TY - JOUR
AU - Sankaranarayanan, Rajan
AB - Abstract Xanthomonas oryzae pv oryzae (Xoo) causes bacterial blight, a serious disease of rice (Oryza sativa). LipA is a secretory virulence factor of Xoo, implicated in degradation of rice cell walls and the concomitant elicitation of innate immune responses, such as callose deposition and programmed cell death. Here, we present the high-resolution structural characterization of LipA that reveals an all-helical ligand binding module as a distinct functional attachment to the canonical hydrolase catalytic domain. We demonstrate that the enzyme binds to a glycoside ligand through a rigid pocket comprising distinct carbohydrate-specific and acyl chain recognition sites where the catalytic triad is situated 15 Å from the anchored carbohydrate. Point mutations disrupting the carbohydrate anchor site or blocking the pocket, even at a considerable distance from the enzyme active site, can abrogate in planta LipA function, exemplified by loss of both virulence and the ability to elicit host defense responses. A high conservation of the module across genus Xanthomonas emphasizes the significance of this unique plant cell wall–degrading function for this important group of plant pathogenic bacteria. A comparison with the related structural families illustrates how a typical lipase is recruited to act on plant cell walls to promote virulence, thus providing a remarkable example of the emergence of novel functions around existing scaffolds for increased proficiency of pathogenesis during pathogen-plant coevolution. The structural study of a rice cell wall–degrading esterase LipA from Xanthomonas oryzae pv oryzae, the causal agent of bacterial leaf blight disease, reveals a unique domain that is essential for LipA function in planta and is conserved across the phytopathogenic genus Xanthomonas. INTRODUCTION The coevolution of plants and their pathogens has given rise to elaborate attack-counterattack strategies. The first barrier faced by plant pathogens is the rigid and chemically complex host cell wall. The ability to secrete specific hydrolytic enzymes to sever different components of the plant cell wall is a major virulence attribute of phytopathogenic fungi and bacteria (Albersheim et al., 1969). While cellulases, xylanases, polygalacturonases, and pectate lyases degrade the main polysaccharide constituents of the plant cell walls, enzymes like pectin esterases cleave the ester cross-links between the polysaccharide fibrils and loosen the cell walls (Esquerre-Tugaye et al., 2000). Cell wall damage not only provides a point of entry to the pathogen but also acts as a mark of invasion that the plant can sense. Degradation products that are released by the action of cell wall–degrading enzymes on plant cell walls can induce potent innate immune responses of plants (Darvill and Albersheim, 1984; Ryan and Farmer, 1991). These defense responses include the synthesis of antimicrobial compounds, such as phytoalexins, and the strengthening of plant cell walls through deposition of callose, as well as programmed cell death reactions. Thus, it appears that the cell wall–degrading enzymes serve dual functions. They are required for virulence, but their activity also induces host defense responses. Successful microbial plant pathogens are able to overcome this difficulty because they have the capability to suppress these innate immune responses (Jha et al., 2005). The genus Xanthomonas includes bacteria that cause more than 300 different diseases on a wide variety of economically important crops, such as rice (Oryza sativa), tomato (Solanum lycopersicum), pepper (Capsicum annuum), cabbage (Brassica oleracea), beans (Phaseolus vulgaris), and lemon (Citrus limon). One member of this genus, Xanthomonas oryzae pv oryzae (Xoo) causes bacterial blight, a serious disease of rice. This bacterium employs a battery of enzymes that includes a lipase/esterase (LipA), cellulase (ClsA), xylanase (XynB), and cellobiosidase (CbsA) to degrade rice cell walls (Rajeshwari et al., 2005; Jha et al., 2007). These enzymes are secreted by a specialized apparatus called the type II secretion (T2S) system (Ray et al., 2000; Sun et al., 2005). The T2S system and the enzymes secreted through it are important for the virulence and pathogenesis of Xoo. Mutations in the T2S system cause a complete loss of virulence on rice, while mutations in the genes for individual T2S secreted proteins, such as LipA and ClsA, cause partial loss of virulence (Ray et al., 2000; Rajeshwari et al., 2005; Jha et al., 2007). A double mutant in the lipA and clsA genes results in a more severe loss of virulence, suggesting a redundancy in the functioning of these proteins. In addition to the above, mutations in the gene for an in planta expressed cellulase (EglXoB) affect Xoo virulence (Hu et al., 2007). Treatment of rice leaves and roots with LipA, ClsA, and CbsA proteins induces rice defense responses, such as callose deposition and programmed cell death. The actual elicitors of the defense responses appear to be soluble cell wall degradation products that are released following the action of these enzymes on rice cell walls (Jha et al., 2007). This is in sharp contrast with the previously described fungal xylanases, which have been shown to function as direct elicitors of defense responses through recognition of a particular sequence motif in the xylanase by a specific plant receptor (Ron and Avni, 2004). LipA is a 42-kD protein with tributyrin and Tween 20–degrading activity, a general indicator of lipase/esterase function (Rajeshwari et al., 2005). LipA homologs are present in several gram-negative bacteria, including all xanthomonads whose genomes have been sequenced. The LipA homologs are poorly characterized and categorized as conserved hypothetical proteins or putative secretory lipases, indicating that these proteins have a lipase/esterase activity different from the feruloyl esterases, pectin esterases, and cutinases that are involved in plant cell wall degradation (Vorwerk et al., 2004; Bauer et al., 2006). In an effort to understand the basis of LipA function, we solved the high-resolution crystal structure of LipA. Analysis of LipA interactions with β-octyl glucoside, a small molecule captured during ligand screening, led to the structure-based identification of a distinct ligand binding module in LipA. We created several point mutations in this module and confirmed its role in ligand binding, virulence, and elicitation of host defense responses using a battery of in vitro and in planta assays, demonstrating the essentiality of this unique module in Xoo pathogenesis. RESULTS General Features of LipA Structure LipA is an α/β hydrolase fold protein with a nine-stranded central mixed β-sheet surrounded by α-helices in a typical α/β hydrolase topology (Nardini and Dijkstra, 1999) and an additional N-terminal short strand and helix. The canonical catalytic triad residues Ser-176, His-377, and Asp-336 (Figure 1A Figure 1. Open in new tabDownload slide Three-Dimensional Structure of LipA. (A) Stereoview of the ligand-bound structure of LipA showing two bound molecules of β-octyl glucoside and the catalytic triad. The hydrolase domain is depicted in pink and the ligand binding domain in green. The ligand molecules (yellow) Ser-176, Asp-336, and His-377 are given in a stick representation. (B) Topology diagram of LipA. The β-sheets are shown as arrows, α-helices as cylinders, and 310 helices as rectangles. The αD set of helices form the ligand binding domain (green). (C) Residues lining the BOG-bound tunnel are shown. Electron density of the two BOG molecules is a 2Fobs-Fcalc map contoured at 1σ value. The amino acids are shown as yellow sticks. Active-site Ser-176 is highlighted. The residues from the LipA structure without BOG are superimposed and shown in gray. Figure 1. Open in new tabDownload slide Three-Dimensional Structure of LipA. (A) Stereoview of the ligand-bound structure of LipA showing two bound molecules of β-octyl glucoside and the catalytic triad. The hydrolase domain is depicted in pink and the ligand binding domain in green. The ligand molecules (yellow) Ser-176, Asp-336, and His-377 are given in a stick representation. (B) Topology diagram of LipA. The β-sheets are shown as arrows, α-helices as cylinders, and 310 helices as rectangles. The αD set of helices form the ligand binding domain (green). (C) Residues lining the BOG-bound tunnel are shown. Electron density of the two BOG molecules is a 2Fobs-Fcalc map contoured at 1σ value. The amino acids are shown as yellow sticks. Active-site Ser-176 is highlighted. The residues from the LipA structure without BOG are superimposed and shown in gray. ) are positioned similar to several other hydrolases. The nucleophilic S176 lies on a strand-turn-helix elbow, forming a G-X-S-X-G motif that is conserved in hydrolases (Jaeger et al., 1999). The distinctive feature of the LipA structure is the presence of a 108–amino acid domain present as an insertion between the β6 and β7 strands, corresponding topologically to the lid insertion position of triacylglycerol lipases (Figure 1B). This insertion domain consists of seven α-helices with one 310 helix. LipA exhibits low sequence identity (≤20%) with characterized hydrolase fold proteins. It shows 17% sequence identity, DALI Z-score of 31.5, and a root mean square deviation of 2.7 Å over 326 Cα atoms with Candida antarctica lipase CalA, indicating a high structural similarity despite low sequence similarity (Holm and Sander, 1998; Ericsson et al., 2008). CalA also has a lid domain equivalent in size to the LipA lid-like domain. A DALI search with the LipA lid-like domain neither identifies the CalA lid nor any other protein of known structure as a structural homolog. However, a manual structural superposition of the two proteins indicates a considerable fold similarity also in the lid region (Figure 2B Figure 2. Open in new tabDownload slide The LipA Ligand Binding Tunnel Has a Carbohydrate-Anchoring Pocket. (A) Cartoon representation of the ligand binding tunnel. The two black filled circles indicate the stretch of residues 28 to 36 for which the electron density is missing. (B) Superimposition of the lid domains of LipA (yellow) and CalA (blue). BOG and PEG are shown as yellow and blue sticks, respectively. Active-site Ser residues for LipA and CalA are shown in yellow and blue, respectively. (C) Important BOG1-specific interactions in the carbohydrate-anchoring pocket of LipA are main-chain mediated and shown as dashed lines with distance in Å. (D) Multiple sequence alignment of LipA ligand binding domain with homologous regions in other bacteria. Residues marked in green are amino acids lining the tunnel, and the red dots indicate the residues in the carbohydrate-anchoring pocket that were mutagenized. The black dots represent 39– to 45–amino acid inserts in the sequences of the corresponding proteins that are not shown in the figure. The black cylinders indicate α-helices. Figure 2. Open in new tabDownload slide The LipA Ligand Binding Tunnel Has a Carbohydrate-Anchoring Pocket. (A) Cartoon representation of the ligand binding tunnel. The two black filled circles indicate the stretch of residues 28 to 36 for which the electron density is missing. (B) Superimposition of the lid domains of LipA (yellow) and CalA (blue). BOG and PEG are shown as yellow and blue sticks, respectively. Active-site Ser residues for LipA and CalA are shown in yellow and blue, respectively. (C) Important BOG1-specific interactions in the carbohydrate-anchoring pocket of LipA are main-chain mediated and shown as dashed lines with distance in Å. (D) Multiple sequence alignment of LipA ligand binding domain with homologous regions in other bacteria. Residues marked in green are amino acids lining the tunnel, and the red dots indicate the residues in the carbohydrate-anchoring pocket that were mutagenized. The black dots represent 39– to 45–amino acid inserts in the sequences of the corresponding proteins that are not shown in the figure. The black cylinders indicate α-helices. ). The sequence identity between the two domains, when considered without the hydrolase domains, is only ∼9% with a root mean square deviation of 3.2 Å over 92 aligned Cα atoms. Other structural hits for the entire LipA are mostly acyl-amino acid peptidases (Protein Data Bank [PDB] code: 1VE6) and dipeptidyl peptidases, (PDB code: 2D5L), which do not possess a similar lid-like domain, and the structural similarity is limited to the hydrolase domain only (see Supplemental Table 1 online) (Bartlam et al., 2004; Ito et al., 2006). An interesting feature in the structure is the presence of at least two Ser-His conserved clusters on the face opposite to the region that contains the canonical catalytic triad. These may be nonfunctional clusters, but their position and partial similarity to general hydrolase active sites is certainly worth noting (see Supplemental Figure 1 online). A Distinct Ligand Binding Domain We screened for putative ligands of LipA using detergents and fatty acid additives that could mimic its in planta substrate(s) using cocrystallization. LipA cocrystallized with the glycoside detergent β-octyl glucoside (BOG) in the same crystallization condition as that of the wild type. The cocrystal structure showed two molecules of bound BOG (Figure 1A). One molecule (referred to as BOG1) has B-factors, indicative of relative mobilities of the atoms, in the range of 18 to 23 Å2, while the other molecule (BOG2) is loosely bound, indicated by higher B-factors (45 to 50 Å2). An electron density map contoured around the ligands is shown in Figure 1C. The ligand acyl chains, placed very close to each other (6.9 Å), reveal a 30-Å tunnel in the LipA structure that passes adjacent to the active-site residues and ends very close to the outer surface of the protein (Figure 2A). The BOG2 glucose moiety hangs out of the tunnel facing the solvent. The proximity of the BOG1 terminal methyl group with the Ser-176 active site residue (3.8 Å) strengthens the idea that this tunnel in LipA could be involved in substrate binding. The putative entry side opening is very broad (20 Å), and a narrow exit point between Ile-287, Val-290, and Ser-220 is also evident. Root mean square deviation upon Cα superimposition of the wild-type and ligand-bound LipA structures is only 0.29 Å, indicating almost no structural changes upon ligand binding (Figure 1C). The side chains also superimpose well between the two structures, indicating a relatively rigid pocket employed for binding BOG. The lid-like domain may also not exhibit any domain motion with respect to the hydrolase-fold upon ligand binding, unlike large movements seen in conventional lid domains of lipases during interfacial activation (Nardini and Dijkstra, 1999). An interesting feature common to both wild-type and BOG-bound structures is the absence of electron density for the region from residues 28 to 36 (Figure 2A, black dots) in the vicinity of the BOG2 site, indicating that these residues are very flexible or unstructured, possibly due to the absence of the natural substrate. This sequence is conspicuously absent in other closely related Xanthomonas strains (see Supplemental Figure 2 online). Considering the position of the missing fragment, which marks the base of the entrance to the tunnel, this region might be involved in holding a long-chain substrate and positioning its entry into the tunnel. The glucoside moiety of BOG1 interacts with the main chain atoms of LipA at the extreme end of the tunnel, ∼15 Å away from the active-site Ser, in a pocket made of three Gly residues and a few other polar residues (Figure 2C). To ascertain whether the pocket would confer carbohydrate specificity, we looked at the possibility of several other sugars besides β-d-glucose occupying the pocket and found that except β-d-xylose, all other sugars that were examined face clashes from the residues surrounding the pocket (see Supplemental Table 2 online). BOG1 β-d-glucose O2 is held, albeit loosely, by main-chain oxygens of Asn-228 and Ser-218, O4 is held by O: Trp-219, and O6 by O: Gly-231 (Figure 2D). Important intramolecular interactions keep Gly-231 and Trp-219 in position even in the LipA structure without BOG, indicating that the pocket is held in place and that a case of substrate-induced fit may not be applicable. Structural coordinates of various sugar conformers with the same pucker as β-d-glucopyranose were superimposed in silico with the glucosyl moiety of the BOG bound in LipA (see Supplemental Table 2 online). Epimers of β-d-glucose (i.e., β-d-mannose and galactose) have axially located O2 and O4, respectively, and face a severe hydrophobic clash when superimposed on the BOG1 sugar ring with the same pucker. Similarly, steric hindrances toward the axial O1 exclude the possibility of binding to α-d- sugars. Isothermal titration calorimetric (ITC) experiments confirmed these structural predictions. The calorimetric binding of LipA versus BOG fits into a one-site model (Figure 3A Figure 3. Open in new tabDownload slide LipA Exhibits BOG Binding and Esterase Activity. (A) Binding isotherm of LipA titrated against BOG using ITC. Top panel, raw titration curve; bottom panel, heats fitted by nonlinear regressional curve fitting using one site binding model. (B) Presence of a zone of clearance indicates LipA activity on short-chain triacylglycerides (1) C4 (tributyrin) and (2) C6 (tricaproin), while no activity is seen on (3) C8 (tricaprylin). Holes punched on the right side contain LipA, and the ones on the left side contain buffer. (C) Loss of esterase activity in the S176A active-site mutant protein. Tributyrin clearance activity of (1) buffer; (2) LipA (wild-type) protein; and mutant LipA proteins (3) S176A, (4) N228W, (5) G231A, (6) G231I, (7) G231F, and (8) G221I. Plates were photographed after 2 h of incubation at room temperature. Figure 3. Open in new tabDownload slide LipA Exhibits BOG Binding and Esterase Activity. (A) Binding isotherm of LipA titrated against BOG using ITC. Top panel, raw titration curve; bottom panel, heats fitted by nonlinear regressional curve fitting using one site binding model. (B) Presence of a zone of clearance indicates LipA activity on short-chain triacylglycerides (1) C4 (tributyrin) and (2) C6 (tricaproin), while no activity is seen on (3) C8 (tricaprylin). Holes punched on the right side contain LipA, and the ones on the left side contain buffer. (C) Loss of esterase activity in the S176A active-site mutant protein. Tributyrin clearance activity of (1) buffer; (2) LipA (wild-type) protein; and mutant LipA proteins (3) S176A, (4) N228W, (5) G231A, (6) G231I, (7) G231F, and (8) G221I. Plates were photographed after 2 h of incubation at room temperature. ), while that of LipA versus β-octyl galactoside could be fitted into a sequential two-site model, indicating that the two modes of binding are different. Since the overall binding affinities can be used as indicators of the strength of binding, we compared LipA versus BOG K d (76.7 μM) to LipA versus β-octyl galactoside K d (739.5 μM). The 10-fold reduction in the binding affinity of β-octyl galactoside compared with BOG is in accordance with our proposition that the tunnel has a sugar-anchoring pocket with a predetermined, though moderate, affinity for β-d-glucose. Conspicuously, the rest of the tunnel is lined with several hydrophobic residues that trace the tunnel from the entrance to the sugar-anchoring pocket. Presence of this hydrophobic pocket suggests that the moderate specificity conferred by the few hydrogen bonds on the sugar moiety of the ligand is sustained by extensive hydrophobic interaction of the acyl chain with the rest of the residues lining the LipA tunnel. Glycoside ligand binding by LipA using recognition pockets for both carbohydrate and acyl chain components provides a rationale for the lack of detection of free sugar binding in ITC assays and in ligand screening experiments. The BOG-bound LipA structure may not be directly relevant in the in planta context since BOG is a nonhydrolyzable compound that has not been identified as a component of plant cell walls. However, the BOG complex provides valuable insights into the possible mode of natural ligand binding and the selectivity role of LipA. LipA-Like Substrate Recognition Evolved in Genus Xanthomonas A search for LipA sequence homologs using BLAST in the nonredundant database of the National Center for Biotechnology Information (NCBI) identifies a wide range of proteo- and actinobacteria with one or more LipA-like proteins in their genomes (with an E-value of 10−7 and lower). A structure-based sequence alignment and a phylogenetic tree was generated using these sequences (see Supplemental Data Set 1 and Supplemental Figure 3 online) using CalA as an outgroup. LipA sequences of Xanthomonas strains and Xylella fastidiosa, the closely related plant pathogen, cluster together with a high bootstrap value. A small subset of these LipA homologs, consisting solely of the genus Xanthomonas and the organisms demarcated as the blue group in Supplemental Figure 3 online, contains a LipA-like ligand binding domain. Interestingly, despite a high conservation of residues surrounding the tunnel region identified in the LipA-like proteins in this subset, it appears that the carbohydrate recognition is present only in genus Xanthomonas. This proposition is based on our structural analysis indicating Gly-231 as the crucial residue for the sugar ring positioning. Any substitution at this position would protrude into the pocket and abolish sugar anchoring. Gly-231 is invariant in all Xanthomonas LipA proteins with one exception (Figure 2D). Xanthomonas campestris pv campestris (strain ATCC 33913) XCC2374 has Thr in place of Gly-231 of LipA, although the strain has an additional LipA-like protein XCC2957 that retains Gly-231. A LipA-like protein in the soil bacterium Ideonella, with 27% identity to LipA, has been functionally characterized recently (Shinohara et al., 2007). Interestingly, it has been shown to degrade β-hydroxy palmitic (C16) acid methyl ester. We generated a homology model of this protein based on the LipA structure (see Supplemental Figure 4 online). The model indicates the presence of a longer hydrophobic tunnel lacking a carbohydrate binding site in the LipA-like protein from Ideonella and is in agreement with its substrate specificity. Structural superposition of LipA with CalA illustrates that the polyethylene glycol (PEG) molecule bound in CalA structure occupies a very different ligand binding pocket (Figure 2B). The analogous region in Xoo LipA is packed with hydrophobic amino acids, and the PEG molecule will have substantial clashes with them. In addition, the CalA region that superimposes on the LipA carbohydrate-anchoring pocket is predominantly occupied by the main chain of a loop, indicating that there is no room for such a pocket in CalA. Therefore, it is clear that the pocket in LipA is unique in nature with a carbohydrate-anchoring site located far away from the solely acyl binding pocket of CalA. This finding, along with the phylogenetic study, advocates for LipA-like proteins to be grouped as a distinct class of cell wall–degrading esterases. LipA Exhibits Esterase and Not Lipase Activity Movement of a small flap-like lid in lipases is implicated in the phenomenon of interfacial activation in response to highly hydrophobic triacylglycerides (Verger, 1997). Esterases act on less hydrophobic substrates and generally lack lid domains (Jaeger et al., 1999). When present, the lid remains permanently open, such as in fungal feruloyl esterases (Hermoso et al., 2004). The lid is absent in certain lipases, cutinases, acetylxylan esterases, and cholesterol esterases (Martinez et al., 1992; Chen et al., 1998; Ghosh et al., 1999; van Pouderoyen et al., 2001). The presence of an unusually large seven-helical domain in place of the usual lid and the absence of any evident movement upon ligand binding in LipA prompted us to check whether LipA has lipase or esterase activity. We found that LipA can degrade smaller chain length triacylglycerides like C4 (tributyrin) and C6 (tricaproin) but lacks activity on C8 (tricaprylin) or longer triacylglycerides (Figure 3B). It is negative for triolein hydrolyzing activity assessed by the olive oil-rhodamine B assay, for which all true lipases are positive. LipA shows maximum enzymatic activity on p-nitrophenyl (pNP) butyrate (C4) compared with pNP-acetate (C2), pNP-hexanoate (C6), and higher chain length pNP esters (see Supplemental Figure 5A online). The specific activity plot of LipA with pNP-butyrate (see Supplemental Figure 5B online) also indicates the esterase activity as the uniform sigmoidal plot is indicative of an absence of interfacial activation (Martinelle et al., 1995). Moreover, this experiment suggests that the natural ligand of LipA is likely to contain aliphatic chains as is also indicated by the presence of several hydrophobic residues in the tunnel leading to the carbohydrate-anchoring site from the catalytic Ser-176. Point Mutations in LipA and Assessment of in Vitro Activity To validate our structural prediction that the tunnel present in LipA structure is involved in substrate recognition, we generated several point mutations in LipA. Gly-231 was identified as a crucial residue mediating main chain interaction of LipA with BOG. The close proximity of this amino acid with the sugar ring of BOG suggests that the smallest replacement at this position would have a severe effect on LipA action; therefore, Gly-231 was mutated to Ala, Ile, and Phe. The residue Asn-228 was mutated to Trp to block the tunnel just below the carbohydrate-anchoring pocket, which would protrude into the acyl-chain binding region. Gly-221, another invariant Gly located in the sugar-anchoring pocket (Figure 2D), was mutated to an Ile. However, Gly-221 has no direct interaction with the sugar ring of BOG (Figure 2C). When tested for tributyrin degrading activity, all these mutants show zone of clearance (Figure 3C). It clearly indicates that a small substrate, such as tributyrin, would not require occupying the tunnel up to the sugar pocket that is located ∼15 Å away from the active-site Ser-176. We also checked for enzyme activity in a S176A mutant of LipA protein, and absence of a zone of clearance on tributyrin plates indicates that the LipA active-site indeed employs the nucleophilic Ser-176. Verification of BOG Binding to LipA and Its Mutants The in vitro binding of BOG to LipA or its mutant proteins was assessed using ITC. LipA, upon titration with BOG, shows a moderate binding affinity with a K d value of 76.7 μM (Figure 3A). S176A mutant LipA protein has K d values of BOG binding similar to that of wild-type LipA, indicating that the tunnel for BOG binding is intact despite a loss in enzymatic activity. All three Gly-231 mutant proteins of LipA display reduced binding of BOG with G231F mutant protein of LipA showing an almost 25-fold reduction from wild-type enzyme (Table 1 Table 1. Binding Parameters and Thermodynamics of BOG Titration with Wild-Type LipA and Mutants Calculated Using ITC Protein . K a (M−1) . K d (μM)a . ΔHapp (Kcal mol−1) . TΔS (Kcal mol−1) . ΔG° (Kcal mol−1) . LipA 1.3 × 104 ± 2.9 × 102 76.7 −5.3 ± 0.05 0.5 −4.8 S176A 1.2 × 104 ± 4.5 × 102 83.3 −3.8 ± 0.07 1.8 −2.0 N228W 8.8 × 103 ± 9.2 × 102 113.6 −2.8 ± 0.15 2.6 −0.2 G231A 1.7 × 103 ± 2.2 × 102 588.2 −0.1 ± 1.0 −5.8 5.7 G231I 1.3 × 103 ± 3.5 × 102 769.2 −0.1 ± 2.3 −6.6 6.5 G231F 5.1 × 102 ± 6.2 × 103 1960.8 −0.01b −100.9 100.9 G221I 8.4 × 103 ± 8.1 × 102 119.0 −4.8 ± 0.30 0.6 −4.2 Protein . K a (M−1) . K d (μM)a . ΔHapp (Kcal mol−1) . TΔS (Kcal mol−1) . ΔG° (Kcal mol−1) . LipA 1.3 × 104 ± 2.9 × 102 76.7 −5.3 ± 0.05 0.5 −4.8 S176A 1.2 × 104 ± 4.5 × 102 83.3 −3.8 ± 0.07 1.8 −2.0 N228W 8.8 × 103 ± 9.2 × 102 113.6 −2.8 ± 0.15 2.6 −0.2 G231A 1.7 × 103 ± 2.2 × 102 588.2 −0.1 ± 1.0 −5.8 5.7 G231I 1.3 × 103 ± 3.5 × 102 769.2 −0.1 ± 2.3 −6.6 6.5 G231F 5.1 × 102 ± 6.2 × 103 1960.8 −0.01b −100.9 100.9 G221I 8.4 × 103 ± 8.1 × 102 119.0 −4.8 ± 0.30 0.6 −4.2 a K d = 1/K a. b Very high error. Open in new tab Table 1. Binding Parameters and Thermodynamics of BOG Titration with Wild-Type LipA and Mutants Calculated Using ITC Protein . K a (M−1) . K d (μM)a . ΔHapp (Kcal mol−1) . TΔS (Kcal mol−1) . ΔG° (Kcal mol−1) . LipA 1.3 × 104 ± 2.9 × 102 76.7 −5.3 ± 0.05 0.5 −4.8 S176A 1.2 × 104 ± 4.5 × 102 83.3 −3.8 ± 0.07 1.8 −2.0 N228W 8.8 × 103 ± 9.2 × 102 113.6 −2.8 ± 0.15 2.6 −0.2 G231A 1.7 × 103 ± 2.2 × 102 588.2 −0.1 ± 1.0 −5.8 5.7 G231I 1.3 × 103 ± 3.5 × 102 769.2 −0.1 ± 2.3 −6.6 6.5 G231F 5.1 × 102 ± 6.2 × 103 1960.8 −0.01b −100.9 100.9 G221I 8.4 × 103 ± 8.1 × 102 119.0 −4.8 ± 0.30 0.6 −4.2 Protein . K a (M−1) . K d (μM)a . ΔHapp (Kcal mol−1) . TΔS (Kcal mol−1) . ΔG° (Kcal mol−1) . LipA 1.3 × 104 ± 2.9 × 102 76.7 −5.3 ± 0.05 0.5 −4.8 S176A 1.2 × 104 ± 4.5 × 102 83.3 −3.8 ± 0.07 1.8 −2.0 N228W 8.8 × 103 ± 9.2 × 102 113.6 −2.8 ± 0.15 2.6 −0.2 G231A 1.7 × 103 ± 2.2 × 102 588.2 −0.1 ± 1.0 −5.8 5.7 G231I 1.3 × 103 ± 3.5 × 102 769.2 −0.1 ± 2.3 −6.6 6.5 G231F 5.1 × 102 ± 6.2 × 103 1960.8 −0.01b −100.9 100.9 G221I 8.4 × 103 ± 8.1 × 102 119.0 −4.8 ± 0.30 0.6 −4.2 a K d = 1/K a. b Very high error. Open in new tab ). Therefore, the presence of the tunnel is crucial for BOG binding, which is not affected in the context of the inactive S176A mutant. N228W and G221I mutant proteins of LipA bind to BOG with affinities similar (less than twofold reduction) to that of wild-type LipA, suggesting that these changes in the sugar binding pocket do not have a significant effect on BOG binding. LipA Ligand Binding Domain Is Essential for Virulence on Rice The Xoo LipA-deficient strain BXO2001 shows reduced virulence on rice compared with BXO43, the wild-type strain (Rajeshwari et al., 2005). We transferred expression constructs for wild-type LipA and various point mutants into BXO2001. Expression of LipA protein in the culture supernatants of these Xoo strains was confirmed using polyclonal rabbit anti-LipA antibodies (see Supplemental Figure 6 online). Upon inoculation on rice, the strain expressing the S176A mutant protein of LipA exhibits reduced virulence (at a level that is comparable to the BXO2001 strain), confirming that the enzymatic activity of LipA is essential for optimal levels of virulence on rice. Remarkably, the Xoo strains expressing G231A, G231I, G231F, and N228W mutant proteins of LipA exhibit a virulence deficiency similar to that observed in the case of BXO2001. However, the strain expressing the G221I mutant protein of LipA exhibits wild-type levels of virulence (Figure 4 Figure 4. Open in new tabDownload slide Virulence Phenotypes of LipA Mutants. (A) Xoo strains were inoculated on rice leaves, and lesion lengths were measured after 7 d. At least 10 leaves were used per strain. Similar results were obtained in another independent experiment. Error bars represent sd. Values marked with an asterisk are significantly different from the values obtained for BXO43 at P < 0.05 in a Student's two-tailed t test for independent means. BXO43 (wild-type Xoo), BXO2001 (LipA− mutant), and various strains that express either wild-type LipA or mutant LipA proteins from the pHM1 plasmid in the BXO2001 background are shown. (B) Photographs of lesions caused by (1) BXO43, (2) BXO2001, and (3) pG231F. The latter two strains exhibit reduced disease symptoms (brown lesions) compared with BXO43. Figure 4. Open in new tabDownload slide Virulence Phenotypes of LipA Mutants. (A) Xoo strains were inoculated on rice leaves, and lesion lengths were measured after 7 d. At least 10 leaves were used per strain. Similar results were obtained in another independent experiment. Error bars represent sd. Values marked with an asterisk are significantly different from the values obtained for BXO43 at P < 0.05 in a Student's two-tailed t test for independent means. BXO43 (wild-type Xoo), BXO2001 (LipA− mutant), and various strains that express either wild-type LipA or mutant LipA proteins from the pHM1 plasmid in the BXO2001 background are shown. (B) Photographs of lesions caused by (1) BXO43, (2) BXO2001, and (3) pG231F. The latter two strains exhibit reduced disease symptoms (brown lesions) compared with BXO43. ). A reduction in virulence caused by either blocking the acyl chain binding region beneath the sugar anchor site (N228W) or obstructing the sugar-anchoring pocket (Gly-231 mutants) specifically proves that the tunnel is indeed involved in natural substrate binding. Virulence proficiency of the G221I mutation indicates that this does not affect binding of the natural substrate in the host as indicated by in vitro binding studies also. Loss of Induction of Rice Innate Immunity LipA treatment elicits rice innate immune responses that are considered as a reaction to LipA-mediated degradation of rice cell walls (Jha et al., 2007). A marker of plant defense responses, in general, is the cell wall fortification by rapid deposition of a β-1,3-glucan polymer called callose around the site of pathogen entry (Bestwick et al., 1995). LipA is known to induce callose deposition at the site of infiltration in rice leaves, and we investigated the ability of various LipA mutants to induce callose deposition. Wild-type and mutant LipA proteins were purified and infiltrated into rice leaves. Callose deposits appear as fluorescent spots upon staining with aniline blue and observation under UV light. Infiltration with the mutant LipA proteins S176A, G231A, G231I, G231F, and N228W results in background levels of callose deposition, similar to that observed in buffer-treated leaves, while leaves infiltrated with the LipA-G221I mutant exhibit levels of callose deposition that are no different from those observed in leaves infiltrated with the strain expressing wild-type LipA (Figure 5 Figure 5. Open in new tabDownload slide LipA Mutant Proteins Are Deficient at Induction of Defense Response–Associated Callose Deposition in Rice Leaves. (A) to (D) Light microscope images (×10 resolution; bars = 100 μm) of rice leaves infiltrated with buffer (A), wild-type LipA (B), S176A LipA (C), and G231F LipA (D). Purified proteins were infiltrated into rice leaves and stained for callose deposition. The callose deposits appear as sharp bright fluorescent spots as against the dull white spots representing the trichomes. (E) Number of callose deposits in leaf zones infiltrated with buffer, wild-type LipA, or various mutant proteins. Mean and sd were calculated for number of callose deposits from a leaf area of 0.60 mm2. Data were collected from four leaves in each experiment (three experiments indicated as ExpI, ExpII, and ExpIII) and four different viewing areas from the infiltrated region of each leaf. Asterisks indicate that the values obtained after treatment with G221I are significantly different at P < 0.05 compared with wild-type LipA values from the same experiment. Figure 5. Open in new tabDownload slide LipA Mutant Proteins Are Deficient at Induction of Defense Response–Associated Callose Deposition in Rice Leaves. (A) to (D) Light microscope images (×10 resolution; bars = 100 μm) of rice leaves infiltrated with buffer (A), wild-type LipA (B), S176A LipA (C), and G231F LipA (D). Purified proteins were infiltrated into rice leaves and stained for callose deposition. The callose deposits appear as sharp bright fluorescent spots as against the dull white spots representing the trichomes. (E) Number of callose deposits in leaf zones infiltrated with buffer, wild-type LipA, or various mutant proteins. Mean and sd were calculated for number of callose deposits from a leaf area of 0.60 mm2. Data were collected from four leaves in each experiment (three experiments indicated as ExpI, ExpII, and ExpIII) and four different viewing areas from the infiltrated region of each leaf. Asterisks indicate that the values obtained after treatment with G221I are significantly different at P < 0.05 compared with wild-type LipA values from the same experiment. ). Localized programmed cell death (PCD) is another important host defense response to pathogen attack (Pennell and Lamb, 1997). Rice roots were treated with the purified wild-type and mutant LipA proteins, stained with propidium iodide (PI) and examined by confocal laser scanning microscopy. PI is excluded from live cells, and its internalization is indicative of cell death. Rice roots treated with wild-type LipA take up PI, and the PI staining material is dispersed within the cell. This is indicative of DNA fragmentation caused by PCD. S176A mutant protein-treated roots showed absence of PCD. Similarly, the Gly-231 mutants and the N228W mutant protein treatments resembled buffer-treated roots with no evidence of LipA-induced PCD (Figure 6 Figure 6. Open in new tabDownload slide LipA Mutant Proteins Are Deficient at Induction of Defense Response–Associated PCD in Rice Roots. Rice roots were treated with purified proteins (either wild type or mutant LipA), stained with PI, and examined under a confocal microscope to assess the extent of DNA fragmentation. The control buffer-treated roots (A) exhibit a prominent cell wall–associated autofluorescence but no internalization of PI into the cells. Treatment with either wild-type (B) or G221I (H) LipA resulted in cell death (intake of PI) accompanied by dispersed intracellular staining, which is indicative of nuclear fragmentation. No cell death is seen in roots treated with S176A (C), N228W (D), G231A (E), G231I (F), or G231F mutant proteins (G). Bars = 20 μm. Figure 6. Open in new tabDownload slide LipA Mutant Proteins Are Deficient at Induction of Defense Response–Associated PCD in Rice Roots. Rice roots were treated with purified proteins (either wild type or mutant LipA), stained with PI, and examined under a confocal microscope to assess the extent of DNA fragmentation. The control buffer-treated roots (A) exhibit a prominent cell wall–associated autofluorescence but no internalization of PI into the cells. Treatment with either wild-type (B) or G221I (H) LipA resulted in cell death (intake of PI) accompanied by dispersed intracellular staining, which is indicative of nuclear fragmentation. No cell death is seen in roots treated with S176A (C), N228W (D), G231A (E), G231I (F), or G231F mutant proteins (G). Bars = 20 μm. ). Treatment of rice roots with the three purified Gly-231 mutant proteins (Gly-231 mutated to Ala, Ile, or Phe) show similar in planta phenotypes, proving that substrate binding in the natural context would be perturbed by any bigger residue substitution that might protrude into the sugar binding pocket. The N228W mutation affects LipA function in planta, indicating that it disrupts binding to the natural substrate. However, this mutation does not affect BOG binding, possibly because BOG might be smaller than the natural substrate. The Gly-221 mutant protein is able to induce PCD. The lack of any in planta phenotypes associated with the G221I mutation suggests that this change can be tolerated, possibly because of the presence of this residue on a less-packed region. G231F and N228W Mutant Structures Two mutant proteins of LipA (G231F and N228W) were cocrystallized with BOG at the same concentration as used for obtaining wild-type LipA-BOG cocrystals. The 2Fobs-Fcalc map contoured at 1σ before the inclusion of the mutated residue in the refinement clearly shows the protrusion of the bulkier residue in the binding pocket. G231F and N228W mutant protein structures could be superimposed with wild-type LipA with a root mean square deviation of 0.10 and 0.38 Å, respectively for 387 Cα atoms. Both structures showed an absence of BOG density in the lid-like domain, clearly due to the obstruction of the carbohydrate pocket caused by G231F (Figure 7A Figure 7. Open in new tabDownload slide The LipA Ligand Binding Tunnel Is Blocked in the G231F and N228W Mutant Proteins of LipA. The mutant structures have been superimposed with the wild-type LipA structure in both figures. Both mutant and wild-type structures are shown as green cartoons and superimpose completely except at the site of mutations. Electron density is shown only around the mutant residues that occlude the region occupied by BOG1 (shown as white stick) in the wild-type structure. (A) An unbiased 2Fobs-Fcalc map contoured at 1σ demonstrates disruption of the glucosyl ring binding site in the G231F mutant. (B) N228W obstructs the tunnel as shown in the unbiased 2Fobs-Fcalc map contoured at 1σ electron density around the mutant residue. The mutant structures have been superimposed with the wild-type LipA structure in both figures. Figure 7. Open in new tabDownload slide The LipA Ligand Binding Tunnel Is Blocked in the G231F and N228W Mutant Proteins of LipA. The mutant structures have been superimposed with the wild-type LipA structure in both figures. Both mutant and wild-type structures are shown as green cartoons and superimpose completely except at the site of mutations. Electron density is shown only around the mutant residues that occlude the region occupied by BOG1 (shown as white stick) in the wild-type structure. (A) An unbiased 2Fobs-Fcalc map contoured at 1σ demonstrates disruption of the glucosyl ring binding site in the G231F mutant. (B) N228W obstructs the tunnel as shown in the unbiased 2Fobs-Fcalc map contoured at 1σ electron density around the mutant residue. The mutant structures have been superimposed with the wild-type LipA structure in both figures. ) or a blocking of the tunnel region by Trp due to N228W mutation (Figure 7B). Together, the mutant structures confirm that the phenotypes associated with these mutants are a direct consequence of blocking the pocket and not due to any major structural changes in the protein. DISCUSSION The plant environment imposes a strong selective pressure on bacterial pathogens for the evolution of a repertoire of divergent and specialized proteins that promote colonization and survival within their hosts. Gain of new functions in proteins is frequently accomplished by modulation of preexisting functionally versatile folds (Thornton et al., 1999; Goldstein, 2008). The α/β hydrolase fold is one of the most promiscuous folds, found in numerous hydrolytic enzymes of diverse substrate specificity and exhibiting extensive sequence variability across several genomes (Ollis et al., 1992; Orengo et al., 1997; Hegyi and Gerstein, 1999; Holmquist, 2000). Esterases and lipases constitute a large group of hydrolase fold proteins. Several insertions and deletions in the basic hydrolase scaffold are found in nature, each change evolving toward hydrolyzing esters found in the local habitats of the pertinent organisms, thereby altering the substrate specificity (Nardini and Dijkstra, 1999). The lid domain of true lipases is a specific adaptation for long-chain triacylglycerols. Association of the catalytic hydrolase domain with additional nonlid noncatalytic domains for specialized substrate binding has been seen in prolyl oligopeptidases and brefeldin A esterase (Wei et al., 1999; Szeltner and Polgar, 2008). Our Xoo LipA crystal structure reveals a hitherto unidentified mode of substrate binding in a canonical α/β hydrolase, fine-tuning esterase activity for a plant-associated function. Structural comparison with CalA clearly demonstrates that the lid domain has been functionally converted into a plant cell wall–degrading esterase by engineering a distinct substrate recognition tunnel. This is a remarkable example of acquisition of a specialized function for increased proficiency in plant pathogenesis using an existing scaffold to act on a complex polysaccharide-rich milieu. The BOG-bound structure of LipA defines a clear path and space for substrate entry in the ligand binding domain, an 18-Å-wide opening that narrows to 8 Å near the active site and remains so up to the end of the tunnel. The BOG complex, although obtained fortuitously, suggests that the natural ligand of LipA is very likely to have a stereochemistry resembling monosaccharide-alkyl chain-ester linkage, since the Gly-231 residue found to be important for BOG–LipA interaction also plays a crucial role in the biological activity of LipA. The fact that the mutation of Asn-228 does not affect BOG binding while it disrupts in planta activity of LipA shows that there are differences between BOG and the natural ligand of LipA. Also supporting this architecture of the natural ligand is the fact that β-octyl galactoside shows much weaker binding to LipA in ITC experiments compared with BOG, indicating that LipA has a higher specificity for β-d-glucoside compared with β-d-galactoside. Therefore, mechanistically, LipA might act on an amphiphilic molecule with a glucose (or perhaps, xylose) moiety attached to a long (substituted) acyl chain (or aryl ring) of a length of 16 to 18 carbons (∼30 Å) with an ester bond situated ∼15 Å from the sugar ring. Presence of an aliphatic chain in the natural ligand is indicated by the stretch of hydrophobic residues lining the LipA ligand binding tunnel just below the carbohydrate anchor site as well as by the activity of LipA on pNP acyl esters. LipA could act on alkyl ester cross-links between long polysaccharide chains, which need to be cleaved to make the polysaccharide chains accessible to other degradative enzymes. LipA may also chew the same chain at several points and thus may be able to accept substrates of varying lengths. Such a compound can be predicted to be a part of the lignin component of the rice cell walls, since short alkyl chains and alkyl groups esterified to aryl moieties are reported in the lignocellulosic cross-links that connect the polysaccharide chains (reviewed in Buranov and Mazza, 2008). Several cyclo-alkyl rings linked to glucose, for example, furostanol glucosides, have been identified from plant cell walls and are possible candidates for LipA action (Arthan et al., 2006). A comparison with the inhibitor-bound structures of other hydrolases in the PDB indicates that, upon binding to the substrate binding tunnel in LipA, the carboxylate part of the ester would probably occupy the tunnel while the alcohol part would hang out of the enzyme. The alcohol part of the ester can be of varied characteristics since the LipA tunnel opens very wide toward the BOG2 binding site. Existing knowledge of the precise ester-linked polymers in plant cell walls, especially rice, is limited owing to their enormous complexity (Carpita, 1996; Somerville et al., 2004; Knox, 2008). However, as our understanding of the architecture of the plant cell wall improves, clarity on the selectivity aspect of LipA function mediated by the substrate binding pocket will also increase. Our study clearly demonstrates that the disruption of either the catalytic Ser or the sugar-anchoring site 15 Å away leads to the abrogation of LipA in planta functions. Therefore, LipA function requires grabbing a cell wall component in the ligand binding tunnel and cleaving the ester linkage at the active site, illustrating the existence of a crucial functional interplay between these two distinct regions of the molecule. The cell wall damage and/or release of soluble elicitors is a cue for induction of host innate defense responses, and a loss of LipA activity leads to an inability to provoke these responses. This substantiates the earlier observations showing that heat-inactivated LipA failed to induce rice innate immunity (Jha et al., 2007). Therefore, the result of LipA action, and not the LipA molecule itself, is required for eliciting rice responses like PCD and callose deposition. Contrastingly, fungal xylanases can elicit plant defense responses through physically binding to a plant receptor (Ron and Avni, 2004). LipA-mediated Xoo pathogenesis is dependent on the presence of the substrate recognition module. It is highly probable that the whole genus Xanthomonas employs this mode of substrate recognition for proficient pathogenesis. The carbohydrate binding domain of LipA is very distinct compared with the carbohydrate binding modules in the Carbohydrate Active Enzymes database (Cantarel, 2008). In general, noncatalytic carbohydrate binding modules are reported to enhance the activity of their catalytic counterparts by facilitating substrate proximity and increasing the efficiency of substrate binding in plants (Boraston et al., 2004). Our data clearly show that the LipA substrate binding domain is an essential modular noncatalytic domain appended to the catalytic hydrolase to recognize alkyl glycoside substrates where the ester bond is located quite far away (15 Å) from the sugar binding site. Taking into account the absence of structural homologs for the glycoside recognition domain and its conservation across the Xanthomonas group, LipA-like proteins constitute a new class of cell wall–degrading enzymes with a unique mode of substrate recognition. Identification or generation of such distinct plant biomass-degrading enzymes is an active area of research (Gilbert et al., 2008). Addition of LipA-like enzymes to the existing cell wall–degrading enzymatic preparations provides a new dimension for the production of clean eco-friendly biofuels from agricultural waste and other sources of plant material. The LipA substrate binding domain may have been acquired by a common ancestor of Xanthomonas and Xylella strains from evolutionarily distant but ecologically coexisting nonpathogenic bacteria like the soil/plant-associated Burkholderia and fine-tuned to act on carbohydrate-linked carboxylic esters. Diverse complex polysaccharides esterified to largely uncharacterized phenolics and aliphatics dominate the plant and soil interface environments, perhaps enforcing evolution of such specializations in the hydrolase fold. Our study offers a glimpse into a plant habitat-specific adaptation of the common esterase activity and provides a solid platform to study secretory esterases from several other important pathogenic bacteria. It might also provide an opportunity to create a battery of small molecule inhibitors for this family of enzymes, which could be developed for potential application in crop fields. METHODS Purification and Crystallization The overexpression, purification, and crystallization of LipA was reported by Aparna et al. (2007). Briefly, for overexpression of LipA in its secreted form, a pHM1 expression construct in which the gene is cloned under the constitutive lac promoter in Xoo strain BXO2008, was used (primers listed in Supplemental Table 4 online). The enzyme was purified to homogeneity from the culture supernatant of BXO2008 using cation exchange and size exclusion chromatography and crystallized at a concentration of 5 mg mL−1 at 298K in two types of monoclinic crystals differing in unit cell dimensions. Plate-shaped type I crystals were obtained in 48% PEG 400, 0.10 M MES, pH 6.0, and diffracted to 1.89 Å resolution. The unit cell parameters were found to be a = 93.1, b = 62.3, c = 66.1 Å, and β = 90.8°. The rod-shaped type II crystals have unit cell parameters of a = 103.6, b = 54.6, c = 66.3Å, and β = 92.6°. The crystals were obtained from a reservoir solution containing 12% PEG 6000 and 0.10 M MES, pH 6.7, and diffracted to 1.86 Å resolution (Aparna et al., 2007). LipA structure was solved by the Multiple Isomorphous Replacement method. The type II crystals, being more reproducible, were used for heavy atom derivatization. A high-throughput crystallization facility (Alchemist screen making system and Minstrel High-throughput crystal imaging system [Rigaku]; Matrix Hydra eDrop [Thermo Fisher Scientific]) was used to prepare crystallization solutions and generate several LipA crystals for heavy atom derivatization. Isomorphous crystals soaked in heavy atoms (Hampton) like 10 mM K2PtCl2 for 2 to 15 min and 10 mM Sm (NO3)2 soaks of 20 to 25 min were used for heavy atom substitution. Long soaks of 1 h were used for the Hg salt thimerosol to obtain substitution. N-Terminal Sequencing LipA sequence was predicted to contain a signal-peptide cleavage site between residues 29 and 30 by the software SIGNAL-P. LipA was purified as indicated above, resolved on a 10% SDS-PAGE gel, and electrotransferred using ECL semidry transfer unit (Amersham Biosciences) on a polyvinylidene fluoride membrane. The band of 39 kD corresponding to LipA detected after Ponceau Red staining of the blot was excised and subjected to Edman N-terminal sequencing using a Procise cLC N-terminal amino acid sequencer (Applied Biosystems) as per the manufacturer's instructions. N-terminal sequencing confirmed the cleavage of a 29–amino acid signal peptide, and the corresponding electron density could not be detected. Electron density for the first residue after the signal peptide (residue 1) and the region from residues 28 to 36 was not detected in the LipA structure solved using either crystal forms. Data Collection, Structure Determination, and Refinement The heavy atom soaked crystal data (see Supplemental Table 3 online) were collected on an in-house MAR Research MAR345dtb image plate detector using Cu Kα x-rays of wavelength 1.54 Å generated by a Rigaku RU-H3R rotating-anode generator. Data were processed using the HKL package (Otwinowski and Minor, 1997). The SOLVE run with four Pt, one Sm, and one Hg data sets gave a figure of merit 0.45 and Z score 40 (Terwilliger and Berendzen, 1999). Phase extension and improvement was performed using DM (Cowtan, 1994). RESOLVE could build only 150 Ala residues and 30 side chain residues out of 397 expected amino acids (Terwilliger, 2003). Iterative rounds of chain building using the experimental map for the rest of the 217 residues and RESOLVE runs for localized loop building were performed. The structure was refined using CNS. Structure visualization and a part of model building were done using the software O (Jones et al., 1991). The final structure was refined to R/Rfree values of 16.1/18.7% at 1.86 Å resolution. Data collection and refinement statistics are listed in Table 2 Table 2. Data Collection and Refinement Statistics . LipA . LipA-BOG . G231F . N228W . Data collection statistics     Unit cell parameters a = 103.7, b = 54.7, c = 66.3 Å, β = 92.6° a = 98.3, b = 64.9, c = 66.0 Å, β = 93.0° a = 103.6, b = 54.7, c = 66.3 Å, β = 92.7° a = 97.9, b = 64.1, c = 66.4 Å, β = 93.0°     Resolution (Å) 25–1.86 (1.93–1.86)a 25–2.1 (2.18–2.1) 25–2.1 (2.18–2.1) 25–2.1 (2.18–2.1)     Unique reflections 30,946 (2,969) 23,231 (1,599) 21,555 (1,898) 23,336 (2,053)     Completeness (%) 99.0 (95.8) 99.5 (95.8) 98.5 (88.2) 97.1 (86.2)     Rsym 3.8 (11.6) 6.2 (18.7) 10.9 (23.4) 5.4 (31.5)     Redundancy 3.5 (3.2) 3.1 (2.1) 3.0 (2.7) 2.7 (2.0)     I/(σ)I 30.5 (8.2) 11 (2.4) 10.2 (3.6) 14.5 (1.7) Refinement statistics     R (%) 16.1 18.8 17.9 20.1     Rfree(%)b 18.7 23.2 22.5 24.5     Mean B factor (Å2)         Protein 9.4 23.3 12.7 34.4         Water 24.9 34.6 24.5 40.8         Ligand – 36.6 – –     Root mean square deviation in         Bond distances (Å) 0.005 0.006 0.005 0.006         Bond angles (°) 1.251 1.253 1.250 1.257     No. of residues 387 390 387 387     No. of ligands 0 2 0 0     No. of atoms 3,470 3,292 3,328 3,154         Protein 2,941 2,946 2,948 2,947         Water 529 306 380 207         Ligand 0 40 0 0 Ramachandran plot statistics     Residues in most favored regions 297 (89.7%) 294 (88.6%) 297 (89.5%) 300 (90.6%)     Residues in additionally allowed regions 32 (9.7%) 36 (10.8%) 33 (9.9%) 29 (8.8%)     Residues in generously allowed regions 0 0 0 0     Residues in disallowed regions 2 (0.6%) 2 (0.6%) 2 (0.6%) 2 (0.6%) . LipA . LipA-BOG . G231F . N228W . Data collection statistics     Unit cell parameters a = 103.7, b = 54.7, c = 66.3 Å, β = 92.6° a = 98.3, b = 64.9, c = 66.0 Å, β = 93.0° a = 103.6, b = 54.7, c = 66.3 Å, β = 92.7° a = 97.9, b = 64.1, c = 66.4 Å, β = 93.0°     Resolution (Å) 25–1.86 (1.93–1.86)a 25–2.1 (2.18–2.1) 25–2.1 (2.18–2.1) 25–2.1 (2.18–2.1)     Unique reflections 30,946 (2,969) 23,231 (1,599) 21,555 (1,898) 23,336 (2,053)     Completeness (%) 99.0 (95.8) 99.5 (95.8) 98.5 (88.2) 97.1 (86.2)     Rsym 3.8 (11.6) 6.2 (18.7) 10.9 (23.4) 5.4 (31.5)     Redundancy 3.5 (3.2) 3.1 (2.1) 3.0 (2.7) 2.7 (2.0)     I/(σ)I 30.5 (8.2) 11 (2.4) 10.2 (3.6) 14.5 (1.7) Refinement statistics     R (%) 16.1 18.8 17.9 20.1     Rfree(%)b 18.7 23.2 22.5 24.5     Mean B factor (Å2)         Protein 9.4 23.3 12.7 34.4         Water 24.9 34.6 24.5 40.8         Ligand – 36.6 – –     Root mean square deviation in         Bond distances (Å) 0.005 0.006 0.005 0.006         Bond angles (°) 1.251 1.253 1.250 1.257     No. of residues 387 390 387 387     No. of ligands 0 2 0 0     No. of atoms 3,470 3,292 3,328 3,154         Protein 2,941 2,946 2,948 2,947         Water 529 306 380 207         Ligand 0 40 0 0 Ramachandran plot statistics     Residues in most favored regions 297 (89.7%) 294 (88.6%) 297 (89.5%) 300 (90.6%)     Residues in additionally allowed regions 32 (9.7%) 36 (10.8%) 33 (9.9%) 29 (8.8%)     Residues in generously allowed regions 0 0 0 0     Residues in disallowed regions 2 (0.6%) 2 (0.6%) 2 (0.6%) 2 (0.6%) a Values in parentheses are for the highest-resolution shell. b Throughout the refinement, 5% of the total reflections were kept aside for Rfree. Open in new tab Table 2. Data Collection and Refinement Statistics . LipA . LipA-BOG . G231F . N228W . Data collection statistics     Unit cell parameters a = 103.7, b = 54.7, c = 66.3 Å, β = 92.6° a = 98.3, b = 64.9, c = 66.0 Å, β = 93.0° a = 103.6, b = 54.7, c = 66.3 Å, β = 92.7° a = 97.9, b = 64.1, c = 66.4 Å, β = 93.0°     Resolution (Å) 25–1.86 (1.93–1.86)a 25–2.1 (2.18–2.1) 25–2.1 (2.18–2.1) 25–2.1 (2.18–2.1)     Unique reflections 30,946 (2,969) 23,231 (1,599) 21,555 (1,898) 23,336 (2,053)     Completeness (%) 99.0 (95.8) 99.5 (95.8) 98.5 (88.2) 97.1 (86.2)     Rsym 3.8 (11.6) 6.2 (18.7) 10.9 (23.4) 5.4 (31.5)     Redundancy 3.5 (3.2) 3.1 (2.1) 3.0 (2.7) 2.7 (2.0)     I/(σ)I 30.5 (8.2) 11 (2.4) 10.2 (3.6) 14.5 (1.7) Refinement statistics     R (%) 16.1 18.8 17.9 20.1     Rfree(%)b 18.7 23.2 22.5 24.5     Mean B factor (Å2)         Protein 9.4 23.3 12.7 34.4         Water 24.9 34.6 24.5 40.8         Ligand – 36.6 – –     Root mean square deviation in         Bond distances (Å) 0.005 0.006 0.005 0.006         Bond angles (°) 1.251 1.253 1.250 1.257     No. of residues 387 390 387 387     No. of ligands 0 2 0 0     No. of atoms 3,470 3,292 3,328 3,154         Protein 2,941 2,946 2,948 2,947         Water 529 306 380 207         Ligand 0 40 0 0 Ramachandran plot statistics     Residues in most favored regions 297 (89.7%) 294 (88.6%) 297 (89.5%) 300 (90.6%)     Residues in additionally allowed regions 32 (9.7%) 36 (10.8%) 33 (9.9%) 29 (8.8%)     Residues in generously allowed regions 0 0 0 0     Residues in disallowed regions 2 (0.6%) 2 (0.6%) 2 (0.6%) 2 (0.6%) . LipA . LipA-BOG . G231F . N228W . Data collection statistics     Unit cell parameters a = 103.7, b = 54.7, c = 66.3 Å, β = 92.6° a = 98.3, b = 64.9, c = 66.0 Å, β = 93.0° a = 103.6, b = 54.7, c = 66.3 Å, β = 92.7° a = 97.9, b = 64.1, c = 66.4 Å, β = 93.0°     Resolution (Å) 25–1.86 (1.93–1.86)a 25–2.1 (2.18–2.1) 25–2.1 (2.18–2.1) 25–2.1 (2.18–2.1)     Unique reflections 30,946 (2,969) 23,231 (1,599) 21,555 (1,898) 23,336 (2,053)     Completeness (%) 99.0 (95.8) 99.5 (95.8) 98.5 (88.2) 97.1 (86.2)     Rsym 3.8 (11.6) 6.2 (18.7) 10.9 (23.4) 5.4 (31.5)     Redundancy 3.5 (3.2) 3.1 (2.1) 3.0 (2.7) 2.7 (2.0)     I/(σ)I 30.5 (8.2) 11 (2.4) 10.2 (3.6) 14.5 (1.7) Refinement statistics     R (%) 16.1 18.8 17.9 20.1     Rfree(%)b 18.7 23.2 22.5 24.5     Mean B factor (Å2)         Protein 9.4 23.3 12.7 34.4         Water 24.9 34.6 24.5 40.8         Ligand – 36.6 – –     Root mean square deviation in         Bond distances (Å) 0.005 0.006 0.005 0.006         Bond angles (°) 1.251 1.253 1.250 1.257     No. of residues 387 390 387 387     No. of ligands 0 2 0 0     No. of atoms 3,470 3,292 3,328 3,154         Protein 2,941 2,946 2,948 2,947         Water 529 306 380 207         Ligand 0 40 0 0 Ramachandran plot statistics     Residues in most favored regions 297 (89.7%) 294 (88.6%) 297 (89.5%) 300 (90.6%)     Residues in additionally allowed regions 32 (9.7%) 36 (10.8%) 33 (9.9%) 29 (8.8%)     Residues in generously allowed regions 0 0 0 0     Residues in disallowed regions 2 (0.6%) 2 (0.6%) 2 (0.6%) 2 (0.6%) a Values in parentheses are for the highest-resolution shell. b Throughout the refinement, 5% of the total reflections were kept aside for Rfree. Open in new tab . The LipA wild-type crystals were soaked with several detergents, and hydrophobic ligands from Hampton additive screen and BOG-bound crystals were obtained using a concentration of 17.5 mM ligand. Molecular replacement using MOLREP (Computational Collaborative Project, Number 4, 1994) on the data set of the BOG-bound crystal resulted in two molecules of BOG binding in the LipA structure. This structure was refined to R/Rfree of 18.8/23.2%, respectively, at 2.1 Å resolution. A similar concentration of BOG and 3 mg mL−1 concentrations of purified proteins of mutants G231F and N228W were used to obtain the mutant cocrystals. Both structures were also solved using molecular replacement as described above. PROCHECK was used to assess the stereochemistry of the structures. Structure and Sequence Analysis Modeling of LipA homologs was done using Modeller9v4 software (Marti-Renom et al., 2000). The DALI server was used for structural homology searches (Sali and Blundell, 1993). The NCBI BLAST server was used for identifying sequence homologs, ClustalW (EBI server) for multiple sequence alignments, and MEGA v.4 was used for preparing phylogenetic trees and bootstrap analysis (Altschul et al., 1990; Thompson et al., 1994; Tamura et al., 2007). MOLE was used as a PYMOL plugin to visualize the tunnel (Petrek et al., 2007). ITC The in vitro binding of BOG to LipA and the LipA point mutants was measured by ITC using a VP-ITC calorimeter (MicroCal). ITC Buffer (50 mM Tris, pH 7.5, and 150 mM NaCl) was used for diluting the protein and dissolving BOG. All samples were degassed prior to titration. The 40 μM protein sample (1.8 mL) was titrated against 150 μL of 2 mM BOG over 50 injections at 303K. The change in heat of the proteins upon titrating with BOG was measured, integrated, and fitted into a one-site binding model for calculation of K a, ΔHapp, and TΔS using the Origin 7.0 software (MicroCal) for curve fitting and data analysis. The parameters K d and ΔG were calculated using formulae K d = K a −1 and ΔG = ΔH – TΔS, respectively. The heat of dilution of BOG was measured by a blank titration of ligand into the buffer, and this was subtracted from the binding isotherms of the wild-type and mutant proteins. The β-octyl galactoside titration was also performed similarly, and a sequential two-site model was used to fit the raw data in this case. Overall K d for β-octyl galactoside was calculated using the equation 1/√K1K2 where K1 was 8.8×103 and K2 was 230. Site-Directed Mutagenesis and Purification of Mutant Proteins The lipA gene was cloned into the pBSKS plasmid (Stratagene) and mutated using the QuickChange site-directed mutagenesis kit (Stratagene; primers listed in Supplemental Table 4 online). The mutant lipA genes were excised as KpnI-HindIII fragments, cloned into the multicloning site of broad-host range vector pHM1, and transformed into Xoo strain BXO2001 that has an insertion mutation in the lipA gene (Innes et al., 1988; Rajeshwari et al., 2005). Presence of the lipA point mutations was confirmed by sequencing of the LipA gene from each strain. Expression of mutant LipA proteins was confirmed using rabbit polyclonal anti-LipA antibodies generated in this study at 1:5000 dilution and 1:10,000 dilution of anti-rabbit alkaline phosphatase–conjugated goat IgG from Sigma-Aldrich. As described previously, Xoo strain BXO2008 containing the lipA gene on the broad-host range vector pHM1 was used as a source of wild-type LipA (Aparna et al., 2007). The LipA mutant proteins were purified to homogeneity using similar protocols. Substrate Clearance Assay and Enzyme Kinetics Tributyrin (C4), tricaproin (C6), tricaprylin (C8), tricaprin (C10), trilaurin (C12), and tripalmitin (C16) (Sigma-Aldrich) were used as substrates in plate assays for LipA activity (Smeltzer et al., 1992). Briefly, 0.5% suspensions of the triglyceride substrates were prepared in a buffer containing 100 mM Tris-Cl, pH 8.0, 25 mM CaCl2, sonicated at 30 W for 3 min to emulsify the substrates, mixed with an equal volume of 2% agarose solution and solidified in Petri plates. Fifty microliters of 0.5 mg mL−1 of purified wild-type and mutant LipA proteins were added to the wells cut into each substrate plate and assayed for a zone of clearance. Triolein (C18:1) activity was assayed using the olive oil-rhodamine B plate assay (Kouker and Jaeger, 1987). In this assay, 2.5% ultrapure olive oil and 0.001% fluorescent Rhodamine B dye were mixed with 1% agarose solution, emulsified using sonication at 30 W for 3 min, and solidified in a Petri plate. Subsequently, 50 μl of 0.5 mg mL−1 of purified wild-type LipA was added to the wells cut into the plate and assayed for appearance of a zone of clearance. All assays were performed at room temperature. LipA substrate specificity on pNP esters was determined using a spectrophotometric assay with pNP acetate (-C2), pNP butyrate (pNP-C4), pNP caprylate (-C8), pNP caprate (-C10), pNP laurate (-C12), pNP myristate (-C14), pNP palmitate (-C16), and pNP stearate (-C18) as substrates. The spectroscopic assays were performed at OD405 after a 10-min incubation of the substrates with LipA in 50 mM Tris-Cl, pH 7.5. Further assessment of the optimal LipA activity was done using pNP butyrate (pNP-C4) as the substrate over 50 mM Tris buffer, pH 5.5 to 9.0, and at various temperatures between 10 and 70°C. LipA enzyme kinetics (see Supplemental Figure 7 online) was also performed using pNP butyrate (pNP-C4) in 50 mM Tris-Cl, pH 7.5, at room temperature. Eight different substrate concentrations (between 10 μM and 1 mM) were used to determine the initial rate of reaction of the enzyme. Virulence Assay The following Xoo strains were used for virulence analysis: BXO43 (wild-type strain; rifampicin resistant), BXO2001 (lipA1∷bla; Ampr derivative of BXO43; Lip− insertion mutant), BXO2008 (lipA1∷bla; pHM1-lipA+; Specr Ampr derivative of BXO2001), and lipA point mutations expressed using pHM1 in the BXO2001 background (Rajeshwari et al., 2005). These strains were grown to saturation in peptone-sucrose medium supplemented with appropriate antibiotics. The cultures were pelleted down by centrifugation and resuspended in sterile water (3 mL) at a concentration of ∼109 cells/mL. Surgical scissors dipped in these bacterial suspensions were used to clip the leaf tips of greenhouse-grown, 40-d-old plants of the Taichung Native-1 (TN-1) rice variety, which is susceptible to Xoo (Kauffman et al., 1973). Lesion lengths were measured 7 d after inoculation. Callose Deposition Assay Purified wild-type and mutant LipA proteins (0.1 mg mL−1) were infiltrated into leaves of 10-d-old TN-1 rice seedlings using the blunt end of a 1-mL syringe as described (Jha et al., 2007). Fourteen hours later, the infiltrated zone (∼1 × 1 cm) was cut from the leaf, destained with 70% ethanol at 65°C, stained with 0.5% aniline blue for 4 h, and observed under an Axioplan2 epifluorescence microscope using a blue filter (excitation wavelength of 365 nm) and ×10 objective (Hauck et al., 2003). PCD Assay Seeds of TN-1 rice cultivar were germinated on 0.5% sterile agar in Petri dishes. After 2 to 3 d, root tips (∼0.5 cm) were excised from the seedlings and treated with 0.5 mg mL−1 of either the wild-type and mutant LipA proteins or buffer (10 mM phosphate buffer, pH 6.0). After incubation for 16 h, roots were washed and stained with 1 mg mL−1 PI for 20 min and mounted in 50% glycerol on glass slides. The samples were observed under an LSM-510 Meta Confocal microscope (Carl Zeiss) using ×63 oil immersion objectives and He-Ne laser at 514-nm excitation as described (Jha et al., 2007). Anti-LipA Antibodies The full-length wild-type LipA protein was used to generate anti-LipA polyclonal antibodies in rabbit. Preimmune serum was collected from the uninjected animal to serve as a negative control for the antibodies following subdermal injection of a 1:1 mixture of 0.5 mg mL−1 of the protein and Freund's incomplete adjuvant (Sigma-Aldrich) and a booster dose of the same composition 10 d later. The serum containing the polyclonal antibodies was collected 15 d after the booster dose and centrifuged at 15,000 rpm for 30 min to remove blood cells. Expression of mutant LipA proteins was confirmed using anti-LipA serum at 1:5000 dilution and 1:10,000 dilution of anti-rabbit alkaline phosphatase–conjugated goat IgG from Sigma-Aldrich. Accession Numbers Coordinates and structure factors of the two wild-type LipA, BOG-bound LipA, G231F, and N228W mutant structures have been deposited at the PDB with codes 3H2G, 3H2J, 3H2K, 3H2H, and 3H2I, respectively. Sequence data used in this article can be found in the GenBank/EMBL database under the following accession numbers: Xanthomonas oryzicola (Xoryp_20705), X. campestris pv vesicatoria (XCV0536), X. axonopodis pv citrii (XAC0501), X. campestris pv campestris strain ATCC 33913 (XCC2957 and XCC2374), Xylella fastidiosa 9a5c (XF0357, XF0358, and XF2151); Burkholderia phytofirmans (Bphyt_4125), B. xenovorans (Bxe_B0552), Ralstonia metallidurans (Rmet_5769), Polaromonas napthalenivorans (Pnap_1828), Streptomyces avermitilis (SAV_5844), and Ideonella sp (BAF64544). Candida antarticus lipase CalA structure used for structural comparison was obtained from the PDB using the accession code 2VEO. Supplemental Data The following materials are available in the online version of this article. Supplemental Figure 1. Surface View of LipA. Supplemental Figure 2. An Interesting Region Missing in the Electron Density of LipA Structure. Supplemental Figure 3. Phylogenetic Analysis of LipA Sequence Homologs. Supplemental Figure 4. Homology Model of the Ideonella LipA-Like Protein. Supplemental Figure 5. LipA Substrate Specificity and Activity Curve. Supplemental Figure 6. Protein Gel Blot Analysis of Wild-type and Mutant LipA Proteins. Supplemental Figure 7. LipA Enzyme Kinetics. Supplemental Table 1. Structural Homology Searches of LipA Using the Dali Server. Supplemental Table 2. Interaction of Various Sugars Found Commonly in Rice Cell Walls with LipA Residues. Supplemental Table 3. Crystallographic Statistics for the Heavy Atom Derivatives Used in the Structure Solution of LipA. Supplemental Table 4. List of Primers Used in This Study. Supplemental Data Set 1. Text File of the Alignment Used for the Phylogenetic Analysis, Corresponding to the Whole LipA Sequence, as Illustrated in Supplemental Figure 3. Acknowledgments G.A. is supported by a senior research fellowship from the Council of Scientific and Industrial Research, India. H. Patel and M.G. Anil are acknowledged for help with microscopy experiments. This study was supported in part by an International Senior Research Fellowship of the Wellcome Trust, UK, Swarnajayanti fellowship, Department of Science and Technology, Government of India to R.S. and by a grant from the Department of Biotechnology, Government of India to R.V.S. References 1. Albersheim, P., Jones, T.M., and English, P.D. ( 1969 ). Biochemistry of the cell wall in relation to the infective processes. Annu. Rev. Phytopathol. 7 : 171 – 194 . Crossref Search ADS PubMed 2. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J. ( 1990 ). Basic local alignment search tool. J. Mol. Biol. 215 : 403 – 410 . Crossref Search ADS PubMed 3. Aparna, G., Chatterjee, A., Jha, G., Sonti, R.V., and Sankaranarayanan, R. ( 2007 ). Crystallization and preliminary crystallographic studies of LipA, a secretory lipase/esterase from Xanthomonas oryzae pv. oryzae. Acta Crystallogr. F63 : 708 – 710 . 4. Arthan, D., Kittakoop, P., Esen, A., and Svasti, J. ( 2006 ). Furostanol glycoside 26-O-β-glucosidase from the leaves of Solanum torvum. Phytochemistry 67 : 27 – 33 . Crossref Search ADS PubMed 5. Bartlam, M., Wang, G., Yang, H., Gao, R., Zhao, X., Xie, G., Cao, S., Feng, Y., and Rao, Z. ( 2004 ). Crystal structure of an acylpeptide hydrolase/esterase from Aeropyrum pernix K1. Structure 12 : 1481 – 1488 . Crossref Search ADS PubMed 6. Bauer, S., Vasu, P., Persson, S., Mort, A.J., and Somerville, C.R. ( 2006 ). Development and application of a suite of polysaccharide-degrading enzymes for analyzing plant cell walls. Proc. Natl. Acad. Sci. USA 103 : 11417 – 11422 . Crossref Search ADS 7. Bestwick, C.S., Bennett, M.H., and Mansfield, J.W. ( 1995 ). Hrp mutant of Pseudomonas syringae pv. phaseolicola induces cell wall alteration but not membrane damage leading to the hypersensitive reaction in lettuce. Plant Physiol. 108 : 503 – 516 . Crossref Search ADS PubMed 8. Boraston, A.B., Bolam, D.N., Gilbert, H.J., and Davies, G.J. ( 2004 ). Carbohydrate-binding modules: Fine-tuning polysaccharide recognition. Biochem. J. 382 : 769 – 781 . Crossref Search ADS PubMed 9. Buranov, A.U., and Mazza, G. ( 2008 ). Lignin in straw of herbaceous crops. Ind. Crops Prod. 28 : 237 – 259 . Crossref Search ADS 10. Cantarel, B.L. ( 2008 ). The Carbohydrate-Active EnZymes database (CAZy): An expert resource for glycogenomics. Nucleic Acids Res. 37 : D233 –D238. PubMed 11. Carpita, N.C. ( 1996 ). Structure and biogenesis of the cell walls of grasses. Annu. Rev. Plant Physiol. Plant Mol. Biol. 47 : 445 – 476 . Crossref Search ADS PubMed 12. Chen, J.C.H., Miercke, L.J.W., Krucinski, J., Starr, J.R., Saenz, G., Wang, X., Spilburg, C.A., Lange, L.G., Ellsworth, J.L., and Stroud, R.M. ( 1998 ). Structure of bovine pancreatic cholesterol esterase at 1.6 Å: Novel structural features involved in lipase activation. Biochemistry 37 : 5107 – 5117 . Crossref Search ADS PubMed 13. Computational Collaborative Project, Number 4 ( 1994 ). The CCP4 Suite: Programs for protein crystallography. Acta Crystallogr. D Biol. Crystallogr. 50 : 760 – 763 . Crossref Search ADS PubMed 14. Cowtan, K. ( 1994 ). DM: An automated procedure for phase improvement by density modification. Joint CCP4 and ESF-EACBM Newsletter on Protein Crystallography 31: 34–38. 15. Darvill, A.G., and Albersheim, P. ( 1984 ). Phytoalexins and their elicitors: A defense against microbial infection in plants. Annu. Rev. Plant Physiol. 35 : 243 – 275 . Crossref Search ADS 16. Ericsson, D.J., Kasrayan, A., Johansson, P., Bergfors, T., Sandstrom, A.G., Backvall, J.E., and Mowbray, S.L. ( 2008 ). X-ray structure of Candida antarctica lipase A shows a novel lid structure and a likely mode of interfacial activation. J. Mol. Biol. 376 : 109 – 119 . Crossref Search ADS PubMed 17. Esquerre-Tugaye, M.T., Boudart, G., and Dumas, B. ( 2000 ). Cell wall degrading enzymes, inhibitory proteins, and oligosaccharides participate in the molecular dialogue between plants pathogens. Plant Physiol. Biochem. 38 : 157 – 163 . Crossref Search ADS 18. Ghosh, D., Erman, M., Sawicki, M., Lala, P., Weeks, D.R., Li, N.Y., Pangborn, W., Thiel, D.J., Jornvall, H., Gutierrez, R., and Eyzaguirre, J. ( 1999 ). Determination of a protein structure by iodination: the structure of iodinated acetylxylan esterase. Acta Crystallogr. D Biol. Crystallogr. 55 : 779 – 784 . Crossref Search ADS PubMed 19. Gilbert, H.J., Stalbrand, H., and Brumer, H. ( 2008 ). How the walls come crumbling down: recent structural biochemistry of plant polysaccharide degradation. Curr. Opin. Plant Biol. 11 : 338 – 348 . Crossref Search ADS PubMed 20. Goldstein, R.A. ( 2008 ). The structure of protein evolution and the evolution of protein structure. Curr. Opin. Struct. Biol. 18 : 170 – 177 . Crossref Search ADS PubMed 21. Hauck, P., Thilmony, R., and He, S.Y. ( 2003 ). A Pseudomonas syringe type III effector suppresses cell wall-based extracellular defense in susceptible Arabidopsis plants. Proc. Natl. Acad. Sci. USA 100 : 8577 – 8582 . Crossref Search ADS 22. Hegyi, H., and Gerstein, M. ( 1999 ). The relationship between protein structure and function: A comprehensive survey with application to the yeast genome. J. Mol. Biol. 288 : 147 – 164 . Crossref Search ADS PubMed 23. Hermoso, J.A., Sanz-Aparicio, J., Molina, R., Juge, N., Gonzalez, R., and Faulds, C.B. ( 2004 ). The crystal structure of feruloyl esterase A from Aspergillus niger suggests evolutive functional convergence in feruloyl esterase family. J. Mol. Biol. 338 : 495 – 506 . Crossref Search ADS PubMed 24. Holm, L., and Sander, C. ( 1998 ). Dictionary of recurrent domains in protein structures. Proteins 33 : 88 – 96 . Crossref Search ADS PubMed 25. Holmquist, M. ( 2000 ). Alpha/beta-hydrolase fold enzymes: Structures, functions and mechanisms. Curr. Protein Pept. Sci. 1 : 209 – 235 . Crossref Search ADS PubMed 26. Hu, J., Qian, W., and He, C. ( 2007 ). The Xanthomonas oryzae pv. oryzae eglXoB endoglucanase gene is required for virulence to rice. FEMS Microbiol. Lett. 269 : 273 – 279 . Crossref Search ADS PubMed 63. Innes, R.W., Hirose, M.A., and Kuempel, P.L. (1988). Induction of nitrogen-fixing nodules on clover requires only 32 kilobase pairs of DNA from Rhizobium trifolii symbiosis plasmid. J. Bacteriol. 170 : 3793 – 3802 . PubMed 27. Ito, K., Nakajima, Y., Xu, Y., Yamada, N., Onohara, Y., Ito, T., Matsubara, F., Kabashima, T., Nakayama, K., and Yoshimoto, T. ( 2006 ). Crystal structure and mechanism of tripeptidyl activity of prolyl tripeptidyl aminopeptidase from Porphyromonas gingivalis. J. Mol. Biol. 362 : 228 – 240 . Crossref Search ADS PubMed 28. Jaeger, K.-E., Dijkstra, B.W., and Reetz, M.T. ( 1999 ). Bacterial biocatalysts: Molecular biology, structure and biotechnological applications of lipases. Annu. Rev. Microbiol. 53 : 315 – 351 . Crossref Search ADS PubMed 30. Jha, G., Rajeshwari, R., and Sonti, R.V. ( 2005 ). Bacterial type two secretion system secreted proteins: double-edged swords for plant pathogens. Mol. Plant Microbe Interact. 18 : 891 – 898 . Crossref Search ADS PubMed 29. Jha, G., Rajeshwari, R., and Sonti, R.V. ( 2007 ). Functional interplay between two Xanthomonas oryzae pv. oryzae secretion systems in modulating virulence on rice. Mol. Plant Microbe Interact. 20 : 31 – 40 . Crossref Search ADS PubMed 31. Jones, T.A., Zou, J.-Y., Cowan, S.W., and Kjeldgaard, M. ( 1991 ). Improved methods for the building of protein models in electron density maps and the location of errors in these models. Acta Crystallogr. A 47 : 110 – 119 . Crossref Search ADS PubMed 32. Kauffman, H.E., Reddy, A.P.K., Hsieh, S.P.Y., and Merca, S.D. ( 1973 ). An improved technique for evaluation of resistance of rice varieties to Xanthomonas oryzae. Plant Dis. Rep. 57 : 537 – 541 . 33. Knox, J.P. ( 2008 ). Mapping the walls of the kingdom: The view from the horsetails. New Phytol. 179 : 1 – 3 . Crossref Search ADS PubMed 34. Kouker, G., and Jaeger, K. ( 1987 ). Specific and sensitive plate assay for bacterial lipases. Appl. Environ. Microbiol. 53 : 211 – 213 . Crossref Search ADS PubMed 35. Martinelle, M., Holmquist, M., and Hult, K. ( 1995 ). On the interfacial activation of Candida antarctica lipase A and B as compared with Humicola lanuginosa lipase. Biochim. Biophys. Acta 1258 : 272 – 276 . Crossref Search ADS PubMed 36. Martinez, C., de Geus, P., Lauwereys, M., Matthyssens, M., and Cambillau, C. ( 1992 ). Fusarium solani cutinase is a lipolytic enzyme with a catalytic serine accessible to solvent. Nature 356 : 615 – 618 . Crossref Search ADS PubMed 37. Marti-Renom, M.A., Stuart, A., Fiser, A., Sanchez, R., Melo, F., and Sali, A. ( 2000 ). Comparative protein structure modeling of genes and genomes. Annu. Rev. Biophys. Biomol. Struct. 29 : 291 – 325 . Crossref Search ADS PubMed 38. Nardini, M., and Dijkstra, B.W. ( 1999 ). α/β Hydrolase fold enzymes: The family keeps growing. Curr. Opin. Struct. Biol. 9 : 732 – 737 . Crossref Search ADS PubMed 39. Ollis, D.L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F., Franken, S.M., Harel, M., Remington, S.J., Silman, I., and Schrag, J. ( 1992 ). The α/β hydrolase fold. Protein Eng. 5 : 197 – 211 . Crossref Search ADS PubMed 40. Orengo, C.A., Michie, A.D., Jones, D.T., Swindells, M.B., and Thornton, J.M. ( 1997 ). CATH: A hierarchic classification of protein domain structures. Structure 5 : 1093 – 1108 . Crossref Search ADS PubMed 41. Otwinowski, Z., and Minor, W. ( 1997 ). Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276 : 307 – 326 . Crossref Search ADS 42. Pennell, R.I., and Lamb, C. ( 1997 ). Programmed cell death in plants. Plant Cell 9 : 1157 – 1168 . PubMed 43. Petrek, M., Kosinova, P., Koca, J., and Otyepka, M. ( 2007 ). MOLE: A Voronoi diagram-based explorer of molecular channels pores and tunnels. Structure 15 : 1357 – 1363 . Crossref Search ADS PubMed 44. Rajeshwari, R., Jha, G., and Sonti, R.V. ( 2005 ). Role of an in planta expressed xylanase of Xanthomonas oryzae pv. oryzae in promoting virulence on rice. Mol. Plant Microbe Interact. 18 : 830 – 837 . Crossref Search ADS PubMed 45. Ray, S.K., Rajeshwari, R., and Sonti, R.V. ( 2000 ). Mutants of Xanthomonas oryzae pv. oryzae deficient in general secretary pathway are virulence deficient and unable to secrete xylanase. Mol. Plant Microbe Interact. 13 : 394 – 401 . Crossref Search ADS PubMed 46. Ron, M., and Avni, A. ( 2004 ). The receptor for the fungal elicitor ethylene-inducing xylanase is a member of a resistance-like gene family in tomato. Plant Cell 10 : 1604 – 1615 . 47. Ryan, C.A., and Farmer, E.E. ( 1991 ). Oligosaccharide signals in plants: A current assessment. Annu. Rev. Plant Physiol. Mol. Biol. 42 : 651 – 674 . Crossref Search ADS 48. Sali, A., and Blundell, T.L. ( 1993 ). Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 234 : 779 – 815 . Crossref Search ADS PubMed 49. Shinohara, M., Nakajima, N., and Uehara, Y. ( 2007 ). Purification and characterization of a novel esterase (β-hydroxypalmitate methyl ester hydrolase) and prevention of the expression of virulence by Ralstonia solanacearum. J. Appl. Microbiol. 103 : 152 – 162 . Crossref Search ADS PubMed 50. Smeltzer, M.S., Hart, M., and Iandolo, J.J. ( 1992 ). Quantitative spectrophotometric assay for staphylococcal lipase. Appl. Environ. Microbiol. 58 : 2815 – 2819 . Crossref Search ADS PubMed 51. Somerville, C., Bauer, S., Brininstool, G., Facette, M., Hamann, T., Milne, J., Osborne, E., Paredez, A., Persson, S., Raab, T., Vorwerk, S., and Youngs, H. ( 2004 ). Toward a systems approach to understanding plant cell walls. Science 306 : 2206 – 2211 . Crossref Search ADS PubMed 52. Sun, Q.H., Hu, J., Huang, G.X., Ge, C., Fang, R.X., and He, C.Z. ( 2005 ). Type-II secretion pathway structural gene xpsE xylanase and cellulase secretion and virulence in Xanthomonas oryzae pv. oryzae. Plant Pathol. 54 : 15 – 21 . Crossref Search ADS 53. Szeltner, Z., and Polgar, L. ( 2008 ). Structure function and biological relevance of prolyl oligopeptidase. Curr. Protein Pept. Sci. 9 : 96 – 107 . Crossref Search ADS PubMed 54. Tamura, K., Dudley, J., Nei, M., and Kumar, S. ( 2007 ). MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24 : 1596 – 1599 . Crossref Search ADS PubMed 55. Terwilliger, T.C. ( 2003 ). Automated main-chain model building by template matching and iterative fragment extension. Acta Crystallogr. D Biol. Crystallogr. 59 : 38 – 44 . Crossref Search ADS PubMed 56. Terwilliger, T.C., and Berendzen, J. ( 1999 ). Automated MAD and MIR structure solution. Acta Crystallogr. D Biol. Crystallogr. 55 : 849 – 861 . Crossref Search ADS PubMed 57. Thompson, J.D., Higgins, D.G., and Gibson, T.J. ( 1994 ). CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22 : 4673 – 4680 . Crossref Search ADS PubMed 58. Thornton, J.M., Orengo, C.A., Todd, A.E., and Pearl, F.M.G. ( 1999 ). Protein folds functions and evolution. J. Mol. Biol. 293 : 333 – 342 . Crossref Search ADS PubMed 59. van Pouderoyen, G., Eggert, T., Jaeger, K., and Dijkstra, B.W. ( 2001 ). The crystal structure of Bacillus subtilis lipase: A minimal α/β hydrolase fold enzyme. J. Mol. Biol. 309 : 215 – 226 . Crossref Search ADS PubMed 60. Verger, R. ( 1997 ). Interfacial activation of lipases: Facts and artifacts. Trends Biotechnol. 15 : 32 – 38 . Crossref Search ADS 61. Vorwerk, S., Somerville, S., and Somerville, C. ( 2004 ). The role of plant cell wall polysaccharide composition in disease resistance. Trends Plant Sci. 9 : 203 – 209 . Crossref Search ADS PubMed 62. Wei, Y., Contreras, J.A., Sheffield, P., Osterlund, T., Derewenda, U., Kneusel, R.E., Matern, U., Holm, C., and Derewenda, Z.S. ( 1999 ). Crystal structure of brefeldin A esterase, a bacterial homolog of the mammalian hormone-sensitive lipase. Nat. Struct. Biol. 6 : 340 – 345 . Crossref Search ADS PubMed Author notes 1 Address correspondence to sankar@ccmb.res.in. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantcell.org) is: Rajan Sankaranarayanan (sankar@ccmb.res.in). Online version contains Web-only data. www.plantcell.org/cgi/doi/10.1105/tpc.109.066886 © 2009 American Society of Plant Biologists This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - A Cell Wall–Degrading Esterase of Xanthomonas oryzae Requires a Unique Substrate Recognition Module for Pathogenesis on Rice
JF - The Plant Cell
DO - 10.1105/tpc.109.066886
DA - 2009-08-05
UR - https://www.deepdyve.com/lp/oxford-university-press/a-cell-wall-degrading-esterase-of-xanthomonas-oryzae-requires-a-unique-NrpbFqKAsu
SP - 1860
EP - 1873
VL - 21
IS - 6
DP - DeepDyve
ER -