The strains used in this study are listed in Table 1. Pseudomonas putida NCIMB 9866 was grown at 30°C in minimal medium (MM; Chapman and Hopper, 1968) on different compound carbon sources. Escherichia coli was grown in lysogeny broth (LB) on a rotary shaker (200rpm) at 37°C under aerobic conditions. The culture medium is usually supplemented with antibiotics at a final concentration of 50μgml⁻¹ for kanamycin sulfate (Kan) and 100μgml⁻¹ for ampicillin sodium (Amp), as necessary. The chemicals used in the experiments were purchased from Aladdin Reagents (Shanghai, China) or Sigma-Aldrich (St. Louis, MO, United States). Taq DNA polymerase and Pfu DNA polymerase were obtained from TaKaRa Bio (Dalian, China). TransEco FastPfu DNA Polymerase was purchased from TransGen Biotech (Beijing, China).

Plasmid preparation and DNA manipulation were carried out as described previously (Green and Sambrook, 2012). The primers used in this study are listed in Table 2 and are shown in Figure 1. All the targeted genes were amplified by PCR using Taq DNA polymerase or TransEco FastPfu DNA Polymerase (Transgen, Beijing, China), according to its manufacturer’s recommendation. The pcaHG98 gene was amplified with PCR with primers pcahg9803 and pcahg9804 from the extracted genomic DNA of strain NCIMB 9866. The PCR reaction volume usually uses 50μl system and consists of 5–30ng Plasmid DNA or 100ng Genomic DNA, 1x TransEco FastPfu Buffer, 250μM dNTP, 0.4μM of each primer, 2.5U of TransEco FastPfu DNA Polymerase or Taq DNA polymerase and ddH₂O up to 50μl. PCR amplification was performed with TransEco FastPfu DNA polymerase as follows: denaturation at 95°C for 5min, 30cycles of 95°C for 1min, 58°C for 1min, and 72°C for 1min (1kb/min for Taq, 2kb/min for FastPfu), and a final extension cycle at 72°C for 5min. The PCR fragments were digested with NdeI/HindIII (Thermo Fisher Scientific, Shanghai, China) before being cloned into pET-28a (+) with T4 DNA ligase (TaKaRa, Dalian, China) to produce pET-28a-pcaHG98e01. The pcaHG98 gene was amplified using primers pcahg9801 and pcahg9802 from plasmid pET-28a-pcaHG98e01 and cloned into NdeI/HindIII sites of pET-28a (+) with T4 DNA ligase to produce pET-28a-pcaHG98e02. The pcaHG98 gene was amplified using primers pcahg9802 and pcahg9803 from plasmid pET-28a-pcaHG98e01 and cloned into NdeI/HindIII sites of pET-28a (+) with T4 DNA ligase to produce pET-28a-pcaHG98e03. The pcaHG98 gene was amplified using primers pcahg9801 and pcahg9804 from plasmid pET-28a-pcaHG98e01 and cloned into NdeI/HindIII sites of pET-28a (+) with T4 DNA ligase to produce pET-28a-pcaHG98e04.

The nucleotide sequence for YFP was synthesized according to the sequences of pcDNA3YFP, which was a gift from Doug Golenbock (Addgene plasmid # 13033; http://n2t.net/addgene:13033; RRID: Addgene_13,033). The YFP gene was amplified using primers pYFP01 and pYFP02 from this synthesized YFP gene (Sangon Biotech, Shanghai, China) and fused to the NdeI/HindIII restriction sites of pET-28a (+) with the ClonExpress II One Step Cloning Kit (Vazyme, Nanjing, China) to produce pET-28a-YFP. The YFP gene was amplified using primers pYFP03 and pYFP04 from pET-28a-YFP and fused to the NdeI/HindIII restriction sites of pET-28a (+) to produce pET-28a-YFPt-PCDS. The YFP gene was amplified by PCR using primers pYFP04 and pYFP05 and fused to the NdeI/HindIII restriction sites of pET-28a (+) to produce pET-28a-YFP-PCDS. PCR amplification conditions are the same as pcaHG98 gene described above.

The eGFP gene was amplified using primers pegfp03 and pegfp04 from the mammalian expression vector pEGFP-N1 and fused to the NdeI/EcoRI restriction sites of pET-28a (+) with the ClonExpress II One Step Cloning Kit to produce pET-28a-eGFP. The primers pET28a-pcas03 and pET28a-pcas04 containing PCDS tag were used to amplify linearized recombinant vector from pET-28a-eGFP, which was then fused by ClonExpress II One Step Cloning Kit to form cyclized vector pET-28a-eGFP-PCDS.

The nucleotide sequence of the resulting plasmid was confirmed by Sangon Biotech (Shanghai, China).

Heterologous expression of the cloned pcaHG gene was accomplished by introducing the constructed plasmid into E. coli BL21 (DE3; Novagen, Madison, WI). The transformed cells were grown at 37°C to an OD₆₀₀ of 0.4 in 100ml LB supplemented with 50gml⁻¹ of kanamycin in 500ml shake flask; then, the protein expression was induced with 0.1mM of isopropyl-β-D-thiogalactopyranoside (IPTG) for approximately 5h at 30°C, resulting in OD₆₀₀ of 2. His₆-PcaHG98e04-PCDS was purified using Ni²⁺-nitrilotriacetic acid agarose chromatography (Novagen) and eluted with 200mM of imidazole. Purified target His₆-PcaHG98e04-PCDS was further dialyzed away from imidazole with a Spectra/Por CE dialysis membrane with a molecular weight cut-off of 10,000Da (Spectrum Laboratories Inc., Shanghai, China) at 4°C for 48h against phosphate buffer (PB) before being further preserved in glycerol at −80°C.

The expression and purification of YFP and eGFP are performed according to that of His₆-PcaHG98e04-PCDS. The total protein amounts of E. coli BL21 (DE3)/pET-28a-YFP, E. coli BL21 (DE3)/pET-28a-YFPt-PCDS and E. coli BL21 (DE3)/pET-28a-YFP-PCDS refer to ultrasound-broken cells after centrifugation collection of 100ml IPTG induced expression. The amounts of purified His₆-YFP, His₆-YFP-PCDS, and His₆-YFPt-PCDS were detected directly after purification. Protein concentrations were determined with the common Bradford assay (Bradford, 1976).

Molecular weight of the purified recombinant His₆-PcaHG98e04-PCDS was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and liquid chromatograph high-resolution accurate mass spectrometry (LC–MS). LC–MS data were obtained using a Thermo Scientific™ Q Exactive™ mass spectrometer (Thermo Fisher Scientific, Waltham, MA, United States). The recombinant His₆-PcaHG98e04-PCDS was purified by SDS-PAGE. The His₆-PcaHG98e04-PCDS sample was dissolved in 2% acetonitrile and 0.1% formic acid. The mixture was vortex mixed and centrifuged at 12,000g for 15min. The sample was then purified by ultra-filtration, concentrated using Amicon Ultra-10K centrifugal filter device and stored in 0.1% formic acid. MS was performed using a Thermo Scientific Q Exactive mass spectrometer operated in the data-dependent/dynamic exclusion mode. A full resolution setting of 70,000 (full width at half maximum, FWHM) was used for full scan MS. All data were processed using Thermo Scientific™ Pinpoint™ software (Thermo Fisher Scientific, Waltham, MA, United States).

PCA-3,4-dioxygenase activities were detected by referring to previously described methods (Stanier and Ingraham, 1954; Gross et al., 1956). For the assay of PCA-3,4-dioxygenase activity, the reaction mixture contained 50mM of PB (pH 7.5) and 200μM of cell extracts or purified His₆-PcaHG98e04-PCDS. All compounds except the substrate were added to the reference cuvette, and the enzyme activity assay was initiated by the addition of 30μg of PCA. The spectra in the range of 220–400nm were recorded every minute. The molar extinction coefficients for PCA at 290nm and 3-carboxy-cis, cis-muconate at 270nm were 3,890M⁻¹ cm⁻¹ (Gross et al., 1956) and 6,390M⁻¹ cm⁻¹ (Iwagami et al., 2000), respectively. One unit of enzyme activity was defined as the protein amount required for the production (or disappearance) of 1μmol of product (or substrate) min⁻¹ at 30°C. The protein amounts of E. coli BL21 (DE3)/pET-28a-pcaHG98e01 and E. coli BL21 (DE3)/pET-28a-pcaHG98e04 refer to the soluble protein after the 100ml IPTG induced cell is ultrasonically broken and centrifuged. The amount of purified His6-PcaHG98e04-PCDS was directly detected. All enzyme activity assays were run in triplicate in three independent experiments.

The BL21 (DE3)/pET-28a(+) was used as a nonfluorescence control. The culture broths of BL21 (DE3)/pET-28a(+), BL21 (DE3)/pET-28a-YFP, BL21 (DE3)/pET-28a-YFP-PCDS, and BL21 (DE3)/pET-28a-YFPt-PCDS grown under the same conditions according to that of YFP expression. Then the cell were pelleted, washed twice with 50mM of Tris-HCl buffer (pH 7.4), resuspended in Tris-HCl buffer, and transferred to a 2-ml round bottom centrifuge tube, respectively. The cell concentration was diluted to 5×10⁴–5×10⁵ and counted. Fluorescence-activated cell sorting (FACS) analysis was carried on a Cytoflex S flow cytometer (Beckman Coulter, IN, United States). The laser beam for cells was 488-nm, and a 529±14-nm FL1 band pass filter used to filter the emission signal. The light signals from the cells are converted to voltage values and the minimum voltage value (400V) was used, with a threshold of 9,200. FACS analysis was run in triplicate in three independent experiments. Data were analyzed using FlowJo v10.0.4 (Tree Star Inc., Ashland, United States) software.

Purified recombinant His₆-YFP, His₆-YFP-PCDS, and His₆-YFPt-PCDS were diluted to 0.1mg/ml with 50mM of Tris-HCl buffer (300mM of NaCl, pH 7.4) and monitored by a Fluorescence Spectrophotometer (F96PRO, Shanghai Kingdak Scientific Instrument, Shanghai, China). The emission and excitation wavelengths of the recombinant protein were 400 and 700nm, respectively. Fluorescent spectral analysis was run in triplicate in three independent experiments.

The statistical analysis used in this study was performed by SPSS version 23.0 software (IBM SPSS Inc., New York, NY, United States). One-way analysis of variance (ANOVA) was used to calculate the values of p for PCA-3,4-dioxygenase activity analyses. Paired-sample tests were used to calculate probability (p) values for the expression of YFP. The values of p of 0.05 and 0.01 were considered as statistically significant and highly statistically significant, respectively.

pcaHG98, encoded a putative protocatechuate 3,4-dioxygenase, was cloned into pET-28a (+). The pcaHG98 gene was amplified by PCR with primers pcahg9801 and pcahg9802 (Figure 1) using strain NCIMB 9866 gDNA or bacteria lysate as a template. But the gene could not be amplified into PCR products using either Taq DNA polymerase or Pfu DNA polymerase. In order to determine this sequence on the genome of strain NCIMB 9866, two primers, pcahg98S01 and pcahg98S02 (Figure 1), were designed for sequencing analysis, and the results show that the sequence was consistent with previous reports (Chen et al., 2014). Then, two primers, pcahg9803 and pcahg9804 were redesigned (Figure 1) successfully cloned the pcaHG98 gene into pET-28a (+) to generate pET-28a-pcaHG98e01.

pET-28a-pcaHG98e01, which contained the amplified pcaHG98 gene from strain NCIMB 9866 gDNA, was expressed in E. coli BL21(DE3). The molecular masses calculated from the nucleotide sequence of PcaG and PcaH were 22.7 and 26.3kDa, respectively, which corresponded to the proteins observed by SDS-PAGE (Figure 2A). The PcaHG9801 protein, however, did not bind to Ni²⁺-nitrilotriacetic acid agarose for chromatography. We obtained the same results three separate times. This may have been because the pcaHG98 gene included 54bp of upstream sequence and 45bp of downstream sequence in pET-28a-pcaHG98e01. The start and stop codons of the pcaHG98 gene still retained with it in this recombinant. Therefore, a His-tag is not fused to the pcaHG98 gene due to the presence of its start and stop codons.

The pcaHG98 gene in vector pET-28a-pcaHG98e02 showed no effective expression, and most of its proteins were insoluble, of which the α-subunit PcaG was not determined by SDS-PAGE (Figure 2A). The pcaHG98 gene was amplified from pET-28a-pcaHG98e01 to produce pET-28a-pcaHG98e03, in which the pcaHG98 gene is expressed in the form of inclusion body (Figure 2B). However, the pcaHG98 gene in pET-28a-PcaHG98e04 was expressed in the form of soluble proteins, which was amplified using primers pcahg9801 and pcahg9804 (Figure 2B).

The PcaHG98e04 protein was constructed using the primers pcahg9801 and pcahg9804, and the primer pcahg9801 (which removed the start codon ATG) starts with the pcaH gene 5′-end, with the lead pcahg9804 on the outside of the pcaG gene (containing 45 bases: GACATCGCCGGGGCGCGCGCGCG). Compared with the insoluble PcaHG98e02, these 45 bases translated into polypeptides make the His₆-PcaHG98e04-PCDS protein soluble, the reasons for which are further explored below (Figure 2B).

His₆-PcaHG98e04-PCDS, purified using Ni²⁺-nitrilotriacetic acid agarose chromatography, was then determined using LC–MS. The molecular weight of the protein detected was 27.6kDa (Figure 3), which contained 22.7kDa encoded by the pcaG gene and 3.3kDa encoded by the 45-bp (the underlined DNA sequence) and the other end sequences (GACATCGCCGGGGCAGCTGCTCCGGCACGAGTACGAAACGCGTCT CAAGCTTGCGGCCGCACTCGAGCACCACCACCACCACCACTGAGATCCGGCTGCTAA). These results indicated that the expressed protein PcaG contained the PCDS tag.

Protocatechuate 3,4-dioxygenase catalyzes the cleavage of PCA into β-carboxy-cis, cis-muconate (Harwood and Parales, 1996). A series of DNA fragments of different lengths containing the complete reading frame of the pcaHG98 gene was ligated into the expression vector pET-28a (+), to make the plasmids pET-28a-pcaHG98e01, pET-28a-pcaHG98e02, pET-28a-pcaHG98e03, and pET-28a-PcaHG98e04. These recombinant plasmids were then individually expressed in E. coli BL21 (DE3). After induction with IPTG, E. coli BL21 (DE3)/pET-28a-pcaHG98e01 and E. coli BL21 (DE3)/pET-28a-pcaHG98e04 cell extracts were found to contain PCA 3,4-dioxygenase with specific activities of 74.7±5.31 and 72.8±4.41U/mg against PCA as a substrate, respectively. No activity was detectable in E. coli BL21 (DE3)/pET-28a-pcaHG98e02 and E. coli BL21 (DE3)/pET-28a-pcaHG98e03, where the expression vector contained no soluble PcaHG98 protein. Purified His₆-PcaHG98e04-PCDS exhibited 205.63±14.23U/mg activity against PCA as a substrate, which was not affected by the PCDS tag. Figure 4 shows the rapid transformation of PCA (λ_max=290nm) to 3-carboxy-cis, cis-muconate (λ_max=270nm) in the cell extract, as described by Harwood and Parales (1996).

It has been reported that the YFP has relatively low expression in E. coli compared to that of codon-optimized YFP (Nguyen et al., 2019). The YFP gene was synthesized according to the sequence of the mammalian expression vector pcDNA3YFP and cloned into the E. coli expression vector pET28a (+) in this study. Based on our results, the YFP gene was expressed in our E. coli system and had a certain amount of solubility (Figure 5B). The 45-bp PCDS tag enhanced the soluble expression of the YFP gene in E. coli after fusing the PCDS tag to the YFP C-terminus (Figure 5B). This YFP-PCDS did not have YFP termination codons, whereas YFPt-PCDS contained YFP termination codons. The PCDS tag enhanced the expression level of YFP in E. coli was analyzed by fluorescence-activated cell sorting (FACS). FACS analysis showed that the fluorescence intensity of BL21 (DE3)/pET-28a-YFP-PCDS and BL21 (DE3)/pET-28a-YFPt-PCDS was higher than that of BL21 (DE3)/pET-28a-YFP, and BL21 (DE3)/pET-28a-YFPt-PCDS was the highest (Figure 5A). BL21 (DE3)/pET-28a-YFP showed weaker fluorescence intensity, as did the control BL21 (DE3)/pET-28a(+). The total protein expression of YFP-PCDS and YFPt-PCDS was significantly amplified up to 1.6-fold and 2-fold compared to that of YFP alone, respectively. Accordingly, His₆-YFP-PCDS and His₆-YFPt-PCDS had 1.6-fold and 3-fold higher soluble expression yield, respectively, than that of His₆-YFP under the same expression and purification conditions (Figure 5C). All protein amounts were determined three times in three independent experiments. His₆-YFP, His₆-YFP-PCDS, and His₆-YFPt-PCDS also showed consistent fluorescence emission spectra, with a peak at 530nm across the scanning range from 400 to 700nm (Figure 5D).