Identification of the amino acid position controlling the different enzymatic activities in walnut tyrosinase isoenzymes (jrPPO1 and jrPPO2)

Polyphenol oxidases (PPOs) are ubiquitously distributed among plants, bacteria, fungi and animals. They catalyze the hydroxylation of monophenols (monophenolase activity) and the oxidation of o-diphenols (diphenolase activity) to o-quinones. PPOs are commonly present as an isoenzyme family. In walnut (Juglans regia), two different genes (jrPPO1 and jrPPO2) encoding PPOs have been identified. In this study, jrPPO2 was, for the first time, heterologously expressed in E. coli and characterized as a tyrosinase (TYR) by substrate scope assays and kinetic investigations, as it accepted tyramine and L-tyrosine as substrates. Moreover, the substrate acceptance and kinetic parameters (kcat and Km values) towards 16 substrates naturally present in walnut were assessed for jrPPO2 (TYR) and its isoenzyme jrPPO1 (TYR). The two isoenzymes prefer different substrates, as jrPPO1 shows a higher activity towards monophenols, whereas jrPPO2 is more active towards o-diphenols. Molecular docking studies performed herein revealed that the amino acid residue in the position of the 1st activity controller (HisB1 + 1; in jrPPO1 Asn240 and jrPPO2 Gly240) is responsible for the different enzymatic activities. Additionally, interchanging the 1st activity controller residue of the two enzymes in two mutants (jrPPO1-Asn240Gly and jrPPO2-Gly240Asn) proved that the amino acid residue located in this position allows plants to selectively target or dismiss substrates naturally present in walnut.


Results and discussion
Genomic DNA extraction and cloning of the jrPPO2 gene. Genomic DNA (gDNA) was isolated from walnut leaves using a cetyltrimethylammonium bromide (CTAB) assisted cell lyses method, which produced a total yield of 250 ng gDNA (~ 40.000 base pairs)/g frozen plant material. The co-extraction of phenolic compounds substantially decreases downstream applicability (PCR) of DNA extracts. Thus, 2% (w/v) PVP (polyvinylpyrrolidone) was added to the DNA extraction buffer 42 as well as 20 mM sodium ascorbate to suppress PPO side reactions since in situ produced quinones oxidize DNA.
The predicted jrPPO2 gene 19 encompasses an N-terminal chloroplast transit peptide, an active domain and a C-terminal domain. Using Q5 High-Fidelity DNA polymerase and specific primers (Table S1) a ~ 1,700 base pair amplicon was obtained, cloned in the pENTRY-IBA51 vector and sequenced to reveal the sequence of the predicted jrPPO2. Compared to the sequence published by Martinez-Garcia et al. 19 , the gene sequenced herein contained the following mutations: Asp256Asn, Phe293Leu, Ser296Pro and Asp477Asn, which are all located www.nature.com/scientificreports/ on the surface of the protein and are a result of different habitats of sampled trees (Vienna, Austria (this study) vs. California 19 ). Minor structural variability has already been reported for jrPPO1, however, enzymatic activity, pH optimum and SDS dependent activation of the latent enzyme has been shown to remain unaffected by these variations 10 .
Based on the sequencing results, a second primer pair was designed binding to the starting region of the active domain and the end of the C-terminal domain (Table S1), which produced an amplicon corresponding to full-length jrPPO2 (active domain and C-terminal domain; Fig. S2). After cloning into a pGEX vector the construct was transformed into competent E. coli BL21 (DE3) cells and used for protein expression. The enzyme was expressed as a fusion protein with an N-terminal GST-tag, which on the one hand facilitates its purification and on the other hand increases the solubility and, thus, reduces the formation of inclusion bodies 21,43 . Expression of jrPPO1, jrPPO1-Asn240Gly, jrPPO2 and jrPPO2-Gly240Asn. As described previously for jrPPO1 10 and other plant PPOs 21,44 , expression at low temperatures (~ 20 °C) in combination with prolonged expression times and the usage of a nutrient-rich medium (2xYT) results in an increased overall yield. jrPPO2 produced the highest yield with 70 mg/l purified, latent enzyme, followed by jrPPO2-Gly240Asn (63 mg/l), jrPPO1 (41 mg/l) and jrPPO1-Asn240Gly (34 mg/l). All enzymes were expressed at a purity level of at least 95% (Fig. S6) and were stored in 50 mM Tris-HCl pH 7.5 and 200 mM NaCl and were immediately used for kinetic measurements.

Molecular mass determination. ESI-LTQ-MS revealed the masses of recombinant jrPPO1, jrPPO1-
Asn240Gly, jrPPO2 and jrPPO2-Gly240Asn. The crystal structure analysis of jrPPO1 9 exhibits one thioether bridge and two conserved disulfide bonds, which are, due to the similar spatial arrangement of the amino acids involved in the formation of the disulfide bonds and the thioether bridge, most probably also in vivo present in jrPPO2 (Fig. S1). The masses of jrPPO1, jrPPO2 and jrPPO2-Gly240Asn matched with the calculated masses corresponding to the formation of the thioether bridge and one of the two disulfide bonds being closed. The mass of jrPPO1-Asn240Gly indicated the formation of the thioether bridge and both disulfide bonds (Table 1 and Fig. S7) being closed. Varying numbers of closed disulfide bonds due to ESI-MS investigations have already been reported for jrPPO1 10 . The formation of the disulfide bonds during the recombinant expression process in E. coli is impeded by the reducing environment of the cytosol 45 . However, disulfide bonds can be present as an artifact of the electrospray ionization process, during which thiyl radicals, formed via one-electron oxidation of thiol groups, dimerize rapidly 46 . The thioether bridge is formed independently in the bacterial cytosol via an autocatalytic process after copper incorporation into the active center 47 . Thus, the disulfide bonds can be attributed to the ionization process whereas the thioether bridge is formed during the expression process.
Characterization of jrPPO2. jrPPO2 was characterized in terms of its pH optimum and activation by SDS using 1 mM dopamine (Fig. S8) as a substrate. Different pH values for maximum activity have been reported for plant PPOs ranging from pH 4.5 (sodium citrate buffer) 48 to pH 8.0 (sodium phosphate buffer) 49 . Thus, the pH dependence was assessed in increments of 0.5 pH units ranging from pH 3.0 to pH 8.0 (pH 3.0-pH 5.5: sodium citrate buffer, pH 6.0-pH 8.0: sodium phosphate buffer). The maximum activity was observed at pH 6.0 (Fig. 2), which follows the pH optimum of jrPPO1 (pH 6.0) 10 .
A general characteristic of plant PPOs is their latency 50 as activity can be measured only in the presence of an additional activator. SDS has been proven suitable in activating plant PPOs 10,21,51 and was previously shown to overcome their latency. Thus, the activation of PPOs is achieved with SDS molarities ranging from 0.35 mM 51 to 4.0 mM 21 . We tested the concentration-dependent activation of jrPPO1 with SDS molarities ranging from 0.5 to 5.0 mM. The highest activity was observed at 2.5 mM SDS (Fig. 2), compared to 2.0 mM for jrPPO1 10 . The respective pH optima and SDS optima of jrPPO1 and jrPPO2 were used for substrate scope assays and the kinetic measurements.
jrPPO2 was characterized kinetically using the monophenolic substrates tyramine and l-tyrosine and the diphenolic substrates dopamine and l-DOPA (Fig. S8). jrPPO2 showed activity towards both monophenolic and both diphenolic substrates and, therefore, was classified as a TYR. k cat (s −1 ) and K m (mM) values were determined for tyramine, l-tyrosine, dopamine and l-DOPA (Table 2).
k cat values were higher for the diphenols (k cat dopamine = 186 s −1 ; k cat l-DOPA = 132 s −1 ) compared to the monophenols (k cat tyramine = 9.14 s −1 ; Table 2). Moreover, the catalytic efficiency (k cat /K m ) of jrPPO2 was considerably higher for the less polar substrate tyramine (k cat /K m = 18.7 s −1 mM −1 ) compared to the carboxylic substrate L-tyrosine (k cat /K m = 0.69 s −1 mM −1 ), which held also true for the diphenolic substrates ( Table 2). www.nature.com/scientificreports/   S5) and the naphthoquinone juglone ( Fig. S5) were tested. Substrate-enzyme combinations leading to a visually detectable change in color within 24 hours were flagged as active, whereas substrate-enzyme combinations remaining colorless after this time were flagged as inactive (Fig. 3).
Kinetic measurements of jrPPO1 and jrPPO2 identify different substrate preferences. To further investigate the kinetic behavior of jrPPO1 and jrPPO2, k cat and K m values were determined for substrates that showed activity towards jrPPO1 and/or jrPPO2. Molar extinction coefficients were reported previously 52 or were determined herein (see supplementary information; Table S2). The flavonoid substrates (quercetin and taxifolin; Fig. S5) were assayed in a solution containing 10% DMSO due to their limited water solubility. The effects of 10% DMSO on the activities of jrPPO1 and jrPPO2 were assessed using dopamine. In the presence of 10% DMSO, jrPPO1 retained 72% activity and jrPPO2 retained 75% activity, compared to enzymatic tests without additional DMSO. Thus, both enzymes are similarly affected by the addition of 10% DMSO.
jrPPO1 and jrPPO2 were more active towards diphenols than towards the corresponding monophenols (Table 2), as in general reported for PPOs 10,17,21,44 . However, jrPPO2 showed a stronger preference for diphenols over monophenols than jrPPO1, as diphenolic and triphenolic substrates showed higher activity values (k cat value) and higher efficiency values (k cat /K m ratio) towards jrPPO2 than towards jrPPO1. The only exception was protocatechuic acid (Fig. S4), which was more active (higher k cat value) with jrPPO1. However, since the K m value for protocatechuic acid (Fig. S4) increased for jrPPO1 (compared to jrPPO2), it showed a substantially higher catalytic efficiency towards jrPPO2 (k cat /K m = 109 s −1 mM −1 ), compared to jrPPO1 (k cat /K m = 3.95 s −1 mM −1 ) ( Table 2). In contrast, all monophenolic substrates (4-hydroxyphenylacetic acid, coumaric acid, l-tyrosine, and tyramine) showed a higher turnover rate towards jrPPO1, compared to jrPPO2. Monophenolase/diphenolase activity ratios (k cat monophenol/k cat diphenol) of corresponding mono-and diphenols were higher for jrPPO1 than for jrPPO2. The activity ratio of tyramine/dopamine for jrPPO1 was 0.27, compared to 0.05 for jrPPO2 ( Table 2). The same trend held true for the monophenolase/diphenolase efficiency ratios ((k cat /K m ) monophenol/ (k cat /K m ) diphenol).
Thus, our data show that jrPPO1 favors monophenolic substrates, whereas jrPPO2 targets diphenolic substrates. This trend also correlates with the flavonoid substrates and was particularly pronounced for the diphenolic substrate taxifolin (Fig. S5), which was 19-times more active towards jrPPO2 (k cat = 19.6 s −1 ), compared to www.nature.com/scientificreports/ jrPPO1 (k cat = 1.04 s −1 ; Table 2). An asparagine in the position of the 1st activity controller residue (Asn240) has previously been proven to increase monophenolase activity 10,25 , which explains the higher activity (k cat values) of jrPPO1 towards monophenols, compared to jrPPO2. To clarify the molecular cause for the increased diphenolase activity of jrPPO2, compared to jrPPO1, docking studies were employed.
Docking studies illustrate the molecular cause of the different reactivities of jrPPO1 and jrPPO2. For the docking studies, a homology model of jrPPO2 was built using the SWISS-MODEL server 53,54 ( Fig. S3) and the crystal structure of jrPPO1 (PDB entry 5CE9) as a template. Molecular docking was performed for jrPPO1 as well as jrPPO2. Binding poses were calculated for all kinetically investigated substrates, which included the standard substrates tyramine, l-tyrosine, dopamine and l-DOPA and the natural substrates pyrogallol, protocatechuic acid, gallic acid, ethyl gallate, 4-hydroxyphenylacetic acid, coumaric acid, caffeic acid, quercetin and taxifolin. The results offered highly valuable information detailing the molecular basis for the different reactivities of jrPPO1 and jrPPO2. The homology model of jrPPO2 exhibited a high level of structural homology, compared to the crystal structure of jrPPO1 (RMSD = 0.487 Å). However, the amino acid in the position of the 1st activity controller residue represents a notable difference between the architectures of the active centers of jrPPO1 and jrPPO2 (Figs. S2 and S3). jrPPO1 features a Gly in this position, whereas jrPPO2 features a spatially more demanding Asn, which is protruding directly into the active center (Fig. S9).
The calculated binding poses clearly show that in jrPPO2 all diphenolic substrates are preferentially oriented in a lying down position (orienting the 3′-hydroxy group toward the di-copper center; Figs. 4, S10, S11), whereas, in jrPPO1, diphenolic substrates have to approach the di-nuclear center in an upright orientation (orienting the 4′-hydroxy group toward the di-copper center; Fig. 4). Orienting diphenolic substrates in a lying down position in jrPPO1 is prevented by Asn240, which overlaps with the tails of diphenolic substrates in jrPPO2. Alternatively, diphenolic substrates can be oriented in jrPPO2 in an upright position as well (data not shown). Thus, orienting substrates into the active center of jrPPO2 with the phenolic ring facing the di-copper center appears to be entropically more favorable, compared to jrPPO1, due to the spatially less demanding 1st activity controller residue (Gly240). This explains the significantly higher turnover rates and efficiency values of diphenolic substrates for jrPPO2, compared to jrPPO1 (Table 2).
In contrast, monophenolic substrates featuring a 4′-hydroxy group are oriented exclusively in an upright position in jrPPO1 and jrPPO2, as demonstrated by the molecular docking poses (Figs. 5, S10, S11). Thus, the entropic advantage of the more spacious active center of jrPPO2 does not come into effect for monophenolic substrates. Moreover, the asparagine present in the position of the 1st activity controller in jrPPO1 has been shown to facilitate monophenolase activity by aiding in the imperative abstraction of the phenolic proton from incoming monophenolic substrates 20 . The resulting phenolate substrate, carrying a negative charge, exhibits an increased affinity towards the positively charged di-copper center, compared to the corresponding not-dissociated phenol 20 . This was first demonstrated for VvPPOcs-3 25 , which features a glycine in the position of the www.nature.com/scientificreports/ 1st activity controller residue (Gly241). Semiquantitative in-gel activity tests demonstrated that the mutant VvPPOcs-3-Gly241Asn (1st activity controller: Asn241) showed increased activity rates towards the monophenolic substrates tyramine and p-tyrosol (compared to the native enzyme; 1st activity controller: Gly241) 25 , which is in accordance with our results. Moreover, it has been demonstrated, based on the crystal structure of the bacterial tyrosinase from Bacillus megaterium (BmTYR) 55 , which also features an Asn in the position of the 1st activity controller (His B1 + 1 = Asn205), that Asn205 forms a polar bond with the first CuB coordinating histidine (His B1 = His204) and, thereby, stabilizes His204 55 . In BmTYR, the N δ1 atom of the imidazole ring of His204 (His B1 ) is located at a distance of 2.7 Å from the amide group of Asn204 (1st activity controller residue). Similarly, the N δ1 atom of the imidazole ring of His239 (His B1 ) in jrPPO1 is located at a distance of 2.9 Å from the amide group of Asn240 (1st activity controller residue) 9 . Thus, in jrPPO1 Asn240 probably shows a stabilizing effect on His239. The combination of these effects explains the higher activity rates of monophenolic substrates with jrPPO1, compared to jrPPO2 (Table 2).
Docking studies were performed for jrPPO1-Asn240Gly und jrPPO2-Gly240Asn using tyramine and L-tyrosine (Fig. 7) and all substrates investigated during the previous docking experiments (Figs. 4, 5, S10-S13). Our data show that diphenolic substrates are oriented in a laying down position in jrPPO1-Asn240Gly (as observed for jrPPO2; Figs. 4, 7, S12 and S13). In contrast, in jrPPO2-Gly240Asn diphenolic substrates must approach www.nature.com/scientificreports/ the di-nuclear copper center in an upright position, since Asn240 now blocks substrates from orienting in the laying down orientation (Figs. 7, S12 and S13). Monophenolic substrates enter the active center of both mutants (jrPPO1-Asn240Gly and jrPPO2-Gly240Asn) in an upright orientation. This data show that the amino acid residue in the position of the 1st activity controller is responsible for the different substrate preferences of jrPPO1 (targeting monophenols) and jrPPO2 (targeting diphenols). The substrate scope is dependent on the amino acid present in the position of the 1st activity controller. Moreover, an asparagine in the position of the 1st activity controller increases monophenolase activity, whereas diphenolase activity is reduced (compared to the presence of glycine in the same position). Substituting asparagine with a spatially less demanding amino acid (such as glycine) increased the k cat value of dopamine considerably. Our results lead to the conclusion that jrPPO1 and jrPPO2 in vivo target different substrates and, thus, most probably fulfill different physiological tasks. The common appearance of plant PPOs as a family of isoenzymes suggests that they are involved in several cellular pathways, covering a diverse spectrum of functionalities. We hope that our work will inspire the deciphering of the different tasks assigned to PPOs, thereby illuminating their elusive reactivities.  www.nature.com/scientificreports/

Conclusions
In this study, jrPPO2 was, for the first time, heterologously expressed, purified and characterized. Activity tests using standard substrates (tyramine, l-tyrosine, dopamine, l-DOPA) clarified that jrPPO2 is a TYR, as it accepted l-tyrosine and tyramine. Moreover, substrate scope assays using 16 natural substrates showed a more expansive substrate scope for jrPPO2 as it accepted 4-hydroxybenzoic acid and gallic acid, compared to jrPPO1, which was inactive with 4-hydroxybenzoic acid and showed only marginal activity towards gallic acid. Kinetic parameters were determined for jrPPO2 and its isoenzymes jrPPO1, which pointed towards differences in substrate preference. jrPPO2 showed a higher catalytic efficiency for diphenols whereas jrPPO1 was more active on monophenols. Docking studies revealed that the amino acid in the position of the 1st activity controller can increase the activity towards monophenolic substrates, as it has previously been proposed, by stabilizing a conserved water molecule or reduce enzymatic activity towards diphenolic substrates by sterically impeding substrate orientation. The two mutants jrPPO1-Asn240Gly and jrPPO2-Gly240Asn proved the key role of the 1st activity controller as jrPPO1-Asn240Gly showed an enzymatic profile similar to jrPPO2, whereas jrPPO2-Gly240Asn resembled jrPPO1.
Our results demonstrate that, in vivo, different PPOs within the same plant target different substrates, which is achieved by the variability of one crucial amino acid residue (1st activity controller). This novel understanding of the functionality of PPO isoenzymes in plants will hopefully allow controlling their reactivity and, thereby, enhance the nutritional and economic value of plant products.

Materials and methods
Isolation of genomic DNA and cloning of the jrPPO2 gene. Walnut leaves were collected from naturally grown trees around Vienna and stored at -80 °C. Two g of frozen leaves were ground in liquid nitrogen. The frozen paste was mixed with 2 ml extraction buffer (100 mM HEPES, 0.5 M NaCl, 20 mM sodium ascorbate, 2% PVP and 1% cetyltrimethylammonium bromide (CTAB), pH 8.0) 42 . The mixture was incubated in a 70 °C water bath for one hour followed by centrifugation at 20.000 rpm for 10 min. The supernatant was extracted with 1 volume of phenol:chloroform:isoamyl alcohol (25:24:1, pH 7.8) and subsequently the aqueous layer was washed two times with 1 volume 100% chloroform. The aqueous layer was precipitated by adding 1 volume of EtOH (96%) and incubation at 0 °C for 2 hours. The pellet resulting from centrifugation at 20.000 rpm for 10 min at 4 °C was washed two times with EtOH (70%) at 0 °C, dried and resuspended in 100 µl TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). The quality and quantity of the DNA extract were checked by 0.6% agarose gel electrophoresis (Fig. S14).
The first pair of primers binding outside of the open reading frame of the active domain and the C-terminal domain was designed (using the NEB Tm calculator v1.12.0) from the sequence of jrPPO2 published previously 19 (Table S1). Q5 High-Fidelity DNA polymerase (NEB, Ipswich, USA) was used for the amplification and a ~ 1,700 base pair amplicon was produced (for detailed PCR setup see supplementary information). The PCR product was cloned into the pENTRY-IBA51 vector and sequenced to reveal the full-length sequence of jrPPO2. Thereafter, the gene encoding jrPPO2 (active and C-terminal domain) was amplified with the second pair of primers (designed based on the sequencing results; Table S1). Using Q5 High-Fidelity DNA polymerase a ~ 1,500 base pair amplicon was obtained, cloned into the pENTRY-IBA51 vector and again sequenced. The sequence-verified construct was sub-cloned into the open reading frame of a pGEX-6P-1 based expression vector using the Esp3I restriction enzyme (Thermo Fisher, Waltham, USA) and transformed into E. coli BL21 (DE3) cells.
Construction of the mutants jrPPO1-Asn240Gly and jrPPO2-Gly240Asn. The genes encoding jrPPO1 and jrPPO2, cloned into the pENTRY-IBA51 donor vector, respectively, served as templates for the mutagenesis experiments. Q5 High-Fidelity DNA polymerase was used to introduce the mutations into the sequence by back to back annealing primers (Table S1) with the forward primer carrying the desired mutation. T4 Polynucleotide Kinase (NEB) and T4 DNA Ligase (NEB) were used to create cyclic plasmids (pENTRY-IBA51). The open reading frames were then sub-cloned into the pGEX-6P-1 expression vector using the Esp3I restriction enzyme and expressed as described previously 10 (see supplementary information).

Molecular mass determination via mass spectrometry. Mass spectra of jrPPO1, jrPPO2, jrPPO1-
Asn240Gly and jrPPO2-Gly240Asn were measured on an LTQ Orbitrap Velos mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) equipped with a nanospray ion source using an ion transfer capillary temperature of 300 °C and an electrospray voltage of 2.1 kV. 5 µl of the sample was loaded on a trap column of an UltiMate 3000 nano HPLC-system (Dionex) using 0.1% trifluoroacetic acid. The separation was carried out at a flow rate of 300 nl/min on a C4 analytical column 50 cm × 75 µm Accucore C4, 150 Å, 2.6 µm (Thermo Fisher Scientific) using mobile phase A (2% acetonitrile, 0.1% formic acid and 98% H 2 O) and mobile phase B (0.1% formic acid, 20% H 2 O and 80% acetonitrile). Full MS scans were acquired in positive ion mode ranging from 400 to 2000 m/z at a resolution of 7,500 (FWHM at 400 m/z).
Characterization of jrPPO2, substrate scope assays and kinetic investigation of jrPPO1, jrPPO2, jrPPO1-Asn240Gly and jrPPO2-Gly240Asn. Kinetic measurements were performed in triplicates. Photometric measurements were all carried out on TECAN infinity M200 (Tecan, Salzburg, Austria) in 96 well plates at 25 °C using the latent enzyme and SDS as an activator. pH and SDS optima were determined for jrPPO2 using the diphenolic substrate dopamine. The highest activities were measured at pH 6.0 (50 mM sodium phosphate buffer) and 2.5 mM SDS. Kinetic measurements of jrPPO2 and jrPPO2-Gly240Asn were performed under these conditions by measuring the increase of the colored reaction products photometrically www.nature.com/scientificreports/ (Fig. S19). For jrPPO1 the optimal conditions were a pH value of 6.0 and 2.0 mM SDS, as published previously 10 . Identical conditions were used for the mutant jrPPO1-Asn240Gly.
To determine which substrates showed activity with jrPPO1 and/or jrPPO2, 100 µg of the purified, latent enzyme were mixed with 1 mM substrate in 50 mM sodium phosphate buffer and 2 mM (jrPPO1 and jrPPO1-Asn240Gly) or 2.5 mM (jrPPO2 and jrPPO2-Gly240Asn) SDS in 200 µl solution at 25 °C. Due to their limited solubility, the assays for the flavonoid substrates (kaempferol, quercetin, taxifolin, myricetin; Fig. S5) and juglone were performed using 0.1 mM substrate, 100 µg enzyme, 50 mM sodium phosphate buffer, 2 mM (jrPPO1 and jrPPO1-Asn240Gly) or 2.5 mM (jrPPO2 and jrPPO2-Gly240Asn) SDS and 10% DMSO in 200 µl at 25 °C. Substrates that exhibited no visually detectable change in color within 24 hours were flagged as inactive. A control assay was performed for each substrate containing 1 mM substrate in 50 mM sodium phosphate buffer and 2.5 mM SDS in 200 µl at 25 °C.
For calculating the kinetic parameters (k cat and K m value), the maximum reaction rate was measured at 7-8 different substrate concentrations in a total volume of 200 µl containing 50 mM sodium phosphate buffer at pH 6.0, 2 mM (jrPPO1 and jrPPO1-Asn240Gly) or 2.5 mM (jrPPO2 and jrPPO2-Gly240Asn) SDS and variable amounts of enzyme (Table S3). For quercetin and taxifolin, DMSO was added to a final concentration of 10% to increase the solubility of the substrates. The data were fitted to the Michaelis-Menten equation by non-linear curve fitting (OriginPro 8 software; Figs. S15-S18).
Molecular docking with jrPPO1 and jrPPO2. Molecular docking was performed using AutoDock Vina 56 . The crystal structure of jrPPO1 (PDB entry 5CE9) was prepared for molecular docking by adding missing side chains using Coot 57 . A homology model of jrPPO2 was built using the SWISS-MODEL server 53,54 . The exhaustiveness was set to 100 and 20 poses were calculated for each target and substrate (Tables S4 and S5). Structures of the substrates were obtained from PubChem and formatted into pdbqt files using AutoDock-Tools (ADT, v.1.5.6) 56 . Docking studies were performed with protonated, semi-protonated and deprotonated hydroxy-phenyl groups (generated by editing the substrate pdbqt files). Binding poses were searched in a grid box enclosing the two copper ions of the active site, the 1st and 2nd activity controller and Phe260 (Figs. S1 and S2). For jrPPO1, jrPPO1-Asn240Gly, jrPPO2 and jrPPO2-Gly240Asn the 1st activity controller residue (jrPPO1 and jrPPO2-Gly240Asn: Asn240, jrPPO2 and jrPPO1-Asn240Gly: Gly240), the 2nd activity controller residue (jrPPO1 and jrPPO1-Asn240Gly: Leu244, jrPPO2 and jrPPO2-Gly240Asn: Ile244) and Phe260, were defined as flexible residues. Poses that significantly deviated from the binding pose of l-tyrosine were flagged as 'unreasonable' poses. All visualizations were created using PyMOL 2.3 58 .