Discovery of novel sulphonamide hybrids that inhibit LSD1 against bladder cancer cells

ABSTRACT Aim:A series of sulphonamide hybrids were designed, synthesised, and identified as potential lysine-specific demethylase 1 (LSD1) inhibitors. Materials and methods: Bladder cancer cell lines were cultured to evaluate the antiproliferative activity. Inhibitory evaluation of sulphonamide hybrids against LSD1 were performed. Conclusion: sulphonamide derivative L8 exhibited the antiproliferative activity against HTB5, HTB3, HT1376, and HTB1 cells with IC50 values of 1.87, 0.18, 0.09, and 0.93 μM, respectively. Compound L8 as a selective and reversible LSD1 inhibitor could inhibit LSD1 with the IC50 value of 60 nM. It effectively inhibited LSD1 by increasing the expression levels of H3K4me1, H3K4me2, and H3K9me2 in HT1376 cells. To the best of our knowledge, this was the first report which showed that sulphonamide–quinoline–dithiocarbamate hybrids potently inhibited LSD1 in bladder cancer cells. Our studies give the potential application of the sulphonamide-based scaffold for developing LSD1 inhibitors to treat bladder cancer.


Introduction
Bladder cancer is the most frequently diagnosed malignancy in the urinary system and has the high morbidity and mortality rates 1 . Chemotherapy plays an important role in the treatment of bladder cancer and it is urgent to develop potent anti-bladder cancer drugs 2,3 . Histone lysine-specific demethylase 1 (LSD1) could catalyse the demethylation of mono and dimethylated H3K4me1/ 2 or H3K9me1/2 and demethylate many other nonhistone substrates 4 . LSD1 is aberrantly expressed in many malignant tumours such as prostate, ovarian, gastric, liver, breast, lung, bladder, neuroblastoma, and blood cancers 5 . The inhibition of LSD1 could prevent tumour cell proliferation, stimulate antitumor immunity, and enhance antitumor efficacy of immune checkpoint blockade 6 . Therefore, LSD1 has been considered as a potential cancer therapeutic target to discover novel anti-bladder cancer agents [7][8][9] . LSD1 and MAO-A/-B were belonged to the monoamine oxidase family, and MAO-A/-B shared the similar enzymatic mechanisms and the same cofactor of LSD1 in the cleavage of the inactivated carbon-nitrogen bonds from their substrates 10 . Although a variety of LSD1 inhibitors have been reported to date, many of them show insufficient selectivity towards LSD1 11 .
Sulphonamide has been proven to be an interesting scaffold and many sulphonamide derivatives are designed as potent antitumor agents for cancer therapy 12 . Phenylpropanoid-based sulphonamide 1 (Figure 1) induced cell cycle arrest at G1/S transition by reducing the expression levels of cyclin D1 and cyclin E in MCF7 cells 13 . Sulphonamide 2 as a potential antitumor agent was a novel tumour-associated isozyme carbonic anhydrase IX inhibitor 14 . Sulphonamide 3 showed a significant antitumor activity against HCT116 human colon carcinoma in vitro and in vivo 15 . Trans-3-(pyridin-3-yl)acrylamide-derived sulphamide 4 showed a single-digit nanomolar antiproliferative activity against DU145, Hela, and H1975 cells and inhibited NAMPT with an IC 50 value of 5.08 nM 16 . Benzenesulphonamide 5 showed the potent inhibitory effects against PC-3 cells with an IC 50 value of 4.08 mM as a potential tubulin polymerisation inhibitor 17 .
Quinolines and dithiocarbamates also represent a large group of anticancer agents and there have been many studies on the structural modification based on quinolines or dithiocarbamates 18,19 . Quinoline 6 ( Figure 2) showed the potently inhibitory activity against Raji cells by inducing changes in the cell cycle 20 . Quinoline derivative 7 triggered p53/Bax-dependent apoptosis by activating p53 transcriptional activity against colorectal cancer HCT-116 cells 21 . Quinoline 8 was a potent antiproliferative agent against HCT-116, RKO, A2780, and Hela cell lines with IC 50 values of 2.56, 3.67, 3.46, and 2.71 lM, respectively 22 . Dithiocarbamate 9 showed the potent cytotoxicity against MGC-803 and HGC-27 cells by the specific and robust inhibition of LSD1 23 . Dithiocarbamate 10 displayed the potent and reversible inhibition against LSD1 with an IC 50 value of 0.39 lM 24 . So, dithiocarbamate moiety might be a promising fragment to design novel LSD1 inhibitors.
Molecular hybridisation is a useful strategy in anticancer drug design based on the combination of different bioactive scaffolds to produce a new molecular architecture 25 . Based on these findings, we proposed that the sulphonamide derivatives with quinoline and dithiocarbamate groups might have the excellent anticancer activity and LSD1 inhibitory activity. Thus, in this study, we designed a series of sulphonamide-quinoline-dithiocarbamate hybrids via a molecular hybridisation strategy and evaluated for their antiproliferative activity and inhibitory activity against LSD1.

Materials and methods
General procedure for the preparation of J1$J7 A solution of aromatic amine (11 or 13, 3.5 mmol) and triethylamine (5.25 mmol) in acetone (10 ml) was cooled to 0 C. Sulphonyl chloride derivative (12 or 14, 5.25 mmol) in acetone (10 ml) was added drop-wise into the solution. After being stirred for 30 min, the reaction mixture was stirred at room temperature for 8 h. The reaction mixture was washed with water (50 ml) and extracted with ethyl acetate (30 ml). The organic extracts were purified by column chromatography on silica (petroleum ether/ ethyl acetate ¼ 10/1) to yield compounds J1~J7.

LSD1 dilution assay
LSD1 recombinant (Cayman Chemical, Fort Annapolis, MI) was incubated with the targeted compound or dimethyl sulphoxide for 60 min. Then, samples were diluted 80 times using a HRP-assay solution containing substrate and coupling reagents. Finally, LSD1 inhibitory activity before and after dilution was examined.

Western blotting
Cells were seeded in four-well plates (Shanghai Yuanye Biotechnology Co., Ltd, Shanghai, China) and exposed to different compounds at different concentrations. Cells were harvested to obtain protein solution by the RIPA Lysis Buffer (DeTie, NanJing, China). The proteins were separated on sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and then transferred to poly(vinylidene fluoride (PVDF) membranes (Shanghai Yuanye Biotechnology Co., Ltd, Shanghai, China). It incubated with the primary antibodies and corresponding HRP conjugated secondary antibodies (Servicebio, Wuhan, China). The electrochemiluminescence kit (Servicebio, Wuhan, China) was used to visualise the bands.

Molecular docking analysis
The protein structure of LSD1 was obtained from Protein Data Bank (PDB code: 2v1d). Sulphonamide derivatives L6~L8 were transferred as PDB files by ChemBio 3 D Ultra 14.0 (CambridgeSoft, Cambridge, MA). Water was removed from LSD1 protein by Pymol. Autodock (The Scripps Research Institute, San Diego, CA) was used to perform antogrid and autodock. The hydrogen-bond interaction between sulphonamide derivatives L6~L8 and amino acid residues of LSD1 was analyzed 26,27 .

Statistical methods
All data of biological experiments were expressed as the mean ± SD. Significant differences between different groups were analysed by GraphPad Prism version 6 (GraphPad Software, La Jolla, CA). Results were considered statistically significant at ÃÃ p< 0.01 verse control, ÃÃÃ p< 0.001 verse control and ÃÃÃÃ p< 0.0001 verse control.

Synthesis of sulphonamide derivatives
Although LSD1 inhibition has been a promising strategy to treat cancer, only few LSD1 inhibitors are currently in clinical trials. Discovery of novel and effective scaffolds to develop LSD1 inhibitors has an important clinical significance. Our aim was to synthesise and identify novel LSD1 inhibitors. Sulphonamide derivatives were synthesised as shown in Scheme 1. Commercially available quinolin-8-amine was treated with different benzenesulfonyl chloride to provide J1$J6 and 3,4,5-trimethoxyaniline was treated with quinoline-8-sulphonyl chloride to afford J7. Compounds K1$K3 were obtained by the alkylation of sulphonamide J7 with 1,2dibromoethane or 1,3-dibromopropane or 1,4-dibromobutane. Sulphonamide intermediates K1$K3 were treated with carbon disulphide and piperazine derivatives in the presence of trisodium phosphate dodecahydrate to obtain sulphonamide-quinolinedithiocarbamate hybrids L1$L8.

LSD1 inhibition and preliminary structure activity relationship
All sulphonamide derivatives in this work were examined in vitro for their inhibitory activity against LSD1, MAO-A, and MAO-B. 2-PCPA (tranylcypromine) and ORY-1001 as LSD1 inhibitors were chosen as reference agents 28 . The inhibitory results of sulphonamides J1$J7 were summarised in Table 1. Sulphonamide derivatives J1$J7 had very weak inhibition effects on MAO-A and MAO-B with IC 50 values of >120 lM. However, all sulphonamide derivatives J1$J7 displayed the moderate activity against LSD1 with IC 50 values ranging from 11.57 to 82.46 lM. Substituent groups on phenyl ring could affect the inhibitory activity against LSD1. Among all these sulphonamides, N-(3,4,5-trimethoxyphenyl)quinoline-8-sulphonamide J7 exhibited the best inhibitory activity, indicating that sulphonamide-quinoline might be a potential scaffold to design LSD1 inhibitors.
Because J7 showed the best inhibitory activity against LSD1, more sulphonamide analogues were designed and synthesised based on compound J7. The in vitro inhibitory activity results of sulphonamides K1$K3 and L1$L8 were listed in Table 2. With the exception of sulphonamide-quinoline-dithiocarbamate hybrids L1$L8, all these sulphonamide analogues bearing a dithiocarbamate fragment exhibit potently inhibitory activity with IC 50 values ranging from 0.06 to 4.92 lM. In comparison with activity results of sulphonamide intermediates K1$K3 without the dithiocarbamate unit, the inhibitory activity of sulphonamide-quinoline-dithiocarbamate hybrids L1$L8 against LSD1 improved obviously, indicating that the dithiocarbamate scaffold play a synergistic role on LSD1 activity. Especially, compound L8 showed the best activity against LSD1, with an IC 50 of 0.06 lM, which is 427 times higher than that of 2-PCPA. During the preliminary structure activity relationship studies, we found that the substituent on the piperazine ring was significant for the LSD1 inhibitory activity showing over 80-fold activity loss, when the hydroxyethyl group was replaced with the ethyl group (compound L8 versus L5). When the substituent on the piperazine ring was tert-butoxycarbonyl group and methyl group, compound L6 and compound L7 displayed the potent LSD1 inhibitory activity with IC 50 values of 0.23 and 0.73 lM, respectively. In addition, the length of carbon linker between the sulphonamide and the dithiocarbamate exhibited a significant role in their activities. With the reduction of the carbon tether length, a decrease of LSD1 inhibitory activity was observed (compound L8 versus L3, or compound L6 versus L1). Based on the enzyme inhibitory results, all sulphonamide derivatives L1$L8 potently inhibited LSD1 with IC 50 values ranging from 4.92 to 0.06 mM. Importantly, they had no significant effects on MAO-A and MAO-B with IC 50 values of >120 mM, indicating that these novel sulphonamide derivatives were selective LSD1 inhibitors. For the preliminary structure activity relationships, dithiocarbamate unit and substituent groups on piperazine of sulphonamide derivatives L1$L8 played significant roles on LSD1 inhibitory activity.  Sulphonamide derivative L8 selectively and reversibly inhibited LSD1 in enzyme-based assays

Sulphonamide derivative L8 exhibited good selectivity between cancer cells and normal cells
From the enzyme inhibitory results in Figure 3(A), sulphonamide derivative L8 weakly inhibited MAO-A and MAO-B with the inhibitory rates of only 10.33% and 9.33% at 100 nM, while it showed about 30.33%, 46.00%, and 68.00% of inhibition against LSD1 at 25, 50, and 100 nM, suggesting that compound L8 selectively inhibited LSD1 in a concentration-dependent. As shown in Figure  3(B), sulphonamide derivative L8 potently inhibited LSD1 in a time-dependent manner. In the dilution assay, the low concentration of sulphonamide derivative L8 by the dilution could result in the recovery of LSD1 inhibitory activity. Results in Figure 3(C) showed that sulphonamide derivative L8 was a reversible LSD1 inhibitor.

Sulphonamide derivative L8 potently inhibited LSD1 in cellbased assays
Before investigating the LSD1 inhibitory mechanisms at the cellular level, we first examined the LSD1 expression levels in four different bladder cancer cell lines (HTB5, HTB1, HTB3, and HT1376). From the results in Figure 4(A), HT1376 cells possessed the highest LSD1 expression levels, followed by HTB3, HTB1, and HTB5. In this work, we used the LSD1 knock-down cells and control cells to investigate the in vitro antiproliferative activity of LSD1 inhibitor L8 (Figure 4(B,C)). Compound L8 inhibited HTB3&shLSD1 cells and HT1376&shLSD1 cells with the IC 50 values of 3.07 and 4.36 lM. However, it significantly inhibited cell proliferation against HTB3&shControl cells and HT1376&shControl cells with the IC 50 values of 0.23 and 0.11 lM, respectively. The great activity discrepancy between LSD1 knock-down cells and control cells showed that the antiproliferative effects of sulphonamide derivative L8 against bladder cancer was dependent on its LSD1 inhibition. Meanwhile, several substrates of LSD1, including H3K4me1, H3K4me2, and H3K9me2 were also investigated to their expression levels with the compound treatment. As shown in Figure  4(D), sulphonamide derivative L8 increased the expression levels of H3K4me1, H3K4me2, and H3K9me2 against HT1376 cells. All these results showed that sulphonamide derivative L8 could inhibit LSD1 in the cellular level.  Molecular docking analysis of sulphonamide derivative L8 targeting LSD1 A molecular docking was carried out to predict the binding interaction between sulphonamide derivatives and LSD1. The protein structure of LSD1 was extracted from Protein Data Bank (PDB code: 2V1D, resolution: 3.1 Å) and was prepared by adding hydrogen atoms and removing flavin adenine dinucleotide. The molecular docking results in Figure 5 indicated that sulphonamide derivatives L6$L8 showed different binding modes through hydrogen bonds in the active pocket of LSD1. The dithiocarbamate group and amide group of sulphonamide derivative L6 formed two hydrogen bonds with LYS14 and GLU559, and the distance was 2.9 and 2.4 Å, respectively. In addition, compound L7 also had hydrogen interactions with LYS9 and ASN383 and the distance was 2.2 and 2.5 Å, respectively. Compared with compound L6 and L7, sulphonamide derivative L8 formed more hydrogen bonds with residues of LSD1. Compound L8 makes five hydrogen bonds with amino acid residues GLN417, GLU413, SER545, ARG688, and ARG526. The docking studies of sulphonamide derivatives were in accordance with the results of LSD1 inhibitory activity.

Conclusion
In conclusion, we have designed and synthesised a new class of sulphonamide-quinoline-dithiocarbamate hybrids as LSD1 inhibitors. Novel sulphonamide derivative L8 selectively and reversibly inhibits LSD1 in a concentration-dependent and time-dependent manner. Importantly, sulphonamide derivative L8 as a potent antiproliferative agent suppresses the proliferation against bladder cancer cells. To the best of our knowledge, there have been no literature reports regarding sulphonamide-quinoline-dithiocarbamate hybrids as LSD1 inhibitors against bladder cancer cells so far. All these findings provide an effective molecular skeleton for the discovery of LSD1 inhibitors, and sulphonamide based LSD1 inhibitors might be potentially anticancer drugs to treat bladder cancer.