Predictive Multivariate Models for Bioorthogonal Inverse-Electron Demand Diels–Alder Reactions

Trans-cyclooctene-a Swiss army knife for bioorthogonal chemistry: exploring the synthesis, reactivity, and applications in biomedical breakthroughs

BACKGROUND

Trans-cyclooctenes (TCOs) are highly strained alkenes with remarkable reactivity towards tetrazines (Tzs) in inverse electron-demand Diels-Alder reactions. Since their discovery as bioorthogonal reaction partners, novel TCO derivatives have been developed to improve their reactivity, stability, and hydrophilicity, thus expanding their utility in diverse applications.

MAIN BODY

TCOs have garnered significant interest for their applications in biomedical settings. In chemical biology, TCOs serve as tools for bioconjugation, enabling the precise labeling and manipulation of biomolecules. Moreover, their role in nuclear medicine is substantial, with TCOs employed in the radiolabeling of peptides and other biomolecules. This has led to their utilization in pretargeted nuclear imaging and therapy, where they function as both bioorthogonal tags and radiotracers, facilitating targeted disease diagnosis and treatment. Beyond these applications, TCOs have been used in targeted cancer therapy through a "click-to-release" approach, in which they act as key components to selectively deliver therapeutic agents to cancer cells, thereby enhancing treatment efficacy while minimizing off-target effects. However, the search for a suitable TCO scaffold with an appropriate balance between stability and reactivity remains a challenge.

CONCLUSIONS

This review paper provides a comprehensive overview of the current state of knowledge regarding the synthesis of TCOs, and its challenges, and their development throughout the years. We describe their wide ranging applications as radiolabeled prosthetic groups for radiolabeling, as bioorthogonal tags for pretargeted imaging and therapy, and targeted drug delivery, with the aim of showcasing the versatility and potential of TCOs as valuable tools in advancing biomedical research and applications.

Computational Organic Chemistry: The Frontier for Understanding and Designing Bioorthogonal Cycloadditions

Computational organic chemistry has become a valuable tool in the field of bioorthogonal chemistry, offering insights and aiding in the progression of this branch of chemistry. In this review, I present an overview of computational work in this field, including an exploration of both the primary computational analysis methods used and their application in the main areas of bioorthogonal chemistry: (3 + 2) and [4 + 2] cycloadditions. In the context of (3 + 2) cycloadditions, detailed studies of electronic effects have informed the evolution of cycloalkyne/1,3-dipole cycloadditions. Through computational techniques, researchers have found ways to adjust the electronic structure via hyperconjugation to enhance reactions without compromising stability. For [4 + 2] cycloadditions, methods such as distortion/interaction analysis and energy decomposition analysis have been beneficial, leading to the development of bioorthogonal reactants with improved reactivity and the creation of orthogonal reaction pairs. To conclude, I touch upon the emerging fields of cheminformatics and machine learning, which promise to play a role in future reaction discovery and optimization.

Synthesis and evaluation of fluorine-18 labelled tetrazines as pre-targeting imaging agents for PET

BACKGROUND

The brain is a challenging target for antibody-based positron emission tomography (immunoPET) imaging due to the restricted access of antibody-based ligands through the blood-brain barrier (BBB). To overcome this physiological obstacle, we have previously developed bispecific antibody ligands that pass through the BBB via receptor-mediated transcytosis. While these radiolabelled ligands have high affinity and specificity, their long residence time in the blood and brain, typical for large molecules, poses another challenge for PET imaging. A viable solution could be a two-step pre-targeting approach which involves the administration of a tagged antibody that accumulates at the target site in the brain and then clears from the blood, followed by administration of a small radiolabelled molecule with fast kinetics. This radiolabelled molecule can couple to the tagged antibody and thereby make the antibody localisation visible by PET imaging. The in vivo linkage can be achieved by using the inverse electron demand Diels-Alder reaction (IEDDA), with trans-cyclooctene (TCO) and tetrazine groups participating as reactants. In this study, two novel 18F-labelled tetrazines were synthesized and evaluated for their potential use as pre-targeting imaging agents, i.e., for their ability to rapidly enter the brain and, if unbound, to be efficiently cleared with minimal background retention.

RESULTS

The two compounds, a methyl tetrazine [18F]MeTz and an H-tetrazine [18F]HTz were radiolabelled using a two-step procedure via [18F]F-Py-TFP synthesized on solid support followed by amidation with amine-bearing tetrazines, resulting in radiochemical yields of 24% and 22%, respectively, and a radiochemical purity of > 96%. In vivo PET imaging was performed to assess their suitability for in vivo pre-targeting. Time-activity curves from PET-scans showed [18F]MeTz to be the more pharmacokinetically suitable agent, given its fast and homogenous distribution in the brain and rapid clearance. However, in terms of rection kinetics, H-tetrazines are advantageous, exhibiting faster reaction rates in IEDDA reactions with dienophiles like trans-cyclooctenes, making [18F]HTz potentially more beneficial for pre-targeting applications.

CONCLUSION

This study demonstrates a significant potential of [18F]MeTz and [18F]HTz as agents for pre-targeted PET brain imaging due to their efficient brain uptake, swift clearance and appropriate chemical stability.

Multiconfigurational photodynamics simulations reveal the mechanism of photodecarbonylations of cyclopropenones in explicit aqueous environments

Gas-evolving photochemical reactions use light and mild conditions to access strained organic compounds irreversibly. Cyclopropenones are a class of light-responsive molecules used in bioorthogonal photoclick reactions; their excited-state decarbonylation reaction mechanisms are misunderstood due to their ultrafast (<100 femtosecond) lifetimes. We have combined multiconfigurational quantum mechanical (QM) calculations and non-adiabatic molecular dynamics (NAMD) simulations to uncover the excited-state mechanism of cyclopropenone and a photoprotected cyclooctyne-(COT)-precursor in gaseous and explicit aqueous environments. We explore the role of H-bonding with fully quantum mechanical explicitly solvated NAMD simulations for the decarbonylation reaction. The cyclopropenones pass through asynchronous conical intersections and have dynamically concerted photodecarbonylation mechanisms. The COT-precursor has a higher quantum yield of 55% than cyclopropenone (28%) because these trajectories prefer to break a σCC bond to avoid the strained trans-cyclooctene geometries. Our solvated simulations show an increased quantum yield (58%) for the systems studied here.

Reaction performance prediction with an extrapolative and interpretable graph model based on chemical knowledge

Accurate prediction of reactivity and selectivity provides the desired guideline for synthetic development. Due to the high-dimensional relationship between molecular structure and synthetic function, it is challenging to achieve the predictive modelling of synthetic transformation with the required extrapolative ability and chemical interpretability. To meet the gap between the rich domain knowledge of chemistry and the advanced molecular graph model, herein we report a knowledge-based graph model that embeds the digitalized steric and electronic information. In addition, a molecular interaction module is developed to enable the learning of the synergistic influence of reaction components. In this study, we demonstrate that this knowledge-based graph model achieves excellent predictions of reaction yield and stereoselectivity, whose extrapolative ability is corroborated by additional scaffold-based data splittings and experimental verifications with new catalysts. Because of the embedding of local environment, the model allows the atomic level of interpretation of the steric and electronic influence on the overall synthetic performance, which serves as a useful guide for the molecular engineering towards the target synthetic function. This model offers an extrapolative and interpretable approach for reaction performance prediction, pointing out the importance of chemical knowledge-constrained reaction modelling for synthetic purpose.

Biotechnology applications of proteins functionalized with DNA oligonucleotides

The functionalization of proteins with DNA through the formation of covalent bonds enables a wide range of biotechnology advancements. For example, single-molecule analytical methods rely on bioconjugated DNA as elastic biolinkers for protein immobilization. Labeling proteins with DNA enables facile protein identification, as well as spatial and temporal organization and control of protein within DNA-protein networks. Bioconjugation reactions can target native, engineered, and non-canonical amino acids (NCAAs) within proteins. In addition, further protein engineering via the incorporation of peptide tags and self-labeling proteins can also be used for conjugation reactions. The selection of techniques will depend on application requirements such as yield, selectivity, conjugation position, potential for steric hindrance, cost, commercial availability, and potential impact on protein function.

Barrier Height Prediction by Machine Learning Correction of Semiempirical Calculations

Different machine learning (ML) models are proposed in the present work to predict density functional theory-quality barrier heights (BHs) from semiempirical quantum mechanical (SQM) calculations. The ML models include a multitask deep neural network, gradient-boosted trees by means of the XGBoost interface, and Gaussian process regression. The obtained mean absolute errors are similar to those of previous models considering the same number of data points. The ML corrections proposed in this paper could be useful for rapid screening of the large reaction networks that appear in combustion chemistry or in astrochemistry. Finally, our results show that 70% of the features with the highest impact on model output are bespoke predictors. This custom-made set of predictors could be employed by future Δ-ML models to improve the quantitative prediction of other reaction properties.

The 1,3-Dipolar Cycloaddition: From Conception to Quantum Chemical Design

The 1,3-dipolar cycloaddition (1,3-DCA) reaction, conceptualized by Rolf Huisgen in 1960, has proven immensely useful in organic, material, and biological chemistry. The uncatalyzed, thermal transformation is generally sluggish and unselective, but the reactivity can be enhanced by means of metal catalysis or by the introduction of either predistortion or electronic tuning of the dipolarophile. These promoted reactions generally go with a much higher reactivity, selectivity, and yields, often at ambient temperatures. The rapid orthogonal reactivity and compatibility with aqueous and physiological conditions positions the 1,3-DCA as an excellent bioorthogonal reaction. Quantum chemical calculations have been critical for providing an understanding of the physical factors that control the reactivity and selectivity of 1,3-DCAs. In silico derived design principles have proven invaluable for the design of new dipolarophiles with tailored reactivity. This review discusses everything from the conception of the 1,3-DCA all the way to the state-of-the-art methods and models used for the quantum chemical design of novel (bioorthogonal) reagents.

Computational insights into the inverse electron-demand Diels-Alder reaction of norbornenes with 1,2,4,5-tetrazines: norbornene substituents' effects on the reaction rate

The study of the reaction rates and mechanism of click chemistry reactions still remains an interesting challenge in organic chemistry. In this regard, the inverse electron demand Diels-Alder (IEDDA) reaction represents a promising metal-free alternative with enhanced reaction rates compared to other reactions of the click chemistry toolbox. Among the different types of dienophiles used in the IEDDA reactions, norbornenes have been widely used given their high stability and fast reaction rates. The inverse electron-demand Diels Alder reaction of 3,6-dipyridin-2-yl-1,2,4,5-tetrazine with a series of norbornene derivatives was studied with quantum mechanical calculations at the M06-2X/6-311+G(d,p) level of theory. The theoretical predictions were confirmed with the experimental data and analyzed with the use of the distortion/interaction model. The obtained results will help in obtaining a better understanding of the factors that affect the relative cycloaddition rates of norbornenes with tetrazines, which are crucial for selectively tuning their efficacy.

Recent Advances in the Development of Tetrazine Ligation Tools for Pretargeted Nuclear Imaging

Tetrazine ligation has gained interest as a bio-orthogonal chemistry tool within the last decade. In nuclear medicine, tetrazine ligation is currently being explored for pretargeted approaches, which have the potential to revolutionize state-of-the-art theranostic strategies. Pretargeting has been shown to increase target-to-background ratios for radiopharmaceuticals based on nanomedicines, especially within early timeframes. This allows the use of radionuclides with short half-lives which are more suited for clinical applications. Pretargeting bears the potential to increase the therapeutic dose delivered to the target as well as reduce the respective dose to healthy tissue. Combined with the possibility to be applied for diagnostic imaging, pretargeting could be optimal for theranostic approaches. In this review, we highlight efforts that have been made to radiolabel tetrazines with an emphasis on imaging.

Amino acids with fluorescent tetrazine ethers as bioorthogonal handles for peptide modification

A set of 3-bromo-1,2,4,5-tetrazines with three distinct substitutions have been used as reagents for late-stage functionalization of small molecules through nucleophilic aromatic substitution. Spectroscopic studies of the products obtained proved that tetrazine ethers are intrinsically fluorescent. This fluorescence is lost upon inverse Electron-Demand Diels–Alder (iEDDA) cycloaddition with strained alkenes. Tetrazine-phenol ethers are rather interesting because they can undergo rapid iEDDA reactions with a second order rate constant (k₂) compatible with bioorthogonal ligations. As a showcase, l-tyrosine was derivatized with 3-bromo-6-methyl-1,2,4,5-tetrazine and coupled to the peptide drug octreotide. This peptide was detected in cellular flow cytometry, and its fluorescence turned off through a bioorthogonal iEDDA cycloaddition with a strained alkene, showing for the first time the detection and reactivity of intrinsically fluorescent tetrazines in a biologically relevant context. The synthesis and characterization of fluorescent tetrazine ethers with bioorthogonal applicability pave the way for the generation of useful compounds for both detection and bioconjugation in vivo.

Octreotide derivatized with the fluorogenic amino acid 6-methyltetratrazinyl tryosine. Emission spectra before and after the iEDDA cycloaddition.

Catalytic asymmetric aromatizing inverse electron-demand [4+2] cycloaddition of 1-thioaurones and 1-azaaurones

Using 1-thioaurones and 1-azaaurones as electron-deficient oxa-dienes, an organocatalytic asymmetric aromatizing inverse electron-demand [4+2] cycloaddition with γ-deconjugated butenolides and azlactones was developed. A wide range of optically active benzothiophene-fused δ-lactones and indole-fused δ-lactones were obtained with desirable outcomes (up to 94% yield, >99 : 1 dr and 99% ee).

Enzymatic non-covalent synthesis of supramolecular assemblies as a general platform for bioorthogonal prodrugs activation to combat drug resistance

Multi-drug resistance (MDR) is one of the leading causes of the anticancer failures. Besides the blockage of the MDR pathways, the development of more potent drugs is with urgent needs, but has been postponed mainly due to an imbalance between safety and efficacy. The recent development of the bioorthogonal prodrug activation strategy has shown immense potential to balance safety and efficacy, while recent studies only focused on few drug entities such as doxorubicin and monomethyl auristatin E, leaving the vast collection of toxins undetermined. Here we have enumerated typical molecular entities ranging from food and drug administration (FDA) approved drugs to a heated antibody drug conjugates (ADC) warhead and a trichothecene toxin to demonstrate that the bioorthogonal caging and specific activation could serve as a general design to increase the therapeutic index of bioactive molecules. These prodrugs can be efficiently activated on-demand by the bioorthogonal activators whose distribution was regulated by the cancer cell specific enzymatic non-covalent synthesis of supramolecular self-assemblies. The prodrug activation not only enhanced the synergistic therapeutic effect within a broad range of dose ratios but also allowed the convenient switching of drug identities to successfully combat MDR tumor in vivo. In general, this strategy might serve as a general platform, which can be readily applicable to enlarge the therapeutic window for various bioactive molecules. We envision that the spatiotemporal controlled bioorthogonal prodrug activation would facilitate the discovery of anticancer drugs.

Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems

Machine learning models are poised to make a transformative impact on chemical sciences by dramatically accelerating computational algorithms and amplifying insights available from computational chemistry methods. However, achieving this requires a confluence and coaction of expertise in computer science and physical sciences. This Review is written for new and experienced researchers working at the intersection of both fields. We first provide concise tutorials of computational chemistry and machine learning methods, showing how insights involving both can be achieved. We follow with a critical review of noteworthy applications that demonstrate how computational chemistry and machine learning can be used together to provide insightful (and useful) predictions in molecular and materials modeling, retrosyntheses, catalysis, and drug design.

Inverse Electron-Demand Aza-Diels-Alder Reaction of Cyclic Enamides with 1,2-Diaza-1,3-dienes in Situ Generated from α-Halogeno Hydrazones: Access to Fused Polycyclic Tetrahydropyridazine Derivatives

An efficient inverse electron-demand aza-Diels-Alder reaction of cyclic enamides and 1,2-diaza-1,3-dienes, which could be readily formed in situ from α-halogeno hydrazones and a base, has been successfully developed. With the developed approach, a wide range of fused polycyclic tetrahydropyridazines were smoothly obtained in up to 99% yield under benign reaction conditions. This reaction concept was also extended to acyclic enamide substrates for accessing 1,4,5,6-tetrahydropyridazines. A gram-scale experiment and further derivatizations of the polycyclic tetrahydropyridazine products were also conducted to verify the practicability of the methodology.

One-Pot and Two-Chamber Methodologies for Using Acetylene Surrogates in the Synthesis of Pyridazines and Their D-Labeled Derivatives

Acetylene surrogates are efficient tools in modern organic chemistry with largely unexplored potential in the construction of heterocyclic cores. Two novel synthetic paths to 3,6-disubstituted pyridazines were proposed using readily available acetylene surrogates through flexible C₂ unit installation procedures in a common reaction space mode (one-pot) and distributed reaction space mode (two-chamber): (1) an interaction of 1,2,4,5-tetrazine and its acceptor-functionalized derivatives with a CaC₂ -H₂ O mixture performed in a two-chamber reactor led to the corresponding pyridazines in quantitative yields; (2) [4+2] cycloaddition of 1,2,4,5-tetrazines to benzyl vinyl ether can be considered a universal synthetic path to a wide range of pyridazines. Replacing water with D₂ O and vinyl ether with its trideuterated analog in the developed procedures, a range of 4,5-dideuteropyridazines of 95-99% deuteration degree was synthesized for the first time. Quantum chemical modeling allowed to quantify the substituent effect in both synthetic pathways.

Oxidation-Induced "One-Pot" Click Chemistry

Click chemistry has been established rapidly as one of the most valuable methods for the chemical transformation of complex molecules. Due to the rapid rates, clean conversions to the products, and compatibility of the reagents and reaction conditions even in complex settings, it has found applications in many molecule-oriented disciplines. From the vast landscape of click reactions, approaches have emerged in the past decade centered around oxidative processes to generate in situ highly reactive synthons from dormant functionalities. These approaches have led to some of the fastest click reactions know to date. Here, we review the various methods that can be used for such oxidation-induced "one-pot" click chemistry for the transformation of small molecules, materials, and biomolecules. A comprehensive overview is provided of oxidation conditions that induce a click reaction, and oxidation conditions are orthogonal to other click reactions so that sequential "click-oxidation-click" derivatization of molecules can be performed in one pot. Our review of the relevant literature shows that this strategy is emerging as a powerful approach for the preparation of high-performance materials and the generation of complex biomolecules. As such, we expect that oxidation-induced "one-pot" click chemistry will widen in scope substantially in the forthcoming years.

How Oriented External Electric Fields Modulate Reactivity

A judiciously oriented external electric field (OEEF) can catalyze a wide range of reactions and can even induce endo/exo stereoselectivity of cycloaddition reactions. The Diels-Alder reaction between cyclopentadiene and maleic anhydride is studied by using quantitative activation strain and Kohn-Sham molecular orbital theory to pinpoint the origin of these catalytic and stereoselective effects. Our quantitative model reveals that an OEEF along the reaction axis induces an enhanced electrostatic and orbital interaction between the reactants, which in turn lowers the reaction barrier. The stronger electrostatic interaction originates from an increased electron density difference between the reactants at the reactive center, and the enhanced orbital interaction arises from the promoted normal electron demand donor-acceptor interaction driven by the OEEF. An OEEF perpendicular to the plane of the reaction axis solely stabilizes the exo pathway of this reaction, whereas the endo pathway remains unaltered and efficiently steers the endo/exo stereoselectivity. The influence of the OEEF on the inverse electron demand Diels-Alder reaction is also investigated; unexpectedly, it inhibits the reaction, as the electric field now suppresses the critical inverse electron demand donor-acceptor interaction.

Organic reactivity from mechanism to machine learning

As more data are introduced in the building of models of chemical reactivity, the mechanistic component can be reduced until 'big data' applications are reached. These methods no longer depend on underlying mechanistic hypotheses, potentially learning them implicitly through extensive data training. Reactivity models often focus on reaction barriers, but can also be trained to directly predict lab-relevant properties, such as yields or conditions. Calculations with a quantum-mechanical component are still preferred for quantitative predictions of reactivity. Although big data applications tend to be more qualitative, they have the advantage to be broadly applied to different kinds of reactions. There is a continuum of methods in between these extremes, such as methods that use quantum-derived data or descriptors in machine learning models. Here, we present an overview of the recent machine learning applications in the field of chemical reactivity from a mechanistic perspective. Starting with a summary of how reactivity questions are addressed by quantum-mechanical methods, we discuss methods that augment or replace quantum-based modelling with faster alternatives relying on machine learning.

Nucleophilicity Prediction via Multivariate Linear Regression Analysis

The concept of nucleophilicity is at the basis of most transformations in chemistry. Understanding and predicting the relative reactivity of different nucleophiles is therefore of paramount importance. Mayr’s nucleophilicity scale likely represents the most complete collection of reactivity data, which currently includes over 1200 nucleophiles. Several attempts have been made to theoretically predict Mayr’s nucleophilicity parameters N based on calculation of molecular properties, but a general model accounting for different classes of nucleophiles could not be obtained so far. We herein show that multivariate linear regression analysis is a suitable tool for obtaining a simple model predicting N for virtually any class of nucleophiles in different solvents for a set of 341 data points. The key descriptors of the model were found to account for the proton affinity, solvation energies, and sterics.

Automated Experimentation Powers Data Science in Chemistry

Data science has revolutionized chemical research and continues to break down barriers with new interdisciplinary studies. The introduction of computational models and machine learning (ML) algorithms in combination with automation and traditional experimental techniques has enabled scientific advancement across nearly every discipline of chemistry, from materials discovery, to process optimization, to synthesis planning. However, predictive tools powered by data science are only as good as their data sets and, currently, many of the data sets used to train models suffer from several limitations, including being sparse, limited in scope and requiring human curation. Likewise, computational data faces limitations in terms of accurate modeling of nonideal systems and can suffer from low translation fidelity from simulation to real conditions. The lack of diverse data and the need to be able to test it experimentally reduces both the accuracy and scope of the predictive models derived from data science. This Account contextualizes the need for more complex and diverse experimental data and highlights how the seamless integration of robotics, machine learning, and data-rich monitoring techniques can be used to access it with minimal human labor.We propose three broad categories of data in chemistry: data on fundamental properties, data on reaction outcomes, and data on reaction mechanics. We highlight flexible, automated platforms that can be deployed to acquire and leverage these data. The first platform combines solid- and liquid-dosing modules with computer vision to automate solubility screening, thereby gathering fundamental data that are necessary for almost every experimental design. Using computer vision offers the additional benefit of creating a visual record, which can be referenced and used to further interrogate and gain insight on the data collected. The second platform iteratively tests reaction variables proposed by a ML algorithm in a closed-loop fashion. Experimental data related to reaction outcomes are fed back into the algorithm to drive the discovery and optimization of new materials and chemical processes. The third platform uses automated process analytical technology to gather real-time data related to reaction kinetics. This system allows the researcher to directly interrogate the reaction mechanisms in granular detail to determine exactly how and why a reaction proceeds, thereby enabling reaction optimization and deployment.

Machine learning meets mechanistic modelling for accurate prediction of experimental activation energies

Accurate prediction of chemical reactions in solution is challenging for current state-of-the-art approaches based on transition state modelling with density functional theory. Models based on machine learning have emerged as a promising alternative to address these problems, but these models currently lack the precision to give crucial information on the magnitude of barrier heights, influence of solvents and catalysts and extent of regio- and chemoselectivity. Here, we construct hybrid models which combine the traditional transition state modelling and machine learning to accurately predict reaction barriers. We train a Gaussian Process Regression model to reproduce high-quality experimental kinetic data for the nucleophilic aromatic substitution reaction and use it to predict barriers with a mean absolute error of 0.77 kcal mol⁻¹ for an external test set. The model was further validated on regio- and chemoselectivity prediction on patent reaction data and achieved a competitive top-1 accuracy of 86%, despite not being trained explicitly for this task. Importantly, the model gives error bars for its predictions that can be used for risk assessment by the end user. Hybrid models emerge as the preferred alternative for accurate reaction prediction in the very common low-data situation where only 100–150 rate constants are available for a reaction class. With recent advances in deep learning for quickly predicting barriers and transition state geometries from density functional theory, we envision that hybrid models will soon become a standard alternative to complement current machine learning approaches based on ground-state physical organic descriptors or structural information such as molecular graphs or fingerprints.

Hybrid reactivity models, combining mechanistic calculations and machine learning with descriptors, are used to predict barriers for nucleophilic aromatic substitution.

Artificial intelligence and automation in computer aided synthesis planning

In this perspective we deal with questions pertaining to the development of synthesis planning technologies over the course of recent years.

Tailored Bioorthogonal and Bioconjugate Chemistry: A Source of Inspiration for Developing Kinetic Target-Guided Synthesis Strategies

Kinetic target-guided synthesis (KTGS) is a promising tool for the discovery of biologically active compounds. It relies on the identification of potent ligands that are covalently assembled by the biological targets themselves from a pool of reagents. Significant effort is devoted to developing new KTGS strategies; however, only a handful of biocompatible reactions are available, which may be insufficient to meet the specificities (stability, dynamics, active site topology, etc.) of a wide range of biological targets with therapeutic potential. This Topical Review proposes a retrospective analysis of existing KTGS ligation tools, in terms of their kinetics and analogy with other biocompatible reactions, and provides new clues to expand the KTGS toolkit. By way of examples, a nonexhaustive selection of such chemical ligation tools belonging to different classes of reactions as promising candidate reactions for KTGS are suggested.