Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2020 Oct 8;5(41):26807–26816. doi: 10.1021/acsomega.0c03911

Combined Approach to Engineer a Highly Active Mutant of Processive Chitinase Hydrolyzing Crystalline Chitin

Akasit Visootsat †,‡, Akihiko Nakamura §, Tak-Wai Wang ∥, Ryota Iino †,‡,*
PMCID: PMC7581260  PMID: 33111007

Abstract

graphic file with name ao0c03911_0007.jpg

Serratia marcescens chitinase A (SmChiA) processively hydrolyzes recalcitrant biomass crystalline chitin under mild conditions. Here, we combined multiple sequence alignment, site-saturation mutagenesis, and automated protein purification and activity measurement with liquid-handling robot to reduce the number of mutation trials and shorten the screening time for hydrolytic activity improvement of SmChiA. The amino acid residues, which are not conserved in the alignment and are close to the aromatic residues along the substrate-binding sites in the crystal structure, were selected for site-saturation mutagenesis. Using the previously identified highly active F232W/F396W mutant as a template, we identified the F232W/F396W/S538V mutant, which shows further improved hydrolytic activity just by trying eight different sites. Importantly, valine was not found in the multiple sequence alignment at Ser538 site of SmChiA. Our combined approach allows engineering of highly active enzyme mutants, which cannot be identified only by the introduction of predominant amino acid residues in the multiple sequence alignment.

Introduction

Chitin is a water-insoluble crystalline polysaccharide and a component of crustacean shell, insect exoskeleton, and the cell wall of many fungi.1−3 Chitin is an abundant biomass on earth next to cellulose, and it is expected as a useful material for industrial application. However, due to its physical and chemical stabilities, crystalline chitin decomposes only at high temperatures and high pressures.4 Processive chitinase is an enzyme that hydrolyzes crystalline chitin under milder conditions,5 at ambient temperature, and under atmospheric pressure. Therefore, as a tool converting crystalline chitin into useful chemicals, chitinase has been studied extensively. Processive chitinase is also known as a linear molecular motor that moves on crystalline chitin surface coupled with processive catalysis.6

The most studied processive chitinase is chitinase A from the Gram-negative bacterium Serratia marcescens (SmChiA). SmChiA is composed of two domains: a binding domain and a catalytic domain (Figure 1).7 Both the chitin-binding surface of the binding domain and the chitin-binding cleft of the catalytic domain have aromatic residues lined along them (Trp33, Trp69, Trp167, Tyr170, Phe232, Trp245, Trp275, Phe396, and Trp539, shown in red and orange in Figure 1). These aromatic residues play important roles in the chitin binding and processive catalysis of the SmChiA. It has been reported that alanine mutations of these aromatic residues (W33A, W69A, W167A, F232A, W245A, W275A, and F396A) result in a decrement of hydrolytic activity of SmChiA against the crystalline chitin.8−10 In the catalytic domain, there are also two essential catalytic residues (Asp313 and Glu315, shown in pink in Figure 1), which are highly conserved among the glycoside hydrolase 18 (GH18) family enzymes.11

Figure 1.

Figure 1

Open in a new tab

Model structure of SmChiA bound to crystalline chitin, important residues for binding and catalysis, and residues mutated in this study. (Top) A crystal structure of SmChiA (PDB entry 1CTN; ribbon model) aligned with the crystalline chitin chains (stick model). Amino acid residues responsible for binding to chitin (red and orange), catalytic residues (pink), and examined mutation sites (blue and green) are highlighted with sphere models. An expanded image around the catalytic site (transparent box in the left structure) is also shown in the right-hand-side box with 45° turn from the left structure. The cyan and yellow bars under the structure indicate the binding and catalytic domains, respectively. (Bottom) Result of multiple sequence alignment for amino acid residues highlighted in the top. The residues of SmChiA are shown in the same color as the top. The alignment result of the whole sequence is shown in Figure S1 of the Supporting Information.

Using single-molecule imaging analysis, we previously determined the kinetic parameters of elementary reaction steps and the chemomechanical coupling mechanism of SmChiA.12−14 Recently, we further revealed that the F232W and F396W mutations at the beginning and end of the substrate-binding catalytic cleft decrease the dissociation rate constant of productively bound enzyme and increase processivity, and result in high catalytic activity.14,15 Based on the multiple sequence alignment of the SmChiA and 257 SmChiA-like proteins, we also found that neither Phe232 nor Phe396 are predominant but tryptophan. Our previous study strongly suggested that amino acid sequence of SmChiA is not optimized for high hydrolytic activity during the process of evolution in the nature.

Site-saturation mutagenesis is a powerful approach that allows the generation of all possible mutations at the target site at once and widely used to improve enzyme thermostability, catalytic activity, and substrate specificity.16−23 Several methods such as the 22c-trick method,24 the Tang method,25 and the Max randomization26 have been developed nowadays to reduce the library size for the site-saturation mutagenesis. In these methods, each amino acid is encoded by one codon (or two codons for valine and leucine in the 22c-trick method). They can reduce the codon redundancy and bias, whereas various kinds of primers are required and can result in high cost as well as complications in primer design and PCR steps. In addition to the primer design methods described above, several molecular cloning techniques, such as QuickChange mutagenesis, In-Fusion,27 In vivo Assembly (IVA) cloning,28 Seamless Ligation Cloning Extract from Escherichia coli (SLiCE),29,30 and Gibson assembly,31,32 have also been developed to improve the mutagenesis procedure. These methods play an important role in simplifying and reducing the cloning processes.

Although the site-saturation mutagenesis is widely used, in general, only a few fractions of generated mutants are found to be improved as most of them show lower stabilities and/or activities than the wild type and low expressions and/or aggregations in the host cells. To overcome this drawback, the bioinformatics approach also plays an important role in mutation-site designation and in reducing the number of mutation trials. For instance, a new method known as PROSS, which uses advanced bioinformatics analyses to design high protein expression and stability, has been reported recently.33 This method can predict an optimal amino acid sequence of a target enzyme based on the multiple sequence alignment and calculation of the Gibbs free-energy difference (ΔΔG) between wild type and single-site mutant using the Rosetta software.34,35 However, it only suggests the amino acid mutations found in the multiple sequence alignment of natural enzymes and does not explore the possibility of other amino acids that do not appear in the alignment.

In the present study, we combined multiple sequence alignment, site-saturation mutagenesis, and automated screening with a liquid-handling robot to further improve the hydrolytic activity of SmChiA against crystalline chitin. For site-saturation mutagenesis, we selected amino acid residues that are not conserved in the multiple sequence alignment and are close to the aromatic residues along the substrate-binding sites in the crystal structure (Figure 1). Using the previously identified high-catalytic-activity mutant (F232W/F396W) as a template, we successfully identified S538V mutant among eight target sites. The F232W/F396W/S538V mutant showed higher hydrolytic activity than F232W/F396W at all crystalline chitin concentrations tried and showed a higher turnover number (kcat) than F232W/F396W. Importantly, valine was not found in the multiple sequence alignment at the Ser538 position of SmChiA. These results indicate that our combination method is an effective approach for engineering high-catalytic-activity mutants of enzymes.

Results and Discussion

To identify amino acid residues that are not conserved in the binding and catalytic domains of SmChiA, we obtained amino acid sequences of the 257 SmChiA-like proteins from the BLAST analysis and aligned them with that of the SmChiA by using Clustal Omega.36 We found that the aromatic residues at the substrate-binding site of the binding and catalytic domains are highly conserved among SmChiA and SmChiA-like proteins. SmChiA has predominant aromatic residues such as tryptophan and tyrosine (Figures 1 and S1, red), except for Phe232 and Phe396 (orange), as recently reported.14 Interestingly, Trp167, Tyr170, Trp275, and Trp539 in the catalytic cleft of SmChiA were completely conserved among the 258 proteins. Furthermore, as expected, two catalytic residues Asp313 and Glu315 (pink) were also completely conserved.

In addition to the conserved and nonconserved aromatic residues described above, we also found several nonconserved amino acid residues in both the binding and catalytic domains of SmChiA. We selected eight nonconserved residues close to the aromatic residues responsible for the chitin binding or close to the catalytic site as potential mutation targets (Figures 1 and S1, blue and green) because mutation in these sites would affect the orientation of the aromatic residues and interaction with the chitin and may result in higher hydrolytic activity. Asn70 is next to Trp69 in the binding domain; Ala238 is near Phe232 at the entrance of the catalytic cleft; Asp397 is next to Phe396 at the product release site; Ser551 is at the surface of the catalytic cleft; and Ser162, Thr441, Ile476, and Ser538 are inside the catalytic cleft. We also confirmed the suggestions by the PROSS for these target sites and compared them with our alignment (Table 1). Except for one site which was not suggested, the amino acid residues suggested by the PROSS were same with the predominant amino acids in the multiple sequence alignment. This result strongly suggests that only the multiple sequence alignment without the Rosetta energy calculation could already be enough for the SmChiA mutation-site designation.

Table 1. Comparison of Predominant Amino Acid Residues in the Multiple Sequence Alignment and Amino Acid Residues Suggested by the PROSS.

SmChiA position predominant amino acid in the alignment PROSS suggestion high-activity mutant
N70 S S  
S162 A A  
A238 P P  
D397 S a  
T441 V V  
I476 V V  
S538 A A Vb
S551 A A  
Open in a new tab
a

Not suggested by the PROSS for this amino acid residue.

b

Newly identified high-activity mutant in the present study.

Figure 2 shows an overview of our combined method of screening. It consists of preparation of site-saturated mutation library, expression in E. coli, protein purification with a liquid-handling robot, enzyme concentration estimation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), measurement of hydrolytic activity with the liquid-handling robot, and sequence analysis using colony PCR. The single cycle of screening for each mutation site can be completed in 5–6 days. To generate the site-saturated mutation library, we used NNB codon (where B = T/G/C) degenerate primers. The primers with degenerate codons were designed and obtained from a supplier that provides less-biased primers, as described previously.37 Since we aimed at covering all 20 amino acids, reducing the probability of stop codon (1/48 for NNB while 1/32 for NNK, where K = G/T),38 and simplifying the protocol as much as possible, the NNB codon degeneracy seemed to be the best way. In our previous study,37 we used SLiCE29,30 in ligation step as it is simple and less expensive. In the present study, we used the NEBuilder HiFi DNA assembly instead of SLiCE as it showed higher efficiency in the ligation step in the process of protocol optimization, although it is more expensive than SLiCE.

Figure 2.

Figure 2

Open in a new tab

Overview of experimental procedures for site-saturation mutagenesis and robot-based screening. See the Experimental Section for details. The images in this figure were obtained from the free domain.

For purification of SmChiA mutants from E. coli colonies and hydrolytic activity measurements of purified mutants, we used the liquid-handling robot, which handles 96-well plates. As we aimed at covering all 20 amino acids at each site in the single screening cycle, we first considered that 4 times the 20 amino acids or 80 colonies should be screened at least. As controls, we also used five wild type and five F232W/F396W mutant colonies in each cycle of screening. As results, in one 96-well plate, we picked 85 colonies expressing the mutants, five colonies of the wild type, five colonies of the F232W/F396W mutant, and one well was left blank. Note that our screening size (85 colonies) is consistent with the calculation of the TopLib analysis;38 the expected screening size of 87 colonies covers 95% probability to find the top one best mutant using NNB codon degeneracy.

After the protein purification by the liquid-handling robot, the purity of soluble protein was confirmed by SDS-PAGE (Figures 3 and S2–S8). Although we used one-step affinity purification of the SmChiA mutants with 6-histidines tag at the carboxyl terminal, the purity of the soluble protein was high. We also checked solubility of each mutant by the SDS-PAGE of insoluble proteins in cell pellets. We found that the mutations near the catalytic site (Ser538, Ser162, and Thr441) showed higher appearance of the full-length insoluble proteins compared to other positions (Figures 3A, S3A, and S6A). As these positions are located inside the catalytic cleft, the space for side-chain packing is limited. The mutation that resulted in a large amino acid side chain might affect its folding ability and cause aggregation.

Figure 3.

Figure 3

Open in a new tab

Screening result of S538 site (A) SDS-PAGE of purified protein (top, soluble) and cell pellet (bottom, pellet). The concentrations of the purified enzyme were estimated form the band intensity by comparing to that of the wild type (WT) purified separately. The arrows indicate the truncated mutants in which the stop codons were incorporated. The stop codon ratio was 4.7% and higher than the expected ratio (2.1%). The molecular mass of the red-colored marker is 75 kDa. (B) Hydrolytic activity was plotted against purified enzyme concentration for each sample. For hydrolytic activity measurement, enzyme concentrations were adjusted to 50 nM. The blue and orange dots represent the wild type and F232W/F396W, respectively. The gray and purple dots represent F232W/F396W/S538X mutants and their high-activity candidates, respectively. Crystalline chitin concentration was 0.5 mg mL–1. Note that the purified enzyme concentration estimation is deemed unreliable at a concentration below 0.5 μM (region indicated by the light blue box), and samples located in this region were excluded from further analysis. The average, standard deviation (SD), and relative standard deviation (RSD) of hydrolytic activity for WT and F232W/F396W are shown on the top.

Then, we also estimated the ratio of stop codon incorporation in the site-saturated mutation library. Note that because we added 6-histidines tag at the carboxyl terminal of the SmChiA, only the full-length proteins with single-site mutation will be purified in the soluble fraction. The truncated mutants generated by the introduction of the stop codon will be recovered in the insoluble fraction as aggregated proteins. As results, we obtained the stop codon ratios of 8.2% (A238X), 11.8% (D397X), 7.1% (T441X), 7.1% (I476X), 4.7% (S538X), and 5.9% (S551X). All of these values were larger than the expected value, 2.1% (1/48 × 100), for the NNB codon. Note that in the case of N70X and S162X, it was difficult to distinguish between the presence of the truncated proteins and the lack of expression as the expected sizes are small (5 and 14.5 kDa, respectively) (Figures S2 and S3). These results are inconsistent with our previous report that the stop codon ratio in the supplier we used was less than the expected value, suggesting that the bias changes depending on the order lot of the primers.37 Another possible reason for the high ratio of stop codon incorporation is PCR error because some truncated mutants showed the sizes different from the expected ones (D397X and T441X; Figures S5 and S6).

After SDS-PAGE, we estimated the purified protein concentration by comparing the band intensity with that of the wild type purified separately from the large-scale culture of E. coli. Then, the protein concentration of each sample was adjusted to 50 nM as the final concentration using the liquid-handling robot to measure hydrolytic activity quantitatively. For measurement, 0.5 mg mL–1 crystalline chitin was used as a substrate. The samples that showed higher hydrolytic activity than the F232W/F396W were selected as the positive candidates for further analysis. Among the eight positions we tried, we identified candidates of high-catalytic-activity mutants for three, Ser162, Ser538, and Ser551 (Figures 3, S3, and S8; purple dots). After the determination of amino acid sequence of the mutants, we conducted the experiments with 12 replications (12 colonies) of each candidate to confirm their hydrolytic activities (Figures 4 and S9).

Figure 4.

Figure 4

Open in a new tab

Reexamination of hydrolytic activity of F232W/F396W/S538X mutants. After amino acid sequence determination, each candidate mutant was cultured again on an agar plate and 12 colonies for each mutant were used to confirm their hydrolytic activities. The open circles represent the individual values for each colony, and the filled circles represent the average values. The F232W/F396W/S538V mutant (green) showed higher hydrolytic activity than F232W/F396W. The crystalline chitin concentration was 0.5 mg mL–1.

As results, for the Ser538 site, we confirmed that additional S538V mutation actually results in higher hydrolytic activity than F232W/F396W, whereas other candidates, S538A, S538G, and S538T, showed lower activity (Figure 4). Interestingly, although alanine is a predominant amino acid at this position, S538A mutant did not improve the activity. More importantly, valine was not found in the multiple sequence alignment at this position. These results indicate that our method enables us to generate non-natural high-activity enzymes, which cannot be identified only by the introduction of predominant amino acid in the multiple sequence alignment. For both the Ser162 and Ser551 sites, we found that the highest-activity mutants (S162A and S551A) just show comparable hydrolytic activities with F232W/F396W (Figure S9). For these positions, the best mutants were both alanine, a predominant amino acid in the multiple sequence alignment.

Next, we measured the hydrolytic activities of F232W/F396W/S538V at various crystalline chitin concentrations (0–6 mg mL–1) and compared them to those of wild type and F232W/F396W (Figure 5). F232W/F396W/S538V showed higher hydrolytic activities than F232W/F396W and wild type at all chitin concentrations tried. The kcat and Michaelis constant (Km) values were estimated by the fitting of the data points at 0–6 mg mL–1 (Figure 5A) or 0–1 mg mL–1 (Figure 5B) chitin concentration range with the Michaelis–Menten equation (Table 2). There were no significant differences between the values obtained by the fittings with these different chitin concentration ranges. Nevertheless, to compare our present results with those of our previous study under the same condition,14 here, we discuss using the values obtained by the fitting with 0–1 mg mL–1 chitin concentration range. The kcat and Km values for F232W/F396W and wild type were almost the same as those identified in our previous report.14 The kcat value for F232W/F396W/S538V was 4.6 ± 0.13 s–1, and 1.2 and 1.6 times higher than those for F232W/F396W (3.9 ± 0.08 s–1) and wild type (2.8 ± 0.12 s–1), respectively. The Km value for F232W/F396W/S538V was 0.14 ± 0.01 mg mL–1 and comparable to that for F232W/F396W (0.13 ± 0.01 mg mL–1) and 1.7 times lower than that for the wild type (0.24 ± 0.03 mg mL–1). These Km values indicate that additional S538V mutation does not change the binding affinity to the crystalline chitin compared to F232W/F396W.

Figure 5.

Figure 5

Open in a new tab

Crystalline chitin concentration dependence of hydrolytic activity. Hydrolytic activities of F232W/F396W/S538V (green) were compared to those of F232W/F396W (orange) and wild type (blue) at various crystalline chitin concentrations. The data points at 0–6 mg mL–1 (A) or 0–1 mg mL–1 (B) chitin concentration range were fitted with the Michaelis–Menten equation to estimate the values of kcat and Km (shown in Table 2). The error bars represent the standard deviations of the triplicate measurements.

Table 2. Turnover Number (kcat) and Michaelis Constant (Km) Estimated by Biochemical Activity Measurement.

  fitted with 0–6 mg mL–1 chitin concentration range
 fitted with 0–1 mg mL–1 chitin concentration range
  kcat (s–1)a Km (mg mL–1)a kcat (s–1)a Km (mg mL–1)a
F232W/F396W/S538V 4.4 ± 0.07 0.12 ± 0.01 4.6 ± 0.13 0.14 ± 0.01
F232W/F396W 3.7 ± 0.05 0.12 ± 0.01 3.9 ± 0.08 0.13 ± 0.01
wild type 2.7 ± 0.04 0.24 ± 0.01 2.8 ± 0.12 0.24 ± 0.03
Open in a new tab
a

The kcat and Km values were determined by the fitting with the Michaelis–Menten equation. The errors shown in the table are the fitting errors.

In our screening, to measure the hydrolytic activity of SmChiA, we used a relatively short reaction time of 30 min and the ferricyanide assay, which estimates water-soluble reducing ends of the sugars. Then, we confirmed reaction time dependence of the hydrolytic activity (Figure S10). For all reaction times tried (30 min, 1 h, 2 h, 4 h, and 8 h), F232W/F396W/S538V showed higher hydrolytic activities than wild type and F232W/F396W. This result indicates that F232W/F396W/S538V retains higher activity at not only the initial but also the later stages of the chitin hydrolysis reaction. We also confirmed the product profile using size exclusion chromatography (Figure S11). After the reaction for 8 h, the product profiles for wild type, F232W/F396W, and F232W/F396W/S538V were the same. For all samples, the major product was N,N′-diacetyl chitobiose, and N-acetyl glucosamine was not detected. This result clearly indicates that the F232W/F396W/S538V and F232W/F396W do not have chitobiase activity as well as wild type, and high hydrolytic activity of F232W/F396W/S538V comes from high production of the N,N′-diacetyl chitobiose.

The Ser538 of SmChiA is located next to Trp539, an aromatic residue inside the catalytic cleft important for substrate binding, and also near Asp313 and Glu315, the catalytic residues (Figure 1). Ser538 is also included in one of the highly conserved motifs (GXXXWXXDXDD) of the GH18 family enzymes,39−41 although Ser538 is not a conserved residue in this motif. Since serine was mutated to valine in S538V, the number of surrounding hydrogen bonds could be reduced due to the removal of the hydroxyl group from the side chain. On the other hand, hydrophobicity could be increased through the introduction of two methyl groups, although the side chain of Ser538 is not facing to the substrate. Structural analysis with X-ray crystallography will be helpful to understand how the S538V mutation changes the structure of SmChiA and interaction with the substrate at the atomic level.

Using single-molecule imaging analysis with fluorescence microscopy and high-speed atomic force microscopy, we have previously revealed that F232W/F396W shows a higher kcat than the wild type because of the lower dissociation rate constant of productively bound enzyme and the higher processivity than the wild type.14 In the present study, we found that F232W/F396W/S538V shows higher kcat than F232W/F396W. Similar single-molecule analysis will be helpful to understand the mechanism of the F232W/F396W/S538V mutant. Considering the location of the S538 and the reaction scheme of SmChiA we proposed previously,14 we expect that additional S538V mutation would increase the rate constant of the processive catalysis (kpc) we described previously.14 In the previous study, the value of kpc was estimated from translational velocity measured by single-molecule imaging and size of the product (1.04 nm). In our model of the processive catalysis of SmChiA, kpc includes several elementary reaction steps such as decrystallization of single chitin chain from the crystal surface, sliding of decrystallized chitin chain into the catalytic cleft, substrate-assisted cleavage of the glycosidic bond, and release of the product.12 To understand the more detailed mechanism of the F232W/F396W/S538V mutant, these elementary reaction steps should be resolved using high-precision and high-speed single-molecule imaging analysis probed with a gold nanoparticle probe.12

In summary, as demonstrated in the present study, our combination method allows engineering of a high-catalytic-activity mutant of the processive chitinase within a few trial residues. Our method will be applicable to other enzymes, although several points could be further improved. For instance, in the current procedure, we used the BugBuster reagent for cell disruption. However, a recent report showed that the BugBuster reagent can cause a high level of contamination of other proteins and decrease the thermostability of the cell extract solutions.42 In that study, hypotonic extraction is suggested to be used instead of the BugBuster or other detergent containing chemical lysis reagents. In the near future, we would like to apply further improved combined approach to engineer other enzymes such as processive cellulases43 and the PETase,44 which directly decomposes recalcitrant solid polymer substrates.

Experimental Section

Multiple Sequence Alignment

The multiple sequence alignment was performed as described previously.14 Briefly, amino acid sequences of SmChiA and 257 SmChiA-like proteins from the different organisms were obtained using the NCBI database protein BLAST tool (http://blast.ncbi.nlm.nih.gov). Before alignment, signal sequences were removed according to the prediction of SignalP 5.045 (http://www.cbs.dtu.dk/services/ SignalP), with the appropriate organism group for prediction. The sequences for which SignalP could not predict the signal sequences were excluded from the alignment. The amino acid sequences of SmChiA and 257 SmChiA-like proteins were aligned and visualized using Clustal Omega36 (Clustalω: http://www.ebi.ac.uk/Tools/msa/clustalo) and WebLogo46 (http://weblogo.threeplusone.com), respectively.

Site-Saturation Mutagenesis

The primers were designed according to our previous report.37 The PCR reactions were carried out using the KOD One PCR Master Mix (TOYOBO, Japan) as follows: 25 μL of KOD One premix, 22 μL of Milli-Q water, 1 μL of each 10 pmol μL–1 forward and reverse primers, and 1 μL of 1 ng μL–1 DNA template plasmids (SmChiA_F232W/F396W with the C-terminal Factor Xa (FaXa) recognition sequence and 6-histidines tag and with signal peptide (23 amino acid residues) at the N-terminal in pET27b). A LifeECO thermocycler (Bioer Technology, China) was used with the following protocol: 30 s at 98 °C of initial denaturation followed by 25 cycles at 98 °C for 10 s, at 55 °C for 5 s, and at 72 °C for 1 min, finalized by a further extension cycle at 72 °C for 2 min. The PCR products were treated with 2 μL of DpnI (New England Biolabs, NEB) for 40 min at 37 °C to eliminate the leftover DNA template. All DpnI-treated PCR products were loaded for electrophoresis on 1% agarose gel. The products of the expected size were extracted and purified using a commercial gel extraction kit (Promega).

The purified DNA fragments were ligated using the NEBuilder HiFi DNA assembly (NEB). The DNA fragments and NEBuilder reagent were mixed at a 1:1 volume ratio (the final DNA concentration was 20 ng μL–1), then incubated at 50 °C for 1 h. One microliter of the ligation reaction was added to 50 μL of Tuner DE3 E. coli-competent cells and put on ice for 1 min before pulsing with an electroporator (Micropulser, Bio-Rad). Then, 200 μL of SOC medium was added to the suspension of transformed cells immediately after the electroporation, and the cells were incubated at 37 °C for 1 h before spreading on LB-Agar plates containing 25 μg mL–1 kanamycin. The plates were incubated overnight (approximately 16 h) at 37 °C.

Small-Scale Culture and Enzyme Purification

Eighty-five colonies of freshly transformed E. coli were picked and cultured in 1 mL of Super Broth (SB) medium (3.2% Bacto tryptone, 2% yeast extract, and 0.5% NaCl) supplemented with 25 μg mL–1 kanamycin in 96-deep-well plates (P-2ML-SQ-C, Axygen). Prior to inoculation into the medium described above, all colonies were also inoculated on an LB-Agar plate containing 25 μg mL–1 kanamycin to prepare a master plate. Five colonies expressing the wild type and five colonies expressing the F232W/F396W were also picked as the controls. The last well was left blank to check contamination across the well. The picked cells were cultured at 30 °C for 12 h by shaking at 1300 rpm. The cells were cooled down on iced water for 10 min before being induced with 0.4 mM isopropyl β-D-1-thiogalactopyranoside (IPTG). Then, the cells were cultured by shaking at 1300 rpm at 20 °C for another 12 h and harvested by centrifugation at 4400 g, at 25 °C for 10 min. The supernatants were discarded.

The cell disruption and protein purification were performed using a Beckman Coulter Biomek 4000 liquid-handling robot (USA). Briefly, 300 μL of the BugBuster reagent (Novagen, Germany) supplemented with 10 units mL–1 of Benzonase (Novagen, Germany) was added to each well to disrupt the cells. Then, the 96-deep-well plate was shaken at 1000 rpm for 20 min at 25 °C. The disrupted cells were then centrifuged at 4400 g at 25 °C for 10 min. The supernatant was transferred into a new 96-deep-well plate, then 100 μL of Ni-NTA agarose (QIAGEN, Germany) was added to each well and mixed by shaking at 1000 rpm for 5 min to facilitate binding of SmChiA mutants with 6-histidines tag. The Ni-NTA mixtures were then transferred to a filter plate (AcroPrep Advance Filter Plates for Aqueous Filtration—30–40 μm PP/PE nonwoven media, PALL) on the collection plate. The liquid was filtered through a filter plate in vacuum. The Ni-NTA agarose was washed twice with 200 μL of 50 mM sodium phosphate pH 7.0 with 100 mM sodium chloride and washed three times with 200 μL of 50 mM sodium phosphate pH 7.0 with 100 mM sodium chloride and 50 mM imidazole. Before elution, the bottom of the filter plate was wiped with a paper and the collection plate was replaced with the new 96-deep-well plate (P-2ML-SQ-C, Axygen). Then, the samples were eluted with 50 mM sodium phosphate pH 7.0 with 100 mM sodium chloride and 100 mM imidazole.

SDS-PAGE, Protein Concentration Estimation

Ten microliters of purified samples from each well was mixed with 5 μL of the sample buffer (62.5 mM Tris-HCl pH 6.8, 2.5% SDS, 0.002% bromphenol blue, 5% β-mercaptoethanol, and 10% glycerol). The cell pellets were shaken in the remaining BugBuster reagent at 1300 rpm for 20 min to resuspend. Two microliters of cell pellet suspension were mixed with 5 μL of the sample buffer. Both purified samples and cell pellets suspension mixtures were boiled at 95 °C for 5 min before loading them on 12% acrylamide gel. One or two micromolar wild-type SmChiA, purified separately from the large-scale culture, was loaded in four lanes of each gel as references for concentration determination.

After SDS-PAGE, gels were stained with a staining solution (100 mg of Coomassie Brilliant Blue G-250 (Tokyo Chemical Industry, Japan), 3 mL of hydrochloric acid, and 1 L of Milli-Q water) and then destained with clean water overnight to reduce the background level. Then, images of the gel were taken using a gel documentation system (WSE-5400 Printgraph Classic, ATTO, Japan), and target protein concentrations were determined using an image analysis software (CS Analyzer 4, ATTO, Japan). The band intensities were analyzed and compared to the wild-type SmChiA references to estimate the concentration of each sample.

Chitin Hydrolytic Activity Measurement of Each Mutant

For chitin hydrolytic activity measurement, protein concentration of each sample was diluted to 100 nM with 100 mM sodium phosphate pH 6.0 in a 96-well low-protein-binding plate (PCR plate 96 LoBind-semi-skirted, Eppendorf, Germany) using the same liquid-handling robot used for cell disruption and protein purification. The hydrolytic activity was also measured using the same liquid-handling robot. Seventy-five microliters of 100 nM enzymes in sodium phosphate pH 6.0 was mixed with 75 μL of 1 mg mL–1 crystalline chitin suspension (highly crystalline β-chitin prepared from the tube of a tubeworm Lamellibrachia satsuma, as described previously6) in a 96-well reaction plate (RTP 6001-20 Bio Chromato, Inc., Japan). Then, the plate was incubated at 25 °C for 30 min without shaking and the reactions were stopped with 200 μL of Schales’ reagent (500 mM sodium carbonate, 1.5 mM potassium ferricyanide). The crystalline chitin was separated on the 96-well 1.2 μm hydrophilic low-protein-binding Durapore membrane filter plate (Merck Millipore, Germany). The filtered solutions were transferred to a PCR-96-well plate (PCR-96-PE2, BMBio, Japan), then incubated at 95 °C for 15 min. Then, 100 μL of the heated samples was transferred into a 384-well clear plate (784061, Greiner Bio-One). The absorbance at 420 nm was measured using a multimode microplate reader (SpectraID3, Molecular Devices). The concentration of reaction products was calculated from the standard curve prepared with N,N′-diacetyl chitobiose. The hydrolytic activities of each sample at each mutation site were plotted against their concentrations and compared to those of the wild type and F232W/F396W. To assure the reliability of the measurement, the samples which showed a purified protein concentration below 0.5 μM were excluded from further analysis. To confirm reproducibility of the measurement, average values and standard deviations of the hydrolytic activities of the wild type and F232W/F396W were determined for each screening. The relative values of the standard deviations (RSD) were less than 15% of the average values.

High-activity mutant candidates were picked from the master plate for colony PCR using T7-promoter and T7-terminator primers. The colony PCR was carried out with KOD One PCR Master Mix (TOYOBO, Japan). The mixture consisted of 25 μL of KOD One premix, 23 μL of Milli-Q water, 1 μL of each 10 pmol μL–1 T7-promoter and T7-terminator primers, and the designated colonies were picked into the mixture. Colony PCR products were purified using a commercial PCR clean-up system (Promega), and the purified DNA sequences were then examined by FASMAC Co., Ltd (Japan).

The hydrolytic activities of the candidates with verified sequences were confirmed by repeating the cultivation, purification, and activity measurements. After the transformation of the plasmid containing each candidate mutant, 12 colonies were picked for each mutant and cultured in the 96-deep-well plate, and used for subsequent purifications and activity measurements using the liquid-handling robot.

Large-Scale Culture and Enzyme Purification

The verified plasmid containing the wild type, F232W/F396W, or F232W/F396W/S538V gene was transformed into Tuner (DE3) competent cells, as described above. Single colonies were inoculated into 10 mL of LB medium supplemented with 25 μg mL–1 kanamycin and incubated overnight at 37 °C and 250 rpm. Then, all of the preculture was added to 1 L of LB medium supplemented with 25 μg mL–1 kanamycin in a 3 L flask and cultured at 37 °C and 130 rpm until O.D.600 = 1.8. Then, the media were cooled on iced water for 10 min before being induced with 500 μM IPTG. The cells were further cultured at 20 °C and 130 rpm overnight. The culture was then centrifuged at 6000 g for 10 min. Ten times volume of the cell weight of 50 mM sodium phosphate pH 7.0 with 100 mM sodium chloride was added and supplemented with protease inhibitor cocktail (cOmplete Mini, EDTA-free, Roche Applied Science, Switzerland). The cell suspension was sonicated on ice for 20 min at 3 s intervals. The disrupted cells were centrifuged at 4 °C and 30 000 g for 20 min. The supernatant was collected, incubated with Ni-NTA agarose (QIAGEN), and equilibrated with 50 mM sodium phosphate pH 7.0 containing 100 mM sodium chloride for 15 min at 25 °C under gentle rotation. The Ni-NTA agarose was packed into an open column and washed with 50 mM sodium phosphate pH 7.0 containing 100 mM sodium chloride and then with 50 mM imidazole in 50 mM sodium phosphate pH 7.0 containing 100 mM sodium chloride. Then, the enzyme was eluted with 100 mM imidazole in 50 mM sodium phosphate pH 7.0 containing 100 mM sodium chloride. The designated eluted fractions were mixed and concentrated to 500 μL using a 30 000 molecular weight cutoff Vivaspin Turbo 50 (Sartorius). To further purify and remove the imidazole in solution, the samples were then injected into a Superdex 200 10/300 GL column (GE Healthcare) and eluted with 50 mM sodium phosphate pH 7.0 containing 100 mM sodium chloride at a flow rate of 0.5 mL min–1. The eluted fractions were mixed, and their concentration was estimated from their absorbance at 280 nm (molar extinction coefficient of 107 050 M–1 cm–1 was used for the wild type, and that of 118 050 M–1 cm–1 was used for both F232W/F396W and F232W/F396W/S538V). The molar extinction coefficients were calculated from the ProtParam in the ExPASy bioinformatics resource portal web service. The samples were then divided into small fractions and stored at −80 °C until further use.

Hydrolytic Activity Measurement to Estimate kcat and Km

The same liquid-handling robot was used to measure the hydrolytic activities of wild type, F232W/F396W, and F232W/F396W/S538V at various concentrations of the crystalline chitin to estimate kcat and Km. The samples were measured in triplicate simultaneously in 96-well plates. The purified enzymes were diluted to 100 nM with 100 mM sodium phosphate pH 6.0 in a low-protein-binding microtube (MS-4215M Sumitomo Bakelite, Japan). In 96-well reaction plates, the diluted enzymes were incubated with crystalline chitin (0-6 mg mL–1) at 25 °C for 30 min in a reaction mixture with the volume of 150 μL (75 μL of enzyme solution and 75 μL of crystalline chitin suspension were mixed) without shaking. The reactions were stopped with 200 μL of the Schales’ reagent. Crystalline chitin was separated on the 96-well 1.2 μm Hydrophil low-protein-binding Durapore membrane filter plate (Merck Millipore). The filtered solution was incubated at 95 °C for 15 min, and 100 μL of the samples were transferred to a 384-well clear plate. Absorbance at 420 nm was measured using a multimode microplate reader (SpectraMax iD3, Molecular Devices). The concentration of reaction products was calculated from the standard curve prepared with N,N′-diacetyl chitobiose. The error bars shown in Figure 5 represent the standard deviations of the triplicated measurements.

Reaction Time Dependence of Hydrolytic Activity

The hydrolytic activities of wild type, F232W/F396W, and F232W/F396W/S538V were measured at various reaction times (30 min, 1 h, 2 h, 4 h, and 8 h). For each condition, the samples were measured in triplicate. The purified enzymes were diluted to 100 nM in 100 mM sodium phosphate pH 6.0 and then mixed with 1 mg mL–1 crystalline chitin in the 400 μL reaction mixture volume (200 μL of enzyme solution and 200 μL of crystalline chitin suspension). The reaction mixture was incubated at 25 °C without shaking for 30 min, 1 h, 2 h, 4 h, and 8 h. The reaction mixture was then centrifuged at 16 000 g for 10 min at 25 °C. Then, 75 μL of supernatant was mixed with 200 μL of the Schales’ reagent. Two hundred microliters of supernatant was stored at −80 °C for the analysis with size exclusion chromatography described below. The supernatant mixed with the Schales’ reagent was then incubated at 95 °C for 15 min, and 100 μL of the samples were transferred to a 384-well clear plate. Absorbance at 420 nm was measured using a multimode microplate reader (SpectraMax iD3, Molecular Devices). The concentration of reaction products was calculated from the standard curve prepared with N,N′-diacetyl chitobiose. The concentration of reaction product and hydrolytic activity were then plotted against the reaction time. The error bars represent the standard deviations of the triplicated measurements.

Size Exclusion Chromatography of Hydrolytic Products from Crystalline Chitin

The oligosaccharide products obtained by the hydrolysis reaction were analyzed using size exclusion chromatography. Sodium chloride (4 M, 2.5 μL) was added to 97.5 μL of the supernatant obtained after 8 h hydrolysis reaction. Then, the mixture was injected to the tandemly connected two YMC-Pack Diol-60 columns (YMC CO., Ltd., Japan). The mobile phase was 50 mM sodium phosphate pH 6.0 with 100 mM sodium chloride, and the flow rate was 1 mL min–1. The oligosaccharide products were detected with the UV detector at 210 nm wavelength. N-acetyl glucosamine and N,N′-diacetyl chitobiose (final concentrations of 1 mM, respectively) were used as controls to determine the peak positions (elution times) in the chromatogram.

Acknowledgments

The authors thank the Japan Agency for Marine Earth Science and Technology (JAMSTEC) for providing the L. satsuma tubes for crystalline chitin preparation. They also thank Yasuko Okuni and Mayuko Yamamoto for their technical help and advice in sample preparation. Thanks also go to all of the lab members for their kind advice and discussion. This work was partially supported by Grants-in-Aid for Scientific Research on Innovative Areas “Molecular Engine” (JP18H05424 to R.I.) and Grants-in-Aid for Scientific Research (JP18H02418, JP18H04755, and JP17K19213 to R.I.) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.

Glossary

Abbreviations

SmChiA

Serratia marcescens chitinase A

WT

wild type

LB

Luria broth

Ni-NTA

nitrilotriacetic acid

PDB

Protein Data Bank

IPTG

isopropyl β-d-1-thiogalactopyranoside

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c03911.

  • Multiple sequence alignment of SmChiA and 257 SmChiA-like proteins (Figure S1); screening result of N70 site (Figure S2); screening result of S162 site (Figure S3); screening result of A238 site (Figure S4); screening result of D397 site (Figure S5); screening result of T441 site (Figure S6); screening result of I476 site (Figure S7); screening result of S551 site (Figure S8); reexamination of hydrolytic activity of F232W/F396W/S162X and F232W/F396W/S551X mutants (Figure S9); reaction time dependence of hydrolytic activity and product concentration (Figure S10); and size exclusion chromatography of product after 8 h reaction (Figure S11) (PDF)

Author Contributions

A.V., A.N., and R.I. designed the experiments. A.V. and A.N. performed the bioinformatics analysis. A.V. and T.-W.W. performed site-saturation mutagenesis and screening. A.V. performed the biochemical analysis. R.I. supervised and coordinated the project and wrote the manuscript with A.V.

The authors declare no competing financial interest.

Supplementary Material

ao0c03911_si_001.pdf (1.7MB, pdf)

References

  1. Merzendorfer H.; Zimoch L. Chitin metabolism in insects: structure, function and regulation of chitin synthases and chitinases. J. Exp. Biol. 2003, 206, 4393–4412. 10.1242/jeb.00709. [DOI] [PubMed] [Google Scholar]
  2. Hamed I.; Özogul F.; Regenstein J. M. Industrial applications of crustacean by-products (chitin, chitosan, and chitooligosaccharides): A review. Trends Food Sci. Technol. 2016, 48, 40–50. 10.1016/j.tifs.2015.11.007. [DOI] [Google Scholar]
  3. Langner T.; Gohre V. Fungal chitinases: function, regulation, and potential roles in plant/pathogen interactions. Curr. Genet. 2016, 62, 243–254. 10.1007/s00294-015-0530-x. [DOI] [PubMed] [Google Scholar]
  4. Deguchi S.; Tsujii K.; Horikoshi K. In situ microscopic observation of chitin and fungal cells with chitinous cell walls in hydrothermal conditions. Sci. Rep. 2015, 5, 11907 10.1038/srep11907. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Einbu A.; Varum K. M. Characterization of chitin and its hydrolysis to GlcNAc and GlcN. Biomacromolecules 2008, 9, 1870–1875. 10.1021/bm8001123. [DOI] [PubMed] [Google Scholar]
  6. Igarashi K.; Uchihashi T.; Uchiyama T.; Sugimoto H.; Wada M.; Suzuki K.; Sakuda S.; Ando T.; Watanabe T.; Samejima M. Two-way traffic of glycoside hydrolase family 18 processive chitinases on crystalline chitin. Nat. Commun. 2014, 5, 3975 10.1038/ncomms4975. [DOI] [PubMed] [Google Scholar]
  7. Perrakis A.; Tews I.; Dauter Z.; Oppenheim A. B.; Chet I.; Wilson K. S.; Vorgias C. E. Crystal structure of a bacterial chitinase at 2.3 A resolution. Structure 1994, 2, 1169–80. 10.1016/S0969-2126(94)00119-7. [DOI] [PubMed] [Google Scholar]
  8. Uchiyama T.; Katouno F.; Nikaidou N.; Nonaka T.; Sugiyama J.; Watanabe T. Roles of the exposed aromatic residues in crystalline chitin hydrolysis by chitinase A from Serratia marcescens 2170. J. Biol. Chem. 2001, 276, 41343–41349. 10.1074/jbc.M103610200. [DOI] [PubMed] [Google Scholar]
  9. Zakariassen H.; Aam B. B.; Horn S. J.; Varum K. M.; Sorlie M.; Eijsink V. G. Aromatic residues in the catalytic center of chitinase A from Serratia marcescens affect processivity, enzyme activity, and biomass converting efficiency. J. Biol. Chem. 2009, 284, 10610–10617. 10.1074/jbc.M900092200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kurašin M.; Kuusk S.; Kuusk P.; Sorlie M.; Valjamae P. Slow Off-rates and Strong Product Binding Are Required for Processivity and Efficient Degradation of Recalcitrant Chitin by Family 18 Chitinases. J. Biol. Chem. 2015, 290, 29074–29085. 10.1074/jbc.M115.684977. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Papanikolau Y.; Prag G.; Tavlas G.; Vorgias C. E.; Oppenheim A. B.; Petratos K. High resolution structural analyses of mutant chitinase A complexes with substrates provide new insight into the mechanism of catalysis. Biochemistry 2001, 40, 11338–11343. 10.1021/bi010505h. [DOI] [PubMed] [Google Scholar]
  12. Nakamura A.; Okazaki K. I.; Furuta T.; Sakurai M.; Iino R. Processive chitinase is Brownian monorail operated by fast catalysis after peeling rail from crystalline chitin. Nat. Commun. 2018, 9, 3814 10.1038/s41467-018-06362-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Nakamura A.; Tasaki T.; Okuni Y.; Song C.; Murata K.; Kozai T.; Hara M.; Sugimoto H.; Suzuki K.; Watanabe T.; Uchihashi T.; Noji H.; Iino R. Rate constants, processivity, and productive binding ratio of chitinase A revealed by single-molecule analysis. Phys. Chem. Chem. Phys. 2018, 20, 3010–3018. 10.1039/C7CP04606E. [DOI] [PubMed] [Google Scholar]
  14. Visootsat A.; Nakamura A.; Vignon P.; Watanabe H.; Uchihashi T.; Iino R. Single-molecule imaging analysis reveals the mechanism of a high-catalytic-activity mutant of chitinase A from Serratia marcescens. J. Biol. Chem. 2020, 295, 1915–1925. 10.1074/jbc.RA119.012078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Liu T.; Chen L.; Zhou Y.; Jiang X.; Duan Y.; Yang Q. Structure, Catalysis, and Inhibition of OfChi-h, the Lepidoptera-exclusive Insect Chitinase. J. Biol. Chem. 2017, 292, 2080–2088. 10.1074/jbc.M116.755330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Buettner K.; Hertel T. C.; Pietzsch M. Increased thermostability of microbial transglutaminase by combination of several hot spots evolved by random and saturation mutagenesis. Amino Acids 2012, 42, 987–996. 10.1007/s00726-011-1015-y. [DOI] [PubMed] [Google Scholar]
  17. Reetz M. T. Biocatalysis in organic chemistry and biotechnology: past, present, and future. J. Am. Chem. Soc. 2013, 135, 12480–12496. 10.1021/ja405051f. [DOI] [PubMed] [Google Scholar]
  18. Zhang X. F.; Yang G. Y.; Zhang Y.; Xie Y.; Withers S. G.; Feng Y. A general and efficient strategy for generating the stable enzymes. Sci. Rep. 2016, 6, 33797 10.1038/srep33797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Yin H.; Pijning T.; Meng X.; Dijkhuizen L.; van Leeuwen S. S. Engineering of the Bacillus circulans beta-Galactosidase Product Specificity. Biochemistry 2017, 56, 704–711. 10.1021/acs.biochem.7b00032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Wu N. C.; Grande G.; Turner H. L.; Ward A. B.; Xie J.; Lerner R. A.; Wilson I. A. In vitro evolution of an influenza broadly neutralizing antibody is modulated by hemagglutinin receptor specificity. Nat. Commun. 2017, 8, 15371 10.1038/ncomms15371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jiao L.; Liu Y.; Zhang X.; Liu B.; Zhang C.; Liu X. Site-saturation mutagenesis library construction and screening for specific broad-spectrum single-domain antibodies against multiple Cry1 toxins. Appl. Microbiol. Biotechnol. 2017, 101, 6071–6082. 10.1007/s00253-017-8347-9. [DOI] [PubMed] [Google Scholar]
  22. Sayous V.; Lubrano P.; Li Y.; Acevedo-Rocha C. G. Unbiased libraries in protein directed evolution. Biochim. Biophys. Acta Proteins Proteomics 2020, 1868, 140321 10.1016/j.bbapap.2019.140321. [DOI] [PubMed] [Google Scholar]
  23. Qu G.; Li A.; Acevedo-Rocha C. G.; Sun Z.; Reetz M. T. The Crucial Role of Methodology Development in Directed Evolution of Selective Enzymes. Angew. Chem., Int. Ed. 2020, 59, 13204–13231. 10.1002/anie.201901491. [DOI] [PubMed] [Google Scholar]
  24. Kille S.; Acevedo-Rocha C. G.; Parra L. P.; Zhang Z. G.; Opperman D. J.; Reetz M. T.; Acevedo J. P. Reducing codon redundancy and screening effort of combinatorial protein libraries created by saturation mutagenesis. ACS Synth. Biol. 2013, 2, 83–92. 10.1021/sb300037w. [DOI] [PubMed] [Google Scholar]
  25. Tang L.; Gao H.; Zhu X.; Wang X.; Zhou M.; Jiang R. Construction of “small-intelligent” focused mutagenesis libraries using well-designed combinatorial degenerate primers. Biotechniques 2012, 52, 149–158. 10.2144/000113820. [DOI] [PubMed] [Google Scholar]
  26. Hughes M. D.; Nagel D. A.; Santos A. F.; Sutherland A. J.; Hine A. V. Removing the redundancy from randomised gene libraries. J. Mol. Biol. 2003, 331, 973–979. 10.1016/S0022-2836(03)00833-7. [DOI] [PubMed] [Google Scholar]
  27. Zhu B.; Cai G.; Hall E. O.; Freeman G. J. In-fusion assembly: seamless engineering of multidomain fusion proteins, modular vectors, and mutations. Biotechniques 2007, 43, 354–359. 10.2144/000112536. [DOI] [PubMed] [Google Scholar]
  28. García-Nafría J.; Watson J. F.; Greger I. H. IVA cloning: A single-tube universal cloning system exploiting bacterial In Vivo Assembly. Sci. Rep. 2016, 6, 27459 10.1038/srep27459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Okegawa Y.; Motohashi K. A simple and ultra-low cost homemade seamless ligation cloning extract (SLiCE) as an alternative to a commercially available seamless DNA cloning kit. Biochem. Biophys. Rep. 2015, 4, 148–151. 10.1016/j.bbrep.2015.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Okegawa Y.; Motohashi K. Evaluation of seamless ligation cloning extract preparation methods from an Escherichia coli laboratory strain. Anal. Biochem. 2015, 486, 51–53. 10.1016/j.ab.2015.06.031. [DOI] [PubMed] [Google Scholar]
  31. Gibson D. G.; Young L.; Chuang R. Y.; Venter J. C.; Hutchison C. A. 3rd; Smith H. O. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods 2009, 6, 343–345. 10.1038/nmeth.1318. [DOI] [PubMed] [Google Scholar]
  32. Gibson D. G.; Glass J. I.; Lartigue C.; Noskov V. N.; Chuang R. Y.; Algire M. A.; Benders G. A.; Montague M. G.; Ma L.; Moodie M. M.; Merryman C.; Vashee S.; Krishnakumar R.; Assad-Garcia N.; Andrews-Pfannkoch C.; Denisova E. A.; Young L.; Qi Z. Q.; Segall-Shapiro T. H.; Calvey C. H.; Parmar P. P.; Hutchison C. A. 3rd; Smith H. O.; Venter J. C. Creation of a bacterial cell controlled by a chemically synthesized genome. Science 2010, 329, 52–56. 10.1126/science.1190719. [DOI] [PubMed] [Google Scholar]
  33. Goldenzweig A.; Goldsmith M.; Hill S. E.; Gertman O.; Laurino P.; Ashani Y.; Dym O.; Unger T.; Albeck S.; Prilusky J.; Lieberman R. L.; Aharoni A.; Silman I.; Sussman J. L.; Tawfik D. S.; Fleishman S. J. Automated Structure- and Sequence-Based Design of Proteins for High Bacterial Expression and Stability. Mol. Cell 2018, 70, 380. 10.1016/j.molcel.2018.03.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Fleishman S. J.; Leaver-Fay A.; Corn J. E.; Strauch E. M.; Khare S. D.; Koga N.; Ashworth J.; Murphy P.; Richter F.; Lemmon G.; Meiler J.; Baker D. RosettaScripts: a scripting language interface to the Rosetta macromolecular modeling suite. PLoS One 2011, 6, e20161 10.1371/journal.pone.0020161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Whitehead T. A.; Chevalier A.; Song Y.; Dreyfus C.; Fleishman S. J.; De Mattos C.; Myers C. A.; Kamisetty H.; Blair P.; Wilson I. A.; Baker D. Optimization of affinity, specificity and function of designed influenza inhibitors using deep sequencing. Nat. Biotechnol. 2012, 30, 543–548. 10.1038/nbt.2214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Li W.; Cowley A.; Uludag M.; Gur T.; McWilliam H.; Squizzato S.; Park Y. M.; Buso N.; Lopez R. The EMBL-EBI bioinformatics web and programmatic tools framework. Nucleic Acids Res. 2015, 43, W580–4. 10.1093/nar/gkv279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Kawai F.; Nakamura A.; Visootsat A.; Iino R. Plasmid-Based One-Pot Saturation Mutagenesis and Robot-Based Automated Screening for Protein Engineering. ACS Omega 2018, 3, 7715–7726. 10.1021/acsomega.8b00663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Nov Y. When second best is good enough: another probabilistic look at saturation mutagenesis. Appl. Environ. Microbiol. 2012, 78, 258–262. 10.1128/AEM.06265-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Chen W.; Qu M.; Zhou Y.; Yang Q. Structural analysis of group II chitinase (ChtII) catalysis completes the puzzle of chitin hydrolysis in insects. J. Biol. Chem. 2018, 293, 2652–2660. 10.1074/jbc.RA117.000119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Arakane Y.; Muthukrishnan S. Insect chitinase and chitinase-like proteins. Cell Mol. Life Sci. 2010, 67, 201–16. 10.1007/s00018-009-0161-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Muthukrishnan S.; Merzendorfer H.; Arakane Y.; Yang Q. Chitin Organizing and Modifying Enzymes and Proteins Involved In Remodeling of the Insect Cuticle. Adv. Exp. Med. Biol. 2019, 1142, 83–114. [DOI] [PubMed] [Google Scholar]
  42. Magnusson A. O.; Szekrenyi A.; Joosten H. J.; Finnigan J.; Charnock S.; Fessner W. D. nanoDSF as screening tool for enzyme libraries and biotechnology development. FEBS J. 2019, 286, 184–204. 10.1111/febs.14696. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Payne C. M.; Knott B. C.; Mayes H. B.; Hansson H.; Himmel M. E.; Sandgren M.; Stahlberg J.; Beckham G. T. Fungal cellulases. Chem. Rev. 2015, 115, 1308–1448. 10.1021/cr500351c. [DOI] [PubMed] [Google Scholar]
  44. Yoshida S.; Hiraga K.; Takehana T.; Taniguchi I.; Yamaji H.; Maeda Y.; Toyohara K.; Miyamoto K.; Kimura Y.; Oda K. A bacterium that degrades and assimilates poly(ethylene terephthalate). Science 2016, 351, 1196–1199. 10.1126/science.aad6359. [DOI] [PubMed] [Google Scholar]
  45. Nielsen H.Predicting Secretory Proteins with SignalP, In Methods in Molecular Biology; Humana Press: New York, NY, 2017; Vol. 1611, pp 59–73. [DOI] [PubMed] [Google Scholar]
  46. Crooks G. E.; Hon G.; Chandonia J. M.; Brenner S. E. WebLogo: a sequence logo generator. Genome Res. 2004, 14, 1188–1190. 10.1101/gr.849004. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao0c03911_si_001.pdf (1.7MB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES