1,2-
disubstituted olefins, and broad functional unit compatibility.6
Although the strategy of a “sacrificial alkyne” has been applied
to a rather limited extent in catalytic nitrile-alkyne cross-
metathesis,27 olefin metathesis has not been the beneficiary of
this operationally simple and promising approach.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.7b06552.
■
S
Experimental details for all reactions and analytic details
for all products (PDF)
The strategies outlined here are likely to resolve other key
and long-standing problems in olefin metathesis. One relates to
the design of kinetically E-selective olefin metathesis
processes,29,30 for which the utility of the same class of
ruthenium complexes has been noted by others, albeit within a
rather confined substrate range.31 The transformations
illustrated in Scheme 12, involving the use of a closely related
AUTHOR INFORMATION
Corresponding Author
*amir.hoveyda@bc.edu
■
ORCID
Amir H. Hoveyda: 0000-0002-1470-6456
Scheme 12. Application to Kinetically Controlled E-Selective
Cross-Metathesis and Macrocyclic Ring-Closing Metathesis
Author Contributions
a
†C.X. and X.S. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by a grant from the National
Science Foundation (CHE-1362763). We are grateful to
Dongmin Xu for experimental assistance and to Thach T.
Nguyen and Elsie C. Yu for experimental data regarding
reactions with Mo MAP complexes and to Ming Joo Koh, Dr.
Sebastian Torker, and Malte S. Mikus for valuable discussions.
We thank Dr. Levente Ondi, Dr. Janos Balazs Czirok, and Dr.
Hasan Mehdi for their support and helpful advice and are
grateful to XiMo, AG for gifts of paraffin tablets that contain
Ru-2a.
REFERENCES
■
(1) Grubbs, R. H., Wenzel, A. G., O’Leary, D. J., Khosravi, E., Eds.
Handbook of Metathesis; Wiley−VCH: Weinheim, 2014.
a
Reactions were carried out under N2 atm. Conversion and Z:E ratios
(2) For reviews on catalyst-controlled Z-selective olefin metathesis,
see: (a) Hoveyda, A. H.; Malcolmson, S. J.; Meek, S. J.; Zhugralin, A.
R. Angew. Chem., Int. Ed. 2010, 49, 34−44. (b) Hoveyda, A. H. J. Org.
Chem. 2014, 79, 4763−4792. (c) Hoveyda, A. H.; Torker, S.; Khan, R.
M. K.; Malcolmson, S. J. In Handbook of Metathesis; Grubbs, R. H.,
O’Leary, D. J., Wentzel, A. G., Khosravi, E., Eds.; Wiley−VCH:
Weinheim, Germany, 2015; Vol. 2, pp 508−562.
(3) For selected Z-selective homocoupling and cross-metathesis
reactions promoted by complexes Ru-1a,b, see: (a) Keitz, B. K.; Endo,
K.; Herbert, M. B.; Grubbs, R. H. J. Am. Chem. Soc. 2011, 133, 9686−
9688. (b) Keitz, B. K.; Endo, K.; Patel, P. R.; Herbert, M. B.; Grubbs,
R. H. J. Am. Chem. Soc. 2012, 134, 693−699. (c) Herbert, M. B.; Marx,
V. M.; Pederson, R. L.; Grubbs, R. H. Angew. Chem., Int. Ed. 2013, 52,
310−314. (d) Quigley, B. L.; Grubbs, R. H. Chem. Sci. 2014, 5, 501−
506.
determined by analysis of 1H NMR spectra of unpurified product
mixtures ( 2%). Yields for purified products ( 5%). Experiments
were run in duplicate or more. See the Supporting Information for
details.
derivative of Ru-2a, dithiolate complex Ru-2b, are representa-
tive. The smaller size of the N-aryl units in the N-heterocyclic
carbene ligand of Ru-2b can accommodate the increased steric
hindrance in the corresponding metallacyclobutane (see II,
Scheme 5).29−31 The reason for larger amounts of readily
removable E-butene is probably its lower reactivity as compared
̀
to the aforementioned Z isomer, as noted previously vis-a-vis
the transformations in eq 1. These findings, which, to the best
of our knowledge, represent the first example of kinetically E-
selective MRCM reaction with a Ru complex, demonstrate that,
(4) For the original report regarding the design and discovery of Ru-
based dithiolate catalysts for Z-selective olefin metathesis, see:
I
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(a) Khan, R. K. M.; Torker, S.; Hoveyda, A. H. J. Am. Chem. Soc. 2013,
135, 10258−10261. For a mechanistic investigation that elucidated
why unlike catecholate Ru complexes the derived catechothiolates
species promote highly Z-selective olefin metathesis reactions, see:
(b) Khan, R. K. M.; Torker, S.; Hoveyda, A. H. J. Am. Chem. Soc. 2014,
136, 14337−14340. For a comprehensive study regarding the
influence of bidentate anionic and chelating ligands on reactivity on
stereoselectivity of olefin metathesis reactions promoted by this class
of catalysts, see: (c) Mikus, M. S.; Torker, S.; Xu, C.; Li, B.; Hoveyda,
A. H. Organometallics 2016, 35, 3878−3892.
(26) Mollica, A.; Guardiani, G.; Davis, P.; Ma, S.-W.; Porreca, F.; Lai,
J.; Mannina, L.; Sobolev, A. P.; Hruby, V. J. J. Med. Chem. 2007, 50,
3138−3142.
(27) Geyer, A. M.; Gdula, R. L.; Wiedner, E. S.; Johnson, M. J. A. J.
Am. Chem. Soc. 2007, 129, 3800−3801.
(28) Ondi, L.; Nagy, G. M.; Czirok, J. B.; Bucsai, A.; Frater, G. E. Org.
Process Res. Dev. 2016, 20, 1709−1716.
(29) For first examples of efficient kinetically controlled E-selective
cross-metathesis reactions, see: Nguyen, T. T.; Koh, M. J.; Shen, X.;
Romiti, F.; Schrock, R. R.; Hoveyda, A. H. Science 2016, 352, 569−
575.
(30) For first examples of kinetically controlled E-selective
macrocyclic ring-closing metathesis reactions, see: Shen, X.; Nguyen,
T. T.; Koh, M. J.; Xu, D.; Speed, A. W. H.; Schrock, R. R.; Hoveyda, A.
H. Nature 2017, 541, 380−386.
(31) Ahmed, T. S.; Grubbs, R. H. J. Am. Chem. Soc. 2017, 139, 1532−
1537.
(5) For Ru-based monochloride/sulfide complexes that have been
used to promote Z-selective homocoupling and cross-metathesis
reactions, see: (a) Occhipinti, G.; Hansen, F. R.; Tornroos, K. W.;
̈
Jensen, V. R. J. Am. Chem. Soc. 2013, 135, 3331−3334. (b) Occhipinti,
G.; Koudriavtsev, V.; Tornroos, K. W.; Jensen, V. R. Dalton Trans.
̈
2014, 43, 11106−11117.
(6) For the design of complex Ru-2a and its use in Z-selective cross-
metathesis reactions that afford a variety of allylic alcohols, see:
(a) Koh, M. J.; Khan, R. K. M.; Torker, S.; Yu, M.; Mikus, M.;
Hoveyda, A. H. Nature 2015, 517, 181−186. For an application in
complex molecule total synthesis, see: (b) Yu, M.; Schrock, R. R.;
Hoveyda, A. H. Angew. Chem., Int. Ed. 2015, 54, 215−220.
(7) Hachem, A.; Roussel, P.; Men
́ ́
ager, E.; Gree, D.; Le Floc’h, Y.;
Gree, R.; Cerletti, C.; Rolland, Y.; Simonet, S.; Verbeuren, T. Bioorg.
́
Med. Chem. Lett. 2002, 12, 2511−2514.
(8) Funk, C. D. Science 2001, 294, 1871−1875.
(9) Chu, Q.; Moellering, R. E.; Hilinski, G. J.; Kim, Y.-W.;
Grossmann, T. N.; Yeh, J. T.-H.; Verdine, G. L. MedChemComm
2015, 6, 111−119.
(10) Kirkland, T. A.; Lynn, D. M.; Grubbs, R. H. J. Org. Chem. 1998,
63, 9904−9909.
(11) (a) Marx, V. M.; Herbert, M. B.; Keitz, B. K.; Grubbs, R. H. J.
Am. Chem. Soc. 2013, 135, 94−97. (b) Rosebrugh, L. E.; Herbert, M.
B.; Marx, V. M.; Keitz, B. K.; Grubbs, R. H. J. Am. Chem. Soc. 2013,
135, 1276−1279.
(12) Quigley, B. L.; Grubbs, R. H. Chem. Sci. 2014, 5, 501−506.
(13) Schrock, R. R.; Hoveyda, A. H. Angew. Chem., Int. Ed. 2003, 42,
4592−4633.
(14) Koh, M. J.; Khan, R. K. M.; Torker, S.; Hoveyda, A. H. Angew.
Chem., Int. Ed. 2014, 53, 1968−1972.
(15) Mikus, M. S.; Torker, S.; Hoveyda, A. H. Angew. Chem., Int. Ed.
2016, 55, 4997−5002.
(16) Luo, S.-X.; Cannon, J. S.; Taylor, B. L. H.; Engle, K. M.; Houk,
K. N.; Grubbs, R. H. J. Am. Chem. Soc. 2016, 138, 14039−14046.
(17) Koh, M. J.; Nguyen, T. T.; Lam, J. K.; Torker, S.; Hyvl, J.;
Schrock, R. R.; Hoveyda, A. H. Nature 2017, 542, 80−85.
(18) Mangold, S. L.; O’Leary, D. J.; Grubbs, R. H. J. Am. Chem. Soc.
2014, 136, 12469−12478.
(19) See the Supporting Information for details.
(20) Conrad, J. C.; Eelman, M. D.; Duarte Silva, J. A.; Monfette, S.;
Parnas, H. H.; Snelgrove, J. L.; Fogg, D. E. J. Am. Chem. Soc. 2007, 129,
1024−1025.
(21) For first examples of catalytic Z-selective macrocyclic ring-
closing metathesis reactions and applications to complex molecule
total synthesis, see: (a) Yu, M.; Wang, C.; Kyle, A. F.; Jakubec, P.;
Dixon, D. J.; Schrock, R. R.; Hoveyda, A. H. Nature 2011, 479, 88−93.
(b) Wang, C.; Yu, M.; Kyle, A. F.; Jakubec, P.; Dixon, D. J.; Schrock, R.
R.; Hoveyda, A. H. Chem. - Eur. J. 2013, 19, 2726−2740.
(22) Nair, S. K.; Henegar, K. E. In Modern Drug Synthesis; Li, J. J.,
Johnson, D. S., Eds.; Wiley: New York, 2010; pp 329−338.
̀
(23) Ungrin, M. D.; Carriere, M.-C.; Denis, D.; Lamontagne, S.;
Sawyer, N.; Stocco, R.; Tremblay, N.; Metters, K. M.; Abramovitz, M.
Mol. Pharmacol. 2001, 59, 1446−1456.
(24) Coulthard, G.; Erb, W.; Aggarwal, V. K. Nature 2012, 489, 278−
281.
(25) Bird, G. H.; Mazzola, E.; Opoku-Nsiah, K.; Lammert, M. A.;
Godes, M.; Neuberg, D. S.; Walensky, L. D. Nat. Chem. Biol. 2016, 12,
845−852.
J
DOI: 10.1021/jacs.7b06552
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX