Novel Sulfonamide Analogs of Sivelestat as Potent Human Neutrophil Elastase Inhibitors

Human neutrophil elastase (HNE) is involved in a number of essential physiological processes and has been identified as a potential therapeutic target for treating acute and chronic lung injury. Nevertheless, only one drug, Sivelestat, has been approved for clinical use and just in Japan and the Republic of Korea. Thus, there is an urgent need for the development of low-molecular-weight synthetic HNE inhibitors, and we have developed a wide variety of HNE inhibitors with various chemical scaffolds. We hypothesized that substitution of the active fragment of Sivelestat into these HNE inhibitor scaffolds could modulate their inhibitory activity, potentially resulting in higher efficacy and/or improved chemical stability. Here, we report the synthesis, biological evaluation, and molecular modeling studies of novel compounds substituted with the 4-(sulfamoyl)phenyl pivalate fragment necessary for Sivelestat activity. Many of these compounds were potent HNE inhibitors with activity in the nanomolar range (IC50 = 19–30 nM for compounds 3a, 3b, 3f, 3g, and 9a), confirming that the 4-(sulfamoyl)phenyl pivalate fragment could substitute for the N-CO group at position 1 and offer a different point of attack for Ser195. Results of molecular docking of the these pivaloyl-containing compounds into the HNE binding site supported the mechanism of inhibitory activity involving a nucleophilic attack of Ser195 from the catalytic triad onto the pivaloyl carbonyl group. Furthermore, some compounds (e.g., 3a and 3f) had a relatively good stability in aqueous buffer (t1/2 > 9 h). Thus, this novel approach led to the identification of a number of potent HNE inhibitors that could be used as leads for the further development of new therapeutics.


INTRODUCTION
Human neutrophil elastase (HNE) is a multifunctional enzyme involved in the killing of pathogens, regulation of inflammatory processes, and tissue homeostasis. HNE is also involved in chemotaxis and the release of inflammatory mediators through the cleavage of adhesion molecules in cellular junctions (Pham, 2006;Korkmaz et al., 2010). Under physiological conditions, HNE is regulated by a group of endogenous protease inhibitors called "serpins" (Silverman et al., 2001;Heutinck et al., 2010). However, when this balance fails in favor of the proteolytic enzyme, excessive HNE activity can cause tissue damage. Among the pathologies associated with increased HNE activity are acute respiratory distress syndrome (ARDS) and acute lung injury (ALI) (Polverino et al., 2017), chronic obstructive pulmonary disease (Pandey et al., 2017), cystic fibrosis (Kelly et al., 2008;Dittrich et al., 2018), and other disorders with an inflammatory component, such as rheumatoid arthritis (Hilbert et al., 2002;Di Cesare Mannelli et al., 2016), atherosclerosis (Henriksen and Sallenave, 2008;Wen et al., 2018), psoriasis, and dermatitis (Marto et al., 2018). HNE has also been implicated in the progression of non-small cell lung cancer (Lerman and Hammes, 2017;. The development of new and selective HNE inhibitors is of great interest, both in the academic and industrial world, for therapeutic development. Despite the discovery of numerous classes of potent HNE inhibitors (Groutas et al., 2011;Von Nussbaum and Li, 2015;Crocetti et al., 2019), only two drugs are currently available on the market: the peptide inhibitor Prolastin R (purified α1-antitrypsin, from Alpha Therapeutic Corp, 2003) and the small molecule Sivelestat. Sivelestat (ONO-5046, Figure 1) developed by Ono Pharmaceutical and marketed in Japan and Korea exclusively as a sodium salt in the injectable formulation Elaspol R 100 for the treatment of ARDS and ALI associated with systemic inflammatory response syndrome and in pediatric surgery to alleviate the inflammatory response induced by cardiopulmonary bypass (Kawabata et al., 1991;Fujii et al., 2010;Inoue et al., 2013). Sivelestat acts as acyl-enzyme inhibitor (IC 50 = 44 nM) (Ohbayashi, 2005), as demonstrated by electrospray ionization mass spectrometry (ESI), which highlighted the formation of a HNE-Sivelestat complex after 0-10 min incubation of the drug with HNE (Figure 2) (Nakayama et al., 2002). We have been working for some time on the design and synthesis of HNE inhibitors with different nitrogen monocyclic and bicyclic scaffolds (Crocetti et al., 2011(Crocetti et al., , 2013Giovannoni et al., 2016Giovannoni et al., , 2018Giovannoni et al., , 2019Vergelli et al., 2017). Figure 3 illustrates the compounds already investigated (A-F) and the range of activity for each series. Many of these compounds have very potent HNE inhibitory activity (IC 50 values in the low nanomolar range), and kinetic experiments have characterized our compounds as competitive and pseudo-irreversible acyl-enzyme inhibitors. Furthermore, molecular modeling studies on compound interactions with HNE indicated that Ser195-OH of the catalytic triad attacks the carbonyl group of the N-CO group at position 1 in the bicyclic compounds (A-E) and the CO at position 5 in the monocyclic derivatives F (Vergelli et al., 2017;Giovannoni et al., 2018).
Since the first patent in which Sivelestat was reported (Ono Pharmaceutical Co. and LTD., 1989), followed by the publication of Kawabata and coworkers (Kawabata et al., 1991), few articles have appeared in the literature describing Sivelestat analogs and (or) derivatives. The patent filed in 2002 (Macias, 2002) reports the introduction of substituents on the benzoyl moiety of Sivelestat, whereas Hwang et al. (2015) performed structural modification of Sivelestat by replacing the glycine moiety with an oxime group and obtained a slightly more potent compound. Recently, we synthesized the isoxazolone derivative G   (Figure 3) containing the 4-(sulfonyl)phenyl pivalate fragment bound to the aniline nitrogen of Sivelestat. This compound exhibited high HNE inhibitory activity (IC 50 = 59 nM) and excellent chemical stability in aqueous buffer (data FIGURE 1 | Sivelestat (ONO-5046) claimed by Ono Pharmaceutical (Kawabata et al., 1991). not shown), which was improved over that of our previously published HNE inhibitors.
In the present studies, we expanded our strategy to evaluate how addition of the 4-(sulfonyl)phenyl pivalate fragment could modulate inhibitory activity of other HNE inhibitor scaffolds to evaluate if an additive effect would occur or whether these modifications could improve chemical stability of these new compounds. Therefore, we selected a number of compounds from our HNE inhibitor library belonging to the series A-E shown in Figure 3 and inserted the active 4-(sulfamoyl)phenyl pivalate fragment of Sivelestat into our scaffolds while leaving/maintaining the best substituents at positions 3 and 5/6. In most of these new bicyclic compounds, the carbonyl group of the N-CO function at position 1 was replaced with the Sivelestat pivalate fragment. In a few compounds, the N-CO function of our original compounds was left unchanged, and the active fragment of Sivelestat was inserted into a different position in order to produce two possible points of interaction with Ser195.

Chemistry
All melting points were determined on a Büchi apparatus (New Castle, DE) and are uncorrected. Extracts were dried over Na 2 SO 4 , and the solvents were removed under vacuum. Merck F-254 commercial plates (Merck, Durham, NC) were used for analytical TLC to follow the course of reactions. Silica gel 60 (Merck 70-230 mesh, Merck, Durham, NC) was used for column chromatography. 1 H NMR and 13 C NMR spectra were recorded on an Avance 400 instrument (Bruker Biospin Version 002 with SGU, Bruker Inc., Billerica, MA). Chemical shifts (δ) are reported in ppm to the nearest 0.01 ppm using the solvent as an internal standard. Coupling constants (J values) are given in Hz and were calculated using TopSpin 1.3 software (Nicolet Instrument Corp., Madison, WI) and are rounded to the nearest 0.1 Hz. Mass spectra (m/z) were recorded on an ESI-TOF mass spectrometer (Brucker Micro TOF, Bruker Inc., Billerica, MA), and reported mass values are within the error limits of ±5 ppm mass units. Microanalyses indicated by the symbols of the elements or functions were performed with a PerkinElmer 260 elemental analyzer (PerkinElmer, Inc., Waltham, MA) for C, H, and N, and the results were within ±0.4% of the theoretical values, unless otherwise stated. Reagents and starting material were commercially available.

Experimental Section
General Procedure for Compounds (3a-c) To a suspension of the substrate 1a-c (1a: Shahidul et al., 2006;1b: DeGraw and Goodman, 1964;1c: Yuen et al., 2013) (0.43 mmol) in 10 mL of anhydrous THF, 0.86 mmol of sodium hydride (60% dispersion in mineral oil) was added while stirring. After 30 min, 0.56 mmol of 4-(chlorosulfonyl)phenyl pivalate 2 (Hwang et al., 2015) was added, and the mixture was stirred at room temperature overnight. After evaporation of the solvent in vacuo, the residue was diluted with ice-cold water (10 mL), neutralized with HCl 6N, and extracted with ethyl acetate (3 × 15 mL). The organic phase was dried over sodium sulfate, and the solvent was evaporated in vacuo to obtain the final compounds 3a-c, which were purified by flash column chromatography using cyclohexane/ethyl acetate 4:1 (for 3a) and 6:1 (for 3c) or hexane/ethyl acetate 5:2 for 3b as eluents.

General Procedure for Compounds (7c-e)
Compounds 7c-e were obtained starting from intermediates 6c-e (6c: Purandare et al., 2014; 6e: Crocetti et al., 2013), respectively, following the same procedure described for 3g. The solvent was concentrated in vacuo to obtain the final compounds 7c-e, which were first purified by flash column chromatography using cyclohexane/ethyl acetate 1:1 as eluent and then by crystallization from ethanol.

Pharmacology
Compounds were dissolved in 100% DMSO at 5 mM stock concentrations. The final concentration of DMSO in the reactions was 1%, and this level of DMSO had no effect on enzyme activity. HNE inhibition assays were performed in black flat-bottom 96-well microtiter plates. Briefly, a solution containing 200 mM Tris-HCl, pH 7.5, 0.01% bovine serum albumin, 0.05% Tween R -20, and 20 mU/mL of HNE (Calbiochem) was added to wells containing different concentrations of each compound. The reaction was initiated by addition of 25 µM elastase substrate (N-methylsuccinyl-Ala-Ala-Pro-Val-7-amino-4-methylcoumarin, Calbiochem) in a final reaction volume of 100 µL/well. Kinetic measurements were obtained every 30 s for 10 min at 25 • C using a Fluoroskan Ascent FL fluorescence microplate reader (Thermo Electron, MA) with excitation and emission wavelengths of 355 and 460 nm, respectively. For all compounds tested, the concentration of inhibitor that caused 50% inhibition of the enzymatic reaction (IC 50 ) was calculated by plotting % inhibition vs. logarithm of inhibitor concentration (at least six points). The data are presented as the mean values of at least three independent experiments with relative standard deviations of <15%.

Molecular Modeling
Structures of Sivelestat (in the form of a carboxylate anion) and compounds 3a, 7b, 7d, and 7e, were created using ChemOffice 2016 software, pre-optimized with the MM2 force field and saved in Tripos MOL2 format. The ligand structures were then imported into the Molegro Virtual Docker 6.0 program (MVD). The structure of HNE complexed with 1-{3-methyl-2-[4-(morpholine-4carbonyl)-benzoylamino]-butyryl}-pyrrolidine-2-carboxylic acid (3,3,4,4,4-pentafluoro-1-isopropyl-2-oxo-butyl)-amide (SEI) ligand was downloaded from the Protein Data Bank (PDB code 1B0F) and also imported into MVD. The co-crystallized water molecules were removed from the 1B0F structure on importing. A search space for docking was defined in the HNE binding site as a sphere of radius 12 Å positioned at the geometric center SCHEME 2 | a Reagents and conditions. (A) for 5a,c,d: CH 3 I, Na 2 CO 3 , anhydrous CH 3 CN, 80 C, 6 h; for 5b: m-Toluoyl chloride, NaH (60% dispersion in mineral oil), anhydrous THF, r.t., 24 h; for 5e: m-Toluoyl chloride, Et 3 N, anhydrous CH 2 Cl 2 , 0 • C, 2 h then r.t., 2 h (B) H 2 (Parr), Pd/C, EtOH 96%, 30 min (for 6c) and 2 h (for 6a,b,d,e); (C) for 7a,b: Et 3 N, anhydrous CH 2 Cl 2 , 0 • C, 2 h then r.t., 2 h; for 7c-e: dry pyridine, r.t., 4 h. of gravity of the SEI ligand, and the investigated compounds were docked into the binding site. MolDock score functions (Thomsen and Christensen, 2006) were applied with a 0.3 Å grid resolution. Ligand flexibility was accounted for with respect to torsion angles auto-detected in MVD. Structure of the protein was considered rigid. The "Internal HBond" and "sp2-sp2 torsions" options were activated in the "Ligand evaluation" menu of the MVD Docking Wizard. Three hundred docking runs were performed for each molecule. Our attempts to enhance number of docking runs up to 600 did not lead to better scored docking poses. The option "Return multiple poses for each run" was enabled, and the post-processing options "Energy minimization" and "Optimize H-bonds" were applied after docking. Similar poses were clustered at a RMSD threshold of 1 Å.

Chemistry
All final compounds were synthesized as reported in Schemes 1-3, and the structures were confirmed on the basis of analytical and spectral data. The 4-(chlorosulfonyl)phenyl pivalate fragment 2 representing the active portion of Sivelestat that was incorporated into all new compounds was synthesized as reported previously (Hwang et al., 2015).

Biological Evaluation
All new products were evaluated for HNE inhibitory activity, and the results are reported in Tables 1, 2 in comparison with Sivelestat. Table 1 presents the results of compounds lacking the 1-N-CO function responsible for activity in our original compounds (Figure 3). These include 3a-g, 3i-l, 4, and 9a, b, which contain the sulfamoyl fragment of Sivelestat at N-1 of the bicyclic nucleus, and compounds 7a, 7c, and 7d, which have the (sulfonyl)phenyl pivalate chain at position five of the nucleus and a methyl group at N-1 (N-methyl derivatives). Many of the new derivatives exhibited very potent HNE inhibitory activity, with IC 50 values between 15 and 78 nM for most compounds, which is comparable to or better that that of Sivelestat (IC 50 = 44 nM). The most potent compounds were the indazole 3f and the cinnoline 9a, which had IC 50 values of 15 and 19 nM, respectively. These results clearly demonstrated that replacement of the NCO function at N-1 with the active fragment of Sivelestat did not affect HNE inhibitory activity (Table 1). Previously, we found that indoles of type B (Figure 3), which were designed as 2-deaza analogs of highly active N-benzoylindazole compounds, were inactive or low activity HNE inhibitors due to the lack of the nitrogen at position two, which forms an important interaction with Gly193 of the catalytic site . Here, we report that introduction of the (sulfonyl)phenyl pivalate chain at N-1 of the indole nucleus into these compounds (i.e., 3a-c) resulted in very potent HNE inhibitors, with IC 50 values of 30, 25, and 49 nM, respectively, suggesting that inclusion of the active fragment of Sivelestat leads to a different interaction of the indole scaffold with HNE. Finally, the N-1 methyl derivatives 7a, c, d bearing the (sulfonyl)phenyl pivalate chain at position five of the nucleus retained some inhibitory activity, although they were not as active as the other compounds described above.
The results reported in the Table 1 indicate that in practice, the pivaloyl fragment of Sivelestat can "replace" the role of the N-CO group at position 1 and offer a different point of attack for Ser195. It is also clear that the selected scaffolds with adequate substitutions are appropriate carriers for the Sivelestat pharmacophore. On the other hand, compounds maintaining the NCO function at position 1 (compounds 7b and 7e) or CO at position five (compound 11) (Table 2) as the point of attack for Ser195, and simultaneously bearing the active pivaloyl fragment of Sivelestat, only exhibited moderate HNE inhibitory activity (IC 50 = 0.29-5.2 µM), with the exception of the previously published compound G (IC 50 = 59 nM), which has activity comparable to its analogs lacking the Sivelestat fragment . However, these data also clearly indicate that this strategy does not produce the expected additive effect, probably due to the increased hindrance of the molecules.
A set of the most potent HNE inhibitors, as well as low activity compound 7b, were evaluated for their chemical stability in aqueous buffer. Spontaneous hydrolysis rates of the inhibitors  -g, 3i-l, 4, 7a,c,d, and 9a,b were measured in phosphate buffer at pH 7.3 and 25 • C. As shown in Table 3, compounds 3a, 3c, 3f, and 7d had a relatively good stability (t 1/2 > 9 h) with high HNE inhibitory activity (IC 50 < 100 nM). The relatively high enzymatic stability of HNE allowed us to evaluate reversibility of the enzyme inhibition over time.
As an example, Figure 4 shows kinetic curves monitoring substrate cleavage catalyzed by HNE over a 10-h period in the presence of selected sulfonamide derivatives (5 µM) and compared to Sivelestat. Persistence of selected HNE inhibitors over an extended period of time (16 h) was also evaluated and showed that the most effective HNE inhibitors over time were 3c, 3d, 3i, 3l, and 9a (Table 4). Inhibitory activities of these compounds were comparable or better than Sivelestat in this assay.

Molecular Modeling
Molecular docking studies of some HNE inhibitors, including Sivelestat and triterpenes, were previously made (Feng et al., 2012(Feng et al., , 2013 based on 1B0F structure from PDB. Hence, we also used this structure in our docking calculations. The MVD program was validated on HNE by confirming the ability of the program to reproduce the position of the co-crystallized SEI ligand contained in the 1B0F structure taken from the Protein Data Bank (Cregge et al., 1998). Independent docking of SEI  . into the HNE binding site was performed, and comparison of the resulting pose with the experimental ligand location demonstrated that the MVD program accurately reproduced the location of the SEI ligand in the HNE binding site (RMSD of nonhydrogen atom positions between the two structures is 1.24 Å) ( Figure S1, see Supplementary Material).
One of the goals of our molecular modeling study consisted in clarifying the possibilities for Nakayama's mechanism for inhibitory action of pivaloyl-containing compounds (Nakayama  (Feng et al., 2012). In this docking pose, Sivelestat forms H-bonds between the oxygen atom of the sulfonamide group and Ser195 and Gly193, as well as an H-bond between the carboxyl group and Ser214. In addition, the Sivelestat pose is near to hydrophobic residues Leu99B, Phe192, His57, Val216, Cys191, and Phe41. All of these features of the ligand location are consistent with the docking results previously reported for Sivelestat (Feng et al., 2012). According to our data and the results obtained by Feng et al. (2012) on the binding of Sivelestat to HNE, Ser195 forms an H-bond with the sulfonamide oxygen atom, hence Ser195 is far from the carbonyl carbon atom of the pivaloyl group, which is a potential reaction center in the reported mechanism (Nakayama et al., 2002). Thus, for the docking pose of Sivelestat, we calculated the distance O(Ser195)····C=O(pivaloyl) to be 9.4 Å. In this regard, the experimental results of Nakayama and co-authors (Nakayama et al., 2002) can be explained by the presence of other possibilities for binding of Sivelestat to HNE using conformations other than the optimal docking pose that we obtained. Indeed, we found another pose for Sivelestat in which the pivaloyl group is located close to the elastase catalytic triad, forming H-bonds with Ser195 and Asp194. In addition, H-bonds were formed between the amide nitrogen atom of the ligand and Val216 and between the carboxyl group and Gly218 and  Gly219 ( Figure 5B). This alternative docking pose of Sivelestat is favorable for nucleophilic attack of the Ser195 oxygen atom on the carbonyl carbon of the pivaloyl group, resulting the distance O(Ser195)····C=O(pivaloyl) of 3.04 Å, which is consistent with the reported mechanism (Nakayama et al., 2002). It should be noted that the MolDock score for this pose is only 1.4 units higher than that for the optimal pose shown in Figure 5A.
Molecule 7d forms H-bonds with Ser195 via participation of both oxygen atoms of the ethoxycarbonyl group. In addition, the carbonyl oxygen of the ethoxycarbonyl substituent forms Hbonds with Gly193 and Asp194, while the sulfonamide nitrogen atom is H-bonded to Ser214 (Figure 6). The docking pose of molecule 7d in the HNE binding site is characterized by a distance of 5.40 Å between the oxygen atom of Ser195 and the carbonyl carbon of the pivaloyl group. This does not exclude the possibility of nucleophilic addition of Ser195 to the C=O group according to the reported mechanism (Nakayama et al., 2002), because as a result of thermal movements of the ligand and receptor, the carbonyl oxygen may be available for nucleophilic attack.   According to our docking results, compound 3a forms strong H-bonds with Ser195 (two possible H-bonds), Asp194, and Gly193 with participation of the carbonyl oxygen atom of the pivaloyl group ( Figure 7A). With this position in the binding site, 3a is quite accessible for attack by Ser195 at the carbonyl carbon atom, i.e., according to the direction of metabolism proposed by Nakayama et al. (2002). The distance O(Ser195)····C=O(pivaloyl) in this case is 3.05 Å, which is comparable to the corresponding distance for Sivelestat (see above). Compound 7e in its docking pose forms several Hbonds with HNE, one of them being a bond between the pyrazole nitrogen and Ser195 ( Figure 7B). Additionally, the amide oxygen atom forms a strong Hbond with Gly193, while the sulfonamide nitrogen H-bonds with Val216. The distance O(Ser195)····C=O(pivaloyl) for the pose of compound 7e is 5.96 Å. Thus, compound 7e is anchored significantly within the binding site. Compound 7b differs from 7e by the presence of a CH group in the 5-membered ring. This reduces opportunities for the formation of H-bonds involving participation of the heterocycle. Accordingly, 7b is slightly shifted away from Ser195 and neighboring residues ( Figure 8A). Molecule 7b is H-bonded to Val216 with participation of the sulfonamide nitrogen atom and also with Asp194 and Gly193 via participation of oxygen atom in the m-methylbenzoyl substituent. In Figure 8B, the poses of 7e and 7b are shown together. Visible amino acid residues lie within 5 Å of the co-crystallized ligand SEI. The distance O(Ser195)····C=O(pivaloyl) for the pose of 7b in the binding site is 6.74 Å, which is the largest value of the investigated compounds (Table 5). Perhaps, due to the remoteness of the pivaloyl group from the key residue Ser195 of the catalytic triad, compound 7b is the least active among the sulfonamides investigated.
Docking scores for the obtained poses of compounds 3a, 7b, 7d, 7e, and Sivelestat anion are equal to −127. 45, −108.94, −114.37, −133.75, and −125.04 MolDock units, respectively. It should be noted that these values did not show any significant correlation with IC 50 indicating that specific protein-ligand interactions rather than total complementarity play role in appearing the inhibitory activity. Indeed, according to our results, the specific mechanism of HNE inhibition proposed by Nakayama et al. (2002), which includes the Ser195 attack on the carbonyl carbon of the pivaloyl group, can be easily achieved for compounds 3a and Sivelestat ( Table 5). The IC50 values obtained for compounds 7b, 7d, and 7e are also in agreement with the geometric characteristics of their docking poses ( Table 5).

CONCLUSIONS
Previously, we demonstrated that the isoxazolone derivative G had high HNE inhibitory activity  and excellent chemical stability in aqueous buffer (data not shown). Since this compound contains the 4-(sulfamoyl)phenyl pivalate fragment that is necessary for Sivelestat activity, we hypothesized that substitution of this active fragment onto other HNE inhibitor scaffolds could modulate their inhibitory activity, potentially resulting in higher efficacy and/or improved chemical stability of these new compounds. Based on this novel approach, we synthesized and characterized a number of new derivatives and demonstrated that the 4-(sulfamoyl)phenyl pivalate fragment could "replace" the role of the N-CO group at position 1 and offer a different point of attack for Ser195. Indeed, results of molecular docking of the these pivaloyl-containing compounds into the HNE binding site supported the mechanism of inhibitory activity involving a nucleophilic attack of Ser195 from the catalytic triad onto the carbonyl group of the pivaloyl moiety. Clearly, the selected scaffolds with adequate substituents can be appropriate carriers for the Sivelestat pharmacophore since many of the new compounds had high inhibitory activity in the nanomolar range, with the most potent inhibitors being 3a, 3b, 3f, 3g, and 9a (IC 50 = 19-30 nM). However, these data also indicate that this strategy does not produce an expected additive effect of inhibitor potency, probably due to increased steric hindrance of the pivaloyl substituent.

DATA AVAILABILITY STATEMENT
All datasets presented in this study are included in the article/Supplementary Material.