Pectin lyase

At first, Klebsiella phages were collected from the GenBank database (retrieved at 15.08.2018). A number of 59 phages were finally analyzed (Supplementary Table S1). From these phages proteins annotated as tail fibers or tail spikes were analyzed with BlastP¹ (Altschul et al., 1990 (link)), Phyre2² (Kelley et al., 2015 (link)), SWISS-MODEL³ (Bordoli et al., 2009 (link); Bordoli and Schwede, 2012 (link)), HMMER⁴ (Finn et al., 2011 (link)) and HHPred⁵ (Zimmermann et al., 2018 (link)) to identify phages that encode RBPs with putative depolymerase activity (Supplementary Table S2). If neither a tail fiber nor a tail spike gene was found in the genome, we analyzed all genes located in the vicinity of annotated structural genes. BlastP (protein–protein Blast) was performed against the non-redundant protein sequences (nr) database using standard parameters (expect threshold: 10, word size: 6, MATRIX: BLOSUM62, Gap cost: existence 11, extension 1, conditional compositional score matrix adjustment). HMMER was used in the quick search mode against: Reference Proteomes, UniProtKB, SwissProt, and Pfam with significance E-values: 0.01 (sequence) and 0.03 (hit). For Phyre2 the normal modeling mode was used. HHPred homology detection structure prediction was run using the PDB_mmCIF70 database and the following parameters [MSA generation method: HHblits uniclust30_2018_08; Maximal no. of MSA generation steps: 3; E-value incl. threshold for MSA generation: 1e-3; minimal sequence identity of MSA hits with query (%): 0; minimal coverage of MSA hits (%) 20; Secondary structure scoring: during alignment; Alignment Mode: Realign with MAC: local:norealign; MAC realignment threshold: 0.3; No. of target sequences: 250; Min. probability in hit list (>10%): 20].
Criteria for the prediction of putative depolymerase activity were (Supplementary Table S2): (1) the protein must be longer than 200 residues; (2) the protein must be annotated as tail fiber/tail spike/hypothetical protein in the NCBI database; (3) the protein must show homology to domains annotated as lyase [hyaluronate lyases (hyaluronidases), pectin/pectate lyases, alginate lyases, K5 lyases] or hydrolase (sialidases, rhamnosidases, levanases, dextranases, and xylanases) with a confidence of at least 40% in Phyre2 or the enzymatic domain should also be recognized by at least SWISS-MODEL, HMMER, or BlastP; (4) the length of homology with one of these enzymatic domains should span at least 100 residues; (5) a typical β-helical structure should be predicted by Phyre2. These RBP depolymerases are indicated without additional labeling in the tables. Proteins possessing experimentally confirmed depolymerizing activity were marked in the tables with (a). When the RBP was only partially fulfilling the above-mentioned criteria, it was indicated with label (b). These putative depolymerases that could only be predicted with a lower probability were fulfilling criteria 1 and 2, but the confidence of the Phyre2 prediction was below 40% or only SWISS-MODEL, HMMER or BLASTP gave a positive prediction. In addition, the homologous domain only spans between 50 and100 amino acids and no β-helical structure could be predicted with Phyre2 (for details see Supplementary Table S2).
All selected Klebsiella phages were then grouped based on gene homology and a conserved gene synteny into KP32viruses, KP34viruses, and KP36viruses and into groups containing only Klebsiella-specific phages similar to phage JD001 (belonging to Jedunavirus), similar to phage Menlow (belonging to Ackermannviridae), similar to phage ΦK64-1 (belonging to Alcyoneusvirus). Within each group, further subdivisions were proposed for the purpose of this study, based on the organization of the RBP gene cluster (number of RBPs, length of different genes, presence of anchor, or branching domains).
When there was one RBP, a domain in the N-terminus of a RBP was annotated as ‘anchor’ when there was at least an identity of 39% (BLASTP) over at least 166 residues starting from the N-terminus of the corresponding protein among phages belonging to the same group. These parameters were set empirically based on the shortest identity region found among all RBPs, specifically in the first RBP of phage IL33, belonging to KP32viruses group B (166 amino acids) and the identity% of the first RBP of phage Kp1. When more than one RBP was present, the anchor domain was annotated in the RBP in which also a T4gp10-like domain was detected. In the other RBP(s) the N-terminal conserved sequence was called ‘conserved peptide,’ which was also generally shorter than the anchor domains. To define consensus sequences of the anchor domains and conserved peptides, multiple sequence or pairwise alignment were used, since these structures are highly conserved among phages from the same group. To identify domains involved in the branching of RBPs, the sequences were analyzed by HHPred performing protein structure prediction⁵ (Zimmermann et al., 2018 (link)) in search for domains homologous to T4gp10 domain 2 and 3 as experimentally confirmed attachment sites (Prokhorov et al., 2017 (link)). WebLogos of the anchor domains and conserved peptides were created with the online available tool⁶ (Crooks et al., 2004 (link)).

Pectobacterium atrosepticum strain SCRI1043 was obtained from the SCRI bacterial collection. The bacteria were grown in a variety of media at 15°C or 27°C as described in Table 3. Pectin, polygalacturonic acid and arabinogalactan were added to two of the growth media in order to trigger different responses in P. atrosepticum gene transcription. Two different types of pectin, citrus pectin (cip) and cabbage pectin (cap) were used, as this could in principle affect the transcriptional response of the bacterium. The temperature of 15°C was selected based on optimal expression of key enzymes involved in the breakdown of the plant cell walls (e.g. pectate and pectin lyase). The temperature of 27°C was selected as the optimum growth temperature for P. atrosepticum [53 (link)]. In addition, gene expression levels at different growth phases were tested, as these may vary during plant infection. Thus, bacteria were sampled from both exponential and stationary phase in LB medium (Table 3). For leaf infiltration, overnight LB-cultures of P. atrosepticum grown at 27°C were pelleted and resuspended in 10 mM MgSO₄. Leaflets from potato cv. Bintje were vacuum infiltrated with a suspension of 10⁷ bacterial cells/ml for ~15 minutes under low vacuum using a water pump. Negative control leaflets were infiltrated with 10 mM MgSO₄ without bacteria. After infiltration, leaflets were placed on moist filter paper in Petridishes (3–5 leaflets per dish), and incubated at 18°C. This temperature was selected to mimic conditions at which P. atrosepticum optimally causes blackleg and soft rotting symptoms [13 ]. Samples were harvested 21 hours after infiltration, at which point the leaves showed clear rotting symptoms, flash-frozen in liquid nitrogen and kept at -80°C until RNA extraction.

The extraction method was optimized to enable industrial scale production and obtain further enrichment of pectic RG-I domains from bell pepper (bpRG-I) and carrot (cRG-I). Commonly used food processing methods were applied including aqueous extraction at 85 °C (bell pepper) or at 45 °C in the presence of Pectinex™ Ultra Mash (Novozymes, Bagsværd, Denmark) food-grade pectinolytic enzymes with pectin lyase and polygalacturonase as main activity (carrot); decanting and filtration to remove non-soluble residues; centrifugation to remove fat (decreaming, bell pepper), ultra- and dia-filtration to remove small molecules (<10 kDa), pasteurization and subsequent spray-drying of the RG-I enriched polysaccharide extracts.

In this study, we mainly referred to the activity assays for pectin lyase reported and the DNS method described in the literature and comprehensively selected the assay suitable for the pectin lyase enzyme activity in this study (Chen et al. 2022 , 2023 ; Zhang et al. 2022 ). Initially, 0.5% pectin was accurately weighed and dissolved in Tris-HCl buffer (50 mM, pH 8.0). The fermentation supernatant was appropriately diluted. Subsequently, 190 µL of the pectin solution was placed in a thermostatic reactor to preheat for 10 min, and 10 µL of sample was added to react for 10 min. 300 µL of DNS solution was added, boiled for 5 min and then cooled down immediately, and 200 µL of the reaction solution was taken to measure the absorbance value at 540 nm. One unit (U) of enzyme activity was defined as the amount of pectin lyase releasing 1 µmol of reducing sugar per minute under the conditions described above.
The optimal pH value of PMGL-Ba was measured by determining the activity between pH 5.0–11.0 in 60 °C. To determine pH stability of PMGL-Ba, purified enzymes were firstly incubated between pH 5.0–12.0 at 4 °C for 12 h. Next, residual activity of PMGL-Ba was further measured using above method. Buffers used were 100 mmol/L PBS buffer (pH 6.0), 50 mmol/L Tris-HCl buffer (pH 7.0–9.0) and Glycine-sodium hydroxide buffer(pH10-11). Optimal temperature of PMGL-Ba was investigated by measuring the activity in a range of 50–70 °C in pH 8.0. To determine the thermostability of PMGL-Ba, enzymes were pre-incubated at 30–90 °C for 1 h pH 8.0.

Gene mining for PMGL was performed with the help of enzyme resource mining software EnzymeMiner (Damborsky et al. 2020 (link)). Two high enzyme activity pectin lyases Q01172 and P94449 of medium alkaline and medium high temperature were used as probe sequences. Additionally, the protein sequences of three PMGLs, Q2TXS4, AY825251, and R9XVW6, were used as auxiliary references for sequence search. The initial set of sequences of potential PMGLs was obtained by performing iterative site-specific homology searches in the NCBI database. The key residues of PMGL were then used as filtering criteria to filter sequences with mismatched key residues in the homology search results. The final selection was based on the sequence similarity and sequence characterization provided by EnzymeMiner, the sequence similarity to the probe template (with a threshold value of 30-70%), the predicted degree of soluble expression of the protein, and the type of microorganism from which the protein was derived. A phylogenetic tree was constructed between the selected sequences and the probe sequences by neighbor-joining method using MEGA-X software to analyze the homology of the sequences. The protein structure of PMGL-Ba was modeled by SwissModel homology. Sequence comparison analysis was performed using Jalview. The basic physicochemical properties of Bacillus licheniformis pectin lyase were predicted using ProtParam.

Pectin methyl esterase (PME) activity was detected by coupled assays using the pectate lyase family 1 enzyme from Aspergillus niger (AnPL1, Megazyme: E-PCLYAN), an enzyme only capable of cleaving unmethylated galacturonan. Assays of 200 µL contained 0.035% pectin (various types supplied from Megazyme, K-PECID) or 1% poly-methylgalacturonan (Biosynth) in 100-mM potassium phosphate buffer at pH 7, 225 U of AnPL1 (Megazyme), and were supplemented with or without 0.5 nmol of PvCE15 or 1:100 of the PME NovoShape from Aspergillus aculeatus (University of Reading, UK). Pectate lyase product formation was observed by measuring the absorbance at 235 nm using a UV-Star UV transparent 96-well microplate (Greiner Bio-One) as described previously (46 ).