Phosphorus fluoride exchange (PFEx): a new frontier in bioconjugation, enzyme inhibition, proximity-induced target-binding and fluorescence probes

1 Introduction

In chemical biology, the demand for biorthogonal reactions that enable covalent modification, precise labelling, and probing of biomolecules under physiological conditions remains high, but the current toolkit of such reactions remains comparatively limited (Bird et al., 2021). Similarly, the number of different chemical reactions that are broadly applied to synthesize compound libraries for biological testing is still limited with most such efforts involving highly reliable reactions such as amide bond formation, cross-coupling reactions, transition-metal-catalyzed reactions, reductive amination, nucleophilic substitution, and sulphonamide formation reactions (Haldón et al., 2015; Brown and Boström, 2016; Wang et al., 2021; Dombrowski et al., 2022; Gesmundo et al., 2024). Over the past 20 years, only a limited number of novel reactions have become routinely implemented for the synthesis of small molecule compound libraries, including photoredox cross-couplings (Gesmundo et al., 2024), click chemistry mediated-copper-catalysed azide-alkyne cycloadditions (CuAAC) and sulphur-fluoride exchange reactions (SuFEx) (Rostovtsev et al., 2002; Tornøe et al., 2002; Haldón et al., 2015; Zhu et al., 2016). Most of these reactions lead to molecules that are rich in Csp² bonds with an overall flat structure that lacks three-dimensional (skeletal) diversity. Such molecules provide compound collections with a lower chance of targeting protein-binding pockets compared to more structurally diverse natural product-like molecules (Schreiber, 2000; Kumar and Waldmann, 2009; Lovering et al., 2009; Lovering, 2013). Over the past two decades, the emergence of click chemistry has provided a conceptual and practical framework for developing reactions that are modular, and high-yielding. The concept of click chemistry has found broad applications in chemical biology, drug development and materials chemistry since its introduction by Sharpless in 1998 (Kolb et al., 2001; Rostovtsev et al., 2002; Tornøe et al., 2002; Killops et al., 2008; Dong et al., 2014; Finn et al., 2022; Sun S. et al., 2023; Hou et al., 2024; Borne et al., 2021; Cruite et al., 2023). SuFEx chemistry was subsequently identified by Sharpless and co-workers as another “click” chemistry reaction (Dong et al., 2014). SuFEx chemistry covers a range of reactions where compounds containing S-F bonds with varying reactivity can be functionalized by a broad range of nucleophiles, potentially providing access to compound libraries with increased chemical diversity of interest in medicinal chemistry (Homer et al., 2023). In addition to SuFEx-based chemistry, the closely related phosphorus fluoride exchange (PFEx) chemistry has recently emerged as a compelling addition to click chemistry (Figure 1) (Sun S. et al., 2023; Homer et al., 2024; Wang et al., 2025; Yang et al., 2025).

Distinguished by the unique reactivity of the P-F bond, PFEx occupies a conceptual space between stability and reactivity. It is inert under ambient conditions but selectively activates in well-defined chemical or biological/cellular microenvironments. This duality is particularly significant in biological systems (Yang et al., 2025), where uncontrolled reactivity can lead to off-target effects, whereas insufficient reactivity limits functional engagement. PFEx elegantly addresses this challenge through latent electrophilicity, enabling proximity-driven and environment-dependent covalent transformations (Wang and Wang, 2022; Sun S. et al., 2023). Beyond its synthetic utility, PFEx exhibits a distinct biomimetic character because the resulting P–O and P–N bonds are ubiquitous in biological systems, including nucleic acids (DNA and RNA), phospholipids, ATP, and phosphoramidate-containing biomolecules, where they play central roles in enzymatic catalysis, signal transduction, and metabolic regulation (Wang and Wang, 2022). In contrast, SuFEx chemistry generally generates sulfur (VI)-based linkages such as sulfonates, sulfates, sulfamides, and sulfonamides, which, although well-established and highly valuable in chemical biology, are less prevalent in native biological systems than phosphorus-based linkages (Cao et al., 2024b). However, comprehensive comparative biological studies evaluating PFEx versus SuFEx in terms of residue selectivity, biocompatibility, and proteome-wide performance remain limited and need further investigation. SuFEx based unnatural amino acids (UAAs) are limited in their ability to form stable linkages with Lys, His and Tyr in proteins (Mukherjee et al., 2017; Yang et al., 2019). In contrast, PFEx show capacity to engage with broader spectrum of nucleophilic amino acid residues, including Lys, His, Tyr, and Cys, thereby expanding its scope in protein engineering and covalent targeting (Cao et al., 2024b). A particularly transformative development has been the integration of PFEx into genetic code expansion technologies (Huang et al., 2025), which enables the site-specific incorporation of phosphoramidofluoridate-based UAAs into proteins, thereby facilitating proximity-enabled PFEx reactions to occur within living systems (Cao et al., 2024b). In parallel, the longstanding use of fluorophosphonate-based probes in activity-based protein profiling (ABPP) (Liu et al., 1999; 2022) and enzyme inhibition highlights how PFEx-like processes can be exploited by biological systems, where enzyme active sites serve as catalytic environments to facilitate selective P–F bond activation. Despite these promising advances, PFEx chemistry remains at an early stage of exploration in chemical biology and exhibits several key challenges. Nevertheless, recent innovations, such as proximity-induced activation (Cao et al., 2024b), and late-stage functionalization strategies (d’Andrea and Townsend, 2019) are beginning to address these limitations, opening new opportunities for the rational design of PFEx-enabled probes and therapeutics.

In this mini-review, we provide an important recognition of PFEx chemistry as an emerging tool in chemical biology. The article highlights its fundamental reactivity, recent methodological advances, integration into biomolecular systems, applications in protein modification, covalent inhibition, and activity-based profiling. Furthermore, current challenges and future directions for leveraging PFEx chemistry in the development of next-generation chemical probes and precision therapeutics have been outlined.

2 Emerging applications of PFEx and a mechanistically similar fluorophosphonate chemistry in chemical biology

PFEx chemistry, a transformation mechanistically linked to fluorophosphonate chemistry has evolved into a versatile platform in chemical biology, enabling precise modification of biomolecules through nucleophilic attack and fluoride displacement. This section highlights its expanding applications, from proximity-driven reactivity to target inhibition and probe design, highlighting the potential of PFEx and fluorophosphonate chemistry to decode and intervene in complex biological pathways.

2.1 Proximity-enabled PFEx reactions

Proximity-enabled PFEx exploits precisely engineered molecular scaffolds and inherent biomolecular recognition elements to spatially localize fluorophosphate functionalities in close vicinity to target nucleophilic amino acid residues, thereby markedly enhancing the kinetics of otherwise intrinsically inactive substitution processes under physiological conditions (Cao et al., 2024b).

2.1.1 Proximity-enabled PFEx chemistry in proteins: role of genetically encodable phosphoramidofluoridate unnatural amino acids

The ability to introduce new covalent bonding functionalities into proteins has emerged as a powerful strategy for advancing protein science, engineering, and development of therapeutics (Cao and Wang, 2022). This is primarily enabled through genetic code expansion (Huang et al., 2025), which allows the site-specific incorporation of latent bio-reactive unnatural amino acids (UAAs) into proteins (Wang et al., 2001; Xiang et al., 2013). These engineered residues undergo proximity-enabled reactions with neighbouring natural amino acid side chains, forming covalent linkages with biomolecules in either intra- or intermolecular fashion (Li et al., 2023; Sun W. et al., 2023). Such type of transformations in case of SuFEx have facilitated the discovery of transient biomolecular interactions, endowed proteins with novel functionalities (Yang et al., 2017; Tang et al., 2018; Liu C. et al., 2021; Liu et al., 2021 J.; Shu et al., 2023; Sun W. et al., 2023), and accelerated the development of covalent protein-based therapeutics (Li et al., 2020; Cao et al., 2021; Yu et al., 2022; Klauser et al., 2023). Despite significant progress, the implementation of latent bio-reactive UAAs (Chen et al., 2014; Furman et al., 2014; Xiang et al., 2014; Hoppmann and Wang, 2016; Xuan et al., 2016; 2017; Wang et al., 2018; Liu et al., 2019; Xu et al., 2022; Yu et al., 2023) in complex biological systems remains challenging, particularly with respect to cytotoxicity and biocompatibility (Wang and Wang, 2022).

In this context, phosphoramidofluoridate-based UAAs, namely, phosphoramidofluoridate tyrosine (PFY) (Figure 2A) and phosphoramidofluoridate lysine (PFK) (Figure 2B) have been developed by Wang et al. (Cao et al., 2024b) as genetically encodable handles for PFEx chemistry in proteins. PFY (Figure 2A) was synthesized via a nucleophilic addition-elimination reaction between dimethylphosphoramidic difluoride and benzyl-protected tyrosine followed by removal of the protecting group. PFK (Figure 2B) was synthesized analogously via nucleophilic addition-elimination reaction between dimethylphosphoramidic difluoride and benzyl paraben. Subsequent deprotection, amide bond formation with benzyl-protected lysine, and final deprotection afforded the desired PFK. These residues bear a phosphoramidofluoridate moiety that remains inert under physiological conditions but becomes reactive upon spatial proximity to nucleophilic amino acid side chains (Figure 3A). PFY and PFK were successfully incorporated into proteins in E. coli and mammalian cells, demonstrating the robustness of this approach across biological platforms. Notably, PFEx reactivity within proteins is intrinsically activated by proximity effects, eliminating the need for external reagents. This feature enables efficient covalent bond formation in aqueous, protic environments and even within the complex cellular environment (Cao et al., 2024b). Mechanistically, PFEx facilitates the formation of stable P–O, P–N, and P–S linkages with nucleophilic residues such as Tyr, Lys, His, and Cys (Cao et al., 2024b; Yang et al., 2025). This broad residue scope distinguishes PFEx from many conventional bioconjugation strategies, such as maleimide-thiol coupling (Fontaine et al., 2015; Gober et al., 2023), N-hydroxysuccinimide ester chemistry (Hansen et al., 2022), SuFEx (Wilson Lucas et al., 2023), etc. Remarkably, thiol groups of Cys previously underexplored in small-molecule PFEx chemistry (Sun S. et al., 2023) were found to participate effectively in protein-based PFEx reactions, particularly in the presence of sodium silicate (Na₂SiO₃), which enhanced cross-linking efficiency with Cys and Tyr while attenuating reactivity with His. These findings suggest new opportunities for extending PFEx chemistry to thiol-containing small molecules (Cao et al., 2024b). Compared to SuFEx, PFEx offers enhanced versatility and biomimetic relevance. While SuFEx predominantly generates S–N and S–O linkages, PFEx enables the formation of P–N and P–O bonds, which are more commonly found in native biomacromolecules such as nucleic acids (DNA and RNA), phospholipids, ATP, and phosphoramidate-containing biomolecules. Furthermore, PFEx expands the reactive amino acid collection to include Cys and introduces unique features, such as thermally reversible P–N bond formation. For instance, efficient cross-linking between PFY and His was observed at lower temperatures, while reversibility occurred at elevated temperatures. The thermally reversible nature of PFY-His conjugation may be associated with the relatively labile phosphoramidate linkage formed between the phosphorus centre and the imidazole nitrogen of histidine, as phosphoramidate bonds are generally more labile than phosphate esters. Elevated temperatures may facilitate hydrolysis or bond dissociation through enhanced proton transfer and increased activation of the reverse reaction pathway, whereas lower temperatures favor formation and stability of the cross-linked adduct. Additionally, PFEx displays an unusual pH profile, exhibiting enhanced reactivity with His under mildly acidic conditions contrasting with the basic pH preference typical of SuFEx and other nucleophilic chemistries. This distinctive behaviour offers promising opportunities for selective targeting of acidic microenvironments, such as those found in tumor tissues (Cao et al., 2024b). The improved PFEx reactivity observed with His under mildly acidic conditions may arise from the unique acid-base properties of the imidazole side chain of His (Raum et al., 2024). Unlike conventional nucleophilic reactions that generally benefit from strongly basic conditions, PFEx may involve significant contribution from proton-transfer processes and stabilization of the departing fluoride ion (Mentz and Modro, 1995). At mildly acidic pH, His exists as a mixture of neutral and protonated species, enabling simultaneous nucleophilic attack and proton-transferring behavior that facilitates P–F bond cleavage (Kasireddy et al., 2017). Furthermore, hydrogen-bond stabilization of the leaving fluoride ion under these conditions may lower the activation barrier for addition-elimination reaction, although direct mechanistic evidence in PFEx systems remains limited.

The proximity-dependent nature of PFEx has been validated across multiple protein systems. In the Z_spa affibody (Afb-Z) protein complex (Figure 3B), PFY incorporated at position 36 of Afb protein enabled selective cross-linking with amino acid residues at position 6 of the Z-protein (Högbom et al., 2003). Among twelve amino acids evaluated, His and Tyr exhibited strong cross-linking, whereas Lys and Cys showed comparatively weaker reactivity. Similarly, in the nanobody SR4 system targeting the SARS-CoV-2 spike protein (Figure 3C), PFY incorporation at position 57 of SR4 resulted in efficient cross-linking with a proximal Tyr505 residue, whereas incorporation at a distal site (position 54) failed to produce cross-linking, thereby confirming strict proximity dependence (Cao et al., 2024b).

Importantly, PFEx reactivity was also demonstrated in living mammalian cells. PFY was incorporated at position 103 of E. coli glutathione-S-transferase (ecGST), a homodimeric protein, to target a neighbouring His106 residue across the dimer interface. Expression of this mutant in HEK-293T cells resulted in efficient covalent dimer formation (Figure 3D). Substitution studies further revealed strong cross-linking with His and Tyr, and weaker interactions with Lys and Cys, consistent with in-vitro observations (Cao et al., 2024b). Additionally, studies with PFK further reinforced the versatility of PFEx-enabled protein cross-linking. In the SARS-CoV-2 targeting nanobody mNb6, incorporation of PFK at position 54 enabled robust cross-linking with Tyr351 of the spike receptor-binding domain, whereas other incorporation sites were ineffective. Additionally, PFK incorporated into ecGST at position 103 facilitated covalent dimerization in HEK-293T cells by targeting proximal residues, including Cys, His, and Tyr (Cao et al., 2024b).

Collectively, these studies establish PFY and PFK as powerful genetically encoded chemical tools that enable proximity-driven PFEx chemistry in proteins across both in-vitro environment and mammalian cells. The ability to selectively target multiple nucleophilic residues, operate under physiological conditions, and form stable linkages highlights the significant potential of PFEx as a next-generation bioconjugation strategy with broad implications in chemical biology, protein engineering, and therapeutic development (Cao et al., 2024b).

2.1.2 PFY-mediated PFEx cross-linking efficiency: promoting role of arginine

Although PFEx reactions between proteins hold significant potential, their current efficiency remains limited (Yu et al., 2023). As noted earlier, these reactions depend on positioning a latent bio-reactive UAA in close proximity to its target residue during protein binding. Their efficiency is affected by several factors, including the binding affinity of the carrier protein, the length and flexibility of the UAA side chain (Liu J. et al., 2021; Cao et al., 2024b), the orientation and inherent reactivity of the UAA warhead (Hoppmann and Wang, 2016; Klauser et al., 2022; Sun W. et al., 2023), and pH conditions (Yu et al., 2023), which influence the nucleophilicity of the target residues.

Suboptimal reaction kinetics within protein environments significantly constrain the practical utility of PFEx chemistry, as conventional catalytic strategies developed for small molecules are generally incompatible with biological systems and in-vivo applications. Consequently, there is need for the development of a broadly applicable and biocompatible strategy to enhance PFEx reaction rates between proteins. In this context, the strategic introduction of an Arg residue proximal to a latent bio-reactive UAAs by Wang et al. (Cao et al., 2024a) was found to significantly accelerate PFEx reactivity (Figure 4) (Cao et al., 2024a). Specifically, Arg was introduced adjacent to the UAA PFY through site-directed mutagenesis (Figure 4A). The modified PFY-containing construct was incorporated into Afb to generate Afb (36PFY mutated withF32R), wherein PFY was installed at position 36 and the native Phe (F) at position 32 was substituted with Arg (R). In parallel, a non-mutated variant, Afb (36PFY), was prepared as a control. These protein variants were subsequently incubated with a maltose binding protein, MBP-Z (6H), that fuses to N-terminus of Z-protein to facilitate the separation between Afb and Z-proteins, (Figure 4B). SDS-PAGE analysis confirmed successful cross-linking between MBP-Z (6H) and both Afb constructs via the PFY-histidine conjugation pathway. Interestingly, incorporation of Arg at position 32 in Afb (36PFYmutated with F32R) was found to result in a 44% enhancement in PFEx reaction rate (Figure 4C) (Cao et al., 2024a).

The Arg-containing mutant Afb (36PFY/F32R) exhibited pronounced cross-linking, with an observed rate constant (k _obs) of 0.0218 h^-1. In contrast, under PFY-MBP-Z (6Y) tyrosine reaction conditions, cross-linking between MBP-Z (6Y) and the control Afb (36PFY) was not detectable. Collectively, these findings highlight the critical role of proximal arginine in promoting PFEx reactivity, thereby offering a viable strategy to enhance protein-protein cross-linking efficiency in biologically relevant environment (Cao et al., 2024a). Although the precise mechanism of Arg-mediated PFEx acceleration remains to be fully elucidated, several mechanistic factors may contribute to the enhanced reactivity. The guanidinium group of arginine is capable of providing strong electrostatic and H-bonding interactions that can stabilize the highly polarized PFEx transition state. In particular, Arg may facilitate P–F bond cleavage by stabilizing the negative charge on the departing fluoride ion and by increasing polarization of the P–F bond through H-bond donation. Additionally, strategic placement of Arg near the latent bio-reactive amino acid may promote local preorganization of the reacting partners, thereby reducing entropic barriers and favoring covalent bond formation (Homer et al., 2023). This effect is dependent on the proper orientation of the guanidinium group, necessitating its precise positioning at a defined site within the protein to enable optimal catalytic interaction. These features are well known to play a critical role in protein-protein interfaces and molecular recognition and likely contribute to stabilization of charge development and improved reaction geometry during PFEx process (Zarrine-Afsar et al., 2006; Balakrishnan et al., 2012; Xu et al., 2017).

2.2 Enzyme inhibition by reaction of nucleophilic amino acid residue with P(V)–F bond

Fluorophosphonate chemistry in chemical biology involves the substitution of a nucleophilic amino acid residue, such as serine (in serine hydrolase) at an electrophilic phosphorus(V) center of a P(V)–F bond. This results in the displacement of fluoride as the leaving group and covalent bond formation with nucleophilic site. Representative examples of fluorophosphonate-based enzyme inhibitors and their applications in chemical biology are discussed herein. Fluorophosphonate chemistry based reactions offer a modular strategy for the development of potent, irreversible inhibitors of enzymes like serine hydrolases (d’Andrea and Townsend, 2019) and glutamine synthetase (Inoue et al., 1999) through the generation of reactive intermediates that trap catalytic residues. These inhibitors provide superior orthogonal selectivity over traditional covalent warheads.

2.2.1 Phosphorus-based covalent warheads: inhibition of serine hydrolase in chemical biology

Serine hydrolases (SHs) are one of the largest and most diverse enzyme superfamilies present across all domains of life (Long and Cravatt, 2011). They utilize specialized catalytic triads or dyads to break down amide, ester, and thioester bonds in both small molecules and complex proteins (Ekici et al., 2008). Because they play critical roles in a vast array of biological functions, ranging from metabolism (Long and Cravatt, 2011) and blood clotting (Coughlin, 2000) to neurological signalling (Colovic et al., 2013), cancer progression (Shields et al., 2010) and immune response (Ortega et al., 2016), these enzymes remain a primary focus for drug discovery and pharmaceuticals. The development of phosphorus-based covalent inhibitors necessitates a delicate balance between synthetic accessibility and electrophilic reactivity (Maślanka and Mucha, 2019), particularly when targeting nucleophilic residues such as the catalytic serine in SHs. Among the most extensively explored phosphorus warheads, diphenylphosphonates (DPPs) and fluorophosphonates (FPs) (Joachimiak and Błażewska, 2018) represent two contrasting yet complementary examples (Figure 5).

DPPs are synthetically robust and exhibit broad functional group tolerance, allowing their incorporation into structurally complex, substrate-mimetic frameworks. However, their relatively lower electrophilicity often limits efficient covalent interaction with less reactive enzyme active sites. In DPPs, the phosphorous centre is less polarized and the aryloxy substituents function as comparatively poor leaving groups, resulting in slower rates of nucleophilic substitution. Consequently, effective covalent bond formation generally requires highly activated catalytic serine residues and favorable active-site environments. In contrast, FPs contain a strongly polarized P–F bond arising from the high electronegativity of fluorine, which increases the electrophilic character of the phosphorus centre and makes it more susceptible to nucleophilic attack. Upon binding within the active site of a serine hydrolase, the serine hydroxyl group is activated through proton abstraction, generating a highly nucleophilic alkoxide species. This activated serine attacks the electrophilic phosphorous centre to form a pentacoordinate phosphorane transition state or intermediate, followed by displacement of fluoride as the leaving group, resulting in stable covalent enzyme inactivation. Thus, the superior inhibitory potency of FPs arises from the combined effects of enhanced phosphorus electrophilicity, favorable transition-state formation, and efficient fluoride displacement within the catalytic pocket. However, despite this high reactivity, the incorporation of FP moieties into complex molecular frameworks remains challenging. This difficulty primarily arises from the poor compatibility of conventional fluorination methods with sensitive functional groups (d’Andrea and Townsend, 2019).

To demonstrate an elegant resolution to this longstanding challenge, Townsend et al. (d’Andrea and Townsend, 2019) developed a late-stage fluorinative hydrolysis strategy that enables the direct conversion of diphenylphosphonate-containing substrates into their corresponding fluorophosphonates under mild conditions (Figure 5) (d’Andrea and Townsend, 2019). This transformation utilizes ammonium fluoride as a nucleophilic fluoride source and proceeds through sequential substitution at the phosphorus centre. Mechanistic studies support a pathway involving initial formation of a monofluorophosphonate intermediate followed by transient generation of a highly reactive difluorophosphonate species. Although this difluoride intermediate is not directly observed, its intermediary involvement is strongly supported by isotopic labelling experiments and the formation of a characteristic P–N–P dimeric byproduct, providing strong indirect evidence (d’Andrea and Townsend, 2019). Importantly, this late-stage transformation strategy overcomes key limitations associated with the de novo synthesis of fluorophosphonate-based probes by enabling direct modification of pre-existing phosphonates. Its practical utility was demonstrated using a β-lactam-containing peptide substrate analog targeting the Nocardicin biosynthetic thioesterase (NocTE). Under physiological conditions, the catalytic serine residue of NocTE engages the peptide substrate analog, facilitating epimerization (L→D) followed by hydrolytic release of the product (pro-nocardicin G), which is subsequently converted into the active antibiotic Nocardicin A (Gaudelli and Townsend, 2014). While the corresponding diphenylphosphonate analogs failed to form covalent enzyme adducts, the fluorophosphonate derivative exhibited rapid and covalent modification of the catalytic serine residue (Figure 5). The electrophilic phosphorus centre of the fluorophosphonate undergoes nucleophilic addition of Ser and elimination of fluoride with cleavage of P–F bond. This observation highlights a fundamental principle in covalent inhibitor design that binding affinity alone is insufficient without appropriately tuned electrophilic reactivity (d’Andrea and Townsend, 2019).

2.2.2 Phosphonofluoridate-based transition-state mimics enabling enzyme-assisted inhibition of glutamine synthetase

Glutamine synthetase (GS) inhibitors have been extensively investigated for their therapeutic potential across a wide spectrum of pathological conditions, including cancer (Kim et al., 2021; Yang et al., 2021; Liu et al., 2024), neurological disorders (Eid et al., 2012; 2019; Jeitner and Cooper, 2014), bacterial infections (Berlicki et al., 2005), tuberculosis (Mowbray et al., 2014), as well as hepatic dysfunction (Brusilow et al., 2010). GS catalyze the ATP-dependent formation of a γ-L-glutamyl peptide bond to produce L-glutamine (Griffith and Mulcahy, 1999). The reaction catalyzed by GS is thought to proceed through the formation of γ-glutamyl phosphate intermediate followed by nucleophilic substitution by substrate amine/ammonia (Figure 6A) (Inoue et al., 1999). An important and conceptually significant example of biologically relevant chemistry is represented by a phosphonofluoridate derivative of L-glutamic acid (Inoue et al., 1999). This compound was rationally designed and synthesized by Sakata et al. (Inoue et al., 1999). To function as a tetrahedral transition-state mimic of the γ-glutamyl phosphate intermediate (Figure 6B), a key species formed during the ATP-dependent amide bond formation catalyzed by GS. Unlike classical phosphonate analogues, phosphonofluoridate derivative contain a P–F bond, which serves as a latent electrophilic handle. This polarized P–F bond enables the phosphorus centre suitable to nucleophilic substitution. Importantly, this substitution occurs in a highly controlled, enzyme-assisted microenvironment. Mechanistically, phosphonofluoridate derivative inhibit the catalytic machinery of GS. Within the active site, nucleophiles such as ammonia or substrate-derived amines are precisely oriented and activated. Under these conditions, nucleophilic attack at the electrophilic phosphorus centre of phosphonofluoridate derivative is facilitated, leading to displacement of F⁻ion and formation of a new P–N bond (Inoue et al., 1999).

From a structural perspective, the effectiveness of phosphonofluoridate derivatives also highlight the importance of the tetrahedral phosphorus centre, which resembles the geometry of high-energy tetrahedral intermediates and transition-state-like species encountered during the GS catalytic cycle. This geometric similarity promotes favorable interactions within the enzyme active site, thereby enhancing binding affinity and facilitating precise positioning of the electrophilic P–F bond in proximity to catalytically relevant nucleophiles. Consequently, molecular recognition and covalent reactivity become intrinsically linked, allowing selective activation of the latent P–F electrophile predominantly within the enzyme microenvironment. This combination of transition-state mimicry, active-site preorganization, and controlled electrophile activation is considered an important feature of modern covalent chemical biology strategies and contributes significantly to the potency and selectivity of phosphonofluoridate-based GS inhibitors (Inoue et al., 1999).

2.3 Fluorophosphonate (FP)-based electrophilic probes for serine hydrolase

The fluorophosphonate-based chemistry plays an important role in the activity-based protein profiling (ABPP). FP-based chemistry has been utilized to design and develop probes that forms a stable adducts with nucleophiles, enabling proteome-wide profiling of hydrolase activities. The tunable reactivity and bio-orthogonal compatibility of these FP-probe molecules, have revolutionized ABPP. ABPP is a powerful chemoproteomic approach that employs active site-directed covalent probes to selectively label catalytically active proteins within complex biological systems. The conceptual foundations of this technology can be traced back to studies in the 1970s investigating the mechanism of action of penicillin and other serine-reactive agents (Blumberg and Strominger, 1974). However, the modern ABPP era emerged in the late 1990s with the advent of activity-based probes (ABPs) compatible with proteomic workflows. A major breakthrough was achieved in 1999 when Cravatt and co-workers reported the first fluorophosphonate-based probe for the global profiling of serine hydrolases (Liu et al., 1999). This seminal advance established the framework for activity-based proteomics and catalyzed the widespread adoption of ABPP in drug discovery, target identification, inhibitor development, and the investigation of fundamental biological processes. Representative examples of FP-based ABPP have been discussed herein.

FP-based electrophilic probes have been extensively utilized in ABPP to selectively label and identify members of the serine hydrolase superfamily (Simon and Cravatt, 2010; Chen et al., 2016; Wang et al., 2019; Davison et al., 2022; Long and Cravatt, 2011). At the core of this strategy lies the unique reactivity of the P–F bond, which, although relatively stable under ambient conditions, becomes suitably reactive for nucleophilic substitution within the confined and catalytically optimized environment of an enzyme active site (d’Andrea and Townsend, 2019; Cao et al., 2024b; Homer et al., 2024; Yang et al., 2025). In serine hydrolases, the catalytic serine residue is activated through a hydrogen-bonding network (Long and Cravatt, 2011). This activated serine performs a nucleophilic attack on the electrophilic phosphorus centre of the fluorophosphonate probe, leading to the formation of a pentacoordinate transition state, followed by the displacement of F⁻ion (Liu et al., 1999; d’Andrea and Townsend, 2019). The result is the formation of a stable, covalent phosphonyl-enzyme adduct, effectively mimicking the tetrahedral intermediate of ester hydrolysis and irreversibly inhibiting the enzyme (Silvaggi et al., 2003). FP-based probes exhibit remarkable selectivity for active serine hydrolases, enabling functional profiling rather than mere detection of protein abundance. This activity dependence has been experimentally validated through the loss of labelling upon enzyme denaturation or mutation of the catalytic serine residue, confirming that covalent modification occurs specifically at the active site (Higson et al., 1999). Structurally, FP probes are modular in design, typically comprising three components: (i) the reactive fluorophosphonate warhead, (ii) a linker that modulates solubility and accessibility, and (iii) a reporter tag or a fluorophore (d’Andrea and Townsend, 2019; Huang et al., 2012).

2.3.1 Biotinylated FP-probe enables covalent labelling and affinity-based capture of serine hydrolases

Diisopropyl fluorophosphate (DFP) is a small organophosphorus compound that covalently inhibits serine esterases/proteases by reacting with the active-site serine residue (Cohen et al., 1967). Building upon this established reactivity, DFP was transformed from a simple inhibitor into a functional chemical probe capable not only of enzyme inactivation but also of detection and isolation. To achieve this, a biotinylated analogue of DFP was designed by Nikolaev et al. (Higson et al., 1999) in which the reactive phosphorofluoridate group was retained as the electrophilic “warhead,” while a biotin moiety was introduced as a reporter tag. Importantly, these two functional elements were separated by a flexible aminohexyl linker, ensuring that the bulky biotin unit would not hinder access of the phosphorus centre to the enzyme active site. This rational design led to the target molecule, 6-N-biotinylaminohexyl isopropyl phosphorofluoridate (Figure 7A). The fluorophosphonate-biotin derivative was synthesized by coupling a biotin ester precursor with 6-aminohexanol, followed by phosphorylation to form the corresponding phosphonate intermediate. Subsequent esterification and oxidative fluorination afforded the target fluorophosphonate product (Figure 7A) (Higson et al., 1999). The probe was found to function through covalent modification of the enzyme. Upon binding within the active site of a serine esterases/proteases, the nucleophilic serine residue attacks the phosphorus centre, displacing F⁻ion and forming a stable P–O bond. This results in irreversible inhibition of the enzyme. The effectiveness of this designed probe was evaluated as an inhibitor of the serine protease α-chymotrypsin. Pre-incubation of α-chymotrypsin with the probe (1 mM) in HEPES-NaOH buffer (pH 7.4) for 30 min resulted in complete loss of enzymatic activity toward the chromogenic substrate N-succinyl-Ala-Ala-Pro-Phe-p-nitroanilide, as determined by absorbance measurements at 410 nm. The absence of detectable enzyme activity following treatment suggests that the phosphorofluoridate-containing compound acts as a potent irreversible inhibitor of α-chymotrypsin. These results support the accessibility and reactivity of the phosphorofluoridate P–F functionality within the enzyme active site (Higson et al., 1999).

However, the effective functional strength of this system lies beyond inhibition. The appended biotin tag enables downstream detection of the labelled enzyme. Due to the extraordinarily strong affinity between biotin and avidin or streptavidin, the modified proteins can be readily visualized using enzyme-linked or radiolabelled detection systems in assays’ such as ELISA or Western blotting (Figure 7B) (Hermanson, 1996). Furthermore, immobilized avidin matrices allow for the selective capture and purification of these labelled enzymes from complex biological mixtures by affinity chromatography (Figure 7B). In this way, the probe serves dual function: identification of active serine esterases/proteases through covalent modification and their isolation for further biochemical analysis.

2.3.2 FP-probes for selective labelling of serine hydrolases and proteomic profiling

Lin et al. (Huang et al., 2012) developed FP-based probes that exhibits high chemoselectivity and efficiency in covalently targeting the catalytic serine residue (Ser102) of PhaZ from Bacillus thuringiensis. This enzyme was thereby validated as representative model of serine hydrolases, as evidenced by the absence of detectable off-target modification at non-catalytic serine sites (Figure 7C). The remarkable site selectivity of FP probes toward catalytic serine residues is largely attributed to the unique microenvironment of serine hydrolase active sites. In these enzymes, the catalytic Ser-His-Asp/Glu triad is generally understood to activate the serine hydroxyl group through a proton-transfer relay, generating a highly nucleophilic alkoxide species that is significantly more reactive than non-catalytic serine residues. Two developed hydrophilic FP-based probes, FP-ethylenedioxy (edo)-fluorescein and FP-edo-biotin, were shown to enable effective labelling of target serine hydrolases as well as affinity-based enrichment via pull-down. Their enhanced aqueous solubility is expected to broaden the coverage of identifiable serine hydrolase targets. Furthermore, FP-edo-fluorescein was employed to profile serine hydrolase activity across diverse murine tissues, thereby providing a comparative framework for enzymatic activity among biological samples. In parallel, pull-down of FP-edo-biotin-labelled proteins followed by mass spectrometric analysis proved effective for the identification of active serine hydrolases in mice liver. Given that a substantial fraction of serine hydrolases remains functionally uncharacterized, application of this strategy to compare normal and diseased states is anticipated to reveal aberrantly regulated enzymes potentially implicated in pathological processes (Huang et al., 2012).

Collectively, fluorophosphonate-based probes illustrate how classical organophosphorus reactivity can be harnessed and refined within biological contexts to achieve selective, covalent modification of enzyme targets. From a medicinal chemistry perspective, these systems provide a strong framework for the development of next-generation covalent inhibitors and chemical probes for precision targeting of functional biomolecules (Huang et al., 2012; Higson et al., 1999).

3 Current limitations and challenges in PFEx chemistry-based applications

The concept of PFEx chemistry has emerged as a promising platform in chemical biology and covalent drug discovery because of its unique reactivity profile and potential for selective bioconjugation. Despite these advances, several critical limitations continue to restrict its broader translational and therapeutic applications. One major challenge arises from the ability of PFEx warheads to react with multiple nucleophilic amino acid residues, including Lys, Tyr, His, and Cys, thereby increasing the risk of nonspecific protein labelling and off-target interactions (Yang et al., 2025). Furthermore, comprehensive chemoproteomic and proteome-wide investigations of PFEx chemistry remain inadequate due to the limited understanding of its residue selectivity, protein scope, and biological targeting preferences relative to the more extensively studied SuFEx concept (Huang and Jones, 2023; Yang et al., 2025). In addition, PFEx warheads exhibit a relatively narrow hydrolytic stability window and finely balanced bioreactivity profile, both of which remain insufficiently explored and consequently restrict their rational utilization in probe and covalent drug design (Yang et al., 2025). Efficient PFEx reactions under physiological conditions generally rely on proximity-driven activation, requiring precise molecular recognition and specific orientation of the phosphorus-based warhead toward the target amino acid residue. In the absence of such spatial arrangement, covalent bond formation may be significantly compromised, thereby limiting the general applicability of PFEx-based approaches (Yang et al., 2025). At present, PFEx-based investigations are largely confined to in-vitro and cellular studies, while systematic in-vivo validation remains limited. Consequently, the pharmacokinetic behaviour, long-term toxicity, and overall biological effects of PFEx-derived covalent modifiers are still less understood, thereby restricting assessment of their true clinical potential (Cao et al., 2024b). Moreover, the presence of P–F bonds in PFEx warheads raises important toxicological concerns because structurally related organophosphorus compounds are frequently associated with acetylcholinesterase inhibition, neurotoxicity, and irreversible enzyme inactivation. Thus comprehensive toxicological evaluations are required before their widespread therapeutic implementation (Yang et al., 2025). Another major limitation is the insufficient understanding of how the local protein microenvironment influences PFEx reactivity and selectivity. Parameters such as pKa, neighbouring amino acid residues, H-bonding networks, and microenvironmental polarity can affect selective covalent modification; however, these factors remain significantly underexplored, particularly in comparison with SuFEx chemistry (Jones and Kelly, 2020).

4 Conclusion

PFEx chemistry has emerged as a powerful and conceptually unique strategy in the field of click chemistry, offering unique opportunities for chemical biology and medicinal chemistry through the controlled reactivity of P(V)–F electrophiles. The ability of PFEx warheads to form biologically relevant P–O and P–N linkages while maintaining an optimal balance between stability and tuneable reactivity has enabled their broad application in site-selective protein modifications. Amongst the most promising directions for PFEx strategy is the development of programmable protein crosslinking that explore the residue selective reactivity for controlled covalent linkage of biomacromolecules. Another major strength of PFEx includes incorporation of phosphorus warheads in the design of the covalent therapeutics. The approachoffers significant potential for targeting enzymes and previously inaccessible binding sites through tuneable P(V)-centred electrophiles. The ability to strategically alter the steric and electronic properties around the P(V)-centre may further enhance selectivity, reaction kinetics, and minimize off-target reactivity in complex biological environments. Importantly, the incorporation of phosphoramidofluoridate-based UAAs into proteins has demonstrated the selective intracellular bioconjugation under biologically compatible conditions. Despite these advancements, PFEx strategy exhibits several challenges, including limited chemoproteomic validation, incomplete understanding of residue selectivity, and the need for broader substrate compatibility under physiological conditions which restrict the translational and therapeutic applications. Addressing these challenges will require interdisciplinary efforts at the interface of synthetic chemistry, chemical biology, structural biology, proteomics, and molecular pharmacology. With continued advancement, PFEx chemistry is well positioned to emerge as an important next-generation strategy of targeted therapeutic innovation for precision bioconjugation, covalent therapeutics, biopharmaceuticals and theranostic development.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Debbie C. Crans, Colorado State University, United States

Reviewed by: Richard C. Holz, Colorado School of Mines, United States