Dithiocarbamates strongly inhibit carbonic anhydrases and show antiglaucoma action in vivo

Abstract

A series of dithiocarbamates was prepared by reaction of primary/secondary amines with carbon disulfide in the presence of bases. These compounds were tested for the inhibition of 4 human (h) isoforms of the zinc enzyme carbonic anhydrase, CA (EC 4.2.1.1), hCA I, II, IX and XII, involved in pathologies such as glaucoma (CA II and XII) or cancer (CA IX). Several low nanomolar inhibitors targeting these CAs were detected. X-ray crystal structure of hCA II adduct with morpholine dithiocarbamate evidenced the inhibition mechanism of these compounds, which coordinate to the metal ion through a sulfur atom from the dithiocarbamate zinc-binding function. Some dithiocarbamates showed effective intraocular pressure lowering activity in an animal model of glucoma.

Introduction

Carbonic anhydrases (CAs, EC 4.2.1.1) are widespread zinc metalloenzymes found in higher vertebrates including humans.^1–3 16 isozymes have been characterized to date, many of which are involved in critical physiological processes. They catalyze the following reaction: CO₂+H₂O ↔H⁺+HCO₃^−.1–3 In humans, CAs are present in a large variety of tissues including the gastrointestinal tract, the reproductive tract, the nervous system, kidneys, lungs, skin and eyes.^2,3 The different isozymes are localized in different parts of the cell with CA I and CA II, important isozymes in normal cells, being localized in the cytosol. ^1–4

Many of the CA isozymes are important therapeutic targets with the potential to be inhibited to treat a range of disorders.^1–7 CA II plays a role in bicarbonate production in the eye and is therefore a target for therapy of eye disease such as glaucoma.^7–9 Indeed, CA inhibitors (CAIs) of the sulfonamide type such as dorzolamide DZA or brinzolamide BRZ are topically used antiglaucoma agents,^7–10 whereas the older drugs, such as acetazolamide AAZ or dichlorophenamide DCP show the same action through systemic administration, which however leads to a wide range of side effects due to inhibition of the enzyme from other organs than the target one, i.e., the eye.¹¹ CA XII, a transmembrane isoform with an extracellular active site, was shown to be overexpressed in glaucomatous patients eyes.¹²

As some solid tumors grow in cancer patients, hypoxic regions are formed, particularly in the interior of the tumor.¹³ The gene expression profile of a hypoxic cancer cell is different from that of other cancer cells in a normally-oxygenated environment, i.e., in normoxic conditions.^13–15 Under hypoxia, the distribution of CA isoforms is altered as compared with normoxic cells.^13,14 As a result, CA isozymes IX and XII are overexpressed in hypoxic tumor cells, in a variety of solid tumors.^13–15 Unlike other CAs, CA IX and CA XII are both extracellularly localized on hypoxic tumor cells. ^13–15 These enzymes play various roles in tumorigenesis, by regulating pH inside and outside the tumor cell,¹⁵ interfering with phosphorylation of various proteins¹⁶ or by playing a role in the cell-cell adhesion.^13–15 They therefore provide a target for cancer therapy because they are relatively specific to the hypoxic tumor cells and appear to be important in their survival and proliferation.¹⁵ Indeed, several antibodies targeting CA IX are in Phase III clinical development for the treatment of solid tumors (or for their imaging)¹⁷ whereas some small molecule inhibitors are also in advanced preclinical evaluation.^15,18,19

The classical carbonic anhydrase inhibitors (CAIs) are the sulfonamides and their isosteres (sulfamates, sulfamides, etc).^1–4 However, most of these compounds indiscriminately inhibit many of the 16 CA isoforms known to date in mammals.^1–3 Thus, efforts have been made to find different CAIs, from the sulfonamide, sulfamate and sulfamide ones. Indeed, recently the coumarins were discovered as mechanism-based inhibitors which act as prodrugs and bind in a very different mode compared to sulfonamides and their isosteres,²⁰ whereas some polyamines (such as spermine),²¹ as well as a range of phenols²² were also investigated and showed interesting such properties and novel mechanisms of inhibition. Among CAIs investigated to date there are also the inorganic anions, which coordinate to the zinc ion from the enzyme active site.^1,23,24 Indeed, trithiocarbonate (CS₃²⁻) an anion similar to carbonate, has recently been investigated and shown to constitute a “lead” for novel CAIs.²³ The X-ray crystal structure for the adduct of trithiocarbonate (CS₃²⁻), bound to hCA II has recently been reported (Fig. 1).²³ The inhibitor binds to the Zn²⁺ in the hCA II active site in a slightly distorted tetrahedral geometry of the metal ion, occupying a position similar to that observed in the case of hCA II-bicarbonate complex.²³ Trithiocarbonate was monocoordinated to the Zn(II) ion by means of one of the sulfur atoms. The same sulfur made a hydrogen bond to the OH of Thr199 whereas a second sulfur atom participated to another hydrogen bond with the NH group of the same amino acid residues, Thr199. This binding mode explains the low micromolar affinity of this inhibitor to many of the CA isoforms investigated to date.²³ Based on this binding mode of a millimolar inhibitor, trithiocarbonate (CS₃²⁻), we hypothesized that compounds incorporating this new zinc-binding function, CS₂⁻, may act as even stronger CAIs. Indeed, we have recently demonstrated in a preliminary communication that dithiocarbamates (DTCs), compounds possessing the general formula R¹R²N-CS₂⁻M⁺ act as highly efficient CAIs.²⁵ Here we report the first detailed study of the DTCs as a class of potent CAIs, with a mechanism of action different of that of the sulfonamides. Furthermore, we prove that some of these highly water soluble compounds possess excellent intraocular pressure (IOP) lowering properties in an animal model of glaucoma, making them interesting candidates for developing antiglaucoma drugs.

Fig 1.

Trithiocarbonate (CS₃²⁻), a recently investigated low micromolar CAI, binds to the Zn²⁺ in the hCA II active site in a slightly distorted tetrahedral geometry of the metal ion, occupying a position similar to that observed in the case of hCA II-bicarbonate complex. The protein zinc ligands (His94, 96 and 119) and Thr199 are shown (PDB code 3K7K).²⁴

Results and Discussion

Chemistry

DTCs are well known metal complexing agents and they also possess interesting biomedical and agricultural applications.^26–29 Although this class of compounds (and their metal complexes) started to be used as fungicides more than 50 years ago,²⁹ few studies investigated their interactions with metalloenzymes.²⁷ Apart studies of DTCs as inhibitors of tyrosinase, a copper enzyme,²⁷ only one such work investigated the inhibition of N,N-diethyl-DTC with bovine CA (bCA).²⁸ By using Co(II)-substituted CA, Morpurgo et al.²⁸ showed that the inhibitor does not extrude the metal ion from the enzyme active site (as it does with the copper ion from the tyrosinase active site)²⁷ and that it binds to it, probably in a trigonal-bipyramidal geometry of Co(II). However, no other DTC were subsequently investigated for their interaction with CAs till our group reported that trithiocarbonate and related compounds containing the new zinc binding group (ZBG) found in it, i.e., CS₂⁻, inhibit several CA isoforms in the low micromolar or submicromolar range.²³ Here we extend those findings, showing that a wide range of DTCs incorporating various aliphatic and/or aromatic moieties at the nitrogen atom, acts as low nanomolar and even subnanomolar CAIs.

We prepared a series of 27 DTCs, of types 1a-27a, by the classical reaction^27,30,31 of dithiocarbamoylation between primary/secondary amines 1-27, and CS₂, in the presence of a base, which most of the time was NaOH, KOH but in the case of more basic amines, the amine itself can act as base (Scheme 1). As shown in Table 1, a large variety of R¹ and R² moieties are present in DTCs reported here, i.e., hydrogen, alkyl, aryl, aralkyl, hetaryl and cyclic such moieties, which lead to the generation of a wide chemical diversity in these compounds. Presumably, this should be reflected also in varied physico-chemical and biological properties of these DTCs. Compounds 1a-27a were characterized by physico-chemical standard procedures (IR, ¹H- an ¹³C-NMR spectroscopy, MS) and were > 99 % pure, as determined by HPLC (see Experimental protocols for details)

Scheme 1.

Preparation of DTCs 1a-27a by reaction of amines 1-27 with carbon disulfide in the presence of bases.

Table 1.

CA I, II, IX and XII inhibition data with dithiocarbamates 1a-27a by a stopped-flow, CO₂ hydrase assay.³²

R1R2N-CSS−M+ 1a-27a
No*,#	R1	R2	KI (nM)#
			hCA I	hCA II	hCA IX	hCA XII
1a	H	Ph	4.8	4.5	4.2	4.3
2a	H	O[(CH2CH2)]2N	4.8	3.6	29.1	9.2
3a	H	MeN[(CH2CH2)]2N	33.5	33.0	22.1	17.5
4a	H	2-butyl	21.1	29.4	4.6	31.7
5a	H	O[(CH2CH2)]2N(CH2)2	31.8	36.3	4.5	4.2
6a	H	N[(CH2CH2)N]3	31.9	13.5	27.4	9.3
7a	H	PhCH2	4.1	0.70	19.2	11.5
8a	H	4-PyridylCH2	3.5	16.6	26.0	24.1
9a	H	[(CH2)5N]CH2CH2	4.5	20.3	3.6	20.5
10a	H	2-thiazolyl	3.9	4.6	12.6	22.0
11a	H	KOOCCH2	13.1	325	57.1	6.7
12a	H	imidazol-1-yl-(CH2)3	8.6	24.7	4.3	6.5
13a	Me	Me	699	6910	714	798
14a	Et	Et	790	3100	1413	1105
15a		(CH2)5	0.96	27.5	70.4	46.1
16a	iso-Bu	iso-Bu	0.97	0.95	4.5	0.99
17a	n-Pr	n-Pr	1838	55.5	53.8	7.0
18a	n-Bu	n-Bu	43.1	50.9	50.3	5.8
19a	n-Hex	n-Hex	48.0	51.3	27.4	16.1
20a	Et	n-Bu	157	27.8	25.9	7.5
21a	HOCH2CH2	HOCH2CH2	9.2	4.0	4.3	4.2
22a	Me	Ph	39.6	21.5	28.2	7.7
23a	Me	PhCH2	69.9	25.4	53.0	3.0
24a		O[(CH2CH2)]2	0.88	0.95	6.2	3.4
25a		NaS(S=C)N[(CH2CH2)]2	12.6	0.92	37.5	0.78
26a		(NC)(Ph)C(CH2CH2)2	48.4	40.8	757	169
27a		(S)-[CH2CH2CH2CH(COONa)]	2.5	17.3	4.1	4.0
AAZ		-	250	12	25	5.7

^*Compounds 1a, 8a and 10a were thriethylamonium salts; 2a-6a, 9a, 11a and 12a were potasium salts, and the remaining ones were sodium salts.

^# Mean from 3 different assays. Errors were within ± 5–10 % of the reported values (data not shown).

^# Inhibition data against hCA I, II and IX for 9 DTCs, i.e. compounds 13a, 14a, 16–18a, 20a, 22a, 23a and 26a, were reported in ref.²⁵

Carbonic anhydrase inhibition

Compounds 1a-27a were assayed³² for the inhibition of four physiologically relevant CA isoforms, hCA I, II, IX and XII. All of them are drug targets: hCA I, II and XII for ophthalmologic diseases, mainly glaucoma,^1,10 whereas CA IX and XII for antitumor drugs/tumor imaging agents.^1,15,17–19 Inhibition data with the sulfonamide, clinically used agent acetazolamide (AAZ) are also reported in Table 1, for comparison reasons.

The following structure-activity relationship (SAR) can be observed for the CA inhibition data with DTCs 1a-27a investigated here:

1. The cytosolic isoform hCA I was strongly inhibited by DTCs 1a-27a investigated here, with K_Is in the range of 0.88 – 1838 nM. It may be observed that irrespective of the nature of R², the DTCs prepared from primary amines (R¹ = H) 1a-12a were highly effective hCA I inhibitors, with inhibition constants in the low nanomolar range (3.5 – 33.5 nM). On the contrary, the compounds prepared from secondary amines showed a more varied biological activity. Thus, the simple dimethyl- and diethyl-DTCs 13a and 14a were weak hCA I inhibitors, with K_Is in the range of 699–790 nM. The same is true for the di-n-propyl derivative 17a (K_I of 1838 nM). However, the cyclic derivative 15a, which differs from 14a by the cyclic structure and an extra carbon atom present in its molecule, is a subnanomolar hCA I inhibitor, showing a dramatic increase of potency of 823 times compared to 14a. It is also interesting to compare the potencies of 16a and 18a, which incorporate iso-butyl and n-butyl moieties (and are isomers), and differ by a factor of 44, unexpectedly, in favor of the compound with a branched scaffold. Increasing the length of the aliphatic chains to C6, as in 19a leads to a slight loss of potency compared to the most active compounds in the aliphatic series, which are 15a and 16a. Furthermore, the presence of hydroxyethyl moieties (instead of the ethyl ones) as in 21a, lead to a steady increase of potency, 21a being 85.9 times a better hCA I inhibitor compared to 14a. The compounds incorporating one alkyl and one aryl moiety at the nitrogen atom of the DTC function, such as 22a and 23a were also effective hCA I inhibitors (K_Is of 39.6–69.9 nM), as were also the heterocyclic derivatives 24a-27a, some of which showed subnanomolar activity (the morpholine DTC 24a had a K_I of 0.88 nM and was the best DTC hCA I inhibitor and also the best hCA I inhibitor ever described, as far as we know). Thus, many of these chemotypes explored here show excellent hCA I inhibitory activity, which range from the subnanomolar to the micromolar. Furthemore, many of the DTCs are much more effective as hCA I inhibitors compared to the sulfonamide AAZ, which has a K_I of 250 nM against this isoform (Table 1).

2. The physiologically dominant cytosolic isoform hCA II also showed an interesting inhibition profile with DTCs 1a-27a. Thus, several primary DTCS (1a, 2a, 7a, and 10a) and several secondary ones (16a, 21a, 24a and 25a) were excellent hCA II inhibitors, with K_Is in the range of 0.70 – 4.6 nM, being more effective (even one order of magnitude) than the clinically used sulfonamide AAZ (Table 1). It may be observed that these compounds incorporate aromatic, arylakyl, hetaryl, alkyl and hydroxyalkyl moieties substituting the nitrogen atom from the DTC moiety. Another rather large group of derivatives, such as 3a-6a, 8a, 9a, 12a, 15a, 17a-20a, 22a, 23a, 26a and 27a, were slightly less effective hCA II inhibitors, but still possessed a high efficacy, with K_Is in the range of 13.5 - 55.5 nM. Again both primary and secondary DTCs are present in this subgroup. They incorporate various types of substituents, such as alkyl, aryl, aralkyl, and hetaryl ones. It is obvious that small structural changes in the DTC scaffold influence dramatically the biological activity. For example, for the aliphatic secondary DTCs, the isomeric pair 16a – 18a, which differ only by the nature of the aliphatic chain (iso-Bu moieties in the first compound and n-Bu in the second derivative) have K_Is which differ by a factor of 53.6. The length of the alkyl chain also strongly influence activity, with compounds possessing a medium chain (e.g., 15a-20a) being more effective than the ones with shorter chains, such as 13a and 14a, which are rather ineffective as hCA II inhibitors (K_Is of 3.1 – 6.9 μM). Also the glycine DTC 11a was a medium potency hCA II inhibitor, with a K_I of 325 nM.

3. The tumor-associated isoform hCA IX was highly inhibited by the DTCs investigated here, with K_Is in the range of 3.6 – 1413 nM. The simple aliphatic secondary DTCs 13a/14a and the bulky cyclic derivative 26a, were the least effective inhibitors (K_Is of 0.714–1.413 μM), and four other compounds (11a, 15a, 17a and 18a) were effective, medium potency inhibitors, with K_Is in the range of 50.3 – 70.4 nM. They incorporate the carboxyalkyl moiety present in glycine (11a), the five-membered aliphatic ring (from 15a) and 3- or 4-carbon atom n-alkyl chains (17a and 18a). All the remaining derivatives showed highly effective hCA IX inhibitory properties, with K_Is < 30 nM. Thus, a rather high structural diversity (aliphatic, aromatic, aralkyl, hetaryl moieties) present in primary/secondary DTCs lead to highly effective hCA IX inhibitors, with minor structural changes drastically affecting enzyme inhibition. Many DTCs were more effective hCA IX inhibitors compared to acetazolamide (Table 1).

4. A rather similar SAR as the one discussed above for hCA IX, was observed for the inhibition of the second transmembrane isoform, hCA XII, with DTCs 1a-27a. Thus, 13a/14a and 26a, were the least effective inhibitors (K_Is in the range of 169 – 1105 nM), whereas the remaining DTCs were highly effective hCA XII inhibitors, with K_Is in the range of 0.78 – 31.7 nM (Table 1). Among the best hCA XII inhibitors (subnanomolar inhibition constants) were the di-isobutyl-DTC 16a and the piparazine-bis-DTC 25a. Again the main conclusion is that a large number of substitution patterns, incorporating varied moieties, lead to highly effective hCA XII inhibitors.

5. The DTCs investigated here showed a rather promiscuous inhibitory activity against all four CA isoforms described here, although each of these CAs had a different inhibition profile with all these compounds. For example, 15a was a subnanomolar inhibitor of hCA I and inhibited the remaining 3 isoforms with K_Is in the range of 27.5–70.4 nM, having thus an acceptable selectivity ratio for the inhibition of hCA I over the remaining three CAs. 25a was a subnanomolar inhibitor of hCA II and XII, and inhibited hCA I and IX with higher K_Is, of 12.6–37.5 nM. Compound 23a showed a rather good selectivity for inhibiting hCA XII over hCA I, II and IX (Table 1). It should be also mentioned that being negatively charged, these compounds show membrane impermeability³³ which may be a favorable pharmacological property in vivo.

X-ray crystallography

In order to explain the potent CA inhibitory properties of the DTCs, which are a new class of CAIs, we resolved the X-ray crystal structure of hCA II in complex with the very potent inhibitor 24a (K_I of 0.95 nM). Compound 24a was well ordered and refined with an occupancy of 1.0 with B-factors that were comparable to the solvent within the active site (Table 2). The compounds is buried deep into the active site, displacing the catalytic zinc-bound solvent, such that one of the sulfur atoms coordinates directly to the zinc ion of the enzyme. The overall zinc coordination (3N from the coordinating histidine residues His94, 96 and 119, and 1S from the inhibitor ligand) can be described as a distorted tetrahedron (Fig. 2). The details of this tetrahedral geometry are provided in Table 3. The zinc bound sulfur also interacts with the O atom of Thr199 in a similar manner to that observed in the more classical clinically used sulfonamides and sulfamates CAIs.^34–36 Compound 24a possesses a puckered ring and binds with a slightly higher B-factor of 23.9 Ų, compared to the two other DTCs for which the X-ray structure in complex with hCA II were reported earlier,²⁵ i.e., 23a and 26a. For a structural comparison the three compounds were superposed onto each other (Fig. 3). For unbound hCA II the side chain conformation of His64 has been shown to be dependent upon the buffer pH, which effects the protonation state of the imidazole ring. It is widely believed that this side chain flips from an “in” to “out” conformation as part of the proton transfer mechanism in hCA II, hence two conformations of the residue are often observed in crystal structures.^34–36 His64 in the hCA II structure in complex with compound 26a (PDB ID 3P5L) and 24a (PDB ID 3P5A) has a dual conformtion. In the case of 26a, the terminal six-membered hydrophobic ring sites close to F131, V135 and P202 at the rim of the active site in a hydrophobic pocket of hCA II. Whereas, for 24a, the six-membered ring does not extend far enough out of the active site to either reach this hydrophobic pocket or close enough to the in-conformation of His64. Hence, compounds 24 and 26a are 5.6 and 5.0 Å respectively from His64 and therefore do not affect its conformation. Whereas in the hCA II – 23a structure (PDB ID 3P58) the six-membered planar ring forms a T-shaped π-stacking with the imidzole ring of His64 and stabilizes this amino acid in the “in” conformation.²⁵ In addition 24a is positioned 3.2 Å from Thr200 but does not form hydrogen bonds with either the protein main- or side-chain. However, the endocyclic oxygen atom in the tail ring of 24a was within hydrogen bond distance of 3.4 and 3.2 Å from water393 and water540, respectively (Fig. 2) which probably also contribute to its high affinity to hCA II.

Table 2.

Crystallographic data refinement and model quality statistics for thr hCA II-24a complex.

PDB accession number	3P5A
Data-collection statistics
Temperature (K)	100
Wavelength (Å)	1.5418
Space group	P21
Unit-cell parameters (Å,°)	a = 42.3; b = 41.2
	c = 72.1; β = 104.2
Total theoretical reflections	39632
Total measured reflections	39395
Resolution (Å)	50.0-1.5 (1.54-1.50)
a Rsym (%); I/σ(I)	6.4 (18.1); 17.5 (5.8)
Completeness; Redundancy	99.4 (95.7); 3.7 (3.4)
Final Model Statistics
b Rcryst (%); c Rfree(%)	0.148; 0.169
Residue Nos.	4–261
d No. of protein atoms	2078
No. of compound atoms	9
No. of H2O molecules	300
R.M.S.D. bond lengths (Å)	0.012
bond angles (°)	1.488
Ramachandran statistics (%)	89.4, 10.7, 0.0
Average B factors (Å2) Main-, side-chain, compound, solvent	13.6, 18.1, 23.9, 29.4

^aR_sym = Σ |I − <I>|/Σ <I>.

^bR_cryst = (Σ |Fo| − |Fc|/Σ |F_obs|) × 100.

^cR_free is calculated in same manner as R_cryst, except that it uses 5% of the reflection data omitted from refinement.

^dIncludes alternate conformations.

^*Values in parenthesis represent highest resolution bin.

Fig. 2.

Stick representation of hCA II active site with compound 24a (green) complexed within it. The active-site zinc is depicted as a blue sphere. Water molecules are depicted as red spheres. The electron density is represented by a Σ-weighted 2Fo − Fc Fourier map (gray mesh). Amino acids involved in binding of inhibitor are also shown. Figure made using PyMOL (DeLano Scientific).

Table 3.

Geometry of the Zn(II) ion and bound S atom (of the DTCs) or compounds 23a, 24a and 26a complexced within the hCA II active site (Angle defined as His-Zn-S).

	23a	24a	26a
Distance (Å)	2.3	2.3	2.3
Angle (°) with His94	111.3	107.3	108.8
Angle (°) with His96	114.2	112.2	112.9
Angle (°) with His119	119.4	128.3	126.1

Fig. 3.

View of compounds 23a (cyan), 24a (green), 26a (magenta), superposed in the active site of hCA II. Left and right panel depict overall and zoomed active site view, respectively. hCA II is depicted as a grey surface representation. The active-site zinc is depicted as a blue sphere. Figure made using PyMOL (DeLano Scientific).

Intraocular pressure lowering in hypertensive rabbits

Two of the new CAIs investigated here, compounds 24a and 25a, which show excellent hCA II and XII inhibitory properties (Table 1) were investigated in vivo, for their ability to lower intraocular pressure (IOP) in carbomer-induced glaucoma in rabbits.^10c Normal IOP in rabbits, like in humans is of around 15–20 mm Hg. In this model, the IOP is quite elevated, mimicking thus the pathologic situation observed in the human disease.¹⁰ Clinically used drugs such as dorzolamide, induce a maximal IOP lowering of 4–5 mm Hg, as reported recently by us.¹⁰ The two dithiocarbamates investigated here in detail were chosen both due to their excellent enzyme inhibitory activity in vitro and also because they show very good water solubility, being formulated at 2 % eye drop solution at neutral pH (due to their salt character, whereas DZA is formulated at pH 5.5, as hydrochloride salt, and induces eye irritation).¹⁰ In fact, water solubility of eye drugs is a significant problem,^8–10 with many classes of drugs achieving an acceptable solubility only as salts with strong acids, such as HCl, which leads to acidic pH values causing eye irritation (dorzolamide DZA is a well known such case).¹⁰

Rabbits were treated with 2 % solutions of DTCs 24a and 25a and their IOP was monitored for 48 h (Fig. 4a). The contralateral eye was treated with vehicle and was used as control (Fig. 4B). As observed from Figure 4, both compounds were effective in reducing elevated IOP time dependently, for a rather long period. The maximal effect (of –6–10 mm Hg) has been observed after 2 hours post administration, and it lasted for up to 4–8 hours, being almost double that reported for DZA (of 4–5 mm Hg, and lasting only for about 3–4 hours).¹⁰ DTC 24a was slightly more effective than 25a as an IOP lowering agent. The IOP in vehicle-treated eyes was rather constant during the entire duration of the experiments, varying between 37–39 mm Hg for animals treated with 25a and between 37–40 mm Hg for animals treated with 24a (the data of the Figures are the mean for 3 different animals and the error range is shown in Fig. 4).

Fig. 4.

A: IOP lowering versus time, of glaucomatous rabbits treated with one drop (50 μL) of a 2 % water solution of DTC 24a and 25a; B: IOP in the eyes treated with vehicle (IOP are the mean from 3 different animals). Error bars are shown in both figures, accounting for an average of 0.8 – 1.5 mm Hg.

Conclusions

We report here that DTCs represent a novel class of highly effective CAIs. DTCs are easy to prepare from simple starting materials, they can incorporate a very high chemical diversity, and act as inhibitors of several physiologically relevant CA isoforms, with potencies from the subnanomolar to the micromolar. SAR for the inhibition of isoforms hCA I, II, IX and XII were straightforward and slightly different, with small modifications in the backbone of the compound leading to dramatic changes of biological activity. The inhibition mechanism of the DTCs was also explained, by resolving the X-ray crystal structure for hCA II complexed with a heterocyclic DTC. The CS₂⁻ moiety present in DTCs represents a new zinc-binding function. It is directly coordinated to the Zn(II) ion from the enzyme active site and also participates in an interaction (hydrogen bond) with the OH moiety of Thr199, an amino acid essential for the binding of many classes of CAIs (and of the substrates). The organic scaffold of the DTC is deeply buried within the enzyme active site and also participates in favorable interactions with it, which leads to a high stabilization of the enzyme-inhibitor adduct. Some of the most potent CAIs detected here showed favorable IOP lowering effects in an animal model of glaucoma. Being water soluble, with pH of the solution in the neutral range, and with duration of action lasting up to 4–8 h, this new class of CAIs may constitute interesting candidates for developing novel antiglaucoma therapies, a field in which no new drug emerged in the last 15 years.

Experimental Protocols

Chemistry

¹H, ¹³C, DEPT, COSY, HMQC and HMBC spectra were recorded using a Bruker Advance III 400 MHz spectrometer. The chemical shifts are reported in parts per million (ppm) and the coupling constants (J) are expressed in Hertz (Hz). For all new compounds DEPT, COSY, HMQC and HMBC were routinely used to definitely assign the signals of ¹H and ¹³C. Infrared spectra were recorded on a Perkin Elmer Spectrum R XI spectrometer as solids on KBr plates. Melting points (m.p.) were measured in open capillary tubes, unless otherwise stated, using a Büchi Melting Point B-540 melting point apparatus and are uncorrected. Thin layer chromatography (TLC) was carried out on Merck silica gel 60 F₂₅₄ aluminium backed plates. Elution of the plates was carried out using ethyl acetate/n-hexane or MeOH/DCM systems. Visualization was achieved with UV light at 254 nm, by dipping into a 0.5 % aqueous potassium permanganate solution, by Hanessian’s Stain solution and heating with a hot air gun or by exposure to iodine.

All other solvents and chemicals were used as supplied from Aldrich Chemical Co., Acros, Fisher, Alfa Aesar or Lancaster Synthesis. Aniline 1, morpholin-4-amine 2, 4-methylpiperazin-1-amine 3, (±) sec-Butylamine 4, 2-Morpholinoethanamine 5, N₁,N₁-bis(2-Aminoethyl)ethane-1,2-diamine 6 (Tris), Benzylamine 7 (CAS 100-46-9), Pyridin-4-ylmethanamine 8 (CAS 3731-53-1), 2′-(Piperidin-1-yl)ethanamine 9, 2-Aminothiazole 10, Glycine 11, 3-(1H-Imidazol-1-yl)propan-1-amine 12, sodium dimethyldithiocarbamate 13a, sodium diethyldithiocarbamate 14a, Pyrrolidine 15, Diisobutylamine 16, Dipropylamine 17, Dibutylamine 18, Dihexylamine 19, Ethylbutyamine 20, Diethanolamine 21, N-methylbenzenamine 22, N,N-Benzylmethylamine 23, Morpholine 24, Piperazine 25, 4-Cyano-4-phenylpiperidine hydrochloride 26, L-Proline 27 were purchased from Sigma-Aldrich (Milan, Italy) and were of the highest available purity. Purity of the prepared DTCS has been determined by HPLC and was > 99 %.

General procedure for the synthesis of compounds 1a-27a.³⁰

Secondary/primary amines 1-27 (1.0 g, 1.0 eq) were treated with a NaOH, KOH or Et₃N (1.0 - 2.2 eq), 4.0 ml of MeOH as co-solvent was used, and the solutions were stirred at 0°C for 20 min (Scheme 1). Then carbon disulfide (1.2 – 2.4 eq) was added dropwise and the mixture was stirred at r.t. until starting material was consumed (TLC monitoring). The solvents were removed under vacuo at r.t. and the residues obtained were dissolved in MeOH, filtered off trough Celite and the filtrate was concentrated in vacuo not exceeding 20 °C.

Synthesis of triethylammonium phenylcarbamodithioate 1a

Aniline 1 (0.5 g, 1.0 eq) was treated with triethylamine (1.0 eq) in benzene (0.5 ml) followed by addition of carbon disulfide (1.0 eq) at 0°C. The mixture was warmed to r.t. and stirred O.N. at r.t.. The solid formed was washed with diethyl ether and dried under vacuo to afford the titled compound as a light yellow solid in 51% yield.

Triethylammonium phenylcarbamodithioate 1a: ν_max (KBr) cm⁻¹, 2960, 2886, 1648, 1599, 1520, 1451; δ_H (400 MHz, DMSO-d₆) 1.13 (9H, t, J 6.8, 3x CH₂CH₃), 2.98 (6H, brs, 3x CH₂CH₃), 6.97 (1H, t, J 8.0, 4-H), 7.22 (2H, dd, J 8.3, 8.0, 2x 3-H), 7.93 (2H, d, J 8.3, 2x 2-H), 9.00 (1H, brs, exchange with D₂O, (CH₃CH₂)₃N⁺-H), 10.10 (1H, brs, exchange with D₂O, N-H); δ_C (100 MHz, DMSO-d₆) 10.0, 46.6, 114.8, 122.9, 128.4, 143.2, 215.5 (C=S); m/z (ESI), 168 [M-Na]⁻.

Diisobutylcarbodithioic acid sodium salt 16a: m.p. 220 °C with dec; ν_max (KBr) cm⁻¹, 2961, 2933, 2867, 1640, 1601, 1520, 1480, 1090; δ_H (400 MHz, DMSO-d₆) 0.84 (12H, d, J 6.8, 4 x CH₃), 2.43 (4H, m, 2 x CH), 3.86 (4H, d, J 7.2, 2 x CH₂); δ_C (100 MHz, DMSO-d₆) 21.2, 27.4, 61.5, 215.4 (C=S); m/z (ESI), 204 [M-Na]⁻. Data are in agreement with reported data.³⁰

Synthesis of morpholinecarbamodithioate sodium salt 24a.

Morpholine 24 (1.0 g, 1.0 eq) was treated according to the general procedure with 1.0 M aqueous solution of NaOH (1.0 eq) followed by addition of carbon disulfide (1.2 eq). The title compound was obtained as a white solid in quantitative yield.

Morpholinecarbamodithioate sodium salt 24a: m.p. 320 °C with dec; ν_max (KBr) cm⁻¹, 2966, 2901, 2854, 1625, 1520, 1416, 1215; δ_H (400 MHz, DMSO-d₆) 3.52 (4H, t, J 8.0 CH₂), 4.33 (2H, t, J 8.0, CH₂); δ_C (100 MHz, DMSO-d₆) 50.6, 67.1, 215.4 (C=S); m/z (ESI), 162 [M-Na]⁻.

CA inhibition

An Applied Photophysics stopped-flow instrument has been used for assaying the CA catalysed CO₂ hydration activity. 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 Na₂SO₄ (for maintaining constant the ionic strength), following the initial rates of the CA-catalyzed CO₂ hydration reaction for a period of 10–100 s.³² The CO₂ 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 15 min at room temperature prior to assay, in order to allow for the formation of the E-I complex. The inhibition constants were obtained by non-linear least-squares methods using PRISM 3, as reported earlier,³³ and represent the mean from at least three different determinations. All CA isofoms were recombinant ones obtained in house as reported earlier.^9,10

Co-crystallization and X-ray data collection of CA II complex

Co-crystals for the hCA II – 24a complex were obtained using the hanging drop vapor diffusion method.³⁷ Drops of 10 μL (0.3 mM hCA II; 0.7 mM DTC 24a; 0.1 % dimethyl sulfoxide; 0.8 M sodium citrate; 50 mM Tris-HCl; pH 8.0) were equilibrated against the precipitant solution (1.6 M sodium citrate; 50 mM Tris-HCl; pH 8.0) at room temperature (~20 °C). Crystals were observed after 5 days. Based on visual selection a crystal of the complex was cryoprotected by quick immersion into 20% glycerol precipitant solution and flash-cooled by exposing to a gaseous stream of nitrogen at 100 K. The X-ray diffraction data was collected using an R-AXIS IV⁺⁺ image plate system on a Rigaku RU-H3R Cu rotating anode operating at 50 kV and 22 mA, using Osmic Varimax HR optics. The detector-crystal distance was set to 80 mm. The oscillation steps were 1° with a 5 min exposure per image. Indexing, integration, and scaling were performed using HKL2000.³⁸

Structure determination of CA II drug complex

Starting phases were calculated from Protein Data Bank (PDB) entry 3KS3³⁹ with waters removed. Refinement using Phenix package,⁴⁰ with 5% of the unique reflections selected randomly and excluded from the refinement data set for the purpose of R_free calculations,⁴¹ was alternated with manual refitting of the model in Coot.⁴² The validity of the final model was assessed by PROCHECK.⁴³ Complete refinement statistics and model quality are included in Table 2.

Animals and glaucoma induction

Adult male New Zealand Albino rabbits weighing 2–2.5 Kg were employed in this study. The animals were utilized in groups of eight for each of the chosen specific treatments. The experimental procedures were conform to those of the Declaration of Helsinki and with the Guideline for the Care and Use of Laboratory Animals as adopted and promulgated by the U.S. National Institute of Health, and were conducted upon authorization of the Italian Regulations on Protection of Animals used for experimental and other scientific purpose (DM 116/1992) as well as with the European Union Regulations (OJ of ECL 358/1, 12/12/1986), and the experimental protocol was approved by the local animal care committee of the University of Florence (Florence, Italy). The rabbits were kept in individual cages; food and water were provided ad libitum. The animals were identified with a tattoo on the ear, numbered consecutively and maintained on a 12–12h light/dark cycle in a temperature controlled room (22°–23°C). All selected animals were examined before the beginning of the study and were determined to be normal on ophthalmic and general examinations. Glaucoma was induced by injection of 0.1 ml 0.25% carbomer (Siccafluid, FarMila - THEA Pharmaceuticals) into anterior eye-chamber bilaterally in New Zeeland albino rabbits] anesthetized with tiletamine and zolazepam (Zoletil 100, 0.05 mg/Kg b.w.) plus xilazine (Xilor 2%, 0.05 ml/Kg b.w.) i.m., by the procedure previously reported.^10c IOP was measured before carbomer injection and after 1, 2 and 4 hours the first day and three times a day until stabilization and then every 24 hours. All rabbits treated with carbomer presented a net increase in IOP. One drop of 0.2% oxybuprocaine hydrochloride (Novesine, Sandoz) diluted 1:1 with sterile saline was instilled in each eye immediately before each set of pressure measurements. IOP was measured using a Tono-Pen XL tonometer (Medtronic Solan, USA) as reported by earlier.¹⁰ The pressure readings were matched with two-point standard pressure measurements at 1, 2, 4 and 8 hours after the instillation of the drug and once a day for the following days using a Digilab calibration verifier. All IOP measurements were done by the same investigators using the same tonometer. As soon as a stable IOP increase was obtained, the animals were treated with the drugs in study. The efficacy of the different drugs in lowering IOP was evaluated after drug administration over four hours, with the following schedule: before and after 30, 60, 90, 120 and 240 minutes after drug administration. The treatment was performed in three animals per drug in one eye and compared to the contralateral eye treated with vehicle. A group of 4 non glaucomatous albino rabbits was treated with the drugs of this study and used as control. At the end of the experiments the animals were killed with a lethal dose of Pentothal (Abbott S.p.A., Campoverde di Aprilia, LT).

Abbreviations

CA

carbonic anhydrase

CAI

carbonic anhydrase inhibitor

DTC

dithiocarbamate