Extending the Inhibition Profiles of Coumarin-Based Compounds Against Human Carbonic Anhydrases: Synthesis, Biological, and In Silico Evaluation

Carbonic anhydrases (CAs, EC 4.2.1.1) catalyze the fundamental reaction of CO2 hydration in all living organisms and are actively involved in the regulation of a plethora of pathological and physiological conditions. A set of new coumarin/ dihydrocoumarin derivatives was here synthesized, characterized, and tested as human CA inhibitors. Their inhibitory activity was evaluated against the cytosolic human isoforms hCA I and II and the transmembrane hCA IX and hCA XII. Two compounds showed potent inhibitory activity against hCA IX, being more active or equipotent with the reference drug acetazolamide. Computational procedures were used to investigate the binding mode of this class of compounds within the active site of hCA IX and XII that are validated as anti-tumor targets.


Drug Design and Chemistry
Five CA inhibition mechanisms have been identified to date, but complete structural binding data are only available for four of them [33]. These are: (i) zinc binders; (ii) inhibitors anchoring to the zinc bound water/hydroxide ion; (iii) inhibitors occluding the entrance to the active site; (iv) inhibitors binding out of the active site; and (v) compounds with unknown inhibition mechanism.
The occlusion of the binding site entrance as a CA inhibition mechanism was evidenced for the first time with a natural product coumarin, isolated from the Australian plant Leionema ellipticum and, therefore, for the simple coumarin [27]. Successively, the antiepileptic drug lacosamide, 5and 6-membered lactones and thiolactones or quinolinones were observed to possess significant CA inhibitory properties probably sharing a common mechanism of action [33]. In detail, X-ray crystallography studies were conducted, which showed that coumarins acts as prodrug -at leastin human CAs being hydrolyzed to the active species 2-hydroxycinnamic acids by the CA esterase activity [33]. The binding of the coumarin active species occurs in regions of the CA active site that most significantly differ among the various human isoforms known to date, furnishing the explanation for the high isoform-selective inhibitory profile shown by such a class of compounds [33].
To extend the structure-activity relationship of coumarins with hCAs, we report here the synthesis of a new set of coumarin-based derivatives to be screened for the inhibition of the ubiquitous hCA I and II and the tumor-associated hCA IX and XII.

CA Inhibition
Coumarin-based compounds 1-12 were evaluated for their inhibition against the cytosolic CA I and II and the membrane-bound IX and XII by using a stopped-flow CO 2 hydrase assay method. The clinically used acetazolamide (AAZ) was used as standard drug in the kinetic evaluation. The following SAR (this is not SAR but results of evaluation) can be worked out from the data reported in Table 1. According to previously reported CA inhibition profiles of coumarin-based derivatives, none of 1-12 inhibited the ubiquitous and cytosolic hCA I and II below 10 µM. In contrast, the main tumor-associated isoform hCA IX was efficiently inhibited by derivatives 1-12 in a low to medium nanomolar range with inhibition constants (K I s) spanning between 9.4 and 243.1 nM. Not surprisingly, coumarin 11, possessing the CA inhibitory scaffold less sterically hindered among the tested compounds, exhibited the best CA inhibitory action of the study with a K I of 9.4 nM. For the same reason, coumarin 10 also acts as an equipotent CAI with the standard drug AAZ (K I of 25.7 nM). Quite unexpectedly because of the steric hindrance produced by the oxime moiety in position 8, coumarin 1 showed a CA IX inhibitory efficacy (K I of 49.3 nM) which is 2-to 5-fold greater than the remaining compounds. The latter's show rather comparable K I s which range from 85.6 to 243.1 nM.
A peculiar inhibition profile was instead measured for the other tumor-associated isoform hCA XII. In fact, unpredictably, most derivatives do not inhibit hCA XII up to 10 µM, whereas the subset composed by 1, 4, 10-12 shows a medium nanomolar inhibition against the isozyme with K I s in the range 432.1-603.8 nM. This might be due to the significant steric hindrance bore by most derivatives on the coumarin or coumarin-like scaffold.

Docking Studies
To rationalize the CA inhibitory profiles of Table 1, docking studies were undertaken with all assayed compounds in the active site of hCA IX and XII. The docking scores of all compounds and their hydrolyzed species in complex with CA IX and XII are shown in Table 2. Docking studies with hCA IX and XII showed that the free energy of binding of the compounds hydrolyzed species (H) is lower than that of the compounds themselves. As a result, we can indicate the hydrolyzed form of the compounds as the responsible for the CA inhibition. For hCA IX, the best docking score was predicted for H11 which was also kinetically reported as the best isozyme inhibitor ( Table 1). The docked poses of compound 11 and its hydrolyzed product (H11) are reported in Figures 2 and 3. H11 formed four hydrogen bonds with residues Gln224, His226, His228 and Thr332. The phenyl ring showed hydrophobic interactions with residues Tyr143, Asn198, Ser201, Val253, Leu331, and Thr333, while the fused rings interact hydrophobically with residues Leu223, Val262, Leu266, Leu272, and Pro335. In contrast, compound 11 forms only two hydrogen bonds within the enzyme binding site with residues Gln203 and Thr332 ( Figure 3). This might explain the better scores reported by H11 in comparison to 11.
Some repulsions taking place in the adduct of compound 4 within the isozyme binding site (red-colored residues Gly144, Gly145 in Figure 4) might account for its weaker hCA IX inhibition with respect to the other derivatives. These repulsive forces between the ligand and the active site residues likely do not allow the ligand to adopt the proper conformation into the cavity.   For hCA XII, the in vitro and docking results were in accordance as all compound showed similar docking scores. Nonetheless, H1 showed a somewhat higher score compared to other derivatives ( Table 2). According to the 2D ligand interaction figure (Figure 5), H1 forms 2 H-bonding interactions with residues Asn64 and Gln89 of the target protein. It also forms hydrophobic interactions with residues Tyr6, His66, Ser67, His93, Val141, Tyr198, Val119, Leu197, Leu139 Thr199, and Pro200.

Chemistry
Solvents, unless otherwise specified, were of analytical reagent grade or of the highest quality commercially available. Synthetic starting materials, reagents and solvents were purchased from InterBioscreen (Chernogolovka, Russia, https://www.ibscreen.com/) and Aldrich Chemie (Steinheimm, Germany). Melting points ( • C) were determined with a Boetius apparatus (Dresden, Germany) without correction. 1 H-NMR spectra of the newly synthesized compounds in DMSO-d 6 solutions were recorded on a Bruker AC 300 instrument (Bruker, Karlsruhe, Germany) at 298 K. Chemical shift (δ) values for 1 H-NMR spectra are reported in parts per million (ppm) with the solvent resonance as the internal standard. The TLC analysis was performed with Merck Silica Gel 60 F254 precoated plates, and each of the synthesized compounds showed a single spot.

CA Inhibition
An Applied Photophysics stopped-flow instrument was used for assaying the CA-catalysed CO 2 hydration activity [36]. Phenol red (at a concentration of 0.2 mM) was used as an indicator, working at the absorbance maximum of 557 nm, with 20 mM Hepes (pH 7.5) as buffer, and 20 mM Na 2 SO 4 (for maintaining constant the ionic strength), following the initial rates of the CA-catalysed CO 2 hydration reaction for a period of 10-100 s. The CO 2 concentrations ranged from 1.7 to 17 mM for the determination of the kinetic parameters and inhibition constants. For each inhibitor, at least six traces of the initial 5-10% of the reaction have been used for determining the initial velocity. The uncatalyzed rates were determined in the same manner and subtracted from the total observed rates. Stock solutions of inhibitor (0.1 mM) were prepared in distilled-deionised water and dilutions up to 0.1 nM were done thereafter with the assay buffer. Inhibitor and enzyme solutions were preincubated together for 6 h at room temperature prior to assay, in order to allow for the formation of the E-I complex (coumarins act as pro-drug inhibitors). The inhibition constants were obtained by nonlinear least-squares methods using PRISM 3 and the Cheng-Prusoff equation, as reported earlier [39], and represent the mean from at least three different determinations. All CA isoforms were recombinant ones obtained in-house, as reported earlier [40][41][42].

Molecular Modeling
Docking calculations were carried out using the AutoDock 4.2 software. The free energy of binding (∆G) of both CA IX and XII complexes with the compounds was generated using this molecular docking program. The crystal structures of CA IX (PDB code 5DVX) and CA XII (PDB code 5MSA) were taken from the Protein Data Bank [43,44]. For the enzymes' preparation, polar hydrogens were added; Kollman United Atom charges and atomic salvation parameters were assigned. For ligands preparation, Gasteiger partial charges were added, non-polar hydrogen atoms were merged, and rotatable bonds were defined. The three-dimensional structures of all the compounds were assembled using Chem3Dultra 12.0 software. The grid size was set to 50 × 50 × 50 xyz points with grid spacing of 0.375 Å. The grid centers were calculated for CA IX (X = 5.741, Y = −15.751 and Z = 8.657) and for CA XII (X = 24.159, Y = 9.619 and Z = 21.1). All parameters used in docking were default. A primary blind docking was performed to find the favored binding sites of the ligand to the receptor. The Lamarckian genetic algorithm was applied for minimization using default parameters. The number of docking runs was 100. After docking, the 100 solutions were clustered into groups with RMS lower than 1.0 E. The discovery studio 2017 R2 silent (BIOVIA, San Diego, CA, USA) was used for the virtualization of the resulting poses and potential interactions.

Conclusions
We reported here the synthesis, characterization, and kinetic/in silico evaluation of a set of new coumarin/dihydrocoumarin derivatives as inhibitors of human CA. In detail, we investigated the compounds for the inhibition of the cytosolic human isoforms hCA I and II and the transmembrane, tumor-associated hCA IX and hCA XII that are validated target for anticancer intervention. Two compounds were identified which showed potent inhibitory activity against hCA IX with K I s of 9.4 and 25.7 nM, thus being more active or equipotent with the reference drug acetazolamide. A computational assessment was performed to gain insights on the binding mode of this class of compounds within the active site of hCA IX and XII. Docking studies with hCA IX and XII revealed that the free energy of binding of the hydrolyzed species (H) is lower than that of the compounds themselves indicating that probably, the hydrolyzed form is responsible for the CA inhibition. For hCA IX, the best docking score was predicted for H11, which was also kinetically reported as the best isozyme inhibitor.

Conflicts of Interest:
The authors declare no conflict of interest.