Luria–Bertani (LB) liquid medium with or without ampicillin (100 μg/mL) or kanamycin (50 μg/mL) was used for the growth and transformation of Escherichia coli at 37°C with 200 rpm of shaking, and LB solid medium (containing 2% agar) was used for screening at 37°C. Yeast Extract-Peptone-Dextrose (YPD) liquid medium with or without G418 (100 μg/mL) was used for the growth and transformation of yeast at 30°C with 200 rpm of shaking, and screening was performed using YPD solid medium (containing 2% agar) incubated at 30°C. Recombinant E. coli containing a temperature-sensitive replicon was grown and transformed using LB liquid medium containing ampicillin (100 μg/mL) at 30°C with 200 rpm of shaking and screened using LB solid medium (containing 2% agar), incubated at 30°C. The fermentation of recombinant yeast strains to induce protein expression was performed under shake flask conditions using Buffered Minimal Glycerol-complex Medium (BMGY) and Buffered Methanol-complex Medium (BMMY) media. The BMGY medium contained the following: 10 g/L glycerol, 10 g/L yeast extract, 20 g/L peptone, 13.4 g/L yeast nitrogen base (YNB), 4 × 10⁻⁴ g/L biotin. The BMMY medium contained the following: 10 g/L yeast extract, 20 g/L peptone, 13.4 g/L YNB, 4 × 10⁻⁴ g/L biotin. All strains and plasmids used in this study are listed in Table 1, and all primers used in this study are listed in Table 2.

In a previous study, we obtained the pectin lyase PGLA-rep4, which showed a 1.13-fold increase in enzymatic activity compared to that of the wild-type, using a rational design, with a fragment substitution technique to replace the unstable region in the gene sequence. To verify the difference in enzymatic activity in different hosts, PGLA-rep4 was ligated into pET28a and pPIC9K vectors and transformed into E. coli BL21(DE3) and P. pastoris GS115 to obtain heterologous expression strains, specifically E. coli BL21/PGLA-rep4 and Pichia pastoris GS115/PGLA-rep4, which were then assayed for enzyme activity.

The AOX1 promoter in P. pastoris is strictly induced by methanol to transcriptionally regulate the expression of related genes, and BMMY is more suitable for propagation of the expression vector pPIC9K, in conjunction with BMGY. After activation, P. pastoris GS115/PGLA-rep4 was transferred to 50 mL of BMMY medium based on a 1% inoculum and incubated overnight in shaking flasks. The BMGY was centrifuged at 3,000 rpm for 10 min, the supernatant was discarded, and the precipitate were resuspended in 50 mL of BMMY medium and then incubated at 30°C with 200 rpm of shaking for 72 h. The concentration of methanol was maintained at 1% by replenishing it every 24 h to induce protein expression. Meanwhile, E. coli BL21/PGLA-rep4 was transferred to 50 mL of LB medium based on a 1% inoculum, when the density of the bacteria reached approximately 0.7 OD₆₀₀, after which isopropyl β-D thiogalactoside (IPTG) at a final concentration of 0.5 mM was added, and cells were incubated at 25°C for 12 h to induce protein expression.

A comparison of different host bacteria was performed based on an alkaline pectin lyase enzymatic activity assay. The supernatant was removed by centrifuging 30 mL of fermentation broth at 12,000 rpm for 10 min. The precipitate was then resuspended in 8 mL of PBS (pH 7.4) and sonicated for 20 min, for E. coli, or 40 min, for P. pastoris, before centrifugation at 12,000 rpm for 10 min. Then, the supernatant was collected as the desired crude enzyme solution. The supernatant of the culture after the induced expression was centrifuged as the crude enzyme solution, and a 25 mL colorimetric tube was used, to which 1.9 mL of glycine-sodium hydroxide buffer (pH 11.0) containing 0.2% pectin substrate was added, followed by 100 µL of the appropriately diluted crude enzyme solution. After mixing, the reaction was carried out at 70°C for 10 min. Finally, 3 mL of 0.03 M phosphoric acid solution was added to terminate the reaction, and the absorbance value of the unsaturated product at 235 nm was measured using a UV spectrophotometer; enzyme activity was determined according to the calculation formula(Yadav et al., 2008; Cerreti et al., 2017; Zhou et al., 2017).

The reaction with PGLA-rep4 was carried out by adding 0.2% pectin in Gly-NaOH buffer (pH 11.0) for 10 min at 70°C. The reaction was then terminated by adding 3 mL of 0.03 M phosphoric acid. The hydrolysates released from pectin were evaluated by performing high-performance liquid chromatography (HPLC). The resulting sugars were separated on a Hi-Plex Ca column (300 mm × 7.7 mm; Agilent, Santa Clara, CA, United States) using a mobile phase of pure water at a flow rate of 0.6 mL/min. The column temperature was maintained at 80°C, and the sample was injected into 10 μL. The peak area of the sugar was measured using a Shimadzu RID-10A refractive index detector. The peak times of the hydrolysate and corresponding standards were compared to determine the type of hydrolysate.

pKD46 is a temperature-sensitive plasmid that relies on the rep101 replication protein; it can replicate normally at 30°C and is essentially unreplicable at 37°C, and it can be utilized to rapidly eliminate the enhanced green fluorescent protein (egfp) gene in mutant strains (Liao et al., 2018). The PCR reaction was performed with the pKD46 plasmid as the template and kd-F and kd-R as primers; here, DNA was pre-denatured at 95°C for 3 min, denatured at 95°C for 15 s, annealed at 60°C for 30 s, pre-extended at 72°C for 90 s, cycled 30 times, then extended at 72°C for 5 min, and finally kept at 4°C. A temperature-sensitive fragment, of 3,285 bp in length, was obtained via gel purification of the replicon. The PCR reaction was then performed with the fragment containing the PGLA-rep4-encoding gene, T7 promoter, and lacⅠ gene with the pET28a-PGLA-rep4 plasmid as the template, using et-F and et-R as primers. Here, the DNA was pre-denatured at 95°C for 3 min, denatured at 95°C for 15 s, annealed at 60°C for 30 s, and pre-extended at 72°C for 90 s, cycled 30 times, and then, extension was performed at 72°C for 5 min; the sample was finally kept at 4°C. The fragment containing the PGLA-rep4-encoding gene, T7 promoter, and lacⅠ gene, with a length of 2,834 bp, was obtained via gel purification. The obtained gene fragment was ligated using a seamless cloning technique to finally obtain the recombinant plasmid pKD46-PGLA-rep4.

The pET28a-pKD46-PGLA-rep4 plasmid was then single endonuclease digestion to obtain the linearized vector. The green fluorescent protein vector pUC57-egfp was provided by GenScript, and the PCR reaction was performed with this plasmid as the template using egf-F and egf-R as primers. Here, the DNA was pre-denatured at 95°C for 3 min, denatured at 95°C for 15 s, annealed at 60°C for 30 s, and pre-extended at 72°C for 90 s, cycled 30 times, and then, extension was performed for 5 min, and the sample was finally kept at 4°C. The green fluorescent protein egfp-encoding gene fragment, with a length of 783 bp, was obtained via gel purification. The obtained linearized pKD46-PGLA-rep4 vector and egfp gene fragment were ligated together using a seamless cloning technique to finally obtain the recombinant plasmid pKD46-PGLA-rep4-egfp. Next, the recombinant plasmid was introduced into E. coli BL21 (DE3), which was cultured in LB containing ampicillin (100 μg/mL) and on LB agar plates; after successful incubation, single colonies were selected and grown in LB liquid medium containing ampicillin (100 μg/mL).

To determine the optimal conditions for UV mutagenesis, 3 mL of a diluted E. coli BL21/E bacterial solution (1.62 × 10⁷ CFU/mL) was placed in a disposable Petri dish, and the suspension was irradiated at a distance of 28 cm from a 36 W UV lamp in the dark for 0, 15, 30, 45, 60, 70, and 90 s. After irradiation, the suspension was diluted to 10^–5, 10^–6, and 10^–7, incubated on LB solid plates and observed. After irradiation, the suspension was diluted to 10^–5, 10^–6, and 10^–7 and then incubated on LB solid plates, and the number of colonies was enumerated to calculate the cell death. The time at which cell death reached >90% was regarded the best time for mutagenesis, as it corresponds to the highest probability of a strain undergoing forward mutation (Kawai et al., 2022).

The fermentation broth was transferred to 50 mL of LB medium after UV irradiation and grown at 30°C with 200 rpm of shaking until the OD₆₀₀ was approximately 0.7. IPTG, at a final concentration of 0.5 mM, was added for induction at 25°C for 12 h. The fermentation broth was centrifuged at 5,000 rpm for 10 min, and then, the supernatant was discarded and the precipitate was washed with 0.9% saline twice. Then, the supernatant was discarded, the precipitate was rinsed 2–3 times with 0.9% saline, and finally, the precipitate was resuspended in saline. The resuspended cytosol was diluted 100-fold and then sorted by performing flow cytometry. Sodium chloride solution, at a concentration of 0.85%, was used as the flow cytometer sheath solution, and fluorescent microspheres were mixed for quality control, with respect to the flow cytometer optical path and liquid flow system. The prepared samples were sorted via flow cytometric analysis using a Moflo XDP model ultra-fast flow cytometer (Beckman Coulter) equipped with a trichrome laser. The maximum excitation wavelength of the green fluorescent protein egfp is approximately 488 nm, and the maximum emission wavelength is approximately 510 nm; therefore, a 485 nm laser was used to efficiently achieve excitation of the fluorescent protein, and the FL 1 channel of the experimental model flow cytometer was selected to determine the emission wavelength. The injector differential pressure was adjusted such that the program was executed at a system pressure of 60 Psi, while keeping the sample pressure 0.1–0.3 Psi higher than the sheath fluid pressure, to ensure the detection of approximately 10,000 target cells per second. Several single cells with uniform size and strong fluorescence were selected, and their fluorescence intensity was measured using a microplate reader after induction.

Tri-generation PacBio sequencing technology was combined with Illumina sequencing technology to perform genome sequencing analysis of the high pectin lyase-producing strain E. coli BL21/1G3, obtained via UV mutagenesis. The starting strain (accession number: NZ_CP053601) and the mutant strain were compared, and the mutated sequences were compared with the genome sequence of E. coli BL21 obtained using BLAST in the NCBI database to obtain information related to the increased expression of pectin lyase in the mutant strain.

After expression was induced in the strain, mRNA was extracted using the Total RNA Extraction Kit (Solarbio). qPCR-specific primers were designed based on the first 500 bases of PGLA-rep4, and a pair of primers was designed by selecting an E. coli reference gene (gapA) as an internal reference (Table 2). qPCR reactions (20 μL) contained 10 μL of SsoAdvanced Universal SYBR Green Supermix (BioRad), 20 μM primers, 10 ng of mRNA extract, and DEPC water. Reactions were measured using a LightCycler 96 Detection System (Roche) with the following thermocycling parameters: one cycle at 95°C for 2 min; 40 cycles at 95°C for 10 s, 59°C, for 20 s, and 72°C for 20 s; and final melting curve analysis (59°C–95°C based on 0.5°C/5 s increments).

The effect of the mutation on transcription efficiency was illustrated by measuring the relative abundance of mRNA in the unmutated strain E. coli BL21/PGLA-rep4 and the mutated strain E. coli BL21/1G3. In qPCR experiments, the mRNA relative abundance in all samples was calculated using the 2^−ΔΔCT method. All measurements shown in the text and figures are based on three independent biological replicates.

Statistical analysis was performed using IBM SPSS Statistics 29 (IBM SPSS, Turkey). One-way ANOVA was used to analyze the significance of mean differences among samples. The significance of differences was estimated with a 0.05 level of confidence.

The host for plasmid expression affects the intensity of enzyme activity. To study the effect of different hosts on pectin lyase expression, plasmids pET28a-PGLA-rep4 and pPIC9K-PGLA-rep4 were constructed (Figure 2; Supplementary Figures S1A-B) and transformed into the host strains E. coli BL21(DE3) and P. pastoris GS115. After the induction of expression in P. pastoris GS115/PGLA-rep4 and E. coli BL21/PGLA-rep4, the fermentation broth was centrifuged, resuspended in PBS, and subjected to ultrasonic fragmentation, and the supernatant was centrifuged to determine the enzyme activity. One unit of enzyme activity was defined as the amount of enzyme required to produce the equivalent of 1 μg of unsaturated oligogalacturonan material per minute from the cleavage of polygalacturonan. To verify that PGLA-rep4 was successfully expressed in E. coli and P. pastoris, we performed SDS-PAGE analysis, and the results proved that PGLA-rep4 was successfully expressed in E. coli (Supplementary Figure S2A) and P. pastoris (Supplementary Figure S2B), with clear and obvious protein bands and a molecular weight of approximately 35 kDa. After the PGLA-rep4 reacted with the pectin substrate at 70°C for 10 min, galacturonic acid was detected as the main hydrolysis product obtained from fructose using HPLC analysis, whereas pectin was barely detectable (Supplementary Figure S3). The changes in enzyme activity of the recombinant strain under 30 L fermenter conditions were determined and are shown in Figure 3. With the E. coli vector, PGLA-rep4 exhibited an activity of 168,058 U/mL, whereas, with the P. pastoris GS115 vector, it was only 26,990 U/mL, which shows that E. coli is a more suitable host for the expression of PGLA-rep4 than P. pastoris GS115. Therefore, E. coli BL21(DE3) was chosen as the host bacterium for PGLA-rep4 expression.

The UV mutagenesis vector E. coli BL21(DE3)/pET28a-pKD46-PGLA-rep4-egfp was constructed (Figure 4; Supplementary Figures S4A-D). By linking the enzyme to a fluorescent protein and coupling the expression of this enzyme to the fluorescence intensity, screening E. coli mutant hosts that are more favorable for enzyme expression after mutagenesis becomes possible. The fluorescence intensity of the fluorescent protein egfp was calculated to determine enzyme activity, and the inserted temperature-sensitive replicon was used to directly avoid the effect of fluorescent protein expression on enzyme activity. Agarose gel electrophoresis showed that the target band was consistent with the size of the target gene, and the sequencing results indicated that the base sequence was complete.

The recombinant strain E. coli BL21/E was UV-irradiated using a 36 W UV lamp a cell death curve was plotted, as shown in Figure 5. When the UV irradiation time reached 15 s, the lethality of the strain reached about 81%; when irradiated for 30 s, the lethality of the strain reached about 83%; and when irradiated for 45 s, the lethality of the strain reached about 97%. We choose UV irradiation for 35 s as the best mutagenesis time.

The UV-irradiated bacterial solution was subjected to primary screening using flow cytometry to screen cells of uniform size with strong fluorescence, and the range of the screening area is shown in Figures 6A–D. Figure 6A shows that the E. coli strain without the egfp gene did not produce fluorescence, whereas Figure 6B shows that the E. coli strain containing the egfp gene exhibited low fluorescence intensity without induction. Figure 6C shows that the E. coli strain containing the egfp gene exhibited improved fluorescence intensity after induction, and Figure 6D shows that the fluorescence intensity of most E. coli strains containing the egfp gene decreased after 35 s of UV irradiation, indicating a reverse mutation. Meanwhile, a small number of strains showed an increase in fluorescence intensity, indicating a forward mutation. The strains with increased fluorescence intensity in the A region, in Figure 6D, were screened, accounting for approximately 1 in 20,000 isolates.

After screening, the fluorescence intensity of each single cell was measured, after growth, multiplication, and expression induction, and the results are shown in Figure 7. The red line indicates the fluorescence intensity produced by the strain without UV irradiation, whereas the orange line represents the fluorescence intensity produced by the strain with UV irradiation. The cells of the five strains for which fluorescence intensity was higher than the highest level of fluorescence intensity of the unmutated strains, in the green box line, were selected and cultured in shake flasks, and their enzyme activity was determined after expression induction. The plasmid of the mutant strain also expressed the fluorescent protein egfp, for which the encoding gene was ligated to the plasmid after the pectin lyase gene sequence. Since this could affect the expression of the enzyme, as well as enzyme activity, additional fluorescent strains obtained via mutagenesis were prepared in competent cells by curing the plasmid at 37°C, and the pET28a-PGLA-rep4 plasmid was transformed into each receptor cell. Then, the enzyme activity was measured after culture and inducing expression. As shown in Figure 8A, among the five mutant strains, E. coli BL21/1H1, E. coli BL21/1G4, E. coli BL21/3E5, and E. coli BL21/3F7 exhibited 0.98- (ns), 1.81- (p < 0.001), 0.98- (p < 0.05), and 1.15- (p < 0.001) fold the enzyme activity of the starting strain, whereas E. coli BL21/1G3 exhibited the highest enzyme activity, which was 2.02- (p < 0.001) fold higher than that of the E. coli BL21/E strain before UV irradiation. Further, this was approximately 1.37- (p < 0.001) fold higher than that of the parental strain E. coli BL21/PGLA-rep4 (Figure 8B).

After the induction of expression in E. coli BL21/PGLA-rep4 and E. coli BL21/1G3, the fermentation broth was centrifuged and resuspended in PBS and subjected to ultrasonic fragmentation. Moreover, the supernatant was centrifuged to determine the enzyme activity. One unit of enzyme activity was defined as the amount of enzyme required to produce the equivalent of 1 μg of unsaturated oligogalacturonan material per minute from the cleavage of polygalacturonan. Changes in the enzyme activity of the recombinant strain under 30 L fermenter conditions are shown in Figure 9. In E. coli BL21/PGLA-rep4, the enzyme activity reached 168,058 U/mL, whereas in the mutagenized strain E. coli BL21/1G3, the enzyme activity reached 230,240 U/mL, which was approximately 1.37- (p < 0.001) fold higher than that in the original strain. This indicated that UV mutagenesis combined with the flow cytometric screening strategy successfully improved the enzymatic activity of PGLA-rep4.

The genome spliced sequence was compared with the genome sequence of E. coli BL21 (DE3), obtained using BLAST from the NCBI database, and one base mutation and one base deletion were identified. The sequencing results have been uploaded to the NCBI SRA database (Registry number: PRJNA988841). Regarding the base mutation, the amino acid was mutated from arginine to leucine, and the corresponding gene was found to encode a ApaH, as determined using BLAST from the NCBI database (Supplementary Figure S5). Regarding the deletion, two bases were missing, and the use of BLAST from the NCBI database showed that the mutated gene encodes RNase E (Supplementary Figure S6). It is thus possible that the mutations in these two enzymes resulted in the increased expression of pectin lyase in the mutant strain.

Differences in the relative abundances of mRNAs were used to show transcript levels in the original strain versus those in the mutant strain. As shown in Supplementary Figure S7, the mRNA abundance in the mutant strain E. coli BL21/1G3 was increased 1.3-fold (p < 0.001) relative to that in the original strain E. coli BL21/PGLA-rep4. The mutant strain also exhibited a higher level of transcription and, therefore, relatively higher expression of PGLA-rep4.