Inhibitory Effects of Sulfonamide Derivatives on the β-Carbonic Anhydrase (MpaCA) from Malassezia pachydermatis, a Commensal, Pathogenic Fungus Present in Domestic Animals

Fungi are exposed to various environmental variables during their life cycle, including changes in CO2 concentration. CO2 has the potential to act as an activator of several cell signaling pathways. In fungi, the sensing of CO2 triggers cell differentiation and the biosynthesis of proteins involved in the metabolism and pathogenicity of these microorganisms. The molecular machineries involved in CO2 sensing constitute a promising target for the development of antifungals. Carbonic anhydrases (CAs, EC 4.2.1.1) are crucial enzymes in the CO2 sensing systems of fungi, because they catalyze the reversible hydration of CO2 to proton and HCO3-. Bicarbonate in turn boots a cascade of reactions triggering fungal pathogenicity and metabolism. Accordingly, CAs affect microorganism proliferation and may represent a potential therapeutic target against fungal infection. Here, the inhibition of the unique β-CA (MpaCA) encoded in the genome of Malassezia pachydermatis, a fungus with substantial relevance in veterinary and medical sciences, was investigated using a series of conventional CA inhibitors (CAIs), namely aromatic and heterocyclic sulfonamides. This study aimed to describe novel candidates that can kill this harmful fungus by inhibiting their CA, and thus lead to effective anti-dandruff and anti-seborrheic dermatitis agents. In this context, current antifungal compounds, such as the azoles and their derivatives, have been demonstrated to induce the selection of resistant fungal strains and lose therapeutic efficacy, which might be restored by the concomitant use of alternative compounds, such as the fungal CA inhibitors.


Introduction
Carbon dioxide (CO 2 ) is ubiquitously generated and released into the atmosphere through cellular respiration and oxidative metabolism [1]. This gas byproduct is typically transported out of cells via passive diffusion. This transport may be aided by CO 2 channels, which are regulated in a CO 2 -dependent way [2,3]. Rather than a waste product, CO 2 has also the potential to act as a physiological stimulant for a variety of cellular signaling pathways that promote microorganism virulence and pathogenicity [1, 3,4]. For example, 2 of 12 Bacteria assist their colonization and infection at CO 2 concentration levels comparable to those found in the host, since bacterial CO 2 sensing mechanisms allow them to adapt and survive in such environments [1]. During their life cycle, fungi are exposed to various environmental variables, including fluctuations in CO 2 levels [1]. Dedicated CO 2 sensing machineries allow fungi to sense the amount of CO 2 present in mammalian tissues (about 5%), compared to the atmospheric levels (about 0.03%), which ultimately stimulates the fungal pathogenicity in the host [1,4].
In microorganisms, proteins involved in sensing CO 2 have been suggested as attractive targets of pharmaceuticals since they modulate cell differentiation and the further production of molecules essential for the pathogen [4,5]. In this context, it has been demonstrated that carbonic anhydrases (CAs, EC 4.2.1.1), catalyzing the reversible hydration of CO 2 to HCO 3 − and H + (CO 2 + H 2 O HCO 3 − + H + ) [6][7][8][9][10][11][12], are crucial enzymes in fungal CO 2 sensing since they produce bicarbonate, which is a promoter (through an adenylate cyclase (AC) intermediate enzyme) of fungal meiosis and sporulation [13]. Indeed, HCO 3 − produced in a CA-dependent manner activates AC and cyclic adenosine monophosphate (cAMP) production, which stimulates the development of filamentous structures (hyphae) needed for fungal virulence, adhesion, and the production of hydrolases, thus triggering cell death in the colonized host [4,5,14,15]. Up to date, eight CA gene families or classes have been identified and designated with Greek letters (α, β, γ, δ, ζ, η, θ, ι) [6][7][8][9][10]. In the fungal kingdom, the typical class is represented by β-CAs, which generally occur with at least one isoform. Conversely, α-CAs are rarely found in fungi [5,[16][17][18]. The catalytic action of fungal CAs triggers a cascade process, which allows the microorganism to adapt into the host, thrive therein, and contribute to its pathogenicity [4,14,15,18,19]. It is readily apparent that CAs can affect fungal growth and thus may represent a potential novel therapeutic target in fungal infections. This is corroborated by the studies of Supuran's group, who demonstrated that typical CA inhibitors (CAIs), namely primary sulfonamides, inhibit the growth of Malassezia globosa when CO 2 availability is limited (i.e., fungus-infected skin surface) [20]. In this context, we have focalized our interest on another such fungus, M. pachydermatis, which has a significant relevance in veterinary and medical sciences, as it has been associated with otitis externa and seborrheic dermatitis in dogs, cats, and wild animals, as well as with fungemia in hospitalized and immunocompromised patients [21][22][23][24]. When the skin microenvironment or the host's defenses are compromised, this opportunistic commensal has the potential to become a disease-causing pathogen [21,22]. In this context, we determined that the genome of the M. pachydermatis contains a single gene encoding a β-CA (acronym MpaCA) that is closely related to β-CAs previously identified by our groups in two other Malassezia species, namely M. globosa and M. restricta, which are responsible for dandruff and seborrheic dermatitis [20,[25][26][27][28][29][30][31][32][33][34].
Here, we have further investigated MpaCA, focusing on its inhibition profile with respect to a series of aromatic or heterocyclic sulfonamides, which are widely used as building blocks for obtaining potent and selective pharmacological agents. Besides, inhibition data on MpaCA have been compared with those of ortholog β-CA enzymes from M. globosa and M. restricta, namely MgCA and MreCA, respectively. Overall, this study tentatively proposes novel potential anti-dandruff and anti-seborrheic dermatitis compounds able to eradicate harmful fungi through the inhibition of CAs. This is potentially relevant since clinically used antifungal drugs, such as azoles and their derivatives, as result of their widespread diffusion, have determined the selection of resistant fungal strains.

Biochemical Characterization of MpaCA
Recombinant M. pachydermatis CA (MpaCA) was overexpressed in E. coli as a fusion protein with a non-natural tail containing six histidines at its molecular N-terminus, and purified by affinity chromatography. The purified enzyme was then subjected to SDS-PAGE and protonography to certify the corresponding molecular mass and ability to elicit a hydratase activity. As shown in Figure 1, SDS-PAGE demonstrated that recombinant MpCA was recovered in the soluble fraction of the bacterial extract in response (after 3 h) to isopropyl-β-D-thiogalactopyranoside (IPTG) induction. Recombinant MpCA showed an experimental molecular mass of about 30 kDa, in agreement with the expected theoretical one (31 kDa  Protonographic analysis (Figure 2) was used to determine whether purified binant MpaCA could catalyze the CO2 hydration reaction (Figures 1 and 2). Two ogous CAs, namely MgCA and MreCA, which are encoded from the genome of M. and M. restricta, respectively, were used as positive controls. As expected, the p gram in all cases exhibited yellow bands migrating at a molecular mass of about ( Figure 2), thus demonstrating a CO2 hydratase activity for all fungal CAs, inc MpaCA.  Protonographic analysis (Figure 2) was used to determine whether purified recombinant MpaCA could catalyze the CO 2 hydration reaction (Figures 1 and 2). Two homologous CAs, namely MgCA and MreCA, which are encoded from the genome of M. globosa and M. restricta, respectively, were used as positive controls. As expected, the protonogram in all cases exhibited yellow bands migrating at a molecular mass of about 30 kDa (Figure 2), thus demonstrating a CO 2 hydratase activity for all fungal CAs, including MpaCA. Protonographic analysis ( Figure 2) was used to determine whether purified binant MpaCA could catalyze the CO2 hydration reaction (Figures 1 and 2). Two ogous CAs, namely MgCA and MreCA, which are encoded from the genome of M. and M. restricta, respectively, were used as positive controls. As expected, the pr gram in all cases exhibited yellow bands migrating at a molecular mass of about ( Figure 2), thus demonstrating a CO2 hydratase activity for all fungal CAs, inc MpaCA.  With the aid of the stopped-flow technique, we further demonstrated that MpaCA exhibits an appreciable CO 2 hydrase activity, with a k cat value of 3.8 × 10 5 s −1 and k cat /K M value of 9.7 × 10 6 M −1 s − . MgCA and MreCA showed a catalytic activity very similar to that of MpaCA. In particular, MreCA showed a k cat value = 1.06 × 10 6 s −1 and k cat /K M value = 1.07 x 10 8 M −1 s −1 [25], while MgCA exhibited a k cat value of 9.2 × 10 5 s −1 and k cat /K M value of 8.3 × 10 7 M −1 s −1 [27][28][29][30][31][32][33].

Inhibition Profile of MpaCA with Sulfonamides
Sulfonamide compounds represent a significant class of synthetic bacteriostatic antibiotics still used today to treat infections caused by bacteria and other microorganisms [35][36][37]. They are also known as sulfa drugs. These compounds are derived from sulfanilamide (compound A in Figure 3) and include synthetic derivatives with the general chemical structure B [38]. Worth mentioning is the fact that often the term sulfonamide is imprecisely referred to antibiotics bearing a sulfonamide moiety, and not all sulfonamides are antibiotics [39]. Sulfonamide antibiotics have two structural characteristics that distinguish them from nonantibiotic counterparts, namely a free amino group at N4 and a nitrogen-containing heterocyclic ring linked to N1 of the sulfonamide group (compounds B) ( Figure 3). Furthermore, the discovery that sulfanilamide A has CA inhibitory properties [40] led to the discovery that corresponding derivatives C act as effective enzyme inhibitors ( Figure 3) [41]. The above-mentioned structural features are essential in mediating allergic reactions to sulfonamide antibiotics [39]. A growing body of clinical evidence indicates no increased risk of reactions to nonantibiotic sulfonamides in patients with a history of allergy to sulfonamide antibiotics [39].

Inhibition Profile of MpaCA with Sulfonamides
Sulfonamide compounds represent a significant class of synthetic bacte antibiotics still used today to treat infections caused by bacteria and other microor [35][36][37]. They are also known as sulfa drugs. These compounds are deriv sulfanilamide (compound A in Figure 3) and include synthetic derivatives general chemical structure B [38]. Worth mentioning is the fact that often sulfonamide is imprecisely referred to antibiotics bearing a sulfonamide moiety all sulfonamides are antibiotics [39]. Sulfonamide antibiotics have two s characteristics that distinguish them from nonantibiotic counterparts, namel amino group at N4 and a nitrogen-containing heterocyclic ring linked to N sulfonamide group (compounds B) ( Figure 3). Furthermore, the discov sulfanilamide A has CA inhibitory properties [40] led to the discovery that corres derivatives C act as effective enzyme inhibitors ( Figure 3) [41]. The above-m structural features are essential in mediating allergic reactions to sulfonamide an [39]. A growing body of clinical evidence indicates no increased risk of rea nonantibiotic sulfonamides in patients with a history of allergy to sulfonamide an [39]. Among nonantibiotic sulfonamides, primary sulfonamides (R'-SO2-NH2) sho most promising results due to their Zn(II) ion-binding properties; thus, they have increased attention due to their capability to specifically inhibit CAs [42]. In fact, t a complex in the enzyme active site with a tetrahedral geometry that is center catalytic Zn (II) ion, with the N atom of the sulfonamide moiety coordinate bivalent metal [4,[36][37][38]43].
In order to investigate the inhibition profile of MpaCA and to compare res that of other enzyme homologues, the interaction of 41 main sulfonamide sulfamate with the enzyme from M. pachydermatis was investigated in vitro. The m structure of these compounds is shown in Figure 4. The derivatives 1-24 and A are either simple aromatic or heterocyclic sulfonamides, and are frequently building blocks to create novel potent and selective pharmaceuticals [43,44]. T AAZ-EPA (see Table 1 for their identification) involves classical CA inhibitors (CA in clinics for managing and treating glaucoma, idiopathic intracranial hype altitude sickness, congestive heart failure, epilepsy, and other diseases [4,[36][37][38]4  Among nonantibiotic sulfonamides, primary sulfonamides (R'-SO 2 -NH 2 ) showed the most promising results due to their Zn(II) ion-binding properties; thus, they have received increased attention due to their capability to specifically inhibit CAs [42]. In fact, they form a complex in the enzyme active site with a tetrahedral geometry that is centered at the catalytic Zn (II) ion, with the N atom of the sulfonamide moiety coordinated to the bivalent metal [4,[36][37][38]43].

2.
Many compounds of the series 1-24 and AAZ-EPA (1, 2 Figure 6B,C provide a graphical representation of these findings, showing sulfonamide inhibitors with 1 µM < K I s < 10 µM for these fungal enzymes.

3.
As mentioned above, many of the chemicals reported in Table 2 were weak inhibitors of MreCA (Ki > 10 µM) and were already demonstrated to be effective and moderate inhibitors of the human isoenzyme II (hCA II) [26], MpaCA and MgCA, respectively. As highlighted above, MreCA showed an inhibition pattern markedly different from those of the other two homologous enzymes MpaCA and MgCA (Table 2). 3. As mentioned above, many of the chemicals reported in Table 2 were weak inhibitors of MreCA (Ki > 10 μM) and were already demonstrated to be effective and moderate inhibitors of the human isoenzyme II (hCA II) [26], MpaCA and MgCA, respectively. As highlighted above, MreCA showed an inhibition pattern markedly different from those of the other two homologous enzymes MpaCA and MgCA (Table 2).

MpaCA Production: Synthetic Gene, Cloning, Heterologous Expression, and Purification
The synthetic MpaCA gene was designed in our labs and produced by Life Technologies (Invitrogen, Carlsbad, CA, USA). Briefly, the MpaCA gene contained NdeI and XhoI restrictions sites at the 5′-and 3′-ends, respectively; it was ligated into the expression vector pET100/D-TOPO (Invitrogen, Carlsbad, CA, USA) to form the expression vector pET100D-Topo/MpaCA, containing a nucleotide sequence encoding for a polypeptide with additional six histidines before the insertion point, for facilitating the purification of the resulting recombinant protein. To overexpress MpaCA, competent E. coli BL21 (DE3)pLysS (Agilent, Santa Clara, CA, USA) cells were transformed with pET100D-Topo/MpaCA, growing them in 1 L of LB broth at 37 °C. Isopropyl-β-D-thiogalactopyranoside (IPTG) was added to a final concentration of 1 mM, and 0.5 mM ZnSO4 was added after incubation for 30 min for uptake in the expressed protein. The incubation period continued for an additional 3 h at 37 °C. To verify the overexpression of MpaCA, the resulting bacterial suspension was tested and analyzed on 12% T SDS-PAGE, according to Laemmli [46]. At 3 h post-induction, the cellular extract was prepared by sonication at 4 °C. Following centrifugation, the supernatant containing the overrepresented MpaCA was purified using a HIS-Select HF Nickel Affinity Gel  [34]. The protein concentration of the obtained active fractions was determined with a Bio-Rad protein assay based on the Bradford method [47]. The enzyme resulted at least 95% pure. About 1.0 mg of final recombinant enzyme was obtained from 1 L of bacterial culture.

Enzyme Protonography
For protonography, SDS-PAGE was performed as described by De Luca et al. [48]. Samples were mixed in a loading buffer without 2-mercaptoethanol, and they were not boiled to avoid protein denaturation. After electrophoresis, the gel was subject to protonography to detect the hydratase activity [48].  [34]. The protein concentration of the obtained active fractions was determined with a Bio-Rad protein assay based on the Bradford method [47]. The enzyme resulted at least 95% pure. About 1.0 mg of final recombinant enzyme was obtained from 1 L of bacterial culture.

Enzyme Protonography
For protonography, SDS-PAGE was performed as described by De Luca et al. [48]. Samples were mixed in a loading buffer without 2-mercaptoethanol, and they were not boiled to avoid protein denaturation. After electrophoresis, the gel was subject to protonography to detect the hydratase activity [48].

Enzyme Assays
An applied photophysics stopped-flow instrument was used for assaying the CAcatalyzed CO 2 hydration activity [49]. Phenol red (at a concentration of 0.2 mM) was used as an indicator in a buffer containing 20 mM Tris (pH 8.3), 20 mM NaClO 4 (for maintaining a constant ionic strength), measuring the absorbance maximum of 557 nm, and following the initial rate of the CA-catalyzed CO 2 hydration reactions for a period of 10-100 s. The CO 2 concentrations values ranged from 1.7 to 17 mM during the determination of the kinetic parameters.

Inhibition Assays
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 performed with the buffer test. Inhibitor and enzyme solutions were preincubated together for 15 min at room temperature prior to the assay, in order to allow the formation of the E-I complex or the eventual active site-mediated hydrolysis of the inhibitor. The inhibition constants, which represent the mean from at least three different determinations, were obtained by the non-linear least-squares methods using PRISM 6 and the Cheng-Prusoff equation, as reported earlier [50]. MgCA and MreCA were recombinant enzymes obtained inhouse. All salts/small molecules were of the highest purity available from Sigma-Aldrich (Milan, Italy).

Conclusions
Fungal MpaCA was generated as a soluble recombinant protein using E. coli cells as the host. SDS-PAGE, protonography, and the stopped-flow experiments showed that MpaCA has a molecular mass of about 30 kDa and an excellent hydratase activity, converting the CO 2 to bicarbonate and protons with a k cat value of 3.8 × 10 5 s −1 . By using the simple aromatic/heterocyclic compounds 1-24 and the therapeutically used drugs AAZ-EPA, the MpaCA sulfonamide inhibition profile was determined. Among the compounds belonging to both series, only 9, 14, 15, 19, 20, 21, 23, AAZ, BRZ, BZA, TMP, CLX, VLX, HCT, and EPA inhibited MpaCA with K I values < 1.0 µM, highlighting these compounds as promising compounds to be further tested for future veterinary and medical applications. The comparative analysis of the sulfonamide inhibition profiles of MpaCA, MreCA, and MgCA highlighted that MpaCA exhibits an inhibitory pattern similar to MgCA, but which is radically different from that of its homolog MreCAs. Considering the sulfonamide inhibition pattern of the human isoforms I and II (hCAI and hCA II) previously determined by our group [26], the above-mentioned fungal enzymes showed significant inhibitory differences with those of the human counterparts.
The differences in the inhibitory effect of the sulfonamides on the three fungal enzymes can be explained considering the structural properties of each biocatalyst here studied. Sulfonamides form an enzyme-inhibitor complex with tetrahedral geometry centered at the Zn(II) ion also involving the N atom of the sulfonamide moiety. An extended network of hydrogen bonds involving amino acids of the enzyme also contributes to the inhibitor molecule anchoring to the metal ion. Besides, an interaction occurs between the aromatic/heterocycle portions of the inhibitor and the hydrophilic and hydrophobic residues present in the catalytic pocket of the enzyme [4,[36][37][38]43,51]. Thus, it is reasonable to speculate that various residues present in the catalytic pocket of the different Malassezia enzymes may be responsible for the observed differences in the calculated K I values measured for the 42 compounds described in this study. Unfortunately, none of the three fungal enzymes were crystallized and, accordingly, no structural data were available for rationalizing the enzyme's behavior versus the investigated sulfonamides. However, these findings are encouraging because they show that, even though these CAs are very similar, there is a good chance for synthesizing inhibitors that can specifically inhibit CAs from the various fungi reported in this study as well as the human isozymes.