Synthesis of Novel Saccharin Derivatives

The synthesis of saccharin (1,2-benzisothiazol-3-one-1,1-dioxide) derivatives substituted on the benzene ring has seen limited development despite the longevity of this compound’s use as an artificial sweetener. This type of saccharin derivative would however present attractive properties for the development of new bioactive, drug-like small molecule compounds. Here we report the derivatisation of the benzene ring of saccharin using Cu(I)-catalyzed azide alkyne cycloaddition (CuAAC) to synthesise a diverse library of novel saccharin-1,2,3-triazole conjugates. All library compounds retain the capability for interactions with biomolecules via the unmodified sulfonamide and lactam groups of the parent saccharin core heterocycle. The compounds also encompass alternate orientations of the 1,2,3-triazole heterocycle, thus further adding diversity to the potential hydrogen bonding interactions of these compounds with biomolecules of therapeutic interest. Our findings demonstrate that specifically functionalized derivatives of saccharin may be prepared from either saccharin azide or saccharin alkyne building blocks in high yield using CuAAC.


Introduction
Several different synthetic routes have been applied to construct the heterocyclic compound 1,2-benzisothiazol-3-one-1,1-dioxide, commonly known as saccharin (1, Figure 1), a synthetic calorie-free sugar substitute [1,2].Alkylation methods to prepare Nand O-substituted (carbonyl oxygen) derivatives of 1 are well established and these have provided researchers with the straightforward synthesis of novel and potentially bioactive molecules [3][4][5][6].In addition to Nand O-alkylation, compounds derivatised on the benzene moiety of 1 are particularly desirable, as these compounds retain both the cyclic sulfonamide and lactam groups which can participate in strong noncovalent interactions with biomolecular targets such as enzymes.The synthesis of the latter derivatives is reliant on the application of synthetic acumen to introduce a latent handle on the benzene moiety of 1 for subsequent derivatisation; this modification's methodology is however much less developed than the straightforward alkylation of 1.
Compound 1 is a weak acid, the measured pK a of the cyclic sulfonamide NH hydrogen is 1.3 [7], and the anion of 1 readily forms complexes with metal(II) cations [8].Klebe and Supuran first demonstrated that the metal binding characteristics of 1 could be utilised to target metalloenzyme inhibition, specifically the zinc metalloenzyme carbonic anhydrase (CA) [9].We further elaborated this finding through the design and synthesis of a small library of compounds based on 1 that retain the cyclic sulfonamide core (allowing for zinc binding) with the addition of a 'tail' group to the benzene ring of 1 [10].The 'tail' approach to develop CA inhibitors is well established, with the physicochemical properties of the 'tail' group as a key driver to CA inhibitor isozyme selectivity across the CA family, of which there are 12 catalytically active isozymes in humans [11].Cu(I)-catalyzed azide alkyne cycloaddition (CuAAC or click chemistry) is the reaction between an azide (R-N 3 ) and an alkyne (R-C≡CH) to form a 1,2,3-triazole [12].We demonstrated that CuAAC is a robust and versatile approach to link a selected tail group to a CA zinc binding pharmacophore [13][14][15][16][17][18].To achieve the synthesis of compounds derivatised on the benzene ring of 1 via CuAAC, we prepared two novel azides, 6-azidosaccharin 2 and t-butyl protected 6-azidosaccharin 3 (Figure 1) [10].One compound from this initial study, the 1,2,3-triazole glycoconjugate 4, exhibited remarkable selectivity for CA IX, a CA isozyme that underpins the survival of hypoxic tumour cells (Figure 1) [10].In the protein X-ray crystal structure of 4 in complex with a CA IX mimic protein, as anticipated, 4 was found coordinated to the active site zinc via the cyclic sulfonamide anion, while the tail moiety of 4 contributed to further interactions with the outer rim of the CA IX-mimic active site residues [19].Here we report further development of the derivatisation of the benzene ring of scaffold 1.Specifically, we have substantially expanded the scope of CuAAC with the design and synthesis of the novel alkyne 5 and bis-alkyne 6, complementary building blocks to the t-butyl protected 6-azidosaccharin 3 (Figure 1).The building blocks 3, 5, and 6 enable access to triazoles with a reversed arrangement of substituents, i.e., with the saccharin fragment as either the 4-substituent or the 1-substituent of the 1,2,3-triazole formed by CuAAC.This is important as the 1,4-disubstituted 1,2,3-triazole is a bioisostere of a Z-amide bond, hence the positioning of the H-bond donor and H-bond acceptor is also reversed with the use of complementary saccharin building blocks [20].Derivatisation of 5 with a diverse panel of azides (a-g), derivatization of 6 with glycosyl azide (d), and the further derivatization of 3 with alkyne partners (h-l) is described.Partner azides (a-g) and alkynes (h-l) were selected to encompass variable physicochemical properties (Figure 2).All compounds retain the sulfonamide and lactam moieties of 1.
Molecules 2017, 22, 516 2 of 19 (R-C≡CH) to form a 1,2,3-triazole [12].We demonstrated that CuAAC is a robust and versatile approach to link a selected tail group to a CA zinc binding pharmacophore [13][14][15][16][17][18].To achieve the synthesis of compounds derivatised on the benzene ring of 1 via CuAAC, we prepared two novel azides, 6-azidosaccharin 2 and t-butyl protected 6-azidosaccharin 3 (Figure 1) [10].One compound from this initial study, the 1,2,3-triazole glycoconjugate 4, exhibited remarkable selectivity for CA IX, a CA isozyme that underpins the survival of hypoxic tumour cells (Figure 1) [10].In the protein X-ray crystal structure of 4 in complex with a CA IX mimic protein, as anticipated, 4 was found coordinated to the active site zinc via the cyclic sulfonamide anion, while the tail moiety of 4 contributed to further interactions with the outer rim of the CA IX-mimic active site residues [19].Here we report further development of the derivatisation of the benzene ring of scaffold 1.Specifically, we have substantially expanded the scope of CuAAC with the design and synthesis of the novel alkyne 5 and bis-alkyne 6, complementary building blocks to the t-butyl protected 6-azidosaccharin 3 (Figure 1).The building blocks 3, 5, and 6 enable access to triazoles with a reversed arrangement of substituents, i.e., with the saccharin fragment as either the 4-substituent or the 1-substituent of the 1,2,3-triazole formed by CuAAC.This is important as the 1,4-disubstituted 1,2,3-triazole is a bioisostere of a Z-amide bond, hence the positioning of the H-bond donor and H-bond acceptor is also reversed with the use of complementary saccharin building blocks [20].Derivatisation of 5 with a diverse panel of azides (a-g), derivatization of 6 with glycosyl azide (d), and the further derivatization of 3 with alkyne partners (h-l) is described.Partner azides (a-g) and alkynes (h-l) were selected to encompass variable physicochemical properties (Figure 2).All compounds retain the sulfonamide and lactam moieties of 1.

Results and Discussion
In our previous study using 6-azidosaccharins, we established that it was preferable to undertake CuAAC reactions with N-t-butyl protected 6-azidosaccharin 3, followed by removal of the t-butyl protecting group, over the more direct synthetic approach of using 6-azidosaccharin 2 [10].Specifically, the ease of synthesis and isolated yield substantially improved when using 3 owing to the simplified reaction workup and product purification.We attributed these advantages to the (R-C≡CH) to form a 1,2,3-triazole [12].We demonstrated that CuAAC is a robust and versatile approach to link a selected tail group to a CA zinc binding pharmacophore [13][14][15][16][17][18].To achieve the synthesis of compounds derivatised on the benzene ring of 1 via CuAAC, we prepared two novel azides, 6-azidosaccharin 2 and t-butyl protected 6-azidosaccharin 3 (Figure 1) [10].One compound from this initial study, the 1,2,3-triazole glycoconjugate 4, exhibited remarkable selectivity for CA IX, a CA isozyme that underpins the survival of hypoxic tumour cells (Figure 1) [10].In the protein X-ray crystal structure of 4 in complex with a CA IX mimic protein, as anticipated, 4 was found coordinated to the active site zinc via the cyclic sulfonamide anion, while the tail moiety of 4 contributed to further interactions with the outer rim of the CA IX-mimic active site residues [19].Here we report further development of the derivatisation of the benzene ring of scaffold 1.Specifically, we have substantially expanded the scope of CuAAC with the design and synthesis of the novel alkyne 5 and bis-alkyne 6, complementary building blocks to the t-butyl protected 6-azidosaccharin 3 (Figure 1).The building blocks 3, 5, and 6 enable access to triazoles with a reversed arrangement of substituents, i.e., with the saccharin fragment as either the 4-substituent or the 1-substituent of the 1,2,3-triazole formed by CuAAC.This is important as the 1,4-disubstituted 1,2,3-triazole is a bioisostere of a Z-amide bond, hence the positioning of the H-bond donor and H-bond acceptor is also reversed with the use of complementary saccharin building blocks [20].Derivatisation of 5 with a diverse panel of azides (a-g), derivatization of 6 with glycosyl azide (d), and the further derivatization of 3 with alkyne partners (h-l) is described.Partner azides (a-g) and alkynes (h-l) were selected to encompass variable physicochemical properties (Figure 2).All compounds retain the sulfonamide and lactam moieties of 1.

Results and Discussion
In our previous study using 6-azidosaccharins, we established that it was preferable to undertake CuAAC reactions with N-t-butyl protected 6-azidosaccharin 3, followed by removal of the t-butyl protecting group, over the more direct synthetic approach of using 6-azidosaccharin 2 [10].Specifically, the ease of synthesis and isolated yield substantially improved when using 3 owing to the simplified reaction workup and product purification.We attributed these advantages to the

Results and Discussion
In our previous study using 6-azidosaccharins, we established that it was preferable to undertake CuAAC reactions with N-t-butyl protected 6-azidosaccharin 3, followed by removal of the t-butyl protecting group, over the more direct synthetic approach of using 6-azidosaccharin 2 [10].Specifically, the ease of synthesis and isolated yield substantially improved when using 3 owing to the simplified reaction workup and product purification.We attributed these advantages to the blockade of metal complex formation between 3 and Cu 2+ (from the CuSO 4 used for CuAAC) by the N-t-butyl protecting group.Building on this experience we selected N-t-butyl-6-ethynyl-1,2-benzisothiazole-3-one-1,1-dioxide (N-t-butyl-protected 6-ethynylsaccharin 5) as the target building block for CuAAC in the present study (Figure 1).Additionally, N-t-butyl-6-N,N-bis(prop-2-yn-1-yl)amino-1,2-benzisothiazole-3-one-1,1-dioxide 6 was selected as the bis-alkyne rather than the unprotected form without the t-butyl protecting group (Figure 1).
The synthetic route to alkynes 5 and 6 and the earlier reported azide 3 have a common precursor, N-t-butyl-6-aminosaccharin 7 [10] (Scheme 1).Iodination of 7 with sodium nitrite and potassium iodide was achieved following a literature procedure used for iodination of similar aromatic and heterocyclic compounds to give the N-protected 6-iodosaccharin 8 in an 83% yield [21].The Sonogashira cross-coupling reaction between 8 and ethynyltrimethylsilane l generated the trimethylsilyl protected alkyne 9 in high yield.Removal of the trimethylsilyl group of 9 under standard conditions of K 2 CO 3 in methanol proceeded, however these conditions additionally caused ring opening at the C-N bond of the heterocycle.The successful removal of this silyl group was instead achieved using mild acidic reaction conditions (tetrabutylammonium fluoride (TBAF)/1% AcOH in tetrahydrofuran (THF)) to afford the target alkyne 5 in an almost quantitative yield.Next, to install the two terminal alkyne groups of 6, the amino saccharin compound 7 [10] was treated with propargyl bromide (2.2 equivalents (equiv)) in the presence of Cs 2 CO 3 .
Molecules 2017, 22, 516 3 of 19 blockade of metal complex formation between 3 and Cu 2+ (from the CuSO4 used for CuAAC) by the N-t-butyl protecting group.Building on this experience we selected N-t-butyl-6-ethynyl-1,2benzisothiazole-3-one-1,1-dioxide (N-t-butyl-protected 6-ethynylsaccharin 5) as the target building block for CuAAC in the present study (Figure 1).Additionally, N-t-butyl-6-N,N-bis(prop-2-yn-1yl)amino-1,2-benzisothiazole-3-one-1,1-dioxide 6 was selected as the bis-alkyne rather than the unprotected form without the t-butyl protecting group (Figure 1).The synthetic route to alkynes 5 and 6 and the earlier reported azide 3 have a common precursor, N-t-butyl-6-aminosaccharin 7 [10] (Scheme 1).Iodination of 7 with sodium nitrite and potassium iodide was achieved following a literature procedure used for iodination of similar aromatic and heterocyclic compounds to give the N-protected 6-iodosaccharin 8 in an 83% yield [21].The Sonogashira cross-coupling reaction between 8 and ethynyltrimethylsilane l generated the trimethylsilyl protected alkyne 9 in high yield.Removal of the trimethylsilyl group of 9 under standard conditions of K2CO3 in methanol proceeded, however these conditions additionally caused ring opening at the C-N bond of the heterocycle.The successful removal of this silyl group was instead achieved using mild acidic reaction conditions (tetrabutylammonium fluoride (TBAF)/1% AcOH in tetrahydrofuran (THF)) to afford the target alkyne 5 in an almost quantitative yield.Next, to install the two terminal alkyne groups of 6, the amino saccharin compound 7 [10] was treated with propargyl bromide (2.2 equivalents (equiv)) in the presence of Cs2CO3.
The reaction of 5 with azidobenzene a [22], benzylazide b [23], and PEG azide c [24] was carried out under typical CuAAC conditions (0.2 equiv of CuSO 4 •5H 2 O and 0.4 equiv of sodium ascorbate, t-BuOH:water 1:1, 45 • C) to generate triazoles 24-26, respectively, in 74%-97% yield (Scheme 2).Subsequent removal of the N-t-butyl group of 24-26 was achieved following reflux in trifluoracetic acid (TFA) for 18 h to furnish the target derivatives of 1, triazoles 10-12, respectively, in high yield (85%-99%).The reaction of alkyne 5 and 2,3,4,6-tetra-O-acetyl-β-D-glucopyranosyl azide d [25] at 50 • C produced triazole 27 in high yield (90%).As we had concerns that the harsh basic conditions promote opening of the saccharin heterocyclic ring, the acetyl groups of 27 were hydrolysed using HCl in MeOH instead of the more usual Zemplén conditions of methoxide in MeOH [26], to afford 28 in a 94% yield, however a lengthy reaction time of 90 h was required (Scheme 2).The N-t-butyl group of 28 was removed with refluxing in TFA for 18 h to yield the target glycoconjugate 13 in high yield.The bis-triazole saccharin glycoconjugate 33 was prepared from bis-alkyne 6 and per-O-acetylated glucosyl azide d [25] in high yield (87%) (Scheme 3).Deacetylation of 33 under acidic conditions (HCl in MeOH, 90 h) gave the free sugar 34, and cleavage of the N-t-butyl group of 34 with TFA furnished the target bis-triazole saccharin glycoconjugate compound 17 (Scheme 3).As the acidic conditions to remove the acetyl groups of 27 and 33 required a prolonged reaction time (90 h), this prompted us to investigate an alternate route to synthesise the glycoconjugates 14 and 15.This route employed free glycosyl azides e and f, instead of the corresponding per-O-acetylated glycosyl azides, to eliminate the need for deprotection of the sugar hydroxyl groups following CuAAC, thus removing the dependence on this synthetic bottleneck [25,[27][28][29].The reaction of alkyne 5 and free glycosyl azides e and f [25,[27][28][29] via CuAAC proceeded smoothly to form intermediates 29 and 30 (Scheme 2).Subsequent removal of the N-t-butyl protecting groups of these intermediates by overnight refluxing in TFA afforded target glycoconjugates 14 and 15, respectively.Next, CuAAC of azidotrimethylsilane (TMSN 3 ) g with alkyne 5 gave the monosubstituted triazole 31 as an inseparable mixture of thermodynamically stable tautomers [30].Cleavage of the N-t-butyl group of 31 using TFA furnished a mixture of 32a and 32b, where 1 H nuclear magnetic resonance (NMR) and high resolution mass spectrometry (HRMS) analysis confirmed triazole N-alkylation, with the t-butyl group on either the N-1 (32a) or N-2 (32b) of the triazole ring.Although the reaction of 5 with azidotrimethylsilane g did not yield the intended target triazole 16, both 32a and 32b are novel compounds that retain the cyclic sulfonamide functional group of 1.
(85%-99%).The reaction of alkyne 5 and 2,3,4,6-tetra-O-acetyl-β-D-glucopyranosyl azide d′ [25] at 50 °C produced triazole 27 in high yield (90%).As we had concerns that the harsh basic conditions promote opening of the saccharin heterocyclic ring, the acetyl groups of 27 were hydrolysed using HCl in MeOH instead of the more usual Zemplén conditions of methoxide in MeOH [26], to afford 28 in a 94% yield, however a lengthy reaction time of 90 h was required (Scheme 2).The N-t-butyl group of 28 was removed with refluxing in TFA for 18 h to yield the target glycoconjugate 13 in high yield.The bis-triazole saccharin glycoconjugate 33 was prepared from bis-alkyne 6 and per-O-acetylated glucosyl azide d′ [25] in high yield (87%) (Scheme 3).Deacetylation of 33 under acidic conditions (HCl in MeOH, 90 h) gave the free sugar 34, and cleavage of the N-t-butyl group of 34 with TFA furnished the target bis-triazole saccharin glycoconjugate compound 17 (Scheme 3).As the acidic conditions to remove the acetyl groups of 27 and 33 required a prolonged reaction time (90 h), this prompted us to investigate an alternate route to synthesise the glycoconjugates 14 and 15.This route employed free glycosyl azides e and f, instead of the corresponding per-O-acetylated glycosyl azides, to eliminate the need for deprotection of the sugar hydroxyl groups following CuAAC, thus removing the dependence on this synthetic bottleneck [25,[27][28][29].The reaction of alkyne 5 and free glycosyl azides e and f [25,[27][28][29] via CuAAC proceeded smoothly to form intermediates 29 and 30 (Scheme 2).Subsequent removal of the N-t-butyl protecting groups of these intermediates by overnight refluxing in TFA afforded target glycoconjugates 14 and 15, respectively.Next, CuAAC of azidotrimethylsilane (TMSN3) g with alkyne 5 gave the monosubstituted triazole 31 as an inseparable mixture of thermodynamically stable tautomers [30].Cleavage of the N-t-butyl group of 31 using TFA furnished a mixture of 32a and 32b, where 1 H nuclear magnetic resonance (NMR) and high resolution mass spectrometry (HRMS) analysis confirmed triazole N-alkylation, with the t-butyl group on either the N-1 (32a) or N-2 (32b) of the triazole ring.Although the reaction of 5 with azidotrimethylsilane g did not yield the intended target triazole 16, both 32a and 32b are novel compounds that retain the cyclic sulfonamide functional group of 1.The target compounds prepared from the reaction of azidosaccharin 3 [10] with alkynes h-l have similar physicochemical diversity to azides a-g reacted with the ethynylsaccharin 5 (Figure 3).We have previously reported the synthesis of the phenyl derivative 18 while all other compounds are novel [10].Reaction of azide 3 with 3-phenyl-1-propyne (benzyl alkyne) i and PEG alkyne j [31] under standard CuAAC conditions gave triazoles 35 and 36, respectively (Scheme 4).Acid mediated cleavage of the t-butyl protecting group in refluxing TFA furnished the target saccharin compounds 19 and 20, respectively.CuAAC of ethynyltrimethylsilane (TMSC≡CH) l and azide 3 [10] gave 37, a 1-substituted 1,2,3-triazole.Consistent with the outcome of deprotection of the related 1-substituted triazole 31, treatment of 37 with TFA removed the N-t-butyl group from the sulfonamide nitrogen, but furnished the alternate N-alkylation product 38, where the t-butyl group is on N-3 of the 1,2,3-triazole instead of the desired target compound 16 (Scheme 4).The target compounds prepared from the reaction of azidosaccharin 3 [10] with alkynes h-l have similar physicochemical diversity to azides a-g reacted with the ethynylsaccharin 5 (Figure 3).We have previously reported the synthesis of the phenyl derivative 18 while all other compounds are novel [10].Reaction of azide 3 with 3-phenyl-1-propyne (benzyl alkyne) i and PEG alkyne j [31] under standard CuAAC conditions gave triazoles 35 and 36, respectively (Scheme 4).Acid mediated cleavage of the t-butyl protecting group in refluxing TFA furnished the target saccharin compounds 19 and 20, respectively.CuAAC of ethynyltrimethylsilane (TMSC≡CH) l and azide 3 [10] gave 37, a 1-substituted 1,2,3-triazole.Consistent with the outcome of deprotection of the related 1-substituted triazole 31, treatment of 37 with TFA removed the N-t-butyl group from the sulfonamide nitrogen, but furnished the alternate N-alkylation product 38, where the t-butyl group is on N-3 of the 1,2,3triazole instead of the desired target compound 16 (Scheme 4).The target compounds prepared from the reaction of azidosaccharin 3 [10] with alkynes h-l have similar physicochemical diversity to azides a-g reacted with the ethynylsaccharin 5 (Figure 3).We have previously reported the synthesis of the phenyl derivative 18 while all other compounds are novel [10].Reaction of azide 3 with 3-phenyl-1-propyne (benzyl alkyne) i and PEG alkyne j [31] under standard CuAAC conditions gave triazoles 35 and 36, respectively (Scheme 4).Acid mediated cleavage of the t-butyl protecting group in refluxing TFA furnished the target saccharin compounds 19 and 20, respectively.CuAAC of ethynyltrimethylsilane (TMSC≡CH) l and azide 3 [10] gave 37, a 1-substituted 1,2,3-triazole.Consistent with the outcome of deprotection of the related 1-substituted triazole 31, treatment of 37 with TFA removed the N-t-butyl group from the sulfonamide nitrogen, but furnished the alternate N-alkylation product 38, where the t-butyl group is on N-3 of the 1,2,3triazole instead of the desired target compound 16 (Scheme 4).Scheme 4. Synthesis of saccharin derivatives from N-t-butyl-protected 6-azidosaccharin 3. Reagents and conditions: (i) 3 (1 equiv), alkyne (1 equiv), sodium ascorbate (0.4 equiv), CuSO4•5H2O (0.2 equiv), 1:1 t-BuOH:water, 45 °C-50 °C, 2 h-overnight; (ii) 3 (1 equiv), TMSC≡CH (2 equiv), sodium ascorbate (0.4 equiv), CuSO4•5H2O (0.2 equiv), 1:1 t-BuOH:water, 45 °C, overnight; (iii) 8% HCl in MeOH, r.t., 90 h; (iv) mCPBA, CH2Cl2, 0 °C to r.t.; (v) TFA, reflux, 18 h.The phenyl derivative 18 as previously reported [10].

General Chemistry
All starting materials and reagents were purchased from commercial suppliers.All solvents were available commercially dried or dried prior to use.Reaction progress was monitored by thin layer chromatography (TLC) using silica gel-60 F254 plates (Merck Millipore, Darmstadt, Germany) with detection by short wave ultraviolet (UV) fluorescence (λ = 254 nm) and staining with 5% w/v dodecamolybdophosphoric acid in ethanol or vanillin staining (5 g of vanillin in a mixture of EtOH:H 2 O:H 2 SO 4 = 85:10:2.75)with subsequent heating.Silica gel flash chromatography was performed using silica gel 60 Å (230-400 mesh) (Merck Millipore, Darmstadt, Germany).NMR ( 1 H, 13 C, 19 F, gradient correlation spectroscopy (gCOSY), and heteronuclear single quantum coherence (HSQC) spectra were recorded on either a 400 or 500 MHz spectrometer at 30 • C. 1 H-NMR spectra were obtained at 500 MHz and were referenced to the residual solvent peak (CDCl 3 δ 7.26 ppm, dimethylsulfoxide (DMSO)-d 6 δ 2.50 ppm). 13C-NMR spectra were recorded at 125 MHz and were referenced to the internal solvent (CDCl 3 δ 77.0 ppm, DMSO-d 6 δ 39.5 ppm). 19F-NMR spectra were recorded at 376 MHz.Multiplicity is indicated as follows: s (singlet); d (doublet); t (triplet); m (multiplet); dd (doublet of doublet); ddd (doublet of doublet of doublet); b (broad).Coupling constants are reported in hertz (Hz).Melting points are uncorrected.Low and high resolution mass spectra (MS) were recorded using electrospray ionization (ESI) in positive ion and/or negative ion modes as stated.All MS analysis samples were prepared as solutions in methanol.The purity of all compounds was ≥95% as determined by HPLC with UV. 1 H-, 13 C-, and 19 F-NMR spectra of all novel compounds are provided in the supporting information.

General Procedure 1-CuAAC
A mixture of azide (1.0 equiv) and alkyne (1.0 equiv) was prepared in t-butyl alcohol/H 2 O (1:1, 6-10 mL).To the mixture was added a solution of sodium ascorbate (0.4 equiv) in water (0.25 mL) followed by a solution of CuSO 4 .5H 2 O (0.2 equiv) in water (0.25 mL).The resulting suspension was stirred vigorously at the temperature and time indicated below.The solvent was removed in vacuo and the residue was purified by column chromatography on silica gel using the eluent conditions described below.
1 H-NMR and HRMS confirmed the formation of 38.Finally, the reaction of saccharin azide 3 and propargyl 2,3,4,6-tetra-O-acetyl-thio-β-Dglucopyranoside k′ [14] using CuAAC gave glycoconjugate 39 in high yield.Acidic cleavage of the acetyl groups of 39 (HCl in MeOH, 90 h) gave the free sugar derivative 40.Given the long 90 h reaction time, triazole 40 was also prepared directly from 3 utilising propargyl thio-β-D-glucopyranoside k [32], as described for 14 and 15.Oxidation of 39 and 40 with m-chloroperbenzoic acid (mCPBA) gave sulfones 41 and 42 in high yields, respectively.Compound 42 was also prepared by the deacetylation of 41 under acidic conditions, and this demonstrated the versatility of protecting group manipulation in the presence of the saccharin core scaffold.The N-t-butyl protecting group of 40 and 42 was removed by refluxing in TFA for 18 h to afford 21 and 22, respectively (Scheme 4).