Abstract
The thiol functionality on the methyl group of a threonine derivative [Thr(SH)] facilitates O‐acylation of the Thr hydroxy group with a thioester. We previously showed that a Thr(SH)‐incorporated peptide thioester can be converted to the corresponding Thr‐containing cyclic depsipeptide through intramolecular thioester exchange (S─S acyl transfer) and subsequent desulfurization of the O‐acyl peptide resulting from intramolecular S─O acyl transfer. Based on our success in depsipeptide synthesis, we applied the Thr(SH)‐facilitated protocol to the synthesis of branched O‐acyl isopeptides. Initial attempts identified two issues. First, the S‐acylation step with a thioester proceeds in an entropically preferential manner in cyclic depsipeptide synthesis, but not in the case of branched isopeptides. Using a highly volatile thiol component in the thioester solved this issue. Second, the intermolecular thioester change step was accompanied by the formation of an S,O‐diacyl intermediate as a major component; this issue was solved by using thioester‐selective activation of the diacyl species with silver(I) salt followed by desulfurization. Ultimately, the optimized Thr(SH)‐mediated protocol facilitated the late‐stage O‐acylation of a Thr residue in peptide sequences. We show that the protocol has wide substrate scope and demonstrate its application to ubiquitination of the Thr residue of HOIL‐1 peptide.
Keywords: intermolecular O‐acylation, S─S and S─O acyl transfers, thiol‐incorporated threonine, threonine depsipeptide, trifluoroethanethiol
The late‐stage intermolecular O‐acylation of the threonine (Thr) side chain was accomplished by reacting thiol‐incorporated Thr with a peptide thioester in open air. The thioester, which contained the volatile trifluoroethanethiol, facilitated the formation of S,O‐diacyl peptides. These were subsequently converted into the desired O‐acyl threonine isopeptide through silver‐mediated thioester selective activation followed by desulfurization.
![]()
1. Introduction
The post translational modification (PTM) of peptides and proteins induces structural and functional changes by attaching a variety of functional units mainly on side chains.[ 1 ] O‐Phosphorylation, N‐acetylation, and N‐ubiquitination represent biologically vital modifications. Esterification of serine (Ser) or threonine (Thr) hydroxy and cysteine (Cys) thiol units with carboxylic acids including peptides or fatty acids has also been recognized as a PTM indispensable for cellular events.[ 2 ] Apart from the O‐esterification PTMs, O‐acyl isopeptide derivatives, including clickable[ 3 ] or function‐changeable molecules,[ 4 ] have shown utility in various peptide‐related disciplines. In practical peptide synthesis, esterification of the secondary hydroxyl group of Thr residue in peptide chains has remained a challenging task. Conventional O‐acylation protocol for the Thr in protected peptides is accompanied not only by low coupling efficiency but also by epimerization. Additionally, no acylation method has achieved late‐stage esterification of the Thr residue in non‐protected peptides possessing various side chain functionalities such as amine, carboxylic acid, or hydroxy group by a peptide due to the issue of residue selectivity. The development of a late‐stage esterification strategy is therefore crucial for facilitating the synthesis of ester‐branched peptides. Such an approach could streamline the process significantly relative to the conventional method that depends on pre‐assembled O‐peptidyl intermediates.[ 5 ] Advancing this technique would not only accelerate research but also unlock new opportunities in the chemical and biological studies associated with the O − acyl iso‐peptide format.
Recently, our group achieved late‐stage macrolactonization[ 6 ] for synthesizing cyclic Thr depsipeptides using a thiol‐incorporated Thr derivative[ 7 ] [Thr(SH) (1)] (Figure 1). The protocol features intramolecular S─S acyl transfer as thioester exchange in the Thr(SH)‐containing peptide thioester 2 and subsequent S─O acyl transfer to afford the corresponding O‐acyl peptide 4 in equilibrium with the S‐acyl peptide 3. Desulfurization of the equilibrium mixture affords the desired cyclic Thr depsipeptide 5, shifting the equilibrium to 4. Based on our success in synthesizing cyclic depsipeptides using this Thr(SH)‐mediated late‐stage macrolactonization strategy, in this study we explored the applicability of 1 to late‐stage intermolecular O‐peptidyl ester formation on a Thr residue via tandem acyl transfers followed by desulfurization.
Figure 1.

Synthesis of cyclic Thr depsipeptide by the late‐stage macrolactonization using tandem acyl transfer followed by desulfurization.
In cyclic depsipeptide synthesis, the first intramolecular thioester exchange step (2 to 3) proceeds smoothly with thiol release. In contrast, intermolecular thioester exchange of 3 with the liberating thiol results in the regeneration of 2, which hardly occurs. On the other hand, both the formation of a branched S‐peptidyl ester and its reverse reaction, involved in the synthesis of an O‐peptidyl Thr peptide, are intermolecular thioester exchange reactions; therefore, no quantitative formation of the branched peptide is achievable. To address this, we attempted to use a thioester made from the highly volatile 2,2,2‐trifluoroethanethiol (TFET; boiling point: 35–37 °C)[ 8 ] with the aim of preventing the reverse thioester exchange.
2. Results and Discussion
The protected Thr(SH) derivative (Fmoc‐Thr(STrt)‐OH, Trt = triphenylmethyl) was prepared according to literature.[ 6 ] In this procedure, the methionine (Met) derivative was converted to vinyl glycine, which was then subjected to epoxidation and subsequent ring opening with Trt thiol to afford Fmoc‐Thr(STrt)‐OH.
Initially, the reaction of a Thr(SH)‐containing peptide [Ac‐EG‐Thr(SH)‐GNY‐NH2 (6a), 1 mM] with nonvolatile sodium mercaptoethanesulfonate (MESNa) or TFET thioester [Ac‐VARYA‐SR: R = MESNa (7a) or TFET (7b), 1 mM each] was conducted in 6 M guanidine hydrochloride (Gn·HCl) − 0.2 M sodium phosphate buffer (NaPB) (pH 7.0) at 37 °C in open air (Figure 2 and Table 1).
Figure 2.

Evaluation of intermolecular thioester exchange (S─S acyl transfer) using MESNa or TFET thioester, followed by S─O acyl transfer. Analytical HPLC conditions: 5%–45% linear gradient of 0.1% TFA/CH3CN in 0.1% TFA/H2O over 30 min; UV detection at 220 nm.
Table 1.
Initial assessment of the reaction consisting of intermolecular S─S and intramolecular S─O acyl transfers.
| Entry[ a ] | Thioester | Ratio of 6a and 7 (mM) | Temp.(°C) | Conversion (%) 6a: 7: 8: 9: 10 |
|---|---|---|---|---|
| 1 | 7a | 1:1 | 37 | 13:38:16:10:8 |
| 2 | 7b | 1:1 | 37 | 6:0:14:10:38 |
| 3 | 7b | 1:1 | 4 | 25:54:18:2:1 |
| 4 | 7b | 2:4 | 37 | 1:0:8:5:67(59) |
[a]All reactions were conducted in 6 M Gn·HCl − 0.2 M NaPB (pH 7.0) for 5 h in open air. The reaction was diluted with H2O fivefold and directly analyzed by HPLC. Conversion (%), or remaining (%), proportions were determined by HPLC analysis with UV detection at 220 nm and calculated using the equation: conversion (%) = 100 [(integ. 6a, 7, 8, 9, or 10) / (integ. all peptidic compounds)], where integ. = integration of the peak area of UV absorption.
Replacement of MESNa with TFET resulted in almost complete consumption of the thioester 7b (<5% remaining), while the MESNa thioester 7a remained as a major component (Table 1, entries 1 and 2 and HPLC in Figure 2). Decreasing the temperature of the reaction of 6a with 7b from 37 to 4 °C increased the amount of 7b remaining (entries 2 and 3). These results indicate that the volatility of the liberating thiol is crucial for the first intermolecular thioester exchange to progress. However, the S,O‐dipeptidyl peptide 10 was identified as a major product, even in the equimolar reaction of compounds 6 and 7b (entry 2). The initial S─S acyl transfer (reaction rate: V inter S─S) may not occur rapidly enough to facilitate predominant formation of the S‐peptidyl intermediate 8; this intermediate is then reversibly converted to the O‐peptidyl intermediate 9 (V intra S─O), which undergoes electrophilic attack by thioesters 7 to produce 10. Ideally, if a significant kinetic difference (V inter S─S >> V intra S─O) could be established between these two reactions, it might lead to preferential formation of an equilibrium mixture of the intermediates 8 and 9 without the formation of 10. This would ultimately facilitate the desired formation of the O‐peptidyl branched peptide through the process of desulfurization.
Based on this consideration, we examined the reaction in concentrated solution; however, no significant improvement was observed, which made us rethink the original strategy of subjecting the equilibrium mixture of 8 and 9 to desulfurization. Thus, we next attempted to subject the S,O‐dipeptidyl peptide 10 to the desulfurization step, followed by thioester selective activation. Gratifyingly, the reaction of 6 (2 mM) with 7b (4 mM) at 37 °C for 5 h in 6 M Gn·HCl − 0.2 M NaPB (pH 7.0) in open air afforded the substrate 10 in acceptable yield (conversion 69%, isolated 59%, entry 4, and Figure S1 in Supporting Information (SI)).
The acylation points of 8 and 9 were confirmed by determining whether the peptides could be changed to the desulfurization material by their reaction under ultrafast desulfurization conditions, as discussed below (Figure S2).
In terms of thioester selective activation, we evaluated the use of silver(I) ion (Scheme 1, Step 1). Hojo and Aimoto reported a pioneering protocol for peptide ligation in which a peptide thioester is selectively activated by silver(I) ion to facilitate subsequent aminolysis.[ 9 ] In our envisioned strategy, one concern is whether undesired O─S peptidyl transfer of 9 might occur after the thioester activation; however, this is unlikely to be the case for silver (I)‐mediated activation because the nucleophilicity of the thiol group is lost upon formation of the sulfur–silver (‐S‐Ag) complex as a silver‐protected thiol. The first trial reaction of 10 (1 mM) with an excess of 2‐hydroxyethylamine in the presence of 2.0 equivalents (eq.) of silver(I) trifluoromethanesulfonate (AgOTf) in N‐methylpyrrolidone (NMP) at 4 °C proceeded with complete disappearance of 10 to give a mixture of the silver‐protected thiol peptide 11 and N‐peptidyl hydroxyethylamine 12a on HPLC analysis of the reaction (Table S1, entry 1, and Figure S3a). From the viewpoint of atom economy, it is desirable that the silver‐mediated activation of the thioester of 10 leads to the recovery of the reusable acyl component 12. A survey of several nucleophiles suggested that N‐hydroxysuccinimide (NHS) might have potential in terms of the consumption of 10 and conversion to the reusable acyl component 12b. However, a non‐negligible amount of epimerization and hydrolysis of the resulting NHS ester was observed (Table S1, entry 2 and Figures S3b and S4). Although the reaction with acetohydroxamic acid[ 10 ] did not satisfactorily afford the acetylazanyl ester 12c (Table S1, entry 3, and Figure S3c), including NHS in the reaction facilitated the conversion of highly activated 12b to moderately activated 12c (Table S1, entry 4 and Figure S3d). The resulting 12c not only resisted epimerization and hydrolysis but also became a thioester upon thiol treatment (discussed below).
Scheme 1.

Scheme for the synthesis of O‐peptidyl branched peptide 13a by silver (I)‐mediated thioester selective activation (Step 1), followed by desulfurization (Step 2).
Having established the protocol for thioester selective activation with recovery of the reusable acyl component 12c, we attempted desulfurization of the silver‐protected thiol‐containing peptide 11. We examined desulfurization in Gn·HCl‐containing aqueous buffer because the chloride anion in the reaction will remove the silver ion from the S‐Ag moiety to form a precipitate of AgCl.[ 11 ]
Our evaluation of desulfurization began with the addition of a conventional mixture of desulfurizing reagents.[ 12 ] [VA‐044 (50 eq.), TCEP·HCl (260 eq.), and 5% (v/v) t‐BuSH] in 6 M Gn·HCl − 0.2 M NaPB (pH 7.0) to the reaction of 10 with AgOTf (2 eq.) in NMP in the presence of NHS (200 eq.) and acetohydroxamic acid (200 eq.). Desulfurization at 37 °C proceeded slowly without satisfactory results even after 3 h (Figure 3a). Next, we tried Li and colleagues’ operationally simple (Add) and superfast (Done) desulfurization reaction mediated by commercially available sodium tetraethylborate (NaBEt4), referred to as Add‐and‐Done Desulfurization (ADD).[ 13 ] The addition of desulfurization buffer containing TCEP·HCl (260 eq.) in 6 M Gn·HCl − 0.2 M NaPB (pH 7.0) and NaBEt4 to the thioester‐activated reaction in NMP enabled the desired desulfurization to go to completion within 5 min at ambient temperature, yielding the desired O‐peptidyl branched peptide 13a and reusable acetohydroxamate 12c as a thioester precursor with no significant side reaction (Figure 3b).
Figure 3.

HPLC examination of desulfurization in the presence of VA‐044 (a) or NaBEt4 (b) described in Step 2 in Scheme 1. Analytical HPLC conditions: 5%–45% linear gradient of 0.1% TFA/CH3CN in 0.1% TFA/H2O over 30 min; UV detection at 220 nm. *The Ag‐protected peptide 11 was eluted as the thiol peptide 9 due to dilution with 6 M Gn·HCl − 0.2 M NaPB buffer before HPLC injection.
The sequence of reactions shown in Figure 2 and Scheme 1 involves, first, a thioester exchange step followed by intramolecular S─O peptidyl transfer; second, silver ion‐mediated thioester activation; and third, desulfurization by the ADD protocol. Across this sequence, an aqueous buffer and nonaqueous conditions are indispensable for the efficient and clean progress of the first and second steps, respectively. Generally, a one‐pot/sequential protocol for synthesizing peptides is desirable to increase isolated yield and decrease experimental operation. We therefore attempted to adjust the reaction sequence to produce 13a in a one‐pot manner (Figure 4 and entry 1 in Table of Figure 4). To allow integrant removal by lyophilization, ammonium acetate was used as the aqueous buffer salt.
Figure 4.

Reactions of one‐pot/thioester selective activation and desulfurization sequence and their results. Analytical HPLC conditions: 5%–45% linear gradient of 0.1% TFA/CH3CN in 0.1% TFA/H2O over 30 min; UV detection at 220 nm. *Ag‐protected peptide 11 was eluted as thiol peptide 9. [a] Acylation was conducted for 15 h. *Nonpeptidic materials.
The thioester exchange of 7b (4 mM) with 6a (2 mM) in 50 mM ammonium acetate buffer (pH 7.0) at 37 °C for 5 h in open air gave the desired S,O‐dipeptidyl material 10. The reaction mixture was then brought to a repeated lyophilization step to remove the volatile integrant. The following thioester‐selective activation was performed by using AgOTf (2 eq.) in the presence of NHS (200 eq.) and acetohydroxamic acid (200 eq.) in NMP at 4 °C for 4 h. Desulfurization was then carried out by adding an equal volume of 6 M Gn·HCl − 0.2 M NaPB (pH 7.0) containing TCEP·HCl (260 eq.) and NaBEt4 (100 eq.) with stirring for 5 min. Lastly, purification of the whole reaction mixture by HPLC afforded the desired O‐acyl peptide 13a and reusable hydrolysis‐resistant acetylazanyl ester 12c in isolated yields of 67% and 64%, respectively, calculated from the starting Thr(SH) peptide 6a (Figure S5). Conversion of the resulting 12c to the TFET thioester by reaction with TFET (100 eq.) in 6 M Gn·HCl − 0.2 M NaPB (pH 7.0) at ambient temperature for 1 h afforded 7b. HPLC analysis of the resulting TFET ester, as compared with an authentic sample of the corresponding C‐terminal D‐Ala TFET thioester, indicated that the recovery process was not accompanied by significant epimerization (Figure S6).
Using the established one‐pot protocol, we next examined substrate scope with a focus on the C‐terminal amino acid of TFET thioesters (Table in Figure 4 and Figure S7). The condensation of peptidyl amino acids is generally accompanied by epimerization of the amino acid except for native chemical ligation (NCL) using peptide thioester in aqueous buffer conditions, which features S─S and irreversible S─N acyl transfers. Our protocol utilizes reversible S─O acyl transfer as a substitute for S─N acyl transfer; therefore, it might carry the risk of epimerization. Comparative examination with L‐Ala and D‐Ala thioester indicated that no significant epimerization occurred during the sequence of reactions (entries 2 and 3, and Figures S7a versus S7b). In addition to Ala thioesters, the one‐pot reaction with a Leu thioester proceeded efficiently to give the O‐leucyl isopeptide 13b in 76% yield (entry 4 and Figure S7c). Generally, acylation of a Thr side chain with protected valine (Val) derivatives is challenging; here, however the attempted O‐acylation of Thr(SH) yielded the desired O‐acyl peptide in 66% isolated yield (entry 5 and Figure S7d). Although the use of C‐terminal Lys thioester was accompanied by formation of the lactam peptide, the desired material was obtained in 60% yield (entry 6 and Figure S7e).
Steric hindrance around the Thr(SH) moiety might affect the efficacy of the reaction, but the reaction of Thr(SH) surrounded by Ala proceeded without a decrease in efficacy (entry 7 and Figure S7f). Furthermore, the attempted one‐pot reaction with the acceptor peptide 15, which possesses a phosphate‐protected highly bulky stimulus‐responsive processing residue (Spr),[ 4a ] yielded the desired model function‐changeable peptide 16 in 34% yield (entry 8 and Figure S7g). Upon hydrolysis of the phosphate ester on the Spr‐containing iso‐peptide 16 by alkaline phosphatase, cleavage of the Spr–Thr amide bond, followed by O─N acyl transfer resulted in the formation of split linear peptides (Ac‐VARYA‐Thr‐GNYG‐NH2 (17) and Ac‐KG‐aminochromanone peptide 18) (Figure S8). To confirm the usefulness of the developed protocol, we attempted to synthesize O‐acyl isopeptide 13e by using 2‐methyl‐6‐nitrobenzoic anhydride (MNBA)‐mediated esterification[ 14 ] on a resin‐bound protected peptide. Esterification of Ac‐Glu(OtBu)‐Gly‐Thr‐Gly‐Asn(Trt)‐Tyr(tBu)‐resin with Ac‐Gly‐Phe‐Ala‐OH in the presence of MNBA, DIPEA, and 4‐dimethylaminopyridine in DMF proceeded slowly yielding only a trace amount of the desired product, including an epimerized isomer (Figure S9).
Having confirmed the utility of our protocol for synthesizing model Thr O‐acyl peptides, we next examined its applicability to the preparation of an O‐ubiquitinated peptide.[ 15 ] Ubiquitination on the Thr side chain has been attracting attention as a function‐unknown PTM; however, O‐ubiquitination has been less investigated than the well‐known N‐ubiquitination. Brik and colleagues reported an elegant method for synthesis of a Thr‐ubiquitinated protein by applying NCL‐based technologies to a pre‐assembled Thr‐esterified isopeptide.[ 16 ] As Cohen and coworkers have shown that HOIL‐1, an E3 ligase, is ubiquitinated on its Thr136 residue,[ 17 ] we attempted to synthesize an O‐ubiquitinated HOIL‐1 sequence encompassing the target Thr136 residue as a model protein. Preliminary evaluation showed that a ubiquitin C‐terminal Gly TFET thioester is not only susceptible to hydrolysis but also sparingly soluble in aqueous buffer. Therefore, we substituted Ala for the hydrolysis‐sensitive Gly residue. We also divided the ubiquitin thioester into two parts between Arg54 and Thr55, aiming to conduct the Thr(SH)‐mediated ligation at the junction (Figure 5).[ 7 ] For the requisite Thr(SH)‐containing peptide thioester, we prepared a Trt‐based solubilizing Tag‐attached Ala TFET thioester 19 [ 18 ] with the intention that the smaller and hydrophilic fragment 19 would facilitate easy formation of the S,O‐diacyl peptide intermediate (Figure S10). Reaction of the acceptor HOIL‐1 (133 − 143) Thr(SH) peptide 20 (1 mM) with the tagged thioester 19 (2 mM) proceeded unambiguously relative to the reaction with a full‐length ubiquitin thioester to give a mixture of the S‐ or O‐monoacyl peptide (21 and 22) and S,O‐diacyl peptide 23 (Figure 5, HPLC profile). However, we did not observe complete disappearance of 21 and 22 even though thioester 19 remained in the reaction, probably due to steric hindrance during the diacylation step. After purification of the mixture by HPLC, each purified peptide (21 as an inseparable mixture of 21 and 19′ (hydrolysis product of 19), 22 (11% isolated yield), or 23 (10% isolated yield)) was subjected to the reaction(s) affording the O‐peptidyl intermediate 24 (Figure 6). Application of the slow desulfurization protocol (Reaction 1) using VA‐044, TCEP·HCl, and t‐BuSH to 21 yielded 24 due to establishment of an equilibrium between 21 and 22 during the desulfurization process (Figure S11a). In contrast, the superfast ADD protocol (Reaction 2) accomplished desulfurization of the O‐acyl intermediate 22 to give 24 in 28% isolated yield without establishing equilibrium (HPLC profile a in Figure 6 and Figure S11b). Conversion of the purified S,O‐diacyl peptide 23 to the desired intermediate 24 was conducted by the optimized thioester‐selective activation and superfast ADD desulfurization sequence (Reaction 3) without any accompanying damage to the S‐Trt tag during the Ag‐mediated activation; however, purification of the resulting crude mixture by HPLC gave an inseparable mixture consisting of 24 and the recovered acetylazanyl ester 25 (Figure S11c). The homogeneous intermediate peptide 24 resulting from 22 was subjected to the one‐pot reaction consisting of tag removal, NCL‐like reaction, and desulfurization (Figure 6 and Figure S12). The treatment of obtained 24 with TFA − TIS resulted in the formation of the branched N‐terminal Thr(SH) ubiquitin fragment‐installed HOIL‐1 peptide 26. After lyophilization of the reaction, the resulting residue was brought to the NCL‐like ligation with ubiquitin (1–54) thioester 27 in 6 M Gn·HCl − 0.2 M NaPB (pH 7.0). After the ultrafiltration of the reaction, the resulting residue was desulfurized by TCEP·HCl and NaBEt4 in 6 M Gn·HCl − 0.2 M NaPB (pH 7.0) to yield the desired Thr‐ubiquitinated HOIL‐1 peptide 28 in 90% isolated yield (HPLC profiles b and c in Figure 6), calculated from substrate 24 resulting from 22. The overall yield of 28 from the starting HOIL‐1 peptide 20 is calculated to be 2.8%.
Figure 5.

Initial peptidylation step for the synthesis of O‐, S‐, and S,O‐acylated HOIL‐1 (133–143) peptides with solubilizing Tag‐attached thioester fragment 19 and HPLC analysis of the reaction. Analytical HPLC conditions: 5%–65% linear gradient of 0.1% TFA/CH3CN in 0.1% TFA/H2O over 30 min; UV detection at 220 nm. 21 was obtained as a mixture of 19′.
Figure 6.

Synthesis of O‐ubiquitinated HOIL‐1 (133–143) peptide 28 and HPLC examination of the reactions from O‐acyl intermediate 22. Analytical HPLC conditions: 5%–65% linear gradient of 0.1% TFA/CH3CN in 0.1% TFA/H2O over 30 min; UV detection at 220 nm.
3. Conclusion
Late‐stage peptidylation of the Thr side chain in peptide sequences has been achieved in this study. The key to the success of this protocol is the use of a peptide TFET thioester as an acyl donor and a thiol‐incorporated Thr(SH) residue as an acyl acceptor, to which intermolecular peptidyl transfer from the TFET thioester occurs with evaporation of the volatile TFET. Subsequently, intramolecular S─O acyl transfer and acylation of the regenerating thiol affords an S,O‐dipeptidyl intermediate, and silver‐mediated thioester activation in the presence of an appropriate nucleophile facilitates the formation of a silver‐protected O‐peptidyl intermediate alongside the recovery of a reusable acyl component. Lastly, superfast desulfurization of the silver‐protected mono acyl intermediate using NaBEt4 in the presence of chloride anion gives the desired branched Thr depsipeptide with no equilibrium shift to S‐acyl species. The established protocol enables late‐stage Thr peptidylation without accompanying epimerization and has been successfully applied to synthesis of an Spr‐containing function‐changeable peptide and to ubiquitination of the Thr residue of HOIL‐1 peptide.
4. Experimental Section
4.1. Representative Experimental Procedure
Preparation of peptide TFET thioester (Ac‐Val‐Ala‐Arg‐Tyr‐Ala‐SCH2CF3 (TFET) (7b))
According to Liu's protocol,[ 19 ] the peptide thioester 7b was prepared by converting the precursor hydrazide peptide to the corresponding thioester. Well swelled 2‐chloro Trt chloride resin (co‐polymer of styrene‐1% divinyl benzene, 100 − 200 mesh, chloride content, 1.6 mmol/g) with DMF was treated with hydrazine monohydrate (4 eq.) and Et3N (3 eq) in DMF at rt. for 2 h. The reaction was quenced by the addition of MeOH and the resulting resin was washed with DMF (3 times) to afford the requisite hydrazine 2‐chloro Trt resin. Coupling of Fmoc amino acids and deprotection of the Fmoc group were perfomred on the hydrazine resin by the procedure described in the section of Fmoc SPPS in the SI. (loading of Fmoc‐Ala‐OH: 0.95 mmol/g). After completion of the elongation of the peptide chain, the completed peptide resin was subjected to the acidic deprotection procedure in TFA − TES─H2O (95: 2.5: 2.5, (v/v)) as mentioned in SI to give crude hydrazide peptide. The resulting crude was purified by semi‐preparative reversed‐phase HPLC column (10 × 250 mm) on 8%–18% linear gradient of solvent B (0.1% TFA in CH3CN) in solvent A (0.1% TFA in CH3CN) over 30 min to afford the desired homogeneous hydrazide peptide in 62% isolated yield calculated from the protected resin.
Peptide hydrazide precursor for 7b
HRMS (ESI‐TOF) Calcd for C28H46F3N10O7 [M + H]+ M/Z = 635.3624, Found 635.3626. Analytical HPLC (5%–45% linear gradient of solvent B in solvent A Over 30 min) Retention Time (RT) = 13.5 Min.
To a solution of crude precursor peptide hydrazide in 6 M Gn·HCl − 0.2 M NaPB (pH 3.5) was added 75 eq. of 10% (w/v) aqueous NaNO2 solution at ‐10 °C with additional stirring for 30 min. Then, a solution of TFET (100 eq.) in 6 M Gn·HCl − 0.2 M NaPB (pH 3.5) was added. Then, 6 M aqueous NaOH solution was added to adjust the pH of the reaction at 7.0 with additional reaction at rt. for 1 h. After addition of TCEP·HCl to the reaction, semi‐preparative HPLC purification on the reversed‐phase HPLC column (10 × 250 mm) on 20%–40% linear gradient of solvent B in solvent A over 30 min afforded the titled peptide thioester 7b in 46% isolated yield calculated from the precursor hydrazide peptide.
7b: HRMS (ESI‐TOF) calcd for C30H45F3N8O7S [M + H]+ m/z = 719.3157, found 719.3187. Analytical HPLC (5%–45% linear gradient solvent B in solvent A over 30 min) RT = 25.5 min.
Incorporation of Thr(SH) residue into peptide chain
Regardless of the method for elongating peptide chain, the Thr(SH) residue was manually incorporated into peptide chain. A solution of Fmoc‐Thr(STrt)‐OH (2.0 eq., 0.2 M) in DMF containing DIPCDI (2.0 eq.) and HOBt·H2O (2.0 eq.) was added to an resin with 12 h shaking of the mixture at rt. The standard acidic deprotection procedure mentioned in the Supporting Information was used for obtaining the Thr(SH)‐containing peptide.
One‐pot/sequential reaction sequence of S,O‐diacylation, Ag‐mediated activation and desulfurization
A mixture of 6a (1.0 µmol) and 7b (2.0 µmol) in 50 mM ammonium acetate buffer (pH 7.0, 500 µL) was incubated at 37 °C for 5 h under open air conditions in fume hood with scrubber. During the reaction, 100 mL of H2O was added every one hour to the mixture to keep the reaction concentration. Then, repeated lyophilization of the mixture (2 or 3 times) afforded a residue. For the thioester selective activation, the resulting residue was dissolved in NMP (1 mL) containing AgOTf (2.0 µmol), NHS (200 µmol), and acetohydroxamic acid (200 µmol) at 4 °C with additional stirring for 4 h at the same temperature. To the reaction was added 6 M Gn·HCl − 0.2 M NaPB (pH 7.0) solution (1 mL) containing TCEP·HCl (260 µmol) and NaBEt4 (100 µmol) at 4 °C for desulfurization. The mixture was allowed to warm to room temperature and was stirred for 5 min. The reaction with precipitation (AgCl) was diluted fivefold with 0.1% TFA − 5% CH3CN in H2O. After filtering off the precipitation, the resulting filtrate was purified by semi‐preparative reversed‐phase HPLC column (10 × 250 mm) on 5%–45% linear gradient of solvent B in solvent A over 30 min. HPLC purification, followed by lyophilization, afforded the desired O‐peptidyl Thr peptide 13a and reusable acetylazanyl ester 12c in 67 and 64% isolated yields, respectively.
13a: HRMS (ESI‐TOF) calcd for C56H82N16O19 [M + H]+ m/z = 1283.6015 m found 1283,6016. Analytical HPLC (5%–45% linear gradient solvent B in solvent A over 30 min) RT = 16.6 min.
12c: HRMS (ESI‐TOF) calcd for C32H47N9O9 [M + H]+ m/z = 678.3570, found 678.3580. Analytical HPLC (5%–45% linear gradient solvent B in solvent A over 30 min) RT = 16.6 min.
Conflict of Interests
The authors declare no conflict of interest.
Supporting information
Supporting Information
Acknowledgements
This research was supported in part by The Canon Foundation, by JSPS KAKENHI (23K18187 and 23K27300) (for A.O.), and by JST SPRING, (JPMJSP2113) (for D.S.).
In memory of Professor Luis Moroder, who made a great contribution to the progress of peptide chemistry
Data Availability Statement
The data that support the findings of this study are available in the supplementary material of this article.
References
- 1. Walsh C. T., Garneau‐Tsodikova S., G. J. Gatto, Jr. , Angew. Chem. Int. Ed. 2005, 44, 7342. [DOI] [PubMed] [Google Scholar]
- 2.a) Jiang H., Zhang X., Chen X., Aramsangtienchai P., Tong Z., Lin H., Chem. Rev. 2018, 118, 919; [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Kaiya H., Kojima M., Hosoda H., Koda A., Yamamoto K., Kitajima Y., Matsumoto M., Minamitake Y., Kikuyama S., Kangawa K., J. Biol. Chem. 2001, 276, 40441. [DOI] [PubMed] [Google Scholar]
- 3.a) Taniguchi A., Sohma Y., Kimura M., Okada T., Ikeda K., Hayashi Y., Kimura T., Hirota S., Matsuzaki K., Kiso Y., J. Am. Chem. Soc. 2006, 128, 696. [DOI] [PubMed] [Google Scholar]; b) Sohma Y., Sasaki M., Hayashi Y., Kimura T., Kiso Y., Chem. Commun. 2004, 1, 124. [DOI] [PubMed] [Google Scholar]
- 4.a) Shigenaga A., Tsuji D., Nishioka N., Tsuda S., Itoh K., Otaka A., ChemBioChem 2007, 8, 1929; [DOI] [PubMed] [Google Scholar]; b) Furutani M., Uemura A., Shigenaga A., Komiya C., Otaka A., Matsuura K., Chem. Commun. 2015, 51, 8020; [DOI] [PubMed] [Google Scholar]; c) Jung D., Sato K., Min K., Shigenaga A., Jung J., Otaka A., Kwon Y., Chem. Commun. 2015, 51, 9670; [DOI] [PubMed] [Google Scholar]; d) Shigenaga A., Yamamoto J., Kohiki T., Inokuma T., Otaka A., J. Pept. Sci. 2017, 23, 505. [DOI] [PubMed] [Google Scholar]; e) Inaba H., Uemura A., Morishita K., Kohiki T., Shigenaga A., Otaka A., Matsuura K., Sci. Rep. 2018, 8, 6243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.a) Stawikowski M., Cudic P., in Peptide Characterization and Application Protocols (Ed.: Fields G. B.), Humana Press, Totowa, NJ, 2007, pp. 321. [Google Scholar]; b) Paquette A. R., Boddy C. N., Org. Biomol. Chem. 2023, 21, 8043. [DOI] [PubMed] [Google Scholar]
- 6. Sato D., Denda M., Tsunematsu H., Tanaka N., Konishi I., Komiya C., Shigenaga A., Otaka A., Chem. Commun. 2022, 58, 2918. [DOI] [PubMed] [Google Scholar]
- 7. Chen J., Wang P., Zhu J., Wan Q., Danishefsky S. J., Tetrahedron 2010, 66, 2277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Thompson R. E., Liu X., Alonso‐García N., Pereira P. J. B., Jolliffe K. A., Payne R. J., J. Am. Chem. Soc. 2014, 136, 8161. [DOI] [PubMed] [Google Scholar]
- 9. Hojo H., Aimoto S., Bull. Chem. Soc. Jpn. 1991, 64, 111. [Google Scholar]
- 10. Dunkelmann O.‐D. L., Hirata Y., Totaro K. A., Cohen D. T., Zhang C., Gates Z. P., Pentelute B. L., Proc. Natl. Acad. Sci. 2018, 115, 3752. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Tamamura H., Matsumoto F., Sakano K., Otaka A., Ibuka T., Fujii N., Chem. Commun. 1998, 1, 151. [DOI] [PubMed] [Google Scholar]
- 12. Wan Q., Danishefsky S. J., Angew. Chem. Int. Ed. 2007, 46, 9248. [DOI] [PubMed] [Google Scholar]
- 13. Sun Z., Ma W., Cao Y., Wei T., Mo X., Chow H. Y., Tan Y., Cheung C. H. P., Liu J., Lee H. K., Tse E. C. M., Liu H., Li X., Chem. 2022, 8, 2542. [Google Scholar]
- 14. Shiina I., Hashizume M., Yamai Y., Oshiumi H., Shimazaki T., Takasuna Y.‐j., Ibuka R., Chem. ‐ Eur. J. 2005, 11, 6601. [DOI] [PubMed] [Google Scholar]
- 15.a) Wang X., Herr R. A., Rabelink M., Hoeben R. C., Wiertz E. J. H. J., Hansen T. H., J. Cell Biol. 2009, 187, 655. [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Pao K.‐C., Wood N. T., Knebel A., Rafie K., Stanley M., Mabbitt P. D., Sundaramoorthy R., Hofmann K., van Aalten D. M. F., Virdee S., Nature 2018, 556, 381. [DOI] [PubMed] [Google Scholar]
- 16. Sun H., Meledin R., Sachitan M. M., Brik A., Chem. Sci. 2018, 9, 1661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Kelsall I. R., Zhang J., Knebel A., Arthur J. S. C., Cohen P., Proc. Natl. Acad. Sci. 2019, 116, 13293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Tsuda S., Mochizuki M., Ishiba H., Yoshizawa‐Kumagaye K., Nishio H., Oishi S., Yoshiya T., Angew. Chem. Int. Ed. 2018, 57, 2105. [DOI] [PubMed] [Google Scholar]
- 19. Fang G.‐M., Li Y.‐M., Shen F., Huang Y.‐C., Li J.‐B., Lin Y., Cui H.‐K., Liu L., Angew. Chem. Int. Ed. 2011, 50, 7645. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information
Data Availability Statement
The data that support the findings of this study are available in the supplementary material of this article.
