Strain release – an old tool for new transformations

Joanna Turkowska a, Jakub Durka ab and Dorota Gryko *a
aInstitute of Organic Chemistry Polish Academy of Sciences, Kasprzaka 44/52, Warsaw 01-224, Poland. E-mail: dorota.gryko@icho.edu.pl
bDepartment of Chemistry, Warsaw University of Technology, Noakowskiego 3, Warsaw 00-664, Poland

Received 7th March 2020 , Accepted 24th April 2020

First published on 24th April 2020


Strain-release driven transformations give access to attractive bioisosteric motifs highly prized by medicinal chemists and they are characteristic of molecules possessing distorted bond lengths and angles. By broadening the chemical space in drug discovery, recently, these compounds have attracted a lot of interest. Their reactivity stems mainly from an increased energy and destabilization. As a result, the opening of the bridging bond occurs under the action of both nucleophiles and electrophiles as well as radical species and transition metals. Though the bridge bond dominates their reactivity, it is also influenced by the substitution pattern. This feature article focuses on strain-release driven strategies paying particular attention to the most recent (year > 2010) advances.


image file: d0cc01771j-p1.tif

Joanna Turkowska

Joanna Turkowska joined the research group of Prof. D. Gryko in 2016. In 2017, she obtained a Master of Science in Chemistry at the Warsaw University. Currently, she is undertaking a PhD at the Institute of Organic Chemistry of the Polish Academy of Sciences. Her research interests include cobalt catalysis in organic synthesis and biomimetic chemistry.

image file: d0cc01771j-p2.tif

Jakub Durka

Jakub Durka was born in Poniatowa, Poland, in 1996. He joined the research group of Prof. D. Gryko in 2017. Two years later, he obtained a Bachelor of Science in Chemical Technology at the Warsaw University of Technology. His current research interests include photoredox catalysis and chemistry of donor–acceptor cyclopropanes.

image file: d0cc01771j-p3.tif

Dorota Gryko

Dorota Gryko obtained a PhD from the Institute of Organic Chemistry at the Polish Academy of Sciences in 1997, under the supervision of Prof. J. Jurczak. After a post-doctoral stay with Prof. J. Lindsey in North Carolina State University (1998–2000), she started an independent career in Poland. In 2009 and 2018, she received the prestigious TEAM grants from the Foundation for Polish Science. Her current research interests are focused on light-induced processes with particular attention being paid to porphyrinoid catalysis as well as vitamin B12 chemistry.


Introduction

The term ‘ring-strain’ was first introduced by Adolf von Baeyer in 1885.1 He postulated that compounds in which bond angles deviate from standard tetrahedral values are less stable because of an increased energy referred to as “strain”. These considerations proved correct for small three- and four-membered rings; however, subsequent studies showed that destabilization of molecules is in fact affected by additional factors: bond length distortion, bond angle distortion, torsional strain, nonbonded interactions and energy changes due to rehybridization.2 Von Baeyer's theory started a new era for the chemistry of strained molecules and the 20th century abounded in reports describing the synthesis, structure and properties of strained molecules. Despite the progress made, this field, however, still remains relatively underexplored.

One of the main characteristics of strained compounds is their enhanced reactivity stemming from an increased energy and destabilization. Although this relationship is not as straightforward as it might appear, generally, the greater the difference in strain energy between a substrate and a product, the greater the driving force for a reaction.2 This feature is particularly well expressed in bi- and tricyclic compounds possessing a bridging bond linking the opposite carbon atoms – as seen in [1.1.1]propellanes (TCP, 1, tricyclo[1.1.1.01,3]-pentanes), bicyclo[1.1.0]butanes (BCB, 2), 1-azabicyclo[1.1.0] butanes (ABB, 3) and bicyclo[2.1.0]pentanes (housanes, 4) (Scheme 1). Early reports on these compounds describe their synthesis and reactivity towards nucleophiles, electrophiles and radicals.3


image file: d0cc01771j-s1.tif
Scheme 1 Characterization and reactivity of strain-release compounds.

Recently, small ring strained molecules have recaptured synthetic chemists’ attention. The real renaissance of strain-release driven transformations was launched by Baran and co-workers, who in 2016 developed a method for amination of strained propellanes (1), BCBs (2), ABBs (3) and housanes (4).4 They also defined these compounds as ‘spring-loaded’ molecules. According to Sharpless, ‘spring-loaded’ reactions have a high thermodynamic driving force resulting from the strain-release.5 This term was attributed to, among others, nucleophilic substitution processes, particularly ring-opening of epoxides and aziridines. Baran's amination method and strategies developed thereafter may be classified similarly since these transformations are driven directly by the strain release. In the following years, the strain-release strategy was utilized for the synthesis of mono- and disubstituted bicyclo[1.1.1]pentanes 5, cyclobutanes 6, azetidines 7, and cyclopentanes 8. Such strategies are particularly desirable in the field of medicinal chemistry where expanding the toolbox for introducing atypical motifs in a straightforward manner during late-stage functionalization is highly prized.6

The main focus of this review is on reactions involving cleavage of the unique strained bridging bond in bi- and tricyclic molecules 1, 2, 3, and 4. We compare the reactivity of spring-loaded compounds in various types of transformations, emphasizing their ability to interact with reagents of different nature. Up-to-date achievements in the chemistry of propellanes (1), bicyclo[1.1.0]butanes (2) and 1-azabicyclo[1.1.0]butanes (3), though described, are scattered over excellent reviews.12,13 In addition, the large number of reports emerging every month enforces continuous updates.

Characterization of strained compounds

The reactivity of [1.1.1]propellanes (1), bicyclo[1.1.0]butanes (2), azabicyclo[1.1.0]butanes (3) and bicyclo[2.1.0]pentanes (4) is governed mainly by strain energies and the properties of the central strained bonds. Although their molecular and electronic structures are reviewed elsewhere,3a,c,d,14 here, we emphasize some essential properties for a better understanding of the described reactions.

[1.1.1]Propellane (1) is characterized by a particularly high strain energy between 98 and 113 kcal mol−1 and an unusual structure involving two bridgehead carbons in a distinct inverted tetrahedral geometry.7,8 Both of these features contribute to the high reactivity of this tricyclic hydrocarbon, which is focused around its central C1–C3 bond. The character of this bridging bond has long been under discussion and, finally, after a series of studies that could not discern between an ionic or a covalent character, it was classified as a charge-shift bond, similar to that observed in difluorine.15 Nonetheless, it still remains a subject of theoretical calculations. What is most important from an experimental point of view is that the reactivity of the σ bond linking bridgehead carbons of [1.1.1]propellane (1) often resembles that of a π bond.16 Disruption of this bond releases about one-third of the strain energy and allows C1 and C3 carbons to adopt the privileged non-inverted geometry, which is the main driving force for its transformations.

Bicyclo[1.1.0]butane (2) resembles [1.1.1]propellane (1) in many ways. First of all, it exhibits a high strain energy of 66.3 kcal mol−1 making it the most strained of all bicyclic molecules.8 Secondly, its C1 and C3 carbons linked by the bridging bond also possess inverted geometries.17 The origin of the increased strain in BCB (2) is still the subject of investigation; however, it is currently explained by destabilizing C1–C3 interactions18 or the angle deformations at these carbon atoms.19 The bridging bent σ bond is 96% formed of two p orbitals, which is responsible for its π-like reactivity.20 At the same time, the bridgehead C–H bonds of BCB (2) are strongly polarized and display mainly s-character, which results in increased acidity of these protons.21 As a consequence of the unusual properties of bridging and bridgehead bonds, BCB (2) exhibits all-round reactivity interacting with nucleophiles, electrophiles, transition metals and radicals.3d

1-Azabicyclo[1.1.0]butane (3) may be considered as an analogue of BCB (2).3c This approximation is based on the calculated bond lengths and bond angles for 2,2,3-triphenyl-1-azabicyclo[1.1.0]butane,22 which closely resemble that of similarly substituted BCB (2).10 The predicted strain energy is very close to that of BCB (2); however, the exact value needs to be determined. Also similarly, the high angle strain results in increased π-character of the central C3–N σ-bond. In fact, in the case of ABB (3), it is considered to be a “pure p” bond formed via overlap of a non-hybridized N(2p) orbital with that of C3(2p). The same factors lie behind the increased s-character of C3–H, which has acidity comparable to that of the terminal C–H bond of acetylene.3c

Although bicyclo[2.1.0]pentane (4) has the lowest strain energy compared to the other described spring-loaded reagents (54.7 kcal mol−1), the disruption of its central C1–C4 bond releases the most energy (nearly 50 kcal mol−1).8 The reason for this lies in the structure of housane (4) – its dihedral angle is much smaller than that of, for example, bicyclo[1.1.0]butane (2) (113° versus 121.7°),9,11 resulting in a shortening of the C1–C4 distance and localization of the strain in the bridge bond rather than throughout the whole molecule.23 As in other described molecules, the central bridging bond is formed mainly of p orbitals, which has a significant impact on its reactivity.23

Nucleophilic addition

The unique properties of the central strained bonds make them reactive towards both nucleophilic and electrophilic reagents though nucleophilic additions dominate the field. At the time of primary interest in strained polycyclic molecules, during which [1.1.1]propellane (1) and bicyclo[1.1.0]butane (2) gained particular attention, a series of nucleophilic ring-opening reactions was developed (Scheme 2).
image file: d0cc01771j-s2.tif
Scheme 2 The early era of nucleophilic additions to strain-release reagents (dppf – 1,1′-bis(diphenylphosphino)ferrocene).

The similarities between the strained bonds of TCP (1) and BCB (2) suggest that they may undergo the same reactions, but the very low boiling temperature of BCB (2) (8 °C) makes it impractical as a substrate.24 As a consequence, BCBs carrying an electron withdrawing group at one end of the bridging bond are often used as substrates in nucleophilic additions instead. The resulting bond polarization enhances its susceptibility to nucleophilic attack in comparison to the central bond of [1.1.1]propellane (1). The pioneering nucleophilic ring-opening reactions of BCBs (2) concerned mainly the addition of nitrogen and oxygen nucleophiles (Scheme 2A).25,26 Gaoni also investigated the addition of organocopper reagents across the central bond of 1-arylsulfonyl bicyclobutanes (13) as well as their reduction to cyclobutane with LiAlH4 (Scheme 2B).27,28 Similar reactions with organometallic reagents were also reported for [1.1.1]propellane (1). The addition of aryl Grignard reagents to TCP (1) is followed by trapping the intermediates with CO229 or subsequent cross-coupling reactions with aryl halides giving disubstituted compound 16 (Scheme 2C).30 Even though many of these reactions were ground-breaking in the field, they often suffered from low yields or low selectivities. In recent years, however, numerous novel or refreshed transformations inspired by these works have been developed.

Reactions with organometallic compounds

In 2013, Fox and co-workers reported the enantioselective synthesis of bicyclobutanes followed by homoconjugate addition (Scheme 3).31 (E)-2-Diazo-5-arylpent-4-enoates 17 treated with Rh2(S-NTTL)4 (tetrakis[N-phthaloyl-(S)-tert-leucinato]dirhodium bis(ethyl acetate) adduct) provide a range of enantiomerically enriched bicyclobutanes 18 bearing the tBu ester moiety at the end of the central strained bond. Subsequent addition of aryl or alkyl Grignard reagents in the presence of a copper(I) catalyst leads to the formation of intermediate enolates. Quenching the reaction with an acid or with electrophiles provides cyclobutanes 19 with quaternary stereocenters. This one-pot methodology gives access to products in very good yields and modest to high diastereoselectivities.
image file: d0cc01771j-s3.tif
Scheme 3 Enantioselective synthesis of bicyclobutanes and their reactions with RMgX (aafter additional epimerization with tBuOK; S-NTTL = tetrakis[N-(1,8-naphthaloyl)-(S)-tert-leucinate]).

A similar concept may be applied to perform alkylation of 1-azabicyclo[1.1.0]butanes (3).32 ABB (3), generated in situ, undergoes addition of alkylmagnesium chlorides across the central bond. Similarly as in the Fox's report, trapping the resulting organometallic azetidine intermediate with an electrophile yields 1,3-difunctionalized products 21 (Scheme 4). In this case, the addition of a copper catalyst is essential to obtain satisfactory yields of alkylated products 21. The optimized conditions enable formation of primary, secondary and tertiary alkyl, vinyl, allyl and benzyl-substituted azetidines 21. Even substituted spirocycle 21f, showing an interesting perspective for the construction of similar, potentially bioactive compounds, can be obtained. Notably, N-Boc-aziridines react in a similar manner to yield N-protected amines (e.g.21g), highlighting the utility of the developed methodology for reactions of strained cyclic amines in general.


image file: d0cc01771j-s4.tif
Scheme 4 Alkylation of 1-azabicyclo[1.1.0]butanes with Grignard reagents – selected examples (E+ – electrophile; a[thin space (1/6-em)]With N-Boc aziridine and BF3·OEt2).

The reaction of organometallic reagents with the strained bond of [1.1.1]propellane (1) was also recently revisited by Knochel and co-workers, who developed an elegant route to 1,3-heteroaryl substituted BCPs (bicyclo[1.1.1]pentanes) via cleavage of the central C–C bond upon treatment with aryl-magnesium halides and subsequent Negishi coupling with aryl and heteroaryl halides.33 Previous reports by Szeimies30 and de Meijere,34 on ring-opening reactions of compound 1 with organomagnesium species, although ground-breaking, suffered from limitations such as a narrow reaction scope and long reaction times (up to 7 days). Knochel's conditions allow for a more efficient process leading to various Grignard reagents 22 in the strain-release ring opening event. Quenching the reaction with ethyl chloroformate affords a range of ethyl esters 23 (Scheme 5A). Moreover, transmetallation of reagent 22a with zinc chloride provides suitable partners for Negishi cross-coupling with aryl and heteroaryl halides leading to bis-arylated bicyclo[1.1.1]pentanes 24, some of which were previously unavailable (Scheme 5B). Importantly, these studies led Knochel's group to consider BCPs as bioisosters of an internal alkynyl group.


image file: d0cc01771j-s5.tif
Scheme 5 Formation of propellane derived Grignard reagent 22 in a strain-release event with subsequent carboxylation or arylation.

BCP–Grignard intermediates 22 are also reactive towards boronic esters 25 (Scheme 6). This reaction leads to BCP boronate complexes 26, suitable substrates for C–C bond forming reactions proceeding via a 1,2-metallate rearrangement.35 Aryl, alkenyl and selected alkyl BCB intermediates 22 undergo Zweifel olefination with alkenylboronic esters giving alkenyl BCP products 27 with generally high yields and excellent stereoselectivities (Scheme 6A).


image file: d0cc01771j-s6.tif
Scheme 6 Selected examples for 1,3-difunctionalizations of [1.1.1]propellane (1) via 1,2-metallate rearrangements of boronate complexes.

The same transformation may be performed using BCP–lithium intermediates. Alkenyl BCP–boronate complexes 26 are also prone to functionalization by cationic or radical electrophiles, as shown by the reaction of complex 26a (Scheme 6B).

Other organometallic reagents such as Hauser bases also react with propellane 1 and may be used for direct synthesis of BCP amines. The need for effective preparation of these compounds arose when the bicyclo[1.1.1]pentyl scaffold was recognized as the bioiostere of the phenyl ring.36 Although from the time of Wiberg's classic BCP–amine synthesis via bicyclo[1.1.1]pentan-1-carboxylic acid as an intermediate,37 some significant advances have been made, particularly by Bunker's group,38 the available methodologies still suffered from some drawbacks limiting their industrial application. For this purpose, Baran's group developed a method utilizing ‘turbo amides’39 – amido magnesium chlorides with LiCl enhancing their reactivity.4,40 Due to their strong nucleophilic character, they readily add across the strained σ bond of [1.1.1]propellane (1), giving access to a wide range of tertiary amines 32 bearing the bicyclo[1.1.1]pentane motif (Scheme 7A). The evident involvement of strain-release as a driving force for this transformation suggested that other valuable small ring precursors should react in a similar manner. Indeed, 1-azabicyclo[1.1.0]butane (3) generated in situ undergoes the reaction with turbo amide 30. The subsequent addition of an electrophile affords stable azetidines 34 with the amino group installed (Scheme 7B).


image file: d0cc01771j-s7.tif
Scheme 7 Selected examples for the amination of strain-release reagents with turbo amides 30.

Baran's methodology greatly improved the synthesis of small-ring substituted amines but it enabled incorporation of the BCP unit only as a terminal group. Gleason and co-workers noticed that addition of turbo amide 30 to TCP (1) affords the metallated intermediate 35, which can be trapped by an electrophile (Scheme 8A).41 Indeed, the reaction of in situ derived Grignard intermediates 35 with alkyl halides gives designed 1,3-disubstituted bicyclo[1.1.1]pentylamines 36. The initially poor efficiency of this transformation was significantly improved by the addition of catalytic amounts of CuI, which catalyzes coupling of alkyl Grignards with alkyl halides. A wide range of 3-alkylbicyclo[1.1.1]pentan-1-amines 36 were obtained in moderate to high yields. The reaction proves suitable for primary, secondary, as well as allyl and benzyl halides and enables the synthesis of pharmaceutically relevant compound 38 in fewer steps and better yields compared to the existing method (Scheme 8B).


image file: d0cc01771j-s8.tif
Scheme 8 Aminoalkylation of [1.1.1]propellane (1).

Reactions with other nucleophiles

Introduction of an electron withdrawing group at the C–C central strained bond adds to its electrophilic character, enabling addition of amines as nucleophilic reagents. Along this line, Gaoni used benzylamine to obtain substituted 1-phenyl or 1-p-tolylbicyclo[1.1.0]butanes.25 Although the obtained yields were high, the reaction was limited to several examples and required high temperature and the use of an amine as a solvent. Building on the synthesis of bicyclo[1.1.1]pentan-1-amines 32, Baran's group expanded their methodology to other pharmaceutically relevant small ring hydrocarbons. 1-Arylsulfonyl bicyclobutanes 13 were recognized as suitable substrates for the strain-release amination yielding aminocyclobutanes 39 (Scheme 9).4,40 Tuning electronic properties of BCB by introducing an appropriate arylsulfonyl group facilitates their reaction with primary and secondary amines as well as anilines. In a similar manner, BCB 13 reacts with thiols in the presence of a base (K2CO3), as shown by peptide labelling studies.
image file: d0cc01771j-s9.tif
Scheme 9 Nucleophilic addition of amines to arylsulfonylbicyclo[1.1.0]butanes 13.

Nucleophiles also react with electrophilic enantiopure housane derivatives 41 leading to 1,3-substituted cyclopentanes 42 in a stereoselective manner (Scheme 10).40 This protocol, usually requiring only the addition of a nucleophile at an elevated temperature, proves suitable for a variety of substrates such as amines, amides, heterocycles, carboxylic acids, thiols and selenols. Alcohols undergo this transformation as well, provided that an appropriate base is added (e.g. NaN(SiMe3)2).


image file: d0cc01771j-s10.tif
Scheme 10 Stereospecific strain-release cyclopentylation (major diastereomers not determined).

These excellent results have opened doors for modification of biologically important molecules. For instance, the synthesis of N-terminally functionalized peptide from sulfonylated bicyclo[2.1.0]pentane 41 and polypeptide on a solid support confirms the suitability of this strategy for the construction of peptides bearing non-native structural motifs. The newly developed cyclopentylation methodology also simplified access to eight different cyclopentane-based targets, improving stereoselectivities or overall yields of the synthetic routes.

Bicyclo[1.1.0]butanes substituted with electron-withdrawing groups also react with other heteroatom nucleophiles. Recently, Milligan et al. developed hydrophosphination of propellane (1) (Scheme 11).42 In this reaction, the spring-loaded BCB nitrile 43 reacts with secondary phosphine borane anions 44 furnishing cyclobutylphosphine borane products 45. Notably, alkyl and aryl 3-substituted bicyclobutane nitriles are well tolerated in this transformation.


image file: d0cc01771j-s11.tif
Scheme 11 Hydrophosphination of bicyclo[1.1.0]butane 43.

Another advance in the strain-release chemistry of spring-loaded molecules was made in 2018 by Walsh's group, who reported the synthesis of BCP benzyl amines.43 For this purpose, they harnessed N-benzyl ketimines 46, which after deprotonation with LiN(SiMe3)2 give ‘super’ electron donating 2-azaallyl anions that react with [1.1.1]propellane 1 furnishing diaryl methanamine analogues 47 (Scheme 12). Both electron-withdrawing and electron-donating aryl substituted ketimines 46 undergo the process. These with an EWG group on the aryl ring react in less than 1 h, while EDG-substituted derivatives need a longer time. The exact mechanism of this transformation is not immediately clear. Based on previous studies, three pathways, of which two involve single electron transfer from the 2-azaallyl anion to give a reactive radical, were proposed. The third involves direct nucleophilic addition of reagent 46 to propellane (1); however, none of these pathways have been definitively confirmed.


image file: d0cc01771j-s12.tif
Scheme 12 Addition of N-benzylketimines to [1.1.1]propellane (1).

Later, a similar concept was applied to 2-aryl-1,3-dithianes 48, which, as pronucleophiles, form anionic species under the action of strong bases such as NaN(SiMe3)2 (Scheme 13).44 They were recognized as promising reagents for introducing an aryl ketone moiety into the bicyclo[1.1.1]pentane scaffold. While previous methods for the construction of such derivatives were characterized by an extremely narrow scope (only two examples of Wiberg's nitrile addition to 1-lithio-BCPs) or formation of oligomeric side products (Wiberg's acyl radical addition),3f the newly developed methodology allows for introduction of the aryl-dithiane moiety into propellane (1), generating products 49 with both electron-donating and electron-withdrawing groups on the aryl ring in good or excellent yields.


image file: d0cc01771j-s13.tif
Scheme 13 Addition of 2-aryl-1,3-dithianes 48 to the bridge bond in [1.1.1]propellane (1) (SDS – sodium dodecylsulfate, NIS – N-iodosuccinimide).

Notably, a range of heterocycles are well tolerated. The obtained products can be deprotected under standard Barik's conditions,45 providing the corresponding aryl-bicyclopentane ketones 50. Alternatively, they can also undergo difluorination giving BCP aryl difluoromethanes 51 in good yields.

The proposed two electron mechanism was supported by computational studies as well as experimental results, including the expected low yield of product 49 in the absence of a proton source (Scheme 14).


image file: d0cc01771j-s14.tif
Scheme 14 Two electron mechanism for the addition of 2-aryl-1,3-dithianes (48) to [1.1.1]propellane (1).

Recently, our group proposed a new approach for the strain-release driven transformations relying on polarity reversal of the spring loaded bicyclo[1.1.0]butanes 2 and bicyclo[2.1.0]pentanes 4.46 Our strategy enables generation of a radical localized at C3 (BCB) or C4 (housane) atoms, exposing them to reactions with electrophiles whilst normally they are vulnerable to nucleophilic addition (Scheme 15).


image file: d0cc01771j-s15.tif
Scheme 15 Proposed reaction mechanism for the generation of the cyclobutyl radical.

Based on our previous experience with vitamin B12, we proposed this unique cobalt catalyst for this purpose. In this reaction, B12's ‘supernucleophilic’ Co(I) form undergoes conjugate addition to the bridge bond of compound 54, leading to strain-release driven ring-opening and, after protonation, generating the Co(III)-alkyl complex 56. The newly formed Co–C bond is then homolytically cleaved under visible-light irradiation giving cyclobutyl radical 57. This intermediate proved a perfect partner for various transformations including reactions with SOMOphiles (Scheme 16) or cross-coupling with electrophiles via transition metal catalysis (Scheme 17). With the use of heptamethyl cobyrinate (60) – a hydrophobic derivative of vitamin B12, the generated crucial intermediate 56 was subsequently harnessed for Giese-type addition to electron deficient alkenes. A wide range of BCBs 54 bearing sulfone, nitrile, ester and amide groups are suitable partners for this reaction. Preliminary studies also showed that the transformation may be successfully applied for the synthesis of bicyclo[2.1.0]pentane derivatives of type 62b from respective housanes.


image file: d0cc01771j-s16.tif
Scheme 16 Reaction of strained molecule-derived radicals with electron deficient alkenes.

image file: d0cc01771j-s17.tif
Scheme 17 Cross-coupling reaction of strained molecule-derived radicals with aryl iodides (DME – 1,2-dimethoxyethane).

The cobalt-based methodology for the generation of cyclobutyl radicals out of spring-loaded compounds combined with nickel catalysis for cross-coupling of strained molecules with aryl iodides 63 gives access to arylcyclobutanes 64 (Scheme 17). A range of BCBs with different electron-withdrawing groups attached to the bridge bond furnished arylated products 64. Representative examples also show applicability of this strategy for the generation of arylated cyclopentanes from bicyclo[2.1.0]pentanes (2). Moreover, employment of an appropriate ligand for the nickel catalyst enables formation of heteroarylated cyclobutyl product 64c. Importantly, the high chemoselectivity of the cobalt-catalyzed transformation allows this strategy to be applied for late-stage functionalizations.

Electrophilic reactions

All described strained molecules undergo electrophilic reactions; however, the reports published in recent years concern only 1-azabicyclo[1.1.0]butanes (3) and propellanes (1). ABB (3) is usually modified across the strained C3–N bond, as a result of a substitution at the N atom and subsequent ring-opening leading to a carbocation 65 at the C3 position (Scheme 18). Typically, various acids or azides are used for these functionalizations.3c
image file: d0cc01771j-s18.tif
Scheme 18 Addition of electrophiles to the bridge bonds.

Lopchuk and co-workers followed this strategy to improve the synthesis of protected 3-haloazetidines.47 1-Azabicyclo[1.1.0]butane (3), generated in situ, was subjected to the simultaneous addition of both nucleophile and electrophile, furnishing 1,3-disubstituted compounds 67 and 68 (Scheme 19).


image file: d0cc01771j-s19.tif
Scheme 19 Strain-release driven synthesis of 3-haloazetidines.

ABB (3) is assumed to react with an electrophile in the first step, generating the N-protected carbocation 65, which is finally trapped by iodide or bromide. The obtained azetidine derivatives can be subsequently diversified to include such functionalities as cyano, carboxyl, hydroxyl and ketone groups.

Reactions involving carbenes

Recently, spring-loaded compounds were recognized for the second time in their history as metal carbene precursors in catalytic cyclopropanation reactions. In previous years, the focus was on bicyclo[1.1.0]butanes 2, which formed metal carbenes under rhodium or nickel catalysis.14 Although strain release is the main driving force for these reactions, the use of [1.1.1]propellane (1) as a substrate was underexplored, due to the isomerization and oligomerization side reactions (Scheme 20). The catalysts investigated included Rh(I), Pd(II), Ag(I), Ir(I), Pt(II) or Pt(0), but none of them enabled effective formation of a cyclopropanation product from compound 1 and an alkene.3f
image file: d0cc01771j-s20.tif
Scheme 20 Transition-metal catalyzed ring-opening of [1.1.1]propellane.

Tolnai's group found that intermediates 72 or 73 derived from propellane (1) in the presence of a copper catalyst do not undergo cyclopropanation but form exocyclic allenic cyclobutanes 74 (Scheme 21).48 In this reaction, the central strained bond of substrate 1 is presumably attacked either by the copper(I) species or by a copper(I)-alkyne intermediate. Nevertheless, both pathways involve the ring-opening of propellane 1 with an electrophile and formation of cyclobutylcarbene. Subsequent rearrangement leads to allene 74. Notably, when 1H-2-silylacetylenes are used as substrates, alkynylcyclobutanes are formed.


image file: d0cc01771j-s21.tif
Scheme 21 The synthesis of exocyclic allenic cyclobutanes through strain release reaction with electrophilic copper species.

A breakthrough in the cyclopropanation reaction of [1.1.1]propellane (1) came when Bedford and Aggarwal applied nickel catalysis for this purpose.49 The Ni–NHC system allows for generation of TCP-derived metal-carbene 76 and its subsequent reaction with various alkenes 75 giving a range of methylenespiro[2.3]hexane derivatives 77 (Scheme 22). Importantly, alkenes undergoing this transformation need to possess an additional aryl or alkene functionality that can coordinate to the Ni catalyst. In order to overcome this limitation, alkenylboronic esters can be utilized as substrates for the cyclopropanation. The mechanism of the reaction was thoroughly investigated, revealing a two-electron pathway to be operative. Both mechanistic experiments and DFT calculations confirmed the generation of metal-carbene 76via concerted ring-opening of a nickel–[1.1.1]propellane complex involving double C–C bond cleavage.


image file: d0cc01771j-s22.tif
Scheme 22 Ni-Catalyzed strain-release cyclopropanation of [1.1.1]propellane (1) (base – LiOMe or NaOtBu; cod – 1,5-cyclooctadiene; SIMes – 1,3-bis(2,4,6-trimethylphenyl)-4,5-dihydroimidazol-2-ylidene).

Radical additions

In general, saturated strained hydrocarbons, including cyclobutanes or cyclopentanes, are not prone to radical addition, although they may undergo hydrogen atom abstraction.14 The situation is a little different for the unique group of compounds: TCP (1), BCB (2), ABB (3) and housane (4) containing a bridging spring-loaded σ bond. The π-like reactivity of this bond makes it susceptible to the addition of radical species giving a stable radical characterized by unique reactivity.3e

Among the spring-loaded compounds described in this review, [1.1.1]propellane (1) has been the most exhaustively investigated with regard to radical addition reactions. The presence of a strained bond makes it reactive towards both nucleophilic and electrophilic radicals. The energy barrier for ring opening of the resulting bicyclo[1.1.1]pent-1-yl radical is as high as ∼26 kcal mol−1, which prevents rearrangements and increases its stability.3a Due to this exceptional reactivity, propellane (1) successfully undergoes radical reactions with iodine, tri- and tetrachloromethane, iodoethane, thiophenol and others.3b Until recently, radical strain-release synthesis of 1- or 1,3-substituted bicyclo[1.1.1]pentanes was challenging, time consuming and costly, because of oligomerization of BCP radicals.

On the other hand, the reactivity of bicyclo[1.1.0]butanes (2) is governed by the character of a substituent present at the end of the central C1–C3 bond. BCBs (2) carrying electron withdrawing groups undergo ring-opening with nucleophilic radicals, whilst those with electron-donating substituents are more prone to electrophilic radical additions. The same reactivity pattern applies to substituted housanes (4). In contrast, the radical addition reactions for azabicyclo[1.1.0]butane (3) have not been reported to date.

In 2011, the Bunker team at Pfizer demonstrated the first radical addition-based synthesis of 1-monosubstituted bicyclo[1.1.1]pentane 79 from propellane 1 relying on transition metal hydrogen atom transfer (TM HAT) (Scheme 23).38


image file: d0cc01771j-s23.tif
Scheme 23 TM HAT strategy for functionalization of propellane (1).

This was built upon Carreira's methodology developed for unactivated olefins.50 According to the postulated reaction mechanism, Mn(dpm)3 reacts with phenylsilane to form a metal-hydride species.51,52 The hydrogen atom derived from this complex adds to the bridging bond of propellane (1), generating a bicyclo[1.1.1]pent-1-yl radical (78), which is trapped by di-tert-butyl azodicarboxylate to give compound 80. Deprotection followed by reductive cleavage of the N–N bond gives the desired 1-aminobicyclo[1.1.1]pentane (81) as a hydrochloride salt (Scheme 24).


image file: d0cc01771j-s24.tif
Scheme 24 Hydrohydrazination reaction of [1.1.1]propellane (dpm – dipivaloylmethane).

The TM HAT methodology is a general tool for generation of bicyclo[1.1.1]pent-1-yl radical 78 in a strain release event. Therefore, Co- or Mn-catalyzed addition of different radical acceptors to this reaction gives access to several 1-substituted bicyclopentanes 79 (Scheme 25).53


image file: d0cc01771j-s25.tif
Scheme 25 Hydrofunctionalization reactions of [1.1.1]propellane.

The bridging bond is also cleaved during radical thiol addition to [1.1.1]propellane (1) in a chain radical process.54 Although the reaction with thiophenol has been known since 1985,55 the addition of other thiols to propellane 1 has remained relatively unexplored. In a newly developed methodology, both aryl 82 and alkyl 84 thiols bearing functional groups (halo, hydroxyl, methoxyl, carboxyl, amino and nitro) may be employed (Scheme 26).


image file: d0cc01771j-s26.tif
Scheme 26 Insertion of thiols into the strained C–C bond.

Bräse's group also showed that in a similar manner, disulfides insert into the strained bond of TCP (1) under light irradiation.56 This method, however, suffers from the formation of n-staffanes as side products.

The ATRA-based (atom transfer radical addition) strategy can be utilized for functionalization of the strained σ bond of [1.1.1]propellane (1) with PhMe2Si-Bpin 86 (Scheme 27).57 The reaction proceeds without any additives at 0 °C, giving compound 87 in 93% yield, even on a gram scale (4.7 g, 80%). Acceleration of the reaction under UV light irradiation as well as its suppression in the presence of 9,10-dihydroanthracene as a radical inhibitor indicated that the silaboration proceeds through a radical chain mechanism.


image file: d0cc01771j-s27.tif
Scheme 27 Silaboration of [1.1.1]propellane (1).

Along the same line, Anderson and co-workers employed ATRA reaction for the cleavage of the central C–C bond in propellane 1.58 The reaction of TCP (1) with various organoiodides 88 in the presence of 10 mol% BEt3 as an initiator gives a wide range of 1,3-disubstituted BCPs 89 in moderate to excellent yields (Scheme 28). Similar methodologies providing such compounds have been developed before but they possess certain limitations.3f,59


image file: d0cc01771j-s28.tif
Scheme 28 Atom transfer radical addition of alkyl iodides 88 to [1.1.1]propellane (1).

Notably, the formation of n-staffanes, which limits the utility of existing protocols, was not observed. This side reaction was, however, the reason for diminished yields in 1-bromo-3-substituted BCP formation. This disparity was explained using DFT calculations, which demonstrated that the activation barrier for iodine abstraction from the substrate by an intermediate BCP radical is much lower than that of TCP capture. In contrast, the barrier for bromine atom abstraction is very close to the TCP (1) capture value, which confirms oligomerization as a competitive reaction pathway.

Although Anderson's methodology gave access to a wide range of 1-iodo-3-substituted-BCPs 89, it still remained limited in terms of certain functional groups such as amine and aldehyde and was incompatible with iodoarenes or heteroarenes. The photoredox-catalyzed ATRA reaction overcame these limitations.60 Several photocatalysts are capable of generating radicals from substrates for which triethylborane was ineffective. The most versatile, fac-Ir(ppy)3 (fac-tris[2-phenylpyridinato-C2,N]iridium(III)), under blue LED irradiation, facilitates the synthesis of a range of aryl and heteroaryl substituted iodo-BCPs 89 as well as benzyl and heterobenzyl derivatives (Scheme 29). Unactivated primary and secondary alkyl iodides give such valuable products as BCP-phenylalanine analogue 89j in an excellent 99% yield. Iodides and bromides carrying electron withdrawing groups at the α position were also taken into account in the scope of the reaction. Notably, alkyl bromides, which performed moderately in the previous report, furnish products in significantly improved yields. As expected, mild photocatalytic conditions are tolerated by sensitive amino and aldehyde functional groups.


image file: d0cc01771j-s29.tif
Scheme 29 1,3-Diffunctionalization of the bridge bond via photocatalyzed ATRA reaction of alkyl iodides 88 with [1.1.1]propellane (1).

In the proposed reaction mechanism, single electron transfer from an excited photocatalyst to a halide substrate gives radical 90, which subsequently adds to the strained bond of propellane (1) furnishing the bicyclopentyl radical 91 (Scheme 30). Two pathways are possible from here to finally generate the desired product: intermediate 91 can either undergo propagation to form the product 89 and regenerate radical 90 or it can be oxidatively quenched by the Ir(IV) complex to bicyclopentyl cation 92. Carbocation 92 can undergo further rearrangements giving the cyclobutyl side product 94 observed in small amounts in the reaction with 2-iodopyridine. Quantum yield measurements suggested that α-EWG iodides undergo the radical propagation pathway, while benzyl and heterocyclic iodides require the active action of a photocatalyst.


image file: d0cc01771j-s30.tif
Scheme 30 Plausible mechanism for the photocatalyzed ATRA reaction.

1,3-Disubstitution of propellane (1) is also accessible via a strategy developed by Kanazawa and Uchiyama et al. Based on DFT calculations, the proposed route to difunctionalized BCP derivatives 97 involves three key steps: generation of a radical, its addition to the strained bond, and trapping of the formed intermediate 95 (Scheme 31).61


image file: d0cc01771j-s31.tif
Scheme 31 Strategy for 1,3-difunctionalization of [1.1.1]propellane (1).

Along this line, the difunctionalized BCP derivative 99 was prepared by means of a radical multicomponent carboamination of [1.1.1]propellane (1) (Scheme 32). The methoxycarbonyl radical 100 is generated by hydrogen abstraction and oxidative denitrogenation of methyl carbazate 98 with tert-butylhydroperoxide (TBHP) and iron(II) phthalocyanine (Fe(Pc)). Its addition to propellane (1) leads to strain-release cleavage of the central C–C bond giving BCP-radical intermediate 102. Similarly to Bunker's work,38 di-tert-butyl azodicarboxylate acts as a radical acceptor to give amidyl radical 103, which after hydrogen abstraction from compound 104 furnishes the 1,3-disubstituted BCP product 99. Importantly, DFT calculations confirm that the C–N bond formation leading to the exceptionally stable radical 103 is kinetically preferred over oligomerization involving propellane (1) as a radical acceptor. After deprotection and hydrogenation, product 99 gives methyl 3-aminobicyclo[1.1.1]pentane-1-carboxylate hydrochloride salt in excellent yield.


image file: d0cc01771j-s32.tif
Scheme 32 Plausible mechanism of carboamination of [1.1.1]propellane (1) (Pc – phtalocyanine, TBHP – tert-butyl hydroperoxide).

This methodology also enables the synthesis of 3-aryl BCP-amine precursors 106 (Scheme 33). Various arylhydrazines 105 with both electron-donating and electron-withdrawing groups are suitable starting materials for this transformation. Notably, the reaction is compatible with halogenated aryls, which usually undergo radical-mediated dehalogenation reactions. Moreover, it is also suitable for the introduction of various heterocyclic groups. Preliminary studies showed that the proposed methodology can also be used for the introduction of alkyl substituents at the C3 position (e.g.106d).


image file: d0cc01771j-s33.tif
Scheme 33 Radical multicomponent carboamination of [1.1.1]propellane (1).

An alternative approach has been recently developed by Sheikh's and Leonori's groups.62 In a strain release event, propellane (1) reacts with electrophilic amidyl radicals generating the bicyclopentyl radical and the following atom/group-transfer reaction affords 1,3-difunctionalized bicyclo[1.1.1]pentanes 108 (Scheme 34).


image file: d0cc01771j-s34.tif
Scheme 34 Radical multicomponent carboamination of [1.1.1]propellane (1) (dtbpy – 4,4′-di-tert-butyl-2,2′-bipyridine; NCS – N-chlorosuccinimide, Nphth – phtalimide).

MacMillan's group used metallaphotoredox catalysis for 1,3-difunctionalization of propellane (1) (Scheme 35).63 In their multicomponent reaction, the π-like central C–C bond is attacked by a radical generated from various precursors, particularly activated carboxylic acids, in a photocatalytic manner. The resultant BCP radical is captured by a copper catalyst and then coupled with various nucleophiles. A number of radical precursors and N-, P- and S-nucleophiles prove suitable for this three component coupling (Scheme 35). Notably, the protocol can be applied to the preparation of pharmaceutical analogs with a bioisosteric motif.


image file: d0cc01771j-s35.tif
Scheme 35 1,3-Difunctionalization of propellane (1) in a copper-mediated multicomponent reaction (BTMG – 2-tert-butyl-1,1,3,3-tetramethylguanidine; NR2 – 6-bromo-4-azaindole; THP – 4-tetrahydropyranyl).

Since the central strained bond of BCB exhibits reactivity similar to that of a double bond, bicyclobutyl–boronate complexes 114 may be considered as analogues of unsaturated boronate complexes. Therefore, they should undergo the radical addition at the bridging bond, as reported by Aggarwal and co-workers.64 The BCB–boronate complexes 114, generated in situ by the addition of t-BuLi to a mixture of sulfoxide (112) and boronic acid pinacol ester 113, react with the radical precursors 88 under blue LED irradiation (Scheme 36). A variety of boronate complexes undergo this reaction, including aromatic ones as well as those bearing an alkyl group, furnishing products 115 with high diastereoselectivities. Additionally, a range of radical precursors were investigated, giving products in high or moderate yields providing that they carried electron withdrawing groups. For nonactivated alkyl iodides 88, the use of sulfones as a traceless activating group turned out to be necessary. The proposed mechanism of this reaction is based on previous studies with vinyl boronate complexes (Scheme 37).65 The alkyl radical 116, generated from iodide 88 under blue LED irradiation, adds to the central bond of strained bicyclobutyl-boronate 114, furnishing the radical anion 117, which undergoes single electron transfer to iodide 88, resulting in the formation of zwitterionic intermediate 118. Subsequent 1,2-metallate rearrangement gives the final product 115. The single electron transfer step is postulated by the authors to influence the stereoselectivity of the reaction giving anti-substituted products.


image file: d0cc01771j-s36.tif
Scheme 36 Radical addition to the strained bonds of BCB–boronate complexes.

image file: d0cc01771j-s37.tif
Scheme 37 Plausible mechanism for the radical addition of organic iodides 88 to the BCB–boronate complexes.

On the other hand, electrophilic BCBs react with radicals of nucleophilic character. Ernouf, Cintrat and co-workers showed that carboxylic acid derived radicals react with electrophilic BCBs 54 in a similar manner (Scheme 38).66 Deprotonation of carboxylic acid (119a) by cesium carbonate and its oxidation by single electron transfer from the excited photocatalyst gives the carboxyl radical (122). Consequent CO2-extrusion generates the nucleophilic radical (123), which inserts into the central bond of strained bicyclobutane 54, leading to the formation of α-EWG radical 124via ring opening. Reduction of this intermediate and protonation of the corresponding anion 125 furnishes the desired 1,3-disubstituted cyclobutane 120. The reaction is sensitive to the character of these radical attacking the central strained bond – only those with highly nucleophilic character efficiently furnish the desired products (Scheme 39).


image file: d0cc01771j-s38.tif
Scheme 38 Plausible mechanism for the cyclobutylation of carboxylic acid-derived radicals 123.

image file: d0cc01771j-s39.tif
Scheme 39 Photocatalyzed radical cleavage of the strained bond with carboxylic acid-derived radicals. a[thin space (1/6-em)]Major diastereomers were not determined, except for the compound 120a (major trans isomer depicted above).

Despite this drawback, a range of α-amino and α-oxy carboxylic acids including natural N-Boc protected amino acids, peptides and drugs afford functionalized cyclobutanes 120 in moderate to very good yields (up to 98%).

Along this line, Jui and co-workers investigated N-aryl amines as a source of radicals (Scheme 40).67 Photochemically generated α-amino radicals add to the central C–C bonds, furnishing 1,3-disubstituted cyclobutanes 127 in decent yields.


image file: d0cc01771j-s40.tif
Scheme 40 α-C–H cyclobutylation of aniline derivatives. a[thin space (1/6-em)]Major diastereomers were not determined, except for compound 127f (major cis isomer depicted).

Alternatively, nucleophilic radicals may also be generated from alkyl chlorides with Ti catalyst.68 The reaction of bicyclo[1.1.0]butane derivatives 129 and 13a with alkyl chloride 128 in the presence of CpTiCl3 as a catalyst, Zn as a reductant and Et3N·HCl as a proton source gives the corresponding cyclobutanes 130 and 131 in moderate to excellent yields (Scheme 41).


image file: d0cc01771j-s41.tif
Scheme 41 Radical alkylation of electrophilic BCB 129 and 13a (major diastereomers not assigned).

Rearrangements

The reactivity of strained bicycles is tuned by the substituents attached to the ends of the bridging C–C bond. For instance, in the case of bicyclo[1.1.0]butanes (2), installation of the arylsulfonyl group on the C1 carbon promotes nucleophilic additions at the C3 position. Conversely, analogous introduction of the pinacol boronic ester fosters ring-opening reactions of BCB (2) with electrophiles such as palladium complexes. The first example of such reaction came from the Aggarwal group, who demonstrated that formation of the bicyclo[1.10]butyl boronate complex (114) enables coordination of an electrophilic palladium(II)–aryl complex at the C3 position of the BCB ring.69 Subsequent transformations result in the formation of highly functionalized 1,1,3-trisubstituted cyclobutanes 133 (Scheme 42).
image file: d0cc01771j-s42.tif
Scheme 42 Palladium-catalyzed strain-release driven functionalization of cyclobutanes (dba – dibenzylideneacetone); dippf – 1,1′-bis(diisopropylphosphino)ferrocene.

In the proposed mechanism, BCB boronate complex 114 is generated in situ as a result of two transformations: first, bicyclo[1.1.0]butyl sulfoxide (112) undergoes sulfoxide-lithium exchange, giving 1-lithio bicyclo[1.1.0]butane, which is subsequently trapped by a pinacol boronic ester (Scheme 43). Subsequent strain release upon reaction with the palladium(II)–aryl complex pre-formed from aryl triflate 132, bis(dibenzylideneacetone)palladium(0) (Pd(dba)2) and a ferrocene based phosphine ligand (dippf) leads to highly functionalized 1,1,3-trisubstituted cyclobutane 133.


image file: d0cc01771j-s43.tif
Scheme 43 Plausible mechanism of carbopalladation of BCB–boronate complexes.

This reaction works well for primary, secondary, tertiary, aryl and heteroaryl boronic esters as well as for a wide range of aryl and vinyl triflates giving products in good to excellent yields. Perfect diastereoselectivity in this reaction is clearly explained by its mechanism. 1,2-Migration of the boronic ester substituent to the α-carbon is possible provided it has an anti-periplanar alignment with the central C–C bond of BCB 114. In this orientation, the approach of the palladium(II)–aryl complex from the endo face is blocked by the pinacol group. The cationic palladium species interacts with the significant orbital density projected by the BCB, favouring its coordination on the exo face. Formation of the new C–Pd bond is simultaneous with the 1,2-migration of the R1 group and the strain-release cleavage of the central bond. Finally, reductive elimination gives the cross-coupling product 133 and regenerates the palladium(0) species. The importance of exceptional strain as a driving force in this transformation was proven by the inertness of the cyclopropyl boronate complex (strain energy lower by >30 kcal mol−1)8 under the same reaction conditions.

A similar concept was used for functionalization of strained azetidines (Scheme 44).70 Preparation of essential azabicyclo[1.1.0]butyl lithium (134) may be accomplished in two ways: either by direct treatment of generated in situ azabicyclo[1.1.0]butane with sec-butyl lithium or by conversion of azabicyclo[1.1.0]butyl sulfoxide (135). Highly nucleophilic spring-loaded lithiated species 134 undergoes reaction with boronic esters giving 1-azabicyclo[1.1.0]butyl boronate complexes 136. N-Protonation with acetic acid is required to make the amine a better leaving group and, consequently, promotes strain-release driven 1,2-metallate rearrangement. Finally, the reaction with an electrophile gives N-protected functionalized azetidines 137. As with the previous bicyclo[1.1.0]butyl-based strategy, the reaction tolerates a wide range of primary, secondary, tertiary alkyl, vinyl and aryl boronic esters. Additionally, bis(boronic esters) undergo regioselective homologation giving azetidines 137 bearing two substituents prone to further functionalizations.


image file: d0cc01771j-s44.tif
Scheme 44 Strain-release rearrangement of ABB–boronate complexes (base – Et3N or N,N-diisopropylethylamine).

Miscellaneous

Recent developments in the generation of bench-stable strain-release substrates enabled revisiting several reactions for the construction of valuable small-ring containing molecules, many of which have pharmaceutical potential. Chiral sulfonyl housanes (138) applied by Baran's group for numerous stereospecific “cyclopentylation” reactions are also used for the formation of functionalized amino alcohols and pyridine-substituted cyclopentanes by means of formal homo [3+2] dipolar cycloaddition (Scheme 45).71 Nitrones 139 or pyridine oxides 142 add across the strained C1–C4 bond giving products 140 and 143, respectively.
image file: d0cc01771j-s45.tif
Scheme 45 Strain-release driven cycloaddition of nitrones (139) and pyridine N-oxides (142) with substituted housanes (138).

The Ma72 and Mykhailiuk73 groups recently progressed in the construction of “ortho/meta-substituted” bioisosters via carbene insertion into strained bonds of bicyclo[1.1.0]butanes 145 and 147. Previous attempts of carbene insertion into the strained BCB bond concerning dichloro- and dibromocarbenes were reported, but they suffered from low yields and multiple by-products formation.74 Taking into account the π-like reactivity of the central σ-bond of the bicyclo[1.1.0]butanes (2), these transformations may be compared to the cyclopropanation of the double bonds. 3-Arylbicyclo[1.1.0]butane-1-carboxylates 145 and 147 undergo reaction with difluorocarbenes generated with trimethylsilyl 2-fluorosulfonyl-2,2-difluoroacetate (TFDA) and sodium fluoride or with trifluoromethyltrimethylsilane (TMSCF3) and sodium iodide, with the latter ones giving slightly higher yields (Scheme 46). The mechanism of this reaction was not proved experimentally, however two possible pathways are postulated. The first scenario involves a stable benzylic carbocation formed as a result of the central BCB bond attack on the electrophilic difluorocarbene. Alternatively, the reaction may proceed via stepwise radical addition of the difluorocarbene to the diradical intermediate being in resonance with the bicyclo[1.1.0]butane.


image file: d0cc01771j-s46.tif
Scheme 46 Difluoro carbene insertion into the strained central bond of substituted BCBs 145 and 147.

Conclusions and outlook

The reported strain-release driven reactions complemented or even revolutionized the fields of nucleophilic, radical, electrophilic or transition metal-catalyzed transformations of strained bi- and tricyclic molecules. This strategy stands out with undeniable advantages desired in modern synthetic chemistry. It offers additional reaction driving force and avoids usage of leaving or directing groups, increasing the atom economy of the performed transformations. Moreover, it enables introduction of atypical three-dimensional structural motifs into complex molecules, giving access to bioisosteres highly prized by the pharmaceutical industry.

While [1.1.1]propellane (1) and bicyclo[1.1.0]butane (2) chemistry was extensively investigated, the reactivity of other spring-loaded compounds remains underexplored. Radical and transition-metal reactions of 1-azabicyclo[1.1.0]butanes (3) are currently unknown and in the case of bicyclo[2.1.0]pentanes (4), there are only a few specific examples. There is still room for improvement, particularly in the fields of electrophilic and transition-metal catalyzed reactions of all strained molecules. It is also anticipated that access to BCB (2)- and housane (4)-derived C-centered radicals will be an inspiration for other polarity-reversal transformations, which will certainly lead to the synthesis of unprecedented complex scaffolds.

Advances made in the strain-release driven transformations are impressive – current methodologies enable mono- and difunctionalizations of spring-loaded molecules by means of strain-release C–C and C–heteroatom bond forming reactions. The construction of Csp3–Csp3 and Csp3–Csp2 bonds are now feasible, but the strain-release Csp3–Csp bond formation remains to be discovered.

Needless to say, spring-loaded molecules are no longer only a scientific curiosity; they have proven useful as bioisosteres in medicinal chemistry. The potential of these molecules, built on their exceptional reactivity and structural properties, is far greater and still needs to be explored. Certainly, a number of exciting, cutting-edge discoveries will be revealed in the coming years, the scope of which will be only limited by our imagination.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Financial support for this work was provided by the Foundation for Polish Sciences (FNP TEAM POIR.04.04.00-00-4232/17-00).

Notes and references

  1. A. Von Baeyer, Ber. Dtsch. Chem. Ges., 1885, 18, 2269 CrossRef.
  2. K. B. Wiberg, Angew. Chem., Int. Ed. Engl., 1986, 25, 312 CrossRef.
  3. For the early reviews see: (a) M. D. Levin, P. Kaszyński and J. Michl, Chem. Rev., 2000, 100, 169–234 CrossRef CAS PubMed; (b) E. W. Delia and I. J. Lochert, Org. Prep. Proced. Int., 1996, 28, 411–441 CrossRef; (c) R. Bartnik and A. P. Marchand, Synlett, 1997, 1029–1039 CrossRef CAS; (d) S. Hoz, in The Chemistry of the Cyclopropyl Group, ed. Z. Rappoport, Wiley, New York, 1987, vol. 1, pp. 1121–1191 Search PubMed; (e) C. Jamieson, J. C. Walton and K. U. Ingold, J. Chem. Soc., Perkin Trans. 2, 1980, 1366–1371 RSC; (f) K. B. Wiberg and S. T. Waddell, J. Am. Chem. Soc., 1990, 112, 2194–2216 CrossRef CAS.
  4. R. Gianatassio, J. M. Lopchuk, J. Wang, C.-M. Pan, L. R. Malins, L. Prieto, T. A. Brandt, M. R. Collins, G. M. Gallego, N. W. Sach, J. E. Spangler, H. Zhu, J. Zhu and P. S. Baran, Science, 2016, 351, 241 CrossRef CAS PubMed.
  5. H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed., 2001, 40, 2004 CrossRef CAS PubMed.
  6. J. Boström, D. G. Brown, R. J. Young and G. M. Keserü, Nat. Rev. Drug Discovery, 2018, 17, 709 CrossRef PubMed.
  7. (a) V. Polo, J. Andres and B. Silvi, J. Comput. Chem., 2007, 28, 857 CrossRef CAS PubMed; (b) A. Ioffe, V. Svyatkin and O. Nefedov, Bull. Acad. Sci. USSR, Div. Chem. Sci., 1988, 37, 1827 CrossRef.
  8. (a) P. R. Khoury, J. D. Goddard and W. Tam, Tetrahedron, 2004, 60, 8103 CrossRef CAS; (b) S. A. Smith, K. E. Hand, M. L. Love, G. Hill and D. H. Magers, J. Comput. Chem., 2013, 34, 558 CrossRef CAS PubMed.
  9. P. Seiler, Helv. Chim. Acta, 1990, 73, 1574 CrossRef CAS.
  10. P. G. Gassman, M. L. Greenlee, D. A. Dixon, S. Richtsmeier and J. Z. Gougoutas, J. Am. Chem. Soc., 1983, 105, 5865 CrossRef CAS.
  11. R. D. Suenram and M. D. Harmony, J. Chem. Phys., 1972, 56, 3837 CrossRef CAS.
  12. (a) A. M. Dilmaç, E. Spuling, A. de Meijere and S. Bräse, Angew. Chem., Int. Ed., 2017, 56, 5684 CrossRef PubMed; (b) J. Kanazawa and M. Uchiyama, Synlett, 2019, 1 CrossRef CAS; (c) G. M. Locke, S. S. Bernhard and M. O. Senge, Chem. – Eur. J., 2019, 25, 4590 CrossRef CAS PubMed; (d) X. Ma and L. Nhat Pham, Asian J. Org. Chem., 2020, 9, 8 CrossRef CAS.
  13. A. Fawcett, Pure Appl. Chem., 2019 DOI:10.1515/pac-2019-1007.
  14. K. B. Wiberg, Strained Hydrocarbons: Structure, Stability, and Reactivity, React. Intermed. Chem., 2004, 717 CAS.
  15. W. Wu, J. Gu, J. Song, S. Shaik and P. C. Hiberty, Angew. Chem., Int. Ed., 2009, 48, 1407 CrossRef CAS PubMed.
  16. V. Galasso, Chem. Phys. Lett., 1994, 230, 387 CrossRef CAS.
  17. K. B. Wiberg, Acc. Chem. Res., 1984, 17, 379 CrossRef CAS.
  18. N. L. Bauld, J. Cessac and R. L. Holloway, J. Am. Chem. Soc., 1977, 99, 8140 CrossRef CAS.
  19. D. Baric and Z. B. Maksic, Theor. Chem. Acc., 2005, 114, 222 Search PubMed.
  20. (a) J. M. Schulman and G. J. Fisanick, J. Am. Chem. Soc., 1970, 92, 6653 CrossRef CAS; (b) M. D. Newton and J. M. Schulman, J. Am. Chem. Soc., 1972, 94, 767 CrossRef CAS.
  21. D. R. Whitman and J. F. Chiang, J. Am. Chem. Soc., 1972, 94, 1126 CrossRef CAS.
  22. R. Bartnik, Z. Cebulska and R. Faure, J. Chem. Soc., Chem. Commun., 1993, 2, 148 RSC.
  23. R. K. Bohn and Y. H. Tai, J. Am. Chem. Soc., 1970, 92, 6447 CrossRef CAS.
  24. G. M. Lampman and J. C. Aumiller, Org. Synth., 1971, 51, 55 CrossRef CAS.
  25. (a) Y. Gaoni, Tetrahedron Lett., 1988, 29, 1591 CrossRef CAS; (b) Y. Gaoni, Org. Prep. Proced. Int., 1995, 27, 185 CrossRef CAS.
  26. S. Hoz and D. Aurbach, Tetrahedron, 1979, 35, 881 CrossRef CAS.
  27. (a) Y. Gaoni, Tetrahedron Lett., 1982, 23, 5215–5218 CrossRef CAS; (b) Y. Gaoni, Tetrahedron, 1989, 45, 2819 CrossRef CAS; (c) Y. Gaoni, A. Tomažič and E. Potgieter, J. Org. Chem., 1985, 50, 2943 CrossRef CAS.
  28. Y. Gaoni and A. Tomažič, J. Org. Chem., 1985, 50, 2948 CrossRef CAS.
  29. K. B. Wiberg and N. McMurdie, J. Am. Chem. Soc., 1994, 116, 11990 CrossRef CAS.
  30. J. D. D. Rehm, B. Ziemer and G. Szeimies, Eur. J. Org. Chem., 1999, 2079 CrossRef CAS.
  31. R. Panish, S. R. Chintala, D. T. Boruta, Y. Fang, M. T. Taylor and J. M. Fox, J. Am. Chem. Soc., 2013, 135, 9283 CrossRef CAS PubMed.
  32. R. Gianatassio and D. Kadish, Org. Lett., 2019, 21, 2060 CrossRef CAS PubMed.
  33. I. S. Makarov, C. E. Brocklehurst, K. Karaghiosoff, G. Koch and P. Knochel, Angew. Chem., Int. Ed., 2017, 56, 12774 CrossRef CAS PubMed.
  34. M. Messner, S. I. Kozhushkov and A. de Meijere, Eur. J. Org. Chem., 2000, 1137 CrossRef CAS.
  35. S. Yu, C. Jing, A. Noble and V. K. Aggarwal, Angew. Chem., Int. Ed., 2020, 59, 3917 CrossRef CAS PubMed.
  36. R. Pellicciari, M. Raimondo, M. Marinozzi, B. Natalini, G. Costantino and C. Thomsen, J. Med. Chem., 1996, 39, 2874 CrossRef CAS PubMed.
  37. K. B. Wiberg and V. Z. Williams, J. Org. Chem., 1970, 35, 369 CrossRef CAS.
  38. K. D. Bunker, N. W. Sach, Q. Huang and P. F. Richardson, Org. Lett., 2011, 13, 4746 CrossRef CAS PubMed.
  39. A. Krasovskiy and P. Knochel, Angew. Chem., Int. Ed., 2004, 43, 3333 CrossRef CAS PubMed.
  40. J. M. Lopchuk, K. Fjelbye, Y. Kawamata, L. R. Malins, C.-M. Pan, R. Gianatassio, J. Wang, L. Prieto, J. Bradow, T. A. Brandt, M. R. Collins, J. Elleraas, J. Ewanicki, W. Farrell, O. O. Fadeyi, G. M. Gallego, J. J. Mousseau, R. Oliver, N. W. Sach, J. K. Smith, J. E. Spangler, H. Zhu, J. Zhu and P. S. Baran, J. Am. Chem. Soc., 2017, 139, 3209 CrossRef CAS PubMed.
  41. J. M. E. Hughes, D. A. Scarlata, A. C. Y. Chen, J. D. Burch and J. L. Gleason, Org. Lett., 2019, 21, 6800 CrossRef CAS PubMed.
  42. J. A. Milligan, C. A. Busacca, C. H. Senanayake and P. Wipf, Org. Lett., 2016, 18, 4300 CrossRef CAS PubMed.
  43. R. A. Shelp and P. J. Walsh, Angew. Chem., Int. Ed., 2018, 57, 15857 CrossRef CAS PubMed.
  44. N. Trongsiriwat, Y. Pu, Y. Nieves-Quinones, R. A. Shelp, M. C. Kozlowski and P. J. Walsh, Angew. Chem., Int. Ed., 2019, 58, 13416 CrossRef CAS PubMed.
  45. N. C. Ganguly and S. K. Barik, Synthesis, 2009, 1393 CrossRef CAS.
  46. M. Ociepa, A. J. Wierzba, J. Turkowska and D. Gryko, J. Am. Chem. Soc., 2020, 142, 5355 CrossRef CAS PubMed.
  47. Y. Ji, L. Wojtas and J. M. Lopchuk, ARKIVOC, 2018, 195 Search PubMed.
  48. D. Lasányi and G. L. Tolnai, Org. Lett., 2019, 21, 10057 CrossRef PubMed.
  49. S. Yu, A. Noble, R. B. Bedford and V. K. Aggarwal, J. Am. Chem. Soc., 2019, 141, 20325 CrossRef CAS PubMed.
  50. J. Waser and E. M. Carreira, Angew. Chem., Int. Ed., 2004, 43, 4099 CrossRef CAS PubMed.
  51. J. Waser, B. Gaspar, H. Nambu and E. M. Carreira, J. Am. Chem. Soc., 2006, 128, 11693 CrossRef CAS PubMed.
  52. S. W. Crossley, C. Obradors, R. M. Martinez and R. A. Shenvi, Chem. Rev., 2016, 116, 8912 CrossRef CAS PubMed.
  53. K. D. Bunker and K. J. Miller, WO2015089170A1, 2015.
  54. R. M. Bär, S. Kirschner, M. Nieger and S. Bräse, Chem. – Eur. J., 2018, 24, 1373 CrossRef PubMed.
  55. K. Semmler, G. Szeimies and J. Belzner, J. Am. Chem. Soc., 1985, 107, 6410 CrossRef CAS.
  56. R. M. Bär, G. Heinrich, M. Nieger, O. Fuhr and S. Bräse, Beilstein J. Org. Chem., 2019, 15, 1172 CrossRef PubMed.
  57. M. Kondo, J. Kanazawa, T. Ichikawa, T. Shimokawa, Y. Nagashima, K. Miyamoto and M. Uchiyama, Angew. Chem., Int. Ed., 2020, 59, 1970 CrossRef CAS PubMed.
  58. D. F. J. Caputo, C. Arroniz, A. B. Dürr, J. J. Mousseau, A. F. Stepan, S. J. Mansfield and E. A. Anderson, Chem. Sci., 2018, 9, 5295 RSC.
  59. P. Kaszynski, A. C. Friedli and J. Michl, J. Am. Chem. Soc., 1992, 114, 601 CrossRef CAS.
  60. J. Nugent, C. Arroniz, B. R. Shire, A. J. Sterling, H. D. Pickford, M. L. J. Wong, S. J. Mansfield, D. F. J. Caputo, B. Owen, J. J. Mousseau, F. Duarte and E. A. Anderson, ACS Catal., 2019, 9, 9568 CrossRef CAS.
  61. J. Kanazawa, K. Maeda and M. Uchiyama, J. Am. Chem. Soc., 2017, 139, 17791 CrossRef CAS PubMed.
  62. J. H. Kim, A. Ruffoni, Y. Al-Faiyz, N. S. Sheikh and D. Leonori, Angew. Chem., Int. Ed., 2020, 59, 2–9 CrossRef.
  63. X. Zhang, R. T. Smith, C. Le, S. J. McCarver, B. T. Shireman, N. I. Carruthers and D. W. MacMillan, Nature, 2020, 580, 220 CrossRef CAS PubMed.
  64. M. Silvi and V. K. Aggarwal, J. Am. Chem. Soc., 2019, 141, 9511 CrossRef CAS PubMed.
  65. M. Silvi, C. Sandford and V. K. Aggarwal, J. Am. Chem. Soc., 2017, 139, 5736 CrossRef CAS PubMed.
  66. G. Ernouf, E. Chirkin, L. Rhyman, P. Ramasami and J. C. Cintrat, Angew. Chem., Int. Ed., 2020, 59, 2618 CrossRef CAS PubMed.
  67. C. J. Pratt, R. A. Aycock, M. D. King and N. T. Jui, Synlett, 2020, 51 CAS.
  68. X. Wu, W. Hao, K. Y. Ye, B. Jiang, G. Pombar, Z. Song and S. Lin, J. Am. Chem. Soc., 2018, 140, 14836 CrossRef CAS PubMed.
  69. A. Fawcett, T. Biberger and V. K. Aggarwal, Nat. Chem., 2019, 11, 117 CrossRef CAS PubMed.
  70. A. Fawcett, A. Murtaza, C. H. Gregson and V. K. Aggarwal, J. Am. Chem. Soc., 2019, 141, 4573 CrossRef CAS PubMed.
  71. S. Clementson, A. Radaelli, K. Fjelbye, D. Tanner and M. Jessing, Org. Lett., 2019, 21, 4763 CrossRef CAS PubMed.
  72. X. Ma, D. L. Sloman, Y. Han and D. J. Bennett, Org. Lett., 2019, 21, 7199 CrossRef CAS PubMed.
  73. R. M. Bychek, V. Hutskalova, Y. P. Bas, O. A. Zaporozhets, S. Zozulya, V. V. Levterov and P. K. Mykhailiuk, J. Org. Chem., 2019, 84, 15106 CrossRef CAS PubMed.
  74. (a) H. K. Hall Jr, C. D. Smith, E. P. Blanchard Jr, S. C. Cherkofsky and J. B. Sieja, J. Am. Chem. Soc., 1971, 93, 121 CrossRef; (b) D. E. Applequist, T. L. Renken and J. W. Wheeler, J. Org. Chem., 1982, 47, 4985 CrossRef CAS.

This journal is © The Royal Society of Chemistry 2020