Development of a Dual-Stage CIM^® CDI Reactor with Immobilized Glucuronan Lyases and Laccases for Sustainable Synthesis of Antioxidant Phenolized Oligoglucuronan

Abstract

Immobilized enzyme reactors (IMERs) are critical tools for developing novel oligosaccharides based on the enzymatic catalysis of polysaccharides. In this paper, a novel glucuronan lyase from Peteryoungia rosettiformans was produced, purified, and then immobilized on a CIM^® CDI disk for cleaving glucuronan. The results showed that around 63.6% of glycuronan lyases (800.9 μg) were immobilized on the disk. The V_max values of immobilized glucuronan lyases did not significantly change (56.9 ± 4.7 μM∙min⁻¹), while the K_m values (0.310 ± 0.075 g∙L⁻¹) increased by 2.5 times. It is worth noting that immobilized glucuronan lyases overcame the catalytic inhibition of free enzymes observed under high glucuronan concentrations (0.5–2 g∙L⁻¹). circumscribed central composite design (CCCD) and response surface methodology (RSM) showed that glucuronan concentration, flow rate, and reaction time significantly affected the yield of oligoglucuronans. The degree of polymerization (DP) of degraded glucuronan ranged from DP 2–8 according to the results obtained by high performance anion exchange chromatography coupled with a pulsed amperometric detector (HPAEC-PAD). The IMER retained 50.9% activity after running 2373 column volumes of glucuronan. Finally, this glucuronan lyase reactor was tentatively connected to an immobilized laccase reactor to depolymerize, and gallic acid (GA) was added to glucuronan. Approximately 8.5 mg of GA was added onto 1 g of initial glucuronan, and the GA–oligoglucuronan conjugates showed notable antioxidant activity.

1. Introduction

Polysaccharides are abundant natural polymers found in plants, animals, bacteria, fungi, algae, and so on. The development and study of these macromolecules have been of primary interest for the last century in the fields of food and nutrition, biomedicine, or well-being applications [1,2,3]. Oligosaccharides, which are low-molecular-weight molecules consisting of 2 to 10 monosaccharide residues, include several functional types that exhibit significant properties, for instance, as dietary fibers, prebiotics, antioxidants, or anti-aging agents [4,5,6]. The demand for developing novel oligosaccharides is urgent and necessary due to their immense potential in various applications. The enzymatic degradation of polysaccharides is a crucial method for producing oligosaccharides, as it offers a green and efficient approach, especially when it is combined with other classical physical processes [7,8,9]. Immobilizing enzymes plays a key role in enhancing enzyme reusability, reducing costs, and also enabling continuous catalysis with controlled conditions [10,11,12]. Monoliths, which are single-piece porous materials with adjustable shapes and channel sizes, can be used for enzyme immobilization and catalytic reactions. They have many advantages, including particularly highly interconnected permeable networks and low back pressures during use, even for viscous aqueous solutions [13,14]. The Convective Interaction Media^® (CIM^®) disk is a type of monolithic column support featuring functional groups that allow for the covalent immobilization of compounds rich in amino groups. Several CIM^® disks have been used for the immobilization of carbohydrate-active enzymes to continuously depolymerize or to molecularly modify polysaccharides, resulting in poly- and oligomers with specific functional bioactivities [15,16].

Glucuronans (β-1,4-polyglucuronic acids) are linear anionic polysaccharides that consist of β-1,4-glycosidic bonds linking glucuronic acid residues, sometimes with variable degrees of O-acetyl substitution at C-2 and/or C-3 positions [17,18]. Natural glucuronan can be extracted from specific bacteria, fungi, and algae, since it acts as a component of extracellular polysaccharides or cell wall structures in microorganisms [19]. Glucuronan lyases (EC 4.2.2.14) are cell wall polysaccharide lyases mostly found in bacteria and fungi [20,21], which are able to cleave the β-1,4-linkage between glucuronic acid residues through a β-elimination mechanism, leading to the formation of unsaturated glucuronan with low molecular weight (oligoglucuronan) [22,23,24]. These anionic oligoglucuronan possess multiple biological activities in fruit preservation, as well as in therapeutic and pharmaceutical applications [25]. Compared to highly acetylated glucuronans, deacetylated glucuronans have been shown to have a stronger affinity to glucuronan lyases and are more easily depolymerized into oligoglucuronans [26,27]. The first work on glucuronan lyase immobilization from Trichoderma sp. GL2 attempted the continuous production of oligosaccharides, yet it showed low immobilized activity and a short use cycle [28]. It is of primary importance to develop more efficient immobilized enzyme reactors (IMERs) to meet the requirements for controlled glucuronan degradation.

In this paper, glucuronan lyase from a Peteryoungia rosettiformans strain [29] was produced, purified, and then covalently immobilized on CIM^® CDI disks. The immobilization yield, the kinetic parameters of free and immobilized enzymes, and the operating stability of the glucuronan lyases were determined. The effects of glucuronan concentration, flow rate, and reaction time on the degradation of deacetylated glucuronan from Sinorhizobium meliloti M5N1CS and the degree of polymerization (DP) of oligoglucuronan were explored. Response surface methodology (RSM) was used to model and analyze the contribution of these parameters to the response to predict the optimized operating conditions. Finally, we tentatively connected the immobilized glucuronan lyase reactor with an immobilized laccase reactor to construct an IMER system for the dual modification of glucuronan, i.e., degradation and phenolization. This research aims to provide an example of the IMER system for multiple exploitation of oligoglucuronan and gallic acid (GA)—glucuronan and oligoglucuronan conjugates—and offer strategies to peers for the construction of IMERs for the exploitation of functional polysaccharides and oligosaccharides.

2. Materials and Methods

2.1. Materials and Chemicals

The CIM^® CDI disks were purchased from Sartorius BIA Separations Biotechnology Company (Ajdovščina, Slovenia). The glucuronan was given by Professor Petit from the BIOPI laboratory (University of Picardie, France) and deacylated according to previous report [17]. Other chemicals and enzymes were analytical grade and purchased from Sigma Aldrich^® (Saint-Quentin-Fallavier, France).

2.2. Production and Purification of Glucuronan Lyase

The production of the recombinant glucuronan lyase was performed as described in [29], and the protein purification was carried out using Immobilized Metal Ion Chromatography Affinity (IMAC) (Protino^® Ni-Ted column, Macherey-Nagel, Düren, France) according to the recommendation of the supplier. Briefly, the Escherichia coli (E. coli) strain containing recombinant glucuronan lyase genes from Peteryoungia rosettiformans was precultured in 15 mL of Lysogeny Broth (LB) medium (Peptone 10 g∙L⁻¹, Yeast extract 5 g∙L⁻¹, NaCl 5 g∙L⁻¹) with 15 μL of kanamycin (30 µg·mL⁻¹) at 37 °C under agitation on a rotary shaker overnight. After 16 h, 4.5 mL of the precultures were cultivated at 37 °C in 150 mL of an LB medium with 150 µL of kanamycin. The growth of the strains was monitored by measuring optical density (UV-1700 spectrophotometer, Shimadzu, Duisburg, Germany). When absorbance reached values ranging from 0.4 to 0.6 at 600 nm, isopropyl β-d-1-thiogalactopyranoside (IPTG) was added (final concentration reached 1 μM) into the cultures, and then the cultures were cultivated at 37 °C for 4 h with agitation to induce the expression of the genes. The culture pellets containing the glucuronan lyase fraction were collected by centrifugation (10,000× g, 5 min) and stored at −20 °C for further purification.

The frozen pellets were subjected to two freeze–thaw cycles before lysis. The pellets were then dissolved in 15 mL of LEW 1× Lysozyme (5 g∙L⁻¹) solution and sonicated using an ultrasonic processor UP50H (50 watts, 30 kHz, Hielscher Ultrasound Technology, Germany) for 2 to 3 min (0.5 cycles, 100% amplitude) to lyse the cells [30,31]. The lysate was then left on ice for 30 min. It was sonicated again (1 cycle, 80% amplitude) for 2 min twice to ensure complete lysis of the cells. The supernatant containing all soluble fractions, including glucuronan lyases, were collected by centrifugation (9000× g, 30 min, 4 °C). Parental and chimeric His-tagged enzymes were purified by Immobilized Metal Ion Chromatography Affinity (IMAC) (Protino^® Ni-Ted column, Macherey-Nagel, France). The solutions of 1X LEW buffer (50 mM NaH₂PO₄, 300 mM NaCl, pH 8), the supernatant fraction, and 1X LEW buffer (50 mM NaH₂PO₄, 300 mM NaCl, 250 mM imidazole, pH 8) were successively passed through the Ni-Ted column. Then, 1X Elution buffer was used to elute covalently bound His-tagged recombinant enzymes. Finally, the purified fraction was concentrated to 25 mL using a Stirred Ultrafiltration Cell (Millipore^®, Molsheim, France) with a 1 kDa ultrafiltration disk. A dialysis membrane (Mw = 3 kDa) along with dialysate (0.1 M sodium phosphate buffer) was used to remove imidazole from the elution buffer. The soluble, purified fractions were analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE, 4% stacking gel and 15% resolving gel, w/v) to analyze the types of proteins.

2.3. Enzymatic Assays of Glucuronan Lyases

The activity and kinetic parameters of glucuronan lyases were quantified through the production of unsaturated products (i.e., oligoglucuronan) within 4 min by measuring the absorbance at 235 nm (UV-1700 spectrophotometer, Shimadzu, Duisburg, Germany). All reactions were performed at room temperature in potassium acetate buffer (50 mM, pH 5.5) with solubilized deacetylated glucuronan and 0.5 µg of glucuronan lyases. The catalytic reaction rate of glucuronan lyases was monitored under substrate concentrations ranging from 0.01 to 2 g∙L⁻¹. The quantity of oligoglucuronan was calculated using a molar extinction coefficient of ∆-(4,5)-unsaturated oligoglucuronan (degree of polymerization of 3), which is assumed to be 4931 M⁻¹ cm⁻¹ [18,21]. One unit (U) of glucuronan lyase activity was defined as the production of unsaturated products per minute. The initial reaction rate (V_i) was calculated based on the stable production of products per minute at the beginning of the reaction. The kinetic parameters of glucuronan lyases, such as the maximal reaction rate (V_max) and the Michaelis constant (K_m), were estimated through Michaelis–Menten fitting (OriginPro 2024, 10.1.0.170) according to the relation between the concentration of substrate and initial reaction rate.

2.4. Immobilization of Glucuronan Lyases

The schematic diagram of the immobilized enzyme reactor system is shown in Figure 1. Purified glucuronan lyases were prepared in sodium phosphate (0.5 M) and sodium chloride (0.5 M) buffer (pH 8). First, the CIM^® CDI disk was flushed with 2 mL of ultrapure water and equilibrated with 2 mL of Na–phosphate buffer with a flow rate of 0.2 mL∙min⁻¹ at room temperature. Next, 20 mL of the prepared glucuronan lyase solution was passed through the CIM^® CDI disk for 24 h at a flow rate of 0.2 mL∙min⁻¹ in a closed loop (Figure 1) with gentle stirring (200 rpm) at room temperature, which corresponded to dynamic immobilization. Subsequently, the CIM^® CDI disk was removed from the system and placed in the glucuronan lyase feed tank for 4 h at room temperature, which corresponded to static immobilization. After that, the CIM^® CDI disk was reconnected to the system and flushed with 2 mL of Na–phosphate buffer (0.5 M, pH 8) to remove non-covalently linked enzymes. The residual CDI groups were deactivated by flushing the disk with 3 mL of ethanolamine (2 M, pH 8) at 0.2 mL∙min⁻¹, and the column was sealed and stored for 24 h at room temperature. Finally, the CIM^® CDI disk was flushed with 2 mL of ultrapure water at 0.2 mL∙min⁻¹ to remove ethanolamine and stored in 20% ethanol at 4 °C for further use.

2.5. Immobilization Yield of Glucuronan Lyases

The immobilization yield of glucuronan lyases was determined by measuring the protein concentration in the feed tank and flushing fluids during immobilization using the Bradford assay with bovine serum albumin as the standard. The protein concentration was quantified by measuring absorbance at 595 nm (UV-1700 spectrophotometer, Shimadzu, Duisburg, Germany). All the measurements were performed in triplicate. The immobilization yield was calculated according to Equation (1):

$$Immobilizationyield\left(\%\right)={\displaystyle \frac{Immobilizedproteinscontents}{Totalproteinscontents}}\times 100$$

(1)

where immobilized protein contents equal to the dynamic immobilized proteins plus static immobilized proteins minus buffer washed proteins minus water washed proteins. Total protein contents are equal to the initial proteins.

2.6. Enzymatic Assays of Immobilized Glucuronan Lyases

The activity and kinetic parameters of immobilized glucuronan lyases were evaluated using the same device system shown in Figure 1 and quantified through the production of unsaturated products within 10 min by measuring the absorbance at 235 nm using the same method as described in Section 2.3 (UV-1700 spectrophotometer, Shimadzu, Duisburg, Germany). Three ml of glucuronan with substrate concentrations ranging from 0.01 to 2 g∙L⁻¹ were fully solubilized in potassium acetate (50 mM, pH 5.5) and passed through the CIM^® CDI disk. The absorbance of the solution in the feed tank was monitored every minute for 10 min. The V_i and kinetic parameters V_max and K_m of immobilized glucuronan lyases were calculated and estimated using the same method as in Section 2.3. The long-term operational stability of immobilized glucuronan lyases was monitored by testing their activity using 0.2% deacetylated glucuronan at a flow rate of 0.2 mL∙min⁻¹.

2.7. Operating Conditions on Immobilized Glucuronan Lyases Reactor

2.7.1. Experimental Design

Multiple external factors can affect the reaction process. Here, based on the capability of the reaction system, factors such as substrate concentration (0.5–1.5%, w/v), flow rate (0.1–0.3 mL∙min⁻¹) of the substrate, and reaction time (30–90 min) were selected to probe their effects on the depolymerization of deacetylated glucuronan. A circumscribed central composite design (α = 1.681) with 3 factors was created using Minitab^® 20.3, and the experimental design table is shown in

Table 1. Experimental design of circumscribed central composite design (CCCD).

Number of Experiments	c of Glucuronan (%)	Flow Rate (mL∙min−1)	Time (min)
1	0.015	0.2	60.0
2	0.05	0.1	30.0
3	0.05	0.1	90.0
4	0.05	0.3	30.0
5	0.05	0.3	90.0
6	0.1	0.03	60.0
7	0.1	0.2	9.5
8	0.1	0.2	60.0
9	0.1	0.2	60.0
10	0.1	0.2	60.0
11	0.1	0.2	110.5
12	0.1	0.36	60.0
13	0.15	0.1	30.0
14	0.15	0.1	90.0
15	0.15	0.3	30.0
16	0.15	0.3	90.0
17	0.18	0.2	60.0

. All the experiments were carried out at room temperature according to Table 1 with 5 mL of deacetylated glucuronan substrate solubilized in potassium acetate (50 mM, pH 5.5). Finally, the production of oligoglucuronan in the feed tank was quantified by directly measuring absorbance at 235 nm (UV-1700 spectrophotometer, Shimadzu, Duisburg, Germany) and calculated using the same method as described in Section 2.3.

2.7.2. Structure Analyses of Oligoglucuronan and Conjugates

High-Performance Anion Exchange Chromatography–Pulsed Amperometric Detection (HPAEC-PAD)

HPAEC-PAD was used to separate oligosaccharides and determine their degree of polymerization (DP) in final productions. The HPAEC analyses were carried out by Dionex^TM ICS3000 ion chromatography with a CarboPac^TM PA1 Analytical Column (4 mm diameter, 250 mm length) and a CarboPac^TM PA1 Guard Column (4 mm diameter, 50 mm length). All samples were filtered using 0.2 μm of a membrane filter, and then 25 μL was injected into the system. The mobile phase was performed under a gradient model at 25 °C with a flow rate of 0.25 mL∙min⁻¹, which contained eluents A (1 M NaOH) and B (1 M sodium acetate in 1 M NaOH) with 4 procedures: pure A for 5 min; gradient 0–100% B in A from 5 to 30 min; pure B for 30 to 40 min; and pure A from 40 to 55 min. Glucuronic acid was used a standard with 1 DP for quantification of the DP of oligosaccharides. The PAD analyses were carried out by pulsed amperometric ED50 detector (Dionex Corp., Sunnyvale, CA, USA). Data analysis was performed using Chromeleon^® software 6.8 (Dionex Corp., Sunnyvale, CA, USA).

Ultra-High-Pressure Liquid Chromatography Coupled to High-Resolution Mass Spectrometry (UHPLC-HRMS)

UHPLC-HRMS was conducted to evaluate the structural features and DP of the final fractions. Prior to analysis, 3 mL of oligoglucuronan solution at a concentration of 0.15% was placed in a 500 Da cut-off dialysis tubing, which had been pre-hydrated and rinsed with ultrapure water. The tubing was hermetically sealed, submerged in ultrapure water, and maintained at 4 °C with gentle shaking for 3 days. The water was changed every 24 h. After dialysis, the oligoglucuronan solution was concentrated using a speed vacuum dryer to obtain the dried-oligoglucuronan sample. Analyses were carried out using an Acquity UPLC H-class system (Waters, Milford, MA, USA) coupled to a high-resolution mass spectrometer (XEVO G2-S QTof) equipped with an ElectroSpray Ionization (ESI) source (Waters, Manchester, UK). The UHPLC system consisted of a quaternary pump (Quaternary Solvent Manager, Waters) and an autosampler (Sample Manager-FTN, Waters) with a 10 µL injection loop. For sample injection, the dried oligoglucuronan sample was resuspended in a mixture of 5% (v/v) solvent A (H₂O/formic acid 99.7:0.3 (v/v), 5 mM ammonium acetate), and 95% (v/v) solvent B (acetonitrile/H₂O/formic acid 97.75:1.75:0.3 (v/v/v), 5 mM ammonium acetate) at a concentration of 1 mg·mL⁻¹. The solution was centrifuged at 13,000 rpm for 5 min and filtered through a 0.22 µm PVDF filter. Ten μL of the filtered solution was injected into an YMC-Triart Diol column (150 mm × 2.1 mm, 1.9 µm) (YMC Europe Gmbh, Dinslaken, Germany) maintained at 30 °C. The system was operated at 0.4 mL·min⁻¹ under the following gradient elution programs involving solvents A and B: 0–2 min, 5% A; 2–16 min, 5–70% A; 16–18 min, 70% A; 18–18.1 min, 70–5% A; 18.1–22 min, 5% A. The column and the autosampler were maintained at 30 °C and 10 °C, respectively. Eluted compounds were ionized in negative mode (ESI⁻) with the following parameters: capillary voltage: 2.0 kV, source temperature: 120 °C, source offset: 80 V, desolvation temperature: 500 °C, cone gas flow: 50 L·h⁻¹, and desolvation gas flow: 800 L·h⁻¹.

The MS and MS/MS experiments were conducted using either MSE 4.2 (Waters, Manchester, UK) or targeted MS/MS approaches in centroid mode with a scan time of 0.1 s, within the range 50–2000 m/z. An MSE experiment involves data acquisitions in a single run, with no collision energy applied in function 1, preventing fragmentation (low-energy function), while function 2 employs a ramped-collision energy to generate fragments ions (high-energy function). MSE software algorithms then associate fragment ion spectra with their corresponding precursor ions by profiling each chromatographic peak and determining their retention time. In contrast, targeted MS/MS experiments are performed in a single run, where the targeted precursor ion is selected, and a specific collision energy is applied to generate fragments ions. MSE function 2 acquisitions were performed with a ramped-collision energy of 15–40 eV, while targeted MS/MS acquisitions employed a ramped-collision energy from 15 to 80 eV, with an ion intensity threshold of 500,000 s⁻¹ (data-dependent analysis). Leucine Enkephalin (MW = 555.62 Da, 1 ng·µL⁻¹) was used as a lock-mass for mass shift correction, and the mass spectrometer was calibrated prior to analysis using a 0.5 mM sodium formate solution. Data analysis was carried out using MassLynx 4.2 (Waters, Manchester, UK) and GlycoWorkbench 2.1 (European Carbohydrates DataBase) [32].

Nuclear Magnetic Resonance (NMR)

Thirty-five milligrams of material (oligoglucuronan or conjugates) were dissolved in 1 mL of D₂O phosphate buffer at pH 7. A volume of 600 μL from the solution was transferred into a 5 mm NMR tube. Proton NMR spectra were acquired at 333 K using a Bruker AVANCE III HD 500 MHz spectrometer with a 5 mm inverse TXI probe (¹H/¹³C/¹⁵N) and a z-gradient coil. For all the measurements, a one-dimensional proton NMR spectrum was recorded employing a ZG sequence. The experiment involved 128 scans with a 90° pulse width of 9 μs, a relaxation delay of 10 s, an acquisition time of 3.28 s, a spectral window of 10,000 Hz, and 65,000 data points, which were zero-filled to 131,000 before Fourier transformation with 0.3 Hz line broadening.

2.7.3. Statistical Analyses

Response surface methodology (RSM) was used to analyze the effects of factors on the catalysis reaction. The response surface design, response optimizer, and contour plot were used to analyze the effects of factors using Minitab^® 20.3.

2.8. Production of Phenolized Oligoglucuronan Using the Dual-Stage CIM^® Disks Reactor and First Investigation of Their Biological Activities

2.8.1. Depolymerization and Phenolization of Glucuronan by the Dual-Stage Reactor

A dual-stage reactor system, comprising an immobilized glucuronan lyase reactor (see Section 2.4) and an immobilized laccase reactor, was designed as shown in Figure 2. The laccase (EC 1.10.3.2) is from Trametes versicolor (38429) (58 kDa, 0.5 U·mg⁻¹)), and the immobilized laccase reactor is prepared as reported in [16]. Fifty ml of glucuronan (0.15% in 50 mM potassium acetate buffer, pH 5.5) was mixed with 300 μL of gallic acid (GA, 50 mM) and passed through the reactor system at a flow rate of 0.3 mL∙min⁻¹ and at room temperature. In total, 500 samples were collected at 0 h and 6 h for UV spectroscopy analysis to quantify the amount of GA, using GA gallic acid as the standard.

2.8.2. DPPH-Radical-Scavenging Ability of GA–Glucuronan and GA–Oligoglucuronan Conjugates

The antioxidant activity of GA–glucuronan and GA–oligoglucuronan conjugates was assessed by measuring their free-radical-scavenging ability against 2,2-Diphenyl-1-picrylhydrazyl (DPPH). Briefly, 0.5 mL of the sample, i.e., ultrapure water, glucuronan (0.25 g·L⁻¹, 0.5 g·L⁻¹, 1 g·L⁻¹, 1.5 g·L⁻¹ and 2 g·L⁻¹), GA–glucuronan and GA–oligoglucuronan (0.05 g·L⁻¹, 0.1 g·L⁻¹, 0.25 g·L⁻¹, 0.5 g·L⁻¹ and 1 g·L⁻¹) were added into 0.5 mL of a DPPH solution (0.1 mM DPPH dissolved in 96% ethanol). After 30 min of incubation, the glucuronan and the GA–glucuronan and –oligoglucuronan conjugate solutions were centrifuged to precipitate the poly- and oligosaccharides (12,000× g, 10 min). The supernatants were collected, and the absorbances were measured at λ = 517 nm (UV-1700 spectrophotometer, Shimadzu, Duisburg, Germany). DPPH-radical-scavenging activity was calculated using the following Equation (2):

$$DPPHradicalinhibition(\%)=100\times \left({Abs}_{control}-{Abs}_{sample}\right)/{Abs}_{control}$$

(2)

where Abs_control is the absorption of the mixture of 0.5 mL H₂O and 0.5 mL DPPH, and Abs_sample is the absorbance of the experimental group. Each group was carried out in triplicate.

3. Results

3.1. Production of the Glucuronan Lyases from Peteryoungia rosettiformans

The glucuronan lyase gene from Peteryoungia rosettiformans strain [29] was inserted into an E. coli expression vector (pET28a) and then subjected to induction for gene expression. Then, the bacterial strains were lysed, and the soluble fraction, which contains all the proteins produced during the expression period, was collected from the supernatant. SDS-PAGE analysis of the soluble fraction, shown in Figure S1 (Supplementary Data), revealed an overexpressed band with a molecular weight around 25 kDa, corresponding to glucuronan lyase. The soluble fraction was then passed through a Protino^® Ni-Ted column to purify the His-tagged glucuronan lyase. As shown in Figure S1 (Supplementary Data), a single band appeared in the purified fraction, highlighting that only one protein was present after purification, which corresponds to the molecular weight of the recombinant glucuronan lyase (25 kDa) [29]. Overall, the SDS-PAGE results confirmed that the purified fraction contains only glucuronan lyases, with an initial concentration of enzymes close to 47.4 μg∙mL⁻¹, which increased fivefold to 251.8 μg∙mL⁻¹ after concentration, as determined by the Bradford assays.

3.2. Immobilization Yields of Glucuronan Lyases

The immobilization of glucuronan lyases was achieved through covalent bonding between imidazole carbamate groups on the CIM^® CDI disk and amino groups of the glucuronan lyases. Samples were collected at various stages of the immobilization process on the CIM^® CDI disk, i.e., the initial glucuronan lyases solution (T₀), after dynamic immobilization (T_d), after static immobilization (T_s), after buffer washing (W₁), and after water washing (W₂). The protein content in each sample was quantified, and the immobilization yields were calculated, as shown in

Table 2. Immobilization yields of glucuronan lyases on the CIM^® CDI disk.

Samples	T0	Td	Ts	W1	W2
Protein mass (μg)	1259.0	386.9	324.6	93.8	39.7
Yield of dynamic immobilization (%)			69.3
Yield of static immobilization (%)			0.07
Loss of buffer washing (%) a			10.8
Loss of water washing (%) b			0.05
Final immobilized mass (μg)			800.92
Final Yield (%)			63.61

^a,b: the percentage of protein loss was calculated based on the mass of immobilized proteins rather than the protein content in the initial glucuronan lyase solution.

. Dynamic immobilization was the most efficient step, with 69.3% of immobilized proteins. Yet, static immobilization was significantly less effective, with only 0.07% of immobilized proteins. This low yield is likely due to high diffusion resistance, as proteins need to migrate from the liquid phase into the CIM^® CDI disk with its 2 μm channels. The loss of immobilized proteins is primarily attributed to physical adsorption rather than covalent bonding, as indicated by the 10.8% loss during the first buffer washing step. Subsequent water washing showed a much lower protein loss of 0.05%, highlighting that covalently linked proteins exhibit strong resistance. Overall, 63.6% of glucuronan lyases were successfully immobilized on the CIM^® CDI disk, corresponding to 800.9 μg of proteins. It is noteworthy that the immobilization yields can be influenced by various effects, including the type and properties of the enzymes, the functional groups of the supports, the immobilization model (flow-through or incubation), the flow rate, or the environmental pH. Previous research described in the literature has reported immobilization yields ranging from 3 to 62.3% [16,33,34].

3.3. Kinetic Parameters of Free and Immobilized Glucuronan Lyases

Kinetic parameters of enzymes describe the rates of enzyme-catalyzed chemical reactions and are critical for understanding enzyme catalysis behaviors after immobilization. The relationship between glucuronan concentration and the production rates of unsaturated glucuronan, for both free and immobilized glucuronan lyases, is shown in Figure 3a,c. In the range of glucuronan concentrations (0–0.375 g∙L⁻¹), the initial reaction rate (V_i) of the free enzymes increased with the rise in substrate concentration. However, in the range of 0.5–2 g∙L⁻¹, the V_i decreased significantly, likely due to catalytic activity inhibition. In contrast, the V_i of immobilized glucuronan lyases did not exhibit this inhibition for the same range of concentrations. This may be attributed to changes in the stability and/or structural rigidity after immobilization, which in turn may alter kinetic parameters and catalytic efficiency toward the substrate [35]. Additionally, comparing the kinetic parameters between free and immobilized enzymes may not be entirely appropriate due to the difference in enzyme availability (U) under the two operating conditions (see

Table 3. Kinetic parameters of free and immobilized glucuronan lyases regarding the model fitting.

Parameters	Free Glucuronan Lyase	Immobilized Glucuronan Lyase
Vmax (μM∙min−1)	56.2 ± 7.41	56.9 ± 4.74
Vmax (U)	0.028	0.171
Km (g∙L−1) a	0.122 ± 0.04	0.310 ± 0.08
Specific activity (U∙mg−1)	56.2	0.21
Kcat (s−1)	5.37	2.14
Catalytic efficiency (s−1·µM−1)	30.8	4.83
R-Square of model	0.95	0.96

^a Estimated molecular weight of glucuronan: 700,000 g∙mol⁻¹ [28].

).

The non-inhibition points of free glucuronan lyases and all points of immobilized glucuronan lyases were used to plot the Michaelis–Menten model fitting in Figure 3b,d. The results presented in Table 3 show R-Square values of 0.95 and 0.96, respectively. The maximum reaction rate (V_max) of glucuronan lyases remained largely unchanged after immobilization, while the Michaelis constant (K_m) of the immobilized glucuronan lyases was around 2.54 times higher than that of the free glucuronan lyases. A higher K_m value indicates that a greater substrate concentration is required to reach half of the V_max, which suggests a reduced affinity between the substrate and the enzymes. This observation aligns with the previous statements and could be due to changes in the rigidity of the catalytic pocket and/or some steric hindrances following the immobilization process. Thus, the immobilized glucuronan lyases with a lower affinity for the substrate can still catalyze the reaction at higher concentrations, on the contrary to the free enzymes which are inhibited for high levels in glucuronan (from 0.5 to 2 g∙L⁻¹). Finally, a 264-fold lower specific activity of immobilized glucuronan lyases was detected, which may be attributed to portions of glucuronan lyases not having the opportunity to contact the substrates as they passed through the reactor. This could be due to an overload (layers) of glucuronan lyases inside the 0.2 mL CIM^® CDI disk.

3.4. Effects of Operating Conditions on Immobilized Glucuronan Lyases Reactor

3.4.1. Effects of Operating Conditions on the Production of Oligoglucuronan

In this study, three factors, i.e., the concentration of glucuronan, flow rate, and reaction time, were selected to explore their influence on the depolymerization of glucuronan by the immobilized glucuronan lyase reactor. The concentration of substrate can impact its physical properties, such as solubility and viscosity (so the flow properties); the flow rate controls the contact time of the substrate and the immobilized enzymes; while the reaction time influences the number of cycles the substrate undergoes through the reactor. The final oligoglucuronan concentration was used as the response variable to evaluate the individual effects of each factor as well as their pairwise interactions on the catalysis process. A full quadratic model was used to describe their relationship, as shown in Equation (3):

$$\eta ={\beta }_0+{\beta }_1{x}_1+{\beta }_2{x}_2+{\beta }_3{x}_3+{\beta }_{11}{x_1}^2+{\beta }_{22}{x_2}^2+{\beta }_{33}{x_3}^2+{\beta }_{12}{x}_1{x}_2+{\beta }_{13}{x}_1{x}_3+{\beta }_{23}{x}_2{x}_3$$

(3)

where η is the concentration of oligoglucuronan (μmol∙L⁻¹), x₁ is the concentration of glucuronan (%), x₂ is the flow rate (mL∙min⁻¹), and x₃ is reaction time (min); the coefficients of β_i and β_ii represent factor’s effects on response.

The coded regression coefficients and analysis of variance are provided in

Table 4. Coded regression coefficients for the response surface and their p-values.

Coefficients	Regression Coefficients	p-Values
β0	767.7	0.000
β1	209.1	0.000
β2	132.6	0.003
β3	271.3	0.000
β11	−72.7	0.058
β22	−36.3	0.295
β33	−27.3	0.424
β12	50.8	0.225
β13	120.6	0.016
β23	48.1	0.247

and

Table 5. Variance analysis of factors on response surface.

Source	Degree of Freedom	Sum Square	Mean Square	F-Value	p-Values
β0		767.7			0.000
Model	9	2,060,444	228,938	19.7	0.000
Linear	3	1,842,637	614,212	52.86	0.000
x1	1	596,922	596,922	51.37	0.000
x2	1	240,155	240,155	20.67	0.003
x3	1	1,005,561	1,005,561	86.53	0.000
Square	3	62,313	20,771	1.79	0.237
x12	1	59,518	59,518	5.12	0.058
x22	1	14,854	14,854	1.28	0.295
x32	1	8388	8388	0.72	0.424
2-Way Interaction	3	155,493	51,831	4.46	0.047
x1x2	1	20,681	20,681	1.78	0.224
x1x3	1	116,304	116,304	10.01	0.016
x2x3	1	18,508	18,508	1.59	0.247
Error	7	81,342	11,620
Lack-of-Fit	5	79,530	15,906	17.55	0.055
Pure Error	2	1813	906

. The coded coefficients help minimize multicollinearity among the factors, with a model-fitting coefficient (R²-sq) of 96.2%. The p-values of regression coefficients β₁, β₂, β₃, and β₁₃ are below 0.05, which indicates that their corresponding items significantly affect the response variable. Their effect ranking is as follows, i.e., reaction time, concentration of deacetylated glucuronan, flow rate, and the interaction of concentration of deacetylated glucuronan and reaction time. The analysis of variance shows that the linear model of x₁, x₂, x₃, and the interaction of x₁x₃ significantly affect the response (p-values < 0.05), which is consistent with the regression coefficients. The coefficients for quadratic terms and the variance of the square model are both not statistically significant (p-values > 0.05), which indicates that the response surface lacks curvature. However, we did not proceed to explore conditions beyond the current parameters, considering the capacity limits of the IMER. Additionally, we do not predict any curvature in the response surface even if parameters are exceeded in this paper.

The pareto chart of standardized effects is shown in Figure S2 (Supplementary Data), which visually displays the standardized effects more clearly, where standardized effect 2.365 corresponds to α = 0.05. The contour plots of the concentration of oligoglucuronan are shown in Figure 4. A high concentration of oligoglucuronan tends to be obtained under the operating conditions of high concentrations of deacetylated glucuronan, a high flow rate, and a long reaction time. Here, the results approach the boundary of the maximum production of oligoglucuronan under all stated continuous variables; higher parameter levels were not measured considering the operability of the CIM^® CDI disk. The optimal operating condition predicted by the model is a deacetylated glucuronan concentration of 0.18%, a flow rate of 0.36 mL∙min⁻¹, and a reaction time of 110.45 min, resulting in a maximum oligoglucuronan concentration of 2034.27 μmol∙L⁻¹, with a model-fitting coefficient of 68.5%.

3.4.2. Effects of Operating Conditions on the Degree of Polymerization of Degraded Glucuronan

Glucuronic acid was selected as the monomer of glucuronan (DP 1). The changes in glucuronan and depolymerized glucuronan distributions under different operating conditions (shown in Table 1) using the IMER are shown in Figure S3 (Supplementary Data). Based on the relationship between elution time and oligoglucuronan molecular weight, the peaks within an elution time ranging from 20 to 40 min were identified as DP 2–8. Mass spectrometry (MS^e and MS-MS) was used to validate the presence of these DPs and calibrate the attributions (Table S1, Supplementary Data). All degraded glucuronans show various peaks and intensities within the 20–40 elution times, while the 0.18% glucuronan condition did not show any peak, indicating the successful depolymerization using the IMER. For all degraded glucuronans, the peak corresponding to DP 3 exhibited the highest intensity. The reaction time influenced the occurrence of DP 2, while the flow rate did not. When 0.1% glucuronan was degraded for 9.5 min at a flow rate of 0.2 mL∙min⁻¹, no peak of DP 2 was observed. Yet, under the same conditions, when the reaction time was increased to 60 and 110.5 min, the peak of DP 2 appeared. Additionally, when 0.1% glucuronan was degraded for 60 min at flow rates of 0.03 and 0.36 mL∙min⁻¹, both conditions exhibited the peak of DP 2. By analyzing each factor, the following conclusions could be drawn, i.e., (i) when degradation occurs for 60 min at 0.2 mL∙min⁻¹, the higher the concentration of substrate (0.015, 0.1, 0.18%), the more intense the oligomer peaks; (ii) when 0.1% glucuronan was degraded for 60 min, the faster the flow rate (0.03, 0.2, 0.36 mL∙min⁻¹), the more intense the oligomer peaks; (iii) when 0.1% glucuronan was degraded at 0.2 mL∙min⁻¹, the longer the reaction time (9.5, 60, 110.5 min), the more intense the oligomer peaks. The effects of substrate concentration, flow rate, and reaction time can be attributed to the increased contact time between glucuronan and the immobilized glucuronan lyase, leading to more extensive cleavage of glucuronan. These results align with the predicted optimal operating parameters described in Section 3.4.1.

On the other hand, the maximal operating parameters are limited by the properties of the polysaccharide and the IMER system. For example, the viscosity of the glucuronan solution increases as concentration increases, which may cause diffusion problems in the CIM^® CDI disk due to the pore size of the polymer support. The flow rate is also limited by the backpressure inside the IMER system. Overall, the balance between efficiency and time cost must be considered.

According to these statements, the contributions of these factors to the production of DP 3 (the most abundant product) and DP 7 (where DP 7–10 corresponds to an interesting structural size for oligosaccharides when investigating biological activities) were examined, as shown in Figure 5. The response surface regression indicated that glucuronan concentration and reaction time significantly (p-value < 0.05) affect the production of DP 3 and DP 7 oligoglucuronan, as reported in Tables S2 and S3 (Supplementary Data). Higher glucuronan concentrations and long reaction times tend to yield more DP 3 and DP 7, with the production of DP 3 being higher than that of DP 7, resulting in two distinct 3D-mesh profiles, as reported for similar DPs in another IMER involving the production of oligodextran [15].

3.5. Storage and Operating Stability

The activity of immobilized glucuronan lyases was monitored based on the volume of substrate passed, measured in column volumes (0.2 mL) of the CIM^® CDI disk, and the number of days, as shown in Figure 6. Overall, the relative activity of immobilized glucuronan lyases decreased with the use of IMER, which may be due to the changes in enzyme conformation during use and storage. After running 1296.5 column volumes of glucuronan, the IMER retained 75.02% of its activity.

However, there was a sharp drop of 21.69% in activity when the column volume reached 1896.5, which may be attributed to the long-term storage (29 days) of the IMER. Finally, the immobilized glucuronan lyase retained 50.9% of its activity compared to the initial level. Examples such as dextranases immobilized on a CIM^® epoxy disk retaining 78% enzymatic activity after processing 5400 column volume of substrate over 78 days [15], and laccases immobilized on a CIM^® CDI disk retaining 91.29% activity after 1200 column volume of substrate [16], demonstrate the effectiveness of these supports for immobilizing enzymes. It is noteworthy that not only the volume of substrate catalyzed, but also the regular use of IMER can affect its activity.

3.6. Depolymerization and Phenolization of Glucuronan

3.6.1. Depolymerization and Phenolization of Glucuronan by the Dual-Stage IMER

As the degradation ability of the immobilized glucuronan lyase reactor has been verified in Section 3.3 and Section 3.4, the glucuronan lyase reactor was tentatively connected with an immobilized laccase reactor, which was previously investigated for adding polyphenol groups to the polysaccharide backbone in our previous research [16]. In this part, the glucuronan solution with gallic acid was continuously passed through the two connected reactors, and the mass of bonded gallic acid was quantified, as shown in Figure 7. After depolymerization and phenolization of glucuronan, the color of the solution changed from transparent to brown. This change demonstrated the function of the laccase reactor, since during the sole depolymerization by the glucuronan lyase reactor, no color change was observed. The same color change was observed using the laccase reactor alone in the production of GA–dextran conjugate in previous research [16]. The UV spectra showed the characteristic peaks of gallic acid at 212 and 260 nm at T0. As the reaction proceeded, the peak intensity at 212 nm decreased at T6h, highlighting the consumption of free gallic acid in the solution, which may have been oxidized and bonded to the glucuronan and oligoglucuronan. The ¹H NMR experiments (see Supplementary Data, Figure S4) of glucuronan and phenolized oligoglucuronan (conditions 0.15%, 0.3 mL/min, 6 h) confirmed the depolymerization of the polysaccharide, visible through the better resolution of the second spectrum and the classical attributions of H-1, H-2, H-3, H-4, and H-6 peaks [27,28]. As reported by other authors, the presence of a signal in the range of 6.2–7.1 ppm, only observed in the GA–oligoglucuronan conjugate spectrum, corresponded to the hydrogen from the aromatic ring of GA, thus confirming the conjugation of GA to the oligoglucuronan [36,37]. By measuring the consumption of free GA, an approximate amount of 8.84 ± 0.10 mg gallic acid was added onto 1 initial g of glucuronan after 6 h. This successful bonding of GA on glucuronan may impart additional biological activities to oligoglucuronan.

3.6.2. DPPH-Radical-Scavenging Ability of GA–Oligoglucuronan Conjugates

The antioxidant activity of GA–oligoglucuronan conjugates was evaluated using a DPPH-radical-scavenging assay, with glucuronan as control, as shown in Figure 8. The half-maximal inhibitory concentration (IC₅₀) of GA–oligoglucuronan was 680 ± 10 mg∙L⁻¹. Compared to classical antioxidants, like ascorbic acid (IC₅₀ = 5.39 ± 0.00 mg∙L⁻¹) and gallic acid (IC₅₀ = 1.60 ± 0.04 mg∙L⁻¹) reported in the literature [16], the IC₅₀ of the GA–oligoglucuronan conjugates is quite low. Most of the mass of these molecules consists of large-molecular-weight glucuronan (Mw = 750,000 to 1,000,000 Da) or oligoglucuronan (DP 2 to 8), which do not contribute to antioxidant activity, apart from the low antioxidant capacities of the hydroxyl groups positioned along the chains.

Additionally, the release study of GA from GA–oligo conjugates showed that no phenolic group have been released into the control solvent (96% ethanol), during 72 h of incubation, highlighting the stability of this conjugation.

4. Conclusions

Glucuronan lyases from Peteryoungia rosettiformans were produced by expressing recombinant genes in E. coli and purified using a Protino^® Ni-Ted column. An IMER was designed by immobilizing the purified glucuronan lyase on a CIM^® CDI disk. The Bradford assay showed that approximately 63.61% (800.92 μg) of glucuronan lyases were successfully immobilized. The immobilized glucuronan lyases overcome the activity inhibition in free glucuronan lyases at relatively high glucuronan concentrations (0.5–2 g∙L⁻¹).

Within the operational limits of the IMER system, high glucuronan concentration, high flow rate, and long reaction time accelerate the cleavage of glucuronan, resulting in the production of oligoglucuronan with DP ranging mainly from 2 to 8. The activity of the immobilized glucuronan lyase reactor retained 50.9% after processing 2373 column volumes of glucuronan. Note that regular use and proper storage of the IMER system may play a significant role in maintaining its activity. Subsequently, a dual-stage IMER was tentatively constructed to both depolymerize and graft GA onto glucuronan backbone. Approximately 8.84 ± 0.10 mg of GA can be bonded to 1 g of initial glucuronan and the GA–oligoglucuronan conjugates exhibit a notable DPPH-radical-scavenging ability with an IC₅₀ of 680 ± 10 mg∙L⁻¹.

Overall, the functionality of the two connected reactors has been preliminarily verified. However, a common challenge in similar reactions is the purification of products, specifically the separation of oligosaccharides with different degrees of polymerization. A separation system, such as a membrane filtration device, would be an appropriate choice for achieving product separation and is expected to be integrated into this kind of IMER system, creating an all-in-one reaction and separation process. Moving forward, we aim to incorporate a membrane filtration device into this reactor system to facilitate the production and separation processes of DP within a comprehensive IMER system, thereby enhancing the exploitation of specific bioactive oligosaccharide molecules. For instance, this integration will enable investigating structure–function relationships of specific DPs in GA–oligoglucuronan conjugates.