Amino acid fluorides: viable tools for synthesis of peptides, peptidomimetics and enantiopure heterocycles

Abstract

This review provides a broad perspective of the uses of amino acid fluorides in the synthesis of peptides and a wide range of other molecules. The topic is discussed with reference to the preparation of N-protected amino acid fluorides, their reactivity, coupling and their synthetic applications. The merits of the use of amino acid fluorides as coupling agents for making difficult peptides and in combating the problem of stereomutation, as well as their successful use in solid phase peptide synthesis, are reported with examples. Recent developments in the application of amino acid fluorides for making amino acid derivatives, peptide conjugates, peptide mimics and heterocycles are also reviewed.

---

Girish Prabhu

Girish Prabhu was born in Karkala, Karnataka, India in 1984. He obtained his M.Sc. in Organic Chemistry from Mangalore University, Mangalore in 2006. He joined Syngenta Biosciences Pvt. Ltd., Goa, India in 2006 as Junior Research Scientist. In January 2010, he joined the research group of Professor V. V. Sureshbabu at the Department of Chemistry, Central College, Bangalore University, Bangalore for his Ph.D. Presently he is pursuing research in peptides and peptidomimetics.

N. Narendra

N. Narendra was born in Bangalore, Karnataka, India in 1981. He completed his M.Sc. in Organic Chemistry at Bangalore University, Bangalore in 2005. He worked on peptides and peptidomimetics under the guidance of Professor V.V. Sureshbabu at the Department of Chemistry, Central College, Bangalore for his Ph.D. Presently he is working as an assistant professor at the University College of Science, Tumkur University.

Basavaprabhu

Basavaprabhu was born in Raichur, Karnataka, India in 1985. He completed his M.Sc. in Organic Chemistry at Bangalore University, Bangalore in 2008. Presently he is pursuing research in peptides and peptidomimetics under the supervision of Professor V. V. Sureshbabu at the Department of Chemistry, Central College, Bangalore.

V. Panduranga

V. Panduranga was born in Vijayawada, Andhra Pradesh, India in 1982. He completed his M.Sc. in Organic Chemistry at K. B. N. College, Nagarjuna University, Guntur in 2005. Presently he is pursuing research in peptides and peptidomimetics under the supervision of Professor V. V. Sureshbabu at the Department of Chemistry, Central College, Bangalore.

Vommina V. Sureshbabu

Professor Vommina V. Sureshbabu was born in Nellore, Andhra Pradesh, India in 1961. He obtained his M.Sc. in Chemistry from Sri Krishnadevaraya University, Anantapur, India in 1983. He was invited by Professor K. M. Sivanandaiah to join him for his Ph.D. at Central College, Bangalore to work in the area of peptide chemistry. After the completion of his Ph.D. in 1989, he was appointed as a Lecturer at the same department. Later, he went to the USA for a postdoctoral assignment to CUNY, New York where he worked on the synthesis of GPCR fragments through native chemical ligation. At present, he is working as a Professor at the Department of Studies in Chemistry, Central College, Bangalore. His research interests include development of new reagents for efficient peptide synthesis, design and synthesis of peptidomimetics, incorporation of unnatural linkages into the peptide backbone, native chemical ligation, C-terminal versus N-terminal modifications for peptidomimetic synthesis and use of Fmoc-groups in solution-phase synthesis. Recently, V.V.S. has contributed a chapter entitled ‘Protection Reactions’ to the volume ‘Amino Acids, Peptides and Proteins in Organic Chemistry’ Vol. 4. Protection Reactions, Combinatorial and Medicinal Chemistry: (Ed. Andrew Hughes), 2011, Wiley International Ltd and a report entitled ‘Total chemical synthesis of polypeptides and proteins: Chemistry of ligation techniques and beyond’ (Tetrahedron, 2012, 68, 9491–9537).

---

1. Introduction

The acid halide method as an activation strategy in peptide chemistry has its origin in Fischer and Otto's first report on the synthesis of a dipeptide, Gly-Gly, employing amino acid chloride in 1901.^1–5 But the emergence of acid halide activation (almost exclusively constituted by acid chlorides) as a method of choice for peptide coupling was ignored as more than one difficulty was experienced in the years that followed. The main challenge was its incompatibility with the, then widely used, N^α-protecting groups namely, tert-butyloxycarbonyl (Boc) and benzyloxycarbonyl (Cbz) groups. The problems were encountered starting from the preparation of the acid chlorides from amino acids protected with acid labile Boc and Cbz groups.^6–12 The N-alkoxycarbonyl amino acid chlorides were deemed to be over-activated species which underwent cyclization reactions to N-carboxyanydrides (NCA, also called Leuch's anhydrides) even at low temperatures.^13–15 Another major problem was that of oxazolone formation leading to the loss of optical purity of the peptides prepared.^16,17

Thus, the stumbling blocks in the widespread usage of the acid chlorides as peptide coupling agents were: stability, premature cleavage of protecting groups¹⁸ and optical fragility. Attempts were made for the development of N^α-amino protection and mild reagents for the preparation of N^α-protected amino acid chlorides, but no major successes were found.^19–21

1.1 Bursting the myth: acid chlorides and Fmoc-urethane chemistry

The discovery of 9-fluorenylmethoxycarbonyl (Fmoc) amino acid chlorides by Carpino et al. in 1982 (ref. 22) led to a paradigm shift in the way acid halides were used in peptide chemistry. It was with the introduction of the acid stable Fmoc group, that acid halide chemistry was reborn and subsequent work quickly transformed them from “obsolete” entities²³ to tailor made reagents for peptide synthesis.^16,17 Carpino et al. prepared and demonstrated the stability of Fmoc-amino acid chlorides²² and employed them in solution as well as in solid phase peptide synthesis (SPPS). The problem of the instability of N-protected amino acid chlorides was solved but that of the optical fragility persisted. The tertiary amine usually employed as the hydrogen chloride acceptor converted the acid chloride into 2-alkoxy-5(4H)-oxazolone which undergoes stereomutation because of the base catalyzed enolization.²⁴ The oxazolone was also susceptible to aminolysis causing it to open into a peptide. This was a rapid reaction in homogeneous or two-phase systems but it became sluggish in solid phase synthesis, thereby leading to more enolization, and thus causing a setback for the usage of Fmoc-amino acid chlorides in SPPS. However, a couple of methods such as the addition of the potassium salt of 1-hydroxy-1H-benzotriazole (HOBt) during coupling and the zinc (Zn) dust mediated no base coupling reaction²⁵ were devised to circumvent the formation of oxazolone. But the problem of stereomutation could not be fully resolved particularly in view of the very high level of optical purity expected of the synthetic peptides and peptide analogues. Furthermore, the success of the orthogonal Fmoc/tert-butyl protection strategy²⁶ could not be extended to peptide couplings via amino acid chlorides because of the incompatibility of the tert-butyl (^tBu) side chain based protections with the acid chloride method.

1.2 Acid fluorides: compatibility with Fmoc/Boc/Cbz urethanes and efficacy of coupling

The breakthrough in the acid halide method of activation came with the introduction of amino acid fluorides as peptide coupling agents. The acid fluorides overcame the limitations of coupling with acyl chlorides by being more stable to hydrolysis, less prone to racemization and more reactive to anionic nucleophiles and amines.^27,28 Unlike the acid chloride activation, the method was compatible with acid labile amino acid protections as well. Thus, the acid fluorides of the Fmoc, Boc, Cbz-amino acids as well as mild acid sensitive ^tBu, benzyl (Bn), and trityl (Trt) moiety based side chain protected amino acids could be easily accessed.^29,30 Despite all these advantages, the reactivity of the acid fluorides was still very much retained at the level of acid chlorides. Within a short period of time, many more classes of acid fluorides were prepared from non-proteinogenic amino acids such as α,α-disubstituted ones {α-aminoisobutyric acid (Aib), isovaline (Iva), α-ethylalanine,^31–33 α,α-diethylglycine (Deg), α,α-dibutylglycine and 1-aminocyclopentane-1-carboxylic acid (Ac₅c), 1-aminocyclohexane-1-carboxylic acid (Ac₆c), 1-aminocycloheptane-1-carboxylic acid (Ac₇c)},³⁴ N-methylated (NMe) amino acids including NMe alanine (Ala), NMe leucine (Leu), NMe valine (Val), sarcosine (Sar or N-methylglycine)³⁴ and N-methylated C^α,α-dimethylamino acids such as NMeAib}.³⁵ Other new N-urethane protecting groups such as 1,1-dioxobenzo[b]thiophene-2-ylmethyloxycarbonyl (Bsmoc) namely Bsmoc-Asn(Trt)-F, Bsmoc-Asn(Dmcp)-F (Asn = asparagine; Dmcp = dimethylcyclopropylmethyl Gln = glutamine), Bsmoc-Gln(Trt)-F, Bsmoc-Gln(Dmcp)-F),^36,37 2-(tert-butylsulfonyl)-2-propyloxycarbonyl (Bspoc),^37,38 2-methylsulfonyl-3-phenyl-1-prop-2-enyloxycarbonyl (Mspoc)³⁹ and non-urethane types including the Trt group⁴⁰ were also employed as N-protecting groups for the preparation of amino acid fluorides.

The acid fluoride method is an easy and efficient way of activating carboxylic acids. The high solubility (in aqueous as well as in organic solvents) and the small size of the leaving group facilitates a rapid reaction. Amino acid fluorides could be used for coupling in the absence of a base and were also sufficiently stable in the presence of tertiary amines. Oxazolone formation was conspicuously reduced in the presence of a tertiary base. In fact, over-activation, which is often a pre-requisite for the coupling of sterically hindered amino acids and assembly of difficult sequences, could be achieved without difficulty. The importance of the acid fluoride method is not limited to its use in the Fmoc/^tBu protection based peptide couplings. It has expanded to accommodate newer protection and coupling strategies. The acid fluoride mediated coupling together with modifications which particularly facilitate the assembly of several difficult peptides and polypeptides in the solid phase. With the availability of reagents for the in situ generation of acid fluorides, the procedures became further simplified, and the acid fluoride method ended up being a solution to multiple problems of peptide coupling that had been persisting for a long period of time.

The objectives of this review are to provide a broad overview of acid fluoride activation in peptide chemistry, which has not been dealt with, with the requisite attention thus far and a full serious paper on this topic has not been found in the literature. A short account of amide bond formation using amino acid fluorides was published in 2005.⁴¹ Carpino et al. described their work on Fmoc-amino acid fluorides in an account on amino acid halides in 1996.¹⁶ El-Faham and Khattab have summarized the applications of tetramethylfluoroformamidinium hexafluorophosphate (TFFH; 2) as a fluorinating agent in peptide as well as in organic chemistry.⁴² Also reviews have been published on peptide coupling agents, but without the essential depth in the treatment of the acid fluoride method.⁴³

Although originally described for the purpose of peptide synthesis, the acid fluorides by virtue of their favorable inherent properties have become the chosen reagents for the preparation of a wide range of molecules, namely, many classes of peptide mimics, carboxamides, Weinreb amides, esters, alcohols and heterocycles. In spite of their remarkable properties, they are underutilized in synthetic chemistry. This paper comprehensively reviews the various aspects of synthesis of acid fluorides in peptide chemistry and beyond. The paper is organised in different sections for fluorinating reagents, N-protected amino acid fluorides (preparation and properties), and coupling employing various protocols and applications in difficult and hindered couplings. In addition, the review highlights recent research where the advantages of acid fluoride activation have been particularly used for the successful assembly of a number of peptides and peptide mimics of biological significance. Attempts have been made to bring out the suitability of the acid fluoride activation for the assembly of peptides bearing C^α,α-dialkyl amino acids, N-methyl amino acids and unnatural amino acids. The modifications in the procedures are outlined whenever they are essential. Solid phase synthesis and segment condensations involving acid fluorides are also discussed. With our expertise in the field of synthetic peptides and peptidomimetic chemistry, we have attempted to show the relevance of amino acid fluorides in the need for the chemical assembly of biologically significant polypeptides in the laboratory, a demanding field which is rapidly expanding.⁴⁴

2. Fluorinating reagents

Beginning with cyanuric fluoride (C₃F₃N₃)^45,46 (1) and followed quickly by Carpino's TFFH^47,48 (2), an array of fluorinating reagents are now available (Fig. 1).

Fig. 1 Fluorinating reagents used in peptide chemistry.

With most of these reagents, the conversion of acid to acid fluorides follows a similar pathway involving an ester intermediate, followed by nucleophilic attack of the fluoride anion (Scheme 1).

Scheme 1 Conversion of acid to acid fluoride using different fluorinating reagents via their respective intermediates.

2.1 Cyanuric fluoride

C₃F₃N₃ is the reagent of choice for the preparation of N-protected amino acid fluorides.²⁷ Olah et al. first reported its use for making acid fluorides from carboxylic acids.⁴⁵ However, C₃F₃N₃ is corrosive, and its use in large scale preparations poses several other problems. Precipitation of insoluble cyanuric acid or formation of suspension or emulsion (in the case of trifunctional amino acids) during large scale synthesis necessitates additional washings and extraction with water, which may lead to a loss of yield.

2.2 DAST and Deoxo-Fluor

Diethylaminosulfur trifluoride (DAST, Et₃NSF₃, 3) was developed as a safer alternative to sulfur tetrafluoride.⁴⁹ The reagent has been used for the conversion of alcohols to fluorides, aldehydes and ketones to geminal difluorides.⁵⁰ Subsequently DAST (3)^51,52 was employed as an alternative to C₃F₃N₃ (Scheme 2). It has several advantages over C₃F₃N₃ in terms of absence of a base, simplified work-up and easy elimination of water soluble by-products.^53,54 In addition no suspension or emulsion forms. Removal of excess DAST, if remaining, can be done by extraction with water. However, DAST is not easy to handle because of its fuming, hygroscopic nature and low thermal stability.

Scheme 2 Mechanism of acid fluoride formation with Deoxo-Fluor and the structure of bis(methoxyethyl)amide side product 21.

Bis(2-methoxyethyl)aminosulfur trifluoride, commonly known as Deoxo-Fluor® [(MeOCH₂CH₂)₂NSF₃] (4),^55–58 is a safer reagent than DAST in large scale reactions and it has enhanced thermal stability. It is a liquid, which fumes in air and has a reactivity similar to that of DAST (Scheme 2). The by-products sulfur dioxide (SO₂) and HN(CH₂CH₂OCH₃)₂ can be evaporated in vacuo. Though the reaction with Deoxo-Fluor could be carried out in the absence of a base, a side reaction involving the formation of bis(methoxyethyl)amide (21) especially with the hindered amines limits its applicability.

2.3 Fluoroformamidinium salts

Fluoroformamidinium salts are the most efficient reagents among the haloformamidinium salts for solid phase couplings.²⁷ In contrast, corresponding phosphonum analogs, namely, fluorotri(pyrrolidino)phosphonium hexafluorophosphate failed to generate any acid fluoride by reaction with protected amino acid anion.²⁷ TFFH^47,59–62 was introduced as a benign fluorinating agent for in situ generation of acyl fluoride and was found to be an efficient alternative to C₃F₃N₃. It is non-hygroscopic, stable and easier to handle. It is postulated that urea formation being the driving force behind the TFFH activation and acid fluoride formation proceeds through the formation of an O-acylisourea intermediate (11) as shown in Scheme 1.

Several variants of TFFH have been developed to improve the reactivity of the reagent and to eliminate the toxic by-products that are generated upon work-up of the reaction mixture (Fig. 2).

Fig. 2 Variants of TFFH.

Bis(tetramethylene)fluoroformamidinium hexafluorophosphate (BTFFH, 22) and 1,3-dimethyl-2-fluoro-4,5-dihydro-1H-imidazolium hexafluorophosphate (DFIH, 23) were developed as new and alternative fluorinating reagents by El-Faham^63,64 and the former was found to be superior to TFFH because it did not form toxic by-products. Other reagents developed include tetraethylfluoroformamidinium hexafluorophosphate (TEFFH, 24), 1,2-dimethyl-3,3-tetramethylenefluoroformamidinium hexafluorophosphate (DMFFH, 25), and 1,2-diethyl-3,3-tetramethylenefluoroformamidinium hexafluorophosphate (DEFFH, 26).⁶⁵ However, these were found to be less stable in dimethylformamide (DMF) solution in the presence of a base because of decomposition. Also they were found to be less reactive than TFFH or BTFFH. N-(Fluoro(morpholino)methylene)-N-methylmethanaminium hexafluorophosphate (DMFH, 27), also introduced by El-Faham and Albericio,⁶⁶ requires only an equimolar quantity of a base and was found to have some advantages over the other fluorinating agents. The oxygen atom in the morpholino moiety acts as a proton acceptor, thereby providing an additional basic site.

2.4 1-Ethyl-2-fluoropyridinium salts

Mukaiyama reagents, namely, 2-chloro- and 2-bromo-pyridinium iodides, are unattractive for peptide synthesis^67–74 because of their poor solubility and the need for rigorous conditions. Li and Xu⁷⁵ developed 1-ethyl-2-fluoro-pyridinium tetrafluoroborate (FEP, 5) and 1-ethyl-2-fluoropyridinium hexachloroantimonate (FEPH, 6) based on 2-halopyridinium salts as an alternative to halouronium type reagents. It was reasoned that the carbocation in halouronium type coupling reagents was stabilized because of the presence of a lone pair of electrons on the amino group. In order to enhance the reactivity of the carbocation intermediate, one of the nitrogens was replaced by atoms without a lone pair of electrons such as carbon. Thus 2-fluoro-pyridinium salts were developed with tetrafluoroborate (BF₄⁻) and hexachloroantimonate (SbCl₆⁻) as non-nucleophilic counter anions to improve the solubility and reactivity of these reagents. These reagents proved to be better than BTFFH (22), bromotri(pyrrolidino)phosphonium hexafluorophosphate (PyBrop), chlorodipyrrolidinocarbenium hexafluorophosphate (PyClU) or N,N′-bis(2-oxo-3-oxazolidinyl)phosphinic chloride (BOP-Cl). Using proton nuclear magnetic resonance (¹H-NMR), infrared (IR) and high performance liquid chromatography (HPLC) studies, the corresponding acyl fluoride and acyloxypyridinium salts of the N-protected amino acid were proposed as major reactive intermediates. However, no attempts were made for isolation and characterization of the intermediate acid fluoride.

2.5 Complex poly(hydrogen fluoride) additives: benzyltriphenylphosphonium dihydrogen trifluoride (PTF) and pyridine-hydrogen fluoride [Py(HF)_n]

Carpino et al. found that PTF [(C₆H₅)₃PCH₂C₆H₅·H₂F₃] (7),⁷⁶ when used together with 1-[bis(dimethylamino)methylene]-1H-benzotriazolium hexafluorophosphate 3-oxide (N-HBTU, 28) or 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo-[4,5-b]pyridinium hexafluorophosphate 3-oxide [N-HATU, (29), Fig. 3]⁷⁷ diverted the course of acid activation to the corresponding acid fluoride. The IR studies showed the presence of both acid fluoride and the respective 7-aza-benzotriazole or benzotriazole derived ester, and the complete conversion to acid fluoride was not obtained. Acid fluorides can be also be generated via dicyclohexylcarbodiimide (DCC) or diisopropylcarbodiimide (DIC)⁷⁸ activation in the presence of PTF or Py(HF)_n where the putative O-acylisourea intermediate is converted to acid fluoride.

Fig. 3 Amino acid fluorides derived from N^α-Fmoc protected Asp/Glu and the structure of Fmoc-pyroglutamic acid tert-butyl ester (32).

2.6 Tetrabutylammonium tetra(tert-butanol)-coordinated fluoride [TBAF(^tBuOH)₄]

Tetrabutylammonium tetra(tert-butanol)-coordinated fluoride [TBAF(^tBuOH)₄ or Bu₄N⁺F⁻(^tBuOH)₄] (8) has a structure that has four (^tBuOH)₄ molecules coordinated around a single fluoride ion through hydrogen bonds.⁷⁹ It is reported to have good nucleophilicity with low basicity and good solubility in organic solvents. A combination of trichloroacetonitrile (Cl₃CCN), triphenylphosphine (Phh₃), and TBAF(^tBuOH)₄ was developed to obtain acid fluorides from Fmoc/Boc/Cbz amino acids with good yields.⁸⁰ The acyl fluorides were formed by halide exchange of in situ generated acid chloride with TBAF(^tBuOH)₄. The reagent, which is similar to DAST and Deoxo-Fluor, does not require a base and is compatible with acid sensitive functionalities.

2.7 Fluolead™ (4-tert-butyl-2,6-dimethylphenylsulfur trifluoride)

Fluolead [(9); 4-(CH₃)₃C-2,6-CH₃C₆H₂SF₃]⁸¹ was recently used in the preparation of N-Fmoc/Boc/Cbz (2S,4S)-4-fluoropyrrolidine-2-carbonyl fluorides in high yields by using stereospecific double fluorination of N-Fmoc/Boc/Cbz (2S,4R)-4-hydroxyproline.⁸² The reagent was a solid with high thermal and hydrolytic stability and was found to be superior to DAST and its analogues.

3. Amino acid fluorides: preparation and properties

The known stability of tert-butyl fluoroformate compared to its chloro counterpart⁸³ caused the examination of amino acid fluorides as an alternative to acid chlorides. Stable amino acid fluorides were prepared and isolated [except for histidine (His), arginine (Arg), acid fluorides which were not stable enough to isolate and thus were generated in situ]. The big advantage was that unlike amino acid chlorides, the acid fluorides were found to be resistant to oxazolone formation. The acid fluorides resembled the activated esters more than acid chlorides or acid bromides in their behaviour during coupling.

3.1 Fmoc chemistry

Fmoc-amino acid fluorides (29) were first reported by Carpino et al.,²⁷ and later by Bertho et al.,²⁸ using cyanuric fluoride as the fluorinating reagent. Carpino's procedure involved refluxing a solution of Fmoc-amino acid in dichloromethane (CH₂Cl₂) under nitorgen (N₂) with cyanuric fluoride and pyridine for two hours (Scheme 3). All the proteinogenic amino acids, except cysteine (Cys), Arg, and His were converted to their corresponding acid fluorides. The acid sensitive tert-butyl ethers of serine (Ser), threonine (Thr), and tyrosine (Tyr), ε-Boc-Lysine (Lys), N-Trt derivatives of Asn, Gln³⁰ were also converted to their respective acid fluorides with good yields. Fmoc-Trp-F (Trp = tryptophan) could be obtained even without the side chain protection.

Scheme 3 Carpino's synthesis of Fmoc-protected amino acid fluorides.

The method was then extended for acid fluorides of β-/γ-tert-butyl esters of aspartate (Asp) and glutamate (Glu), i.e., Fmoc-Asp/Glu(O^tBu)-F (30) and β- and γ-acid fluorides of Fmoc-Asp/Glu-O^tBu³⁰ (31) (Fig. 3). For Fmoc-Glu-O^tBu, the reaction was carried out at −30 to −20 °C, to avoid the formation of a pyroglutamate by-product (32).

In another protocol, DAST was employed in the absence of a base for making Fmoc-amino acid fluorides including those containing acid labile side chain protection.^53,54 The reaction was easy and rapid (completed within 5 min).⁵⁴ Products were isolated as crystalline solids after recrystallization. Fmoc-O,O-(dimethylphospho)-L-tyrosyl fluoride was prepared in high yields (>95%) using DAST and employed as an alternative to circumvent low yield coupling of phosphotyrosine residues to a peptide chain.⁸⁴

Fmoc-amino acid fluorides were shown to be highly soluble and stable in organic solvents (>1 M in DMF) at room temperature (rt) for long periods. When the acid fluorides derived from Fmoc-proline (Pro), isoleucine (Ile), Ala, Val, Thr(^tBu), Leu, Ser(^tBu), Lys(Boc), Gln(Trt), Gly, phenylalanine (Phe), Glu(^tBu), Aib, Tyr(^tBu), Trp, methionine (Met), Asp(^tBu), Cys(^tBu), Asn(Trt) were stored in DMF (dried over molecular sieves for three days) and aliquots of the solutions were diluted with methanol/5% pyridine and analyzed using HPLC at 220 nm after 24 hours. Only the corresponding methyl ester could be observed and the free acids were detected to a small extent. Unlike acid chlorides,²² amino acid fluorides containing tert-butyl, Boc and Cbz as well as other acid labile side chain protecting groups were quite stable for chromatographic purification.

Although most of the acid fluorides from 20 proteinogenic amino acids were found to be stable, sulfonamide protected Fmoc-Arg(Pbf)-OH (Pbf = 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl), after activation with cyanuric fluoride, was found to undergo an easy deactivation via lactam formation [(33); Fig. 4].⁸⁵ Also, Fmoc-His-(Trt)-F was found to be unstable on storage. In such cases, TFFH can be employed in the presence of a tertiary amine⁴⁸ for the in situ generation and coupling of these compounds. Complete conversion to the acid fluoride occurs within 15–20 min and the protocol is generic including the acid sensitive functionalized side chains. For hindered amino acids such as Aib, the reaction requires an extended duration of 1–2 h. In addition, Fmoc-L-(α-Me)Tyr(PO₃Bn₂)-F, Fmoc-L-(α-Me)Phe(CO₂^tBu)-F, and Fmoc-L-(α-Me)Phe(CH₂CO₂^tBu)-F have been generated in situ using the TFFH technique.⁸⁶ TFFH mediated acid fluorides may also be isolated and purified as desired.

Fig. 4 Structure of γ-lactam formed from Fmoc-Arg(Pbf)-F.

BTFFH, DFIH, and DMFH are also useful as fluorinating agents for the in situ generation of acid fluorides including His and Arg.^63–66 However, BTFFH proved superior to the other reagents tested. The IR analysis of the reaction of Fmoc-Arg(Pbf)-OH with BTFFH and N,N-diisopropylethylamine (DIEA) (1 : 1 : 2) in DMF showed an acid fluoride peak (1845 cm⁻¹) within 2 min. Although cyclization to the corresponding lactam (1749 cm⁻¹) occurred slowly, a significant amount of the acid fluoride was obtained even after 60 min.⁶⁴

3.2 Boc and Cbz chemistry

The success of Fmoc amino acid fluorides as effective coupling agents has prompted the examination of the compatibility of fluoride activation with other N^α-protecting groups including Boc and Cbz. One of the main limitations of the acid chloride method of activation is its inability to provide Boc and Cbz protected stable amino acid chlorides (see previously). But with the acid fluoride method of activation, this limitation was overcome. Following the first usage by Mukaiyama et al. of 2-fluoro-1,3-dimethylpyridinium salt^67–73 for the preparation of an acid fluoride from carboxylic acid in the presence of triethylamine,⁷⁴ Picard et al. reported the synthesis of Cbz-protected glutamyl difluoride (34) and cystyl difluoride [(35); Fig. 5] and used them in the synthesis of macrocyclic dilactones.⁸⁷

Fig. 5 Structures of protected cystyl and glutamyl difluorides.

Subsequently, Carpino et al. reported the preparation of Boc-/Cbz-amino acid fluorides (Table 1) employing cyanuric fluoride at low temperatures (−20 to −10 °C) for 1 h.²⁷ Boc-amino acid fluorides were obtained in crystalline form except for those derived from Val, Ile, Pro, and Thr(Bn). These could be stored for a long duration of time under refrigeration. However, Cbz derivatives showed a lesser tendency to crystallize (Val, Pro, Asp(O^tBu), Glu(O^tBu), Lys(Boc) and Ser(^tBu) were found to be oils). Even the racemization prone Boc/Cbz-Phg-OH (where Phg is phenyl glycine) could form optically pure Boc/Cbz-L/D-Phg-F in satisfactory yields.²⁹ Boc/Cbz-amino acid fluorides showed a characteristic IR absorption near 1847 cm⁻¹.

Table 1 List of Boc- and Cbz-amino acid fluorides^a

Acid fluoride	Yield (%)	mp (°C)	Acid fluoride	Yield (%)	mp (°C)
a * decomposes; mp: melting point.
Boc-Gly-F	89.0	52–54	Boc-Met-F	59.8	32–34
Boc-Ala-F	65.4	62–64	Boc-Phg-F	75.0	54–56
Boc-Val-F	86.8	36–38	Boc-D-Phg-F	72.7	56–58
Boc-Leu-F	68.0	55–57	Cbz-Gly-F	83.3	45–47
Boc-Ile-F	77.3	Oil	Cbz-Ala-F	78.0	39–41
Boc-Pro-F	87.6	Oil	Cbz-Val-F	81.0	Oil
Boc-Phe-F	75.8	66–68	Cbz-Pro-F	82.9	Oil
Boc-D-Phe-F	73.8	67–69	Cbz-Phe-F	73.3	87–88
Boc-Trp-F	78.4	114*	Cbz-D-Phe-F	78.3	84–86
Boc-Ser(Bn)-F	65.8	37–38	Cbz-Met-F	82.5	Oil
Boc-Thr(Bn)-F	93.2	Oil	Cbz-Phg-F	81.4	75–77
Boc-Tyr(Bn)-F	80.0	58–60	Cbz-D-Phg-F	84.7	76–78
Boc-Lys(Cbz)-F	78.0	84–86	Cbz-Lys(Boc)-F	87.5	Oil
Boc-Cys(Bn)-F	67.0	52–53	Cbz-Ser(^tBu)-F	89.6	Oil

---

Similarly, reactions of β-/ω-^tBu esters of Boc/Cbz substituted Asp and Glu (37, 39) with cyanuric fluoride at −30 to −20 °C gave the corresponding acid fluorides (40, 41; Scheme 4). Traces of the formation of Leuch's anhydrides (42) were observed in the case of Boc protection. The pure products were then isolated from the mixture by recrystallisation.

Scheme 4 Synthesis of N^α-Boc/Cbz protected Asp and Glu derived acid fluorides.

Furthermore, N^α-Boc/Cbz protected β- and γ-acid fluorides of α-benzyl Asp and Glu (43 and 44) were also synthesized (Fig. 6).³⁰ Thus, all the four isomers of acid fluorides from Boc/Cbz-Asp and Glu acid esters were made available.

Fig. 6 Side chain acid fluorides derived from benzyl esters of N^α-Boc/Cbz-protected Asp/Glu.

Later, Boc/Cbz-amino acid fluorides were reported in absence of a base using DAST (3) with a yield of 95–80%.⁵⁴ In a recent report, acid fluorides of Boc/Cbz-amino acids were synthesized using a halide exchange reaction of in situ generated acid chlorides using a combination of reagents: trichloroacetonitrile/PPh₃ and TBAF(^tBuOH)₄ (Scheme 5).⁸⁰ In another report, N-Boc/Cbz (2S,4S)-4-fluoropyrrolidine-2-carbonyl fluorides were synthesized in high yields using a double fluorination of N-Boc/Cbz (2S,4R)-4-hydroxyproline using Fluolead reagent (9).⁸²

Scheme 5 Synthesis of protected amino acid fluorides using TBAF(^tBuOH)₄.

3.2.1 N,N-Dialkoxycarbonyl amino acid fluorides [N,N-bis(Boc)/N,N′-Boc,Cbz-amino acid fluorides]. Savrda and Wakselman reported the preparation of N,N-bis(alkoxycarbonyl) amino acid fluorides (47, 48) (N,N-diurethane protected amino acid fluorides or U₂AaaF) from N,N-bis protected amino acids (45, 46) (U₂AaaOH) employing cyanuric fluoride at −50 °C to −30 °C for 90 min (Table 2).^88–90 As these compounds are devoid of exchangeable H atoms on the protected α-amino group, they are useful for the acylation of anionic nucleophiles. Carpino et al. have also reported the synthesis of bis(Boc)-amino acid fluorides and utilized them in the synthesis of [Leu⁵] enkephalin.⁹¹

Table 2 List of N,N-dialkoxycarbonyl amino acid fluorides

Entry	Amino acid	Pg^1	Pg^2	Yield (%)	mp (°C)
1	Gly	Boc	Boc	76 (ref. 88) (80)^91	50–52 (ref. 88) (42–44)^91
2	Phe	Boc	Boc	82 (ref. 88) (80.4)^91	43–45 (ref. 88) (36–38)^91
3	D-Phe	Boc	Boc	70 (ref. 91)	36–38
4	Tyr(Bn)	Boc	Boc	65 (ref. 91)	Oil
5	Asp(OBn)	Boc	Boc	62 (ref. 90)	Oil
6	Lys(Cbz,Boc)	Boc	Boc	85 (ref. 90)	Oil
7	Phe	Boc	Cbz	79 (ref. 88)	Oil
8	Gly	Boc	Cbz	76 (ref. 90)	36–38
9	Glu(O^tBu)	Boc	Cbz	89 (ref. 90)	Oil
10	Leu	Bz	Boc	91 (ref. 90)	38–40

---

N,N-Dialkoxycarbonyl amino acid fluorides were found to withstand purification on silica gel and were isolated as stable compounds. Small and variable amounts of the urethane-N-carboxyanhydrides (UNCAs) formed during acid fluoride formation may also be separated by cooling the hexane treated crude residue in a refrigerator followed by filtration. However, at temperatures above −30 °C, the N-protected NCA by-product becomes significant. As with the acid chloride activation of U₂AaaOH, spontaneous cyclization to the corresponding protected amino acid UNCA was not recorded either in the presence of a base or upon heating. With regard to the stability, the hydrolysis of mono- and bis(Boc)-protected glycyl fluorides was followed using ¹H-NMR spectroscopy.⁹¹ Whereas the mono derivative, Boc-Gly-F in deuterated DMF containing a small amount of water slowly hydrolyzed to the acid, (Boc)₂Gly-F was resistant to hydrolysis over a period of seven days in the same solvent composition.

3.3 Other urethane and non-urethane protected amino acid fluorides

To extend the repertoire of amino acid fluorides as coupling agents and to attain a better combination of reactivity and stability, various other acid sensitive as well as base sensitive N^α-protecting groups have been examined.

Considering the ability of the N^α-Trt group to confer configurational stability to chiral α-amino acid derivatives, Karygiannis et al. reported the synthesis of N^α-Trt amino acid fluorides [(49); Fig. 9]. N^α-Trt amino acids were converted to the corresponding acid fluorides using cyanuric fluoride at −10 °C for 1 h with a yield of 55–75%, with triphenylmethanol being the major by-product (Fig. 9).⁴⁰ The yield could be improved by using Trt resin linked amino acids {o-chlorotrityl chloride resin [PTrt(Cl)-Cl] was used to anchor the amino acids via their alpha amino group}. Thus, PTrt(Cl)-Phe-F was obtained under similar conditions with a high yield (98%). It is important to note that N-tritylpyroglutamic acid fluoride [Trt-Glp-F; Glp = pyroglutamic acid (pGlu); 50; Fig. 7] could be obtained, which is valuable for the incorporation of the Glp moiety in thyrotropin releasing hormone (TRH) and its derivatives. These fluorides are quite soluble in diethylether and they exhibit IR carbonyl band frequencies at around 1834–1842 cm⁻¹. They were crystalline with melting points in the range of 148–170 °C except for Trt-Val-F and Trt-Leu-F, which were found to be oils.

Fig. 7 Trityl-protected amino acid fluorides and the structure of Trt-Glp-F (50).

Bsmoc (51; Table 3),^36,37 Bspoc (52; Fig. 8)^37,38 and Mspoc (53; Fig. 8)³⁹ protected amino acid fluorides with base labile sulfonyl-based N-protecting groups have also been reported using cyanuric fluoride. Mspoc amino acid fluorides³⁹ (53) could be obtained in the crystalline form with greater ease than the foams or amorphous materials isolated in other cases.

Table 3 List of Bsmoc-amino acid fluorides

Amino acid	Yield (%)	mp (°C)	Amino acid	Yield (%)	mp (°C)
Gly	82.7	132–133	Pro	95	Oil
Ala	79.9	118–120	Thr(^tBu)	78	118–120
Phe	85.6	132–133	Asn(Trt)	88.9	Foam
D-Phe	80	Amorphous	Gln(Trt)	87	Foam
Val	81.7	114–115	Trp	80	Foam
Asp(O^tBu)	79.7	107–110	Phg	78	Foam
Glu(O^tBu)	75	Amorphous	D-Phg	81	Foam
Lys(Boc)	93	Foam	Ser(^tBu)	82	Amorphous
Tyr(^tBu)	79.7	Foam	Met	75	Amorphous
Aib	66.7	126–127	Asn(Dmcp)	87.8	Amorphous
Leu	90	Oil	Gln(Dmcp)	93.7	Amorphous
Ile	85	68–70

---

Fig. 8 Structures of Bspoc-/Mspoc-protected amino acid fluorides.

The list of other acid fluorides reported in the literature include N₃-Phe-F (54),⁹² N-formyl-Ser(O^tBu)-F (55), benz[f]indene-3-methyloxycarbonyl-Phe-F (56), N-benz[e]indene-1-methoxycarbonyl-Phe-F (57) and N-benz[e]indene-3-methoxycarbonyl-Phe-F (58) (Fig. 9).³⁸

Fig. 9 Other categories of protected amino acid fluorides.

3.4 C^α,α-dialkyl and N-methyl amino acid fluorides

A list of C^α,α-dialkyl and N-methyl amino acid fluorides is given in Table 4 and these were prepared by similar methods to those described previously and have been utilized successfully in sterically hindered coupling.

Table 4 List of C^α,α-dialkyl and N-methyl amino acid fluorides

Entry	Protecting group	Reaction condition	Acid fluoride
1	Fmoc	Cyanuric fluoride, CH2Cl2, rt	Aib (yield = 74%, mp = 118–120 °C), Nva (yield = 78%, mp = 131 °C), Deg (yield = 74%)^32
			NMeAib (yield = 64%, mp = 174–178 °C), NMeVal (yield = 73%, mp = 158–160 °C), (yield = 68%, mp = 94–96 °C)^35
			Iva (yield = 78%, mp = 131 °C)^33
		DAST, CH2Cl2, 0 °C to rt	Ac6c, Ac5c, Deg^34
			NMeLeu, NMeVal^34
		TFFH, DIEA CH2Cl2	L-(α-Me)Tyr(PO3Bn2), L-(α-Me)Phe(CO2^tBu), L-(α-Me)Phe(CH2CO2^tBu)^86
2	Cbz	Cyanuric fluoride, pyridine, CH2Cl2, rt	L-(αMe)-Val^93
		TFFH, pyridine, CH2Cl2, 0 °C	Aib, L-(αMe)-Val, α-methyl-tert-Leu (yield = 84%, oil)^94
		DAST, CH2Cl2, 0 °C to rt	NMeVal^34 L-Nva, D-Nva^54
		FEP, DIEA, CH2Cl2, rt	NMeVal^75 NMePhe, Iva^95
3	Boc	FEP, DIEA, CH2Cl2, rt	NMeLeu^75
		DAST, CH2Cl2, 0 °C to rt	NMeLeu^34
4	Bsmoc	Cyanuric fluoride, CH2Cl2, pyridine under N2 atm. at rt	Aib (yield = 66.7%, mp = 126–127 °C)^37
5	Tosyl	Cyanuric fluoride, CH2Cl2, rt	Aib^96
6	Azido	DAST, CH2Cl2, 0 °C to rt	Aib^92

---

4. Coupling

The amino acid fluorides have shown remarkable reactivity regardless of the type of N^α-protecting group chosen. They can be termed as near ideal candidates for solution as well as SPPS. They have good solubility in both polar and non-polar solvents and thus, couplings can be carried out in CH₂Cl₂ in place of DMF, as and when necessary.^16,97 The acid fluorides are less reactive^98–101 to water or methanol, whereas in the case of amines, hydroxide or alkoxide ions the increased stabilization of the tetrahedral intermediate at the transition state for substitution because of the enhanced C–F dipole effect will supposedly lead to a higher reactivity. Two-phase couplings of protected amino acid fluorides can be effected in the presence of sodium bicarbonate (NaHCO₃) or sodium carbonate and one-phase solution or solid phase syntheses using a tertiary organic amine as base. The couplings are usually fast and efficient and can be carried out even in the absence of any base.¹⁰² Unlike acid chlorides, side reactions such as oxazolone formation are almost negligible even when the coupling was carried out in presence of a tertiary base.^97,103 Unlike many of the coupling agents and methods, the acid fluorides are particularly useful for the insertion of hindered amino acids into peptides. For hindered coupling, the rate of coupling of acid fluorides can be further accelerated by the use of N-silyl amines.^104–107 As a precedence, silyl derivatives of various aliphatic and aromatic amines have been coupled to acyl fluorides in the presence of catalytic amount of TBAF.¹⁰⁴

4.1 Solution phase synthesis

Protected amino acid fluorides can be coupled to a hydrogen chloride (HCl) salt of amino acid esters (59) in the presence of tertiary bases such as N-methylmorpholine (NMM), DIEA (i-Pr₂NEt) in an organic solvent or under Schotten–Baumann conditions in the presence of an inorganic base in biphasic medium (10% aqueous solution of NaHCO₃) (Scheme 6).²⁶

Scheme 6 Solution phase coupling of protected amino acid fluorides. CHCl₃: choroform.

Carpino studied the possibility of racemization during the coupling of highly sensitive Boc/Cbz-Phg-F with the Ala-OMe using ¹H-NMR.¹⁰⁸ It was found that with either a two-phase or a one-phase coupling protocol, no epimerization (<1%) was detected in the peptides prepared. Also the coupling of Boc-Phg-F or Cbz-Phg-F with Aib in a two-phase system led to no significant racemization (<1%) {peptides were examined for their optical purity using a chiral shift reagent, tris[3-((trifluoromethyl)hydroxymethylene)-(+)-camphorato]europium(III)}.²⁹ However, considerable loss in the optical purity was observed when the same reaction was carried out in a one-phase system in the presence of DIEA as base (13.2–15.3% of Cbz-D-Phg-Aib-OMe). Thus, the two-phase method proved better for difficult couplings and both methods could be used in the case of normal couplings. Also, for hindered couplings, amino acid fluorides have been shown to couple without significant racemization when compared to the use of DCC/dimethylaminopyridine (DMAP), DCC/HOBt, isobutyl chloroformate and N,N′-bis(2-oxo-3-oxazolidinyl)phosphinic chloride (BOP-Cl) methods.¹⁰⁹

4.1.1 Coupling employing N^α-Trt amino acid fluorides. N^α-Trt amino acid fluorides were shown to be efficient acylating agents in DMF in the presence of triethylamine (TEA). No racemization was detected when Trt-L-Phe-F and Trt-D-Phe-F were coupled with H-Ala-NHMe·TsOH (TsOH = p-toluenesulfonic acid) for 10 min at 0 °C and the epimeric dipeptides were obtained in a yield of 85–88%. Furthermore, the coupling of polymeric PTrt(Cl)-Phe-F with H-Ala-NHMe·TsOH followed by detritylation with 20% trifluoroacetic acid (TFA) in CH₂Cl₂ and retritylation with Trt-Cl/TEA resulted in a 93% yield of the expected dipeptide. In another case, Trt-Glp-F (50)⁴⁰ was successfully coupled to H-His-OMe·2HCl (61) in 30 min with a 57% yield of the desired peptide (62) (Scheme 7). Thus, the moiety can be incorporated efficiently into peptide chains enabling the synthesis of hormone TRH and its derivatives.

Scheme 7 Coupling of Trt-Glp-F to H-His-OMe·2HCl.

4.1.2 Coupling using U₂AAFs. Using U₂AAFs could be advantageous in potential racemization problems because the presence of the double N-protection precludes oxazolone formation. However, with the possibility of α-H becoming more acidic because of the presence of a second inductive electron withdrawing carbonyl function (in the form of N-alpha protection) there can still be the risk of racemization. A further consequence is that a severe steric constraint will result in sluggishness in the coupling as seen with the active ester, mixed anhydride, or carbodiimide methods and thus, the activated component might be prone to racemization. On the other hand, the steric factor in diurethane protected amino acid fluorides is compensated for by the small size of the fluorine atom which leads to a fast coupling reaction.

U₂AAFs are particularly useful for the efficient acylation of pyrrole-2 derivatives, in which case the standard coupling procedures including mixed anhydride, DCC or UNCAs have been proved to be inefficient. Thus, U₂AAFs could be efficiently coupled to the sodium salts of pyrrole-2 derivatives to obtain N-(Boc₂-phenylalanyl)-2-pyrrole carboxylic acid benzyl ester (63a), N-(Boc,Cbz-phenylalanyl)-2-pyrrolecarboxylic acid tert-butyl ester (63b), N-(Boc₂-phenylalanyl)-2-pyrrolecarbaldehyde (64) and the depsipeptide ester (65).⁹⁰ These pyrrole peptide analogues possess easily removable protecting groups, thus they can be employed in chain elongation (Fig. 10).

Fig. 10 N-Acylated pyrrole derivatives synthesized using U₂AAFs.

The possibility of racemization under standard conditions of peptide coupling was evaluated using Young's test which involved the coupling of N-Bz-N-Boc-L-Leu-F with H-Gly-OEt·HCl (Bz = benzoyl), followed by removal of the Boc group. The enantiomeric excess (ee) for N-Bz-L-Leu-Gly-OEt was found to be 91.2% and 96.0% in the presence of NMM and TEA, respectively (Table 5).⁹⁰

Table 5 Coupling of N-Bz-N-Boc-Leu-F with HCl·H-Gly-OEt: Racemization study

Entry	Solvent	Base (2.0 equiv.)	Duration of coupling	Yield (%)	EE of L-isomer (%)
1	CH2Cl2	NMM	2 h	67	91.2
2	CH2Cl2	TEA	2 h	60	95.3
3	DMF	TEA	2 h	59	96.0

---

Bis(Boc)-amino acid fluorides have been explored for the successful synthesis of [Leu⁵]enkephalin with a yield of 60%,⁹¹ whereas other methods including active ester, mixed anhydride, carbodiimide procedures were rather slow and had the risk of racemization. The final step involved coupling of (Boc)₂-Tyr(Bn)-F with H-Gly-Gly-Phe-Leu-OBn and the protected pentapeptide [(Boc)₂-Tyr(Bn)-Gly-Gly-Phe-Leu-OBn] was isolated as a oil after purification using column chromatography. No hydantoin by-product was observed that was predominant in previous syntheses. In a racemization study, (Boc)₂Phe-F and its D-enantiomer were separately coupled with the Ala-OMe. Because of the interference of the Boc group signals with the diagnostic methyl group signals, the dipeptide was converted to N-benzoyl dipeptide methyl ester. Examination of the ¹H-NMR spectra showed no DL- or LL-diastereomers in the products (<1%). However, Boc-L-Phg-OBn and its enantiomer both suffered loss of configuration upon conversion to the bis(Boc) derivative.

4.1.3 Rapid, continuous solution-phase synthesis. Fmoc-amino acid fluorides can be used in Fmoc/tris(2-aminoethyl)amine (TAEA) rapid, continuous solution phase synthesis of peptide segments as demonstrated previously for Fmoc-acid chlorides. In continuous solution synthesis of peptides, neither protected nor free peptide fragments would be isolated. Sequential coupling by direct addition of the appropriate acid fluoride would be carried out and both the removal of the excess acid fluoride as well as the deblocking step can be accomplished by the use of TAEA. A dibenzofulvene (DBF) adduct of TAEA and its by-products can be removed by extraction with phosphate buffer (pH 5.5) and the growing peptide chain will be retained in the organic phase. Gram scale synthesis of short and medium sized (up to 22 mer) peptides [human parathyroid hormone (hPTH) fragment 13–34, H-Lys-His-Leu-Asn-Ser-Met-Glu-Arg-Val-Glu-Trp-Leu-Arg-Lys-Lys-Leu-Gln-Asp-Val-His-Asn-Phe-NH₂] were made by using amino acid fluorides in a continuous synthesis.¹¹⁰ The study involved preparation of up to 119 g of acylated tetrapeptide: Ac-Ser-Asp-Lys-Pro-OH, 39 mg of an octapeptide: H-D-PFPhe-Gln-Trp-Ala-Val-D-Ala-His-Leu-OMe (PFPhe = pentafluorophenylalanine) and 37 mg of hPTH 13–34 (70.56% yield, found to be 98% pure using reverse phase HPLC (RP-HPLC) analysis). Also, Fmoc/TAEA rapid solution phase synthesis of the acyl carrier protein (ACP) pentapeptide fragment 65–69 (H-Val-Gln-Ala-Ala-Ile-OH) has been demonstrated under no base conditions (2 mmol scale) with a 74% yield with high purity.¹⁰²

Bsmoc amino acid fluorides were also efficiently employed in rapid continuous solution phase synthesis following a similar protocol to that used for Fmoc amino acid fluorides. The octapeptide, LED-CC-II, which is an insect hormone (Glp-Leu-Thr-Phe-Thr-Pro-Asn-Trp-NH₂; 73) was assembled starting from Bsmoc-Trp-NHDmcp (68) using TAEA for Bsmoc deprotection, except at the stage of the heptapeptide (69) where there was considerable loss of Bsmoc-deblocked free amine intermediate into the aqueous phase.³⁷ In order to overcome this problem, TAEA was replaced with solid supported piperazine (15 equiv.) (70) for deblocking of the heptapeptide, which led to a simple work-up involving filtration and evaporation of solvent (Scheme 8). Coupling with Boc-Glp-F (71) in the next step gave the protected octapeptide (72) with a 58% yield. A study of racemization using ¹H-NMR analysis, involving coupling of Bsmoc-Phg-F with methyl alaninate, was found to be epimerization free.

Scheme 8 Continuous solution phase synthesis of the octapeptide, LED-CC-II.

In addition, several short peptides including Bsmoc-Tyr(^tBu)-Gly-Gly-Phe-Leu-O^tBu (49.1%), Bsmoc-Tyr(^tBu)-Ile-Asp(O^tBu)-Gly-O^tBu (87%), Bsmoc-Tyr(^tBu)-Gly-Gly-Phe-Leu-O^tBu (49%), Bsmoc-Phe-Phe-Val-Gly-Leu-Met-OBn (37%), Fmoc-Ile-Thr(^tBu)-Arg(Pbf)-Gln(Trt)-Arg(Pbf)-Tyr(^tBu)-ODcpm (Dcpm = dicyclopropylmethyl; yield = 40.3%) and Bsmoc-Gly-Phe-Leu-O^tBu (91%) were assembled using the continuous solution phase method. Furthermore, Bsmoc-Leu-Leu-Leu-OFm (Fm = 9-fluorenylmethyl; yield = 67.4%), Bsmoc-Leu-Phe-OFm (73.4%)³⁷ and the synthesis of [Leu⁵]enkephalin using Mspoc-amino acid fluorides were also demonstrated using this method.³⁸

4.1.4 Other coupling protocols. Fmoc-amino acid fluorides can be coupled in the absence of base and even more effectively by silylation of the amine component using a silylating agent such as N,O-bis(trimethylsilyl)acetamide (BSA). Silylation of amines is known to increase the nucleophilicity of the nitrogen nucleophiles.^105–107 BSA has been used for the silylation of amines whenever the use of an organic base needed to be avoided. It acts as a non basic acid scavenger.^105,107 Rajeswari et al. demonstrated amide bond formation under mild conditions by the reaction of N-silylamines with acid fluorides in presence of a catalytic fluoride ion.¹⁰⁴ Kim et al. showed that hydrazinyl methyl ester or a hindered secondary amine (73) can be effectively coupled with N₃-Phe-F (54) (2 equiv.) in the presence of BSA (6 equiv.) at room temperature resulting in an azido protected dipeptide (74) with yields of 46–60% (Scheme 9).¹¹¹ The method consistently gave better yields when compared to HATU and significantly improved the yield of coupled product (74), in the case of BocNHCH(CH₃)₂CH₂-Leu-OMe from 6% to 44%.

Scheme 9 Acylation of sterically hindered secondary amine and acyl hydrazides using a silylation technique.

Scheme 10 Tandem deprotection and coupling of N^α-Alloc-amino acid esters with acid fluorides.

Several other protocols have been developed to alleviate the use of base during the coupling of acid fluorides. Sureshbabu and Ananda employed activated Zn dust for the preparation of di- and tri-peptides using Fmoc-, Boc- and Cbz-amino acid fluorides under non Schotten–Baumann conditions.³⁴ Zn dust was employed as an alternative to the tertiary amine base so that the couplings could essentially be carried out in a neutral condition. The rate of acylation of Fmoc-Val-F to Ile-OMe was found to be quite similar to that when two equiv. of BSA was employed.³⁴ The couplings were fast and yielded optically pure peptides which were determined using ¹H-NMR studies of Cbz-L-Phg-Phe-OMe and Cbz-D-Phg-Phe-OMe.³⁴

In another method, tandem deprotection and coupling of N^α-allyloxycarbonyl (N^α-alloc)-amino acid esters (75) with acid fluorides (29) in the presence of phenylsilane (2 equiv.) and catalytic Pd⁰ [tetrakis-(triphenylphosphine)palladium] led to the desired products (76) in a yield of 72–96% under near neutral conditions (Scheme 10).¹¹²

4.2 Solid phase peptide synthesis

In a remarkable contrast to the sluggish reactivity of Fmoc amino acid chlorides in the solid phase synthesis, Fmoc amino acid fluorides were found to be well suited for SPPS on account of their stability towards basic co-reagents. The rapid acylating ability can be ascribed to the lesser tendency towards oxazolone formation, which aminolizes at a much slower rate leading to the loss of optical integrity. Carpino et al. demonstrated that preformed Fmoc amino acid fluorides were suitable reagents for SPPS through a synthesis of prothrombin (1–9) (H-Ala-Asn-Lys-Gly-Phe-Leu-Glu-Glu-Val-OH; 79).²⁷ The synthesis was carried out manually on a batch synthesizer in DMF solution using 1 g of a TFA sensitive polyamide resin. Except for Asn, which was used as its pentafluorophenyl ester, all the amino acids were incorporated as Fmoc-amino acid fluorides (4 equiv. of acid fluoride, 0.08 M in DMF, 4 equiv. of DIEA) (Scheme 11). Final deblocking and cleavage from the resin yielded 74 mg of the TFA salt of the peptide. Bertho et al. also demonstrated the solid phase synthesis of Leu-Ala-Val-Gly (Merrifield's model tetrapeptide) with a yield of 95% (based on the resin capacity) and a chromatographic purity of 99% using Wang resin.²⁸

Scheme 11 SPPS of prothrombin using preformed Fmoc-amino acid fluorides.

4.2.1 Assembly of difficult sequences: ACP fragment 65–74, magainin-II amide, and human-corticotropin-releasing factor (h-CRF). Acyl carrier protein (ACP) fragment 65–74 (H-VQAAIDYING-OH), a decapeptide sequence, is generally selected as a model sequence to demonstrate the efficacy of newly developed coupling protocols.¹¹³ This sequence is considered to be difficult to make, and is a good example of the sequence dependent problems in the SPPS method.¹¹⁴ Several syntheses have been reported for this model peptide based on activation using DCC, mixed anhydride, p-nitrophenyl ester, trichlorophenyl ester, N-hydroxysuccinimide ester with varied success. However, efficient SPPS for this fragment was demonstrated using Fmoc-amino acid fluorides in a short time using TentaGel® S PHB resin with a good yield of 73% with a purity similar to that obtained using 2-[2-oxo-1(2H)-pyridyl]-1,1,3,3-bis(pentamethylene)uronium tetrafluoroborate (TOPPipU) reagent.³²

Magainin-II amide (H-GIGKFLHGAKKFGKAFVGEIMNS-NH₂) is another example³² reported previously to be a difficult sequence to assemble.¹¹⁵ It was also assembled using Fmoc/acid fluoride strategy with TentaGel® S RAM (TG S RAM) resin in a batch reactor and the yield of crude peptide was found to be 82% (for His and Arg, activation by TOPPipU and two equiv. of DIEA was used).^116,117

Subsequently, a 41 amino acid residue peptide, h-CRF (H-SEEPPISLDLTFHLLREVLFMARAEQLAQQAHSNRKLMFII-NH₂) was assembled and a yield of 76% was obtained using TG S RAM resin and Fmoc-amino acid fluorides.³² Also, a partial sequence (chloroac-Gly-Lys-Phe-Ile-Gly-Phe-Gly-Thr-Asp-Ser-Trp-Val-Tyr-Pro-Asn-Val-Ser-Asn-Pro-Glu-Tyr-Gly-NH₂) of the cyclic nucleotide-gated channel protein chloroac-Gly-BOVTESTIS¹¹⁸ has been assembled via the acid fluoride-multiple peptide synthesis (MPS) technique in similar quality to that of a conventionally made peptide.¹¹⁹ This peptide contains many of the trifunctional amino acids, and thus, its successful synthesis also demonstrated the stability of trifunctional Fmoc-amino acid fluorides for long storage in DMF or during coupling in SPPS through the MPS technique. In other reports, use of the Bsmoc amino acid fluorides were successfully demonstrated in the SPPS of peptides including toxin 2(1–6) of the scorpion Androctonus australis Hector, [Leu⁵]enkephalin, and a fragment of ACP 65–74.^37,120

4.2.2 Peptides containing sterically hindered Aib residues. Incorporation of Aib residues into peptide sequences at the desired position (either at the N- or C-terminus) by conventional methods is often not satisfactory. The protected amino acid fluorides stand out as obvious choice for the insertion of Aib units because of their advantageous small size of the fluoride leaving group and their low tendency to oxazolone formation. However, there is a possibility of side reactions in the presence of tertiary bases because of the longer coupling duration required for hindered acid fluorides. Based on the fact that aliphatic amines readily bind three equivalents of hydrofluoride (HF) (as in the case of Et₃N·3HF),¹²¹ it was reasoned that acid fluoride coupling could be accomplished with one-third of an equivalent of a tertiary amine. Control experiments showed little difference in the rates of coupling of Fmoc-Aib-F with H-Aib-OMe with 0.5, 1 or 2 equiv. of DIEA, although the reaction was somewhat faster and nearer to completion in the latter cases (Table 6).¹⁰² The coupling carried out in the absence of any tertiary base also proceeded at a similar rate, contrary to the presumption that the presence of an equivalent of base was essential for the complete acylation with acid halides. The reaction rate was faster in CH₂Cl₂ than in DMF.

Table 6 Synthesis of Fmoc-Aib-Aib-Me with no base and DIEA

Reactants	Condition	Yield (%)
Fmoc-Aib-F + H-Aib-OMe	No base; 6 h	∼75
	DIEA (0.2 equiv.); 6 h	∼75
	DIEA (0.5 equiv.); 6 h	∼85–90

---

In addition, several other protocols have produced satisfactory yields for the coupling between two Aib residues to form Aib-Aib dipeptides as listed in Table 7.

Table 7 Coupling of Pg-Aib-F with H-Aib-OMe under different conditions

Peptide	Condition of coupling	Yield (%) (ref.)
a Alloc-Aib-OMe as amino component.b Acid fluoride was generated in situ.
Fmoc-Aib-Aib-OMe	Pd/PhSiH3, CH2Cl2, rt, 12 h^a	88 (ref. 104)
	Zn dust, THF, rt	85 (ref. 34)
	2 equiv. DIEA, CH2Cl2, rt, 2 h	>95 (ref. 96)
Cbz-Aib-Aib-OMe	FEP, DIEA, CH2Cl2, rt^b	95 (ref. 75)

---

The very first solid phase synthesis of a hindered amino acid sequence with four consecutive Aib residues was that of decapeptide h-CRF analogue [Aib^32–35]h-CRF fragment 32–41 using TG S RAM resin.³² In a comparative study³² (Table 8) involving the synthesis of [Aib^32–35]h-CRF fragment 32–41 using Fmoc-Aib-F with other methods including a symmetrical anhydride, UNCA, and PyBroP activation¹²² (previously recommended coupling methods for the coupling of sterically hindered residues), only the fluoride protocol was found to successfully incorporate four Aib residues. Other methods led to the heptapeptide with a single Aib unit as the major product under almost identical conditions [for His and Arg, activation by TOPPipU and 2 equiv. of DIEA were used].

Table 8 Synthesis of [Aib^32–35]h-CRF fragment 32–41 using S RAM resin: a comparative study³²

Peptide	Acid fluoride	Amino acid/PyBrop	UNCA	Symmetric anhydride
Coupling concentration	0.2 M DMF	0.3 M in DMF	0.2 M DMF	0.2 M DMF
Base	DIEA (1 equiv.)	DIEA (1.8 equiv.), DMAP (0.2 equiv.)	No base	DIEA (1 equiv.)
Equivalents of amino acid	3	3	3	3
Duration of coupling	15 min	15 min (3 min preactivation)	15 min	15 min
Deprotection condition	20% piperidine/DMF, 15 min	20% piperidine/DMF, 15 min	20% piperidine/DMF, 15 min	20% piperidine/DMF, 15 min
Yield (%)	74	Not reported	Not reported	Not reported

---

4.2.3 Alamethicin and other peptaibols. Peptaibols are peptides with about 20 amino acids and characterized by the presence of a high content (up to 60%) of α,α-dialkyl amino acids such as Aib or Iva. The N-terminus is usually acetylated and an amino alcohol such as phenylalaninol (Pheol), valinol (Valol) or leucinol (Leuol) occupies the C-terminus. The membrane modifying properties of these peptaibols leads to antibiotic activity and has aroused significant interest in their chemical synthesis. Alamethicin, (eight Aib and two Pro residues) the most common among the peptaibols, induces voltage dependent ion-conductivities in lipid bilayer membranes and will lead to cell lysis at high concentrations. The major difficulties encountered for the synthesis of peptaibols through SPPS were incomplete couplings leading to truncated sequences, racemization, instability of the acid labile Aib-Pro linkages and lack of an efficient method for the loading of the C-terminal amino alcohols on to the resins. Thus, earlier syntheses involved tedious procedures involving stepwise solution techniques with chemical or enzymic segment condensations.¹²³

The encouraging result obtained with the synthesis of the h-CRF analogue [Aib^32–35]h-CRF fragment 32–41, has led to the efficient synthesis of alamethicin acid^31,32 (neither of them assembled previously using SPPS). Coupling conditions involved single couplings for 15 min using 2 equiv. of acid fluoride derivative, except for the first amino acid, which was loaded by double coupling (30 min) to TentaGel S AC resin. The crude peptide was obtained with a yield of 84% with remarkable purity, as determined using HPLC and electrospray mass spectrometry (ES-MS). Amino acid analysis of the crude peptide showed a D-content not greater than 0.25%. A comparative study involving alamethicin acid synthesis via symmetric anhydride, UNCA and PyBroP activation produced rather inferior HPLC profiles of crude peptide with no significant amount of the desired peptide. SPPS of alamethicin F30 has also been demonstrated under no base conditions using o-Cl-Trt-resin loaded with Fmoc-Pheol with a good yield and purity similar to that obtained in the presence of an equimolar quantity of DIEA.¹⁰²

Following the successful syntheses of these two difficult peptides, [Aib^32–35]h-CRF 36–41 and alamethicin acid, several peptaibols were assembled on an automated Millipore 9050 peptide synthesizer. Automated SPPS resulted in high purity crude peptaibols with no siginificant racemization. The peptides alamethicin F30 (yield = 78%) (81, Scheme 12) and alamethacin F50 (yield = 75%), saturnisporin SA III (yield = 71%) and trichotoxin A50 (component J; yield = 60%) showed satisfactory purities similar to those obtained with the analogous C-terminal acid analogs synthesized on TentaGel or PEG-PS resins [PEG-PS = poly(ethylene glycol)-poly(styrene)] (Table 9).

Scheme 12 Stepwise automated solid phase synthesis of alamethicin F30 using Fmoc-amino acid fluorides.

Table 9 Solid phase synthesis of naturally occurring peptaibols using Fmoc amino acid fluorides³³

	(a) Alamethicin F30	(d) Saturnisporin SA III acid	(f) Trichotoxin A-50 acid
	(b) Alamethicin acid F30	(e) Saturnisporin SA III	(g) Trichotoxin A-50
	(c) Alamethicin F50
Resin	(a) TG S AC	(d) Fmoc-Phe-PEG-PS	(f) Fmoc-Val-PEG-PS
	(b/c) Fmoc-Pheol-o-Cl-Trt-resin	(e) Fmoc-Pheol-o-Cl-Trt-resin	(g) Fmoc-Valol-o-Cl-Trt-resin
Amino acid fluoride equivalence	(a) 3 equiv.	8 equiv.	8 equiv.
	(b/c) 4.5 equiv.
Base for coupling	(a) 1 equiv. DIEA, 15 min	1 equiv. DIEA, 30 min	1 equiv. DIEA, 30 min
	(b) 1 equiv. DIEA, 30 min
Crude yield (%)	(a) 84	(d) 74	(f) 66
	(b) 78	(e) 71	(g) 60
	(c) 75
Alamethicin F30 Ac-Aib-Pro-Aib-Ala-Aib-Ala-Gln-Aib-Val-Aib-Gly-Leu-Aib-Pro-Val-Aib-Aib-Glu-Gln-Pheol
Alamethicin F50 Ac-Aib-Pro-Aib-Ala-Aib-Ala-Gln-Aib-Val-Aib-Gly-Leu-Pro-Aib-Val-Aib-Aib-Gln-Gln-Pheol
Saturnisporin SA III Ac-Aib-Ala-Aib-Ala-Aib-Aib-Gln-Aib-Aib-Leu-Aib-Gly-Aib-Aib-Pro-Val-Aib-Aib-Gln-Gln-Pheol
Trichotoxin A-50 Ac-Aib-Ala-Aib-Leu-Aib-Gln-Aib-Aib-Aib-Ala-Aib-Aib-Pro-Leu-Aib-Iva-Glu-Valol

---

A study of racemization during peptaibol synthesis by Kusomoto et al. (hydrolysis in the presence of deuterium chloride followed by derivatization and analysis on a chiral gas chromatographic column)¹²⁴ indicated less than 0.8% of the D-form for any of the amino acids. Another study of racemization involving the reaction of Fmoc-Leu-F with H-Pro-NH₂ in DMF/DIEA resulted in 0.25% of the DL-dipeptide, whereas no detectable loss of configuration was observed in DMF/collidine. The use of weakly basic acid scavenger BSA in DMF led to 0.29% racemization. As urethane protected amino acid fluorides are averse to racemization in the presence of tertiary amines, the observed epimerization was attributed to the attack of base on the α-hydrogen atom of the acid fluoride itself. To validate this, coupling of highly sensitive Cbz-Phg-F with H-Pro-NH₂ in DMF was carried out, which resulted in 19.0% DL isomer with DIEA and 11.6% DL form with 2,4,6-collidine.

The remarkable reactivity and high solubility of Fmoc-amino acid fluorides were then exploited for the simultaneous synthesis of various peptaibols using the MPS approach.¹¹⁹ Thus, alamethicin F30 and its six analogues (Fig. 11) were assembled on an ACT 348 multiple peptide synthesizer and the protected peptides were obtained in high purity. MPS was also performed on the ACT 348 multiple peptide synthesizer. Furthermore, solid phase synthesis of an alamethicin analogue, [Ala¹⁴] alamethicin has also been reported using a 100 mg batch of o-Cl-trityl resin.

Fig. 11 Sequences of alamethicin F30 and its six analogues assembled by the MPS approach.

In another study, Nagaoka et al. studied the role of Gln⁷ in the ion channel forming properties of peptaibol trichosporin-B-VIa (TS-B-VIa; Ac-UAUAUUQUIUGLUPVUUQQPheol; 83) by replacing the Gln⁷ residue with Ala, in which the C-terminal fragment involving Aib residues was made through preformed acid fluorides by SPPS (Scheme 13).¹²⁵

Scheme 13 SPPS of a C-terminal fragment of trichosporin-B-VIa.

4.2.4 Microwave enhanced automated synthesis of peptaibols. The duration of coupling during the synthesis of peptaibols can be significantly reduced by the application of microwaves (MWs). A rapid assembly of peptaibols was reported using the MW assisted automated SPPS. In this approach, O-(benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU) was employed for coupling normal amino acids and the acid fluoride technique was used for the coupling of α,α-dialkylated amino acids.¹²⁶ Eight different full length peptaibols with 5–20 residues which include Aib, Iva, and hydroxyproline, were assembled efficiently in short periods of 12 h or less. The peptides synthesized include peptaibolin (Ac-LULU-Phol), trichogin A IV (Oc-UGLUGGLUGI-Lol), trikoningin KB I (Oc-UGVUGGVUGI-Lol), cervinin I (Ac-LUPULUPAUPV-Lol), bergofungin D (Ac-VUUVGLUUOQUOU-Phol), ampullosporin I (Ac-WAUULUQUUUQLUQ-Lol), tylopeptin A (Ac-WVUJAQAUSUALUQ-Lol) and alamethicin F30 (Ac-UPUAUAQUVUGLUPVUUEQ-Phol) (Scheme 14). Thus, with the application of MWs, duration for the total assembly of alamethicin F30 was reduced to 12 h compared to long hours required in a conventional automated synthesizer or under no base conditions.

Scheme 14 Microwave enhanced SPPS of peptaibols.

4.2.5 SPPS using acid fluorides generated in situ. Fluoroformamidinium salts can be used as agents for the in situ generation and coupling of amino acid fluorides in the solid phase. For the couplings involving optically sensitive as well as sterically hindered amino acids, in situ generation of acid fluorides is an effective method to render efficient coupling. In addition, unstable acid fluorides of His and Arg [Fmoc-His-(Trt)-F and Fmoc-Arg(Pbf)-F] could be generated as transient intermediates and then coupled efficiently. The superiority of fluoro derivatives among the haloformamidinium salts was illustrated by the coupling of Fmoc-Val-OH to resin bound Ile (on PEG-PS), only TFFH resulted in 100% coupling (as measured by ultraviolet analysis) after 10 min as compared to its analogs tetramethylchloroformamidinium hexafluorophosphate (TCFH; 86% coupling) and tetramethylbromoformamidinium hexafluorophosphate (TBFH; 79% coupling).

Fmoc-amino acid fluorides generated in situ using TFFH were first explored for the difficult synthesis of pentapeptide [Aib^2,3, Leu⁵]enkephalin (H-Tyr-Aib-Aib-Phe-Leu-NH₂), where the peptide was obtained in a yield of 88% (purity of crude product = 92%).²⁷ Furthermore, H-Tyr-Pro-Pro-Phe-Leu-NH₂ (yield = 78.1%; purity = 97.6%), ACP fragment 65–74 (yield = 87%; purity = 92%), prothrombin 1–9 (yield = 75%; purity = 95%) and magainin II amide (yield = 72%; purity = 82%) were assembled using TFFH (4–6 equiv.) with amino acid (4–8 equiv.), DIEA (10 equiv.) and 30 min of coupling time. However, the use of preformed, protected acid fluorides was found to provide products in higher yield and purity, particularly in the case of hindered amino acids.

Other variants of TFFH have also been utilized for the in situ generation and coupling of acid fluorides in SPPS. BTFFH has been successfully employed for the synthesis of magainin I amide, bradykinin amide, ACP fragment 65–74, prothrombin amide, human preproenkephalin 100–111, insulin B-chain 19–25, and substance P.⁶⁴ DMFH was found to be superior to HATU, for the coupling of hindered Aib amino acid to a tripeptide, Aib-Phe-Leu and the tetrapeptide was synthesized on a solid phase with a yield of 99% compared to a yield of 68% for HATU.

As noted previously, for the coupling of α,α-dialkylated amino acids, e.g., Fmoc-Aib-OH, the activation by TFFH alone proved inefficient when compared to the use of isolated amino acid fluorides. The deficiency was traced to the incomplete conversion to acid fluoride, accompanied by the formation of symmetric anhydride and oxazolone in the presence of two equiv. of DIEA. These side reactions could be avoided with the use of PTF together with TFFH.⁷⁶ Thus, the coupling of Fmoc-Aib-OH to H-Aib-Gln-Gln-Phe-TG-SAC with TFFH (5 equiv.)/DIEA (2 equiv.) at a concentration of 0.3 M led to a yield of 89.7% of the target peptide and a yield of 10.3% of the des-Aib peptide. Whereas, the use of TFFH/PTF resulted in a 97.2% yield of the desired peptide and only a 2.8% yield of des-Aib peptide was formed. Also assembly of H-Tyr-Aib-Aib-Phe-Leu-NH₂ using TFFH/PTF gave a pentapeptide with a purity similar to that obtained via isolated acid fluorides.⁷⁶

In other cases, anionic poly(hydrogenfluoride) additives (PTF) in combination with the peptide coupling agents [N-HBTU, N-HATU, 1-(1-pyrrolidinyl-1H-1,2,3-triazolo[4,5-b]pyridin-1-ylmethylene)pyrrolidinium hexafluorophosphate N-oxide (N-HAPyU) and carbodiimide] have been employed to generate acid fluorides in situ.⁷⁶ It has been found that reagents including N-HBTU or N-TBTU employed in combination with PTF resulted in peptides of equal or greater quality than those obtained via N-HATU. Thus, synthesis of pentapeptide H-Tyr-Aib-Aib-Phe-Leu-NH₂, Aib^67,68ACP(65–74) amide (H-Val-Gln-Aib-Aib-Ile-Asp-Tyr-Ile-Asn-Gly-NH₂) and alamethicin amide (Ac-Aib-Pro-Aib-Ala-Aib-Ala-Glu-Aib-Val-Aib-Gly-Leu-Aib-Pro-Val-Aib-Aib-Glu-Gln-Phe-NH₂) were successfully obtained by this method and the purity of the peptides was of equal or better quality than those obtained with the use of O-(7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HATU) or acid fluorides alone.

Fluoride additives were also found to divert the reaction pathway for carbodiimide mediated couplings. The base accelerated stereomutation in such couplings can be controlled using PTF in combination with carbodiimide. In the presence of PTF, the acid fluorides are generated via the labile o-acylisourea intermediate (16; Scheme 15), which would otherwise get converted to the active species symmetric anhydride (86). Excellent synthesis of [Aib^67,68]ACP decapeptide amide fragment 65–74 was carried out using DIC/PTF activation in CH₂Cl₂, followed by coupling in DMF. Use of DIC activation in the absence of PTF resulted only in the des-Aib nonapeptide.⁷⁶

Scheme 15 Carbodiimide method of peptide activation and diversion to acid fluoride using the PTF additive.

4.3 Coupling involving extremely hindered amino acids

C^α-tetrasubstituted amino acids are particularly important because of their ability to induce helical conformation even in short sequences (two helical turns for seven to eight amino acids).^93,127–129 Recently, Polese et al. synthesized terminally blocked, isotactic homopeptides from L-(αMe)Val {X-[L-(αMe)Val]_n-O^tBu; n = 2–8, X = Cbz, Ac, pBrBz} via preformed Cbz-L-(αMe)Val-F and exploited them as rigid and precise molecular rulers in spectroscopic studies.⁹³ The acid fluoride method gave higher yields for the difficult coupling between two sterically demanding L-(αMe)Val residues than the EDC/HOBt [EDC: 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride], EDC/HOAt (HOAt: 1-hydroxy-7-azabenzotriazole) or the symmetrical anhydride method.

The coupling of moderately hindered amino acid fluorides could be facilitated by the no base condition as this precludes the premature deblocking of the base labile Fmoc group and oxazolone formation. Consequently, coupling duration can also be extended to ensure efficient as well as complete coupling. However, for the incorporation of sterically hindered amino acids, couplings were found to be very slow, for example, coupling of Fmoc-Deg-F with HF·H-Aib-OMe in CH₂Cl₂ under no base conditions led to a yield of only 0.5% of the product after 2 h.³⁵

Fiammengo et al. studied the mechanistic course of the reaction during the formation and coupling of extremely hindered amino acid fluorides via TFFH activation and found that two paths were operative.⁹⁴ The study was carried out in dichloromethane-d2 (CD₂Cl₂) at 25 °C in the presence of a slight excess of pyridine and 1,2-dichloroethane as internal standard for ¹H-NMR.

The reaction of TFFH with proteinogenic Cbz-protected amino acids was found to proceed smoothly to give acyl fluoride as the only product. However, in the case of α-methyl-α-alkyl substituted amino acids, an influence of the bulkiness of the substituents at C^α on the rate (Me < iso-propyl < tert-butyl) and accumulation of oxazolidinone and NCA was observed. Determination of the rate constants indicated that the formation of acid fluoride in the case of C^α-tetrasubstituted amino acids was appreciable only with the Cbz-Aib (a favorable gem dialkyl effect), whereas oxazolone formation is largely favored with the Cbz-α-methyl-Val and Cbz-α-methyl-tert-Leu. However, the course of TFFH mediated fluorination of Cbz-(αMe)Val (Scheme 16) as profiled using ¹H-NMR studies revealed that the addition of a slight excess of HF/pyridine (1 : 1) complex after 21 min accelerated the conversion of oxazolone (91) into an acid fluoride at a higher rate. Here the acid catalyzed nucleophilic F⁻ attack on the oxazolone was highly accelerated compared to the acid catalyzed formation of the corresponding NCA (89).

Scheme 16 Reaction pathways in the TFFH activation and coupling of Cbz-(αMe)Val.

A more demanding test for the efficacy of acid fluoride activation would be its application in the assembly of very difficult peptides that contain combinations of C^α,α-dialkyl amino acids, N-alkyl amino acids and N-alkyl-α,α-disubstituted amino acids.³⁵ The problem of peptide coupling with extremely hindered amino acids has been approached through various protocols such as silylation of the amino acid esters prior to acylation,³⁵ Zn dust mediated reaction,³⁴ tandem deprotection and coupling of (N^α-Alloc)-amino acid esters,¹¹² and coupling of acid fluorides with free amino acids in hexafluoroisopropanol (HFIP).¹³⁰

4.3.1 Coupling of C^α,α-dialkyl amino acid fluorides bearing hindered carboxyl group. Carpino et al. studied the coupling of severely hindered α,α-dialkylated and N-methylated amino acid fluorides with the moderately hindered H-Aib-OMe by the application of the silylation technique.³⁵ Although the coupling of the acid fluorides of Aib and NMeGly proceeded without difficulty in the presence of 2 equiv. of DIEA, the rate of acylation was significantly reduced when Fmoc-Iva-F and Fmoc-NMeVal-F were employed and considerable premature deblocking of the Fmoc group (9.2–9.6%) was observed. With the more hindered Deg and NMeAib very decreased yields were observed (Fmoc-Deg-Aib-OMe, yield = 18%; Fmoc-NMeAib-Aib-OMe, yield = 12%). Under these circumstances, because of the low coupling rate, premature Fmoc-deblocking occurred as a prominent side reaction to an extent of 25.5% and 40.7%, for Deg and NMeAib, respectively, after 4 h. The reduction in the efficiency of coupling for highly hindered systems was traced to the formation of oxazolone (15% in the case of Deg after 4 h as revealed by IR studies) and this was attributed to the gem dialkyl effect. Also, α,α-disubstituted amino acid fluorides were found to undergo ready conversion to the corresponding oxazolones in the presence of fluoride ions acting as the base.

Consequently, the HCl salt of the amino component was first converted into the corresponding HF salt (94)³⁵ and the couplings of Fmoc-Deg-F and Fmoc-NMeAib-F to H-Aib-OMe were carried out under no base conditions in CH₂Cl₂. Although Fmoc-deprotection was reduced, the condition proved inefficient with a very slow rate of coupling. Addition of 1 equiv. of DIEA could accelerate the rate of coupling without additional Fmoc-deprotection, albeit with a yield of <20% of dipeptide after 2 h. The modification of conditions involving the prior treatment of the amino acid ester HF salt (94) with 4 equiv. of BSA (95) overnight resulted in a simple silyl-proton exchange reaction (Scheme 17) and led to the efficient coupling resulting in a ∼50% yield of Fmoc-Deg-Aib-OMe. The use of BSA almost eliminated the DBF formation (0% after 24 h). Following this method Fmoc-Iva-Aib-OMe, Fmoc-NMeGly-Aib-OMe, Fmoc-NMeVal-Aib-OMe, and Fmoc-NMeAib-Aib-OMe were synthesized (yields were not given).

Scheme 17 Coupling of extremely hindered amino acid fluorides to a silylated amino acid ester using BSA.

Incorporation of sterically hindered α,α-dialkyl amino acids such as Ac₅c, Ac₆c, and Deg into peptides using activated Zn dust method has been reported (Table 10). The couplings were facile and high yielding, without any side reactions leading to the formation of oxazolone or DBF.³⁴

Table 10 List of sterically hindered peptides prepared through coupling of C^α,α-dialkyl amino acid fluorides under different conditions

Entry	Peptide	Condition of coupling	mp (°C)	Yield (%) (ref.)
1	Fmoc-Deg-Aib-OMe	4 equiv. BSA, CH2Cl2, rt, 2 h		∼52 (ref. 35)
2	Fmoc-Ac5c-Ac5c-OMe	Zn dust, THF, rt	159–160	75 (ref. 34)
3	Fmoc-Ac6c-Ac6c-OMe	Zn dust, THF, rt		65 (ref. 34)
4	Fmoc-Deg-Deg-OMe	Zn dust, THF, rt	122–125	70 (ref. 34)
5	Fmoc-(Ac6c)3-OMe	Zn dust, THF, rt	185–188	69 (ref. 34)

---

4.3.2 Coupling of acid fluorides to hindered amino groups of N-akyl amino acids/esters. Coupling to the amino group of NMeAib is at least an order of magnitude more difficult than coupling to its carboxyl function. The silylation method has been extended for the efficient coupling of less hindered acid fluorides to hindered amino groups such as H-NMeAib-OMe.³⁵ Thus, Fmoc acid fluorides of Ala, Val, Phe were efficiently coupled to silylated H-NMeAib-OMe in the presence of 1 equiv. of DIEA. The previous methods involved high temperatures (50 °C) or long reaction times (seven days) in combination with a large excess of UNCAs.¹³¹ Similarly, the Zn dust method could be applied for coupling involving N-methyl amino acid esters to obtain good yields of dipeptides.³⁴ (N^α-Alloc)-NMe-amino acid esters were coupled with Fmoc-amino acid fluorides using Pd(PPh₃)₄ as catalyst in the presence of PhSiH₃ and this gave good yields. High yields (>90%) for dipeptides containing N-methyl amino acids have been reported using FEP reagent as well (see Table 11 for a list of peptides prepared under these conditions).

Table 11 List of sterically hindered peptides prepared through coupling acid fluorides to N-alkyl- and N-alkyl-C^α,α-dialkyl amino acid/esters^a

Entry	Peptide	Condition of coupling	Yield (%) (ref.)
a THF = tetrahydrofuran; Tic = tetrahydroisoquinoline-3-carboxylic acid.
1	Fmoc-Ala-NMeAib-OMe	DIEA, 2 equiv. BSA, CH2Cl2, rt, 2 h	>90 (ref. 35)
		Cat. Pd^0/PhSiH3, CH2Cl2, rt, 12 h	76 (ref. 111)
2	Fmoc-Phe-NMeAib-OMe	DIEA, 2 equiv. BSA, CH2Cl2, rt, 2 h	>90 (ref. 35)
		Cat. Pd^0/PhSiH3, CH2Cl2, rt, 12 h	65 (ref. 111)
3	Fmoc-Val-NMeAib-OMe	DIEA, 2 equiv. BSA, CH2Cl2, rt	>80 (ref. 35)
4	Boc-Val-NMeLeu-Ala-OBn	Zn dust, THF, rt	76 (ref. 34)
5	Cbz-NMeVal-NMeVal-OMe	Zn dust, THF, rt	75 (ref. 34)
6	Boc-Val-NMeVal-OMe	FEP, DIEA, CH2Cl2, rt, 4 h	92 (ref. 75)
7	Fmoc-NMeLeu-NMeVal-OMe	FEP, DIEA, CH2Cl2, rt, 4 h	95 (ref. 75)
8	Fmoc-NMeVal-Sar-OH	HFIP, 55 °C, 5 min	78 (ref. 130)
9	Fmoc-Ac5c-Sar-OH	HFIP, 55 °C, 45 min	78 (ref. 130)
10	Fmoc-NMeVal-NMeVal-OH	HFIP, 55 °C, 5 min	78 (ref. 130)
11	Fmoc-Ac5c-NMeVal-OH	HFIP, 55 °C, 45 min	74 (ref. 130)
12	Fmoc-Aib-NMeAib-OH	HFIP, 55 °C, 1 h	60 (ref. 130)
13	Fmoc-Ac5c-NMeAib-OH	HFIP, 55 °C, 45 min	60 (ref. 130)
14	Fmoc-NMeVal-(S)-Tic-OH	HFIP, 55 °C, 5 min	80 (ref. 130)
15	Fmoc-Aib-(S)-Tic-OH	HFIP, 55 °C, 1 h	86 (ref. 130)
16	Fmoc-Ac5c-(S)-Tic-OH	HFIP, 55 °C, 45 min	79 (ref. 130)

---

In an interesting study, Brown and Schafmeister investigated an intramolecular acyl transfer approach for the synthesis of extremely hindered peptide acids by coupling Fmoc-amino acid fluorides with the weakly nucleophilic amines of N-alkyl amino acids.¹³⁰ Fmoc-amino acid fluoride (99) was combined with an excess (4 equiv.) of unprotected amino acid (100) dissolved in HFIP.

The amino acid solution in HFIP was warmed to 55 °C and then added to the Fmoc amino acid fluoride and the mixture was maintained at 55 °C until the completion of the reaction (5–60 min) (Scheme 18). The product (101) was isolated by subjecting the reaction mixture to reverse phase-HPLC (RP-HPLC). HFIP is used as the solvent in these reactions, as it dissolves both the amino acid and Fmoc-amino acid fluoride. However, HFIP reacts with the acid fluoride to produce small amounts of HFIP ester. Thus, it is important to avoid base in these reactions, as its presence will lead to quantitative esterification with the solvent. A competition experiment involving the reaction of Fmoc-Aib-F with 4 equiv. each of H-NMeVal-OH and H-NMeVal-OMe·HCl in HFIP in one pot resulted in the products: Fmoc-Aib-NMeVal-OH/Fmoc-Aib-NMeVal-OMe in the ratio of 97 : 3.

Scheme 18 Coupling of hindered amino acid fluorides to unprotected amino acids in HFIP solvent.

Based on the observation that the degree of steric crowding of the amino group of the amino acid had little effect on the yield of the dipeptide, Brown and Schafmeister have proposed the mechanism involving the attack of the carboxylic acid of the amino acid (102) on the acid fluoride (103) to form a transient amino-anhydride (104) similar to that of an Ugi intermediate (106).¹³² The anhydride then spontaneously rearranges through a five membered ring transition state to form the hindered amide bond (105) in an intramolecular O,N-acyl transfer (Scheme 19).

Scheme 19 Mechanism of extremely hindered peptide acid synthesis via a Ugi type intermediate.

4.3.3 Limiting cases of extremely hindered coupling: acid fluorides versus acid chlorides. It is worth recognizing that the effectiveness of the carboxyl activation is a consequence of the factors, which arise from the nature of the carboxyl activation as well as the type of the N^α-protecting group. Simple couplings proceed without experiencing the effect of the N^α-protecting group. However, when the couplings are sterically disfavored, such as those which involve severely constrained amino acids, this effect gets magnified and becomes limiting.

Carpino et al. made significant observations on the comparative reactivities of differentially N-protected acid fluorides and acid chlorides in the context of couplings, nameley, Aib to NMeAib and NMeAib to NMeAib.⁹⁶ The coupling of Fmoc-Ala-Cl to H-NMeAib-OMe in the presence of BSA was completed within 30 min to 1 h, which otherwise is greatly inhibited by the corresponding fluoride. However, the Aib-Aib coupling via the acid chloride was highly sluggish because of the oxazolone effect when compared to the easy coupling using acid fluoride under the same conditions. Thus, for less hindered amino acid chlorides, the oxazolone formation overpowers the coupling reaction, whereas for highly hindered substrates, chlorides clearly dominate the coupling reaction compared to fluorides, in which case the reaction is sluggish.

As described before, when Fmoc-NMeAib-F, was made to couple to H-Aib-OMe, 40.7% of DBF was formed. However this coupling could be successfully achieved in the presence of a silylating agent. However, no coupling could be achieved between Fmoc-Aib-F or even Fmoc-Aib-Cl and H-NMeAib-OMe in the presence of either DIEA or BSA. Thus, the limit for the urethane-based amino acid chloride coupling has been set with the hindered Fmoc-Aib-Cl coupling to H-NMeAib-OMe in the presence of DIEA or BSA. So the reactivity of urethane protected acid chlorides with highly hindered amino acid coupling was reconsidered. This led to the use of the inductive effect of the sulfonyl protecting groups to enhance the reactivity of acid chlorides and prevent oxazolone formation. As expected, Ts-Aib-Cl (108) reacted readily with the H-NMeAib-OMe (109) provided that non-aqueous media was used to avoid degradation, but the corresponding acid fluoride (111) failed to acylate H-NMeAib-OMe (Scheme 20). The use of sulfonyl protector is finally exemplified by the assembly of extremely hindered Pbf-NMeAib-NMeAib-OMe with a yield of 63% using the more easily deblocked Pbf group.

Scheme 20 A comparison of the reactivity of Ts-Aib-Cl versus Ts-Aib-F with H-NMeAib-OMe.

5. Peptide acid fluorides

5.1 Segment condensation

The use of peptide acid fluorides in segment condensation has not been explored to its full potential yet. In the past, the acid azide method, which involved preparation of an activated peptide acid through a multi-step protocol, was predominantly employed for segment condensations.¹³³ However, segment condensation could be effected easily once the in situ conversion of peptide acid to its acid fluoride was possible.

In the limited work carried out in this direction, TFFH has been employed with an equimolar quantity of HOAt. The latter is added to prevent the epimerization (from the easy formation of oxazolone intermediates) which is common in activation of peptide acids. Coupling of Cbz-Phe-Val-OH to H-Ala-OMe·HCl in DMF using TFFH alone in the presence of DIEA, NMM, PS or 2,2,6,6-tetramethylpiperidine (TMP), showed a high degree of epimerization (25%, 23%, 8% and 6%, of the LDL isomer was found, respectively). Epimerization could be reduced to 2.0% by using the TFFH/HOAt/DIEA system or to a minimum of less than 0.1% with the use of TFFH/HOAt/TMP.^47,134

In a synthesis of peptidyl methylcoumarin esters for use as substrates and active site titrants for the prohormone processing proteases Kex2 and proprotein convertase 2 (PC2),¹³⁵ attempts to couple the methylcoumarin (α-amino) esters (amino MCEs) to tripeptides under standard segment coupling conditions led to a poor yield and high levels of epimerization. After an extensive survey of coupling reagents and protocols, the optimal results were obtained by the TFFH activation of tripeptide and the epimerization could be brought down to less than 13%. The protocol was then employed successfully in the synthesis of a number of tetrapeptidyl MCEs, including Cbz-Nle-Tyr-Lys-Lys-MCE, Cbz-Ala-Tyr-Lys-Lys-MCE (Nle = norleucine; MCE = 4-methylcoumarin ester), Cbz-Nle-Tyr-Lys(Boc)-Arg(Mtr)-MCE (Mtr = 4-methoxy-2,3,6-trimethylphenylsulfonyl), Cbz-Nle-Tyr-Lys-(D-Lys)-MCE, and Cbz-(D-Nle)-Tyr-Lys-Lys-MCE.

Li and Xu have demonstrated 2-halo-pyridinium based reagents FEP and FEPH for the [2 + 1] segment condensation of Cbz-Gly-Phe-OH with H-Val-OCH₃·HCl.⁷⁵ Both these reagents showed higher reactivities and lower levels of racemization than the halogenated uronium and phosphonium salts, such as PyBroP, PyClU, BTFFH and BOP-Cl. The use of HOAt as an additive, further suppressed the loss of configuration because of its anchimeric assistance effect. Similar segment condensation using FEP in the presence of HOAt gave Cbz-Val-Leu-Ala-O^tBu with a yield of 97%.

In a study pertaining to the synthesis and conformation of Ac-Tyr-dmP-Asn (dmP = 5,5-dimethylproline), where dmP was substituted for Pro to lock the conformation in cis form in a tripeptide fragment 92–94 Tyr-Pro-Asn of bovine pancreatic ribonuclease A, coupling of Ac–Y(O^tBu)dmP-OH to an equimolar amount of NH₂–N(Trt)NovaSyn® resin using the TFFH/DIPEA method gave a higher yield (75%) than the O-benzotriazol-1-yl-1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU)/HOBt method (yield = 58%).¹³⁶

Peptide acid fluorides were also generated and isolated by employing pyridine-hydrogen fluoride Py(HF)_n in tandem with DIC. Cbz-Gly-Ala-F (113) was obtained from Cbz-Gly-Ala-OH (112) by using five equiv. of Py(HF)_n, which was stable for several hours at room temperature. Coupling to leucine p-nitroanilide (114) in the presence of BSA (1 equiv.) for only 2 min resulted in a tripeptide anilide (115) with a yield of 97% (Scheme 21). Only 0.3% of the D-Ala form was observed in the product, whereas in the absence of Py(HF)_n, there was 1% of the D-Ala form.⁷⁶

Scheme 21 Coupling of Cbz-Gly-Ala-F to leucine p-nitroanilide in the presence of BSA.

An application of polyfluoride additive was also reported in the [4 + 6] segment condensation leading to the assembly of a protected ACP decapeptide amide fragment 65–74 (118) (Scheme 22). The desired peptide was obtained successfully using 10 equiv. of HF complex with 1-(1-pyrrolidinyl-1H-1,2,3-triazolo[4,5-b]pyridin-1-ylmethylene)pyrrolidinium hexafluorophosphate N-oxide (N-HAPyU)/DIEA with only 0.5% loss of configuration at Ala (as determined by the method of Kusomoto et al.¹²⁴) in contrast to 1.5% stereomutation in the absence of Py(HF)_n.

Scheme 22 Assembly of a protected ACP decapeptide amide fragment 65–74 via [4 + 6] segment condensation.

5.2 Macrocyclisation

The use of acid fluorides in macrocyclisation has been illustrated in the preparation of cyclosporine A derivatives¹³⁷ which exhibited non-immunosuppressive and anti-HIV activity and in the synthesis of a 14 membered cyclic enamide of C3-epimauritine D.¹³⁸ Three macrocyclic dilactones have also been made using Cbz-glutamyl difluoride as well as Cbz-cystyl difluoride.⁸⁷

5.2.1 Cyclosporine A derivatives. Hubler et al.¹³⁷ reported an efficient synthesis of [NEtXaa⁴] cyclosporin A analogues (120) starting from cyclosporine A in 10 steps and with overall yields of up to 27%. In an initial strategy, the cyclization was attempted by the coupling of a sterically hindered N-ethyl residue to the activated sarcosine residue. The powerful coupling reagents used for hindered couplings such as BOP-Cl, PyBOP, [(7-azabenzotriazol-1-yl)oxy]tris(pyrrolidino)-phosphonium hexafluorophosphate (PyAOP), TFFH and alternative bases including DIEA, 1,8-diazabicyclo[5.4.0]undec-7-ene, and DMAP proved unsuccessful for cyclization. Thus, the strategy was modified such that the penultimate step involved the cyclisation of the linear peptide (119) through the coupling of the Val⁵ residue to the activated ester of the protected N-ethyl residue employing TFFH in combination with sym.-collidine as a weak base (Scheme 23). This protocol resulted in efficient cyclisation without any detectable epimerization. On the other hand, cyclisation of undecapeptide to activated N-ethyl-Ile using PyAOP/DMAP resulted in a partial epimerization at the N-ethyl residue.

Scheme 23 Synthesis of [NEtXaa⁴] cyclosporin A analogue via acid fluoride mediated macrocyclization. NaOCH₃ = sodium methoxide CH₃OH = methanol.

5.2.2 Cyclic enamide of C3-epimauritine D. Kim et al. reported the synthesis of a key intermediate cyclic enamide of C3-epimauritine D by macrocyclisation at the C1–N14 site using TFFH in the presence of HOAt.¹³⁸ Cyclization of the intermediate (121), was attempted (after removal of the Cbz group) using activated pentafluorophenyl (Pfp) ester, but the method failed. Other reagents including BOP-Cl, diphenylphosphoryl azide and BOP did not give the expected product either. The desired macrolactam (122) was then obtained in a yield of 75% using an in situ generated acid fluoride using TFFH/DIEA mediated by HOAt at 0 °C (Scheme 24). However, the product was obtained as a mixture of two diastereomers (the ratio of the diastereomers was not given).

Scheme 24 Synthesis of a cyclic enamide of C3-epimauritine D by macrocyclization using TFFH.

6. Applications

6.1 Acylating agents

Protected amino acid fluorides serve as efficient acylating reagents for the preparation of various amino acid derivatives. These include Weinreb amides, hydroxamic acids, α-amino boronic acid derived peptides, nucleosides, glycopeptides, and so on (Scheme 25).

Scheme 25 Protected amino acid fluorides used as acylating agents in the synthesis of various amino acid derivatives. DIPEA = DIEA.

6.1.1 Weinreb amides, hydroxamic acids, anilides and azides. Georg's group^57,58 employed Deoxo-Fluor for an easy one-pot preparation of amides, Weinreb amides, and peptides with good yields. Boc-amino acids were reacted with Deoxo-Fluor and then treated in situ with N,O-dimethylhydroxylamine and amino esters to yield Weinreb amides (123) and peptides, respectively, with good yields. The preparation was reported to be devoid of racemization.

Singh and Umemoto reported the stereospecific synthesis of N-protected (2S,4S)-4-fluoropyrrolidine-2-carbonyl fluorides (129) in high yields starting from (2S,4R)-4-hydroxyproline (128) using Fluolead (9) (Scheme 26).⁸² 4-Fluoropyrrolidine derivatives were then employed for the synthesis of optically pure Weinreb amides (130), 4-fluoropyrrolidine-2-carboxamides (131a), carboxylate esters (131b) and carbonitriles. The 4-fluoropyrrolidine derivatives are useful intermediates for the synthesis of medically important compounds such as dipeptidyl peptidase IV inhibitors.

Scheme 26 Synthesis of a 4-fluoropyrrolidine derived Weinreb amide, amides, and esters.

Hydroxamic acids (124) were also synthesized from in situ generated acid fluorides using TFFH, with or without PTF, and hydroxylamine hydrochloride in the presence of TEA or DIEA.⁴⁸ The use of PTF resulted in a marginal increase in the yield of the products. The products were isolated with good yields and high optical purity. Also, Cbz or Bz protected amino acids were converted to their respective anilides (127) using TFFH activation.¹³⁹

6.1.2 Lysine analogues. Lysine analogues have been introduced into pseudopeptides via the acyl fluoride.¹⁴⁰ Various 6-amino-2-substituted hexanoic acid derivatives have been prepared from lysine and lysyl-proline via a triflate of 6-(benzyloxycarbonylamino)-2-hydroxyhexanoic acid derivatives. These triflate derivatives were known to be more reactive than the corresponding bromide derivatives as demonstrated by Effenberger et al., thus, their nucleophilic substitution would not cause racemization.¹⁴¹ The triflates were then treated with various N-nucleophiles to give the 2-substituted derivatives. The secondary amines thus obtained were coupled with Boc-Asp(OBn)-OH using TFFH as an in situ fluorinating reagent to obtain various pseudopeptides. Only the acid fluoride protocol gave consistently good yields (60–80%) whatever the amino component, as compared to other activation methods screened (PyBroP, PyBOP, mixed anhydride).

6.1.3 α-Amino boronic acid derived peptides. Hong and Morken prepared enantiomerically enriched α-amino boronates using a platinum-catalysed asymmetric addition of B₂(pin)₂ across in situ generated silyl amines. Here the intermediate diboration adduct could be readily acylated to N-acyl-α-amino boronic acid derivatives.¹⁴² Thus, a tandem diboration/acylation sequence employing Boc-amino acid fluoride in the presence of pivalic acid led to peptidic amino boronates (126) with good yields (58–76%) and a high diastereomeric ratio (10 : 1 or 20 : 1). The resulting pinacol boronates underwent an easy conversion to the corresponding boronic acids on treatment with boron trichloride.

6.1.4 Aminoacylation of nucleosides. Aminoacylated nucleosides and oligonucleotides are the key intermediates in the synthesis of aminoacylated tRNAs and show pharmacological activity. Methods for the aminoacylation of nucleosides and nucleoside derivatives are known to be tedious and low yielding. Oliver and Oyelere found that Fmoc amino acid fluorides proved to be superior reagents for this purpose. Thus, the acylation of 2-deoxyribonucleosides (132) and ribonucleosides (134) proceeded in high yields when two equivalents of Fmoc amino acid fluorides were used in the presence of pyridine (Scheme 27).¹⁴³ In the case of ribonucleoside, acylation at the 2′- or 3′- position gave a single monoacylated product (135) (with the amino acid rapidly exchanging between the two hydroxyl groups). The yield of monoacylated product increased with the increase of acid to the extent of 1.5 equivalents and above this amount formation of the diacylated nucleoside (136) became significant. The mono- and diacylated products could be separated by chromatography.

Scheme 27 Amino acylation of 2-deoxyribonucleosides and ribonucleosides.

6.1.5 Glycopeptides. Acid fluoride activation was employed successfully in sterically demanding coupling of amino acids to carbohydrates during the synthesis of glycopeptides. Glycosyl azide or silyl carbamate (as the latent glycosyl amine component) was reacted with Fmoc/Boc-aspartic acid γ-fluoride resulting in a N-glycosidic linkage between N-acetylglucosamine (GlcNAc) and Asn (Scheme 28).¹⁴⁴ Lindlar's catalyst and Me₃SiOMe (to trap HF) under H₂ was employed in the reaction of the GlcNAc derived azide (137) with appropriately protected fluorides (Fmoc/^tBu, Cbz/Bn, or Boc/^tBu) to obtain the glycosylated amino acids (140) with good yields. In an alternative route, GlcNAc derived azide (137) was converted to glycosyl silyl carbamate (139) and then coupled with the requisite acid fluoride in the presence of a catalytic amount of TBAF. Further elongation of the peptide linkage was carried out by coupling N^α-Fmoc and N^α-Boc protected glycosylated Asn to a tripeptide [NH₂-Thr(Bn)-Thr(Bn)-Asp(OBn)-OAll; All = allyl] via the in situ generated acid fluoride using TFFH to give the glycopeptides in yields of 67% and 91%, respectively. In another application, β-mannoside containing trisaccharide azide was coupled with Fmoc-Asp(F)-O^tBu to give a yield of 70%. ¹H-NMR analysis of the diasteromeric glycopeptides prepared using Fmoc-D-Asp(F)-O^tBu revealed that the products were optically pure.

Scheme 28 Synthesis of glycopeptides.

6.1.6 Loading of sulfonamide safety catch linkers. Fmoc amino acid fluorides were employed efficiently for the acylation of sulfonamide safety catch (SCL) linkers in high loadings and low levels of epimerization. This enables the efficient solid phase synthesis of thioesters by a SCL strategy. The use of PyBOP/DIEA (a reagent generally used for acylation of alcohols) required long reaction times and careful selection of reaction conditions for adequate loadings. However, acylation of SCL-PS and SCL-TentaGel (TG) resins (141) (Champion I and NovaSyn TG, respectively) with various Fmoc-amino acid fluorides in the presence of 2 equiv. of DIEA in CH₂Cl₂, resulted in high loadings in 1 h with minimal epimerization (Scheme 29).¹⁴⁵ With Fmoc-Pro-F, double coupling was employed to obtain a 66–75% yield.

Scheme 29 Acylation of sulfonamide safety catch (SCL) linkers.

6.1.7 Anchoring to resins used in SPPS. Green and Bradley found that low polarity solvents such as CH₂Cl₂, THF with pyridine as acylation catalyst as well as co-solvent provided suitable conditions for the esterification of hydroxyl functionalized resins (143) with Fmoc-amino acids.¹⁴⁶ This was demonstrated via both the acid fluoride (Scheme 30) and DIC/HOBt activation methods. Acid sensitive amino acids such as Trp, Met and amino acids bearing tert-butyl-based side chains could be incorporated efficiently. Hindered amino acid fluorides, Fmoc-Ile-F, Fmoc-Thr(^tBu)-F and Fmoc-Val-F required longer reaction times. However, Fmoc-Arg(Pmc)-F (Pmc = 2,2,5,7,8-pentamethylchroman-6-sulfonyl) underwent cyclisation to the lactam, Fmoc-Cys(Trt)-F and Fmoc-His(Trt)-F suffered premature detritylation.

Scheme 30 Acylation of a p-alkoxybenzyl acohol resin using Fmoc-amino acid fluorides.

In another detailed study, Granitza et al.¹⁴⁷ used Fmoc amino acid fluorides (3 equiv.) for the esterification of the hydroxy resin in nonpolar solvents such as CH₂Cl₂, toluene or THF and in the presence of DMAP as the base. High loading levels were achieved within 30 min with low racemization. Despite the high basicity and perceived tendency towards inducing side reactions, DMAP could be employed for loading hydroxy resins with amino acids. Most of the proteinogenic, side chain protected amino acids and sterically hindered amino acids such as NMeVal and Aib could be loaded with little or no racemization (Fmoc-His(Trt)-F was an exception). This was attributed to short reaction time in nonpolar medium and the high reactivity of acid fluorides. The loading of Ile required double coupling with a duration of 90 min each. The loading of sensitive Fmoc-Cys(^tBu)-OH in toluene using two equiv. of DMAP led to a yield of 5.9% of the D isomer. However, the use of the weaker and hindered base, 2,4,6-collidine effected high loading with little loss of configuration (0.05%). Fmoc cleavage was very slow (<0.2% DBF after 1 h) as monitored by HPLC and no dipeptide formation was observed using ES-MS.

6.1.7.1 Anchoring to TG S RAM and Aib-TG S RAM resins. Acid fluorides were also found to be efficient in loading hindered Fmoc-Aib-OH on to TG S RAM resin (96%) and Aib-TG S RAM resin (55%) using 3 equiv. of acid fluoride in a 0.2 M solution of DMF with 1 equiv. of DIEA.³² Other methods including PyBroP, UNCA and symmetric anhydride methods fared poorly in coupling involving Fmoc-Aib-OH onto Aib-TG S RAM resin.
6.1.7.2 Anchoring to PAC-PEG-PS resin. Bsmoc amino acid fluorides could be loaded on to PAC-PEG-PS resin³⁷ (PAC = 4-hydroxymethylphenoxy acetic acid) in CH₂Cl₂ with 4 equiv. of 2,4,6-collidine or 3,4-lutidine as base in loadings of 0.17 mequiv. g⁻¹ and the percentage of the D-form, as determined by the application of Marfey's reagent,¹⁴⁸ was found to be less than 0.1%.

6.1.8 Conversion to other functionalities.
6.1.8.1 N-Protected amino alcohols. N-Protected amino alcohols (145) were prepared by the in situ reduction of acid fluorides (generated using cyanuric fluoride and pyridine) with sodium borohydride (NaBH₄)/CH₃OH (Scheme 31).¹⁴⁹ The method displays a high level of functional group compatibility including Fmoc/Boc/Cbz protecting groups as well as side chain benzyl ether and benzyl esters of Ser and Glu respectively. The method has the advantage of chemoselectivity when compared to the I₂-NaBH₄ method, which cannot be applied to N-protected peptides and to substrates containing N-acyl-type or ester protecting groups. On the other hand, the borane-tetrahydrofuran method gives N-protected amino alcohols in low chemical yields and optical purity. ¹H-NMR analysis of Mosher esters of (R)- and (RS)-Boc-phenylalaninol [prepared using (R)-(+)-α-methoxy-α-(trifluromethyl)phenylacetic acid] confirmed the absence of the other diastereomeric proton signals because the NH, methoxy, and methylene protons of CH₂OH are all well resolved.

Scheme 31 Synthesis of N-protected amino alcohols.

In other reports, N^α-trityl amino acid fluorides were reduced with NaBH₄ in diglyme at 0 °C. The reduction of N^α-Trt-Glp-F thus yielded a corresponding amino alcohol (146) with a yield of 85% after 10 min (Fig. 12).⁴⁰ The resulting alcohol is a useful intermediate for the synthesis of chiral γ-aminobutyric acid derivatives.

Fig. 12 Structure of N^α-Trt-Glp-CH₂OH.

El-Faham et al. demonstrated the preparation of Boc-/Cbz-/Phe-CH₂OH and Fmoc-Leu/Glu(OBn)-CH₂OH by a NaBH₄ reduction of in situ generated carboxylic acid fluorides generated using TFFH alone or by the addition of PTF in the presence of TEA or DIEA.⁴⁸

6.1.8.2 Wittig olefination of N^α-trityl amino acid fluorides and trifluoromethyl ketones. Wittig olefination of N^α-trityl amino acid fluoride (147) with the stabilized phosphorane Ph₃P=C(Me)CO₂Me in refluxing acetonitrile (CH₃CN) for 10 h led to the formation of an alkene (148) with a yield of 42% (Scheme 32).

Scheme 32 Synthesis of trifluoromethyl ketones and Wittig olefination of N^α-trityl amino acid fluorides.

Trifluoromethyl ketones, which are known to act as serine protease inhibitors were obtained from N^α-trityl amino acid fluorides in two steps using Ruppert's reagent (Me₃SiCF₃). The reaction of acid fluoride with Me₃SiCF₃ in the presence of a catalytic amount of TBAF·H₂O at 0 °C in THF gave silyl ether (149) as the main product (48%). The isolated silyl ether was then treated with equimolar TBAF hydrate at 0 °C for 5 min to obtain the desired ketone (150) in a yield of 87%.⁴⁰ The silyl ether when treated with excess TBAF in situ, led to the formation of a mixture of trifluoromethyl ketone (150) and an alkene (151). However, the ketone could not be obtained directly from acid fluoride even with the use of an equimolar quantity of Me₃SiCF₃.

6.2 Heterocycles

6.2.1 1,4-Benzodiazepines. Ellman and Bunin reported solid phase synthesis of 1,4-benzodiazepine derivatives (155) (Scheme 33), where Fmoc-amino acid fluorides were found to couple extremely well with PS supported 2-aminobenzophenones (152) in the presence of an acid scavenger, 4-methyl-2,6-di-tert-butylpyridine (153).¹⁵⁰ Other protocols including carbodiimides/HOBt or pentafluorophenyl ester gave poor yields. The resulting amides (154) were cyclised (after the removal of acid protection) using 5% acetic acid in DMF, yielding 1,4-benzodiazepine derivatives (155) with various steric and electronic properties. Racemization did not occur during the entire synthesis as shown by the chiral HPLC analysis of benzodiazepines obtained from (R)- and (S)-N^α-Fmoc-Ala-F.

Scheme 33 Synthesis of enantiopure 1,4-benzodiazepine derivatives.

6.2.2 Chiral imidazolines. Chiral 2-imidazolines (158) were prepared using intramolecular aza-Wittig ring closure of N-acylated azido sulfonamides (157) under neutral conditions, with complete regiocontrol.¹⁵¹ N-Acylation of secondary sulfonamides (156) was carried out using preformed Boc-amino acid fluorides in the presence of DIEA in good yields and purity (diastereomeric ratio (dr) > 97 : 3), (Scheme 34) as compared to carbodiimide or PyBroP activation which resulted in low yields and/or extensive racemization. Although the uronium salt of HATU in the presence of caesium carbonate (Cs₂CO₃) gave better results for a range of substrates, for Ala and Phe, the conversion was unsatisfactory. This was attributed to the premature cleavage of the intermediate active ester by the nucleophilic Cs₂CO₃ or the formation of unreactive oxazolone. However, preformed acid fluorides proved superior in these cases because of the high electrophilicity of the acyl donor and the deprotonation of sulfonamide by an optimal base.

Scheme 34 N-Acylation of secondary sulfonamides using Boc-amino acid fluorides and their aza-Wittig reaction to chiral imidazolines.

6.2.3 Substituted hydroxyproline based 2,5-diketopiperazines. Bianco et al. reported solid phase synthesis of highly substituted hydroxyproline-based 2,5-diketopiperazines (Scheme 35).¹⁵² The key step involved the difficult acylation of pyrrolidine nitrogen which was carried out using two approaches. In the first, Fmoc-amino acid fluoride (31) was coupled to the hindered amino function of pyrrolidines (167) by silylation using BSA. Coupling was repeated two to three times and the formation of the 2,5-diketopiperazine was initiated immediately after the cleavage of Fmoc-group. In a second approach, azido acid chloride was employed to acylate the nitrogen of the hydroxyproline. The azide was then reduced using a suspension of tin chloride, TEA, and thiophenol in THF. The resin was finally heated in DMF in the presence of potassium cyanide as catalyst to achieve the cyclisation to 2,5-diketopiperazine (169).

Scheme 35 Solid phase synthesis of highly substituted hydroxyproline-based 2,5-diketopiperazines.

6.2.4 2-Acyl-1-aryl-1,2-dihydroisoquinolines. (S)-α-Cbz-aminoacyl fluorides were employed as chiral auxiliaries in the stereoselective 1,2-additions to isoquinolines. The treatment of (S)-α-Cbz-amino acid fluorides with isoquinoline (162) and anhydrous aluminium chloride (AlCl₃) led to the intermediate N-acyliminium salts (163) (Scheme 36). A subsequent reaction with electron rich arenes or heteroarenes resulted in a Mannich type reaction yielding 2-acyl-1-aryl-1,2-dihydroisoquinolines (164) in modest to high diastereomeric ratio (dr = 6 : 1).¹⁵³ However, in some cases tricyclic imidazo-isoquinolines (165) were formed by an intramolecular attack of the N-atom at the iminium carbon. Grignard reagents/diorgano zinc compounds could be used with electron deficient arenes, albeit with moderate yields and stereoselectivity.

Scheme 36 Synthesis of 2-acyl-1-aryl-1,2-dihydroisoquinolines.

6.2.5 5-N-Acetylardeemin [multi-drug resistant (MDR) reversal agent]. Total synthesis of 5-N-acetylardeemin, a potential MDR reversal agent was demonstrated by Depew et al. The crucial chain elongation step during the synthesis of 5-N-acetylardeemin (169) employed acid fluoride activation for the amide bond formation between the modified amino acid (166) and the amino acid ester (Scheme 37).⁴⁵ Other routes employing BOP-Cl/TEA, DCC/HOBt and DCC/DMAP, all led to a partial racemization. When the acid fluoride was coupled with amino acid esters (H-D-Ala-OMe, H-Gly-OMe, H-D-Phe-OMe) under Schotten–Baumann conditions, protected peptides (168) were obtained with good yields without any racemization.

Scheme 37 Synthesis of MDR reversal agent, 5-N-acetylardeemin.

6.3 Biologically active molecules

6.3.1 Culicinin D. In a recent report, Hung et al. assembled the peptaibol framework of an anticancer agent, culicinin D (180; Fig. 13) which was found to suppress proliferation of the phosphatase and tensin homolog negative MDA468 breast tumor cells.¹⁵⁴ SPPS involved HATU/DIPEA mediated couplings on a 2-chlorotrityl functionalized aminomethyl PS resin, except for the difficult coupling between two Aib residues. Coupling between Fmoc-Aib-OH and a peptidyl resin led to an unknown by-product with a yield of 50%. The problem was overcome by the application of TFFH/DIEA which led to the formation of the desired peptide via the Fmoc-Aib-F generated in situ. Finally, less than 1 mg of culicinin D was obtained, with >95% purity.

Fig. 13 Anticancer agent culicinin D and the site of TFFH mediated acid fluoride activation.

6.3.2 β-Subunit fragment of human chorionic gonadotropin (hCG). Sakamoto et al. developed an elegant protocol for circumventing the diketopiperazine (DKP) formation that occurs during the synthesis of peptides having a C-terminal proline benzyl or allyl ester.¹⁵⁵ The procedure involved the use of a pentafluorophenyl ester of triisopropylsilyloxycarbonyl amino acid (Tsoc) and Fmoc-amino acid fluoride as second and third amino acid, respectively. The combination of silyl carbamate (172) and acid fluoride (171) led to the instantaneous capture of the amine (liberated during Tsoc deprotection effected by a catalytic fluoride ion, 0.1 equiv. of TBAF) by the highly reactive acid fluoride, thereby eliminating the DKP formation. The protocol was employed in the solution phase synthesis of a tripeptide (174) (Fmoc-Ala-Leu-Pro-Oall ester linker) with a yield of 84%. The strategy was later used in the SPPS of 37–53 fragment of the β-subunit of hCG (175) with a yield of 70% (Scheme 38).

Scheme 38 Synthesis of an Fmoc-Ala-Leu-Pro-OAll ester linker and the sequence of the β-subunit 37–53 fragment of hCG.

6.3.3 Phosphotyrosine peptide stat91^695−708 and peptide mimetics containing phosphotyrosine analogues. Fretz reported the synthesis of Fmoc-O,O-(dimethylphospho)-L-tyrosyl fluoride (176) and used it for the efficient coupling of phosphotyrosine to Aib or Ac₆c.⁸⁴ The acid fluoride was subsequently employed in the synthesis of the difficult phosphotyrosine peptide, stat91^695−708 (178). The BOP/HOBt activation proved inefficient for the coupling of N^α-Tyr(PO₃H₂)–OH to Aib and Ac₆c. Three peptides (177) were assembled on the Rink amide MBHA-PS resin (MBHA = 4-methylbenzhydrylamine) using an acid fluoride protocol as well as the 2-(2-pyridon-1-yl)-1,1,3,3-tetramethyluronium tetrafluoroborate (TPTU) method, with quantitative yields and good purity (>95%). However, TPTU mediated incorporation of Fmoc-Tyr(PO₃H₂)-OH resulted in considerable pyrophosphate formation and led to poor yields and low purity of the peptides. A further example of the efficacy of the acid fluoride method is given by the synthesis of stat91^695−708 (Ac-GPKGTGpYIKTELIS-NH₂) (178) on TentaGel S RAM resin (Rapp polymer; Scheme 39). The phosphate methyl groups were deprotected by using trimethylsilyl iodide in CH₃CN after peptide assembly.⁸⁴

Scheme 39 Synthesis of stat91^695−708 using Fmoc-O,O-(dimethylphospho)-L-tyrosyl fluoride.

A series of phosphotyrosines containing oligopeptides with the sequence mAZ-pTyr-Xaa-Asn-NH₂ (mAZ = m-aminobenzyloxycarbonyl) were synthesized as inhibitors of the Grb2 SH2 domain, an important element in the signal transduction pathway which directs cell proliferation and differentiation.⁸⁶ Xaa denotes phosphotyrosine and α-methylphosphotyrosine [(α-Me)pTyr] or its carboxylic mimetics (α-Me)Phe(4-CO₂H) or (α-Me)Phe(4-CH₂CO₂H) as hydrophilic residues. α-Methylphosphotyrosine analogues [Fmoc-L-(α-Me)Tyr(PO₃Bn₂)-OH, Fmoc-L-(α-Me)Phe(CO₂^tBu)-OH and Fmoc-L-(α-Me)Phe(CH₂CO₂^tBu)-OH] and their adjacent phosphotyrosines were coupled using an in situ acid fluoride technique using TFFH, whereas other amino acids were introduced by the standard coupling methods. A peptide with (α-Me)pTyr as Xaa was found to have the highest binding affinity for Grb2 and the best inhibitory activity to displace PSpYVNVQN-Grb2 interaction in an enzyme-linked immunosorbent assay (ELISA) test.

6.4 Peptidomimetics

6.4.1 Trifluoromethyl substituted imidazolines. Peptidyl trifluoromethyl ketones (TFMKs) have found applications as inhibitors of serine, aspartic, metallo, and cysteine proteases. 4-Trifluoromethyl-Δ³-imidazoline (181), as a latent form of TFMK, was synthesized using a 1,3-dipolar cycloaddition reaction between an azomethine ylide (generated from Fmoc-amino acid fluoride and an α-silyl amine) and a dipolarophile, trifluoroacetonitrile (CF₃CN) (Scheme 40). The cycloaddition reaction involved the heating of Fmoc-amino acid fluoride (179), benzylidinetrimethylsilylmethylamine (180), and CF₃CN at 75–80 °C resulting in yield of 70–77% of 4-trifluoromethyl-Δ³-imidazoline (181). The latter was then hydrolyzed to the required TFMK (182), or else it could be converted to a pseudotripeptide under standard Fmoc deprotection and peptide coupling conditions with imidazoline as ketone protector, followed by hydrolysis to yield the tripeptide TFMK (183).¹⁵⁶ The isolation of a single diastereomer indicates that the steromutation does not occur during either the formation of acid fluoride or the 1,3-dipolar cycloaddition.

Scheme 40 Synthesis of 4-trifluoromethyl-Δ³-imidazoline and peptidyl TFMKs.

6.4.2 δ-Conotoxin EVIA peptide analogues: 2,2-dimethylthiazolidine as cis-amide replacement for proline. Chierici et al. employed 2,2-dimethylthiazolidine as locked cis-proline amide bond to study the cis–trans isomerism of the Leu¹²-Pro¹³ bond in δ-conotoxin EVIA (Scheme 41).¹⁵⁷ In the targeted analogue, Pro was replaced by 2,2-dimethylated thiazolidine in the Leu-Pro-Ile sequence. The pseudoproline was coupled to Fmoc-Leu which was activated using DAST, to obtain the desired building block Fmoc-Leu-Cys(Ψ^Me,MePro) (186) with a yield of 75%. The dipeptide was then subjected to segment coupling with the resin bound peptide, followed by standard Fmoc-SPPS to arrive at the target molecule (187).

Scheme 41 Synthesis of Fmoc-Leu-Cys(Ψ^Me,MePro) and its use in SPPS of δ-conotoxin EVIA.

6.4.3 Fmoc-aminoacyl-N-ethyl-S-triphenylmethylcysteine. Hojo et al.¹⁵⁸ employed the Brown and Schafmeister protocol¹³⁰ in the synthesis of Fmoc-aminoacyl-N-ethyl-S-triphenylmethylcysteine (190), an N- to S-migratory device for the preparation of peptide thioesters. Fmoc-Leu-F (184) was coupled to N-ethylcysteine (189) in HFIP under no base conditions at 55 °C for 20 h with a yield of 69% (Scheme 42). The protocol led to a minor amount of the diastereomer. The pentafluorophenyl activated ester in the presence of 3-hydroxy-1,2,3-benzotriazin-4(3H)-one (HOOBt) at 55 °C for 3 h resulted in a better yield of 78%. The addition of HOOBt was found to be essential to increase the yield. However, activation by HATU resulted in a complex mixture of the desired dipeptide and a tripeptide by-product.

Scheme 42 Synthesis of Fmoc-aminoacyl-N-ethyl-S-triphenylmethylcysteine using the Brown and Schafmeister protocol.

6.4.4 α-Ureido peptides and urea tethered glycosylated amino acids. Hemantha et al. reported a one pot synthesis of ureido peptides and urea tethered glycosylated amino acids using in situ generated acid fluorides.¹⁵⁹ Fmoc/Boc/Cbz-amino acids were converted to their respective acid fluorides using Deoxo-Fluor/TEA, which were then treated with trimethylsilyl azide (TMSN₃) and an amino acid ester in one pot under ultrasonication to obtain the desired α-ureido peptides (191) (Scheme 43). The reaction sequence involved the formation of an acid fluoride, conversion into an azide, then rearrangement into an isocyanate, and finally coupling with the amine component to yield α-ureido peptide. The replacement of the amino acid ester by 2,3,4,6-tetra-O-acetylglycosyl-1-amine (192) resulted in urea linked glycosylated amino acids (193) with good yields. The conditions employed were found to be racemization free as shown by ¹H-NMR analysis.

Scheme 43 Synthesis of ureido peptides and urea tethered glycosylated amino acids.

6.4.5 1,2,4-Oxadiazole, 1,3,4-thiadiazole and 1,3,4-oxadiazole tethered dipeptide mimetics. 1,2,4-Oxadiazole linked orthogonally N-protected (Fmoc/Boc/Cbz) dipeptide mimetics (195) were reported by Sureshbabu et al., by a reaction between N-protected amino acyl fluoride and an amino acid derived amidoxime (194) through an O-acyl amidoxime intermediate.¹⁶⁰ N-Protected amino acid fluoride was reacted with amino acid derived amidoxime in the presence of NMM in ethanol for 15 min, followed by the addition of an equimolar quantity of sodium acetate and then refluxing for 3 h (Scheme 44). The protocol could be extended to tetramer synthesis as well by using peptide acid fluorides.

Scheme 44 Synthesis of 1,2,4-oxadiazole, 1,3,4-thiadiazole and 1,3,4-oxadiazole tethered dipeptide mimetics.

Fmoc/Boc/Cbz-amino acid fluorides were employed with the Boc/Cbz-protected acyl hydrazides (196), to yield diacylhydrazine intermediates, which were then used in the synthesis of orthogonally protected 1,3,4-thiadiazole or 1,3,4-oxadiazole tethered dipeptide mimetics (197).¹⁶¹ The diacylhydrazines were treated with Lawesson's reagent (LR) followed by dehydrosulfurization under reflux to yield 1,3,4-thiadiazoles. They could also be cyclized using EDC to give the corresponding 1,3,4-oxadiazoles.

6.4.6 N-Carboxylalkyl and N-amino alkyl functionalized dipeptides. Reissmann's group synthesized N-carboxylalkyl and N-amino alkyl dipeptide building blocks (201, 202) in order to avoid incomplete couplings of the bulky amino acid to an N-alkylated amino acid in solid phase synthesis, and to improve the assembly of cyclic peptides, particularly bradykinin and somatostatin analogues.^162,163 N-Carboxylalkyl and N-amino alkyl amino acid tert-butyl esters were refluxed with Fmoc-amino acid fluorides in the presence of collidine to give the corresponding dipeptide esters (Scheme 45). The dipeptide tert-butyl esters were then deprotected with TFA. The desired dipeptide acids were obtained with good yields (32–44%) and purity, and they could be used directly in standard Fmoc solid phase synthesis. A similar strategy was employed for dipeptide units with N-terminal arginine, namely, Cbz-L-Arg(Cbz)₂Ψ{CO–N[(CH₂)₂CO₂All]}Phe-OH and Cbz-L-Arg(Cbz)₂Ψ{CO–N[(CH₂)₃NHAlloc]}Phe-OH except that the acid fluoride was generated in situ using TFFH.¹⁶³

Scheme 45 Synthesis of N-carboxylalkyl and N-amino alkyl functionalized dipeptide units.

7. A summary of the advantages of using acid fluorides and a few exceptions to using them

(1) Acid fluorides are considerably less reactive towards water and methanol and thus, are more resistant to hydrolysis and solvolysis. In fact, the reactivity of acid fluorides resembles more that of active esters than acid chlorides or acid bromides.¹⁶⁴

(2) The activation of the carbonyl group is high because of the small size and high electronegativity of the fluorine atom.

(3) Acid fluorides of amino acids bearing acid sensitive tert-butyl based protecting groups are shelf stable.

(4) Acid fluorides have a lesser tendency to form oxazolones when exposed to tertiary amines [DIEA, NMM, pyridine, 1,8-bis(dimethylamino)naphthalene (PS), and so on], and thus, yield optically pure products. No base coupling is also feasible.

(5) Acid fluorides are highly stable in solvents such as DMF making them suitable to be used in couplings in the solid phase as well as in reactions that require longer durations. Thus, they are ideal intermediates for both solution as well as solid phase synthesis.

(6) They are efficient agents for the coupling of hindered amino acids, unlike most of the common coupling protocols/reagents.

(7) In situ generation of amino acid fluorides is possible with the availability of reagents such as TFFH.

Although amino acid fluorides have proved to be comparatively valuable for the syntheses of a wide range of molecules, the other coupling agents/methods being better than acid fluorides cannot be ruled out, at least in some special and isolated cases. However, enough work has not been done on a comparative study to determine this. The study of aminolysis of benzoyl fluoride has revealed that the reactivity of acid fluorides is similar to that of active esters unlike with other benzoyl halides (chlorine, bromine and iodine).¹⁶⁴ Anhydrides have been found to be more reactive than acid fluorides in some cases. NCAs⁸⁹ and acid chlorides¹⁶⁵ of less hindered amino acids were found be more reactive than acid fluorides, when coupled to a hindered Aib residue. Also it has been reported that N-phthalimido-γ-benzyl-L-aspartic acid fluoride could not be coupled to methyl (S)-1,2,3,4-tetrahydroquinoline-2-carboxylate (203).¹⁶⁶ When the same coupling was carried out using acid chloride activation (Scheme 46) with either aqueous NaHCO₃ or 2,6-di-tert-butylpyridine, the target product was obtained with a high yield (up to 86%) and diastereomeric purity (up to a dr of 99.4%), whereas in the presence of NMM, a 1 : 1 mixture of diastereomers resulted.

Scheme 46 Reaction of Pht-Asp(Bn)-Cl with methyl (S)-1,2,3,4-tetrahydroquinoline-2-carboxylate.

8. Conclusions

The acid halides, which were introduced by Emil Fischer at the very beginning of peptide synthesis, were reborn as peptide coupling agents with the discovery of Fmoc-amino acid chlorides by Louis Carpino. This renaissance was completed by the introduction of Boc-/Cbz-amino acid fluorides. With the advent of acid fluorides, several of the intriguing concerns about difficult peptide synthesis have been addressed satisfactorily thereby asserting them as the reagents of choice especially for the coupling of sterically hindered amino acids. Also, the problem of chemical and optical stability often associated with acid chlorides has been alleviated with use of acid fluorides. The possibility of in situ generation of acid fluorides, similar to mixed anhydrides, has made them convenient to use and increased the scope of their applicability. This is of particular significance in the activation of peptide acids. They are also efficiently employed in automated SPPS. In addition, their compatibility with acid labile protecting groups and the convenience of no base coupling has increased their synthetic use from mere peptide coupling agents to reagents for the synthesis of several heterocycles, peptidomimetics and biologically important molecules. Thus, although a large number of activating agents have been found previously, acid fluorides have undoubtedly survived the competition by virtue of their critical advantages. They continue to be one of the first choices as carboxyl activating agents, thus adding an appendix to Fischer's legacy.

Acknowledgements

V. V. Sureshbabu is grateful to the late Professor K. M. Sivanandaiah, Bangalore University, Bangalore for his guidance in the initial stages of the establishment of the peptide research group at Central College, Bangalore (CCB) and to the late Professor B. S. Sheshadri for his untiring support in providing laboratory space at CCB in 1998. VVS is ever thankful to Dr H. N. Gopi and Dr K. Ananda for their work in establishing the group during its infancy. VVS would also like to offer sincere thanks to all former students, in particular, Dr R. Venkataramanarao, Dr Chennakrishnareddy, Dr S. A. Shankar and Dr H. P. Hemantha for their sincere involvement and hard work during their time in the group. VVS is grateful to Professor M. S. Thimmappa and Dr N. Prabhudev, formerly Vice-Chancellors of Bangalore University, Bangalore and in particular, Professor B. Thimme Gowda (present Vice-Chancellor) for extending academic support for the maintenance of the research group at CCB. The Department of Science and Technology, Council of Scientific and Industrial Research, Department of Biotechnology, University Grants Commission, Government of India, New Delhi and the Board of Research in Nuclear Sciences, Department of Atomic Energy, Mumbai are thanked for their generous financial support.

Notes and references

1. E. Fischer and E. Otto, Ber. Dtsch. Chem. Ges., 1903, 36, 2106–2116.
2. E. Fischer, Ber. Dtsch. Chem. Ges., 1907, 40, 1754–1767.
3. E. Fisher, Ber. Dtsch. Chem. Ges., 1903, 36, 2094–2106.
4. E. Fischer, Ber. Dtsch. Chem. Ges., 1905, 38, 605–619.
5. E. Fisher and P. Bergell, Ber. Dtsch. Chem. Ges., 1903, 36, 2592–2608.
6. M. Bodanszky, Principles of Peptide Synthesis, Springer Verlag, Berlin, 1984, p. 10.
7. M. Bodanszky, Y. S. Klausner and M. A. Ondetti, Peptide Synthesis, Wiley, New York, 1976, vol. 13–17, pp. 23–24, 102.
8. M. Bodanszky and A. Bodanszky, The practice of peptide synthesis, Springer Verlag, Berlin, 1984, p. 89.
9. J. Jones, The Chemical Synthesis of Peptides, Clarendon Press, Oxford, 1991, pp. 17–22.
10. B. Gutte, Peptides- Synthesis, structure and applications, Academic Press, San Diego, 1995, p. 511.
11. P. D. Bailey, An introduction to peptide chemistry, John Wiley & Sons, Chichester, 1990, p. 125.
12. E. Schroder and K. Lubke, The peptides, Methods of peptide synthesis, Academic Press, New York, 1965, vol. 1, p. 77.
13. H. Leuchs, Ber. Dtsch. Chem. Ges., 1906, 39, 857–861.
14. (a) H. Leuchs and W. Manasse, Ber. Dtsch. Chem. Ges., 1907, 40, 3235–3249; (b) H. Leuchs and W. Geiger, Ber. Dtsch. Chem. Ges., 1908, 41, 1721–1726.
15. K. Poduska and H. Gross, Ber, 1961, 94, 527–537.
16. L. A. Carpino, M. Beyermann, H. Wenschuh and M. Bienert, Acc. Chem. Res., 1996, 29, 268–274.
17. M. Beyermann, M. Bienert and L. A. Carpino, Acid halides, Houben Weyl: Synthesis of peptides and peptidomimetics, George Thieme Verlag, Stuttgart, New York, 2003, vol. E22a, pp. 475–494.
18. E. Gross and J. Meienhofer, The Peptides, Academic, New York, 1979, vol. 1, pp. 1–284.
19. G. Losse and K.-D. Wehrstedt, Z. Chem., 1981, 21, 148–153.
20. (a) U. Schmidt, M. Kroner and U. Beutler, Synthesis, 1988, 475–477; (b) For an earlier report on the preparation of dipeptides of Z-Cys(Bn), Cbz-Pro and Cbz-Gly using Vilsmeier-Hack reagent see: M. Zaoral and Z. Arnold, Tetrahedron Lett., 1960, 14, 9–12; (c) A. C. Farthing, J. Chem. Soc., 1950, 3213–3217; (d) S. Goldschmidt, W. Lautenschlager, B. Kolb and G. Zumach, Chem. Ber., 1964, 94, 2434–2438.
21. F. Munyemana, A.-M. Frisque-Hesbain, A. Devos and L. Ghosez, Tetrahedron Lett., 1989, 30, 3077–3080.
22. (a) L. A. Carpino, B. J. Cohen, K. E. Stephens Jr, Y. Sadat-Aalaee, J. H. Tien and D. C. Langridge, J. Org. Chem., 1986, 51, 3732–3734; (b) M. Beyerman, M. Bienert, H. Niedrich, L. A. Carpino and D. Sadat-Aalaee, J. Org. Chem., 1990, 55, 721–728; (c) L. A. Carpino, H. G. Chao, M. Beyermann and M. Bienert, J. Org. Chem., 1991, 56, 2635–2642; (d) For the synthesis of peptides via Fmoc-Xaa-Cl/tris(2-aminoethyl)amine approach see: L. A. Carpino, D. Sadat-Aalaee and M. Beyermann, J. Org. Chem., 1990, 55, 1673–1675.
23. J. H. Jones, E. Gross and J. Meienhofer, Peptides, Academic Press, New York, 1979, vol. 1, p. 65.
24. For a detailed mechanistic treatment and in depth analysis on stereomutation during peptide synthesis see: N. L. Benoiton, Biopolymers, 1996, 40, 245–254.
25. (a) For the coupling of acid chlorides under no base conditions (i) mediated by KOBt see: K. M. Sivanandaiah, V. V. Sureshbabu and S. C. Shankaramma, Int. J. Pept. Protein Res., 1994, 44, 24–30 (ii) For the use of KOAt see: H. N. Gopi and V. V. Sureshbabu, J. Indian Chem. Soc., 1998, 75, 511–513 (iii) For the use of TBDMS-OBt see: S. J. Tantry and V. V. Sureshbabu, Lett. Pept. Sci., 2002, 9, 35–41; (b) For the utility of Zn dust mediated coupling circumventing the premature deblocking of Fmoc group and thereby extending the coupling duration see: H. N. Gopi and V. V. Sureshbabu, Tetrahedron Lett., 1998, 39, 9769–9772.
26. L. A. Carpino, Acc. Chem. Res., 1987, 20, 401–407.
27. L. A. Carpino, D. Sadat-Aalaee, H. G. Chao and R. H. DeSelms, J. Am. Chem. Soc., 1990, 112, 9651–9652.
28. J.-N. Bertho, A. Loffet, C. Pinel, F. Reuther and G. Sennyey, Tetrahedron Lett., 1991, 32, 1303–1306.
29. L. A. Carpino, E.-S. M. E. Mansour and D. Sadat-Aalaee, J. Org. Chem., 1991, 56, 2611–2614.
30. L. A. Carpino and E.-S. M. E. Mansour, J. Org. Chem., 1992, 57, 6371–6373.
31. H. Wenschuh, M. Beyermann, E. Krause, L. A. Carpino and M. Bienert, Tetrahedron Lett., 1993, 34, 3733–3736.
32. H. Wenschuh, M. Beyermann, E. Krause, M. Brudel, R. Winter, M. Schumann, L. A. Carpino and M. Bienert, J. Org. Chem., 1994, 59, 3275–3280.
33. H. Wenschuh, M. Beyermann, H. Haber, J. K. Seydel, E. Krause and M. Bienert, J. Org. Chem., 1995, 60, 405–410.
34. V. V. Sureshbabu and K. Ananda, Lett. Pept. Sci., 2000, 7, 41–46.
35. H. Wenschuh, M. Beyermann, R. Winter, M. Bienert, D. Ionescu and L. A. Carpino, Tetrahedron Lett., 1996, 37, 5483–5486.
36. L. A. Carpino, M. Philbin, M. Ismail, G. A. Truran, E.-S. M. E. Mansour, S. Iguchi, D. Ionescu, A. El-Faham, C. Riemer, R. Warrass and M. S. Weiss, J. Am. Chem. Soc., 1997, 119, 9915–9916.
37. L. A. Carpino, M. Ismail, G. A. Truran, E.-S. M. E. Mansour, S. Iguchi, D. Ionescu, A. El-Faham, C. Riemer and R. Warrass, J. Org. Chem., 1999, 64, 4324–4338.
38. L. A. Carpino and D. Sadat-AaLaee, US Pat., US5360928, 1994.
39. L. A. Carpino and E.-S. M. E. Mansour, J. Org. Chem., 1999, 64, 8399–8401.
40. G. Karygiannis, C. Athanassopoulos, P. Mamos, N. Karamanos, D. Papaioannou and G. W. Francis, Acta Chem. Scand., 1998, 52, 1144–1150.
41. For a short report on amide bond formation employing amino acid fluorides as coupling agents see: J. W. Lippert III, ARKIVOC, 2005, xiv, 87–95.
42. For an account describing the use of TFFH in peptides as well as organic synthesis see: A. El-Faham and S. N. Khattab, Synlett, 2009, 886–904.
43. (a) F. Albericio, Curr. Opin. Chem. Biol., 2004, 8, 211–221; (b) S. Y. Han and Y. A. Kim, Tetrahedron, 2004, 60, 2447–2467; (c) E. Valeur and M. Bradley, Chem. Soc. Rev., 2009, 38, 606–631; (d) C. A. G. N. Montalbetti and V. Falque, Tetrahedron, 2005, 61, 10827–10852; (e) A. El-Faham and F. Albericio, Chem. Rev., 2011, 111, 6557–6602; (f) V. R. Pattabiraman and J. W. Bode, Nature, 2011, 480, 471–479.
44. For a review on the ribosomal and post translationally modified peptides and their biological role see: P. G. Arnison, M. J. Bibb, G. Bierbaum, A. A. Bowers, T. S. Bugni, G. Bulaj, J. A. Camarero, D. J. Campopiano, G. L. Challis, J. Clardy, P. D. Cotter, D. J. Craik, M. Dawson, E. Dittmann, S. Donadio, P. C. Dorrestein, K. D. Entian, M. A. Fischbach, J. S. Garavelli, U. Goransson, C. W. Gruber, D. H. Ha, T. K. Hemscheidt, C. Hertweck, C. Hill, A. R. Horswill, M. Jaspars, W. L. Kelly, J. P. Klinman, O. P. Kuipers, A. J. Link, W. Liu, M. A. Marahiel, D. A. Mitchell, G. N. Moll, B. S. Moore, R. Muller, S. K. Nair, I. F. Nes, G. E. Norris, B. M. Olivera, H. Onaka, M. L. Patchett, J. Piel, M. J. Reaney, S. Rebuffat, R. P. Ross, H. G. Sahl, E. W. Schmidt, M. E. Selsted, K. Severinov, B. Shen, K. Sivonen, L. Smith, T. Stein, R. D. Sussmuth, J. R. Tagg, G. L. Tang, A. W. Truman, J. C. Vederas, C. T. Walsh, J. D. Walton, S. C. Wenzel, J. M. Willey and W. A. van der Donk, Nat. Prod. Rep., 2013, 30, 108–160.
45. (a) For the wider utility of cyanuric fluoride see: G. H. Olah, M. Nojima and I. Kerekes, Synthesis, 1973, 487–488; (b) K. M. Depew, S. P. Marsden, D. Zatorska, A. Zatorski, W. G. Bornmann and S. J. Danishefsky, J. Am. Chem. Soc., 1999, 121, 11953–11963.
46. J. M. Schkeryantz, J. C. G. Woo, P. Siliphaivanh, K. M. Depew and S. J. Danishefsky, J. Am. Chem. Soc., 1999, 121, 11964–11975.
47. L. A. Carpino and A. El-Faham, J. Am. Chem. Soc., 1995, 117, 5401–5402.
48. A. El-Faham, S. N. Khattab and M. A. Ghani, ARKIVOC, 2006, xiii, 57–63.
49. (a) W. J. Middleton, J. Org. Chem., 1975, 40, 574–578; (b) L. N. Markovsku, V. E. Pashinnik and A. V. Kirsanov, Synthesis, 1973, 787–789.
50. (a) M. Hudlický, Org. React., 1988, 35, 513–637; (b) R. P. Singh, D. T. Meshri and J. M. Shreeve, in Advances in Organic Synthesis Vol. 2, Modern Organofluorine Chemistry-Synthetic Aspects, ed. Atta-Ur-Rahman and K. K. Laali, Bentham Science Publishers Ltd., Hilversum, Netherlands, 2006, pp. 291−326.
51. L. N. Markovskiji, V. E. Pashinnik and A. V. Kirsanov, Synthesis, 1973, 787–789.
52. A. C. O'Sullivan, F. Struber and S. V. Ley, J. Org. Chem., 1999, 64, 6252–6256.
53. C. Kaduk, H. Wenschuh, M. Beyermann, K. Forner, L. A. Carpino and M. Bienert, Lett. Pept. Sci., 1995, 2, 285–288.
54. V. V. Sureshbabu, H. N. Gopi and K. Ananda, Indian J. Chem., Sect. B: Org. Chem. Incl. Med. Chem., 2000, 39, 384–385.
55. G. S. Lal, G. P. Pez, R. J. Pesaresi, F. M. Prozonic and H. Cheng, J. Org. Chem., 1999, 64, 7048–7054.
56. G. S. Lal, G. P. Pez, R. J. Pesaresi and F. M. Prozonic, Chem. Commun., 1999, 215–216.
57. A. R. Tunoori, J. M. White and G. I. Georg, Org. Lett., 2000, 2, 4091–4093.
58. J. M. White, A. R. Tunoori, B. J. Turunen and G. I. Georg, J. Org. Chem., 2004, 69, 2573–2576.
59. For a “dry” protocol regarding the preparation of TFFH see: T. Vojkovsky and B. drake, Org. Prep. Proced. Int., 1997, 29, 497–498.
60. For an alternative “dry” procedure see: U. Boas, B. Pederson and B. C. Christensen, Synth. Commun., 1998, 28, 1223–1231.
61. For the wider utility of TFFH see: (a) K. C. Nicolaou, H. J. Mitchell, H. Suzuki, R. M. Rodríguez, O. Baudoin and K. C. Fylaktakidou, Angew. Chem., Int. Ed., 1999, 38, 3334–3339; (b) U. Boas, B. Pedersen and J. B. Christensen, Synth. Commun., 1998, 28, 1223–1231; (c) U. Boas, H. Gertz, J. B. Christensen and P. M. H. Heegaard, Tetrahedron Lett., 2004, 45, 269–272; (d) A. El-Faham, M. Chebbo, M. Abdul-Ghani and G. Younes, J. Heterocycl. Chem., 2006, 43, 599–606.
62. K. C. Nicolaou, R. M. Rodríguez, K. C. Fylaktakidou, H. Suzuki and H. J. Mitchell, Angew. Chem., Int. Ed., 1999, 38, 3340–3345.
63. A. El-Faham, Org. Prep. Proced. Int., 1998, 30, 477–481.
64. A. El-Faham, Chem. Lett., 1998, 671–672.
65. A. El-Faham, S. N. Khattab, M. R. Abdul-Ghani and F. Albericio, Eur. J. Org. Chem., 2006, 1563–1573.
66. A. El-Faham and F. Albericio, J. Org. Chem., 2008, 73, 2731–2737.
67. T. Mukaiyama, M. Usui, E. Shimada and K. Saigo, Chem. Lett., 1975, 1045–1048.
68. E. Bald, K. Saigo and T. Mukaiyama, Chem. Lett., 1975, 1163–1166.
69. T. Mukaiyama, S. Ikeda and S. Kobayashi, Chem. Lett., 1975, 1159–1162.
70. T. Mukaiyama, H. Toda and S. Kobayashi, Chem. Lett., 1976, 13–14.
71. T. Mukaiyama, M. Usui and K. Saigo, Chem. Lett., 1976, 49–50.
72. T. Mukaiyama, T. Aikawa and S. Kobayashi, Chem. Lett., 1976, 57–60.
73. T. Mukaiyama and K. Hojo, Chem. Lett., 1976, 893–896.
74. For the reaction of carboxylic acids with 2-fluoropyridinium salts in presence of TEA see: T. Mukaiyama and T. Tanaka, Chem. Lett., 1976, 303–306.
75. P. Li and J.-C. Xu, Tetrahedron, 2000, 56, 8119–8131.
76. L. A. Carpino, D. Ionescu, A. El-Faham, M. Beyermann, P. Henklein, C. Hanay, H. Wenschuh and M. Bienert, Org. Lett., 2003, 5, 975–977.
77. L. A. Carpino, H. Imazumi, A. El-Faham, F. J. Ferrer, C. Zhang, Y. Lee, B. M. Foxman, P. Henklein, C. Hanay, C. Mugge, H. Wenschuh, J. Klose, M. Beyermann and M. Bienert, Angew. Chem., Int. Ed., 2002, 41, 441–445.
78. For the first report on fluorination of carboxylic acids to acyl fluorides via a reaction with hydrogen fluoride-pyridine and 1,3-dicyclohexylcarbodiimide in CH₂Cl₂ see: C. Chen, C.-T. Chien and C.-H. Su, J. Fluorine Chem., 2002, 115, 75–77.
79. D. W. Kim, H.-J. Jeong, S. T. Lim and M.-H. Sohn, Angew. Chem., Int. Ed., 2008, 47, 8404–8406.
80. J.-G. Kim and D. O. Jang, Synlett, 2010, 3049–3052.
81. T. Umemoto, R. P. Singh, Y. Xu and N. Saito, J. Am. Chem. Soc., 2010, 132, 18199–18205.
82. R. P. Singh and T. Umemoto, J. Org. Chem., 2011, 76, 3113–3121.
83. (a) E. Schnabel, H. Herzog, P. Hoffman, E. Klauke and I. Ugi, Justus Liebigs Ann. Chem., 1968, 716, 175–185; (b) L. A. Carpino, K. N. Parameswaran, R. K. Kirkley, J. W. Spiewak and E. Schmitz, J. Org. Chem., 1970, 35, 3291–3295.
84. H. Fretz, Lett. Pept. Sci., 1997, 4, 171–176.
85. For the studies related to the lactam formation upon activation of Fmoc-Arg derivatives see: L. Juliano, M. A. Juliano, A. De Miranda, S. Tsuboi and Y. Okada, Chem. Pharm. Bull., 1987, 35, 2550–2553.
86. W.-Q. Liu, M. Vidal, N. Gresh, B. P. Roques and C. Garbay, J. Med. Chem., 1999, 42, 3737–3741.
87. C. Picard, L. Cazaux and P. Tisner, Tetrahedron Lett., 1984, 25, 3809–3812.
88. J. Savrda and M. Wakselman, J. Chem. Soc., Chem. Commun., 1992, 812–813.
89. M. Wakselman, J.-P. Mazaleyrat and J. Savrda, Amino Acids, 1994, 7, 67–77.
90. J. Savrda, L. Chertanova and M. Wakselman, Tetrahedron, 1994, 50, 5309–5322.
91. L. A. Carpino, E.-S. M. E. Mansour and A. El-Faham, J. Org. Chem., 1993, 58, 4162–4164.
92. M. Meldal, M. A. Juliano and A. M. Jansson, Tetrahedron Lett., 1997, 38, 2531–2534.
93. A. Polese, F. Formaggio, M. Crisma, G. Valle, C. Toniolo, G. M. Bonora, Q. B. Broxterman and J. Kamphuis, Chem.–Eur. J., 1996, 2, 1104–1111.
94. R. Fiammengo, G. Licini, A. Nicotra, G. Modena, L. Pasquato, P. Scrimin, Q. B. Broxterman and B. Kaptein, J. Org. Chem., 2001, 66, 5905–5910.
95. R. Gratias, R. Konat, H. Kessler, M. Crisma, G. Valle, A. Polese, F. Formaggio, C. Toniolo, Q. B. Broxterman and J. Kamphuis, J. Am. Chem. Soc., 1998, 120, 4763–4770.
96. L. A. Carpino, D. Ionescu, A. El-Faham, P. HenKlein, H. Wenschuh, M. Bienert and M. Beyermann, Tetrahedron Lett., 1998, 39, 241–244.
97. For a detailed study on the premature deblocking of Fmoc group employing Fmoc-amino acid fluorides as coupling agents during coupling, in particular as and when duration of coupling is to be extended to ensure completion of acylation see: H. Wenschuh, M. Beyermann, M. Bienert, A. El-Faham, S. Ghassemi and L. A. Carpino, Peptides, Chemistry, Structure and Biology, ed. P. T. P. Kaumaya and R. S. Hodges, Mayflower Scientific Ltd., Kingswinford, England, U. K, 1996; pp. 33–35.
98. For a comparative account of reactivity of acid chlorides versus acid fluorides see:^92–94 C. A. Bunton and J. H. Fendler, J. Org. Chem., 1966, 31, 2307–2312 .
99. M. L. Bender and J. M. John, J. Org. Chem., 1962, 27, 3771–3774.
100. G. A. Olah and S. J. Kuhn, J. Org. Chem., 1961, 26, 225–227.
101. C. G. Swann and C. B. Scott, J. Am. Chem. Soc., 1953, 75, 246–248.
102. H. Wenschuh, M. Beyermann, A. El-Faham, S. Ghassemi, L. A. Carpino and M. Bienert, J. Chem. Soc., Chem. Commun., 1995, 669–670.
103. In the case of Fmoc-NMe-Aib-F, dibenzofulvene (DBF) was formed to the extent of 40.7% within 4h of coupling reaction. For details see: H. Wenschuh, M. Beyermann, M. Bienert, A. El-Faham, S. Ghassemi and L. A. Carpino, Peptides: Chemistry, Structure and Biology, Proceedings of the 14th American Peptide Society, ed. P. T. P. Kaumaya and R. S. Hodges, Mayflower Scientific Ltd., Kingswinford, England, 1995, p. L-041.
104. S. Rajeswari, R. J. Jones and M. P. Cava, Tetrahedron Lett., 1987, 28, 5099–5102.
105. For the preparation of amides and peptides using N-silylamines circumventing the use of an equimolar quantity of a base during coupling see:^99,100 (a) H. H. Anderson, J. Am. Chem. Soc., 1952, 74, 1421–1423; (b) J. Pump and U. Wannagat, Monatsh. Chem., 1962, 93, 352–359.
106. (a) J. R. Bowser, P. J. Williams and K. Kurz, J. Org. Chem., 1983, 48, 4111–4113; (b) S. Raucher and D. S. Jones, Synth. Commun., 1985, 15, 1025–1031; (c) J. F. Klebe, H. Finkbeiner and D. M. White, J. Am. Chem. Soc., 1966, 88, 3390.
107. For the N-silylation of amines as well as amino acid esters using TMS-Cl/Zn dust under non Schotten–Baumann conditions (as an alternative to the use of BSA) and subsequent utility in peptide synthesis see: V. V. Sureshbabu, G. R. Vasanthakumar and S. J. Tantry, Tetrahedron Lett., 2005, 46, 4099–4102.
108. L. A. Carpino, J. Org. Chem., 1988, 53, 875–877.
109. S. P. Marsden, K. M. Depew and S. J. Danishefsky, J. Am. Chem. Soc., 1994, 116, 11143–11144.
110. L. A. Carpino, S. Ghassemi, D. Ionescu, M. Ismail, D. Sadat-Aalaee, G. A. Truran, E.-S. M. E. Mansour, G. A. Siwruk, J. S. Eynon and B. Morgan, Org. Process Res. Dev., 2003, 7, 28–37.
111. H.-O. Kim, B. Gardner and M. Kahn, Tetrahedron Lett., 1995, 36, 6013–6016.
112. N. Thieriet, P. Gomez-Martinez and F. Guibe, Tetrahedron Lett., 1999, 40, 2505–2508.
113. (a) K. M. Sivanandaiah, S. Gurusiddappa and V. V. Sureshbabu, Indian J. Chem., Sect. B: Org. Chem. Incl. Med. Chem., 1991, 30, 302–305; (b) W. S. Hancock, D. J. Prescott, P. R. Vagelos and G. Marshall, J. Org. Chem., 1973, 38, 774–781; (c) E. Atherton, D. L. Clive and R. C. Sheppard, J. Am. Chem. Soc., 1975, 97, 6584–6585; (d) R. Arshady, E. Atherton, D. L. J. Clive and R. C. Sheppard, J. Chem. Soc., Perkin Trans. 1, 1981, 529–537; (e) E. Atherton and R. C. Sheppard, J. Chem. Soc., Chem. Commun., 1985, 165–166.
114. D. Hudson, J. Org. Chem., 1988, 53, 617–624.
115. C. Barris, A. Brass, B. Robson and G. Tomalin, Innovation and perspectives in solid phase synthesis, ed. R. Epton, SPCC (UK) Ltd., Birmingham, 1990, p. 441.
116. P. Henklein, M. Beyennann, M. Bienert and R. Knorr, Peptides, 1990, ed. E. Giralt and D. Andreu, ESCOM Leiden, 1991, p. 67.
117. For a comparison of purity and ES-MS data see: M. Beyermann, H. Wenschuh, P. Henklein and M. Bienert, Innovation and Perspectiues in Solid Phase Synthesis, ed. R. Epton, Wolverhampton, 1992, p. 349.
118. I. Weyand, M. Godde, S. Frings, J. Weiner, F. Muller, W. Altenhofen, H. Hatt and U. B. Kaupp, Nature, 1994, 368, 859–863.
119. H. Wenschuh, M. Beyermann, S. Rothemund, L. A. Carpino and M. Bienert, Tetrahedron Lett., 1995, 36, 1247–1250.
120. C. Gundala, S. J. Tantry, S. A. Naik and V. V. Sureshbabu, Protein Pept. Lett., 2009, 16, 105–111.
121. R. Franz, J. Fluorine Chem., 1980, 15, 423–434.
122. (a) L. A. Carpino, A. El-Faham and F. Albericio, Lett. Pept. Sci., 1994, 1, 283–290; (b) B. F. Gisin, D. G. Davis, Z. K. Borowska, J. E. Hall and S. Kobayashi, J. Am. Chem. Soc., 1981, 103, 6373–6377; (c) C. Peggion, I. Coin and C. Toniolo, Pept. Sci., 2004, 76, 485–493; (d) C. Baldini, M. Bellanda, C. Peggion, A. Djontu, C. Atagua, S. Mammi and C. Toniolo, Chem. Biodiversity, 2007, 41, 1129–1143.
123. For previous reports on the synthesis of alamethicin see: (a) R. Nagaraj and P. Balaram, Tetrahedron, 1981, 37, 1263–1270; (b) T. M. Balasubramanian, N. C. E. Kendrick, M. Taylor, G. R. Marshall, J. E. Hall, I. Vodyanon and F. Reusser, J. Am. Chem. Soc., 1981, 103, 6127–6132; (c) H. Schmitt and G. Jung, Liebigs Ann. Chem., 1985, 321–344; (d) B. F. Gisin, D. G. Davis, Z. K. Bolowska, J. E. Hall and S. Kobayashi, J. Am. Chem. Soc., 1981, 103, 6373–6377; (e) H. Heimgartner, Angew. Chem., Int. Ed. Engl., 1991, 30, 238–264.
124. S. Kusomoto, M. Matsukura and T. Shiba, Biopolymers, 1981, 20, 1869–1875.
125. Y. Nagaoka, A. Iida, T. Kambara, K. Asami and T. Fujita, Chem. Commun., 1996, 1079–1080.
126. C. V. Hjorringgaard, J. M. Pederson, T. Voseggaard, N. C. Nielsen and T. Skrydstrup, J. Org. Chem., 2009, 74, 1329–1332.
127. I. L. Karle and P. Balaram, Biochemistry, 1990, 29, 6747–6756.
128. C. Toniolo, M. Crisma, F. Formaggio, G. Valle, G. Cavicchioni, G. Precigoux, A. Aubry and J. Kamphuis, Biopolymers, 1993, 33, 1061–1072.
129. M. D. Fletcher and M. M. Campbell, Chem–Eur. J., 1996, 2, 1104–1111.
130. Z. Z. Brown and C. E. Schafmeister, J. Am. Chem. Soc., 2008, 130, 14382–14383.
131. J. R. Spencer, N. Antonenko, G. J. Delaet and M. Goodman, Int. J. Pept. Protein Res., 1992, 40, 282–293.
132. The proposed transient amino-anhydride intermediate which rearranges through a five membered transition state is similar to O,N-acyl transfer. For similar mechanistic studies see: (a) I. Ugi, Angew. Chem., Int. Ed., 1962, 1, 8–21; (b) A. Domling and I. Ugi, Angew. Chem., Int. Ed., 2000, 39, 3168–3210; (c) O. Mumm, Chem. Ber., 1910, 43, 886–893; (d) P. E. Dawson, T. W. Muir, I. Clark-Lewis and S. B. H. Kent, Science, 1994, 266, 776–779; (e) M. Bergmann, E. Brand and F. Weinmann, Hoppe-Seyler's Z. Physiol. Chem., 1923, 131, 1–17; (f) S. Dos Santos, A. Chandravarkar, B. Mandal, R. Mimna, K. Murat, L. Saucede, P. Tella, G. Tuchscherer and M. Mutter, J. Am. Chem. Soc., 2005, 127, 11888–11889; (g) D. S. Kemp and R. I. Carey, J. Org. Chem., 1993, 58, 2216–2222.
133. For the detailed description of acid azide protocol in segment condensation illustrating the total synthesis of bovine pancreatic ribonuclease (R Nase) see: (a) N. Fujii and H. Yajima, J. Chem. Soc., Perkin Trans. 1, 1981, 789–796; (b) N. Fujii and H. Yajima, J. Chem. Soc., Perkin Trans. 1, 1981, 797–803; (c) N. Fujii and H. Yajima, J. Chem. Soc., Perkin Trans. 1, 1981, 804–810; (d) N. Fujii and H. Yajima, J. Chem. Soc., Perkin Trans. 1, 1981, 811–818; (e) N. Fujii and H. Yajima, J. Chem. Soc., Perkin Trans. 1, 1981, 819–830; (f) N. Fujii and H. Yajima, J. Chem. Soc., Perkin Trans. 1, 1981, 831–841; (g) H. Yajima and N. Fujii, J. Am. Chem. Soc., 1981, 103, 5867–5871.
134. Z-Phe-Val was coupled with H-Pro-NH₂ employing TFFH/TCFH/BTFFH/BTCFH/DFIH/DCIH and 2,4,6-trimethylpyridine (TMP; collidine) with/without HOAt. Thus obtained products were analysed on a C-18 column for RP-HPLC analysis to estimate the preservation of configuration (i.e., the presence of LLL vs. LDL forms of the test peptide).
135. N. C. Rockwell, D. J. Krysan and R. S. Fuller, Anal. Biochem., 2000, 280, 201–208.
136. S. S. A. An, C. C. Lester, J.-L. Peng, Y.-J. Li, D. M. Rothwarf, E. Welker, T. W. Thannhauser, L. S. Zhang, J. P. Tam and H. A. Scheraga, J. Am. Chem. Soc., 1999, 121, 11558–11566.
137. F. Hubler, T. Ruckle, L. Patiny, T. Muamba, J.-F. Guichou, M. Mutter and R. Wenger, Tetrahedron Lett., 2000, 41, 7193–7196.
138. Y.-A. Kim, H.-N. Shin, M.-S. Park, S.-H. Cho and S.-Y. Han, Tetrahedron Lett., 2003, 44, 2557–2560.
139. A. El-Faham and M. Abdul-Ghani, Org. Prep. Proced. Int., 2003, 35, 369.
140. I. Weber, P. Potier and J. Thiery, Tetrahedron Lett., 1999, 40, 7083–7086.
141. F. Effenberger, U. Burkard and J. Villfahrt, Liebigs Ann. Chem., 1986, 2, 314–333.
142. K. Hong and J. P. Morken, J. Am. Chem. Soc., 2013, 135, 9252–9254.
143. J. S. Oliver and A. Oyelere, J. Org. Chem., 1996, 61, 4168–4171.
144. Y. Ito, M. Gerz and Y. Nakahara, Tetrahedron Lett., 2000, 41, 1039–1042.
145. R. Ingenito, D. Dreznjak, S. Guffler and H. Wenschuh, Org. Lett., 2002, 4, 1187–1188.
146. J. Green and K. Bradley, Tetrahedron, 1993, 49, 4141–4146.
147. D. Granitza, M. Beyermann, H. Wenschuh, H. Haber, L. A. Carpino, G. A. Truran and M. Bienert, J. Chem. Soc., Chem. Commun., 1995, 2223–2224.
148. P. Marfey, Carlsberg Res. Commun., 1984, 49, 591–596.
149. G. Kokotos and C. Noula, J. Org. Chem., 1996, 61, 6994–6996.
150. (a) B. A. Bunin and J. A. Ellman, J. Am. Chem. Soc., 1992, 114, 10997–10998; (b) M. J. Plunkett and J. A. Ellman, J. Am. Chem. Soc., 1995, 117, 3306–3307.
151. P. Loss, M. Riedrich and H.-D. Arndt, Chem. Commun., 2009, 1900–1902.
152. A. Bianco, C. P. Sonksen, P. Roepstorff and J.-P. Briand, J. Org. Chem., 2000, 65, 2179–2187.
153. C. Bender and J. Liebscher, ARKIVOC, 2009, vi, 111–136.
154. K.-Y. Hung, P. W. R. Harris and M. A. Brimble, Org. Lett., 2012, 14, 5784–5787.
155. K. Sakamoto, Y. Nakahara and Y. Ito, Tetrahedron Lett., 2002, 43, 1515–1518.
156. C. W. Derstine, D. N. Smith and J. A. Katzenellenbogen, J. Am. Chem. Soc., 1996, 118, 8485–8486.
157. S. Chierici, M. Jourdan, M. Figuet and P. Dumy, Org. Biomol. Chem., 2004, 2, 2437–2441.
158. H. Hojo, H. Kobayashi, R. Ubagai, Y. Asahina, Y. Nakahara, H. Katayama, Y. Ito and Y. Nakahara, Org. Biomol. Chem., 2011, 9, 6807–6813.
159. H. P. Hemantha, G. Chennakrishnareddy, T. M. Vishwanatha and V. V. Sureshbabu, Synlett, 2009, 407–410.
160. V. V. Sureshbabu, H. P. Hemantha and S. A. Naik, Tetrahedron Lett., 2008, 49, 5133–5136.
161. G. Nagendra, R. S. Lamani, N. Narendra and V. V. Sureshbabu, Tetrahedron Lett., 2010, 51, 6338–6341.
162. B. Muller, D. Besser, P. Kleinwachter, O. Arad and S. Reissmann, J. Pept. Res., 1999, 54, 383–393.
163. C. Schumann, L. Seyfarth, G. Greiner and S. Reissmann, J. Pept. Res., 2000, 55, 428–435.
164. M. Jedrzejczak, R. E. Motie, D. P. N. Satchell, R. S. Satchell and W. N. Wassef, J. Chem. Soc., Perkin Trans. 2, 1994, 1471–1479.
165. E. Vedejs, S. Lin, A. Klapars and J. Wang, J. Am. Chem. Soc., 1996, 118, 9796–9797.
166. J. A. Robl, D. S. Karanewsky and M. A. Asaad, Tetrahedron Lett., 1995, 36, 1593–1596.

---

Footnote

† Dedicated to Professor Padmanabhan Balaram, Indian Institute of Science, Bangalore, on the occasion of his retirement.

---