Covalent inhibitors generally refer to irreversible covalent inhibitors; however, in detailed studies, they can be divided into irreversible covalent inhibitors (ICIs) and reversible covalent inhibitors (RCIs).
Advantages of covalent inhibitors
- Prolongs the residence time of the inhibitor on the targeted protein, providing resistance to drug resistance.
- Inhibits targets traditionally considered “undruggable,” including those with extensive or shallow pockets, or even targets without pockets, as well as large and flat functional interfaces involved in protein-protein interactions (PPIs).
- Dual-mode selection of protein targets, imparting additional affinity to drug affinity elements, enhancing targeting efficacy. This is because covalent inhibition generally involves two steps: first, like non-covalent inhibition, the inhibitor binds to the target protein with affinity; then, reactive amino acid residues on the target form covalent bonds, either through irreversible suicide inhibition or by forming covalent adducts to increase affinity, making the inhibitor more competitively inhibitory.
Covalent “warheads” increase the stickiness of inhibitors to target proteins. The reaction between inhibitors and target amino acid residues can be either permanently irreversible or temporarily reversible. Irreversible covalent inhibitors permanently glue the affinity elements and target proteins together like glue, while reversible covalent inhibitors are like tying a “live buckle” between the inhibitor and the target, which can be restored. Therefore, factors affecting the balance constants of NCI, ICI, and RCI differ.
The advantages of covalent inhibition mentioned above have led to its revival. However, the inherent characteristics of covalency also bring some concerns to researchers.
- Structurally, covalent inhibitors consist of affinity elements and reactive warheads. Therefore, there are concerns about their excessive activity, off-target effects, harming “innocent” protein targets. ICIs cause permanent damage, while RCIs also “harm the innocent,” but their reversibility kinetics mean they can also “release the innocent.”
- Concerns about their stability, inability to withstand the complex environments of plasma or in vivo metabolism, premature “inactivation,” or even toxicity from reactive functional groups during metabolism. Some components containing active groups entering the bloodstream and the body generally exhibit poor pharmacokinetics and lower stability.
- Concerns about immune activation, such as allergenic reactions caused by adducts or degradation products formed by inhibitors and proteins.
These concerns once limited the development of covalent drugs, with researchers mostly following an empirical golden rule-developing reversible drugs, i.e., non-covalent inhibitors. However, even non-covalent inhibitors have been withdrawn from the stage of history due to toxicity. There are also many covalent drugs playing pivotal roles in treatment. Therefore, concerns detrimental to covalent inhibition can completely turn into targets for design and overcoming in research and development.
In drug development, researchers are dedicated to expanding the therapeutic efficacy of drugs while reducing their side effects. For covalent inhibitors, off-target issues, pharmacokinetic issues, and metabolic stability issues are inevitable and need to be addressed in the drug development process.
Chemical structure and advantages of reversible covalent inhibitors
The electrophilic warheads of covalent inhibitors are often quite evident, possessing significant structural features such as β-lactams with ring strain, α,β-unsaturated carbonyl compounds, etc. Some are concealed in prodrugs, only revealing their covalent structure after metabolism, like omeprazole. There are also cases where the structural observation is not clear but confirmed as covalent mechanisms in mechanistic studies, such as 3-nitropropionic acid as a covalent inhibitor of aconitase 3, and Wortmannin as a covalent inhibitor of phosphatidylinositol-3 kinase.
For reversible covalent inhibitors, reaction kinetics enable reversible, non-permanent modification of “innocent” targets. Ultimately, the inhibitor will release from both the inhibitor and the “innocent” target, thus mitigating or avoiding off-target effects and associated toxicity. The warheads suitable for reversible covalent inhibition have also seen some development in recent years.
The mechanism of action targeting amino acid residues of the target protein differs and can be mainly classified into three categories:
Targeting cysteine residues: Mainly includes α-cyanoacrylamide and its derivatives, cyanide derivatives, aldehyde derivatives, α-fluoro-substituted acrylamide esters, fluoro-chloro-substituted methyl ketones, thiomethyl tetrazines, etc.
Targeting serine/threonine residues: Mainly includes α-ketoamides, boronic acids, etc.
Targeting lysine residues and N-terminal: Mainly includes aldehydes, with adjacent hydroxyl groups providing hydrogen bond stabilization, or boronic acids providing hydrogen bond coordination to stabilize imine structures.
In recent years, several reversible covalent inhibitors have been reported, with red markings indicating reversible covalent warheads. The process of reversible covalent inhibition consists of two steps: affinity binding and reversible bonding. Clearly, reversible covalent inhibition does not permanently alter protein targets, thus reducing the risk of off-target effects and immune activation to some extent. The reactivity of such warheads is weaker, and their stability in plasma and in vivo is relatively improved compared to ICIs (Irreversible Covalent Inhibitors). In the equation, K represents the equilibrium constant, τ represents the residence time of the inhibitor-target adduct, and researchers aim to obtain covalent inhibitors with the smallest Ki* and the largest τ possible.
Biochemical analysis of reversible covalent warheads
How to determine if an inhibitor is a reversible covalent inhibitor? Mechanistic studies and confirmation cannot solely rely on empirical judgment from chemical structure. It requires validation of the reversibility of the inhibitor warhead with target residues, activity determination, residence time measurement, off-target effect assessment, etc. All of these require certain technical support.
Reversibility analysis: Confirming reversibility is relatively straightforward and can generally be achieved through techniques such as:
Nuclear Magnetic Resonance (NMR): By comparing the NMR chemical signals of RCI-Target adducts with unbound RCIs, it’s easy to distinguish between them. Diluting the NMR sample of the covalent adduct, if it’s a reversible covalent binding, the system will re-equilibrate, reducing the concentration of the adduct and obtaining more unbound RCIs. Additionally, if the complex or RCIs have significant UV/visible spectroscopic signals, this method can also determine Kd.
Mass Spectrometry (MS): Injecting the covalent adduct into a mass spectrometer, during the MS experiment, the adduct will decompose, releasing the inhibitor and the native protein, which can be observed on the MS spectrum.
Bioluminescence Resonance Energy Transfer (BRET): Using competitive inhibition assays with BRET, in conjunction with washout assays, to verify reversible covalency. Firstly, the protein target is saturated with the covalent inhibitor, then a fluorescence tracking reagent is introduced to replace the covalent inhibitor. If it can competitively replace it, then the covalent inhibition is reversible. Furthermore, the degradation or denaturation of the covalent inhibitor-protein adduct, and subsequent recovery of the inhibitor, further prove reversible covalent inhibition.
Residence time analysis: The residence time of reversible covalent inhibitors (RCIs) can be determined through fluorescence competition assays or pharmacokinetic/pharmacodynamic (PK/PD) analysis.
Off-target effect assessment: Due to the inevitable need to address and resolve the toxicity of off-target effects in covalent inhibition, assessing off-target effects of covalent inhibitors is crucial. Currently, there are mainly three methods reported.
Classic methods include: Incubating with radioactively labeled inhibitors, followed by gel filtration, electrophoresis, gel excision, Edman sequencing, and identification of amino acid-inhibitor adducts.
Modern methods include mass spectrometry: After separating the target protein of interest (e.g., chromatography, electrophoresis), covalent modification analysis is conducted (if the modification sites are known, protein digestion may not be necessary; if the modification sites are unknown, protein digestion is required).
Chemical proteomics is another approach for detecting off-target effects of covalent inhibitors: By labeling reversible covalent inhibitors (e.g., with affinity, fluorescence, or radioactive labels), adducts can be separated and analyzed (e.g., using affinity tags for separation from gels, or using fluorescence or radioactive labels for separation from gels). This method has been applied to identify target compounds in drug discovery and can also be used to identify off-target proteins.
It should be noted that the presence of covalent adducts does not necessarily mean that associated effects are adverse. The identification of off-target adducts can provide evidence for assumptions of side effects, but if the components of off-target adducts are unknown, further exploration must be conducted, which may lead to unexpected discoveries!
Design of reversible covalent inhibitors
Developing effective reversible covalent inhibitors requires considering a series of issues:
- There must be a covalent warhead.
- The covalent warhead must have some selectivity for specific nucleophilic residues.
- The covalent adduct must be reversible.
- Covalent drugs must have good affinity binding to the desired target.
A good reversible covalent inhibitor exhibits certain general characteristics, thus a generalized process for the design of optimal reversible covalent inhibitors is summarized.
Step 1: Computational analysis and virtual modeling. Ensure that the target has a nucleophilic residue suitable for irreversible covalent inhibition (such as serine, threonine, lysine, cysteine, or N-terminal amino groups). However, computational analysis and virtual modeling are not essential for the successful development of RCIs, especially in cases where high-quality structural data of the target are not available. There are already some modeling frameworks for RCIs, as well as modeling experiments using quantum mechanics (QM), QM/molecular mechanics (MM), or QM/MM combined with molecular dynamics methods.
Step 2: Describe the equilibrium state of products and reactants using the law of mass action. Introduce the law of mass action to evaluate the binding kinetics of inhibitor-target through dynamic equilibrium and residence time assessment. Equilibrium favors the formation of covalent adducts; the thermodynamic favorability of the reverse reaction of off-target effects; if the dissociation constant Ka is small, there is no difference between reversible and irreversible inhibitors; optimize the residence time of both target and off-target enzymes.
Step 3: Similarity of inhibitor binding kinetics to reversible covalent inhibitors. This includes the characterization of inhibitors along the kinetic description, spectral assays after biochemical characterization of histone deacetylase 4, among other methods such as tracking after washing (e.g., JAK3 inhibitors), inhibitor recovery after trypsin digestion, etc.
Step 4: Inhibitor-glutathione adducts should not accumulate for an extended period. Glutathione is associated with electrophilic stress; high reversibility is necessary to avoid stress reactions; competitive glutathione digestion assays are conducted to address stress reactions.
Step 5: Minimize global stress responses. Covalent inhibition may lead to protein misfolding, and if the covalent inhibitor is nonspecific, it may cause global protein misfolding stress responses. Protein misfolding stress caused by covalent inhibition is observed to ensure limited off-target effects.
Step 6: Impact on immune cells. Besides effective target binding, another important consideration in developing covalent inhibitors is their impact on immune cells. Immune cells are highly sensitive to electrophilic stress, which can cause either immune stimulation or immune suppression. Therefore, testing the effects of inhibitors on B cells, T cells, and macrophages is conducted to determine whether the inhibitors induce pro-inflammatory or anti-inflammatory responses. The expression and secretion of cytokines are important for avoiding potential complications of immune activation.
Case study of designing reversible covalent inhibitors
Protein kinase – JAK3 inhibitors
Janus kinases (JAKs) are key regulatory enzymes that mediate cellular signal responses, especially in the secretion of cytokine signals involving immune cell responses. There are four classes of JAK subtypes in humans, including JAK1, JAK2, JAK3, and TYK2, each highly similar, making the development of highly selective drugs challenging.
Broad-spectrum JAK inhibitors have been developed clinically, such as FDA-approved tofacitinib, but their effects on multiple subtypes are not clear, especially considering the different roles of different JAK subtypes in immune health. Research has found that JAK3 has a unique cysteine residue (Cys909), while the residues at the same position in other subtypes are serine. Therefore, designing selectively targeted covalent inhibitors has great potential.
Using newly developed irreversible covalent inhibitors 1-3, known irreversible JAK3 covalent inhibitor compound 4, and two known non-covalent JAK3 inhibitors 5 and 6 shown in the figure, researchers characterized and determined a novel irreversible JAK3 covalent inhibitor with good activity (low nM) and selectivity (>500-fold).
Inhibitors 1-3 exhibit good in vitro inhibitory activity, but their electrophilic warheads undergo off-target reactions before attacking JAK3, leading to adverse reactions. Therefore, researchers designed reversible covalent inhibitors based on the polycyclic scaffold of compound 7 and compared them with saturation equivalent groups of warheads. They found that the inhibitory activity of reversible covalent inhibitor design was more than 4 times that of the saturation equivalent. Compounds 13 and 14 containing reversible covalent inhibitory warheads show good inhibitory activity against JAK3 (127 pM, 154 pM) and high selectivity relative to other subtypes (400-fold to 5800-fold).
Protein Kinase—BTK Inhibitors
BTK is a key enzyme in the immune regulatory response associated with antibody signaling and immune cell recruitment, expressed in most hematopoietic cells, playing a crucial role in the signaling cascade of B cells, and is an important immunotherapeutic target. BTK also has a cysteine residue, while similar kinases have serine residues, which enhances the congenital condition for designing selective targeted covalent inhibitors. As shown above, ibrutinib, acalabrutinib, zanubrutinib, and tirabrutinib are approved irreversible covalent BTK inhibitors targeting the Cys481 residue. Although they have strong inhibitory effects on BTK, kinases with cysteine residues similar to Cys481 are also prone to irreversible inhibition, leading to off-target effects. In response to this, researchers developed the reversible covalent inhibitor rilzabrutinib. Through off-target effect testing, its off-target side effects have been well controlled, and it has now entered phase II/III clinical trials.
Protease—EV71 cPro Inhibitors
Proteases (also known as proteases) exist in all life structures, and protease inhibitors have been widely used in the treatment of cancer, bacterial infections, hypertension, type 2 diabetes, Alzheimer’s disease, non-alcoholic fatty liver disease, and fibrosis. Proteases are generally classified according to their catalytic reactive residues, and different types have different mechanisms of action, so reversible covalent inhibitors can serve as effective tools for designing selective targeted inhibitors.
There have been many successful developments of reversible covalent inhibitors for proteases in clinical practice. Examples include reversible covalent inhibitors targeting serine proteases such as Bortezomib and Ixazomib for serine proteases; Boceprevir, Saxaglibtin, and Telaprevir for serine proteases; and Nirmatrelvir for cysteine proteases. The latest example of the development of a reversible covalent inhibitor targeting a cysteine protease is Enterovirus 71 (EV71)—used to treat one of the pathogens associated with hand, foot, and mouth disease.
Rupintrivir is a known irreversible covalent inhibitor of EV71 homologous human rhinovirus, with a covalent warhead targeting the cysteine residue of acrylic acid ethyl ester. Researchers modified the covalent warhead of Rupintrivir to obtain covalent compounds 15-22, hoping to obtain highly efficient, reversible, and selective covalent inhibitors against EV71 3C protease. Among them, α-carbonyl-N-propylacetamide, aldehyde, acrylamide methyl ester/methylamine, α-halogenated acrylamide methyl ester, α-cyano-substituted acrylamide methylamine t-butyl ester, and α-cyano-substituted acryloyl t-butyl ester were selected as covalent warheads. Researchers used MS to detect the reversibility of inhibitors and kinetics fitting data to detect the activity of the target.
Reference
Patel, Disha, Zil E. Huma, and Dustin Duncan. “Reversible Covalent Inhibition─ Desired Covalent Adduct Formation by Mass Action.” ACS Chemical Biology (2024).