Natural Products That Inhibit Carbonic Anhydrase

Abstract

The chemical diversity, binding specificity and propensity to interact with biological targets has inspired many researchers to utilize natural products as molecular probes. Almost all reported carbonic anhydrase inhibitors comprise a zinc binding group in their structure of which the primary sulfonamide moiety (-SO₂NH₂) is the foremost example and to a lesser extent the primary sulfamate (-O-SO₂NH₂) and sulfamide (-NH-SO₂NH₂) groups. Natural products that comprise these zinc binding groups in their structure are however rare and relatively few natural products have been explored as a source for novel carbonic anhydrase inhibitors. This chapter will highlight the recent and growing interest in carbonic anhydrase inhibitors sourced from nature, demonstrating that natural product chemical space presents a rich source of potential alternate chemotypes for the discovery of novel drug-like carbonic anhydrase inhibitors.

Susan C. Frost and Robert McKenna (eds.). Carbonic Anhydrase: Mechanism, Regulation, Links to Disease, and Industrial Applications

1 Introduction

Carbonic anhydrases (CAs) are zinc metalloenzymes that catalyze the reversible hydration of carbon dioxide to bicarbonate and a proton [1]. The active site zinc cation is the implied target for small molecule inhibitors to block the endogenous CA catalyzed reaction. Almost all reported CA inhibitors comprise a zinc binding group (ZBG) of which the primary sulfonamide moiety (-SO₂NH₂) is the foremost example, and to a lesser extent the primary sulfamate (-O-SO₂NH₂) and primary sulfamide (-NH-SO₂NH₂) groups. Compounds sourced from nature that comprise either a primary sulfonamide, sulfamate or sulfamide moiety in their structure are exceedingly rare. A literature search of the Dictionary of Natural Products (DNP) database [2] (a comprehensive and fully-edited database on natural products (NPs)) revealed just two NP primary sulfonamide compounds, (−)-altemicidin 1 and psammaplin C 2, and five NP primary sulfamate compounds, nucleocidin 3 [3], 5′-O-sulfamoyl adenosine 4 [4], 5′-O-sulfamoyl 2-chloroadenosine 5 [5], 5′-O-sulfamoyl 2-bromoadenosine 6 [5, 6] and 5′-O-sulfamoyl tubercidin 7 (Fig. 16.1) [7, 8]. (−)-Altemicidin 1 is a marine alkaloid isolated from the actinomycete strain Streptomyces sioyaensis [9]. This compound exhibited potent acaricidal activity as well as strong inhibition of tumor cell growth [10]. A total synthesis of 1 as well as the isolation of two secondary sulfonamide analogues of 1 have subsequently been reported [9]. Psammaplin C 2 is a bromotyrosine amino acid derivative isolated from the marine sponge Pseudoceratina purpurea [11, 12]; no bioactivity for this alkaloid has been reported to date. The sulfamate nucleosides 3–7 were isolated from actinomycete species belonging to the genus Streptomyces. These structurally related sulfamates are reported to have a range of biological effects including cytotoxicity[4], herbicidal activity [6–8, 13], inhibition of blood platelet aggregation [5], antibacterial activity [5, 14] and antitrypanosomal activity [14]. Nucleocidin 3, isolated from the fermentation broth of Streptomyces calcus, is of particular note since it was the first NP to contain either a fluorinated carbohydrate or a primary sulfamate group. This molecule became an attractive synthetic target owing to its novel structural features, with the first total synthesis reported in 1976 [15]. Whilst NP 3 has been shown to exhibit broad spectrum antibacterial effects as well as potent antitrypanosomal activity, its potential use in the clinic has been limited due to toxicity [16]. The NPs 1–7 have not been investigated for CA inhibition properties, however it is likely that these compounds would inhibit CA activity owing to the presence of an unhindered primary sulfonamide or primary sulfamate moiety within their structure. Finally, our search of the DNP failed to identify any NP primary sulfamides.

Fig. 16.1

Natural products (NPs) that comprise a primary sulfonamide or sulfamate moiety in their structure. (a) NP primary sulfonamides: (−)-altemicidin 1 and psammaplin C 2. (b) NP primary sulfamates: nucleocidin 3, 5′-O-sulfamoyl adenosine 4, 5′-O-sulfamoyl 2-chloroadenosine 5, 5′-O-sulfamoyl 2-bromoadenosine 6 and 5′-O-sulfamoyl tubercidin 7

At the time of writing the Protein Data Bank (PDB) contained X-ray structures of ∼160 sulfonamide ligands (R-SO₂NH₂) in complex with hCA II (h = human). The binding mode of the sulfonamide anion (R-SO₂NH^-) to the Zn²⁺ cation is invariant in these structures, with the sulfonamide anion coordinated to the active site Zn²⁺. Primary sulfamates and sulfamides, ZBG isosteres of primary sulfonamides, contribute an additional ∼35 X-ray structures of ligands in complex with hCA II in the PDB. Of the remaining PDB protein-ligand structures most comprise very simple ligands such as anions or small organic molecules, there are however several structures comprising more complex alternate CA ligands, some of which are NPs.

NPs comprise a vast collection of diverse chemical structures and have proven to be an invaluable source of new chemotherapies [17–22]. Plant NPs have been the basis of traditional medicine for thousands of years and continue to actively contribute to contemporary drug discovery [22]. In more recent times, marine macro- and micro-organisms along with terrestrial microbes have been the source of numerous lead molecules or drugs [23]. The significance of NPs in drug discovery is most evident in the anticancer and anti-infective therapeutic areas [20, 24–26]. For example, between 1940 and 2011 48.6 % of all new anticancer small molecule therapies approved by the FDA were either NPs or NP derivatives [26]. Furthermore between 1981 and 2011 75 % of all antibacterial new chemical entities were either NPs or their derivatives [26]. The success of NPs and their semi-synthetic derivatives as therapeutic agents is intrinsically linked to the fact that NPs have been biologically pre-validated and selected during evolution to bind to biosynthetic enzymes [27–32]. It has been hypothesized that this inherent capacity to bind in biological space allows NPs to also recognize human therapeutic targets [27, 29, 32]. Furthermore, computational studies have shown that NPs occupy complementary areas of chemical space compared with synthetic compounds, and thus should be implemented to increase the chemical complexity and drug-likeness of screening libraries [28–31]. The chemical diversity, binding specificity and efficiency, and propensity to interact with biological targets have inspired many researchers to utilize NPs as molecular probes. These studies go beyond the identification of potential new lead or drug molecules, and have increased our understanding of biological pathways and systems [33, 34]. While the primary sulfonamide and sulfamate moieties are poorly represented in NP chemical space, this space does provide a rich source of alternate chemotypes for the discovery of CA inhibitors with a different enzyme binding mode to the typical ZBGs. The remainder of this chapter will highlight the recent and growing interest in novel CA inhibitors sourced from nature.

2 Natural Products That Inhibit Carbonic Anhydrase

Several classes of novel CA inhibitors have been identified from screening collections of NPs, most notable are coumarin and phenol containing NPs [35]. NPs comprising these fragments display diverse profiles of CA inhibition when compared to classical ZBGs and these will be discussed further in the following sections. In addition a selection of NP-derived CA inhibitors have been synthesized where the NP scaffold has been synthetically modified to incorporate the classical ZBG of CA inhibitors i.e. sulfonamide, sulfamate or sulfamide. A summary of these hybrid molecules will also be presented.

2.1 Coumarins

Coumarin compounds are abundant secondary metabolites in plants and are found to a lesser extent in microorganisms and animal sources. Plant coumarins are phytoalexins, defense compounds produced when the plant is under threat from other organisms, and have attracted interest owing to a range of biological activities including antimicrobial, molluscicidal, acaricidal, antiviral, anticancer, antioxidant and anti-inflammatory properties [36]. The coumarin structure comprises a benzopyrone core, with NP coumarins categorized as (a) simple coumarins, (b) furanocoumarins, (c) pyranocoumarins, (d) bis- and triscoumarins, or (e) coumarinolignans [36]. Simple coumarins, including coumarin 8 (Fig. 16.2) are highly abundant in several plant species belonging to the taxonomic families Umbelliferae, Rutaceae and Compositae [36]. A recent review of coumarin-based drugs highlights the growing interest in the coumarin compound class to deliver new therapeutics, [37] which is driving efforts towards the isolation and the structural characterization of further novel bioactive coumarin derivatives. In recent years it was discovered that NP coumarins inhibit CA enzymes via an alternate and unprecedented mechanism to classical sulfonamides [38], these findings are described next.

Fig. 16.2

NP coumarin CA inhibitors. (a) Coumarin 8. (b) CA hydrolysis of coumarin 9 to form the cinnamic acid compound 10

Nature Bank is a unique drug discovery resource that encompasses a diverse collection of >50,000 biota samples of plants, fungi and marine invertebrates collected from Australia, China and Papua New Guinea along with biota extracts, semi-purified fractions and pure compounds [39]. A selection of Leionema ellipticum (family Rutaceae) extracts was sourced from Nature Bank and screened using Fourier transform ion cyclotron resonance electrospray ionization mass spectrometry (FTICR ESI MS) for binding to bovine CA II (bCA II) [38]. From this study the NP coumarin, 6-(1S-hydroxy-3-methylbutyl)-7-methoxy-2H-chromen-2-one 9 was identified as a ligand for bCA II as it formed a noncovalent complex that could be detected by ESI MS. In follow on studies it was demonstrated that coumarin 9 inhibits a spectrum of human CAs in an unprecedented time dependent manner [40]. The usual enzyme assay conditions to investigate small molecule inhibition of CA activity is to incubate the test compound with the CA protein of interest for 15 min prior to monitoring the effect on CA-mediated CO₂ hydration. Under these conditions this coumarin had only weak CA inhibition prompting us to extend the pre-incubation time. Following 6 h of pre-incubation with hCA II the K _i of coumarin 9 dropped to 60 nM (with a similar reduction in K _i at other CA isozymes also observed). As the coumarin chemotype lacked a classic ZBG typical of known small molecule CA inhibitors and displayed unusual time-dependent inhibition it was important to understand how this chemotype binds to and inhibits CAs. Using protein X-ray crystallography the crystal structure of hCA II with this NP was obtained at a resolution of 2.0 Å (Fig. 16.3). The coumarin 9 was not observed, instead the hydrolysis product of 9, the cinnamic acid derivative 10 was identified [40]. Esterase activity is known for CAs [41–43] and the observation of cinnamic acid 10 rather than NP coumarin 9, although unexpected, could be rationalized as a consequence of hCA II esterase activity leading to hydrolysis of the lactone of 9. The bulky hydrolysis product 10 then plugged the CA active site cavity entrance, exhibiting no interactions with the catalytic zinc ion. This unusual inhibition mode is previously unobserved for CAs and together with the coumarin pedigree in medicinal chemistry is suggestive of a potential new avenue for drug development compared to the ZBGs of classical CA inhibitors. Reactive Michael acceptors are a general structural alert in drug discovery, however it has been demonstrated that simple coumarins exhibit poor protein binding characteristics compared to other carbonyl containing Michael acceptors [44, 45]. The lower reactivity of the coumarin double bond compared with other Michael acceptors, has been attributed to it being part of a pseudoaromatic system [46].

Fig. 16.3

Detailed interactions between hCA II and NP coumarin 9 hydrolysis product 10 from a protein X-ray crystal structure. The catalytic site showing the tetrahedral Zn²⁺ cation (violet sphere) with the three coordinated His ligands (His94, His96, and His119) and a water molecule (red sphere). The cinnamic acid 10 (gold) interacts with three active site ordered water molecules (red spheres), with Phe131 and Asn67 (CPK colors) from the active site as well as with Glu238sym (yellow) from a symmetry related enzyme molecule. The proton shuttle residue His64 is shown (CPK colors) (Reprinted with permission from Maresca, A.; Temperini, C.; Vu, H.; Pham, N. B.; Poulsen, S.-A.; Scozzafava, A.; Quinn, R. J.; Supuran, C. T. J. Am. Chem. Soc. 2009, 131, 3057. Copyright 2009 American Chemical Society)

Following the findings outlined above we performed a substructure search of the Nature Bank [39] pure compound repository against the bare coumarin scaffold 8. A set of 81 coumarins were identified and from this a subset of 27 coumarins were sourced in sufficient quantity and purity for follow up evaluation as CA inhibitors [47]. These NP coumarins, compounds 11–37 (Fig. 16.4) comprise 24 plant coumarins (compounds 11–34) and three marine coumarins (compounds 35–37). Specifically, the plant NPs comprise avicennin 11 [48, 49], trans-avicennol 12 [50, 51], calanolide B 13 [39, 52], dihydrogeiparvarin 14 [53], geiparvarin 15 [53, 54], dehydromarmin 16 [53], xanthyletin 17 [55], xanthoxyletin 18 [50, 55], ceylantin 19 [56], alloxanthoxyletin 20 [55], fraxidin 21 [57], fraxin 22 [58], scopoletin 23 [59], 6,7,8-trimethoxycoumarin 24 [60], 5,7,8-trimethoxycoumarin 25 [60], 7-hydroxy-8-methoxycoumarin 26 [59], isoscopoletin 27 [61], fraxoside 28 [62], scopolin 29 [63], murralongin 30 [64], (+)-isomurralonginol nicotinate 31 [65], isophellodenol C 32 [66], ellagic acid 33 [67] and nasutin B 34 [68]. The ascidian NP coumarins include lamellarins E 35 [69], B 36 [70], and G 8-sulfate 37 [71]. A variety of bioactivities have been reported for these coumarins, for example calanolide B 13 isolated from the tropical rainforest tree Calophyllum lanigerum, displayed protection against HIV-1 replication and cytopathicity (EC₅₀ = 0.4 μM) [52]. Dihydrogeiparvarin 14 and geiparvarin 15, both isolated from Geijera parviflora [53, 54], possessed significant in vitro activity against human carcinoma of the nasopharynx [72, 73]. Xanthoxyletin 18 [50, 55], purified from a variety of Citrus species, acts as a DNA-damaging agent [74], while several synthetic derivatives have been shown to exhibit toxicity towards L-1210 leukemia cells with IC₅₀ values ranging from 0.9 to 60.3 μM [74]. The inhibition activity data for the NP coumarins 9 and 11–37 against hCA I and II (off-target isozymes), as well hCA IX and XII (isozymes of interest in cancer drug development) is presented in Table 16.1. Data for the simplest coumarin 8 and standard CA inhibitors (Fig. 16.5) is included for development of structure-activity relationships (SAR).

Fig. 16.4

NP coumarin library (11–37) (sourced from Nature Bank [39, 47])

Table 16.1 Inhibition data for coumarins 8, 9, 11–37 against hCA isozymes I, II, IX and XII [47]. Standard inhibitors (AZA, ZNS and TPM) are included for comparison [75–77]

Fig. 16.5

Standard CA inhibitors: acetazolamide AZA, zonisamide ZNS and topiramate TPM

Since the discovery of the NP coumarin 9 synthetic libraries of coumarins and thiocoumarins have been prepared and evaluated as CA inhibitors [79–81]. The complexity and diversity of NP coumarin structures far exceeds that described for synthetic coumarin CA inhibitors. Coumarin 8, the simplest coumarin, is not an appreciable inhibitor of CA IX or XII however it is a weak inhibitor of off-target CA I and CA II, with K _is of 3.1 and 9.2 μM, respectively. The NP coumarins are substituted at any of six available sites, with many fused to form tricyclic, tetracyclic or larger ring systems. This diversity does not readily allow simple SAR to be defined, however several trends surrounding CA inhibition are evident. Most obvious is that the NP coumarin library members are very weak CA II inhibitors, most have K _is > 100 μM, the only exception being the trimethoxycoumarin 25 (K _i = 9.65 μM). When compared to the structurally related methoxy/hydroxy coumarins 21–24, 26 and 27, compound 25 differs only in the pattern of substituents, this SAR indicates that it may be a combination of interacting substituents that directs the CA inhibition profile at CA II. At CA I, IX and XII many of the NP coumarins have K _is in the range of 1–10 μM, this tight grouping of K _is reflects minimal isozyme selectivity with these coumarins, however there are a few outliers to this general trend and these compounds represent interesting structures owing to their CA isozyme selectivity characteristics. At CA I there was one stand out compound being compound 24, a nanomolar CA I inhibitor. This trimethoxy coumarin is the most potent of any of the NP coumarins at CA I and is a structural isomer of 25, the only potent CA II coumarin of the study. Around half of the NP coumarins have submicromolar inhibition of the isozymes CA IX and XII, some of these coumarins (11, 12, 13, 14, 15, 18 and 25) are submicromolar at both CA IX and XII, while the remainder are submicromolar at either CA IX (19, 21–23, 26, 28 and 31) or CA XII (16, 29 and 32). This subset of NP coumarins has viable selectivity characteristics that warrant further studies in cell-based models of CA in cancer.

2.2 Phenols

The first single crystal X-ray structure of phenol 38 and a CA protein (hCA II) was reported in 1994 and identified that 38 binds in an unprecedented way within the enzyme active site [82]. It was shown that the phenolic OH interacts with the zinc-bound water molecule/hydroxide ion through a hydrogen bond while a second hydrogen bond formed between the phenolic OH and the NH amide of Thr199, an amino acid critical for the catalysis and inhibition of various CAs. NPs containing the phenol fragment 38 are highly abundant in nature. A substructure search of the DNP [2] against the phenol fragment identified >50,000 NPs from the 246,994 database entries that contain this fragment (∼20 % of all entries). Early CA inhibitory studies focused on simple, commercially available mono-, di- or tri-substituted phenols that are also found in nature [2] such as pyrocatechol 39, resorcinol 40, hydroquinol 41, salicylic acid 42, p-hydroxybenzoic acid 43, p-coumaric acid 44, caffeic acid 45, ferulic acid 46, gallic acid 47 and syringic acid 48 (Fig. 16.6) [83, 84]. A number of phenolic-NPs containing more complex scaffolds 49–57 have since been sourced from the Davis open-access compound repository housed at the Queensland Compound Library (QCL) [85], and screened against selected CAs (Fig. 16.6) [86, 87]. These phenolic-derivatives include the endophytic fungal metabolites, (−)-xylariamide A 49 [88], and its synthetic enantiomer (+)-xylariamide A 50 [88], xanthones 51 and 52 [89], the marine ascidian-derived alkaloids, polyandrocarpamines A 53 and B 54 [90, 91] and the plant secondary metabolites, endiandrins A 55 [92] and B 56, [93] and (−)-dihydroguaiaretic acid 57 [92, 93].

Fig. 16.6

NP phenols 38–57 tested as CA inhibitors [83, 84, 86, 87]

The phenols 38–57 have been evaluated for their inhibition of human cytosolic isoforms CA I and II (off-target) and mitochondrial isozymes CA VA and CA VB, Table 16.2. The latter have been recognized as potential targets for designing anti-obesity agents that act with a novel mechanism of action [94, 95]. The simple phenolic secondary metabolites 38–48 have also been tested against hCA III, IV, VI, VII, IX, XII, XIII and XIV [83, 84]. These data (not shown) indicate that the phenol class of NP CA inhibitor exhibits complex SAR, with small chemical changes leading to large effects on CA enzyme inhibition. The chemical diversity of phenolic NPs is vast; so far investigation of this chemotype for its interaction with CAs is in its infancy.

Table 16.2 Inhibition data for phenolic NPs 38–57 against hCA isozymes I, II, VA and VB [83, 84, 86, 87]. Standard inhibitors AZA, ZNS and TPM are included for comparison [75–77]

The β-CAs from Helicobacter pylori, Candida albicans, Candida glabrata, Cryptococcus neoformans and Brucella suis are essential for growth and have proven susceptible to inhibition with several compound classes including sulfonamides, carboxylates and boronic acids [96–103]. A positive correlation from enzyme assays to a cell-based anti-infective phenotype assay demonstrates that the β-CAs from these pathogens are potential druggable targets for anti-infective therapies. Mammals possess only α-CAs, whilst many pathogenic organisms, such as bacteria and fungi encode β-CAs. Similarly to α-CAs, a zinc cation defines the location of the active site of the β-CA enzymes. Phenols 49–57 [92, 93] along with the fungal NP phenols 58–62 [104–106] (Fig. 16.7) have been screened for enzyme inhibition against selected pathogen β-family CAs, Table 16.3. CAs from Mycobacterium tuberculosis, Candida albicans and Cryptococcus neoformans were studied and selectivity towards the pathogen isozymes over human CAs was assessed.

Fig. 16.7

Phenolic NPs 58–62 with activity against mycobacterial and fungal CAs [86]

Table 16.3 Enzyme inhibition of pathogenic M. tuberculosis β-CA isozymes Rv3273 and Rv1284, C. albicans isozyme Nce103, C. neoformans isozyme Can2 and human α-CA isozymes I and II with the NP phenols 49–62 [86] and standard CA Inhibitors AZA, ZNS, and TPM

These studies showed that several phenolic NPs were selective inhibitors of mycobacterial and fungal β-CAs, with the two best performing NPs identified as (−)-dihydroguaiaretic acid 57 and 3-chloro-4-hydroxyphenylacetamide 62. Specifically, 57 was a sub-micromolar β-CA inhibitor with up to 495-fold selectivity over hCA I and 371-fold selectivity over hCA II. Compound 62 was also a low micromolar inhibitor of the fungal CAs and displayed 130–280-fold selectivity over the two human CAs. These compounds were the first non-sulfonamide inhibitors that display β over α CA selectivity. In order to determine how the phenolic-based NPs 38–62 interacted with CAs, soaking and co-crystallization studies were undertaken with the readily available protein hCA II. While the most selective NPs 57 and 62 did not yield co-crystals with CAs suitable for X-ray diffraction studies, compound 50 [(+)-xylariamide A] did at a resolution of 2.0 Å. While it was predicted that the phenolic moiety present in 50 would play a major role in the hCA II binding it was discovered that instead the ester carbonyl of 50 interacts with a zinc-bound water molecule and is further engaged in a hydrogen bond donated by the backbone amide group of Thr198 (Fig. 16.8). In this crystal structure the electron density of the inhibitor is well defined, allowing unambiguous placement of the ligand. This was a totally new binding mode to CAs.

Fig. 16.8

The ligand-protein interactions are illustrated schematically for hCA II in complex with compound 50 [(+)-xylariamide A]. Protein residues are shown in grey; hydrogen bond interactions are shown as dotted lines, van der Waals interactions as dashed lines. Distances are given in Å. Two conformations of the phenol moiety were observed, giving rise to the appearance of the chloro substituent on both sides of the phenolic hydroxyl group; the refined occupancies for both positions are noted at each position (Reprinted with permission from Davis, R. A.; Hofmann, A.; Osman, A.; Hall, R. A.; Mühlschlegel, F. A.; Vullo, D.; Innocenti, A.; Supuran, C. T.; Poulsen, S. A. J. Med. Chem. 2011, 54, 1682. Copyright 2011 American Chemical Society)

2.3 Polyamines

Polyamines belong to an alkaloid structure class and have been reported from various natural sources including terrestrial and marine animals, plants, fungi and bacteria [2]. Two of the simplest polyamines isolated to date include spermine 63 and spermidine 64 (Fig. 16.9). A substructure search of the DNP against 63 and 64 identified >400 NPs from the 246,994 database entries that comprise these alkaloid fragments [2]. The polyamine chemotype has been shown to modulate multiple biological processes including gene expression, cell proliferation, translation, cell signaling, membrane stabilization and ion channel inhibition as well as antibacterial activity [107–113]. Despite the myriad bioactivities reported for polyamine NPs until recently no CA inhibition had been reported. Carta et al. showed that 63, 64 and several semi-synthetic polyamine analogues inhibited hCA I-XIV with K _i values ranging from low nanomolar to millimolar [114]. A single crystal X-ray structure of spermine 63 with hCA II (at a resolution of 2.0 Å) was also reported [114] showing compound 63 anchored to the zinc bound water ligand (as for phenol 38) through a network of hydrogen bonds. The terminal amine moiety of 63 is hydrogen bonded with residues Thr200 and Pro201. Notably 63 binds differently to hCA II when compared to either sulfonamides, phenols or coumarins and thus polyamines have the potential for the identification and development of additional CA inhibitors with a unique mechanism of binding and CA selectivity profile. This alkaloid structure class warrants further investigation and we expect that NPs will provide future opportunities to study additional polyamine alkaloids for CA inhibition.

Fig. 16.9

NP polyamine CA inhibitors spermine 63 and spermidine 64 [114]

2.4 Semi-synthetic NPs Modified to Incorporate a ZBG and Inhibit CA

2.4.1 Carbohydrate-ZBG Hybrid Molecules

Carbohydrates represent an abundant group of NPs and a selection of naturally occurring mono- and disaccharides have been modified to incorporate CA recognizing ZBG’s to give glycosyl primary sulfonamides (sugar-SO₂NH₂) [115], glycosyl primary sulfamides (sugar-NH-SO₂NH₂) [116], and glycoconjugate sulfamates (sugar-O-SO₂NH₂) [117]. Compounds 65–74 derived from the monosaccharides d-glucose, d-galactose, d-mannose and the disaccharide maltose are shown (Fig. 16.10). An aromatic group, which is typical for classical CA inhibitors, is absent from these compounds and instead they comprise the hydrophilic mono- or disaccharide fragment directly attached to the ZBG. These NP-ZBG hybrid molecules have been evaluated as CA inhibitors, Table 16.4. All carbohydrate-ZBG hybrid compounds behaved as weak inhibitors of hCA I, with the anomeric sulfonamides 65–67 and sulfamides 68–70 also weak micromolar inhibitors of hCA II, IX and XII. In contrast the C-6 sulfamates 71–74 delivered good activity, particularly monosacchrides 71–73, which showed K _i <10 nM against hCA XII. The glucose sulfamate 71 also has a K _i <10 nM at hCA IX and displayed selectivity for inhibiting the tumor-associated isoforms CA IX and XII over cytosolic CA I and II. The interested reader is directed to crystal structures for anomeric sulfonamides and glycoconjugate sulfamates in complex with hCA II in the PDB (accession codes: 3HKN, 3HKQ, 3HKT, 3HKU, 3T82, 3T83, 3T84, and 3T85).

Fig. 16.10

Carbohydrate-ZBG hybrid molecules 65–74: glycosyl primary sulfonamides (sugar-SO₂NH₂), glycosyl primary sulfamides (sugar-NH-SO₂NH₂) and glycoconjugate sulfamates (sugar-O-SO₂NH₂) [115–117]

Table 16.4 Inhibition data of hCA Isozymes I, II, IX and XII and cLog P values for carbohydrate-ZBG hybrid molecules 65–74 [115–117]

The membrane permeability properties were measured for selected carbohydrate-ZBG hybrid CA inhibitors, the results confirm that the compounds are expected to have poor passive membrane permeability. cLog P is an indicator of passive diffusion through cell membranes and values <0 are indicative of molecules with poor membrane permeability. The cLog P values of the hybrid molecules 65–74 range from −2.7 for monosaccharides to −5.5 for dissacharides, Table 16.4. The compound design, employing a deliberate approach towards CA IX/XII isozyme selectivity by changing the physicochemical properties to impart poor membrane permeability, is consistent with these cLog P values. Membrane permeable ester ‘prodrugs’ of the carbohydrate-ZBG hybrids were also synthesized, this allows for potential oral administration, with the polar carbohydrate-ZBG hybrid molecules ‘unmasked’ in vivo enabling targeting of extracellular CA IX and XII.

2.4.2 Coumarin-ZBG and Steroid-ZBG Hybrid Molecules

A selection of NP-derived sulfamates are potent inhibitors of the cancer drug target steroid sulfatase (STS) and are being developed as a therapy for hormone-dependent breast cancer [118]. This includes the steroidal sulfamate oestrone-3-O-sulphamate (EMATE) 75 and two coumarin based sulfamates, COUMATE-667 76 and STX-118 77 (Fig. 16.11), which at a simpler structural level may also be considered phenolic sulfamates. These NP hybrids, modified with the sulfamate ZBG, are also potent CA inhibitors, Table 16.5 [119–121]. It is hypothesized that dual steroid sulfatase/CA inhibitors may represent a novel method for treating hormone dependent breast cancer tumors, with the reversible binding of the sulfamates to erythrocyte CA II increasing the metabolic stability of the compounds by protecting the sulfamate moiety from rapid degradation [122]. This indirect improvement of biopharmaceutical properties may persist alongside the direct effect of modulating the activity of cancer-associated CA IX and XII. The X-ray crystal structure of hCA II with both 75 [123] and 76 [122] are reported. These structures conform to the classical ZBG interactions with the sulfamate moiety binding to the active site Zn²⁺ cation. The steroid fragment of 75 and the coumarin fragment of 76 interact with the residues in the hydrophobic half of the CA II active site.

3 Conclusion

Contemporary drug discovery is under increased pressure to identify more suitable small molecules as chemical starting points for drug development and finding novel compounds as starting points for optimisation is one of the major challenges in drug discovery research. NPs already provide a significant portion of FDA approved drugs and have emerged as an effective way to sample chemical diversity. The chemical diversity within NPs is vast and while the investigation of NP chemotypes for interaction with CAs is in its infancy, an encouraging start has been made. The NP compounds presented here (phenols, coumarins and polyamines) are suggestive of a tremendous opportunity that NPs provide for the discovery of novel chemotypes for selectively targeting either human or pathogen CAs. It will be imperative for future efforts to further evaluate the NP or NP-hybrid compounds in cell-based models of CA associated disease alongside classical control compounds for validation. The identification of unique CA binding for any NPs might offer possibilities for future rational drug discovery design and development. Thus the use of NPs in the search for new CA inhibitors has a strategic advantage since nature’s unique chemical diversity has only been superficially explored in this particular field of research. We predict additional NP structures classes will be identified as binding to and perturbing CA function as further research is undertaken.

Fig. 16.11

Steroidal sulfamate oestrone-3-O-sulphamate (EMATE) 75 and two coumarin based sulfamates, COUMATE-667 76 and STX-118 77

Table 16.5 Inhibition data of hCA Isozymes I, II, IX and XII for steroid- and coumarin-ZBG hybrid molecules 75–77