Insights into the Activity and Substrate Binding of Xylella fastidiosa Polygalacturonase by Modification of a Unique QMK Amino Acid Motif Using Protein Chimeras

Polygalacturonases (EC 3.2.1.15) catalyze the random hydrolysis of 1, 4-alpha-D-galactosiduronic linkages in pectate and other galacturonans. Xylella fastidiosa possesses a single polygalacturonase gene, pglA (PD1485), and X. fastidiosa mutants deficient in the production of polygalacturonase are non-pathogenic and show a compromised ability to systemically infect grapevines. These results suggested that grapevines expressing sufficient amounts of an inhibitor of X. fastidiosa polygalacturonase might be protected from disease. Previous work in our laboratory and others have tried without success to produce soluble active X. fastidiosa polygalacturonase for use in inhibition assays. In this study, we created two enzymatically active X. fastidiosa / A. vitis polygalacturonase chimeras, AX1A and AX2A to explore the functionality of X. fastidiosa polygalacturonase in vitro. The AX1A chimera was constructed to specifically test if recombinant chimeric protein, produced in Escherichia coli, is soluble and if the X. fastidiosa polygalacturonase catalytic amino acids are able to hydrolyze polygalacturonic acid. The AX2A chimera was constructed to evaluate the ability of a unique QMK motif of X. fastidiosa polygalacturonase, most polygalacturonases have a R(I/L)K motif, to bind to and allow the hydrolysis of polygalacturonic acid. Furthermore, the AX2A chimera was also used to explore what effect modification of the QMK motif of X. fastidiosa polygalacturonase to a conserved RIK motif has on enzymatic activity. These experiments showed that both the AX1A and AX2A polygalacturonase chimeras were soluble and able to hydrolyze the polygalacturonic acid substrate. Additionally, the modification of the QMK motif to the conserved RIK motif eliminated hydrolytic activity, suggesting that the QMK motif is important for the activity of X. fastidiosa polygalacturonase. This result suggests X. fastidiosa polygalacturonase may preferentially hydrolyze a different pectic substrate or, alternatively, it has a different mechanism of substrate binding than other polygalacturonases characterized to date.


Introduction
Pierce's disease (PD) of grapevine is caused by the gram negative plant pathogenic bacterium Xylella fastidiosa [1][2][3]. To induce disease, X. fastidiosa must be able to spread from one xylem element to another [4]. X. fastidiosa accomplishes this by enzymatically degrading the xylem pit membranes that separate adjacent elements; this degradation is primarily accomplished by an endopolygalacturonase that is encoded by a single gene in the bacterial genome [5][6][7][8][9]. Previous research showed that if the gene encoding X. fastidiosa polygalacturonase was disrupted, the resulting polygalacturonase deficient mutant was non-pathogenic in grapevines [5]. Additionally, work by Aguero et al. (2005) shows that a polygalacturonase inhibiting protein (PGIP) from pear is able to reduce PD severity when expressed in transgenic grapevines [10]. These results suggest that if grapevines expressed sufficient amounts of an inhibitor of X. fastidiosa polygalacturonase, then X. fastidiosa cells could not degrade the pit membranes separating xylem elements. This would prevent X. fastidiosa from systemically colonizing the plant and prevent PD symptom development.
Unfortunately, producing enough active soluble X. fastidiosa polygalacturonase has been a major impediment for performing polygalacturonase inhibition assays. Previous efforts in our lab and others to produce soluble active X. fastidiosa polygalacturonase have suffered from its insolubility in heterologous expression systems. In all recombinant protein expression systems tried to date, X. fastidiosa polygalacturonase is produced in an insoluble and inactive form. Fast protein liquid chromatography (FPLC) was utilized to attempt to solubilize and purify insoluble recombinant X. fastidiosa polygalacturonase produced in Escherichia coli, however, this method did not produce an enzymatically active form of X. fastidiosa polygalacturonase.
Another issue complicating the ability to produce enzymatically active X. fastidiosa polygalacturonase in vitro is the fact that X. fastidiosa polygalacturonase appears to have a unique a distinct substrate binding amino acid motif compared to other characterized active endopolygalacturonase enzymes. [6,[11][12][13][14]. Whereas X. fastidiosa polygalacturonase possesses all the conserved catalytic amino acids implicated in the hydrolysis of 1, 4-alpha-D-galactosiduronic linkages, it has very different amino acids in the substrate binding domains [6]. Specifically, the R/ (I/L/V)/K motif that is possessed by all other known active endopolygalacturonases, is QMK (380-382) in the X. fastidiosa polygalacturonase [11]. It was shown in previous studies with Aspergillus niger polygalacturonase that mutation of the R residue in the RIK (256-258) motif to an A results in only 14% polygalacturonase activity compared to the wild type polygalacturonase. More importantly, changing R256 to Q resulted in only 6.5% residual polygalacturonase activity [15,16]. Thus, it is possible that the presence of the Q residue in the X. fastidiosa polygalacturonase could result in substantially lower activity compared to other characterized polygalacturonases, or alternatively it might be reasonably active but has a different substrate specificity than most other polygalacturonases. The experiments conducted in this study focused on two questions: first, can we produce active, soluble X. fastidiosa polygalacturonase in vitro and second, do the altered amino acids in the substrate binding motif of the X. fastidiosa polygalacturonase result in reduced enzyme activity, or different substrate specificity. We decided to address both of these questions through the creation of a protein chimera in which the catalytic and substrate binding domains of X. fastidiosa polygalacturonase were exchanged with those in an active polygalacturonase enzyme from a different prokaryotic plant pathogen. Previous research has shown that protein chimeras can be useful for studying enzymatic activity and substrate binding motifs [17,18]. Agrobacterium vitis polygalacturonase is the same size as X. fastidiosa polygalacturonase, and is important in the virulence of A. vitis to grapevines [19,20]. Relevant to this study, a soluble active form of A. vitis polygalacturonase can be readily produced in recombinant E. coli expression systems [20,21].
The N-terminal and C-terminal regions of both A. vitis and X. fastidiosa polygalacturonase contain the greatest sequence variation while the middle (amino acids~290-390), which contains the active site and substrate binding amino acids, is highly conserved between the two. [6,20]. This suggests that the variable regions might be responsible for the inability to produce X. fastidiosa polygalacturonase in an active and soluble form in in vitro expression systems. Thus, a chimeric protein containing the variable N terminal and C terminal regions from the soluble and active A. vitis polygalacturonase, combined with the catalytic and substrate binding regions from the X. fastidiosa polygalacturonase might produce a soluble and biologically active chimeric protein in E.coli. Such a chimera would allow us to assess the catalytic site amino acids of X. fastidiosa polygalacturonase and confirm these amino acids are responsible for catalyzing the degradation of polygalacturonic acid substrate. Furthermore, producing a second chimera, including both the catalytic amino acids as well as the QMK substrate binding domain amino acids, would determine if, indeed, the QMK domain specific amino acids are enzymatically relevant for substrate binding. Moreover, if the X. fastidiosa polygalacturonase QMK motif was mutated to the RLK motif found in all other characterized active polygalacturonases, then a more biologically active enzyme might be observed. If this was the case these active chimeric proteins could function as surrogates for native X. fastidiosa polygalacturonase in in vitro enzyme inhibition assays.

Materials and Methods
Bacterial strains, plasmids, and PCR primers All bacterial strains, plasmids and PCR primers used in this study are presented in Table 1.
Polygalacturonase gene cloning X. fastidiosa was grown on solid PD3 [22] medium at room temperature for 7 days. Colony PCR amplification of the pglA gene (gene ID:1142827) was performed using Platinum taq polymerase (Invitrogen, Carlsbad, CA) and pglA specific primers XFPGF and XFPGRH with the following PCR parameters: 94°C (2 min) followed by 35 cycles of 94°C (1 min) 60°C (1 min) 72°C (2 min) and a final extension of 72°C (6 min). The resulting 1,635 bp PCR product was gel purified with the QIAquick gel extraction kit (Qiagen, Venlo, The Netherlands), cloned into pCR2.1 plasmid using the TOPO-TA cloning kit (Invitrogen, Carlsbad, CA) and transformed into E. coli TOP10 cells. The resulting plasmid, pCR2.1XFPG, was digested with NdeI and XhoI (NEB, Ipswich, MA) restriction enzymes and ligated with T4 DNA ligase (NEB, Ipswich, MA) into the NdeI and XhoI sites of pET30B (EMD Milipore, Billerica, MA) such that a 6X His tag sequence was present on the 3' end of the gene, creating pET30BXFPG. A. vitis strain S4 (kindly provided by T. Burr, Cornell, NY) was grown on solid 523 medium [23] at room temperature for 48 hours. Colony PCR amplification and cloning of the pehA gene (Gen-Bank: ACM36064.1) was performed exactly as stated above for X. fastidiosa except the pehA specific primers, AVPGF and AVPGRH, were used in the PCR reaction. The resulting 1,635 bp PCR amplicon was gel-purified as described above, cloned into pCR2.1 using the TOPO-TA cloning kit and transformed into E. coli TOP10 cells (Invitrogen, Carlsbad, CA). The resulting plasmid pCR2.1AVPG, was digested with NdeI and SalI restriction enzymes (NEB, Ipswich, MA) and ligated with T4 DNA ligase (NEB, Ipswich, MA) into the NdeI and XhoI sites of the pET30B (EMD Milipore, Billerica, MA) resulting in the plasmid pET30BAVPG.

X. fastidiosa/A. vitis chimera creation
The pBS-AvPGbx plasmid was generated as the backbone for the two chimeras. This plasmid was created in the following manner; the X. fastidiosa pglA gene was cloned into pBSSK via the shuttle vector pFLAG-ATS. Plasmid pET30bAVPG was digested with NdeI and XhoI restriction enzymes and the resulting full length A. vitis pehA gene was ligated into the NdeI and XhoI sites of pFLAG-ATS creating pFA-AvPG. The pFA-AvPG plasmid was then digested with BamHI and XhoI enzymes and ligated into the BamHI and XhoI sites of pBSSK, creating pBS-AvPGbx. To create the chimeric genes and ligate the products into the pET30b expression vector, an overlap extension PCR procedure was performed. Two sections of the pglA gene, X1 and X2, were PCR-amplified from plasmid pET30BXFPG with the following primers: the catalytic domain "X1" (877-1044 bp) was amplified with the XPG1R_Nco and ntOLF primers and the catalytic and substrate binding domain "X2" (877-1191 bp) was amplified with the XPG2R_Sph and ntOLF primers. Similarly, a portion 'A' of the pehA gene (536-880bp) from the pBS-AvPGbx plasmid was amplified with the APG1F_Cla and ntOLR primers. Both PCR reactions were performed using the following parameters: PCR was performed using Herculase II fusion DNA polymerase (Agilent, Santa Clara, CA) and the cycling parameters were 94°C (1 min) followed by 35 cycles of 94°C (15 sec) 60°C (15 sec) 72°C (1 min) and a final extension of 72°C (6 min). These two PCR reactions generate fragments with a region of 100% sequence identity between the fragments that can then be used in an overlap extension PCR which uses homologous recombination to individually fuse the X1 and X2 portions of the pglA gene to that of the "A" portion of the pehA gene piece. The overlap extension PCR using the above described primers was performed with Herculase II enzyme with the following cycling parameters: 94°C (  subsequently ligated into pET30b creating the chimeric expression plasmids pET30b-AX1APG and pET30b-AX2APG, respectively.

Point mutation creation
All mutants created in this study were generated by creating point mutations in the plasmid pET30b-AX2APG. Point mutations were created with the QuikChange Lightning Multi Site-Directed Mutagenesis Kit (Agilent, Santa Clara, CA) according to the manufacturer's instructions. Briefly, 50 ng of pET30b-AX2APG plasmid were incubated with 125 ng of the mutation primer pair in a reaction tube containing appropriate buffers and QuikChange Lightning enzyme. This reaction mixture was used in the following PCR reaction: 95°C (2 min) followed by 10 cycles of 95°C (20 sec) 60°C (10 sec) 72°C (4 min) and a final extension of 72°C (5 min). Following amplification, the samples were incubated with the restriction enzyme Dpn I. During this step any parental plasmids are degraded, whereas plasmids containing point mutations are not digested because these plasmids are not dam methylated and, thus, resistant to Dpn I digestion. The resulting plasmids with the introduced mutations were then transformed into E. coli and screened by sequencing with primer AVPGM to identify plasmids that contained genes having the desired mutations. Plasmids with the proper mutations were then transformed into BL21 (DE3) for subsequent protein expression. The following primers were used in the point mutation reaction described above to create the desired mutated plasmids. To create a glutamine to arginine mutation at position 380 the SXPGa1139g and ASXPGa1139g primer pair was used as described above to generate pET30b-AX2APG Q380R. To create an isoleucine mutation at position 381, the SXPGg1143a and ASXPGg1143a primer pair was used as described above to generate pET30b-AX2APG M381I. Lastly, to create the double mutant glutamine to arginine mutation at position 380 and methionine to isoleucine mutation at position 381, the SXPGa1139g_g1143 a and ASXPGa1139g_g1143a primer pair was used as described above to generate pET30b-AX2APG Q380R M381I. The nucleotide sequences of all plasmids were confirmed by sequencing with the T7 promoter and terminator primers, as well as, the AVPGM primer.

Protein expression and purification
Proteins containing the previously described mutations were expressed in the BL21 (DE3) E. coli strain. BL21(DE3) transformed with each plasmid was grown on an orbital shaker 250 RPM at 37°C for 16 hours in 500 ml Luria broth plus kanamycin 50 μg/ml (LB + kan50). Exponentially growing cells, OD600 = 0.8, were induced by the addition of 1mM IPTG and incubated on an orbital shaker set to 250 RPM and 37°C for 3 hours. Cells were harvested by centrifugation at 4,000 × g, suspended in 15 ml of lysis buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 10 mM imidazole pH 8.0, 0.25 mg/ml lysozyme), incubated on ice for 30 min. The cells were disrupted by sonication with a model 100 sonic dismembrator (ThermoFisher, Waltham, MA) at a power setting of 8 using six repetitions of ten second bursts. Cells were then centrifuged at 12,000 × g for 10 minutes at 4°C, and the soluble fraction was removed and incubated with 1 ml of NI-NTA agarose beads (Qiagen, Venlo, The Netherlands) for 1 hour at 4°C. The NI-NTA agarose beads were then loaded onto a chromatography column and washed with 15 ml of wash buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 50 mM imidazole pH 8.0). Proteins were eluted from the column in 500 μl aliquots of elution buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 250 mM imidazole pH 8.0). Fractions containing 58 kDa recombinant polygalacturonase, as determined by SDS polyacrylamide gel electrophoresis, were used in subsequent experiments.   column purified polygalacturonase enzyme adjusted where applicable to 5 U/ml (One unit will release 1.0 μmole of reducing sugar per min at pH 5.0 at 37°C). Reactions were incubated at 37°C for 1-16 hours. Polygalacturonase activity was determined by measuring the accumulation of reducing sugars using the 2-cyanoacetamide method [24]. The reaction was stopped by mixing 100ul of enzymatic reaction with 450 μl 100 mM borate buffer pH 9.0 and 100 μl of a 1% 2-cyanoacetamide solution and incubated at 95°C for 15 min. The solution was cooled to room temperature at which point the concentration of reducing sugars was estimated by measuring the absorbance of the solution at 276 nm.

X. fastidiosa/A. vitis protein chimeras
Two X. fastidiosa / A. vitis polygalacturonase chimeras (AX1A, AX2A) were generated to investigate if the X. fastidiosa polygalacturonase active site amino acids mediated the hydrolysis of polygalacturonic acid, and to determine the biological relevance of the unique QMK substrate binding site of X. fastidiosa polygalacturonase (Fig 1). The X. fastidiosa / A. vitis polygalacturonase chimera AX1A contained the N-terminal and C-terminal portions of A. vitis polygalacturonase and the X. fastidiosa polygalacturonase catalytic amino acids but did not include the unique QMK substrate binding motif of X. fastidiosa polygalacturonase. The AX1A chimera was constructed to specifically test if recombinant chimeric protein, produced in E. coli, is soluble and if the X. fastidiosa polygalacturonase catalytic amino acids are able to hydrolyze polygalacturonic acid. The AX2A chimera contained the Nterminal and C-terminal portions of A. vitis polygalacturonase and the X. fastidiosa polygalacturonase catalytic amino acids as well as the unique QMK substrate binding motif of X. fastidiosa polygalacturonase and was constructed to evaluate the ability of the unique QMK motif of X. fastidiosa polygalacturonase to bind to and allow the hydrolysis of polygalacturonic acid. Chimeras AX1A and AX2A were produced in an E. coli protein expression system and both chimeric proteins were present and purified from the soluble fraction. The elutions containing soluble AX1A and AX2A chimeras were then used in 2-cyanoacetamide reducing sugar activity assays to determine solubility and activity. The results of the AX1A chimera activity assay showed that the active site amino acids of X. fastidiosa polygalacturonase hydrolyzed 1, 4-alpha-D-galactosiduronic linkages of polygalacturonic acid (Fig 2). Additionally, the results of the AX2A chimera activity assay shows that the AX2A chimera displays a markedly reduced catalytic activity compared to the AX1A chimera (Fig 2). This suggests that the QMK binding motif of X. fastidiosa polygalacturonase is less efficient in mediating enzymatic activity as previous research suggested when a R to Q mutation in A. niger polygalacturonase resulted in only 6.5% residual activity [15,16].
To further investigate the role of the unique QMK substrate binding motif in X. fastidiosa polygalacturonase enzymatic activity, three mutants were constructed using the AX2APG chimera to determine if changing this motif to the conserved RIK motif would result in increased enzymatic activity (Fig 3).
The first point mutation constructed, AX2APG Q380R, changed the Q residue in the motif to a R residue, resulting in a RMK motif. The second point mutation, AX2APG M381I, changed the M residue to an I residue, resulting in a QIK motif. The third point mutation generated, AX2APG Q380R M381I, changed the Q residue to an R residue and the M residue to an I residue, resulting in the conserved RIK motif. The RIK motif was chosen, as this is the conserved substrate binding motif in A. vitis polygalacturonase. Each point mutation construct was expressed in an E. coli protein expression system, and each of the mutant proteins were isolated and purified from the soluble fraction (Figs 4 and 5). The results of the activity assays with the mutants were surprising in that none of the mutants showed any activity in 2-cyanoacetamide reducing sugar activity assays (Fig 6). This is exactly the opposite result that would be expected from previous polygalacturonase mutational studies which suggest a mutation of an R residue to a Q residue would reduce activity to 6.5% of wild type [15,16]. These results suggest that the QMK motif may be required for the architecture of the X. fastidiosa polygalacturonase binding cleft and subsequent enzymatic activity.
Interestingly, rhamnogalacturonase, another plant cell wall degrading enzyme responsible for degradation of "hairy pectin" where the rhamnogalacturonan has a large number of branched side chains possesses a K residue (K253 Aspergillus aculeatus numbering) that is thought to be involved in carbohydrate substrate binding. Specifically the K253 residue of rhamnogalacturonase is thought be involved in binding galacturonic acid as well as rhamnose in the rhamnogalacturonan chain [25,26]. This result led us to re-evaluate the amino acids of the active cleft of X. fastidiosa polygalacturonase to determine if X. fastidiosa polygalacturonase may have other substrate binding motifs located in different positions. Indeed, it was determined that there is in fact a RLK motif near the C-terminal end of the binding site. This motif is in a position where it could possibly interact with the substrate and is not found in this 2-cyanoacetamide reducing sugar assay measuring the increase in reducing ends produced via hydrolysis of PGA substrate produced by polygalacturonase enzymatic activity. None of the polygalacturonase mutant chimeras, AX2APG Q380R, AX2APG M380I, or AX2APG Q380R M381I, shows the ability to hydrolyze the PGA substrate. Only the AX2APG chimera containing the wild-type X. fastidiosa QMK motif shows enzymatic activity significantly greater than the negative control. Positive control is A. vitis polygalacturonase and negative control consists of enzyme mixture with no polygalacturonase added. Data was analyzed using one-way analysis of variance and multiple pairwise comparisons using the Tukey test P = 0.05 (SigmaPlot). Data are represented as the mean of three replicates and error bars represent the 95% confidence interval.
doi:10.1371/journal.pone.0142694.g006 position in any other polygalacturonases. While we cannot rule out the possibility that this chimera is constraining the substrate binding cleft in some manner, leading to reduced activity, the fact that AX2A is active and none of the mutants show activity supports our conclusion that the QMK motif is biologically important for the activity of X. fastidiosa polygalacturonase. These results also suggest that X. fastidiosa polygalacturonase could have a preference for a different pectic substrate or, at least, that it has a different manner of substrate binding than other polygalacturonase enzymes characterized to date.
In conclusion, these experiments showed that it was possible to generate X. fastidiosa / A. vitis polygalacturonase chimeras that are active and soluble proteins in E. coli expression systems. Additionally, the catalytic amino acids of X. fastidiosa polygalacturonase are able to hydrolyze 1, 4-alpha-D-galactosiduronic linkages of polygalacturonic acid. Some evidence is also presented that the X. fastidiosa polygalacturonase may possess unique substrate binding amino acids. These chimeras should be able to be used in phage panning experiments to select for inhibitors that target the active site of the enzymes.