Catalytic C–H amination at its limits: challenges and solutions

Damien Hazelard , Pierre-Antoine Nocquet and Philippe Compain *
Laboratoire de Synthèse Organique et Molécules Bioactives (SYBIO), Université de Strasbourg/CNRS (UMR 7509), Ecole Européenne de Chimie, Polymères et Matériaux (ECPM), 25 rue Becquerel, 67087 Strasbourg Cedex 2, France. E-mail:

Received 3rd July 2017 , Accepted 4th September 2017

First published on 12th September 2017

Catalytic C–H amination reactions enable direct functionalization of non-activated C(sp3)–H bonds with high levels of regio-, chemo- and stereoselectivity. As a powerful tool to unlock the potential of inert C–H bonds, C–H amination chemistry has been applied to the preparation of synthetically challenging targets since major simplification of synthetic sequences are expected from this approach. Pushing C–H amination to its limits has led to a deeper understanding of the reaction mechanism and scope. In this review, we present a description of the specific challenges facing catalytic C–H amination in the synthesis of natural products and related compounds, as well as innovative tactics created to overcome them. By identifying and discussing the major insights gained and strategies designed, we hope that this review will stimulate further progress in C–H amination chemistry and beyond.

image file: c7qo00547d-p1.tif

Damien Hazelard

Damien Hazelard obtained his PhD in 2005 under the supervision of Dr Fadel (Paris-Sud University). In 2006, he performed a post-doctoral training in the field of organocatalysis in the group of Pr Hayashi at the Tokyo University of Science. Then he joined the group of Pr Colobert to work on total synthesis at the University of Strasbourg. He was appointed in 2010 as assistant professor at the same university in the group of Pr. Compain. His current research interests deal with the development of new synthetic methodologies for the synthesis of original nitrogen heterocycles related to glycomimetics.

image file: c7qo00547d-p2.tif

Pierre-Antoine Nocquet

Pierre-Antoine Nocquet was awarded his PhD in chemistry in 2013 at the University of Strasbourg under the supervision of Prof. Compain. He worked as a postdoctoral fellow successively under the supervision of Prof. Opatz (University of Mainz), Prof. Cossy (ESPCI Paris), and then Dr Carboni (University of Rennes 1). In 2017 he joined Dr Oudeyer's group (IRCOF, Rouen) as a postdoctoral fellow. His current interest is in the development of new methods for the synthesis of nitrogen heterocycles.

image file: c7qo00547d-p3.tif

Philippe Compain

Philippe Compain gained his Engineer degree at CPE Lyon. In 1998, he was awarded the Dina Surdin Prize from French Chemical Society for his PhD research on alkaloid synthesis (UCBL, Lyon). After a postdoctoral stay at Montreal with Pr Hanessian, he was appointed Chargé de Recherche at CNRS (ICOA, Orléans). In 2008, he accepted a full professorship at the University of Strasbourg. His research interests span from synthetic methodologies to glycomimetics of interest, from square sugars to multivalent sweet giants. Pr. Compain is a fellow of the Royal Society of Chemistry and a member of the Institut Universitaire de France.

1 Introduction

In less than two decades, catalytic amination of C(sp3)–H bonds has established itself as a powerful tool for the synthesis of nitrogen-containing natural products.1 Unlocking the potential of inert C–H bonds with high levels of regio-, chemo- and stereoselectivity is indeed a strategy of choice to achieve a major simplification of synthetic sequences.2,3 Based on these unprecedented chemical possibilities, synthetic chemists have challenged the C–H amination process with more and more demanding substrates (Fig. 1).4,5 The total synthesis of complex natural molecules such as (−)-tetrodotoxin, a densely functionalized polycyclic compound, has for example superbly demonstrated the synthetic power of catalysed C–H amination.6 Beyond apparent structural complexity which is intuitively related to molecular size or intricate atomic composition,7 a number of specific synthetic challenges have to be faced in catalytic C–H insertion. Because of the highly electrophilic nature of the metal nitrene generated, synthetic targets containing C[double bond, length as m-dash]C bonds and/or a high density of electron-rich C–H bonds, such as carbohydrate derivatives, represent an extremely challenging class of compounds for direct C–H amination. For these substrates, inventive synthetic solutions have to be found to achieve a predictable and high level of regio- and chemoselectivity. C–H amination can be impeded also by conformational effects or electronically disfavoured C–H bonds.
image file: c7qo00547d-f1.tif
Fig. 1 Selected examples of synthetically challenging products obtained by way of catalytic C–H amination reactions.

Pushing C–H amination to its limits represents a powerful driving force for further progress in the field (Fig. 1). There are many insights to be gained from such experiments. Counterintuitive selectivity, unexpected reactivity and failures may indeed offer a deeper understanding of the reaction mechanism and scope. This may lead to a better control of reaction outcomes by skilful tailoring of the substrate, the catalyst or the reaction conditions. The purpose of this review is to stimulate further research in this area by providing a description of the specific challenges facing catalytic C–H amination in the synthesis of natural products and related compounds, as well as tactics created to overcome them whenever such data are available. The reactivity of substrates with structural features challenging C–H amination technology will be presented. Most of the examples involve intramolecular nitrene insertion since the impact of the corresponding intermolecular process in organic synthesis is still rather limited. The development of site-selective intermolecular C–H amination is indeed a challenge in itself as highlighted in many reviews.1,4 Consequently, this technology has been applied to structurally simple substrates with a few exceptions, including the late stage intermolecular functionalization of natural products.

2. Essentials in C(sp3)–H amination

2.1 Catalytic systems and nitrenoid sources

Several catalytic systems and nitrenoid sources have been studied for C(sp3)–H amination, in particular in its intramolecular version (Fig. 2).1,8–15 The more popular systems were based on the in situ generation of metal nitrene from carbamates and sulfamic esters in the presence of a rhodium catalyst and an oxidant. Catalytic systems using dirhodium species have indeed proven to be highly successful for intramolecular C–H amination since the seminal work of Du Bois (Fig. 2a).1,4,8 In addition, cyclic sulfamidates, the corresponding C–H amination products obtained from sulfamic esters, are easily functionalized in the presence of various nucleophiles.16 According to the structure of the substrates, oxathiazinanes may be seen as an electrophilic azetidine equivalent or as a masked iminium ion (N,O acetal or aminal products when Z = OR or NRR′, Fig. 2b). After activation of the oxathiazinane by the introduction of an electron-withdrawing group on the nitrogen atom, the azetidine equivalent generated may be subsequently opened by nucleophilic attack on the oxygen-bearing carbon. The scope of C(sp3)–H amination reaction has also been expanded to other nitrenoid sources17–19 including ureas, guanidines and carbamimidates. To further improve the efficiency of the C(sp3)–H amination process, various metals have been evaluated as catalysts including silver,10 ruthenium,11 manganese,12 copper,13 iron14 and cobalt.15
image file: c7qo00547d-f2.tif
Fig. 2 Typical features for intramolecular C–H amination.

2.2 C–H insertion selectivity

The main advantage of intramolecular C(sp3)–H amination resides in the high level of regio- and stereocontrol usually observed for this process. The reaction proceeds with complete retention of configuration at the insertion site. General rules may be formulated to predict C–H amination regioselectivity: reactions performed with sulfamic esters lead generally to geometrically favoured 6-membered sulfamates, whereas carbamate substrates afford almost exclusively five-membered rings.

Electronic factors play also a decisive role. As highly electrophilic species, metal nitrenes have generally a strong preference for electron-rich C–H bonds (Fig. 2c). Sites adjacent to electron-donating group as well as allylic, benzylic and tertiary C–H bonds are favoured whereas sites adjacent to electron-withdrawing groups as well as primary C–H bonds are usually disfavoured. The order of reactivity for C–H bond insertion may be roughly formulated as follow: allylic > α-ethereal ≈ 3° > benzylic > 2° ≫ 1°.1 Non-classical regioselectivity trends are however observed in C–H amination (see section 4). Sulfamate substrates may form, for example, large rings (>6) whereas carbamates may be converted to ketones via 4-membered oxazetidinone intermediates. Recently, Schomaker et al. reported examples of tunable differentiation between two types of C–H bonds (tertiary and benzylic) in ligand-controlled intramolecular silver-catalysed C–H amination.10d,e The same group described also the first amination of propargylic C–H bonds and showed that these bonds are favoured over benzylic C–H bonds.8d More interestingly, the group of White12b has recently developed manganese catalysts that selectively aminate primary C–H bonds over secondary and tertiary C–H bonds (see section 5.3). An important limitation in C(sp3)–H amination has thus been overcome; manganese nitrenoids are able to react selectively with strong primary C–H bonds displaying usually high bond dissociation energy without interfering with weaker secondary/tertiary C–H bonds and reactive functional groups such as olefins. Chemoselectivity is indeed an important issue for unsaturated substrates as aziridination is also a possible reaction (see section 3). It is important to note that the regioselectivity observed might be different whether the amination reaction is inter- or intramolecular. A correlation between site selectivity and reaction pathway20 has been recently postulated to explain the fact that rhodium-catalysed inter- and intramolecular C–H aminations display an inverse selectivity in benzylic-to-tertiary site competitive reactions.20f This theoretical investigation further suggest that inter- and intramolecular C–H bond aminations might proceed according to different pathways (see below).

2.3 Mechanistic aspects

Several mechanistic investigations have been studied in particular for the intramolecular C–H amination catalysed by Rh species.1,4,8–15,20 It is suggested that C–H amination operates via a concerted asynchronous pathway in most catalytic C–H amination systems via metallonitrene intermediates.1a,20 On the other hand, stepwise radical C–H abstraction/rebound processes have been also observed with base metals including iron, manganese and cobalt.12b,14a,15c Recent mechanistic studies based on kinetic isotopic effect (KIE) experiments have shown that intermolecular rhodium-catalysed C–H amination of tertiary C–H bond may proceed via a stepwise pathway20e whereas C–H insertion mode mediated by manganese catalysts may lie between the stepwise and concerted insertion pathway.12b Despite the large body of work on the subject, further research is still needed to have a complete, unambiguous picture of the C–N bond forming process.

3. Chemoselectivity challenges: C–H amination versus C[double bond, length as m-dash]C aziridination

The C–H amination of unsaturated compounds is particularly challenging since alkenes may react with electrophilic nitrenoids to provide the corresponding aziridines. Several studies have indeed shown that unsaturated sulfamic esters and carbamates afforded predominantly aziridination using rhodium catalysts.1b,19a–c,20a,21 To achieve high chemoselectivity in favour of allylic C–H amination, several catalytic systems have been studied (see infra).

3.1 Sulfamate substrates

In 2007, during their studies en route to the synthesis of (+)-saxitoxin, a neurotoxic agent, Du Bois et al. planned to introduce the nitrogen atom at C-5 by way of intramolecular C–H amination (Scheme 1).22 The first strategy was to perform an allylic C–H amination from δ,ε-unsaturated sulfamate 1 since an alkene function was required for further functionalization. However, treatment of 1 with Rh2(esp)2 catalyst provided aziridine 3 in 30% yield along with the desired allylic amines 2 (40% yield).
image file: c7qo00547d-s1.tif
Scheme 1 Allylic C–H amination of 1 in the context of the synthesis of (+)-saxitoxin.

Considering the low chemo- and stereoselectivity of the reaction, an alternative approach based on the functionalization of N,O acetal 5 used as an iminium ion surrogate was evaluated (Scheme 2). Compound 5 was formed by C–H amination of sulfamic ester 4 in good yields. Coupling of sulfamidate 5 with appropriate alkynyl zinc reagent afforded compound 6 in excellent diastereoselectivity. The synthesis of (+)-saxitoxin was completed in 16 further steps from intermediate 6 which was obtained in multigram scale (Scheme 2).

image file: c7qo00547d-s2.tif
Scheme 2 Alternative approach for the synthesis of (+)-saxitoxin.

This example demonstrates the difficulty of planning allylic C–H amination in total synthesis. Nevertheless, it has been shown, shortly after the synthesis of (+)-saxitoxin, that allylic C–H amination can been obtained as the major or unique pathway depending on the catalytic system used. During their studies on enantioselective intramolecular C–H amination, Du Bois et al. observed serendipitously that Rh2(S-nap)4[thin space (1/6-em)]23 may display high chemoselectivity for allylic amination of sulfonamides 7–10 compared to other Rh-catalysts (Scheme 3).20a,24,25 Quite surprisingly, the related catalyst Rh2(PTPI)4 afford opposite chemoselectivity compared to Rh2(S-nap)4 for the conversion of substrate 7.23

image file: c7qo00547d-s3.tif
Scheme 3 Chemoselective allylic C–H amination using Rh2(S-nap)4.

A marked difference of reactivity may be observed between sulfamic esters and sulfonamides.19 Treatment of hex-5-ene-1-sulfonamide (11) with Rh2(OAc)4 provided only C–H amination product 1219b whereas the conversion of the corresponding sulfamate 8 afforded a 1[thin space (1/6-em)]:[thin space (1/6-em)]1 mixture of the C–H insertion and the aziridination products (Schemes 3 and 4).

image file: c7qo00547d-s4.tif
Scheme 4 Synthesis of cyclic sulfonamide 12 by C–H amination.

Although cyclic sulfonamides may seem less synthetically useful as reactive intermediates than the corresponding carbamate and sulfamate analogues, functionalization of these motifs has been nevertheless reported in the literature.26 Furthermore cyclic sulfonamides have been used as scaffolds for drug development.27

The choice of the catalyst metal is also an important key to achieve high chemoselectivity in favour of C–H amination. Indeed, several studies showed than ruthenium, manganese and iron catalysts afford mainly insertion into allylic C–H bonds.11a,c,12b,14a In 2012, the group of White developed [FePc] (Pc = phtalocyanito) as a new, inexpensive and nontoxic catalyst for allylic C–H amination (Schemes 5 and 6).14a In contrast to Rh2(OAc)4, [FePc] displays high chemoselectivity since only trace amounts of aziridines were observed. Furthermore, this catalyst allows also high regioselectivity in favour of allylic amination over C–H insertion into tertiary, ethereal or benzylic C–H bonds (Scheme 5).14a

image file: c7qo00547d-s5.tif
Scheme 5 Allylic C–H amination reactions catalysed by [FePc] and Rh2(OAc)4.

image file: c7qo00547d-s6.tif
Scheme 6 Comparison of different catalytic systems for C–H amination of selected unsaturated sulfamates.

Ru2(hp)4Cl and Mn(tBuPc) catalysts developed by the groups of Du Bois and White, respectively, also display high chemoselectivity in favour of allylic C–H amination over aziridination (Scheme 6).11c,12b Direct comparison between [FePc], Ru(hp)4Cl and Mn(tBuPc) catalysts showed that the manganese-based catalyst provides the best yields and chemoselectivity. Excellent C–H insertion[thin space (1/6-em)]:[thin space (1/6-em)]aziridination ratios have been also observed by Zhang et al. for the cobalt-catalysed C–H amination of unsaturated sulfamoyl azides.15

3.2 Carbamate substrates

As for sulfamates (see section 3.1), competition between aziridination and C–H amination is likewise observed for unsaturated carbamates.9a,b,13a,28 However in the case of carbamates, fewer catalytic systems allowing chemoselective C–H amination have been reported in the literature. During their studies on the C–H amination of homoallenic carbamates, Schomaker et al. observed that the reaction chemoselectivity was strongly dependent of the substrate structure (Scheme 7).29 For example treatment of allene 13a with Rh2(esp)2 yielded predominantly aziridine 14a, whereas under the same conditions allene 13b afforded the C–H amination product 15b in 80% yield. A series of catalysts has been screened to improve the insertion[thin space (1/6-em)]:[thin space (1/6-em)]aziridination ratio.29c Quite unexpectedly and in contrast with the allylic C–H amination of sulfamic esters, Ru2(hp)4Cl and [FePc]Cl failed to achieve this purpose. Conversely, silver triflate provided very interesting results. While treatment of 13a and 13b with AgOTf resulted in low conversion yields, the addition of a heteroaromatic ligand was found to have a dramatic effect. Thus, the addition of 1.25 equivalent of phenanthroline (phen) with respect to AgOTf gave excellent yields and high chemoselectivity in favour of aziridines 14. Remarkably, the chemoselectivity of the process can be effectively inverted when the catalyst[thin space (1/6-em)]:[thin space (1/6-em)]ligand ratio is increased to 1[thin space (1/6-em)]:[thin space (1/6-em)]3 (Scheme 7). Indeed, carbamates 15a and 15b were obtained in up to 81% yields under these conditions. The authors reported similar results for the C–H amination of the corresponding homoallylic carbamates.29d
image file: c7qo00547d-s7.tif
Scheme 7 C–H amination of homoallenic carbamates.

To explain the chemoselectivity observed, the authors suggested that the catalytic system at play depends on the catalyst[thin space (1/6-em)]:[thin space (1/6-em)]ligand ratio used. Ag(phen)OTf may favour aziridination, whereas Ag(phen)2OTf may promote the formation of the allylic C–H amination product.29d Experimental and computational studies led very recently to the hypothesis that suppression of the arizidination reactions may be due to the steric congestion increase around the silver centre in Ag(phen)2OTf.29e Furthermore, a decrease of the C–H insertion rate is observed with Ag(phen)OTf by comparison to Ag(phen)2OTf.

4. Regioselectivity challenges

Regioselectivity in C–H amination is generally controlled by electronic factors. The order of reactivity for C–H bond insertion could be roughly summarized as follow: allylic > α-ethereal ≈ 3° > benzylic > 2° ≫ 1°. Considering intramolecular reactions, sulfamates afford mainly 6 membered rings whereas carbamates give only 5 membered rings, with very few exceptions. The matter may be, however, further complicated by subtle stereoelectronic effects or by substrates having a high density of activated C–H bond. Regioselectivity challenges may be nevertheless overcome by various tactics including the fine tuning of protecting groups, the use of conformation effects or the clever exploitation of kinetic isotope effect (C–H versus C–D bond insertion).

4.1 Counter-intuitive regioselectivity

While intramolecular C–H amination of sulfamates or carbamates generally afforded 5- or 6-membered rings, some examples of formation of 7-, 8-, or 9-membered rings have been reported in the literature, leading to unanticipated regioselectivity.

In the context of the synthesis of FR900482 and mitomycin derivatives, Trost et al. planned to modify the oxidation state at C-8 in compounds 16 and 19 by way of C–H amination (Scheme 8).30 However, unexpected results were obtained. Thus, the intramolecular C–H amination of a 1[thin space (1/6-em)]:[thin space (1/6-em)]1 mixture of C-7 epimers 16 led to the formation of the undesired five-membered C–H insertion product 17 and, more surprisingly, to the intriguing nine-membered ring 18. It is the unique example of the formation of a large ring by intramolecular C–H amination of a carbamate. This result may be explain by the concave shape of the pyrrolizidine ring system that may place the nitrene reactive center in a favorable position with respect to the reactive C–H bond in α-position to the endocyclic nitrogen atom. Conversion of the corresponding sulphonamides 19 gave also puzzling results and two products were obtained as a 1[thin space (1/6-em)]:[thin space (1/6-em)]1 mixture in 42% yield. Again, a nine-membered ring resulting from C–H insertion at C-11 was obtained along with the even more surprising stable zwitterionic product 20 formed by attack of the aniline nitrogen atom. It is noteworthy that no product corresponding to the insertion into the C–H bond at C-8 was detected. Yet, such regioselectivity was expected to be strongly promoted by the formation of a geometrically favoured 6-membered sulfamates and by the fact that the insertion would occur into a tertiary C–H bond.

image file: c7qo00547d-s8.tif
Scheme 8 Unexpected regioselectivity in C–H amination of carbamates 16 and sulfamates 19 in the context of the synthesis of FR900482, mitomycins and related compounds.

C–H amination reactions31 of various square sugar32 derivatives have been reported during studies devoted to the synthesis of constrained spiranic iminosugars (see also sections 4.3 and 5.4).31 Surprisingly, no C–H insertions were observed by exposure of 22 to standard Du Bois's conditions despite the presence of activated tertiary and α-ethereal C–H bonds on the cyclobutane ring (Scheme 9a).31b Alcohol 23 was the only product isolated (32% yield). The formation of 23 was explained by the formation of a 9-membered cyclic sulfamidate intermediate. C–H insertion into the activated methylene of the benzylic ether would generate hemiaminal 24, leading to alcohol 23 after an acidic workup. The significantly higher s-character of the exocyclic bonds on four-membered hydrocarbon ring may explain the absence of C–H insertions on the cyclobutane ring, leading to 24 as the only C–H amination product.31

image file: c7qo00547d-s9.tif
Scheme 9 Synthesis of 9-membered rings by C–H amination.

Another example of the formation of a 9-membered ring by C–H amination has been reported by Blakey et al. during their studies on enantioselective allylic C–H amination (Scheme 9b).11a Treatment of sulfamate 25 with chiral catalyst 28 afforded the desired oxathiazinane 26 along with the 9-membered product 27. It is noteworthy that no competing aziridination was observed during this reaction. En route to the synthesis of (−)-muraymycin D2,33 an antibacterial nucleoside natural product, Ichikawa et al. formed the L-epi-capreomycidine core by way of C–H amination of sulfamates 29 (Scheme 10).33a The desired C–H insertion at C-3 proved difficult for substrates 29a–c whatever the nature of the amino protecting groups used.33a Treatment of 29a or 29b in the presence of catalytic amount of Rh2(esp)2 afforded the C-3 insertion products in moderate yields and low diastereoselectivity towards the undesired diastereoisomers 31. Interestingly, when the amine at C-2 was protected as a phtalimido group, formation of the undesired isomer 31 was totally suppressed. These results suggest that the free hydrogen at the 2-amino group may influence the diastereoselectivity of the reaction. For all substrates, the low yields observed for the synthesis of 30 and 31 was explained by the formation of other rather unexpected C–H amination products. Indeed, C–H insertions at C-2, C-4 and C-5 to form 5-, 7- and 8-membered rings, respectively, were also observed. Whereas the formation of N-sulfonyl imines 35 and 36 is explained by the presence an amino group at C-2 which increases the electronic density of the adjacent C–H bond, the formation of compounds 32c, 33c and 34 is more counter-intuitive. Indeed, the synthesis of 7 and 8-membered rings by C–H amination is uncommon (for other examples see Schemes 12 and 29). The formation of such products is favoured by the presence of the NHBoc group which increases the electronic density at C-5 and, to a lesser extent, at C-4. Structure of compound 34 obtained from 29a is intriguing and could proceed by C–H insertion at C-5 position to form a 8-membered ring (for a similar example see structure 20 in Scheme 8). The formation of 7- and 8-membered rings was suppressed by using a more electron-withdrawing phtalimido (Phth) group to protect the amino group at C-5 (compound 29b).

image file: c7qo00547d-s10.tif
Scheme 10 Counter-intuitive regioselectivity of C–H amination in the context of the synthesis of (−)-muraymycin D2.

Other examples of counter-intuitive regioselectivity in C–H amination have been reported in the context of the synthesis of 1,2-disubstituted diamondoids designed as conformationally rigid analogues of the antihyperglycemic agent vildagliptin® (Scheme 11).34 Whereas C–H amination of tertiary C–H bonds is generally favoured, treatment of noradamantane derivative 37 in the presence of Rh2(OAc)4 produced exclusively compound 38. The authors explained this unexpected regioselectivity by ring strain in the 5-membered ring of the noradamantane cage, leading to an increase of the s-character of the tertiary C–H bond. The same authors also described the C–H amination of diamondoid derivative 39. In this case, two regioisomers were obtained with a ratio in favour of oxathiazinane 41 corresponding to C–H insertion into the geometrically accessible tertiary C–H bond over the secondary one. Product 40 was however formed in significant yield highlighting the relative lower reactivity of tertiary C–H bonds in such systems.

image file: c7qo00547d-s11.tif
Scheme 11 C–H amination of adamantane derivatives.

4.2 Impact of conformational and/or stereoelectronic effects

As described in part 4.1, synthesis of heterocycles with unusual ring sizes over 6 atoms by C–H amination has been reported in the literature. It has been shown by Compain et al. that the formation of such large rings could be controlled by a combination of conformational and stereoelectronic effects.35,36 First insights were obtained with piperidines 42 used as models to develop a general access to mimetics of N-acetylhexosaminidase substrates by C–H amination (Scheme 12).35,36 Treatment of piperidine 42a in the presence of Rh2(OAc)4 was expected to afford the products corresponding to the insertion at C-3 position (with formation of a favoured six-membered ring) and/or at C-2 position (with formation of a five-membered ring by insertion into a favoured tertiary α-amino C–H bond). C–H amination of sulfamate 42a led surprisingly to the formation of bicyclic aminal 43a,35 the first seven-membered ring ever obtained by way of the Du Bois reaction (for other examples see Schemes 10 and 29). Remarkably, 8-membered rings can also be formed as demonstrated by the reaction performed from sulfamate 42b, the higher homologue of 42a. It is noteworthy that the reactivity of the N-tosyliminium ion precursor 43a has been exploited to develop an original bond-construction strategy based on iterative functionalization of nonactivated C–H bonds in nitrogen-containing heterocycles.36 In sharp contrast to piperidines 42, pyrrolidine 44a, pyran 44b and tetrahydrofuran 44c led only to 5-membered cyclic products 45a–45c corresponding to the insertion into the C–H bond at C-2 (Scheme 12).35
image file: c7qo00547d-s12.tif
Scheme 12 Conformational effects in the C–H amination of piperidines.

The marked difference of regioselectivity obtained in the pyran and the piperidine series was explained by conformational effects (Fig. 3).35,37 The pyran ring in 44c is likely to adopt a chair conformation in which the sulfamoyloxymethyl substituent is equatorial. This conformation makes the addition into the reactive axial C–H bond at C-2 highly favourable and prevents amination of the C-6 position. The C–H bond weakening is indeed reached when the adjacent heteroatom lone pair is aligned or closely aligned with the σ*(C–H) orbital for maximum overlap and minimal when they form a 90° angle (Fig. 3). In contrast to pyran 44c, the 6-membered ring of piperidine 42a may be close to a chair conformation in which the sulfamoyloxymethyl substituent occupies an axial position to minimize the pseudo 1,3-allylic strain caused by the partial double bond character of the S–N sulphonamide bond. This conformation favoured the insertion into the reactive axial C–H bond at C-6 to form a 7-membered ring in which the values observed for the S–N and S–O bond lengths and the N–S–O angle (∼106°) are very close to those obtained for oxathiazinane derivatives.1,8b Amination at C-2 would lead to the formation of a less favoured 5-membered ring. More importantly, this position is unactivated since the adjacent nitrogen lone pair is orthogonal to the equatorial σ*(C2–H).

image file: c7qo00547d-f3.tif
Fig. 3 Conformational and stereoelectronic effects in C–H amination of piperidine 42a and pyran 44c.

Stereoelectronic effects have been also observed by the same authors for the C–H amination of C-glycosides 46β and 46α (Scheme 13).38 C–H insertion into the favoured axial “anomeric” C–H bond of 46β provided after protection oxathiazolidine 47 in 63% yield. In sharp contrast, 46α, the C-1 epimer of 46β, led to a mixture of unidentifiable products. For stereoelectronic reasons described above (Fig. 3), the C–H insertion into the pseudo anomeric equatorial C–H bond is disfavoured over the two activated axial α-oxygenated C–H bonds at C-3 and C-5. Insertion into these positions may lead to the formation of unstable hemi-aminal derivatives and then to degradation.37,38

image file: c7qo00547d-s13.tif
Scheme 13 Difference of reactivity in C–H amination of pseudo anomers 46.

Another example of unusual regioselectivity attributed to stereoelectronic factors was observed during the C–H amination of 48 (Scheme 14).39 It was expected that insertion occurred into the C–H bond at C-2 to form the corresponding 6-membered sulfamate. Rather unexpectedly, the 5-membered oxathiazolidine 49 was the only product obtained in the amination process. In addition only C–H insertion at C-3 was obtained whatever the nature of the protecting groups used;39 no conversion was observed from the deprotected analogue of 48.

image file: c7qo00547d-s14.tif
Scheme 14 Unexpected regioselectivity upon C–H amination of 48.

Several factors may explain the regioselectivity observed. The rigid conformation of 48 may present the C–H bond at C-3 in a much more favourable position for nitrene insertion than the C–H bond at C-2 and C-5 as suggested by the crystal structure of 48.39 Stereoelectronic effects can also be considered as the dominating effect to explain the, at first, counter-intuitive specificity observed for the nitrenoid attack at C-3 over C-5 (Fig. 4). Both tertiary C–H bonds show indeed a similar spatial orientation and both are in α-position to an oxygen atom, the insertion product leading to a 5-membered ring in both cases. Attack at C-3 is however likely to be favoured by hyperconjugaison40 from the adjacent oxygen lone pair which is held in a suitable conformation to overlap well with the C3–H σ* orbital.

image file: c7qo00547d-f4.tif
Fig. 4 Stereoelectronic analysis of regioselectivity observed in C–H amination of 48.

The diastereoselectivity of C–H amination reaction may be also controlled by “pure” conformational effect as demonstrated by Lebel et al. in a recent study (Scheme 15).9c While cis-4-tert-butyl-cyclohexanol derivative 50a led to the desired cis-oxazolidinone 51a in good yields and high diastereoselectivity, the conversion of the corresponding trans derivative 50b proved difficult and led to trans-oxazolidinone 51b in poor isolated yields. These results that have been confirmed by DFT calculations indicated clearly that insertion into the less hindered equatorial C–H bond is preferred over the axial C–H bond. In trans-4-tert-butyl-cyclohexanol derivative 50b, the carbamate moiety is forced into an equatorial position leading to the formation of the constrained trans-oxazolidinone 51b by insertion into the adjacent equatorial C–H bond.

image file: c7qo00547d-s15.tif
Scheme 15 Conformational bias in C–H amination of cycloxanol derivatives.

4.3 Fine tuning of protecting groups

In the context of the synthesis of conformationaly constrained iminosugars, Compain et al. studied the C–H amination of polyoxygenated cyclobutanes (Schemes 16–18, see also Schemes 9a and 38).31 In addition to regioselective issues induced by the high density of activated α-ethereal C–H bonds, this study was a rare example of C–H amination of cyclobutanic C–H bonds.10c,d,41 Reaction of carbamate 52a with catalytic amount of Rh2(esp)2 in the presence of PhI(OAc)2 and MgO afforded hemiaminal 54a in 61% yields as the only regioisomer (Scheme 16).31b The presence of the ester group is indeed expected to deactivate the position at C-4. To disfavour the insertion at C-2 and reverse the reaction regioselectivity, the benzyloxy group was switched by an electron-withdrawing protecting group. However, no C–H amination product was obtained from cyclobutane 52b. The presence of four electron-withdrawing groups around the cyclobutane ring eventually abolished the reactivity of the cyclobutane C–H bonds.
image file: c7qo00547d-s16.tif
Scheme 16 Regioselectivity in C–H amination of 52.

image file: c7qo00547d-s17.tif
Scheme 17 Synthesis and evaluation of carbamate 56 as C–H amination substrate.

image file: c7qo00547d-s18.tif
Scheme 18 Influence of protecting groups in the C–H amination of carbamates 58.

To generate the desired C–N bond at C-4, the authors planned to take advantage of the C–H amination product 54a. The objective was to cleave the oxazolidinone ring to afford a hemi-aminal function, as a masked ketone, and an alcohol at C-3 that may be converted into the corresponding carbamate to perform a second C–H amination reaction (Scheme 17). After N-Boc protection, oxazolidinone 54a was reacted with cesium carbonate to provide alcohol 55 which was converted into the corresponding carbamate 56 in 51% yield for the three steps. Unfortunately, despite several attempts using typical rhodium-catalysed conditions, no C–H amination product could be obtained from 56, although there are theoretically limited regioselectivity issues with this substrate (Scheme 17).31b

To favour the desired regioisomer the authors performed C–H amination's studies on subtrates 58 having a vinylic group at the C-4 position (Scheme 18).31 C–H amination of allylic C–H bonds was indeed shown to be favoured over α-oxygenated C–H bonds in several studies.14a,42 Substrate 58a was subjected to Rh2(esp)2-catalysed C–H amination and in this case desired regioisomer 59a was obtained in 17% yield. However undesired compound 60a was still the major product. Other catalytic systems were studied but afforded lower yields. To overcome this unexpected regioselectivity, the electronic density of the α-oxygenated C–H bond at C-2 was reduced by using electron-withdrawing protecting group. When carbamates 58b–d were subjected to C–H amination, the formation of product 60 was suppressed and the desired regioisomer 59 was obtained in yields up to 40%. This moderate yield could be explained by the presence of three electron-withdrawing groups around the cyclobutane ring which may also reduce the reactivity of the C–H bond at C-4. Carbamate 59c was used as an intermediate for the synthesis of unprecedented constrained iminosugars 61–63.31

Del Valle et al. have also shown the influence of protecting groups on the regioselectivity of the C–H amination of carbamates 64 (Scheme 19).43 Reaction of 64a with catalytic amount of Rh2(OAc)4 led to exclusive insertion at the ethereal carbon to give 6-membered compound 66a. This result was unexpected in light of the well-established preference of carbamates to afford 5-membered rings. The authors explained this regioselectivity by the hindrance of the silyl protecting group which block the β face of the cyclopentane ring. When the reaction was carried out with the less sterically demanding MOM protecting group, desired carbamate 65b was obtained in 30% yield, the 6-membered carbamate 66b being still the major product.

image file: c7qo00547d-s19.tif
Scheme 19 Influence of protecting groups on the regioselectivity of C–H amination of carbamates 64.

Steric hindrance generated by TBS protecting group may also explain the high regioselectivity observed in the C–H amination of a precursor of (−)-cytoxazone, a cytokine-modulating agent isolated from Streptomyces sp. (Scheme 20).9c In this case, C–H amination catalysed by chiral valine derived Rh2[(S)-ntv]4[thin space (1/6-em)]19d into the benzylic C–H bonds was shown to be highly favoured over the more hindered α-ethereal C–H bond, both C–H bonds being β to the N-mesyloxycarbamate.

image file: c7qo00547d-s20.tif
Scheme 20 Regioselective C–H amination of N-mesyloxycarbamate 67.

4.4 The carbamate issue

In contrast to sulfamates, only few examples of C–H amination yielding large rings have been reported with carbamate substrates (for an example of 9-membered ring see Scheme 8). However, carbamates may be converted to ketones by C–H amination probably via the formation of a 4-membered ring.44–52 For example, reaction of 69 in the presence of Rh2(OAc)4 afforded only ketone 70 in 58% yield (Scheme 21).45 The use of dirhodium catalysts Rh2(esp)2 and Rh2(OCOCPh3)4 having bulkier ligands allowed the formation of the desired compound 71, albeit in low yields, ketone 70 being still the major product. Compound 70 may be generated via the formation of 4-membered ring 72. Direct hydrogen abstraction by nitrenoid intermediates from the carbon atom in α position to the carbamate may be also considered.45,46 Carbamate 71 was converted to pachastrissamine (jaspin B), a natural product isolated from marine sponge which displays cytotoxic activities against several human carcinoma cell lines.45
image file: c7qo00547d-s21.tif
Scheme 21 Formation of ketone 70 by C–H amination en route to the synthesis of jaspin B.

The influence of the catalytic system in the formation of ketones by C–H amination of carbamates has also been observed by the group of Nemoto (Scheme 22).47 For their approach to the core of Pactamycin, a natural product with potent antiproliferative activity against bacteria, viruses or protozoa, Nemoto et al. studied the C–H amination of carbamate 73.47 Using Rh2(esp)2 catalyst, the desired compound 74 was isolated in only 7% yield along with ketone 75 obtained as the major product. Screening of Rh catalysts and oxidants revealed that Rh2(OAc)4 in the presence of PhI(OCOt-Bu)2 afforded desired compound 74 in 70% yield. The authors examined also the C–H amination of 73 in the presence of silver triflate.10b However in these conditions the desired product 74 was not obtained.47

image file: c7qo00547d-s22.tif
Scheme 22 Approach to the pactamycin core via C–H amination.

In contrast to 73 (Scheme 22), the use of AgOTf was crucial for the formation of carbamate 77a from challenging substrate 76a (Scheme 23). Indeed, treatment with Rh catalysts furnished ketone 78 rather than the desired product 77a.48a In contrast, conversion of 76a using 0.5 equivalent of AgOTf in the presence of bathophenantroline and PhI(OAc)2 afforded carbamate 77a as a major product with concomitant formation of ketone 78.48 More interestingly, the authors showed that the formation of ketone 78 could be limited using deuterium kinetic isotope effects (Scheme 23);48b the introduction of a deuterium in place of a hydrogen atom at C-10 significantly reduced the unwanted formation of the carbonyl function at this position. The C–H amination of deuterated carbamate 76b afforded compound 77b in 60% yield. This compound was further transformed into welwitindolinones 79a–79d. These naturals products are known to display promising activity against drug-resistant cancer cells.48 Remarkably, no C–H insertion was observed at C-9 position although allylic C–H bonds are considered to be more reactive than benzylic position.1,14a The low reactivity of the C–H bond at C-9 may be explained by the presence of a proximate gem-dimethyl group which could hindered the approach of the bulky nitrene intermediate. This explanation is highlighted by further observations of the same authors in their studies on the synthesis of N-methylwelwitindolinone B isothiocyanate (Scheme 24).49 Similar results in terms of yields and regioselectivity were obtained with 80 and its close analogue 76b showing that the low reactivity of the C–H bond at C-9 is irrespective of whether the hydrogen atom is allylic or not.49 This example further highlights the efficiency of the deuterium kinetic isotope effect strategy, since no ketone is formed at C-10 during the process.47–51 C–H amination of compound 82, the C-10 epimer of 80, provided 83 as the only product in 35% yield. In contrast to 80, no insertion into the favoured benzylic position at C-11 was observed presumably due to its neopentyl character. The subtlety here is that both the C-11 and C-9 position are neopentylic. The presence of bulky substituents on the same side of the metal-nitrene may favour conformations in which the nitrene precursor is in closer proximity to the reacting C–H bond. Steric hindrance at C-12 led thus to insertion at C-9 for 82, whereas for the corresponding C-10 epimer 80, the presence of a proximate gem-dimethyl group favoured the insertion at C-11 over C-9. It is noteworthy that the sensitive alkyl chloride unit is compatible with the Ag-mediated C–H amination conditions.

image file: c7qo00547d-s23.tif
Scheme 23 Deuterium kinetic effect in C–H amination en route to the synthesis of welwitindolinones.

image file: c7qo00547d-s24.tif
Scheme 24 Difference of regiosectivities in C–H amination of C-10 epimers 80 and 82.

In 2012 Hatakeyama et al. reported the synthesis of (−)-kaitocephalin, a potent antagonist of ionotropic glutamate receptors, via a sequence involving benzylic and allylic C–H aminations.52 The authors showed that subtle, yet unknown effects may favour the formation of a ketone function via nitrene C–H insertion. Indeed, treatment of carbamate 84 and the less constrained analogue 86 with Rh2(esp)2 gave rise to a completely different scenario (Scheme 25). As expected, allylic C–H amination of 84 to give the corresponding oxazolidone 85 proceeded in very good yields. In contrast, all attempts to perform allylic C–H amination of 86 was unsuccessful and only ketone 87 could be obtained in low yield.

image file: c7qo00547d-s25.tif
Scheme 25 Synthesis of (−)-kaitocephalin via C–H amination.

5. Reactivity challenges

Electronic factors often dominate reactivity in C–H amination substrates (see Fig. 2). The metal nitrenes behave as highly electrophilic species and therefore have a strong preference for electron-rich C–H bonds. Sites adjacent to electron-withdrawing groups are consequently considered to be electronically deactivated towards nitrenoid reactions. In some cases, subtle stereoelectronic effects may dominate pure electronic factors. For example, full activation of a C–H bond by hyperconjugation from an adjacent heteroatom lone pair is reached only when the lone pair is held in a suitable conformation to be aligned well with the σ*(C–H) orbital for maximum overlap (see section 4.2). Conversely, an orthogonal arrangement of the orbitals minimizes C–H activation. Late stage C–H amination of drug molecules or complex intermediates in total synthesis raise also many synthetic challenges because such compounds typically contain a variety of reactive functionalities. Finally, unexpected reactivity issues may be caused by remote protecting groups or by unexplained low conversion rates.

5.1 Electron-deficient C–H bonds

En route to the synthesis of (+)-conagenin, a natural immunomodulator, Yakura et al. prepared the α-methylserine fragment by way of direct intramolecular C–H amination of ester 88 (Scheme 26) based on Du Bois’ chemistry.53–55
image file: c7qo00547d-s26.tif
Scheme 26 En route to the total synthesis of (+)-conagenin.

As expected, the insertion into the targeted electronically deactivated C–H bond proved difficult. After extensive optimization, the best procedure involved the use of 10 mol% of Rh2(esp)2 in refluxing CH2Cl2 and provided the desired carbamate 89 in 44% yield along with 35% of recovered 88. C–H amination of N-tosyloxycarbamate 90 following Lebel's procedure9 led to a lower yield with no recovery of starting material. The use of sulfamide or O-sulfamoyl-N-alkylhydroxylamine substrates was found to significantly increase the Rh-catalysed insertion into electronically deactivated tertiary C–H bonds as shown by the conversion of 91a and 93 into 92 and 94, respectively (Scheme 27).18 However, the C–H amination yields were found to be drastically reduced when insertion occurred into a secondary C–H bond as shown by the reaction of N-Boc N-alkyl sulfamide derivative 91b.

image file: c7qo00547d-s27.tif
Scheme 27 Metal nitrene insertion into unactivated α-C–H bonds of esters.

Lu, Zhang et al. have shown recently that challenging electron-deficient C(sp3)–H substrates, such as secondary α-C–H bonds of esters could be aminated efficiently via Co(II)-based metalloradical catalysis.15b Treatment of sulfamoyl azides 95 with Co(II) catalyst 97 afforded the expected 5-membered cyclic sulfamides 96 in 89 to 99% yields (Scheme 28). Yields were significantly reduced when the insertion occurred in secondary α-C–H bonds of amides as shown by conversion of sulfamoyl azides 98.15b

image file: c7qo00547d-s28.tif
Scheme 28 Metal nitrene insertion into unactivated α-C–H bonds of esters and amides.

5.2 Conformational control of reactivity

As described in section 4.2, intramolecular C–H amination in α-position to endocyclic nitrogen atom in azacycloalkanes generates aminals that may react with various nucleophiles upon treatment with Lewis acid.36 In addition to introducing structural diversity, this process regenerates the metal nitrene precursor that may be used again for further C–H amination reactions. With the objective of developing new strategies for the diversity-oriented synthesis of azacycloalkanes by iterative scaffold decoration, piperidine and pyrrolidine systems in which a sulfamoyloxy group was connected to the endocyclic nitrogen were explored (Scheme 29).56 A marked difference of reactivity between 5- and 6-membered azacycles was observed. Quite surprisingly, the introduction of an electronically favored allylic site did not improve the insertion process as shown with tetrahydropyridine 100b.
image file: c7qo00547d-s29.tif
Scheme 29 Conformational effects in intramolecular C–H amination of azacycles 100 and 101.

As suggested by X-ray analysis and 1H NMR of 100a, the well-defined chair conformation of the piperidine ring and the planar arrangement induced by amide conjugation may place the nitrene reactive center in an unfavorable position with respect to the more reactive axial C–H bond in α-position to the nitrogen atom (Fig. 5a). In contrast, the pyrrolidine ring adopt a more flexible planar conformation in which the two geminal α-amino C–H bonds are both activated with a dihedral angle with the p-type lone pair orbital of the nitrogen of ca. 30°.

image file: c7qo00547d-f5.tif
Fig. 5 Conformational and stereoelectronic effects in intramolecular C–H amination of azacycles 100 and 101.

In the pyrrolidine series, the introduction of a cyclopropyl group in the spacer arm between the sulfamoyloxy group and the endocyclic nitrogen dramatically improves the efficiency of the C–H amination process from 47 to 86% yields. This result was rationalized by the “reactive rotamer” concept, i.e. the cyclopropyl group favours a reactive conformation in which the nitrene precursor is in close proximity to the reacting C–H bond (Fig. 5b).56

5.3 Late stage C–H amination

Late stage C–H amination is emerging as a promising strategy in total synthesis of complex natural products and novel drug candidates.57 This approach combines indeed synthetic efficiency with the streamlining of structural diversification. In addition, the introduction of amino groups in the late stages of a multi-step synthesis may also limit unwanted side-reactions due to amine intrinsic reactivity. While conceptually attractive, late stage C–H amination strategy has to face many challenging hurdles. The presence of a variety of functional groups such as basic amines or alkenes at late stages in target-oriented synthesis greatly increases the level of difficulty in controlling regio- and chemoselectivity. The total synthesis of (−)-tetrodotoxin in 2003, shortly after the disclosure of the Du Bois’ C–H amination process has however beautifully demonstrated the feasibility of such a strategy (Scheme 30).6
image file: c7qo00547d-s30.tif
Scheme 30 Synthesis of (−)-tetrodotoxin by way of intramolecular C–H amination.

Despite the high density of α-oxygenated C–H bonds and the presence of reactive functions, treatment of 104 with catalytic amount of Rh2(HNCOCF3)4, PhI(OAc)2 and MgO provided the expected C–H insertion product 105 in 77% yield. It is noteworthy that the same reaction performed under conditions originally reported for Du Bois reaction8c furnished only trace amounts of the desired product 105.6 The same group reported an approach to the assembly of the B/C/D ring system of (+)-aconitine, a toxin produced by the Aconitum plant (Scheme 31).58

image file: c7qo00547d-s31.tif
Scheme 31 Approach to the B/C/D ring system of (+)-aconitine.

The strategy was based on a late stage C–H amination reaction that would provide an oxathiazinane N,O-acetal used as a direct iminium precursor en route to the A/B spiro-bicycle of (+)-aconitine. This approach was evaluated with sulfamate 106, a substrate encumbered with potential regioselectivity issues due to the presence of activated tertiary or benzylic C–H bonds. Remarkably, the C–H amination reaction performed with Rh2(esp)2 was found to be completely regioselective. In particular, no 6-membered product corresponding to insertion into the tertiary C–H bond at C-8 was observed. This result may be due in part to the presence of the adjacent strongly electron-withdrawing trifluoroacetyl protecting group. The divergent synthesis of various amino analogues of artemisinin from carbamates or sulfamic esters has been described by Wong and Che (Scheme 32).44 This research was motivated by the antimalarial and cytotoxicity activity of this famous natural sesquiterpene lactone.59 Late stage amination of artemisinin derivatives was performed from sulfamate 109 and carbamate 111 generated in few steps from readily available 10-dihydroartemisinin 108. Remarkably, the reaction of sulfamate 109 with Rh2(OAc)4 catalyst and MgO provided oxathiazinane 110 in 85% yield as a single regioisomer. Insertion into the secondary C–H bond at C-8 was thus highly preferred over insertion into more electron-rich C–H bonds; no insertion was observed into the ethereal Cα–H bond at C-10 to form a 5-membered ring or into the acetal C–H bond at C-12 to form a 7-membered ring. Intramolecular Cα–H amination of 111, the corresponding carbamate analogue of 109, proved more difficult despite the presence of only one electron-rich site of insertion at C-10 leading to a geometrically favoured 5-membered carbamate. Good conversion (77%) and acceptable yields were obtained only with Rh2(O2CC3F7)4 catalyst, no reaction being observed when [Rh(TTP)Cl] or [Ru(TTP)CO] were used. Among the artemisinin analogues prepared with an amino group at C-8 or C-10, compound 113 was found to display cytotoxic activity whereas carbamate 112, sulfamate 110 and aminal 114 were found to be inactive (Scheme 32).

image file: c7qo00547d-s32.tif
Scheme 32 Synthesis of artemisinin analogues by way of intramolecular C–H amination.

In a recent work, Romo et al. used intermolecular C–H amination for simultaneous structure–activity studies and arming of bioactive natural products.60 Steroids61 and terpenes bearing activated allylic or benzylic sites were regio- and chemoselectively functionalized under oxidative treatment with Rh2(esp)2 catalyst in the presence of a trichloromethyl substituted sulfamate (Scheme 33). The process efficiency was found to be satisfactory considering the yields that would be obtained by de novo synthesis.

image file: c7qo00547d-s33.tif
Scheme 33 Products obtained after intermolecular site-selective C–H amination of some terpenes and steroids.

In 2015, White et al. reported intramolecular late stage amination using a novel manganese C–H amination catalyst (Scheme 34).12b For example, the functionally dense picrotoxinin derivative 115 was converted to the corresponding oxathiazinane 116 by C–H insertion into a tertiary C–H bond in good yields. Interestingly, under manganese tert-butylphthalocyanine [Mn(tBuPc)] catalysis (see Scheme 6), amination to form a geometrically favoured 6-membered ring is generally preferred regardless of the relative bond strength. In contrast to rhodium catalysts, [Mn(tBuPc)] may thus aminate efficiently primary C–H bonds. This unusual selectivity and reactivity was for example applied to dihydropleuromutilone sulfamate ester 117.

image file: c7qo00547d-s34.tif
Scheme 34 Late stage diversification of complex molecules by way of intramolecular Mn-catalysed C–H amination.

In this challenging substrate, intramolecular C–H insertion can feasibly occur into primary C–H bonds at C-16 to form a 6-membered ring or into secondary C–H bonds at C-8 or C-6 to form a 5-membered ring. Remarkably, the Mn-catalyzed C–H amination process provided the oxathiazinane 118 as a single regioisomer in 84% yield.12b [Mn(tBuPc)] catalyst is thus able to unlock the usual low reactivity observed for primary aliphatic C–H bonds which is due in part to their higher bond dissociation energy compared to secondary or tertiary C–H bonds.

5.4 Unexpected reactivity

Sometimes remote protecting groups may have a decisive and unexpected influence on the C–H amination step as shown by the Shindo's synthesis of the core skeleton of stemonamide.62 The synthesis of the desired spirocyclic compound 120 was first generated in almost quantitative yield by treatment of carbamate 119 with Rh2(OAc)4 catalyst (Scheme 35). Since the deprotection of the terminal methoxy group was unsuccessful, various protecting groups were evaluated. No C–H amination product was obtained with alcohol derivatives 119b,c protected as silyl ethers (TIPS or TMS) regardless of the relative steric hindrance. The Ms-protected substrate 119d failed also to provide the expected C–H amination product. The problem could be solved by using an acetate-protected substrate and, more counter-intuitively, by removing the protecting group. This is another example that, sometimes, the simpler the solution, the better.
image file: c7qo00547d-s35.tif
Scheme 35 Towards the synthesis of the core skeleton of stemonamide by way of intramolecular Rh-catalysed C–H amination.

Remote protecting groups were also found to have an influence on the enantioselective intermolecular C–H amination of allylic bond of silyl enol ethers, as shown with compounds 121 (Scheme 36). This challenging insertion process was developed by Hashimoto et al. to achieve a concise, catalytic asymmetric route to (−)-pancracine, which displays a variety of biological activities.63–65 The best yield was obtained for silyl enol ether 121d, the triethylsilyl group representing a good balance between stability and bulkiness. The trimethyl silyl functionality was found to be partially unstable under the reaction conditions and more robust bulky silyl groups may reduce substrate reactivity towards metal nitrenes. It is noteworthy that application of Hashimoto's protocol to cyclopentanone and cycloheptanone derivatives afforded a complex mixture of products, including β-amino ketones, but also α-amino ketones arising from enol ether aziridinations.

image file: c7qo00547d-s36.tif
Scheme 36 Enantioselective C–H amination of silyl enol ethers 121.

Catalytic C–H amination of benzylic bonds is generally a favoured process. Steric effects may however dramatically reduce amidation efficiency as showed by the striking absence of reactivity of o-methoxyphenyl sulfamate ester 124a when treated with the ruthenium porphyrin catalyst [Ru(F20-TPP)(CO)] (Scheme 37).11b The impact of the steric hindrance was also nicely highlighted by the correlation between C–H insertion efficiency and the positions of the methoxy substituent in the phenyl ring, from ortho to meta and para (Scheme 37). In sharp contrast, exposure of ortho-substituted substrate 124a to cationic ruthenium(II) pybox catalyst 28 (Scheme 9) led to the formation of the expected oxathiazinane 125a in 71% yields.11a

image file: c7qo00547d-s37.tif
Scheme 37 Ru-catalysed C–H amination of sulfamate esters 124.

Unexplained absence of reactivity may also occur with carbamate substrates. Despite the reactivity of tertiary C–H bond towards metal nitrenes and the strong bias of carbamates to form 5-membered ring, no C–H amination products was detected by treatment of squaranose32126 with Rh-catalysts (Scheme 38).31b After chromatography on silica gel, substrates 126 protected (Ac, TBS) or not were the only compounds that could be isolated. Formation of aldehyde derivatives, which would involve nitrene insertion into the secondary C–H bond in α position to the carbamate function, was not observed. This process is likely to be less favoured than the conversion of carbamates 69, 73, 76a or 86 into the corresponding ketones by nitrene insertion into a tertiary C–H bond (see section 4.4).

image file: c7qo00547d-s38.tif
Scheme 38 Evaluation of carbamates 126 as the substrate for C–H amination.

Rh-Catalysed intramolecular C–H amination of carbamate 128 synthesized from pleuromutilin, a natural antibacterial drug, led to low yields with many side reactions (Scheme 39).66 The insertion process suffered also from incomplete conversion of starting material. Changing the metal catalyst from rhodium to silver led to a marked improvement and the byproduct formation was significantly suppressed.

image file: c7qo00547d-s39.tif
Scheme 39 Synthesis of pleuromutilin derivatives by way of C–H amination.

6 Conclusion

Since the seminal works of Du Bois in the early 2000s, the pace of discovery in the field of metal-catalysed C–H amination has been breath-taking. Not surprisingly, this synthetic tool has been applied to the total synthesis of many compounds of interest, given the high prevalence of the amino group in natural products and synthetic pharmaceuticals.67 Chemist's confidence in the high potential of the C–H amination methodology to unlock inert C–H bonds has been demonstrated by its application to more and more challenging substrates. This has been a powerful drive for progress in the field. New valuable insights have been gained allowing, for example, a better regiochemical control via stereoelectronic and/or conformational effects. Innovative strategies have been discovered to direct the insertion event in substrates bearing a large degree of attendant functionality. Solutions have spanned from the elegant exploitation of kinetic isotope effects to the tactical use of protecting groups with different sizes or electronic characteristics. Systematic exploration of different catalytic systems has been also performed leading to the opening of new possibilities in C–H amination technology. Manganese-based catalysts have thus given rise to nitrenoids that overcome the low reactivity of primary aliphatic C–H bonds without interfering with weaker secondary/tertiary C–H bonds. Despite these impressive achievements, much remains to be done. Counterintuitive selectivity and unexplained reactivity should serve as a reminder that further studies are highly needed to understand in depth catalytic C–H amination chemistry. Many challenges remain on the way, from basic to applied research. A clear mechanistic view based on definitive evidence concerning the details of the C–N bond forming process would undoubtedly facilitate the rational design of efficient catalytic systems leading to higher regio-, chemio- and stereoselectivity. In particular, the quest for site-selective C–H amination through catalyst control has to be pursued.10d,e In this context, the development of efficient intermolecular C–H amination process still represents a major challenge and upcoming advancements are expected to increase the impact of this technology in organic synthesis. Future progress made in the field of catalytic C–H amination chemistry might also lead to industrial-scale applications in the next decade. It is likely that total synthesis of synthetically challenging targets related to natural products will continue to be a powerful driving force towards this goal.

Conflicts of interest

There are no conflicts to declare.


The authors are grateful to financial supports from the CNRS (UMR 7509), the French department of research, the University of Strasbourg, the association Vaincre La Mucoviscidose and the International Centre for Frontier Research in Chemistry (icFRC).

Notes and references

  1. For recent reviews on catalytic C–H amination, see: (a) C. G. Espino and J. Du Bois, in Modern Rhodium-catalyzed Organic Reaction, ed. P. A. Evans, Wiley-VCH, Weinheim, 2005, pp. 379–416 Search PubMed; (b) J. L. Roizen, M. E. Harvey and J. Du Bois, Acc. Chem. Res., 2012, 45, 911–922 CrossRef CAS PubMed; (c) P. Dauban and R. Dodd, in Amino Group Chemistry: From Synthesis to the Life Sciences, ed. A. Ricci, Wiley-VCH, Weinheim, 2007, pp. 55–92 Search PubMed; (d) A. R. Dick and M. S. Sanford, Tetrahedron, 2006, 62, 2439–2463 CrossRef CAS; (e) Z. Li and C. He, Eur. J. Org. Chem., 2006, 4313–4322 CrossRef CAS; (f) F. Collet, R. H. Dodd and P. Dauban, Chem. Commun., 2009, 5061–5074 RSC; (g) P. Compain and S. Toumieux, in Targets in Heterocyclic systems, Chemistry and Properties, ed. O. A. Attanasi and D. Spinelli, SCI, Rome, 2007, vol. 11, pp. 338–364 Search PubMed; (h) G. Dequirez, V. Pons and P. Dauban, Angew. Chem., Int. Ed., 2012, 51, 7384–7395 CrossRef CAS PubMed; (i) J. L. Jeffrey and R. Sarpong, Chem. Sci., 2013, 4, 4092–4106 RSC; (j) J.-P. Wan and Y. Jing, Beilstein J. Org. Chem., 2015, 11, 2209–2222 CrossRef CAS PubMed; (k) J. Buendia, G. Grelier and P. Dauban, Adv. Organomet. Chem., 2015, 64, 77–118 CrossRef; (l) J. Du Bois, Chemtracts, 2005, 18, 1–13 CAS; (m) R. K. Rit, M. Shankar and A. K. Sahoo, Org. Biomol. Chem., 2017, 15, 1282–1293 RSC; (n) J. Du Bois, Org. Process Res. Dev., 2011, 15, 758–762 CrossRef CAS PubMed; (o) Y. Park, Y. Kim and S. Chang, Chem. Rev., 2017, 117, 9247–9301 CrossRef CAS PubMed.
  2. (a) J. Yamaguchi, A. D. Yamaguchi and K. Itami, Angew. Chem., Int. Ed., 2012, 51, 8960–9009 CrossRef CAS PubMed; (b) H. M. L. Davies and J. R. Manning, Nature, 2008, 451, 417–424 CrossRef CAS PubMed.
  3. A. K. Mailyan, J. A. Eickhoff, A. S. Minakova, Z. Gu, P. Lu and A. Zakarian, Chem. Rev., 2016, 116, 4441–4557 CrossRef CAS PubMed.
  4. B. Darses, R. Rodrigues, L. Neuville, M. Mazurais and P. Dauban, Chem. Commun., 2017, 53, 493–508 RSC.
  5. For clarity's sake, the new C–N bond formed is drawn in red colour and the nitrogen atom involved in the C–H amination process in blue colour in the synthetic schemes.
  6. A. Hinman and J. Du Bois, J. Am. Chem. Soc., 2003, 125, 11510–11511 CrossRef CAS PubMed.
  7. J. Li and M. D. Eastgate, Org. Biomol. Chem., 2015, 13, 7164–7176 CAS.
  8. For Rh catalyst: (a) R. Breslow and S. H. Gellman, J. Am. Chem. Soc., 1983, 105, 6728–6729 CrossRef CAS; (b) C. G. Espino, P. M. Wehn, J. Chow and J. Du Bois, J. Am. Chem. Soc., 2001, 123, 6935–6936 CrossRef CAS; (c) C. G. Espino and J. Du Bois, Angew. Chem., Int. Ed., 2001, 40, 598–600 CrossRef CAS; (d) R. D. Grigg, J. W. Rigoli, S. D. Pearce and J. M. Schomaker, Org. Lett., 2012, 14, 280–283 CrossRef CAS PubMed.
  9. For Rh catalyst using N-tosyl- or N-mesyloxycarbamates see: (a) H. Lebel, K. Huard and S. Lectard, J. Am. Chem. Soc., 2005, 127, 14198–14199 CrossRef CAS PubMed; (b) K. Huard and H. Lebel, Chem. – Eur. J., 2008, 14, 6222–6230 CrossRef CAS PubMed; (c) H. Lebel, L. Mamani Laparra, M. Khalifa, C. Trudel, C. Audubert, M. Szponarski, C. Dicaire Leduc, E. Azek and M. Ernzerhof, Org. Biomol. Chem., 2017, 15, 4144–4158 RSC.
  10. For Ag catalysts: (a) See ref. 1e. ; (b) Y. Cui and C. He, Angew. Chem., Int. Ed., 2004, 43, 4210–4212 CrossRef CAS PubMed; (c) R. J. Scamp, J. G. Jirak, N. S. Dolan, I. A. Guzei and J. M. Schomaker, Org. Lett., 2016, 18, 3014–3017 CrossRef CAS PubMed; (d) J. M. Alderson, A. M. Phelps, R. J. Scamp, N. S. Dolan and J. M. Schomaker, J. Am. Chem. Soc., 2014, 136, 16720–16723 CrossRef CAS PubMed; (e) J. R. Corbin and J. M. Schomaker, Chem. Commun., 2017, 53, 4346–4349 RSC.
  11. For Ru catalysts: (a) E. Milczek, N. Boudet and S. Blakey, Angew. Chem., Int. Ed., 2008, 47, 6825–6828 CrossRef CAS PubMed; (b) J.-L. Liang, S.-X. Yuan, J.-S. Huang and C.-M. Che, J. Org. Chem., 2004, 69, 3610–3619 CrossRef CAS PubMed; (c) M. E. Harvey, D. G. Musaev and J. Du Bois, J. Am. Chem. Soc., 2011, 133, 17207–17216 CrossRef CAS PubMed.
  12. For Mn catalysts: (a) J. Zhang, P. W. H. Chan and C.-M. Che, Tetrahedron Lett., 2005, 46, 5403–5408 CrossRef CAS; (b) S. M. Paradine, J. R. Griffin, J. Zhao, A. L. Petronico, S. M. Miller and M. C. White, Nat. Chem., 2015, 7, 987–994 CrossRef CAS PubMed.
  13. For Cu catalyst: (a) D. N. Barman and K. M. Nicholas, Eur. J. Org. Chem., 2011, 908–911 CrossRef CAS; (b) R. T. Gephart and T. H. Warren, Organometallics, 2012, 31, 7728–7752 CrossRef CAS.
  14. For Fe catalysts: (a) S. M. Paradine and M. C. White, J. Am. Chem. Soc., 2012, 134, 2036–2039 CrossRef CAS PubMed; (b) Y. Liu, X. Guan, E. L.-M. Wong, P. Liu, J.-S. Huang and C.-M. Che, J. Am. Chem. Soc., 2013, 135, 7194–7204 CrossRef CAS PubMed; (c) L. Zhang and L. Deng, Chin. Sci. Bull., 2012, 57, 2352–2360 CrossRef CAS.
  15. For Co catalysts: (a) H. Lu, H. Jiang, Y. Hu, L. Wojtas and X. P. Zhang, Chem. Sci., 2011, 2, 2361–2366 RSC; (b) H. Lu, K. Lang, H. Jiang, L. Wojtas and X. P. Zhang, Chem. Sci., 2016, 7, 6934–6939 RSC; (c) V. Lyaskovskyy, A. I. O. Suarez, H. Lu, H. Jiang, X. P. Zhang and B. de Bruin, J. Am. Chem. Soc., 2011, 133, 12264–12273 CrossRef CAS PubMed.
  16. (a) R. S. Meléndez and W. D. Lubell, Tetrahedron, 2003, 59, 2581–2616 CrossRef; (b) J. F. Bower, J. Rujirawanich and T. Gallagher, Org. Biomol. Chem., 2010, 8, 1505–1519 RSC; (c) J. M. Alderson and J. M. Schomaker, Chem. – Eur. J., 2017, 23, 8571–8576 CrossRef CAS PubMed.
  17. (a) M. Kim, J. V. Mulcahy, C. G. Espino and J. Du Bois, Org. Lett., 2006, 8, 1073–1076 CrossRef CAS PubMed; (b) G. Grelier, R. Rey-Rodriguez, B. Darses, P. Retailleau and P. Dauban, Eur. J. Org. Chem., 2017, 1880–1883 CrossRef CAS.
  18. For synthesis of 1,3- and 1,2-diamines using Rh catalysts, see: (a) T. Kurokawa, M. Kim and J. Du Bois, Angew. Chem., Int. Ed., 2009, 48, 2777–2779 CrossRef CAS PubMed; (b) D. E. Olson and J. Du Bois, J. Am. Chem. Soc., 2008, 130, 11248–11249 CrossRef CAS PubMed; (c) D. Olson, D. A. Roberts and J. Du Bois, Org. Lett., 2012, 14, 6174–6177 CrossRef CAS PubMed.
  19. For synthesis of cyclic sulfonamides via C–H amination see: (a) P. Dauban, L. Sanière, A. Tarrade and R. H. Dodd, J. Am. Chem. Soc., 2001, 123, 7707–7708 CrossRef CAS PubMed; (b) J.-L. Liang, S.-X. Yuan, P. W. H. Chan and C.-M. Che, Org. Lett., 2002, 4, 4507–4510 CrossRef CAS PubMed; (c) A. Padwa, A. C. Flick, C. A. Leverett and T. Stengel, J. Org. Chem., 2004, 69, 6377–6386 CrossRef CAS PubMed; (d) C. Fruit and P. Müller, Helv. Chim. Acta, 2004, 87, 1607–1615 CrossRef CAS.
  20. For mechanistic investigation on C–H amination using Rh species see: (a) K. W. Fiori, C. G. Espino, B. H. Brodsky and J. Du Bois, Tetrahedron, 2009, 65, 3042–3051 CrossRef CAS; (b) R. H. Perry, T. J. Cahill III, J. L. Roizen, J. Du Bois and R. N. Zare, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 18295–18299 CrossRef CAS PubMed; (c) X. Lin, C. Zhao, C.-M. Che, Z. Ke and D. L. Phillips, Chem. – Asian J., 2012, 2, 1101–1108 CrossRef PubMed; (d) A. Varela-Alvarez, T. Yang, H. Jennings, K. P. Kornecki, S. N. Macmillan, K. M. Lancaster, J. B. C. Mack, J. Du Bois, J. F. Berry and D. G. Musaev, J. Am. Chem. Soc., 2016, 138, 2327–2341 CrossRef CAS PubMed; (e) J. L. Roizen, D. N. Zalatan and J. Du Bois, Angew. Chem., Int. Ed., 2013, 52, 11343–11346 CrossRef CAS PubMed; (f) J. Wang, K. Zheng, B. Lin and Y. Weng, RSC Adv., 2017, 7, 34783–34794 RSC.
  21. For examples of chemoselective aziridination using nitrenes, see: (a) K. Guthikonda, P. M. When, B. J. Caliando and J. Du Bois, Tetrahedron, 2006, 62, 11331–11342 CrossRef CAS; (b) C. Fruit and P. Müller, Tetrahedron: Asymmetry, 2004, 15, 1019–1026 CrossRef CAS.
  22. J. J. Fleming, M. D. McReynolds and J. Du Bois, J. Am. Chem. Soc., 2007, 129, 9964–9975 CrossRef CAS PubMed.
  23. D. N. Zalatan and J. Du Bois, J. Am. Chem. Soc., 2008, 130, 9220–9221 CrossRef CAS PubMed.
  24. K. P. Kornecki and J. F. Berry, Chem. Commun., 2012, 48, 12097–12099 RSC.
  25. For mechanistic investigation of dirhodium-catalysed intramolecular allylic C–H amination versus alkene aziridination see: X. Zhang, H. Xu and C. Zhao, J. Org. Chem., 2014, 79, 9799–9811 CrossRef CAS PubMed.
  26. See for examples: (a) C.-B. Yu, D.-W. Wang and Y.-G. Zhou, J. Org. Chem., 2009, 74, 5633–5635 CrossRef CAS PubMed; (b) V. O. Rogachev, S. Merten, T. Seiser, O. Kataeva and P. Metz, Tetrahedron Lett., 2008, 49, 133–136 CrossRef CAS.
  27. See for examples: (a) A. Casini, A. Scozzafava and C. T. Supuran, Expert Opin. Ther. Pat., 2002, 12, 1307–1327 CrossRef CAS; (b) L. H. Silver, Am. J. Ophthalmol., 1998, 126, 400–408 CrossRef CAS PubMed; (c) B. J. Gates, T. T. Nguyen, S. M. Setter and N. M. Davies, Expert Opin. Pharmacother., 2005, 6, 2117–2140 CrossRef CAS PubMed.
  28. C. J. Hayes, P. W. Beavis and L. A. Humphries, Chem. Commun., 2006, 4501–4502 RSC.
  29. (a) L. A. Boralsky, D. Martson, R. D. Grigg, J. C. Hershberger and J. M. Schomaker, Org. Lett., 2011, 13, 1924–1927 CrossRef CAS PubMed; (b) R. D. Grigg, J. M. Schomaker and V. Timokhin, Tetrahedron, 2011, 67, 4318–4326 CrossRef CAS; (c) J. W. Rigoli, C. D. Weatherly, B. T. Vo, S. Neale, A. R. Meis and J. M. Schomaker, Org. Lett., 2013, 15, 290–293 CrossRef CAS PubMed; (d) J. W. Rigoli, C. D. Weatherly, J. M. Alderson, B. T. Vo and J. M. Schomaker, J. Am. Chem. Soc., 2013, 135, 17238–17241 CrossRef CAS PubMed; (e) C. Weatherly, J. M. Alderson, J. F. Berry, J. E. Hein and J. M. Schomaker, Organometallics, 2017, 36, 1649–1661 CrossRef CAS.
  30. B. M. Trost, B. M. O'Boyle, W. Torres and M. K. Ameriks, Chem. – Eur. J., 2011, 17, 7890–7903 CrossRef CAS PubMed.
  31. (a) P.-A. Nocquet, R. Hensienne, J. Wencel-Delord, E. Wimmer, D. Hazelard and P. Compain, Org. Biomol. Chem., 2015, 13, 9176–9180 RSC; (b) P.-A. Nocquet, R. Hensienne, J. Wencel-Delord, E. Laigre, K. Sidelarbi, F. Becq, C. Norez, D. Hazelard and P. Compain, Org. Biomol. Chem., 2016, 14, 2780–2796 RSC.
  32. For a review on square sugars see: D. Hazelard and P. Compain, Org. Biomol. Chem., 2017, 15, 3806–3827 CAS.
  33. (a) T. Tanino, S. Ichikawa, M. Shiro and A. Matsuda, J. Org. Chem., 2010, 75, 1366–1377 CrossRef CAS PubMed; (b) T. Tanino, S. Ichikawa and A. Matsuda, Org. Lett., 2011, 13, 4028–4031 CrossRef CAS PubMed.
  34. R. Hrdina, F. M. Metz, M. Larrosa, J.-P. Berndt, Y. Y. Zhygadlo, S. Becker and J. Becker, Eur. J. Org. Chem., 2015, 6231–6236 CrossRef CAS.
  35. S. Toumieux, P. Compain, O. R. Martin and M. Selki, Org. Lett., 2006, 8, 4493–4496 CrossRef CAS PubMed.
  36. S. Toumieux, P. Compain and O. R. Martin, J. Org. Chem., 2008, 73, 2155–2162 CrossRef CAS PubMed.
  37. P. Compain, Synlett, 2014, 1215–1240 CrossRef.
  38. S. Toumieux, P. Compain and O. R. Martin, Tetrahedron Lett., 2005, 46, 4731–4735 CrossRef CAS.
  39. F. J. Wyszynski, A. L. Thompson and B. G. Davis, Org. Biomol. Chem., 2010, 8, 4246–4248 CAS.
  40. For another example of C–H amination regioselectivity controlled by hyperconjugaison see: J. D. St. Denis, C. F. Lee and A. K. Yudin, Org. Lett., 2015, 17, 5764–5767 CrossRef CAS PubMed and ref. 35, 38 and 56.
  41. M. Mazurais, C. Lescot, P. Retailleau and P. Dauban, Eur. J. Org. Chem., 2014, 66–79 CrossRef CAS.
  42. K. A. Parker and W. Chang, Org. Lett., 2005, 7, 1785–1788 CrossRef CAS PubMed.
  43. S. Ranatunga and J. R. Del Valle, Tetrahedron Lett., 2009, 50, 2464–2466 CrossRef CAS PubMed.
  44. Y. Liu, W. Xiao, M.-K. Wong and C.-M. Che, Org. Lett., 2007, 9, 4107–4110 CrossRef CAS PubMed.
  45. T. Yakura, S. Sato and Y. Yoshimito, Chem. Pharm. Bull., 2007, 55, 1284–1286 CrossRef CAS PubMed.
  46. An alternative mechanism was proposed for the formation of ketones via C–H amination of allylic carbamate: (a) B. Hurlocker, N. C. Abascal, L. M. Repka, E. Santizo-Deleon, A. L. Smenton, V. Baranov, R. Gupta, S. E. Bernard, S. Chowdhury and C. M. Rojas, J. Org. Chem., 2011, 76, 2240–2244 CrossRef CAS PubMed; (b) R. Bodner, B. K. Marcellino, A. Severino, A. L. Smenton and C. M. Rojas, J. Org. Chem., 2005, 70, 3988–3996 CrossRef CAS PubMed; (c) R. Gupta, K. M. Sogi, S. E. Bernard, J. D. Decatur and C. M. Rojas, Org. Lett., 2009, 11, 1527–1530 CrossRef CAS PubMed.
  47. M. Yamaguchi, M. Hayashi, Y. Hamada and T. Nemoto, Org. Lett., 2016, 18, 2347–2350 CrossRef CAS PubMed.
  48. (a) A. D. Huters, K. W. Quasdorf, E. D. Styduhar and N. K. Garg, J. Am. Chem. Soc., 2011, 133, 15797–15799 CrossRef CAS PubMed; (b) K. W. Quasdorf, A. D. Huters, M. W. Lodewyk, D. J. Tantillo and N. K. Garg, J. Am. Chem. Soc., 2012, 134, 1396–1399 CrossRef CAS PubMed.
  49. N. A. Weires, E. D. Styduhar, E. L. Baker and N. K. Garg, J. Am. Chem. Soc., 2014, 136, 14710–14713 CrossRef CAS PubMed.
  50. E. D. Styduhar, A. D. Huters, N. A. Weires and N. K. Garg, Angew. Chem., Int. Ed., 2013, 52, 12422–12425 CrossRef CAS PubMed.
  51. S. Sato, M. Shibuya, N. Kanoh and Y. Iwabuchi, Chem. Commun., 2009, 6264–6266 RSC.
  52. K. Takahashi, D. Yamaguchi, J. Ishihara and S. Hatakeyama, Org. Lett., 2012, 14, 1644–1647 CrossRef CAS PubMed.
  53. T. Yakura, Y. Yoshimoto and C. Ishida, Chem. Pharm. Bull., 2007, 55, 1385–1389 CrossRef CAS PubMed.
  54. T. Yakura, Y. Yoshimoto, C. Ishida and S. Mabuchi, Synlett, 2006, 930–932 CrossRef CAS.
  55. T. Yakura, Y. Yoshimoto, C. Ishida and S. Mabuchi, Tetrahedron, 2007, 63, 4429–4438 CrossRef CAS.
  56. M. S. T. Morin, S. Toumieux, P. Compain, S. Peyrat and J. Kalinowska-Tlusik, Tetrahedron Lett., 2007, 48, 8531–8535 CrossRef CAS.
  57. For reviews on late stage C–H functionalization see: (a) J. He, L. G. Hamann, H. M. L. Davies and R. E. J. Beckwith, Nat. Commun., 2015, 6, 5943,  DOI:10.1038/ncomms6943; (b) T. Cernak, K. D. Dykstra, S. Tyagarajan, P. Vachal and S. W. Krska, Chem. Soc. Rev., 2016, 45, 546–576 RSC.
  58. R. M. Conrad and J. Du Bois, Org. Lett., 2007, 9, 5465–5468 CrossRef CAS PubMed.
  59. A. Robert, O. Dechy-Cabaret, J. Cazelles and B. Meunier, Acc. Chem. Res., 2002, 35, 167–174 CrossRef CAS PubMed.
  60. J. Li, J. S. Cisar, C.-Y. Zhou, B. Vera, H. Williams, A. D. Rodríguez, B. F. Cravatt and D. Romo, Nat. Chem., 2013, 5, 510–517 CrossRef CAS PubMed.
  61. For related examples on steroids see: S. Yamashita, M. Himuro, Y. Hayashi and M. Hirama, Tetrahedron Lett., 2013, 54, 1307–1308 CrossRef CAS.
  62. K. Yaji and M. Shindo, Tetrahedron Lett., 2010, 51, 5469–5472 CrossRef CAS.
  63. M. Anada, M. Tanaka, N. Shimada, H. Nambu, M. Yamawaki and S. Hashimoto, Tetrahedron, 2009, 65, 3069–3077 CrossRef CAS.
  64. For a related work devoted to asymmetric C–H amination of enol triflates see ref. 41.
  65. For other examples of applications of intermolecular C–H aminations to the synthesis of natural products see: (a) S. P. Lathrop, M. Pompeo, W.-T. T. Chang and M. Movassaghi, J. Am. Chem. Soc., 2016, 138, 7763–7769 CrossRef CAS PubMed; (b) S. P. Lathrop and M. Movassaghi, Chem. Sci., 2014, 5, 333–340 RSC.
  66. D. P. Uccello, S. M. Miller, N. A. Dieterich, A. F. Stepan, S. Chung, K. A. Farley, B. Samas, J. Chen and J. I. Montgomery, Tetrahedron Lett., 2011, 52, 4247–4251 CrossRef CAS.
  67. Amino Group Chemistry: From Synthesis to the Life Sciences, ed. A. Ricci, Wiley-VCH, Weinheim, 2007 Search PubMed.


Present address: COBRA UMR 6014-Equipe Hétérocycles, Bâtiment IRCOF, 1 Rue Tesnière, 76821 Mont St Aignan Cedex, France.

This journal is © the Partner Organisations 2017