In 2011, Gutekunst and Baran described both the first example of sequential
catalytic C(sp
3 )–H arylations in total synthesis and of transition-metal-catalyzed
activation of C–H bonds of a cyclobutane ring [29]. This iterative C–H functionalization strategy provided an efficient access to unsymmetrical cyclobutanes of
biological interest while avoiding pitfalls of classical cyclobutane synthesis via
photoinduced [2+2] cross-dimerization: head-to-head and head-to-tail additions,
homodimerization, and E/Z isomerization of olefin precursors, generally leading to
the uncontrolled production of a complex mixture of regio- and stereoisomers [30,
31]. To achieve the synthesis of piperarborenines (Scheme 4), Baran and co-worker
selected the [2-(methylthio)phenyl]carbamoyl derivative 16 instead of the
8-aminoquinoline directing group, because the hydrolysis of the former was
reported to occur under milder conditions [32]. After optimization, they found
that the addition of HFIP and pivalic acid [33] is critical to perform the first C
(sp
3 )–H arylation with complete regio- and diastereoselectivity, to give cis-configured product 17 on gram scale. Taking advantage of divergent epimerization to
obtain diastereoisomers 18 and 19 in good yield and stereoselectivity, they successfully performed the second diastereoselective C(sp
3 )–H arylation under similar
conditions, thus affording tetrasubstituted cyclobutanes 20 and 21.
Piperarborenines B and D were then synthesized from these intermediates in 2–3
steps via transformation of the amide and ester groups into carboxylic acids,
followed by condensation with dihydropyridone. Overall, this iterative C–H
functionalization strategy allowed to access both natural products in 6–7 steps, 7–
12% overall yield. Later on, Baran and co-workers reported an extension of this
strategy to the sequential C–H arylation/alkenylation of cyclobutanes, which
allowed to synthesize pipercyclobutanamide A, another congener of the same
family of natural products [34, 35].
During their studies on the total synthesis of podophyllotoxin based on a
Pd-catalyzed C(sp
3 )–H arylation strategy (Scheme 5) [36], Ting and Maimone
reported subtle conformational effects on reductive elimination pathways. Indeed,
when precursor 22 was engaged in directed C(sp
3 )–H arylation under usual conditions, β-lactam 23 was unexpectedly isolated as the major product. In the past few
years, the direct C–N bond reductive elimination of the nitrogen atom of the amide
directing group has been well documented [37–42]. To understand and suppress
this undesired pathway, the authors carried out X-ray diffraction analyses of the
acetonitrile-bound Pd
II complex arising from the C–H activation step. They identified that the environment of this palladacycle is highly congested and affected by
the conformation of the cyclohexene ring. As a consequence, they prepared the
conformationally distinct substrate 24 as new a precursor of the C–H activation
process. After significant optimization including the use of dibenzylphosphate as an
additive [43–45], the desired C–C bond formation was performed in 58% yield. To
finish, a simple treatment of the arylated product with a TFA/THF/H 2 O mixture
afforded podophyllotoxin and C-4 epi-podophyllotoxin, due to the epimerizable
character of the C4 stereogenic center. This strategy provided a five-step synthesis
of podophyllotoxin from commercially available bromopiperonal and a straightforward entry into novel arene-modified analogues.
138
D. Dailler et al.
catalytic C(sp
3 )–H arylations in total synthesis and of transition-metal-catalyzed
activation of C–H bonds of a cyclobutane ring [29]. This iterative C–H functionalization strategy provided an efficient access to unsymmetrical cyclobutanes of
biological interest while avoiding pitfalls of classical cyclobutane synthesis via
photoinduced [2+2] cross-dimerization: head-to-head and head-to-tail additions,
homodimerization, and E/Z isomerization of olefin precursors, generally leading to
the uncontrolled production of a complex mixture of regio- and stereoisomers [30,
31]. To achieve the synthesis of piperarborenines (Scheme 4), Baran and co-worker
selected the [2-(methylthio)phenyl]carbamoyl derivative 16 instead of the
8-aminoquinoline directing group, because the hydrolysis of the former was
reported to occur under milder conditions [32]. After optimization, they found
that the addition of HFIP and pivalic acid [33] is critical to perform the first C
(sp
3 )–H arylation with complete regio- and diastereoselectivity, to give cis-configured product 17 on gram scale. Taking advantage of divergent epimerization to
obtain diastereoisomers 18 and 19 in good yield and stereoselectivity, they successfully performed the second diastereoselective C(sp
3 )–H arylation under similar
conditions, thus affording tetrasubstituted cyclobutanes 20 and 21.
Piperarborenines B and D were then synthesized from these intermediates in 2–3
steps via transformation of the amide and ester groups into carboxylic acids,
followed by condensation with dihydropyridone. Overall, this iterative C–H
functionalization strategy allowed to access both natural products in 6–7 steps, 7–
12% overall yield. Later on, Baran and co-workers reported an extension of this
strategy to the sequential C–H arylation/alkenylation of cyclobutanes, which
allowed to synthesize pipercyclobutanamide A, another congener of the same
family of natural products [34, 35].
During their studies on the total synthesis of podophyllotoxin based on a
Pd-catalyzed C(sp
3 )–H arylation strategy (Scheme 5) [36], Ting and Maimone
reported subtle conformational effects on reductive elimination pathways. Indeed,
when precursor 22 was engaged in directed C(sp
3 )–H arylation under usual conditions, β-lactam 23 was unexpectedly isolated as the major product. In the past few
years, the direct C–N bond reductive elimination of the nitrogen atom of the amide
directing group has been well documented [37–42]. To understand and suppress
this undesired pathway, the authors carried out X-ray diffraction analyses of the
acetonitrile-bound Pd
II complex arising from the C–H activation step. They identified that the environment of this palladacycle is highly congested and affected by
the conformation of the cyclohexene ring. As a consequence, they prepared the
conformationally distinct substrate 24 as new a precursor of the C–H activation
process. After significant optimization including the use of dibenzylphosphate as an
additive [43–45], the desired C–C bond formation was performed in 58% yield. To
finish, a simple treatment of the arylated product with a TFA/THF/H 2 O mixture
afforded podophyllotoxin and C-4 epi-podophyllotoxin, due to the epimerizable
character of the C4 stereogenic center. This strategy provided a five-step synthesis
of podophyllotoxin from commercially available bromopiperonal and a straightforward entry into novel arene-modified analogues.
138
D. Dailler et al.