Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Nov 6;127(45):9697–9709. doi: 10.1021/acs.jpcb.3c04293

Catalytic Mechanism of Collagen Hydrolysis by Zinc(II)-Dependent Matrix Metalloproteinase-1

Ann Varghese †, Sodiq O Waheed †, Koteswararao Gorantla †, Isabella DiCastri ‡, Ciara LaRouche ‡, Brendan Kaski §, Gregg B Fields ∥, Tatyana G Karabencheva-Christova †,*
PMCID: PMC10659029  PMID: 37931179

Abstract

graphic file with name jp3c04293_0015.jpg

Human matrix metalloproteinase-1 (MMP-1) is a zinc(II)-dependent enzyme that catalyzes collagenolysis. Despite the availability of extensive experimental data, the mechanism of MMP-1-catalyzed collagenolysis remains poorly understood due to the lack of experimental structure of a catalytically productive enzyme–substrate complex of MMP-1. In this study, we apply molecular dynamics and combined quantum mechanics/molecular mechanics to reveal the reaction mechanism of MMP-1 based on a computationally modeled structure of the catalytically competent complex of MMP-1 that contains a large triple-helical peptide substrate. Our proposed mechanism involves the participation of an auxiliary (second) water molecule (wat2) in addition to the zinc(II)-coordinated water (wat1). The reaction initiates through a proton transfer to Glu219, followed by a nucleophilic attack by a zinc(II)-coordinated hydroxide anion nucleophile at the carbonyl carbon of the scissile bond, leading to the formation of a tetrahedral intermediate (IM2). The process continues with a hydrogen-bond rearrangement to facilitate proton transfer from wat2 to the amide nitrogen of the scissile bond and, finally, C–N bond cleavage. The calculations indicate that the rate-determining step is the water-mediated nucleophilic attack with an activation energy barrier of 22.3 kcal/mol. Furthermore, the calculations show that the hydrogen-bond rearrangement/proton-transfer step can proceed in a consecutive or concerted manner, depending on the conformation of the tetrahedral intermediate, with the consecutive mechanism being energetically preferable. Overall, the study reveals the crucial role of a second water molecule and the dynamics for effective MMP-1-catalyzed collagenolysis.

1. Introduction

Peptide bond enzymatic hydrolysis is critical for many vital biological processes and, in addition, has numerous biotechnological and industrial applications; therefore, hydrolytic zinc enzymes are a target for protein design and thorough mechanistic studies.1–4 A critical proteolytic reaction is collagenolysis.5–7 The main family of enzymes responsible for collagenolysis are the zinc(II)-dependent matrix metalloproteinases (MMPs).7,8 Among the 24 human MMPs, MMP-1 is the most ubiquitously expressed interstitial collagenase. It is involved in the migration and invasion of tumor cells in several types of cancers, including lung, colorectal, bladder, breast, and oral cancers.9–12 Human MMP-1 is a multidomain enzyme comprising an N-terminal catalytic (CAT) domain connected via a 17-residue linker region to a C-terminal hemopexin-like (HPX) domain (Figure 1).13,14 While catalysis occurs in the CAT domain, the HPX domain is crucial in orienting the CAT domain for collagen triple-helical peptide (THP) binding and hydrolysis.15,16 The CAT domain comprises catalytic and structural zinc(II) centers.17,18 Spectroscopic studies suggest that the catalytic zinc(II) ion is pentacoordinated.19 The catalytic zinc(II) coordinates to His218, His222, His228, a water molecule (wat1), and the carbonyl group of the scissile bond (Figure 1) for collagenolysis, while the structural zinc(II) coordinates to His168, Asp170, His183, and His196 (Figure S1). Collagenolysis occurs at the catalytic zinc(II) center, while the structural zinc(II) center provides structural stability to the CAT domain.20,21 The catalytic zinc(II)-coordinated wat1 has been proposed to assist in peptide bond cleavage.22,23 The S10′ exosite in blade I (Figure 1) of the HPX domain participates in multiple interactions with the THP required for substrate recognition and cleavage.16

Figure 1.

Figure 1

Open in a new tab

Structure of the MMP-1·THP complex. The zoomed-in panel on the left shows the catalytic zinc(II) coordinates to the scissile bond carbonyl Gly775, His218, His222, and His228, and a water molecule (wat1) polarized by the base Glu219.

Before collagenolysis, the scissile bond carbonyl oxygen coordinates with the catalytic zinc(II). In the first step of the proposed catalytic mechanism of MMPs, a conserved second coordination sphere (SCS) glutamate residue polarizes wat1, abstracting a proton and generating a nucleophilic hydroxide anion. The significance of this residue has been validated through mutagenesis experiments.24 As the next step, a water-mediated nucleophilic attack on the scissile bond carbonyl forms a tetrahedral intermediate. The lifetime of this intermediate is a subject of discussion, and further experimental studies are required to provide definite clarity. Following this, a rearrangement of the hydrogen bonding network around glutamate allows easy access to the cleavage site. The glutamate acts as a proton shuttle, transferring the proton abstracted from wat1 to the scissile amide group, resulting in the cleavage of the scissile C–N bond (Scheme S1).25–27 The final step involves the release of the product.26,28 Terahertz and X-ray absorption spectroscopy studies suggest that water dynamics can play an important role in the enzyme activity of MMPs.29 Kinetic studies show that MMP-1 hydrolyzes α1(I) THP with a rate constant (kcat) of 0.11 s–1.30

Several computational studies have explored MMP-catalyzed collagenolysis using the “water-promoted pathway” involving the zinc(II)-bound water molecule; however, they used a small cluster model or the enzyme with a small single-stranded model peptide substrate.25–27,31 Mechanistic studies on MMP-225,26,32 by Díaz et al. and Vasilevskaya et al. conclude that protonation of the scissile amide group or product release can be the rate-determining step. Pelmenschikov and Siegbahn propose a two-step mechanism for MMP-331 with the nucleophilic attack as the crucial rate-determining step, whereas recent study on MMP-927 and snake venom metalloproteinase33 by Chen et al. and Castro-Amorim et al. indicate that hydrogen-bond rearrangement and proton transfer are the respective rate-determining steps of the overall catalytic cycle. Furthermore, although the computational studies described above of other MMPs provide critical mechanistic insights, they do not incorporate the full-length THP and the dynamical properties of the productive MMP-1·THP complex and the tetrahedral intermediate.

The available X-ray crystallographic structure of MMP-1 with full-length THP (MMP-1·THP complex) is in an unproductive substrate binding mode as the collagen cleavage site is positioned away from the catalytic zinc(II) ion.34 NMR and small-angle X-ray scattering studies by Bertini et al. suggest that the formation of the catalytically productive MMP-1·THP complex includes a complex conformational transition from the open to the closed form.35,36 Our recent multilevel study, using classical molecular dynamics (MD), metadynamics (MetD), quantum mechanics/molecular mechanics (QM/MM) MD, and QM/MM MetD, revealed the conformational and free-energy changes associated with the formation of the catalytically productive MMP-1·THP complex and the related changes in the coordination state of the catalytic zinc(II) during the process.37

Despite the progress in earlier studies, the catalytic mechanism of MMP-1 remains completely unknown, stable reaction intermediates are yet to be detected, and the effects of the full-length THP and conformational dynamics on the catalytic mechanism are unexplored. SCS residues sensitively contribute to metalloenzyme catalysis;38–42 however, the role of the SCS residues in MMP-1’s reaction mechanism has not been revealed.

To complement the missing knowledge, the present study utilizes the computationally modeled structure of the productive MMP-1·THP complex, MD, and combined QM/MM methods to reveal, for the first time, the catalytic mechanism of MMP-1, including the whole multidomain enzyme, and the full-length THP substrate (that contains 114 residues). This study offers new insights into the key catalytic role of an auxiliary (second) water molecule in the active site and elucidates the importance of the conformational dynamics for the reactivity of the tetrahedral intermediate and its conversion to the final product complex. Furthermore, this study delineates the critical role of SCS residues and their correlated motions for the catalytic process.

2. Methods

The initial structure for this study was chosen from our prior QM/MM MetD simulations of the productive MMP-1·THP complex.37 1 μs classical MD was performed on the MMP-1·THP complex and the tetrahedral intermediate to account for large-scale conformational changes. The simulated system consisted of a total of 90,673 atoms (7214 protein atoms, 8 Cl– counterions, and 83,451 water atoms), and that of the tetrahedral intermediate contained a total of 89,563 atoms (7214 protein atoms, 8 Cl– counterions, and 82,341 water atoms). The THP substrate was numbered based on our previous studies.43 More details on the methodology are provided in the Supporting Information (Pages S6 and S7).

The initial structure for the QM/MM reaction path calculations was selected from the QM/MM MetD trajectory.37 The QM/MM calculations were performed with the ChemShell44 package integrated with the TURBOMOLE45 program for the QM and DL_POLY46 for the MM calculations. The QM region consists of the five-coordinated (5C) zinc(II) center containing the side chains of His218, His222, and His228, the catalytic water molecule wat1, and the second water molecule wat2; the side chain of the SCS residue Glu219; the substrate’s scissile bond residues Gly775 and Ile776 and the peptide carbonyl and the backbone of the adjacent Gln774 and Ala777 (76 atoms in total) (Figure 2).

Figure 2.

Figure 2

Open in a new tab

QM region used in the QM/MM calculations.

The QM part contained 76 atoms with a net charge +1 and was modeled using density functional theory47,48 with the B3LYP functional.49–51 The stationary points of the rate-determining step were also optimized at B3LYP with Grimme’s D3 dispersion with Becke–Johnson (BJ) damping.52 A closed-shell singlet ground state was employed in the MMP-1·THP complex and the tetrahedral intermediate. All amino acid residues and water molecules within 2 Å of the QM part were included in the flexible MM region, and the rest of the protein residues and water molecules were kept in the fixed part of the MM region. There were 27,259 MM atoms in the MMP-1·THP complex and 32,068 MM atoms in the tetrahedral intermediate. In addition, QM/MM studies of the rate-determining step were performed with both 2 and 6 Å flexible MM regions. Furthermore, we tested functionals M06 and PW6B95 for the rate-determining step. More details on the QM/MM methodology are provided in the Supporting Information (Pages S6 and S7). In addition, we evaluated an extended QM region (99 atoms) for the rate-determining step, which contained the immediately adjacent residues to the substrate’s scissile bond (Gln774 and Ala777).

As the TURBOMOLE output file does not list the free-energy corrections, the thermal free-energy corrections are not added to the electronic energy barriers. Therefore, the B3LYP/MM activation energies with zero-point energy (ZPE) corrections approximate the respective free energies. In addition, the ZPE values are provided in the Supporting Information (Table S1). Previous QM/MM studies have reported B2 (single-point energy) + ZPE values and show accurate energy barriers in agreement with experimental values.53–55

3. Results

3.1. Conformational Dynamics of the Productive MMP-1·THP Complex

The catalytically productive MMP-1·THP reactant complex (RC) was generated by MM and QM/MM MetD simulations56–60 and equilibrated by 1 μs MD. The rmsd and radius of gyration (RoG) are presented in Figure S2. In the RC, the THP scissile bond carbonyl oxygen is coordinated to catalytic zinc(II) (which acts as a Lewis acid) along with three histidine residues (His218, His222, and His228) and the catalytic water molecule, wat1. Wat1 maintains averaged distances of 3.12 and 3.23 Å with OE1 and OE2 carboxylate oxygens of Glu219 (Figure S3). The MD simulations reveal the presence of another water molecule (wat2) in addition to the zinc(II)-coordinated wat1 as indicated by the radial distribution function (RDF), which showed sharp peaks at 2.65 and 3.75 Å from the carboxylate carbon (CD) of Glu219 (Figure S4A). Although the presence of wat2 is identified in 40% of the MD trajectory, it is important to note that the most populated conformation is not always the catalytically productive one.61,62 The wat2 is not observed in the crystal structure of the unproductive MMP-1·THP complex.34

MD reveals an extensive network of hydrogen bonding interactions involving catalytic zinc(II) first coordination sphere (FCS) residues (His218, His222, and His228) (Supporting Information, Page S7), favorably orienting them for optimal metal coordination and also contributing to the stability of the catalytic site. For example, His228 backbone nitrogen is hydrogen bonded to Gln774 carboxylate oxygen OE1 of the THP leading (L) chain (present in 55% of the MD trajectory), thus contributing to the proper orientation of the scissile bond to the catalytic zinc(II). The backbone nitrogen and oxygen atoms of Glu219 participate in hydrogen bonds with residues belonging to the same α2 helix, Ser223 (59%) and Val215 (53%) (Figure S5A,C). These interactions preserve the orientation of Glu219 to act as a proton shuttler and to stabilize the α2 helix. The THP L-chain’s Ala777 backbone nitrogen and oxygen, near the scissile bond, make hydrogen bonds with the backbone oxygen of Pro238 (77%) and backbone nitrogen of Tyr240 (28%) (Figure 3A,B) of the key S1′ substrate specificity loop (that lines the shallow S1′ pocket), assisting substrate binding in the catalytic site. Many leading and middle-chain THP residues, namely, L-chain Gln774, Ala777, and Gly778 and M-chain Val783, Gly784, Hyp786 (hydroxyproline786), and Hyp789, participate in a network of multiple hydrogen bonding interactions with both CAT and HPX domain residues [His228 (55%), Pro238 (77%), Tyr240 (28%), Ser239 (39%), Arg291 (71%), Arg304 (68%), and Ala335 (29%)], thus enhancing THP substrate binding and orientation for catalysis (Figure 3A).

Figure 3.

Figure 3

Open in a new tab

(A) Residues involved in hydrogen bonding between THP and the CAT and HPX domains in the RC. (B) Zoomed-in view of the hydrogen bonding interactions between the L-chain residues Gln774, Ala777, and Gly778 (orange) and the CAT domain residues (tan). The hydrogen atoms not involved in interactions are hidden for clarity.

The Dynamic cross-correlation analysis (DCCA) explored long-range correlated motions involving catalytically essential residues. The catalytic and structural zinc(II) ions positively correlate with THP scissile bond residues Gly775 and Ile776 and adjacent Gln774 and Ala777 (Figure S6A). The CAT domains’ VB loop63 (Figure S7) involved in collagenolysis shows correlated motions with the S1′ specificity loop,64 which controls the shape of the S1′ specificity pocket. The VB loop participates in correlated motions with the S1′ specificity loop, and both show correlations with the THP L chain; both loops also correlate with the L-chain scissile bond residues Gly775 and Ile776 (Figure S6A). These correlated motions might facilitate the entry of the scissile bond residues into the active site cleft. Principal component analysis (PCA) shows the THP scissile bond and surrounding residues moving toward the CAT domain and the remaining terminal substrate residues moving in the opposite direction, i.e., away from the CAT domain (Figure S6B), facilitating the formation of productive conformation for catalysis.

3.2. Reaction Mechanism of MMP-1-Catalyzed Collagenolysis

3.2.1. Proton Abstraction by Glu219

The initial RC for QM/MM reaction path calculations corresponds to the state of the substrate bound to the catalytic zinc(II). The wat2 participating in the reaction mechanism was present near wat1 and Glu219 during the MD simulation (Figure S8). We performed multiple calculations with different initial structures and reaction coordinates with a single water molecule only, which were unproductive. We, therefore, invoked the second water molecule present in the active site during the QM/MM MetD and MD. The participation of a second water molecule has also been demonstrated in MMP-365 and MT1-MMP;66 thus, it is not uncommon in MMP-based catalysis. Wat2 stabilizes the coordinated wat1 via a hydrogen bonding interaction at a distance of 1.68 Å, and the oxygen atom of wat2 is hydrogen bonded to the Glu219 carboxylate oxygen at a distance of 1.60 Å (Figure 4).

Figure 4.

Figure 4

Open in a new tab

Stationary point geometries for proton abstraction by Glu219. The corresponding distances (Å) are given in the blue boxes next to each structure. Hydrogen atoms of the substrate other than the scissile bond residues are hidden for clarity.

In the first step of the MMP-1 catalytic mechanism, the conserved negatively charged Glu219 plays the role of a base, abstracting a proton from wat2 that, in turn, accepts a proton from wat1, generating a nucleophilic wat1 hydroxide anion in intermediate IM1 (Scheme 1). The distances between Ow2 and Hw1b, OE1(Glu219) and Hw2b, Ow1 and Hw1b, and Ow2 and Hw2b were chosen as the reaction coordinate. The proton abstraction step proceeds through a transition state (TS1) with a very low activation energy of 1.6 kcal/mol (B3 energy [B2 + ZPE]; Figure 5). The TS1 frequency calculation shows one imaginary value of 617i cm–1. The Ow1–Hw1b bond of wat1 is partly broken (increased from 1.01 Å in RC to 1.23 Å in TS1), and the Ow2–Hw1b is partly formed (decreased from 1.68 Å in RC to 1.19 Å in TS1). In addition, the distance between Glu219 OE1 and Hw2b decreases from 1.60 Å in RC to 1.25 Å in TS1, showing partial protonation of the carboxylate. In the slightly exothermic IM1 (−1.4 kcal/mol) (with respect to the RC), the bonds Hw2b–OE1 (Glu219) and Hw1b–Ow2 are completely formed, and the bonds Ow1–Hw1b and Ow2–Hw2b are completely broken (Figure 4).

Scheme 1. Proposed Reaction Mechanism of MMP-1 with the Participation of Two Water Molecules (wat1 and wat2).

Scheme 1

Open in a new tab
Figure 5.

Figure 5

Open in a new tab

Potential energy profile of the MMP-1 reaction mechanism. The relative B3 (B2 + ZPE) energies are shown in red. The yellow oval-shaped denotation indicates that MD simulations were run on both RC and IM2. Energy values are in kcal/mol.

We further analyzed Mulliken charges in the RC, TS1, and IM1. Glu219 OE1 Mulliken charge in RC is −0.37 and becomes less negative in the corresponding TS1 (−0.34) and IM1 (−0.21), indicating reduced electron density on OE1 upon proton abstraction from wat2. In contrast, the Mulliken charge of wat1 Ow1 becomes more negative (from −0.42 in RC to −0.52 in TS1 and −0.56 in IM1), thus showing an increase in the nucleophilic character of the oxygen atom, which is essential for the next step—the nucleophilic attack on the scissile bond carbon atom. Previous QM/MM studies on MMP-2 by Vasilevskaya et al. showed that the enzyme–substrate (ES) complexes with deprotonated Glu219 (C1) and protonated Glu219 (C2) are isoenergetic (negligible energy barrier <1 kcal/mol). At the same time, Díaz et al. concluded that even though C1 and C2 were structurally and energetically similar, C1 was slightly favored.25,26

3.2.2. Nucleophilic Attack

In the next step of the catalytic mechanism, the nucleophilic attack, the zinc(II)-coordinated hydroxide anion of IM1 makes a nucleophilic attack on the carbon atom of the scissile peptide bond. This leads to the formation of the tetrahedral intermediate IM2, in which the oxygen atom of the zinc(II)-coordinated hydroxide anion binds to the carbon atom of the scissile bond. The distance between Ow1 and C(Gly775) was selected as the reaction coordinate. The reaction proceeds through TS2 with an activation energy barrier of 22.3 kcal/mol (compared to IM1) at the B3 level. It leads to an endothermic IM2 (20.5 kcal/mol) (Figure 5). B1 (B3LYP/def2-SVP/MM) energies were also calculated by incorporating Grimme’s D3 dispersion with BJ damping52 to account for dispersion interactions (Table S2). The energies were quite similar to the one calculated without dispersion correction. The QM/MM calculations were performed on two additional snapshots with B1 barriers of 18.8 and 21.1 kcal/mol, respectively, which is consistent with the original barrier (21.6 kcal/mol). The calculations on the additional snapshots also confirm the nucleophilic attack as an endothermic reaction (reaction energies of 18.0 and 19.6 kcal/mol). The high endothermicity of IM2 prompts further exploration of the lifetime of the intermediate. However, accurate lifetime detection is often complicated, and to our knowledge no experimental data explicitly address this aspect within MMPs.

A 2 Å flexible MM region was used in the study, as the calculations with a larger flexible MM region (e.g., 6 Å) were computationally expensive. However, the size of the optimized (flexible) MM part might be important;67 we, therefore, optimized the stationary points of the nucleophilic attack step only with an extended 6 Å MM region, and the results proved consistent (17.1 kcal/mol activation energy and 15.7 kcal/mol reaction energy) with the calculations with the flexible 2 Å MM region (21.6 kcal/mol activation energy and 18.6 kcal/mol reaction energy). Overall, the 2 and 6 Å MM region calculations are consistent; however, they reassert the importance of accounting conformational relaxation to calculate the activation and reaction energies. To assess the sensitivity of the mechanism to the choice of the density functional, geometry optimizations at M06/def2-SVP/MM and PW6B95/def2-SVP/MM were also carried out, and the results were comparable (Table S3). To evaluate the choice of the size of the QM region, we performed calculations with an extended QM region, additionally including the adjacent residues to the scissile bond (Gln774 and Ala777), which provided the activation energy (22.6 kcal/mol) and reaction energy (18.4 kcal/mol) consistent with the values obtained from the QM/MM calculation with the standard QM region (21.6 and 18.6 kcal/mol respectively). Moreover, the MP2 level of theory has been successfully applied to zinc(II) systems;68,69 therefore, we performed calculations of the rate-determining step using RI-MP2/MM level of theory, which provided activation energies within 3.6 kcal/mol from B3LYP/MM calculations and less endothermic reaction energies (Table S4).

Our calculations identified the nucleophilic attack as the rate-determining step of the overall reaction mechanism. The TS2 has a single imaginary frequency of 187.37i cm–1. The distance between Ow1 and C(Gly775) decreases from 3.09 Å in IM1 to 1.75 Å, and that of the carbonyl bond increases from 1.23 Å in IM1 to 1.31 Å in TS2. In the IM2, the bond between Ow1 and C(Gly775) is formed (1.34 Å), and the carbonyl bond further weakens to 1.34 Å (Figure 6).

Figure 6.

Figure 6

Open in a new tab

Stationary point geometries for the nucleophilic attack. The corresponding distances (Å) are given in the blue box next to each structure. Hydrogen atoms other than the scissile bond ones are hidden for clarity.

Mulliken analysis indicates that TS2 (−0.47) and IM2 (−0.36) have a less negative Mulliken charge on the wat1 Ow1 atom compared to IM1 (−0.56), indicating a charge migration from the nucleophile to the electrophilic carbonyl group. Similarly, the Mulliken charge on the carbonyl carbon of Gly775 increases from +0.12 in IM1 to +0.24 in TS2 to +0.28 in IM2. Previous QM/MM studies of the MMP-2 proteolysis mechanism on different peptide substrates by Díaz et al. reported comparatively lower energy barriers in the 7.0–16.5 kcal/mol range.25,32 Similar studies on MMP-2 by Vasilevskaya et al. and MMP-9 by Chen et al. have shown activation energy barriers of ∼10.0 and 16.0 kcal/mol, respectively, for the nucleophilic attack, also including the simultaneous migration of a proton from wat1 to Glu219 as wat1 attacks the substrate.26,27 The barrier of the rate-determining step can vary as a function of conformational changes, as also shown in other cases.33,70–74 As we have demonstrated, the barrier of the rate-determining step can be lowered by 3.5 kcal/mol in our additional studies with multiple snapshots. Furthermore, our studies include the full-size multidomain MMP-1 enzyme and a full-size THP substrate in contrast to the earlier studies of MMP-2 and MMP-925–27 which include only a small single-chain peptide model, and the conformational dynamics of THP might play an additional role.

3.2.2.1. Interactions Stabilizing the TS

Stabilization of the nucleophilic attack TS was achieved mainly by favorable hydrogen bonding interactions involving the FCS residue His228 and SCS residues Glu219 and Ala777. The backbone nitrogen of His228 maintains a hydrogen bonding interaction with the hydroxyl group of Ser227 (Figure 7). The backbone oxygen and nitrogen atoms of Ala777 of the THP L chain participate in hydrogen bonds with the backbone nitrogen of Tyr240 and the oxygen of Pro238, respectively. DCCA shows that all key SCS residues (Ser227, Tyr240, and Pro238) are positively correlated with each other and the FCS residues including the substrate, Glu219, and the S1′ specificity loop, further supporting their involvement in TS stabilization. Pro238 is a part of the highly conserved “Met-turn” which is important for the stability of the active site. Pro238 and Tyr240 show long-range correlated movements with the S-loop (an S-shaped loop above the active site), implying the role of long-range interactions in stabilizing the TS.

Figure 7.

Figure 7

Open in a new tab

TS stabilizing SCS residues (yellow) of the rate-determining nucleophilic attack. The hydrogen atoms not involved in the interactions are hidden for clarity.

3.2.3. Hydrogen-Bond Rearrangement and Proton Transfer

First, we explored the hydrogen-bond rearrangement and proton transfer using an initial structure IM2, as obtained from the QM/MM calculations of the nucleophilic attack (Figures S9 and S10); however, the QM/MM reaction profile showed a high energy barrier, which ultimately led to an endothermic final product (after the C–N bond breakage). Therefore, we hypothesized that IM2 conformational flexibility might play a role in the hydrogen-bond rearrangement and proton transfer.

3.2.3.1. Conformational Dynamics of the Tetrahedral Intermediate (IM2)

The tetrahedral intermediate plays a central role in zinc(II)-dependent proteolysis; however, its structural flexibility, which might influence the reaction mechanism, is entirely unexplored. To obtain this knowledge, we performed 1 μs MD of IM2 (after MD, IM2 is renamed as IM2′). The flexibility of the hydrogen bond network might require shorter time scales; however, to account for the long-range correlated interactions resulting from complex domain motions at larger scales, longer simulations are required. In addition, the flexibility of the THP substrate can correlate with the dynamics of the protein. Such a network of interactions, in addition, might influence the local interactions in the active site (via long-range correlated motions). The trajectory reveals that the zinc(II)-coordinating histidine residues and Glu219 participate in hydrogen bonding interactions with SCS residues Ser223 and Val215 as in the RC (Figure S5B). Detailed analysis is provided in the Supporting Information (Figure S11 and Pages S14 and S15). The carboxylate oxygens (OE1 and OE2) of Glu219 are at averaged distances of 4.19 and 4.06 Å from wat1 (Figure S12), and the Glu219 CD is at an averaged distance of 5.62 Å from the scissile nitrogen atom of Ile776 (Figure S13). The RDF profile of water molecules showed a reduced density within 4 Å of the CD of Glu219 and an increased density within 5 Å of the scissile nitrogen atom of Ile776 compared to the RC (Figure S4).

Differential DCCA shows changes in the correlated motions in IM2 compared to RC. In particular, an increased number of anticorrelated motions among S-loop residues and helix α1 of the CAT domain, blades II and III of the HPX domain, and correlated motions between both domains are observed in the IM2 (Figure S14A). The VB loop and S1′ specificity loop increase anticorrelated motions with THP L-chain’s scissile bond neighboring residues 777–787. PCA shows that in contrast to the RC, the CAT domain exhibits more intensive motions than the HPX domain, which might be needed for adapting the CAT domain for THP cleavage (Figure S14B).

Before exploring the reaction mechanism of the hydrogen-bond rearrangement and proton transfer, we compared the initial IM2 structure obtained from QM/MM reaction path calculations of the nucleophilic attack with the IM2′ structure produced by the MD simulation to delineate how conformational flexibility affects the IM2 structure. The two structures’ superimposition shows that wat2 and the acetate fragment of Glu219 adopt different orientations in both the IM2 and IM2′ complexes (Figure S15). In the initial IM2, wat2 is stabilized in the active site by hydrogen bonding interactions with Glu219 (2.53 Å), Gln774 (2.01 Å), and wat1 (2.89 Å). In contrast, in the IM2′, the oxygen atom of Glu219, holding the proton, is better oriented for proton transfer, and wat2 is located much closer to the scissile bond residues, thus allowing wat2 to hydrogen bond with the nitrogen atom of Ile776 (2.86 Å) instead of Gln774 in the initial structure. This rearranged hydrogen bonding network in IM2′, as observed in MD, might be critical for creating a more favorable productive conformation of the active site residues involved in the subsequent steps of the catalytic mechanism. Thus, the reaction mechanism of the hydrogen-bond rearrangement and proton transfer can be influenced by the conformational flexibility of IM2, and we assume that IM2 stabilizes after a conformational change; however, further studies are needed to confirm this proposal. We, therefore, explored this reaction step by performing QM/MM calculations on four randomly selected snapshots from the MD trajectory of IM2.

3.2.3.2. Reaction Mechanism of the Hydrogen-Bond Rearrangement and Proton Transfer

Since the MD-derived IM2′ contains a different number of atoms (32,144) than IM2 obtained from QM/MM calculations (27,335), both structures are incompatible for energy comparison. Therefore, we set the energy of the IM2′ for the hydrogen-bond rearrangement and proton transfer to zero, as often done in similar cases.70,75,76 The distances between N(Ile776) and Hw2a, OE2(Glu219) and Hw1, OE1(Glu219) and HE1(Glu219), and Ow2 and Hw2a were chosen as reaction coordinates for hydrogen-bond rearrangement and proton transfer. In three of the calculated reaction paths, the hydrogen-bond rearrangement and proton transfer proceeded in consecutive steps (Figure 8A). The three reaction paths for the hydrogen-bond rearrangement are characterized by low activation energies ranging between 0.1 and 1.4 kcal/mol at the B3 level. Because of the flatness of the potential energy surface, it is difficult to characterize the TS3 vibrational frequency, as also reported in studies on MMP-2.26 The proton transfer occurs after the hydrogen-bond rearrangement and proceeds with B3 activation barriers of 2.3, 4.1, and 4.7 kcal/mol (with respect to IM3). The IM4 is exothermically stabilized by −0.5 to −4.5 kcal/mol. The fourth calculated reaction path shows a concerted process for hydrogen-bond rearrangement and proton transfer with a higher activation energy of 10.9 kcal/mol and an endothermic intermediate of 4.4 kcal/mol, indicating that the concerted pathway is less energetically viable when compared to the consecutive route (Figure 8B). Geometric parameters for snapshots 6594, 4642, and 8234 are shown in the Supporting Information (Tables S5–S7).

Figure 8.

Figure 8

Open in a new tab

Potential energy profile (kcal/mol) for hydrogen-bond rearrangement and proton transfer. (A) Snapshots 5818 (black), 6594 (pink), and 4642 (blue) show consecutive hydrogen-bond rearrangement and proton transfer. (B) Snapshot 8234 (green) shows a concerted mechanism. The relative B3 energies are shown for the four snapshots.

In snapshot 5818, the oxygen of wat2 (Ow2) is hydrogen bonded to the scissile bond nitrogen (Ile776) (Figure 9) and the oxygen of wat1 (Ow1) (2.74 Å) (Scheme 1). The Ow2 is also hydrogen bonded to the protonated carboxylate (OE1) of Glu219 (2.75 Å). IM2′ passes through TS3, forming IM3 in which the interaction between wat1 and wat2 is lost, and the distance between Ow2 and Hw1 increases from 1.94 Å in IM2 to 1.97 Å in the TS3 and finally to 2.70 Å in IM3. In IM3, a new hydrogen bond is formed between Ow1 and the deprotonated carboxylate oxygen (OE2) of Glu219 (the distance between OE2 and Hw1 decreases from 2.67 Å in IM2′ to 1.92 Å in IM3). The IM2′ is present in 87.7% of the snapshots, and IM3 is present in 12.3% of the snapshots from the MD of the tetrahedral intermediate.

Figure 9.

Figure 9

Open in a new tab

Stationary point geometries of the hydrogen-bond rearrangement of the lowest barrier snapshot (5818). The corresponding distances (Å) are given in the blue box next to each structure. Hydrogen atoms of the substrate other than the scissile bond residues are hidden for clarity.

From IM3 starts the proton transfer process from Glu219 to the scissile nitrogen atom of Ile776 through wat2 (Scheme 1). The process proceeds via TS4, where the hydrogen bond between the hydrogen of wat2 (Hw2a) and the nitrogen of Ile776 (1.40) is shorter compared to IM3 (1.88 Å). As Hw2a is being transferred to the nitrogen (N) of Ile776, Ow2 abstracts a proton from the carboxylate oxygen (OE1) of Glu219. The TS4 shows the partial formation of this bond (1.60 Å in IM3 to 1.15 Å in TS4). In the IM4, the proton is completely transferred to the nitrogen of Ile776 (1.05 Å) and separated from the Ow2 of wat2 by 1.76 Å, and the proton from OE1 is completely transferred to the Ow2 (1.00 Å), thus resulting in deprotonated Glu219 (Figure 10).

Figure 10.

Figure 10

Open in a new tab

Stationary point geometries of proton transfer (snapshot 5818). The corresponding distances (Å) are given in the blue boxes next to each structure. Hydrogen atoms of the substrate other than the scissile bond residues are hidden for clarity.

The less negative Mulliken charge on the nitrogen atom (N) of Ile776 in the IM4 (−0.1) compared to that of TS4 (−0.27) and IM3 (−0.27) appears to be due to the reception of the hydrogen atom from wat2 and the subsequent reduction in the electron density. Computational studies on MMP-2 and MMP-9 revealed varying activation energies for hydrogen-bond rearrangement and proton transfer. Hydrogen-bond rearrangement was assigned as the rate-determining step in MMP-9 proteolysis with a comparatively higher energy barrier of 11.27 kcal/mol.27 Díaz et al. predicted the proton transfer as the rate-determining step.25 Vasilevskaya et al. and Díaz et al. reported activation energies of 0.6 and 5.7 kcal/mol for hydrogen-bond rearrangement and 0.2 and 2.2 kcal/mol for proton transfer by MMP-2.25,26 Both the hydrogen-bond rearrangement and proton-transfer reactions were reported as exothermic by Chen et al. (reaction energies ∼ −23.34 and −7.94 kcal/mol) and Vasilevskaya et al. (reaction energies ∼ −2.5 and −5.3 kcal/mol), whereas Díaz et al. suggested an endothermic hydrogen-bond rearrangement (reaction energy, 12.98 kcal/mol) followed by exothermic proton transfer (reaction energy, −4.94 kcal/mol). QM studies of MMP-3 by Pelmenschikov and Siegbahn demonstrated an endothermic (reaction energy, 6.5 kcal/mol) concerted nucleophilic attack and proton transfer mechanism with a rate-determining barrier of 22.8 kcal/mol.31

In the IM2′ of snapshot 8234, wat2 participates as a donor of hydrogen bond with wat1, whereas wat2 switches to a hydrogen bond acceptor in the IM2′ of snapshot 5818 (Figure 11). This results in the orientation of wat2 away from the nitrogen atom of Ile776 in snapshot 8234, as observed by the increased distance between Hw2b and N(Ile776) (2.46 Å), whereas in snapshot 5818 the distance is 1.90 Å. Because the Hw1 of wat1 is not involved in other hydrogen bonding interactions in snapshot 8234, it is in a favorable orientation for hydrogen bond with the OE2 of Glu219 (OE2–Hw1 = 2.20 Å and < OE2–Hw1–Ow1 = 150.85°), which is a prime hydrogen bonding interaction that takes place during the hydrogen-bond rearrangement. A weaker interaction was observed in snapshot 5818 (OE2–Hw1 = 2.67 Å and < OE2–Hw1–Ow1 = 127.79°). Overall, the wat1 and wat2 molecules in snapshot 8234 are in a relatively better orientation for a rapid hydrogen-bond rearrangement, which might have led to a concerted hydrogen-bond rearrangement and proton-transfer path; however, this concerted process led to a higher activation barrier.

Figure 11.

Figure 11

Open in a new tab

Tetrahedral intermediates (IM2′) from (A) snapshot 5818 (representing snapshots 5818, 6594, and 4642) and (B) snapshot 8234, demonstrating the hydrogen bonding interactions in the catalytic site leading to the consecutive and concerted hydrogen-bond rearrangement/proton-transfer reaction step. The hydrogens of the carbon atoms are hidden for clarity.

The consecutive hydrogen-bond rearrangement and proton-transfer TSs (snapshots 5818, 6594, and 4642) are stabilized by an extended network of hydrogen bonds with residues in the SCS. For example, hydrogen bonds between the His218 side chain nitrogen and the backbone oxygen of Leu235, the backbone nitrogen of His228 and the hydroxyl oxygen of Ser227, the backbone nitrogen of His222 and the backbone oxygen of Glu219, the backbone oxygen of His222 and the backbone nitrogen atom of Gly225, the backbone nitrogen of THP Ala777 and the backbone oxygen atom of Pro238, and the backbone oxygen of THP Gln774 and the backbone nitrogen atom of Ala184 (Figure 12A,B) further facilitate the consecutive hydrogen-bond rearrangement and proton-transfer processes. In the concerted reaction (snapshot 8234), Ala777 lacks stabilization by Pro238, and the backbone nitrogen of His218 is additionally stabilized by the hydrogen bonding interaction with the backbone oxygen atom of Arg214. Such differences in the SCS interactions might contribute to the preferential consecutive hydrogen-bond rearrangement and proton transfer. Similar to the observed correlation pattern of the TS stabilizing residues in the rate-determining step, the SCS residues demonstrate correlated motions with each other and the FCS residues, including the substrate and the S1′ specificity loop. Ala184 and Pro238 show correlated motions with structural zinc(II) and the S-loop.

Figure 12.

Figure 12

Open in a new tab

TS stabilizing residues for the (A) consecutive hydrogen-bond rearrangement and proton transfer in snapshots 5818, 6594, and 4642 and (B) concerted mechanism in snapshot 8234 of the IM2'.

Notably, the calculations imply that after IM2 formation, a conformational change occurs that rearranges IM2 for productive hydrogen-bond rearrangement and proton transfer. Furthermore, the QM/MM calculations using different IM2′ structures from MD simulations show that the hydrogen-bond rearrangement and proton-transfer processes can occur in a consecutive or concerted manner, with the consecutive mechanism being energetically more preferable. In addition, tunneling can further contribute to lower the barriers.77,78

3.2.4. C–N Bond Cleavage

The final step of the catalytic mechanism is scissile C–N bond cleavage, resulting in the product complex (P) formation. The distance between C(Gly775) and N(Ile776) was the reaction coordinate. The C–N bond cleavage is triggered upon protonation of the scissile nitrogen atom. The reaction proceeds barrierlessly, producing a highly exothermic product (−16.2 kcal/mol). The IM4 showed a weak C–N bond with a bond distance of 1.69 Å (Figure 13). At P, the C–N bond is completely cleaved (3.07 Å), significantly strengthening the C–Ow1 double bond character (1.27 Å). The increase of the C–Ow1 double bond character is combined with the transfer of Hw1 to the OE2 of Glu219, as confirmed by the 1.02 Å distance between the Hw1 and OE2 in P, with respect to 1.48 Å in IM4.

Figure 13.

Figure 13

Open in a new tab

Stationary point geometries of the C–N bond cleavage. The corresponding distances (Å) are given in the blue boxes next to each structure. Hydrogen atoms of the substrate other than the scissile residues are hidden for clarity.

In the final product complex P, the nitrogen atom of the N-terminal product makes a hydrogen bond with wat2 (3.00 Å), and the carboxyl group oxygen atoms of the C-terminal product coordinates zinc(II) with Zn–O distances of 2.03 and 2.32 Å, respectively. The Mulliken charge of the nitrogen atom (N) of Ile776 is more negative in P (−0.21) than in IM4 (−0.11), and the charge on Ow1 is also more negative in P (−0.44) compared to IM4 (−0.40). However, the charge on the carbon atom (C) of Gly775 (0.23 in IM4) turns slightly less positive (0.21) in P. Our calculated activation energy for this step agrees well with prior mechanistic studies on MMP-2, MMP-3, and MMP-9 that have reported very low energy barriers for C–N bond cleavage in the range of 0.5–5 kcal/mol. Moreover, these studies also indicate that an exothermic reaction drives the C–N cleavage (reaction energies, −1 to −50 kcal/mol).25–27,31

4. Conclusions

Despite the availability of extensive experimental data, the mechanism of MMP-1-catalyzed collagenolysis remained poorly understood due to the lack of an experimental structure of a catalytically productive ES complex of MMP-1. The present study capitalized on a computationally generated catalytically competent MMP-1·THP complex of multidomain MMP-1 with the large and conformationally flexible THP substrate.

The calculated mechanism involves proton transfer followed by nucleophilic attack to scissile bond carbonyl carbon, forming a tetrahedral intermediate. Afterward, a hydrogen-bond rearrangement facilitates proton transfer from wat2 to the scissile bond amide nitrogen, resulting in C–N bond cleavage. Notably, the study revealed the critical role of a second water molecule (wat2) in the catalytic site and the critical role of the conformational dynamics of the tetrahedral intermediate for proper orientation of the hydrogen bonding network in the reaction mechanism. The calculations show the water-mediated nucleophilic attack with an activation energy barrier of 22.3 kcal/mol as the rate-determining step, in agreement with the experimental kinetic data and earlier computational studies of MMP-2 and MMP-9.

The study underlines the essential role of the interactions with the SCS residues Ala184, Gly225, Ser227, Leu235, Pro238, and Tyr240 in stabilizing the TSs along the reaction path. Furthermore, the results illustrate the important role of the correlated motions of SCS residues with long-range partnering residues, such as the S1′ specificity loop, the VB loop of the CAT domain and blades of the HPX domain.

Overall, the study revealed the catalytic mechanism of MMP-1 including in the model the complete enzyme and the full-length THP, revealing the key roles of a second water molecule, the conformational dynamics, and the interactions with the SCS, thus providing a mechanistic background for the design of more effective MMP-1 inhibitors.

Acknowledgments

This study was supported by the NIH grant 2R15GM132873-02 to TKC and GBF.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jpcb.3c04293.

  • Additional experimental details, including the computational methodology, MD analysis, QM/MM reaction path analysis, and Cartesian coordinates (PDF)

The authors declare no competing financial interest.

Supplementary Material

jp3c04293_si_001.pdf (1.5MB, pdf)

References

  1. Wezynfeld N. E.; Frączyk T.; Bal W. Metal Assisted Peptide Bond Hydrolysis: Chemistry, Biotechnology and Toxicological Implications. Coord. Chem. Rev. 2016, 327–328, 166–187. 10.1016/j.ccr.2016.02.009. [DOI] [Google Scholar]
  2. Grant K.; Kassai M. Major Advances in the Hydrolysis of Peptides and Proteins by Metal Ions and Complexes. Curr. Org. Chem. 2006, 10 (9), 1035–1049. 10.2174/138527206777435535. [DOI] [Google Scholar]
  3. Parkin G. Synthetic Analogues Relevant to the Structure and Function of Zinc Enzymes. Chem. Rev. 2004, 104 (2), 699–768. 10.1021/cr0206263. [DOI] [PubMed] [Google Scholar]
  4. Zastrow M. L.; Pecoraro V. L. Designing Hydrolytic Zinc Metalloenzymes. Biochemistry 2014, 53 (6), 957–978. 10.1021/bi4016617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Borel J. P.; Monboisse J. C.. Collagenolysis. Handbook Methods For Oxygen Radical Research, 1st ed.; CRC Press, 1985; p 6. [Google Scholar]
  6. Fields G. B. Interstitial Collagen Catabolism. J. Biol. Chem. 2013, 288 (13), 8785–8793. 10.1074/jbc.R113.451211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Amar S.; Smith L.; Fields G. B. Matrix Metalloproteinase Collagenolysis in Health and Disease. Biochim. Biophys. Acta, Mol. Cell Res. 2017, 1864 (11), 1940–1951. 10.1016/j.bbamcr.2017.04.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Birkedal-Hansen H.; Moore W. G. I.; Bodden M. K.; Windsor L. J.; Birkedal-Hansen B.; DeCarlo A.; Engler J. A. Matrix Metalloproteinases: A Review. Crit. Rev. Oral Biol. Med. 1993, 4 (2), 197–250. 10.1177/10454411930040020401. [DOI] [PubMed] [Google Scholar]
  9. Arakaki P. A.; Marques M. R.; Santos M. C. L. G. MMP-1 Polymorphism and its Relationship to Pathological Processes. J. Biosci. 2009, 34 (2), 313–320. 10.1007/s12038-009-0035-1. [DOI] [PubMed] [Google Scholar]
  10. Xu X.; Lu X.; Chen L.; Peng K.; Ji F. Downregulation of MMP1 Functions in Preventing Perineural Invasion of Pancreatic Cancer through Blocking the NT −3/TrkC Signaling Pathway. Clin. Lab. Anal. 2022, 36 (11), e24719 10.1002/jcla.24719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Leite M. F. F.; Santos M. C. L. G.; de Souza A. P.; Line S. R. P. Osseointegrated Implant Failure Associated with MMP-1 Promotor Polymorphisms (−1607 and −519). Int. J. Oral Maxillofac. Implants 2008, 23 (4), 653–658. [PubMed] [Google Scholar]
  12. Peng Z.; Konai M. M.; Avila-Cobian L. F.; Wang M.; Mobashery S.; Chang M. MMP-1 and ADAM10 as Targets for Therapeutic Intervention in Idiopathic Pulmonary Fibrosis. ACS Pharmacol. Transl. Sci. 2022, 5 (8), 548–554. 10.1021/acsptsci.2c00050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Pardo A.; Selman M. MMP-1: The Elder of the Family. Int. J. Biochem. Cell Biol. 2005, 37 (2), 283–288. 10.1016/j.biocel.2004.06.017. [DOI] [PubMed] [Google Scholar]
  14. Hideaki N.; Woessner J.. Matrix Metalloproteinases. Zinc Metalloproteases In Health And Disease, 1st ed.; CRC Press, 1996; p 52. [Google Scholar]
  15. Lauer-Fields J. L.; Chalmers M. J.; Busby S. A.; Minond D.; Griffin P. R.; Fields G. B. Identification of Specific Hemopexin-like Domain Residues that Facilitate Matrix Metalloproteinase Collagenolytic Activity. J. Biol. Chem. 2009, 284 (36), 24017–24024. 10.1074/jbc.M109.016873. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Arnold L. H.; Butt L. E.; Prior S. H.; Read C. M.; Fields G. B.; Pickford A. R. The Interface between Catalytic and Hemopexin Domains in Matrix Metalloproteinase-1 Conceals a Collagen Binding Exosite. J. Biol. Chem. 2011, 286 (52), 45073–45082. 10.1074/jbc.M111.285213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Murphy G.; Nagase H. Progress in Matrix Metalloproteinase Research. Mol. Aspects Med. 2008, 29 (5), 290–308. 10.1016/j.mam.2008.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Whittaker M.; Floyd C. D.; Brown P.; Gearing A. J. H. Design and Therapeutic Application of Matrix Metalloproteinase Inhibitors. (Chem. Rev. 1999, 99, 2735–2776. Published on the Web September 8, 1999). Chem. Rev. 2001, 101 (7), 2205–2206. 10.1021/cr0100345. [DOI] [PubMed] [Google Scholar]
  19. Yang H.; Makaroff K.; Paz N.; Aitha M.; Crowder M. W.; Tierney D. L. Metal Ion Dependence of the Matrix Metalloproteinase-1 Mechanism. Biochemistry 2015, 54 (23), 3631–3639. 10.1021/acs.biochem.5b00379. [DOI] [PubMed] [Google Scholar]
  20. Wetmore D. R.; Hardman K. D. Roles of the Propeptide and Metal Ions in the Folding and Stability of the Catalytic Domain of Stromelysin (Matrix Metalloproteinase 3). Biochemistry 1996, 35 (21), 6549–6558. 10.1021/bi9530752. [DOI] [PubMed] [Google Scholar]
  21. Díaz N.; Suarez D. Molecular Dynamics Simulations of Matrix Metalloproteinase 2: Role of the Structural Metal Ions. Biochemistry 2007, 46 (31), 8943–8952. 10.1021/bi700541p. [DOI] [PubMed] [Google Scholar]
  22. Tallant C.; Marrero A.; Gomis-Rüth F. Matrix Metalloproteinases: Fold and Function of their Catalytic Domains. Biochim. Biophys. Acta, Mol. Cell Res. 2010, 1803 (1), 20–28. 10.1016/j.bbamcr.2009.04.003. [DOI] [PubMed] [Google Scholar]
  23. Bertini I.; Calderone V.; Fragai M.; Luchinat C.; Maletta M.; Yeo K. J. Snapshots of the Reaction Mechanism of Matrix Metalloproteinases. Angew. Chem., Int. Ed. 2006, 45 (47), 7952–7955. 10.1002/anie.200603100. [DOI] [PubMed] [Google Scholar]
  24. Crabbe T.; Zucker S.; Cockett M. I.; Willenbrock F.; Tickle S.; O’Connell J. P.; Scothern J. M.; Murphy G.; Docherty A. J. P. Mutation of the Active Site Glutamic Acid of Human Gelatinase A: Effects on Latency, Catalysis, and the Binding of Tissue Inhibitor of Metalloproteinases-1. Biochemistry 1994, 33 (21), 6684–6690. 10.1021/bi00187a039. [DOI] [PubMed] [Google Scholar]
  25. Díaz D. N.; Suárez D. D. Peptide Hydrolysis Catalyzed by Matrix Metalloproteinase 2: A Computational Study. J. Phys. Chem. B 2008, 112 (28), 8412–8424. 10.1021/jp803509h. [DOI] [PubMed] [Google Scholar]
  26. Vasilevskaya T.; Khrenova M. G.; Nemukhin A. V.; Thiel W. Mechanism of Proteolysis in Matrix Metalloproteinase-2 Revealed by QM/MM Modeling. J. Comput. Chem. 2015, 36 (21), 1621–1630. 10.1002/jcc.23977. [DOI] [PubMed] [Google Scholar]
  27. Chen B.; Kang Z.; Zheng E.; Liu Y.; Gauld J. W.; Wang Q. Hydrolysis Mechanism of the Linkers by Matrix Metalloproteinase-9 Using QM/MM Calculations. J. Chem. Inf. Model. 2021, 61 (10), 5203–5211. 10.1021/acs.jcim.1c00825. [DOI] [PubMed] [Google Scholar]
  28. Elsässer B.; Goettig P. Mechanisms of Proteolytic Enzymes and their Inhibition in QM/MM Studies. Int. J. Mol. Sci. 2021, 22 (6), 3232. 10.3390/ijms22063232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Grossman M.; Born B.; Heyden M.; Tworowski D.; Fields G. B.; Sagi I.; Havenith M. Correlated Structural Kinetics and Retarded Solvent Dynamics at the Metalloprotease Active Site. Nat. Struct. Mol. Biol. 2011, 18 (10), 1102–1108. 10.1038/nsmb.2120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lauer-Fields J. L.; Tuzinski K. A.; Shimokawa K.; Nagase H.; Fields G. B. Hydrolysis of Triple-Helical Collagen Peptide Models by Matrix Metalloproteinases. J. Biol. Chem. 2000, 275 (18), 13282–13290. 10.1074/jbc.275.18.13282. [DOI] [PubMed] [Google Scholar]
  31. Pelmenschikov V.; Siegbahn P. E. M. Catalytic Mechanism of Matrix Metalloproteinases: Two-Layered ONIOM Study. Inorg. Chem. 2002, 41 (22), 5659–5666. 10.1021/ic0255656. [DOI] [PubMed] [Google Scholar]
  32. Díaz N.; Suárez D.; Suárez E. Kinetic and Binding Effects in Peptide Substrate Selectivity of Matrix Metalloproteinase-2: Molecular Dynamics and QM/MM Calculations. Proteins 2010, 78 (1), 1–11. 10.1002/prot.22493. [DOI] [PubMed] [Google Scholar]
  33. Castro-Amorim J.; Oliveira A.; Mukherjee A. K.; Ramos M. J.; Fernandes P. A. Unraveling the Reaction Mechanism of Russell’s Viper Venom Factor X Activator: A Paradigm for the Reactivity of Zinc Metalloproteinases?. J. Chem. Inf. Model. 2023, 63, 4056–4069. 10.1021/acs.jcim.2c01156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Manka S. W.; Carafoli F.; Visse R.; Bihan D.; Raynal N.; Farndale R. W.; Murphy G.; Enghild J. J.; Hohenester E.; Nagase H. Structural Insights into Triple-Helical Collagen Cleavage by Matrix Metalloproteinase 1. Proc. Natl. Acad. Sci. U.S.A. 2012, 109 (31), 12461–12466. 10.1073/pnas.1204991109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Bertini I.; Fragai M.; Luchinat C.; Melikian M.; Toccafondi M.; Lauer J. L.; Fields G. B. Structural Basis for Matrix Metalloproteinase 1-Catalyzed Collagenolysis. J. Am. Chem. Soc. 2012, 134 (4), 2100–2110. 10.1021/ja208338j. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Cerofolini L.; Fields G. B.; Fragai M.; Geraldes C. F. G. C.; Luchinat C.; Parigi G.; Ravera E.; Svergun D. I.; Teixeira J. M. C. Examination of Matrix Metalloproteinase-1 in Solution. J. Biol. Chem. 2013, 288 (42), 30659–30671. 10.1074/jbc.M113.477240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Waheed S. O.; Varghese A.; DiCastri I.; Kaski B.; LaRouche C.; Fields G. B.; Karabencheva-Christova T. G. Mechanism of the Early Catalytic Events in the Collagenolysis by Matrix Metalloproteinase-1. ChemPhysChem 2023, 24, e202200649 10.1002/cphc.202200649. [DOI] [PubMed] [Google Scholar]
  38. Bullock R. M.; Dey A. Introduction: Catalysis beyond the First Coordination Sphere. Chem. Rev. 2022, 122 (14), 11897–11899. 10.1021/acs.chemrev.2c00428. [DOI] [PubMed] [Google Scholar]
  39. Van Stappen C.; Deng Y.; Liu Y.; Heidari H.; Wang J.-X.; Zhou Y.; Ledray A. P.; Lu Y. Designing Artificial Metalloenzymes by Tuning of the Environment beyond the Primary Coordination Sphere. Chem. Rev. 2022, 122 (14), 11974–12045. 10.1021/acs.chemrev.2c00106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Reek J. N. H.; de Bruin B.; Pullen S.; Mooibroek T. J.; Kluwer A. M.; Caumes X. Transition Metal Catalysis Controlled by Hydrogen Bonding in the Second Coordination Sphere. Chem. Rev. 2022, 122 (14), 12308–12369. 10.1021/acs.chemrev.1c00862. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Khare S. D.; Kipnis Y.; Greisen P. J.; Takeuchi R.; Ashani Y.; Goldsmith M.; Song Y.; Gallaher J. L.; Silman I.; Leader H.; Sussman J. L.; Stoddard B. L.; Tawfik D. S.; Baker D. Computational Redesign of a Mononuclear Zinc Metalloenzyme for Organophosphate Hydrolysis. Nat. Chem. Biol. 2012, 8 (3), 294–300. 10.1038/nchembio.777. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Singh M. K.; Chu Z. T.; Warshel A. Simulating the Catalytic Effect of a Designed Mononuclear Zinc Metalloenzyme that Catalyzes the Hydrolysis of Phosphate Triesters. J. Phys. Chem. B 2014, 118 (42), 12146–12152. 10.1021/jp507592g. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Karabencheva-Christova T. G.; Christov C. Z.; Fields G. B. Conformational Dynamics of Matrix Metalloproteinase-1·Triple-Helical Peptide Complexes. J. Phys. Chem. B 2018, 122 (21), 5316–5326. 10.1021/acs.jpcb.7b09771. [DOI] [PubMed] [Google Scholar]
  44. Metz S.; Kästner J.; Sokol A. A.; Keal T. W.; Sherwood P. ChemShell—A Modular Software Package for QM/MM Simulations. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2014, 4 (2), 101–110. 10.1002/wcms.1163. [DOI] [Google Scholar]
  45. Ahlrichs R.; Bär M.; Häser M.; Horn H.; Kölmel C. Electronic Structure Calculations on Workstation Computers: The Program System Turbomole. Chem. Phys. Lett. 1989, 162 (3), 165–169. 10.1016/0009-2614(89)85118-8. [DOI] [Google Scholar]
  46. Smith W.; Yong C. W.; Rodger P. M. DL_POLY: Application to Molecular Simulation. Mol. Simul. 2002, 28 (5), 385–471. 10.1080/08927020290018769. [DOI] [Google Scholar]
  47. Kohn W.; Sham L. J. Self-Consistent Equations Including Exchange and Correlation Effects. Phys. Rev. 1965, 140 (4A), A1133–A1138. 10.1103/PhysRev.140.A1133. [DOI] [Google Scholar]
  48. Ziegler T. Approximate Density Functional Theory as a Practical Tool in Molecular Energetics and Dynamics. Chem. Rev. 1991, 91 (5), 651–667. 10.1021/cr00005a001. [DOI] [Google Scholar]
  49. Becke A. D. Density-Functional Exchange-Energy Approximation with Correct Asymptotic Behavior. Phys. Rev. A 1988, 38 (6), 3098–3100. 10.1103/PhysRevA.38.3098. [DOI] [PubMed] [Google Scholar]
  50. Lee C.; Yang W.; Parr R. G. Development of the Colle-Salvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B 1988, 37 (2), 785–789. 10.1103/PhysRevB.37.785. [DOI] [PubMed] [Google Scholar]
  51. Becke A. D. Density-Functional Thermochemistry. III. The Role of Exact Exchange. J. Chem. Phys. 1993, 98 (7), 5648–5652. 10.1063/1.464913. [DOI] [Google Scholar]
  52. Grimme S.; Antony J.; Ehrlich S.; Krieg H. A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements H-Pu. J. Chem. Phys. 2010, 132 (15), 154104. 10.1063/1.3382344. [DOI] [PubMed] [Google Scholar]
  53. Shaik S.; Cohen S.; Wang Y.; Chen H.; Kumar D.; Thiel W. P450 Enzymes: Their Structure, Reactivity, and Selectivity—Modeled by QM/MM Calculations. Chem. Rev. 2010, 110 (2), 949–1017. 10.1021/cr900121s. [DOI] [PubMed] [Google Scholar]
  54. Dutta Dubey K.; Stuyver T.; Kalita S.; Shaik S. Solvent Organization and Rate Regulation of a Menshutkin Reaction by Oriented External Electric Fields are Revealed by Combined MD and QM/MM Calculations. J. Am. Chem. Soc. 2020, 142 (22), 9955–9965. 10.1021/jacs.9b13029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Ramanan R.; Chaturvedi S. S.; Lehnert N.; Schofield C. J.; Karabencheva-Christova T. G.; Christov C. Z. Catalysis by the JmjC Histone Demethylase KDM4A Integrates Substrate Dynamics, Correlated Motions and Molecular Orbital Control. Chem. Sci. 2020, 11 (36), 9950–9961. 10.1039/D0SC03713C. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. De Vivo M.; Masetti M.; Bottegoni G.; Cavalli A. Role of Molecular Dynamics and Related Methods in Drug Discovery. J. Med. Chem. 2016, 59 (9), 4035–4061. 10.1021/acs.jmedchem.5b01684. [DOI] [PubMed] [Google Scholar]
  57. Sutto L.; Marsili S.; Gervasio F. L. New Advances in Metadynamics: New Advances in Metadynamics. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2 (5), 771–779. 10.1002/wcms.1103. [DOI] [Google Scholar]
  58. Barducci A.; Bussi G.; Parrinello M. Well-Tempered Metadynamics: A Smoothly Converging and Tunable Free-Energy Method. Phys. Rev. Lett. 2008, 100 (2), 020603. 10.1103/PhysRevLett.100.020603. [DOI] [PubMed] [Google Scholar]
  59. Brunk E.; Rothlisberger U. Mixed Quantum Mechanical/Molecular Mechanical Molecular Dynamics Simulations of Biological Systems in Ground and Electronically Excited States. Chem. Rev. 2015, 115 (12), 6217–6263. 10.1021/cr500628b. [DOI] [PubMed] [Google Scholar]
  60. Raich L.; Nin-Hill A.; Ardèvol A.; Rovira C.. Enzymatic Cleavage of Glycosidic Bonds. Methods in Enzymology; Elsevier, 2016; Vol. 577, pp 159–183. 10.1016/bs.mie.2016.05.015. [DOI] [PubMed] [Google Scholar]
  61. Lodola A.; Mor M.; Zurek J.; Tarzia G.; Piomelli D.; Harvey J. N.; Mulholland A. J. Conformational Effects in Enzyme Catalysis: Reaction via a High Energy Conformation in Fatty Acid Amide Hydrolase. Biophys. J. 2007, 92 (2), L20–L22. 10.1529/biophysj.106.098434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Ranaghan K. E.; Mulholland A. J. Conformational Effects in Enzyme Catalysis: QM/MM Free Energy Calculation of the ‘NAC’ Contribution in Chorismate Mutase. Chem. Commun. 2004, (10), 1238–1239. 10.1039/B402388A. [DOI] [PubMed] [Google Scholar]
  63. Chung L.; Shimokawa K.; Dinakarpandian D.; Grams F.; Fields G. B.; Nagase H. Identification of the (183)RWTNNFREY(191) Region as a Critical Segment of Matrix Metalloproteinase 1 for the Expression of Collagenolytic Activity. J. Biol. Chem. 2000, 275 (38), 29610–29617. 10.1074/jbc.M004039200. [DOI] [PubMed] [Google Scholar]
  64. Lukacova V.; Zhang Y.; Mackov M.; Baricic P.; Raha S.; Calvo J. A.; Balaz S. Similarity of Binding Sites of Human Matrix Metalloproteinases. J. Biol. Chem. 2004, 279 (14), 14194–14200. 10.1074/jbc.M313474200. [DOI] [PubMed] [Google Scholar]
  65. Feliciano G. T.; da Silva A. J. R. Unravelling the Reaction Mechanism of Matrix Metalloproteinase 3 Using QM/MM Calculations. J. Mol. Struct. 2015, 1091, 125–132. 10.1016/j.molstruc.2015.02.079. [DOI] [Google Scholar]
  66. Decaneto E.; Vasilevskaya T.; Kutin Y.; Ogata H.; Grossman M.; Sagi I.; Havenith M.; Lubitz W.; Thiel W.; Cox N. Solvent Water Interactions within the Active Site of the Membrane Type I Matrix Metalloproteinase. Phys. Chem. Chem. Phys. 2017, 19 (45), 30316–30331. 10.1039/C7CP05572B. [DOI] [PubMed] [Google Scholar]
  67. Calixto A. R.; Ramos M. J.; Fernandes P. A. Influence of Frozen Residues on the Exploration of the PES of Enzyme Reaction Mechanisms. J. Chem. Theory Comput. 2017, 13 (11), 5486–5495. 10.1021/acs.jctc.7b00768. [DOI] [PubMed] [Google Scholar]
  68. Navrátil V.; Klusák V.; Rulíšek L. Theoretical Aspects of Hydrolysis of Peptide Bonds by Zinc Metalloenzymes. Chem.—Eur. J. 2013, 19 (49), 16634–16645. 10.1002/chem.201302663. [DOI] [PubMed] [Google Scholar]
  69. Bauzá A.; Quiñonero D.; Deyà P. M.; Frontera A. Long-Range Effects in Anion-π Interactions: Their Crucial Role in the Inhibition Mechanism of Mycobacterium Tuberculosis Malate Synthase. Chem.—Eur. J. 2014, 20 (23), 6985–6990. 10.1002/chem.201304995. [DOI] [PubMed] [Google Scholar]
  70. Waheed S. O.; Ramanan R.; Chaturvedi S. S.; Lehnert N.; Schofield C. J.; Christov C. Z.; Karabencheva-Christova T. G. Role of Structural Dynamics in Selectivity and Mechanism of Non-Heme Fe(II) and 2-Oxoglutarate-Dependent Oxygenases Involved in DNA Repair. ACS Cent. Sci. 2020, 6 (5), 795–814. 10.1021/acscentsci.0c00312. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Claeyssens F.; Ranaghan K. E.; Lawan N.; Macrae S. J.; Manby F. R.; Harvey J. N.; Mulholland A. J. Analysis of Chorismate Mutase Catalysis by QM/MM Modelling of Enzyme-Catalysed and Uncatalysed Reactions. Org. Biomol. Chem. 2011, 9 (5), 1578. 10.1039/c0ob00691b. [DOI] [PubMed] [Google Scholar]
  72. Rifayee S. B. J. S.; Chaturvedi S. S.; Warner C.; Wildey J.; White W.; Thompson M.; Schofield C. J.; Christov C. Catalysis by KDM6 Histone Demethylases—A Synergy between the Non-Heme Iron(II) Center, Second Coordination Sphere, and Long-Range Interactions. Chem.—Eur. J. 2023, 29, e202301305 10.1002/chem.202301305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Christov C. Z.; Lodola A.; Karabencheva-Christova T. G.; Wan S.; Coveney P. V.; Mulholland A. J. Conformational Effects on the pro—S Hydrogen Abstraction Reaction in Cyclooxygenase-1: An Integrated QM/MM and MD Study. Biophys. J. 2013, 104 (5), L5–L7. 10.1016/j.bpj.2013.01.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Álvarez-Barcia S.; Kästner J. Atom Tunneling in the Hydroxylation Process of Taurine/α-Ketoglutarate Dioxygenase Identified by Quantum Mechanics/Molecular Mechanics Simulations. J. Phys. Chem. B 2017, 121 (21), 5347–5354. 10.1021/acs.jpcb.7b03477. [DOI] [PubMed] [Google Scholar]
  75. Ramanan R.; Waheed S. O.; Schofield C. J.; Christov C. Z. What Is the Catalytic Mechanism of Enzymatic Histone N-Methyl Arginine Demethylation and Can It Be Influenced by an External Electric Field?. Chem.—Eur. J. 2021, 27 (46), 11827–11836. 10.1002/chem.202101174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Waheed S. O.; Varghese A.; Chaturvedi S. S.; Karabencheva-Christova T. G.; Christov C. Z. How Human TET2 Enzyme Catalyzes the Oxidation of Unnatural Cytosine Modifications in Double-Stranded DNA. ACS Catal. 2022, 12 (9), 5327–5344. 10.1021/acscatal.2c00024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Krishtalik L. I. The Mechanism of the Proton Transfer: An Outline. Biochim. Biophys. Acta, Bioenerg. 2000, 1458 (1), 6–27. 10.1016/S0005-2728(00)00057-8. [DOI] [PubMed] [Google Scholar]
  78. Cukier R. I.; Nocera D. G. Proton-Coupled Electron Transfer. Annu. Rev. Phys. Chem. 1998, 49 (1), 337–369. 10.1146/annurev.physchem.49.1.337. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

jp3c04293_si_001.pdf (1.5MB, pdf)

Articles from The Journal of Physical Chemistry. B are provided here courtesy of American Chemical Society

RESOURCES