Low-Racemization Control in Peptide Synthesis: An Inverse N→C Peptide Synthesis Strategy via Activated α-Aminoesters

Introduction

What is commonly described as “low racemization” in peptide synthesis is, in essence, the effort to minimize loss of configuration at the amino acid stereocenter during bond formation, especially at the α-position of the activated carboxylic acid component. Recent reviews have repeatedly pointed out that the risk of epimerization/racemization is closely related to the type of activation intermediate, the way base is involved, the pre-activation time, the intrinsic sensitivity of the substrate, and the reaction temperature.

Against this background, the inverse bond-forming strategy proposed by the Campagne and de Figueiredo groups is methodologically significant. Instead of following the traditional main route of “activating the carboxyl group first, then coupling with the amine,” it uses CDI [N,N′-carbonyldiimidazole] to convert the amino ester side into a couplable activated amino component, thereby exploring peptide synthesis in the N→C direction [from the amino terminus to the carboxyl terminus, N-to-C]. What truly changes here is where the high-energy intermediate is generated before bond formation.

1. Low racemization is not merely a matter of optimizing carboxyl activation systems

Traditional peptide synthesis has long been discussed around systems such as HATU, HBTU, PyBOP, COMU, and DIC/Oxyma because the mainstream peptide-bond-forming route itself is built on carboxyl activation. As long as this pathway is retained, experimental optimization typically takes the form of identifying carboxyl activation conditions that are faster, cleaner, and more compatible with sensitive substrates.

However, much of the epimerization observed in peptide synthesis is also closely tied to this very stage. Once the carboxyl group is activated, if the system forms intermediates that are more prone to inducing epimerization, while also involving a suitable basic environment, a prolonged residence time, or intrinsically sensitive substrates, configurational risk is more easily amplified. Recent reviews have likewise emphasized that low racemization is first and foremost a stereochemical management issue at the activation stage, rather than merely a question of reagent replacement.

Therefore, as long as bond formation still centers on carboxyl activation, one must manage the stereochemical cost associated with that activation mode. An instructive direction is to change the default reaction partner that is activated during bond formation. The importance of the inverse bond-forming route lies in the fact that it pushes this idea one step further: low racemization need not rely solely on continued optimization of carboxyl activation conditions; it may also be achieved by redesigning which component is activated in the bond-forming event.

1.1 Several levels of judgment that should be distinguished in studies of low-racemization peptide synthesis

Level of observation	Common question	More critical question
Reagent level	Which coupling reagent gives lower racemization?	Under what activation framework does this system operate?
Mechanistic level	Can the reaction be made to work?	At which step is configurational loss most likely to occur?
Condition level	Should the additive be changed?	Are the base, pre-activation time, and temperature amplifying the risk?
Method level	Is there a milder carboxyl activation reagent?	Is it possible to stop treating the carboxyl group as the default activation site?

2. How the inverse bond-forming route was progressively established: from the proposal of a new entry point in 2014 to its development and refinement from 2017 to 2022

The 2014 work first introduced the most crucial change in this route: unlike the traditional strategy of activating the carboxylic acid first, the authors instead used activated α-aminoesters as the entry point for peptide synthesis, thereby enabling inverse N→C peptide construction [from the amino terminus to the carboxyl terminus, N-to-C]. The abstract also explicitly states that this method is not only applicable to dipeptide synthesis, but that no racemization was detected in a sensitive Cys [cysteine] example, and that the feasibility of inverse N→C elongation was demonstrated through a model tetrapeptide. For studies on low racemization in peptide synthesis, the importance of this step is that it showed for the first time that the entry point to peptide bond formation does not necessarily have to be built on carboxyl activation; changing the activated component itself may open up new space for stereochemical control.

In 2017, the authors further developed the originally stepwise route, which required preparation and isolation of the activated intermediate, into a sequential one-pot synthesis. According to the abstract, this method relies on the in situ generation of a transient CDI-protected α-aminoester, and the coupling can be carried out under mild, neutral conditions without the addition of extra base. This step shows that the inverse bond-forming strategy began to move from “conceptually feasible” to “experimentally more practical.”

The 2024 Organic Syntheses discussion addendum provided a clearer summary of the condition logic at this stage: in the sequential one-pot method, activation is first carried out in dichloromethane with 1.5 equivalents of CDI [N,N′-carbonyldiimidazole] at room temperature for 30 minutes, followed by addition of the second amino acid component together with catalytic CuBr2 [copper bromide] and HOBt [1-hydroxybenzotriazole] to complete the coupling. Unlike the early stepwise protocol, this route no longer requires added base during the CDI activation stage; the authors also explicitly noted that adding base at this stage reduces the yield of the target dipeptide from 87% to 7%. This indicates that in this method, base is not merely an ordinary auxiliary condition, but a factor that directly affects whether the reaction can proceed as intended.

By 2022, the route had advanced another step. The abstract of the Journal of Organic Chemistry paper shows that the authors developed a Cu(II)/HOBt-catalyzed, microwave-accelerated method for the synthesis of dipeptides and general amides, and demonstrated gram-scale synthesis; at the same time, the method still achieved coupling with no detected racemization under conditions involving sensitive amino acids. The Organic Syntheses addendum further summarizes that this step compressed a coupling process that had previously required about 20 hours down to 30 minutes. By this stage, the route was no longer merely a new idea that “changes the activated partner,” but had become a continuous methodological chain: proposing a new entry point, verifying configurational retention, developing a one-pot protocol, then accelerating the process and extending its applicability.

2.1 Key milestones in the development of the inverse bond-forming route from a new entry point to a refined method

Time	Core advance	What this step demonstrated
2014	First proposal of inverse peptide synthesis through activated α-aminoesters	Demonstrated that “amine activation” can serve as an entry point to peptide synthesis, instead of assuming that only the carboxyl group can be activated
2014	No racemization detected in a model containing the sensitive Cys residue	Showed that configurational retention was built into the value judgment of the method from the very beginning
2017	Establishment of a sequential one-pot protocol, eliminating isolation of the activated intermediate	The method began to move from proof of concept toward a more executable experimental workflow
2017 / 2024 summary	No additional base during the CDI activation stage, and the addition of base significantly suppresses the reaction	Indicates that the condition logic of this route is not the same as that of conventional coupling methods
2022	Coupling time shortened from 20 h to 30 min under microwave or conventional heating, with extension to general amides	Addressed the early problem of slow kinetics and broadened the scope of the method
2022	Demonstration of gram-scale examples	The methodological value was no longer limited to small-scale validation on a few model substrates

3. Why the risk of epimerization may change once the activated partner is changed

Existing literature points to an important issue: when the key intermediate governing bond formation no longer arises from the traditional carboxyl activation pathway, the main source of epimerization risk may also change accordingly. For studies of low racemization in peptide synthesis, this is more important than simply comparing which coupling reagent performs better, because the issue is not one of local condition optimization, but of where the stereochemical risk at the bond-forming stage actually originates.

The 2024 Organic Syntheses discussion addendum provides a more specific summary of the mechanistic understanding of this route. Based on experimental observations, NMR [nuclear magnetic resonance] monitoring, and subsequent DFT [density functional theory] analysis, the authors consider that the reaction is more likely to proceed through a carbamic mixed anhydride intermediate, followed by decarboxylative 1,3-acyl migration to form the amide. At the same time, the available experimental results and literature precedents both indicate that the main productive pathway of this reaction does not proceed through oxazolone or acyl imidazole intermediates to form the amide.

In traditional discussions of peptide synthesis, oxazolone-related pathways have long been regarded as one of the major sources of epimerization risk. If the principal reaction channel in this inverse bond-forming route no longer depends on such intermediates, then a more reasonable interpretation of its good configurational retention with certain substrates is that the key intermediates traversed during bond formation have changed, and therefore the main source of stereochemical risk may also have changed.

4. For which problems should this route be prioritized, and how should its practical position be understood?

This inverse bond-forming route is better suited to research questions that are concerned not only with whether bond formation is possible, but also with bond-forming directionality, configurational retention, mild conditions, and methodological scalability. Existing literature has already shown that it can be used for inverse N→C peptide synthesis, has reported no detected racemization with sensitive substrates, and has subsequently been developed into a sequential one-pot method, accelerated by microwave or conventional heating, and extended to general amides and gram-scale synthesis.

If the research focus is to compare whether bond-forming behavior and stereochemical risk change once the activation site is altered, then this route deserves higher priority than continuing to screen only among traditional coupling reagents. By contrast, if the goal is simply to complete routine short-peptide assembly, existing mainstream routes can already meet that need reliably. Although this inverse bond-forming route now has clear methodological value, it has not yet developed to the point where it can be generally and robustly adopted across most peptide synthesis tasks in the same way as mainstream routes.

The 2024 Organic Syntheses addendum indicates that this route has already been clearly validated for inverse N→C tetrapeptides and general amides; however, the current evidence is still concentrated mainly on short peptides and model systems. A 2025 review further points out that for N→C peptide synthesis to move toward broader application, issues such as anchoring, purification, and chain elongation still need to be addressed.

4.1 Which experimental objectives are better suited to prioritizing this inverse bond-forming route?

Current experimental objective	Is it worth prioritizing?	Main reason
To investigate the source of low-racemization behavior in peptide synthesis, rather than merely screen coupling reagents	Worth prioritizing	This route directly changes the activation site and allows comparison of how changing the activated partner affects configurational retention
To examine whether N→C peptide synthesis is feasible	Worth prioritizing	The method itself is a representative route established and developed specifically around inverse N→C peptide synthesis
To examine mild coupling conditions with sensitive amino acids	Deserves focused attention	Existing literature reports no detected racemization with sensitive substrates and demonstrates that coupling can be completed under mild conditions
To carry out only routine short-peptide assembly	Not necessarily a priority	Existing mainstream routes are already relatively mature, whereas the more distinctive value of this method at present lies in the change of activation site itself
To directly replace automated platforms for long-peptide synthesis	At present, usually not a priority	The current evidence mainly comes from short peptides, model systems, and methodological demonstrations, and is still insufficient to support treating it as a mature routine method

5. Product Selection Guide for Low-Racemization Peptide Synthesis and the “Activated Amine” Inverse Bond-Forming Route (Tables 1–3)

Current research or experimental objective	Recommended table to consult first	Why this table should be consulted first	Recommended companion table(s)	Guidance
To first establish the basic experimental route of “activate the amine first, then couple with the carboxylic acid,” and clarify how to choose the core reagents, solvent, base, and cocatalytic additives	Table 1	Table 1 focuses on the core activation-system and condition-control components used in this route, such as CDI, CuBr2, HOBt, dichloromethane, DIPEA, and triethylamine. It is the most suitable starting point for building a clear operational framework for inverse bond formation.	Then see Table 2	It is easier to make the route work once the activation system and condition window are established first, and then substrate and protecting-group combinations are screened.
To compare the suitability of different amino acid ester substrate classes and determine how methyl ester, ethyl ester, and tert-butyl ester substrates should be used in the activated-amine route	Table 2	Table 2 focuses on activated amino acid ester substrates and protected carboxylic acid coupling partners, directly corresponding to practical selection questions such as “how to pair substrates,” “how to match protecting groups,” and “how to change the coupling partner.”	Consult together with Table 1	Differences among substrates ultimately need to be evaluated again in the context of activation conditions and order of addition, so Table 2 is best read together with Table 1.
To compare where Boc-, Fmoc-, and Cbz-protected carboxylic acid components fit into this route and build a protecting-group selection strategy	Table 2	Table 2 places Boc-L-phenylalanine, Fmoc-Gly-OH, and Cbz-Gly-OH in the same group, making it easier to compare coupling entry points and downstream deprotection planning from the perspective of protection strategy.	Consult together with Table 1	The choice of protecting group affects how the coupling is organized, the pretreatment workflow, and the downstream route connection, so it should be evaluated together with Table 1.
To perform a parallel comparison between the traditional “activated carboxyl” route and the “activated amine” route and observe differences in condition design and configurational retention	Table 2	Table 2 retains both the substrate platform used in the CDI route and PyBOP as a conventional activated-carboxyl control reagent, making it the most suitable table for direct comparison of these two activation strategies.	Consult together with Tables 1 and 3	For comparison, one should examine both the condition system in Table 1 and whether the model residues in Table 3 are sufficient to reveal meaningful differences.
To validate the compatibility of this route with sensitive residues, sulfur-containing residues, aromatic residues, and sterically hindered residues	Table 3	Table 3 focuses on representative model residues such as Cys, Met, Trp, Phe, and Val, making it suitable for experiments centered on residue compatibility, side-chain tolerance, and model short-peptide construction.	Consult together with Table 1	Residue compatibility cannot be judged independently of the specific activation conditions, and should in particular be evaluated together with the base, solvent, and cocatalytic additives listed in Table 1.
To focus on configurational retention and low-racemization performance and establish a more discriminating model system	Table 3	The residues in Table 3 are more suitable for designing experiments that evaluate configurational retention, especially because steric effects, sulfur-containing side chains, and aromatic side chains can be incorporated into the same assessment framework.	Consult together with Tables 1 and 2	First use Table 3 to define the model system, then use Table 1 to choose conditions and Table 2 to choose substrate/protecting-group combinations; this leads to a more complete experimental assessment.
To carry out initial validation with model dipeptides or short peptides and determine which product group is the most efficient starting point	Table 2	Table 2 most directly corresponds to the amino acid ester precursors and protected carboxylic acid components required for model dipeptide construction, making it the most suitable for setting up the first round of substrate combinations.	Consult together with Table 1	Whether the model dipeptide can be formed smoothly still ultimately depends on whether the activation and coupling conditions listed in Table 1 are appropriate.
To optimize a “one-pot” protocol or compare the effects of different pretreatment bases and determine which components should be prioritized in condition screening	Table 1	Table 1 retains triethylamine, DIPEA, CDI, CuBr2, HOBt, and the reaction solvent in the same set, making it most suitable for systematic optimization around salt liberation in pretreatment, the activation stage, and the coupling stage.	Consult together with Table 2	Condition optimization is best carried out with a clearly defined substrate combination, so Table 1 normally needs to be used together with Table 2.

Table 1. Core inverse bond-forming activation system and condition-control components

Category	CAS No.	Aladdin Cat. No.	Name	Specification / Purity	Product Features and Applications
Condition-control base / comparison component for salt liberation	121-44-8	T140677	Triethylamine	Anhydrous, ≥99.5%, Water ≤50 ppm	Commonly used for liberating amino acid ester hydrochlorides and for pretreatment; also suitable as a tertiary amine base control for comparing how different bases affect solubility, addition sequence, and side-reaction control during the coupling stage.
Common reaction solvent for inverse bond formation	75-09-2	D116152	Dichloromethane	Suitable for peptide synthesis, ≥99.8% (GC), stabilized with 50–150 ppm isoamylene	Commonly used for CDI-mediated generation of activated amino intermediates and the subsequent coupling stage; the clear solvent framework facilitates observation of the activation process, the order of addition of catalytic components, and one-pot operations.
Pretreatment base for hydrochloride salt liberation / basic-condition control component	7087-68-5	D109322	N,N-Diisopropylethylamine	Redistilled, ≥99.5%	Commonly used to liberate amino acid ester hydrochlorides, facilitating formation of a free amino ester system suitable for dropwise addition; also suitable for parallel comparison with triethylamine to examine how base sterics and the basic environment affect the organization of the activation stage.
Cu(II) cocatalytic component	7789-45-9	C105401	Cupric bromide	AR, ≥99%	Forms a cocatalytic system with HOBt to promote coupling between the activated amino component and the carboxylic acid; suitable for establishing inverse N→C bond-forming conditions and comparing conversion efficiency under catalytic versus non-catalytic conditions.
Core reagent for activated amine generation	530-62-1	C109315	N,N'-Carbonyldiimidazole (CDI)	≥99%	Core activating reagent that converts amino acid esters into couplable activated amino components; suitable for establishing the inverse bond-forming route of “activate the amine first, then couple with the carboxylic acid,” and for examining the reactivity of different ester-type substrates.
Coupling-promoting additive / cocatalytic component	123333-53-9	H106176	1-Hydroxybenzotriazole Monohydrate	≥97%	Commonly used together with CuBr2 to improve performance at the coupling stage; suitable for optimizing coupling efficiency, crude-product quality, and the condition window in the activated-amine route.

Table 2. Activated amino acid ester substrates and protected carboxylic acid coupling components in inverse bond formation

Category	CAS No.	Aladdin Cat. No.	Name	Specification / Purity	Product Features and Applications
Activated amino acid ester platform substrate / ethyl ester type	623-33-6	G105477	Glycine ethyl ester hydrochloride	≥99%	A typical ethyl ester amino acid ester precursor, suitable for generating ethyl ester-type activated amino components and for comparing how different ester groups affect activation efficiency, coupling rate, and downstream deprotection planning.
Conventional activated-carboxyl control reagent	128625-52-5	P109336	1H-Benzotriazol-1-yloxytripyrrolidinophosphonium Hexafluorophosphate	≥98%	A classic phosphonium-type carboxyl-activation reagent, suitable for parallel comparison with the CDI-mediated activated-amine route in order to examine differences in condition design, coupling efficiency, and configurational retention between traditional carboxyl activation and inverse bond formation.
Protected carboxylic acid coupling component / Boc	13734-34-4	B105463	Boc-L-phenylalanine	≥98%	A commonly used Boc-protected amino acid carboxylic acid component, suitable for coupling with activated amino acid esters to construct model dipeptides and for comparing coupling performance and purification convenience under different N-protection strategies.
Protected carboxylic acid coupling component / Fmoc	29022-11-5	F103019	Fmoc-Gly-OH	≥98%	A commonly used Fmoc-protected amino acid carboxylic acid component, suitable for evaluating the compatibility of the inverse bond-forming route with the Fmoc protection strategy and for providing reference points for subsequent solution-phase or solid-phase extension.
Activated amino acid ester platform substrate / methyl ester type	2491-20-5	A100815	L-Alanine methyl ester hydrochloride	≥98%	A typical chiral α-amino acid ester precursor that can be used to establish methyl ester-type activated amino components; suitable for investigating configurational retention during coupling, activation stability, and model dipeptide construction.
Protected carboxylic acid coupling component / Cbz	1138-80-3	Z110912	N-Carbobenzyloxyglycine	≥98%	A commonly used Cbz-protected amino acid carboxylic acid component, suitable for parallel comparison with Boc and Fmoc in order to assess how different N-protecting groups affect coupling efficiency, purification routes, and downstream deprotection planning.
Activated amino acid ester platform substrate / tert-butyl ester type	27532-96-3	G109218	Glycine tert-butyl ester hydrochloride	≥98%	A tert-butyl ester amino acid ester precursor, suitable for comparing how bulky ester groups affect the stability of activated amino intermediates, coupling reactivity, and downstream deprotection strategy.

Table 3. Representative residue models and components for configurational-retention evaluation

Category	CAS No.	Aladdin Cat. No.	Name	Specification / Purity	Product Features and Applications
Representative sterically hindered residue model	72-18-4	L755656	L-Valine	UltraBio™, ≥99.5% (NT)	A common amino acid with substantial branched-chain steric hindrance, suitable for constructing sterically sensitive model systems to investigate bond-forming efficiency, condition tolerance, and sequence-extension performance.
Representative aromatic heterocyclic residue model	73-22-3	L755707	L-Tryptophan	UltraBio™, ≥99.5% (NT)	An aromatic heterocyclic residue containing an indole side chain, suitable for evaluating the compatibility of inverse coupling with aromatic heterocyclic residues as well as downstream analytical and purification behavior.
Representative sensitive thiol-containing residue model	52-90-4	L755721	L-Cysteine	UltraBio™, ≥98.5% (RT)	A more sensitive model residue containing a thiol group, suitable for testing mild coupling conditions, configurational-retention capability, and compatibility with sulfur-containing side chains.
Representative neutral sulfur-containing residue model	63-68-3	M101130	L-Methionine	Moligand™, ≥99%	A neutral residue containing a thioether side chain, suitable for evaluating the compatibility and stability of the activated-amine route toward sulfur-containing substrates that do not contain a free thiol group.
Representative aromatic hydrophobic residue model	63-91-2	P110424	L-Phenylalanine	Moligand™, ≥99%	A common aromatic hydrophobic residue, suitable for constructing model dipeptides or short peptides with a clear analytical window for comparing bond-forming efficiency and separation performance under different activation modes.

Note: The above are representative Aladdin products. For additional product specifications, please refer to the product list at the end of the article or search by “product name/CAS/catalog number” on the Aladdin website.

References

[1] Suppo JS, Subra G, Bergès M, de Figueiredo RM, Campagne JM. Inverse Peptide Synthesis via Activated α-Aminoesters. Angew Chem Int Ed. 2014;53(21):5389-5393. doi:10.1002/anie.201402147.

[2] de Figueiredo RM, Suppo JS, Midrier C, Campagne JM. Sequential One-Pot Synthesis of Dipeptides through the Transient Formation of CDI-N-Protected α-Aminoesters. Adv Synth Catal. 2017;359(11):1963-1968. doi:10.1002/adsc.201700034.

[3] Tosi E, Campagne JM, de Figueiredo RM. Amine Activation: “Inverse” Dipeptide Synthesis and Amide Function Formation through Activated Amino Compounds. J Org Chem. 2022;87(18):12148-12163. doi:10.1021/acs.joc.2c01288.

[4] Campagne JM, de Figueiredo RM. Discussion Addendum for: Dipeptide Syntheses via Activated α-Amino Esters. Org Synth. 2024;101:508-523. doi:10.15227/orgsyn.101.0508.

[5] Duengo S, Muhajir MI, Hidayat AT, Musa WJA, Maharani R. Epimerisation in Peptide Synthesis. Molecules. 2023;28(24):8017. doi:10.3390/molecules28248017.

[6] Ghosh K, Lubell WD. N- to C-Peptide Synthesis, Arguably the Future for Sustainable Production. J Pept Sci. 2025;31(6):e70019. doi:10.1002/psc.70019.

[7] Suppo JS, de Figueiredo RM, Campagne JM. Dipeptide Syntheses via Activated α-Aminoesters. Org Synth. 2015;92:296-308. doi:10.15227/orgsyn.092.0296.

For more related articles, see below:

The Methodological Value of Boc-Oxyma: Low-Racemization Coupling, Route Continuity, and Extended Applications

Low Racemization in Peptide Synthesis Is Not Just About Reagent Choice: Stereochemical Risk During the Activation Stage and Practical Strategies for Experimental Control

From High Reactivity to Comprehensive Evaluation: Using CTSOAt as an Example to Understand the Design Logic of New Coupling Reagents

Understanding ATD-DMAP: From Reagent Design to Esterification, Amidation, and Stereochemical Integrity

A New Balance in Coupling Reagents: How DMT-TU Balances Reactivity, Racemization Control, and Thermal Hazard

The Positioning of TBTU in Practice: Racemization Risk, Extended Uses, and Selection Criteria

The Methodological Value of HATU in Difficult Couplings: Reactivity Advantages, Stereochemical Retention, and Condition Management

From “High Yield” to “Configuration Retention”: Activation-Stage Risks, Route Differences, and Selection Principles for Coupling Systems