Enzymatic Hydrolysis of Triacetin and L-Lactide in Emulsified Microparticles Within a Cellulose Hydrogel Dispersion

Abstract

Triacetin (TA) is a solvent commonly used in pharmaceutical and food applications, and as a plasticizer in bioplastics such as poly(lactic acid) (PLA) and cellulose acetate (CA). L-lactide is the monomer used in the ring-opening polymerization of PLA. The structure of TA emulsions stabilized by a cellulose hydrogel (CH) was imaged in this study. The emulsions were prepared by mechanical homogenization or a two-step process with subsequent high-pressure homogenization (HPH). The two-step process yielded smaller TA droplets and a more homogeneous CH dispersion. The images demonstrate that emulsion stabilization is due to CH particles adsorbed at the TA–water interface. The ester hydrolysis of TA and a lactide/TA solution by two industrially important lipases, from Candida rugosa (CRL) and Burkholderia cepacia (BCL), was investigated, assessing the effect of CH as an emulsion stabilizer. Mechanically homogenized TA emulsions were effectively hydrolyzed. Lactide was found to inhibit the enzymatic hydrolysis of TA. This inhibition was mitigated by CH for CRL-catalyzed hydrolysis but not for BCL catalysis. These results indicate a synergistic effect of CH stabilization on the interfacial activation of CRL. Thise effect may also be relevant for the biodegradation of bio-derived plastics and their fibrous cellulose composites.

1. Introduction

Triacetin (TA, glycerol triacetate) is the smallest triglyceride. TA and its hydrolysis products, diacetin and monoacetin (DA and MA, glycerol diacetate and monoacetate, respectively), have many useful applications [1,2]. Their low toxicity and regulatory approval provide enhanced value in the medical [3], food [4], cosmetic [5], and pharmaceutical industries [6]. Glycerol acetates are also useful as additives in gasoline and diesel fuels [7]. TA and DA are widely used as biodegradable plasticizers for plastic materials, especially bio-derived ones such as poly(lactic acid) (PLA) [8]. TA’s water solubility is 70 gL⁻¹ at 25 °C, and it is also miscible with various organic solvents such as alcohols, diethyl ether and aromatic hydrocarbons. Its high boiling point (258 °C) and low vapor pressure make it suitable for high-temperature processing [9]. TA undergoes hydrolysis in water, catalyzed by acids or bases [10], and is readily enzymatically hydrolyzed by various microorganisms as well as immobilized enzymes. TA hydrolysis is often used to test enzymatic activity as a model system [11].

Lipases (acylglycerol acylhydrolase, EC 3.1.1.3) catalyze the hydrolysis of triglycerides to free fatty acids and glycerol. Lipase from the Candida rugosa yeast (CRL) is used extensively because it exhibits high activity as well as specificity for the L-enantiomer [12,13,14]. Another lipase, from Burkholderia cepacia (BCL), was identified with activity toward the L-isomer, indicating its potential for the hydrolysis of L-lactide [15]. Lipases are activated by oil–water interfaces [16,17]. This was considered to be due to a conformational change upon the adsorption of the enzyme to the interface, exposing the buried hydrophobic active site [18,19,20]. Enzymatic activity is significantly enhanced when the substrate is present at an interface of dispersed or emulsified droplets [21,22]. Lower activity is observed on substrates that dissolve in water molecularly, due to passing the solubility limit in water, or when dissolved in a hydrophobic solvent [17]. TA hydrolysis catalyzed by CRL was studied with respect to the effects of a water–hydrophobic interface, which exists either beyond its aqueous solubility limit or due to an added hydrophobic liquid. It was found that, when the aqueous interface is exposed to TA droplets, the hydrolysis yield is less than that at an interface to hexane, which indicates a difference in the structure of the active site [23]. Interest in TA is due also to the possibility of the regio-selective synthesis of diacetin by hydrolysis in position 1 or 3 using immobilized lipase [24].

Ester-hydrolyzing enzymes are well known to function in the environmental biodegradation of organic and polymer pollutants [25,26]. The use of bio-derived plastics for pollution mitigation as well as waste valorization is highly pursued [27]. PLA and lactic acid copolymers are of particular interest, despite the relative recalcitrance of PLA to degradation in natural environments. Thus, much research has focused on PLA-degrading microorganisms and the enzymes, natural and engineered, derived from them [28]. Cellulose acetate (CA) is another important bio-derived polymer. The most widely used CA plastics have a degree of acetylation of each glucose unit of 2–2.5 [29], which renders them significantly more resistant to enzymatic degradation, thus causing a serious pollution and health hazard [30,31,32]. In the context of the present study, it is of interest to note that the enzymatic degradation of a small molecule, glucose pentaacetate, was used to model the hydrolysis of the polymer CA, as it resembles its hydrophobic repeat unit [33]. Lactide, the hydrophobic cyclic dimer of lactic acid ester, is the monomer used in the ring-opening polymerization of PLA [34]. It is also formed during the thermal or alkaline degradation of PLA by intramolecular transesterification [35]. Its biodegradation may also serve as a model for that of PLA.

The need for plasticizers in the use of bio-derived thermoplastics, such as CA and PLA, to facilitate processing and achieve desired properties has boosted the demand for natural, biodegradable plasticizers that are highly compatible [36]. Considered as such, TA and DA are effectively utilized as plasticizers for PLA [8,37,38,39,40] and CA [33,41,42,43]. Furthermore, water-soluble plasticizers were shown to accelerate the degradation of biopolymers by improving the penetration of water and enzymes [41,44]. As TA possesses antimicrobial properties, it can interfere with microbial degradation [45]. Despite the advancement in bioplastics’ application, many challenges remain, especially their limited biodegradation in open or anaerobic conditions [46], resulting in the generation of microplastics and their associated environmental and health hazards [47,48]. Embedding enzymes within the bioplastic during processing is emerging as a possible solution [49,50,51].

Fibrous biopolymers are often used as reinforcing fillers in bioplastics to enhance the material’s strength, thermal stability, and gas and water barrier properties [52]. Cellulose, the most abundant and renewable biopolymer, is used in a variety of forms, from nano- and micro-sized crystals to macroscopic fibers [53]. The incorporation of cellulose fibers into a PLA matrix, along with additives such as TA, was shown to create heterogeneous structures that improve the PLA’s features and facilitate its biodegradation [45,54]. The cellulose chain exhibits amphiphilicity capable of stabilizing emulsions [55]. A cellulose hydrogel (CH) formed by dissolution and regeneration can be used for emulsification, forming a continuous cellulose coating [56]. A unique encapsulating structure was discovered by emulsification with a cellulose solution in an ionic liquid [57] or an aqueous dispersion of CH particles [58], in which an inner porous CH shell was formed between the hydrophobic core and the outer continuous layer. This inner interface serves as a suitable microenvironment for the activity of lipases that are spontaneously incorporated from the aqueous media into the encapsulated microparticles [59,60]. Lipase CRL was immobilized on CH microspheres by physical adsorption, exhibiting enhanced activity and stability compared to free lipase [61]. These results highlight the effective microenvironment provided by CH for lipase activity.

The objective of this work was to assess the emulsification of TA by a CH dispersion and to investigate the enzymatic hydrolysis of TA and L-lactide as a model for interfacial biocatalysis in emulsions based on a cellulose-hydrogel system. It is hypothesized that the amphiphilic character of CH and its open structure can facilitate the interfacial activation of the enzyme. Moreover, by analyzing the enzymatic activity toward these small molecules in the CH medium, this study may provide insight into the biodegradability of plasticized PLA composites with cellulose-based reinforcement.

2. Results

2.1. Microscopy Analysis

A colloidal system of TA droplets incorporated into a CH dispersion at a 1:14.1 cellulose-to-TA wt. ratio was prepared via either one-step (mechanical homogenization) or two-step (mechanical and HPH) emulsification. Fluorescence microscopy was used to obtain images of the TA droplets (red-stained with Nile Red) and the cellulose hydrogel dispersion (CH, green-stained with Calcofluor White). The one-step process of mechanical homogenization produced a non-homogeneous CH structure exhibiting globular aggregates and smaller fibrous structures (Figure 1a, right panel). A merged fluorescence image (Figure 1a, left panel) indicates the existence of significant (dark) aqueous regions. The two-step process yielded a markedly more uniform CH dispersion (Figure 1b), as expected. Unexpectedly, dark rings are visible around the large TA droplets, indicating a lower CH content and regions without CH around them. Furthermore, the presence of CH in the regions of the TA droplets is indicated by the green fluorescence from these droplets, in contrast with the image for one-step emulsification. This effect may be interpreted as the adsorption of CH microparticles from the surrounding CH onto the TA droplets, as in Pickering-type emulsions. The droplet size distributions for both methods, as assessed from Figure 1, are very broad yet seem to be comparable. The 3D renderings (Videos S1 and S2) provide a detailed visualization and confirm improved integration in the sample prepared by two-step emulsification.

Figure 1.

Confocal 3D rendering micrographs of TA droplets (red) in CH (green) dispersion. Each trio of images contains (from left to right) the merged red and green channels, the red channel alone and the green channel alone. (a) One-step homogenization; (b) two-step homogenization. Volume: 714 (X) × 714 (Y) × 100 (Z) µm.

Confocal cross-sectional plane images attained from the 3D rendering of Figure 1 are presented in Figure 2. The larger square images on the left are XY-plane images from around the mid-planes of the one- and two-step processes. The smaller rectangular images below and beside the XY image are of XZ (below) and YZ (right) cross-sections at locations indicated by white lines in the XY images. Successive enlarged XY cross-section images of one droplet from each process, taken at 5 μm intervals along the Z-axis, are presented on the right. XZ and YZ panels are also shown in the enlarged image in the far-right images. Of particular interest are the yellow regions that appear around the droplet in the enlarged images from both processes. The yellow fluorescence signal in the merged images is due to combined red and green fluorescence [62]. This indicates the penetration of CH into the TA droplets, due to interfacial contact between cellulose and TA, which stabilizes the TA dispersion. An enhanced effect may be indicated by the CH depletion rings around the droplets in the HPH process, due to adsorbed and penetrated CH. Dynamic imaging (Videos S3 and S4) confirmed a uniform droplet distribution across the sample depth.

Figure 2.

Confocal cross-sectional XY (white lines), YZ (right) and XZ (bottom) views showing integration of TA droplets (red) in CH (green). The rectangles indicate identical fields of view in the images beside (from left to right, XY slices captured at 5 µm intervals). (a) One-step homogenization; (b) two-step homogenization.

LM images of samples from the one-step process (Figure 3a) also exhibit TA droplets with a very broad size distribution, in the range of 1–100 µm, dispersed in CH. Approximately three types of droplets can be classified by size: very large droplets (tens of microns), mid-range droplets (about 10 µm) and small ones (~1 µm). Some of the large and mid-range droplets exhibit small water inclusions. The observed CH structure in this case is inhomogeneous, with several aqueous regions without cellulose. The coarse appearance of the CH also exhibits some fibrous structures. The samples from the two-step process (Figure 3b) displayed a higher prevalence of small TA droplets (~1 μm), and the CH appears less coarse.

Figure 3.

LM images of TA droplets in CH: (a) one-step homogenization; (b) two-step homogenization. Images on the right are enlargements of regions from those at left, as indicated. Arrows point to particular structures: small-size TA droplets; mid-size TA droplets; TA droplets with water inclusions.

Cryo-SEM images of the fracture surface of rapidly frozen TA/CH dispersions made using the two-step process are presented in Figure 4. Large- and small-sized TA droplets are shown embedded in the frozen aqueous medium in Figure 4a and Figure 4b, respectively. A smooth interfacial boundary of the droplets is observed. It is difficult to discern CH particles, yet some regions in the enlargement of Figure 4b, which are indicated by black arrows, may be due to CH at the edge of the small droplet. The radial features in the large-sized droplet in Figure 4a may be due to TA crystallization, whereas the small droplet may have vitrified to a more amorphous structure. Figure 4c presents a combined secondary electron image and EDS elemental mapping of the large droplet (shown in Figure 4a). The EDS spectrum and elemental map are included in the supplemental information (Figure S1). Blue-colored inclusions in the red-colored particle indicate oxygen-rich water inclusions in the carbon-rich TA droplet. Inclusion features can also be seen in the image of the small-sized particle, as indicated by the white arrows in the enlargement of Figure 4b. An image of a dispersed cellulose gel, exposed by the specimen fracture, is shown in Figure 4d and its enlargement, revealing interconnected fibrous cellulose morphology.

Figure 4.

Cryo-SEM images of the fracture surface of rapidly frozen TA/CH dispersions made by the two-step process: (a) a large TA droplet; (b) a small droplet. White arrows point to water inclusions. Black arrows indicate possible CH particles at the droplet edge. (c) A combined secondary electron image and EDS elemental mapping of the large droplet shown in (a). Red color: carbon-rich regions; blue color: oxygen-rich regions. (d) An image of a dispersed cellulose gel particle, exposed by the specimen fracture. Images on the right are enlargements of the areas indicated in (b,d).

2.2. Nuclear Magnetic Resonance (NMR) Analysis

In this study, the biodegradation of the cyclic lactic acid dimer, L-lactide, was studied as a model for the effective depolymerization of PLA and its composites. Of particular interest is lactide biodegradation under the effects of the biodegradable plasticizer/solvent triacetin, often used as a plasticizer [37,38,39,40], and of the natural biopolymer cellulose, which can be found as fibrous reinforcements [53,63,64]. The concentrations of substrates and products were measured by ¹H NMR spectroscopy. Figure 5 shows the NMR spectra of the components: TA, L-lactide, and their mixture (measured in CDCl₃), and TA as well as TA/L-lactide mixture (measured in D₂O). The chemical shifts of the measured peaks are provided in Table 1. Comparing the peaks of L-lactide in CDCl₃ alone and pre-dissolved in TA (N 2 and N 3, respectively), it can be seen that the -CH₃ chemical shift (D) decreases from 1.67 to 1.60 ppm, whereas the -CH peak (E) is nearly unchanged. In D₂O, the -CH₃ peak of L-lactide in the TA solution is shifted to 1.5 ppm, due to the effect of water. In this case, the -CH peak shifts significantly to 5.3 ppm. The effect of water renders L-lactide accessible to hydrolysis; hence, the NMR spectrum exhibits peaks (F–I) of the dimeric structure, due to partial ring opening by hydrolysis for 10 min at ambient temperature. Spectrum N 6 in Figure 5 is of 1% commercial L-LA (90 wt.%), in D₂O, showing peaks M1 and M2 of L-LA, as well as F, G, H and I of the LA-dimer and T1 and T2 of the trimer.

Figure 5.

¹H NMR spectra (295 K, 400 MHz) of samples: (N 1) TA in CDCl₃; (N 2) L-lactide in CDCl₃; (N 3) L-lactide/TA solution in CDCl₃; (N 4) the aqueous phase of L-lactide/TA mixture in D₂O; (N 5) TA in D₂O. (N 6) 1 wt.% of commercial 90 wt.% of L-LA solution in D₂O. The solution concentrations are given in Table 1.

Table 1.

¹H NMR chemical shifts (ppm) of TA, L-lactide and their solution (measured in CDCl₃, 7.26 ppm), and of L-LA, TA and L-lactide/TA solution (measured in D₂O, 4.7 ppm).

Sample ID	Sample Composition	Solvent	1H NMR Chemical Shifts 3 (ppm)
N 1	5.5 wt.% TA	CDCl3	-CH3 (A)—2.04 -CH2 (B)—4.1-4.27 -CH (C)—5.2
N 2	1.1 wt.% of L-lactide		Ring -CH3 (D)—1.67 -CH (E)—5.04
N 3	5.5 wt.% TA with 0.85 wt.% L-lactide		Ring -CH3 (D)—1.6 -CH (E)—5.05
N 4	5.1 wt.% TA with 0.9 wt.% L-lactide and 0.37 wt.% dimeric structure 2	D2O	Ring -CH3 (D)—1.5 -CH (E)—5.3 Open ring -CH3 (F,G)—1.36, 1.43 -CH (H,I)—4.37, 5
N 5	5.8 wt.% TA 1		-CH3 (A)—2.02 -CH2 (B)—4.18-4.28 -CH (C)—5.2
N 6	1 wt.% of commercial 90 wt.% L-LA		Monomers -CH3 (M1)—1.29 -CH (M2)—4.255 Dimers -CH3 (F,G)—1.33, 1.4 -CH (H,I)—4.35, 4.98 Trimers -CH3 (T1)—1.455 -CH (T2)—5.13

^1,2 D₂O phase, separated by centrifugation from 1:2 vol. ratio mixture of (1) TA or (2) 15 wt.% L-lactide solution in TA, added to D₂O, after 10 min of vortexing at ambient temperature. The concentrations were calculated from intensity integrations relative to the DMSO peak at 2.61 ppm. ³ For the A-I peaks’ identification, see the molecular structures in Figure 5.

2.3. Evaluation of Lipase-Catalyzed Hydrolysis

2.3.1. Enzymatic Hydrolysis of TA

The lipase-catalyzed hydrolysis of TA is well studied, even as a model system. It was shown that the TA/water interface, beyond the limit of TA solubility, is less effective in lipase activation than either lipase immobilization [65] or the interface formed with hexane [23]. The hydrolysis of TA proceeds through the sequential cleavage of three ester bonds, producing diacetin (DA), monoacetin (MA), glycerol and acetic acid (Figure 6). All the control samples without enzymes exhibited separation into aqueous and TA phases. This indicates that the aqueous phases retained full TA saturation (~7 wt.%). The reduction in the TA phase in the biphasic structure and subsequent reduction in the measured TA concentration in water confirm progressive enzymatic hydrolysis. The activity of several lipases was studied—CRL, BCL, PPL, MJL and AOL (see Section 4)—but only CRL and BCL showed significant hydrolysis, as shown in Figure S2. Figure 7 presents the NMR spectra of the products of TA’s enzymatic hydrolysis by the lipases CRL or BCL (5 mg mL⁻¹) in aqueous TA emulsions, prepared by mechanical homogenization with and without CH dispersion, after incubation at 45 °C for 120 h. The TA conversion and glycerol yield were calculated from integrals of the ¹H NMR signals from protons at the second (methine) carbon atom of the glycerol part (position C in Figure 6, with more details in Table S1), which provides non-overlapping signals, and the methyl protons of acetic acid, as presented in Figure 7. The total conversion of TA by enzymatic hydrolysis to DA, MA and glycerol, and the yield of glycerol formation, were calculated based on the final TA and glycerol concentrations evaluated from the NMR spectra presented in Table 2, and the initial TA concentration (presented in Table 4 in Section 4). For the control samples, the data of the produced acetic acid was used due to its biphasic nature.

Figure 6.

Sequential hydrolysis of TA showing progressive formation of diacetin, monoacetin, glycerol and acetic acid. Annotated proton C positions (green) identify corresponding peaks in the ¹H NMR spectra.

Figure 7.

¹H NMR spectra of products of TA’s enzymatic hydrolysis by lipases CRL and BCL in aqueous TA emulsions and emulsions prepared with CH dispersion, after incubation at 45 °C for 120 h. The emulsions were prepared by mechanical homogenization.

Table 2.

Evaluation of ¹H NMR spectra shown in Figure 7 and Figure 9. Peak integrals and calculated concentrations.

Catalyst 5 mg mL−1	Emulsion Medium	1H NMR Intensity of Integral (×10−2)			Concentration (wt.%)
		TA	Acetic Acid	Glycerol	TA	Acetic Acid	Glycerol
CRL	TA in water	0.3 ± 0.3	40 ± 2	3.6 ± 0.5	0.4 ± 0.4	5.6 ± 0.4	2.2 ± 0.3
	TA +lactide in water	5.6 ± 0.3	15 ± 1	0.72 ± 0.07	8.6 ± 0.4	2.1 ± 0.1	0.45 ± 0.04
	TA in CH dispersion	0	43 ± 1	4.5 ± 0.2	0	5.9 ± 0.2	2.8 ± 0.2
	TA +lactide in CH	1 ± 0.7	31 ± 3	1.9 ± 0.5	1.6 ± 1.0	4.2 ± 0.4	1.2 ± 0.3
BCL	TA in water	1.2 ± 0.2	22 ± 1	1.41 ± 0.08	1.9 ± 0.2	3.1 ± 0.2	0.88 ± 0.05
	TA +lactide in water	2.9 ± 0.8	16 ± 2	1.2 ± 0.1	4.5 ± 1.3	2.3 ± 0.3	0.75 ± 0.07
	TA in CH dispersion	1.6 ± 0.1	24 ± 1	1.44 ± 0.04	2.5 ± 0.2	3.3 ± 0.2	0.90 ± 0.03
	TA +lactide in CH	3.7 ± 0.3	17 ± 1	1.2 ± 0.1	5.7 ± 0.4	2.3 ± 0.2	0.73 ± 0.07
Control	TA in water	4.2 ± 0.8	1.5 ± 0.3	0.18 ± 0.09	6.3 ± 1.2	0.20 ± 0.04	0.11 ± 0.05
	TA +lactide in water	4.6 ± 0.2	2.6 ± 0.7	0.16 ± 0.05	7.0 ± 0.4	0.36 ± 0.10	0.10 ± 0.03
	TA in CH dispersion	4.11 ± 0.08	1.44 ± 0.04	0.10 ± 0.01	6.3 ± 0.1	0.2	0.06
	TA +lactide in CH	4.2 ± 0.5	2.9 ± 0.5	0.16 ± 0.03	6.4 ± 0.7	0.40 ± 0.07	0.10 ± 0.02

Figure 8 depicts the TA conversion and glycerol yield produced by enzymatic hydrolysis, as well as TA autohydrolysis in control samples without enzymes, for both the aqueous solution and CH dispersion. CRL and BCL were effective in TA hydrolysis, with CRL exhibiting a higher yield than BCL. Full elimination of TA was achieved in the CH-stabilized emulsion. The benefit of CH stabilization is also evident in the higher yield of glycerol. Altogether, the presence of lactide was observed to inhibit TA hydrolysis. The stabilizing effect of CH in achieving higher TA conversion in the presence of lactide is significant. The autohydrolysis of TA evaluated in aqueous media without enzymes (control samples) under both acidic (pH 3.4) and neutral (pH 7) conditions (after 24 h at 45 °C, as shown in Figure S3) indicated that the reaction proceeded more slowly at the lower pH.

Figure 8.

Enzymatic conversion of TA after incubation at 45 °C for 120 h with 5 mg mL⁻¹ of lipases BCL and CRL versus control samples, with or without L-lactide, in both aqueous and CH dispersion: (a) total conversion to diacetin, monoacetin and glycerol; (b) yield of TA complete hydrolysis to glycerol. Emulsions were prepared by the one-step mechanical homogenization process.

Experiments were also conducted at a higher enzyme loading of 10 mg mL⁻¹ and with a more extensive two-step homogenization. As expected, the yield of CRL-catalyzed hydrolysis increased, reaching 100%, yet hydrolysis by BCL was only slightly improved. Additional high-pressure homogenization further improved TA hydrolysis by BCL. The inhibitory effect of lactide was also less prominent. Higher enzyme loading and extensive homogenization also increased the yield of full hydrolysis to glycerol. These expected results are presented in Table S2 and Figures S4–S6.

2.3.2. Enzymatic Hydrolysis of L-Lactide

The enzymatic hydrolysis of L-lactide was evaluated in the presence of TA, a common lactide solvent, in the aqueous emulsion state stabilized by cellulose hydrogel (CH). This also serves as an indicator for the biodegradation of the polyester PLA, derived from lactide, into water-soluble products. TA is a common solvent and plasticizer for PLA, and cellulose fibrils can be found as a PLA reinforcement or in surrounding waste. Figure 9 presents the NMR spectra of the products of L-lactide’s enzymatic hydrolysis by lipases CRL or BCL (5 mg mL⁻¹) in aqueous TA emulsion, prepared by mechanical homogenization with and without CH dispersion, after incubation at 45 °C for 120 h. The initial compositions (wt.%) of the emulsion subjected to enzymatic hydrolysis were about 13% TA and 1% lactide, with 1% cellulose in emulsions containing CH, as detailed in Table 4 in Section 4. The yields of L-lactide’s complete hydrolysis to L-LA monomer and partial hydrolysis to L-LA dimer are presented in Figure 10. These were determined from the concentrations of L-LA monomers and dimers quantified by analyzing the methyl proton signals in the NMR spectra, which provide distinct, non-overlapping peaks for each compound as presented in Figure 9. Table 3 summarizes the relevant NMR integrals and the calculated product concentrations. Increasing the CRL load to 10 mg mL⁻¹ enhanced lactide hydrolysis, with increased yield of LA monomers, as shown in Table S3 and Figure S7.

Figure 9.

¹H NMR spectra of the products of TA’s and L-lactide’s enzymatic hydrolysis by lipases CRL and BCL in aqueous TA emulsions and emulsions prepared with CH dispersion, after incubation at 45 °C for 120 h. The emulsions were prepared by mechanical homogenization.

Figure 10.

Enzymatic hydrolysis of L-lactide by CRL and BCL lipases (5 mg mL⁻¹), in emulsions prepared by a one-step mechanical homogenization process, after 120 h at 45 °C, compared to lipase-free controls, in both aqueous solution and CH dispersion.

Table 3.

Evaluation of ¹H NMR spectra shown in Figure 9. Peak integrals and the calculated LA product concentrations.

Catalyst 5 mg mL−1	Emulsion Medium	1H NMR Intensity of Integral (×10−2)		Concentration (wt.%)
		LAmonomer	LAdimer	LAmonomer	LAdimer
CRL	Water	1.6 ± 0.1	4.2 ± 0.3	0.32 ± 0.02	0.79 ± 0.06
	CH dispersion	2.3 ± 0.3	3.5 ± 0.4	0.47 ± 0.06	0.65 ± 0.07
BCL	Water	1.4 ± 0.2	3.5 ± 0.4	0.29 ± 0.04	0.65 ± 0.08
	CH dispersion	1.5 ± 0.1	3.8 ± 0.4	0.32 ± 0.03	0.71 ± 0.07
Control	Water	1.1 ± 0.3	3.4 ± 0.7	0.23 ± 0.06	0.64 ± 0.13
	CH dispersion	1.2 ± 0.2	3.4 ± 0.5	0.24 ± 0.03	0.63 ± 0.09

3. Discussion

TA is commonly used as a solvent and is generally recognized as safe. It has been shown to have a stabilizing effect in the emulsification of compounds that are not easily emulsified, such as pharmaceuticals [66] and food flavorings [4]. It is often applied as a plasticizer in conjunction with cellulose-based systems such as with CA or with fibrillar reinforcement in PLA [63,64]. Unmodified cellulose was shown to effectively emulsify organic solvents and oils [55,56,57,58]. The structure of TA emulsions stabilized by a cellulose hydrogel was imaged in this study. Stable emulsions were formed by mechanical homogenization. As expected, subsequent high-pressure homogenization yielded smaller TA droplets and a more homogeneous CH dispersion. In both cases, the images demonstrate that stabilization is due to CH particles adsorbed at the TA–water interface. The observation of adsorbed CH particles that penetrate the interface highlights their amphiphilic character, which can provide a synergetic interaction with the interfacial activation of hydrolyzing enzymes.

In this study, the ester hydrolysis of TA and a lactide/TA solution by two industrially important lipases was investigated, assessing the effect of CH as an emulsion stabilizer: CRL, the extracellular lipase of the yeast Candida rugosa, and BCL, the extracellular lipase of the bacterium Burkholderia cepacia. The potential of cellulose hydrogel microspheres for CRL immobilization by physical adsorption, affording enhanced catalytic activity and enabling the reuse of the enzymes, was demonstrated [61]. Immobilization is important for the many industrial applications of CRL, providing broad substrate specificity, high activity and strong interfacial activation [13]. BCL is utilized extensively in free and immobilized forms due to its versatile substrate recognition, heat resistance and stability in many solvents. Moreover, BCL was shown to hydrolyze some biodegradable aliphatic polyester fibers or films, yet lipase-catalyzed polymer biodegradation is a formidable challenge [67]. Ester-hydrolyzing enzymes expressed by microorganisms can degrade microplastic pollutants, which are a major environmental concern even with the use of bio-derived plastics such as PLA and CA, especially in water [68].

The lipase-catalyzed hydrolysis of TA in aqueous solution is slower than when the enzymes are adsorbed on hydrophobic polymer microparticles [65]. The emulsification of TA enhances its enzymatic hydrolysis, yet it is still lower than that in emulsified TA solution in a hydrophobic solvent [23]. Several types of lipases can hydrolyze PLA, depending on molecular weight and stereochemical form [28]. However, the activity is not high. For example, a study on the enzymatic degradation of PLA and its composite with cellulose nanocrystals showed that CRL does not degrade PLA significantly, whereas proteinase K does so very efficiently [69]. Lactide, the building block of PLA, may be viewed as its small-molecule analog, given their similarity in chemical structure and hydrophobicity. The activity of lipase on lactide was studied in high-boiling organic solvents and in ionic liquids, demonstrating ring-opening polymerization to PLA [70,71]. The polymerization of hydrophobic large-ring lactones in emulsion was also reported [72]. The activity of lipase on lactide in an emulsion, especially when dissolved in TA, was not investigated.

The results of this study show, as expected, that both enzymes hydrolyze emulsified TA, with CRL exhibiting a higher conversion. This is not surprising because the common CRL form is known to exhibit high catalytic activity in the hydrolysis of short-chain substrates [73]. The new finding is that lactide significantly hinders TA hydrolysis, at an initial concentration of about 14 wt.% (assuming that all lactide is in the emulsified TA droplets, beyond its ~7 wt.% solubility in water). Lactide’s inhibition of TA hydrolysis is more prominent in CRL-catalyzed hydrolysis, such that about two-thirds of the initial TA content remains after incubation under the experimental conditions, compared to the nearly full conversion without lactide. However, the results clearly show that emulsion stabilization by CH improves TA hydrolysis, close to its value without lactide. The inhibitory effect of lactide is less noticeable in BCL-catalyzed hydrolysis, and the effect of CH in this case is not significant. The presence of lactide has a strong effect on the complete hydrolysis of TA to glycerol catalyzed by CRL but not on hydrolysis catalyzed by BCL. Again, CH stabilization significantly alleviates the CRL-catalyzed conversion to glycerol. The higher sensitivity of CRL to the inhibitory effect of lactide, compared to BCL, may be hypothesized to be due to partial blocking of the active site by the smaller and more hydrophobic lactide, which is more effective in the enzyme more attuned to smaller molecules. Under the experimental conditions, the lactide ring was hydrolyzed to a significant extent in all samples, mostly to the opened ring dimer form, with the enzymes providing higher conversion. This differs from TA hydrolysis, which was insignificant without enzymes, and may be due to some straining of the lactide ring compared to the triglyceride structure [74]. CRL was effective in the nearly full hydrolysis of lactide to water-soluble LA dimers and monomers, and the highest yield of LA monomers was achieved by CRL in the CH-stabilized emulsion. The beneficial effect of CH on CRL activity is in line with previous reports on the enhanced activity of CRL immobilized on a cellulose-based support [75,76].

The observation that the CH stabilization of TA emulsions enhanced hydrolysis by CRL but not that by BCL may be due to their different structures, particularly the structure of the so-called lid (or flap). This structural feature is a mobile domain that covers the active site. Its “closed” conformation in water blocks access to the inner active site, while transition to an “open” conformation provides access for substrates to the active site. This is considered the basis for interfacial activation in aqueous two-phase systems or specifically immobilized enzymes, or lipase activity in some organic solvents [18]. The lid structure of CRL, as used in this study, was reported to be large and complex, consisting of two α-helices [20,77,78,79], whereas that of BCL is smaller, consisting of a single mobile α-helix [67,80]. The larger size of the CRL lid can interact better with the interface with the TA droplet, assisted by parts of the CH that are submerged at the TA–water interface. Such penetration of CH into the TA droplets’ interface was observed microscopically in this study. A schematic description of this interaction is presented in Figure 11.

Figure 11.

A schematic presentation of an emulsion of TA droplets containing dissolved lactide, with adsorbed lipases, stabilized by CH microparticles, demonstrating CH-assisted activation of the lipase active site at the triacetin–water interface. The purple circles represent enzymes with an open or closed lid. The enlargement on the right indicates (schematically) the role of CH in promoting the open-lid conformation that enables the access of triacetin and lactide to the inner cavity and the catalytic site. The solid purple contour lines represent the lipase’s shape, with large orange dashes marking the inner cavity and small red dashes the open lid (the sketch is not to scale and does not reflect accurate molecular structures).

4. Materials and Methods

4.1. Materials

Triacetin (≥99% GC; Honeywell Fluka™, Charlotte, NC, USA), sodium hydroxide (Bio-Lab Ltd., Jerusalem, Israel), dipotassium hydrogen phosphate (K₂HPO₄, Spectrum Chemical, New Brunswick, NJ, USA), potassium dihydrogen phosphate (KH₂PO₄, Sigma Aldrich, Rehovot, Israel) and L-Lactic acid (90 wt.% in water; ThermoFisher Scientific, Madrid, Spain) were used in this study. Dimethyl sulfoxide (DMSO; NMR standard, Honeywell Riedel-de Haën™, Seelze, Germany), deuterium oxide (D₂O, 99.9%; Sigma Aldrich, Rehovot, Israel) and deuterated chloroform (CDCl₃, 99.8%; ZEOtope, Rüti, Switzerland) were employed for NMR spectroscopy. Microcrystalline cellulose (MCC; Lot #MKCJ3230), L-lactide (98%) and the fluorescent dyes Nile Red and Calcofluor White were obtained from Sigma Aldrich (Rehovot, Israel). Porcine pancreatic lipase (PPL), Amano lipase PS from Burkholderia cepacia (BCL), Amano Lipase M from Mucor javanicus (MJL), lipase from Candida rugosa (CRL) and lipase from Aspergillus oryzae (AOL) were also purchased from Sigma Aldrich (Rehovot, Israel) and used as received.

4.2. Preparation of Cellulose Hydrogel (CH) Dispersion

The CH dispersion was prepared as described previously [81]. MCC powder (4 wt.%) was dispersed in aqueous sodium hydroxide (7 wt.%) at room temperature, followed by cooling to −10 °C using a chiller bath under continuous mechanical stirring. CH formation was induced by gradually introducing the dissolved cellulose solution into a 16-fold-larger volume of deionized water. This gradual regeneration produced a dispersion of soft amorphous hydrogel particles. The regenerated CH was repeatedly washed with deionized water until its conductivity fell below 1 mS cm⁻¹. Excess water was removed by filtration to obtain a cellulose content of approximately 1.5–2 wt.% in the dispersion. The cellulose concentration was determined gravimetrically in triplicate.

4.3. Fabrication and Characterization of TA Colloidal Systems in Both Aqueous and CH Dispersion

Four colloidal systems, either with or without CH dispersion containing 8 wt.% TA were prepared by magnetic stirring for 10 min under ambient conditions. Additional TA, either alone or containing 15 wt.% L-lactide, was then introduced with continuous stirring for another 10 min. The compositions of these primary systems are presented in Table 4. Two homogenization processes were employed in this study, mechanical homogenization and high-pressure homogenization (HPH), to assess the effect of vigorous mixing, especially regarding the integration of TA within the CH dispersion.

• One-step homogenization: mechanical homogenizer (ULTRA-TURRAX^® T 18 digital, IKA Works Inc., Staufen, Germany), for 10 min at 20 krpm.

• Two-step homogenization: mechanical homogenization followed by high-pressure homogenization (HPH; LM10 Microfluidizer^®, Microfluidics, Newton, USA) at 10 kPSI for 5 min at controlled temperature of 15–20 °C.

Table 4.

Prescription of components in four tested colloidal systems.

Colloidal Systems	Components	System 1	System 2	System 3	System 4
		TA in Water	TA + Lactide in Water	TA in CH	TA + Lactide in CH
Primary	TA (wt.%)	14.16	13.16	14.16	13.16
	Lactide (wt.%)	-	1	-	1
	Cellulose (wt.%)	-	-	1	1
For enzymatic hydrolysis (after buffer addition)	TA (wt.%)	13.8	12.85	13.8	12.85
	Lactide (wt.%)	-	0.95	-	0.95
	Cellulose (wt.%)	-	-	0.95	0.95

4.4. Enzymatic Hydrolysis of TA and L-Lactide

The four primary colloidal systems, after homogenization by either method, were distributed into centrifuge tubes and mixed with 0.05 M phosphate buffer (pH 7) that was saturated with TA (~7 wt.%) to minimize changes in TA–water equilibrium. Their composition is presented in Table 4. Each colloidal system (6 g) was divided as follows: one control (no enzyme) and others containing one kind of lipase (CRL, BCL, PPL, MJL and AOL) at concentrations of 5 or 10 mg mL⁻¹. The enzyme-to-substrate wt.% ratio was 1:27.6 or 1:13.8. Samples were incubated horizontally at 45 °C and 150 rpm for 120 h.

4.5. Quantification of Hydrolysis Products

After incubation, samples were centrifuged (6 krpm, 10 min), and the aqueous phase was collected for ¹H NMR analysis. The enzymatic hydrolysates of TA and L-lactide were analyzed using an NMR spectrometer (Bruker Avance III™ HD 400 MHz, Fällanden, Switzerland). The samples consisted of 480 μL of D₂O, 120 μL of aqueous phase and 10 μL of DMSO as an internal standard. Spectra were recorded at room temperature with a 3 s delay between scans. Chemical shifts were referenced to D₂O (4.7 ppm). The spectra were processed, and the integrals were calculated using the TopSpin (4.3.0) program (Bruker, Rheinstetten, Germany). Products were quantified using signal integrals relative to the DMSO standard: C_X = (C_DMSO × N_DMSO × I_X)/(I_DMSO × N_X), where C_X is the molar concentration of component X, I_X is its signal integral, N_X is the number of contributing protons, I_DMSO = 1, N_DMSO = 6 and C_DMSO = 0.23 mmol mL⁻¹.

4.6. Characterization of TA Colloidal Systems

4.6.1. Spinning Disk Confocal Microscopy

Fluorescent imaging was carried out with a spinning disk confocal inverted fluorescence microscope (Eclipse Ti2-E, Nikon, Tokyo, Japan) equipped with a scanner (CSU-W1, Yokogawa, Tokyo, Japan). Samples were stained with Nile Red (1 g L⁻¹ in TA) and Calcofluor White (1 g L⁻¹ in water). For systems prepared by the one-step process, dyes were added before homogenization; for two-step-processed samples, dyes were incorporated after HPH. Dual-channel imaging (2 × 2 binning, 1024 × 1024 px, 16-bit) was conducted using a 20× objective (NA 0.7) with 101 optical sections across 100 μm. The images were created in a cavity glass slide and visualized using dual fluorescent labeling with the NIS-Elements Imaging Software (5.22.00). Fluorescence excitation/emission settings: 405/447 ± 30 nm (Calcofluor White) and 561/708 ± 37.5 nm (Nile Red). Gamma corrections (green = 2.0, red = 3.0) were applied to enhance low-intensity structures.

4.6.2. Light Microscopy

Phase-contrast and interference-contrast images were acquired using a light microscope (LM, Olympus BH2, Tokyo, Japan) with 10× and 40× objectives and a CCD camera (Nikon DS-Fi2, Tokyo, Japan). Image analysis was performed using the NIS-Elements software, as above. Interference contrast provided optical depth, while phase contrast enhanced droplet boundaries and visualized TA droplets without staining.

4.6.3. Cryogenic Scanning Electron Microscopy (Cryo-SEM) and Elemental Analysis

The morphology of the TA-CH system 3 (Table 4) was examined using a cryogenic scanning electron microscope (Zeiss Ultra Plus, Oberkochen, Germany) with a Gemini SEM 360 column and Schottky field-emission source. Samples were vitrified using a Leica ICE system (Leica Microsystems, Wetzlar, Germany), fractured and etched at −100 °C for 3 min 30 s to reveal internal structures. Specimens were sputter-coated with 6 nm Pt/C prior to imaging. Imaging was conducted at 2 kV, with a 4.7–5 mm working distance, using an Everhart–Thornley secondary electron detector (SE2, Zeiss, Oberkochen, Germany). The elemental composition was analyzed by energy-dispersive X-ray spectroscopy (EDS) at 10 kV with a 6.1 mm working distance using a silicon drift detector (SDD, Zeiss, Oberkochen, Germany).

5. Conclusions

The hydrolysis of TA, a partially water-soluble short triglyceride, was enhanced by the lipases CRL and BCL under mild hydrolytic conditions, with CRL exhibiting a higher conversion. The presence of lactide was found to inhibit the enzymatic hydrolysis of TA. This lactide-mediated inhibition of TA hydrolysis was mitigated in TA emulsions stabilized by CH for CRL-catalyzed hydrolysis but not for BCL catalysis. In these integrated cellulose-based systems, TA microparticles were stabilized through interfacial contact with the CH microparticles. Microscopic observation of adsorbed CH particles that penetrate the interface highlights their amphiphilic character and indicates possible synergism with the interfacial activation of the enzymes. A more extensive emulsification process with CH, involving mechanical and high-pressure homogenization, resulted in small droplets and a more uniform CH dispersion. However, mechanical homogenization was sufficiently effective in TA degradation and the hydrolysis of lactide into water-soluble dimers and monomers. These results indicate that catalytic activity is driven by the combined effects of a stabilized interface and anchoring of the enzyme to the TA droplets with a more prominent “open lid” conformation, which is assisted by the CH dispersion. The synergistic effect of CH stabilization on CRL activity compared with BCL may be due to its larger and more complex lid structure. These results may also be relevant for the biodegradation of bio-derived plastics and their fibrous cellulose composites to water-soluble molecules that can be further eliminated by bio-organisms.

Abbreviation

The following abbreviations are used in this manuscript:

LA	Lactic acid
PLA	Polylactic acid
CA	Cellulose acetate
TA	Triacetin
DA	Diacetin
MA	Monoacetin
CH	Cellulose hydrogel
MCC	Microcrystalline cellulose
HPH	High-pressure homogenization
CRL	Candida rugosa lipase
BCL	Amano Lipase PS, from Burkholderia cepacia
PPL	Porcine pancreatic lipase
MJL	Amano Lipase M, from Mucor javanicus
AOL	Aspergillus oryzae lipase

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijms27041799/s1. Reference [82] was cited in the Supplementary Materials.

Author Contributions

M.A., Y.C., G.A. and D.M.R. developed the concept and experimental design. M.A. conducted the experiments, analyzed the results and prepared the original draft. M.A. and Y.C. prepared the manuscript. G.A., D.M.R. and Y.C. supervised the research. G.A. and D.M.R. performed review and editing. Y.C. and G.A. were responsible for funding acquisition. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Conflicts of Interest

The authors declare that they have no competing interests.

Funding Statement

This project was funded by the Israel Science Foundation Grant no. 449/23, and by the Russell Berrie Nanotechnology Institute for equipment use.