Novel 6- and 7-Substituted Coumarins with Inhibitory Action against Lipoxygenase and Tumor-Associated Carbonic Anhydrase IX

A series of carboxamide derivatives of 6- and 7-substituted coumarins have been prepared by an original procedure starting from the corresponding 6- or 7-hydroxycoumarins which were alkylated with ethyl iodoacetate, and the obtained ester was converted to the corresponding carboxylic acids which were thereafter reacted with a series of aromatic/aliphatic/heterocyclic amines leading to the desired amides. The new derivatives were investigated as inhibitors of two enzymes, human carbonic anhydrases (hCAs) and soy bean lipoxygenase (LOX). Compounds 4a and 4b were potent LOX inhibitors, whereas many effective hCA IX inhibitors (KIs in the range of 30.2–30.5 nM) were detected in this study. Two compounds, 4b and 5b, showed the phenomenon of dual inhibition. Furthermore, these coumarins did not significantly inhibit the widespread cytosolic isoforms hCA I and II, whereas they were weak hCA IV inhibitors, making them hCA IX-selective inhibitors. As hCA IX and LOX are validated antitumor targets, these results are promising for the investigation of novel drug targets involved in tumorigenesis.


Introduction
Vertebrates, including humans, encode for a multitude of metalloenzymes belonging to the carbonic anhydrase (CA, EC 4.2.1.1) family of proteins [1][2][3][4]. Although seven CA genetic families are known to date (α-, β-, γ-, δ-, ζ-, ηand θ-CAs) [2,5], only α-CAs are present in humans, but as 15 different isoforms, 12 of which are catalytically active and involved in a multitude of physiologic functions [3][4][5][6][7][8][9]. By catalyzing the reversible hydration of CO 2 to bicarbonate, with the release of a hydronium ion, in humans CAs are involved in pH regulation, biosynthetic reactions, electrolyte secretion, excretion, tumorigenesis, etc. [3,4,[6][7][8][9]. CA inhibitors (CAIs) are in pharmacological/clinical use for decades for the treatment of glaucoma [6,7], for the imaging and treatment of hypoxic tumors [3,4,8,9], as anti-obesity agents [10], or as diuretics [11]. Recently these pharmacological agents were validated for the management of neuropathic pain [12], but the sulfonamides, which are the main class of CAIs [11][12][13] possess a rather large number of side effects, as they indiscriminately inhibit all catalytically active CA isoforms, and not only the ones targeted for a specific application [1-3, [13][14][15][16][17]. Thus, alternative classes of CAIs to the sulfonamides and their isosteres were explored in the last The two carboxylic acids 3a and 3b were converted to the corresponding amides by reaction with aromatic, aliphatic and heterocyclic primary amines, as shown in Scheme 2, by using carbodiimide chemistry. The nature of the various amines was chosen in such a way as to generate the widest possible chemical diversity (Scheme 2). All compounds were extensively characterized by spectral and other physico-chemical procedures which proved their structure (see Experimental part for details).
As seen from the data in Table 1, like other coumarins investigated by our group these derivatives also do not inhibit the cytosolic isoforms hCA I and II up to 10 µM concentration of inhibitor in the assay system. hCA IV was also poorly inhibited, with most compounds being inactive whereas few of them showed activity in the high nanomolar range (e.g., 8a, 8b and 11a, with KIs in the range of 350.4-848.3 nM). Several other coumarins, incuding 2a, 3a and 7b, were micromolar hCA IV inhibitors, with KIs in the range of 2.65-8.48 µM. Thus, the 4-fluoroanilides of both 2-((2-oxo-2Hchromen-7-yl)oxy)acetic acid as well as its 6-isomer led to the best inhibitors of this isoform.
hCA IX on the other hand was effectively inhibited by most new coumarins reported here, except for 12a and 12b which were not hCA IX inhibitors up to 10 µM (Table 1). These compounds incorporate the morpholine-ethylamide moiety which is obviously inappropriate for obtaining effective CAIs in that position of the coumarin ring and with this type of substitution pattern. The The two carboxylic acids 3a and 3b were converted to the corresponding amides by reaction with aromatic, aliphatic and heterocyclic primary amines, as shown in Scheme 2, by using carbodiimide chemistry. The nature of the various amines was chosen in such a way as to generate the widest possible chemical diversity (Scheme 2). All compounds were extensively characterized by spectral and other physico-chemical procedures which proved their structure (see Experimental part for details).
As seen from the data in Table 1, like other coumarins investigated by our group these derivatives also do not inhibit the cytosolic isoforms hCA I and II up to 10 µM concentration of inhibitor in the assay system. hCA IV was also poorly inhibited, with most compounds being inactive whereas few of them showed activity in the high nanomolar range (e.g., 8a, 8b and 11a, with KIs in the range of 350.4-848.3 nM). Several other coumarins, incuding 2a, 3a and 7b, were micromolar hCA IV inhibitors, with KIs in the range of 2.65-8.48 µM. Thus, the 4-fluoroanilides of both 2-((2-oxo-2Hchromen-7-yl)oxy)acetic acid as well as its 6-isomer led to the best inhibitors of this isoform.
hCA IX on the other hand was effectively inhibited by most new coumarins reported here, except for 12a and 12b which were not hCA IX inhibitors up to 10 µM (Table 1). These compounds incorporate the morpholine-ethylamide moiety which is obviously inappropriate for obtaining The two carboxylic acids 3a and 3b were converted to the corresponding amides by reaction with aromatic, aliphatic and heterocyclic primary amines, as shown in Scheme 2, by using carbodiimide chemistry. The nature of the various amines was chosen in such a way as to generate the widest possible chemical diversity (Scheme 2). All compounds were extensively characterized by spectral and other physico-chemical procedures which proved their structure (see Experimental part for details).
As seen from the data in Table 1, like other coumarins investigated by our group these derivatives also do not inhibit the cytosolic isoforms hCA I and II up to 10 µM concentration of inhibitor in the assay system. hCA IV was also poorly inhibited, with most compounds being inactive whereas few of them showed activity in the high nanomolar range (e.g., 8a, 8b and 11a, with K I s in the range of 350.4-848.3 nM). Several other coumarins, incuding 2a, 3a and 7b, were micromolar hCA IV inhibitors, with K I s in the range of 2.65-8.48 µM. Thus, the 4-fluoroanilides of both 2-((2-oxo-2H-chromen-7-yl)oxy)acetic acid as well as its 6-isomer led to the best inhibitors of this isoform.
hCA IX on the other hand was effectively inhibited by most new coumarins reported here, except for 12a and 12b which were not hCA IX inhibitors up to 10 µM (Table 1). These compounds incorporate the morpholine-ethylamide moiety which is obviously inappropriate for obtaining effective CAIs in that position of the coumarin ring and with this type of substitution pattern. The remaining compounds showed an interesting hCA IX inhibitory patterns, with several compounds being quite effective inhibitors, with K I s in the range of 30.2-30.5 nM, similar to AAZ (K I of 25 nM). These compounds, 4b and 5b, are the phenethylamide and benzylamide derivatives of 2-((2-oxo-2H-chromen-6-yl)oxy)acetic acid 3b and they are much more effective CA IX inhibitors compared to the corresponding 6-isomers 4a and 5a (Table 1). However, this was not always the case, as for other pairs of isoforms, the 7-isomer was a better hCA IX inhibitor compared to the corresponding 6-isomer (e.g., 2a, which is a better inhibitor than 2b; 11a, a much more effective CA IX inhibitor compared to its isomer 11b, etc.). Many other coumarins were slightly less effective hCA IX inhibitors, with K I s in the range of 83.7-290.2 nM. They include derivatives 2a, 3a, 3b, 4a, 5a, 7a, 7b, 8a, 8b, 11a, 13b (Table 1). It is thus obvious that apart from the position in the coumarin ring where the substituent is appended, the most important factor influencing hCA IX inhibition is the nature of the moiety present on the amide part of the functionality. Indeed, the effective hCA IX inhibitors incorporate amides obtained from phenethylamine, benzylamine, aniline and substituted anilines. The only heterocyclic derivative leading to effective inhibitors was 4-pyridylmethylamine and piperidin-1-yl-ethylamine. The remaining amides (2b, 6a, 6b, 9a, 9b, 10a, 10b, 11b, 13a) were micromolar hCA IX inhibitors, with K I s in the range of 1.96-2.73 µM (Table 1).
An important feature of many coumarins reported here is that they are highly selective hCA IX versus hCA I/II/IV inhibitors, and in many cases also very effective in inhibiting the tumor-associated isoform hCA IX without inhibition of the widespread cytosolic/membrane-bound isoforms I; II and IV. For example 4b and 5b are equipotent to acetazolamide as hCA IX inhibitors but do not inhibit at all hCA I, II and IV, whereas AAZ inhibits these three isoforms significantly (Table 1).
In vitro inhibition of soybean lipoxygenase (LOX) has also been investigated with the new coumarins reported here (Table 2). Eicosanoids are oxygenated metabolites of arachidonic acid with a broad implication in a diversity of diseases among which are included the pathogenesis of neutrophil-mediated inflammatory diseases with a marked relation to the severity of cardiovascular diseases, asthma and cancer [36].  [38]. e nt, not tested (IC 50 values not found due to the fact that it may be >100 µM).
In this context, we evaluated the synthesized compounds of Table 2 for their ability to inhibit soybean LOX by the UV absorbance based enzyme assay [34] using compounds samples with concentrations from 0.1-100 µM. Most of the LOX inhibitors are antioxidants or free radical scavengers. LOXs contain a non-heme iron per molecule in the enzyme active site as high-spin Fe 2+ in the native state and the high spin Fe 3+ in the activated state [35]. Some studies suggest a relationship between LOX inhibition and the ability of the inhibitors to reduce Fe 3+ at the active site to the catalytically inactive Fe 2+ , whereas several LOX inhibitors are excellent ligands for Fe 3+ [35]. Nordihydroguaiaretic acid (NDGA), a known inhibitor of soybean LOX, has been used as a reference compound (IC 50 0.45 µM/93% at 100 µM) and as a positive control in our experiments [35]. We determined the IC 50 inhibition values for compounds 1a, 3a-b, 4a-b, 6a, 7a, 9b, 10a, 11a, 12a-b, 13a. We did not succeed to evaluate the IC 50 values for the rest of the compounds, since they were not active LOX inhibitors at 100 µM (11-46%). The most potent % inhibition at 100 µM is shown by compound 4a (4a > 4b > 10a~11a > 12b~6a).
Perusal of the IC 50 's inhibition values (Table 2) shows that the most potent, and equipotent, inhibitors are compounds 4a and 4b (10 µM) followed by 10a (15 µM) and 12b (16.5 µM). It is interesting to note that attachment on the coumarin ring, e.g., in the 6-/7-for compounds 4a and 4b, does not seem to play any role. Replacement of phenyl (4a) by a 2-pyridyl group (10a) or by a morpholinyl group (12a) leads to a reduction of the inhibitory activity, which is highly significant for 12a (42.5 µM). The presence of a 2-pyridyl group in compound 10b significantly decreased activity (by 40%) compared to 4b. In a similar manner, the presence of a 4-pyridyl group (11a) resulted in significant loss of inhibitory activity (27 µM) compared to compound 10a. The replacement by a morpholinyl group (12b) does not induce a considerable loss in activity.
The length of the chain between the aromatic ring and the NHCO-group [(CH 2 ) n ], influenced the biological response, since compound 4a (10 µM) with n = 2, is more potent compared to 7a (100 µM) in which n = 0 and 5a (45%) in which n = 1. The same is seen for 7b and 5b. The F-substitution allows an improved inhibitory activity compared to the unsubstituted compound: for example 6a has an IC 50 of 47 µM, whereas 5a only presents 45% at a concentration of 100 µM ( Table 2). As concerns the acids 3a and 3b they appear to present some inhibitory activities ( Table 2). Although lipophilicity is referred to as an important physicochemical property for LOX inhibitors [35], herein the theoretically calculated log P values did not always support this observation. The most potent compounds 4a and 4b showed the third higher lipophilicity values (2.60) in this series (Table 2). Furthermore, compounds with comparable lipophilicities showed in many cases striking different LOX inhibitory activities ( Table 2).

General Information
All biochemical reagents were of analytical grade and purchased from commercial sources. Soybean lipoxygenase, sodium linoleate, and NDGA were obtained from Sigma Chemical, Co. (St. Louis, MO, USA).

Chemistry
3.2.1. General Procedure for the Synthesis of Compounds 3a-b [39] A mixture of 7-hydroxycoumarin (1a) or 6-hydroxycoumarin (1b) (1 eq.) and potassium carbonate (3 eq.) was dissolved in dry DMF (5 mL) and the mixture was stirred at room temperature for 15 min. Then, ethyl 2-iodoacetate (a, 1.5 eq.) was added dropwise to the mixture under nitrogen atmosphere and heated to 100 • C for 30 min. After completion of the reaction (TLC monitoring) the mixture was cooled to room temperature and quenched with water and 1M aqueous HCl solution. The precipitated products 2a-b were collected by filtration and washed with water, and used without further purification. Perusal of the IC50's inhibition values (Table 2) shows that the most potent, and equipotent, inhibitors are compounds 4a and 4b (10 µΜ) followed by 10a (15 µΜ) and 12b (16.5 µΜ). It is interesting to note that attachment on the coumarin ring, e.g., in the 6-/7-for compounds 4a and 4b, does not seem to play any role. Replacement of phenyl (4a) by a 2-pyridyl group (10a) or by a morpholinyl group (12a) leads to a reduction of the inhibitory activity, which is highly significant for 12a (42.5 µΜ). The presence of a 2-pyridyl group in compound 10b significantly decreased activity (by 40%) compared to 4b. In a similar manner, the presence of a 4-pyridyl group (11a) resulted in significant loss of inhibitory activity (27 µΜ) compared to compound 10a. The replacement by a morpholinyl group (12b) does not induce a considerable loss in activity.
The length of the chain between the aromatic ring and the NHCO-group [(CH2)n], influenced the biological response, since compound 4a (10 µΜ) with n = 2, is more potent compared to 7a (100 µΜ) in which n = 0 and 5a (45%) in which n = 1. The same is seen for 7b and 5b. The F-substitution allows an improved inhibitory activity compared to the unsubstituted compound: for example 6a has an IC50 of 47 µM, whereas 5a only presents 45% at a concentration of 100 µΜ ( Table 2). As concerns the acids 3a and 3b they appear to present some inhibitory activities ( Table 2). Although lipophilicity is referred to as an important physicochemical property for LOX inhibitors [35], herein the theoretically calculated log P values did not always support this observation. The most potent compounds 4a and 4b showed the third higher lipophilicity values (2.60) in this series (Table 2). Furthermore, compounds with comparable lipophilicities showed in many cases striking different LOX inhibitory activities ( Table 2).

General Information
All biochemical reagents were of analytical grade and purchased from commercial sources. Soybean lipoxygenase, sodium linoleate, and NDGA were obtained from Sigma Chemical, Co. (St. Louis, MO, USA).

Chemistry
3.2.1. General Procedure for the Synthesis of Compounds 3a-b [39] A mixture of 7-hydroxycoumarin (1a) or 6-hydroxycoumarin (1b) (1 eq.) and potassium carbonate (3 eq.) was dissolved in dry DMF (5 mL) and the mixture was stirred at room temperature for 15 min. Then, ethyl 2-iodoacetate (a, 1.5 eq.) was added dropwise to the mixture under nitrogen atmosphere and heated to 100 °C for 30 min. After completion of the reaction (TLC monitoring) the mixture was cooled to room temperature and quenched with water and 1M aqueous HCl solution. The precipitated products 2a-b were collected by filtration and washed with water, and used without further purification. The crude products 2a or 2b (2.7 mmol) were dissolved in an aqueous solution of 5% NaOH (5 mL) in ethanol (15 mL) and the mixture was stirred at room temperature for 5 min. The residue quenched with water and acidified with aqueous 6 M solution HCl. The precipitated white solid was filtered off and subsequently washed with cool water and DCM to give compounds 3a-b, respectively.

Soybean Lipoxygenase Inhibition Studies
A DMSO solution of the tested compound was incubated with sodium linoleate (0.1 mM) and 0.2 mL of soybean LOX solution (1/9 × 10 −4 w/v in saline) in buffer pH 9 (tris) and at room temperature. The conversion of sodium linoleate to 13-hydroperoxylinoleic acid was recorded at 234 nm and compared with the standard inhibitor NDGA (IC 50 = 0.45 µM). The results are given in Table 2 expressed as IC 50 values or % inhibition at 100 µM [34].

CA Inhibition Assay
An Applied Photophysics (Oxford, UK) stopped-flow instrument has been used for assaying the CA catalysed CO 2 hydration activity [36]. Phenol red (at a concentration of 0.2 mM) has been used as 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-catalyzed 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-deionized water and dilutions up to 0.01 nM were done thereafter with distilled-deionized water. Inhibitor and enzyme solutions were preincubated together for 6 h at 4 • C prior to assay, in order to allow for the formation of the E-I complex and for the active site mediated hydrolysis of the inhibitor [18,19]. Data reported in Table 1 show the inhibition after 6 h incubation, which led to the completion of the in situ hydrolysis of the coumarin and formation of the hydroxycinnamic acid [18,19]. The inhibition constants were obtained by non-linear least-squares methods using PRISM 3, as reported earlier [42][43][44][45][46][47][48][49][50] and represent the mean from at least three different determinations. The four CA isoforms were recombinant proteins obtained as reported earlier in our laboratory [42][43][44][45][46][47][48][49][50].

Conclusions
We report here a series of carboxamide derivatives of 6-and 7-substituted coumarins. They have been prepared by an original procedure starting from the corresponding 6-or 7-hydroxycoumarins which were alkylated with ethyl iodoacetate, then the obtained ester was converted to the corresponding carboxylic acid which was thereafter reacted with a series of aromatic/aliphatic/heterocyclic amines leading to the desired amides. The present study shows that these compounds represent a promising class of multi-targeting derivatives which can interact with several biological targets, in this case, lipoxygenase and carbonic anhydrases. Compounds 4a and 4b were potent LOX inhibitors, whereas many effective hCA IX inhibitors (K I s in the range of 30.2-30.5 nM) were detected in this study. Two compounds 4b and 5b showed the phenomenon of dual inhibition. Furthermore, these coumarins did not significantly inhibit the widespread cytosolic isoforms hCA I and II, whereas they were weak hCA IV inhibitors, making them hCA IX-selective inhibitors. As hCA IX and LOX are validated antitumor targets, these results are promising for the investigation of novel drug targets involved in tumorigenesis.