A SEQUENTIAL DUAL CLEAVAGE OF THE ARYLSULFAMATE LINKER TO PROVIDE BOTH SULFAMATE AND PHENOL DERIVATIVES

Tyramine sulfamate was linked to the trityl chloride resin and this polymeric solid support used to introduce two levels of molecular diversity by formation of peptide bonds. A dual cleavage strategy next generated in a sequential way (without resin split) two different types of compounds (phenol and arylsulfamate derivatives), which are therapeutically attractive types of compounds. Here, we used tyramine as a general scaffold, but other arylsulfamate derivatives could be judiciously used to extend the nature of synthesized compounds.


Introduction
Sulfamate derivatives have been known for several years as artifi cial sweeteners (cyclamates) [1], but therapeutic applications are numerous and have signifi cantly broadened in recent years [2,3].They were fi rst known as anticonvulsants and the drug topimarate, discovered in 1980, is still used in clinical settings for the treatment of refractory epilepsy [4].Sulfamate derivatives were also found to be potent inhibitors of carbonic anhydrases (CAs) [5,6].This could explain the high oral availability of these drugs since they were shown to bind reversibly to CA II in red blood cells, which allows them to avoid degradation in the liver.This property could lead to new therapeutic approaches against cancer because CA IX and CA XII are highly expressed in tumours which need them to maintain pH and eliminate CO 2 .
Arylsulfamates have shown high potency as inhibitors of steroid sulfatase (STS) [7][8][9][10], an important therapeutic target for the treatment of hormone-sensitive diseases such as breast, endometrium and prostate cancers [11,12] in addition to acne and alopecia [13].STS inhibitors could also have potential in the treatment of Alzheimer's disease through an increase in the level of dehydroepiandrosteronesulfate, a substrate of STS in the brain [14].On the other hand, phenol derivatives are present in many biologically active molecules.For example, many natural antioxidants, such as fl avonoids and polyphenols, contain the phenol moiety.These compounds are known to exert a protective effect on cardiovascular health through the lowering of low-density lipoproteins.Many similar compounds were synthesized with the hope of optimizing such properties [15].Phenols are also very abundant in essential oils, contributing to their aroma and antimicrobial properties [16].Small phenolic molecules have shown potential as analgesics [17] and Presented at the XVIII-th International Conference "Physical Methods in Coordination and Supramolecular Chemistry" (Dedicated to the memory of the professor Constantin Turta and professor Mihail Revenco), 8-9 October 2015, Chisinau, Republic of Moldova _________________ some have been tested as non-steroidal anti-infl ammatory drugs [18] for their potential anti-rejection properties.Moreover, the phenolic group is essential in selective-estrogen receptor modulator drugs [19], which are used in the treatment of hormone-sensitive breast cancer and osteoporosis.Phenol derivatives are also known as inhibitors of 17β-hydroxysteroid dehydrogenases [20][21][22] and reversible inhibitors of STS [23,24] (although less potent than sulfamate derivatives).
The ability to generate libraries of phenol and arylsulfamate derivatives from the same resins, through rapid parallel synthesis, is thus of high interest for medicinal chemists.Herein, we describe a strategy (Figures 1 and 2) to generate two different kinds of compounds (sulfamate and phenol derivatives) from commercially available building blocks (amino acids and carboxylic acids) and the multidetachablesulfamate linker [25]. .Solid-phase chemical synthesis of libraries of phenol and arylsulfamate derivatives using the sequential approach of cleavage.See Figure 1, Table 1 and Table 2 for the building blocks used in the preparation of libraries and the representation of all library members.

Synthesis of libraries
The solid-phase strategy providing two types of diversifi ed molecules, the phenol and sulfamate derivatives, was represented in Figure 2. The primary amine of tyramine (1) was fi rst protected as a 9-fl uorenylmethoxycarbonyl (Fmoc) derivative, whereas the phenol group was next transformed by the sulfamoyl chloride in N,N-dimethyacetamide (DMA) used as base and solvent [26] to provide 2. This arylsulfamate was linked to a polystyrene solid support by a reaction with trityl chloride resin in presence of diisopropylethylamine (DIPEA), thus providing the resin 3. The mass increase suggested a quantitative yield for the coupling reaction.The characteristic band of Fmoc (1696 cm -1 ) and sulfamoyl (1350 and 1156 cm -1 ) groups were observed in FTIR spectra.A gel-phase 13 C NMR analysis of resin 3 showed all carbon signals associated with tyramine moiety.Finally, a micro-cleavage under acid conditions released the sulfamate 2. All these results confi rmed the presence of a tyramine residue linked on the trityl resin and consequently the formation of 3, the precursor of all library members.
The synthesis of libraries started by removing the Fmoc protecting group of 3 to generate the corresponding free primary amine, which was submitted to a coupling reaction using benzotriazol-1-yl-oxy-tris-pyrrolidino-phosphonium hexafl uorophosphate (PyBOP), 1-hydroxybenzotriazole (HOBt) and an amino acid.A series of fi ve amino acids (Figure 1) protected as N-Fmoc derivative was the key element used to introduce the fi rst level of molecular diversity.The formation of resins 4a-4e was supported by the presence in FTIR spectra of a new amide (NC=O) band in the range of 1656-1660 cm -1 .To introduce the second level of diversity, the fi ve resins 4a-4e were split in 5 groups (25 samples) and submitted to a sequence of two steps, the cleavage of Fmoc group and the coupling of fi ve carboxylic acids using PyBrOP and HOBt, thus producing resins 5aa-5ee.

Cleavage strategy (recovering the fi nal compounds)
The sequential dual cleavage consisted in performing, on the same resin sample, fi rst a partial release of a phenol derivative with a nucleophile, and next a cleavage of the remaining linked material with an acid to obtain an arylsulfamate derivative.This approach avoids the step of splitting the resin, which is time-consuming.
The type of cleavage we call «nucleophilic» here is not strictly speaking nucleophilic, but this type of reaction was studied in detail by Spillane et al. [27][28][29][30].The mechanism is a two-step base-catalysed E1cb-type mechanism that probably involves a bimolecular complex between a base and the sulfamate NH as intermediate, which is then attacked by a nucleophile to release the phenol.In this case, we carried out an incomplete nucleophilic cleavage at room temperature with 30% of diethylamine (DEA), which released about 50% of the linked material as the phenol derivatives 6-30 (Table 1).

Table 1
Representation of all library members (phenol derivatives 6-30).W: Weight of released compound; CY: crude yield; P: purity determined by quantitative 1

H NMR. (*) R 1 : The residue of amino acids used as building blocks. (**) R 2 : The residue of carboxylic acids used as building blocks.
The acidic cleavage probably proceeds through protonation of the sulfamate NH and subsequent formation of a trityl anion on the resin (which takes on an intense red color).In the fi rst assays, the acidic cleavage used to produce the arylsulfamate derivative was done using a 5% trifl uoroacetic acid (TFA) solution in CH 2 Cl 2 , which gave better yields with shorter reaction times, but lower purity.Since we intended our solid-phase synthesis products for screening purposes, and thus wanted rapid production with no purifi cation step, milder acid conditions (30% hexafl uoroisopropanol (HFIP) in CH 2 Cl 2 ) were preferred to generate the arylsulfamate derivatives 31-55 (Table 2).
After we generated two libraries of phenol derivatives (Table 1) and arylsulfamate derivatives (Table 2), all the library members were analysed by thin-layer chromatography and showed a good homogeneity (mostly one spot).A sampling of both libraries (10 members by library) was performed and the compounds tested by NMR analysis.We chose quantitative NMR over HPLC for purity assessment because it allows the detection of the real quantity of the D. Fournier et al. / Chem.J. Mold. 2015, 10(2), 68-76 desired product in a given mass even if an impurity is insoluble or invisible to NMR.On the contrary, since purity assessment with HPLC depends on what is visible to the detector, insoluble or detector-invisible material is not taken into account, and thus artifi cially high purity values are read.
It should be noted that no purifi cation steps whatsoever are used in the procedure described above.Nonetheless, we think the average purity of the products (phenol derivatives: 39 to 82%, average 60%; arylsulfamate derivatives: 38 to 66%, average 53%) could be increased by optimizing the coupling steps.Moreover, we noticed that some resins developed a red-brownish color after an amino-acid coupling cycle, and this color could not be washed away.After cleavage, the products showed low purity.Impurities were sometimes visible on 1 H NMR spectra as direct capping (resulting from incomplete amino-acid coupling) or phosphorus derivatives (from PyBOP and PyBrOP coupling agents).These results do not seem to be systematic since different amino acids are affected in different trials.They were not observed in two compounds previously generated as models, as in those cases average purity of the crude product was 80% and excellent purity and yields were obtained after chromatography.The lower purities could have many causes.For example, it is known that PyBOP and PyBrOP produce deleterious esters when left in solution in DMF for long periods (more than one hour).These esters probably form a large part of the impurities and could be avoided by shortening coupling cycles (one hour each), by choosing other coupling reagents, or by using DMA instead of DMF (PyBOP and PyBrOP are stable for several days in DMA).The low purity could also result from cross-contamination when using the synthesizer, which was not used in the preparation of the two model compounds.However, products with higher purity could be obtained by performing a silica gel fi ltration or a fl ash chromatography [31].Libraries of steroidal sulfamates have also been successfully generated in high yields and purities [31][32][33].

Conclusions
The multidetachablesulfamate linker used herein allowed the preparation of both phenol and arylsulfamate derivatives, which are therapeutically attractive types of compounds.The loading step on trityl chloride resin is quantitative and the peptide coupling reactions are compatible with this linker.The dual cleavage strategy allowed us to generate two different types of compounds in a sequential way (without resin split).Here, we used the Fmoctyramine sulfamate (2) as a general scaffold to introduce two levels of diversifi cation, but other arylsulfamates could be judiciously used to extend the nature of the synthesized compounds.The sulfamate linker thus represents a valuable addition to the chemical tools available to the medicinal and organic chemists.

General remarks
9-Fluorenylmethyl succinimidyl carbonate (FmocOSu) and Fmoc-protected amino-acids were purchased from Advanced ChemTech, (Louisville, KY, USA).PyBOP, PyBrOP, anhydrous DMF and trityl resin were purchased from Novabiochem (EMD Biosciences, San Diego, CA, USA).Other reagents were purchased from Aldrich (Milwaukee, WI, USA).The sulfamoyl chloride (moisture sensitive) was prepared from chlorosulfonyl isocyanate and concentrated HCl according to a known procedure [34].All reagents were used as provided.A Jouan RC1010 SpeedVac apparatus (Winchester, VA, USA) was used for the solvent evaporation of the fi nal library compounds.FTIR spectra were obtained on a Perkin-Elmer 1600 spectrophotometer (Norwalk, CT, USA). 13C NMR spectra were recorded at 75.5 MHz on a Bruker AC/F 300 spectrometer (Billerica, MA, USA). 1 H NMR spectra with and without internal reference were recorded at 400 MHz on a Bruker Avance 400 Spectrometer (Billerica, MA, USA).The NMR purity of the cleaved products was determined using the external reference method.The reference compound (1,2,4-triazole) was dissolved in DMSO-d 6 and placed in a WGS-5BL coaxial insert (WILMAD, Buena, NJ, USA).

Library synthesis: Introducing the fi rst level of molecular diversity
A 96-well Tefl on reaction vessel from Advanced ChemTechLabtech Manual Organic Synthesizer-Platform IV was used for the solid-phase synthesis.Small-volume peptide fl asks or PD-10 columns (Amersham Biosciences, Upsala, Sweden) fi tted with a three-way stopcock (Bio-Rad, Hercules, CA, USA) could alternatively be used.In 25 wells of a 96-well reaction block loaded with resin 3 (25 x 120 mg) was added a solution of 20% piperidine in CH 2 Cl 2 (1.5 mL) and the mixture was stirred at room temperature to hydrolyze the Fmoc.After 1 h, the resins were fi ltered to remove the solvent, washed with CH 2 Cl 2 (5 x 2 mL) and dried under a vacuum for 2 h.Five stock solutions, each containing PyBOP (450 mg, 0.870 mmol) and HOBt (120 mg, 0.870 mmol), as well as one of the Fmoc-protected amino acids from the L series (alanine, valine, leucine, isoleucine or phenylalanine) (0.870 mmol), were prepared in DMF (7.5 mL) and reacted 2 min with DIPEA (0.3 mL, 1.74 mmol).Next, one of the stock solutions was added (in equal proportions of 1.5 mL) in D. Fournier et al. / Chem.J. Mold. 2015, 10(2), 68-76 each of the 25 reaction vessels (fi ve wells for each amino acid) containing the resin with the tyramine.The resins were stirred under argon for 3 h, then fi ltered, washed with DMF (3 x 2 mL) and CH 2 Cl 2 (5 x 2 mL), and dried overnight under a vacuum to afford fi ve groups (5 x 5) of resins 4a-4e.FTIR (KBr matrix) υ, cm -1 : 1708-1709 (C=O, Fmoc group), 1656-1662 (C=O, amide) and 1350 and 1156 (sulfonamide).

Cleavage strategy providing phenol derivatives (Nucleophilic cleavage)
To each of the 25 resins 5aa-5ee in their reaction vessels was added a solution of 30% diethylamine (DEA) in THF (1 mL) and the resins were stirred under argon.After 24 h at room temperature, 1 mL of 30% DEA/THF solution was added and the resins were stirred for 24 additional hours.The resins were then fi ltered under vacuum, washed with 30% DEA in THF (2 x 1.5 mL), and the fi ltrates were collected in pre-weighted glass tubes.The solvent was evaporated in a SpeedVac apparatus, THF (3 mL) was then added to each tube and the solutions were evaporated again (2x).The procedure was repeated with Et 2 O.The crude products were dried for 48 h under vacuum pump to afford phenol derivatives 6-30.A sampling of 10 compounds from the 25 library members was characterized by 1 H NMR and MS.Purity was also assessed by quantitative 1 H NMR (Table 1).

Cleavage strategy providing sulfamate derivatives (Acidic cleavage)
To each of the 25 resins 5aa-5ee previously submitted to nucleophilic cleavage was added a solution of 30% hexafl uoroisopropanol (HFIP) in CH 2 Cl 2 (1.5 mL) and the resins were stirred under argon.After 6 h at room temperature, the resins were fi ltered under vacuum, washed with 30% HFIP in CH 2 Cl 2 (2 x 1 mL) and THF (2 x 1 mL), and the fi ltrate was collected in pre-weighted tubes.The solvent was evaporated in a SpeedVac apparatus, THF (2 x 3 mL) was then added to each tube and the solutions were evaporated again.The procedure was repeated with Et 2 O.The crude products were dried 48 h under vacuum pump to afford sulfamate derivatives 31-55.A sampling of 10 compounds from the 25 library members was characterized by 1 H NMR and MS.Purity was also assessed by quantitative 1 H NMR (Table 2).

Figure 2
Figure2.Solid-phase chemical synthesis of libraries of phenol and arylsulfamate derivatives using the sequential approach of cleavage.See Figure1, Table1and Table2for the building blocks used in the preparation of libraries and the representation of all library members.