#8 Drug design IV: covalent inhibitors

Introduction to Covalent Drugs

The field of drug discovery extensively explores molecules that form strong interactions with their biological targets. Among these, covalent drugs represent a distinct class that forms stable covalent bonds with specific residues on enzymes or receptors.

• These interactions are critical for their mechanism of action, often leading to prolonged or irreversible inhibition of the target protein.

• A historical perspective often viewed covalent drugs with caution due to concerns about their potential for non-specific binding and toxicity, leading to a period of hesitance in their development.

• Baxdalen, an N-methylamine, serves as a notable early example within the broader discussion of drugs utilizing covalent bonds, highlighting the early recognition of such mechanisms.

Understanding Covalent Drugs

• Definition of Covalent Drugs:

  • Covalent drugs are defined by their ability to form a chemical bond, specifically a covalent bond, with their biological targets, which can include various receptors and enzymes. This bond is formed through an electrophilic 'warhead' on the drug reacting with a nucleophilic residue on the target protein.

  • This covalent bond formation is a key distinguishing feature, leading to a highly stable, often irreversible, attachment of the drug to its target. The stability of the bond prevents easy dissociation, unlike non-covalent interactions.

  • The consequence of covalent bonding is a prolonged inhibition of the target protein. This means the drug's effect persists even after the unbound drug has been cleared from the system.

  • Recovery from inhibition requires de novo synthesis of new target proteins (enzymes or receptors) to replace the inhibited ones, making the process time-dependent on protein turnover.

• Long-lasting inhibition:

  • The duration of action for covalent drugs is directly contingent on the turnover rate of the target protein. For targets with slow turnover, the drug's effect can last for days or even weeks after the drug itself has been cleared from the system.

  • This characteristic can lead to advantages such as less frequent dosing and sustained therapeutic effects. However, it also poses challenges if off-target effects are encountered or if dose-limiting toxicities occur, as the adverse effects would also be long-lasting and difficult to reverse without new protein synthesis.

Historical Context and Examples of Covalent Drugs

Historically, there was a prevalent perception that covalent drugs were inherently more toxic due to their irreversible nature and potential to react with off-target proteins. This perception led to a significant hesitance in their active pursuit during early periods of drug discovery, with many in the field viewing them as undesirable.

Despite this historical apprehension, many widely used and successful drugs function through covalent mechanisms:

• Aspirin (Acetylsalicylic acid):

  • One of the oldest known covalent inhibitors, aspirin acetylates a serine residue (Ser-530) in cyclooxygenase-1 (COX-1) and COX-2 enzymes, irreversibly inhibiting prostaglandin synthesis. This action underlies its anti-inflammatory, analgesic, antipyretic, and anti-platelet aggregation effects. Its covalent mechanism was elucidated significantly after its initial widespread use.

• Penicillin:

  • A revolutionary class of antibiotics that acts by forming a covalent bond with the active site serine residue of bacterial transpeptidases, also known as penicillin-binding proteins (PBPs). This irreversible acylation inhibits the final cross-linking step in bacterial cell wall synthesis, leading to bacterial death. The discovery of its precise covalent mechanism further legitimized this approach in drug design.

• Other older covalent inhibitors:

  • Omeprazole (and other proton pump inhibitors): Form covalent bonds with cysteine residues on the H+/K+-ATPase pump in gastric parietal cells, irreversibly inhibiting acid secretion.

  • Mao Inhibitors (e.g., Phenelzine): Form covalent adducts with monoamine oxidase enzymes, preventing the breakdown of neurotransmitters.

Contemporary applications have seen a resurgence in the development of covalent inhibitors, particularly in oncology and infectious diseases. For example, various new kinase inhibitors are specifically designed as covalent inhibitors, targeting mutant kinases or unique cysteine residues to enhance selectivity and efficacy.

Mechanism of Action of Covalent Drugs

• Mechanism of Action:

  • The fundamental mechanism involves an electrophilic drug molecule reacting with a nucleophilic amino acid residue within the target protein. This interaction forms a stable covalent bond.

  • The drug typically contains an electrophilic 'warhead' (e.g., an acrylamide, nitrile, epoxide, or aldehyde) which is a functional group that is electron-deficient and seeks an electron-rich site for reaction.

  • Common nucleophilic amino acids in target proteins that are exploited by covalent inhibitors include:

    • Serine (Ser): The hydroxyl group ($-OH$) of serine can act as a nucleophile, especially when activated within an enzymatic active site. This is often seen in serine proteases and esterases. Examples include aspirin and penicillin.

    • Cysteine (Cys): The thiol group ($-SH$) of cysteine is highly nucleophilic due to the relatively low $pKa$ of its proton (around 8.0-8.5 in many protein environments, meaning a significant portion exists as the deprotonated thiolate form, $-S^-$ at physiological pH). Its larger size and polarizability compared to oxygen also contribute to its reactivity. It is a favored target for many targeted covalent inhibitors due to its relatively lower abundance and unique reactivity profile in some protein pockets, which can confer specificity.

    • Lysine (Lys): The primary amine group ($-NH2$) of lysine can also be nucleophilic. However, it is less commonly exploited than serine or cysteine because its amine group is typically protonated ($-NH3^+$) at neutral pH, making it lessnucleophilic. Its global $pKa$ is usually around 10-10.5, so only a small fraction is deprotonated and reactive at physiological pH ($\text{pH } 7.4$).

    • Histidine (His): The imidazole nitrogen can also act as a nucleophile but is generally less reactive than cysteine or serine under physiological conditions. Its $pKa$ is around 6-7, meaning it can be protonated or deprotonated near physiological pH, offering some nucleophilicity depending on its microenvironment.

• Catalysis:

  • Enzymatic reactions often involve a nucleophilic residue within their active sites to facilitate catalysis (e.g., the catalytic triad in serine proteases: Ser-His-Asp). Covalent inhibitors frequently exploit these enzymatic mechanisms, mimicking natural substrates or transition states to ensure precise targeting within the active site.

Interaction Dynamics in Covalent Inhibition

For the effective covalent modification of a protein by a small molecule inhibitor, a multi-step process must be understood and optimized. This process is often described by two key kinetic phases:

1. Noncovalent Interaction (Equilibrium):

   • The initial step involves the rapid and reversible formation of a noncovalent complex between the inhibitor (I) and the enzyme (E). This is a rapid binding step driven by reversible interactions such as hydrogen bonds, van der Waals forces, hydrophobic interactions, and electrostatic interactions.

   • This binding is characterized by a specific equilibrium dissociation constant ($KI$), which reflects the affinity of the inhibitor for the target before covalent bond formation. A lower $KI$ indicates stronger initial noncovalent binding.

   • The equation for this step is: $E + I \leftrightharpoons E \cdot I$

2. Covalent Bond Formation:

   • Following the initial noncovalent binding, the electrophilic warhead on the inhibitor reacts with the strategically positioned nucleophilic residue in the target protein within the pre-formed $E \cdot I$ complex.

   • This step is governed by a rate constant ($k_2$), which defines the intrinsic rate of covalent bond formation (units are $min^{-1}$ or $s^{-1}$).

   • This second step is typically irreversible for many covalent drugs, leading to the formation of a stable, long-lived $E-I_{cov}$ adduct. However, some covalent inhibitors are designed to form reversible covalent bonds (e.g., through Michael addition to thiols which can be reversed by displacement), where the bond can eventually hydrolyze, providing a more tunable duration of action.

   • The overall reaction pathway is: $E + I \leftrightharpoons E \cdot I \xrightarrow{k2} E-I{cov}$

• Importance of drug positioning:

  • Optimal positioning is crucial. The noncovalent interactions must effectively orient the electrophilic warhead adjacent to the target nucleophile within the binding pocket. This precise alignment is essential for the $k2$ step to occur efficiently. Poor initial noncovalent binding ($KI$ is too high) or improper orientation can lead to slow or no covalent modification, even if the electrophilic warhead is intrinsically reactive.

• Determining $KI$ and $k2$ :

  • These parameters are critical for characterizing the efficiency and mechanism of a covalent inhibitor. The overall rate of inactivation of the enzyme by a covalent inhibitor can be described by an observed pseudo-first-order rate constant ($k_{obs}$).

  • $k{obs}$ is typically measured by incubating the enzyme with various concentrations of the inhibitor and monitoring the decrease in enzyme activity over time. Since the covalent inactivation process follows pseudo-first-order kinetics (assuming [I] >> [E]), a plot of $ln(\text{activity})$ versus time yields a straight line with a slope equal to $-k{obs}$ .

  • When inhibitor concentration ([I]) is much greater than the enzyme concentration ([E]), the reactions appear pseudo-first-order. The relationship between $k{obs}$, $KI$, $k2$, and [I] is given by the hyperbolic equation: $k{obs} = \frac{k2 \cdot [I]}{KI + [I]}$

  • By plotting $k{obs}$ against [I] and fitting the data to this hyperbolic equation, one can extract the individual values for $KI$ (the dissociation constant of the noncovalent complex, representing binding affinity prior to covalent bond formation) and $k_2$ (the maximal rate of covalent bond formation at saturating inhibitor concentrations).

  • The efficiency of a covalent inhibitor is often expressed as the ratio $k2/KI$ (units $M^{-1}s^{-1}$ or $M^{-1}min^{-1}$), known as the efficiency constant or inactivation efficiency. A higher $k2/KI$ value indicates a more potent and efficient covalent inhibitor, reflecting both strong initial binding ($KI$) and rapid covalent bond formation ($k2$ ).

Strategies in Developing Covalent Inhibitors

Two primary strategies guide the rational design of covalent inhibitors, each with distinct advantages and challenges regarding specificity and reactivity:

1. Mechanism-Based Inhibitors (Suicide Inhibitors):

   • These inhibitors are often substrate analogues that are initially unreactive or minimally reactive. Their 'warhead' only becomes electrophilic after being chemically transformed by the target enzyme during its normal catalytic cycle.

   • The enzyme utilizes its own catalytic machinery (e.g., a nucleophilic amino acid or co-factor) to process the inhibitor, generating a highly reactive species (e.g., an electrophile) within its active site. This reactive species then forms an irreversible covalent bond with a nearby residue on the enzyme, thereby 'suicidally' inactivating it.

   • Advantages: Offers a high degree of specificity because the inhibitor must be recognized and processed by the enzyme itself. It conceptually relies on the enzyme's unique catalytic properties.

   • Challenges: Can potentially yield specificity challenges if other enzymes share similar catalytic mechanisms or active site residues capable of activating the inhibitor. Design can be complex to ensure activation only occurs inside the target active site.

   • Examples: Clavulanic acid (a beta-lactamase inhibitor), some ornithine decarboxylase inhibitors.

2. Targeted Covalent Inhibitors (TCIs):

   • These inhibitors are designed to selectively target specific, accessible nucleophilic residues (most commonly cysteine due to its unique reactivity profile) that may or may not be part of the enzyme's immediate active site, but are often located within or adjacent to the binding pocket.

   • TCIs typically consist of two main components: a noncovalent recognition element that provides potent and selective binding to the target protein and an electrophilic 'warhead' precisely positioned to react with the target nucleophile.

   • They rely on strong initial noncovalent interactions with the protein's binding pocket ($KI$) to achieve high affinity and precisely orient the electrophilic warhead. This initial binding facilitates efficient covalent bond formation ($k2$ ) with a nearby, often unique, nucleophilic residue.

   • Advantages: Can achieve high selectivity by targeting residues that are unique or differentially exposed in related proteins (e.g., a non-catalytic cysteine unique to a specific isoform or a mutant protein). This avoids pan-reactivity often feared with highly reactive electrophiles.

   • Challenges: Requires careful design to balance warhead reactivity (must be reactive enough but not overly so) and noncovalent affinity. Identification of suitable, unique nucleophilic residues can be challenging. An over-reactive warhead can lead to off-target effects.

   • Examples: Osimertinib, Ibrutinib, Afatinib (all kinase inhibitors).

Challenges and Considerations for Covalent Inhibitors (Safety Issues)

Despite their therapeutic advantages, covalent inhibitors present several inherent challenges that must be carefully addressed during drug development:

• Selectivity and Toxicity Concerns:

  • Off-target reactivity: Overly reactive electrophiles risk binding indiscriminately to numerous off-target nucleophilic proteins and biomolecules (e.g., glutathione, serum albumin, other enzymes). This promiscuous reactivity can result in unwanted side effects, systemic toxicity, and reduced therapeutic index. The electrophilic warhead must be tuned to possess sufficient reactivity for the target while minimizing reactivity with off-targets.

  • Irreversibility: The prolonged or irreversible nature of covalent binding means that any off-target effects will also be long-lasting and difficult to reverse, potentially exacerbating adverse events.

  • Haptenization: Covalent attachment of a small molecule drug to endogenous proteins can modify these proteins, making them appear 'foreign' to the immune system. This can elicit an unwanted immune reaction, potentially leading to drug-induced hypersensitivity reactions or other immunogenic responses. This is a significant safety concern for many covalent therapeutics.

• Metabolic Activation:

  • Some drugs may not be inherently covalent but are metabolized in vivo into reactive electrophilic species. These reactive metabolites can then form covalent adducts with proteins, DNA, and other biomolecules, leading to idiosyncratic toxicity.

  • Examples include acetaminophen, which at high doses is metabolized to a reactive quinone imine that can cause hepatotoxicity through covalent binding to liver proteins.

  • Understanding the metabolic profile of a drug candidate is crucial to predict and mitigate such risks.

• Pharmacokinetic Challenges:

  • Covalent warheads can sometimes react with metabolic enzymes (e.g., CYPs) or transporters, complicating drug clearance and potentially leading to drug-drug interactions.

  • The stability of the electrophilic warhead in vivo (e.g., against hydrolysis or reaction with glutathione) can affect bioavailability and in situ target engagement.

Covalent Kinase Inhibitors

Kinases are a highly important class of enzymes involved in signal transduction, making them prime targets for drug development, especially in oncology. Covalent kinase inhibitors represent a significant advancement in this field.

• Why Kinases are Good Targets: Kinases often have an ATP-binding site that can accommodate small molecules. Many kinases also possess a conserved cysteine residue in or near the active site, making them amenable to targeted covalent inhibition, particularly with electrophiles like acrylamides.

• Mechanism: These inhibitors typically bind noncovalently to the ATP binding pocket and then form a covalent bond with a specific cysteine residue, often Cys797 in EGFR or Cys481 in BTK (Bruton's Tyrosine Kinase). This covalent modification leads to irreversible inhibition of kinase activity.

• Advantages in Kinase Inhibition:

  • Enhanced Selectivity: By targeting specific cysteine residues that may be unique to a particular kinase, or to a mutant form of a kinase (e.g., drug-resistant mutants), covalent inhibitors can achieve superior selectivity over pan-kinase inhibitors, reducing off-target effects.

  • Overcoming Resistance Mutations: Some resistance mutations in kinases involve altering the gatekeeper residue in a way that allows the enzyme to escape non-covalent inhibitors but presents a unique opportunity for covalent binders.

  • Improved Duration of Action: Irreversible binding can lead to sustained target inhibition, even after drug clearance, which is beneficial for diseases like cancer where continuous suppression of signaling pathways is desired.

• Example: Osimertinib (Tagrisso):

  • Osimertinib is a third-generation EGFR (Epidermal Growth Factor Receptor) tyrosine kinase inhibitor used for treating non-small cell lung cancer (NSCLC) with specific EGFR mutations (e.g., L858R, Exon 19 deletion, and T790M resistance mutation).

  • It acts as a targeted covalent inhibitor, forming an irreversible covalent bond with Cysteine 797 (Cys797) in the EGFR active site. This residue is present in both wild-type and the T790M mutant EGFR. However, Osimertinib achieves selectivity for the mutant forms due to its structural design, which leverages the altered conformation of the ATP-binding pocket caused by the resistance mutations, allowing for highly specific non-covalent interactions that precede the covalent bond formation.

  • Its design allows it to selectively inhibit the mutant forms of EGFR, including those with the T790M resistance mutation, while sparing wild-type EGFR to a greater extent, leading to improved efficacy and reduced side effects compared to earlier generation inhibitors.

Activity-Based Protein Profiling (ABPP)

ABPP is a powerful chemical biology technique used to monitor the functional state of enzymes and other protein classes within complex biological systems (e.g., cell lysates, intact cells, or whole organisms). It is particularly useful for studying covalent inhibitor interactions.

• Mechanism:

  • ABPP involves using small molecule electrophilic probes (called activity-based probes or ABPs). These probes typically consist of three components:

    1. An electrophilic warhead: designed to react with specific nucleophilic residues (e.g., serine, cysteine, lysine) on target enzymes, forming a covalent bond.

    2. A recognition element: often mimics a natural substrate or inhibitor to guide the probe to the active site or specific binding pocket of target proteins.

    3. A reporter tag: typically a fluorophore for direct visualization or a 'tag' (like biotin or an alkyne for click chemistry) that allows for subsequent purification, detection, and identification of the covalently labeled proteins.

  • When an ABP reacts with an active enzyme, it forms a covalent adduct, thus 'tagging' the active enzyme. The intensity of the tag is proportional to the enzyme's activity/occupancy.

• Applications:

  • Target identification and validation: Identifying unknown protein targets of drugs.

  • Mechanism of action studies: Elucidating how drugs interact with their targets, especially for covalent modifiers.

  • Biomarker discovery: Identifying active enzymes involved in disease pathophysiology.

  • Drug occupancy and selectivity: Quantifying the extent to which a covalent drug occupies its intended target relative to off-targets in treated cells or tissues. By pre-treating cells with a covalent inhibitor and then applying an ABP, reduced ABP labeling indicates target engagement by the inhibitor. This allows for quantification and comparison of drug interactions between treated and untreated samples.

  • Profiling reactivity of nucleophiles: Identifying and characterizing the reactivity of specific nucleophilic residues across the proteome.

Metallo-enzymes

Metallo-enzymes are a class of enzymes that contain a metal ion (e.g., zinc, iron, copper, magnesium) in their active site, which is directly involved in their catalytic mechanism. These metal ions can act as Lewis acids, stabilize reactive intermediates, or participate in redox reactions. Examples include matrix metalloproteinases (MMPs), alcohol dehydrogenase, carbonic anhydrase, and angiotensin-converting enzyme (ACE).

• Challenges for Covalent Inhibitors: While some covalent inhibitors may target residues adjacent to the metal center, directly forming a stable covalent bond with the metal ion for irreversible inhibition is rare for traditional organic electrophiles. Instead, many metallo-enzyme inhibitors (MEIs) function by chelating the metal ion, disrupting its catalytic role through strong non-covalent or dative interactions.

• Indirect Covalent Strategies: However, some strategies for inhibiting metallo-enzymes can involve covalent modification of residues around the metal center. For example, some inhibitors might react with a cysteine or histidine residue that is part of the metal-binding site or critical for the enzyme's mechanism, leading to irreversible inactivation. Additionally, mechanism-based inhibitors could be designed to be processed by the metallo-enzyme, generating an electrophilic species that then reacts covalently with a nearby amino acid.

• Example (Indirect): Certains inhibitors of zinc-dependent enzymes like histone deacetylases (HDACs) may not be covalent themselves, but some experimental approaches have explored suicide substrates that, upon enzymatic processing involving the metal, generate reactive species.

HIV Integrase

HIV integrase is a critical enzyme required for the replication of the Human Immunodeficiency Virus (HIV). After reverse transcription, integrase incorporates the viral DNA (reverse transcribed from RNA) into the host cell's genome, a process essential for the virus to establish a permanent infection and replicate.

• Mechanism: HIV integrase catalyzes two key enzymatic steps:

  1. 3'-processing: Cleavage of dinucleotides from each 3' end of the viral DNA.

  2. Strand transfer: Covalently joining the processed viral DNA ends into the host cell's chromosomal DNA.

• Inhibitor Strategy: Integrase inhibitors are a major class of antiretroviral drugs. The predominant and clinically successful strategy involves Integrase Strand Transfer Inhibitors (INSTIs). These are non-covalent inhibitors that bind to the integrase-viral DNA complex and prevent the strand transfer step by forming stable non-covalent interactions with the active site.

• Covalent Inhibitors for HIV Integrase: While INSTIs are highly effective non-covalent drugs (e.g., Raltegravir, Dolutegravir, Bictegravir), covalent inhibitors for HIV integrase have not reached clinical success and are not a primary strategy in current HIV treatment. Research has explored various approaches, but significant challenges exist in designing covalent inhibitors that achieve sufficient specificity and avoid toxicity, given the widespread nucleophiles in a cell and the complex activity of integrase. The precise and potent non-covalent binding of INSTIs has proven to be a highly effective and safer approach.

Case Study: Discovery of Paxlovid (Nirmatrelvir) for SARS-CoV-2 Proteases

The COVID-19 pandemic necessitated rapid drug development. The main protease ($\text{M}^{pro}$ or 3CL protease) of SARS-CoV-2 was identified as a critical drug target due to its essential role in processing viral polyproteins into functional units necessary for viral replication.

• Rapid Development Timeline: Initial development relied heavily on knowledge from previous research on proteases of other coronaviruses (SARS-CoV and MERS-CoV), which shared structural and mechanistic similarities.

• Design Strategy: Covalent Inhibition of Cysteine Protease:

  • SARS-CoV-2 $\text{M}^{pro}$ is a cysteine protease, meaning it utilizes a catalytic cysteine residue (Cys145) for peptide bond cleavage. This makes it a prime target for covalent inhibitors containing electrophilic warheads that can react with the thiol group of cysteine.

  • The covalent inhibitor concept centered on mimicking the tetrahedral intermediate associated with peptide bond cleavage. During catalysis, the catalytic cysteine attacks the carbonyl carbon of the scissile peptide bond, forming a tetrahedral intermediate.

  • Nirmatrelvir (part of Paxlovid) was designed as a potent and specific targeted covalent inhibitor. It features a nitrile as its electrophilic warhead.

  • Mechanism of Nirmatrelvir: The nitrile group in Nirmatrelvir covalently reacts with the catalytic Cys145 of the $\text{M}^{pro}$ via a reversible Michael addition. Specifically, the Cys145 attacks the electrophilic carbon of the nitrile, forming a thioketal intermediate. This reversible covalent bond provides potent inhibition while potentially allowing for a more manageable safety profile compared to irreversibly binding warheads.

  • Utilized structure-based drug design principles to optimize compound structure, ensuring high affinity for the protease's active site and optimal positioning of the nitrile for reaction with Cys145.

• Paxlovid Formulation:

  • Paxlovid is a co-packaged combination of two antiviral medications: Nirmatrelvir (the SARS-CoV-2 $\text{M}^{pro}$ inhibitor) and Ritonavir.

  • Ritonavir is a potent cytochrome P450 3A (CYP3A) inhibitor. It is included not for its antiviral activity against SARS-CoV-2, but as a pharmacokinetic booster. It inhibits the metabolic breakdown of Nirmatrelvir, leading to higher and more sustained plasma concentrations of Nirmatrelvir, thereby enhancing its antiviral efficacy and allowing for less frequent dosing.

Key Lessons from the Case Study

• Iterative Optimization:

  • Drug development is an iterative process. Initial lead compounds for Nirmatrelvir underwent extensive modifications to enhance key drug properties, including improved cell permeation, reduced susceptibility to efflux pumps, and maintenance or improvement of affinity for the target enzyme.

  • Structural features, such as the P1' group (responsible for hydrophobic interactions) and various substitutions, were adjusted to optimize interactions, minimize solvent exposure where beneficial, and enhance pharmacokinetic properties without compromising specificity.

• Pharmacokinetics and Delivery Considerations:

  • The final drug candidate often requires combination with pharmacokinetic boosters (like Ritonavir in Paxlovid's case) to achieve sufficient oral bioavailability, sustained plasma levels, and efficacy in vivo.

  • Considerations of drug metabolism and potential drug-drug interactions are crucial, especially for compounds used in a patient population often on multiple medications.

Conclusion

Covalent inhibitors represent a potent and increasingly sophisticated avenue in drug design. While necessitating careful consideration of specificity, off-target effects, and potential immunogenicity, a growing understanding of their mechanisms, combined with advanced design strategies like targeted covalent inhibition and tools like ABPP, enhances the prospects for developing highly effective and selective therapies. This approach has proven particularly valuable in addressing challenging targets, overcoming drug resistance mutations, and rapidly responding to emerging infectious diseases, as exemplified by the success of drugs like Osimertinib and Paxlovid.