Polyethyleneimine modified Cry protein and application thereof

Abstract

The present application provides a system for immobilizing recombinantly produced proteins, in particular enzymes useful for catalyzing various chemical reactions, and cofactors required for the reactions, by polyethyleneimine (PEI) modified protein crystals. Related compositions and methods of making and using the immobilized proteins and cofactors are also described.

Description

[0001] Related applications

[0002] This application claims priority to U.S. Provisional Application No. 63 / 636,314, filed April 19, 2024, the contents of which are hereby incorporated by reference in their entirety for all purposes. Background of the Invention

[0003] In many medical and industrial environments, the use of recombinant proteins is becoming more and more important. Recombinant proteins, particularly the production, separation, use and potential reuse of enzymes still need to be improved to obtain better quality, efficiency and stability. For example, enzyme immobilization can help to easily remove enzymes and reuse enzymes subsequently in multiple rounds of catalytic processes. In many cases, immobilization has also improved the stability of enzymes to many industrial conditions such as high temperature and organic solvents. Fixation by embedding is particularly attractive because it does not involve any modification to the enzyme structure, increases the possibility of high activity retention and native enzyme conformation. Therefore, new and effective means are needed to produce recombinant proteins, for example enzymes, in immobilized form. The present application has achieved this need and other related needs. Summary of the Invention

[0004] In a first aspect, the present application provides polyethyleneimine (PEI)-modified protein crystals. The protein crystals can be formed by a Cry protein, a crystal-forming variant of a Cry protein, a crystal-forming fragment of a Cry protein, or a fusion polypeptide comprising the Cry protein, the crystal-forming variant, or the crystal-forming fragment.

[0005] In some embodiments, the Cry protein in the claimed PEI-modified protein crystal is Cry3Aa or CrylAb. In some embodiments, the PEI-modified protein crystal is formed from a fusion polypeptide comprising (1) a Cry protein, a crystal-forming variant of the Cry protein, or a crystal-forming fragment of the Cry protein fused to (2) a heterologous protein. In some embodiments, the PEI-modified protein crystal is formed from a Cry protein, a crystal-forming variant of the Cry protein, or a crystal-forming fragment of the Cry protein and comprises a heterologous protein encapsulated within the protein crystal. In some embodiments, the PEI-modified protein crystal comprises a fused or embedded heterologous protein that is an enzyme, such as formate dehydrogenase (FDH), leucine dehydrogenase (LDH), human arginase (hArg), ornithine decarboxylase (ODC), alcohol dehydrogenase (ADH), mandelate dehydrogenase (MDH), carbonyl reductase (CBR), or reductive aminase (RedAm) or cytochrome P450 (CYP or P450), including functional variants of any of the foregoing. In some embodiments, the fusion polypeptide in the claimed PEI-modified protein crystal comprises a Cry protein, a crystal-forming variant of the Cry protein, or a crystal-forming fragment of the Cry protein fused to an enzyme. For example, the fusion polypeptide is a Cry3Aa-ADH fusion, a Cry3Aa-FDH fusion, a Cry3Aa-LDH fusion, a Cry3Aa-MDH fusion, a Cry3Aa-P450 fusion, a CrylAb-hArg fusion, a CrylAb-ODC fusion, or a CrylAb-P450 fusion, including Cry fusions of functional variants of these enzymes. In some embodiments, the PEI-modified protein crystal is associated with a cofactor required for the chemical reaction catalyzed by the enzyme. In some embodiments, the cofactor is a phosphorylated cofactor. For example, the phosphorylated cofactor is nicotinamide adenine dinucleotide (NADH), flavin adenine dinucleotide (FADH), or pyridoxal phosphate (PLP). In some embodiments, the PEI-modified protein crystal comprises an enzyme that is ADH, FDH, LDH, MDH, CBR, RedAm, or P450 and is associated with the phosphorylated cofactor NADH. In some embodiments, the PEI-modified protein crystal comprises the enzyme, ODC, or a functional variant of ODC and is associated with the phosphorylated cofactor PLP. In some embodiments, the PEI-modified protein crystal comprises two fusion polypeptides, such as Cry3Aa-FDH and Cry3Aa-ADH; or Cry3Aa-FDH and Cry3Aa-LDH.

[0006] In a second aspect, the application provides a composition comprising a PEI-modified protein crystal described above and herein. In some embodiments, the protein crystal is embedded in agarose, e.g., agarose beads. In some embodiments, the composition comprises a PEI-modified protein crystal comprising an enzyme capable of catalyzing a substrate in a chemical reaction and a co-factor required for the reaction, and wherein the composition further comprises the substrate. For example, the protein crystal is formed from a fusion polypeptide comprising a Cry protein fused to ADH, FDH, LDH, MDH, or P450. In some embodiments, the PEI-modified protein crystal is formed from a fusion polypeptide comprising a Cry protein fused to hArg or ODC. In some embodiments, the PEI-modified protein crystal is formed from two fusion polypeptides, e.g., Cry3Aa-FDH and Cry3Aa-ADH; or Cry3Aa-FDH and Cry3Aa-LDH. In some embodiments, the PEI-modified protein crystal binds to NADH. In some embodiments, the PEI-modified protein crystal binds to PLP.

[0007] In a third aspect, the application provides a method of performing a chemical reaction. The method comprises the step of incubating a PEI-modified protein crystal of the application as described above and herein (which comprises an enzyme) with a substrate for the enzyme under conditions that allow the substrate to be catalyzed by the enzyme in the chemical reaction. For example, the enzyme is formate dehydrogenase (FDH), leucine dehydrogenase (LDH), human arginase (hArg), ornithine decarboxylase (ODC), alcohol dehydrogenase (ADH), mandelate dehydrogenase (MDH), carbonyl reductase (CBR), or reductive aminase (RedAm) or cytochrome P450 (CYP or P450), including functional variants of any of the foregoing. In some embodiments, the PEI-modified protein crystal comprises an enzyme that is ADH, FDH, LDH, MDH, CBR, RedAm, or P450, including functional variants thereof, and the protein crystal binds to NADH. In some embodiments, the PEI-modified protein crystal comprises the enzyme, ODC, or a functional variant thereof, and the protein crystal binds to PLP. In some embodiments, the PEI-modified protein crystal comprises two fusion polypeptides, e.g., Cry3Aa-FDH and Cry3Aa-ADH, or Cry3Aa-FDH and Cry3Aa-LDH. In some embodiments, the method further comprises the step of, after the reaction is complete, removing the reaction product and reusing the crystal in a subsequent reaction.

[0008] In a fourth aspect, the present application provides an in vitro method for immobilizing a co-factor by incubating said co-factor with a PEI-modified protein crystal of the present application as described above and herein in solution and allowing said co-factor to bind to said protein crystal. In some embodiments, said co-factor is NADH or PLP. In some embodiments, the method further comprises, after said co-factor binds to said protein crystal, a step of shaking or stirring or otherwise agitating (e.g., by mechanical or manual means) said solution to release said co-factor from said protein crystal. In some embodiments, the method further comprises, after said co-factor is released from said protein crystal, a step of stopping shaking or stirring or otherwise agitating said solution to allow said co-factor to bind again to said protein crystal.

[0009] In a fifth aspect, the present application provides a method of delivering a protein of interest to a target cell by contacting said target cell with a pre-formed complex between said protein of interest and a PEI-modified protein crystal as described herein (in particular in the first aspect of the present application as described above). In some embodiments, said protein crystal is a PEI-modified Cry3Aa protein crystal. In some embodiments, said protein of interest is a negatively charged protein or a protein comprising a negatively charged domain or linked to a negatively charged moiety, such as a Cas9 ribonucleoprotein (RNP) pre-loaded with at least one small guide RNA (sgRNA). In some embodiments, such a delivery method further comprises a step of incubating said protein of interest with said PEI-modified protein crystal prior to said contacting step to allow them to bind to each other to form a complex. In some embodiments, said protein of interest is a nucleoprotein, and wherein said contacting step is performed in the presence of an endosomolytic agent, such as chloroquine. In some embodiments, said method is a method for in vitro use. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 , PEI-mediated Cry3Aa crystal cationization. Figure 1 (A) shows the distribution of negatively charged residues (Asp + Glu) in red and large volume shape throughout the Cry3Aa crystal (PDB code: 1DLC). Figure 1 (B) represents the coupling of EDC / NHS-mediated PEI to carboxylic acids present in the Cry3Aa crystal (Asp, Glu and C-terminal).

[0011] Figure 2 , characterization of 3A-PEI crystals: Figure 2(A-D), SEM imaging of Cry3Aa (a) and 3A-PEI (b) crystals at low magnification (5000x), SEM imaging of (c) single Cry3Aa crystal and (d) single 3A-PEI crystal at 150000x magnification. Figure 2 (E) Size distribution of Cry3Aa crystals before and after PEI-mediated cationization and Figure 2 (F) Zeta potential of Cry3Aa crystals before and after PEI-mediated cationization using MALVERN Zetasizer Nano.

[0012] Figure 3 Binding and retention of co-factors in 3A-PEI crystals: Figure 3 A shows NADH binding capacity of 3A-PEI, Cry3Aa and Pos3Aa crystals. Figure 3 B shows retention of NADH bound to 3A-PEI crystals after multiple wash cycles with NaPB (10 mM, pH: 7.0). Finally, NADH bound to 3A-PEI crystals was completely removed using 1 M NaPB (pH: 7.0). Figure 3 C represents binding capacity of co-factors (PLP, NAD + , NADP + , NADPH, FADH2, CoA, riboflavin and TPP) compared to NADH. Among the co-factors, PLP showed the strongest binding affinity to 3A-PEI crystals (650 ± 2 μmol / g support) while TPP showed the weakest binding affinity to 3A-PEI crystals (160 ± 1 μmol / g support).

[0013] Figure 4 Confocal fluorescence microscopy of NADH-loaded 3A-PEI crystals. Cry3Aa protein was labeled with Alexa568 (red), PEI polymer was labeled with Alexa488 (green) and bound NADH molecules showed blue fluorescence. Crystals were grown under in vitro conditions.

[0014] Figure 5 Effect of pH and ionic strength on NADH binding to 3A-PEI crystals. Figure 5 A represents the effect of pH on NADH binding capacity of 3A-PEI crystals in 25 mM Na-PB. Figure 5 B shows the effect of different concentrations of NaCl (0-800 mM) in Na-PB buffer (25 mM, pH: 7.0) on the release rate of NADH from 3A-PEI crystals.

[0015] Figure 6 Effect of orbital shaking on cofactor binding and release in 3A-PEI crystals. Figure 6 A shows the effect of different mechanical agitation (0-2000 RPM) on the release rate of NADH from 3A PEI crystals in different solutions. The buffer concentration of all buffers was normalized to 10 mM and the pH was adjusted to 7.0, except for ammonium formate which was adjusted to a pH of 8.0. It has been shown that by increasing the mechanical agitation, the NADH release rate increases significantly, while when no mechanical agitation is applied, about 15% release can be seen for phosphate based buffers, and close to 10% release for HEPES and ammonium formate buffers, due to leaching. Figure 6 B represents the release of phosphorylated cofactor from 3A-PEI crystals into solution in sodium phosphate buffer (10 mM, pH 7.0) under mechanical shaking (1000 rpm). Figure 6 C shows the correlation of relative force to (K d ) 2 The results show that in Na-PB buffer (10 mM, pH: 7.0), when the relative force reaches almost 1.8 mN, 50% of the bound NADH molecules can be released from 3A PEI crystals, while when the force is removed, the value of (K d ) 2 decreases significantly, indicating the correlation. Figure 6 D illustrates the rebinding of NADH molecules in the same solution after eliminating the mechanical force by reducing the shaking speed from 1000 rpm to 0 rpm. The release and rebinding of NADH molecules can be adjusted by the shaking speed, but the effect of temperature on the binding stability should not be neglected. Figure 6 E represents the rechargeability of 3A-PEI crystals after completely eliminating the bound NADH molecules by using high concentrations of salt.

[0016] Figure 7 Comparison of NADH binding capacity to different SB-PEI beads. Figure 7 A represents the binding capacity of NADH to PEI-SB1, PEI-SB2 and PEI-SB3 beads with the same amount of anchoring PEI. It is clear that PEI-SB3 beads have almost twice as high NADH binding capacity compared to the other beads. Figure 7 B shows the relationship between PEI anchored to SB and their NADH immobilization capacity. By increasing the amount of bound PEI from 0 mg / g support to 59 mg / g support, NADH binding increases significantly, and by further increase (more than 60 mg / g) NADH immobilization can decrease as the volume occupied by NADH can be less.

[0017] Figure 8It is illustrated that NADH molecules are regenerated by oxidizing formic acid to CO2.

[0018] Figure 9 Recyclability of 3AFL + 3A-PEI-NADH crystals for L-tert-leucine biosynthesis. 3AFL with free NADH is represented by striped bars, while 3AFL + 3APEI-NADH mixture is represented by solid bars. Reactions were carried out in a thermomixer (Eppendorf) at 25 °C in a reaction buffer containing 25 mM TMP in 30 mM ammonium formate (pH: 8.0) under mechanical stirring (1000 RPM), followed by resting at 10 °C for 30 min. The amount of NADH added to each cycle in 3AFL reactions was 0.5 pmol, while for 3AFL + 3APEI-NADH mixture, this value was 0.5 pmol in the 1stcycle and 0.2 pmol in the 7thand 14thcycles. C) in a reaction buffer containing 25 mM TMP in 30 mM ammonium formate (pH: 8.0) at 25 °C under mechanical stirring (1000 RPM), followed by resting at 10 °C for 30 min. The amount of NADH added to each cycle in 3AFL reactions was 0.5 pmol, while for 3AFL + 3APEI-NADH mixture, this value was 0.5 pmol in the 1stcycle and 0.2 pmol in the 7thand 14thcycles.

[0019] Figure 10 Scheme of hArg and ODC-mediated conversion of L-arginine to putrescine. In this pathway, hArg acts to convert L-arginine to produce L-ornithine, which is then converted to putrescine by PLP-dependent ODC.

[0020] Figure 11 A-B, Characterization of co-immobilization of GFP and mCherry mediated by Cry1Ab. Figure 11 A: Confocal microscopy was performed to examine the fluorescence signal distribution of co-expressed [GFP-Cry1Ab / mCherry-Cry1Ab] particles (upper panel) and a mixture of separately immobilized GFP-Cry1Ab and mCherry-Cry1Ab particles (lower panel). Yellow small spots indicate the overlap of GFP and mCherry signals. Figure 11 B: Corresponding fluorescence intensity distribution curves of co-expressed [GFP-Cry1Ab / mCherry-Cry1Ab] particles (left) and a mixture of separately immobilized GFP-Cry1Ab and mCherry-Cry1Ab particles (right) were measured along solid magenta lines a and b in (A), respectively, using ImageJ software. Green curve and red curve represent GFP signal and mCherry signal, respectively. Green: GFP-Cry1Ab particles; Red: mCherry-Cry1Ab particles.

[0021] Figure 12 C: Activity of separately immobilized particles mediated by Cry1Ab. Figure 12 A: ODC-Cry1Ab and Cry1Ab-ODC as well as ODC mutActivity comparison of Cry1Ab particles. 1 mM of Cry1Ab ODC fusion particles (0.21 mg / mL) or ODC enzyme (0.079 mg / mL) were used in the reaction. The activity of ODC was based on the consumption of L-ornithine. The activity of free ODC and Cry1Ab were considered as positive and negative controls. Double mutations (I163T / E165T) were introduced into wild-type ODC by site-directed mutagenesis to generate ODC mut Cry1Ab. Figure 12 B: Activity of hArg-Cry1Ab particles. 1 mM of hArg-Cry1Ab fusion particles (0.165 mg / mL) or hArg enzyme (0.035 mg / mL) were used in the reaction. Their hArg activity was evaluated based on the amount of urea produced. The activity of free hArg and Cry1Ab were considered as positive and negative controls. All experiments were performed in triplicate. Data are represented as mean ± standard deviation.

[0022] Figure 13 Characterization of Cry1Ab immobilized enzyme. Figure 13 A: DLS of Cry1Ab immobilized particles. 25 pg / mL of particles resuspended in aqueous solution were used for the measurement. Figure 13 B: SEM of Cry1Ab immobilized particles at 12,000x (top panel) and 80,000x magnification (bottom panel). The scale bars of the top and bottom panels are 10 pm and 1 pm, respectively.

[0023] Figure 14 Activity of co-immobilized particles. Figure 14 A: Determination of putrescine produced by indirect method. Co-immobilized [hArg-Cry1Ab / ODC mut Cry1Ab] particles were incubated with 30 mM L-arginine for 20 min at 40 °C. The supernatant was used to determine the urea produced and the remaining L-ornithine as previously described. Co-immobilized [hArg-Cry1Ab / ODC mut Cry1Ab] particles were incubated with 30 mM L-arginine for 20 min at 40 °C. The supernatant was used to determine the urea produced and the remaining L-ornithine as previously described. Co-immobilized [hArg-Cry1Ab / ODC mut Activity of hArg and ODC in co-immobilized [hArg-Cry1Ab / ODC Figure 14 B: Direct detection of putrescine produced by HPLC method. The reaction solution used in (A) was derivatized with OPA and then analyzed by HPLC. At 0 min and 20 min, the samples reached a peak at 8.8 min and 23.4 min, which corresponded to the elution profile of L-arginine and putrescine standards, respectively.

[0024] Figure 15, comparison of the enzymatic activity of the immobilized particles. At 40°C, 30 mM L-arginine was used as substrate for 0.5 mg / mL co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles (red squares), 0.27 mg / mL hArg-CrylAb and 0.23 mg / mL ODC mut -CrylAb particles (blue circles) and 0.057 mg / mL hArg and 0.087 mg / mL ODC mut mixture of enzymes (green triangles). Activity was based on the production of putrescine, which was determined by an indirect method as described before. All experiments were performed in triplicate. Data are represented as mean ± standard deviation.

[0025] Figure 16 Thermal stability of CrylAb immobilized particles and free enzymes. Figure 16 A: Thermal stability of hArg enzyme and hArg-CrylAb particles. 1 μΜ hArg enzyme (0.035 mg / mL) and hArg-CrylAb fusion particles (0.165 mg / mL) were incubated at 50°C for 1 h and then residual activity was measured. Residual activity at 50°C was considered as 100%. Figure 16 B: Thermal stability of ODC and ODC mut enzymes and their CrylAb immobilized counterparts. In the study, 1 μΜ ODC and ODC mut enzymes (0.079 mg / mL) and their CrylAb immobilized counterparts (0.21 mg / mL) were used. Residual activity at 30°C was considered as 100%. All data were fitted to a sigmoidal model using GraphPad Prism software. All experiments were performed in triplicate. Data are represented as mean ± standard deviation.

[0026] Figure 17 Recyclability of co-immobilized multi-enzyme or single enzyme and their cofactors. Figure 17 A: Recyclability of CrylAb mediated co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles. In each cycle, 2 mg / mL particles and 30 mM L-arginine in 50 mM HEPES buffer (pH 8.0) containing 0.1 mM PLP were incubated at 40°C and 750 rpm for 30 min. After 12 cycles, the total turnover number (TTN) of the rate-limiting enzyme, ODC mut reached 73284, indicating that CrylAb mediated co-immobilized particles are highly recyclable. Figure 17 B: PEI modified ODCmut -Recyclability of Cry1Ab. mut -Cry1Ab particles, bound to 0.14 μmol of PLP before recycling studies, whereas for the control group (ODC mut -CrylAb+PLP), in each cycle, 0.14 μmol of PLP was added to a reaction solution containing 4 mg / mL particles and 10 mM L-ornithine in 10 mM HEPES buffer (pH 7.0). After 25 cycles, the TTN of PLP for PEI-modified particles and unmodified particles was 1736 and 71, respectively, indicating that the particles after PEI modification are cost-effective and highly reusable. All experiments were performed in triplicate. Data are expressed as mean ± standard deviation.

[0027] Figure 18 , FTIR of the produced nylon. The in situ produced nylon was washed and vacuum dried before FTIR analysis. At 3300 cm -1 、2940cm -1 and 2870cm -1 、1640cm -1 The absorption peaks at 1540 cm indicate the presence of NH, CH, and C=O. -1 The peak at corresponds to CN stretching vibration, indicating that a chemical reaction has occurred between putrescine and adipoyl chloride.

[0028] Figure 19 Characterization of PEI-3A-FDH / 3A-LDH particles. SEM images of 3A-FDH / 3A-LDH particles and PEI-3A-FDH / 3A-LDH particles were taken at 10,000× (A, B) and 150,000× (C, D), respectively. (E) Size distribution and (F) zeta potential of 3A-FDH / 3A-LDH particles before (dark blue) and after (red) PEI cationization.

[0029] Figure 20 , Co-immobilization and retention of NADH in PEI-3A-FDH / 3A-LDH. Figure 20 A: Confocal image of fluorescent dye-labeled PEI-3A-FDH / 3A-LDH-NADH particles. The red signal represents Alexa 568-labeled 3A-FDH / 3A-LDH particles. NADH exhibits autofluorescence in the DAPI channel. Figure 20 B: L-tert-Leucine production by 3A-FDH / 3A-LDH-NADH particles before and after modification with PEI. Figure 20C: Recyclability of PEI-3A-FDH / 3A-LDH particles and 3A-FDH / 3A-LDH that only provides NADH at the beginning of the first cycle.

[0030] Figure 21 Confocal microscopy analysis of fluorescently labeled PEI-3A-FDH / 3A-LDH-NADH-encapsulated agarose beads. Alexa 568-labeled PEI-3A-FDH / 3A-LDH particles exhibit uniform distribution throughout the beads. NADH molecules observed in the DAPI channel overlap with particle fluorescence, indicating their immobilization within the PEI-3A-FDH / 3A-LDH particles.

[0031] Figure 22 , Biosynthesis of L-tert-leucine in a continuous flow reactor. Figure 22 A: Catalytic conversion of TMP to L-tert-leucine using PEI-3A-FDH / 3A-LDH particles embedded in agarose beads. The FDH assembly serves as an NADH regenerator. Figure 22 B: Changes in the conversion rate of PEI-3A-FDH / 3A-LDH-encapsulated agarose beads over time. The beads were initially loaded with 2 μmol of NADH, and then 0.2 μmol of NADH was added every 5 days. Figure 22 C:TTN NADH Gradual accumulation as a function of time. TTN NADH The values ​​decreased slightly on days 6, 11, 16, 21, and 26 due to the addition of exogenous NADH but then continued to increase over the next 5 days. Figure 22 D: Illustrative schematic diagram of the continuous flow reaction system. Figure 22 E: PEI-3A-FDH / 3A-LDH agarose beads used for continuous flow biosynthesis.

[0032] Figure 23 , Preparation of free CadA and CadA-Cry1Ab. Figure 23 A: Nickel resin column purification of free CadA. Lane 1, flow-through; lanes 2-6, wash with 0-50 mM imidazole; lane 7, elution with 100 mM imidazole; lanes 8-10, elution with 250 mM imidazole. Figure 23 B: SDS-PAGE of purified CadA-Cry1Ab particles.

[0033] Figure 24A and 24B , CadA( Figure 24A ) and CadA-Cry1Ab( Figure 24B ) Orbitrap mass spectrometry.

[0034] Figure 25 SEC of free CadA. Figure 25 A: SEC of standards with different molecular weights. Figure 25 B: Standard curve of molecular weight vs. retention volume obtained from A. Figure 25 C: SEC of CadA.

[0035] Figure 26 L-lysine conversion by CadA-CrylAb. The conversion of CadA was taken as a positive control.

[0036] Figure 27 Thermal stability of CadA-CrylAb and free CadA. The residual activity at 25 °C was taken as 100%.

[0037] Figure 28 Optimization of CadA and CadA-CrylAb reaction pH. The activity of CadA and CadA-CrylAb reached a maximum at pH 6, which was taken as 100%.

[0038] Figure 29 Lifetime of CadA-CrylAb particles at pH 6 and pH 7.

[0039] Figure 30 Recyclability of CadA-CrylAb particles at high concentrations of L-lysine substrate. In each cycle, 45 mg of particles were reacted with 1 M L-lysine HC1 in 10 mL reaction buffer (1 M MES, 0.1 mM PLP, pH 6.0) at 37 °C for 1 h.

[0040] Figure 31 Binding capacity of CadA-CrylAb before and after PEI modification. The maximum PLP binding capacity of PEI-modified particles was estimated to be about 1300 pmol / g, while unmodified particles exhibited a binding capacity of about 100 pmol / g.

[0041] Figure 32 Recyclability of PEI-modified CadA-CrylAb particles. PEI-CadA-CrylAb particles efficiently converted L-lysine to cadaverine, with only 0.1 pmol of PLP added prior to the experiment. In contrast, unmodified CadA-CrylAb particles required a continuous addition of 0.1 pmol of PLP before each cycle to achieve 100% conversion and did not work in the absence of PLP.

[0042] Figure 33ζ-potential of CadA-CrylAb before and after PEI modification and upon PLP binding. The ζ-potential of CadA-CrylAb particles after PEI modification shifted significantly from -50 mV to +54 mV. When PLP was bound to PEI-CadA-CrylAb particles, the ζ-potential decreased to +2 mV, indicating electrostatic interactions between PLP and PEI-modified particles.

[0043] Figure 34 Hydrodynamic size distribution of CadA-CrylAb particles before and after PEI modification. Upon PEI binding, the particle size increased slightly from 485 nm to 520 nm, which can be due to the attachment of PEI molecules to the particle surface.

[0044] Figure 35 Confocal microscopy analysis of agarose bead-embedded PEI-CadA-CrylAb-PLP. Particles were labeled with Alexa 568, while PLP molecules exhibited intrinsic fluorescence in the DAPI channel. Imaging revealed uniform distribution of particles within agarose beads and efficient PLP immobilization.

[0045] Figure 36 Conversion rate comparison between CadA-CrylAb, PEI-CadA-CrylAb, agarose bead-embedded PEI-CadA-CrylAb, and physically disrupted beads. The conversion rate of CadA-CrylAb was considered 100%.

[0046] Figure 37 Continuous flow biosynthesis of cadaverine using PEI-CadA-CrylAb-agarose beads. At the start of the experiment, only 4 μmol of PLP was used, and cadaverine was continuously produced over 19 days without the need for exogenous PLP. The flow rate was set to 50 μL / min, and the reaction buffer consisted of 50 mM L-lysine in 100 mM MES buffer (pH 6.0). The product was collected and measured every 24 hours. A total of 75 mg of particles was used. After 19 cycles, a decrease in activity of about 15% was observed. Upon addition of 1 μmol of PLP, the enzyme activity did not recover effectively and gradually decreased, indicating enzyme inactivation after about 3 weeks.

[0047] Figure 38A, RNP binding to PEI-modified Cry3Aa. RNP was incubated with Cry3Aa or PEI-modified Cry3Aa (PEI-3A) or Pos3Aa crystals in OptiMEM without serum at 4°C. The reaction mixture was centrifuged to separate the pellet (p) containing RNP / crystal complex from the supernatant (sn). The pellet was washed once with OptiMEM and resuspended in 20 μΐ of OptiMEM for subsequent SDS-PAGE analysis. Cas9 protein without gRNA complexation was set up at the same time for comparison and as a control. Figure 38 B, Intracellular delivery of RNP mediated by PEI3A in MDA-MB-231. 37.5 nM equivalent of Cas9 (1 : 1.1 molar ratio of Cas9:sgRNA) in RNP was used for all experimental groups. MDA-MB-231 cells in serum-free DMEM supplemented with 1% P / S and 10 μg / mL gentamicin were treated with PEI3A / RNP and RNP mixed in Lipofectamine (L2K / RNP) for 4 h. The samples were then fixed with 4% PFA and blocked with 5% FBS before incubation with anti-Cas9, followed by secondary antibody Alexa-488 anti-MS IgG (H+L) overnight. Cell nuclei were stained using Hoechst 33342. L2K represents Lipofectamine 2000.

[0048] Figure 39 A, Effect of chloroquine on PEI3A / RNP gene editing efficiency in HEK293T. Editing efficiency evaluation of PEI3A-mediated gene editing component delivery in the presence of 60 μΜ chloroquine. After a 1 h incubation with 60 μΜ chloroquine, fresh serum-free DMEM was used to serially dilute the media, and then PEI / RNP was added to the cells. After a 4 h incubation, the FBS concentration was adjusted to 10%, without removing PEI3A / RNP. The media was further diluted with regular media after 18 h. Editing efficiency was detected using T7E1 DNA mismatch cleavage assay. Figure 39 B, Optimization of PEI3A / RNP combined with chloroquine for gene editing efficiency in HEK293T. Investigation of PEI3A:RNP at different molar ratios for gene editing efficiency in the presence of 60 μΜ chloroquine. RNP was fixed at 1 pmol and incubated with different concentrations of PEI3A. The resulting PEI3A / RNP was then added to HEK293T cells pretreated with 60 μΜ chloroquine for 1 h. Editing efficiency was detected using T7E1 assay.

[0049] Figure 40X-ray analysis of PEI-3A crystals. The structure of PEI-3A was superimposed onto the originally reported Cry3Aa structure (PDB ID: 1DLC) using the ChimeraX program to an RMSD of 0.586 across all 584 pairs. The red color in residues between 290-294 shows significant differences, which can imply some structural variation, flexibility, or distinction between the two structures.

[0050] Definitions

[0051] As used herein, the term "Cry protein" refers to any one of a class of crystalline proteins produced by Bacillus thuringiensis (Bt) strains. Some examples of "Cry proteins" include, but are not limited to, Cry1Aa, Cry1Ab, Cry2Aa, Cry3Aa, Cry4Aa, Cry4Ba, Cry11Aa, Cry11Ba, and Cry19Aa. Their amino acid sequences and polynucleotide coding sequences are known and can be found in publications such as U.S. Patent Application Publication No. US 2010 / 0322977. Their GenBank Accession Numbers are:

[0052] Cry1Aa AY197341.1

[0053] Cry1Ab AY847289.1

[0054] Cry2Aa AF273218.1

[0055] Cry3Aa AJ237900.1

[0056] Cry4Aa AB513706.1

[0057] Cry4Ba AB161456.1

[0058] Cry11Aa AL731825.1

[0059] Cry11Ba LC153032.1

[0060] Cry19Aa Y07603.1.

[0061] In addition to wild-type Cry proteins, the term "Cry protein" also includes functional variants that (1) share at least 80%, 81%, 82%, 83%, 84%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 99.5% amino acid sequence identity with the polypeptide sequence of any one of the Cry proteins listed in US2010 / 0322977; and (2) retain the ability to spontaneously form crystals in the host cell, as can be confirmed by known methods, such as electron micrographs (see, for example, Park et al., Appl Environ Microbiol, 1998, 64, 3932-3938; Schnepf et al., Microbiol Mol Biol Rev, 1998, 62, 775-806; Whiteley and Schnepf, Annu Rev Microbiol, 1986, 40, 549-576; and Nair et al., PLoS One, 2015, 10, e0127669). For example, “Cry protein” includes a Cry protein domain II (e.g., the 295 to 499 segment of SEQ ID NO: 4, and the corresponding segments in other known Cry proteins, e.g., those of WO2020 / 216322) replaced by negatively charged amino acids such as aspartic acid (D) and glutamic acid (E). Figure 27 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more positively charged or neutral amino acids within the Cry3Aa residue (e.g., those shown in ), any variant that imparts an increased negative charge to the resulting protein. An example of such a mutant is NegCry3Aa, which is a Cry3Aa variant containing the following mutations: K384E, N391D, N395D, S425E, Q430E, TQ436437EE, KR442443EE, T461D and K467E. See SEQ ID NO: 6 in WO2020 / 216322. The term "Cry protein" also includes modified Cry proteins, which are described in WO2020 / 216322. Figure 22 The 8 specific areas shown in and in WO2020 / 211782 Figure 18One or more of the regions in their respective regions in other Cry proteins shown in Table 1 (e.g., at least 1, 2, 3, 4, 5, 6, 7, or all 8 regions) comprising an amino acid insertion or deletion (e.g., an insertion of at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids, or a deletion of at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more amino acids) can be effective to respectively reduce or enlarge the channel size of the protein, and thus can result in the crystal-forming protein variant to more effectively entrap the target protein (e.g., enzyme) with increased retention and decreased diffusion of the target protein. Typically, such channel size-altered Cry protein mutants contain at least 3, 4, or 5 but no more than 10, 12, or 15 point mutations (deletions or insertions) throughout the protein. One example of such a Cry3Aa mutant of this nature is referred to as 3A2-2 (SEQ ID NO: 7 in WO2020 / 216322). In addition, the term “Cry protein” also includes Cry protein mutants that have been modified at one or more residues to improve the stability of the resulting protein or its crystal structure, e.g., by inserting or substituting one or more cysteine residues to form one or more additional disulfide bonds to stabilize the protein or crystal structure. One such example is a double mutant of Cry3Aa (3AS145C, H161R) having the amino acid sequence of SEQ ID NO: 9 shown in WO2020 / 216322.

[0062] Similarly, a "crystal forming fragment" of a Cry protein is any fragment of a known Cry protein (i.e., less than the full length of a wild-type Cry protein) that still retains the ability to self-crystallize, as evidenced by crystallization of the fragment alone and by causing a fusion protein to self-crystallize when the fragment is present in the fusion protein with another protein of interest (e.g., an enzyme). In addition to being a truncated form of a Cry protein, a "crystal forming fragment" can also include one or more modifications, such as insertions, deletions, or substitutions, to the native amino acid sequence, especially conservative modifications, such that the resulting "crystal forming fragment" shares at least 80%, 81%, 82%, 83%, 84%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 99.5% amino acid sequence identity to the polypeptide sequence of the corresponding fragment of a wild-type Cry protein. Exemplary crystal forming fragments of Cry proteins have been described in earlier publications, such as WO2018 / 028371, WO2020 / 211782, and WO2020 / 216322. In the present disclosure, the term "protein crystal" is used to describe an insoluble protein aggregate, which can involve one or two or more different proteins (e.g., as part of a fusion protein), at least one of which comprises a crystal forming protein or a crystal forming fragment thereof. The term "protein crystal" does not require that any particular crystal structure be present in an organized manner.

[0063] The term "nucleic acid" or "polynucleotide" refers to deoxyribonucleic acid (DNA) or ribonucleic acid (RNA) in either single- or double-stranded form, and polymers thereof. Unless specifically limited, the term encompasses nucleic acids containing known analogues of natural nucleotides that have similar binding properties to the reference nucleic acid and are metabolized in a manner similar to naturally occurring nucleotides. Unless otherwise indicated, a particular nucleic acid sequence also implicitly encompasses conservatively modified variants thereof (e.g., degenerate codon substitutions), alleles, orthologs, SNPs, and complementary sequences as well as the sequence explicitly indicated. Specifically, degenerate codon substitutions can be achieved by generating sequences in which the third position of one or more selected (or all) codons is substituted with a mixture of bases and / or a deoxyinosine residue (Batzer et al., Nucleic Acid Res. 19:5081 (1991); Ohtsuka et al., J. Biol. Chem. 260:2605-2608 (1985); and Rossolini et al., Mol. Cell. Probes 8:91-98 (1994)). The term "nucleic acid" can be used interchangeably with gene, cDNA, and mRNA encoded by a gene.

[0064] The term "gene" means a segment of DNA involved in producing a polypeptide chain. It can include regions before and after the coding region (leader and trailer) as well as intervening sequences between individual coding regions (introns).

[0065] The term "amino acid" refers to naturally occurring and synthetic amino acids, as well as amino acid analogs and amino acid mimetics that function in a manner similar to the naturally occurring amino acids. Naturally occurring amino acids are those encoded by the genetic code, such as those listed above, as well as those modified after translation, such as hydroxyproline, gamma-carboxyglutamate, and O-phosphoserine. Amino acid analogs refer to compounds that have the same basic chemical structure (i.e., an alpha carbon bonded to a hydrogen, a carboxyl group, an amino group, and an R group) as a naturally occurring amino acid, but having modified R group, such as homoserine, norleucine, methionine sulfoxide, methionine methylsulfonium. Such analogs have the same basic chemical structure of a naturally occurring amino acid, but have modified R groups (e.g., norleucine) or modified peptide backbone, but otherwise function in a manner similar to a naturally occurring amino acid. "Amino acid mimetics" refers to chemical compounds that have a structure that is different from the general chemical structure of an amino acid, but that functions in a manner similar to a naturally occurring amino acid.

[0066] There are a variety of methods known in the art that allow for the site-specific incorporation of non-natural amino acid derivatives or analogs into a polypeptide chain, see, e.g., WO 02 / 086075.

[0067] Amino acids can be referred to herein by either their commonly known three letter symbols or by the one-letter symbols recommended by the IUPAC-IUB Biochemical Nomenclature Commission. Nucleotides, likewise, can be referred to by their commonly accepted single-letter codes.

[0068] "Conservatively modified variants" applies to both amino acid and nucleic acid sequences. With respect to particular nucleic acid sequences, "conservatively modified variants" refers to those nucleic acids that encode identical or essentially identical amino acid sequences, or where the nucleic acid does not encode an amino acid sequence, to essentially identical sequences. Because of the degeneracy of the genetic code, a large number of functionally identical nucleic acids encode any given protein. For instance, the codons GCA, GCC, GCG and GCU all encode the amino acid alanine. Thus, at every position where an alanine is specified by a codon, the codon can be altered to any of the corresponding codons described without altering the encoded polypeptide. Such nucleic acid variations are "silent variations," which are one species of conservatively modified variations. Every nucleic acid sequence herein that encodes a polypeptide also describes every possible silent variation of the nucleic acid. One of skill will recognize that each codon in a nucleic acid (except AUG, which is ordinarily the only codon for methionine, and TGG which is ordinarily the only codon for tryptophan) can be modified to yield a functionally identical molecule. Accordingly, each silent variation of a nucleic acid that encodes a polypeptide is implicit to each described sequence.

[0069] As to amino acid sequences, the skilled artisan will recognize that individual substitutions, deletions or additions to a nucleotide, peptide, polypeptide, or protein sequence (that alter, add or delete a single amino acid or a small percentage of amino acids in the encoded sequence) are "conservatively modified variations" where the alteration results in the substitution of an amino acid with a chemically similar amino acid. Conservative substitution tables providing functionally similar amino acids are well known in the art. Such conservatively modified variations are in addition to the polymorphic variants, interspecies homologs, and alleles of the application, and are not precluded from the scope of the application.

[0070] The following eight groups each contain amino acids that are conservative substitutions for one another:

[0071] 1) Alanine (A), Glycine (G);

[0072] 2) Aspartic acid (D), Glutamic acid (E);

[0073] 3) Asparagine (N), Glutamine (Q);

[0074] 4) Arginine (R), Lysine (K);

[0075] 5) Isoleucine (I), Leucine (L), Methionine (M), Valine (V);

[0076] 6) Phenylalanine (F), Tyrosine (Y), Tryptophan (W);

[0077] 7) Serine (S), Threonine (T); and

[0078] 8) Cysteine (C), Methionine (M)

[0079] (See, e.g., Creighton, Proteins, W. H. Freeman and Co., N.Y. (1984)).

[0080] Amino acids can be referred to herein by either their commonly accepted single-letter codes or by the IUPAC-IUB Biochemical Nomenclature Commission recommended three-letter codes. Likewise, nucleotides can be referred to by their commonly accepted single-letter codes.

[0081] In the present application, amino acid residues are numbered according to their relative position from the left-most residue (numbered 1) in the unmodified wild-type polypeptide sequence.

[0082] As used herein, in the context of describing two or more polynucleotide or amino acid sequences, the terms "identical" or percent "identity" mean that two or more sequences or subsequences that are the same or have a specified percentage of amino acid residues or nucleotides that are the same, when aligned for maximum correspondence over a comparison window or designated region, as measured using one of the following sequence comparison algorithms or by manual alignment and visual inspection (e.g., a crystal forming fragment of a Cry protein or Cry protein sequence included in a fusion protein of the present application has at least 80% identity, preferably 85%, 90%, 91%, 92%, 93, 94%, 95%, 96%, 97%, 98%, 99%, or 100% identity to a reference sequence, e.g., the amino acid sequence of the corresponding wild-type Cry protein or fragment). These sequences are then said to be "substantially identical." With respect to polynucleotide sequences, the definition also includes the complement of the test sequence. Preferably, the identity exists over a region that is at least about 50 amino acids or nucleotides in length, or more preferably over a region that is 75-100 amino acids or nucleotides in length.

[0083] For sequence comparison, typically one sequence acts as the reference sequence, to which test sequences are compared. When using a sequence comparison algorithm, test and reference sequences are entered into a computer, subsequence coordinates are designated if necessary, and sequence algorithm program parameters are designated. Default program parameters can be used, or alternative parameters can be designated. The sequence comparison algorithm then calculates the percent sequence identities for the test sequences relative to the reference sequence, based on the program parameters. For sequence comparison of nucleic acids and proteins, BLAST and BLAST 2.0 algorithms are used with the default parameters, as discussed below.

[0084] As used herein, "comparison window" includes reference to a segment of any one of the multiple contiguous positions selected from between 20 to 600, typically about 50 to about 200, more typically about 100 to about 150, wherein the sequence can be compared to a reference sequence of the same number of contiguous positions after the two sequences are optimally aligned. Methods of sequence alignment for comparison are well known in the art. Optimal alignment of sequences for comparison can be conducted by, for example, the local homology algorithm of Smith & Waterman, Adv. Appl. Math. 2:482 (1981), by the homology alignment algorithm of Needleman & Wunsch, J. Mol. Biol. 48:443 (1970), by the search for similarity method of Pearson & Lipman, Proc. Nat'l. Acad. Sci. USA 85:2444 (1988), by computer implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Dr., Madison, WI), or by manual alignment and visual inspection, see, e.g., Current Protocols in Molecular Biology (Ausubel et al., eds. 1995 supplement).

[0085] An example of an algorithm that is suitable for determining percent sequence identity and sequence similarity are the BLAST and BLAST 2.0 algorithms, which are described in Altschul et al., (1990) J. Mol. Biol. 215:403-410 and Altschul et al. (1977) Nucleic Acids Res. 25:3389-3402, respectively. Software for performing BLAST analyses is publicly available through the National Biotechnology Information web site at ncbi.nlm.nih.gov. The algorithm involves first identifying high scoring sequence pairs (HSPs) by identifying short words of length W in the query sequence that either match or satisfy some positive-valued threshold score T when aligned with a word of the same length in a database sequence. T is referred to as the neighborhood word score threshold (Altschul et al., supra). These initial neighborhood word hits act as seeds for initiating searches to find longer HSPs containing them. The word hits are then extended in both directions along each sequence for as far as the cumulative alignment score can be increased. Cumulative scores are calculated using, for nucleotide sequences, the parameters M (reward score for a pair of matching residues; always > 0) and N (penalty score for mismatching residues; always < 0). For amino acid sequences, a scoring matrix is used to calculate the cumulative score. Extension of the word hits in each direction are halted when: the cumulative alignment score falls off by the quantity X from its maximum achieved value; the cumulative score goes to zero or below, due to the accumulation of one or more negative-scoring residue alignments; or the end of either sequence is reached. The BLAST algorithm parameters W, T, and X determine the sensitivity and speed of the alignment. The BLASTN program (for nucleotide sequences) uses as defaults a wordlength of 28, an expectation value of 10, M=1, N=-2 and a comparison of both strands. For amino acid sequences, the BLASTP program uses as defaults a wordlength of 3, an expectation value of 10 and the BLOSUM62 scoring matrix (see Henikoff & Henikoff, Proc. Natl. Acad. Sci. USA 89:10915 (1989)).

[0086] The BLAST algorithm also performs a statistical analysis of the similarity between two sequences (see, e.g., Karlin & Altschul, Proc. Nat'l. Acad. Sci. USA 90:5873-5787 (1993)). One measure of similarity provided by the BLAST algorithm is the smallest sum probability (P(N)), which provides an indication of the probability by which a match between two nucleotide or amino acid sequences would occur by chance. For example, a nucleic acid is considered similar to a reference sequence if the smallest sum probability in a comparison of the test nucleic acid to the reference nucleic acid is less than about 0.2, more preferably less than about 0.01, and most preferably less than about 0.001.

[0087] Two nucleic acid sequences or polypeptides are indicated to be substantially identical if a polypeptide encoded by a first nucleic acid is immunologically cross-reactive with an antibody raised against a polypeptide encoded by a second nucleic acid, as described below. Thus, for example, a polypeptide is typically substantially identical to a second polypeptide if the two peptides differ only by conservative substitutions. Two nucleic acid sequences are also indicated to be substantially identical if the two molecules, or their complements, hybridize to each other under stringent conditions, as described below. Two nucleic acid sequences are yet further indicated to be substantially identical if the sequences can be amplified using the same primers.

[0088] “Polypeptide,” “peptide,” and “protein” are used interchangeably herein to refer to polymers of amino acid residues. All three terms apply to amino acid polymers in which one or more amino acid residues are artificial chemical mimics of corresponding naturally occurring amino acids, as well as to naturally occurring and non-naturally occurring amino acid polymers. As used herein, the terms include amino acid chains of any length, including full-length proteins, in which the amino acid residues are linked via covalent peptide bonds.

[0089] The term “recombinant,” when used in connection with, for example, a cell or nucleic acid, protein, or vector, means that the cell, nucleic acid, protein, or vector has been modified by the introduction of a heterologous nucleic acid or protein, or alteration of a native nucleic acid or protein, or that the cell is derived from a cell so modified. Thus, for example, a recombinant cell expresses genes that are not found in the native (non-recombinant) form of the cell, or express native genes that are otherwise abnormally expressed, under-expressed, or not expressed at all.

[0090] An “expression cassette” is a recombinantly or synthetically produced nucleic acid construct having a specific set of nucleic acid elements that allow for the transcription of a particular polynucleotide sequence in a host cell. An expression cassette can be a plasmid, a viral vector derived from a viral genome, or a portion of a nucleic acid fragment / construct. Typically, an expression cassette includes a polynucleotide to be transcribed operably linked to a promoter. Other elements that can be present in an expression cassette include elements that enhance transcription (e.g., enhancers) and terminate transcription (e.g., terminators), as well as elements that confer certain binding affinities or antigenic properties to the recombinant protein produced by the expression cassette.

[0091] A "promoter" is defined as a series of nucleic acid control sequences that direct transcription of a polynucleotide sequence. As used herein, a promoter includes the necessary polynucleotide sequences about the transcription initiation site, e.g., in the case of a polymerase II type promoter, including the TATA element. A promoter also optionally includes distal enhancer or suppressor elements, which can be located as far as several kilobases from the transcription initiation site. A "constitutive" promoter is a promoter that is active in most environments and developmental conditions. An "inducible" promoter is a promoter that is active under environmental or developmental regulation. The term "operably linked" refers to functional linkage between a polynucleotide expression control sequence (e.g., a promoter or array of transcription factor binding sites) and a second polynucleotide sequence, wherein the expression control sequence directs transcription of the polynucleotide sequence corresponding to the second sequence.

[0092] The term "heterologous" as used in the context of describing the relative positions of two elements, e.g., a polynucleotide sequence (e.g., a promoter or protein / polypeptide coding sequence) or a polypeptide sequence (e.g., a Cry protein sequence or another polypeptide sequence), means that the two elements do not naturally occur in the same relative position. Thus, a "heterologous promoter" of a gene means a promoter that is not naturally operably linked to the gene. Similarly, a "heterologous protein," "heterologous polypeptide," or "heterologous polynucleotide" of a Cry protein or its coding sequence or fragment thereof is derived from a source other than the Cry protein, or in the case of a fragment / coding sequence of a Cry protein, can be derived from another portion of the same Cry protein or coding sequence, but is not naturally directly linked to the fragment in the same manner. Fusion of a fragment of a Cry protein (or its coding sequence) with a heterologous polypeptide (or polynucleotide sequence) does not result in a longer polypeptide or polynucleotide sequence that can naturally occur in a wild-type Cry protein.

[0093] By "host cell" is meant a cell that contains an expression vector and supports the replication or expression of a coding sequence carried in the expression vector. Host cells can be prokaryotic cells, such as Bacillus thuringiensis (Bt), Bacillus subtilis (Bs), or E. coli; or eukaryotic cells, such as yeast, insect, amphibian, or mammalian cells, such as CHO, HeLa, and the like, e.g., cultured cells, explants, and in vivo cells.

[0094] In the present application, the term "polyethylenimine-modified protein" or "PEI-modified protein" refers to a protein that has been reacted with a sufficient amount of PEI under conditions that allow the formation of a chemical bond between any one free carboxyl group (-COOH) on the protein and any one free amine (-NH2) group on the PEI.

[0095] As used herein, a “cofactor” or “coenzyme” is a non-protein organic or inorganic compound required by an enzyme to function properly as a catalyst for a chemical reaction. Organic cofactors containing at least one phosphate group (-PO4) are also referred to as “phosphorylated cofactors”.

[0096] The term “about” as used herein denotes a range of + / - 10% of the reference value. For example, “about 10” defines a range of 9 to 11. DETAILED DESCRIPTION

[0097] I. INTRODUCTION

[0098] Due to the high reactivity, excellent regio- and enantio-specificity, and low environmental toxicity of enzymes, the use of enzymes to catalyze industrial reactions has gained increasing interest. In order to compete economically with chemical catalysis, biocatalysts are optimized so that they can be recycled multiple times. Furthermore, biocatalysts are often optimized so that they can tolerate high concentrations of organic solvents - conditions that can promote substrate solubility and enzyme activity. In addition to improving enzyme stability to harsh industrial conditions, enzyme immobilization can also facilitate easy removal of enzymes and subsequent reuse of enzymes in multiple rounds of catalysis. To this end, the first method of enzyme immobilization by way of recombinantly engineering and producing fusion proteins of enzymes fused to crystal-forming proteins or crystal-forming fragments of said proteins has successfully produced protein crystals with retained enzyme activity, see e.g. WO2018 / 028371 and WO2020 / 211782. The second method, enzyme immobilization by entrapment of these enzymes, which includes the simultaneous entrapment of multiple enzymes in the same protein crystal, is particularly attractive as it does not involve any modification to the enzyme structure, increasing the likelihood of high activity retention and native enzyme conformation, see e.g. WO2020 / 216322.

[0099] The present disclosure provides further improvements to enzyme immobilization and general methods of practical application. The inventors of the present application have discovered that Cry proteins can tightly bind and thus efficiently immobilize the cofactor NADH after modification with polyethylenimine (PEI). This discovery supports a new approach to performing chemical reactions in which both the enzyme(s) catalyzing the reaction, by way of being a fusion partner to a Cry protein, and the essential cofactor, which is adsorbed to the PEI-modified Cry protein, are immobilized by the same protein crystal. This important improvement allows further minimization of time, process, and materials to perform various chemical reactions, making the product greener and more cost-effective.

[0100] II. Construction of Recombinant Proteins

[0101] A. General recombinant techniques

[0102] Basic texts disclosing general methods and techniques in the field of recombinant genetics include Sambrook and Russell, Molecular Cloning, A Laboratory Manual (3rd ed. 2001); Kriegler, Gene Transfer and Expression: A Laboratory Manual (1990); and Ausubel et al., eds., Current Protocols in Molecular Biology (1994).

[0103] For nucleic acids, sizes are given in kilobases (kb) or base pairs (bp). These are estimates derived from agarose or acrylamide gel electrophoresis, from sequenced nucleic acids, or from published DNA sequences. For proteins, sizes are given in kilodaltons (kDa) or number of amino acid residues. Protein sizes are estimated from gel electrophoresis, from sequenced proteins, from derived amino acid sequences, or from published protein sequences.

[0104] Noncommercially available oligonucleotides can be chemically synthesized, e.g., using the solid phase phosphoramidite triester method first described by Beaucage & Caruthers, Tetrahedron Lett. 22: 1859-1862 (1981), using an automated synthesizer, as described in Van Devanter et. al., Nucleic Acids Res. 12: 6159-6168 (1984). Purification of oligonucleotides is performed using any art-recognized strategy, e.g., native acrylamide gel electrophoresis or anion-exchange HPLC, as described in Pearson & Reanier, J. Chrom. 255: 137-149 (1983).

[0105] Sequences of interest genes, e.g., polynucleotide sequences encoding enzymes such as alcohol dehydrogenase (ADH), formate dehydrogenase (FDH), leucine dehydrogenase (LDH), human arginase (hArg), or ornithine decarboxylase (ODC), polynucleotides encoding Cry proteins or crystal-forming fragments thereof, and synthetic oligonucleotides, can be verified after cloning or subcloning using, e.g., the chain termination method for sequencing double-stranded templates of Wallace et al., Gene 16: 21-26 (1981).

[0106] B. Coding sequences

[0107] Polynucleotide sequences encoding Cry proteins, fragments, or fusion proteins for use in the present application can be readily constructed using the corresponding coding sequences for Cry proteins, fragments, or the coding sequences of the component fusion partners (e.g., Cry3Aa or CrylAb proteins and hArg or ODC). Sequences of Cry proteins and enzymes are generally known and can be obtained from commercial suppliers.

[0108] In addition to using full-length wild-type Cry proteins to produce crystal-forming proteins for use in the present application, fragments of Cry proteins and / or variants of Cry proteins can also be useful. DNA sequences encoding Cry proteins can be modified to produce fragments or variants of Cry proteins. As long as these fragments and variants retain the ability to spontaneously form crystals when expressed in host cells, especially Bacillus bacterial cells, they can be used to produce protein crystals, either by themselves or by way of fusion proteins that are capable of undergoing spontaneous crystallization, and thus produce protein crystals comprising one or more recombinant proteins (e.g., one or more enzymes) embedded therein. Generally, variants have a high percentage of sequence identity to wild-type Cry protein sequences (e.g., at least 80%, 85%, 90%, 95%, 97%, 98%, 99%, or more), while fragments can be substantially shorter than full-length Cry proteins, e.g., removal of some amino acids (e.g., 10-300 or 20-200 or 50-100 amino acids) from the N or C terminus of a full-length Cry protein. For example, useful Cry3Aa fragments can be as short as the first 290 amino acids from the N terminus, including domain I of the protein. Other examples of such fragments include Cry protein fragments that have the first 57 amino acids removed from the N terminus and Cry protein fragments that have the C terminus 18 amino acids removed. The ability of recombinantly produced Cry proteins, fragments thereof, or fusion proteins comprising Cry proteins or fragments to undergo spontaneous crystallization can be verified by electron micrographs, while the enzymatic activity of recombinantly produced enzymes, including in the form of fusion proteins with Cry proteins or fragments thereof, can be confirmed by assays that have been established for each particular enzyme. Exemplary Cry protein fragments that are capable of self-crystallization can be found in the inventors’ earlier publications, e.g., WO 2018 / 028371.

[0109] In the case of a fusion protein, one or more peptide linkers or spacers can be used to join the coding sequences of the Cry protein / fragment and the one or more heterologous polypeptides. One purpose is to ensure the proper reading frame of the fusion protein so that the coding sequences of both the Cry protein / fragment and the heterologous polypeptides are in frame. Another purpose is to provide the proper spatial relationship between the Cry protein / fragment and the heterologous polypeptides so that each component of the fusion protein can retain its original function: the Cry protein / fragment is able to cause self-crystallization of the fusion protein, and the heterologous polypeptide, such as an enzyme, remains active in its catalytic ability. In addition, one or more linkers can be placed at the very beginning and / or end of the open reading frame for proper initiation and termination of translation of the coding sequence. Such joining amino acid sequences are typically short and are usually no longer than 100 or 50 amino acids, for example, 1 to 100, 1 or 2 to 50, 2 or 3 to 25, 3 or 4 to 10 amino acids.

[0110] C. Sequence modifications to favor codon usage in the host organism

[0111] The polynucleotide sequence encoding the recombinant protein to be expressed according to the methods of the present application can be further altered to coincide with the preferred codon usage of a particular host. For example, the preferred codon usage of one strain of bacterial cells can be used to obtain a polynucleotide encoding a recombinant polypeptide of the present application and comprising the codons preferred by that strain. The frequency of preferred codon usage exhibited by a host cell can be calculated by averaging the frequency of preferred codon usage in a large number of genes expressed by the host cell (e.g., a calculation service is available from the website of the Kazusa DNA Research Institute in Japan). The analysis is preferably limited to genes that are highly expressed by the host cell.

[0112] Upon completion of the modifications, the coding sequence is verified by sequencing, and then subcloned into an appropriate expression vector for recombinant production of the protein (e.g., enzyme, such as a lipase) with the Cry protein, the crystal-forming fragment of the Cry protein, or the fusion protein comprising the Cry protein or the crystal-forming fragment, such that the protein (e.g., enzyme) is produced and embedded in a protein crystal formed by the Cry protein, fragment, or fusion protein.

[0113] III. Expression and Isolation of Recombinant Proteins

[0114] Upon verification of the coding sequence, co-expression of the protein of interest (e.g., enzyme) and the Cry protein, the crystal-forming fragment thereof, or the fusion protein of the present application comprising the Cry protein or the fragment can be produced according to the polynucleotide sequences encoding the Cry fusion proteins disclosed herein using routine techniques in the art of recombinant genetics.

[0115] A. Expression systems

[0116] To obtain high level expression of a polynucleotide sequence encoding a recombinant protein, the polynucleotide encoding the protein in the correct reading frame is typically subcloned into an expression vector containing a strong promoter for directing transcription, a transcription / translation terminator, and a ribosome binding site for translation initiation. Suitable bacterial promoters are well known in the art and described in, for example, Sambrook and Russell, supra, and Ausubel et al., supra. Bacterial expression systems for expressing polypeptides are available in, for example, E. coli, Bacillus sp., Salmonella, and Caulobacter. Kits for such expression systems are commercially available.

[0117] The promoter used to direct expression of a recombinant protein depends on the particular host cell used for production of the recombinant protein. For example, to efficiently express in a Bacillus bacterial strain, such as a Bacillus thuringiensis (Bt) or Bacillus subtilis (Bs) cell, a promoter should be selected that is known to direct robust protein expression in these particular bacterial cells. The promoter is optionally positioned about the same distance from the transcription start site of the exogenous or recombinant protein as it is in its natural environment from the transcription start site. However, as is known in the art, some variation in this distance can be accommodated without loss of promoter function. In some cases, a constitutive promoter is used, while in other cases, an inducible promoter is preferred over a constitutive promoter. Further, placing a common promoter or multiple separate promoters (which can be the same or different kinds of promoters) in a common expression cassette (e.g., an expression vector, e.g., a plasmid), and / or placing one or more separate ribosome binding sites that separate different segments of coding sequence (each encoding a recombinant protein or a crystal formation protein) can allow for different expression ratios of the recombinant protein to the crystal formation protein, thereby maximizing the percentage of recombinant protein that is embedded or immobilized in protein crystals formed from the crystal formation protein, e.g., a Cry protein, a crystal formation fragment thereof, or a fusion protein comprising said Cry protein or said fragment.

[0118] In addition to the promoter, the expression vector will typically also include a transcription unit or expression cassette containing all additional elements required for expression of the recombinant protein and crystal formation protein in the host cell. Thus, a typical expression cassette contains a promoter operably linked to a polynucleotide sequence encoding the recombinant protein and / or crystal formation protein and signals required for efficient polyadenylation of the transcript, ribosome binding site, and translation termination. The polynucleotide coding sequence can be linked to a cleavable signal peptide sequence to facilitate secretion of the polypeptide by the transformed cell. Such signal peptides include those from tissue plasminogen activator, insulin, and neuron growth factor, as well as juvenile hormone esterase of Heliothis virescens, among others. Other elements of the cassette can include enhancers, and, if genomic DNA is used as the structural gene, introns with functional splice donor and acceptor sites.

[0119] In addition to the promoter sequence, the expression cassette should contain a transcription termination region downstream of the coding sequence to provide for efficient termination. The termination region can be obtained from the same gene as the promoter sequence, or can be obtained from different genes. Placement of a commonly shared termination site in combination with placement of multiple promoters directing transcription of multiple coding sequences can also be used to modulate the expression ratio of the recombinant protein to the crystal formation protein.

[0120] The particular expression vector used to transport the genetic information into the cell is not of particular significance. Any conventional vector for expression in eukaryotic or prokaryotic cells can be used, particularly those suitable for expression in cells of Bacillus spp. such as Bt and Bs. Standard bacterial expression vectors include plasmids, such as pBR322-based plasmids, pSKF, pET23d, and fusion expression systems such as GST and LacZ.

[0121] Elements that are typically included in the expression vector also include a replicon that functions in the bacteria, such as Bacillus and E. coli, a gene encoding antibiotic resistance to allow selection of bacteria containing the recombinant plasmid, and unique restriction sites in non-essential regions of the plasmid to allow insertion of coding sequences. The particular antibiotic resistance gene selected is not critical, any of a number of resistance genes known in the art are suitable. Similar to antibiotic resistance selection markers, metabolic selection markers based on known metabolic pathways can also be used as a means of selecting transformed host cells.

[0122] B. Transfection methods

[0123] Bacterial, mammalian, yeast, insect, or plant cell lines expressing large amounts of the recombinant fusion proteins of the application are generated using standard transfection methods, and the fusion proteins are then purified using standard techniques (see, e.g., Colley et al., J. Biol. Chem. 264: 17619-17622 (1989); Guide to Protein Purification, in Methods in Enzymology, vol. 182 (Deutscher, ed., 1990)). Transformation of eukaryotic and prokaryotic cells is performed according to standard techniques (see, e.g., Morrison, J. Bact. 132: 349-351 (1977); Clark-Curtiss & Curtiss, Methods in Enzymology 101: 347-362 (Wu et al., eds, 1983)).

[0124] Any well-known method of introducing an exogenous polynucleotide sequence into a host cell can be used. These methods include the use of calcium phosphate transfection, polybrene, protoplast fusion, electroporation, liposomes, microinjection, plasmid vectors, viral vectors, and any other well-known method for introducing cloned genomic DNA, cDNA, synthetic DNA, or other exogenous genetic material into a host cell (see, e.g., Sambrook and Russell, supra). Only the particular genetic engineering procedure used needs to be able to successfully introduce at least one gene into a host cell that is capable of expressing a recombinant protein, as well as a crystal formation protein (e.g., a Cry protein, a crystal formation fragment thereof, or a fusion protein comprising a Cry protein or a crystal formation fragment thereof), according to the methods of the application. As described above and herein, the recombinant protein(s) (e.g., enzyme(s)) and the crystal formation protein can be contained within one single expression cassette, e.g., in the same expression vector, where all coding sequences can be under the control of one single promoter, or each coding sequence can be under the control of a separate promoter (which optionally can be different from each other). In the alternative, each of the coding sequences can be present in separate expression cassettes (e.g., expression vectors). In either alternative, different ratios of recombinant protein to crystal formation protein can be achieved by using a single copy or multiple copies of any one of the coding sequences, or by placing separate or commonly shared ribosome binding sites and / or termination sites in the expression cassettes.

[0125] C. Isolation of crystal-embedded recombinant proteins

[0126] Once expression of the recombinant protein and the crystal forming protein (e.g., a Cry protein, a crystal forming fragment thereof, or a fusion protein comprising a Cry protein or a crystal forming fragment thereof) in the transfected host cell has been confirmed, e.g., by electron micrographs for detecting protein crystals or immunoassays such as Western blots, the host cell is then cultured on a suitable scale for the purpose of purifying or isolating the recombinant protein embedded within the protein crystals formed by the Cry protein, a crystal forming fragment thereof, or a fusion protein comprising a Cry protein or a crystal forming fragment thereof.

[0127] For example, upon induction of the promoter, the recombinant protein and the crystal forming protein are produced in large quantities by the transformed bacteria, and the recombinant protein is embedded within the crystals formed by the crystal forming protein. In other words, the recombinantly produced protein exists in a crystalline form or insoluble aggregates within the host cell. Thus, they can be easily isolated from the cell lysate by techniques such as centrifugation and density gradient separation based on the different densities of the crystals, followed by one or more washing steps to further remove contaminants from the protein crystals.

[0128] There are several protocols suitable for purifying protein inclusion bodies. For example, purification of the aggregated protein (hereinafter referred to as inclusion bodies) typically involves extracting, isolating and / or purifying the inclusion bodies by disrupting the bacterial cells, e.g., by incubation in a buffer of about 100-150 μg / ml lysozyme and 0.1% Nonidet P40, a non-ionic detergent. The cell suspension can be ground using a Polytron grinder (Brinkman Instruments, Westbury, NY). Alternatively, the cells can be sonicated on ice. Other methods of lysing the bacteria are described in Ausubel et al. and Sambrook and Russell, both supra, and will be apparent to those skilled in the art.

[0129] The cell suspension is typically centrifuged, and the pellet containing the inclusion bodies is resuspended in a buffer that does not solubilize the inclusion bodies but washes the inclusion bodies, e.g., 20 mM Tris-HCl (pH 7.2), 1 mM EDTA, 150 mM NaCl and 2% Triton-X 100, a non-ionic detergent. It can be necessary to repeat the washing step to remove as much cellular debris as possible. The remaining inclusion body pellet can be resuspended in an appropriate buffer, e.g., 20 mM sodium phosphate, pH 6.8, 150 mM NaCl. Other appropriate buffers will be apparent to those skilled in the art.

[0130] After isolation, the recombinant protein is recovered from the host cells in the form of protein crystals, which can be used directly according to their inherent biological activity. For example, lipase encapsulated in Cry protein crystals can be used to hydrolyze triglycerides. Due to its insoluble crystalline form, the lipase has a high level of resistance to harsh environmental conditions such as high temperature, extreme pH, and organic solvents, allowing repeated cycles of cleaning and reuse.

[0131] In an alternative, after the washing step, the inclusion bodies are dissolved to release the embedded recombinant protein by adding a solvent that is both a strong hydrogen acceptor and a strong hydrogen donor (or a combination of solvents each having one of these characteristics). The protein from the inclusion bodies can then be renatured by diluting or dialysis with a compatible buffer. Suitable solvents include, but are not limited to, urea (about 4M to about 8M), formamide (at least about 80%, based on volume / volume), and guanidine hydrochloride (about 4M to about 8M). Some solvents that can dissolve aggregate-forming proteins, such as SDS (sodium dodecyl sulfate) and 70% formic acid, may not be suitable for this procedure because the protein may be irreversibly denatured, with immunogenicity and / or active deficiency. Although guanidine hydrochloride and similar reagents are denaturants, this denaturation is not irreversible, and renaturation may occur after removing (for example, by dialysis) or diluting the denaturant, thereby allowing the reformation of the target immune and / or biologically active protein. After dissolving, the protein can be separated from other bacterial proteins by standard separation techniques. For further description of purification of recombinant polypeptides from bacterial inclusion bodies, see, eg, Patra et al., Protein Expression and Purification 18:182-190 (2000).

[0132] Although protein crystals tend to remain insoluble at lower or neutral pH, placing them in alkaline solutions at a pH equal to or greater than 10 or 11 can often effectively solubilize the protein. Once solubilized, the protein can then be analyzed by gel separation (e.g., on an SDS gel) and immunoassay to confirm its identity based on appropriate molecular weight and immunoreactivity.

[0133] D. Polyethyleneimine (PEI) modification of crystal-forming recombinant proteins

[0134] Once the recombinant expression of a crystal-forming protein (e.g., a Cry protein, a crystal-forming fragment thereof, or a fusion protein comprising a Cry protein or a crystal-forming fragment thereof) in the transfected host cells is confirmed, for example, by electron micrographs for detecting protein crystals or immunoassays such as Western blot analysis, the host cells are then cultured on an appropriate scale for the purpose of producing and then purifying or isolating protein crystals, and the protein crystals are then modified with polyethyleneimine (PEI).

[0135] PEI is a cationic polymer containing primary, secondary, and tertiary (if branched) amino groups with strong positive charges (Hernandez-Montelongo et al., Mater. Sci. Eng. C. 2017, 71, 718-724). PEI typically has a molecular weight of 700 Da to 1000 kDa depending on the degree of polymerization (Vicennati et al., A. Curr. Med. Chem. 2008, 15, 2826-2839). Using standard conjugation methods, PEI can be covalently attached to proteins through reaction with amino acid side chains. Because PEI has a high cationic charge density, limited coupling to PEI provides sufficient cationic charge to the protein without causing a severe drop in protein function or activity.

[0136] Conjugation of PEI to the protein crystals of the present application can be performed using methods known in the art or described herein. Briefly, the protein crystals are produced recombinantly in culture of transfected cells. After harvesting and lysing the host cells, the protein crystals are isolated by centrifugation of the cell lysate and appropriate washing steps. The protein crystals are then resuspended in a reaction solution containing an appropriate amount of PEI (e.g., 10-100 pg / mL, pH 6.5-7.5) and at least one coupling reagent (e.g., about 0.1-10 mg / ml of crystals) capable of coupling PEI to the protein crystals in a molar ratio of about 1 :0.5-5. An additional option is to use the coupling reagent DCC in an organic solvent. After incubation (optionally with agitation) for a sufficient period of time (e.g., about 1 to 10 hours), the coupling reaction is terminated, and the PEI-conjugated protein crystals are recovered from the solution (e.g., by centrifugation) and washed again (to remove any unbound PEI and other molecules) before using the PEI protein crystals according to the present application.

[0137] IV. Immobilization of cofactors

[0138] In another aspect, the present application relates to the use of a co-factor or co-enzyme in the context of performing a biosynthetic reaction catalyzed by an enzyme, said co-factor or co-enzyme being necessary for the enzymatic activity of the enzyme and having been co-immobilized with the enzyme comprised within a protein crystal. As described above, recombinant proteins, in particular enzymes, can be produced according to the methods of the present application such that the recombinant protein is immobilized within a protein crystal either by entrapment when co-expressed with a crystal formation protein, or by participating in protein crystal formation as a fusion partner to another polypeptide to produce a crystal formation fusion protein. Based on this, the inventors of the present application further found that PEI-modified protein crystals with highly ordered PEI natural scaffolds allow the reversible binding of phosphorylated co-factors (e.g. NADH) that are essential for the catalytic activity of certain enzymes by applying or stopping mechanical stirring of a solution containing the PEI-modified protein crystals loaded with the co-factor. This finding supports a reaction scheme in which the enzyme and its co-factor are co-immobilized by the protein crystal and thus can be used in a recyclable and washable manner for multiple rounds of enzyme-catalyzed reactions.

[0139] More specifically, the inventors of the present application found that PEI-modified protein crystals are able to bind co-factors such as NADH at high capacity (e.g. about 550 ± 5 pmol / g) and retain more than 50% of the bound NADH molecules even after 20 wash cycles. Thus, these protein crystals are ideal reservoirs for storing co-factors, in particular phosphorylated co-factors such as NADH, in a stirred batch reaction - the co-factor is released from the PEI-modified crystals under mechanical stirring, allowing the co-factor to participate in the co-factor dependent reaction process, while the co-factor is reabsorbed back onto the PEI-modified protein crystals when stirring is stopped. This reversible nature of the PEI-modified protein crystals in their release and reabsorption of co-factors (controlled by the application or stop of mechanical stirring) contributes to an important feature of the reaction scheme of the present application in which the step of cleaning / washing the immobilized enzyme and co-factor after each round of reaction can be conveniently performed before the next round of reaction.

[0140] V. Use of immobilized proteins

[0141] Another aspect of the application relates to the use of a recombinant protein such as an enzyme, embedded and / or immobilized in a protein crystal, particularly a PEI-modified protein crystal, produced according to the methods described herein, to exert the intrinsic biological activity of the protein to, for example, carry out a reaction normally catalyzed by the enzyme present in the protein crystal, such as hydrolysis, esterification, ligation, proteolysis, biosynthesis, etc., when the protein crystal containing the enzyme is incubated in a reaction solution having at least one substrate for the enzyme, each substrate present in an effective amount under conditions that allow the reaction to proceed to give the desired product. In some sequential reaction schemes, multiple immobilized enzymes can be placed in a single reaction mixture with initial substrates to be catalyzed by a first enzyme in a first reaction to produce a first product that in turn serves as a substrate for a second enzyme to be catalyzed in a second reaction to produce a second product, and so on.

[0142] The presence of PEI (highly positively charged moiety) on the protein crystal allows for the immobilization of one or more co-factors or co-enzymes necessary for the function of the immobilized enzyme, thus further ensuring enzymatic activity for catalyzing chemical reactions. Since organic solvents are generally able to facilitate these reactions, and the immobilized recombinant proteins produced by the methods of the application are highly resistant to the presence of organic solvents, the reactions carried out using the immobilized proteins according to the application are generally not only water-based solvents, but also one or more organic solvents, such as ethanol, methanol, acetonitrile, and dimethylformamide.

[0143] The inventors of the present application have discovered that immobilizing a recombinant protein within a crystallized Cry protein or fusion protein results in the protein having a higher level of resistance to organic solvents as well as a higher level of thermal stability, potentially allowing for the enzyme activity to be retained for more reaction cycles. The presence of PEI on the protein crystal provides further options for immobilizing co-factors or co-enzymes required for the immobilized reaction, which can be relatively easily loaded, unloaded, or reloaded onto the PEI-modified protein crystal. For example, incubating a phosphorylating co-factor (such as NADH) with a PEI-modified protein crystal will result in the co-factor binding or loading onto the crystal, thus becoming immobilized; while the co-factor-loaded PEI-modified protein crystal is present in a fluid environment with mechanical agitation, the co-factor will be released or unloaded from the crystal, thus becoming available to facilitate enzymatic reactions. Upon cessation of agitation, the co-factor will again bind or be reloaded onto the PEI-modified protein crystal.

[0144] In some cases, the reaction process includes a cleaning step performed after completion of a round of reaction and removal of reaction products and any remaining substrate, during which the protein crystals containing the recombinant protein are flushed or washed in preparation for their reuse with new substrate in a subsequent round of reaction. Similarly, a cleaning step can be performed between any two rounds of reaction in order to reuse, recycle, or replenish a cofactor or coenzyme for a successive series of reactions, the cleaning step including reloading of the cofactor or coenzyme onto the PEI portion of the PEI protein crystals after completion of a round of enzymatic reaction, followed by flushing of the crystals.

[0145] Examples

[0146] The following examples are offered by way of example only and not by way of limitation. One of skill in the art will readily recognize a variety of noncritical parameters that can be changed or modified in order to yield essentially the same or similar results.

[0147] Example 1. NADH-dependent enzyme crystals and Cry3Aa-PEI crystals: a novel interface for molecular exchange under mechanical agitation

[0148] NAD(H)-dependent enzymes play a key role in the biosynthesis of pharmaceuticals and fine chemicals. However, the requirement of enzymes for NAD(H) cofactor is a significant obstacle for their application in industrial biosynthesis due to the high cost of the cofactor. This study introduces polyethylenimine (PEI)-modified Cry3Aa (3A-PEI) crystals as highly ordered natural scaffolds that recycle NADH cofactor for multiple catalytic cycles by controlled binding and release of NADH cofactor under mechanical stirring. The 3A-PEI crystals exhibit an impressive NADH binding capacity of 550 ± 5 μmol / g and retain more than 50% of the bound NADH molecules even after 20 wash cycles. Thus, these 3A-PEI crystals can be used as ideal reservoirs for storing NAD(H) cofactor in stirred batch reactions - they release NADH molecules under mechanical stirring, allow the NADH molecules to participate in NADH-dependent processes, and rebind NADH when stirring is stopped. To demonstrate the practicality of this NADH recycling technology in real applications, we have produced a second Cry3Aa protein pellet containing genetically co-immobilized formate dehydrogenase (FDH) and leucine dehydrogenase (LDH) that can facilitate the synthesis of L-tert-leucine. Here, FDH regenerates NADH concomitant with the oxidation of sacrificial formate, while LDH performs the catalytic reaction of interest. A reaction solution containing a mixture of NADH-loaded 3A-PEI and Cry3Aa-FDH / Cry3Aa-LDH crystals allows the efficient biosynthesis of L-tert-leucine for more than 20 days with minimal addition of NADH (NADH turnover number: 284). This innovative approach takes an important step towards addressing the challenge of NAD(H) cofactor utilization in industrial NAD(H)-dependent biosynthesis by providing a sustainable and efficient means to maintain a continuous NADH supply for multiple reaction cycles.

[0149] INTRODUCTION

[0150] NADH-dependent enzymes have shown outstanding efficiency and enantioselectivity in the biosynthesis of pharmaceuticals, making them very important for practical applications [1]. However, the high cost of NAD(H) production (3000 USD / mol in bulk) [2] and the need for stoichiometrically supplemented enzyme catalysis are major economic challenges that limit its large-scale utilization. Therefore, regeneration and recycling of NAD(H) are two important technical obstacles for the practical use of NAD(H)-dependent enzymes to produce chiral pharmaceutical compounds.

[0151] Although several technologies have been developed to regenerate NADH [1-3], enzymatic NADH regeneration is the main method used in the industry sector because it is compatible with other biosynthetic systems, has mild reaction conditions, high selectivity and specificity for NAD(H), and low energy demand [4]. In this regard, formate dehydrogenase (FDH) is an attractive enzyme for this purpose because it can catalyze the conversion of NAD+ to NADH with concomitant oxidation of formate to produce CO2, a byproduct that can be easily separated from the supernatant during work-up [1, 4].

[0152] Recycling of NAD(H) cofactor for multiple reaction cycles is also an important challenge because it eliminates the need to add NADH at the beginning of each reaction cycle, thereby greatly reducing the amount of expensive NADH cofactor required. Several methods for NAD(H) recycling have been reported, including ionophores [5], covalent immobilization [6], encapsulation [7, 8], and membrane filtration [8], however, these methods each have different limitations, including weak binding [9], support defects [9], diffusion problems [7], low support capacity for cofactor preservation [6], and the need for complex molecular assembly and design [10, 11], which hinder their universal application. Therefore, it would be of great benefit to develop a simple and practical solution for NADH recycling.

[0153] We have used the Cry3Aa protein crystals produced in Bacillus thuringiensis (Bt) for a variety of applications. The unique bio-self-assembly properties of the Cry3Aa protein enable the embedding of co-expressed protein targets within the Cry3Aa lattice, or direct immobilization of genetically fused functional enzymes or proteins in a way that stabilizes the fusion protein against thermal and solvent denaturation [12, 13]. We successfully utilized the latter strategy to produce Cry3Aa-lipase A and Cry3Aa-PML fusion crystals, both of which showed high activity and recyclability in the conversion of plant oil to FAME biodiesel. Direct embedding of PML in Cry3Aa crystals was demonstrated in vitro as well as in bacteria

[12] . The embedded PML was found to be even more active than the Cry3Aa-PML fusion crystal and showed suitability for the recyclable conversion of waste cooking oil to FAME biodiesel

[14] .

[0154] Due to the initial observation of NADH binding affinity to Cry3Aa crystals, we became interested in the cofactor entrapment in Cry3Aa crystals. This observation can be due to the long nanochannel in Cry3Aa crystals, which can facilitate the uptake of NADH molecules. The level of NADH binding is suboptimal for practical applications, so we directed our exploration toward protein cationization as a prospective strategy to increase the binding capacity of the crystals for the negatively charged cofactor. Our lab's previous studies have created positively charged variants of Cry3Aa, Pos3Aa by mutating multiple residues on the surface of domain II of Cry3Aa to lysine to create a positively charged surface

[15] . Although Pos3Aa crystals can bind more NADH than Cry3Aa, the level is still too low to be used for biocatalysis studies.

[0155] Therefore, we sought a strategy to achieve higher levels of cationization, and since polyethylenimine (PEI) has a high positive charge and flexibility, we decided to explore the modification of Cry3Aa crystals with polyethylenimine. Here, we demonstrate that cationization of Cry3Aa crystals with PEI indeed effectively facilitates high levels of NADH binding, in amounts sufficient to be used as an NAD + / NADH reservoir to facilitate NADH-dependent enzyme reactions. A simple strategy involving mechanical stirring was subsequently developed to facilitate the reversible release of NADH from Cry3Aa-PEI crystals to modulate the release of NADH molecules from Cry3-A-PEI crystals into the reaction solution and their re-binding. Finally, we combined this NADH capture technology with a more advanced version of our Cry3Aa enzyme immobilization technology to produce a biocatalyst that contains co-immobilized formate dehydrogenase (FDH) and leucine dehydrogenase (LDH), which can convert NADH released from the Cry3Aa-PEI crystals into NAD + and the leucine dehydrogenase (LDH) can then use the NADH to biosynthesize L-tert-leucine, a chiral precursor that has been widely used in the pharmaceutical industry to produce various types of therapeutics.

[0156] Methods and Materials

[0157] 3-aminopropyl-triethoxysilane (APTMS), O-phthaldialdehyde (OPA), 1-fluoro-2,4- dinitrophenyl-5-L-alanine amide, nicotinamide adenine dinucleotide (NAD(H)), trimethylpyruvic acid (TMP), L-tert-leucine and D-tert-leucine were obtained from TCI America. Ammonium formate, polyethylenimine (PEI) and other buffer salts were also purchased from Sigma-Aldrich. Additional compounds and chemicals are given in brackets below.

[0158] Preparation of 3A-PEI crystals

[0159] Cry3Aa crystals were produced by transforming B. thuringiensis BT407G cells with pre-prepared PHT-Cry3Aa plasmid and growing the cells for three days. Cells were harvested and lysed with lysozyme, and after washing with 10% hexane to remove spores, Cry3Aa crystals were purified by low speed centrifugation (6000 rpm)

[16] . 0.5 mg / mL Cry3Aa crystal solution was mixed in PEI solution (50 pg / mL, pH: 7.0) and then EDC / NHS coupling reagent (molar ratio 1 :2) was gradually added under magnetic stirring. After 5 hours of reaction, the reaction was quenched using a solution of 100 mM Tri buffer (pH 7.0). Subsequently, the sample was centrifuged, washed with NaCl (100 mM) and then with ddH2O several times to clean up any non-crosslinked or weakly bound PEI molecules.

[0160] Quantification of PEI bound to Cry3Aa crystals

[0161] To quantify PEI bound to Cry3Aa crystals, PEI standard solutions were prepared by dissolving 500 mg of PEI solution (50% v / v) in 50 ml of ddH2O for 2 hours, then adjusting the pH to 7.0 by gradually adding 5 M HC1. PEI was quantified using PEI and Cu 2 + The fact that a blue complex is formed, fixing 130 mM of copper (II) sulfate in the presence of different amounts of PEI (0-50 pg / ml) to determine a standard curve of the PEI quantification in solution. Subsequently, this standard curve was used to evaluate the PEI bound to the crystals using the PEI residual amount after the cationization process [17, 18]. All determinations were carried out at room temperature, with a copper (II) sulfate solution as a blank control.

[0162] Binding of NADH to 3A-PEI crystals

[0163] The binding capacity of NADH to 3A-PEI crystals was determined by mixing 50 pg of dried 3A-PEI crystals and NADH (0.25 mM) at 10 °C for 30 min under gentle rotation. Then, the amount of NADH bound to 3A-PEI crystals was quantified by measuring the absorbance of the supernatant at 340 nm wavelength using an Eppendorf BioSpectrometer (Eppendorf AG, Germany) after centrifugation at 15000 rpm for 5 min. The NADH bound into the crystals was then defined as the number of moles of NADH bound per unit of dry weight of 3A-PEI crystals, which is shown in 3A-PEI-NADH crystals (dry weight of 3A-PEI, number of moles of NADH bound) in this study. The binding of NADH molecules was further investigated by using a Leica SP8 confocal microscope in the DAPI channel. The excitation and emission wavelengths of NADH fluorescence were adjusted to 335-350 nm and 440-470 nm, respectively [4]. To compare the binding capacity of 3A-PEI crystals with Cry3Aa and Pos3Aa crystals, the same experiment was performed as a control.

[0164] In vitro growth of Cry3Aa crystals

[0165] Cry3Aa protein crystals were dissolved in a solution containing 50 mM Na2CO3, 150 mM NaCl and adjusted to pH 11. For solubilization and purification of Cry3Aa protein, a high-resolution size exclusion column (Bio-rad ENrich™ SEC 650) was used with gel filtration technique. In vitro growth of Cry3Aa crystals was achieved using sitting-drop vapor-diffusion method at 10 °C. This involved mixing 1 pL of protein solution (8.3 mg / mL) with 1 pL of well solution (20% v / v 1,4-butanediol, 100 mM HEPES / NaOH, 200 mM NaCl) in a 96-well crystallization plate (Hampton)

[15] . The obtained Cry3Aa crystals were gently transferred to a solution containing 100 mM HEPES / PEI (50 pg / mL) and EDC / NHS coupling reagents (5 mM and 10 mM, respectively) at pH 7.5. The samples were incubated at room temperature for 2 h, followed by treatment with 100 mM Tri buffer (pH 7.5) for 30 min to inactivate any residual EDC / NHS crosslinking reagents. Subsequently, the samples were centrifuged and washed with HEPES (100 mM) and then with ddH2O multiple times to remove excess salt, and then stored at 4 °C for further use.

[0166] Crystallization, data collection, phase determination and refinement

[0167] 3A-PEI crystals were flash frozen in liquid nitrogen and used 30% glycerol as cryoprotectant. Subsequently, data sets were collected on the microfocus beamline 05A at NSRRC using a RAYONIX MX-300HS detector. Data processing and scaling were performed using the HKL2000 software

[19] . To generate the initial structure, PHASER was used under the CCP4i interface, where the Cry3Aa structure (PDB ID: 1DLC) was used as the search model

[20] . Iterative cycles of modeling and refinement were performed using the Coot, Rafmac5 and Phenix programs, respectively, to improve the quality of the model [14, 21-23].

[0168] Scanning electron microscopy (SEM) of 3A-PEI crystals

[0169] Scanning electron microscopy (SEM) was used to study the morphology of 3A and 3A-PEI grown in vitro and in vivo. After incubation of the samples in ddH2O (0.2 mg / ml), 5 pL was placed on carbon glass and allowed to dry gradually overnight, followed by coating with 80 / 20 Pd / Pt using a Q150T ES TURBOMOLECULAR PUMP COATER (Quorum). Images were captured using a Hitachi SU8000 instrument.

[0170] Distribution of PEI and NADH molecules in 3A-PEI crystals grown in vitro

[0171] The localization of PEI polymers in Cry3Aa crystals was assessed by first labeling the crystals grown in vitro by mixing with Alexa Fluor TM 568C5 maleimide (1 mM, Thermo Fisher Scientific) in 100 mM HEPES buffer (pH 7.5) for 1 h under gentle rotation. Subsequently, the crystals were washed thoroughly with the same buffer multiple times to remove any unbound Alexa 568 molecules. In the meantime, a PEI solution (50 pg / mL) was incubated with 100 mM HEPES buffer (pH 7.5) containing Alexa Fluor TM 488 NHS ester (succinimidyl ester) (Thermo Fisher Scientific) (1 mM) under gentle rotation for 1 h. After this, a centrifugal filter (3 kDa, Amicon® Ultra-0.5 mL, Millipore) was used to separate the crystals from the solution. The crystals were washed with the same buffer multiple times to remove any unbound Alexa 488 molecules. The crystals were then mounted on a carbon-coated grid and imaged using a JEOL 1400 Plus transmission electron microscope at 80 kV. Ultra) with 100 mM HEPES buffer multiple times to obtain labeled PEI-Alexa488 polymer. Finally, labeled Cry3Aa-Alexa568 and PEI-Alexa488 polymers were used in the cationization process to produce Cry3Aa-Alexa568-PEI-Alexa488 crystals. Then, the obtained crystals were incubated in ddH2O in the presence of NADH (0.25 mM) for 30 min at 10 °C under slow rotation. After several ddH2O washes to remove unbound NADH molecules, Cry3Aa-Alexa568-PEI-Alexa488-NADH crystals were analyzed by Leica SP8 confocal microscope using DAPI, Alexa568 and Alexa488 channels.

[0172] Dynamic light scattering (DLS)

[0173] ζ-potential and hydrodynamic diameter of Cry3Aa and 3A-PEI crystals were measured at 25 °C using a Malvern Zetasizer Nano ZS90 (Malvern Instruments, UK). 100 pg / mL crystals were prepared in ddH2O solution for size distribution measurements and in PBS (10 mM) for ζ-potential.

[0174] Size distribution of PEI polymers bound in Cry3Aa crystals

[0175] Size distribution of PEI solution within Cry3Aa crystals was measured by immersing Cry3Aa crystals (5 mg / ml) in PEI solution (50 pg / ml, pH 7.0) while gently stirring using a magnetic stirrer. Following this, EDC / NHS crosslinker was gradually added. Then, the sample was centrifuged at 15000 rpm for 10 min, allowing the crystals to separate from the surrounding solution. The resulting supernatant was then analyzed using dynamic light scattering (DLS) to determine the size distribution of PEI molecules present. For comparison, PEI solution (50 pg / ml, pH 7.0) was included in the experiment as a control.

[0176] Retention of NADH on 3A-PEI crystals over multiple wash cycles

[0177] The capacity of 3A-PEI crystals to retain bound NADH was measured by incubating NADH-loaded 3A-PEI crystals (10 mg, 216 ± 3 nmol NADH) in sodium phosphate buffer (Na-PB, 10 mM, pH 7.0) at 10 °C under slow rotation. Then, the samples were centrifuged at 15000 rpm for 5 minutes and the amount of released NADH in the supernatant was measured spectrophotometrically at 340 nm. This procedure was repeated for a total of 20 cycles. Subsequently, the samples were resuspended in 1 M Na-PB and vortexed vigorously to release all bound NADH.

[0178] Effect of pH on NADH binding to 3A-PEI crystals

[0179] To determine the effect of pH on the NADH binding capacity of 3A-PEI crystals, 10 mg of 3A-PEI crystals were prepared individually in Na-PB (25 mM) at different pH (6.0, 7.0 and 8.0) and then incubated in the presence of 2 mM NADH at 10 °C under slow rotation. Then, the samples were centrifuged at 15000 rpm for 5 minutes and the amount of NADH bound to 3A-PEI crystals was measured spectrophotometrically at 340 nm. NADH solution (2 mM) was treated at all pH as a control.

[0180] Effect of salt concentration on NADH binding to 3A-PEI crystals

[0181] 3A-PEI-NADH crystals (10 mg, 183 ± 2 nmol NADH) were prepared in Na-PB (25 mM, pH: 7.0) and then incubated in the presence of gradually increasing NaCl (0-800 mM) at 25 °C for 10 minutes, respectively. Then, the samples were centrifuged at 15000 rpm for 5 minutes and the amount of NADH released from 3A-PEI crystals was measured spectrophotometrically at 340 nm.

[0182] Release of NADH from 3A-PEI crystals by mechanical agitation

[0183] The release of NADH from 3A-PEI-NADH crystals (10 mg, 2 μmol) was performed by resuspending them in sodium phosphate buffer (Na-PB, 10 mM, pH 7.0) and incubating them at 25 °C under different shaking speeds (0, 400, 800, 1200, 1600 and 2000 rpm) To investigate the potential influence of mechanical stirring on the release of NADH molecules from 3A-PEI crystals, the samples were incubated for 10 min at 1000 rpm and 25 °C in a thermomixer (Eppendorf Thermomixer Comfort, Eppendorf AG, Germany). Subsequently, the samples were immediately centrifuged for 5 min at 15000 rpm and the amount of released NADH was determined spectrophotometrically at 340 nm using a UV-Vis spectrophotometer (Eppendorf BioSpectrometer, Eppendorf AG, Germany).

[0184] Rebinding of NADH molecules to 3A-PEI crystals

[0185] To investigate the rebinding of NADH molecules to 3A-PEI crystals, 3A-PEI-NADH crystals (10 mg, 2 pmol NADH) in Na-PB buffer (10 mM, pH: 7.0) were first subjected to mechanical stirring at 1000 rpm and 25 °C for 10 min. Subsequently, the crystals were immediately centrifuged for 5 min at 15000 rpm and the amount of released NADH was quantified. Then, the supernatant was returned to the original tube to resuspend the precipitated crystals. The suspension was then incubated for 30 min at 10 °C under gentle rotation. Next, the sample was centrifuged and the amount of NADH bound to the crystals was measured.

[0186] Binding of NADH to 3A-PEI crystals as a function of mechanical stirring speed

[0187] To determine the K d and B max constants for the binding of NADH to 3A-PEI crystals under different mechanical shaking conditions, a fixed amount of 3A-PEI crystals (0.1 mg) was incubated with different concentrations of NADH (0-1000 mM) under different mechanical shaking conditions (0, 400, 800, 1200, 1600, and 2000 rpm) for 5 min at 25 °C. Then, the samples were subsequently centrifuged for 5 min at 15000 rpm to measure the amount of NADH bound to the crystals at 340 nm.

[0188] Preparation of PEI-modified mesoporous silica beads

[0189] PEI-modified mesoporous silica beads with a diameter of 1 pm or 150 nm and a pore size of 2 nm or 3 nm were used. ​Different mesoporous silica beads (Glantreo Company, Ireland) were used to prepare PEI particles for comparison with 3A-PEI crystals. Each silica bead was treated with APTMS alone to expose the amino groups by treating the surface hydroxyl groups [24, 25]. This treatment was achieved by immersing the beads in a 0.2% APTMS solution in dry toluene under vacuum for 15 hours. The beads were then washed with a mixture of toluene and methanol (50% v / v) to remove residual methanol. The beads were then dried at 80°C for 4 hours. The amino-functionalised silica beads were gently rotated and immersed in a solution of glutaraldehyde (0.5% wt in PBS, 100 mM, pH: 7.0) as a cross-linking agent for 30 minutes at 25°C. Subsequently, a PEI solution (0.6 mg / mL) was added to the functionalised beads in PBS buffer (100 mM, pH: 7.0) and allowed to gently rotate for 2 hours at 25°C. To block any remaining free amine groups, the beads were rinsed with ethanolamine-HCl (0.5 M, pH 8.5). The amount of PEI remaining in solution was used to determine the amount of PEI bound to the beads. The samples were then washed with ddH2O multiple times and incubated with a solution of 2.0 mM NADH in ddH2O at 10°C for 30 minutes to determine the binding capacity. The amount of NADH bound in the beads was determined at 340 nm. NADH immobilisation on the beads was further analysed using confocal fluorescence microscopy.

[0190] Co-immobilisation of GFP and mCherry proteins mediated by Cry3Aa

[0191] The plasmids pHT-Cry3Aa-mCherry and pHT315-Cry3Aa-GFP were previously provided

[26] . To prepare the pHT315-Cry3Aa-GFP-RBS-Cry3Aa-mCherry plasmid, the Cry3Aa-mCherry gene was amplified by Kappa HiFi DNA polymerase (Kappa Biosystems) using primers with a ribosome binding site (RBS) upstream (Table 1). Subsequently, the Cry3Aa-GFP gene was amplified using primers with a RBS upstream and a stop codon downstream (Table 1). The two PCR products were then purified and mixed together in a 1 : 1 ratio. The mixture was then used to transform E. coli DH5a cells. The plasmid was purified from the colonies and the sequence was confirmed by Sanger sequencing. The pHT315-Cry3Aa-GFP plasmid was amplified in super competent cells and then digested with Kpnl after purification by Takara MiniBEST kit 4.0. The amplified Cry3Aa-mCherry gene was fused to the linearized pHT315-Cry3Aa-GFP downstream using Gibson Assembly Reaction (NEB). After sequencing the plasmid, it was transformed into the ER2925 strain by heat shock for non-methylation. Then, the plasmid was purified and used for transformation into B. thuringiensis BT407G cells by electroporation. Cells were expressed in SSM medium in the presence of erythromycin (50 pg / ml) for 72 hours at 25 °C and 220 rpm (Schaeffer sporulation) and then harvested by centrifugation at 8000 rpm. After incubation of the cells in chicken lysozyme (Sigma) (1 mg / ml) overnight, samples were sonicated in ice / water bath for 3 cycles of 2 seconds on, 2 seconds off at 40% power. The samples were then washed with 0.5 M NaCl, 0.25 M NaCl and ddH20 for fluorescence microscopy studies using a Nikon TE300 microscope. Pearson correlation coefficient values were obtained by ImageJ software.

[0192] Cry3Aa NADH-dependent enzyme fusion crystals

[0193] To produce Cry3Aa fusion NADH-dependent enzyme fusion crystals, pHT315-Cry3Aa-FDH, pHT315-Cry3Aa-LDH and pHT315-Cry3Aa-FDH / Cry3Aa-LDH plasmids were constructed as follows: To construct expression plasmids for soluble formate dehydrogenase (FDH), leucine dehydrogenase (LDH) and alcohol dehydrogenase (ADH), the fdh gene from Thiobacillus sp. KNK65MA and the adh gene from Bartonella apis were synthesized by Integrated DNA Technologies, while the LDH gene was amplified from the genome of Bacillus thuringiensis 407G using Kappa HiFi DNA polymerase (Kappa Biosystems) following standard protocols. Subsequently, the respective genes were cloned into the pET28b vector downstream orientation using XhoI and NdeI restriction sites by Gibson assembly reaction (NEB). This resulted in the insertion of a C-terminal His-tag for FDH and a N-terminal His-tag for LDH and ADH. To construct plasmids expressing Cry3A-FDH, Cry3A-LDH and Cry3Aa-ADH, the FDH, LDH and ADH genes were first amplified by PCR reaction and then inserted into the pHT315 vector downstream of the Cry3A gene using Gibson Assembly at BamHI and Kpnl restriction sites. To provide plasmids for co-expression of Cry3A-FDH and Cry3Aa-LDH or Cry3Aa-FDH and Cry3Aa-ADH, the pHT315-Cry3Aa-FDH plasmid was first amplified in XL The pHT315-Cry3Aa-FDH plasmid was amplified in super competent cells and then digested by Kpnl after purification by Takara MiniBEST Kit 4.0. Then, the Cry3A-LDH and Cry3A-ADH genes were individually amplified with primers containing ribosome binding sites (RBS). Finally, using pHT315-Cry3Aa-FDH linearized at Kpnl restriction site and the amplified Cry3A-FDH and Cry3A-ADH DNA fragments, co-expression plasmids pHT315-Cry3A-FDH / Cry3A-LDH, pHT315-Cry3A-FDH / Cry3A-ADH and pHT315-Cry3A-FDH / Cry3Aa-FDH / Cry3A-ADH were constructed using Gibson assembly reaction (NEB). A detailed list of primers used to produce the different clones is presented in Table 1. The XL Supercompetent cells were used for all DNA modification experiments. Plasmid purification was performed using Takara MiniBEST Kit 4.0 and successful plasmids were sequenced by BGI.

[0194] Table 1. List of primers used for cloning (5’-3’). Restriction sites are underlined.

[0195]

[0196]

[0197] The plasmids were then introduced into B. thuringiensis BT407G cells by electroporation, respectively. Then, the cells were cultured in 500 mL SSM medium containing sporulation salts (Schaeffer sporulation) and 50 pg / ml erythromycin at 25 °C and 220 rpm for 72 h. After the culture, the cells were collected by centrifugation at 8000 rpm for 8 min using an Avanti J25 Ultracentrifuge (Beckman Coulter). Then, the cell pellets were resuspended in ddH2O in the presence of chicken lysozyme (1 mg / ml, Sigma) at room temperature overnight, followed by sonication in ice-water bath for 30 min to achieve complete cell lysis. After that, the crystals were purified with 0.5 M NaCl, 0.25 M NaCl and ddH2O. Then the proteins were characterized using 10% SDS-PAGE gel. The concentration of the crystals was determined using Bradford assay (Bio-Rad).

[0198] Combination of 3A-PEI-NADH and Cry3Aa-FDH / Cry3Aa-LDH crystals for L-tertiary leucine biosynthesis

[0199] In this study, the aim was to investigate the recycling and retention of NADH in the coupled system of NADH-loaded 3A-PEI and Cry3Aa-FDH / Cry3Aa crystals for L-tert-Leu biosynthesis. For this purpose, 12 mg of 3A-PEI crystals were first incubated in the presence of NADH (0.5 mM) in ddH2O for 10 min at 10 °C under slow rotation. Then, the amount of bound NADH was quantified using spectroscopy at 340 nm. After washing the crystals several times with ddH2O, the provided 3A-PEI crystals (0.5 pmol of loaded NADH) were mixed with Cry3Aa-FDH / Cry3Aa-LDH crystals (12 mg / ml) in a reaction buffer containing 30 mM ammonium formate (pH 8.0) and TMP (25 M). Then, the samples were subjected to mechanical stirring (1000 rpm) for 23.5 h at 25 °C. Then, the samples were removed and incubated at 10 °C for 30 min before starting the next cycle. The samples were centrifuged at 7000 rpm for 10 min, and the supernatant was analyzed by HPLC. Cry3Aa-FDH / Cry3Aa-LDH crystals were used as a control, and 0.5 pmol of NADH was added at each cycle under the same conditions. When a decrease in activity was observed relative to the control, a partial amount of NADH was added to the mixture of 3A-PEI and Cry3Aa-FDH / Cry3Aa crystals to maintain the conversion at the highest rate. This process was repeated for 21 cycles.

[0200] Results

[0201] Preparation and characterization of 3A-PEI crystals

[0202] Cry3Aa is a self-crystallizing protein naturally produced by certain Bt strains. Our team has used crystals and particles of Cry3Aa to entrap positively charged small peptides and proteins due to the exposed surface of domain II of Cry3Aa, which forms the solvent channel ( Figure 1A)) the presence of a large number of negatively charged residues, aspartate (Asp) and glutamate (Glu) residues. We have engineered a protein called Pos3Aa, which is derived from Cry3Aa with all of these residues replaced with positively charged arginine (Arg) and lysine (Lys) residues in domain II, which provides a positively charged channel within the crystal. We hypothesized that these crystals would be able to ionically entrap negatively charged molecules such as phosphorylated co-factors within the channel. Thus, first, we tested the ability of Pos3Aa crystals to bind NADH. Although NADH binding was achievable, the amount of NADH binding was deemed insufficient for practical implementation. Thus, we explored alternative methods by considering cationizing the crystals using polyethyleneimine (PEI). This polymer (43 kDa) has 3 to 3.6 times more protonated nitrogen per molecular weight unit (one positive charge per 157 and 129 molecular weight, respectively) than polyArg and polyLys. Thus, the cationization process was expected to yield highly positively charged crystals.

[0203] To prepare PEI-modified Cry3Aa crystals (3A-PEI), Cry3Aa crystals were purified using standard protocols. The purity of the crystals was then analyzed using a 10% SDS-PAGE gel. The presence of a clear band at 73 kDa confirmed the high purity of the obtained crystals. Cry3Aa crystals were then incubated in the presence of PEI polymer at room temperature for 30 minutes under slow rotation. This incubation was postulated to allow the penetration of PEI molecules into the channels of Cry3Aa crystals. EDC / NHS was subsequently added for the permanent cross-linking of PEI to the crystals Figure 1 B).

[0204] PEI with Cu 2+ ions forms a deep blue color, which can be used for visualization and quantification

[27] . The obtained crystals were resuspended and washed several times with NaCl (100 mM) and ddH2O to remove all non-covalently bound PEI or other molecules, and then stored at 4 °C for further use. When 3A-PEI was reacted with Cu 2+ ions, an immediate color change from white to a distinct deep blue hue occurred, confirming the binding of PEI in the Cry3Aa crystals. In contrast, when 3A-PEI was reacted with Cu 2+In the presence of ions, Cry3Aa crystals alone did not show any color change. The concentration of PEI in Cry3Aa crystals was determined to be about 12 ± 0.3 mg PEI / g Cry3Aa crystals. According to our calculations, the molar ratio should be 0.001 PEI:Cry3Aa monomer. Assuming the size of Cry3Aa crystals to be 1.6 pm x 0.80 pm x 0.80 pm

[49] , this indicates that one single Cry3Aa crystal should have 6425 ± 2102 PEI molecules when the binding and distribution of all crystals are the same.

[0205] Scanning electron microscope (SEM) images showed that Cry3Aa crystals are rod-like crystals with an average length and width of 1.6 x 0.8 pm, which is similar to the average length and width of 3A-PEI crystals (1.6 x 0.8 pm) (A-D). Figure 2 A-D). The size distribution of Cry3Aa and 3A-PEI crystals determined by dynamic light scattering (DLS) agreed with their average hydrodynamic diameters of about 825 nm (PDI: 0.08) and about 914 nm (PDI: 0.102), respectively (E). We assume that the slight increase in 3A-PEI crystals compared to Cry3Aa crystals observed in the DLS study can be attributed to the binding of some PEI molecules on the crystal surface. In summary, the results indicate that the binding of PEI to Cry3Aa crystals has no significant effect on the crystal size, crystallinity, and morphology. However, the zeta potential analysis of Cry3Aa and 3A-PEI crystals showed a drastic change from the zeta potential of -35 mV of the original crystals to a positive value of +45 mV of 3A-PEI crystals (F). These results prove the successful PEI-mediated cationization of Cry3Aa crystals. Figure 2 Figure 2

[0206] Binding and retention of cofactors to 3A-PEI crystals

[0207] After incubation of NADH solution with the crystals, the NADH binding was evaluated using spectrometry at 340 nm. The NADH binding capacity of 3A-PEI crystals was determined to be 550 ± 5 pmol / g 3A-PEI crystals. Although Cry3Aa crystals and their positively charged variant, Pos3Aa crystals, were able to bind NADH (6 ± 2 and 11.2 ± 0.7 pmol / g, respectively), their utilization was not suitable for practical applications due to the low binding capacity (A). Figure 3 In other words, 3A-PEI crystals showed a 91-fold and 50-fold higher NADH binding capacity than Cry3Aa and Pos3Aa crystals, respectively, proving the superior NADH immobilization ability of 3A-PEI crystals. To evaluate the retention of NADH by 3A-PEI crystals, a washing test was performed. As Figure 3 ​​B, after 9 wash cycles with Na-PB (10 mM, pH: 7.0), less than 50% of the NADH bound to the 3A-PEI crystals was released, while the bound NADH remained stable up to 20 wash cycles without significant loss. These results indicate that the 3A-PEI crystals have a high capacity for NADH molecule retention. To confirm that the binding between NADH molecules and 3A-PEI crystals is an electrostatic interaction, all NADH molecules were released from the crystals after 20 wash cycles using 1 M Na-PB (pH: 7.0). The NaPB buffer consists of phosphate (PO4 3- ) ions that can compete with the NADH molecules for binding to the PEI molecules.

[0208] To evaluate whether 3A-PEI crystals can bind other phosphorylated cofactors, PLP, NAD + , NADP + , NADPH, FADH2, CoA, riboflavin, and TPP (each at a concentration of 0.25 mM) were incubated with 20 μg of 3A-PEI crystals at 10 °C under slow rotation for 30 min. Subsequently, the crystals were centrifuged, and the amount of bound cofactor within the crystals was quantified. As shown in Figure 3 C, 3A-PEI crystals can also retain these other phosphorylated cofactors. Among the cofactors studied, PLP exhibited the strongest binding affinity to 3A-PEI crystals (650 ± 2 μmol / g support), while TPP exhibited the weakest binding affinity to 3A-PEI crystals (160 ± 1 μmol / g support). These data indicate that 3A-PEI crystals can be considered as a general platform for cofactor retention in biocatalytic systems.

[0209] Structural details of 3A-PEI crystals

[0210] Cry3Aa crystals have a consistent packing interface in vitro and in vivo, exhibiting the same structure under both conditions. We took advantage of this to investigate the effect of PEI modification on the crystal structure. For this purpose, Cry3Aa crystals were purified and dissolved, and then used to produce micro-sized crystals under in vitro conditions. The crystals grown with PEI modification were then used in the same procedure as previously described. The structure of these PEI-modified in vitro-grown Cry3Aa crystals was then determined using X-ray crystallography. Analysis of the structure showed no evidence of any altered or specific interactions between the PEI polymer and the crystals (Table A and Figure 40 ). These data indicate that the binding of PEI molecules within the Cry3Aa crystals occurs in a random manner.

[0211] Table A. Data collection and refinement statistics for PEI-3A crystals

[0212]

[0213] Although confocal imaging of PEI-modified in vivo grown Cry3Aa crystals is feasible, these crystals are too small to determine the localization and distribution of NADH and PEI molecules within the crystals. Therefore, PEI-modified in vitro grown Cry3Aa crystals were chosen for this analysis. Here, we used the blue fluorescence of NADH in the DAPI channel to show the binding pattern of NADH molecules within the crystals using confocal fluorescence microscopy. For this purpose, PEI-modified in vitro grown Cry3Aa crystals were incubated in the presence of NADH solution and then used for confocal fluorescence microscopy studies after washing several times with ddH2O. NADH molecules were distributed and localized throughout the crystals. Z-stack analysis confirmed the localization of NADH molecules within the crystals.

[0214] The binding of PEI molecules in Cry3Aa crystals is another important question that needs to be answered. To address this question, Cry3Aa crystals were labeled with Alexa 568 fluorophore (in red) while PEI molecules were independently labeled with Alexa 488 fluorophore (in green). Subsequently, the PEI modification process was performed as previously described. PEI molecules were located throughout the crystals. Z-stack analysis showed that the crystals were filled with PEI molecules, demonstrating that PEI polymers can penetrate the crystals. After comparing the size distribution of PEI polymers before and after labeling with Cry3Aa crystals, it was realized that PEI polymers with a size distribution less than 5 nm were more likely to attach to the crystals. These data support the concept that PEI molecules enter and label the interior of these channels. To obtain 3D reconstruction of NADH-loaded 3A-PEI crystals, PEI-Alexa488-modified Cry3Aa-Alexa568 crystals were incubated in the presence of NADH molecules and then analyzed simultaneously in the Alexa 488, Alexa 568, and DAPI channels Figure 4 ).

[0215] Stability of 3A-PEI-NADH crystals

[0216] The interaction between NADH molecules and 3A-PEI crystals can be affected by two main factors including pH and ionic strength. pH is a key element that can affect the protonation state of amine groups on PEI molecules (pKa value of ~9.6). These amine groups become protonated in acidic environments, leading to an increase in positive charge, while the positive charge decreases under basic conditions due to deprotonation. The pH range of Na-PB buffer is 5.8 to 8.0, which can be used to study the effect of pH on NADH binding into 3A-PEI crystals. As shown in Figure 14, the intensity of NADH fluorescence in the DAPI channel decreased as the pH of the Na-PB buffer increased. This result indicates that the binding of NADH molecules to 3A-PEI crystals is affected by the pH of the buffer. The intensity of NADH fluorescence in the DAPI channel was higher at pH 5.8 than at pH 8.0, indicating that the binding of NADH molecules to 3A-PEI crystals is affected by the pH of the buffer. This result indicates that the binding of NADH molecules to 3A-PEI crystals is affected by the pH of the buffer. Figure 5As shown in Table A, the NADH binding capacity of 3A-PEI crystals in NaPB (25 mM, pH: 7.0) was 412 ± 4 μmol / g. However, when the pH was decreased to 6.0 in the same buffer, the NADH binding capacity reached 453.5 ± 2.8 μmol / g. On the other hand, when the pH was adjusted to 8.0, the NADH binding capacity decreased to 350.2 ± 3.6 μmol / g. These results indicate that the stability of NADH molecules and 3A-PEI crystals is dependent on the pH.

[0217] The second key factor, ionic strength, is described as the concentration and charge of ions in a solution. In this study, the release of NADH molecules from 3A-PEI-NADH crystals was measured using a fixed concentration of NaPB (25 mM, pH: 7.0) and different concentrations of NaCl solutions (ranging from 0 to 800 mM). Figure 5 Table B shows that the release rate of NADH significantly increased from 5 ± 2.4% to 94 ± 5.3% by increasing the concentration of NaCl from 0 mM to 800 mM. These results indicate that the increase in ionic strength leads to a decrease in the stability of NADH molecules bound within 3A-PEI crystals due to increased competition for anion binding into the crystals.

[0218] Strategies for releasing NADH from 3A-PEI crystals by mechanical agitation

[0219] Several strategies have been developed to achieve the release of molecules by using the responsiveness of polymers to dynamic mechanical environments. Several studies have used mechanical force as a stimulus for the release of molecules from polymeric matrices, self-assembled polymeric micelles, or phospholipid liposomes induced by mechanical perturbation [28-31]. A promising study by Jen et al. (2016) showed that fluorescent mechanophores generated by Michael addition reactions and stabilized by protonated amines via electrostatic interactions could be reversibly released by external mechanical forces

[32] . In another similar study, Imato et al. (2020) reported that fluorescent pyrene and adjacent naphthalimide-linked polymer films locally fold through charge transfer, which can be released by mechanical stretching

[33] . In sufficiently soft materials such as polyurethane, it has been shown that upon relaxation, pyrene can be reassociated and refolded by the polymer, indicating a reversible behavior

[34] . However, in amorphous polycaprolactone, it was irreversible, possibly due to strain-induced ordering and crystallization of the polymeric matrix

[33] . An interesting advance in polymer technology has led to the generation of mechanocatalysts with outstanding responsiveness when subjected to mechanical agitation

[35] . Although few mechanically activated catalysts have been reported, most of them are metal-ligand coordination complexes incorporated into the polymer backbone [36, 37]. A good example is the ruthenium-silver-N-heterocyclic carbene complex, which can be activated by stretching forces generated by their separation. Copper-silver-N-heterocyclic carbene complexes have also been reported as promising catalysts controllable by ultrasound technology

[38] .

[0220] These encouraging findings prompted us to explore the possibility of modulating the release and binding of cofactor molecules from 3A-PEI crystals by applying mechanical agitation. To explore this hypothesis, we first used Eppendorf C The stability of 3A-PEI-NADH crystals was examined in NaPB buffer (10 Mm, pH: 7.0) at various levels of orbital shaking (ranging from 0 to 2000 RPM). As shown in Figure 6 A The release rate of NADH molecules from 3A-PEI crystals was significantly increased from 5 ± 2.4% in the absence of orbital shaking to more than 83 ± 2.1% at 2000 rpm. These release rates were considered apparent NADH release rates due to the possibility of NADH rebinding during the centrifugation process prior to NADH quantification. This indicates that the application of mechanical shaking not only accelerated the competition of anions in the buffer with NADH molecules for 3A-PEI binding, but also helped the dissociation of NADH molecules in 3A-PEI crystals by shear forces.

[0221] To understand whether this strategy can be applied to other cofactors, we also studied the release of other phosphorylated cofactors from 3A-PEI crystals. Figure 6 As shown in Figure B, after 2 hours of orbital shaking (1000 rpm) at 25°C, more than 80% of the bound NAD in the crystals was + , NADP + The cofactors, NADPH, FADH2, CoA, and TPP, could be released into solution. However, due to the higher binding affinity of the PLP and riboflavin cofactors, only approximately 40 ± 1% and 65 ± 3% of them, respectively, could be released under the same conditions. These findings indicate that mechanical shaking can effectively release bound cofactors within 3A-PEI crystals, making them available for subsequent use as catalysts in bioprocessing applications.

[0222] Kinetic studies of NADH binding to 3A-PEI crystals showed that the maximum NADH binding capacity (B max ) decreased from 540±5.9nmol / mg to 152±4.8nmol / mg, while K d The values ​​(equilibrium dissociation constants) showed an increase from 111 ± 0.6 μM to 565 ± 6.3 μM (Table 2). These findings indicate that most NADH molecules dissociate under increased mechanical agitation. To determine the relative force required to release NADH molecules from 3A-PEI crystals, the following formula was used:

[0223] Relative force = (RPM) 2 ×1.118×10 -5 ×r

[0224] where relative centrifugal force (RCF) is the radial force generated by the spinning rotor, r is the distance of rotation (1.5 mm), and RPM is considered to be the rotation speed per minute.

[0225] like Figure 6 C, 50% of the NADH molecules bound to the 3A-PEI crystals were released into the solution by increasing the relative force from 0 mN to 1.8 mN. This relative force is an indication of the minimum force required to release 50% of the bound NADH molecules from the 3A-PEI crystals.

[0226] The elimination of mechanical stirring is the trigger for the re-incorporation of NADH molecules into 3APEI crystals. Figure 6As shown in D, after the release of NADH molecules from 3A-PEI crystals by using mechanical agitation, the NADH molecules can be stabilized into 3A-PEI crystals at low temperature by electrostatic interaction when the mechanical agitation is removed. This strategy provides a simple kinetics for the regulation of NADH binding and release in the reaction, ensuring the continuous supply of free cofactor for the reaction under mechanical agitation, and the subsequent cycle by eliminating mechanical agitation at low temperature. Figure 6 The rechargeability of 3A-PEI crystals as shown in E indicates that NADH can be replenished within the crystals even if all the bound NADH molecules are removed. Thus, the crystals retain their ability to be used and utilized effectively.

[0227]

[0228] Table 2. Kinetic parameters of NADH binding to 3A-PEI crystals under different mechanical shaking. The results show that by increasing the mechanical agitation, B max values significantly decreased (almost 3-fold), while K d values increased to about 5-fold. This data represents the importance of mechanical agitation on NADH binding to 3A-PEI crystals or release from 3A-PEI crystals.

[0229] Importance of support for PEI modification

[0230] To explore the importance of the support for PEI modification for NADH binding, three types of PEI-modified silica beads were provided, including PEI-SB1 (size of 1 pm, pore size of 4 nm), PEI-SB2 (size of 1 pm, pore size of 10 nm), and PEI-SB3 (size of 150 nm, pore size of 4 nm). Each type of bead was prepared with an equal amount of anchoring PEI (52 ± 5 mg PEI / g silica beads). Then, the NADH binding capacity of the beads was evaluated. The binding of NADH molecules to PEI-SB was analyzed by using the fluorescence emission of the bound NADH molecules with a confocal microscope. It was shown that PEI labeled with Alexa488 was successfully anchored into SB beads. This observation supports the successful modification of PEI polymers into SB to facilitate the uptake and capture of NADH.

[0231] Interestingly, when comparing beads with the same size but different pore sizes (PEI-SB1 and PEI-SB2), it was found that PEI-SB1 exhibited about two-fold higher NADH binding capacity (60 ± 1.3 pmol / g) Figure 7A). These data suggest that an increase in pore size results in a decrease in the surface area of the silica beads, limiting the availability of binding sites for the interaction between NADH and PEI [39, 40]. In addition, a smaller pore size provides a more confined space for the NADH molecules, promoting stronger interactions with PEI, whereas a larger pore size creates a more open and less confined environment that can lose NADH molecules after binding [41, 42]. Additionally, when comparing beads with the same pore size but different bead sizes (PEI-SB1 and PEI-SB3), it was observed that the smaller beads exhibited a significantly higher NADH immobilization capacity (about 1.9-fold increase) (114 ± 0.9 μmol / g) Figure 7 A). This suggests that smaller beads have a higher surface-to-volume ratio, providing more available binding sites for the interaction between NADH and PEI, resulting in a higher immobilization capacity [43-45].

[0232] To investigate the effect of PEI concentration on the NADH binding capacity of the silica beads (SB), we labeled the SB with different amounts of PEI polymer. Figure 7 B shows that increasing the amount of PEI attached to the beads (0-60 ± 0.6 mg PEI / g SB) resulted in a significant increase in the amount of bound NADH from 0 to 105 ± 1.2 μmol / g. This suggests a positive correlation between the amount of bound NADH and the number of cross-linked PEI macromolecules. However, it is worth noting that a further increase in the amount of PEI can result in a decrease in the NADH binding capacity. These findings suggest that an increase in the concentration of bound PEI on the SB beads results in a higher availability of NADH binding sites. This enhanced availability allows for more binding interactions between NADH molecules and anchoring PEI, resulting in a larger amount of NADH being bound. However, beyond a certain point (specifically, when the amount of PEI-SB1 exceeds 60 mg / g of beads), a further increase in the amount of PEI can decrease the capacity for NADH binding Figure 7 B). This can be due to factors such as overcrowding of PEI molecules, a decrease in the accessible volume for NADH binding, or possible conformational changes that limit binding efficiency.

[0233] Based on the data provided, it was observed that 3A crystals exhibited a maximum binding capacity of 540 ± 1.2 pmol / g Cry3Aa-PEI crystals when anchored to PEI (12 ± 0.3 mg / g), which is much higher than PEI-modified SB beads. However, it is important to note that the densities of SB beads and 3A crystals are different. The results obtained using the pycnometer method to measure the densities of 3A-PEI crystals and PEI-SB indicate that 3A-PEI crystals have a lower density, approximately 60%, compared to beads. Despite the fact that PEI-SB beads have a higher density than 3A-PEI crystals, it is clear that 3A-PEI crystals can achieve approximately 5 times higher NADH binding capacity. It is suggested that the distribution and arrangement of PEI polymers in the solvent channel can affect this NADH binding capacity.

[0234] Recyclability of 3A PEI-NADH + 3A FL mixture

[0235] To explore the feasibility of 3A-PEI crystals in facilitating binding and release in enzymatic reactions, it is crucial to develop biocatalysts that can effectively utilize NADH for the biosynthesis of fine chemicals and regenerate NADH molecules. Our research group previously immobilized enzymes by embedding or gene fusion methods using Cry3Aa technology [12, 13, 46]. Based on the crystallization of Cry3Aa fusion proteins in Bt, we hypothesized that the co-expression of various mixtures of Cry3Aa fusions could contribute to considerable intermolecular interactions, resulting in co-immobilization and co-localization of proteins in crystals. Therefore, to evaluate the feasibility of Cry3Aa fusion crystals for co-immobilization and co-localization of different proteins, we constructed a pHT315-Cry3Aa-GFP-RBS-Cry3Aa-mCherry plasmid in which each Cry3Aa fusion protein has its independent RBS. Then, it was used to co-express Cry3Aa-GFP / Cry3Aa-mCherry crystals in Bt. The distribution and localization of GFP and mCherry fluorescent reporter proteins were evaluated by using a fluorescence microscope. The co-localization of these fluorescent reporter proteins within the same crystal was evident by the dominance of the combined signal (yellowish) from both green (GFP) and red (mCherry) due to their close proximity. Image analysis indicated that the co-expressed crystals showed a high Pearson correlation coefficient of 0.956, indicating a strong association between the distribution and intensity of the corresponding fluorescent signals. These findings suggest that Cry3Aa has the potential to serve as an effective platform for direct co-immobilization of multifunctional enzymes.

[0236] In the present study, Cry3Aa-FDH / Cry3Aa-LDH (3AFL) crystals were expressed and produced in Bt. The crystals were then used to catalyze the conversion of trimethylpyruvic acid (TMP) to L-tert leucine (L-tert Leu) by using ammonia and NADH as cofactors. These crystals can regenerate NADH molecules by oxidizing formic acid to CO2, as shown in Figure 8 .

[0237] Although the NADH regeneration system in 3AFL crystals eliminates the need for continuous addition of NADH, the need for NADH supplementation in each reaction cycle presents a significant challenge for practical applications. To address this limitation, we utilized the capacity of 3APEI crystals to effectively minimize the costs associated with NADH supplementation. As shown in Figure 9As shown, it indicates that 3AFL crystals can maintain 100% conversion over 15 cycles by adding free NADH in each cycle. The turnover number (TNN) of NADH and crystals were calculated as 44 and 9271, respectively. However, 3AFL + 3APEI-NADH mixture indicates that the conversion can be maintained at 100% over 6 reaction cycles without the need for further NADH addition (except for the first cycle (1 μmol of NADH)). The decrease in production rate to 70% at the 7th cycle indicates leaching of NADH due to the inevitable buffer ion effect in the system. Therefore, 0.3 μmol of NADH was added to the mixture to maintain the highest conversion rate. This recharging step enables the reaction to proceed for an additional 6 cycles without the addition of NADH. The subsequent 30% decrease in crystal activity at the 14th cycle requires recharging of 0.3 μmol to continue the reaction until the activity decreases to less than 65% at the 21st cycle. The TNN of NADH increases to 284, while the TNN of the crystals remains at 8409. These results indicate that the capacity of 3APEI crystals to preserve NADH molecules in the reaction is more than 6 times higher than the control. Without the recharging step, this number can reach more than 659, but the conversion rate will decrease to less than 50% after the 12th cycle. The dual system of 3AFL + 3APEI-NADH mainly requires mechanical stirring for catalyzing the reaction, as evidenced by the release of NADH molecules during shaking. The NADH recombination step at the end of each reaction cycle is crucial for the reattachment of the released NADH molecules to the 3APEI crystals. In addition, in the absence of mechanical stirring, the reaction rate of 3AFL crystals is reduced by about five times compared to when mechanical stirring is applied. It is understood that the use of mechanical stirring can accelerate the reaction rate as well as reduce the amount of catalytic agent required for the highest conversion rate and reaction time. In addition, the system allows free access to NADH molecules by mechanical stirring, which is advantageous compared to co-immobilized systems where NADH molecules are usually restricted. Interestingly, due to the consumption of ammonium formate used as a substrate in the system, it causes the pH of the reaction to decrease from 8.0 to 7.75 ± 0.05. This decrease in pH helps NADH recombination at the end of the reaction due to the increase in protonation of the PEI polymer. The assessment of NADH binding in the reaction buffer before and after the reaction indicates that the NADH binding rate almost increases by 30% when the reaction is over. This can be attributed to the consumption of ammonium formate buffer, which reduces the buffer ion effect; and the slight acidification at the end of the reaction, which enhances the positive charge in the PEI polymer, thus facilitating NADH recombination.

[0238] DISCUSSION

[0239] Immobilization of cofactor-dependent enzymes is an important technology for the production of fine chemicals as it allows enzymes to be bound to solids to improve reaction control and maintain stability of heterologous biocatalysts for long-term use

[47] . However, the high cost of certain cofactors, such as NADH, is a major obstacle to the implementation of these biocatalysts, which hinders their widespread use in industry [9]. One possible strategy to address this issue is to co-immobilize enzymes and cofactors very close to each other in a porous support. However, this approach has limitations due to internal diffusion of the cofactor, which can hinder the productivity of the biocatalyst

[48] . Studies have shown that enzymes exhibit reduced activity towards reversibly or irreversibly tethered cofactors on solid materials, highlighting the challenges associated with this approach [49-51]. Furthermore, the design and chemical construction of the support can be complex and expensive as it requires uniform and close positioning of the cofactor and enzyme within the porous structure. This involves strict regulation of the enzyme distribution within the particle by controlling the immobilization rate through the chemical properties of the enzyme / cofactor-support interface. Ensuring diffusion of the cofactor within the pores and full utilization of the three-dimensional (3D) surface of the support, while achieving fast immobilization, constitutes a major challenge [26, 52, 53]. This challenge is amplified by the involvement of various spatiotemporal phenomena at multiple scales, from the nanoscale to the macroscale, such as structural deformation of the enzyme, limited substrate access, and insufficient reactor environment such as inadequate stirring, thus reducing catalytic efficiency [54, 55]. Some studies have explored the use of porous supports with improved size, chemical properties, and structure to improve the positioning and effectiveness of cofactor and enzyme co-immobilization [53, 56, 57]. However, the use of hydrophobic porous materials, such as hydrophobic porous methacrylate beads, still presents challenges related to internal transport of the cofactor

[53] . Although PEI-modified hydrophilic porous supports show enhanced association and dissociation equilibrium of phosphorylated cofactors, improving cofactor access to active sites, the use of this system requires careful handling due to the instability of the complex in aqueous solution

[53] . In a study, phosphorylated cofactors (NAD + , FAD + , and PLP) and enzymes were successfully immobilized on PEI-activated agarose microbeads. This method produced catalysts with similar activity to soluble and co-immobilized NAD + , which could be used in batch mode for up to 4 reaction cycles without the need for additional NAD + . In a flow reactor, they reached a total number of NAD + turnovers (TTN) of 85 after 107 h [9]. However, they faced challenges during the washing process with NAD +The challenge of rapid loss limits their method's practicality. In another study, the same strategy was used to co-immobilize a thermophilic (S)-3-hydroxybutyryl-CoA dehydrogenase from Thermus thermophilus HB27 (Tt27-HBDH) and NADH PEI-modified 6BCL agarose to perform asymmetric reduction of ethyl acetoacetate (EAA). By increasing the enzyme loading (32.3 mg enzyme / g support), NADH and Tt27-HBDH achieved good TTNs of 145 and 5223, respectively, after 10 reaction cycles

[58] . In another strategy, the cofactor is covalently immobilized to the polymeric particles, while the enzyme is separately immobilized with the cofactor. The collision between these particles facilitates the interaction between the cofactor and the enzyme, enabling the desired biotransformation reaction to take place. This catalytic system has been tested for the enzymatic production of methanol from carbon dioxide, which has been explored for greenhouse gas fixation and renewable fuel production. The efficient regeneration and reuse of cofactors are crucial to exploit the potential of the enzymatic biotransformation of CO2. However, due to the limited availability of the cofactor NADH, this system exhibited low product yield after five cycles of recycling. The reaction using free cofactor maintained a constant production rate for each reuse

[49] .

[0240] Inspired by nature, this study utilized 3A crystals with self-crystallization properties as a scaffold for enzymes and co-factors. The introduction of 3A scaffolds as natural supports offers different advantages, including low production cost, easy modification, eliminating the need for complex support synthesis, non-toxic properties, and biocompatibility. By immobilizing NAD(H)-dehydrogenase (FDH) within 3A crystals along with NADH-dependent catalysts (LDH or ADH) that can utilize this co-factor within the same crystals, an efficient NADH regeneration system was established for the asymmetric reduction of alcohols or ketones. This system showed high stability for batch-mode reactions under mechanical stirring for nearly 20 days, achieving the highest conversion rates. The continuous need for NADH supplementation was addressed by introducing 3A PEI-NADH crystals, which produce highly stable complexes. Data showed that more than 50% of the bound NADH remained in the crystals even after more than 20 wash cycles with different buffers (NaPB, PBS, HEPES, or ammonium formate), highlighting the high capacity of these crystals for NADH retention. This capacity (540 ± 5.9 μmol / g) exceeded the capacity of other reported systems such as PEI-modified mesoporous agarose (18 μmol / g) [9] and PEI-modified hydrophilic porous 6% agarose beads (200 μmol / g)

[53] , while the use of PEI polymer was minimized (12 ± 0.3 mg / g support). This difference can be attributed to the unique orientation and distribution of functionalized carboxyl groups (Asp and Glu residues) within the solvent channels, which help PEI labeling and provide highly ordered cationized solvent channels. This study presents an innovative strategy using mechanical stirring as a trigger for NADH molecule release and re-binding in NAD(H)-dependent systems. This approach allows the release of NADH when it is needed for the catalytic reaction by applying mechanical stirring, and subsequently re-binding of the co-factor by eliminating mechanical stirring. This simple method provides an effective means of controlling NADH availability in biocatalytic reactions, which can improve results and retain NADH molecules for the next cycle. Kinetic studies exploring the binding and release of NADH from 3A PEI crystals under different mechanical stirring showed a significant decrease in the B max value and an increase in the K dThe increase in the value indicates the dissociation of NADH molecules from the crystal under mechanical agitation. The subsequent re-association of NADH molecules to the 3APEI crystals after removal of mechanical agitation represents an interesting mechanical response phenomenon. The applicability of this method was tested for 3AFL crystals to effectively produce L-tert-leucine with minimal amount of NADH. This study indicates the need for mechanical agitation to effectively diffuse the substrate or cofactor molecules into the crystal, ultimately increasing the production rate. Considering that we have also demonstrated the efficient binding and release of other phosphorylated cofactors by 3A-PEI crystals, this simple and convenient method can be applicable to other cofactor-dependent catalytic systems.

[0241] Conclusion

[0242] This study presents an effective strategy for controlling the accessibility of NADH molecules in an aqueous environment to allow their participation in NAD(H)-dependent reactions by applying mechanical agitation. This method enables the 3APEI-NADH dual system connected to 3AFL and 3AFFA to catalyze multiple cycles of biosynthesis of chiral compounds such as L-tert-leucine and S-EHB with minimal amounts of NADH. The recyclability and rechargeability of this dual system ensure its long-term availability for continuous reactions.

[0243] References

[0244] 1. Roche, J., et al., NADH regenerated using immobilized FDH in a continuously supplied reactor - Application to l-lactate synthesis. Chemical Engineering Journal, 2014. 239: p. 216-225.

[0245] 2. Saba, T., et al., NADH Regeneration: A Case Study of Pt-Catalyzed NAD+ Reduction with H2. ACS Catalysis, 2021. 11(1): p. 283-289.

[0246] 3. Ali, I., B. Soomro, and S. Omanovic, Electrochemical regeneration of NADH on a glassy carbon electrode surface: The influence of electrolysis potential. Electrochemistry Communications, 2011. 13(6): p. 562-565.

[0247] 4. Wang, X., et al., Cofactor NAD(P)H Regeneration Inspired by Heterogeneous Pathways. Chem, 2017. 2(5): p. 621-654.

[0248] 5. Velasco-Lozano, S., A.I. Benitez-Mateos, and F. Lopez-Gallego, Co-immobilized Phosphorylated Cofactors and Enzymes as Self-Sufficient Heterogeneous Biocatalysts for Chemical Processes. Angew Chem Int Ed Engl, 2017. 56(3): p. 771-775.

[0249] 6. Ji, X., et al., Tethering of Nicotinamide Adenine Dinucleotide Inside Hollow Nanofibers for High-Yield Synthesis of Methanol from Carbon Dioxide Catalyzed by Coencapsulated Multienzymes. ACS Nano, 2015. 9(4): p. 4600-4610.

[0250] 7. Li, P., et al., Hierarchically Engineered Mesoporous Metal-Organic Frameworks toward Cell-free Immobilized Enzyme Systems. Chem, 2018. 4(5): p. 1022-1034.

[0251] 8. Liang, J., et al., Hierarchically Porous Biocatalytic MOF Microreactors as a Versatile Platform towards Enhanced Multienzyme and Cofactor-Dependent Biocatalysis. 2021. 60(10): p. 5421-5428.

[0252] 9. Velasco-Lozano, S., A.I. Benitez-Mateos, and F. Lopez-Gallego, Co-immobilized Phosphorylated Cofactors and Enzymes as Self-Sufficient Heterogeneous Biocatalysts for Chemical Processes. 2017. 56(3): p. 771-775.

[0253] 10. Fu, J., et al., Multi-enzyme complexes on DNA scaffolds capable of substrate channelling with an artificial swinging arm. Nat Nanotechnol, 2014. 9(7): p. 531-6.

[0254] 11. Hartley, C.J., et al., Engineered enzymes that retain and regenerate their cofactors enable continuous-flow biocatalysis. Nature Catalysis, 2019. 2(11): p. 1006-1015.

[0255] 12. Heater, B. S., et al., Directed evolution of a genetically encoded immobilized lipase for the efficient production of biodiesel from waste cooking oil. Biotechnol Biofuels, 2019. 12: p. 165.

[0256] 13. Heater, B. S., et al., In Vivo Enzyme Entrapment in a Protein Crystal. Journal of the American Chemical Society, 2020. 142(22): p. 9879-9883.

[0257] 14. Emsley, P. and K. Cowtan, Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr, 2004. 60(Pt 12 Pt 1): p. 2126-32.

[0258] 15. Yang, Z., M. M. M. Lee, and M. K. Chan, Efficient intracellular delivery of p53 protein by engineered protein crystals restores tumor suppressing function in vivo. Biomaterials, 2021. 271: p. 120759.

[0259] 16. Mounsef, J. R., et al., A simple method for the separation of Bacillus thuringiensis spores and crystals. J Microbiol Methods, 2014. 107: p. 147-9.

[0260] 17. Ungaro, F., et al., Spectrophotometric determination of polyethylenimine in the presence of an oligonucleotide for the characterization of controlled release formulations. Journal of Pharmaceutical and Biomedical Analysis, 2003. 31(1): p. 143-149.

[0261] 18. Wen, T., et al., A facile, sensitive, and rapid spectrophotometric method for copper(II) ion detection in aqueous media using polyethyleneimine. Arabian Journal of Chemistry, 2017. 10: p. S1680-S1685.

[0262] 19. Otwinowski, Z. and W. Minor, Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol, 1997. 276: p. 307-26.

[0263] 20. McCoy, A.J., et al., Phaser crystallographic software. J Appl Crystallogr, 2007. 40(Pt 4): p. 658-674.

[0264] 21. Winn, M.D., An overview of the CCP4 project in protein crystallography: an example of a collaborative project. J Synchrotron Radiat, 2003. 10(Pt 1): p. 23-5.

[0265] 22. Murshudov, G.N., et al., REFMAC5 for the refinement of macromolecular crystal structures. Acta Crystallogr D Biol Crystallogr, 2011. 67(Pt 4): p. 355-67.

[0266] 23. Liebschner, D., et al., Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr D Struct Biol, 2019. 75(Pt 10): p. 861-877.

[0267] 24. Goscianska, J., A. Olejnik, and I. Nowak, APTES-functionalized mesoporous silica as a vehicle for antipyrine-adsorption and release studies. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2017. 533: p. 187-196.

[0268] 25. Bourkaib, M.C., et al., APTES modified SBA15 and meso-macro silica materials for the immobilization of aminoacylases from Streptomyces ambofaciens. Microporous and Mesoporous Materials, 2021. 323: p. 111226.

[0269] 26. Nair, M.S., et al., Cry Protein Crystals: A Novel Platform for Protein Delivery. PLOS ONE, 2015. 10(6): p. e0127669.

[0270] 27. Semenova, A., et al., Copper-Binding Properties of Polyethylenimine-Silica Nanocomposite Particles. Langmuir, 2022. 38(34): p. 10585-10600.

[0271] 28. Lee, K.Y., M.C. Peters, and D.J. Mooney, Controlled Drug Delivery from Polymers by Mechanical Signals. 2001. 13(11): p. 837-839.

[0272] 29. Lin, H.-Y. and J.L. Thomas, PEG-Lipids and Oligo(ethylene glycol) Surfactants Enhance the Ultrasonic Permeabilizability of Liposomes. Langmuir, 2003. 19(4): p. 1098-1105.

[0273] 30. Küng, R., et al., Mechanochemical Release of Non-Covalently Bound Guests from a Polymer-Decorated Supramolecular Cage. 2021. 60(24): p. 13626-13630.

[0274] 31. White, S.R., et al., Autonomic healing of polymer composites. Nature, 2001. 409(6822): p. 794-7.

[0275] 32. Li, Z.a., et al., Highly Sensitive Built-In Strain Sensors for Polymer Composites: Fluorescence Turn-On Response through Mechanochemical Activation. 2016. 28(31): p. 6592-6597.

[0276] 33. Imato, K., et al., Fluorescent supramolecular mechanophores based on charge- transfer interactions. Chemical Communications, 2020. 56(57): p. 7937- 7940.

[0277] 34. Aerts, A., et al., Pyranine Based Ion-Paired Complex as a Mechanophore in Polyurethanes. 2021. 42(1): p. 2000476.

[0278] 35. Michael, P. and W.H. Binder, A Mechanochemically Triggered “Click” Catalyst. 2015. 54(47): p. 13918-13922.

[0279] 36. Groote, R., et al., Unfolding and Mechanochemical Scission of Supramolecular Polymers Containing a Metal-Ligand Coordination Bond. Macromolecules, 2011. 44(23): p. 9187-9195.

[0280] 37. Piermattei, A., S. Karthikeyan, and R.P. Sijbesma, Activating catalysts with mechanical force. Nature Chemistry, 2009. 1(2): p. 133-137.

[0281] 38. Michael, P., S.K. Sheidaee Mehr, and W.H. Binder, Synthesis and characterization of polymer linked copper(I)bis(N-heterocyclic carbene) mechanocatalysts. 2017. 55(23): p. 3893-3907.

[0282] 39. Shan, C. a., et al., Effects of nano-pore system characteristics on CH4 adsorption capacity in anthracite. Frontiers of Earth Science, 2019. 13(1): p. 75-91.

[0283] 40. Liu, W., et al., Effect of Pore Size Distribution and Amination on Adsorption Capacities of Polymeric Adsorbents. Molecules, 2021. 26(17).

[0284] 41. Suresh Kumar, P., et al., Effect of pore size distribution and particle size of porous metal oxides on phosphate adsorption capacity and kinetics. Chemical Engineering Journal, 2019. 358: p. 160-169.

[0285] 42. Ferrati, S., et al., Leveraging nanochannels for universal, zero-order drug delivery in vivo. Journal of Controlled Release, 2013. 172(3): p. 1011-1019.

[0286] 43. Zhang, B., et al., New insights into the effects of porosity, pore length, pore shape and pore alignment on drug release from extrusion based additive manufactured pharmaceuticals. Additive Manufacturing, 2021. 46: p. 102196.

[0287] 44. Gao, Y., et al., Effect of pore size of three-dimensionally ordered macroporous chitosan-silica matrix on solubility, drug release, and oral bioavailability of loaded-nimodipine. Drug Dev Ind Pharm, 2016. 42(3): p. 464-72.

[0288] 45. Li, J., et al., Effects of pore size on in vitro and in vivo anticancer efficacies of mesoporous silica nanoparticles. RSC Advances, 2018. 8(43): p. 24633-24640.

[0289] 46. Sun, Q., et al., Cry3Aa*SpyCatcher Fusion Crystals Produced in Bacteria as Scaffolds for Multienzyme Coimmobilization. Bioconjugate Chemistry, 2022. 33(2): p. 386-396.

[0290] 47. Sheldon, R.A. and S. van Pelt, Enzyme immobilisation in biocatalysis: why, what and how. Chem Soc Rev, 2013. 42(15): p. 6223-35.

[0291] 48. Rocha-Martín, J., et al., Rational Co-Immobilization of Bi-Enzyme Cascades on Porous Supports and their Applications in Bio-Redox Reactions with In Situ Recycling of Soluble Cofactors. 2012. 4(9): p. 1279-1288.

[0292] 49. El-Zahab, B., D. Donnelly, and P. Wang, Particle-tethered NADH for production of methanol from CO(2) catalyzed by coimmobilized enzymes. Biotechnol Bioeng, 2008. 99(3): p. 508-14.

[0293] 50. DiCosimo, R., et al., Industrial use of immobilized enzymes. Chem Soc Rev, 2013. 42(15): p. 6437-74.

[0294] 51. Ji, X., et al., Enabling multi-enzyme biocatalysis using coaxial-electrospun hollow nanofibers: redesign of artificial cells. J Mater Chem B, 2014. 2(2): p. 181-190.

[0295] 52. Bolivar, J.M., et al., Modulation of the distribution of small proteins within porous matrixes by smart-control of the immobilization rate. J Biotechnol, 2011. 155(4): p. 412-20.

[0296] 53. Benitez-Mateos, A.I., et al., Design of the Enzyme-Carrier Interface to Overcome the O2 and NADH Mass Transfer Limitations of an Immobilized Flavin Oxidase. ACS Applied Materials & Interfaces, 2020. 12(50): p. 56027-56038.

[0297] 54. Cantone, S., et al., Efficient immobilisation of industrial biocatalysts: criteria and constraints for the selection of organic polymeric carriers and immobilisation methods. Chemical Society Reviews, 2013. 42(15): p. 6262-6276.

[0298] 55. Liese, A. and L. Hilterhaus, Evaluation of immobilized enzymes for industrial applications. Chemical Society Reviews, 2013. 42(15): p. 6236-6249.

[0299] 56. Zhou, N., et al., Self-sufficient biocatalysts constructed using chitin-based microspheres. Chemical Engineering Journal, 2023. 459: p. 141660.

[0300] 57. Wang, Z., et al., Durable cofactor immobilization in sol-gel bio-composite thin films for reagentless biosensors and bioreactors using dehydrogenases. Biosensors and Bioelectronics, 2012. 32(1): p. 111-117.

[0301] 58. Orrego, A.H., et al., Self-sufficient asymmetric reduction of beta-ketoesters catalysed by a novel and robust thermophilic alcohol dehydrogenase co-immobilised with NADH. Catal Sci Technol, 2021. 11(9): p. 3217-3230.

[0302] Example 2. Genetically encoded multi-enzyme particles for the biosynthesis of putrescine from L-arginine

[0303] Immobilized enzyme biocatalysts have a wide range of industrial applications, but the multiple steps involved in their preparation significantly increase the time, effort, and cost of producing them. Our group has previously developed a platform for the direct immobilization of enzymes based on the crystal-forming Cry3Aa protein. The application of this platform has been demonstrated for the direct immobilization of single enzymes, with a molecular size limit of 50 kDa. However, the biosynthesis of a large number of compounds is achieved through cascades of multiple enzymes of various sizes. In this paper, we report the use of a larger Cry homolog, CrylAb, for the co-immobilization of two enzymes involved in the biosynthesis of putrescine - human arginase I (hArg, 35 kDa) and ornithine decarboxylase (ODC, 79 kDa) in vivo. We show that CrylAb-mediated immobilization enhances the thermal stability of these enzymes, allowing efficient and recyclable conversion of arginine to putrescine - in the first 7 cycles, 30 mM putrescine was produced in 30 min with near 100% conversion per cycle, which can be directly converted to the thermally stable polyamide nylon-4,6.

[0304] PEI polymer modification of gene-up immobilized ODC mut CrylAb, which enabled further improvement of the system that allowed for the recycling of the cofactor PLP and eliminated the need for its exogenous supplementation in each cycle. As a result, PEI-modified ODC mut CrylAb particles were able to achieve a 24-fold increase in TTN of PLP.

[0305] INTRODUCTION

[0306] Multi-enzyme cascade reaction systems have tremendous potential for a wide range of biomedical and chemical applications 1 — especially for industrial applications where chemical synthesis occurs under harsh and environmentally harmful conditions 2At this point, given the inherent specificity and catalytic properties of enzymes, which produce less toxicity and fewer unwanted byproducts, enzymes not only provide a greener solution, but also a more sustainable and economical production process 3,4 . More importantly, in a confined microenvironment, coupling of multiple enzymes in a cascade, such as co-immobilizing enzymes onto a single support, not only increases the local concentration of intermediate substrates by bringing the active sites spatially close together 5-7 , but also improves the overall reaction efficiency by enabling the use of reactions involving unstable intermediates, which can be easily converted by subsequent enzymes in the enzymatic pathway 6,8 .

[0307] Putrescine is a polyamine found in many organisms, and its biosynthesis can be carried out by several different enzymatic pathways 9-11 . In commerce, putrescine has many industrial applications, including as a precursor to the production of polyamide polymers 12 , such as nylon-4,6, which is produced by the polycondensation of putrescine and adipic acid 13 , and is commonly used in the production of biomedical devices and textiles, as well as automotive components 14,15 . However, most of the putrescine currently used is chemically produced by the hydrogenation of succinonitrile, which is neither sustainable nor environmentally friendly, as it uses non-renewable propylene as a feedstock, and it requires the in situ production of the toxic intermediate hydrogen cyanide 16,17 . To minimize the negative impact on the environment as much as possible, the biosynthesis of putrescine has been extensively explored, with almost all the focus on microbial engineering, with Escherichia coli being the most commonly used system. In many of these whole-cell systems, extensive genetic engineering was performed to overexpress enzymes related to two key putrescine synthesis pathways—(1) the ornithine decarboxylase (ODC) pathway, which utilizes ODC to directly convert L-ornithine to putrescine, and (2) the arginine decarboxylase (ADC) pathway, which utilizes ADC and agmatinase to convert L-arginine to putrescine 13,16,18-23 . To enhance the production of putrescine, additional genetic modifications were also performed, such as the deletion of putrescine degradation pathways and the disruption of competing pathways for the utilization of the precursor L-ornithine 16,18-20,23,24 . However, the amount produced is usually low, partly because high concentrations of putrescine are toxic to the organism 16,23 .

[0308] One strategy to bypass the limitations of the above-mentioned approaches is to use immobilized enzymes for the cell-free biosynthesis of putrescine. In industry, enzyme immobilization is commonly used to achieve higher thermal, mechanical, and solvent stability, as well as better recyclability 4,25-30Furthermore, the enhancement of enzyme selectivity towards non-natural substrates can be achieved by enzyme immobilization, which can induce conformational changes that alter the interaction between the substrate and the active site of the enzyme. 27,28 However, the cost of purifying the free enzyme and its subsequent immobilization limits its widespread use. 4,25,31 .

[0309] Our group has developed a novel enzyme immobilization platform based on Cry proteins, a class of crystal-forming proteins that naturally form crystals during expression in Bacillus thuringiensis (Bt) bacteria. 32-34 We subsequently demonstrated that fusion of lipases to Cry3Aa, a member of the Cry protein family, did not destroy their crystal-forming ability and produced immobilized lipases with high activity and high thermostability, enabling their use in the recyclable production of biodiesel. 35,36 Notably, this approach eliminates the need for column purification of the target enzyme and subsequent immobilization of the enzyme on a solid support and reduces these steps to simple purification of fusion protein crystals by density gradient centrifugation.

[0310] Recently, we further developed this system for enzyme co-immobilization. In our proof-of-concept studies, we genetically fused Cry3Aa protein to SpyCatcher to generate Cry3Aa-SpyCatcher crystals. We demonstrated the successful co-immobilization of various soluble enzyme-SpyTag proteins on these crystals and ultimately demonstrated their enhanced thermal stability and recycling using menaquinone biosynthetic enzymes as a model system. 37 Although co-immobilized crystals exhibit higher reaction rates and are more thermally stable than their free enzyme counterparts, additional steps are required to purify the enzyme-SpyTag, and the efficiency of SpyTag-SpyCatcher conjugation is limited by the size of the target protein. 37 The latter is consistent with our observations to date that the maximum size of the subunits that have been successfully fused to the Cry3Aa protein to produce active fusion crystals in vivo is less than 50 kDa. 35,38,39 .

[0311] As a method to directly immobilize the larger enzyme, we decided to switch from Cry3Aa to a different crystal protein, Cry1Ab, as the immobilization platform. Although Cry1Ab contains the same three-domain core as Cry3Aa (N-terminal α-helical domain, variable domain, and conserved domain), Cry1Ab is larger (130-140 kDa) due to the presence of the C-terminal protoxin crystallization domain. 40 Further motivation came from the fact that previous studies had shown that Cry1Ab expressed in E. coli could form well-shaped particles.41,42 whereas the particles of Cry3Aa are irregular aggregates 35 Considering that E. coli is the most widely studied bacterium, we surmised that its utilization in biocatalyst production would enhance the general applicability of our immobilization technology. In this paper, we describe our efforts in developing an in vivo multi-enzyme immobilization method for one-step production of immobilized multi-enzyme biocatalysts. We report the successful co-immobilization of two key enzymes in the putrescine synthesis pathway, hArg and ODC, in Cry1Ab, leading to the generation of a biocatalytic system capable of cell-free synthesis of putrescine in high yield Figure 10 ). We further demonstrate that the in situ generated putrescine can be directly used for the synthesis of nylon-4,6, a heat-resistant polyamide used for a variety of commercial applications 43 .

[0312] Materials and Methods

[0313] Construction of expression plasmids

[0314] To construct the plasmids expressing ODC and hArg, the speC gene (Accession No: NP_417440.4) encoding ODC and the hArg gene (Accession No: NM_000045.4) were each amplified by PCR and inserted into the pET28 vector by Gibson assembly (NEB). The resulting recombinant plasmids were transformed into E. coli XL10. All positive colonies were verified by DNA sequencing (BGI).

[0315] The plasmids for expressing Cry3Aa fusions and Cry1Ab fusions were prepared by inserting the amplified SpeC, hArg, gfp (Accession No: OQ870305.1) or mCherry gene (Accession No: MK160997.1) into the pHT315-Cry3Aa vector or pET28-Cry1Ab vector, respectively.

[0316] The double mutants of ODC and ODC-Cry1Ab (I163T / E165T) were generated by introducing the corresponding mutations into pET28-ODC and pET28-ODC-Cry1Ab by site-directed mutagenesis using the mutagenic primers listed in Table 3. The resulting plasmids were digested with Dpnl and then transformed into E. coli XL10 as described above.

[0317] Table 3. Primers used for construction of expression plasmids.

[0318]

[0319] To construct the plasmid for co-expression of GFP-CrylAb and mCherry-CrylAb, the mCherry-CrylAb gene was amplified with a primer containing a ribosome binding site (RBS) (AAGGAG) and then inserted into linearized pET28-GFP-CrylAb to generate the plasmid pET28-GFP-CrylAb-RBS-mCherry-CrylAb. Similarly, to co-express hArg-CrylAb and ODC mut -CrylAb, the speCmut-CrylAb was amplified with a primer containing a RBS (AAGGAG) and then inserted into linearized pET28-hArg-CrylAb to generate the plasmid pET28-hArg-CrylAb-RBS-ODC mut -CrylAb.

[0320] All primers used for gene cloning and the amino acid sequences of the proteins used in this study are listed in Table 3 and the Sequence Listing.

[0321] Expression and purification of Cry3Aa fusion crystals

[0322] To produce Cry3Aa-ODC crystals, the pHT315-Cry3Aa-ODC plasmid was transformed into Bt 407G cells by electroporation and the transformed cells were cultured at 25 °C for 72 h. The cell pellet was lysed with lysozyme in an aqueous solution overnight, then sonicated, and washed with NaCl and ddH2O. The expression and purity of the Cry3Aa-ODC crystals were verified by SDS-PAGE. Multiple independent trials of Cry3Aa-ODC crystal expression and purification under the same conditions have been performed and no significant changes in expression levels and purity were observed.

[0323] Expression and purification of CrylAb fusion particles

[0324] Expression of individual CrylAb fusions, such as GFP-CrylAb or mCherry-CrylAb, and CrylAb fusion pairs, such as GFP-CrylAb and mCherry-CrylAb ([GFP-CrylAb / mCherry-CrylAb]) or hArg-CrylAb and ODC mut -CrylAb ([hArg-CrylAb / ODC mutCo-expression of ODC and hArg free enzymes was achieved by transforming their respective plasmids into E. coli BL21(DE3) by heat shock and the resulting colonies were inoculated into 500 mL Terrific Broth (TB, IBI Scientific) supplemented with 4 mL glycerol and 50 pg / mL kanamycin. Cells were then grown at 37 °C until OD 600 0.6-0.8, at which point 2 mM isopropyl beta-D-l-thiogalactopyranoside (IPTG, IBI Scientific) was added and cells were further incubated at 30 °C for 48 h. Cell pellets were resuspended in lysis buffer (50 mM Tris-HCl, 50 mM EDTA, 15% sucrose, pH 8.0) and incubated with lysozyme overnight, followed by sonication and centrifugation the next day. The collected pellets were washed with Crystal Wash I solution (0.5 M NaCl, 2% Triton X-100) and then with 0.5 M NaCl and ddH2O. Expression and purity of the target fusion protein pellets were confirmed by SDS-PAGE. Multiple independent trials of CrylAb fusion pellet expression and purification under the same conditions have been performed and no significant variations in their expression levels and purity were observed.

[0325] Expression and purification of ODC and hArg free enzymes

[0326] For expression of free enzymes, the respective plasmids were transformed into E. coli BL21(DE3) by heat shock. The resulting colonies were cultured in 500 mL Lennox Broth (LB, IBI Scientific) containing 50 pg / mL kanamycin at 37 °C until OD 600 0.6-0.8 was reached. Then 0.2 mM IPTG was added to induce soluble enzyme expression at 18 °C for 20 h. Cell pellets were resuspended in a buffer containing 20 mM Tris (pH 7.5), 300 mM NaCl, 1 mM benzamidine hydrochloride (TCI Chemicals), and 1 mM phenylmethylsulfonyl fluoride (PMSF, Cayman Chemical) and lysed by sonication. Cell lysates were clarified by centrifugation, the resulting supernatant was filtered, and loaded onto a nickel resin column (BioRad) for affinity purification. The protein of interest was eluted with 250 mM imidazole and the purity of the eluted fractions was analyzed by SDS-PAGE.

[0327] Quantification of enzymes in CrylAb immobilized pellets

[0328] Bradford protein assay (BioRad) was used to obtain all protein concentrations 44To quantify the amount of enzyme in CrylAb fusion particles, the total protein concentration (mg / mL) of the particles was converted to its molarity by dividing by the respective combined molecular weight of the enzyme and CrylAb. For fusion particles, the molarity of the enzyme component is equivalent to the molarity of the total particle. Meanwhile, the mass / volume concentration (mg / mL) of the enzyme component can be calculated by multiplying the molarity by its molecular weight.

[0329] Confocal microscopy

[0330] Fluorescence images of co-expressed [GFP-CrylAb / mCherry-CrylAb] particles resuspended in ddH2O as well as mixtures of individually expressed GFP-CrylAb and mCherry-CrylAb particles were acquired using a Leica TCS SP8 confocal microscope. Fluorescence intensity curves and Pearson’s correlation coefficient values were generated using ImageJ software.

[0331] hArg activity assay

[0332] The activity of hArg was determined based on the production of urea following the method of Rahmatullah et al 45 with slight modifications. Briefly, 1 mM of hArg-CrylAb fusion particles (0.165 mg / mL) or hArg enzyme (0.035 mg / mL) were added to a reaction solution containing 30 mM L-arginine in 50 mM HEPES (pH 8.0), 0.1 mM Mn 2+ buffer, and incubated at 40 °C with shaking at 750 rpm for 10 min. The reaction was terminated by 2 M HCIO4, and the produced urea was measured using diacetyl monoxime (DAMO)-thiosemicarbazide (TSC) reagent. The reaction between urea and DAMO-TSC reagent in boiling water for 5 min resulted in the formation of a pink product with an absorption peak at 525 nm. The protein concentration of all constructs used for the assay was determined using the Bradford protein method (BioRad), and the quantification of hArg in hArg-CrylAb fusion particles was performed as previously described. All experiments were performed in triplicate. Multiple independent experiments of activity assays of freshly expressed and purified hArg-CrylAb fusion particles were performed, and no significant variation in activity was observed.

[0333] ODC activity assay

[0334] The activity of ODC was determined following the method of Chinard et al 46The activity of ODC was determined based on the conversion of L-ornithine using the indantrione method. Reaction solutions containing 1 mM of CrylAb ODC fusion particles (0.21 mg / mL) or Cry3Aa ODC fusion crystals (0.152 mg / mL) or ODC enzyme (0.079 mg / mL) and 30 mM L-ornithine in 50 mM HEPES (pH 8.0), 0.1 mM PLP were incubated at 40 °C with shaking at 750 rpm for 30 min. The reaction was terminated by the addition of 2 M HCIO4. The remaining L-ornithine was determined by further reacting the reaction solution with 20 mg / mL indantrione dissolved in 2-methoxyethanol at 90 °C with further shaking at 750 rpm for 30 min. The converted L-ornithine was obtained by subtracting the remaining L-ornithine from the input amount of L-ornithine. Assuming a 1 : 1 stoichiometry of the reaction of L-ornithine to putrescine, the amount of putrescine produced is equal to the converted L-ornithine. The protein concentration was determined for all constructs from the assay using the Bradford protein assay and the quantification of ODC in the Cry ODC fusion particles was performed as previously described. All experiments were performed in triplicate. Multiple independent assays of the activity determination of freshly expressed and purified CrylAb ODC fusion particles were performed and no significant variation in their activity was observed.

[0335] Quantitative analysis of the produced putrescine

[0336] Putrescine produced by co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles

[0337] For the indirect method, the determination of the amount of putrescine is determined by subtracting the remaining L-ornithine at the end of the reaction (which can be conveniently measured by the indantrione method) from the total amount of urea produced from the first reaction step catalyzed by hArg (which can be measured by the DAMO-TSC method). Here, the assumption is based on the fact that the first reaction produces a 1 : 1 molar ratio of urea and L-ornithine. Calibration curves for urea and L-ornithine were obtained by plotting the absorbance at 525 nm and 515 nm, respectively, against their respective concentrations.

[0338] To verify the effectiveness of using the indirect method to quantify the produced putrescine, reaction solutions containing 0.5 mg / mL of co-immobilized particles and 30 mM L-arginine were incubated at 40 °C for 20 min and then analyzed by the indirect method and HPLC.

[0339] For HPLC measurements, the supernatant of the resulting reaction solution was diluted 25-fold with 0.1 M HC1 and then analyzed by HPLC using a SUPELCORP-C 18 column (250 mm x 4.6 mm, 5 μm) by HPLC. Prior to loading, samples were derivatized with o-phthaldialdehyde (OPA). OPA-derivatized products were detected by monitoring the elution of the OPA-derivatized products for fluorescence (excitation wavelength: 335 nm; emission wavelength: 440 nm) 47 . Retention times of derivatized L-arginine, L-ornithine and putrescine were obtained by HPLC measurements of their respective standards. And calibration curves for putrescine were obtained by plotting peak areas versus putrescine concentrations and then used for quantification.

[0340] Kinetic studies

[0341] 0.1 mg / mL CrylAb immobilized particles, 3.5 μg / mL hArg and 7.9 μg / mL ODC mut for kinetic studies. Kinetic parameters of hArg enzymes, immobilized hArg-CrylAb particles and co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles were measured at 40°C using 50 mM HEPES (pH 8.0) containing 0-500 mM L-arginine, 0.1 mM Mn 2+ buffer as substrate (for the particles, kinetic parameters were based on the concentration of hArg) while ODC mut enzymes, immobilized ODC mut -CrylAb particles and co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles were determined at 40°C using 50 mM HEPES (pH 8.0) containing 0-100 mM L-ornithine, 0.1 mM PLP buffer as substrate (for the particles, kinetic parameters were based on the concentration of ODC mut . Urea production was determined by the DAMO-TSC method as previously described 45 . Putrescine production was measured by HPLC. One unit of enzyme activity was defined as the amount of total particles or enzyme required to produce 1 μmol of urea (for hArg) or putrescine (for ODC mut ) per minute at 40°C. Protein concentrations of all constructs used for the assays were determined using the Bradford protein assay (BioRad) and quantification of hArg or ODC mut in CrylAb fusion particles was done as previously described. All experiments were performed in triplicate.

[0342] Co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles, hArg-CrylAb + ODC mut -CrylAb particles and hArg + ODC mut Reaction rate comparison of enzyme mixtures

[0343] Separate immobilization of hArg-CrylAb or ODC mut Calibration curves (concentration vs. band intensity on SDS-PAGE) for hArg-CrylAb or ODC mut -CrylAb particles, (2) 0.27 mg / mL hArg-CrylAb and 0.23 mg / mL ODC mut -CrylAb particles, or (3) 0.057 mg / mL hArg and 0.087 mg / mL ODC mut Reactions of enzyme mixtures were performed at 40 °C and 750 rpm. Reactions were sampled at specific time points (5, 10, 20, 30, and 60 min) to allow determination of putrescine produced using the indirect method as previously described. Concentrations of particles and free enzymes were measured using the Bradford protein assay. All experiments were performed in triplicate.

[0344] Effect of urea on hArg and ODC mut activity

[0345] 1 μΜ hArg (0.035 mg / mL) or ODC mut (0.079 mg / mL) were incubated in different concentrations (0, 250, 500, and 1000 mM) of urea solution at 40 °C with shaking at 750 rpm for 6 h. Residual activity was measured and compared as previously described. Protein concentrations were obtained using the Bradford protein assay. All experiments were performed in triplicate.

[0346] Orbitrap mass spectrometry

[0347] To confirm the identity of the two distinct bands obtained from SDS-PAGE of co-expressed [hArg-CrylAb / ODC mut -CrylAb] particles indeed are hArg-CrylAb and ODC mut- Cry1Ab, the bands were excised from the Coomassie blue-stained gels and cut into small pieces. The gel pieces were destained and dehydrated, then digested with trypsin. The digested samples were analyzed by Orbitrap Fusion Lumos Tribrid Mass Spectrometer (Thermo Fisher Scientific).

[0348] Scanning Electron Microscopy (SEM)

[0349] Samples were prepared by resuspending 20 pg of each particle in 1 mL ddH2O, then sonicated for 1 min. 1 pL of the sonicated sample was loaded onto a copper stub and air-dried overnight. Afterwards, the samples were coated with conductive Au / Pd using a S150B sputter coater (Edwards) and then imaged on a Prisma E (Thermo Scientific) scanning electron microscope at 20 kV and a working distance of 8.0 mm to 8.1 mm. SEM was performed on multiple independent experiments of freshly expressed and purified Cry1Ab immobilized particles and no significant changes were observed in their size and morphology.

[0350] Dynamic Light Scattering (DLS)

[0351] The size and dispersity of immobilized hArg-Cry1Ab particles, ODC mut -Cry1Ab particles and co-immobilized [hArg-Cry1Ab / ODC mut -Cry1Ab] particles were measured using a Malvern Zetasizer Nano ZS90 at 25 °C. DLS was performed on multiple independent experiments of freshly expressed and purified Cry1Ab immobilized particles and no significant changes were observed in their size.

[0352] Thermal stability

[0353] In the thermal stability study, 1 pM hArg enzyme (0.035 mg / mL), hArg-Cry1Ab fusion particles (0.165 mg / mL), ODC or ODC mut enzyme (0.079 mg / mL) and ODC-Cry1Ab or ODC mut -Cry1Ab fusion particles (0.21 mg / mL) were used. For hArg-Cry1Ab fusion particles and hArg enzyme, they were incubated at 50 mM HEPES (pH 8.0), 0.1 mM Mn 2+The cells were incubated in a reaction buffer composed of 50 μL of PBS at 50°C with shaking at 750 rpm for 1 h. 100 μL aliquots of the reaction mixture were then transferred to individual tubes and incubated at different temperatures with shaking at 750 rpm for 1 h, followed by cooling on ice. The residual activity was then measured at 40°C as previously described. The residual activity at 50°C was considered 100%.

[0354] ODC and ODC mut The thermal stability of the proteins and the corresponding Cry1Ab fusion particles was performed in the same reaction buffer as the activity assay and the procedure was similar to that described above for hArg and its Cry1Ab fusion counterparts, except that the residual activity at 30°C was considered as 100%.

[0355] The data were fitted to a sigmoidal curve and T was calculated using GraphPad Prism software. 50 Value. T 50 It is defined as the temperature at which 50% activity is retained after 1 hour incubation. Protein concentration was measured using the Bradford protein assay. All measurements were performed in triplicate.

[0356] Optimization of catalytic conditions

[0357] 0.5 mg / mL co-immobilized [hArg-Cry1Ab / ODC mut [-Cry1Ab] particles and 30mM L-arginine substrate and a reaction time of 10 minutes were used to optimize the temperature and pH of the catalytic reaction. To identify the optimal reaction temperature, the particles and substrate were resuspended in a pH 8 buffer and incubated for the indicated 10 minutes at a range of incubation temperatures. The amount of putrescine produced was determined as previously described. The optimal pH was identified by performing the reaction in different pH buffers (pH 6: 0.1M BIS-Tris buffer; pH 7-8: 0.1M Tris buffer; pH 9: 0.1M CAPSO buffer; pH 10: 0.1M CAPS buffer) and incubating at 40°C for the indicated times.

[0358] To determine whether pH affects the solubility of the fusion protein particles and thus their recyclability, 2 mg / mL of co-immobilized particles were mixed in the same buffer as listed above in the pH optimization experiment as well as an additional alkaline buffer (0.1 M CAPS pH 11, which is known to dissolve Cry crystals) at 40°C in a thermomixer. 35 At the end of the incubation period, the suspension was centrifuged at 15000 rpm for 5 minutes, and the supernatant was transferred for subsequent protein concentration measurement by the Bradford method.

[0359] All experiments were performed in triplicate.

[0360] Recyclability of co-immobilized particles for conversion of L-arginine to putrescine

[0361] Recyclability of co-immobilized particles for conversion of L-arginine to putrescine mut Recyclability of co-immobilized particles for conversion of L-arginine to putrescine 2+ Recyclability of co-immobilized particles for conversion of L-arginine to putrescine mut Recyclability of co-immobilized particles for conversion of L-arginine to putrescine mut Recyclability of co-immobilized particles for conversion of L-arginine to putrescine

[0362] In situ production of nylon-4,6 from putrescine

[0363] To demonstrate the in situ production of nylon-4,6, 4 mg / mL of co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles were incubated with 500 mM of the substrate L-arginine under optimal reaction conditions for 12 hours to convert L-arginine to putrescine. The supernatant was then mixed with NaOH and then reacted with adipoyl chloride in hexane. Nylon membranes were withdrawn by continuously pulling them out as a strand from the interface of the reaction mixture and then washed with 50% MeOH and deionized water and then vacuum dried overnight. The dehydrated nylon was ground into pieces before analysis by FTIR. Fourier transform infrared spectroscopy (FTIR)

[0364] Samples were prepared for FTIR analysis by mixing nylon pieces with potassium bromide (KBr) to make KBr pellets. Infrared spectra were collected by recording the transmittance at different wavenumbers between 4000 and 400 cm -1

[0365] Results and discussion

[0366] Co-immobilization of GFP and mCherry mediated by CrylAb

[0367] We have previously demonstrated that Cry3Aa-hArg crystals are active48 Thus, our initial plan was to co-immobilize hArg and ODC on Cry3Aa. However, it was found that full-length Cry3Aa-ODC crystals were poorly expressed and inactive. We speculated that this was due to the large size of ODC making it difficult for Cry3Aa to form stable interactions present in Cry3Aa crystals 32,49 Indeed, during the development of Cry3Aa immobilization technology, we have observed a general trend that only proteins smaller than 50 kDa can be successfully expressed as highly active Cry3Aa fusion crystals in vivo 35,36,39,49 Thus, we sought another strategy to facilitate its immobilization.

[0368] Since CrylAb is much larger than Cry3Aa, we wondered if it might be better suited for the immobilization of large proteins such as ODC in terms of maintaining enzymatic activity. Moreover, we speculated that since CrylAb forms crystals in Bt, there should be a tendency to form similar interactions in the particles produced by CrylAb-fusions, suggesting that a mixture of CrylAb-fusions might produce co-immobilized particles. To this end, we first investigated the suitability of CrylAb as an immobilization platform for multiple proteins. To facilitate the analysis of the co-localization of different immobilized proteins, we selected the fluorescent reporter proteins GFP and mCherry as fusion partners to CrylAb to produce GFP-CrylAb and mCherry-CrylAb fusions, respectively. We produced GFP-CrylAb and mCherry-CrylAb particles expressed individually and [GFP-CrylAb / mCherry-CrylAb] particles co-expressed in E. coli and characterized by confocal microscopy Figure 11 A). For the co-expressed [GFP-CrylAb / mCherry-CrylAb] sample, the co-localization of the two CrylAb-reporter proteins within the same particle was evident as evidenced by the rich yellow punctate signal in the merged image, in contrast to the separate green and red punctate dots shown in the merged image of the GFP-CrylAb and mCherry-CrylAb particle mixtures Figure 11 A). Moreover, the fluorescence intensity curve of the co-expressed [GFP-CrylAb / mCherry-CrylAb] particles showed a strong correlation between the distribution and intensity of the two fluorescence signals (Pearson correlation coefficient = 0.874), while the fluorescence intensity curve of the mixture of individually immobilized GFP-CrylAb and mCherry-CrylAb showed a different distribution pattern and uncorrelated intensities (Pearson correlation coefficient = 0.162) Figure 11 B).

[0369] Preparation of active ODC-Cry1Ab particles

[0370] Having validated the feasibility of Cry1Ab-fusion particles production in E. coli containing co-immobilized two proteins, GFP and mCherry, our focus shifted to the production of relevant constructs involving the pathway related to the biosynthesis of putrescine from L-arginine Figure 10 ). Considering the problems observed with Cry3Aa-ODC, we prioritized the activity of ODC fused to Cry1Ab. To determine whether the position of ODC in the Cry1Ab fusion would affect its enzymatic activity, ODC was fused to the C-terminal or N-terminal gene of Cry1Ab to produce Cry1Ab-ODC and ODC-Cry1Ab particles, respectively, and their activities were evaluated. The expression of Cry1Ab-ODC and ODC-Cry1Ab in E. coli BL21(DE3) was confirmed by SDS-PAGE. Measurement of their ODC activity showed that while Cry1Ab-ODC particles were inactive, ODC-Cry1Ab particles exhibited activity, albeit lower than that of the free enzyme Figure 12 A). Thus, the N-terminal fusion of ODC to Cry1Ab was used in subsequent experiments.

[0371] Previous studies have shown that ODC is the rate-limiting enzyme for the two-step enzymatic conversion of L-arginine to putrescine via hArg and ODC 11,50 , and Choi et al. generated a double mutant ODC (I163T / E165T) that catalytic efficiency exhibited a 60-fold increase to overcome this limitation 11 . Therefore, we introduced a similar mutation into ODC-Cry1Ab to generate ODC mut -Cry1Ab with the aim of obtaining more active particles. The expression of ODC mut -Cry1Ab in E. coli BL21(DE3) was verified by SDS-PAGE, the activity of the isolated ODC mut -Cry1Ab particles was evaluated, and compared to that of ODC-Cry1Ab particles. As expected, ODC mut -Cry1Ab particles exhibited a greatly improved conversion rate, showing >2.5-fold enhancement Figure 12 A). Preparation of active hArg-Cry1Ab particles

[0372] We then proceeded to prepare and characterize the partner enzyme hArg in the multi-step reaction. The hArg gene was fused to the N-terminal of Cry1Ab, and hArg-Cry1Ab particles were successfully expressed according to SDS-PAGE. The resulting hArg-Cry1Ab particles were also determined to be active (Figure 12 C). Co-immobilized [hArg-CrylAb / ODC mut Preparation and characterization of co-immobilized [hArg-CrylAb / ODC

[0373] Having confirmed that the individually immobilized particles were active, we further prepared co-immobilized particles of ODC mut -CrylAb and hArg-CrylAb. hArg-CrylAb and ODC mut -CrylAb were co-expressed by SDS-PAGE based on the molecular weights corresponding to hArg-CrylAb (165 kDa, lower band) and ODC mut -CrylAb (210 kDa, upper band) were verified, subsequently excised and trypsin digested for mass spectrometry analysis. Their respective identities were confirmed with identified peptides covering >40% of the corresponding sequence.

[0374] The size and morphology of the immobilized particles were then investigated by DLS and SEM. DLS data indicated that hArg-CrylAb, ODC mut -CrylAb and [hArg-CrylAb / ODC mut -CrylAb] particles were 573 nm, 902 nm and 477 nm Figure 13 A). SEM images show that the natively immobilized particles have a spherical shape and exhibit a uniform size distribution, which is an important attribute for biocatalytic reactions Figure 13 B).

[0375] Co-immobilized [hArg-CrylAb / ODC mut Activity of co-immobilized [hArg-CrylAb / ODC

[0376] The activity of co-immobilized [hArg-CrylAb / ODC mut -CrylAb] was then investigated based on its ability to produce putrescine and quantified using the indirect method described in the methods and verified by HPLC. As Figure 14 shown, co-immobilized [hArg-CrylAb / ODC mut -CrylAb] was able to convert 30 mM of L-arginine substrate to 25.1 mM (based on the indirect method) or 26.1 mM (based on HPLC) of putrescine in 20 min at 40 °C, corresponding to about 84% conversion. The agreement of these two results supports the use of the indirect method to determine the putrescine produced in subsequent experiments. It is worth noting that both methods demonstrated that [hArg-CrylAb / ODC mutCry1Ab] particles are active and can facilitate the easy conversion of L-arginine to putrescine Figure 14 ).

[0377] Kinetic studies of Cry1Ab immobilized particles and free enzymes

[0378] To better understand the effect of Cry1Ab-mediated co-immobilization on hArg and ODC mut activities, their kinetic parameters were determined. The apparent K m (appK m ) values of the immobilized enzymes and their corresponding free enzyme counterparts showed close similarities (Table 4), indicating that immobilization did not significantly affect the binding affinity of the substrates to the enzymes. However, the k mut values of the [hArg-Cry1Ab / ODC cat -Cry1Ab] particles were found to be about 50% and 60% of the values of free hArg and ODC mut , respectively (Table 4). This observation indicates a decrease in enzyme activity, presumably due to conformational changes induced by the immobilization process or slower exchange rates of substrates and products within the particle confines.

[0379] Table 4. Kinetic parameters of Cry1Ab-mediated immobilized enzymes and their corresponding free enzymes.

[0380]

[0381] Activity and kinetic parameters were based on the concentration of hArg in the co-immobilized particles. Activity and kinetic parameters were based on the concentration of ODC mut in the co-immobilized particles. *1 U was defined as the amount of total particles or free enzymes required to produce 1 μmol urea or putrescine per minute at 40 °C. 0.1 mg / mL Cry1Ab immobilized particles, 3.5 μg / mL hArg, and 7.9 μg / mL ODC mut were used for the kinetic studies. All protein concentrations were determined using the Bradford method. Quantification of the amount of enzymes in the particles was performed as described in “Materials and Methods”. The k cat values were obtained by dividing the V max values by the corresponding enzyme concentrations, assuming that the substrates were in steady state and in excess. All experiments were performed in triplicate. Data are presented as mean ± standard deviation.

[0382] Reaction rates of co-immobilized particles compared to the mixture of individually immobilized particles and the mixture of free enzymes

[0383] Previous confocal microscopy studies of [GFP-Cry1Ab / mCherry-Cry1Ab] particles (Figure 11 CrylAb-mediated co-immobilization has been shown to result in the co-localization of multiple proteins. By inference, the enzymes within the [hArg-CrylAb / ODC mut -CrylAb] particle should have a similar distribution, which should bring their active sites in close proximity and in turn result in an increased reaction rate. To verify this supposition, the activity of the co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particle was compared to the individually immobilized hArg-CrylAb and ODC mut -CrylAb particle mixture as well as the free hArg and ODC mut protein mixture. Since it is known that urea inactivates enzymes at high concentrations 51,52 , our first task was to determine whether the urea produced by the arginase in the first step of the cascade would affect the reaction rate of the enzymes. The activity of free hArg and ODC mut was not affected even after 6h incubation at 1M urea.

[0384] After ruling out urea as a factor affecting the activity of the enzymes under the conditions used, we proceeded with the enzymatic conversion of L-arginine to putrescine. For both the individually immobilized particle mixture and the free enzyme mixture, a molar ratio of 1.5:1 of hArg to ODC mut was used. This ratio was based on the stoichiometry of the [hArg-CrylAb / ODC mut -CrylAb] particle, which was derived from the densitometric analysis of the band intensities of the SDS-PAGE of the co-expressed [hArg-CrylAb / ODC mut -CrylAb] as well as the calibration curves (concentration versus band intensity) of the individually immobilized hArg-CrylAb and ODC mut -CrylAb. Based on our previous studies on Cry3Aa-mediated co-immobilized enzyme systems 37 , our expectation was that the co-immobilized enzymes would exhibit a higher reaction rate due to the higher local substrate concentration (in this case L-ornithine for the second enzyme ODC mut ). As expected, although the rate of each reaction was slower, the co-immobilized particle produced 1.25 times more putrescine than the mixture of free hArg and ODC mut after 20 minutes of reaction. Figure 15 These data indicate that CrylAb-mediated co-immobilization of multiple enzymes results in an increased reaction rate for a multi-step reaction.

[0385] Our previous study on the Cry3Aa-mediated co-immobilized enzyme system found that the mixture of individually immobilized particles was much slower than the co-immobilized enzyme system. 37 Similarly, with co-immobilized [hArg-Cry1Ab / ODC mut -Cry1Ab] particles alone immobilized hArg-Cry1Ab and ODC mut -Cry1Ab particles showed lower activity - producing 2.5 times less putrescine after 20 minutes ( Figure 15 We speculate that this slower rate stems from the inefficient escape of the intermediate L-ornithine from the particles, which is different from the hArg and ODC in which L-ornithine is freely mobile in solution. mut This is in contrast to the enzyme mixture and the co-immobilized particles where L-ornithine only needs to react within the same particle.

[0386] Thermal stability of Cry1Ab-immobilized particles

[0387] Previous studies have shown that Cry3Aa-mediated immobilization can lead to improved thermal stability. 35,37 Therefore, we studied the effect of Cry1Ab protein on enzyme stability. hArg is Mn 2+ A metalloenzyme that catalyzes the hydrolysis of L-arginine to L-ornithine and urea 53 , and it has been shown that by incubating at 50 °C for a specific time with Mn 2+ Strong coordination 54 Therefore, before studying the thermal stability of hArg and hArg-Cry1Ab, we first investigated the thermal stability of hArg and hArg-Cry1Ab by incubating particles and free proteins with Mn 2+ Incubate at 50 °C for different lengths of time to identify Mn 2+ The incubation time required for optimal integration into the enzyme was then determined, and their residual activities were then determined. After 60 min of incubation, the activities of free hArg and its immobilized form reached a peak and remained almost constant, indicating that Mn 2+ is strongly incorporated into the enzyme. Then, by firstly reacting hArg and hArg-Cry1Ab in a Mn-containing 2+ The thermal stability of ODC and ODC was evaluated by incubating them in the reaction buffer at 50 °C for 60 min and then incubating them at different temperatures for 1 h. mut As well as the thermal stability studies of their corresponding Cry1Ab-immobilized particles, no such preincubation was required, and the enzymes were directly incubated at different temperatures.

[0388] From hArg-Cry1Ab, ODC-Cry1Ab and ODC mut -T obtained by thermal stability study of Cry1Ab50 values of 70.68°C ± 0.05, 57.98°C ± 0.42, and 56.43°C ± 0.26, respectively. Figure 16 In comparison, the free enzyme counterparts hArg, ODC, and ODC mut values of 65.89°C ± 0.11, 51.81°C ± 1.86, and 48.85°C ± 0.91, respectively. Figure 16 These results indicate that the immobilized particles are more thermostable than their free enzyme counterparts and suggest that CrylAb-mediated immobilization leads to enzyme stabilization - possibly due to partial confinement of the protein within the particle hindering un-folding and aggregation. At the same time, it was observed that ODC mut -CrylAb is less stable than hArg-CrylAb, with the smallest loss of activity ranging from 30°C to 40°C Figure 16 B). This indicates that for the recyclability of the co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles, it is best not to exceed this temperature range.

[0389] Optimization of reaction conditions

[0390] Determination of the optimal temperature for maximum activity of the co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles indicates that 45°C yields the highest amount of putrescine, but considering that ODC mut -CrylAb is thermostable in the range of 30°C to 40°C, 40°C was chosen as the optimal reaction temperature during the recyclability study to extend the biocatalyst lifetime. Studies of the co-immobilized particle activity as a function of pH indicate that both pH 8 and pH 9 yield the maximum amount of putrescine. Considering that Cry crystals, including CrylAb, have been shown to dissolve 35 at alkaline pH, and considering that the recyclability of the enzymes can be negatively affected by including losses from dissolved enzymes 4 , the stability of CrylAb was therefore studied by determining the amount of protein released into the supernatant when the co-immobilized enzymes were incubated at different pH. In agreement with previous studies on the stability of most Cry crystals, the co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles are stable to dissolution from pH 6.0 to pH 8.0. Overall, these data indicate that pH 8 is the optimal pH at which the enzymes and the immobilization platform are least affected.

[0391] Recyclability of the co-immobilized particles

[0392] Recyclability of biocatalysts is one of the most important features to evaluate their usability in industrial applications 55as its lifetime directly affects the production cost. Therefore, we tested the ability of 2 mg / mL co-immobilized [hArg-CrylAb / ODC mut -CrylAb] particles to recycle 30 mM L-arginine to putrescine under optimized conditions (40 °C and pH 8.0). More than 80% conversion was maintained even after 9 reaction cycles Figure 17 (A), resulting in a total amount of 321 mM putrescine and a yield of 0.451 g / g L-arginine over 12 cycles. By calculation, the TTN of ODC mut inside the co-immobilized particles is 73284. However, in this case, exogenous PLP is required in each cycle, which significantly increases the total cost of the biosynthetic process. To make the cofactor PLP more cost-effective and reusable, the genetically immobilized ODC mut -CrylAb was modified with a positively charged PEI polymer, which allows the phosphorylated PLP to be tethered, thus recyclable. As a result, PEI-ODC mut -CrylAb-PLP particles were able to facilitate the recyclable biosynthesis of putrescine without the addition of exogenous cofactor, with a TTN PLP of 1736 Figure 17 (B) after 25 cycles. In contrast, without PEI modification of the ODC mut -CrylAb particles, PLP is required in each cycle, with a TTN PLP of only 71.

[0393] Production of putrescine and its application in nylon-4,6 synthesis

[0394] Due to the excellent mechanical properties and thermal stability of nylon-4,6 compared to other nylons, it becomes an attractive material for various industries, and the conversion of putrescine to nylon-4,6 is of great significance 14 ,56 ,57 Bio-based putrescine would be an attractive starting material for nylon-4,6 production, as it can provide a sustainable and environmentally friendly alternative to the traditional petrochemical-based process 17— especially if nylon-4,6 can be produced directly from the reaction solution in which putrescine is biosynthesized from biologically available reactants. To this end, putrescine was biosynthesized in a single-cycle reaction using 500 mM L-arginine as substrate. After 12 hours, about 430 mM (37.9 g / L) of putrescine was produced, with a productivity of 3.16 g / L / h and a total product yield of 0.435 g / g L-arginine. This performance is comparable to that reported for whole cells, as in the case of Bacillus amyloliquefaciens engineered with a synthetic ODC pathway in the study of Li et al., which produced 5.51 g / L of putrescine after 48 h, with a productivity of 0.11 g / L / h and a total product yield of 0.14 g / g xylose 19 , or as in the case of engineered E. coli cells overexpressing enzymes of the ADC pathway in the study of Yang et al., which obtained a productivity of 8.56 g / L / h and a total product yield of 0.41 g / g L-arginine after 2 hours 22 . It is worth noting that in the latter case a high concentration of the co-factor PLP was used (1 mM vs. 0.1 mM in our system) 22 .

[0395] After centrifugation of the reaction mixture and removal of the co-immobilized [hArg-Cry1Ab / ODC mut -Cry1Ab] biocatalyst, which can be reused in subsequent reaction cycles, the resulting putrescine (about 430 mM) was recovered and directly mixed with adipoyl chloride to synthesize nylon-4,6-polymer. The resulting nylon was characterized by FTIR Figure 18 ). In agreement with previous reports 56,58 , absorption peaks at 3300 cm-1, 2940 cm-1 and 2870 cm-1, 1640 cm-1, respectively attributed to the stretching vibrations of N-H, C-H, C=0, and a peak at 1540 cm-1 corresponding to N-H deformation and C-N stretching (amide II) were observed. These data support the reaction synthesis of nylon-4,6 from adipoyl chloride and putrescine directly produced by the [hArg-Cry1Ab / ODC mut -Cry1Ab] biocatalyst.

[0396] Conclusions

[0397] We have developed a Cry1Ab-based platform and demonstrated its use for the direct co-immobilization of hArg and ODC mut. Notably, this multi-enzyme catalyst exhibits enhanced thermal stability and higher reaction rate than its free enzyme counterpart mixture and individual immobilized particle mixture, and is able to perform recyclable biosynthesis of putrescine. Moreover, the produced reaction solution containing putrescine can be directly used for subsequent reaction with adipoyl chloride to produce nylon-4,6. To the best of our knowledge, this is the first demonstration of biosynthesis of putrescine from L-arginine using a cell-free method, and both productivity and yield are comparable to those by whole-cell synthesis through engineered bacteria. Notably, these Cry1Ab-mediated co-immobilized multi-enzyme particles can be recycled through hArg-Cry1Ab and ODC mut -Cry1Ab in E. coli cells and subsequent isolation of particles by ultrasonication and centrifugation at high yield (about 100 mg co-immobilized enzyme particles / L TB medium) at low cost and easily. More broadly, these results highlight the potential use of Cry1Ab protein particles as a cost-effective and robust platform to directly co-immobilize multiple enzymes involved in multi-step cascade reactions for productive biosynthesis of useful chemicals.

[0398] We demonstrated that ODC mut -Cationization of Cry1Ab with PEI polymers further improved our technology for recycling of the expensive cofactor PLP. We demonstrated a self-sufficient PEI-ODC mut -Cry1Ab-PLP can facilitate biosynthesis of putrescine in multiple cycles without addition of exogenous PLP, where TTN PLP was significantly increased (1736 vs. ODC mut -Cry1Ab particles can be recycled for multiple cycles without loss of activity (71).

[0399] References

[0400] 1. Li, D.; Xiong, Q.; Liang, L.; Duan, H. Multienzyme nanoassemblies: from rational design to biomedical applications. Biomater. Sci. 2021, 9, 7323-7342

[0401] 2. Bugada, L. F.; Smith, M. R.; Wen, F. Engineering spatially organized multienzyme assemblies for complex chemical transformation. ACS Catal. 2018, 8, 7898- 7906

[0402] 3. Adrio, J. L.; Demain, A. L. Microbial enzymes: tools for biotechnological processes. Biomolecules 2014, 4, 117-139

[0403] 4. Sheldon, R. A.; Basso, A.; Brady, D. New frontiers in enzyme immobilisation: robust biocatalysts for a circular bio-based economy. Chem. Soc. Rev. 2021, 50, 5850-5862

[0404] 5. S. ; Carballares, D.; Morellon-Sterlling, R.; Berenguer-Murcia, Alcántara, A. R.; Rodrigues, R. C.; Fernandez-Lafuente, R. Enzyme co- immobilization: Always the biocatalyst designers' choice…or not? Biotechnol. Adv. 2021, 51, 107584

[0405] 6. Santiago-Arcos, J.; Velasco-Lozano, S.; Lopez-Gallego, F. Multienzyme co- immobilization on triheterofunctional supports. Biomacromolecules 2023, 24, 929-942

[0406] 7. Zhang, G.; Quin, M. B.; Schmidt-Dannert, C. Self-assembling protein scaffold system for easy in vitro coimmobilization of biocatalytic cascade enzymes. ACS Catal. 2018, 8, 5611-5620

[0407] 8. Ricca, E.; Brucher, B.; Schrittwieser, J. H. Multi-enzymatic cascade reactions: overview and perspectives. Adv. Synth. Catal. 2011, 353, 2239-2262

[0408] 9. Schneider, J.; Wendisch, V. F. Putrescine production by engineered Corynebacterium glutamicum. Appl. Microbiol. Biotechnol. 2010, 88, 859-868

[0409] 10. Handa, A. K.; Fatima, T.; Mattoo, A. K. Polyamines: bio-molecules with diverse functions in plant and human health and disease. Front. Chem. 2018, 6

[0410] 11. Choi, H.; Kyeong, H.-H.; Choi, J. M.; Kim, H.-S. Rational design of ornithine decarboxylase with high catalytic activity for the production of putrescine. Appl. Microbiol. Biotechnol. 2014, 98, 7483-7490

[0411] 12. Chung, H.; Yang, J. E.; Ha, J. Y.; Chae, T. U.; Shin, J. H.; Gustavsson, M.; Lee, S. Y. Bio-based production of monomers and polymers by metabolically engineered microorganisms. Curr. Opin. Biotechnol. 2015, 36, 73-84

[0412] 13. Li, G. H.; Huang, D. X.; Wang, L.; Deng, Y. Highly efficient whole-cell biosynthesis of putrescine by recombinant Escherichia coli. Biochem. Eng. J. 2021, 166

[0413] 14. Roerdink, E.; Warnier, J. M. M. Preparation and properties of high molecular mass nylon-4,6: a new development in nylon polymers. Polymer 1985, 26, 1582-1588

[0414] 15. Shakiba, M.; Rezvani, G.; Khosravi, F.; Jouybar, S.; Bigham, A.; Zare, M.; Abdouss, M.; Moaref, R.; Ramakrishna, S. Nylon-A material introduction and overview for biomedical applications. Polym. Adv. Technol. 2021, 32, 16

[0415] 16. Qian, Z. G.; Xia, X. X.; Lee, S. Y. Metabolic engineering of Escherichia coli for the production of putrescine: a four carbon diamine. Biotechnol. Bioeng. 2009, 104, 651-662

[0416] 17. Sanders, J.; Scott, E.; Weusthuis, R.; Mooibroek, H. Bio-refinery as the bio-inspired process to bulk chemicals. Macromol. Biosci. 2007, 7, 105-117

[0417] 18. Kong, C.-C.; Wei, X.; Liu, G.-L.; Chi, Z.-M.; Chi, Z. Metabolic engineering of Aureobasidium melanogenum for the overproduction of putrescine by improved L-ornithine biosynthesis. Microbiol. Res. 2022, 260, 127041

[0418] 19. Li, L.; Zou, D.; Ji, A.; He, Y.; Liu, Y.; Deng, Y.; Chen, S.; Wei, X. Multilevel metabolic engineering of Bacillus amyloliquefaciens for production of the platform chemical putrescine from sustainable biomass hydrolysates. ACS Sustain. Chem. Eng. 2020, 8, 2147-2157

[0419] 20. Thongbhubate, K.; Irie, K.; Sakai, Y.; Itoh, A.; Suzuki, H. Improvement of putrescine production through the arginine decarboxylase pathway in Escherichia coli K-12. AMB Express 2021, 11

[0420] 21. Hui, H.; Bai, Y.; Fan, T.; Zheng, X.; Cai, Y. Biosynthesis of putrescine from L-arginine using engineered Escherichia coli whole cells. Catalysts 2020, 10, 947

[0421] 22. Yang, S. C; Ting, W. W.; Ng, I. S. Effective whole cell biotransformation of arginine to a four-carbon diamine putrescine using engineered Escherichia coli. Biochem. Eng. J. 2022, 185

[0422] 23. Yang, F. Y.; Qiu, K. L.; Zhu, Y. C; Zhang, X.; Yang, T. W.; Yi, G. F.; Xu, M. J.; Rao, Z. M. Production of putrescine in metabolic engineering Corynebacterium crenatum by mixed sugar fermentation. ACS Sustain. Chem. Eng. 2022, 10, 14407-14416

[0423] 24. Schneider, J.; Eberhardt, D.; Wendisch, V. F. Improving putrescine production by Corynebacterium glutamicum by fine-tuning ornithine transcarbamoylase activity using a plasmid addiction system. Appl. Microbiol. Biotechnol. 2012, 95, 169-178

[0424] 25. Bolivar, J. M.; Woodley, J. M.; Fernandez-Lafuente, R. Is enzyme immobilization a mature discipline? Some critical considerations to capitalize on the benefits of immobilization. Chem. Soc. Rev. 2022

[0425] 26. Rodrigues, R. C; Berenguer-Murcia, Carballares, D.; Morellon-Sterling, R.; Fernandez-Lafuente, R. Stabilization of enzymes via immobilization: Multipoint covalent attachment and other stabilization strategies. Biotechnol. Adv. 2021, 52, 107821

[0426] 27. Mateo, C.; Palomo, J. M.; Fernandez-Lorente, G.; Guisan, J. M.; Fernandez- Lafuente, R. Improvement of enzyme activity, stability and selectivity via immobilization techniques. Enzyme Microb. Technol. 2007, 40, 1451-1463

[0427] 28. Rodrigues, R. C.; Ortiz, C.; Berenguer-Murcia, Torres, R.; Fernandez-Lafuente, R. Modifying enzyme activity and selectivity by immobilization. Chem. Soc. Rev. 2013, 42, 6290-6307

[0428] 29. Guisan, J. M.; Fernandez-Lorente, G.; Rocha-Martin, J.; Moreno-Gamero, D. Enzyme immobilization strategies for the design of robust and efficient biocatalysts. Curr. Opin. Green Sustain. Chem. 2022, 35, 100593

[0429] 30. Maghraby, Y. R.; El-Shabasy, R. M.; Ibrahim, A. H.; Azzazy, H. M. E.-S. Enzyme immobilization technologies and industrial applications. ACS Omega 2023, 8, 5184-5196

[0430] 31. Olcucu, G.; Klaus, O.; Jaeger, K.-E.; Drepper, T.; Krauss, U. Emerging solutions for in vivo biocatalyst immobilization: tailor-made catalysts for industrial biocatalysis. ACS Sustain. Chem. Eng. 2021, 9, 8919-8945

[0431] 32. Li, J.; Carroll, J.; Ellar, D. J. Crystal structure of insecticidal delta-endotoxin from Bacillus thuringiensis at 2.5 A resolution. Nature 1991, 353, 815-821

[0432] 33. Schnepf, E.; Crickmore, N.; Van Rie, J.; Lereclus, D.; Baum, J.; Feitelson, J.; Zeigler, D. R.; Dean, D. Bacillus thuringiensis and its pesticidal crystal proteins. Microbiol. Mol. Biol. Rev. 1998, 62, 775-806

[0433] 34. Loutfi, H.; Fayad, N.; Pellen, F.; Le Jeune, B.; Chakroun, M.; Benfarhat, D.; Lteif, R.; Kallassy, M.; Le Brun, G.; Abboud, M. Morphological study of Bacillus thuringiensis crystals and spores. Appl. Sci. 2021, 11, 155 ​

[0434] 35.Heater,B.S.;Lee,M.M.;Chan,M.K.Direct production of a genetically-encoded immobilized biodiesel catalyst.Sci.Rep.2018,8,1-10

[0435] 36.Heater,B.S.;Chan,W.S.;Lee,M.M.;Chan,M.K.Directed evolution of agenetically encoded immobilized lipase for the efficient production ofbiodiesel from waste cooking oil.Biotechnol.Biofuels 2019,12,1-14

[0436] 37.Sun,Q.;Heater,B.S.;Li,T.L.;Ye,W.;Guo,Z.;Chan,M.K.Cry3Aa*SpyCatcherfusion crystals produced in bacteria as scaffolds for multienzyme coimmobilization.Bioconjug.Chem.2022,33,386-396

[0437] 38.Nair,M.S.;Lee,M.M.;Bonnegarde-Bernard,A.;Wallace,J.A.;Dean,D.H.;Ostrowski,M.C.;Burry,R.W.;Boyaka,P.N.;Chan,M.K.Cry protein crystals:a novelplatform for protein delivery.PLoS ONE 2015,10,e0127669

[0438] 39.Sun,Q.;Cheng,S.W.;Cheung,K.;Lee,M.M.;Chan,M.K.Cry protein crystal-immobilized metallothioneins for bioremediation of heavy metals fromwater.Crystals 2019,9

[0439] 40. A.; Grande, R.; Sanchez, J.; Tabashnik, B. E.; Bravo, A.; Soberon, M.; Gomez, I. The C-terminal protoxin region of Bacillus thuringiensis CrylAb toxin has a functional role in binding to GPI-anchored receptors in the insect midgut. J. Biol. Chem. 2018, 293, 20263-20272

[0440] 41. Oeda, K.; Inouye, K.; Ibuchi, Y.; Oshie, K.; Shimizu, M.; Nakamura, K.; Nishioka, R.; Takada, Y.; Ohkawa, H. Formation of crystals of the insecticidal proteins of Bacillus thuringiensis subsp. aizawai IPL7 in Escherichia coli. J. Bacteriol. 1989, 171, 3568-3571

[0441] 42. Lee, T. F. Study on Escherichia coli-expressed CrylAb towards the development of a targeted protein delivery system. Master's thesis, The Chinese University of Hong Kong, 2017.

[0442] 43. Kudo, K.; Mochizuki, M.; Kiriyama, S.; Watanabe, M.; Hirami, M. Studies on the structure and properties of nylon 46 fiber. I. Dimensional stability. J. Appl. Polym. Sci. 1994, 52, 861-867

[0443] 44. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 1976, 72, 248-254

[0444] 45. Rahmatullah, M.; Boyde, T. R. C. Improvements in the determination of urea using diacetyl monoxime; methods with and without deproteinisation. Clin. Chim. Acta 1980, 107, 3-9

[0445] 46. Chinard, F. P. Photometric estimation of proline and ornithine. J. Biol. Chem. 1952, 199, 91-95

[0446] 47. Pereira, V.; Pontes, M.; Ferreira, I. M.; Marques, J. C. Simultaneous analysis of free amino acids and biogenic amines in honey and wine samples using in loop orthophthalaldeyde derivatization procedure. J. Chromatogr. A 2008, 1189, 435-443

[0447] 48. Heater, B. S. Cry3Aa-inspired enzyme immobilization: from biocatalysis to therapeutics. Doctoral thesis, The Chinese University of Hong Kong, 2018.

[0448] ​49. Heater, B. S.; Yang, Z.; Lee, M. M.; Chan, M. K. In Vivo Enzyme Entrapment in a Protein Crystal. J. Am. Chem. Soc. 2020, 142, 9879-9883

[0449] 50. Alarcon, R.; Orellana, M. S.; Neira, B.; Uribe, E.; Garcia, J. R.; Carvajal, N. Mutational analysis of substrate recognition by human arginase type I- agmatinase activity of the N130D variant. FEBS J. 2006, 273, 5625-5631

[0450] 51. Mashino, T.; Fridovich, I. Effects of urea and trimethylamine-N-oxide on enzyme activity and stability. Arch. Biochem. Biophys. 1987, 258, 356-360

[0451] 52. Chilaka, F. C.; Nwamba, C. O. Kinetic analysis of urea-inactivation of β- galactosidase in the presence of galactose. J. Enzyme Inhib. Med. Chem. 2008, 23, 7-15

[0452] 53. Grobben, Y.; Uitdehaag, J. C. M.; Willemsen-Seegers, N.; Tabak, W. W. A.; de Man, J.; Buijsman, R. C.; Zaman, G. J. R. Structural insights into human Arginase-1 pH dependence and its inhibition by the small molecule inhibitor CB-1158. J. Struct. Biol.: X 2020, 4

[0453] 54. Stone, E. M.; Glazer, E. S.; Chantranupong, L.; Cherukuri, P.; Breece, R. M.; Tierney, D. L.; Curley, S. A.; Iverson, B. L.; Georgiou, G. Replacing Mn 2+ with Co2+in human arginase I enhances cytotoxicity toward L-arginine auxotrophic cancer cell lines. ACS Chem. Biol. 2010, 5, 333-342

[0454] 55. Basso, A.; Serban, S. Industrial applications of immobilized enzyme - A review. Mol. Catal. 2019, 479, 110607

[0455] 56. Zhang, Z.; Huang, K.; Liu, Z. Synthesis of high molecular weight nylon 46 in supercritical carbon dioxide. Macromolecules 2011, 44, 820-825

[0456] 57. Yamanobe, T.; Kurihara, Y.; Uehara, H.; Komoto, T. Structure and characterization of nylon 46. J. Mol. Struct. 2007, 829, 80-87

[0457] 58. Cong, P.; Xiang, F.; Liu, X.; Li, T. Morphology and microstructure of polyamide 46 wear debris and transfer film: In relation to wear mechanisms. Wear 2008, 265, 1100-1105

[0458] Example 3. Simple strategy for the co-immobilization of NADH-dependent enzymes with NADH for efficient biosynthesis in continuous flow reactors

[0459] The application of NADH-dependent enzymes in continuous flow reactors presents significant challenges due to the high cost of NADH and its inevitable loss during reactor operation. To overcome these limitations, we developed a novel platform comprising polyethylenimine (PEI)-modified Cry3Aa-enzyme fusion particles and demonstrated its application in two catalytic processes. In the first system, two NAD(H)-dependent enzymes, formate dehydrogenase (FDH) and leucine dehydrogenase (LDH), were co-immobilized genetically in Cry3Aa particles, which were then modified with PEI to facilitate NADH co-immobilization. To make the system suitable for continuous flow reactors, the PEI-modified particles were entrapped in agarose beads, loaded with NADH, and then used to catalyze L-tert-leucine biosynthesis under continuous flow conditions for 30 days with minimal NADH re-supplementation, achieving LDH and NADH turnover numbers of 21348 and 6544, respectively. The platform was then validated with another NADH-dependent multi-enzyme system comprising FDH and alcohol dehydrogenase (ADH) and used for the continuous conversion of ethyl acetoacetate to (R)-3-hydroxybutyrate ethyl ester. After 30 days, ADH and NADH turnover numbers of 15074 and 3256 were obtained. These examples illustrate the great potential of this technology to facilitate NADH-dependent enzyme-mediated biosynthesis in continuous flow reactors.

[0460] Methods and results

[0461] NAD(H)-dependent enzymes are promising biocatalysts for the production of pharmaceutical precursors due to their high selectivity and enantiopreference. However, the application of NAD(H)-dependent enzymes is limited due to the high cost of NADH and the necessity of its continuous regeneration and replenishment. Numerous strategies have been implemented to address these issues for continuous flow systems. Common solutions to enable NADH regeneration include the use of coupling enzymes such as formate dehydrogenase (FDH) [1] While a variety of strategies have been explored for NADH retention, including covalent binding to a support [2,3] entrapment in polymeric materials [4,5] adsorption to a solid support [6] and NADH encapsulation in nanostructures [7] .

[0462] Continuous flow reactors offer multiple advantages for industrial synthesis, improving the efficiency and catalytic performance of high-throughput reactions, including excellent control over reaction conditions, ease of scaling up to large-scale reactors, and continuous product removal that minimizes product inhibition [8] However, the implementation of NADH-dependent enzymes in such reactors presents significant challenges, most notably the need to maintain a continuous supply of NADH in the reactor [9]It is logical that one of the simplest strategies to address this issue for continuous flow systems involves co-immobilization of NADH-dependent enzymes with NADH on solid supports coated with cationic polymers such as polyethylenimine (PEI) [10,11] This strategy relies on electrostatic interactions to bind the cofactor without altering its chemical structure, thus preserving its native functionality. The free mobility of the cofactor enables seamless shuttling between the active site of the enzyme and the immobilized support, thus preserving the functionality of the enzyme while ensuring continuous enzyme activity [12,13] Furthermore, if the NADH cofactor is lost or inactivated, it can be replenished. However, previously reported systems exhibit several limitations, such as low NADH binding capacity, low NADH total turnover number (TTN NADH ), unproven NADH rechargability, and the need for labor-intensive enzyme purification prior to cofactor co-immobilization. For example, in the asymmetric reduction of acetophenone, co-immobilization of NAD+, ethanol dehydrogenase (ADH), and formate dehydrogenase (FDH) onto PEI-coated agarose beads resulted in a low TTN (TTN NADH : 85) for 107 hours of operation

[14] Other systems, such as the one for the asymmetric reduction of trifluoroacetaldehyde, exhibit a better TTN (TTN NADH : 1076) for NADH, but the operation time is still only 120 hours

[11] These limitations demonstrate the need and opportunity for new technologies to facilitate NADH-dependent enzyme reactions in continuous flow systems.

[0463] Previously, we reported the preparation of in vivo immobilized Cry3Aa-enzyme fusion beads for the efficient and multi-cycle biosynthesis of L-tert-leucine using two different Cry3Aa particles in batch mode

[15] Here, one particle contains two co-immobilized NADH-dependent enzymes, the Cry3Aa-leucine dehydrogenase (3A-LDH) component that drives L-tert-leucine (L-tert Leu) biosynthesis and the Cry3Aa-formate dehydrogenase (3A-FDH) component that facilitates NADH regeneration. The second particle consists of PEI-modified Cry3Aa (PEI-3A) crystals that can be used as a mechanical responsive NADH sponge for NADH recycling. Under high agitation, NADH is released from the PEI-3A crystals where it can mix with enzyme crystals and reactants to facilitate L-tert-leucine biosynthesis. Between cycles, agitation is stopped and NADH binds to the insoluble PEI-3A crystals, allowing for the exchange of the reactant solution without loss of NADH.

[0464] While this strategy is ideal for a batch reaction, it is not efficient for a continuous flow reactor due to the migration of NADH between the particles, which can lead to the cofactor passing by the product. Therefore, we sought a way to adapt it to a continuous flow reaction. We hypothesized that a simple way to achieve our goal could be to modify the co-immobilized 3A-LDH / 3A-FDH particles with PEI at a level that would enable co-immobilization of NADH without loss of enzyme activity. Our hypothesis was that this dual enzyme and NADH co-immobilization would allow the cofactor to migrate between the enzymes in the particle microenvironment without being released into the aqueous medium.

[0465] To test this hypothesis, co-immobilized 3A-FDH / 3A-LDH particles were prepared in vivo as previously reported

[15] and then modified with PEI. At low concentrations of EDC / NHS or PEI, the activity of the 3A-FDH / 3A-LDH particles remained almost unchanged. After PEI modification, the particles retained their ovoid shape and were only slightly larger in size - probably due to the labeling of some PEI on the surface of the particles Figure 19 A-E), the zeta potential shifted from -15 mV to +40 mV Figure 19 F), strongly supporting its PEI-mediated cationization.

[0466] Confocal microscopy was used to monitor the attachment of dye-modified PEI to the 3A-FDH / 3A-LDH particles, while the co-immobilization of NADH to these particles was confirmed by taking advantage of the intrinsic fluorescence of NADH Figure 20 A). The NADH binding capacity of the PEI-3A-FDH / 3A-LDH particles was 226 ± 5 μmol NADH / g particles as spectrometrically determined by measuring the absorbance of NADH at 340 nm, while the particles without PEI retained only about 4.2 ± 0.3 μmol NADH / g.

[0467] The effect of PEI modification on the conversion efficiency of 3A-FDH / 3A-LDH for L-tert-leucine production was determined and showed that the modification resulted in a small decrease of about 10% in the product produced Figure 20 B). To gain further insight, the kinetic parameters of the 3A-FDH / 3A-LDH and PEI-3A-FDH / 3A-LDH particles were determined and showed that the main effect of PEI modification was a slight increase in the K m values of FDH and LDH, thus slightly reducing the catalytic efficiency. Overall, these data showed that PEI modification had only a very small effect on the function of Cry3Aa immobilized enzymes.

[0468] To evaluate the ability of PEI-3A-FDH / 3A-LDH particles to retain NADH co-factor for catalysis, we tested their application in the production of L-tert-leucine in a multi-cycle batch reaction under slow shaking (80 rpm). PEI-3A-FDH / 3A-LDH-NADH particles were able to promote complete conversion of substrate to product in 10 cycles without the addition of exogenous NADH, while unmodified 3A-FDH / 3A-LDH-NADH particles showed a significant (about 80%) decrease in product yield after the first cycle Figure 20 C). These data demonstrate that PEI modification significantly enhances NADH retention in 3A-FDH / 3A-LDH particles.

[0469] The next step was to optimize the application of PEI-3A-FDH / 3A-LDH-NADH particles in a continuous flow system. Given the small size of the co-immobilized enzyme particles, two concerns were their retention in the column under flow conditions and their potential to cause clogging. To overcome these problems, we embedded PEI-modified particles in 0.5% (w / v) agarose beads produced according to the reported method

[16] to generate pores smaller than 200 nm (which would thus be suitable for embedding micrometer-sized PEI-3A-FDH / 3A-LDH-NADH particles while allowing moderate access to substrates and products) [17-19] The produced beads were spherical, 20 mm in diameter (wet) (about 2 mm in diameter (air-dried)), with about 7.1 ± 0.2 mg PEI-3A-FDH / 3A-LDH particles per g of wet beads. Confocal microscopy of PEI-3A-FDH / 3A-LDH-Alexa 568-agarose beads showed a uniform distribution of particles within the agarose beads Figure 21 ).

[0470] To confirm that the NADH-binding ability of the particles remained intact after embedding into agarose beads, we checked for the presence of NADH molecules in the beads by confocal microscopy using the auto-fluorescence of the co-factor. Images showed co-localization of NADH with PEI-3A-FDH / 3A-LDH particles within agarose beads Figure 21 ), indicating the effectiveness of the system to co-embed co-factor and enzyme particles

[14] .

[0471] To assess the impact of entrapment in the particles of the agarose beads on the activity of the co-immobilized enzymes, the production efficiency of the particles for L-tertiary leucine production was evaluated before and after bead formation. These studies showed that entrapment in agarose under the conditions used resulted in a significant decrease in the production of L-tertiary leucine to 23% of the original PEI-3A-FDH / 3A-LDH particles. Considering that crushing the agarose beads into fragments doubled the amount of L-tertiary leucine produced compared to the intact beads, it was postulated that this change was mostly due to the substrate diffusion limitations imposed by the beads. These findings are in agreement with previous studies that have found that immobilization of enzymes in porous materials often results in a decrease in activity due to slower substrate and product diffusion rates

[20] .

[0472] The suitability of the PEI-3A-FDH / 3A-LDH agarose beads for continuous flow reactions was demonstrated by placing the beads in a column and testing their performance under typical flow conditions. The retention of the PEI-3A-FDH / 3A-LDH particles within the beads was verified by demonstrating that there was no detectable protein in the outlet solution at high flow rates (1 mL / min) and no loss of catalytic activity after 60 minutes. Furthermore, highlighting their potential as a robust immobilization platform, the beads maintained their structural integrity and 86% of their catalytic activity for L-tertiary leucine production even after three months of storage in ddH2O at 4°C.

[0473] After characterizing the beads, we then evaluated their utility for continuous flow biosynthesis. Two pmol of NADH were supplied to the agarose beads containing 220 mg of PEI-3A-FDH / 3A-LDH particles placed in a column with an effective reactor volume of 150 mL, and then the agarose beads were used to facilitate the complete conversion of trimethylpyruvic acid (TMP) to L-tertiary leucine over a period of 5 days without the addition of exogenous NADH Figure 22 ). To maintain high conversion rates for longer periods of time, the column was re-supplied with 0.2 pmol of the co-factor every 5 days. After 30 days, the TTN of NADH and LDH in the co-immobilized particles were 6544 and 21348, respectively, with a productivity of 0.066 mmol / h at room temperature.

[0474] To test the versatility of this approach for other enzyme processes, we utilized our Cry3Aa-mediated immobilization strategy to co-immobilize FDH

[21] and B. bombi alcohol dehydrogenase (ADH)

[22] —two enzymes that can act together to facilitate the conversion of ethyl acetoacetate (EAA) to (R) 3-hydroxybutyric acid ethyl ester (R-EHB), a drug precursor used for the synthesis of antibiotics, anti-inflammatory agents, and antiviral drugs

[23] Given the slower reaction rate of FDH compared to ADH, two copies of 3A-FDH were used to increase its relative concentration in the particle. The resulting co-immobilized (3A-FDH)2 / 3A-ADH particles were then modified with PEI, embedded in agarose beads, and evaluated for continuous flow conversion of 5 mM EAA to R-EHB at a flow rate of 0.1 mL / min and room temperature. Agarose beads containing 180 mg of PEI-(3A-FDH)2 / 3A-ADH particles pre-incubated with 2 pmol of NADH were loaded onto a continuous flow column and could be used to facilitate the biosynthesis of R-EHB for 5 days without the addition of exogenous NADH. The conversion rate was 100% for the first 4 days but dropped on the fifth day, presumably due to leaching of NADH - full activity could be restored by re-supplementing the beads with 0.5 pmol of NADH. After 30 days, the TTN of the NADH and ADH components in the (3A-FDH)2 / 3A-ADH crystals were recorded as 3256 and 15074, respectively, with a productivity of 0.023 mmol / h.

[0475] The PEI-3A-FDH / 3A-LDH and PEI(3A-FDH)2 / 3A-ADH systems described herein illustrate the simplicity of our PEI-modified Cry3Aa-enzyme / NADH co-immobilization technology, which we believe is a significant advance over existing immobilization methods. While covalent attachment of cofactors provides robust cofactor immobilization and resistance to leaching, they often require complex modifications and support designs

[24] For example, NADH was covalently attached to a flexible arm + The study achieved a high TTN (turnover number) of 10839, but the reaction time was limited to 24 hours

[25] This requires frequent preparation of biocatalysts to extend usage, increasing production costs. Encapsulation methods, such as Pickering emulsion droplets, show promise for cofactor recycling, but are limited to specific reactions. This limitation arises because both substrate and product must pass through the oil phase, resulting in very low flow rates due to the fragility of the droplets. In addition, there is lower control over reaction conditions and substrate retention issues, which limit productivity

[26] Although hydrogel encapsulation [27,28] and other ion adsorption methods [11,14] have been reported, they face challenges stemming from enzyme and cofactor leaching that limit their useful lifetime.

[0476] In this study, we present a system designed for long-term cofactor retention and rechargability, supporting continuous biosynthesis for up to 30 days. Despite some cofactor leaching, the stability of operation of this system makes it a robust platform for continuous biocatalysis. The system retains NADH for up to 5 days without the need for exogenous addition and is able to recover activity through recharging. This method provides superior system longevity (30 days) compared to other methods. The simple method for preparation of biocatalysts for continuous flow reactors facilitates their scalability by simplifying implementation, reducing operational complexity, and ensuring consistent biocatalyst performance over extended use. These features make it a useful system for a variety of NADH-dependent enzyme systems as well as other cofactor-dependent enzyme systems.

[0477] References

[0478] [1] B. Binay, D. D. Yildirim, A. S. S. Tükel, Highly Stable and Reusable Immobilized Formate Dehydrogenases: Promising Biocatalysts for In Situ Regeneration of NADH, Beilstein J. Org. Chem. 2016, 12, 271-277.

[0479] [2] S. Kim, K. Kwon, G. Tae, I. Kwon, Nano-Entrapping Multiple Oxidoreductases and Cofactor for All-In-One Nanoreactors, ACS Sustain. Chem. Eng. 2021, 9, 6741-6747.

[0480] [3] X. Ji, Z. Su, P. Wang, G. Ma, S. Zhang, Integration of Artificial Photosynthesis System for Enhanced Electronic Energy-Transfer Efficacy: A Case Study for Solar-Energy Driven Bioconversion of Carbon Dioxide to Methanol, Small. 2016, 12, 4753-4762.

[0481] [4] J. Grunwald, B. Wirz, M. P. Scollar, A. M. Klibanov, Asymmetric Oxidoreductions Catalyzed by Alcohol Dehydrogenase in Organic Solvents, J. Am. Chem. Soc. 1986, 108, 6732-6734.

[0482] [5] J. Wykes, P. Dunnill, M. J. B. Lilly, Bioengineering, Cofactor Recycling in an Enzyme Reactor: A Comparison Using Free and Immobilized Dehydrogenases with Free and Immobilized NAD, Bioeng. 1975, 17, 51-68.

[0483] [6] C.-H. Wong, Nicotinamide Cofactor-Requiring Enzymatic Synthesis in Organic Solvent-Water Biphasic Systems, J. Org. Chem. 1987, 29, 197-208.

[0484] [7] P. Wang, G. Ma, L. Liao, F. Gao, Construction of Multienzyme Bioactive systems, J. Biotechnol. 2005, 22-41.

[0485] [8] S. Wu, R. Snajdrova, J. C. Moore, K. Baldenius, U. T. Bornscheuer, Bicatalysis: Enzymatic Synthesis for Industrial Applications, Catalysts. 2021, 60, 88-119.

[0486] [9] S. Mordhorst, J. N. Andexer, Round, Round We Go - Strategies for Enzymatic Cofactor Regeneration, Nat. Prod. Rep. 2020, 37, 1316-1333.

[0487]

[10] T. Montes, V. Grazu, I. Manso, B. Galan, F. Lopez-Gallego, R. Gonzalez, J. A. Hermoso, J. L. Garcia, J. M. Guisan, R. Fernandez-Lafuente, Improved Stabilization of Genetically Modified Penicillin G Acylase in the Presence of Organic Cosolvents by Co-Immobilization of the Enzyme with Polyethyleneimine, Adv. Synth. Catal. 2007, 349, 459-464.

[0488]

[11] A. I. Benitez-Mateos, E. San Sebastian, N. Rios-Lombardia, F. Moris, J. Gonzalez-Sabin, F. Lopez-Gallego, Asymmetric Reduction of Prochiral Ketones by Using Self-Sufficient Heterogeneous Biocatalysts Based on NADPH-Dependent Ketoreductases, Chem. Eur. J. 2017, 23, 16843-16852.

[0489]

[12] E. S. da Silva, V. Gomez-Vallejo, J. Llop, F. Lopez-Gallego, Efficient nitrogen-13 radiochemistry catalyzed by a highly stable immobilized biocatalyst, Catal. Sci. Technol. 2015, 5, 2705-2713.

[0490]

[13] A. I. Benitez-Mateos, C. Huber, B. Nidetzky, J. M. Bolivar, F. Lopez-Gallego, Design of the Enzyme-Carrier Interface to Overcome the O2 and NADH Mass Transfer Limitations of an Immobilized Flavin Oxidase, ACS Appl. Mater. Interfaces. 2020, 12, 56027-56038.

[0491]

[14] S. Velasco-Lozano, A. I. Benitez-Mateos, F. Lopez-Gallego, Co-immobilized Phosphorylated Cofactors and Enzymes as Self-Sufficient Heterogeneous Biocatalysts for Chemical Processes, Angew. Chem. Int. Ed. Engl. 2017, 56, 771-775.

[0492]

[15] R. Yekta, X. Xiong, J. Li, B. S. Heater, M. M. Lee, M. K. Chan, Mechanoresponsive Protein Crystals for NADH Recycling in Multicycle Enzyme Reactions, J. Am. Chem. Soc. 2024, 146, 18817-18822.

[0493]

[16] A. Alehosseini, E.-M. Gomez del Pulgar, M. J. Fabra, L. G. Gomez-Mascaraque, A. Benitez-Paez, M. Sarabi-Jamab, B. Ghorani, A. Lopez-Rubio, Agarose-Based Freeze-Dried Capsules Prepared By The Oil-Induced Biphasic Hydrogel Particle Formation Approach For The Protection Of Sensitive Probiotic Bacteria, Food. Hydrocoll. 2019, 87, 487-496.

[0494]

[17] N. Pernodet, M. Maaloum, B. Tinland, Pore Size of Agarose Gels by Atomic Force Microscopy, Electrophoresis 1997, 18, 55-58.

[0495]

[18] N. C. Stellwagen, Electrophoresis of DNA in Agarose Gels, Polyacrylamide Gels and in Free Solution, Electrophoresis 2009, 30, 188-195.

[0496]

[19] J. Narayanan, J.-Y. Xiong, X. Y. Liu, Determination of Agarose Gel Pore Size: Absorbance Measurements vis a vis Other Techniques, J. Phys. Conf. Ser. 2006, 1, 120-128.

[0497]

[20] L. Bayne, R. V. Ulijn, P. J. Halling, Effect of Pore Size on the Performance of Immobilised Enzymes, Chem. Soc. Rev. 2013, 42, 9000-9010.

[0498]

[21] H. Choe, J. C. Joo, D. H. Cho, M. H. Kim, S. H. Lee, K. D. Jung, Y. H. Kim, Efficient CO2-Reducing Activity of NAD-Dependent Formate Dehydrogenase from Thiobacillus sp. KNK65MA for Formate Production from CO2 Gas, PLoS One. 2014, 9, 103-111.

[0499]

[22] Y. H. Zhu, C. Y. Liu, S. Cai, L. B. Guo, I. W. Kim, V. C. Kalia, J. K. Lee, Y. W. Zhang, Cloning, Expression and Characterization of a Highly Active Alcohol Dehydrogenase for Production of Ethyl(S)-4-Chloro-3-Hydroxybutyrate, Indian. J. Microbiol. 2019, 59, 225-233.

[0500]

[23] J.-Y. He, L.-M. Zhou, P. Wang, L. Zu, Microbial Reduction of Ethyl Acetoacetate to Ethyl(R)-3-Hydroxybutyrate in an Ionic Liquid Containing System, Process. Biochem. 2009, 44, 316-321.

[0501]

[24] E. Diamanti, S. Velasco-Lozano, D. Grajales-Hernandez, A. H. Orrego, D. DiSilvio, J. M. Fraile, F. Lopez-Gallego, Self-Sufficient Heterogeneous Biocatalyst through Boronic Acid-Diol Complexation of Adenylated Cofactors, ACS Sustain. Chem. Eng. 2023, 11, 14409-14421.

[0502]

[25] C. J. Hartley, C. C. Williams, J. A. Scoble, Q. I. Churches, A. North, N. G. French, T. Nebl, G. Coia, A. C. Warden, G. Simpson, A. R. Frazer, C. N. Jensen, N. J. Turner, C. Scott, Engineered Enzymes that Retain and Regenerate their Cofactors Enable Continuous-Flow Biocatalysis, Nat. Catal. 2019, 2, 1006-1015.

[0503]

[26] W. Wei, R. Ettelaie, X. Zhang, M. Fan, Y. Dong, Z. Li, H. Yang, Co-compartmentalization of Enzymes and Cofactors within Pickering Emulsion Droplets for Continuous-Flow Catalysis, Angew. Chem. Int. Ed. Engl. 2022, 61, e202211912.

[0504]

[27] T. Peschke, P. Bitterwolf, S. Gallus, Y. Hu, C. Oelschlaeger, N. Willenbacher, K. S. Rabe, C. M. Niemeyer, Self-Assembling All-Enzyme Hydrogels for Flow Biocatalysis, Angew. Chem., Int. Ed. Engl. 2018, 57, 17028-17032.

[0505]

[28] Q. Chen, Y. Wang, G. Luo, Recycling of Cofactors in Crude Enzyme Hydrogels as Co-immobilized Heterogeneous Biocatalysts for Continuous-Flow Asymmetric Reduction of Ketones, ChemSusChem. 2023, 16, e202201654.

[0506] Example 4: Self-assembled protein system for scalable and sustainable cadaverine biosynthesis

[0507] The growing demand for sustainable and cost-effective biocatalysts for bio-based chemicals (as alternatives to petroleum-based feedstocks) production drives the need for innovative approaches. This study shows the development of a novel biocatalytic platform for cadaverine production using CrylAb fusion technology-mediated enzyme self-assembly protein system. The resulting fusion particles exhibit high catalytic efficiency for cadaverine synthesis comparable to that of free enzymes, while eliminating the need for complex purification and immobilization steps that can be recycled. To eliminate the dependency on exogenous PLP cofactor, the fusion particles were modified with polyethyleneimine (PEI) enabling stable co-immobilization of PLP. These modified particles were subsequently embedded in agarose and integrated into a continuous flow biosynthesis system, achieving high production efficiency. The system maintained robust cadaverine synthesis for 19 days with minimal PLP supplementation, showing excellent stability and catalytic longevity. The developed platform achieved turnover numbers (TTN) of 16938 and 190609 for PLP and CadA, respectively. This innovative approach simplifies enzyme utilization, improves operational stability, and paves the way for scalable and cost-effective bio-based chemical production, highlighting its potential for large-scale biocatalytic applications.

[0508] INTRODUCTION

[0509] The production of plastics and a wide range of chemicals primarily relies on petroleum-based feedstocks as the main source, which amounts to approximately 5 million tons per year. This significant dependence on fossil resources raises concerns about resource depletion and environmental pollution [1]. In response to this issue, the industrial sector is shifting towards environmentally friendly and renewable bio-based resources. It is predicted that more than 75% of petroleum-based materials will be replaced by bio-based materials by 2035, highlighting their significant popularity and economic benefits [1].

[0510] Cadaverine, also known as 1,5-pentanediamine, is a key precursor for the production of bio-based polymers in various industries, particularly in biochemistry and materials engineering [2]. Nylon 5X polymers synthesized from cadaverine exhibit high-performance polyamides with perfect thermal and mechanical properties. For example, bio-nylon 5,6 derived from cadaverine shows excellent mechanical strength, flame retardancy, moisture absorption, and air permeability, making it a sustainable and advantageous choice for plastic and fiber engineering [3] . Another important application of cadaverine is in the synthesis of polyurethanes, a versatile polymer widely used in various industries due to its durability, flexibility, and resistance to chemicals and abrasion. This polymer is traditionally synthesized from petroleum-derived hexamethylene diisocyanate as the preferred precursor [4,5]However, biobased polyurethanes synthesized from cadaverine offer a sustainable and environmentally friendly alternative, characterized by lower resource consumption, high reactivity, and enhanced weather resistance [5 ,6] In addition to polymer production, cadaverine has important applications in medicine, with potential efficacy in treating arrhythmia, hypoglycemia, and dysentery [1 ,7 ,8] In agriculture, it functions as a valuable biostimulant, increasing crop yield, and contributing to modern and sustainable agricultural strategies [9,10] These diverse properties and applications make cadaverine a promising green alternative for the manufacture of a wide range of valuable polymers and chemicals, highlighting the importance of developing sustainable and environmentally friendly large-scale synthesis methods for cadaverine to meet global demand.

[0511] There are two main strategies for the synthesis of cadaverine from L-lysine: (i) chemical methods and (ii) biological biosynthesis. Chemical catalysis offers high conversion rates, simple separation by centrifugation, and high reusability through the use of centrifugation; however, due to the high reactivity of L-lysine leading to side reactions, its productivity (2-5 g L -1 h -1 ) and selectivity (30-60%) are low [1,11] The use of strong acids in this method often leads to corrosion problems, while the instability of cadaverine in acidic and metal-rich environments further exacerbates selectivity challenges [1] On the other hand, the enzymatic biosynthesis of cadaverine using lysine decarboxylase (CadA) is highly selective, with 100% conversion rates, and shows high productivity (e.g., 204 g L -1 h -1 ) [12,13] This method has significant advantages, including the direct synthesis of cadaverine under mild conditions such as room temperature, resulting in lower energy consumption and production costs [2] The industrial application of enzymatic biosynthesis of cadaverine faces several challenges, including enzyme instability, reduced activity due to the toxicity of cadaverine, and competition for food resources in whole-cell systems. One major problem is the increase in pH during decarboxylation, which leads to enzyme instability and limits recycling and overall effectiveness [14,15] Another significant challenge is the dependence of CadA on 5'-phosphor-pyridoxal (PLP) as a co-factor. Continuous supply of PLP is required, which poses economic and feasibility problems and complexity of downstream purification [2,9]The purification of free CadA is also expensive, labor-intensive, and time-consuming, further adding economic barriers. Therefore, it is crucial to develop techniques that eliminate the complex purification process of cell-free biocatalysts and explore strategies for co-immobilization near the enzyme and immobilization of PLP to enhance its stability and reusability.

[0512] CrylAb is a crystal-forming protein from the Cry protein family produced by Bacillus thuringiensis (Bt) that kills insects by forming pores in the midgut cells of insects

[16] Recently, our group successfully used this protein as an immobilization platform for the in vivo direct co-immobilization of E. coli ornithine decarboxylase and human arginase

[17] This approach eliminates the need for conventional enzyme purification, simplifies the process of cell lysis, and the fusion particles can be harvested after lysis. Direct enzyme immobilization in intracellular CrylAb particles also eliminates the need for post-immobilization steps such as attaching the enzyme to a support or carrier for industrial recyclability. CrylAb-mediated immobilization has been shown to significantly improve the thermal stability of enzymes, enabling sustainable and efficient biocatalytic reactions

[17] .

[0513] In this study, we used CrylAb-mediated in vivo immobilization technology to directly immobilize CadA inside the cell. To eliminate the continuous need for PLP supplementation, CadA-CrylAb fusion particles were chemically modified with a highly cationic polymer, polyethyleneimine (PEI), to facilitate the co-immobilization of the negatively charged PLP cofactor through ionic interactions. The biological system was further expanded to a packed-bed reactor, in which the particles were embedded within agarose beads, enabling the continuous and efficient production of cadaverine in a continuous flow mode.

[0514] Materials and Methods

[0515] 5'-Phosphorylriboflavin (PLP), L-lysine hydrochloride, cadaverine dihydrochloride, o-Phthaldehyde (OPA), and MES (2-(N-morpholino)ethanesulfonic acid) were supplied by TCI Corporation. Certified molecular biology agarose from Bio-Rad was provided. Polyethyleneimine (PEI) (P3143) in 50% (w / v) in H20 and all other buffer salts were purchased from Sigma-Aldrich.

[0516] Construction of expression plasmids

[0517] To construct the expression plasmid for CadA (Accession number: NP_418555), the cadA gene was amplified by PCR and inserted into the pET28 vector by Gibson assembly (NEB). The resulting recombinant plasmid was transformed into E. coli XL10. All positive colonies were verified by DNA sequencing (BGI).

[0518] The plasmid for expressing the CadA-CrylAb fusion was prepared by inserting the amplified cadA gene into the pET28-CrylAb vector via Gibson assembly (NEB). The resulting recombinant plasmid was transformed into E. coli XL10. All positive colonies were verified by DNA sequencing (BGI).

[0519] Expression and purification of CadA free enzyme

[0520] To express free CadA, the corresponding plasmid was transformed into E. coli BL21(DE3) by heat shock. The resulting colonies were cultured at 37°C in 500 mL Lennox Broth (LB, IBI Scientific) containing 50 pg / mL kanamycin until OD 600 0.6-0.8. Then, 0.5 mM isopropyl β-D-l-thiogalactopyranoside (IPTG, IBIScientific) was added to induce soluble enzyme expression at 18°C for 20 h. Cells were harvested by centrifugation and the pellet was resuspended in buffer A (10 mM HEPES, 300 mM NaCl, pH 7.5) supplemented with 1 mM benzamidine hydrochloride (TCI Chemicals) and 1 mM phenylmethylsulfonyl fluoride (PMSF, Cayman Chemical) and lysed by sonication. The cell lysate was clarified by centrifugation, the resulting supernatant was filtered and loaded onto a nickel resin column (BioRad) for affinity purification. The target protein was eluted using 250 mM imidazole and the purity of the eluted fractions was evaluated by SDS-PAGE. Subsequently, the eluted high purity fractions were dialyzed twice against 2 L of buffer A using SnakeSkin dialysis tubing (MWCO: 10 kDa; Thermo Scientific). The dialyzed protein was concentrated using centrifugal filters (MWCO: 30 kDa; Merck Millipore) and then stored at -80°C after freezing with liquid nitrogen.

[0521] Expression and purification of CadA-CrylAb fusion particles

[0522] Expression of CadA-CrylAb fusion protein was achieved by transforming the encoding plasmid into E. coli BL21(DE3) by heat shock and the resulting colonies were inoculated into 500 mL Terrific Broth (TB, IBI Scientific) supplemented with 4 mL glycerol and 50 pg / mL kanamycin. Cells were then grown at 37°C until an OD 600 of 0.6-0.8, at which point 2 mM IPTG was added and the cells were further incubated at 30°C for 48 hours. The cell pellet was resuspended in lysis buffer (50 mM Tris, 50 mM EDTA, 15% sucrose, pH 8.0) and incubated with 1 mg / mL lysozyme overnight. The lysed cell pellet was washed twice with ddH2O and then resuspended in Crystal Wash I solution (0.5 M NaCl, 2% Triton X-100) followed by sonication and centrifugation. The collected pellet was washed twice with Crystal Wash I solution, twice with 0.5 M NaCl, once with 0.25 M NaCl, and three times with ddH2O. Expression and purity of the target fusion protein pellet was confirmed by SDS-PAGE.

[0523] All protein concentrations were obtained using the Bradford protein assay (BioRad). To quantify the amount of CadA in the CadA-CrylAb fusion pellet, the total protein concentration (mg / mL) of the pellet was converted to its molarity by dividing by the combined molecular weight of CadA (81 kDa) and CrylAb (130 kDa). As a fusion pellet, the molarity of the enzyme component is equivalent to the molarity of the total pellet. Conversely, the mass volume concentration (mg / mL) of the CadA component can be calculated by multiplying its molarity by its molecular weight (81 kDa).

[0524] Size Exclusion Chromatography (SEC)

[0525] To characterize the oligomerization of CadA, 1 mg / mL protein in running buffer (10 mM HEPES, 300 mM NaCl, pH 7.5) was injected into a 30 pL injection loop and run at 0.1 mL / min. A mixture of SEC standards (thyroglobulin: 670 kDa; gamma-globulin: 158 kDa; ovalbumin: 44 kDa; myoglobin: 17 kDa; vitamin B12: 1.35 kDa) was diluted two-fold in running buffer and then injected into the injection loop and run at 0.1 mL / min.

[0526] Orbitrap Mass Spectrometry

[0527] To confirm the identity of the major bands from SDS-PAGE of free CadA and CadA-CrylAb particles, the bands were excised from the Coomassie blue-stained gels and fragmented into small pieces. The gel pieces were destained, dehydrated, and subsequently digested with trypsin. The processed samples were analyzed by an Orbitrap Fusion Lumos Tribrid Mass Spectrometer (Thermo Fisher Scientific).

[0528] Assay for quantification of lysine consumption for CadA activity

[0529] The activity of CadA was determined based on the conversion of L-lysine following the ninhydrin method of Chinard et al. Reaction solutions containing 1 mM of CadA-CrylAb fusion particles (0.211 mg / mL) or CadA enzyme (0.081 mg / mL), 0.1 mM PLP, and 50 mM L-lysine HC1 were incubated at 37°C with 1000 rpm shaking for 6 min. The reaction was terminated by the addition of 2 M HC104. The remaining L-lysine was determined by further reacting the reaction solution with an equal volume of 20 mg / mL ninhydrin dissolved in 2-methoxyethanol at 90°C with 1000 rpm shaking for 30 min. The converted L-lysine was obtained by subtracting the remaining L-lysine from the input amount of L-lysine. Assuming a 1 : 1 stoichiometry of the reaction of L-lysine conversion to cadaverine, the amount of cadaverine produced is equal to the amount of converted L-lysine. Bradford protein assay was used to determine all protein concentrations, and the quantification of CadA in CadA-CrylAb fusion particles was performed as previously described. All experiments were performed in triplicate.

[0530] Thermal stability

[0531] Two hundred μL of 2 μM CadA enzyme or CadA-CrylAb particles prepared in 0.1 mM PLP and 50 mM HEPES (pH 8.0) were incubated at different temperatures (25°C, 30°C, 37°C, 45°C, 50°C, 60°C, 70°C) with 1000 rpm shaking for 1 h. After this time, their residual activity was determined as previously described. The particles were sonicated for 10 seconds prior to the activity measurement. The residual activity at 25°C was considered 100%. The data were fitted to a sigmoidal curve and T 50 values were calculated using GraphPad Prism software. T 50 defined as the temperature that retains 50% activity after 1 hour of incubation. Protein concentration was measured using Bradford protein assay. All experiments were performed in triplicate.

[0532] Optimization of reaction pH

[0533] Fifty μL of 1 μΜ CadA enzyme or 2 μΜ CadA-CrylAb particles were incubated with an equal volume of 100 mM L-lysine HCI in different reaction buffers (pH 4-5: 0.1 M citric acid / sodium citrate; pH 6: 0.1 M MES; pH 7-8: 0.1 M HEPES; pH 9: 0.1 M CAPSO) containing 0.1 mM PLP at 37°C with 1000 rpm shaking for 6 min. Transformed L-lysine was quantified as described before, assuming 100% transformation of L-lysine at pH 6. All experiments were performed in triplicate.

[0534] Lifetime of CadA-CrylAb particles

[0535] Five hundred μL of 2 μΜ CadA-CrylAb were incubated at pH 6 (0.1 M MES, 0.1 mM PLP) and pH 7 (0.1 M HEPES, 0.1 mM PLP) and 25°C. Residual activity was measured at day 0, day 1, day 5 and day 15 as described before. The initial activity at day 0 was considered 100%. All experiments were performed in triplicate.

[0536] Recyclability of CadA-CrylAb particles for L-lysine transformation into cadaverine

[0537] Recyclability of co-immobilized particles was evaluated by resuspending 45 mg of CadA-CrylAb with 10 mL of reaction buffer (1 M MES, 0.1 mM PLP, pH 6.0) containing 1 M L-lysine HCI and incubating in a 50 mL centrifuge tube at 37°C with 250 rpm shaking for 1 h. At the end of the incubation, the reaction solution was centrifuged at 8000 rpm for 5 min and the remaining L-lysine in the supernatant was determined as described before. The pellet containing PEI-CadA-CrylAb-PLP particles was resuspended with 10 mL of reaction buffer containing fresh 1 M L-lysine HCI to start the next reaction cycle. All experiments were performed in triplicate.

[0538] PEI modification of CadA-CrylAb fusion particles

[0539] To prepare PEI-modified CadA-CrylAb fusion particles, 1 mg / mL of particles were incubated with PEI solution (pH 7.0) at room temperature for 2 hours under gentle rotation. Subsequently, EDC / NHS crosslinker (10 mM, molar ratio 1 :2) was gradually added to the mixture and allowed to react at 25 °C for 2 hours under continuous rotation. To quench the reaction, 200 mM Tris-HCl (pH 7.0) was added at a 1 : 1 (v / v) ratio, followed by an additional 30 minutes of rotation at 25 °C. The samples were then centrifuged at 8000 rpm for 10 minutes and the precipitated particles were washed with 100 mM NaCl and ddH2O multiple times to remove any unbound PEI molecules. The amount of PEI bound to the particles was quantified using the copper sulfate assay [19,20] .

[0540] Effect of PEI modification on CadA-CrylAb particle activity

[0541] To evaluate the effect of PEI modification on CadA-CrylAb activity, the conversion efficiency of the particles before and after PEI modification was compared by incubating a reaction solution containing 1 mM CadA-CrylAb or PEI-CadA-CrylAb, 0.1 mM PLP and 50 mM L-lysine HC1 in 0.1 M MES (pH 6) at 37 °C for 10 minutes with gentle shaking at 100 rpm. The consumption of L-lysine HC1 was quantified as described above, assuming 100% consumption of L-lysine HC1 for CadA-CrylAb. All experiments were performed in triplicate.

[0542] Dynamic light scattering (DLS)

[0543] To examine the zeta potential and hydrodynamic diameter of CadA-CrylAb fusion particles before and after PEI binding, a Malvern Zetasizer Nano ZS90 (Malvern Instruments, UK) was used. Particles (100 pg / mL) were prepared in ddH2O and PBS (10 mM, pH 7.0) for size distribution and zeta potential analysis, respectively. All measurements were performed at 25 °C. Binding capacity of PLP to PEI-CadA-CrylAb fusion particles

[0544] For binding analysis, PEI-CadA-CrylAb fusion particles (50 pg / mL) were incubated with increasing concentrations of PLP (0-2 mM) for 25 min at room temperature under gentle rotation. After incubation, particles were centrifuged at 8000 rpm for 10 min and the supernatant was quantified for unbound PLP. A calibration curve was generated from PLP standards treated under the same conditions to determine the amount of PLP bound to the particles. To stabilize the absorption peak of PLP after formation of the Schiff base with the protein, 1 M HC1 was added to the unbound PLP supernatant in a 1 : 1 (v / v) ratio. After 5 min of incubation, the absorbance of unbound PLP was measured at 294 nm using an Eppendorf Bio-Spectrometer (Eppendorf AG, Germany). Based on the Beer-Lambert law, the concentration of PLP was determined using a standard curve generated for various amounts of PLP treated with 1 M HC1.

[0545] Batch recyclability of PEI-CadA-CrylAb v.s. CadA-CrylAb

[0546] To investigate the recyclability of PEI-CadA-CrylAb particles or CadA-CrylAb particles, 1 mg / mL of each particle type was added separately to a reaction buffer containing 0.5 mM L-lysine as a substrate and 1 M MES buffer (pH 6.0). The reaction was then carried out at 45 °C for 20 min under 1000 rpm shaking. Before the first cycle, 0.1 pmol of PLP was fixed. To evaluate the effect of PEI modification, two control groups were included: (i) CadA-CrylAb particles, where 0.1 pmol of PLP was added before each cycle, and (ii) CadA-CrylAb particles, where 0.1 pmol of PLP was added only before the first cycle. L-lysine consumption was measured for all samples as described above. The conversion achieved in the first cycle was considered 100%. All experiments were performed in triplicate.

[0547] Oil-induced entrapment of PEI-modified CadA-CrylAb fusion particles in agarose beads

[0548] To prepare PEI-CadA-CrylAb fusion particles suitable for continuous flow reactor applications, the particles were encapsulated in agarose polymers using an oil-two-phase approach

[21] First, a 2 wt.% agarose solution was prepared in ddH2O and heated in the microwave for 5 min. Then the solution was cooled to 40 °C in an incubator. Meanwhile, PEI-CadA-CrylAb fusion particles (1 mg / mL) were suspended in ddH2O and gently shaken in an incubator at 37 °C. Then the two solutions were mixed in the appropriate ratio and dropped into the oil / water two-phase system using a syringe. Upon contact with the oil phase, the droplets were spherical and solidification occurred as the mixture cooled. The resulting beads were collected in the water phase and washed to remove the oil. The beads were then rinsed several times with 100 mM NaCl and ddH2O. After this, the beads (containing 200 mg of particles) were incubated with PLP (0.1 mM) at 10 °C for 1 h under slow rotation. The amount of PLP bound to the beads was measured as described above.

[0549] Confocal microscopy

[0550] To confirm the immobilization of PLP molecules on agarose beads embedding PEI-CadA-CrylAb fusion particles, PEI was first labeled with Alexa Fluor 488 NHS ester and the fusion particles were labeled with Alexa Fluor 568 C5 maleimide as previously reported TM 488NHS ester and the fusion particles were labeled with Alexa Fluor 568 C5 maleimide as previously reported TM 488NHS ester and the fusion particles were labeled with Alexa Fluor 568 C5 maleimide as previously reported

[19] After separation of PEI-Alexa 488 molecules by centrifugation, they were used to modify Alexa 568 labeled CadA-CrylAb particles. The resulting PEI-Alexa 488-CadA-CrylAb-Alexa 568 particles (0.1 mg / mL) were embedded in agarose beads and then incubated with PLP (1.0 mM) for 10 min under rotation, then centrifuged and washed with ddH2O. The beads were placed on a glass slide for confocal microscopy analysis. Using a Leica SP8 confocal microscope, the signal from the PLP molecules bound inside the beads was captured through the DAPI channel and the distribution of the Alexa 488 labeled PEI molecules was visualized.

[0551] Comparison of cadaverine production efficiency of PEI-CadA-CrylAb v.s. PEI-CadA-CrylAb-agarose beads

[0552] To compare the efficiency of PEI-CadA-CrylAb particles and PEI-CadA-CrylAb-agarose beads in cadaverine production, a fixed amount of free or agarose-embedded particles (0.211 ± 0.05 mg) was incubated in a reaction buffer containing 0.1 mM PLP and 50 mM L-lysine HC1 in 0.1 M MES (pH 6) for 10 min at 37 °C under gentle shaking at 100 rpm. Then the consumption of L-lysine HC1 was quantified. Cadaverine production from PEI-CadA-CrylAb particles was set to 100%. To investigate the effect of agarose hydrogel on substrate diffusion, the same experiment was performed using physically broken agarose beads. All experiments were performed in triplicate.

[0553] Continuous flow biosynthesis of cadaverine

[0554] PEI-CadA-CrylAb-PLP-embedded agarose beads (475 mg of particles containing 4 pmol of bound PLP) were first loaded into a column and washed with ddH2O at a flow rate of 0.5 mL / min for 60 min. After this, a reaction solution (50 mM L-lysine, 0.1 M MES, pH 6.0) was continuously fed into the column at a rate of 50 pL / min, and the product was collected.

[0555] Results

[0556] We previously demonstrated the successful immobilization and co-immobilization of enzymes (human argininease I and PLP-dependent ornithine decarboxylase) and a fluorescent reporter protein using the CrylAb platform in E. coli

[17] This approach enhanced the stability of the fusion enzyme, enabling efficient and recyclable conversion of arginine to cadaverine, which was subsequently used to synthesize a thermostable polyamide nylon-4,6. In this study, we extended this technology to the in vivo immobilization of another PLP-dependent enzyme, CadA, on CrylAb particles for cadaverine biosynthesis. Further modification of the immobilized particles with PEI and embedding them in agarose beads resulted in a simple, highly stable, and efficient system for the continuous production of cadaverine in a flow reactor.

[0557] Preparation and characterization of CadA-CrylAb

[0558] To generate CadA-CrylAb particles, CadA was fused to the N-terminal gene of CrylAb, and its expression in E. coli BL21 (DE3) was confirmed by SDS-PAGE Figure 23). The purified CadA-CrylAb fusion particles were then analyzed using orbitrap mass spectrometry to assess their structural integrity and sequence coverage. Mass spectrometry analysis showed that using the CadA protein sequence as a reference, free CadA showed about 85% sequence coverage, while using the CadA-CrylAb fusion sequence as a reference, CadA-CrylAb showed 58% sequence coverage Figure 24A and 24B ). The detection of CadA peptides within the fusion particles confirmed that the fusion was successfully achieved and that CadA was effectively integrated into the CrylAb particle. These results validated the successful expression and purification of the CadA-CrylAb fusion particles. To compare free CadA and CadA-CrylAb particles, free CadA was also expressed, purified, and verified using SDS-PAGE analysis. Additionally, size exclusion chromatography (SEC) was performed to assess the oligomeric state of CadA Figure 25 ). The SEC profile showed a clear elution peak corresponding to the expected molecular weight of CadA, indicating its correct folding.

[0559] After producing free CadA and CadA-CrylAb fusion particles, their catalytic activity was evaluated. As shown in Figure 26 , both exhibited comparable levels of cadaverine production, with the fusion particles showing only a slight decrease in activity (about 10%) relative to the free enzyme. Thermal stability analysis showed that the fusion particles exhibited comparable stability at elevated temperatures Figure 27 . To determine the optimal reaction conditions, the enzyme activity of free CadA and CadA-CrylAb particles was evaluated at different pH values. The results indicated that the optimal pH range for cadaverine production was 6.0-7.0 Figure 28 . The long-term stability of CadA-CrylAb particles was further examined at pH 6.0-7.0 over a 15-day period, showing high stability without significant loss of enzyme activity Figure 29 . The recycling of the particles was also investigated. The particles were effectively recovered by simple centrifugation and reused for 10 cycles in the conversion of 1 M L-lysine, with only a slight decrease in activity observed after the 8th cycle Figure 30 . However, the addition of the cofactor PLP was required in each cycle. PEI modification of the CadA-CrylAb fusion particles

[0560] To more efficiently use the cofactor and also reduce the downstream separation burden of the target product cadaverine, a cationization method using the highly cationic polymer PEI was explored to retain PLP. As Figure 31As shown, PEI modification significantly increased the PLP binding capacity of the particles from approximately 100 μmol / g to approximately 1300 μmol / g—a 13-fold enhancement. The modified particles displayed prolonged stability, maintaining PLP immobilization for more than 10 cycles in MES buffer (100 mM, pH 6.0), indicating effective cofactor retention ( Figure 32 Dynamic light scattering (DLS) analysis revealed that PEI modification altered the zeta potential of the particles, increasing from -50 mV for unmodified particles to +54 mV for PEI-coated CadA-Cry1Ab particles ( Figure 33 ), while the overall size distribution remains unchanged ( Figure 34 Furthermore, the binding of PLP to PEI-modified particles resulted in a decrease in the zeta potential from +54 mV to +2 mV, confirming the ionic nature of the PLP-PEI-CadA-Cry1Ab particle interaction ( Figure 33 ).

[0561] Entrapment of PEI-CadA-Cry1Ab fusion particles in agarose beads

[0562] Because the decarboxylation reaction of L-lysine increases the reaction pH and CadA is highly sensitive to alkaline conditions that can inactivate the enzyme, PEI-modified particles were embedded in agarose beads to obtain an appropriate size for use in a continuous flow reactor, facilitating continuous collection of the product and thereby enhancing enzyme stability.

[22] To evaluate the distribution of PEI-CadA-Cry1Ab particles within agarose beads and to assess the immobilization of PLP, fluorescently labeled PEI-Alexa 488-CadA-Cry1Ab-Alexa568 particles were used for embedding. Cross-sections of the beads were then analyzed using confocal microscopy. Figure 35 As shown, the particles were evenly distributed throughout the beads and confirmed the successful co-immobilization of PLP in the particles within the beads. Although the activity of the beads decreased by almost 60% after immobilization due to substrate diffusion limitations ( Figure 36 ), but this approach offers significant advantages. The ease and convenience of using immobilized beads in a bioreactor simplifies the process while also preventing issues with backpressure and particle leakage through the column, which are common challenges associated with free enzyme systems.

[0563] Continuous flow biosynthesis of cadaverine

[0564] After characterization, beads containing 75 mg of immobilized particles were placed in a continuous flow reactor. A 50 mM L-lysine solution in 0.1 M MES buffer (pH 6.0) was continuously injected at a flow rate of 50 μL / min to achieve continuous production of cadaverine. The conversion rate of L-lysine to cadaverine was evaluated at the beginning of the process, confirming complete substrate conversion. The amount of PLP bound within the same amount of beads was determined to be 4 μmol. Figure 37 As shown, the continuous flow bioreactor operated efficiently for almost three weeks without the need for additional PLP supplementation, achieving maximum substrate conversion efficiency. After 19 days of continuous operation, exogenous PLP was added to determine whether the observed loss of activity was due to PLP depletion or enzyme inactivation. The results showed that conversion recovered but then rapidly declined, indicating that enzyme inactivation, rather than cofactor depletion, was the primary cause. This extended enzyme stability and operational lifespan represents a significant advancement in biocatalyst development for continuous cadaverine production, demonstrating strong potential for industrial scalability. After 19 days, the turnover number for PLP reached 16,938, while the turnover number for CadA was 190,609, highlighting the high efficiency of the bioreactor system.

[0565] discuss

[0566] Co-immobilization of pyridoxal phosphate (PLP) and enzyme stabilization are key steps in optimizing the biocatalysis of cadaverine synthesis. Wei et al

[23] reported for the first time that chitosan powders can be used for the co-immobilization of PLP-dependent L-lysine decarboxylase and PLP. However, they face challenges related to limited surface area and irregular morphology, which restrict the adsorption of PLP and hinder their application in continuous flow systems. To address these challenges, Zhou et al.

[24] developed chitin-chitosan microspheres for co-immobilization of PLP and three different PLP-dependent enzymes (L-lysine decarboxylase, L-dopa decarboxylase, and transaminase). In this system, deacetylation of the support played a key role in enhancing PLP adsorption and enzyme stabilization. They reported that although high deacetylation significantly improved PLP retention by increasing the number of exposed primary amines, it also weakened the interaction between the chitin binding domain-fused enzyme and the support. Further improvement was achieved by modifying the microspheres with polyethyleneimine (PEI) polymer, which enhanced the PLP loading capacity due to the presence of additional amine groups. However, leakage of the enzyme from the support remains a major problem in batch mode reactions. The conversion to a continuous flow reactor significantly improved the stability of the enzyme, but the biocatalyst retained only 85% of its initial activity after 8 hours. In another study, Mi et al. [2]His-tagged L-lysine decarboxylase and PLP were co-immobilized in PEI-modified alginate barium hydrogel microspheres using a hybrid system. This system simplified enzyme purification through His-tag and facilitated the co-immobilization of PLP with PEI. They applied this system in a microreactor and achieved 2.16 g of cadaverine production after 6 h at a flow rate of 0.07 mL / min in a 10 mL column. However, a significant decrease in activity was observed only after 3 h of reaction, highlighting the need for further improvement to enhance its long-term stability and sustainability.

[0567] In this study, we developed an enzyme self-assembly protein system mediated by CrylAb fusion technology to produce CadA-CrylAb fusion granules in E. coli as an expression host. The granules were obtained by simple sonication and showed excellent potential as an effective biocatalyst for cadaverine production. However, the need for exogenous PLP cofactor was identified, prompting further modification of the granules with PEI polymer to achieve stable co-immobilization of PLP and CadA-CrylAb fusion granules. These modified granules were subsequently embedded in a Sepharose matrix and used for continuous flow biosynthesis of cadaverine. The resulting biological system showed high production efficiency, maintaining robust cadaverine synthesis (TTN PLP : 16938) for almost three weeks with minimal additional PLP, demonstrating its sustained catalytic performance. This simple preparation, purification, and PLP co-immobilization method highlights the economic feasibility of this technology for large-scale cadaverine production. Considering these promising results, this platform has great potential for scale-up and can be extended to other PLP-dependent enzymes, thereby expanding its applicability for efficient biocatalytic production of a wide range of value-added chemicals. The versatility and stability of this platform make it an attractive choice for industrial biotransformation, where long-term stability and high catalytic conversion are crucial for cost-effective production processes.

[0568] References

[0569] [1] Y. Huang, X. Ji, Z. Ma, M. Y. Xue, H. Zhao, Green chemical and biological synthesis of cadaverine: recent development and challenges, RSC Advances 2021, 11, 23922-23942.

[0570] [2] J. Mi, S. Liu, Y. Du, H. Qi, L. Zhang, Cofactor self-sufficient by co-immobilization of pyridoxal 5'-phosphate and lysine decarboxylase for cadaverine production, Bioresource Technology Reports 2022, 17, 100939.

[0571] [3] S. Kind, S. Neubauer, J. Becker, M. Yamamoto, M. G. Abendroth, O. Zelder, C. Wittmann, From zero to hero-production of bio-based nylon from renewable resources using engineered Corynebacterium glutamicum, Metabolic engineering 2014, 25, 113-123.

[0572] [4] W. Ma, K. Chen, Y. Li, N. Hao, X. Wang, P. Ouyang, Advances in Cadaverine Bacterial Production and Its Applications, Engineering 2017, 3, 308-317.

[0573] [5] A. Delavarde, G. Savin, P. Derenne, M. Boursier, R. Morales-Cerrada, B. Nottelet, J. Pinaud, S. Caillol, Sustainable polyurethanes: toward new cutting-edge opportunities, Progress in Polymer Science 2024, 151, 101805.

[0574] [6] S. Samanta, S. Selvakumar, J. Bahr, D. S. Wickramaratne, M. Sibi, B. J. Chisholm, Synthesis and Characterization of Polyurethane Networks Derived from Soybean- Oil-Based Cyclic Carbonates and Bioderivable Diamines, ACS Sustainable Chemistry & Engineering 2016, 4, 6551-6561.

[0575] [7] L. L. Cruz, B. S. F. da Silva, G. G. Araujo, T. Leal-Silva, V. G. Paula, M. R. Souza, T. S. Soares, R. Q. Moraes-Souza, G. C. Monteiro, G. P. P. Lima, D. C. Damasceno, G. T. Volpato, Phytochemical and antidiabetic analysis of Curatella americana L. aqueous extract on the rat pregnancy, Journal of Ethnopharmacology 2022, 293, 115287.

[0576] [8] R. Chen, C. Liao, Q. Guo, L. Wu, L. Zhang, X. Wang, Combined systems pharmacology and fecal metabonomics to study the biomarkers and therapeutic mechanism of type 2 diabetic nephropathy treated with Astragalus and Leech, RSC Adv 2018, 8, 27448-27463.

[0577] [9] S. Ozmen, S. Tabur, S. Oney-Birol, Alleviation role of exogenous cadaverine on cell cycle, endogenous polyamines amounts and biochemical enzyme changes in barley seedlings under drought stress, Scientific Reports 2023, 13, 17488.

[0578]

[10] F. Cassan, S. Maiale, O. Masciarelli, A. Vidal, V. Luna, O. Ruiz, Cadaverine production by Azospirillum brasilense and its possible role in plant growth promotion and osmotic stress mitigation, European Journal of Soil Biology 2009, 45, 12-19.

[0579]

[11] X. Lv, Z. Ma, X. Li, Y. Zhang, Y. Huang, T. Li, Highly efficient decarboxylation of L-lysine to cadaverine catalyzed by supported ruthenium oxide, Catalysis Communications 2021, 158, 106339.

[0580]

[12] H. T. Kim, K.-A. Baritugo, Y. H. Oh, S. M. Hyun, T. U. Khang, K. H. Kang, S. H. Jung, B. K. Song, K. Park, I.-K. Kim, M. O. Lee, Y. Kam, Y. T. Hwang, S. J. Park, J. C. Joo, Metabolic Engineering of Corynebacterium glutamicum for the High-Level Production of Cadaverine That Can Be Used for the Synthesis of Biopolyamide 510, ACS Sustainable Chemistry & Engineering 2018, 6, 5296-5305.

[0581]

[13] H. T. Kim, K.-A. Baritugo, Y. H. Oh, K.-H. Kang, Y. J. Jung, S. Jang, B. K. Song, I.-K. Kim, M. O. Lee, Y. T. Hwang, K. Park, S. J. Park, J. C. Joo, High-Level Conversion of l-lysine into Cadaverine by Escherichia coli Whole Cell Biocatalyst Expressing Hafnia alvei l-lysine Decarboxylase, 2019, 11, 1184.

[0582]

[14] Y. M. Moon, S. Y. Yang, T. R. Choi, H. R. Jung, H. S. Song, Y. H. Han, H. Y. Park, S. K. Bhatia, R. Gurav, K. Park, J. S. Kim, Y. H. Yang, Enhanced production of cadaverine by the addition of hexadecyltrimethylammonium bromide to whole cell system with regeneration of pyridoxal-5'-phosphate and ATP, Enzyme and microbial technology 2019, 127, 58-64.

[0583]

[15] U. Kanjee, I. Gutsche, E. Alexopoulos, B. Zhao, M. El Bakkouri, G. Thibault, K. Liu, S. Ramachandran, J. Snider, E. F. Pai, W. A. Houry, Linkage between the bacterial acid stress and stringent responses: the structure of the inducible lysine decarboxylase, The EMBO journal 2011, 30, 931-944.

[0584]

[16] R. Adalat, F. Saleem, N. Crickmore, S. Naz, A. R. Shakoori, In Vivo Crystallization of Three-Domain Cry Toxins, 2017, 9, 80.

[0585]

[17] X. Xiong, B. S. Heater, W. Y. Ho, H. K. Lee, M. M. Lee, M. K. Chan, Genetically Encoded Multienzyme Particles for the Biosynthesis of Putrescine from l-Arginine, ACS Sustainable Chemistry & Engineering 2024, 12, 773-784.

[0586]

[18] M. M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 1976, 72, 248-254.

[0587]

[19] R. Yekta, X. Xiong, J. Li, B. S. Heater, M. M. Lee, M. K. Chan, Mechanoresponsive Protein Crystals for NADH Recycling in Multicycle Enzyme Reactions, J. Am. Chem. Soc. 2024, 146, 18817-18822.

[0588]

[20] T. Wen, F. Qu, N. B. Li, H. Q. Luo, A facile, sensitive, and rapid spectrophotometric method for copper(II) ion detection in aqueous media using polyethyleneimine, Arabian Journal of Chemistry 2017, 10, S1680-S1685.

[0589]

[21] A. Alehosseini, E.-M. Gomez del Pulgar, M. J. Fabra, L. G. Gomez-Mascaraque, A. Benitez-Paez, M. Sarabi-Jamab, B. Ghorani, A. Lopez-Rubio, Agarose-Based Freeze-Dried Capsules Prepared by the Oil-Induced Biphasic Hydrogel Particle Formation Approach for the Protection of Sensitive Probiotic Bacteria, Food Hydrocoll. 2019, 87, 487-496.

[0590]

[22] A. Alehosseini, E.-M. Gomez del Pulgar, M. J. Fabra, L. G. Gomez-Mascaraque, A. Benitez-Paez, M. Sarabi-Jamab, B. Ghorani, A. Lopez-Rubio, Agarose-based freeze-dried capsules prepared by the oil-induced biphasic hydrogel particle formation approach for the protection of sensitive probiotic bacteria, Food Hydrocolloids 2019, 87, 487-496.

[0591]

[23] G. Wei, Y. Chen, N. Zhou, Q. Lu, S. Xu, A. Zhang, K. Chen, P. Ouyang, Chitin biopolymer mediates self-sufficient biocatalyst of pyridoxal 5'-phosphate and L-lysine decarboxylase, Chemical Engineering Journal 2022, 427, 132030.

[0592]

[24] N. Zhou, G. Wei, X. Chen, B. Wu, H. Li, Q. Lu, X. Cao, A. Zhang, K. Chen, P. Ouyang, Self-sufficient biocatalysts constructed using chitin-based microspheres, Chemical Engineering Journal 2023, 459, 141660.

[0593] Example 5: PWI-Cry3Aa delivery of SpCas9 ribosome nucleoprotein

[0594] We previously showed that Pos3Aa crystals are useful tools for delivering proteins, so we decided to look at whether PEI-Cry3Aa (PEI3A) crystals, which are more positively charged, could be used as potentially better protein delivery vehicles. To test this, we decided to try Cas9, an important protein used in gene editing. Here, rather than as a fusion, the negatively charged Cas9 ribonucleoprotein (RNP) was bound to the positively charged PEI3A crystals via electrostatic interactions.

[0595] We first tested the binding of RNP to PEI3A crystals. Cas9 protein was pre-complexed with a single guide RNA (sgRNA) to form an RNP in reduced serum medium at room temperature. PEI3A crystals, equilibrated in the same medium, were then added to the RNP at a 10: 1 (PEI3A:RNP) molar ratio and incubated at 4°C for 1 hour to form PEI3A / RNP complexes. As shown in FIG. 1A, PEI3A crystals can bind a large amount of RNP, and they can be pelleted with PEI3A by centrifugation. The binding is almost eliminated in the absence of the polyanionic sgRNA complexed with the Cas9 protein, indicating that the interaction is charge-based. Figure 38 A). We next evaluated the efficiency of intracellular delivery of RNP to human breast cancer cells MDA-MB0231 by PEI3A. RNP mixed with commercially available Lipofectamine 2000 (L2K RNP), a cationic lipid previously reported to mediate efficient RNP delivery for gene editing, was established in parallel as a positive control. Immunostaining of treated cells using Cas9 antibody showed that the delivery efficiency of PEI3A was comparable to that of L2K RNP (FIG. 1B).

[0596] We next evaluated the efficiency of intracellular delivery of RNP to human breast cancer cells MDA-MB0231 by PEI3A. RNP mixed with commercially available Lipofectamine 2000 (L2K RNP), a cationic lipid previously reported to mediate efficient RNP delivery for gene editing, was established in parallel as a positive control. Immunostaining of treated cells using Cas9 antibody showed that the delivery efficiency of PEI3A was comparable to that of L2K RNP (FIG. 1B). Figure 38 B).

[0597] We next evaluated the efficiency of intracellular delivery of RNP to human breast cancer cells MDA-MB0231 by PEI3A. RNP mixed with commercially available Lipofectamine 2000 (L2K RNP), a cationic lipid previously reported to mediate efficient RNP delivery for gene editing, was established in parallel as a positive control. Immunostaining of treated cells using Cas9 antibody showed that the delivery efficiency of PEI3A was comparable to that of L2K RNP (FIG. 1B). Figure 39 A). Further optimization experiments showed that using a 10: 1 PEI3A:RNP molar ratio to generate PEI3A / RNP complexes resulted in the highest level of editing efficiency (FIG. 1B). Figure 39 B).

[0598] All patents, patent applications, and other publications, including GenBank Accession Numbers, cited herein are hereby incorporated by reference in their entirety for all purposes.

[0599] SEQUENCE LISTING

[0600] SEQ ID NO: 1 (amino acid sequence of hArg)

[0601] MSAKSRTIGIIGAPFSKGQPRGGVEEGPTVLRKAGLLEKLKEQECDVKDYGDLPFADIPNDSPFQIVKNPRSVGKASEQLAGKVAEVKKNGRISLVLGGDHSLAIGSISGHARVHPDLGVIWVDAHTDINTPLTTTSGNLHGQPVSFLLKELKGKIPDVPGFSWVTPCISAKDIVYIGLRDVDPGEHYILKTLGIKYFSMTEVDRLGIGKVMEETLSYLLGRKKRPIHLSFDVDGLDPSFTPATGTPVVGGLTYREGLYITEEIYKTGLLSGLDIMEVNPSLGKTPEEVTRTVNTAVAITLACFGLAREGNHKPIDYLNPPK

[0602] SEQ ID NO: 2 (amino acid sequence of ODC mut

[0603] ​MKSMNIAASSELVSRLSSHRRVVALGDTDFTDVAAVVITAADSRSGILALLKRTGFHLPVFLYSEHAVELPAGVTAVINGNEQQWLELESAACQYEENLLPPFYDTLTQYVEMGNSTFACPGHQHGAFFKKHPAGRHFYDFFGENVFRADMCNADVKLGDLLTHTGSAKDAQKFAAKVFHADKTYFVLNGTSAANKVVTNALLTRGDLVLFDRNNHKSNHHGALIQAGATPVYLEASRNPFGFIGGIDAHCFNEEYLRQQIRDVAPEKADLPRPYRLAIIQLGTYDGTVYNARQVIDTVGHLCDYILFDSAWVGYEQFIPMMADSSPLLLELNENDPGIFVTQSVHKQQAGFSQTSQIHKKDNHIRGQARFCPHKRLNNAFMLHASTSPFYPLFAALDVNAKIHEGESGRRLWAECVEIGIEARKAILARCKLFRPFIPPVVDGKLWQDYPTSVLASDRRFFSFEPGAKWHGFEGYAADQYFVDPCKLLLTTPGIDAETGEYSDFGVPATILAHYLRENGIVPEKCDLNSILFLLTPAESHEKLAQLVAMLAQFEQHIEDDSPLVEVLPSVYNKYPVRYRDYTLRQLCQEMHDLYVSFDVKDLQKAMFRQQSFPSVVMNPQDAHSAYIRGDVELVRIRDAEGRIAAEGALPYPPGVLCVVPGEVWGGAVQRYFLALEEGVNLLPGFSPELQGVYSETDADGVKRLYGYVLK

[0604] SEQ ID NO: 3 (amino acid sequence of CrylAb)

[0605]

[0606] SEQ ID NO: 4 (amino acid sequence of Cry3Aa)

[0607] MNPNNRSEHDTIKTTENNEVPTNHVQYPLAETPNPTLEDLNYKEFLRMTADNNTEALDSSTTKDVIQKGISVVGDLLGVVGFPFGGALVSFYTNFLNTIWPSEDPWKAFMEQVEALMDQKIADYAKNKALAELQGLQNNVEDYVSALSSWQKNPVSSRNPHSQGRIRELFSQAESHFRNSMPSFAISGYEVLFLTTYAQAANTHLFLLKDAQIYGEEWGYEKEDIAEFYKRQLKLTQEYTDHCVKWYNVGLDKLRGSSYESWVNFNRYRREMTLTVLDLIALFPLYDVRLYPKEVKTELTRDVLTDPIVGVNNLRGYGTTFSNIENYIRKPHLFDYLHRIQFHTRFQPGYYGNDSFNYWSGNYVSTRPSIGSNDIITSPFYGNKSSEPVQNLEFNGEKVYRAVANTNLAVWPSAVYSGVTKVEFSQYNDQTDEASTQTYDSKRNVGAVSWDSIDQLPPETTDEPLEKGYSHQLNYVMCFLMQGSRGTIPVLTWTHKSVDFFNMIDSKKITQLPLVKAYKLQSGASVVAGPRFTGGDIIQCTENGSAATIYVTPDVSYSQKYRARIHYASTSQITFTLSLDGAPFNQYYFDKTINKGDTLTYNSFNLASFSTPFELSGNNLQIGVTGLSAGDKVYIDKIEFIPVN

[0608] SEQ ID NO: 5 (amino acid sequence of wild-type ODC)

[0609] MKSMNIAASSELVSRLSSHRRVVALGDTDFTDVAAVVITAADSRSGILALLKRTGFHLPVFLYSEHAVELPAGVTAVINGNEQQWLELESAACQYEENLLPPFYDTLTQYVEMGNSTFACPGHQHGAFFKKHPAGRHFYDFFGENVFRADMCNADVKLGDLLIHEGSAKDAQKFAAKVFHADKTYFVLNGTSAANKVVTNALLTRGDLVLFDRNNHKSNHHGALIQAGATPVYLEASRNPFGFIGGIDAHCFNEEYLRQQIRDVAPEKADLPRPYRLAIIQLGTYDGTVYNARQVIDTVGHLCDYILFDSAWVGYEQFIPMMADSSPLLLELNENDPGIFVTQSVHKQQAGFSQTSQIHKKDNHIRGQARFCPHKRLNNAFMLHASTSPFYPLFAALDVNAKIHEGESGRRLWAECVEIGIEARKAILARCKLFRPFIPPVVDGKLWQDYPTSVLASDRRFFSFEPGAKWHGFEGYAADQYFVDPCKLLLTTPGIDAETGEYSDFGVPATILAHYLRENGIVPEKCDLNSILFLLTPAESHEKLAQLVAMLAQFEQHIEDDSPLVEVLPSVYNKYPVRYRDYTLRQLCQEMHDLYVSFDVKDLQKAMFRQQSFPSVVMNPQDAHSAYIRGDVELVRIRDAEGRIAAEGALPYPPGVLCVVPGEVWGGAVQRYFLALEEGVNLLPGFSPELQGVYSETDADGVKRLYGYVLK

[0610] SEQ ID NO: 6 (amino acid sequence of wild-type alcohol dehydrogenase ADH)

[0611] MRVKDKVAIVTGGAKGLGEATSRLLAKEGAKVVLTDLDEVQGKKVEAEINKAGGIAKYMKHDVANEQQWKKVVDDTVKLFGKLDILVNNAGVGHSASAEDETLEKWRFLQSINLDAVFLGTREAIRVMKNQKSGSIINLSSIEGLVGDPTLAAYNASKGGVRLFTKSAALHCAKSGYGIRVNSVHPGYIWTPMVEESAKESGDVEERKAALTALHPIGKLGVPDDIAYGILYLASDESGFVTGSELVIDGGYTAQ

[0612] SEQ ID NO: 7 (amino acid sequence of wild-type formate dehydrogenase FDH)

[0613] MAKILCVLYDDPVDGYPKTYARDDLPKIDHYPGGQTLPTPKAIDFTPGQLLGSVSGELGLRKYLEANGHTFVVTSDKDGPDSVFEKELVDADVVISQPFWPAYLTPERIAKAKNLKLALTAGIGSDHVDLQSAIDRGITVAEVTYCNSISVAEHVVMMILGLVRNYIPSHDWARKGGWNIADCVEHSYDLEGMTVGSVAAGRIGLAVLRRLAPFDVKLHYTDRHRLPEAVEKELGLVWHDTREDMYPHCDVVTLNVPLHPETEHMINDETLKLFKRGAYIVNTARGKLADRDAIVRAIESGQLAGYAGDVWFPQPAPKDHPWRTMKWEGMTPHISGTSLSAQARYAAGTREILECFFEGRPIRDEYLIVQGGALAGTGAHSYSKGNATGGSEEAAKFKKAG

[0614] SEQ ID NO: 8 (amino acid sequence of wild-type leucine dehydrogenase LDH)

[0615] MTLEIFEYLEKYDYEQVVFCQDKESGLKAIIAIHDTTLGPALGGTRMWTYDSEEAAIEDALRLAKGMTYKNAAAGLNLGGAKTVIIGDPRKDKSEAMFRALGRYIQGLNGRYITAEDVGTTVDDMDIIHEETDFVTGISPSFGSSGNPSPVTAYGVYRGMKAAAKEAFGTDNLEGKVIAVQGVGNVAYHLCKHLHAEGAKLIVTDINKEAVQRAVEEFGASAVEPNEIYGVECDIYAPCALGATVNDETIPQLKAKVIAGSANNQLKEDRHGDIIHEMGIVYAPDYVINAGGVINVADELYGYNRERALKRVESIYDTIAKVIEISKRDGIATYVAADRLAEERIASLKNSRSTYLRNGHDIISRR

Claims

1. A protein crystal formed from a Cry protein, a crystal-forming variant thereof, a crystal-forming fragment thereof, or a fusion polypeptide comprising said Cry protein, said variant, or said fragment fused to a heterologous protein, wherein said protein crystal is modified with polyethylenimine (PEI).

2. The protein crystal of claim 1, wherein said Cry protein is Cry3Aa or CrylAb.

3. The protein crystal of claim 1, formed from a fusion polypeptide comprising a Cry protein, a crystal-forming variant of said Cry protein, or a crystal-forming fragment of said Cry protein fused to a heterologous protein.

4. The protein crystal of claim 1, formed from a Cry protein, a crystal-forming variant of said Cry protein, or a crystal-forming fragment of said Cry protein, and comprising a heterologous protein embedded within.

5. The protein crystal of claim 3 or 4, wherein said heterologous protein is an enzyme.

6. The protein crystal of claim 5, wherein said enzyme is formate dehydrogenase (FDH), leucine dehydrogenase (LDH), human arginase (hArg), ornithine decarboxylase (ODC), alcohol dehydrogenase (ADH), mandelate dehydrogenase (MDH), carbonyl reductase (CBR), reductive aminase (Red Am), or cytochrome P450 (CYP or P450).

7. The protein crystal of claim 3, wherein said fusion polypeptide comprises a Cry protein, a crystal-forming variant of said Cry protein, or a crystal-forming fragment of said Cry protein fused to an enzyme.

8. The protein crystal of claim 7, wherein said fusion polypeptide is a Cry3Aa-ADH fusion, a Cry3Aa-FDH fusion, a Cry3Aa-LDH fusion, a Cry3Aa-MDH fusion, a Cry3Aa-P450 fusion, a CrylAb-hArg fusion, a CrylAb-ODC fusion, or a CrylAb-P450 fusion.

9. The protein crystal of any one of claims 5-8, wherein said protein crystal is bound to a cofactor required for a chemical reaction catalyzed by said enzyme.

10. The protein crystal of claim 9, wherein said cofactor is a phosphorylated cofactor.

11. The protein crystal of claim 10, wherein said phosphorylated cofactor is nicotinamide adenine dinucleotide (NADH), flavin adenine dinucleotide (FADH), or pyridoxal phosphate (PLP).

12. The protein crystal of claim 9, wherein said enzyme is ADH, FDH, LDH, MDH, CBR, or Red Am, and said phosphorylated cofactor is NADH.

13. The protein crystal of claim 9, wherein said enzyme is ODC, and said phosphorylated cofactor is PLP.

14. The protein crystal of claim 1, comprising two fusion polypeptides.

15. A composition comprising the protein crystal of any one of claims 1-14.

16. The composition of claim 15, wherein the protein crystal comprises an enzyme capable of catalyzing a substrate in a chemical reaction and a co-factor required for the reaction, and wherein the composition further comprises the substrate.

17. The composition of claim 15, wherein the protein crystal is formed from a fusion polypeptide comprising a Cry protein fused to ADH, FDH, LDH, MDH, or P450.

18. The composition of claim 15, wherein the protein crystal is formed from a fusion polypeptide comprising a Cry protein fused to hArg or ODC.

19. The composition of claim 15, wherein the protein crystal is formed from two fusion polypeptides.

20. The composition of claim 17 or 19, wherein the protein crystal binds to NADH.

21. The composition of claim 18, wherein the protein crystal binds to PLP.

22. A method of performing a chemical reaction, comprising the steps of: incubating the protein crystal of any one of claims 5-12 with a substrate for the enzyme under conditions that allow the substrate to be catalyzed by the enzyme in the chemical reaction.

23. The method of claim 22, wherein the enzyme is formate dehydrogenase (FDH), leucine dehydrogenase (LDH), human arginase (hArg), ornithine decarboxylase (ODC), alcohol dehydrogenase (ADH), mandelate dehydrogenase (MDH), carbonyl reductase (CBR), or reductive aminase (RedAm) or cytochrome P450 (CYP or P450).

24. The method of claim 22, wherein the enzyme is ADH, FDH, LDH, MDH, CBR, or RedAm, and the protein crystal binds to NADH.

25. The method of claim 22, wherein the enzyme is ODC, and the protein crystal binds to PLP.

26. The method of claim 22, wherein the protein crystal comprises two fusion polypeptides.

27. The method of any one of claims 22-26, further comprising, after the reaction is complete, removing the reaction product and reusing the crystal in a second reaction.

28. An in vitro method of immobilizing a co-factor by incubating the co-factor with the protein crystal of any one of claims 1-14 in solution and allowing the co-factor to bind to the protein crystal.

29. The method of claim 28, wherein the co-factor is NADH or PLP.

30. The method of claim 28, further comprising, after the co-factor binds to the protein crystal, agitating the solution to release the co-factor from the protein crystal.

31. The method of claim 28, further comprising, after the co-factor is released from the protein crystal, ceasing agitation of the solution to allow the co-factor to bind to the protein crystal again.

32. A method of delivering a protein of interest to a target cell, comprising contacting the target cell with a preformed complex between the protein of interest and a protein crystal of any one of claims 1-14.

33. The method of claim 32, wherein the protein crystal is a PEI-Cry3Aa protein crystal, and wherein the protein of interest is a negatively charged protein or a protein comprising or linked to a negatively charged moiety.

34. The method of claim 32, further comprising incubating the protein of interest with the protein crystal to form a complex prior to the contacting.

35. The method of claim 32, wherein the protein of interest is a nucleic acid, and wherein the contacting is performed in the presence of an endosomolytic agent.

Images