Skip to main content
NCBI home page
Protein Science : A Publication of the Protein Society logoLink to Protein Science : A Publication of the Protein Society
. 2024 Jul 19;33(8):e5122. doi: 10.1002/pro.5122

Evidence of isochorismate channeling between the Escherichia coli enterobactin biosynthetic enzymes EntC and EntB

Xue Bin 1, Peter D Pawelek 1,✉
PMCID: PMC11258883  PMID: 39031458

Abstract

Enterobactin is a high‐affinity iron chelator produced and secreted by Escherichia coli and Salmonella typhimurium to scavenge scarce extracellular Fe3+ as a micronutrient. EntC and EntB are the first two enzymes in the enterobactin biosynthetic pathway. Isochorismate, produced by EntC, is a substrate for EntB isochorismatase. By using a competing isochorismate‐consuming enzyme (the E. coli SEPHCHC synthase MenD), we found in a coupled assay that residual EntB isochorismatase activity decreased as a function of increasing MenD concentration. In the presence of excess MenD, EntB isochorismatase activity was observed to decrease by 84%, indicative of partial EntC‐EntB channeling (16%) of isochorismate. Furthermore, addition of glycerol to the assay resulted in an increase of residual EntB isochorismatase activity to approximately 25% while in the presence of excess MenD. These experimental outcomes supported the existence of a substrate channeling surface identified in a previously reported protein‐docking model of the EntC‐EntB complex. Two positively charged EntB residues (K21 and R196) that were predicted to electrostatically guide negatively charged isochorismate between the EntC and EntB active sites were mutagenized to determine their effects on substrate channeling. The EntB variants K21D and R196D exhibited a near complete loss of isochorismatase activity, likely due to electrostatic repulsion of the negatively charged isochorismate substrate. Variants K21A, R196A, and K21A/R196A retained partial EntB isochorismatase activity in the absence of EntC; in the presence of EntC, isochorismatase activity in all variants increased to near wild‐type levels. The MenD competition assay of the variants revealed that while K21A channeled isochorismate as efficiently as wild‐type EntB (~ 15%), the variants K21A/R196A and R196A exhibited an approximately 5‐fold loss in observed channeling efficiency (~3%). Taken together, these results demonstrate that partial substrate channeling occurs between EntC and EntB via a leaky electrostatic tunnel formed upon dynamic EntC‐EntB complex formation and that EntB R196 plays an essential role in isochorismate channeling.

Keywords: automated docking, enterobactin, protein–protein interaction, siderophore, site‐directed mutagenesis, substrate channeling

1. INTRODUCTION

Iron is essential for many metabolic processes and cell growth in living organisms, including most bacteria. In an iron‐depleted environment, Gram‐negative bacteria synthesize and secrete siderophores to scavenge scarce extracellular Fe3+. Enterobactin, a siderophore produced by Escherichia coli and Salmonella typhimurium, has extraordinarily high affinity for ferric iron (KD ~ 10−35 M at physiological pH). (Carrano & Raymond, 1979) Upon acquisition of extracellular iron, the ferric enterobactin complex is then imported via TonB‐dependent uptake (Raymond et al., 2003).

Biosynthesis of enterobactin in E. coli involves the action of six cytoplasmic enzyme activities: EntC, EntB (IC domain), EntA, EntE, EntB (ArCP domain), and EntF. These activities catalyze reactions in two functional arms of the enterobactin biosynthetic pathway: (1) the 2,3‐dihydroxybenzoate (DHB) biosynthetic arm and (2) the non‐ribosomal peptide synthetase (NRPS) arm. The initial substrate, chorismate, is first converted to isochorismate by EntC (isochorismate synthase; EC 5.4.4.2) (E1, Figure 1) and then cleaved into 2,3‐dihydro‐2,3‐dihydroxybenzoate (2,3‐dihydro‐DHB) and pyruvate by the N‐terminal isochorismatase (IC) domain of EntB (EC 3.3.2.1) (E2, Figure 1). The oxidoreductase EntA catalyzes the oxidation of 2,3‐dihydro‐DHB to 2,3‐DHB. The NRPS enzyme EntE (2,3‐dihydroxybenzoate‐AMP ligase) then activates 2,3‐DHB for the NRPS arm of the pathway via adenylation, for subsequent transfer to the C‐terminal aryl carrier protein domain of EntB. In the NRPS arm of the pathway, three molecules of activated 2,3‐DHB are condensed with L‐serine as catalyzed by the large (~ 145 kDa) multifunctional enzyme EntF, followed by EntF‐catalyzed cyclization and release (Lai et al., 2006).

FIGURE 1.

FIGURE 1

Open in a new tab

Scheme of reactions catalyzed in the reported coupled assay system. The coupled assay utilized EntC, EntB (IC domain), and LDH (E1, E2, E3, respectively). Substrate channeling assays utilized the above as well as E. coli MenD (E4). 2,3‐diDHB: 2,3‐dihydro‐dihydroxybenzoic acid; LDH: Lactate dehydrogenase; SEPHCHC: 2‐succinyl‐5‐enolpyruvyl‐6‐hydroxy‐3‐cyclohexene‐1‐carboxylate.

Regulation of bacterial iron uptake processes is essential for maintaining intracellular iron homeostasis. Under iron‐replete conditions, the expression of genes encoding the enterobactin biosynthetic enzymes is down‐regulated by the Fur (Ferric Uptake Regulator) repressor protein. Upon iron starvation, Fe2+ dissociates from Fur, which in turn dissociates from the consensus Fur box sequence upstream of the ent gene cluster to promote gene expression. Once sufficient intracellular iron has been acquired, holo‐Fur once again binds to the Fur box and represses ent gene expression. Beyond this, very little remains known about how enterobactin biosynthesis is regulated (Andrews et al., 2003; Brickman et al., 1990).

At the protein level, we know that the Ent proteins extensively engage in pairwise protein interactions (PPIs). Such interactions have now been reported among all enzymes involved in enterobactin biosynthesis, such as EntB‐EntA, EntA‐EntE, EntE‐EntB(ArCP), and EntF‐EntB(ArCP) (Khalil & Pawelek, 2009; Khalil & Pawelek, 2011; Pakarian & Pawelek, 2016a; Pakarian & Pawelek, 2016b). Recently, our lab also reported an in vivo PPI between EntC and EntB via a bacterial two‐hybrid assay, demonstrating that all of the enzymes responsible for DHB biosynthesis and activation (EntCBAE) participate in extensive PPIs with each other (Ouellette et al., 2022). While the necessity for PPIs within the downstream NRPS portion of the pathway is well understood, the function of the PPIs within the upstream DHB biosynthetic portion of the pathway and its connection to the NRPS apparatus (EntC‐EntB, EntB‐EntA, EntA‐EntE) are not known. Enzymes often associate with static or dynamic complexes to enhance the efficiency of metabolic processes in which they are involved. Substrate channeling is a process through which the product of a reaction is transferred from an active site to a second active site that catalyzes a sequential reaction without diffusing into the bulk solvent (Arentson et al., 2012). Channeling may occur between active sites on a single multifunctional enzyme, or between enzymes that dynamically associate via PPIs. Evidence of substrate channeling has now been reported for the enzymes involved in many metabolic processes including the TCA cycle, proline catabolism, glycolysis, and gluconeogenesis (Bulutoglu et al., 2016; Sanyal et al., 2015; Svedružić et al., 2020). Biological advantages of channeling include: (i) increasing catalytic efficiency by limiting the diffusion of intermediates into the bulk medium; (ii) chemoprotection of labile intermediates; (iii) controlling the routing of intermediates into competing metabolic pathways (Anderson, 1999; Miles et al., 1999; Spivey & Ovádi, 1999). While intermediates within the NRPS arm have been shown to be directly transferred between Ent protein domains, no substrate channeling within the upstream DHB arm of the pathway has been reported to date. Given the relative instabilities of the DHB biosynthetic intermediates isochorismate and 2,3‐dihydro‐DHB, substrate channeling within this earlier part of the pathway would be advantageous both in terms of enhancing catalysis as well as chemoprotection of labile intermediates.

Here we investigate the hypothesis that the EntC‐EntB interaction enhances catalytic efficiency in the production of 2,3‐dihydro‐DHB (an EntB isochorismatase product) from chorismate (an EntC substrate) via direct channeling of isochorismate. We employed a commonly used approach to investigate substrate channeling: the introduction of a competing enzymatic activity. We used E. coli MenD (2‐succinyl‐5‐enolpyruvyl‐6‐hydroxy‐3‐cyclohexene‐1‐carboxylate synthase; EC: 2.2.1.9) (E4, Figure 1a) as a competitor of EntB for isochorismate. We found that in the presence of excess MenD, a residual amount of EntC‐produced isochorismate was still available to EntB in vitro, demonstrating leaky EntCB channeling. Finally, we investigated a proposed electrostatic channeling surface predicted using automatic docking to be at the interaction interface of the EntC‐EntB complex. Site‐directed mutagenesis of EntB interface residues K21 and R196 to alanine or aspartate resulted in a decrease or complete abrogation of intrinsic EntB isochorismatase activity, respectively. Furthermore, the variants R196A and K21A/R196A exhibited a 5‐fold loss in channeling efficiency compared to wild‐type EntB, underlining the key role of EntB R196 in isochorismate channeling within the EntC‐EntB complex.

2. MATERIALS AND METHODS

2.1. Reagents

Glycerol, 2,3‐DHB, porcine lactate dehydrogenase (LDH), and chorismic acid were purchased from Sigma‐Aldrich (St. Louis, MO). All other reagents were purchased from Bioshop Canada, Inc. (Burlington, ON).

2.2. Protein expression and purification

The genes encoding EntB and MenD with in‐frame N‐terminal hexahistidine tags were expressed in E. coli AG‐1 cells harboring pCA24N plasmids obtained from the ASKA repository (Kitagawa et al., 2005). The gene encoding EntC was PCR‐amplified from the pCA24N‐entC plasmid from the same collection. The PCR product was subcloned into pET24b vector with an in‐frame C‐terminal hexahistidine tag and then transformed into E. coli BL21(DE3) cells. Cells were grown in LB (or 2xYT for EntC) broth containing 30 μg/mL chloramphenicol (or kanamycin for EntC) at 37 °C to a final OD600 between 0.6 and 0.8. Overexpression of recombinant proteins was induced by 0.5 mM isopropyl β‐D‐1‐thiogalactopyranoside (IPTG) overnight at 16–22°C.

Cells were lysed by ultrasonic homogenizer (Biologics) in a buffer containing 50 mM HEPES (pH 8.0), 500 mM NaCl, 5% glycerol, 10 mM imidazole, 2 mM β‐mercaptoethanol, and protease inhibitor cocktail. Cell lysate was centrifuged for 1.5 h at 47,850g and the supernatant was collected. Clarified lysate was applied to a metal ion affinity column (Profinity IMAC resin, Bio‐Rad Laboratories) connected to a BioLogic DuoFlow FPLC system (Bio‐Rad Laboratories). Target proteins were eluted by linear gradient imidazole (10–400 mM imidazole) in a buffer containing 50 mM HEPES (pH 8.0), 500 mM NaCl, 5% glycerol, and 2 mM β‐mercaptoethanol. Pooled fractions containing purified proteins were dialyzed against 50 mM HEPES (pH 8.0), 150 mM NaCl, 1.0 mM tris(2‐carboxyethyl)phosphine (TCEP) and 15% (v/v) glycerol. Dialyzed samples were stored at −20°C. Protein concentrations were determined by measuring absorbance at 280 nm and using molar extinction coefficients predicted from primary amino acid sequences (ε = 51,910, 31,970, and 107,285 M−1 cm−1 for EntB, EntC, and MenD, respectively) (Gasteiger et al., 2005).

2.3. EntC‐EntB‐LDH coupled assay

Enzymological assays were carried out at 21°C using a Genesys 10 Spectrophotometer. Reaction mixtures containing 0.05–10 μM EntC, 0.05–10 μM EntB (WT or variants), 1.50 unit/mL LDH, 5 mM MgCl2 and 0.35 mM NADH in Buffer A (50 mM HEPES (pH 8.0), 150 mM NaCl and 1 mM TCEP) were preincubated at room temperature for 10 min. Chorismate was added to the mixtures to initiate reactions. The consumption of NADH by LDH was continuously monitored at 340 nm for 10 min. For competition assays, MenD was added to the reaction mixtures (final concentration range: 0.00 to 6.25 μM) in Buffer A containing 0.05 μM EntC, 1.25 μM EntB, 50 μM thiamine pyrophosphate and 660 μM 2‐ketoglutarate (the latter two reagents being MenD co‐substrates). Given that apparent turnover numbers for EntC and EntB under our assay conditions were roughly similar (67 min−1 and 49 min−1, respectively (data not shown)), a 20:1 molar ratio of EntB to EntC was used in the competition assay in order to compensate for the possibility that EntB activity might be limiting. For some channeling assays, the viscogen glycerol was added to a final concentration of 30% (v/v). In all cases, initial velocities were calculated by linear regression of OD340 decrease using the VISIONlite Rate software (Thermo Scientific). Kinetics data were fit to steady‐state enzyme kinetic models using GraphPad Prism.

2.4. EntB variants

Site‐directed mutagenesis was performed by a commercial service (Top Gene, St‐Laurent, Quebec) using purified pCA24N‐entB as a template. All EntB mutations (K21A, K21D, R196A, R196D, and K21A/R196A) were verified by DNA sequencing. Mutagenic constructs encoding EntB variants were transformed into AG‐1 competent cells following established protocols (Sambrook & Russell, 2006).

Kinetic assays of EntB‐LDH coupled reactions were carried out using conditions similar to those of the EntC‐EntB‐LDH coupled assay, but in the absence of EntC. Due to the lack of commercial availability of isochorismate, we prepared an equilibrium mixture of chorismate:isochorismate by incubating 256.5 μM chorismate with 0.50 μM EntC in Buffer A containing 5 mM MgCl2 in an ice bath for 10 min, which was expected to yield a chorismate:isochorismate at ratio of approximately 3:2 (Hubrich et al., 2014). Once the reaction reached equilibrium, the chorismate:isochorismate mixture was purified by centrifugation at 11,000g for 8 min in a 0.5 mL centrifugal filter with 3 K MWCO (Pall Corporation). The chorismate:isochorismate mixture was used as freshly prepared. Reaction mixtures containing 0.10 μM EntB WT or variants, 1.50 unit/mL LDH and 0.35 mM NADH in Buffer A were preincubated at room temperature for 10 min. The reactions were initiated by the addition of chorismate:isochorismate equilibrium mixture to a final concentration of 25.7 μM (or 51.3 μM for EntB‐EntA coupled assay), and initial velocities were measured as decrease in OD340 upon LDH consumption of NADH.

2.5. MenD activity assay

MenD activity assay was similar to that employed by Jiang et al. (2007). The reaction mixtures contained 66 nM MenD (or 130 nM for assays with 30% glycerol), 50 μM thiamine pyrophosphate and 660 μM 2‐ketoglutarate, 5 mM MgSO4 in Buffer A were preincubated at room temperature for 10 min. The reactions were initiated by the addition of chorismate:isochorismate equilibrium mixture to a final concentration of 3.08, 6.16, and 10.26 μM, and the initial velocities were measured as decrease in OD278 upon MenD consumption of isochorismate.

3. RESULTS AND DISCUSSION

3.1. EntC‐EntB‐LDH coupled assay

Protein–protein interactions between Ent enzymes have long been known to occur in the NRPS arm of the pathway (Koglin & Walsh, 2009; Lai et al., 2006; Sundlov et al., 2012). Such protein interactions (EntB‐EntE, EntB‐EntF) are considered obligate since they are required for the NRPS mechanism. Protein interactions in the upstream DHB arm of the pathway, as well as at the junction between the DHB and NRPS arms (i.e., the EntA‐EntE interaction (Pakarian & Pawelek, 2016a)), have also been reported by our group (Khalil & Pawelek, 2009; Ouellette et al., 2022; Pakarian & Pawelek, 2016b). We have long hypothesized that such interactions may facilitate substrate channeling (Khalil & Pawelek, 2009; Khalil & Pawelek, 2011). In order to investigate a possible role for the EntC‐EntB interaction in promoting the channeling of the EntC product isochorismate to the EntB isochorismatase active site, here we devised a coupled assay in which LDH converts pyruvate produced by EntB into L‐lactate with the concomitant oxidation of NADH to NAD+ (Figure 1, upper reaction); the decrease in NADH absorbance at 340 nm in this assay reports on the combined activities of EntC and EntB.

The EntC‐EntB‐LDH coupled reaction was performed at 21°C in the presence of limiting concentrations of EntC, excess LDH, and varying ratios of EntB to EntC. For the EntC substrate chorismate, we obtained an apparent Michaelis–Menten constant (K m) value of 53 μM when assaying EntC and EntB at a 1:1 ratio (Figure 2). This value is on the order of chorismate K m values previously reported for EntC (14 μM; Liu et al., 1990, 42 μM; Jiang & Guo, 2007), demonstrating that the coupled assay produces outcomes consistent with previous kinetic studies. A previously reported EntB K m value for isochorismate (14.7 μM; Rusnak et al., 1990) also lies within this substrate range. A discontinuous quenched assay (Zhu et al., 2010) that measured the increase of the EntB product pyruvate was also consistent with the coupled assay (data not shown).

FIGURE 2.

FIGURE 2

Open in a new tab

Michaelis–Menten plot of EntC‐EntB‐LDH coupled activity assay. The reaction mixture contained 0.1 μM EntC, 0.1 μM EntB, 1.5 unit/mL LDH, 5 mM MgCl2, 0.35 mM NADH and various amounts of chorismate in Buffer A (see Methods). Initial velocities were measured in triplicate. Error bars indicated the standard deviation from the mean values.

3.2. MenD competition assay

MenD is an enzyme in the menaquinone biosynthesis pathway in E. coli, catalyzing the 1,4‐addition of α‐ketoglutarate to isochorismate to form 2‐succinyl‐5‐enolpyruvyl‐6‐hydroxy‐3‐cyclohexene‐1‐carboxylate (SEPHCHC) and carbon dioxide (Jiang et al., 2007; Kurutsch et al., 2009). Unlike the enterobactin biosynthetic pathway, which is up‐regulated by Fur de‐repression upon intracellular iron deprivation under aerobic conditions, the menaquinone pathway is activated under anaerobic conditions (Buss et al., 2001). Although the menaquinone biosynthetic pathway begins with the conversion of chorismate to isochorismate by the upstream isochorismate synthase enzyme MenF, there are no other common metabolites between the two pathways. Moreover, it has been reported that overexpression of either MenF or EntC in a menF − entC − background strain resulted in biosynthesis of both menaquinone and enterobactin (Buss et al., 2001). This demonstrated that the isochorismate formed by the overexpressed heterologous isochorismate synthase could be utilized in vivo by both the menaquinone and the enterobactin pathways. Finally, the reported isochorismate K m for E. coli MenD is 53 nM (Jiang et al., 2007), whereas the reported isochorismate K m for EntB is 14.7 μM (Rusnak et al., 1990). Given the above, we reasoned that MenD could serve as an effective competitor for isochorismate when added to the EntC‐EntB‐LDH activity assay. Continued production of NADH in the presence of excess exogenous MenD would be indicative of substrate channeling between EntC and EntB, since this would represent the portion of residual isochorismate that remained accessible to EntB isochorismatase due to direct transfer of the intermediate between the EntC and EntB active sites (and thus inaccessible to MenD).

As seen in Figure 3, MenD successfully competed with EntB for isochorismate in the in vitro reaction mixture during the coupled assay. The coupled EntC‐EntB‐LDH activity decreased in a nonlinear fashion as a function of increasing MenD concentration and remained at a constant level of 16% in the presence of a saturating concentration of MenD. Taken together, these data demonstrate while most isochorismate produced by EntC is diffused into the bulk medium in vitro, an observable portion is directly channeled to EntB. We therefore concluded that mixture of EntC and EntB in vitro, in the presence of appropriate substrates, resulted in observed partial (i.e., “leaky”) channeling (Ovádi et al., 1989) of isochorismate. To further explore the leaky nature of the observed channeling phenomenon, we added glycerol into the assay solution to emulate the crowded and viscous intracellular environment. With addition of glycerol to final concentration of 30% (v/v), the observed that EntC‐EntB‐LDH residual activity in the presence of excess MenD increased from 16% to 25%, which is consistent with the approximate halving of the assay solution's diffusion coefficient at the elevated glycerol concentration (Rausch et al., 2017). This increase in observed substrate channeling efficiency in vitro is likely due to the suppression of isochorismate diffusion into the more viscous bulk medium, and/or higher interaction affinity between EntC and EntB in the presence of glycerol—both outcomes would result in higher channeling efficiency. As a control, assays of individual enzymatic activities in the presence and absence of 30% glycerol were performed to ensure that observed channeling efficiencies were not due to glycerol‐induced changes in the kinetic parameters of individual enzymes. The addition of glycerol was observed to not significantly affect the chorismate K m value for EntC, nor the isochorismate K m values for EntB, or MenD. In the presence of glycerol, the apparent V max of MenD remained relatively unchanged, while the apparent V max values of EntC and EntB were each observed to decrease approximately 5‐fold; since both enzymes decreased in activity roughly proportionally, glycerol‐induced changes to individual enzyme activities were determined not to impact observed isochorismate channeling efficiencies.

FIGURE 3.

FIGURE 3

Open in a new tab

Effect of addition of excess MenD on EntC‐EntB‐LDH coupled assay. Data shown are the mean of triplicate readings. Error bars represent the standard deviation from the mean. Fit of data to one phase decay curve (continuous line) in order to extract a plateau value. The coupled activity assays were in the absence (dark circles) and presence (dark triangles) of 30% (v/v) glycerol respectively.

We previously reported an automated docking model of the EntC‐EntB complex in which a potential electrostatic channeling surface was predicted to occur between the active sites of the interacting enzymes (Ouellette et al., 2022). In that study, we identified a 42 Å surface in the docked model using Caver (Stourac et al., 2019)   that connected the EntC and EntB active sites. Further inspection of this putative channeling surface revealed that the isochorismatase active site of EntB is partially surface‐exposed in the EntCB docked model, which is consistent with our observation of leaky isochorismate channeling. In addition, the distribution of residues along this predicted channeling surface exhibited expected asymmetry: positively charged residues mainly aligned on one side of the channel towards the protein interaction interface and the EntB active site, whereas the neutral and negatively charged residues were found to cluster near the EntC active site. This arrangement, as revealed in the model, suggested a gradient of increasing electropositivity towards the EntB active site, capable of facilitating direct transfer of negatively charged isochorismate.

Partial channeling within an electrostatic channeling system has been reported previously. For example, the investigation of a fusion protein between malate dehydrogenase and citrate synthase revealed that negatively charged oxaloacetate could channel between the MDH and CS active sites with a transfer efficiency of approximately 45% (Bulutoglu et al., 2016; Elcock & McCammon, 1996). In this case, the distance between the MDH and CS active sites was approximately 60 Å, similar to the distance between EntC and EntB active sites observed in our docked model (Ouellette et al., 2022). Metabolic channeling of isochorismate has also been reported between E. coli MenF and MenD (Buss et al., 2001). The isochorismate synthase MenF shares approximately 25% identity with E. coli EntC. Our observation of isochorismate channeling between EntC and EntB is consistent with these other reports.

3.3. The roles of EntB K21 and R196 in electrostatic guiding of isochorismate

EntB residues K21 and R196 were found near the exit of the tunnel and proximal to the EntB active site. To further investigate these residues and their possible role in isochorismate channeling, we generated five E. coli EntB variants in which charges were either reversed (K21D, R196D), or removed altogether (K21A, R196A, and K21A/R196A). Fluorescence emission spectra of WT EntB and those of the purified variant proteins were superimposable, indicating that the folding of EntB variants was not significantly affected by mutagenesis at residue positions 21 and 196 (supplemental material, Figure S1). We also confirmed that the dimeric states of the variants were consistent with WT EntB via native polyacrylamide gel electrophoresis (PAGE) (supplemental material, Figure S2). In addition to confirming the relative purities of the variants, subtle migratory differences in the variants were observed that were consistent with charge substitutions affecting native gel migration. Given that the overall mobilities of the variants were similar to dimeric wild‐type EntB, none of the introduced mutations were determined to have disrupted the quaternary structures of the variant proteins.

We measured the isochorismatase activities of the EntB variants using an equilibrium mixture of chorismate:isochorismate and LDH. Reversal of charges at positions 21 and 196 resulted in an almost complete abrogation of intrinsic EntB isochorismatase catalytic activity (93% and 95% loss in activity for K21D and R196D, respectively). These two residues were predicted by our docked model to be located near the entry of the active site of EntB. The pronounced loss of activity observed for the aspartate substitutions was likely due to charge repulsion between the negatively charged isochorismate and the negatively charged EntB variant residues K21D and R196D. Alanine substitution at position 196 (R196A) resulted in approximately 50% loss of the catalytic activity of EntB, while alanine substitution at position 21 (K21A) resulted in only a 5% loss of activity. As expected, the double variant (K21A/R196A) exhibited an approximate 75% loss of activity. In the case of the alanine variants, residual electropositivity near the EntB active site was likely sufficient to allow for isochorismate entry into the EntB active site, albeit less efficiently. Taken together, this experimental evidence confirms the predictions generated from our docked model, demonstrating that positive charges at positions 21 and 196 serve as an electrostatic guide for negatively charged isochorismate, either for direct entry from the bulk medium or for guided channeling within the EntC‐EntB complex.

When the EntB alanine‐substituted variants were assayed in the presence of EntC using chorismate as a substrate, we observed an almost complete restoration of EntB isochorismatase activity for the EntB variants R196A and K21A/R196A (Figure 4). This EntC‐dependent increase in EntB isochorismatase activity is consistent with EntC‐EntB complexation providing steric protection from isochorismate diffusion into the bulk medium: formation of the transient EntC‐EntB complex can sterically trap exogenous isochorismate, promoting its catalysis by EntB over diffusion to the bulk medium. In contrast, no change in EntB variant activity was observed when EntA/NAD+ was added to the coupled assay mixture instead of EntC, suggesting that the observed effects on EntB activity were EntC‐specific (supplemental material, Figure S3). Interestingly, the EntB K21A variant was observed to be insensitive to the presence of EntC. These results suggest that the EntC‐EntB interaction itself stabilizes EntB, especially in the region of residue position 196, which is proximal to the EntB active site and at the terminus of the putative electrostatic channeling surface reported previously (Ouellette et al., 2022).

FIGURE 4.

FIGURE 4

Open in a new tab

The effect of EntC addition on the relative isochorismatase activities of EntB WT and variants. Reaction mixtures in the absence of EntC (solid) contained 0.1 μM EntB, 1.5 unit/mL LDH and fed with an equilibrium mixture of chorismate:Isochorismate. Reaction mixture in the presence of EntC (stripes) contained 0.05 μM EntC, 1.25 μM EntB, 1.5 unit/mL LDH; reactions were initiated with the addition of chorismate. Relative activity is defined here as the initial rate of EntB isochorismatase (wild type or variant, in the presence or absence of EntC) divided by the initial rate of wild‐type EntB isochorismatase as measured under identical conditions. Data are shown with the mean and standard deviation of triplicate independent measurements.

3.4. The roles of EntB K21 and R196 on substrate channeling in the EntC‐EntB complex

We employed our MenD competition assay to investigate substrate channeling between EntC and the K21 and R196 EntB variants. Given that K21D and R196D exhibited an almost complete loss of EntB isochorismate activity, we decided to investigate channeling efficiencies of the alanine variants only. As shown in Figure 5, residual EntB isochorismatase activity of the coupled assay reactions with the EntB K21A did not change as the concentration of MenD increased. In contrast, the variants EntB R196A and K21A/R196A were found to have residual isochorismatase activity decreased to 3% (in contrast to 16% residual activity observed for wild‐type EntB in the presence of excess MenD). This approximately 5‐fold decrease in isochorismatase channeling is consistent with the inability of the R196A variants to electrostatically guide isochorismate produced by EntC to the EntB active site. A single Ala substitution at position 21 did not have a dramatic effect on channeling while substitution at position 196 resulted in an abrogation of channeling. This suggests that the inability of R196A to channel isochorismate at wild‐type levels is not simply due to loss of a single‐point charge. Given that R196 occurs at the terminus of the channeling surface near the EntB active site, we hypothesize that the residue plays an additional mechanistic role in escorting isochorismate to the EntB active site. Taken together, our results demonstrate that EntC‐EntB complexation, and the concomitant stabilization of EntB R196, is necessary for efficient EntC‐EntB isochorismate channeling.

FIGURE 5.

FIGURE 5

Open in a new tab

MenD competition for isochorismate in the presence of EntB variants K21A and R196A. Coupled assay reactions were performed in the presence of increasing concentrations of MenD. Residual EntB isochorismatase activities in the presence of MenD are shown as percentages. Wild‐type EntB: Blue, EntB K21A: Orange, EntB R196A: Magenta; EntB K21A/R196A: Green. Data shown are the mean values of triplicate readings. The degree of channeling was calculated by dividing the activity of EntC‐EntB‐LDH coupled reaction in the presence of 3 μM MenD over that in the absence of MenD using wild‐type EntB.

4. CONCLUSIONS

Coupled EntC‐EntB‐LDH activity assays in the presence and absence of excess MenD, a competitor of EntB for isochorismate, demonstrated that isochorismate is partially channeled between EntC and EntB, presumably upon dynamic pairwise complex formation in vitro. Even though the observed in vitro channeling efficiency was low, this is the first reported evidence of substrate channeling within the DHB biosynthetic arm of the pathway. Under viscous conditions that emulate an intracellular environment, substrate channeling was observed to increase, as has been shown previously in other systems (Pareek et al., 2021). Given that EntC, EntB, EntA, and EntE likely participate in a multienzyme complex in vivo (Ouellette et al., 2022), the partial channeling that we observed may also be due to the absence of EntA and EntE in our in vitro pairwise assays; assembly of additional proteins into an EntCBAE multienzyme complex in vivo may further inhibit isochorismate diffusion into the bulk medium, thus resulting in increased channeling efficiency. Our experimental coupled assay data using EntB variants showed that two electropositive EntB residues, K21 and R196, are involved in electrostatically guiding isochorismate to the EntB active site. Furthermore, EntB residue R196 was found to play a key role in the direct channeling of isochorismate within the EntC‐EntB complex. A graphical representation of the EntC‐EntB complex with the positions of EntB K21 and R196 relative to the EntC and EntB active sites is shown in Figure 6. While we have demonstrated that the binary EntC‐EntB complex can partially channel isochorismate in vitro, this model suggests that these residues are somewhat solvent‐exposed, which would limit the possibility of enhancing isochorismate channeling via steric trapping in addition to being guided along an electrostatic surface. This could explain the leaky channeling that we report here. Recent research from our lab (manuscript in preparation) has demonstrated that the EntC and EntB proteins may be part of a larger multienzyme complex comprised of EntCBAE. Through the use of bacterial two‐hybrid experiments, we previously reported that EntB and EntA interact in vivo (Pakarian & Pawelek, 2016b). Could such an interaction assist in enhancing isochorismate channeling between EntC and EntB within a larger complex? We are now investigating the possibility that the presence of additional proteins (i.e., EntA, EntE) may assist in increasing EntC‐EntB channeling efficiency through further occlusion of this partially exposed surface upon multienzyme complex assembly.

FIGURE 6.

FIGURE 6

Open in a new tab

Docked model of the EntC‐EntB complex showing positions of K21 and R196 relative to the EntB active site. The proteins EntC (light blue) and EntB (green) are shown in cartoon representation. EntB residues K21 and R196 are displayed in stick representation. EntC active site residue positions are shaded yellow; EntB active site residue positions are shaded magenta. The gray space‐filling volume indicates the position of the proposed electrostatic channeling tunnel as reported previously (Ouellette et al., 2022). Molecular graphic generated using PyMOL.

AUTHOR CONTRIBUTIONS

Xue Bin: Writing – original draft; investigation; conceptualization; methodology; validation; visualization; writing – review and editing; software. Peter D. Pawelek: Conceptualization; investigation; funding acquisition; writing – original draft; methodology; writing – review and editing; software; project administration; supervision; resources.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Figure S1. Characterization of purified wild‐type EntB and variants by fluorescence spectroscopy. Normalized fluorescence emission spectra (λex = 280 nm) of solutions containing 2 μM of purified protein in Buffer A. EntB: red; EntB K21A: light green; EntB K21D: dark green; EntB R196A: light blue; EntB R196D: dark blue; EntB K21A/R196A: magenta.

Figure S2. Purified wild‐type EntB and variants resolved on a 12% native‐PAGE, stained with Coomassie Blue.

Figure S3. The effect of EntA addition on the relative isochorismatase activity of EntB wild‐type and variants. EntB‐EntA coupled reactions (solid) were in comparison with the EntB‐LDH coupled reactions (stripes). Reaction mixtures containing 0.1 μM EntB (wild‐type and variants), 15 μM EntA, and 10 mM NAD+ were assayed as of EntB‐LDH coupled assay. The data shown were the mean and standard deviation of triplicated independent measurements. Relative activity is defined here as the initial rate of EntB isochorismatase (wild type or variant, in the presence or absence of EntA) divided by the initial rate of wild‐type EntB isochorismatase as measured under identical conditions. Data are shown with the mean and standard deviation of triplicate independent measurements.

PRO-33-e5122-s001.docx (3.1MB, docx)

ACKNOWLEDGMENTS

This work was supported by Discovery Grant 341983 from the Natural Sciences and Engineering Research Council of Canada to P.D. Pawelek. Additional financial support for X. Bin was provided by J.W. McConnell Memorial Doctoral Fellowship. We acknowledge Dr. Eric Brown at McMaster University for providing menD ASKA strain. We thank Mary Judith Kornblatt for helpful discussusions regarding substrate channeling. We also thank Sylvie Ouellette for helpful comments related to the manuscript.

Bin X, Pawelek PD. Evidence of isochorismate channeling between the Escherichia coli enterobactin biosynthetic enzymes EntC and EntB . Protein Science. 2024;33(8):e5122. 10.1002/pro.5122

Review Editor: Lynn Kamerlin

REFERENCES

  1. Anderson KS. Fundamental mechanisms of substrate channeling. Enzyme kinetics and mechanism part E: energetics of enzyme catalysis. Methods in Enzymology Volume 308. Cambridge: Academic Press; 1999. p. 111–145. [DOI] [PubMed] [Google Scholar]
  2. Andrews SC, Robinson AK, Rodríguez‐Quiñones F. Bacterial iron homeostasis. FEMS Microbiol Rev. 2003;27(2–3):215–237. [DOI] [PubMed] [Google Scholar]
  3. Arentson BW, Sanyal N, Becker DF. Substrate channeling in proline metabolism. Front Biosci (Landmark Ed). 2012;17(1):375–388. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brickman TJ, Ozenberger BA, McIntosh MA. Regulation of divergent transcription from the iron‐responsive fepB‐entC promoter‐operator regions in Escherichia coli. J Mol Biol. 1990;212(4):669–682. [DOI] [PubMed] [Google Scholar]
  5. Bulutoglu B, Garcia KE, Wu F, Minteer SD, Banta S. Direct evidence for metabolon formation and substrate channeling in recombinant TCA cycle enzymes. ACS Chem Biol. 2016;11(10):2847–2853. [DOI] [PubMed] [Google Scholar]
  6. Buss K, Müller R, Dahm C, Gaitatzis N, Skrzypczak‐Pietraszek E, Lohmann S, et al. Clustering of isochorismate synthase genes menF and entC and channeling of isochorismate in Escherichia coli. Biochim Biophys Acta—Gene Struct Expr. 2001;1522(3):151–157. [DOI] [PubMed] [Google Scholar]
  7. Carrano CJ, Raymond KN. Ferric ion sequestering agents. 2. Kinetics and mechanism of iron removal from transferrin by enterobactin and synthetic tricatechols. J Am Chem Soc. 1979;101(18):5401–5404. [Google Scholar]
  8. Elcock AH, McCammon JA. Evidence for electrostatic channeling in a fusion protein of malate dehydrogenase and citrate synthase. Biochemistry. 1996;35(39):12652–12658. [DOI] [PubMed] [Google Scholar]
  9. Gasteiger E, Hoogland C, Gattiker A, et al. In: Walker JM, editor. Protein identification and analysis tools on the ExPASy server BT—the proteomics protocols handbook. Totowa, NJ: Humana Press; 2005. p. 571–607. [Google Scholar]
  10. Hubrich F, Müller M, Andexer JN. In vitro production and purification of isochorismate using a two‐enzyme cascade. J Biotechnol. 2014;191:93–98. [DOI] [PubMed] [Google Scholar]
  11. Jiang M, Cao Y, Guo Z‐F, Chen M, Chen X, Guo Z. Menaquinone biosynthesis in Escherichia coli: identification of 2‐Succinyl‐5‐enolpyruvyl‐6‐hydroxy‐3‐cyclohexene‐1‐carboxylate as a novel intermediate and Re‐evaluation of MenD activity. Biochemistry. 2007;46(38):10979–10989. [DOI] [PubMed] [Google Scholar]
  12. Jiang M, Guo Z. Effects of macromolecular crowding on the intrinsic catalytic efficiency and structure of Enterobactin‐specific Isochorismate synthase. J Am Chem Soc. 2007;129(4):730–731. [DOI] [PubMed] [Google Scholar]
  13. Khalil S, Pawelek PD. Ligand‐induced conformational rearrangements promote interaction between the Escherichia coli Enterobactin biosynthetic proteins EntE and EntB. J Mol Biol. 2009;393(3):658–671. [DOI] [PubMed] [Google Scholar]
  14. Khalil S, Pawelek PD. Enzymatic adenylation of 2,3‐dihydroxybenzoate is enhanced by a protein−protein interaction between Escherichia coli 2,3‐Dihydro‐2,3‐dihydroxybenzoate dehydrogenase (EntA) and 2,3‐dihydroxybenzoate‐AMP ligase (EntE). Biochemistry. 2011;50(4):533–545. [DOI] [PubMed] [Google Scholar]
  15. Kitagawa M, Ara T, Arifuzzaman M, Ioka‐Nakamichi T, Inamoto E, Toyonaga H, et al. Complete set of ORF clones of Escherichia coli ASKA library (a complete set of E. Coli K‐12 ORF archive): unique resources for biological research. DNA Res an Int J Rapid Publ Reports Genes Genomes. 2005;12(5):291–299. [DOI] [PubMed] [Google Scholar]
  16. Koglin A, Walsh CT. Structural insights into nonribosomal peptide enzymatic assembly lines. Nat Prod Rep. 2009;26(8):987–1000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kurutsch A, Richter M, Brecht V, Sprenger GA, Müller M. MenD as a versatile catalyst for asymmetric synthesis. J Mol Catal B: Enzym. 2009;61(1):56–66. [Google Scholar]
  18. Lai JR, Fischbach MA, Liu DR, Walsh CT. A protein interaction surface in nonribosomal peptide synthesis mapped by combinatorial mutagenesis and selection. Proc Natl Acad Sci. 2006;103(14):5314 LP–5319 LP. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Liu J, Quinn N, Berchtold GA, Walsh CT. Overexpression, purification, and characterization of isochorismate synthase (EntC), the first enzyme involved in the biosynthesis of enterobactin from chorismate. Biochemistry. 1990;29(6):1417–1425. [DOI] [PubMed] [Google Scholar]
  20. Miles EW, Rhee S, Davies DR. The molecular basis of substrate channeling. J Biol Chem. 1999;274(18):12193–12196. [DOI] [PubMed] [Google Scholar]
  21. Ouellette S, Pakarian P, Bin X, Pawelek PD. Evidence of an intracellular interaction between the Escherichia coli enzymes EntC and EntB and identification of a potential electrostatic channeling surface. Biochimie. 2022;202:159–165. [DOI] [PubMed] [Google Scholar]
  22. Ovádi J, Tompa P, Vértessy B, Orosz F, Keleti T, Welch GR. Transient‐time analysis of substrate‐channelling in interacting enzyme systems. Biochem J. 1989;257(1):187–190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Pakarian P, Pawelek PD. Subunit orientation in the Escherichia coli enterobactin biosynthetic EntA‐EntE complex revealed by a two‐hybrid approach. Biochimie. 2016a;127:1–9. [DOI] [PubMed] [Google Scholar]
  24. Pakarian P, Pawelek PD. Intracellular co‐localization of the Escherichia coli enterobactin biosynthetic enzymes EntA, EntB, and EntE. Biochem Biophys Res Commun. 2016b;478(1):25–32. [DOI] [PubMed] [Google Scholar]
  25. Pareek V, Sha Z, He J, Wingreen NS, Benkovic SJ. Metabolic channeling: predictions, deductions, and evidence. Mol Cell. 2021;81(18):3775–3785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Rausch MH, Heller A, Fröba AP. Binary diffusion coefficients of glycerol–water mixtures for temperatures from 323 to 448 K by dynamic light scattering. J Chem Eng Data. 2017;62(12):4364–4370. [Google Scholar]
  27. Raymond KN, Dertz EA, Kim SS. Enterobactin: an archetype for microbial iron transport. Proc Natl Acad Sci. 2003;100:3584–3588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Rusnak F, Liu J, Quinn N, Berchtold GA, Walsh CT. Subcloning of the enterobactin biosynthetic gene entB: expression, purification, characterization, and substrate specificity of isochorismatase. Biochemistry. 1990;29(6):1425–1435. [DOI] [PubMed] [Google Scholar]
  29. Sambrook J, Russell DW. Preparation and transformation of Competent E. Coli using calcium chloride. Cold Spring Harb Protoc. 2006;2006(1):pdb.prot3932. [DOI] [PubMed] [Google Scholar]
  30. Sanyal N, Arentson BW, Luo M, Tanner JJ, Becker DF. First evidence for substrate channeling between proline catabolic enzymes: a validation of domain fusion analysis for predicting protein‐protein interactions. J Biol Chem. 2015;290(4):2225–2234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Spivey HO, Ovádi J. Substrate channeling. Methods. 1999;19(2):306–321. [DOI] [PubMed] [Google Scholar]
  32. Stourac J, Vavra O, Kokkonen, P , et al., “Caver Web 1.0: identification of tunnels and channels in proteins and analysis of ligand transport.” Nucleic Acids Res. 2019;47(W1):W414–W422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sundlov JA, Shi C, Wilson DJ, Aldrich CC, Gulick AM. Structural and functional investigation of the intermolecular interaction between NRPS adenylation and carrier protein domains. Chem Biol. 2012;19(2):188–198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Svedružić ŽM, Odorčić I, Chang CH, Svedružić D. Substrate channeling via a transient protein‐protein complex: the case of D‐Glyceraldehyde‐3‐phosphate dehydrogenase and L‐lactate dehydrogenase. Sci Rep. 2020;10(1):10404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Zhu A, Romero R, Petty HR. A sensitive fluorimetric assay for pyruvate. Anal Biochem. 2010;396(1):146–151. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1. Characterization of purified wild‐type EntB and variants by fluorescence spectroscopy. Normalized fluorescence emission spectra (λex = 280 nm) of solutions containing 2 μM of purified protein in Buffer A. EntB: red; EntB K21A: light green; EntB K21D: dark green; EntB R196A: light blue; EntB R196D: dark blue; EntB K21A/R196A: magenta.

Figure S2. Purified wild‐type EntB and variants resolved on a 12% native‐PAGE, stained with Coomassie Blue.

Figure S3. The effect of EntA addition on the relative isochorismatase activity of EntB wild‐type and variants. EntB‐EntA coupled reactions (solid) were in comparison with the EntB‐LDH coupled reactions (stripes). Reaction mixtures containing 0.1 μM EntB (wild‐type and variants), 15 μM EntA, and 10 mM NAD+ were assayed as of EntB‐LDH coupled assay. The data shown were the mean and standard deviation of triplicated independent measurements. Relative activity is defined here as the initial rate of EntB isochorismatase (wild type or variant, in the presence or absence of EntA) divided by the initial rate of wild‐type EntB isochorismatase as measured under identical conditions. Data are shown with the mean and standard deviation of triplicate independent measurements.

PRO-33-e5122-s001.docx (3.1MB, docx)

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society

RESOURCES