Abstract
The increasing use of methacrylate-based materials in tissue engineering and dental restorations demands detailed evaluation of enzymolysis of these materials due to toxicity, durability, and biocompatibility concerns. The objective of this study is to develop tools for assessing and ranking the enzymolysis kinetics of dimethacrylate (DMA) compounds. Triethyleneglycol DMA and diurethane DMA are employed as model DMAs for kinetic studies of 2-step enzymolysis by 2 esterases, pseudocholine esterase and cholesterol esterase. In addition, the intermediate hydrolysis products, mono-methacrylates (mono-MAs), are prepared via esterases. The kinetics of DMA enzymolysis are evaluated per the concentrations of DMA. The enzymolysis products are quantified by high-performance liquid chromatography. Additionally, stoichiometric analysis and a Berkeley Madonna model are employed to compare the efficacy of esterases in DMA enzymolysis. The chemical structure of mono-MAs is verified by proton and heteronuclear single quantum coherence (2D 1H-13C) nuclear magnetic resonance spectroscopy and mass spectrometry. In evaluating the ratio of sequential and simultaneous degradations of DMA and mono-MA, the stoichiometric analysis draws the same conclusions without using [mono-MA] as the experimental observation using [mono-MA]. The majority of the 4 esterase-DMA combinations undergo the sequential enzyme-catalyzed hydrolysis, from DMA to mono-MA to diol. However, cholesterol esterase is more effective than pseudocholine esterase in maintaining sequential degradation until >90% of DMA is decomposed. Both enzymolysis steps are first-order reactions. The mono-MAs are more hydrolysis resistant than DMAs. Moreover, esterase efficacy and selectivity on DMA enzymolysis are presented. The stoichiometric analysis provides valuable tools in assessing DMA enzymolysis when mono-MA is difficult to be obtained. The resistance of mono-MAs to enzymolysis suggests a need for thorough toxicity evaluations of these intermediate compounds. It also advocates the alternative approaches in designing and developing durable and biocompatible materials.
Introduction
Enzyme-catalyzed ester hydrolysis is one of the most ubiquitous enzymatic actions in human tissues (Pocker and Stone 1967). Due to the critical roles of enzymes in biological and metabolic processes, material enzymolysis is drawing attention to achieve stimuli-responsive, targeted, and controlled drug delivery (Pauletti et al. 1997; Patel et al. 2008; Xu et al. 2008; Tian et al. 2012; Ma et al. 2015; Lian et al. 2017; Chung et al. 2018). The high selectivity and efficacy of these enzyme-catalyzed degradations have been applied to prepare complex compounds with high yield and purity in mild conditions (e.g., body temperature, neutral pH and aqueous solution) (Theil 1995; Bhangale et al. 2012). However, the enzyme-material interactions may cause material degradation and failures of medical devices (Santerre et al. 1999; Finer and Santerre 2004; Lin et al. 2005; Gregson et al. 2012). Consequently, serious concerns exist regarding the toxicity, durability, and biocompatibility of materials used in tissue engineering and dental restorations.
Pseudocholine esterase (PCE) and cholesterol esterase (CE) are the 2 most commonly used esterases for the biostability evaluation of dental resins (Santerre et al. 1999; Finer and Santerre 2004; Gonzalez-Bonet et al. 2015). In addition, cariogenic bacteria (e.g., Streptococcus mutans) also exhibit esterase-like activities (Bourbia et al. 2013; Huang et al. 2018). Esterases hydrolyze ester groups and split them into acids and alcohols through enzymolysis (Santerre et al. 1999; Finer and Santerre 2004; Lin et al. 2005). Dimethacrylate (DMA)–based resins are the most popular materials for dental adhesives, sealants, and composites. The DMA enzymolysis creates mechanical and biological challenges for medical devices in the oral cavity (Ferracane 2006; Park et al. 2009; Bettencourt et al. 2010; Bourbia et al. 2013; Cai et al. 2014; Delaviz et al. 2014), and enzymolysis products cause toxicity concerns (Bakopoulou et al. 2009; Chang et al. 2010; Visalli et al. 2013). Precisely characterizing these interactions is vital for the success of materials development. So far, no quantitative information is available to compare the kinetics of DMA enzymolysis and the enzyme-catalyzed hydrolysis of the corresponding unsymmetrical product, mono-methacrylate (mono-MA) (Yourtee et al. 2001). The challenges in obtaining information stems from the following: 1) some DMAs (e.g., bisphenol A glycidyl DMA [Bis-GMA]) do not form mono-MA during enzymolysis by either PCE or CE (Finer and Santerre 2003), and 2) some DMAs contain isomers (e.g., diurethane DMA [UDMA]) (Polydorou et al. 2009). In addition, the dental polymers are resin networks containing unconverted methacrylates. Precisely determined kinetics of monomer enzymolysis are vital to understand the biostability of dental polymers and identify the causative factors that lead to a compromised durability of resin restoratives. High-performance liquid chromatography (HPLC) has been used extensively for assessing biostability of dental resin composites. However, due to the limitation of instruments and the availability of degradation products (e.g., mono-MAs or diols), current HPLC methods do not differentiate the methacrylate monomer enzymolysis and polymer degradation except for Bis-GMA, which is always used with other methacrylate monomers because of its high viscosity. Such 1-pot HPLC methods provide results of polymers’ biostability, which need to be clarified by well-resolved enzymolysis kinetics of each component, including monomers’.
In this study, we chose triethyleneglycol DMA (TEGDMA) and UDMA as model DMAs because they are commonly used with Bis-GMA (Ferracane 1995, 2011; Jandt and Sigusch 2009) and also produce diols that require special columns and detectors for concentration determination. The hypothesis is that precisely determined monomer enzymolysis kinetics is achievable without using HPLC-measured [diol] or [mono-MA]. Given the selectivity of enzymes to DMAs, both CE and PCE are employed for the kinetic studies and as catalysts to produce mono-MAs through sequential hydrolysis in which mono-MAs are hydrolyzed after the majority of DMA is decomposed (Fig. 1) (Finer and Santerre 2003; Hsu et al. 2012; Gonzalez-Bonet et al. 2015). The chemical structure of these mono-MAs is verified by nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry. We used the ratio of methacrylic acid (MA)/mono-MA to quantify the competition between sequential and simultaneous decompositions of DMA and mono-MA. By determining the enzymolysis kinetics of TEGDMA, UDMA, and mono-MAs, the efficacy and selectivity of enzymes on DMAs and mono-MAs hydrolysis are compared. This in-depth study of monomer enzymolysis is achieved by using HPLC methods without determining [diol], which is supported by stoichiometric analysis and Berkeley Madonna models. All these tools may be used in other systems that involve human saliva, bacteria, and/or polymers. The quantitative information generated in this study is vital for understanding polymer biostability and polymer biocompatibility.

Scheme of consecutive enzyme–catalyzed hydrolysis (pseudocholine esterase or cholesterol esterase) of triethyleneglycol dimethacrylate and diurethane dimethacrylate monomers. TEGDMA, triethyleneglycol dimethacrylate; UDMA, diurethane dimethacrylate.
Results and Discussion
Synthesis and Characterization of Asymmetric Multifunctional Intermediate Products
Precise stoichiometry analysis and kinetic studies require high-purity mono-MAs, which are commercially unavailable. In addition, the intermediate products prepared by enzymolysis have a specific composition of isomers. These isomer compositions might change if prepared via other means, including acid- or base-catalyzed hydrolysis. As an example, UDMA is a mixture of 2 isomers (Fig. 1). The UMA is a mixture of 4 isomers, which are associated with 2 adjacent HPLC peaks due to the slight difference in hydrophilicity, in terms of logarithm of partition coefficient (LogP). Compounds with lower LogP are relatively more hydrophilic and less strongly retained in the reverse-phase HPLC column. Consequently, these compounds have a shorter retention time. As indicated in Figure 1, UMA1.1 and 1.2 and UMA2.1 and 2.2 from the same UDMA isomers (UDMA1 or UDMA2) share the same LogP (2.31 or 2.37), respectively. Correspondingly, 2 adjacent HPLC UMA peaks are identified at 13.2 min and 13.4 min (Appendix Fig. 1). In this study, the UMA isomers were characterized and quantified together.
The intermediate products, TEGMA and UMA, were synthesized through enzymolysis and purified by HPLC as described in Appendix Materials. In this study and for the first time, we fully characterize these 2 compounds using mass spectroscopy and NMR (Appendix Figs. 2–5). These compounds were previously identified by mass spectrometry and HPLC (Santerre et al. 2001; Hsu et al. 2012), but no NMR results were reported.
Well-Maintained Sequential Enzymolysis: Experimental Observation
We use the equimolar production of MA and mono-MA (TEGMA or UMA) as an indicator for the sequential enzymolysis of DMA and mono-MA. The well-maintained equimolar ratio of MA/mono-MA indicates that the decomposition of mono-MA takes place after most DMAs are decomposed, suggesting that the 2 ester groups on 1 DMA are hydrolyzed sequentially. Specifically, data in Appendix Table 1 highlight the molar ratios of MA/mono-MA when <20% and >90% of DMA are hydrolyzed. In DMA-PCE combinations, the equimolar MA/mono-MA was maintained (1.00 ± 0.10) when <20% of DMA is degraded. However, when >90% of DMAs are decomposed, the mean ± SD MA/mono-MA ratios were 1.59 ± 0.01 and 2.02 ± 0.01 for TEGDMA-PCE and UDMA-PCE, respectively. The molar ratios >1 indicate the hydrolysis of mono-MA. In contrast, the equimolar ratio of MA/mono-MA was maintained in all DMA-CE combinations. In addition, the MA/mono-MA ratio during the full course of DMA degradation is illustrated in Figure 2. In comparison, CE was more effective than PCE in maintaining the MA/mono-MA molar ratio at 1. Both MA/TEGMA and MA/UMA are equimolar (ratio ≤1.10) until >90% of TEGDMA and UDMA was hydrolyzed by CE, which indicates that no mono-MA is decomposed and DMA and mono-MA are hydrolyzed sequentially. Such a well-controlled sequential reaction has great potential to produce asymmetrical multifunctional compounds with high purity. However, when PCE serves as the catalyst, the MA/mono-MA ratios are approximately 1.25 during the major portion of DMA enzymolysis and increase significantly when >80% of DMA is decomposed, which suggests that a fraction of mono-MA is being hydrolyzed simultaneously with the DMA.

Experimental observation of well-maintained sequential enzymolysis of TEGDMA and UDMA. (
Stoichiometric Analysis of 2-Step Enzymolysis
In going from DMA to diols, the DMA and mono-MA will be hydrolyzed by enzymes either sequentially or simultaneously (see equations 1 and 2 in the Appendix Materials). To assess the relative ratios of sequential and simultaneous ester group hydrolysis, we used the percentage of mono-MA (mono-MA%) that underwent the second step of enzymolysis to form diols: mono-MA% of 0 represents the sequential 2-step ester hydrolysis, and mono-MA% of 100 corresponds to simultaneous DMA/mono-MA degradation with no mono-MAs. By employing the stoichiometric analysis, we were able to determine the mono-MA% without involving the [mono-MA].
In the following equations (equations 3 to 8), y1, y2, and y3 correspond to the concentration of decomposed DMA, [MA], and [mono-MA], respectively, and y2a and y2b represent the concentrations of MA obtained by splitting the first and second ester functional groups, respectively. [DMA]0, [DMA], [MA], and [mono-MA] were determined by HPLC from the peak area values and the calibration curves of the corresponding compounds.
Figure 3 shows the mono-MA% during the progression of DMA degradation (DMA deg.%). The stoichiometric calculations agree very well with the experimental observations: CE is more effective than PCE in maintaining the sequential degradation, and approximately 13% of mono-MA are hydrolyzed simultaneously with DMAs (between 20 and 80 DMA deg.%) when PCE is used as the catalyst. Moreover, the mono-MA% is calculated per [MA] alone, showing that the stoichiometric approach is a valuable tool in studying DMA enzymolysis especially when 2 ester groups are hydrolyzed simultaneously.

The percentage of mono-MA hydrolyzed during the course of DMA enzymolysis: (
First-Order Enzymolysis of the First Ester Group on DMAs
Figure 4 depicts reaction time–dependent changes in substrate (DMA) concentration expressed as natural log of [DMA], ln[TEGDMA], or ln[UDMA]. Strong linear correlation between the ln concentration and reaction time indicates the first-order reaction kinetics in both TEGDMA and UDMA systems. Slopes of the straight lines correspond to the reaction rate constant (k1) and indicate that, for both TEGDMA and UDMA, the PCE-catalyzed hydrolysis was more rapid than the CE-catalyzed one. PCE was more effective in TEGDMA preparations resulting in 5-times faster enzymolysis as compared with UDMA. In contrast, CE is more effective on UDMA than on TEGDMA (approximately by a factor of 2).

First order of enzyme-catalyzed hydrolysis of the first ester group of TEGDMA (
First-Order Enzymolysis of the Second Ester Group on DMAs
The decomposition of mono-MAs becomes a dominating process at the later stage of enzymolysis when most of the DMAs are already hydrolyzed. Within 24 h, only PCE enzymolysis of TEGMA and UMA had reached this condition and exhibited the linear correlation between the ln[mono-MA] and reaction time, indicative of the first-order reaction kinetics with respect to TEGMA and UMA. Slow reaction rate of CE-catalyzed hydrolysis of TEGMA and UMA precluded us from confirming their reaction order. To resolve this, we applied nonlinear regression fitting (Berkeley Madonna model) and estimated the apparent reaction rate constants (k1′ and k2′) per the analytic data sets for [DMA], [mono-MA], [MA], and calculated [diol] per the stoichiometry of DMA enzymolysis (equation 9):
The 2-step DMA enzymolysis was simplified according to equation 10. We used the rate law equations (equations 10 to 15) for consecutive first-order reactions to determine reaction rate constants k1′ and k2′ and predict the concentration changes in substrate and products.
Integrating [DMA] gives
Substituting this into equation 13 gives
The solution of this differential equation is
Finally, using equation 10, we find
The reaction rate coefficients that are determined by linear fitting (k1) and by the nonlinear regression fitting via the multicompartment model (k1′ and k2′) are listed in Appendix Table 2. Statistical comparison of the observed reaction rate constants (k1 and k1′) of the first enzymolysis step revealed no significant difference (P > 0.05) for all 4 substrate-enzyme combinations. Changes in substrates and product concentrations, which are calculated with apparent rate constants (k1′ and k2′), are in good agreement with the experimentally determined values (Fig. 5). Concentrations of both DMA substrates exposed to esterases decreased exponentially. DMA half-lives in the examined substrate-enzyme combinations decreased in the following order: TEGDMA-PCE (14.8 ± 1.2 min) > UDMA-PCE (113.6 ± 14.9 min) > UDMA-CE (239.0 ± 24.7 min) > TEGDMA-CE (495.1 ± 35.4 min). Both ester groups in TEGDMA and UDMA were completely hydrolyzed by PCE within the experimental time frame, while the ester decompositions were not completed by CE.

Nonlinear regression data fitting of concentrations as a function of time with a Berkeley Madonna 4-compartment mathematic model for different substrate-enzyme combinations. (
The [mono-MA]s increased, reached maximum, and then decayed exponentially in the TEGMA-PCE and UMA-PCE systems. Within the experimental time frame, TEGMA-CE and UMA-CE only began to approach the maximum, and no significant decline was observed. As indicated in Appendix Table 2, k1′ > k2′ for all substrate-enzyme combinations; specifically, k1′/k2′ ratios are ≥10 for all but the UDMA-PCE system (e.g., k1′/k2′ = 3). The completion of mono-MA enzymolysis, expressed as a half-life of mono-MA, is in the following order TEGMA-PCE (2.5 ± 0.1 h) > UMA-PCE (6.4 ± 0.4 h) > UMA-CE (140.9 ± 55.0 h) > TEGMA-CE (641.8 ± 320.9 h). The large standard deviations of k2′ (i.e., 39% and 50% for UMA and TEGMA exposed to CE, respectively) are likely due to the extremely low extent of their enzymolysis. These estimates are useful for ranking their degradation rate.
The enzymolysis levels of the examined DMA-enzyme systems are as follows: TEGDMA-PCE (completed cleavage of both ester bonds) > UDMA-PCE > UDMA-CE (completed hydrolysis of DMA) > TEGDMA-CE (partial hydrolysis of DMA). PCE is more efficient on shorter-chain linear aliphatic monomers TEGDMA and TEGMA, while CE is more effective on higher molecular weight UDMA and UMA esters. The observed selective activity of enzymes agrees well with the literature reports: PCE has been reported to preferentially hydrolyze low molecular weight esters (e.g., TEGDMA) (Yourtee et al. 2001), while CE has shown higher activities toward long-chain fatty acid esters, including Bis-GMA (Sutton et al. 1991; Feaster et al. 1996; Gonzalez-Bonet et al. 2015). In addition, we tested the enzymolysis of UDMA with different enzyme concentrations at a short time (e.g., 2 h) (Appendix Table 3). In the UDMA/PCE combination, a linear correlation of UDMA deg.% and enzyme concentration forms when the enzyme concentration doubles from 2.5 unit/mL to 5 unit/mL and to 10 unit/mL. This correlation suggests that the hydrolysis power of PCE (10 unit/mL) is much greater than that of the medium, and so does CE (1 unit/mL). In addition, CE reaches its maximum enzymolysis efficacy at approximately 2.5 unit/mL. A higher CE concentration does not lead to greater UDMA degradation in 2-h incubation. Furthermore, as opposed to when these 2 are used together, UDMA deg.% ≥ sum UDMA deg.% when PCE and CE are incubated individually with UDMA. Such findings highlight the need for thorough studies of monomer enzymolysis to understand the synergistic effects of CE and PCE on the more complex polymer degradation (Finer et al. 2004), which includes the splitting of ester groups in unpolymerized methacrylate and cross-linked resin network.
Correlation of DMA/Mono-MA Enzymolysis with Restoration Toxicity and Biostability
The DMAs provide cross-linked resin networks for dental resin restorations through curing of monomers. The completion of monomer polymerization is quantified by the degree of vinyl conversion (DC). For a traditional Bis-GMA/TEGDMA composite, DC is approximately 70%, which suggests that 30% of vinyl groups from the feeding monomers were not converted. Some of these vinyl groups are in DMAs, which will leach out into the oral environments, and some of them are attached in the resin network, with 1 of the 2 vinyl groups converted.
Since both TEGDMA and UDMA monomers have been used extensively in oral devices, the observed resistance of mono-MAs (TEGMA and UMA) to enzymolysis raises concerns regarding material toxicity and biocompatibility. The unpolymerized DMAs may leach into oral cavities within hours. While general toxicity and biocompatibility studies focus on the evaluation of DMAs (Santerre et al. 2001), there are no cytotoxicity reports on the corresponding mono-MAs, partially due to the difficulty in synthesizing the exact mono-MAs (e.g., the isomers of UMA). The longer enzymolysis half-lives of mono-MAs, as compared with DMAs, strongly suggest that they should be evaluated in toxicity and biocompatibility studies.
Sequential enzymolysis is a 2-step degradation from DMA to mono-MA to diol (2D-DMAs). The mono-MAs are more stable than their 2D-DMAs. Moreover, the interchain ester groups in TEGDMA homopolymers (Yourtee et al. 2001) and TEGDMA/UDMA copolymers (Wang et al. 2018) undergo even slower hydrolysis. Furthermore, increasing DC of TEGDMA/UDMA copolymers enhanced their biostability. In summary, the biostability of these monomers and their copolymers increases in the following order: 2D-DMA << mono-MA << low DC (≈70%) polymer << mid DC polymer (≈85%) < high DC polymer (≈96%) ≈ no degradation. In contrast, the simultaneous enzymolysis (e.g., enzymolysis of Bis-GMA by CE) is a pseudo 1-step degradation from DMA directly to diol (1D-DMAs) (Finer and Santerre 2003). Consequently, the order of biostability is 1D-DMA ≥ mono-MA. Will such dramatic variation in DMA/mono-MA biostability lead to different trends in polymer biostability when increasing the DC of polymers? Such in-depth evaluations were overlooked but are essential to identifying causative factors that lead to a compromised durability of medical devices. They also offer strong scientific rationale for new materials design and development. In addition, the presented stoichiometric analysis and the employed mathematical model provide unique tools for identifying and quantifying differences in biostability of the building blocks for the dental resin networks without assessing [diol]s using HPLC. These tools may be applied for evaluating monomer and polymer degradation by human saliva and bacteria.
Conclusion
In summary, we successfully prepared and thoroughly characterized 2 asymmetric multifunctional mono-MAs, which represented the exact isomer compositions during sequential enzymolysis of TEGDMA or UDMA. We also determined and compared the kinetics of 8 enzyme/substrate combinations by tracking the temporal changes of [DMA]s and [mono-MA]s simultaneously without assessing [diol]s using HPLC. All DMA/mono-MA enzymolysis reactions evaluated in this study were first-order reactions. The mono-MAs were more stable than their corresponding DMAs, which emphasized the need for toxicity evaluation of mono-MAs. These quantitative kinetic evaluations of monomer enzymolysis may provide the overlooked but essential in-depth understanding of dental polymer degradation, which includes the hydrolysis of unpolymerized methacrylate and in-network ester groups. Overall, this study delivers valuable knowledge on enzymolysis kinetics and material biostability while providing new tools for preparation and characterization of multifunctional compounds.
Author Contributions
S. Frukhtbeyn, contributed to design, data acquisition, analysis, and interpretation, drafted and critically revised the manuscript; K. Van Dongen, contributed to data acquisition, drafted the manuscript; J. Sun, contributed to conception, design, data analysis, and interpretation, drafted and critically revised the manuscript. All authors gave final approval and agree to be accountable for all aspects of the work.
Supplemental Material
DS_10.1177_0022034519858975 – Supplemental material for Stoichiometry and Kinetics of Sequential Dimethacrylate Enzymolysis
Supplemental material, DS_10.1177_0022034519858975 for Stoichiometry and Kinetics of Sequential Dimethacrylate Enzymolysis by S. Frukhtbeyn, K. Van Dongen and J. Sun in Journal of Dental Research
Footnotes
Acknowledgements
The authors thank the American Dental Association for its support. J. Sun thanks Dr. Michael Nelson for his help on NMR data collection and analysis and Dr. Drago Skrtic for his critical reviews on the manuscript.
A supplemental appendix to this article is available online.
This work was funded by the National Institute of Dental and Craniofacial Research (U01DE023752). Financial support was also provided through the ADA Foundation.
The authors declare no potential conflicts of interest with respect to the authorship and/or publication of this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
