Development of a Versatile Method to Construct Direct Electron Transfer-Type Enzyme Complexes Employing SpyCatcher/SpyTag System

The electrochemical enzyme sensors based on direct electron transfer (DET)-type oxidoreductase-based enzymes are ideal for continuous and in vivo monitoring. However, the number and types of DET-type oxidoreductases are limited. The aim of this research is the development of a versatile method to create a DET-type oxidoreductase complex based on the SpyCatcher/SpyTag technique by preparing SpyCatcher-fused heme c and SpyTag-fused non-DET-type oxidoreductases, and by the in vitro formation of DET-type oxidoreductase complexes. A heme c containing an electron transfer protein derived from Rhizobium radiobacter (CYTc) was selected to prepare SpyCatcher-fused heme c. Three non-DET-type oxidoreductases were selected as candidates for the SpyTag-fused enzyme: fungi-derived flavin adenine dinucleotide (FAD)-dependent glucose dehydrogenase (GDH), an engineered FAD-dependent d-amino acid oxidase (DAAOx), and an engineered FMN-dependent l-lactate oxidase (LOx). CYTc-SpyCatcher (CYTc-SC) and SpyTag-Enzymes (ST-GDH, ST-DAAOx, ST-LOx) were prepared as soluble molecules while maintaining their redox properties and catalytic activities, respectively. CYTc-SC/ST-Enzyme complexes were formed by mixing CYTc-SpyCatcher and SpyTag-Enzymes, and the complexes retained their original enzymatic activity. Remarkably, the heme domain served as an electron acceptor from complexed enzymes by intramolecular electron transfer; consequently, all constructed CYTc-SC/ST-Enzyme complexes showed DET ability to the electrode, demonstrating the versatility of this method.


Introduction
The principles of current electrochemical enzyme sensors are categorized into three generations [1,2]: the first generation utilizes oxygen as an electron acceptor, the second generation uses synthetic electron acceptors or mediators, and the third generation employs enzymes capable of direct electron transfer (DET). Enzyme sensors employing DET-type oxidoreductases do not require oxygen or any electron mediators and can be operated under a low oxidation potential, which is advantageous for minimizing the impact of electrochemically active ingredients. Furthermore, toxic synthetic mediators are not necessary for the electrochemical measurements, and errors due to variations in the concentration of oxygen in the biological samples are eliminated. Hence, DET-type oxidoreductase-based enzyme sensors are ideal for continuous and in vivo monitoring. However, the number and types of DET-type oxidoreductases, which are inherently able to transfer electrons from their redox cofactor to the electrode, are limited. Scheme 1. The architecture to create DET-type enzyme complex facilitated by SpyCatcher/SpyTag complex. In formed complex, heme c bound in electron transfer protein works as electron transfer domain from oxidoreductase to electrode with high proximity to cofactor of enzyme.
A heme c containing an electron transfer protein derived from Rhizobium radiobacter (CYTc) was selected. This molecule contained mono heme c covalently attached to the heme-binding motif and functioned as an electron acceptor of FAD-dependent glucoside 3-dehydrogenase [43]. Three non-DET-type oxidoreductases were selected as candidates for the SpyTag-fused enzyme as shown in Table 1: an engineered FAD-GDH, an engineered FAD-dependent D-amino acid oxidase (DAAOx), and an engineered FMN-dependent L-lactate oxidase (LOx) derived from Aspergillus flavus [44,45], Rhodotorula gracilis [18,46], and Aerococcus viridans [15,47], respectively.  [15] CYTc-SpyCatcher (CYTc-SC) and SpyTag-Enzymes (ST-GDH, ST-DAAOx, ST-LOx), which contain flavin cofactors, were expressed separately. Both CYTc-SpyCatcher and each SpyTag-Enzyme were prepared as soluble molecules while maintaining their redox properties and catalytic activities, respectively. The CYTc-SC/ST-Enzyme complexes were formed by mixing the CYTc-SpyCatcher and SpyTag-Enzyme, and the complexes retained their original enzymatic activity. Remarkably, the heme domain served as an electron acceptor from complexed enzymes by intramolecular electron transfer; consequently, all constructed CYTc-SC/ST-Enzyme complexes showed DET ability to the electrode, demonstrating the versatility of this method.

Recombinant Expression and Characterization of CYTc-SC
The expression vector for CYTc-SC (Supplementary Materials, Figure S1) was transformed into the E. coli BL21(DE3) host strain containing pEC86, which was inserted by ccm operon [36]. CYTc-SC was expressed as a soluble protein and purified using Ni 2+ affinity chromatography. The SDS-PAGE analysis of purified CYTc-SC showed a band at the expected molecular weight of 23 kDa ( Figure 1A). This result confirms the expression of CYTc-SC as a soluble fusion molecule in the host strain.
formed by mixing the CYTc-SpyCatcher and SpyTag-Enzyme, and the complexes retained their original enzymatic activity. Remarkably, the heme domain served as an electron acceptor from complexed enzymes by intramolecular electron transfer; consequently, all constructed CYTc-SC/ST-Enzyme complexes showed DET ability to the electrode, demonstrating the versatility of this method.

Recombinant Expression and Characterization of CYTc-SC
The expression vector for CYTc-SC (Supplementary Materials, Figure S1) was transformed into the E.coli BL21(DE3) host strain containing pEC86, which was inserted by ccm operon [36]. CYTc-SC was expressed as a soluble protein and purified using Ni 2+ affinity chromatography. The SDS-PAGE analysis of purified CYTc-SC showed a band at the expected molecular weight of 23 kDa ( Figure 1A). This result confirms the expression of CYTc-SC as a soluble fusion molecule in the host strain. The redox properties of heme c in CYTc-SC were analyzed by investigating the absorbance spectra in the oxidized and reduced states ( Figure 1B). Depending on its redox state, CYTc exhibits different absorption spectra, with two absorption peaks observed in the reduced state that are not observed in the oxidized state, which is typical for heme c [43]. The oxidized CYTc-SC showed a peak at 409 nm, corresponding to the Soret band. The reduced CYTc-SC showed a peak at 415 nm and two peaks at 521 nm and 551 nm corresponding to Q bands derived from the reduction of heme c. The absorbance spectra of oxidized/reduced CYTc-SC were similar to the previous observation with CYTc [43]. These results indicate that CYTc-SC was expressed with properly matured heme c with the original redox property upon fusion with SpyCatcher.

Complex Formation between CYTc-SC and ST-Enzymes
ST-GDH, ST-DAAOx, and ST-LOx were recombinantly expressed and purified using Ni 2+ affinity chromatography. Each ST-Enzyme was prepared as a soluble molecule. Thus, the prepared samples were subjected to an enzyme analysis, which revealed that SpyTagfusion did not have a significant negative impact on the selected enzymes (Table 2). To The redox properties of heme c in CYTc-SC were analyzed by investigating the absorbance spectra in the oxidized and reduced states ( Figure 1B). Depending on its redox state, CYTc exhibits different absorption spectra, with two absorption peaks observed in the reduced state that are not observed in the oxidized state, which is typical for heme c [43]. The oxidized CYTc-SC showed a peak at 409 nm, corresponding to the Soret band. The reduced CYTc-SC showed a peak at 415 nm and two peaks at 521 nm and 551 nm corresponding to Q bands derived from the reduction of heme c. The absorbance spectra of oxidized/reduced CYTc-SC were similar to the previous observation with CYTc [43]. These results indicate that CYTc-SC was expressed with properly matured heme c with the original redox property upon fusion with SpyCatcher.

Complex Formation between CYTc-SC and ST-Enzymes
ST-GDH, ST-DAAOx, and ST-LOx were recombinantly expressed and purified using Ni 2+ affinity chromatography. Each ST-Enzyme was prepared as a soluble molecule. Thus, the prepared samples were subjected to an enzyme analysis, which revealed that SpyTagfusion did not have a significant negative impact on the selected enzymes (Table 2). To observe the complex formation between CYTc-SC/ST-Enzymes, both samples were mixed in an equivalent molar ratio and incubated on ice for 2 h. The complex formation was confirmed by band shifting using an SDS-PAGE analysis ( Figure 2). By mixing CYTc-SC with ST-Enzymes, bands corresponding to the expected molecular weights were observed:  observe the complex formation between CYTc-SC/ST-Enzymes, both samples were mix in an equivalent molar ratio and incubated on ice for 2 h. The complex formation w confirmed by band shifting using an SDS-PAGE analysis ( Figure 2). By mixing CYTc-S with ST-Enzymes, bands corresponding to the expected molecular weights we observed: 87 kDa ( After the purification of each CYTc-SC/ST-Enzyme complex, the enzyme activit were measured, which revealed that the complex formation with CYTc-SC did not hav significant negative impact on enzyme activities. The determined KM and Vmax values each ST-Enzymes or complexes shown in Table 2   After the purification of each CYTc-SC/ST-Enzyme complex, the enzyme activities were measured, which revealed that the complex formation with CYTc-SC did not have a significant negative impact on enzyme activities. The determined K M and V max values of each ST-Enzymes or complexes shown in Table 2

Investigation of Intramolecular Electron Transfer in CYTc-SC/ST-Enzyme Complexes
To investigate the ability of intramolecular electron transfer from the flavin cofactor to heme c in the CYTc-SC/ST-Enzymes, changes in the absorbance spectra in the presence of the substrate were observed. The time-dependent absorbance spectra are shown in Figure 3 and Supplementary Materials Figure S3. After the addition of the substrate, a shift of the peak from 409 nm to 415 nm, corresponding to the Soret band of heme c, was observed. Simultaneously, an increase in the absorbance peaks at 521 nm and 551 nm derived from the Q bands of reduced heme c was observed in all cases of CYTc-SC/ST-Enzyme complexes. These results suggest that the addition of substrates for each complex resulted in the reduction of the flavin cofactor, followed by the intramolecular electron transfer from the reduced flavin cofactor to heme c. Consequently, the characteristically reduced heme c spectrum was observed. Thus, the CYTc-SC/ST-Enzyme complexes acquired the intramolecular electron transfer ability from flavin cofactor to heme with substrate oxidation by the complexed ST-Enzyme. resulted in the reduction of the flavin cofactor, followed by the intramolecular electron transfer from the reduced flavin cofactor to heme c. Consequently, the characteristically reduced heme c spectrum was observed. Thus, the CYTc-SC/ST-Enzyme complexes acquired the intramolecular electron transfer ability from flavin cofactor to heme with substrate oxidation by the complexed ST-Enzyme.

Electrochemical Evaluation of CYTc-SC/ST-Enzyme Complexes
The DET ability of each CYTc-SC/ST-Enzyme was then investigated by chronoamperometry measurements using a CYTc-SC/ST-Enzyme complex-immobilized gold disk electrode. Figure 4 shows the representative current responses with the successive addition of the substrate: Figure 4A and D, CYTc-SC/ST-GDH-immobilized electrode; Figure 4B and E, CYTc-SC/ST-DAAOx-immobilized electrode; and Figure 4C and F, CYTc-SC/ST-LOx-immobilized electrode. These results revealed that the electrodes with CYTc-SC/ST-Enzyme complexes exhibited current increases in the sample addition,

Electrochemical Evaluation of CYTc-SC/ST-Enzyme Complexes
The DET ability of each CYTc-SC/ST-Enzyme was then investigated by chronoamperometry measurements using a CYTc-SC/ST-Enzyme complex-immobilized gold disk electrode. Figure 4 shows the representative current responses with the successive addition of the substrate: Figure 4A Table 3). Similarly, CYTc-SC/ST-DAAOx and CYTc-SC/ST-LOx showed a current response depending on the addition of the substrate with I max app as 7.5 nA ( Figure 4B,E, Table 3) or 410 nA ( Figure 4C,F, Table 3 Table 2). In the CYTc-SC/ST-LOx-immobilized electrode, K M app from the current response was slightly lower than the K M value (K M app = 1.8 mM, K M = 3.3 mM; Tables 2 and 3). These results revealed that heme c in complexed CYTc-SC acted as an electron transfer domain from the flavin cofactor in ST-Enzyme to the electrode, achieving the DET between the enzyme and electrode. Thus, by preparing the SpyCatcher-fused heme c and SpyTagfused non-DET-type oxidoreductases and by forming CYTc-SC/ST-Enzyme complexes in vitro, DET-type oxidoreductase complexes were successfully prepared. and the current increases were dependent upon the substrate concentration, even in the absence of an external electron acceptor. In contrast, the current response of each ST-Enzyme-immobilized electrode was hardly observed. The CYTc-SC/ST-GDHimmobilized electrode showed a current response with an apparent Imax app of 55 nA ( Figure 4A,D, Table 3). Similarly, CYTc-SC/ST-DAAOx and CYTc-SC/ST-LOx showed a current response depending on the addition of the substrate with Imax app as 7.5 nA ( Figure  4B,E, Table 3) or 410 nA ( Figure 4C,F, Table 3). The current responses of CYTc-SC/ST-GDH-and CYTc-SC/ST-DAAOx-immobilized electrodes were saturated at a lower concentration according to their KM values, which reflects the much lower KM app values (5.6 mM for CYTc-SC/ST-GDH and 0.3 mM for CYTc-SC/ST-DAAOx; Table 3) compared with their KM values (44 mM for CYTc-SC/ST-GDH and 1.9 mM for CYTc-SC/ST-DAAOx; Table 2). In the CYTc-SC/ST-LOx-immobilized electrode, KM app from the current response was slightly lower than the KM value (KM app = 1.8 mM, KM = 3.3 mM; Table 2 and Table 3). These results revealed that heme c in complexed CYTc-SC acted as an electron transfer domain from the flavin cofactor in ST-Enzyme to the electrode, achieving the DET between the enzyme and electrode. Thus, by preparing the SpyCatcher-fused heme c and SpyTag-fused non-DET-type oxidoreductases and by forming CYTc-SC/ST-Enzyme complexes in vitro, DET-type oxidoreductase complexes were successfully prepared.

Discussion
In this study, we demonstrated a novel and versatile method to create a DET-type enzyme complex based on an in vitro protein complex formation technique, which is an alternative to conventional gene fusion techniques. The formation of DET-type oxidoreductase complexes was confirmed by the observation of the intramolecular electron transfer between the flavin cofactor in ST-Enzyme and CYTc-SC and by the observation of the electron transfer from the CYTc-SC/ST-Enzyme to the electrode on the substrate addition. As we expected, our proposed method solved the addressed inherent issues in the preparation of DET-type oxidoreductases by genetic fusion techniques such as protein folding and post-translational modification. The in vitro complex formation between the heme protein and oxidoreductase by the SpyCatcher/SpyTag system waived the construction of the large fusion molecule between oxidoreductases and heme-containing electron transfer proteins. Instead, the fusion of the SpyTag peptide, which consists of only 13 amino acids, to oxidoreductases did not cause any significant impact on protein folding, and oxidoreductases with the fused SpyTag were produced as soluble proteins without a significant decrease in enzyme activities (Table 2). It was also demonstrated that the SpyCatcher-fused heme c molecule was also recombinantly produced as a soluble protein, and heme c was properly matured showing its characteristic two peaks at 521 nm and 551 nm, corresponding to the Q bands derived from the reduction of heme c (Figure 1). The in vitro complex formation between the ST-Enzyme and CYTc-SC was successively achieved with high yield for the CYTc-SC/ST-GDH complex and for the CYTc-SC/ST-DAAOx complex within 2 h (Figure 2). Since the formation of the CYTc-SC/ST-LOx complex was not completed in this period, further optimization for this case will be required. Consequently, this method enables us to create flavin-dependent oxidoreductases complexed with heme c in the bacterial host, which should be proceeded by different post-translational modifications from the one for the flavin enzymes.
The intramolecular electron transfer ability of each CYTc-SC/ST-Enzyme was confirmed by the observed increase in the Q bands of the reduced heme c when the enzyme substrate was added, which reduced the flavin. Since the catalytic activities of St-Enzymes and CYTc-SC/ST-Enzymes were confirmed (Table 2), the reduction of the flavin cofactor was expected to occur. Therefore, the increase in Q bands was due to the intramolecular electron transfer between flavin and heme c in CYTc-SC. A difference in the heme c reduction speed during the intramolecular electron transfer was observed, which depends on the oxidoreductases employed to prepare ST-Enzyme-CYTc-SC/ST-LOx > CYTc-SC/ST-DAAOx > CYTc-SC/ST-GDH (faster to slower). These results suggest that the intramolecular electron transfer from the flavin cofactor to heme c depends on the redox enzyme used for the complex formation with CYTc-SC. Furthermore, these trends are in good agreement with the DET abilities of CYTc-SC/ST-Enzymes, where CYTc-SC/ST-LOx showed the highest catalytic current.
Considering that all CYTc-SC/ST-Enzymes employed the same electron transfer protein, CYTc, the most crucial feature in the design of the complex would be the intramolecular electron transfer ability between ST-Enzyme and CYTc-SC. The efficiency of the electron transfer depended on the distance, orientation, and affinity/propulsion between ST-Enzyme and CYTc-SC. Another consideration was the quaternary structure of the oxidoreductases used in this study. While FAD-GDH is a monomeric enzyme, DAAOx is a homodimer, and LOx is a homotetramer in the C4 symmetrical structure. Considering that CYTc-SC/ST-LOx showed the highest and CYTc-SC/ST-DAAOx showed the second highest intramolecular electron transfer, the electron transfer between flavin and heme c must have occurred not only in the same "subunit", but also by the inter-subunit electron transfer, which occurs between flavin and heme c in different subunits in homo-oligomers. Therefore, the observed increase in the Q bands is the sum of both the intra-subunit and inter-subunit electron transfers. Considering that the homotetrameric structure of LOx may provide more chances for inter-subunit electron transfer than homodimeric DAAOx and monomeric GDH, the quaternary structure of the original oxidoreductases will have a significant impact on the intramolecular electron transfer ability and, consequently, the DET ability of the CYTc-SC/ST-Enzymes.
To the best of our knowledge, this is the first report to achieve the internal electron transfer between the catalytic cofactor of oxidoreductase and the redox cofactor of an electron transfer protein in a complex formed by the SpyCatcher/SpyTag reaction. With respect to the design of DET-type enzymes, the ideal electron transfer domain should meet the requirements of small size and appropriate electrostatic surface charge to be suitable sterically and electrostatically as an electron acceptor to the catalytic domain or as an electron donor to the electrode. Indeed, our recent engineering approaches showed an enhanced DET ability of DET-type enzymes by downsizing the electron transfer subunit [48], or by mutating the amino acids at domain interface to suppress the electrostatic repulsion between the catalytic domain and electron transfer domain interfaces [49]. Additionally, the lower redox potential of the electron transfer domain was favorable for attaining a lower operating potential to avoid interference. The novel approach proposed in this study will provide a wide variety of design methods for DET-type enzyme molecules by selecting the heme c protein with respect to its size, redox potential, and electrostatic surface charge, regardless of the issues related to the post-translational modification of cofactors.

Construction of the Expression Vectors
The amino acid sequences of the CYTc and SpyCatcher have been previously described [43,50]. To fuse the SpyCatcher to the C-terminus of CYTc, the following genes were synthesizedo: the structural gene of CYTc (residues 16-130) without its original signal sequence with an NcoI-pelB signal sequence HisTag at the 5 end, and the structural gene of SpyCatcher (residues 21 to 104) with HindIII at the 3 end was subsequently coded. This gene fragment, NcoI-pelB-HisTag-CYTc-SpyCatcher-HindIII, was amplified by a polymerase chain reaction (PCR) and inserted into pET30c(+) (Merck KGaA, Darmstadt, Germany) to construct an expression vector of the CYTc-SpyCatcher. To construct the ST-Enzyme expression vector, the structural gene of the SpyTag was added at the 5 side of the structural gene of the FAD-dependent GDH derived from Aspergillus flavus [44,45], a G52V mutant of DAAOX derived from Rhodotorula gracilis [18,46], and an A96L/N212K double mutant of LOX derived from Aerococcus viridans [15,47]
The cells were collected, washed with 0.85% NaCl, and disrupted by ultrasonication in 20 mM potassium phosphate buffer (pH 7.0) containing 20 mM imidazole and 500 mM NaCl. Purification was performed using the following chromatographic steps. The supernatant of the disrupted cells was applied to a HisTrap HP column (Cytiva, Marlborough, MA, USA) equilibrated with 20 mM potassium phosphate buffer (pH 7.0) containing 20 mM imidazole and 500 mM NaCl. The adsorbed protein was eluted using a linear imidazole gradient (20-500 mM) in 20 mM potassium phosphate buffer (pH 7.0) containing 500 mM NaCl. Eluted proteins were dialyzed against the 20 mM potassium phosphate buffer (pH 7.0). The eluted fractions were analyzed using SDS-PAGE.

Preparation of Oxidized or Reduced Sample of CYTc-SC
To prepare the oxidized CYTc-SC sample, the purified sample (0.1 mM) was incubated with 1 mM potassium ferricyanide and then dialyzed against the 20 mM potassium phosphate buffer (pH 7.0). The reduction of CYTc-SC was investigated by adding excess sodium dithionite. Each sample was analyzed using a spectrometer to confirm the peaks derived from heme c and to confirm the redox properties of heme.

Formation of CYTc-SpyCatcher/SpyTag-Enzyme
To prepare the complex formed by the SpyCatcher/SpyTag system, 0.15 mM oxidized CYTc-SpyCatcher was mixed with 0.15 mM ST-GDH, ST-DAAOx, or ST-LOx and incubated on ice for 2 h. After incubation, each complex was purified by size-exclusion chromatography using Superdex 200 Increase 10/300 GL (Cytiva, Marlborough, MA, USA) with 20 mM potassium phosphate buffer (pH 7.0) containing 150 mM NaCl as the elution buffer to remove the unreacted molecules. Eluted samples were collected and analyzed by SDS-PAGE to determine the expected molecular weight of the SpyCatcher/SpyTag complex.

Enzyme Activity and Absorption Spectrum Analysis of the Complexes
The dye-mediated enzymatic activities of ST-Enzymes (ST-GDH, ST-DAAOx, and ST-LOx) and CYTc-SC/ST-Enzymes (CYTc-SC/ST-GDH, CYTc-SC/ST-DAAOx, and CYTc-SC/ST-LOx) were determined. The enzyme sample was incubated in a 20 mM potassium phosphate buffer (pH 7.0) containing 0.6 mM PMS, 0.06 mM DCIP, and various concentrations of substrate (glucose for GDH, D-serine for DAAOx, and L-lactate for LOx). The decrease in absorbance at 600 nm derived from the reduction of DCIP was monitored using a UV-1280 spectrometer (Shimadzu, Kyoto, Japan). In this assay, the reduction of 1 µmol of DCIP in 1 min was defined as 1 unit of dehydrogenase activity, which corresponded to the oxidation of 1 µmol of the substrate.
To observe the internal electron transfer from the flavin cofactor to heme c in each CYTc-SC/ST-Enzyme, the absorbance spectra of CYTc-SC/ST-Enzymes were measured using a UV-1800 spectrometer (Shimadzu, Kyoto, Japan). The enzyme solution was prepared in a 20 mM potassium phosphate buffer (pH 7.0). The substrate solution was then applied to the complex solution, and the absorbance spectra were recorded every 1 min for 20 min after the addition of the substrate.

Electrochemical Measurement
Gold disk electrodes (7 mm 2 ) were polished with alumina slurry and electrochemically cleaned with 0.5 M NaOH and 0.5 M H 2 SO 4 , as described previously [52]. The enzyme ink was prepared by mixing 80 µg of mesoporous carbon particles (Cnovel P(4)050, Toyo Tanso . Five microliters of each enzyme ink was dropped onto the cleaned Au electrode and dried. After drying, the ink-deposited electrode was cross-linked with 25% glutaraldehyde vapor for 30 min. The enzyme ink layer was then coated with 1 µL of 1% (w/v) Nafion solution in 20 mM potassium phosphate buffer and dried. The electrode was stored in 20 mM potassium phosphate buffer (pH 7.0) until use. The DET properties of the enzyme and complex immobilized electrode were evaluated by chronoamperometry measurements with an Ag/AgCl electrode and Pt wire as the reference electrode and counter electrode, respectively. Each electrode was immersed in 10 mL of 20 mM potassium phosphate buffer (pH 7.0) with continuous stirring. All chronoamperometric measurements were performed using a VSP Electrochemical Measurement System (Bio-Logic Science Instruments, Seyssinet-Pariset, France). A bias potential of +400 mV (vs. Ag/AgCl) was applied, and the current response was recorded using an electrochemical measurement system with the successive addition of the substrate solution. The output amperogram was analyzed using a band-reject filter with a Welch window. The rejected band frequencies ranged from 20 mHz to 50 mHz to remove the noise caused by continuous stirring. This smoothing analysis was performed using the EC-Lab software (Bio-Logic Science Instruments, Seyssinet-Pariset, France).

Conclusions
Owing to the benefits of bioelectrochemical applications and the low availability of DET-type enzymes in nature, oxidoreductases, which are capable of DET from the catalytic cofactor to the electrode, have been developed. Conventional methods for constructing DET-type enzymes are based on the recombinant expression of fusion proteins with an oxidoreductase and an electron transfer domain that harbors heme b or heme c. However, the application of this approach is sometimes restricted because of the issue related to the recombinant expression host, such as the difficulty of fusion protein expression as soluble and active molecules or the inherent problem with the post-translational modification pathway of cofactors. To overcome issues or limitations arising from the recombinant expression of fusion proteins, this study proposed an alternative and versatile method known as the in vitro protein complex formation technique to construct a DET-type enzyme complex facilitated by the SpyCatcher/SpyTag system.
Each CYTc-SC (as a heme c containing an electron transfer domain and ST-Enzymes as a catalytic domain) was separately prepared with its redox properties or enzymatic activities. CYTc-SC/ST-Enzyme complexes were formed by mixing, and the formed complex maintained its dehydrogenase activity. In addition, heme c in the complexed CYTc-SC functioned as an electron acceptor from the complexed enzyme via an internal electron transfer. Therefore, the CYTc-SC/ST-Enzyme complexes showed DET to the electrode compared to each ST-Enzyme alone. This novel method for creating DET-type enzyme complexes was demonstrated in three different oxidoreductases, indicating the versatility of this method.