Anion Inhibition Studies of the Beta-Carbonic Anhydrase from Escherichia coli

The interconversion of CO2 and HCO3− is catalyzed by a superfamily of metalloenzymes, known as carbonic anhydrases (CAs, EC 4.2.1.1), which maintain the equilibrium between dissolved inorganic CO2 and HCO3−. In the genome of Escherichia coli, a Gram-negative bacterium typically colonizing the lower intestine of warm-blooded organisms, the cyn operon gene includes the CynT gene, encoding for a β-CA, and CynS gene, encoding for the cyanase. CynT (β-CA) prevents the depletion of the cellular bicarbonate, which is further used in the reaction catalyzed by cyanase. A second β-CA (CynT2 or Can or yadF), as well as a γ and ι-CAs were also identified in the E. coli genome. CynT2 is essential for bacterial growth at atmospheric CO2 concentration. Here, we characterized the kinetic properties and the anion inhibition profiles of recombinant CynT2. The enzyme showed a good activity for the physiological CO2 hydratase reaction with the following parameters: kcat = 5.3 × 105 s−1 and kcat/KM = of 4.1 × 107 M−1 s−1. Sulfamide, sulfamate, phenylboronic acid, phenylarsonic acid, and diethyldithiocarbamate were the most effective CynT2 inhibitors (KI = 2.5 to 84 µM). The anions allowed for a detailed understanding of the interaction of inhibitors with the amino acid residues surrounding the catalytic pocket of the enzyme and may be used as leads for the design of more efficient and specific inhibitors.


Introduction
Escherichia coli is a bacterium discovered in 1885 by the German bacteriologist Theodor Escherich who isolated it from the feces of a newborn [1]. This microorganism was initially named Bacterium coli, and lately, the term was modified to Escherichia coli to honor Escherich [1]. Escherichia coli is a harmless microbe, which typically colonizes the infant gastrointestinal tract within the first hours of life, establishing a mutual benefit with its host [2][3][4]. However, when the gastrointestinal mucosa is damaged by various factors affecting its integrity, the harmless microbe disseminates and provokes infection in the body, becoming a pathogen, which provokes a wide spectrum of diseases [2,5]. Although its discovery dates back to the previous century, only in 1935 E. coli was identified as the etiological Molecules 2020, 25 agent responsible for the outbreak of diarrhea among infants [1]. The common genus Escherichia coli contains a broad variety of different forms: (i) pathogenic microorganisms, which can lead to death, or triggering severe disease outbreaks worldwide as well as serious infections, such as watery diarrhea, bloody diarrhea, urinary tract infection, meningitis, and sepsis [6][7][8]; (ii) opportunistic pathogens, which can cause disease if the host defenses are weakened [9]; and (iii) commensal microorganisms that innocuously colonize the healthy intestine of warm-blooded animals, including humans, with mutual benefits [10][11][12]. During their growth, bacteria need carbon dioxide (CO 2 ) and bicarbonate (HCO 3 − ), which are necessary for supporting the central metabolism [13,14]. The interconversion of inorganic CO 2 and HCO 3 − is naturally and correctly balanced to maintain the equilibrium between dissolved CO 2 and HCO 3 − [15][16][17][18]. The naturally occurring reaction of interconversion of CO 2 [13,14,21].
The catalyzed reaction has a k cat ranging from 10 4 to 10 6 s −1 [22,23] and is carried out by a superfamily of ubiquitous metalloenzymes known as carbonic anhydrases (CAs, EC 4.2.1.1) [24][25][26][27][28]. The CA superfamily includes eight genetically distinct families (or classes), named with the Greek letters, α, β, γ, δ, ζ, η, θ, and ι [13,20,29]. The last three classes were only recently discovered [30,31]. Up to now, the exploration of the bacterial genome revealed only four of the eight CA-classes: α, β, γ, and ι [20,28,29,[32][33][34][35][36], showing an intricate gene pattern distribution since the genome of some bacteria encodes for one, two, or even three different CA-families [13,20]. A fourth class, ι-CAs, recently discovered, was identified by our groups in the genome of Burkholderia territorii, which is a Gram-negative bacterium found in soil and water, which often shows resistance to common antibiotics [29,37]. In the genome of Escherichia coli, the cyn operon gene includes the CynT gene, encoding for a β-CA, and CynS gene, encoding for the cyanase, which catalyzes the reaction of cyanate with bicarbonate to give ammonia and carbon dioxide [38][39][40]. It has been hypothesized that the β-CA (CynT), catalyzing the CO 2 hydration, prevents the depletion of the cellular bicarbonate, which thereafter will be used in the reaction catalyzed by the enzyme cyanase. Exploring the E. coli genome, a second β-CA (CynT2 or Can or yadF) was identified [41], whereas a γ-CA and a ι-CA (annotated as SgcJ/EcaC family oxidoreductase) were discovered by our groups (unpublished data from our laboratory and manuscript in preparation). CynT2 was characterized for its three-dimensional structure and for its essential role in allowing bacterial growth at atmospheric pCO 2 [21,41]; no such information is available on the γ-and ι-CAs. In the latter context, it seems that the activity of E. coli CAs can promote bacterial growth and adaptation in the host. This observation is corroborated by the in vivo results demonstrating that CAs are crucial macromolecules for survival, pathogenicity, and virulence of several species of human pathogens, such as Helicobacter pylori [42][43][44], Vibrio cholerae [45], Brucella suis [46][47][48][49], Salmonella enterica [50], and Pseudomonas aeruginosa [51]. In this context, here, using a stopped-flow technique, we investigated the kinetic constants of the recombinant CynT2, a β-CA identified in the genome of Escherichia coli, for which the kinetic characterization has not yet been reported. Furthermore, since CynT2 is essential for bacterial growth at atmospheric CO 2 concentration, its inhibition profile has been explored with a broad range of inorganic metal-complexing anions. These inhibitors are among the classical CA inhibitors (CAIs) and are very attractive because they are small molecules/ions, which can be efficiently transported in the body; in addition, they can exploit oxidation and ligand substitution reactions [52,53]. The aim of this work was to identify in vitro efficient inhibitors of CynT2, whose inactivation could impair the microbe diffusion in the host. Moreover, since E. coli is a microorganism that can be handled without risk in the laboratory, it might represent a bacterial study model to be used safely in vitro for cell-based tests, diversely from other human and animal pathogens, which require the use of particular levels of protection for reducing risks of contaminations. Moreover, we believe that the results of this study will be useful in further exploring novel approaches for the inhibition of bacterial CAs, which may lead to alternatives in the use of standard antibiotics for contrasting the growth and virulence of both human and animal pathogens.

Primary Structure Analysis
The genome of Escherichia coli was inspected with BLAST (Basic Local Alignment Search Tool) to identify the CA-classes encoded by the bacterial genome, using individually the amino acid sequences belonging to the eight CA-classes (α, β, γ, δ, ζ, η, θ, and ι) as query sequences. Table 1 summarizes the results obtained with the BLAST analysis. Three CA classes were identified in the E. coli genome: β, γand ι-CAs; furthermore, different isoforms were detected for each class of CAs.  Table 1 shows that no representative of the α-class was detected in the E. coli genome. In our previous works, it has been documented that the bacterial α-CAs identified in the Gram-negative bacteria are characterized by the presence of a short secretory or signal peptide at the N-terminal end of the polypeptide chain [13,19,29]. This short peptide allows the translocation of the neo-synthesized protein into the bacterial periplasmic space, which is a typical feature of the Gram-negative bacteria. Lately, this secretory signal was also identified in the amino terminus of some polypeptide chains of β-, γ-, and ι-CAs from Gram-negative bacteria [13,29]. These findings prompted us to investigate the N-terminal portion of the two E. coli β-CA sequences reported in Table 1 (CynT and CynT2, with the GenBank IDs WP_033547590.1 and EEW0221051.1, respectively) for the presence of a putative signal peptide. Figure 1 shows the results obtained using the bioinformatics tool "SignalP 4.1" (http://www.cbs.dtu.dk/services/SignalP/), which is a software optimized for the prediction of a signal peptide in Bacteria, Archaea, and Eukarya.
From Figure 1, it is readily apparent that all the scores represented in the SignalP output were very low, close to the value of 0.1. Thus, E. coli β-CAs seemed characterized by the absence of a secretory signal at the amino terminus of their sequence. The translocation of CAs in the space between the two bacterial membranes (outer and inner layers) of the Gram-negative bacteria guarantees the rapid conversion of the periplasmic CO 2 to bicarbonate, avoiding its depletion. Furthermore, the bicarbonate transported in the cytoplasm will be used by the cytoplasmic CA-classes (β and γ) to produce the CO 2 /HCO 3 − for the central bacterial metabolism [13,29]. From this consideration, it can be assumed that in E. coli, the lack of the α-CA is compensated by the presence of the ι-CA, whose polypeptide chain is typified by a secretory signal at the N-amino terminal [29]. This assumption is also corroborated by the existence of βor γ-CAs with a signal peptide in Gram-negative bacteria, whose genome doesn't encode for any α-CA. Figure 2 shows the multialignment of the two β-CAs amino acid sequences reported in Table 1 with a representative bacterial CA sequence belonging to the same family.
β-, γ-, and ι-CAs from Gram-negative bacteria [13,29]. These findings prompted us to investigate the N-terminal portion of the two E. coli β-CA sequences reported in Table 1 (CynT and CynT2, with the GenBank IDs WP_033547590.1 and EEW0221051.1, respectively) for the presence of a putative signal peptide. Figure 1 shows the results obtained using the bioinformatics tool "SignalP 4.1" (http://www.cbs.dtu.dk/services/SignalP/), which is a software optimized for the prediction of a signal peptide in Bacteria, Archaea, and Eukarya.  From Figure 1, it is readily apparent that all the scores represented in the SignalP output were very low, close to the value of 0.1. Thus, E. coli β-CAs seemed characterized by the absence of a secretory signal at the amino terminus of their sequence. The translocation of CAs in the space between the two bacterial membranes (outer and inner layers) of the Gram-negative bacteria guarantees the rapid conversion of the periplasmic CO2 to bicarbonate, avoiding its depletion. Furthermore, the bicarbonate transported in the cytoplasm will be used by the cytoplasmic CAclasses (β and γ) to produce the CO2/HCO3 − for the central bacterial metabolism [13,29]. From this consideration, it can be assumed that in E. coli, the lack of the α-CA is compensated by the presence of the ι-CA, whose polypeptide chain is typified by a secretory signal at the N-amino terminal [29]. This assumption is also corroborated by the existence of β-or γ-CAs with a signal peptide in Gramnegative bacteria, whose genome doesn't encode for any α-CA. Figure 2 shows the multialignment of the two β-CAs amino acid sequences reported in Table 1 with a representative bacterial CA sequence belonging to the same family. The isoform CynT showed an amino acid sequence identity of 28%, when compared with the isoform CynT2. The identity of CynT or CynT2 with the VchCA_beta from Vibrio cholerae was of 28% and 61%, respectively. This means that the two isoforms are characterized by multiple amino acid substitutions, even if the catalytic triad (two Cys and on His) is perfectly conserved ( Figure 2). All the catalytically active CAs contain, independently of the genetic groups, a metal ion cofactor, which is necessary for enzyme catalysis [13,20,28,34]. The β-CAs use as catalytic metal a Zn 2+ ion, which is coordinated by three amino acid residues; the fourth ligand is a water molecule/hydroxide ion acting as the nucleophile in the catalytic enzyme cycle, or an Asp residue in Type II β-CAs as CynT2 [41]. Figure 3 shows the X-ray crystal structure of the CynT2 from E. coli, which has been solved at 2.0 Å resolution; it reveals a dimeric arrangement of the protein, which is a tetramer, as two dimers interact with each other in the crystal packing [41]. The enzyme structure was determined as the closed conformation (type II β-CA) in which the zinc is tetrahedrally coordinated to Cys42, Asp44, His98, and Cys101 [41] ( Figure 3A,B). The Asp44 side chain replacing the fourth ligand (water molecule) forms a non-canonical CA active site, which does not allow the CO2 hydration activity. This inactive form (closed active site) is present at pH values of less than 8.0 [41]. At pH values of ≥8.3, an incoming water molecule replaces the carboxylate moiety of the Asp residue, generating the nucleophile used in the typical catalytic cycle of the CAs [22,23]. Thus, the closed active site is converted to the open and active form (type I β-CA). As shown in Figure 3B, the closed structure is stabilized by HCO3 − , which occurs in a non-catalytic binding pocket close to the zinc ion, as reported for other few β-CAs, such as those from and Haemophilus influenzae and Vibrio cholerae [54,55]. The isoform CynT showed an amino acid sequence identity of 28%, when compared with the isoform CynT2. The identity of CynT or CynT2 with the VchCA_beta from Vibrio cholerae was of 28% and 61%, respectively. This means that the two isoforms are characterized by multiple amino acid substitutions, even if the catalytic triad (two Cys and on His) is perfectly conserved (Figure 2). All the catalytically active CAs contain, independently of the genetic groups, a metal ion cofactor, which is necessary for enzyme catalysis [13,20,28,34]. The β-CAs use as catalytic metal a Zn 2+ ion, which is coordinated by three amino acid residues; the fourth ligand is a water molecule/hydroxide ion acting as the nucleophile in the catalytic enzyme cycle, or an Asp residue in Type II β-CAs as CynT2 [41]. Figure 3 shows the X-ray crystal structure of the CynT2 from E. coli, which has been solved at 2.0 Å resolution; it reveals a dimeric arrangement of the protein, which is a tetramer, as two dimers interact with each other in the crystal packing [41]. The enzyme structure was determined as the closed conformation (type II β-CA) in which the zinc is tetrahedrally coordinated to Cys42, Asp44, His98, and Cys101 [41] (Figure 3A,B). The Asp44 side chain replacing the fourth ligand (water molecule) forms a non-canonical CA active site, which does not allow the CO 2 hydration activity. This inactive form (closed active site) is present at pH values of less than 8.0 [41]. At pH values of ≥8.3, an incoming water molecule replaces the carboxylate moiety of the Asp residue, generating the nucleophile used in the typical catalytic cycle of the CAs [22,23]. Thus, the closed active site is converted to the open and active form (type I β-CA). As shown in Figure 3B, the closed structure is stabilized by HCO 3 − , which occurs in a non-catalytic binding pocket close to the zinc ion, as reported for other few β-CAs, such as those from and Haemophilus influenzae and Vibrio cholerae [54,55].

Production of Recombinat β-CA (CynT2)
We produced the bacterial CynT2 encoded in the E. coli genome, since a detailed investigation of its kinetic parameters is lacking to date, although, as mentioned above, the CynT2 threedimensional structure has been solved [41]. Furthermore, this enzyme seems to be essential for bacterial growth at atmospheric pCO2 [21]. The purified enzyme, throughout the purification steps, was monitored following the Wilbur-Anderson Units (WAU) as described previously by Capasso's group [56]. Figure 4 shows the purity of the recombinant CynT2 after the affinity column. Three biochemical techniques were used to verify the heterologous overexpression and purification of the bacterial enzyme, namely SDS-PAGE, Western Blot (WB), and protonography. SDS-PAGE and WB reported in Figure 4, (lane 1 and 2) indicated that the CynT2 fusion protein was purified to the homogeneity as a subunit with an apparent molecular weight of about 29.0 kDa. Moreover, the developed protonogram obtained by the protonography analysis evidenced that the recombinant enzyme was catalytically active (lane 3). It evidenced a yellow band due to the production of ions (H + ) during the enzymatic CO2 hydration reaction at the molecular weight of 29.0 kDa, which corresponds to the mass of the recombinant CynT2 (Figure 4, lane 3).

Production of Recombinat β-CA (CynT2)
We produced the bacterial CynT2 encoded in the E. coli genome, since a detailed investigation of its kinetic parameters is lacking to date, although, as mentioned above, the CynT2 three-dimensional structure has been solved [41]. Furthermore, this enzyme seems to be essential for bacterial growth at atmospheric pCO 2 [21]. The purified enzyme, throughout the purification steps, was monitored following the Wilbur-Anderson Units (WAU) as described previously by Capasso's group [56]. Figure 4 shows the purity of the recombinant CynT2 after the affinity column. Three biochemical techniques were used to verify the heterologous overexpression and purification of the bacterial enzyme, namely SDS-PAGE, Western Blot (WB), and protonography. SDS-PAGE and WB reported in Figure 4, (lane 1 and 2) indicated that the CynT2 fusion protein was purified to the homogeneity as a subunit with an apparent molecular weight of about 29.0 kDa. Moreover, the developed protonogram obtained by the protonography analysis evidenced that the recombinant enzyme was catalytically active (lane 3). It evidenced a yellow band due to the production of ions (H + ) during the enzymatic CO 2 hydration reaction at the molecular weight of 29.0 kDa, which corresponds to the mass of the recombinant CynT2 (Figure 4, lane 3).

Determination of the Kinetic Parameters Using the Stopped-Flow Technique
Using CO 2 as a substrate, the recombinant CynT2 was subject to a stopped-flow analysis for the determination of the kinetic constants for the CO 2 hydratase activity. As shown in Table 2, CynT2 showed a good biocatalyst activity for the physiological CO 2 hydratase reaction to bicarbonate and protons, with k cat of 5.3 × 10 5 s −1 and catalytic efficiency (k cat /K M ) of 4.1 × 10 7 M −1 s −1 . In fact, the CynT2 kinetic constants were similar to those obtained for other bacterial CAs, as well as for hCA I. CynT2 was also inhibited by the sulfonamide acetazolamide (K I = 227 nM), which is a well-known pharmacological CA inhibitor (Table 2). bacterial enzyme, namely SDS-PAGE, Western Blot (WB), and protonography. SDS-PAGE and WB reported in Figure 4, (lane 1 and 2) indicated that the CynT2 fusion protein was purified to the homogeneity as a subunit with an apparent molecular weight of about 29.0 kDa. Moreover, the developed protonogram obtained by the protonography analysis evidenced that the recombinant enzyme was catalytically active (lane 3). It evidenced a yellow band due to the production of ions (H + ) during the enzymatic CO2 hydration reaction at the molecular weight of 29.0 kDa, which corresponds to the mass of the recombinant CynT2 (Figure 4, lane 3).  The results of Table 2 show that CynT2 was sensitive to 5-acetamido-1,3,4-thiadiazole-2sulfonamide (acetazolamide, AZA) inhibition, similarly to the human isoform I (K I = 250 nM), but it was 19 times less inhibited than the human isoform hCA II (K I = 12 nM). Again, CynT2 with respect to VchCA_alpha from V. cholerae was 33 times less sensitive to AZA inhibition and 3.5 times less compared to the ι-CA from B. territorii (K I = 65 nM). The comparison with the β-CA from V. cholerae (K I = 451 nM) showed that the E. coli enzyme with a K I of 227 nM is two times more susceptible to AZA inhibition. These results are of extreme importance in the field of the inhibition of bacterial CAs because they prove that, even if these enzyme catalyze the same reaction, they can show a different inhibition pattern, whose investigation can lead to the discovery of novel inhibitors, which may impair the microbial growth as well as their virulence [57]. Table 2. CynT2 kinetic parameters for the CO 2 hydration reaction. The CynT2 calculated kinetic constants were compared with those determined for the two human isoforms hCA I and II (α-class), the α-, β-, γ-CAs from Vibrio cholerae and ι-CAs from Burkholderia territorii. The reaction was performed at 25 • C, in 20 mM Tris buffer and 20 mM NaClO 4 , pH 8.3. Inhibition data with the clinically used acetazolamide are also provided.

Inhibition Profile of Inorganic Metal-Complexing Anions
Since CAs are crucial for bacterial growth as well as for their virulence, we decided to investigate the inhibition profile of CynT2 with a broad range of inorganic metal-complexing anions (Table 1).
Anions and small molecules, such as diethyldithiocarbamate, iminodisulfonate, sulfamide, sulfamate, phenylboronic and phenlylarsonic acids, can complex the metal ion (orange sphere of the Figure 3) of the enzyme catalytic pocket, hindering the hydration of the CO 2 to bicarbonate and protons. They bind the Zn 2+ ion of the enzyme either in a tetrahedral geometry or as trigonal-bipyramidal adducts of the metal ion [59]. Generally, these inhibitors show an inhibition constant (K I ) in the millimolar range. Even if the anion inhibitors are usually less effective than sulfonamides (K I is nM range), their investigation is essential for two fundamental aspects: (i) to design more efficient and selective inhibitors for the various CA-classes and their isoforms; (ii) potential for clinical applications for the treatment of diseases caused by pathogens, including bacteria. Here, we report the inhibition profile of CynT2 comparing these data with those obtained for the two human α-CAs, the isoforms hCA I and hCAII, and the bacterial enzyme (VchCA_beta), which have been previously investigated [58]. Table 3 lists the obtained results, which prompted us to elaborate the following observations:  Table 3). Within this group, VchCA_beta was inhibited with a K I ranging from 5.7 to 9.0 mM (Table 3), except for OsO 5 2− , PF 6 − , and CF 3 SO 3 − , whose K Is were even not measurable (Table 3). Moreover, N 3 − and P 2 O 7 2− showed a K I > 10 mM for the CynT2, while ClO 4 − with a K I > 200 mM was an ineffective inhibitor of CynT2 as well as the V. cholerae enzyme. Interesting to note that some of these anions resulted to be quite effective inhibitors, when tested on the two human CAs. For example, cyanide and azide inhibited the human isoform h CA I with K I of 0.5 and 12 µM, respectively (Table 3). Thus, low concentrations of N 3 − and CN − poison the two human enzymes as well as humans, whereas the bacterial CAs may better tolerate such toxic anions.

2.
Another group of anion inhibitors, which weakly inhibited CynT2, showed a K I in the range between 1.  Table 1). The differences in the values of the inhibition constants can be due to the amino acid residues surrounding the catalytic pocket, which influence the interaction of the anion with the enzyme, even if the three-dimensional structure of the β-CAs from E. coli and V. cholerae are very similar. Intriguing is the result obtained using bicarbonate and carbonate as inhibitors. These two anions are not effective inhibitors (K I = 12-85 mM) for the two human isoforms (hCA I and hCA II), as well as for the β-CA from V. cholerae (K I = 5.9-6.7 mM). These results may reflect an evolutionary adaptation of the human and Vibrio CAs due to their continuous exposition to the high concentration of these two anions. The human enzymes are adapted to the high concentration of carbonate and bicarbonate present in the plasma, while V. cholerae colonizes the upper part of the small intestine characterized by high concentrations of bicarbonate, which is also a potent inducer of the expression of the genes involved in the virulence of the pathogen [45].
that anions investigated here show the same mechanism for the inhibition of CynT2 as the anions crystallized with various other αand β-CAs.

Chemicals and Instruments
All the chemicals used in this study were of reagent grade and purchased from Sigma (Milano, Italy). The Affinity column (His-Trap FF) and the AKTA-Prime purification system were bought from GE Healthcare (Chicago, IL, USA). The SX20 Stopped-Flow was obtained by the AppliedPhotophysics. SDS-PAGE and Western-Blot apparatus were procured by BioRAD (Hercules, CA, USA).

Cloning, Expression, and Purification
The synthetic Escherichia coli gene encoding for the CynT2 was synthesized by the Invitrogen GeneArt (ThermoFisher Scientific, Waltham, MA, USA), a company specialized in gene synthesis, and cloned into the expression vector pET100D-Topo/CynT2. Briefly, the gene was designed to produce the recombinant CynT2 as fusion proteins with a tag containing nucleotides encoding for six histidines (His-Tag) at the amino terminus of neosynthesized recombinant protein. Competent E. coli BL21 (DE3) codon plus cells (Agilent) were transformed as described by Del Prete et al. [68]. Isopropyl β-D-1-thiogalactopyranoside (IPTG) at the concentration of 1 mM was added to the cellular culture to overexpress the recombinant CynT2. After growth, the cells were harvested and disrupted by sonication. Cellular extract was purified using a nickel affinity column (His-Trap FF), which allows the interaction between the matrix functionalized with Ni 2+ ion and the His-Tag at the N-terminus of the protein. The HisTrap column (1 mL) was equilibrated with a 20-mL equilibration buffer (50 mM Tris, 20 mM imidazole and 150 mM sodium chloride, pH 7.5) at 1 mL/min. The supernatant from the cellular lysate was loaded onto the column at 1 mL/min, connected with AKTA Prime. The recombinant CynT2 was eluted from the column by fluxing a linear gradient of imidazole (0-300 mM) at a flow of 0.5 mL/min in a buffer composed of 50 mM Tris and 300 mM sodium chloride, pH 7.5. The recovered CynT2 was 90% pure. The protein quantification was carried out by Bradford method (BioRAD) [69].

Carbonic Anhydrase Assay for Monitoring the Recombinant Enzyme during the Purification Steps
CA activity assay was performed as described by Capasso et al. [56]. Briefly, the assay was based on the monitoring of pH variation due to the catalyzed conversion of CO 2 to bicarbonate. Bromothymol blue was used as the indicator of pH variation. The assay was performed at 0 • C and a CO 2 -satured solution was used as substrate. The enzyme activity was calculated by measuring the time required for Bromothymol blue to change from blue to yellow. This time is inversely related to the quantity of enzyme present in the sample and allows the calculation of the Wilbur-Anderson units as described previously [56].

SDS-PAGE
A 12% Sodium Dodecyl Sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) prepared as described by Laemmli [70] was used, loading on the gel the recovered CynT2 from the affinity column. The gel was stained with Coomassie Brilliant Blue-R.

Western Blot
CynT2 was subjected to a 12% (w/v) SDS-PAGE, and then was transferred to a PVDF (polyvinylidene fluoride) membrane with transfer buffer (25 mM Tris, 192 mM glycine, 20% methanol) using Trans-Plot SD Cell (Bio-Rad, Hercules, CA, USA). His-Tag Western blot was carried out using the Pierce Fast Western Blot Kit (Thermo Scientific, Waltham, MA, USA). Blotted membrane was placed in the wash blot solution Fast Western 1 Wash Buffer to remove transfer buffer. Primary Antibody Working Dilution was added to the blot and incubated for 30 min at room temperature (RT) with shaking. Invitrogen anti-His antibody (1:10,000) was used. Afterwards, the blot was removed from the primary antibody solution and incubated for 10 min with the FastWestern Optimised HRP ReagentWorking Dilution. Subsequently, the membrane was washed two times in about 20 mL of FastWestern 1 Wash Buffer. Finally, the membrane was incubated with the detection reagent working solution and incubated for 1 min, at room temperature, and then developed with X-ray film.

Protonography
To perform the protonography, wells of 12% SDS-PAGE gel were loaded with samples mixed with loading buffer not containing 2-mercaptoethanol and not subjected to boiling, in order to avoid protein denaturation. The gel was run at 150 V until the dye front ran off the gel. Following the electrophoresis, the 12% SDS-PAGE gel was subject to protonography to detect the yellows bands due to the hydratase activity on the gel as described previously [71][72][73][74].

Kinetic Parameters and Inhibition Constants Determined by the Stopped-Flow Technique
The CO 2 hydration activity performed by the BteCAι was monitored using an Applied Photophysics stopped-flow instrument [59]. Phenol red (at a concentration of 0.2 mM) was used as indicator, working at the absorbance maximum of 557 nm, with 20 mM TRIS (pH 8.3) as buffer, and 20 mM NaClO 4 (for maintaining constant the ionic strength), following the initial rates of the CA-catalyzed CO 2 hydration reaction for a period of 10-100 s. To determine the kinetic parameters by Lineweaver-Burk plots and the inhibition constants, a concentration of CO 2 between 1.7 to 17 mM was used. At least six measurements of the original 5-10% reaction were used to assess the initial velocity for each inhibitor. The uncatalyzed rates were identically determined and detracted from the total observed rates. Stock inhibitor solutions (10-100 mM) were prepared in distilled-deionized water and dilutions up to 0.01 mM were done with the buffer test. Inhibitor and enzyme solutions were preincubated together for 15 min at room temperature prior to assay, in order to allow for the formation of the E-I complex or for the eventual active site mediated hydrolysis of the inhibitor. The inhibition constants were obtained by non-linear least-squares methods using PRISM 6 and the Cheng-Prusoff equation, as reported earlier [58,60,61], and represent the mean from at least three different determinations. All CA isoforms were recombinant ones obtained in-house.

Conclusions
The E. coli genome encodes only for three of the eight CA-classes reported in the literature: β-, γ-, and ι-CAs. In the present paper, we produced the recombinant isoform CynT2, a CA belonging to the β-class. The enzyme kinetic parameters and its anion inhibition profile were determined, using the stopped-flow technique. CynT2 resulted to be a good biocatalyst for the CO 2 hydration reaction showing a k cat = 5.3 × 10 5 and a k cat /K M = 4.1 × 10 7 . These values are similar to those obtained for other enzymes belonging to the β-CA class. Moreover, we also investigated the inhibition profile of CynT2 with a broad range of inorganic metal-complexing anions, a well-known group of CA inhibitors capable of blocking the enzyme activity. Small molecules such as sulfamide, sulfamate, phenylboronic acid, phenylarsonic acid, and diethyldithiocarbamate resulted to be the most effective CynT2 inhibitors (K I = 2.5 to 84 µM). Despite these metal-complexing anions showed inhibition constants in the millimolar range, this anion investigation study is relevant, because it allowed a better understanding of the interaction of the CA inhibitors with the amino acid residues surrounding the enzyme catalytic pocket as well as the design of more efficient and specific inhibitors. Besides, it is important to keep in mind that the physiological role of CAs is to balance pH, CO 2 , and bicarbonate inside the bacterial cell, ensuring the right amounts of these molecules/ions to the bacterial metabolism. Thus, inhibition of such enzymes can impair bacterial growth as well as their virulence through a metabolic pathway, which is different from those used by the common pharmacological antibiotics. These could open a new solution to antibiotic resistance, which is occurring worldwide. Finally, E. coli is a microorganism that can easily be manipulated in the laboratory, avoiding the risk connected to the handling of pathogenic bacteria. Considering this aspect, E. coli could be considered as a model study organism for testing CA inhibitors in vitro, in cell-based assays, evaluating the effect of the inhibition on bacterial cell growth in a safe and facile way. To address the issues mentioned above, it is necessary to analyze in vitro the effect of CA inhibitors on all CA-classes encoded by the E. coli genome. That's what we are doing.
Funding: This research received no external funding.