PLpro inhibitor

Understanding Viral Replication and Substrate Specificity

By studying how PLpro inhibitors work, scientists can gain valuable insights into the mechanisms of viral replication. This knowledge can be used to develop more targeted therapies that disrupt specific steps in the viral lifecycle. For instance, a recent study employed activity profiling to identify the molecular determinants of PLpro substrate specificity []. This information can pave the way for designing inhibitors that specifically target PLpro's interaction with crucial viral proteins.

Drug Discovery and Repurposing

A major thrust of PLpro inhibitor research is the identification of potential drugs for COVID-19 treatment. Computational methods like computer-aided drug design (CADD) are being used to virtually screen existing drugs and libraries of novel compounds for their ability to bind and inhibit PLpro []. Additionally, research is underway to discover new small-molecule inhibitors with high selectivity for PLpro, minimizing the risk of side effects by avoiding interaction with human proteins.

Another promising avenue is drug repurposing, where existing medications approved for other diseases are investigated for their potential effectiveness against COVID-19. In silico studies have identified several drugs, including ribavirin and valganciclovir, that exhibit binding affinity to PLpro, suggesting their potential use as antiviral treatments [].

Papain-like protease, commonly referred to as PLpro, is an essential enzyme found in coronaviruses, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). It plays a dual role as both a peptidase and a deubiquitinating enzyme, which allows it to regulate host protein responses and evade the immune system by removing ubiquitin and interferon-stimulated gene 15 modifications from host proteins. This activity is crucial for viral replication and survival, making PLpro a prime target for antiviral drug development .

• PLpro inhibitors target the active site of PLpro, where the enzyme binds and cleaves viral proteins. By fitting into this active site, the inhibitor prevents the virus from cleaving the proteins it needs to replicate [, ].
• Some PLpro inhibitors may also affect PLpro's ability to interfere with the host's immune response [].

• As PLpro inhibitors are under development, their safety profiles are still being investigated.
• Generally, any new drug candidate undergoes rigorous testing to assess potential side effects, toxicity, and interactions with other medications [].

PLpro catalyzes the hydrolysis of peptide bonds in proteins that contain the LXGG motif, cleaving between glycine and alanine or lysine residues. This enzymatic activity is vital for processing viral polyproteins into functional proteins necessary for the virus's life cycle. Additionally, PLpro can hydrolyze ubiquitin and ISG15 modifications, which are critical for regulating immune responses in host cells . The reaction involves the formation of a tetrahedral intermediate stabilized by the enzyme's catalytic triad, which includes a cysteine residue that acts as a nucleophile .

The biological activity of PLpro is characterized by its ability to modulate host immune responses. By cleaving ubiquitin and ISG15 from host proteins, PLpro effectively suppresses the innate immune response, allowing the virus to replicate more efficiently. Studies have shown that inhibiting PLpro can significantly reduce viral loads in infected cells, highlighting its potential as a therapeutic target . Furthermore, PLpro has been shown to interact with various substrates, influencing cellular signaling pathways and immune modulation .

The synthesis of PLpro inhibitors has focused on designing small molecules that can effectively bind to the active site of the enzyme. Various approaches include:

• Covalent Inhibitors: These compounds typically feature electrophilic groups that form irreversible bonds with the catalytic cysteine residue (Cys111) of PLpro. Examples include derivatives based on GRL0617, which have been modified to enhance potency and selectivity .
• Non-Covalent Inhibitors: These inhibitors rely on non-covalent interactions such as hydrogen bonding and hydrophobic interactions to bind to PLpro without forming permanent modifications. Research has identified several promising non-covalent inhibitors through high-throughput screening methods .
• Novel Structural Classes: Recent studies have uncovered new classes of inhibitors, including thiophene, cyanofuran, and triazoloquinazoline compounds, which exhibit significant inhibitory activity against PLpro .

PLpro inhibitors are primarily being explored for their potential use in treating infections caused by coronaviruses such as SARS-CoV-2. By targeting PLpro, these compounds aim to restore host immune responses and reduce viral replication. Additionally, due to their ability to modulate cellular signaling pathways, there is potential for these inhibitors in broader therapeutic applications beyond viral infections .

Interaction studies involving PLpro have utilized techniques such as X-ray crystallography and mass spectrometry to elucidate binding mechanisms with various inhibitors. These studies have revealed critical interactions between inhibitors and specific residues within the active site of PLpro. For instance, hydrogen bonds between inhibitor amides and key amino acids like Asp164 have been documented, providing insights into how structural modifications can enhance inhibitor efficacy . Furthermore, conformational changes in the enzyme upon inhibitor binding have been characterized, aiding in the rational design of more effective compounds .

Several compounds have been identified that share similarities with PLpro inhibitors but differ in their mechanisms or target sites:

PLpro inhibitors are unique due to their specific targeting of the papain-like protease's dual function in viral replication and immune evasion. Unlike other antiviral agents that may target different viral enzymes or processes, these inhibitors focus on disrupting a critical aspect of coronavirus biology.

The structural characterization of Severe Acute Respiratory Syndrome Coronavirus 2 papain-like protease has been achieved through extensive crystallographic studies, providing detailed insights into its three-dimensional architecture [1] [6]. The protein crystallizes in multiple space groups, with the most common being C2, yielding crystals with cell dimensions of approximately a = 58.4 Å, b = 189.7 Å, c = 63.1 Å, and β = 98.7° [14]. High-resolution structures have been determined to resolutions ranging from 2.23 Å to 2.9 Å, providing sufficient detail for comprehensive structural analysis [3] [6].

The overall architecture of Severe Acute Respiratory Syndrome Coronavirus 2 papain-like protease exhibits the characteristic "thumb-palm-fingers" arrangement that is conserved among papain-like cysteine proteases [1] [2]. This three-domain organization creates a cleft-like active site that accommodates substrate binding and catalysis. The thumb domain contains the N-terminal region and contributes the catalytic cysteine residue, while the palm domain houses the majority of the active site machinery [7]. The fingers domain, positioned opposite to the thumb, contains a structurally essential zinc-binding site coordinated by four cysteine residues [2] [11].

Crystallographic studies have revealed that papain-like protease adopts different packing arrangements depending on crystallization conditions [6]. The unliganded structure commonly crystallizes with three closely packed monomers in the asymmetric unit, while certain mutant forms, such as the catalytically inactive Cys111Ser variant, exhibit alternative packing with two dimers per asymmetric unit [6]. This unique crystal form demonstrates higher solvent content (56%) and looser packing, making it particularly suitable for fragment-based screening approaches [6].

The structural integrity of papain-like protease is maintained by several key architectural features [2]. The zinc-binding site in the fingers domain is absolutely essential for maintaining the overall fold and catalytic competence of the enzyme [2] [11]. The coordination of the zinc ion by Cys189, Cys192, Cys224, and Cys226 creates a stable tetrahedral geometry that is crucial for structural stability [11]. Disruption of this zinc coordination leads to complete loss of catalytic activity and structural collapse [2].

Catalytic Triad and Active Site Configuration

The active site of Severe Acute Respiratory Syndrome Coronavirus 2 papain-like protease contains a classical catalytic triad consisting of Cys111, His272, and Asp286 [1] [7]. This triad is positioned at the interface between the thumb and palm subdomains, creating an optimal geometric arrangement for catalysis [2] [7]. The catalytic cysteine (Cys111) is located at the base of α-helix α4 in the thumb domain, positioned 3.7 Å from the catalytic histidine (His272) [7].

The catalytic histidine (His272) is situated at the base of the palm domain, adjacent to the flexible blocking loop 2 [2]. The side chain of His272 is positioned to act as a general base during the catalytic mechanism, abstracting a proton from the catalytic cysteine to enhance its nucleophilicity [7]. The third member of the triad, Asp286, is located within hydrogen-bonding distance (2.7 Å) of His272, where it functions to stabilize the proper ionization state of the histidine residue [7].

The oxyanion hole is formed by the backbone amide groups of Gly143 and Cys145, along with the indole nitrogen of Trp106 [2] [13]. This structural feature is essential for stabilizing the negatively charged tetrahedral intermediates that form during the catalytic cycle [2]. The Trp106 residue shows remarkable conformational flexibility, adopting different orientations depending on substrate binding status and the specific step of the catalytic reaction [13].

The active site exhibits significant plasticity, particularly in the region surrounding the catalytic triad [13]. Crystal structures of papain-like protease bound to different inhibitors reveal that the oxyanion hole can exist in various active and inactive conformations [13]. This conformational variability is thought to be important for accommodating different substrates and may represent different snapshots of the catalytic cycle [13].

The spatial arrangement of the catalytic triad in Severe Acute Respiratory Syndrome Coronavirus 2 papain-like protease is highly similar to that observed in other papain-like proteases, including those from Severe Acute Respiratory Syndrome Coronavirus and human ubiquitin-specific proteases [15]. However, subtle differences in the precise positioning of catalytic residues contribute to differences in substrate specificity and catalytic efficiency between these enzymes [13].

Substrate Binding Pockets (S1-S4)

The substrate binding region of Severe Acute Respiratory Syndrome Coronavirus 2 papain-like protease is organized into four distinct subsites designated S1, S2, S3, and S4, which accommodate the corresponding P1, P2, P3, and P4 positions of substrate peptides [8] [14]. These binding pockets are essential for recognizing the conserved leucine-X-glycine-glycine cleavage motif found in viral polyproteins and host protein substrates [9] [14].

The S1 subsite is the most restrictive binding pocket, exhibiting absolute specificity for glycine residues at the P1 position [8]. This pocket is formed by the side chains of Trp106 and Asn109, creating a narrow cavity that can only accommodate small, non-branched residues [14]. The strict glycine requirement at P1 is enforced by steric constraints that would clash with any larger side chain [8].

The S2 subsite similarly demonstrates high specificity for glycine residues at the P2 position [8]. This pocket is delineated by Leu163, Pro248, and Tyr264, forming another restrictive binding environment [14]. The dual glycine specificity at P1 and P2 positions creates significant challenges for inhibitor design, as these pockets offer limited opportunities for high-affinity interactions [8].

The S3 subsite exhibits considerably more flexibility compared to S1 and S2, capable of accommodating various amino acid residues at the P3 position [14]. Key residues forming this pocket include Asp164, Tyr264, and Tyr268 [8]. Substrate specificity profiling has revealed that papain-like protease can tolerate hydrophobic, polar, and even some charged residues at this position, although leucine remains strongly preferred [14].

The S4 subsite represents the most expansive and well-characterized binding pocket, forming a deep hydrophobic cavity that preferentially accommodates leucine residues [8] [14]. This pocket is bounded by Asp164, Tyr264, Tyr268, Tyr273, and Thr301, creating a substantial binding surface [8]. The S4 pocket has emerged as the primary target for inhibitor development due to its size and accessibility [3] [8].

Structural studies have revealed that the S4 pocket can be subdivided into distinct regions with different chemical properties [8]. The deeper portion of the pocket contains an acidic patch formed by Asp164, Tyr273, and Thr301, while the entrance region is dominated by aromatic residues [14]. This chemical diversity within the S4 pocket provides multiple opportunities for inhibitor optimization through structure-based drug design [8].

The substrate binding pockets undergo conformational changes upon ligand binding, particularly involving the flexible blocking loop 2 region [12]. These induced-fit mechanisms contribute to the substrate selectivity and catalytic efficiency of papain-like protease [12]. Understanding these dynamic aspects of substrate recognition is crucial for the development of effective inhibitors [12].

Blocking Loop 2 (BL2) Dynamics

Blocking loop 2 is a highly flexible six-residue β-hairpin structure (residues 267-272) that plays a critical role in substrate recognition and inhibitor binding [8] [12]. This loop exhibits remarkable conformational plasticity, adopting different conformations depending on the binding state of the enzyme [12]. In the unbound state, blocking loop 2 predominantly exists in an open conformation, while ligand binding induces closure of the loop [12].

The loop contains an unusual β-turn formed by Tyr268 and Gln269, which undergoes significant conformational rearrangement upon inhibitor binding [8]. Crystallographic studies have revealed that the binding of inhibitors such as GRL-0617 induces a dramatic repositioning of these residues, effectively sealing the entrance to the substrate binding cleft [3] [8]. This conformational change creates a more enclosed binding environment that contributes to inhibitor potency [8].

Molecular dynamics simulations have provided detailed insights into the dynamic behavior of blocking loop 2 [12]. Principal component analysis of simulation trajectories reveals that the loop samples multiple conformational states, with the first two principal components accounting for over 50% of the motion variability [12]. The loop dynamics are coupled to conformational changes in distant regions of the protein, suggesting allosteric communication networks [12].

The conformational flexibility of blocking loop 2 has been quantified using distance measurements between key residues [12]. In the presence of ligands, the distances between Tyr268 and stable reference points such as Arg166 and Pro299 are consistently smaller compared to the unbound state [12]. This loop closure is accompanied by changes in the χ1 dihedral angles of Tyr268 and Gln269, reflecting the reorientation of their side chains [12].

The dynamics of blocking loop 2 exhibit inhibitor-specific patterns [12]. Different classes of inhibitors induce characteristic conformational signatures in the loop, suggesting that the precise mechanism of loop closure depends on the chemical structure of the bound ligand [12]. These observations have important implications for inhibitor design, as compounds that can effectively induce and stabilize the closed loop conformation may exhibit enhanced potency [8].

The temporal aspects of blocking loop 2 dynamics have been investigated using surface plasmon resonance kinetic studies [8]. Inhibitors that engage the loop region exhibit significantly slower dissociation rates compared to those that do not interact with this structural element [8]. This correlation between loop engagement and residence time suggests that targeting blocking loop 2 dynamics could be a viable strategy for developing long-acting inhibitors [8].

Ubiquitin-Like Domain and Its Structural Implications

The ubiquitin-like domain of papain-like protease encompasses the N-terminal region (residues 1-60) and adopts a canonical ubiquitin fold characterized by a β-grasp structure [11]. This domain is connected to the catalytic core through a flexible linker region, allowing for relative motion between the two structural units [11]. Despite its structural similarity to ubiquitin, the ubiquitin-like domain does not participate directly in substrate binding or catalysis [11].

Mutational analysis has demonstrated that the ubiquitin-like domain is largely dispensable for proteolytic activity [11]. Deep mutational scanning experiments reveal that residues within this domain are generally tolerant of amino acid substitutions, in contrast to the highly conserved catalytic core [11]. Truncation studies confirm that removal of the ubiquitin-like domain results in minimal impact on catalytic activity against small peptide substrates [11].

The structural role of the ubiquitin-like domain appears to be primarily regulatory rather than catalytic [11]. Nuclear magnetic resonance studies suggest that the domain may influence the conformational dynamics of the catalytic core through long-range allosteric effects [11]. However, the precise mechanism by which these allosteric interactions modulate enzyme function remains unclear [11].

Comparative analysis with other papain-like proteases reveals significant variability in the structure and function of ubiquitin-like domains [11]. While the basic fold is conserved, the specific amino acid sequences and surface properties differ substantially between coronavirus species [11]. These differences may contribute to species-specific variations in substrate recognition and regulatory mechanisms [11].

The ubiquitin-like domain may also play a role in protein-protein interactions within the viral replication complex [11]. Although direct evidence for such interactions is limited, the presence of this domain in all coronavirus papain-like proteases suggests an evolutionary pressure to maintain this structural feature [11]. Future studies will be needed to fully elucidate the functional significance of the ubiquitin-like domain in the context of viral replication [11].

Comparative Analysis with PLpro from Other Coronaviruses

Structural and functional comparisons between papain-like proteases from different coronavirus species reveal both conserved and divergent features that reflect evolutionary adaptation to different hosts and pathogenic strategies [4] [13] [18]. The overall fold and catalytic mechanism are highly conserved, but significant differences exist in substrate specificity and regulatory mechanisms [4] [13].

The most extensively studied comparison is between Severe Acute Respiratory Syndrome Coronavirus 2 and Severe Acute Respiratory Syndrome Coronavirus papain-like proteases, which share 83% sequence identity [4]. Despite this high similarity, significant differences exist in their substrate preferences [4]. Severe Acute Respiratory Syndrome Coronavirus 2 papain-like protease preferentially cleaves interferon-stimulated gene 15 conjugates, while Severe Acute Respiratory Syndrome Coronavirus papain-like protease shows stronger activity against K48-linked polyubiquitin chains [4] [13].

Structural analysis reveals that these functional differences arise from subtle changes in the S1 ubiquitin-binding site [13]. Specifically, variations in the fingers domain between the two enzymes create different binding environments for ubiquitin and interferon-stimulated gene 15 substrates [13]. These differences have been mapped to specific residues that modulate the affinity and orientation of substrate binding [13].

The comparison with Middle East Respiratory Syndrome Coronavirus papain-like protease reveals more substantial structural differences, reflecting the greater evolutionary distance between these viruses [18]. The sequence identity drops to approximately 30%, with corresponding changes in substrate specificity and inhibitor sensitivity [18]. The blocking loop 2 region shows reduced flexibility in Middle East Respiratory Syndrome Coronavirus papain-like protease compared to Severe Acute Respiratory Syndrome coronaviruses [17].

Cross-reactivity studies with inhibitors designed for Severe Acute Respiratory Syndrome Coronavirus papain-like protease demonstrate variable efficacy against other coronavirus proteases [5] [18]. Inhibitors such as GRL-0617 show excellent activity against both Severe Acute Respiratory Syndrome coronavirus variants but reduced potency against more distantly related enzymes [5]. This pattern highlights the importance of species-specific structural features in inhibitor recognition [18].

The comparative analysis extends to the deubiquitinating and deISGylating activities of different coronavirus papain-like proteases [21]. Functional studies reveal that the relative importance of these activities varies significantly between species [21]. While disruption of deubiquitinating activity has minimal impact on Severe Acute Respiratory Syndrome Coronavirus 2 replication, the same mutations severely attenuate Middle East Respiratory Syndrome Coronavirus [21].

Polyprotein Processing Activity

The papain-like protease exhibits essential enzymatic functions in the processing of coronavirus polyproteins, cleaving at specific junctions to release mature nonstructural proteins required for viral replication. The enzyme demonstrates exquisite substrate specificity through recognition of conserved sequence motifs and precise catalytic mechanisms.

Cleavage at nonstructural protein 1/2, nonstructural protein 2/3, and nonstructural protein 3/4 Junctions

Papain-like protease processes the coronavirus polyprotein at three distinct cleavage sites, each exhibiting unique recognition characteristics and processing efficiencies [1] [2]. The enzyme cleaves between nonstructural protein 1 and nonstructural protein 2 at the sequence LNGG↓AYTR, between nonstructural protein 2 and nonstructural protein 3 at LKGG↓APTK, and between nonstructural protein 3 and nonstructural protein 4 at LKGG↓KIVN [3] [4].

The cleavage efficiency varies significantly among these sites, with the nonstructural protein 1/nonstructural protein 2 junction being processed most rapidly, followed by the nonstructural protein 2/nonstructural protein 3 site, while the nonstructural protein 3/nonstructural protein 4 junction requires extended enzyme domains for efficient processing [5] [4]. Structural analysis reveals that the nonstructural protein 3/nonstructural protein 4 cleavage site necessitates additional hydrophobic domain downstream of the catalytic domain, suggesting a more complex recognition mechanism [5].

The differential processing rates at these junctions have important biological implications, as the nonstructural protein 2-3 intermediate may play specific roles in viral replication, as evidenced by pulse-chase studies demonstrating slower processing kinetics in virus-infected cells [4]. This temporal regulation of polyprotein processing ensures proper assembly and function of the viral replication complex.

Recognition of LXGG Motif

The papain-like protease demonstrates stringent substrate specificity through recognition of the LXGG motif at the P4-P1 positions of substrate sequences [6] [7]. This tetrapeptide motif represents a critical determinant for enzyme-substrate interaction, with each position contributing differently to binding affinity and catalytic efficiency.

The P4 position specifically accommodates hydrophobic residues, with leucine being the preferred amino acid across all coronaviral papain-like proteases [7] [8]. However, structural flexibility exists at this position, as demonstrated by the accommodation of valine at the nonstructural protein 1/nonstructural protein 2 cleavage site and isoleucine at the Middle East respiratory syndrome coronavirus nonstructural protein 3/nonstructural protein 4 site [3] [9]. The S4 binding pocket of the enzyme consists of hydrophobic residues that form favorable interactions with these branched aliphatic amino acids [7].

The P3 position exhibits broad substrate specificity, accepting various amino acids including lysine, arginine, and asparagine in natural cleavage sites [7]. This flexibility allows the enzyme to recognize diverse substrate sequences while maintaining overall specificity through the other positions of the motif. The P2 and P1 positions strictly require glycine residues, with any substitutions at these positions resulting in complete loss of enzymatic activity [3] [4].

Kinetic Parameters of Proteolytic Processing

Comprehensive kinetic analysis reveals distinct catalytic parameters for papain-like protease across different coronavirus species and substrate types. The enzyme exhibits varying catalytic efficiencies depending on the specific substrate and viral origin, providing insights into evolutionary adaptations and functional optimization.

For the peptide substrate Z-RLRGG-AMC, SARS-CoV-2 papain-like protease demonstrates a Km value of 6.9 ± 1.4 μM and kcat of 0.11 ± 0.03 s⁻¹, resulting in a catalytic efficiency (kcat/Km) of 0.016 μM⁻¹ s⁻¹ [10] [11]. In comparison, SARS-CoV papain-like protease shows higher catalytic efficiency with kcat/Km of 0.3 ± 0.1 μM⁻¹ min⁻¹, while MERS-CoV papain-like protease exhibits significantly lower efficiency at 0.003 ± 0.0001 μM⁻¹ min⁻¹ [12].

The kinetic parameters demonstrate that papain-like protease exhibits substrate-dependent catalytic behavior, with higher efficiencies observed for full-length protein substrates compared to peptide substrates. This finding suggests that the enzyme has evolved specific recognition mechanisms for its natural protein substrates, utilizing additional binding sites beyond the active site for optimal catalysis [12].

Deubiquitinating Activity

The papain-like protease possesses significant deubiquitinating enzyme activity, functioning as a cysteine protease that removes ubiquitin modifications from cellular proteins. This activity represents a crucial mechanism for viral immune evasion and cellular hijacking, distinguishing coronavirus papain-like proteases from typical viral proteases.

Structural Basis for Ubiquitin Recognition

The structural architecture of papain-like protease enables specific recognition and binding of ubiquitin through a sophisticated binding site arrangement. The enzyme adopts a canonical right-handed thumb-palm-fingers architecture reminiscent of human ubiquitin-specific proteases, with the ubiquitin molecule positioning in the palm domain and stabilization by the zinc-binding fingers subdomain [1] [13].

The ubiquitin recognition mechanism involves a bivalent binding mode, where the enzyme utilizes both S1 and S2 binding sites for optimal substrate interaction [14] [15]. The S1 site accommodates the proximal ubiquitin molecule, while the S2 site engages the distal ubiquitin in polyubiquitin chains. This dual-site recognition mechanism explains the enzyme's preference for polyubiquitin substrates over mono-ubiquitin [14].

Crystal structure analysis of papain-like protease bound to ubiquitin-aldehyde reveals a protein-protein interface of approximately 900 Ų, significantly smaller than the 3600 Ų interface observed in human deubiquitinating enzymes such as HAUSP [14] [15]. This reduced binding surface area contributes to the enzyme's distinct substrate specificity profile compared to human deubiquitinases.

The catalytic site architecture features a flexible glycine-hinged β-turn that blocks access to the active site in the unbound state, undergoing conformational changes upon substrate binding to facilitate catalysis [14] [15]. This induced-fit mechanism ensures substrate specificity and prevents non-specific proteolysis.

Cleavage of K48- and K63-linked Ubiquitin Chains

Papain-like protease demonstrates remarkable specificity for different ubiquitin chain linkages, with distinct preferences varying among coronavirus species. The enzyme exhibits strong preference for K48-linked polyubiquitin chains over K63-linked chains, with this specificity mediated by the S2 binding site architecture [14] [16].

For SARS-CoV papain-like protease, K48-linked pentaubiquitin chains are rapidly degraded through sequential cleavage, while K63-linked chains show only moderate processing activity [14]. Time-course experiments demonstrate that the enzyme processively cleaves K48-linked chains from the distal end, generating mono-ubiquitin as the primary product. The preference for K48-linkages stems from optimal positioning of the distal ubiquitin molecule in the ridge region of the thumb domain [14].

SARS-CoV-2 papain-like protease exhibits similar K48-linkage specificity, though with reduced overall deubiquitinating activity compared to SARS-CoV [17] [18]. The diminished activity results from subtle amino acid differences in the fingers domain that affect S1 ubiquitin binding site interactions. Despite reduced mono-ubiquitin processing, the enzyme maintains efficient cleavage of K48-linked polyubiquitin chains through the S2 site mechanism [16].

MERS-CoV papain-like protease shows broader linkage specificity, efficiently cleaving K6-, K11-, K33-, K48-, and K63-linked diubiquitin chains with similar reaction rates [18]. This promiscuous deubiquitinating activity distinguishes MERS-CoV from SARS coronaviruses and may contribute to its distinct pathogenesis profile.

Comparison with Human Deubiquitinases

Papain-like protease shares structural and functional similarities with human deubiquitinating enzymes, particularly those of the ubiquitin-specific protease family, yet exhibits distinct characteristics that enable selective inhibition. The enzyme contains a catalytic triad consisting of cysteine, histidine, and aspartate residues, similar to human deubiquitinases, but differs in key structural elements that determine substrate specificity [1] [19].

The most significant distinction lies in the oxyanion hole composition, where papain-like protease utilizes tryptophan 107 for oxyanion stabilization, while human ubiquitin-specific proteases typically employ glutamine or asparagine residues [1] [19]. This structural difference contributes to the enzyme's unique substrate preferences and enables selective inhibition by compounds such as GRL0617, which do not affect human deubiquitinases [19].

Kinetic analysis reveals that papain-like protease exhibits comparable catalytic efficiency to human deubiquitinases for ubiquitin substrates, with kcat/Km values in the range of 1-2 μM⁻¹ min⁻¹ [12]. However, the enzyme demonstrates distinct substrate preferences, showing higher activity for ISG15 substrates compared to most human deubiquitinases, which typically prefer ubiquitin [12].

The enzyme's compact binding interface and reduced surface area interactions distinguish it from human deubiquitinases such as HAUSP and USP14, which form extensive protein-protein contacts with ubiquitin [15]. This structural difference enables the development of selective inhibitors that exploit the unique binding pocket architecture of papain-like protease without affecting human enzyme function.

DeISGylation Activity

The papain-like protease demonstrates exceptionally high deISGylating activity, removing interferon-stimulated gene 15 modifications from cellular proteins with remarkable efficiency and specificity. This activity represents a critical mechanism for viral immune evasion, as ISG15 conjugation is a key component of the host antiviral response.

Interferon-Stimulated Gene 15 Recognition Mechanisms

Papain-like protease recognizes ISG15 through sophisticated molecular mechanisms that involve interactions with both ubiquitin-like domains of the ISG15 protein. The enzyme binds ISG15 with nanomolar affinity, significantly higher than its affinity for ubiquitin, through a dual-domain recognition mechanism [20] [21].

Crystal structure analysis reveals that papain-like protease engages both the amino-terminal and carboxy-terminal ubiquitin-like domains of ISG15, with the carboxy-terminal domain positioned in the S1 binding site and the amino-terminal domain contributing to binding through alternative interactions [13] [20]. This bivalent binding mode explains the enzyme's high affinity for ISG15 compared to mono-ubiquitin.

The recognition mechanism involves specific interactions between the enzyme's palm domain and both ISG15 domains, with the ridge region of the thumb domain providing additional contacts [20]. Nuclear magnetic resonance studies and cross-linking mass spectrometry confirm that both ISG15 domains are essential for high-affinity binding, as isolated domains exhibit only micromolar dissociation constants [21].

The ISG15 carboxy-terminal domain contains the conserved LRGG motif required for enzymatic cleavage, while the amino-terminal domain provides specificity determinants that distinguish ISG15 from ubiquitin [22]. The enzyme's ability to recognize and cleave ISG15 CTD-containing substrates with 160-fold higher efficiency than ubiquitin demonstrates the importance of these domain-specific interactions [16].

Preferential Binding to Interferon-Stimulated Gene 15 over Ubiquitin

SARS-CoV-2 papain-like protease exhibits extraordinary selectivity for ISG15 substrates, demonstrating 350-fold higher catalytic efficiency for ISG15-TAMRA compared to ubiquitin-TAMRA in quantitative fluorescence polarization assays [16] [23]. This remarkable preference stems from optimized binding interactions with the ISG15 structure and enhanced recognition of the ISG15 carboxy-terminal domain.

The preferential binding results from specific amino acid differences between ISG15 and ubiquitin that create favorable interactions with the enzyme's S1 binding site [17]. The ISG15 carboxy-terminal domain exhibits distinct surface properties compared to ubiquitin, enabling tighter binding to the enzyme through hydrophobic and electrostatic interactions [24].

Kinetic analysis demonstrates that ISG15-AMC substrates are processed with kcat/Km values of 29 ± 5.3 μM⁻¹ min⁻¹ for SARS-CoV papain-like protease, compared to 1.5 ± 0.3 μM⁻¹ min⁻¹ for ubiquitin-AMC [12]. This 20-fold difference in catalytic efficiency reflects both improved binding affinity (lower Km) and enhanced turnover rates (higher kcat) for ISG15 substrates.

The enzyme's preference for ISG15 has important biological implications, as ISG15 conjugation is rapidly induced during viral infections and plays crucial roles in antiviral immunity [22]. By preferentially targeting ISG15 modifications, papain-like protease can effectively counteract host immune responses while maintaining some capacity for ubiquitin processing.

Structural Determinants of DeISGylation Specificity

The molecular basis for papain-like protease's ISG15 specificity lies in distinct structural determinants that differentiate ISG15 recognition from ubiquitin binding. These specificity determinants involve both the primary binding site and secondary recognition elements that collectively determine substrate preference.

The primary determinant of ISG15 specificity resides in the S1 binding site, where specific amino acid residues create favorable interactions with the ISG15 carboxy-terminal domain [17] [16]. The enzyme's fingers domain contains variant residues compared to SARS-CoV that enhance ISG15 binding while reducing ubiquitin affinity, explaining the shift in substrate preference between these viruses [17].

Mutagenesis studies reveal that residues in the ridge region of the thumb domain are critical for ISG15 recognition but dispensable for mono-ubiquitin cleavage [14]. These residues, including specific positions in the α7 helix, form contacts with the ISG15 amino-terminal domain and contribute to the dual-domain recognition mechanism. Mutation of these residues significantly reduces ISG15-AMC hydrolysis while maintaining viral protease activity [14].

The structural determinants of ISG15 specificity also involve the enzyme's ability to accommodate both ISG15 domains simultaneously, requiring specific conformational flexibility in the binding sites [20]. Cross-linking mass spectrometry and structural modeling reveal that ISG15 adopts a distinct binding mode compared to K48-linked diubiquitin, with different domain orientations and contact patterns [21].

Computational analysis using alanine scanning mutagenesis identifies key hotspot residues that contribute differentially to ISG15 versus ubiquitin binding [24]. These residues, particularly E167 and F69/Y264, partition between proximal and distal binding sites and show substrate-dependent binding contributions that explain the enzyme's remarkable ISG15 selectivity.