1.1. History of Covalent Drugs
Covalent drugs contain a reactive functional group, or “warhead,” that can form a strong chemical bond with the biological target (
Figure 1) [
1]. This definition includes prodrugs that are metabolised inside the body to produce reactive species in their active form. The warheads of covalent drugs are usually electrophilic in nature, ranging from mildly reactive (e.g. acrylamides, aziridines, esters, nitriles) to highly reactive (e.g. chloroethylamines, nitrogen mustards, epoxides). This electrophilic reactivity is complementary to the nucleophilic functional groups commonly found within biological macromolecules, such as the cysteine residues of proteins or the nitrogen atoms of DNA bases.
The simple act of forming a covalent bond between a drug and its target has a significant effect on the drug’s pharmacodynamic properties. Permanent blockage of the binding site usually forces the target to undergo resynthesis before its activity can be re-established, leading to a longer therapeutic effect and improved potency of the drug [
2,
3]. Covalent drugs can be advantageous for treating diseases in which high target occupancy is important, such as cancer and bacterial infections [
2,
3]. It may be possible to administer covalent drugs at lower, less frequent doses, which can reduce toxicity and improve patient comfort and compliance. Finally, covalent drugs can successfully address what would otherwise be considered “undruggable targets”, i.e. intractable proteins that have shallow binding pockets where reversible drugs cannot bind [
4].
Covalent drugs have a long history in the pharmaceutical industry, stretching back to the discovery of aspirin in 1899 for the treatment of pain and inflammation (
Figure 2). Aspirin remains the most widely-used medication today [
4], and covalent drugs now account for approximately 7% of all small-molecule drugs approved by the FDA [
5]. Numerous review articles have highlighted the sustained interest in designing novel covalent drugs over recent decades [
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12].
Many historical covalent drugs were discovered without any knowledge of their mechanism of action. In the case of aspirin (
Figure 2), it was found only much later that the therapeutic effect is attributable to inhibition of the enzyme, cyclooxygenase [
13]. The ester moiety of aspirin acts as an acyl transfer reagent, which irreversibly acetylates Ser530 of the enzyme. Another type of acylating drug is the β-lactam class of antibiotics, e.g. ampicillin (
Figure 2). The ring strain of the lactam (a cyclic amide), compounded by the presence of a fused ring, forces the nitrogen into a trigonal pyramidal geometry. This makes the adjacent carbonyl more electrophilic and prone to ring-opening by nucleophiles [
14]. β-Lactam antibiotics inhibit important enzymes responsible for building cell walls in both Gram-positive and Gram-negative bacteria [
15]. The lactone (cyclic ester) variant is present in the drug orlistat (
Figure 2). Orlistat is used to treat obesity by inhibiting fatty acid synthase, but it has been recently investigated for the treatment of cancer, as fatty acid synthase is often overexpressed in cancer.
Some drugs, such as 5-fluorouracil and decitabine (
Figure 2), can harness enzymes to form covalent bonds with DNA. Such drugs are known as antimetabolites; they are structural analogues of purines and pyrimidines and can thus act as atypical DNA building blocks. The generation of aberrant / damaged DNA makes these drugs useful in chemotherapy to kill rapidly-dividing tumour cells [
16].
Irreversible DNA binding is further exploited with the reactive nitrogen mustards (
Figure 2). Nitrogen mustards contain the bis(2-chloroethyl)amino functional group, which spontaneously expels chloride to form an aziridinium intermediate that can alkylate the nucleophilic sites on DNA bases [
17]. Repetition of this process with the second chloroethyl group of the nitrogen mustard allows a second covalent bond to be formed with DNA, leading to crosslinks which prevent DNA replication and ultimately result in apoptosis of the cell [
18]. A prominent nitrogen mustard, cyclophosphamide, was developed in the 1950s. Bendamustine was discovered soon after in East Germany, but was not approved by the FDA until half a century later in 2008 [
19]. Carmustine, approved in 1977, is a related structure. Despite the known toxicity of these compounds, they are still considered acceptable in chemotherapy due to the gravity of cancer as a disease.
Functionally similar to the nitrogen mustards are the aziridines, e.g. mitomycin C (
Figure 2). Aziridines become activated by protonation, and the resulting aziridinium resembles the activated intermediate derived from nitrogen mustards. However, aziridines are subtly less reactive than mustards, because the charge of the protonated aziridinium is somewhat dissipated by solvation. Therefore, aziridines are more stable and less likely to be inactivated by off-target nucleophiles like water and glutathione.
Reversible covalent bonding groups, which strike a balance between the benefits of non-covalent and covalent drugs, have also been used [
20]. The boron-containing bortezomib (
Figure 2) is a proteasome inhibitor designed to treat multiple myeloma. The boron reacts with a threonine hydroxyl group on the 20S proteasome to form a boronate [
6].
Michael acceptors (
Figure 2) are another important category of electrophilic warheads, typically targeting cysteine residues within protein binding sites [
21,
22]. Exemplifying this category are the drugs ibrutinib and afatinib, which are tyrosine kinase inhibitors, and sotorasib, which is a GTPase inhibitor. During the development of these drugs there was a strong emphasis on optimising the noncovalent binding interactions, in order to maximise selectivity for the desired target over off-targets. Drugs that emerge from such an approach are sometimes referred to as targeted covalent inhibitors (TCIs).
The final category of electrophilic warhead depicted in
Figure 2 is the nitrile, as seen in the drugs saxagliptin and nirmatrelvir. Saxagliptin is a dipeptidyl peptidase-4 (DPP-4) inhibitor and anti-diabetic, and has potential to treat Alzheimer’s disease [
23]. Nirmatrelvir, which is an antiviral drug that targets the main protease of SARS-CoV-2, was discovered by an electrophile-first approach: instead of building from a known reversible inhibitor, an electrophile was chosen and the rest of the structure was expanded from it [
6].
1.2. Disadvantages of Covalent Drugs
The primary disadvantage of covalent drugs is their potential to form irreversible bonds with off-target proteins, which can lead to unpredictable downstream effects [
24]. In some cases, unexpected drug-protein adducts can induce idiosyncratic immune responses that are harmful to patients [
25,
26]. The negative consequence of off-target binding is compounded by the fact that less drug will reach the desired target. As was discussed above with TCIs, it is possible to impart some selectivity for the desired target by optimising the non-covalent interactions, but the issue of off-target binding remains a concern.
Another disadvantage of covalent drugs is their susceptibility to metabolism. Due to their reactive nature, covalent drugs can be easily degraded and inactivated. For example, increased expression of glutathione is a significant factor in cancer drug resistance: partly due to this, the nitrogen mustards bendamustine and carmustine both have short half-lives of around 30 minutes [
17]. Meanwhile, afatinib suffers from significant extrahepatic metabolism by reactivity with glutathione [
27].
For a time, these disadvantages caused the development of covalent drugs to be seen as a risky endeavour. During the advent of high-throughput screening of drug candidates in the 1980s, compounds that covalently bind to proteins were generally excluded from compound libraries due to fears that they could bind to random proteins and cause toxicity [
28,
29]. The overall hesitancy of the pharmaceutical industry to invest in covalent drug research means that covalent drugs may be yet to reach their full potential [
2,
4,
30].
1.3. Nanoparticles as a Possible Solution
Drug delivery systems are a useful way of mitigating some of the problems of drugs by protecting them until they are released at their destination in a controlled, sustained manner. Research has progressed from conventional delivery systems such as tablets and capsules, to controlled-release hydrogels and matrices, and recently to more advanced technologies like nanomedicine [
31]. Nanoparticles are useful in that they are able to carry a payload of drugs, while being small enough to cross biological barriers be distributed locally and avoid embolisms [
32].
Nanoparticles can be constructed from a range of materials, such as lipids, polymers, carbohydrates, proteins and inorganic substances [
33]. They can form various structures like liposomes, micelles, dendrimers and worm-like particles, and can easily be modified to be imaged
in vitro and
in vivo. Polymers have been widely used in the development of drug delivery systems, owing to their ability to self-assemble into many sizes and shapes (
Figure 3). Many polymers are biocompatible, meaning they are non-toxic, are metabolised or hydrolysed into non-toxic compounds, and can be efficiently expelled from the body once they release their payload. Many of these materials can act as treatments themselves, potentially bypassing multi-drug resistance [
34]. Commonly-used polymers include polyethylene glycol (PEG), polylactic acid (PLA), polydopamine, poly(lactic-co-glycolic acid) (PLGA), polyvinyl alcohol (PVA), polycaprolactone (PCL) and chitosan.
It is important that the drug and material used to formulate the nanoparticles are compatible. Strong van der Waals and hydrogen bonding interactions between the two can increase the drug loading capacity and delay the rate of release. With respect to covalent drugs, the warheads need to be compatible with any potentially reactive moieties within the nanoparticle. Finally, drugs can be conjugated to nanoparticles and so appropriate linker groups need to be considered to connect the two entities together. For example, the carboxylic acid side chains of bendamustine allow for easy conjugation to polymers [
35].
Nanoparticles provide several key benefits to drug delivery. First, they can improve the solubility of hydrophobic drugs (Figure 4, “solubility”). Second, they can enhance drugs’ ability to cross biological membranes such as the intestine and the blood-brain barrier, BBB (Figure 4, “permeability”). This can be achieved in conditions across a range of pH values. Third, the half-life of drugs can be extended by preventing metabolism and inactivation of the covalent warheads, allowing more circulation time within the body (Figure 4, “lifetime”). Fourth, the rate at which the drug reaches its target can be fine-tuned by the composition of the nanoparticle, which can further prolong the therapeutic effect while reducing side effects (Figure 4, “controlled release”). These four benefits can be considered together under the umbrella idea of bioavailability.
Figure 2.
Benefits of nanoparticles for drug delivery.
Figure 2.
Benefits of nanoparticles for drug delivery.
Further advantages are offered besides bioavailability. A fifth benefit is that nanoparticles can prevent the non-specific binding by the covalent warheads and allow tissue selectivity through active targeting, the latter of which is highly important in cancer and infectious diseases (Figure 4, “selectivity”). Finally, a sixth benefit offered by nanoparticles is the opportunity for co-delivery of drugs (Figure 4, “co-delivery”). Drug treatment can often be more effective when two or more drugs are administered simultaneously. This is seen especially in the case of cancer with combination therapy. This multi-targeted approach can decrease the likelihood of drug resistance developing over the course of treatment [
36]. For this to be successful, the correct ratio of drugs must arrive at the target site within the same timeframe, otherwise toxicity issues may result. Nanoparticles have been developed to deliver multiple drugs at an optimised ratio.
Each of these benefits will be examined in detail in
Section 2 of this review.