Combining carbonic anhydrase and thioredoxin reductase inhibitory motifs within a single molecule dramatically increases its cytotoxicity

Abstract A hypothesis that simultaneous targeting cancer-related carbonic anhydrase hCA IX and hCA XII isoforms (whose overexpression is a cancer cell’s defence mechanism against hypoxia) along with thioredoxin reductase (overexpressed in cancers as a defence against oxidative stress) may lead to synergistic antiproliferative effects was confirmed by testing combinations of the two inhibitor classes against pancreatic cancer cells (PANC-1). Combining both pharmacophoric motifs within one molecule led to a sharp increase of cytotoxicity. This preliminary observation sets the ground for a fundamentally new approach to anticancer agent design.


Introduction
Growing malignant tumours possess numerous defence mechanisms which ensure survival of aberrantly proliferating cancer cells while normal cells in the surrounding tissues are being killed and replaced with the neoplasm. 1 Among these mechanisms, overexpression of membrane-bound isoforms of carbonic anhydrase (hCA IX and XII) protect tumours from hypoxia and provide defence against acidification of the environment surrounding the growing tumour. 2 Selective targeting of hCA IX and XII has been confirmed to lead to retardation of tumour growth and, ultimately, reduction of tumour size. 3 Recently, we reported on the discovery of potent 1,2,4-oxadiazole-based inhibitors 1-3 of these two cancer-related isoforms of carbonic anhydrase. 4,5 These compounds possess subnanomolar to double-digit nanomolar potency against hCA IX and XII (Table 1), which makes them ideal tool compounds for further investigation of the role of these carbonic anhydrase isoforms in tumour growth.
Another important mechanism of tumour survival which can be considered a target for anticancer agent design, 6 is that providing tumour cell defence against oxidative stress (reactive oxygen species or ROS). In particular, tumour cells have been shown to overexpress thioredoxin reductase (TrxR, EC 1.8.1.9) which contributes to their resistant phenotype characterised by higher levels of ROS. 7 Thus, targeting TrxR1 (the most widespread cytosolic isoform of human TrxR) is an emerging approach to selective killing of cancer cells. 8 TrxR is a selenocysteine (Sec) enzyme which, along with NADPH and thioredoxin (Trx), constitutes the thioredoxin system responsible for maintaining Trx in the reduced bis-sulfhydryl state. 9 Among several classes of inhibitors possessing a variable degree of electrophilicity towards the catalytic Sec residue (reviewed by Bellelli 10 and Fang 11 ), we found covalent Michael acceptor inhibitors (such as Ugi-type adducts 4-6 which we dubbed "Ugi Michael Acceptors" or UMAs) to be particularly efficacious. 12 From the mechanistic standpoint, the inhibitory action of UMAs towards TrxR1 likely is exerted via the irreversible covalent trapping of the catalytic Sec residue (which exists in the ionised form at physiological pH 13 ) by the electrophilic b-benzoyl acrylamide (or acrylate) moiety present in 4-6.
The inhibitory potency of these compounds was established at the submicromolar to low micromolar level (Table 1) on a recombinant TrxR1 enzyme and compound 6 (DVD-445) was nominated a lead for further optimisation.
In this work, we aimed at verifying whether combinations of the said inhibitors from different classes (i.e. any of 1-3 with any of 4-6) would have stronger antiproliferative effects on tumour cells in comparison with a given single agent. With this idea in mindalong with the general understanding of the pharmacophoric moieties present in the carbonic anhydrase inhibitors (the zinc-binding primary sulphonamide group, 3 ZBG) and thioredoxin reductase inhibitor (the a,b-unsaturated Michael acceptor moiety 12 )we also aimed to combine these critical inhibitory motifs in the structure of a single agent and verify whether this will lead to potentiation of its cytotoxicity compared to inhibitors 1-6. Herein, we present the results of these studies.

Chemical synthesesgeneral
All reagents and solvents were obtained from commercial sources and used without purification. All reactions implemented in an open flask without any protection from CO 2 and H 2 O. Reactions were monitored by analytical thin-layer chromatography (TLC) Macherey-Nagel, TLC plates Polygram V R Sil G/UV254. Visualisation of the developed chromatograms was performed by fluorescence quenching at 254 nm. 1 H and 13 C NMR spectra were measured on Table 1. Chemical structures and inhibitory profile of carbonic anhydrase inhibitors 1-3 and thioredoxin reductase inhibitors 4-6 employed in this study (the pharmacophoric inhibitory motifs are highlighted red and blue, respectively).
To a solution of amidoxime 9 (1.0 mmol) and methyl cyclopropanecarboxylate (1.5 mmol) in DMSO (1.0 ml) powdered NaOH (60 mg, 1.5 mmol) was rapidly added. The reaction mixture was stirred at room temperature for 2 h. The reaction mixture was diluted with cold water (20 ml). The resulting precipitate was filtered off, washed with water (15 ml) and dried in air at 50 C.

General procedure for the preparation of compounds 4-6
Aldehyde (1 mmol) and amine (1 mmol) were combined in methanol (3 ml) and the resulting solution was stirred for 15 min at r. t. Isocyanide (1 mmol) and carboxylic acid (1 mmol) were added and the stirring continued for 24 h. The crystalline precipitate (formed in case of products 5a, 5i and 5j) was collected by filtration and washed with ether (3 ml); in all other cases, the solution was concentrated to dryness and the residue was purified by column chromatography on silica gel using dichloromethane-methanol (200:1) mixture as eluent.

Cell viability assay
Human cell lines were maintained at 37 C in humidified atmosphere containing air and 5% CO 2 as previously described. 15 Pancreas ductal adenocarcinoma cells PANC-1 were obtained from Russian collection cell cultures at the RAS Institute of Cytology (Saint Petersburg, Russian Federation). Cell line were grown in Dulbecco's Modified Eagle's Medium-F12 (BioloT) containing 10% (v/v) heat-inactivated foetal calf serum (FCS, HyClone Laboratories, UT), 1% L-glutamine, 1% sodium pyruvate, 50 U/ml penicillin, and 50 lg/ml streptomycin (BioloT). Cytotoxicity of carbonic anhydrase inhibitors was evaluated using a routine colorimetric method with tetrazolium dye À 3-(4,5-dimethylthiazol-2-yl)À2,5-diphenyltetrazolium bromide (MTT). The cell lines were incubated for 48 h under normoxia and in the presence of the hypoxia-mimicking agent 50 lM CoCl 2 with medium containing different concentrations of carbonic anhydrase inhibitors. Following treatment, Dulbecco's Modified Eagle's Medium-F12 (100 ll/well) and 20 ll of a 2.5 mg/ml MTT solution were added and cells were incubated for 1 h at 37 C. The used cell density was 5 Â 103 cells/200 ll/well in 96-well microtiter plates. After aspiration of the supernatants, the MTT-formazan crystals formed by metabolically active cells were dissolved in dimethyl sulphoxide (100 ll/well) and absorbance was measured at 540 nm and 690 nm in Varioskan LUX TM Multimode Microplate Reader (Thermo Scientific, Waltham, MA). Values measured at 540 nm were subtracted for background correction at 690 nm, and the data were plotted as a percent of control untreated samples.

Chemistry
Compounds 1 and 2 were synthesised from common precursor À 5-sulphamoylthiophene-3-carboxylic acid (7) as described previously. 5 Carboxylic acid 7 was activated by treatment with CDI and the resulting imidazolide was reacted with amidoximes 8a-b. Addition of solid NaOH formed a superbase solution in DMSO which triggered cyclodehydration yielding compounds 1 and 2 in good yields (Scheme 1).
Diamides 4 and 5 were prepared via the traditional Ugi reaction. However, employing secondary b-(methylamino)ethanol as the amine component in the preparation of compound 6 (DVD-445) produced a different, amide ester scaffold (Scheme 3). 12 For the synthesis of UMA/primary sulphonamide ZBG hybrid 10, the following synthetic strategy was adopted. Known 14 sulphonyl chloride 11 was converted to bis-DMB-protected (DMB ¼ 2,4dimethoxybenzyl) aldehyde 12. The latter was involved in the Ugi reaction with cyclopropylamine, tert-butyl isocyanide, and p-toluoyl acrylic acid conducted at ambient temperature in methanol to give adduct 13 which was used in the next step without further purification. Removal of the DMB protecting groups with TFA gave the target UMA/ZBG hybrid 10 in modest yield (Scheme 4).

Biological evaluation
Initially, compounds 1-3 (100 lM) and 4-5 (1 lM) were tested as single agents against pancreas ductal adenocarcinoma cell line (PANC-1) to establish the basal cytotoxicity levels of the two groups of compounds exerting their cytotoxicity via entirely different mechanisms (hCA IX/XII inhibition and TrxR inhibition, respectively). As the result, only compounds 2 and 4 had a pronounced antiproliferative effect reducing the cell viability by >50% and Scheme 1. Preparation of compounds 1 and 2.
Next, we proceeded to text carbonic anhydrase inhibitors 1-3 (100 lM) in combination with thioredoxin reductase inhibitors 4-6 (1 lM). To our delight, in all cases, TrxR inhibitor 4 produced a strong potentiation of the carbonic anhydrase inhibitor's antiproliferative activity. Addition of TrxR inhibitors 5 and 6 seemed to make little difference for the antiproliferative effect of carbonic anhydrase inhibitors 1 and 3. However, all three TrxR inhibitors 4-6 noticeably potentiated the cytotoxicity of carbonic anhydrase inhibitor 2 (Figure 2).
Encouraged by these findings, we also tested UMA/ZBG hybrid 10 against the same cell line. Pre-emptively expecting higher cytotoxicity of this compound compared to 1-3 and 4-5 taken as single agents or even in combination with each other, we tested compound 10 in the 1 À 100 lM concentration range. On increasing the concentration of compound 10 from 1 to 10 lM a sharp decrease in the number of viable cells was observed. To confirm this cytotoxic effect being dose-dependent, a full dose-response curve was obtained from which an IC 50 value of 1.8 ± 0.4 lM was deduced (Figure 3).

Conclusions
We hypothesised that simultaneous targeting of two cancer cells' defence mechanismthose against hypoxia and oxidative stressby cooperative inhibiting carbonic anhydrase IX/XII and thioredoxin reductase, respectively, would have a synergistic effect. This was confirmed by testing several combinations of the previously reported CA IX/XII and TrxR inhibitors. These results led to an idea of combining CA and TrxR inhibitory pharmacophores (a primary sulphonamide zinc-binding groups and a Michael acceptor moiety, respectively) in the structure of a single agent. Testing a pilot compound conforming to this design principle demonstrated that the compound has a pronounced cytotoxic effect against pancreas ductal adenocarcinoma cells PANC-1 (IC 50 ¼ 1.8 ± 0.4 lM). This observation validated a new approach to the design of potential anticancer agents. Further studies are underway in our laboratories which are aimed at further investigating the structure-activity relationships of structurally diverse zinc-binding group-containing Michael acceptors as antiproliferative agents and to correlate the phenotypic assay data with the inhibition profile against the isolated recombinant enzymes. The results of these studies will be reported in due course.
Detailed experimental procedures for synthesis, complete characterization data and copies of 1 H and 13 C NMR spectra are available online (Supplementary material).