Enantiocontrolled Preparation of ϒ-Substituted Cyclohexenones: Synthesis and Kinase Activity Assays of Cyclopropyl-Fused Cyclohexane Nucleosides

The enantioselective preparation of the two isomers of 4-hydroxy-2-cyclohexanone derivatives 1a,b was achieved, starting from a common cyclohexenone, through asymmetric transfer hydrogenation (ATH) reactions using bifunctional ruthenium catalysts. From these versatile intermediates, a stereoselective route to a cytosine analogue built on a bicyclo [4.1.0]heptane scaffold is described. Nucleoside kinase activity assays with this cyclopropyl-fused cyclohexane nucleoside, together with other related nucleosides (2a–e), were performed, showing that thymine- and guanine- containing compounds have affinity for herpes simplex virus Type 1 (HSV-1) thymidine kinase (TK) but not for human cytosolic TK-1, thus pointing to their selectivity for herpetic TKs but not cellular TKs.


Introduction
Optically active Υ-substituted cycloalkenones are compounds of synthetic importance that are used as precursors in the synthesis of natural products and pharmaceutically active molecules [1][2][3][4]. In particular, both enantiomers of 4-hydroxy-2-cyclohexenone 1a and their O-protected derivatives, such as 1b,c, have been extensively used in organic synthesis ( Figure 1) [5][6][7]. These chiral cycloalkenones have been prepared by different methodologies that involve enzymatic transformations [8], kinetic resolutions [9], chiral auxiliaries [10], asymmetric catalysis [11], or chiral pool compounds [12,13]. However, most of these approaches do not provide access to both enantiomers. Therefore, the development of efficient routes to such relevant building blocks is of considerable interest. In order to perform these enzymatic assays, to complete the 5′-hydroxymethylbicyclo [4.1.0]heptanyl family 2 by preparing the cytosine analogue 2e, and considering our work on the synthesis of bioactive products bearing a cyclohexane unit [16][17][18], general and easy access to both enantiomers of 1a and 1b was required. We envisaged that the In a previous work, we reported the synthesis of the nucleoside analogues (NAs) 2a-d built on a bicyclo [4.1.0]heptane scaffold, starting from cyclohexenone 3 bearing a dihydrobenzoin moiety as the chiral auxiliary [14]. These NAs were designed as potential antiherpetic agents by a molecular modeling study on the HSV-1 TK active site that is involved in the first phosphorylation step of the activation process. However, none of the compounds showed significant antiviral activity at subtoxic concentrations (~250 µM). The lack of activity indicated that these functionalized NAs might, eventually, fail to reach the HSV DNA polymerase interaction step. Therefore, further studies were needed to identify the molecular basis of the antiherpetic inactivity of the prepared compounds; in particular, single enzymatic assays were required to test the suitability of these compounds to pass the usual rate-limiting first phosphorylation step [15].
In order to perform these enzymatic assays, to complete the 5 -hydroxymethylbicyclo [4.1.0]heptanyl family 2 by preparing the cytosine analogue 2e, and considering our work on the synthesis of bioactive products bearing a cyclohexane unit [16][17][18], general and easy access to both enantiomers of 1a and 1b was required. We envisaged that the asymmetric transfer hydrogenation (ATH) on 1,4-cyclohexenedione monoethylene ketal 4 could be applied to prepare such chiral cyclohexenones [19]. ATH with bifunctional ruthenium catalysts has become one of the most practical and versatile tools for accessing enantiomerically enriched alcohols in organic synthesis, due to its excellent selectivity, practical simplicity, and wide substrate scope [20,21]. In this article, we disclose a concise entry to both enantiomers of Υ-hydroxycyclohexenone by ATH via catalyst control, their use as precursors in the preparation of the conformationally restricted cytosine analogue 2e, and the results of the enzymatic affinity of compounds 2a-e on HSV-1 thymidine kinase.

Results and Discussion
The synthesis of both enantiomers of 1a and 1b (Scheme 1) started with enone 4, which was easily prepared on a multigram scale from the commercially available 1,4cyclohexanedione monoethylene acetal in 88% yield, through a two-step optimized dehydrogenation protocol that involved the formation of the corresponding silyl enol ether followed by oxidation with IBX·MPO complex [17]. The asymmetric reduction of enone 4 was first carried out in a biphasic medium (CH 2 Cl 2 /H 2 O 1:1), using (R,R)-Noyori-I, (R,R)-5 as catalyst, TBAC as a phase-transfer agent, and HCOONH 4 as the hydrogen source [22][23][24]. Under these experimental conditions, allylic alcohol (R)-6 was obtained, along with its totally hydrogenated derivative 7 in a 6:1 ratio and 67% yield. The reduction also resulted in the removal of the ketal protecting group, providing variable amounts of 1a. However, when the reaction was performed using HCOONa [25], the desired allylic alcohol (R)-6 was isolated in 79% yield. Next, deprotection of the carbonyl group was achieved by treatment with montmorillonite K-10 in dichloromethane at room temperature, affording the volatile 4-hydroxy-2-cyclohexenone (R)-1a in 64% yield and 92% ee (determined by CHPLC). Protection of the alcohol 4 as a silyl ether with imidazole and TBSCl in dichloromethane and subsequent hydrolysis of the ketal using montmorillonite K-10 led to (R)-1b in 87% yield for the two steps.
The satisfactory results achieved in the ATH step prompted us to examine the scope of the reaction with other cyclohexenones 8-10 using the optimized conditions ((R,R)-5, TBAC and HCOONa in CH2Cl2/H2O 1:1, Table 1). The ATH reaction on 2-cyclohexenone 8 delivered a 2:1 mixture of the corresponding alcohol 11 and the fully hydrogenated cyclohexanol 12 in good yield with an enantiomeric excess of 92% (entry 1). The ATH on cyclohexenone 9 bearing a methyl group at the β position, provided a 10:1 mixture of the corresponding allylic alcohol 13 and its fully hydrogenated derivative 14, from which the major compound 13 could be isolated by column chromatography (entry 2). The enantioselectivity of the reaction was slightly lower (88% ee). Finally, the ATH reaction with 1-tetralone 10 (entry 3) delivered allylic alcohol 15 in good yield with the highest enantioselectivity (94% ee). Overall, the ATH reaction on 8-10 in the presence of (R,R)-5 proved to be highly effective, providing access to the corresponding alcohols as the major products with excellent levels of enantioselectivity. The enantio-and chemoselectivities of the ATH reaction were sensitive to the structure of the cyclohexenones. In cyclohexanone 8, the formation of the fully hydrogenated compound was detrimental for the yield. The presence of a vinylic methyl group enhances the chemoselectivity of the ATH reaction. These chiral cyclohexanols are useful platforms for the synthesis of more elaborate products [26][27][28]. The ATH reaction has also been studied on cycloalkenones, with five-and seven-membered rings obtaining lower values of enantio-and chemoselectivity (see the Supplementary Materials). The satisfactory results achieved in the ATH step prompted us to examine the scope of the reaction with other cyclohexenones 8-10 using the optimized conditions ((R,R)-5, TBAC and HCOONa in CH 2 Cl 2 /H 2 O 1:1, Table 1). The ATH reaction on 2-cyclohexenone 8 delivered a 2:1 mixture of the corresponding alcohol 11 and the fully hydrogenated cyclohexanol 12 in good yield with an enantiomeric excess of 92% (entry 1). The ATH on cyclohexenone 9 bearing a methyl group at the β position, provided a 10:1 mixture of the corresponding allylic alcohol 13 and its fully hydrogenated derivative 14, from which the major compound 13 could be isolated by column chromatography (entry 2). The enantioselectivity of the reaction was slightly lower (88% ee). Finally, the ATH reaction with 1-tetralone 10 (entry 3) delivered allylic alcohol 15 in good yield with the highest enantioselectivity (94% ee). Overall, the ATH reaction on 8-10 in the presence of (R,R)-5 proved to be highly effective, providing access to the corresponding alcohols as the major products with excellent levels of enantioselectivity. The enantio-and chemoselectivities of the ATH reaction were sensitive to the structure of the cyclohexenones. In cyclohexanone 8, the formation of the fully hydrogenated compound was detrimental for the yield. The presence of a vinylic methyl group enhances the chemoselectivity of the ATH reaction. These chiral cyclohexanols are useful platforms for the synthesis of more elaborate products [26][27][28]. The ATH reaction has also been studied on cycloalkenones, with five-and seven-membered rings obtaining lower values of enantio-and chemoselectivity (see the Supplementary Materials).  Next, we turned our attention to the use of chiral cyclohexanol R-(6) as the starting material in the preparation of the cytosine analogue 2e, following an adaptation of our earlier work [14]. Accordingly, the synthesis of 2e (Scheme 2) started with the cyclopropanation reaction on (R)-6, which was first attempted using Shi's carbenoid (CF3COOZn CH2I) [29,30]. Under these conditions, the cyclopropanation took place with concomitant removal of the ketal to deliver the known bicyclic keto alcohol 16 albeit in  Next, we turned our attention to the use of chiral cyclohexanol R-(6) as the starting material in the preparation of the cytosine analogue 2e, following an adaptation of our earlier work [14]. Accordingly, the synthesis of 2e (Scheme 2) started with the cyclopropanation reaction on (R)-6, which was first attempted using Shi's carbenoid (CF3COOZn CH2I) [29,30]. Under these conditions, the cyclopropanation took place with concomitant removal of the ketal to deliver the known bicyclic keto alcohol 16 albeit in  Next, we turned our attention to the use of chiral cyclohexanol R-(6) as the starting material in the preparation of the cytosine analogue 2e, following an adaptation of our earlier work [14]. Accordingly, the synthesis of 2e (Scheme 2) started with the cyclopropanation reaction on (R)-6, which was first attempted using Shi's carbenoid (CF3COOZn CH2I) [29,30]. Under these conditions, the cyclopropanation took place with concomitant removal of the ketal to deliver the known bicyclic keto alcohol 16 albeit in  Next, we turned our attention to the use of chiral cyclohexanol R-(6) as the starting material in the preparation of the cytosine analogue 2e, following an adaptation of our earlier work [14]. Accordingly, the synthesis of 2e (Scheme 2) started with the cyclopropanation reaction on (R)-6, which was first attempted using Shi's carbenoid (CF3COOZn CH2I) [29,30]. Under these conditions, the cyclopropanation took place with concomitant removal of the ketal to deliver the known bicyclic keto alcohol 16 albeit in  Next, we turned our attention to the use of chiral cyclohexanol R-(6) as the starting material in the preparation of the cytosine analogue 2e, following an adaptation of our earlier work [14]. Accordingly, the synthesis of 2e (Scheme 2) started with the cyclopropanation reaction on (R)-6, which was first attempted using Shi's carbenoid (CF3COOZn CH2I) [29,30]. Under these conditions, the cyclopropanation took place with concomitant removal of the ketal to deliver the known bicyclic keto alcohol 16 albeit in  Next, we turned our attention to the use of chiral cyclohexanol R-(6) as the starting material in the preparation of the cytosine analogue 2e, following an adaptation of our earlier work [14]. Accordingly, the synthesis of 2e (Scheme 2) started with the cyclopropanation reaction on (R)-6, which was first attempted using Shi's carbenoid (CF3COOZn CH2I) [29,30]. Under these conditions, the cyclopropanation took place with concomitant removal of the ketal to deliver the known bicyclic keto alcohol 16 albeit in

87
-94 a Determined by 1 H-NMR. b Determined by chiral high-pressure liquid chromatography (CHPLC). c From the mixture of compounds 11 and 12.
Next, we turned our attention to the use of chiral cyclohexanol R-(6) as the starting material in the preparation of the cytosine analogue 2e, following an adaptation of our earlier work [14]. Accordingly, the synthesis of 2e (Scheme 2) started with the cyclopropanation reaction on (R)-6, which was first attempted using Shi's carbenoid (CF3COOZn CH2I) [29,30]. Under these conditions, the cyclopropanation took place with concomitant removal of the ketal to deliver the known bicyclic keto alcohol 16 albeit in very low yield (16%). Better results were obtained using Furukawa's procedure [31,32] (Et 2 Zn and ICH 2 Cl), which, after purification by column chromatography, afforded 16 in 78% yield. This compound was highly volatile; therefore, the next protection step was performed immediately after the cyclopropanation reaction without further purification. To continue with the synthesis, two protecting groups were evaluated. First, protection of the alcohol as a tert-butyldiphenylsilyl ether furnished 17 in 22% yield for the two steps. All attempts to increase the yield failed. On the other hand, benzyl protection under standard conditions (BnBr, NaH, THF) did not provide better results, delivering only unidentified decomposition products. After some experimentation, it was found that the O-benzyl-protected 18 could be obtained in 62% overall yield over the two steps, using BnBr and Ag 2 O. Next, olefination of the ketone via a Wittig reaction led to the expected alkene 19, which was rapidly submitted to the next step, as it easily isomerizes to the endocyclic isomer 20. Accordingly, the subsequent hydroboration (9-BBN) -oxidation (H 2 O 2 ) process furnished a chromatographically inseparable 2:1 mixture of diastereomers 21 and 22 in 95% yield. After several purifications by column chromatography, an enriched fraction of the main product was obtained and analyzed by NMR. The anti relative configuration of C-2 of the main product was determined by a NOESY experiment that showed cross peaks between H-7 endo and H-2 , indicating that the approach of the borane through the syn face was favored, due to the steric hindrance between the alkylborane and the cyclopropane in the four-center transition state that resulted from the approach of the borane to the more accessible anti face. After purification by column chromatography, benzoyl protection, followed by the removal of the benzyl-protecting group, delivered alcohol 23 in 62% yield and its diastereomer 24 in 29% yield. The primary amine 25 required for the nucleobase construction was accomplished in 84% yield, starting from 23 through a Mitsunobu reaction using diphenylphosphoryl azide (DPPA) [33], followed by catalytic hydrogenation and hydrochloride salt formation.
The cytosine nucleoside was prepared through the amination of the corresponding uridine derivative, which was prepared by a two-step protocol that involved the addition of the amine 25 to the isocyanate 26, followed by acid-mediated cyclization, to furnish the protected uracil nucleoside 27 [34]. Then, a one-pot amination in the presence of TsCl, Et 3 N and N-methylpiperidine at 0 • C [35], followed by ammonolysis with 30% NH 4 OH solution in water and, finally, removal of the benzoyl protection (CH 3 NH 2 , 33% in EtOH), afforded, after purification by column chromatography, the desired cytosine nucleoside analogue 2e.
Compound 2e was examined for antiherpetic activity (herpes simplex virus-1 (HSV-1; strain KOS) and herpes simplex virus-2 (strain G) in human embryonic lung (HEL) cell cultures. Unfortunately, however, it did not show significant antiviral activity at subtoxic concentrations (~100 µM) (see Supplementary Materials). These results were similar to those obtained for the previously synthesized compounds 2a-d. As previously mentioned, the lack of significant activity against herpes simplex virus by the 5 -hydroxymethylbicyclo [4.1.0]heptanyl NAs may arise from a low affinity (if any) or from substrate activity for the viral-encoded or cellular nucleoside kinases. Therefore, to evaluate this last point, we performed further studies to test the suitability of these compounds to pass the usual rate-limiting first phosphorylation step. Thus, the affinity of the synthesized compounds 2a-e for the cellular and HSV-1 thymidine kinases (TKs) were investigated ( Table 2). Based on these results, it can be concluded that compounds 2b and, to a lesser extent, 2d, were nicely recognized by herpes simplex virus-TK (IC 50 = 1.6 ± 0.1 µg/mL and 52 ± 28 µg/mL, respectively), in agreement with our modelling studies [14]. Only compound 2b showed some marginal affinity for mitochondrial TK-2, in addition to its excellent affinity for HSV-1 TK. The 2b analogue was also evaluated for substrate activity by HPLC technology, and it was found to be a good substrate for HSV-1 TK with an efficient conversion to the monophosphate metabolite. The cytosine derivative 2e did not display affinity for HSV-1 TK. It was expected that this cytosine analogue might show affinity for HSV-1 TK, as the natural nucleoside (2-deoxycytidine, 2-dC) is known to be recognized by several kinases, including HSV-1 TK; however, that was apparently not the case. Interestingly, none of the analogues showed significant affinity to human cellular kinases. Therefore, because some of the synthesized analogues were recognized by the virus-encoded nucleoside kinase, the lack of antiherpetic activity could be attributed to the lack of further conversion of the 5 -monophosphate derivative to the 5 -triphosphate or to a low affinity, if any, of the 5 -triphosphate metabolite to the virus-encoded DNA polymerase. REVIEW 5 of the amine 25 to the isocyanate 26, followed by acid-mediated cyclization, to furnis protected uracil nucleoside 27 [34]. Then, a one-pot amination in the presence of Et3N and N-methylpiperidine at 0 °C [35], followed by ammonolysis with 30% NH solution in water and, finally, removal of the benzoyl protection (CH3NH2, 33% in Et afforded, after purification by column chromatography, the desired cytosine nucleo analogue 2e.

Materials and Methods
General Methods. Commercially available reagents were used as received. Solvents were dried by distillation over the appropriate drying agents. All of the reactions were monitored by analytical thin-layer chromatography (TLC), using silica gel 60 F254 precoated aluminum plates (0.25 mm thickness). TLC spots were detected under UV light and/or by charring with a KMnO 4 /KOH aqueous solution or vanillin solution. Flash column chromatography was performed using silica gel (230-400 mesh). 1 H NMR spectra were recorded using 250 or 400 MHz and were referenced to the residual proton signals of CDCl 3 , 7.26 ppm, and MeOH-d 4 , 3.31 ppm. 13 C [ 1 H] NMR spectra were recorded at 90 MHz or 100 MHz and were referenced to the residual 13 C signal of CDCl 3 , 77.16 ppm, and MeOH-d 4 , 49.00 ppm. NMR signals were assigned with the help of COSY, HSQC, HMBC, and NOESY experiments. Melting points were determined on a hot stage and were uncorrected. Optical rotations were measured at 20 ± 2 • C at the sodium D line (589 nm) in a microcell (0.1 dm). Infrared spectra were recorded on a spectrophotometer equipped with a Golden Gate Single Refraction Diamond ATR (attenuated total reflectance) accessory. High-resolution mass spectra were recorded using electrospray ionization (ESI). [36]. The activity of recombinant thymidine kinase 1 (TK-1), TK-2, and herpes simplex virus-1 (HSV-1) TK, and the 50% inhibitory concentration of the test compounds, were assayed in a 50-µL reaction mixture containing 50 mM Tris/HCl, pH 8.0, 2.5 mM MgCl 2 , 10 mM dithiothreitol, 0.5 mM CHAPS, 3 mg/mL bovine serum albumin, 2.5 mM ATP, 1 µM [methyl-3 H]dThd, and enzyme. The samples were incubated at 37 • C for 30 min in the presence or absence of different concentrations (five-fold dilutions) of the test compounds. At this time, the enzyme reaction still proceeded linearly. Aliquots of 45 µL of the reaction mixtures were spotted on Whatman DE-81 filter paper disks (Whatman, Clifton, NJ, USA). The filters were washed three times for 5 min each in 1 mM ammonium formate, once for 1 min in water, and once for 5 min in ethanol. The radioactivity was determined by scintillation counting.

4R)-4-((tert-Butyldimethylsilyl)oxy)cyclohex-2-en-1-one [(R)-1b].
To a stirred solution of (R)-6 (102 mg, 0.65 mmol) in CH 2 Cl 2 (0.5 mL), imidazole (56 mg, 0.82 mmol) and a solution of TBSCl (124 mg, 0.82 mmol) in CH 2 Cl 2 (0.4 mL) were added at room temperature. The mixture was allowed to stir overnight. Then, a saturated aqueous NaHCO 3 solution (1 mL) was added, the aqueous layer was extracted with CH 2 Cl 2 (3 × 1 mL), and the organic layers were dried and concentrated under vacuum. The crude was dissolved in CH 2 Cl 2 (4 mL) and montmorillonite K-10 (513 mg) was added. The reaction mixture was allowed to stir at room temperature for 1 h. Then, it was filtered, and the solvent was removed under vacuum to furnish enone (R)-1b (128 mg, 0.56 mmol, 87% yield) as a yellowish oil: The mixture was allowed to stir overnight. Then, a saturated aqueous NaHCO 3 solution (1 mL) was added, the aqueous layer was extracted with more CH 2 Cl 2 (3 × 1 mL), and the organic layers were dried and concentrated under vacuum. The crude was dissolved in CH 2 Cl 2 (4 mL) and montmorillonite K-10 (294 mg) was added. The reaction mixture was allowed to stir at room temperature for 1 h. Then, it was filtered, and the solvent was removed under reduced pressure to afford enone (S)-1b (64 mg, 0.28 mmol, 88% yield) as a yellowish oil: were sequentially added at room temperature. The reaction mixture was allowed to stir for 24 h. Then, CH 2 Cl 2 (10 mL) and water (10 mL) were added, the two phases were separated, and the aqueous layer was extracted with CH 2 Cl 2 (3 × 5 mL). The volatiles of the organic phase were removed under reduced pressure and the resulting residue was purified by column chromatography (CHCl 3 ), affording an inseparable mixture of allylic alcohol (R)-11 and cyclohexanol 12 (210 mg, 2.14 mmol, 82% yield) in a 2:1 ratio as a yellowish oil that was analyzed by NMR spectroscopy and CHPLC (Daicel Chiralpak IC): 92% ee; 1  13 mmol) were sequentially added at room temperature. The reaction mixture was allowed to stir for 72 h. Then, CH 2 Cl 2 (10 mL) and water (10 mL) were added, the two phases were separated, and the aqueous layer was extracted with CH 2 Cl 2 (3 × 5 mL). The volatiles of the organic phase were removed under reduced pressure and the resulting residue was purified by column chromatography (CHCl 3  were sequentially added at room temperature. The reaction mixture was allowed to stir for 24 h. Then, CH 2 Cl 2 (2 mL) and water (2 mL) were added, the two phases were separated, and the aqueous layer was extracted with CH 2 Cl 2 (3 × 1 mL). The volatiles of the organic phase were removed under reduced pressure and the resulting residue was purified by column chromatography (CHCl 3  To a stirred solution of allylic alcohol (R)-6 (314 mg, 2.01 mmol) in anhydrous CH 2 Cl 2 (10 mL), Et 2 Zn (4 mL, 4.02 mmol, 1 M in hexane) at 0 • C was added and the mixture was stirred for 5 min at this temperature. Then, a solution of ICH 2 Cl (600 µL, 8.30 mmol) in CH 2 Cl 2 (3 mL) was added dropwise via syringe at 0 • C, and the mixture was stirred overnight, allowing it to warm to room temperature. Then, a saturated aqueous NaHCO 3 solution (10 mL) was added, and the aqueous layer was extracted with CH 2 Cl 2 (3 × 10 mL). The organic layer was dried (Na 2 SO 4 ), concentrated under reduced pressure (with vacuum controller) and purified by column chromatography (hexane-EtO 2 , 4:1) furnishing ketone 16 (198 mg, 1.57 mmol, 78% yield) as a colorless oil. Due to the volatility of compound 16, the crude was used without further purification prior to the next step.
To a solution of crude of the hydroxyacetone 16 in CH 2 Cl 2 (4 mL), Ag 2 O (560 mg, 2.41 mmol) and benzyl bromide (340 µL, 2.81 mmol) were added. The mixture was stirred at room temperature for 24 h, then it was filtered through a Celite ® pad and concentrated in vacuo. The crude was purified by column chromatography (hexanes-EtOAc, 6:1) to furnish 18 (270 mg, 1,24 mmol, 62% yield from (R)-6) as a pale oil. To a stirring solution of Ph 3 PCH 3 I (1.452 g, 3.59 mmol) in anhydrous THF (4 mL) at 0 • C, t-BuOK (402 mg, 3.58 mmol) was added, under nitrogen atmosphere, and the resulting yellow mixture was allowed to react for 1 h. Then, a solution of ketone 18 (155 mg, 0.71 mmol) in dry THF (1 mL) was added and the mixture was allowed to warm to room temperature and stirred for 3 h. Then, diethyl ether (10 mL) was added and the crude was filtered through a short pad of silica and Celite ® , using additional diethyl ether as eluent. The volatiles were removed under vacuum to obtain an orange oil of alkene 19 that was used for the next step without further purification, as it isomerizes to the more stable endocyclic regioisomer 20 at room temperature.
The crude of alkene 19 was rapidly dissolved in anhydrous THF (7 mL) and 9-BBN (4.30 mL, 2.15 mmol, 0.5 M in THF) was added at −10 • C. The mixture was allowed to warm to room temperature and stirred overnight. Then, water (1.2 mL), NaOH (1.5 mL, 3 M in water), and H 2 O 2 (1.4 mL, 30% in water) were added at 0 • C. After stirring for 15 min, the mixture was diluted with brine (15 mL) and CH 2 Cl 2 (15 mL) and the aqueous phase was extracted with additional CH 2 Cl 2 (2 × 10 mL). The organic layers were dried (Na 2 SO 4 ),