Glucopyranosylidene-Spiro-Thiazolinones: Synthetic Studies and Determination of Absolute Configuration by TDDFT-ECD Calculations

Reactions of O-peracylated C-(1-bromo-β-d-glucopyranosyl)formamides with thioamides furnished the corresponding glucopyranosylidene-spiro-thiazolin-4-one. While O-debenzoylations under a variety of conditions resulted in decomposition, during O-deacetylations the addition of MeOH to the thiazolinone moiety was observed, and with EtOH and water similar adducts were isolated or detected. The structure and stereochemistry of the new compounds were established by means of NMR and electronic circular dichroism (ECD) data supported by time-dependent density functional theory ECD (TDDFT-ECD) calculations. TDDFT-ECD calculations could efficiently distinguish the proposed epimeric products having different absolute configuration in the spiro heterocyclic ring.

The first spirocyclic GP inhibitor (GPI), a glucopyranosylidene-spiro-hydantoin (A, X = O in Chart 1) was disclosed in the mid-nineties [17] followed by the more easily synthesized thiohydantoin counterpart (A, X = S) [18,19].Both compounds were active as GPIs in the low micromolar concentration range.Glucopyranosylidene-spiro-oxathiazoles (B, X = S) [20,21] and isoxazolines (B, X = CH 2 ) [22] identified some ten years later proved to be submicromolar GPIs and still belong in the top ten most efficient glucose-derived inhibitors of the enzyme.For the strong binding of compounds A to the enzyme, X-ray crystallography of the enzyme-inhibitor complexes revealed the importance of the H-bond donor β-NH, as well as that of the α-carbonyl groups highlighted in red [23,24].Similar studies with compounds B concluded that bulky substituents such as the 2-naphthyl moiety (marked in blue) better contributed to the binding even in the absence of the hydrogen bonding β-NH and the carbonyl in the α-position [22].An attempt was made to unify these properties by the study of compounds C [3]. Surprisingly, these derivatives proved rather weak inhibitors and this was attributed to the =N-C=O linker between the spirocycle and the aryl groups resulting in an unfavorable orientation of the latter for the binding.To circumvent this problem the synthesis of compounds D was envisaged, and the results of these studies are presented in this paper.

Results and Discussion
In our previous study dealing with the preparation of glucopyranosylidene-spiroiminothiazolidinone C [3] bromo-amide derivative 1 was found to be the substrate of choice in reactions with thiourea, therefore its transformation with thiobenzamide was investigated first (Table 1) to find the best conditions for providing the present target compounds.In refluxing EtOH, that was advantageous earlier, only 6 [25] could be obtained (entry 1).Since acetone also proved to be a suitable solvent previously, the higher boiling diethylketone was tried next, but 7 [26] could be isolated as the only product (entry 2).Much less time was needed for the complete consumption of the starting 1 in pyridine or DMF (entries 3 and 4), however, only the formation of 8 [27] or 9 [28] was observed, respectively.Following this, less reactive solvents such as toluene and m-xylene (entries 5 and 6) were tested affording similar yields of the expected compound 2 with a significantly shorter reaction time for the latter.In ether type solvents (entries 7-9) and nitromethane (entry 10) the yields of 2 remained the same, therefore, the use of m-xylene was further optimized.Applying an inert atmosphere (entry 11) or addition of a silver salt (entry 12) left the yield practically unchanged, while microwave heating resulted in a considerable increase in the yield of 2 (entry 13).This was further improved by raising the temperature (entry 14), however even these conditions did not allow diminishing the amount of thiobenzamide (entry 15).Other thioamides gave the expected spiro derivatives 3-5 in acceptable yields (entries [16][17][18].The formation of the depicted molecules 2-5 with an inversion of configuration at the newly formed spiro carbon is consistent with a presumable bimolecular nucleophilic substitution mechanism and concurs with earlier observations on the stereochemical outcome of the synthesis of glucopyranosylidene-spiro-iminothiazolidinone C [3]. Attempts to remove the benzoyl protecting groups [29] under a series of base (NaOMe/MeOH, 1 day; KCN MeOH, 4 weeks decomposition) or acid catalyzed (AcCl/MeOH, 4 weeks no reaction) transesterification conditions or under basic hydrolysis conditions (4 equiv.LiOH/MeOH, 0 °C decomposition) proved unsuccesful.

Results and Discussion
In our previous study dealing with the preparation of glucopyranosylidene-spiroiminothiazolidinone C [3] bromo-amide derivative 1 was found to be the substrate of choice in reactions with thiourea, therefore its transformation with thiobenzamide was investigated first (Table 1) to find the best conditions for providing the present target compounds.In refluxing EtOH, that was advantageous earlier, only 6 [25] could be obtained (entry 1).Since acetone also proved to be a suitable solvent previously, the higher boiling diethylketone was tried next, but 7 [26] could be isolated as the only product (entry 2).Much less time was needed for the complete consumption of the starting 1 in pyridine or DMF (entries 3 and 4), however, only the formation of 8 [27] or 9 [28] was observed, respectively.Following this, less reactive solvents such as toluene and m-xylene (entries 5 and 6) were tested affording similar yields of the expected compound 2 with a significantly shorter reaction time for the latter.In ether type solvents (entries 7-9) and nitromethane (entry 10) the yields of 2 remained the same, therefore, the use of m-xylene was further optimized.Applying an inert atmosphere (entry 11) or addition of a silver salt (entry 12) left the yield practically unchanged, while microwave heating resulted in a considerable increase in the yield of 2 (entry 13).This was further improved by raising the temperature (entry 14), however even these conditions did not allow diminishing the amount of thiobenzamide (entry 15).Other thioamides gave the expected spiro derivatives 3-5 in acceptable yields (entries [16][17][18].The formation of the depicted molecules 2-5 with an inversion of configuration at the newly formed spiro carbon is consistent with a presumable bimolecular nucleophilic substitution mechanism and concurs with earlier observations on the stereochemical outcome of the synthesis of glucopyranosylidene-spiro-iminothiazolidinone C [3]. Attempts to remove the benzoyl protecting groups [29] under a series of base (NaOMe/MeOH, 1 day; KCN MeOH, 4 weeks decomposition) or acid catalyzed (AcCl/MeOH, 4 weeks no reaction) transesterification conditions or under basic hydrolysis conditions (4 equiv.LiOH/MeOH, 0 • C decomposition) proved unsuccesful.After this experience the use of the more readily removable O-acetyl protecting groups was envisaged.To this end, the O-benzoyl groups of 10 [28] were removed by using the Zemplén protocol and the resulting amide 11 was treated with Ac 2 O under acidic catalysis (Scheme 1).Since N-acetylation also took place with an excess of the acetylating agent to give 12, four equivalents of Ac 2 O in AcOH had to be used to yield the expected 13 [30].This compound was subjected to radical-mediated bromination conditions [31,32] to afford 14 [19] which could be transformed to thiazolinone 15 with thiobenzamide.It is to be noted, that in this latter transformation the conditions used to get the O-perbenzoylated 2-5 resulted in only 14-18% yields due to decomposition, but at lower temperature in toluene 46% of 15 could be achieved.After this experience the use of the more readily removable O-acetyl protecting groups was envisaged.To this end, the O-benzoyl groups of 10 [28] were removed by using the Zemplén protocol and the resulting amide 11 was treated with Ac2O under acidic catalysis (Scheme 1).Since Nacetylation also took place with an excess of the acetylating agent to give 12, four equivalents of Ac2O in AcOH had to be used to yield the expected 13 [30].This compound was subjected to radicalmediated bromination conditions [31,32] to afford 14 [19] which could be transformed to thiazolinone 15 with thiobenzamide.It is to be noted, that in this latter transformation the conditions used to get the O-perbenzoylated 2-5 resulted in only 14-18% yields due to decomposition, but at lower temperature in toluene 46% of 15 could be achieved.For the removal of the O-acetyl protective groups [29] of 15 a wide range of conditions recommended in the literature were attempted (Table 2).Under classical Zemplén conditions or transesterifications catalyzed by K2CO3 or KCN (entries 1-3, respectively) only the formation of multicomponent mixtures could be observed.Similar outcomes were achieved by using nucleophilic ester cleaving reagents (entries 4 and 5) or the non-nucleophilic base DBU (entry 6).However, LiOH in MeOH (entry 7) gave a mixture of 16 and 17, and these compounds were obtained in better yields with a substoichiometric amount of LiOH (entry 8).From the attempted acid catalyzed transesterification conditions (entries 9-11), it was only the use of KHSO4 (entry 11) that resulted in a moderate yield of 18 with the 2'-O-acetyl group remaining in the molecule.
The presence of the MeOH addition products 17 and 18 in the deacetylation mixtures prompted us to investigate the reaction of 2 and 15 with MeOH and EtOH without any other additive (Table 3).It turned out that both alcohols added to the thiazolinone moiety to furnish 19-22 (entries 1-4), respectively, as the major stereoisomers indicated by the characteristic signals in the NMR spectra (see detailed structural elucidation below).Removal of the solvent alcohol at 30 °C allowed us to characterize these addition products.The presence of minor components (<10%) could be observed in the 1 H-NMR spectra (see Supporting Information), however, their separation and purification proved infeasible due to the same chromatographic mobility of the substances.Therefore, the formation of both possible diastereomeric addition products cannot be excluded.This addition reaction proved to be reversible since by heating of 19 (at 80 °C) or 21 (at 60 °C) in toluene thiazolinones 2 and 15 were recovered, respectively.To the best of our knowledge, similar alcohol addition was reported only for 2-perfluoroalkyl-5-methoxycarbonylmethylene-thiazoline-4-one type compounds [33,34].Our findings indicate that the presence of strongly electron withdrawing substituents, such as the perfluoroalkyl groups are not required for the ROH addition.For the removal of the O-acetyl protective groups [29] of 15 a wide range of conditions recommended in the literature were attempted (Table 2).Under classical Zemplén conditions or transesterifications catalyzed by K 2 CO 3 or KCN (entries 1-3, respectively) only the formation of multicomponent mixtures could be observed.Similar outcomes were achieved by using nucleophilic ester cleaving reagents (entries 4 and 5) or the non-nucleophilic base DBU (entry 6).However, LiOH in MeOH (entry 7) gave a mixture of 16 and 17, and these compounds were obtained in better yields with a substoichiometric amount of LiOH (entry 8).From the attempted acid catalyzed transesterification conditions (entries 9-11), it was only the use of KHSO 4 (entry 11) that resulted in a moderate yield of 18 with the 2'-O-acetyl group remaining in the molecule.
The presence of the MeOH addition products 17 and 18 in the deacetylation mixtures prompted us to investigate the reaction of 2 and 15 with MeOH and EtOH without any other additive (Table 3).It turned out that both alcohols added to the thiazolinone moiety to furnish 19-22 (entries 1-4), respectively, as the major stereoisomers indicated by the characteristic signals in the NMR spectra (see detailed structural elucidation below).Removal of the solvent alcohol at 30 • C allowed us to characterize these addition products.The presence of minor components (<10%) could be observed in the 1 H-NMR spectra (see Supporting Information), however, their separation and purification proved infeasible due to the same chromatographic mobility of the substances.Therefore, the formation of both possible diastereomeric addition products cannot be excluded.This addition reaction proved to be reversible since by heating of 19 (at 80 • C) or 21 (at 60 • C) in toluene thiazolinones 2 and 15 were recovered, respectively.To the best of our knowledge, similar alcohol addition was reported only for 2-perfluoroalkyl-5-methoxycarbonylmethylene-thiazoline-4-one type compounds [33,34].Our findings indicate that the presence of strongly electron withdrawing substituents, such as the perfluoroalkyl groups are not required for the ROH addition.The addition of water was also tested with 15 (entry 5) and the formation of 23 was observed in a mixture with the starting compound (15:23~1:0.4).The appearance of 23 made it clear that a similar addition may occur during the enzymatic tests necessarily carried out under aqueous conditions, thereby thwarting the study of these compounds as enzyme inhibitors.The addition of water was also tested with 15 (entry 5) and the formation of 23 was observed in a mixture with the starting compound (15:23~1:0.4).The appearance of 23 made it clear that a similar addition may occur during the enzymatic tests necessarily carried out under aqueous conditions, thereby thwarting the study of these compounds as enzyme inhibitors.(C-4) (for numbering see Chart 2).Molecular masses determined by MS corroborated the formation of the spirocycles 2-5 and 15.The configuration of the 1',5-spiro carbon in 2 was established by determining the three-bond heteronuclear coupling constant between the axial H-2' ( 4 C1 conformation of the sugar ring followed from the vicinal proton-proton couplings in the 1 H-NMR spectra, see experimental) and C-4.The observed 6 Hz value indicated trans diaxial arrangement of these nuclei thereby proving the (R) configuration of the spiro center.For 3-5, the similarities of the chemical shifts for the sugar ring protons lead to the assumption that the spiro configuration is the same.
Both 1 H-and 13 C-NMR spectra of 17-22 exhibited characteristic signals for the OMe and OEt groups (see experimental section for specifics).The 1 H-NMR spectra showed resonances for the exchangable NH protons (6.6-7.2 ppm), and in the 13  Since no NMR method seemed suitable to determine the configuration of C-2 in the addition products 17-22, ECD measurements and TDDFT-ECD calculations were also performed for 21.These studies were extended to 15 in order to confirm the NMR results independently (Chart 2 shows the studied stereoisomeric structures).For the determination of the absolute configuration of the C-1' chirality center, the solution TDDFT-ECD method was utilized on the (1'R) and (1'S) diastereomers of 15 (Chart 2) [35,36].Based on the literature data for other acetylated glucopyranose derivatives, the ECD spectrum was expected to be governed by the aglycon part containing the C-1' chirality center as part of a 2-phenylthiazolin-4-one unit [37].The initial conformers generated for both stereoisomers with the OPLS-2005 force field were reoptimized at the B3LYP/6-31G(d) in vacuo and the ωB97XD/TZVP [35,38] PCM/MeCN levels (see Supporting Information).ECD calculations were performed with various functionals (B3LYP, BH&HLYP, CAM-B3LYP and PBE0) and the TZVP basis set for the low-energy conformers over 1% Boltzmann population.The computed averaged ECD spectrum of the (1'R) diastereomer reproduced all the transitions of the experimental ECD spectrum well (Figure 1a).In contrast, the Since no NMR method seemed suitable to determine the configuration of C-2 in the addition products 17-22, ECD measurements and TDDFT-ECD calculations were also performed for 21.These studies were extended to 15 in order to confirm the NMR results independently (Chart 2 shows the studied stereoisomeric structures).
For the determination of the absolute configuration of the C-1' chirality center, the solution TDDFT-ECD method was utilized on the (1'R) and (1'S) diastereomers of 15 (Chart 2) [35,36].Based on the literature data for other acetylated glucopyranose derivatives, the ECD spectrum was expected to be governed by the aglycon part containing the C-1' chirality center as part of a 2-phenylthiazolin-4-one unit [37].The initial conformers generated for both stereoisomers with the OPLS-2005 force field were reoptimized at the B3LYP/6-31G(d) in vacuo and the ωB97XD/TZVP [35,38] PCM/MeCN levels (see Supporting Information).ECD calculations were performed with various functionals (B3LYP, BH&HLYP, CAM-B3LYP and PBE0) and the TZVP basis set for the low-energy conformers over 1% Boltzmann population.The computed averaged ECD spectrum of the (1'R) diastereomer reproduced all the transitions of the experimental ECD spectrum well (Figure 1a).In contrast, the (1'S) diastereomer had a much poorer mirror-image agreement of the experimental Cotton effects (CEs) below 325 nm (Figure 1b  In the case of 21, both the C-1' and the C-2 chirality centers were expected to contribute to the ECD spectrum and ECD calculations were performed for the (2R,1'R) and (2S,1'R) diastereomers (Chart 2).The computed ECD spectra of (2R,1'R)-21 and (2S,1'R)-21 showed opposite CEs in the 190-290 nm spectral range with acceptable mirror image agreement (Figure 2a) or good reproduction of the experimental ECD spectrum (Figure 2b), respectively.However, the computed ECD of (2S,1'R)-21 had two well-separated negative CEs, the shape of which were much more similar to the experimental ECD than that of the broad positive computed CE of (2R,1'R)-21.Thus, the good overall agreement of (2S,1'R)-21 allowed the unambiguous elucidation of the absolute configuration at C-2 as (S).Compounds 19-22 were formed as preponderant diastereomers (e.g., (2S,1'R)-21 in Chart 2) although nucleophilic attack of the alcohol could have been expected from both sides of the planar thiazolinone ring.This observation may be explained by a steric hindrance of the C-2 center by the 2'-OAc or 2'-OBz groups facilitating the addition reaction from the opposite side.This can be visualized by the computed conformers of (1'R)-15 (Figure S1b in the Supporting Information).In the case of 21, both the C-1' and the C-2 chirality centers were expected to contribute to the ECD spectrum and ECD calculations were performed for the (2R,1'R) and (2S,1'R) diastereomers (Chart 2).The computed ECD spectra of (2R,1'R)-21 and (2S,1'R)-21 showed opposite CEs in the 190-290 nm spectral range with acceptable mirror image agreement (Figure 2a) or good reproduction of the experimental ECD spectrum (Figure 2b), respectively.However, the computed ECD of (2S,1'R)-21 had two well-separated negative CEs, the shape of which were much more similar to the experimental ECD than that of the broad positive computed CE of (2R,1'R)-21.Thus, the good overall agreement of (2S,1'R)-21 allowed the unambiguous elucidation of the absolute configuration at C-2 as (S).In the case of 21, both the C-1' and the C-2 chirality centers were expected to contribute to the ECD spectrum and ECD calculations were performed for the (2R,1'R) and (2S,1'R) diastereomers (Chart 2).The computed ECD spectra of (2R,1'R)-21 and (2S,1'R)-21 showed opposite CEs in the 190-290 nm spectral range with acceptable mirror image agreement (Figure 2a) or good reproduction of the experimental ECD spectrum (Figure 2b), respectively.However, the computed ECD of (2S,1'R)-21 had two well-separated negative CEs, the shape of which were much more similar to the experimental ECD than that of the broad positive computed CE of (2R,1'R)-21.Thus, the good overall agreement of (2S,1'R)-21 allowed the unambiguous elucidation of the absolute configuration at C-2 as (S).Compounds 19-22 were formed as preponderant diastereomers (e.g., (2S,1'R)-21 in Chart 2) although nucleophilic attack of the alcohol could have been expected from both sides of the planar thiazolinone ring.This observation may be explained by a steric hindrance of the C-2 center by the 2'-OAc or 2'-OBz groups facilitating the addition reaction from the opposite side.This can be visualized by the computed conformers of (1'R)-15 (Figure S1b in the Supporting Information).Compounds 19-22 were formed as preponderant diastereomers (e.g., (2S,1'R)-21 in Chart 2) although nucleophilic attack of the alcohol could have been expected from both sides of the planar thiazolinone ring.This observation may be explained by a steric hindrance of the C-2 center by the 2'-OAc or 2'-OBz groups facilitating the addition reaction from the opposite side.This can be visualized by the computed conformers of (1'R)-15 (Figure S1b in the Supporting Information).
While Odebenzoylations could not be achieved, during the O-deacetylation by the Zemplén protocol addition of MeOH onto the thiazolinone moiety was observed.This addition could be extended to EtOH and even water and the reactions proved to be reversible.Electronic circular dichroism spectroscopy aided by TDDFT-ECD computations could differentiate the possible diastereomers and confirmed the absolute configuration assigned by NMR analysis of the new compounds and also allowed the determination of the absolute configuration of the new stereogenic center formed upon the addition of alcohols.The formation of the water addition product under aqueous conditions prevented the test of these compounds as enzyme inhibitors.

General Methods
Melting points were measured on a Kofler hot-stage and are uncorrected.Optical rotations were determined with a Perkin-Elmer 241 polarimeter at r.t.NMR spectra and were recorded with Bruker 360 (360/90 MHz for 1 H/ 13 C) or Bruker 400 (400/100 MHz for 1 H/ 13 C) or Avance DRX 500 (500/125 MHz for 1 H/ 13 C) spectrometers (Bruker, Karlsruhe, Germany; Billerica, MA, USA).Chemical shifts are referenced to the internal TMS ( 1 H), or to the residual solvent signals ( 13 C).In the NMR spectra complete signal assignments are based on COSY, HSQC, and HSQMBC correlations.Mass spectra were obtained by a Bruker micrOTOF-Q instrument.TLC was performed on DC-Alurolle Kieselgel 60 F 254 (Merck, Darmstadt, Germany), and the plates were visualised under UV light and by gentle heating (generally no spray reagent was used but, if more intense charring was necessary, the plate was sprayed with the following solution: abs.EtOH (95 mL), cc.H 2 SO 4 (5 mL) anisaldehyde (1 mL)).For column chromatography Kieselgel 60 (Merck, particle size 0.063-0.200mm) was used.Organic solutions were concentrated in vacuo at 40-60 • C (water bath).Xylene and toluene were distilled from P 4 O 10 and stored over sodium wires.MeOH was purified by distillation after refluxing for a couple of hours with magnesium turnings and iodine.CHCl 3 was distilled from P 4 O 10 and stored over 4Å molecular sieves.Thioamides were purchased from Sigma-Aldrich (Steinheim, Germany).C-(2,3,4,6-tetra-O-benzoyl-1-bromo-β-D-glucopyranosyl)formamide (1) and C-(2,3,4,6-tetra-O-benzoyl-β-D-glucopyranosyl)formamide (10) were synthesized according to published procedures [28].Microwave-assisted reactions were carried out using a CEM-Discover Focused microwave synthesis system (2450 MHz).[28] and the corresponding thioamide (2.0 equiv.)were mixed in dry xylene (10 mL/mmol), in a sealed vial and the reaction mixture was stirred for 1.5 h at 140 • C (200 W).Then the mixture was evaporated, and the residue was purified by column chromatography.

Computational Methods
Mixed torsional/low-frequency mode conformational searches were carried out by means of the Macromodel 10.8.011 software [39] using OPLS-2005 (Optimized Potential for Liquid Simulations) Force Field [40,41] with the implicit solvent model for CHCl 3 applying a 21 kJ/mol energy window yielding 53-151 conformers.Geometry optimizations [B3LYP/6-31G(d) in vacuo and ωB97XD/TZVP [35,38] with a PCM solvent model for MeCN] and TDDFT calculations were performed with Gaussian 09 using various functionals (B3LYP, BH&HLYP, CAM-B3LYP and PBE0) and TZVP basis set [42].ECD spectra were generated as the sum of Gaussians with 3000 cm −1 half-height width (corresponding to ca. 15 nm at 220 nm), using dipole-velocity computed rotational strengths [43].Boltzmann distributions were estimated from the ZPVE-corrected B3LYP/6-31G(d) energies in the gas-phase calculations and from the ωB97XD/TZVP energies in the solvated ones.The MOLEKEL software package was used for visualization of the results [44].

Chart 1 .
Selected spirobicyclic inhibitors of rabbit muscle glycogen phosphorylase A, B and C (Ki dissociation constant of the enzyme-inhibitor complex) and the target compounds of the present study (D).

Chart 1 .
Selected spirobicyclic inhibitors of rabbit muscle glycogen phosphorylase A, B and C (K i dissociation constant of the enzyme-inhibitor complex) and the target compounds of the present study (D).

H 23 2 1
The formation of the other possible diastereomers cannot be excluded although their isolation failed; 2 Partial addition (15:23~1:0.4).The addition of water was also tested with 15 (entry 5) and the formation of 23 was observed in a mixture with the starting compound (15:23~1:0.4).The appearance of 23 made it clear that a similar addition may occur during the enzymatic tests necessarily carried out under aqueous conditions, thereby thwarting the study of these compounds as enzyme inhibitors.Structural elucidation of the O-perbenzoylated spirocycles 2-5 was carried out by NMR and MS methods.The 13 C-NMR spectra contained signals in the range of 194-195 ppm and 187-188 ppm for C-2 and C-4, respectively, while similar resonances for 15 appeared at 194.7 ppm (C-2) and 187.0 ppm (C-4) (for numbering see Chart 2).Molecular masses determined by MS corroborated the formation of the spirocycles 2-5 and 15.The configuration of the 1',5-spiro carbon in 2 was established by determining the three-bond heteronuclear coupling constant between the axial H-2' ( 4 C 1 conformation of the sugar ring followed from the vicinal proton-proton couplings in the 1 H-NMR spectra, see experimental) and C-4.The observed 6 Hz value indicated trans diaxial arrangement of these nuclei thereby proving the (R) configuration of the spiro center.For 3-5, the similarities of the chemical shifts for the sugar ring protons lead to the assumption that the spiro configuration is the same.
C-NMR spectra signals appeared for C-2 in the range of 90-100 ppm instead of the 194-195 ppm resonances of the starting materials.
(1'S) diastereomer had a much poorer mirror-image agreement of the experimental Cotton effects (CEs) below 325 nm (Figure 1b) confirming the decisive role of the C-1' chirality center in the ECD pattern.The highest-wavelength positive n-π* CE could not be reproduced by the computed ECD of the (1'S) epimer, since similar to the other transitions, opposite CE should have been calculated.Based on the good agreement of the (1'R) diastereomer, the (1'R) absolute configuration of 15 was unambiguously determined in concord with the NMR analysis of 2. Chart 2. Possible stereoisomers of 15 and 21.Both 1 H-and 13 C-NMR spectra of 17-22 exhibited characteristic signals for the OMe and OEt groups (see experimental section for specifics).The 1 H-NMR spectra showed resonances for the exchangable NH protons (6.6-7.2 ppm), and in the 13 C-NMR spectra signals appeared for C-2 in the range of 90-100 ppm instead of the 194-195 ppm resonances of the starting materials.
) confirming the decisive role of the C-1' chirality center in the ECD pattern.The highest-wavelength positive n-π* CE could not be reproduced by the computed ECD of the (1'S) epimer, since similar to the other transitions, opposite CE should have been calculated.Based on the good agreement of the (1'R) diastereomer, the (1'R) absolute configuration of 15 was unambiguously determined in concord with the NMR analysis of 2.

Table 3 .
Addition of alcohols and water onto spiro-thiazolinones 2 and 15.

Table 3 .
Addition of alcohols and water onto spiro-thiazolinones 2 and 15.

Table 3 .
Addition of alcohols and water onto spiro-thiazolinones 2 and 15.