If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Amino acids and peptides play an important role in nature as well as in the pharmaceutical industry. Therefore, transformation of amino acids and peptides into their unnatural derivatives via step and atom economical ways is highly demanding. Although, a number of methods exist for selective reactions in proximal α- or β-C(sp3)-H bond, selective functionalization of distal bonds in amino acids remains a challenge. In recent years, transition metal catalyzed C–H functionalization has come up as a superior tool for C–H bond transformations. In this review, we intend to provide the reader a detailed overview of the current state of art in the utilization of site selective C–H functionalization technology for transformation of distal C–H bonds of amino acids and peptides. Various methods and synthetic tools have been discussed to deliver a broader idea about the field. A systematic compilation has been presented here to provide insights into the recent developments and future challenges in the field.
] Although nature only utilizes a small set of amino acids for the synthesis of peptides and proteins, the option of building innovative structures that have improved properties are unlimited with unnatural amino acids. Recent years have witnessed an upsurge of bioactive unnatural amino acids [
]. These unnatural amino acids are extremely valuable candidates for the synthesis of drug molecules, as they often exhibit superior bioactivity and pharmacokinetics compared to their parent natural derivative [
]. From a synthesis point of view, multiple strategies have been applied to synthesize unnatural amino acids in the last few decades. On the other hand, the modification of side chains of readily available amino acids and peptides has also come up as a fruitful approach. Traditionally, functional group transformations and cross coupling methodologies have been extensively applied for side chain modification of amino acids [
]. Nonetheless, these approaches have some inherent problems such as need of prefunctionalized substrates, generation of unnecessary waste, etc. In recent years, transition metal catalyzed direct C–H functionalization has emerged as one of the alternative approaches for amino acids side chain modification. This approach offers significant advantages over other traditional approaches to transform amino acids side chains like: 1) shorter reaction routes as no prefunctionalized substrates are necessary and therefore offers better step and atom economy, 2) Utilization of chiral α-amino acids as chiral pool of substrates 3) late-stage functionalization of complex peptides and proteins, etc. For the same reason, various metal catalyzed C–H activation reactions have been extensively utilised in the past decade to transform the amino acids chains. Notably, the majority of reactions performed on amino acids were restricted to proximal β-C(sp3)-H bonds of amino acids [
]. This is due to the intricacies involved in activating the distal C(sp3)-H bonds in amino acids. Activation of a β-C(sp3)-H bond in amino acids requires intermediacy of a five membered metallacycle that is stable, whereas, the activation of γ- or δ-C(sp3)-H bond requires the formation of thermodynamically less stable six membered metallacycle. This intrinsic difficulty in forming higher membered metallacycles presents difficulty in reaching out to distal positions of amino acids that has γ- or δ-C(sp3)-H bonds. In spite of this challenge, researchers in this area are continuously designing new strategies to for functionalization of C(sp3)–H bonds of amino acids beyond proximity. Multiple review and accounts have been already written to describe the state of art in functionalization of proximal α- and β-C(sp3)-H bonds of amino acids [
]. In this highlight we intend to provide a deep insight into the current scenario in transforming the distal C(sp3)–H bonds of amino acids (Fig. 1). The discussion has been divided into two sections as γ-C(sp3)–H functionalization and δ-C(sp3)–H functionalization in amino acids. The concepts and details presented in the highlight are expected to provide the readers with fruitful insights of this field.
2. Distal C(sp3)-H functionalizations of amino acid derivatives
2.1 Gamma functionalization of amino acid derivatives
A number of natural amino acids such as leucine, isoleucine, valine, etc. and many unnatural amino acids contains γ-C(sp3)–H bonds in their side chain. Transformation of their γ-C-H bonds have been the subject of numerous synthetic strategies. The various strategies applied and transformations achieved by activating the γ-C(sp3)–H bond in amino acids and peptides are described below. The sections are divided in the order of various functionalizations achieved in distal locations of amino acids.
In 2006, Corey and colleagues first discovered a novel method to carry out arylation at the γ-C(sp3)-position of phthalimide-protected α-amino acids using a N-linked directing group 8-aminoquinoline and a palladium catalyst [
]. They achieved this by using reaction parameters similar to β-arylation which was described in the same report, but the reaction time was increased to 2.5 h. Under the reaction conditions, the isoleucine derivative underwent arylation with 87% yield (Scheme 1). Another amino acid tert-Leucine was also converted into a mixture of mono- and di-arylated products. The substrate scope was further extended to include a valine derivative, which provided yield of 85%.
Chen and co-workers have done some pioneering work by employing picolinamide (PA) as DG for multifold Pd-catalyzed C(sp3)-H functionalization reactions. In their first report in 2011, they demonstrated γ-C(sp3)-H arylation of a number of substrates such as cyclohexyl amino esters, threonine methyl ester, and isoleucine derivatives [
]. Interestingly, they were able to achieve this transformation under milder reaction conditions as compared to the Daugulis group. A number of aryl iodides were compatible with reaction (Scheme 2). Various electronic substitutions such as -OMe, o-TIPS, –NO2, tosyl amide, and –CO2Me groups were incorporated in good yields. A free phenol group gave the product in moderate yield but pyridine substrate failed to provide any arylated product. The presence of an electron-withdrawing substituent on the pyridine ring of PA led to a decrease in the arylation yield, whereas, the presence of an electron-donating substituent increased the yield. Selectively monoarylated 6-membered cyclic amino esters were obtained in all cases and no bisarylated products were observed. Furthermore, threonine methyl ester cleanly underwent arylation with the ortho-substituted aryl iodide, affording the product in a moderate yield. The isoleucine derived substrate predominantly provided the monoarylated product, with the minor bisarylated product formed on prolonging reaction time at 60 °C. Although the exact mechanism remains elusive, it is believed that this reaction proceeds via a Pd(II)/Pd(IV) catalytic cycle, consisting of sequential C–H activation/Ar–I coupling pathway. A threonine derivative was utilised as starting product to carry out the synthesis of a natural product, (+)-obafluorin, which utilised a modified DG and hence, was easily removable under in a mildly acidic environment.
In 2012, an efficient method for carrying out Pd-catalyzed γ-C(sp3)-H arylation of a number of amino acid derivatives using the N-(2-pyridyl)sulfonyl auxiliary was showcased by Carretero and co-workers (Scheme 3) [
]. A variety of amino acid derivatives including α-quaternary amino acids and β-amino acids were chosen as substrates for this reaction and γ-monoarylated products were obtained with very high diastereoselectivities. Predominant bisarylation occurred on increasing the amount of aryl iodide from 2.5 equivalents to 5 equivalents. The derivatives of valine, allo-isoleucine, protected threonine, homoalanine, and α-quaternary amino acid afforded the monoarylated product in good yields, whereas, the isoleucine derivative gave a significantly lower yield. Both electron withdrawing and electron donating substituents on the aryl ring worked well in the reaction. However, changing the auxiliary to p-tolylsulfonamide or N-methyl sulfonamide failed to give any product at all. Similarly, alanine and leucine (lacking the γ-methyl group) proved to be completely unreactive as substrates under the standard reaction condition. Mechanistic studies indicated the formation of a bimetallic five-membered palladacycle intermediate, wherein the PdII atom is coordinated to the sulfonamide nitrogen, γ-carbon of the substrate, as well as the pyridine nitrogen, and one of the sulfonylic oxygens. The N-(2-pyridyl) sulfonyl auxiliary was removed on treatment with Zn/NH4Cl after completion of the arylation process providing as high as 98% ee.
The Ma group in 2013 disclosed a different protocol for the γ-C(sp3)-H arylation of various 2- aminobutanoic acid substrates, using a covalently tethered 2-methoxyiminoacetyl (MIA) auxiliary [
]. Using Pd(OAc)2 as the catalyst and AgOAc as the silver salt, they observed the best yield could be obtained by using hexafluoroisopropanol (HFIP) as the solvent and PivOH as an additive (Scheme 4). The simplest unsubstituted aminobutanoic acid derivative provided a highly enantiopure product (99% ee) indicating that racemization did not take place during the reaction. A range of aryl iodides were screened using the unsubstituted aminobutanoic acid derivative and the arylated products were obtained in moderate to good yields (54–88%). However, better yields were provided by ortho-substituted aryl iodides as compared to their para- and meta-counterparts, probably because of the additional binding to palladium complexes provided by ester and nitro groups. It was found that the derivatives of D-allo-isoleucine and α-quaternary amino acid are compatible with the reaction but the threonine derivative did not furnish better yield, presumable due to steric factors. Interestingly, the l-valine derivative was able to produce a γ-γ’ bisarylated product in a high enantiomeric excess along with the monoarylated product. The derivatives of methyl 2-aminopentanoate and methyl 2-amino-4-methylpentanoate proved to be unsuccessful substrates indicating that γ-arylation of the methylene unit and δ-arylation of the terminal –CH3 group cannot be carried out under these reaction condition. It is hypothesized that a double five-membered palladium chelate is formed upon C–H insertion that undergoes oxidative addition to the aryl iodide followed by reductive elimination to deliver the product. The MIA auxiliary could be easily removed on treatment with KOH and the resulting amino acids were protected using (Boc)2O, providing excellent yields and enantiomeric excess. Some of these products are known precursors in the synthesis of valuable bioactive molecules. The authors also emphasized upon the synthetic utility of the MIA auxiliary by transforming it into a glycine moiety that could be utilised for further transformations.
In 2014, Chen and co-workers devised a Pd-catalyzed γ-C(sp3)-H bond arylation strategy to synthesize an important natural product, Hibispeptin A [
]. A number of directing groups were tested for this protocol such as picolinamide (PA), pyridyl sulfonamide (PS), aminoquinoline (AQ), pyridyl methylamine (PM), 2-pyridyl ethylamine (PE), and gem-dimethyl substituted pyridyl methylamine. Finally, a pyridyl methylamine-based directing group (PR) was found to be best for this synthesis. The PR-protected Isoleucine derivative was arylated at the γ-methyl position by using a sterically hindered aryl iodide in one of the steps of this total synthesis scheme.
In 2014, Yu and colleagues reported a Pd-catalyzed cross coupling reaction, wherein, the γ-C(sp3)-H bonds of triflyl protected aliphatic amino acids were arylated (Scheme 5) [
]. Arylboron compounds were utilizedas arylation source. The authors postulated the role of mono-protected amino acid (MPAA) ligand is to acceletate the reaction. It was suggested that the triflimide could form an imidate-like moiety as a weakly-coordinating σ-donor when deprotonated under basic conditions which then with the help of the MPAA ligand could accelerate the reaction. The importance of the ligand was established as the reaction did not proceed in its absence. Evaluation of the MPAA ligands revealed that the l-amino ester provided higher yields when D-enantiomers of the MPAA ligands were used. The ligand Ac-D-tLeu-OH was found to give the best yields when utilised with Pd(OTf)2(MeCN)4 as catalyst. Ester groups at the meta- and para-positions of the phenyl ring of the arylboron reagents provided moderate to very good yields (64–82%). Fluorinated and trifluoromethylated aryl rings were also compatible with the protocol provided yields in the range of 60–73%. However, arylboronates containing furan-, pyridine, and pyrazole-type motifs turned out to be ineffective in the reaction. Screening of amino acids revealed that valine and t-leucine derivatives could be arylated to produce a mixture of mono- and bisarylated products in excellent yields. However, the isoleucine derivative provided only a moderate yield under the same conditions.
In 2015, the Liu group came up with an excellent strategy for Pd-catalyzed selective γ-C(sp3)-H arylation of α-aminobutanoic acid derivatives using 5-methylisoxazole-3-carboxamide (MICA) as the auxiliary [
]. They tested various auxiliaries and found that the methyl group at the 5-position of the isoxazole ring was extremely important, presumably because it provides stability and a balanced electron distribution to the MICA auxiliary when coordinating with the metal center. Toluene was found to be the ideal solvent for the reaction. A wide range of aryl iodides with substituents such as methyl-, methoxy-, halogens, and nitro-afforded the arylated products in good to excellent yields (67–93%). Bulky aryl iodides such as naphthalene iodide and pinacoborane-bearing phenyl iodide were also well-tolerated in this protocol. The scope of the amino acid derivatives were extended to MICA-protected l-Valine, l-Isoleucine and l-Threonine methyl esters which provided excellent yields (Scheme 6). It is worth mentioning that the l-Valine derivative predominantly formed monoarylated product in good yields but also produced 15% of bisarylated product under mild conditions (60 °C, 36 h). It is presumed that coordination by the isoxazole moiety promotes the formation of a less sterically hindered trans-palladacycle that undergoes further transformation to give the product. The auxiliary used here could be easily removed under mild conditions.
In 2016, the Yu group thoroughly investigated the influence of ligand structure on reactivity by carrying out Pd-catalyzed γ-C(sp3)-H arylation of natural amino acids by using quinoline based ligands combined with a weakly coordinating amide directing group (CONHArF, ArF = 2,3,5,6 tetrafluoro-4-(trifluoromethyl)phenyl) (Scheme 7) [
]. Ligands such as pyridine, quinoline, acridine, and, phenanthroline derivatives were screened and electron-rich tricyclic quinolines provided the highest yields. The reaction was carried out in an environment of molecular oxygen which is elemental to the reactivity. While using amino acids as substrates, the ligand/palladium ratio was maintained at 1:1 and an electron-donating tert-butyl substituent was tethered to the 7-position of the tricyclic quinoline ligand. These parameters, coupled with AgO2CnBu as the additive, afforded the maximum yield. Under these conditions, a variety of aryl iodides were compatible with the reaction with a valine derivative as the substrate. Aryl iodides containing methyl- and methoxy-substituents produced good to excellent yields, while substituents such as halides, trifluoromethyl, ester, ketone, and acetate afforded moderate to good yields. Isoleucine was successfully arylated with 85% yield and tert-leucine produced the bisarylated product with 44% yield. The high diastereoselectivity (dr > 20:1) is worth mentioning due to the highly favoured trans-conformation of the methyl group with respect to the phthalimide group.
Our group in 2017 devised a strategy to arylate the γ-C(sp3)-H bonds of N-protected amino acids using a Pd catalyst and 8-aminoquinoline as a bidentate chelating auxiliary (Scheme 8) [
]. Here, the amine and quinoline nitrogens both coordinate to the metal center. Mechanistic studies indicated that the reaction proceeds via a 6-membered cyclometallation and this presents the need for a donor group to stabilize the less favourable metallacycle intermediate. Derivatives of l-Valine and l-Isoleucine underwent arylation and provided good to excellent yields despite the absence of β-hydrogen. Aryl iodides containing both electron-donating and electron-withdrawing substituents provided monoarylated products with excellent diastereomeric ratio.
Later in the same year, the Yu group demonstrated a Pd-catalyzed and ligand-accelerated γ-C(sp3)-H cross-coupling of Nosyl-protected α-amino acids using arylboron reagents (Scheme 9) [
]. Out of all the screened ligands, an acetyl-protected aminomethyl oxazoline (APAO) ligand was found to be most compatible with the reaction. Both electron-donating and electron-withdrawing substituents on the aryl ring of Ar-Bpin were amenable to the protocol as was the unsubstituted phenylboronate ester, provided good yields and extremely high diastereoselectivities. Further, 2-naphthylboronic acid pinacol ester, meta-substituted and di-substituted arylboron reagents, and heteroaryl boronate ester were also able to furnish good yields. As far as the scope of the amino acid derivatives is concerned, L-tert-Leucine and L-allo-Isoleucine derivatives were amenable to these reaction conditions and the former also produced some diarylated product. Considerably lesser yield of the monoarylated product was given by the l-Isoleucine derivative, however, it formed a minor amount of δ-arylated product as well. The simplest unsubstituted aminobutanoic acid could afford only 30% yield, presumably because of the absence of Thorpe-Ingold effect. The l-Valine benzyl ester derivative was also reactive, but more sterically hindered substrates gave considerably less yields, owing to repulsion from the APAO ligand. Interestingly, the effect of the disparity between the chiral carbon center of the amino acid derivative and the chiral ligand was demonstrated when the d-Valine derivative gave only 11% yield, despite its l-Valine counterpart displaying excellent reactivity. This protocol was further applied to synthesis of novel bioactive peptides.
In 2018, Jana and colleagues highlighted that a number of C(sp3)-H activation strategies suffer from harsh reaction parameters, use of expensive ligands/directing groups, and use of toxic and super-stoichiometric silver salts. To overcome this problem, they came up with a protocol for γ-C(sp3)-H arylation of amino acid derivatives utilising a cheap and commercially available 1,10-phenanthroline ligand, easily obtained and convenient to remove picolinamide auxiliary, and lastly the earth-abundant and economical Mn(OAc)3 salt (Scheme 10) [
]. The idea was based on promoting Pd(II)-catalyzed γ-C(sp3)-H arylation through the formation of a stable five-membered palladacycle via N-linked directing group such as picolinamide. PA-protected valine methyl ester underwent reaction in presence of various ligands and it was observed that the “conformational restriction” of 1,10-phenanthroline ligand was beneficial for this method. Evaluation of aryl iodides revealed that both electron-donating and electron-withdrawing substituents on the aryl ring were compatible with the reaction. In particular, methoxy-, keto-, nitro-, and trifluoromethyl-substituted aryl iodides afforded good yields of the monoarylated product. Notably, electron rich aryl iodides were more reactive in comparison to electron-deficient aryl iodides. The derivatives of D-2-aminobutyric acid and l-Isoleucine afforded good to moderate yields. Further, monoarylated products underwent further arylation to produce unsymmetrical diarylated products in good to moderate yields. Although the exact mechanism is elusive, preliminary mechanistic studies indicate that aryl iodide could undergo oxidative addition via a single-electron transfer mechanism, leading to the formation of a palladacycle intermediate.
In 2019, the Yao group explored free amino group directed, Pd-catalyzed γ-C(sp3)-H arylation of α-amino esters mediated by diaryliodonium triflates [
]. Studies on the model substrate alanine ethyl ester revealed that Ph2IOTf worked best as the arylating agent, combined with Ag2O as the oxidant, Pd(OAc)2 as catalyst, and HFIP/AcOH mixture as the solvent system (Scheme 11). A number of derivatives of α-alkyl-α-ethyl glycine ethyl esters were successfully arylated using Ph2IOTf, giving respectable yields (46–72%). Linear, branched, and cyclic alkyl groups were well tolerated. γ-aryl-, ethoxycarbonyl- and methoxyl-functional groups were amenable as well. α-methylated derivatives of Valine and Isoleucine provided good to moderate yields of the arylated product. It is worth noting that the α-tertiary carbon center is necessary for this protocol, since the simplest unsubstituted amino ester failed to give any product at all. Both electron-withdrawing and electron-donating aryl groups worked well as arylating agents. Methoxyl-, alkyl-, trifluoromethyl-, halides, methoxycarbonyl-, and cyano-groups were tolerated as substituents on the aryl ring. Unsymmetrical diaryliodonium salts were compatible as well. Here the selectivity was found to be controlled by steric factors and not the electronic properties of the arylating partners. Presence of kinetic isotope effect indicated that the C–H bond breakage process could be the turnover-determining step. Mechanistic studies show that an alkyl-Pd(II) intermediate C was generated during the reaction, which underwent oxidative addition to Ar2IOTf, followed by anion exchange, forming complex D. Further reductive elimination finally provided the desired product 2 (Scheme 12).
In the same year, Shi and co-workers reported a γ-C(sp3)-H arylation of the sterically hindered tert-Leucine derivatives by a ligand mediated pathway [
]. No external directing group was employed and the protocol was directed by the weakly coordinating carboxylate moiety of amino acids (Scheme 13). Phthalimide protected tert-Leucine derivatives underwent reaction with various aryl iodides, in presence of Ac-Tle-OH as the ligand, and Ag3PO4 as the halide scavenger. It is also hypothesized that Ag3PO4 could act as a heteronuclear active species to promote the cleavage of the C–H bond. Both electron-withdrawing and electron-donating substituents on the aryl iodide were found to favour the reaction, producing good yields of the monoarylated product. Acetyl-, ethoxycarbonyl- and formyl groups tethered to para or meta position of the aryl iodide resulted in very good yields. Aryl iodides with multiple substituents also gave the arylated product in moderate yields. The substrate when reacted with 1-iodo-4-methoxybenzene, generated 15% of the diarylated product as well. A gram-scale synthesis of monoarylated tert-Leucine derivative was also carried out to showcase the chiral integrity of the product remains intact in this protocol.
In another report by the Yao group in 2019, they put forward the idea that if the metallacycle intermediate formed during the reaction could undergo anion exchange with a suitable anion, it could help to facilitate the reactivity of α-amino esters towards γ-C(sp3)-H arylation. Later, they demonstrated a novel method for free amino group directed site selective γ-C(sp3)-H arylation of α-amino acids using aryl iodides (Scheme 14) [
]. The optimized reaction parameters for ethyl L-tert-Leucinate as substrate include Pd(OAc)2 as the catalyst, AgOAc as the anion source, and TfOH in HFIP/AcOH as the solvent system. The scope of aryl iodides for the same substrate revealed that both electron-rich and electron-deficient aryl iodide species are comfortably incorporated, producing the γ-arylated products in moderate to good yields (60–84%). Ortho-, meta- and para-substituted aryl iodides all gave similar yields, signifying that steric factors does not play a role in the reactivity. Evaluation of various α-amino esters showed that the ethyl 2-aminobutanoate moiety underwent predominant racemization but formed only 37% of the product. Both ethyl l-Valinate and l-Isoleucinate generated products in moderate yields (50–58%), along with a moderate diastereomeric ratio. Notably, L-tert-Leucine was found to be unreactive, probably because of the formation of inactive η [
]. Picolinamide was employed as the auxiliary for this reaction. Since Ag+ ions are known for their I− scavenging ability, Ag2CO3 was used as an Ag + ion source. Organic phosphate additives were surveyed and a catalytic amount of (BnO)2PO2H was employed as a Phase Transfer Catalyst (PTC) to prevent the decomposition of the electrophile due to high concentration of Ag + ions. MeI and α-iodoacetic ester were found to be the best alkylating agents that afforded the products in high yields (Scheme 15 A). With MeI in toluene/t-amylOH, using NaI led to an increase in the yield. In case of α-iodoacetic ester in t-amylOH, the initial yields were moderate to high which got reduced considerably on addition of CuCl2 (Scheme 15 B). It was observed that substrates such as protected threonine, allo-isoleucine, and β-homothreonine furnished the alkylated products in good to excellent yields. Further, the alkylated threonine derivative could be transformed into β-hydroxylated amino acids that are an integral component of various natural products. Although the exact mechanism remains unclear, the formation of a palladacycle intermediate via a concerted palladation/deprotonation mechanism, followed by oxidative addition via an SN2 mechanism has been suggested by the authors.
Later in the same year, Shi and co-workers reported the Pd-catalyzed γ-C(sp3)-H alkylation of a Valine derivative, using milder conditions as compared to the Chen group (Scheme 16). The directing group used in this case was 8-aminoquinoline. α-bromoacetate ester and α-iodoacetate ester were used as alkylating agents and (BnO)2PO2H was used as the additive. However, this protocol failed to generate good yields and only a single example of γ-alkylation has been showcased in the report [
]. The standard conditions for the alkylation protocol were kept same as that of arylation and alkyl iodides were utilised as alkylating agents. Isoleucine-, threonine- and valine-derived esters underwent the reaction very comfortably with ethyl iodoacetate, furnishing excellent yields of the mono-alkylated product (Scheme 17). Using this protocol, MICA-protected tert-butyl ester of threonine was transformed into a natural product, (−)-banalol.
In 2017, the Yu group outlined a scheme for the Pd-catalyzed, ligand-accelerated γ-C(sp3)-H alkylation of Nosyl-protected α-amino acids, along with the previously summarized arylation protocol [
]. Alkylboron reagents such as potassium trifluoroborate salts (R–BF3K) and Li2CO3 were used to promote the reaction (Scheme 18). Contrary to the case of arylation, the APAO (acetyl-protected aminomethyl oxazoline) ligands were found to be relatively very less effective in case of alkylation. Hence, MPAA (mono protected amino acid) ligands were screened and N-acetyl-d-Valine was found to be the best ligand for this protocol. Scope of the alkylboron reagents was investigated and tert-Leucine was chosen as the substrate for the screening. Alkylating partners bearing simple and linear alkyl chains were comfortably incorporated to substrates with good to excellent yields. Notably, cyclobutylmethyl boron reagent was furnished good yield. Alkyl moieties containing a number of functional groups such as ester, phenyl-, trifluoromethyl-, chloro-, and acetoxy-were amenable to the reaction, giving respectable yields. The alkylated products bearing these functional groups could be utilised further for a variety of synthetic transformations.
2.1.3 Alkenylation. Olefination reactions constitute an important part of synthetic organic chemistry. However, owing to the inert nature of C(sp3)-H bonds, only a few example of directed γ-C(sp3)-H alkenylation reactions have been reported in literature. In 2011, the Chen group carried out a Pd-catalyzed γ-C(sp3)-H alkenylation of picolinamide-directed six-membered cyclic amino esters using various disubstituted cyclic vinyl iodides, which yielded the alkenylated product in moderate to good yields (Scheme 19) [
]. Notably, the reaction parameters were differed from that of arylation which was also demonstrated in the same report. The silver salt used was AgOAc, contrary to Ag2CO3 in case of arylation. It was observed that the ring size of the alkenyl iodides played an important role, since the seven-membered alkenyl iodide was found to give the greatest yield. Interestingly, some improvement in the yields were observed when benzoquinone was added, although its role is still under investigation. Through mechanistic studies, it has been hypothesized that a palladacycle intermediate could be formed on the insertion of the vinyl iodide into the Pd–C bond that could lead to the formation of the desired product on trans β-halogen elimination.
Later in 2016, Yu and colleagues showcased Pd-catalyzed γ-C(sp3)-H olefination of a number of triflyl- (Tf) and 4-nitrobenzenesulfonyl- (Ns) protected α-amino ester substrates and emphasized on the importance of quinoline or pyridine-based ligands for this transformation (Scheme 20) [
]. Detailed analysis revealed that the olefinated species further went through an intramolecular aza-Wacker oxidative cyclization to form pyrrolidines. Olefins such as electron-deficient alkenes and styrenes were employed as coupling partners. Electron-deficient ligands proved to be beneficial for this reaction and 3,4-bis(trifluoromethyl)pyridine was found to furnish the best yield. Tf-protected l-Valine, L-tert-Leucine and L-allo-Isoleucine derivatives gave the corresponding products in extremely good yields. However, the l-Isoleucine derivative afforded a considerably lower yield of 35%. Protected l-threonine ester derivatives underwent this protocol as well, giving moderate yields. Notably, the γ-methyl C(sp3)-H bond was selectively activated with a very good yield when both γ-methyl and γ-methylene C(sp3)-H bonds were present in the substrate. Using L-tert-Leucine as the model substrate, the screening of olefins was carried out. Acrylates were found to give the best yields (91–93%), followed by styrenes which provided the olefinated product in moderate yields. Similarly, the Ns-protected L-tert-Leucine substrate also afforded very good yields with acrylates and moderate to good yields with coupling agents such as methyl vinyl ketone and acrylonitrile. The protecting groups could be comfortably removed as well.
A method to activate the γ-C(sp3)-H bonds of phthalimide-protected amino acid derivatives using 8-aminoquinoline as the directing group was introduced by our group in 2017 (Scheme 21) [
]. The protocol was mediated by a palladium catalyst and alkenyl iodides were used as olefinating agents. AgOAc and toluene played crucial roles as the oxidant and solvent respectively. The phthalimide-protected Valine derivative produced the best yield on reaction with benzyl trans-3-iodoacrylate, giving an excellent diastereomeric ratio as well. However, it also formed a negligible amount of cyclized product. The same Valine derivative also successfully underwent reactions with cyclic ketoiodides which are electron deficient in nature. The products were obtained in poor to moderate yields, with good diastereomeric ratios, and with retention of stereochemistry. However, ethyl cis-iodoacrylate afforded a mixture of stereoisomers in a poor yield which suggests that a Heck-type Pd-alkyl intermediate might be formed in the course of the reaction. Various olefinating partners such as hexenyl iodides and 6-membered cyclic alkenyl iodides were amenable to the reaction, producing good yields with Valine as substrate. Benzyl protected 3-Iodo-2-propen-1-ol underwent reaction with both Valine and L-tert-Leucine. Nonetheless, dialkenylated product turned out to be the major product for the latter. Selective functionalization at the γ-CH3 position occurred for the Isoleucine substrate, hinting at the possibility that the bis-alkenylated product could not form due to steric factors. A plausible reaction mechanism presented in the report suggests the formation of a 6-membered palladacycle intermediate B (upon C–H activation) with the help of 8-aminoquinoline as the directing group, followed by coordination of the olefin with the intermediate, forming complex C. Subsequent migratory insertion results in the formation of intermediate D and finally β-hydride elimination generates the final product (Scheme 22).
2.1.3 Intramolecular amination
Having extensively worked on picolinamide directed C(sp3)-H functionalization, Chen group sought to diversify the utility of this method beyond C–C bond formation. In 2012, they came up with an efficient strategy to transform γ-C(sp3)-H bonds of various amine and amino acid derivatives into C–N bonds [
]. Using a palladium catalyst and PhI(OAc)2 as the oxidant, they synthesized a range of azetidines from amine and amino acid derivatives (Scheme 23). The reaction could be successfully carried out in an Ar atmosphere and the presence of AcOH further increased the yield. Amino acid derivatives containing a β-substituent such as L-tert-Leucine and Threonine methyl esters afforded the cyclized product in excellent yields (79–91%). However, the simplest unsubstituted ester derivative of 2-aminobutanoic acid furnished only 25% of the cyclized azetidine product as the major product in this case was the γ-acetoxylated product. The formation of the seemingly unfavourable cyclized product could be facilitated by the torsional strain imposed by the β-substituent on the β-position of the substrate and the neighbouring γ-C-H bond during the process of the out-of-palladacycle-plane C–O formation. This strain might be enabling the C–N bond formation over C–O bond formation even though the resulting azetidine product is ring-strained. They also suggested that a PdIV intermediate could be formed and the final product could be obtained after subsequent reductive elimination.
In 2013, the first example of a Pd-catalyzed 1,2,3-triazole-directed intramolecular amination of γ-C(sp3)-H bonds of amino acid derivatives was reported by the Shi group (Scheme 24) [
]. Although the 1,2,3-triazole moiety is highly electron-deficient, but it was found to perform better than other hetero aromatic compounds such as imidazole, furan, and pyrazole. The reaction was carried out in presence of Pd(OAc)2 as catalyst, PhI(OAc)2 as the oxidant, and DCE as the solvent. This work reported only two examples of amino acid derivatives as substrates, and in both cases cyclized azetidines were obtained as the major products in moderate to very good yields and diastereomeric ratios.
Later in the same year, the Chen group synthesized pyrrolidones via a Pd-catalyzed carboxamide-directed intramolecular amination of γ-C(sp3)-H bonds of amines and amino acid derivatives [
]. They utilised two types of carboxamide-based directing groups: 8-aminoquinoline (AQ) group and 2-pyridylmethyl amine (PM) group. Pd(OAc)2 was employed as the catalyst, PhI(OAc)2 as the oxidant in toluene solvent to generate γ-lactams as products (Scheme 25). An atmosphere of Ar was preferred for this reaction. The AQ-coupled Isoleucine substrate was smoothly cyclized, producing the γ-lactam in 86% yield. The same substrate when coupled with the PM group, provided an identical yield of the cyclized product, albeit at a slightly higher temperature. The ester derivative of Valine gave the cyclized product with both the directing groups in excellent yields and high diastereoselectivities. Other amino acid derivatives containing a β-substituent such as the derivatives of threonine and tert-leucine furnished excellent yields with both the directing groups. The substrates lacking a β-substituent did not give any cyclized product at all. The authors predicted that a Pd(II)/Pd(IV) cycle is operative in this case. To increase the practicality of this transformation, a modified directing group, 8-amino-5-methoxyquinoline (MQ) was used since it was found to be easily removable under mild conditions.
While working on developing a strategy for C–H amination reaction, the Chen group had reported the formation of an acetoxylated side product [
]. In 2012, they explored the possibility of obtaining the acetoxylated product as the sole product by carrying out the reaction in AcOH, omitting the use of toluene entirely. They were successful in their attempt and further noted that the concentration of OAc ligands in the reaction mixture affects the reaction pathway and hence has a direct effect on product distribution as well. This indicated that the OAc ligand could dissociate from the PdIV intermediate and subsequently the resulting intermediate could be attacked by a nucleophilic species such as –OR. Finally, the alkoxylated product could form after the reductive elimination of C-OR [
]. To test this hypothesis, they performed the reaction by replacing toluene with MeOH and they were successful in obtaining the desired alkoxylated product (Scheme 26). A mixture of alcohol and xylene turned out to be the best medium after optimization for γ-C(sp3)-H bonds of amino acid derivatives. A number of amino acid derivatives were successfully methoxylated with excellent yields, including picolinamide protected ester derivatives of 2-aminobutanoic acid, Valine, and Isoleucine. The Valine derivative also produced a bis-methylated product. The ester derivative of 2-aminobutanoic acid was also alkoxylated using 2-chloroethanol and tert-butyl alcohol, giving excellent yields.
In synthetic chemistry, strategies to build C–B bonds have always been actively sought after by chemists. In 2014, Shi and co-workers introduced a Pd-catalyzed and picolinamide-directed protocol to produce alkyl boronic esters from amino acid derivatives by the activation of γ-C(sp3)-H bonds [
]. Pd(OAc)2 was employed as the catalyst and bis(pinacolato)diboron (B2pin2) was used as the borylating agent (Scheme 27). Moreover, an oxidant effective enough to prevent the oxidation of the aliphatic C–B bond formed after C–H functionalization was essential for the reaction to work. An alkaline environment was found to be indispensable for stabilizing the C–B bond and to complete the catalytic cycle. iPr2S worked well as an exogeneous ligand and helped in increasing the yield. The reaction worked well under an atmosphere of O2. In case of the model substrate Valine, the borylated product was obtained in 68% yield with a diastereomeric ratio of 83:17. Ester derivatives of Isoleucine, l-threonine, and tert-Leucine also provided very good yields (61–84%) and borylation occurred selectively at the primary γ-C(sp3) position in case of Isoleucine. Moreover, the configuration of the starting material in case of Isoleucine did not affect the yield. It was suggested that this protocol could proceed via two fused palladacycles as intermediates. Both Pd0/PdII and PdII/PdIV catalytic cycles were predicted to be viable pathways for this reaction.
2.1.6 Silylation and germanylation
Organosilicon moieties are integral components of various therapeutic agents and the addition of a silyl group into amino acids can significantly alter their chemical properties. However, due to the steric hindrance inflicted by the presence of a trimethylsilyl-group, intermolecular C(sp3)-H silylation is quite challenging. In 2017, we explored transition-metal catalyzed γ-C(sp3)-H silylation and germanylation of amino acid derivatives and other aliphatic carboxylic acids (Scheme 28) [
]. A directing group approach was deemed necessary for this protocol and the bidentate directing group 8-aminoquinoline was found to work well. Pd(OPiv)2 was employed as the catalyst, (SiMe3)2 as the silylating agent and the reaction worked well in the presence of Ag2CO3 and NaHCO3 in tBuOH or dioxane. Various exogeneous ligands were also screened and the best yield was obtained on utilising 2-chloroquinoline as the ligand. Amino acids such as l-Valine, l-Isoleucine, and tert-butyl Leucine were compatible with the reaction conditions and gave the γ-silylated products in good yields (63–70%) and high diastereoselectivities. Notably, l-Isoleucine and tert-butyl Leucine were selectively silylated at the primary γ-C(sp3) position. When (GeMe3)2 was used instead of (SiMe3)2 under the previous reaction parameters, the corresponding γ-germanylated products were afforded by the same amino acid substrates in good yields (59–69%) and excellent diastereoselectivities.
In 2015, Zhao and co-workers came up with a novel strategy for the carbonylation of γ-C(sp3)-H bonds of aliphatic amine and amino acid substrates to generate pyrrolidones as products [
]. For this process, they developed a bidentate directing group oxalyl amide. Pd(OAc)2 was utilised as the catalyst. 3-(trifluoromethyl)benzoic acid was found to be the best additive, presumably because the electron-deficient arene was stabilizing the Pd(0) species and could further lead to its re-oxidation to Pd(II). AgOAc was established as the most suitable oxidant. Other directing groups such as Cbz, triflamide, and picolinamide could afford only a trace yield of the product. A number of amino acid derivatives were included in the substrate scope for this reaction. It was found that presence of a β-substituent in the substrate greatly aids the yield. Thus, the tert-Leucine derivative provided the corresponding pyrrolidone in a yield of 85%. The Valine derivative afforded the product in a good yield. However, the simplest and unsubstituted 2-amino butanoic acid derivative could produce only 47% yield. Amino acid derivatives bearing a cyclopropyl ring were also carbonylated at the γ-methylene position. The ester derivative of cyclopropylalanine produced a mixture of 5- and 6-membered products, but with unsatisfactory yields. Interestingly, the cyclopropylglycine derivative gave one of the highest yields (80%), but cyclobutyl- and cyclopentylglycine could afford the carbonylated product in trace amounts only. Similarly, cyclohexyl amino ester also furnished only a trace amount of product (Scheme 29). The directing group could be easily removed under mild conditions at 50 °C.
In an attempt to functionalize the γ-C(sp3)-H bonds of aliphatic carboxylic acids, the challenges that were presented by a thermodynamically less favoured 6-membered metallacycle intermediate proved to be difficult to overcome. In this regard, our group explored the feasibility of a C-heteroatom bond formation strategy using a bidentate directing group 8-aminoquinoline in 2018. A wide variety of 8-aminoquinolamides of aliphatic carboxylic acids and amino acids were subjected to γ-thioarylation and selenoarylation using dichalcogenides (Scheme 30) [
]. One of the significant problems that was identified prior to this study was the tendency of the sulfur to bind with the metal ions, thereby leading to degradation of the catalyst. However, dichalcogenides were successful in preventing the poisoning of the catalyst. Pd(OAc)2 was found to be the ideal catalyst for this protocol and Ag2CO3 was found to be the best oxidant. 2-Chloroquinoline as a ligand and NaHCO3 as the base drove the reaction to completion, affording the maximum possible yield. When diphenyl disulfide was employed as the coupling partner, amino acid derivatives of L-tert-Leucine, l-Valine, and l-Isoleucine provided the γ-thioarylated products in good yields. There were no racemization at the α-carbon center. Tethering a -Cl group at both para positions of diphenyl disulfide afforded the products in good yields and high diastereoselectivities with both Valine and Isoleucine derivatives. Using Pd(OPiv)2 as the catalyst and keeping the remaining conditions identical, the same substrates produced a γ-selenoarylated product with diphenyl diselenide in good yields. TEMPO was utilised as a radical scavenger and it indicated that a single-electron transfer process is unlikely to be involved.
In 2014, the Chen group disclosed a procedure for the Pd-catalyzed acetoxylation of the γ-C(sp3)-H bonds of various alkylamines and amino acid derivatives [
]. In a previous report published in 2011, they discovered that the Pd-catalyzed intramolecular amination of the same substrates in presence of PhI(OAc)2 could result in the formation of an acetoxylated side product as well [
]. They successfully fine-tuned the reaction conditions to afford the acetoxylated product as the major product (Scheme 31). The reaction was carried out with Pd(OAc)2 as the catalyst in the presence of PhI(OAc)2. The popular directing group picolinamide was used to selectively functionalize the γ-position. A mixture of AcOH and xylene turned out to be the ideal solvent and Li2CO3 worked well as an additive by suppressing the C–N cyclization to the maximum possible extent. The methyl ester of 2-aminobutanoic acid gave a very good yield of the acetoxylated product, however, the threonine derivative failed to furnish a better yield presumably due to the presence of a β-substituent.
In 2016, Liu and colleagues outlined a very interesting procedure for synthesizing unnatural amino acids. Using 5-Methylisoxazole-3-carboxamide (MICA) as the directing group, Pd(OAc)2 as the catalyst, and PhI(OAc)2 as the oxidizing agent they first successfully acetoxylated the γ-C(sp3)-H bonds of various amino acid derivatives. Further, toluene was used as the solvent and the reaction required the presence of AcOH. Methyl- and tert-Butyl ester derivatives of Valine, Isoleucine, and Threonine were all successfully acetoxylated in very good yields (Scheme 32) [
]. These acetoxylated products were further used as precursors to synthesize γ-mercaptoamino acids by treating with necessary reagents. Using these thiol-modified unnatural amino acids, native chemical ligation (NCL) was successfully performed. In this regard, NCL was demonstrated using γ-mercaptoisoleucine in the first step of a reaction protocol to synthesize Xenopus Histone H3 protein.
In 2017, Shi and co-workers designed a strategy for the acetoxylation of the γ-C(sp3)-H bonds of primary alkyl amines in which the free amino group was utilised as the directing group [
]. One of the major concerns they successfully addressed is the formation of a stable bis(amine)-metal complex that could deactivate the catalyst. They sought to suppress the formation of the bis(amine)-metal complex by protonating the amine and thus preventing it from forming a complex with the metal. Pd(OAc)2 was used as the catalyst, PhI(OAc)2 as the oxidizing agent, and Ac2O as an additive in AcOH solvent (Scheme 33). The α-methyl derivative of homoserine afforded the γ-acetoxylated product in moderate yield and the same yield was provided by a β-amino acid ester as well.
In pursuit of directing groups that can be removed under mild conditions, Xuan and co-workers outlined a strategy for the Pd-catalyzed acetoxylation of the γ-C(sp3)-H bonds of various amines and amino acid derivatives in 2018 [
]. They used Benzothiazole-2-sulfonyl (Bts) as the N-linked directing group, owing to the stability of sulfonamide and the relatively mild conditions involved in the protection and deprotection steps. Pd(OAc)2 was employed as the catalyst, PhI(OAc)2 as the oxidant, and Ac2O as an additive in toluene solvent. The methyl-, ethyl- and benzyl-ester derivatives of l-Valine afforded the γ-acetoxylated product in good yields and diastereoselectivities which is presumably a consequence of the formation of a favourable α,β-trans palladacycle intermediate (Scheme 34). The methyl ester of 2-aminobutanoic acid also gave a good yield with no racemization. The ester derivatives of l-Isoleucine, Threonine, and L-tert-Leucine were also amenable to the reaction and the primary γ-CH3 position was selectively activated in case of Isoleucine.
Later in the same year, Jia et al. also reported the first ever procedure for the Pd-catalyzed acetoxylation of the γ-C(sp3)-H bonds of a wide range of triflyl-protected amino acid derivatives and alkyl amines, wherein the reaction is supported by pyridine-based ligands [
]. Pd(OAc)2 was used as the catalyst and PhI(OAc)2 as the oxidant. Further, DCE was found to be the best solvent and the reaction required the presence of a ligand for which 2,6-lutidine was chosen. N-protected methyl-, ethyl-, and benzyl-esters of l-Valine gave moderate yields of the mono-acetoxylated product, although with high diastereoselectivities (Scheme 35). However, the tert-butyl ester of l-Valine was found to be ineffective and afforded only trace amounts of the product. The Isoleucine derivative furnished an unsatisfactory yield of the desired product. Interestingly, allo-isoleucine was extremely amenable to the reaction, giving the product in 87% yield. The reason for this was presumed to be the greater accessibility of the C(sp3)-H bond during C–H palladation in case of allo-isoleucine. O-tBu and O-TBS protected threonine derivatives were also acetoxylated, although the yield for the latter was very low. The yield increased significantly when O-TBS protected allo-threonine derivative was used. Based on various kinetic studies, the authors suggested that an organopalladium(IV) complex might be formed after γ-C(sp3)-H activation followed by oxidative addition. Further, the acetoxylated product is obtained after a C–O reductive elimination of the aforementioned complex, along with the regeneration of the catalyst.
In 2016, Shi and co-workers outlined a strategy for the stereoselective alkoxycarbonylation of unactivated γ-C(sp3)-H bonds of amino acid derivatives [
]. Using alkyl chloroformates as the carbonylation agent, Pd(OAc)2 as the catalyst, Ag2CO3 as the silver salt, and I2 as the additive. They successfully obtained the alkoxycarbonylated product for a wide range of aliphatic carboxamides. 8-aminoquinoline was employed as the auxiliary for this protocol. However, most of the substrate scope consisted of β-C(sp3)-H activation and just three cases of γ-C(sp3)-H activation have been reported. It is to be noted that the γ-C(sp3)-H bond activation occurred only when no reactive C–H bonds were present at the β-position. N-phthaloyl protected L-tert-Leucine, l-Isoleucine, and l-Valine derivatives were amenable to the protocol, however, they afforded the γ-alkoxycarbonylated product in lower yields (Scheme 36).
2.2 Delta C(sp3-H) functionalization of amino acid derivatives
Introducing functionalities in the δ-C(sp3)-H bond of amino acids is a profound challenge as it requires formation of less stable six membered metallacycle. To surmount this challenge, researchers have invented various strategies. Here are some approaches and transformations of δ-C(sp3)-H bonds in amino acid derivatives.
2.2.1 Intramolecular amination
In 2012, the Daugulis group reported the first δ-C(sp3)-H bond activation in amine substrates [
]. They synthesized pyrrolidines via intramolecular C–N bond formation where Pd(OAc)2 was utilised as the catalyst, PhI(OAc)2 as the oxidant, and toluene as the solvent. However, only one amino acid derivative was used as the substrate. The picolinamide protected tert-butyl ester of Leucine afforded the cyclized product in 36% yield (Scheme 37).
In the same year, the Chen group successfully showcased the intramolecular amination of γ-C(sp3)-H bonds of various amino acid derivatives and alkyl amines. In the same report, they also demonstrated the formation of 5-membered pyrrolidines via δ-C(sp3)-H bond activation which would require the intermediacy of a kinetically less favoured 6-membered palladacycle [
]. Pd(OAc)2 was used as the catalyst and PhI(OAc)2 as the oxidant in presence of AcOH and toluene (Scheme 38). Picolinamide worked well as the directing group. They successfully carried out the transformation of the ester derivative of Leucine and obtained the corresponding cyclized pyrrolidine product in a good yield. Further, the ester derivative of 4-methylleucine also yielded the desired product in a good yield. However, the Norvaline derivative afforded only 17% of the product and over 70% of the starting material was recovered after the reaction.
One of the reasons why functionalization of δ-C-H bonds is challenging is because the reaction needs to go through a kinetically unfavourable six-membered palladacycle intermediate. In 2016 Shi and co-workers reported the first example of alkenylation of the δ-C(sp3)-H bonds of amino acid derivatives and amines despite the availability of equally accessible γ-C(sp3)-H bonds in the substrate [
]. Pd(OAc)2 was chosen as the transition-metal catalyst and the reaction was directed by the picolinamide directing group. 2,6-Dimethylbenzoquinone, which serves as both ligand and co-oxidant was found to promote the reaction. The yield was significantly reduced in the absence of NaHCO3. LiF was used as an additive and a mixture of tetrachloroethane (TCE) and hexafluoroisopropanol (HFIP) was utilised as the solvent.1,2-disubstituted alkynes were employed as alkenylating agents and an O2 atmosphere was found to be the most conducive for the reaction. Isoleucine methyl ester was chosen as the model substrate and a variety of alkynes were screened. Diarylacetylenes containing both electron-donating groups and electron-withdrawing groups were incorporated at the δ-position, affording moderate to good yields. Dialkylacetylenes and symmetrical alkynes such as 3-hexyne and 4-octyne were also amenable to the protocol, giving moderate yields. Unsymmetrical alkynes were also well-tolerated and the alkenylated products were found to be a mixture of regioisomers. Ester derivatives of l-Isoleucine, D-allo-Isoleucine, and Norvaline were successfully alkenylated but the yields remained moderate (Scheme 39). Other aliphatic amines bearing γ-methoxy or δ′-ester provided unsatisfactory yields. Based on previous studies and deuteration experiments, it was confirmed that the cleavage of γ-C-H bonds was reversible but there was no deuterium incorporation at the δ-position. This indicated that either δ-C-H activation was irreversible or the migratory insertion of the alkyne was considerably faster than the reversibility of the δ-C-H activation step. Referring to the Curtin-Hammett principle, they suggested that high steric repulsion could prevent the palladacycles C-1 and C-2 from undergoing migratory insertion (Scheme 40). However, the intermediate C (formed via δ-C-H activation and subsequent alkyne coordination) forms an eight-membered cyclic intermediate through migratory insertion. Further protonation finally yields the desired product D, and regenerates the catalyst (Scheme 40, path C). They also suggested a different pathway wherein palladacycle E could form by 1,4-Palladium migration of B-1 (Scheme 40, path A). A third possible scenario suggested that the complex B-2 could undergo β-hydride elimination, leading to the formation of palladacycle B by bond rotation/insertion (Scheme 40, path B). However, deuteration experiments and 1H NMR studies leaned highly in favour of Path C. The picolinamide directing group could be easily removed on treatment with PCl5/Lutidine, followed by methanol. They also successfully put this protocol to application by synthesizing chiral piperidines, which are integral parts of various bioactive natural products.
Later in 2018, Shi group again reported an elegant procedure for the Pd(II) catalyzed, picolinamide-directed alkylation of δ-C(sp3)-H bonds of amino acid and amine derivatives, in presence of equally accessible γ-CH3 bonds [
]. Maleimides, a class of widely used organic reagents, have been used as the alkylating agent for this protocol. Pd(OAc)2 was employed as the catalyst and benzoquinone was found to aid the reaction significantly. The authors proposed that it might be acting as both ligand and co-oxidant for this reaction. Further, the procedure required the presence of 1-AdCOOH and Na2CO3 as additives and 1,1,2-trichloroethane worked well as the solvent. A wide range of N-substituted maleimides, such as N-aryl and N-alkyl maleimides were screened with l-Isoleucine methyl ester which was chosen as the model substrate. Further, the aryl rings present in the maleimides were also substituted at different positions with various functional groups and δ-alkylated products were obtained in all cases with moderate to excellent yields (53–88%, 13 examples). It was shown that ethyl-, benzyl- and allyl-esters of Isoleucine were amenable to the reaction and gave good yields. Other Isoleucine derivatives such as γ-arylated, γ-alkoxylated, and γ-alkylated Isoleucine esters were also alkylated at the δ-position. The Norvaline ester also yielded the desired product, albeit in a low yield (Scheme 41). Similar to their previous alkenylation study, they cited the Curtin-Hammett principle to explain the reason for δ-selectivity over γ-selectivity in this case. It was suggested that the functionalization of the five-membered palladacycle resulting from γ-C-H activation does not occur due to the high energy barrier and therefore, a six-membered palladacycle could form via δ-C-H activation and then undergo further functionalization.
In 2019, our group in collaboration with Paton and co-workers disclosed a procedure for the arylation of the δ-C(sp3)-H bonds of amino acid derivatives and other amines [
]. The choice of ligand plays a crucial role in this reaction as evident from DFT studies. For the same reason we screened a number of ligands to find the best one for the transformation. The ethyl ester of Leucine was the first substrate to be screened and we found that the combination of Pd(OPiv)2 as catalyst, 4-Benzyloxy-2(1H)-pyridone as the ligand, Ag2CO3 as the silver salt, CF3COONa as the additive, and tert-butyl methyl ether (TBME) as the solvent worked best for the reaction. Picolinic acid turned out to be the best-performing directing group. A range of aryl iodides were utilised as coupling partners. The scope of amino acid substrates included the methyl and ethyl ester derivatives of leucine as well as the methyl ester of homoleucine, all of which provided good yields of the δ-arylated product. Further, both electron-withdrawing and electron-donating groups present on the aryl ring of the coupling partners were well-tolerated and a high degree of diastereoselectivity was observed in case of l-Leucine methyl ester (Scheme 42). Removal of the directing group was comfortably achieved under mild conditions. Various experimental and computational studies using DFT provided insights into the mechanism for this protocol.
So far, most of the attempts that have been made to carry out the activation of δ-C(sp3)-H bonds have been dependent on substrates which either have a high conformational restraint, rendering the γ-C-H bond inaccessible, or substrates in which the γ-position has been blocked by various substituents. In 2021, Carretero and colleagues designed a method for the Pd-catalyzed arylation of δ-C(sp3)-H bonds of α-amino acid derivatives by overcoming the need for γ-substitution [
]. In this regard, instead of using a directing group having a carbonyl connector, they chose to utilize a directing group with a sulfonyl connector such as 2-pyridylsulfonyl (-SO2Py) group, owing to different geometrical and electronic properties from the commonly used picolinic acid. The –SO2Py group was found to give a greater control over the δ-selectivity when tested on the ester derivative of α-methyl norvaline. Pd(OAc)2 was used as the catalyst, AgOAc as the oxidant, and 1,4-dioxane as the solvent for this transformation, while aryl iodides acted as the coupling partner. Amino acid substrates containing α-quaternary carbon centers with unbranched substituents provided δ-selective arylation products in moderate yields (Scheme 43). The unsubstituted methyl ester of Norvaline gave a very low yield of 16%. However, the yield was incremented when a β-substituent was added to the methyl ester of Norvaline (52–70%). Lactamization at a moderate yield was observed β-substituent was a (methoxycarbonyl)methyl group. A wide range aryl iodides containing functional groups at the para- and meta-positions such as, -Cl, -Br, –I, -Me, -OMe, CF3 were amenable to the reaction, furnishing moderate to very good yields (46–74%). Furthermore, heteroaromatic systems were also compatible to this reaction and complete δ-selectivity was observed even though a γ-CH2 moiety was present. Further, aryl iodides of higher structural and functional complexity could also be comfortably incorporated into the substrate, confirming that unnatural amino acids could be synthesized from natural products and drug derivatives via this protocol. H/D scrambling experiments and DFT studies suggested that a concerted metalation-deprotonation pathway could facilitate the δ-C(sp3)-H bond activation, followed by the formation of a six-membered palladacycle intermediate. Afterwards, the intermediate could go through a AcOH/Ph-I ligand-exchange, leading to the oxidative addition of Ph-I, forming a Pd(IV) intermediate. Subsequent reductive elimination could then furnish the desired product. The superior performance of the sulfonyl based directing group has been attributed to absence of the conjugation of nitrogen with sulfur which leads to the sulfur atom orienting out of the plane. This permits the palladacycle intermediate to adopt a conformation in which the distortion of the coordination plane of Pd(II) is as little as possible. On the contrary, the conformation of the palladacycle in case of a picolinamide-directed substrate is more rigid and thus, it leads to a more distorted geometry around Pd(II).
Direct C–H functionalization has come up as one of the potent strategies to modify amino acid derivatives and peptides. Given the inert and fluxional nature of C–H bonds in amino acid alkyl chains, the progress made in the past decade in this domain is commendable. Especially, the challenging area of remote C(sp3)-H activation in amino acid derivatives has witnessed great successes in recent years. In this regard, multifarious directing groups, ligands, etc. have been introduced for the selective activation of γ- and δ-C(sp3)-H bonds of amino acid and peptide side chains. A number of functionalities have been introduced into remote C–H bonds with high regioselectivities as well as diastereoselectivities. To incorporate various functional groups, strong coordination of an external directing group has been utilised in most cases. Nonetheless, the recent trends have shown that the weaker coordination of the carboxylate present in the amino acid could also be utilised for carrying out distal C–H functionalization. Even after these great successes, there remain ample opportunities to further refine distal C–H functionalization of amino acids. Most of the examples discussed in this highlight require the usage of an external auxiliary which also ends up in the product, stoichiometric amount of oxidants/or necessity of cocktails of reagents, high temperatures, etc. Also, the transformation of C(sp3)-H bonds most often requires a high catalyst loading to achieve efficient activation/functionalization. Pd catalysts mainly dominated the remote C(sp3)-H functionalization of amino acid. Usage of other inexpensive earth abundant metals like Ni, Mn, Fe, etc. will make these reactions more economical. Further, considering the late-stage functionalization of complex peptides and proteins, future attempts must be made to develop reactions in aqueous medium. Asymmetric synthesis could be merged with C(sp3)-H functionalization of amino acids and peptides to harness the full potential of such reactions. Transition metal catalyzed remote C(sp3)-H activation has already become a potent arsenal for synthesizing unnatural amino acids and peptides in an efficient manner. Further honing of this field is required to make these reactions the approach of choice for selective and sustainable transformation of amino acid derivatives.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
This activity has been supported by a research grant SERB India ( CRG/2018/003951 ). Financial support received from CSIR-India (J.D) is gratefully acknowledged.
a)Hughes A.B. Amino Acids, Peptides and Proteins in Organic Chemistry: Building Blocks, Catalysis and Coupling Chemistry. Wiley-VCH,
☆Professor Debabrata Maiti, the corresponding author on this paper is a member of the advisory board but this author had no involvement in the peer review process used to assess this work submitted to Tetrahedron Chem. This paper was assessed and the corresponding peer review managed by Dr Jessica Pancholi, a scientific editor working on Tetrahedron Chem.