Synthesis of Cyclohexyl 6-O-Trityl-alpha-D-Threo-Hexopyranosid-4-ulo-(2,3:3',4')-2-Pyrazoline

Article

Synthesis of Cyclohexyl 6-O-Trityl-a-D-Threo-Hexopyranosid-4-ulo-(2,3:3',4')-2-Pyrazoline

Rajendra M. Srivastava ^{* a}, Ana Maria A. de Souza^b,Ladjane P. da Silva^a, João R. de Freitas Filho^a, Fernando Hallwass^a and Bogdan Doboszewski^a

^aDepartamento de Química Fundamental, Universidade Federal de Pernambuco, Av. Prof. Luis Freire, s/n, 50740-540, Recife - PE, Brazil

^bDepartamento de Engenharia Química, Universidade Federal de Pernambuco, Av. Prof. Artur de Sá, s/n, 50740-521, Recife - PE, Brazil

A reação de tri-O-acetil-D-glucal (1) com cicloexanol (2) usando o método de Ferrier forneceu o glicosídeo cicloexílico insaturado 3, que por sua vez produziu o composto 12 através da seqüência envolvendo hidrólise, tritilação seletiva no oxigênio do C-6, oxidação alílica e cicloadição dipolar 1,3 com diazometano. O produto 17 foi obtido em quatro etapas a partir de 6. Cálculos semi-empíricos de orbitais moleculares (AM1) para o composto 12 forneceram a conformação estável e também apoiaram a existência do efeito anomérico. Foi feito um esforço para abrir o anel pirazolínico dos compostos 12 e 17, porém não tivemos sucesso. A carga negativa parcial no C-3 foi responsável pela falta de reatividade deste carbono frente a redução da ligação C=N.

Reaction of tri-O-acetyl-D-glucal (1) with cyclohexanol (2) using Ferrier's method provided the unsaturated cyclohexyl glucoside 3, which furnished the title compound 12 in a sequence involving hydrolysis, selective tritylation of oxygen at C-6, allylic oxidation with MnO₂ and 1,3-dipolar cycloaddition with diazomethane. Starting from 6, product 17 was obtained in four steps. Semi-empirical molecular orbital calculations (AM1) were carried out for compound 12 which gave an idea about its stable conformation, and also supported the existence of the anomeric effect. Attempts to open the pyrazoline ring in 12 and 17 failed. Partial negative charge at C-3 was responsible for the lack of reactivity of 12 towards reduction of C=N bond.

Keywords: aminodeoxy sugars, Ferrier's rearrangement, 1,3-dipolar cycloaddition, diazomethane, allylic oxidation

Introduction

The isolation of the antibiotic streptomycin in 1944¹ and the presence of 2-deoxy-2-methylamino-L-glucosamine² as one of its components gave an impetus to the study of amino sugar synthesis. Other antibiotics of this family are kanamycins,³ neomycins⁴ and gentamycins.⁵ Many antibiotics composed primarily of carbohydrates have been obtained from microorganisms and are called aminoglycoside antibiotics because they contain amino groups in their glycoside moieties.⁶ Prumycin, 4-(D-alanylamino)-2-amino-2,4-dideoxy-L-arabinose, having antitumor activity⁷ was synthesized⁸ from glycine and L-serine. Another amino sugar derivative has been shown to completely inhibit the derivation of tumor necrosis factor induced by 100 mg/mL Escherichia coli endotoxin.⁹ 5-Amino-5-deoxy sugars are also known to be glycosidase inhibitors.^10,11

The importance of amino sugars described above led us to undertake the synthesis of cyclohexyl 3-amino-2-aminomethyl-2,3-dideoxy-a-D-talopyranoside 18 starting from 3,4,6-tri-O-acetyl-D-glucal 1.¹² However, this communication reports the synthesis and characterization of 12 and 17 only.

Experimental

General methods

Melting points were determined with the Thomas Hoover (Unimelt) apparatus and are uncorrected. Elemental analyses were carried out at the Instituto de Química, Universidade de São Paulo and in the Department of Fundamental Chemistry of this University. IR spectra were recorded (n_max, cm^-1) on a Perkin-Elmer models 467 and 237B spectrophotometers. NMR spectra were measured on a Varian 300 MHz or Bruker 200 MHz instruments in CDCl₃ solution unless otherwise stated, and the chemical shifts are reported in parts per million downfield from TMS. Beckmann DB-G or Carl Zeiss-Jena instrument was used for recording the UV spectrum. Specific rotations were obtained on JASCO Model DIP-370 polarimeter. TLC was performed on plates coated with silica gel 60-G (E. Merck). Exposure of the plates to iodine vapors was used to reveal the spots. Silica gel G 70-200 mesh was used for gravitational column chromatography. Anhydrous magnesium sulfate was used to dry the extracts.

Computational method

The semi-empirical molecular orbital calculations (AM1)¹³ were carried out using MOPAC 93 program^14,15 on an IBM RISC 6000 computer of this Department. Complete optimization of the geometry was achieved and the gradient norm dropped to 0.02.

Cyclohexyl 4,6-di-O-acetyl-2,3-dideoxy-a-D-erythro -hex-2-enopyranoside (3)

Tri-O-acetyl-D-glucal 1 (5.000 g, 18.38 mmol), cyclohexanol 2 (2.82 g, 28.2 mmol) in dry benzene (50 cm³) were stirred at room temperature, and boron trifluoride etherate (1 cm³) was added. The mixture was stirred for 40 min. During this time the solution turned dark brown. TLC (CHCl₃-EtOAc 9:1) showed the presence of two spots: R_f = 0.25 (product) and other R_f = 0.18. The latter could either be due to unreacted 1 or to the b-anomer 4. The mixture was washed with a saturated aqueous solution of sodium bicarbonate (5 x 10 cm³), water (2 x 10 cm³), and dried over Na₂SO₄. Filtration and evaporation of the volatiles provided a brown-colored syrup, which was purified by chromatography using hexane-CHCl₃ gradient to give 3 as a colorless syrup (3.29 g, 57%); [a]_D²⁵ + 111 ± 2° (c 3.3, CHCl₃); IR n_max/cm^-1 1750 and 1740 (COO) (neat); ¹H NMR (200 MHz) d 5.7-5.9 (m, 2 H, H-2 and H-3), 5.30 (ddd, ³J_4-5 9.0 Hz, ³J_4-3 1.0 Hz, ⁴J_4-2 1.0 Hz, 1 H, H-4), 5.17 (m, W/2 » 4.5 Hz, 1 H, H-1), 4.1-4.3 (m, 3 H, H-5, H-6 and H-6'), 3.64 (m, 1 H, O-CH), 2.08 (s, 3 H, OAc), 2.09 (s, 3 H, OAc), 1.1-2.1 (m, 10 H, 5 CH₂). Elemental analysis: Found: C, 61.66; H, 7.86. Calc. for C₁₆H₂₄O₆: C, 61.52; H, 7.74.

Cyclohexyl 4,6-di-O-acetyl-2,3-dideoxy-a-D-erythro -hexopyranoside (5)

Compound 3 (0.05 g, 1.6 mmol) in ethyl acetate (5 cm³) in the presence of PtO₂ (0.01 g) was hydrogenated at 1 atm. for 12 h at room temperature. Filtration and solvent evaporation yielded 5 as a syrup in almost quantitative yield (one spot on TLC with the same R_f value in CHCl₃-EtOAc 9:1 as that of 3); [a]_D^30.5 + 100° (c 0.8, CHCl₃); ¹H NMR (300 MHz): d 4.99 (s,br, W/2 » 4.5 Hz, 1 H, H-1), 4.6-4.8 (m, 1 H, H-4), 4.23 (dd, ²J_6-6' 11.7 Hz, ³J_6-5 5.6 Hz, 1 H, H-6), 4.10 (dd, ²J_6'-6 11.7 Hz, ³J_6'-5 2.3 Hz, 1 H, H-6'), 4.01 (ddd, ³J_5-4 10.1 Hz, ³J_5-6 5.6 Hz, ³J_5-6' 2.3 Hz, 1 H, H-5), 3.56 (m, 1 H, O-CH), 2.08 (s, 3 H, OAc), 2.05 (s, 3 H, OAc), 1.2-2.0 (m, 14 H, 7 CH₂). Elemental analysis: Found: C, 61.18; H, 8.42. Calc. for C₁₆H₂₆O₆: C, 61.13; H, 8.34.

Cyclohexyl 2,3-dideoxy-a-D-erythro-hex-2-enopyranoside (6)

A flask containing 3 (3.00 g, 9.62 mmol) in a solution (140 cm³) of MeOH-H₂O-Et₃N (9:6:1)¹⁶ was kept at room temperature for 1.5h. TLC in CHCl₃-AcOEt-CH₃OH (1.0:0.25:0.05) showed the disappearance of the starting material (R_f = 0.75) and the appearance of a new spot with R_f = 0.20. Evaporation and chromatography using CHCl₃-hexane (1.0:0.43) gave 1.9g (86.7%) of 6 as hygroscopic crystals: m.p. 67-68^oC (from EtOAc-cyclohexane); [a]_D²⁵ = + 46.0 ± 0.9^o (c 3.4, CHCl₃); IR n_max/cm^-1 3600-3100 (OH) (Nujol); ¹H NMR (300 MHz): d 5.95 (dt, ³J_3-2 10.2 Hz, ⁴J_3-1 » 1.3 Hz, ³J_3-4 » 1.3 Hz, 1 H, H-3), 5.73 (ddd, ³J_2-3 10.2 Hz, ⁴J_2-4 2.4 Hz, ³J_2-1 2.7 Hz, 1 H, H-2), 5.13 (m, ³J 1.3 Hz, 1 H, H-1), 4.20 (d,br ³J_4-5 9 Hz, 1 H, H-4), 3.85 (d, ³J_6-5 = ³J_6'5 3.9 Hz, 2 H, H-6, H-6'), 3.75 (dt, ³J_5-4 9.0 Hz, ³J_5-6 = ³J_5-6' 3.9 Hz, 1 H, H-5), 3.62 (m, 1 H, O-CH), 2.63 and 2.37 (s,br, exchangeable, 2 H, 2 OH), 2.04-1.10 (m, 10 H, 5CH₂). Elemental analysis: Found: C, 61.19; H, 8.57. Calc. for C₁₂H₂₀O₄.1/2H₂O : C, 60.72; H, 8.93.

Cyclohexyl 2,3-dideoxy-a-D-erythro-hexopyranoside (7)

Compound 6 (0.22 g, 0.93 mmol) in ethanol (20 cm³) and 5% Pd/C (0.02 g) was hydrogenated at room temperature and at atmospheric pressure for 24 h. TLC in CHCl₃-benzene (19.0:1.0) showed a new product with R_f = 0.46. The product was purified by column chromatography using benzene initially followed by benzene-chloroform as gradient. Chloroform-benzene (1:1) eluted product 7 as a viscous liquid (0.10 g, 45% yield ): ¹H NMR (300 MHz): d 4.92 (apparent dd, ³J_1-2 1.5 Hz, ³J_1-2¢ 3.0 Hz, 1 H, H-1), 3.84-3.71 (m, 2 H), 3.65-3.48 (m, 3 H), 2.69 (bs, s, 2 H, 2 OH), 1.96-1.10 (m, 14 H, 7 CH₂).

Cyclohexyl 2,3-dideoxy 6-O-trityl-a-D-erythro-hex-2-enopyranoside (8)

Compound 6 (1.00 g, 4.38 mmol) in dry pyridine (5 cm³) and trityl chloride (1.43 g, 5.13 mmol) were stirred at room temperature under nitrogen for 72 h following the reported procedure.¹⁷ TLC showed only a trace of unreacted 6. Excess of the trityl chloride was destroyed by putting crushed ice into the flask and letting the contents to warm up to room temperature. The contents of the flask were diluted with CHCl₃, transferred to a separatory funnel, washed with a saturated solution of sodium bicarbonate (3 x 5 cm³), water (2 x 5 cm³), and finally dried (Na₂SO₄). Solvent removal gave a viscous material which was chromatographed using initially benzene and followed by benzene-CHCl₃ (19:1) to give 1.9 g (92.2%) of 8 as a semisolid material; [a]_D²⁵ = + 10^o (c, 1; CHCl₃); IR n_max/cm^-1 3640-3120 (OH) and 1600 (C = C aromatic) (Nujol); ¹H NMR (300 MHz) d 7.50-7.20 (m, 15 H, Ph-H), 5.89 (ddd, ³J_3-2 10.1 Hz, ⁴J_3-1 1,3 Hz, ³J_3-4 1.7 Hz, 1 H, H-3), 5.72 (ddd, ³J_2-3 10.1 Hz, ³J_2-1 2.7 Hz, ⁴J_2-4 2.1 Hz, 1 H, H-2), 5.10 (m, W/2 » 4 Hz, 1 H, H-1), 4.04 (m, 1 H, H-4), 3.90 (ddd, ³J_5-6 5.5Hz, ³J_5-6' 5.1 Hz, ³J_5-4 9.0 Hz, 1 H, H-5), 3.67 (m, 1 H, O-CH), 3.42 (dd, ²J_6'-6 9.7 Hz, ³J_6'-5 5.1Hz, 1H, H-6'), 3.34 (dd, ²J_6-6' 9.7 Hz, ³J_6-5 5.5 Hz, 1 H, H-6), 2.30 (s, br,1 H, exchangeable, OH), 2.10-1.00 (m, 10 H, 5 CH₂). Elemental analysis: Found: C, 77,51; H, 7.20. Calc. for C₃₁H₃₄O₄^. 1/2H₂O: C, 77.63; H, 7.36.

Cyclohexyl 2,3-dideoxy-a-D-glycero-hex-2-enopyranoside -4-ulose (9)

Compound 6 (1.80 g, 7.90 mmol) was dissolved in dry CH₂Cl₂ (450 cm³) and freshly prepared activated MnO₂¹⁸ was added (18.50 g, 212.8 mmol). The mixture was stirred at room temperature for 4 hours. TLC (CHCl₃) showed a new spot with R_f = 0.59 and the presence of some starting alcohol (R_f = 0.38). The mixture was filtered through diatomaceous earth and the clear filtrate was evaporated to give a semi-solid material. Chromatography using hexane initially followed by hexane-chloroform eluted the fast moving compound to furnish 1.0 g (56%) of 9: m.p. 99°-100 °C (from ether-petroleum ether); IR n_max/cm^-1: 3600-3100 (OH, H bonded), 1690 (C = O conjugated) (KBr); ¹H NMR (300 MHz): d 6.87 (dd, ³J_2-3 10.2 Hz, ³J_2-1 3.6 Hz, 1 H, H-2), 6.11 (d, ³J_3-2 10.2 Hz, 1 H, H-3), 5.17 (d, ³J_1-2 3.6 Hz, 1 H, H-1), 4.53 (t, ³J_5-6 4.2 Hz, 1 H, H-5), 3.92 (dd, ²J_6-6' 11.8 Hz, ³J_6-5 4.2 Hz, 1 H, H-6), 4.02 (dd, ²J_6'-6 11.8 Hz, ³J_6'-5 4.2. Hz, 1 H, H-6'), 3.62-3.80 (m, 1 H, O-CH), 2.22 (s,br, exchangeable, 1 H, OH), 2.20-0.80 (m, 10 H, 5-CH₂). Elemental analysis: Found: C, 63.42; H, 8.07. Calc. for C₁₂H₁₈O₄: C, 63.72; H, 7.96.

Cyclohexyl 2,3-dideoxy-6-O-trityl-a-D-glycero-hex-2 -enopyranosid-4-ulose (10)

To a solution of 8 (1.00 g, 2.13 mmol) in CH₂Cl₂ (250 cm³) was added freshly prepared and dried MnO₂¹⁸ (5.00 g, 57.5 mmol). The mixture was stirred for 4 h. TLC (CHCl₃) showed the disappearance of 8. Filtration and evaporation of the solvent left a viscous material which was chromatographed using a mixture of benzene-CHCl₃ (1.0:0.43) to give amorphous 10 (0.72 g, 72.4% yield): IR n_max/cm^-1 1695 (C=O conjugated) (KBr); ¹H NMR (300 MHz): d 7.50 - 7.20 (m, 15 H, Ph-H), 6.86 (dd, ³J_2-3 10.2 Hz, ³J_1-2 3.4 Hz, 1 H, H-2), 6.07 (d, ³J_3-2 10.2 Hz,1 H, H-3), 5.49 (d, ³J_1-2 3.4 Hz, 1 H, H-1), 4,70 (dd, ³J_5-6 7.2 Hz, ³J_5-6' 2.4 Hz, 1 H, H-5), 3.86 (m, 1 H, O-CH), 3.63 (dd, ²J_6'-6 10.2 Hz, ³J_6'-5 2.4 Hz, 1 H, H-6'), 3.44 (dd, ²J_6-6' 10.2 Hz, ³J_6-5 7.2 Hz, 1 H, H-6), 2.20-1.10 (m, 10 H, 5 CH₂). Compound 10 was also prepared by tritylation of 9 in 86% yield.

Cyclohexyl 6-O-trityl-a-D-threo-hexopyranoside-4-ulo -(2,3,:3',4')-2-pyrazoline (12)

Compound 10 (0.10 g, 0.21 mmol) was dissolved in ether (5.0 cm³), and the freshly prepared diazomethane¹⁹ in ether was added dropwise at room temperature until the yellow color persisted.²⁰ The crystals appeared gradually. Decantation of the ethereal layer provided 0.08 g (70.0%) of the crude product. Recrystallization from a large quantity of hot CH₂Cl₂ gave 12 (0.07g , 61.9%): m.p. 184°-186 °C; [a]_D²⁰ -200^o (c 0.2, pyridine); IR n_max/cm^-1 3322 (NH) and 1660 (C=O) (KBr); ¹H NMR (300 MHz, pyridine-d₅) d 9.92 (s,br, 1 H, N-H, exchangeable), 7.8 ¾ 7.1 (m, 15 H, Ph-H), 5.51 (d, ³J_1-2 4.2 Hz, 1 H, H-1), 4.91 (t, ³J_5-6 ³J_5-6' 4.8 Hz, 1 H, H-5), 4.24 (d, ³J 2.1 Hz, 1 H, H-7'), 4.03 (m, 1 H, O-CH), 3.90 (d, ³J_6-5 ³J_6'-5 4.8 Hz, 2 H, H-6 and H-6'), 3.79 (unresolved, 2 H, H-2, H-7"), 2.2-1.1 (m, 10 H, cyclohexyl). Elemental analysis: Found: C, 74.40; H, 6.68; N, 5.25. Calc. for C₃₂H₃₄N₂O₄ .¹/₄H₂O: C, 74.61; H, 6.75; N, 5.43.

Cyclohexyl 1'-N-acetyl-4,6-di-O-acetyl-a-D-lyxo-hexo pyranoside-(2,3:3',4')-2-pyrazoline (17)

Compound 9 (0.27 g, 1.2 mmol) was dissolved in ether (10 cm³) and to this solution was added diazomethane in ether dropwise at room temperature until the greenish yellow color persisted. TLC (CH₂Cl₂-EtOAc, 9:1) showed the disappearance of the substrate after 5 min. Solvent evaporation left 15 which was hydrogenated (10 atm) in EtOAc (10 cm³) at room temperature overnight in the presence of 0.020 g of PtO₂. Filtration and solvent evaporation under vacuum furnished intermediate 16, which was acetylated and purified by chromatography (hexane-EtOAc 7:3) to give 17 (0.21 g, 45.7%): m.p. 117-118 °C (hexane- ether); ¹H NMR (300 MHz): d 5.85 (d, ³J_5-4 5.1 Hz, 1 H, H-4), 4.85 (d, ³J_1-2 6.3 Hz, 1 H, H-1), 4.57 (ddd, ³J_5-4 = ³J_5-6 4.3 Hz, ³J_5-6' 8.4 Hz, 1 H, H-5), 4.33 (dd, ³J_6-5 7.8 Hz, ²J_6-6' 12.3 Hz, 1 H, H-6), 4.25-4.10 (m, 3 H, H-6', H-7'', OCH), 3.73 (dd, ³J_7'-2 7.5 Hz, ²J_7'-7'' 12.0 Hz, 1 H, H-7'), 3.28 (quintet, ³J » 6 Hz, 1 H, H-2), 2.25 and 2.17 (two s, 6 H, OAc), 2.06 (s, 3 H, NAc), 1.90-1.20 (m, 10 H, 5 CH₂). Elemental analysis: Found: C, 57.16; H, 7.37; N, 6.87. Calc. for C₁₉H₂₈O₇N₂ : C, 57.56; H, 7.12; N, 7.06.

Results and Discussion

Reaction of tri-O-acetyl-D-glucal 1 with cyclohexanol 2 and a catalytic quantity of boron trifluoride etherate using the previously reported procedure,²¹ provided cyclohexyl 4,6-di-O-acetyl-2,3-dideoxy-a-D-erythro -hex-2-enopyranoside 3 as the main product (Scheme 1). The b-anomer which was probably formed as the minor product was inseparable from the unreacted 1 and no effort was made to isolate it. Column chromatography furnished 3 as a thick viscous liquid. Its ¹H NMR agreed with the proposed structure but the configuration a became clear only after catalytic reduction of the C-2-C-3 double bond (3®5). The J_1-2 coupling constants in the a- and b-2,3-unsaturated hexopyranosyl glycosides are known to be similar.²² Consequently, evaluation of the anomeric configuration on the basis of these couplings alone is difficult. The 300 MHz ¹H NMR spectrum of the reduced product 5 showed a signal of the anomeric proton as a broad singlet having a half-width of ca 4.5 Hz. This value together with a coupling constant J_4-5 10.1 Hz proves that 5 is an a anomer which adopts ⁴C₁ conformation. Observed couplings are incompatible with the b configuration.

Deacetylation of 3 using triethylamine in methanol¹⁶ gave 6. Tritylation at the primary OH group¹⁴ furnished 8, which in turn was subjected to the allylic oxidation at C-4 with activated MnO₂ to give the enone 10. The same compound was obtained by performing the allylic oxidation first (6®9) followed by tritylation (9®10), however the former sequence gave a better yield. Compound 6 was also hydrogenated to give cyclohexyl 2,3-dideoxy-a-D-erythro-hexopyranoside 7.

Addition of diazomethane to C-2-C-3 double bond in 10 occurred from the opposite side of the cyclohexyl function to give a transient product 11 which rapidly tautomerized to 12 (Scheme 2).²⁰ An alternative approach of CH₂N₂ from the same side is unlikely due to the bulky aglycon. It has already been shown earlier²⁰ that much smaller ethoxy group in ethyl 6-O-acetyl-2,3-dideoxy-a-D-glycero-hex-2-enopyranoside-4-ulose 13 controlled the approach of CH₂N₂ in the same way as in our case to furnish 14 exclusively (Scheme 3). The UV spectrum of 12 (l_max = 325 nm) is similar to that of 14 (l _max = 326 nm).²⁰

Following the procedure described above, compound 9 was treated with diazomethane to give 15. Attempted reduction of 15 with the hope of obtaining a cyclohexyl 3-amino-2-aminometyl-2,3-dideoxy-a-D-talopyranoside 18 unexpectedly didn't go beyond the reduction of the keto function at the C-4 atom. The pyrazoline ring remained unaltered using either sodium borohydride as a reductant or hydrogenation of 15 in the presence of PtO₂ using different conditions (solvent, temperature, pressure). Compound 16 which invariably resulted in all these reactions was isolated and characterized as its triacetate 17. Similarly, attempted cleavage of the pyrazoline ring in 12 also failed. The inertness of 12 and 15 towards reducing agents to furnish 18 can be rationalized in terms of the partial charges on the carbon atoms 3 and 4 (Table 2). According to the semi-empirical molecular orbital calculations using AM1 method, C-4 in compound 12 has a charge of +0.2885 e.u. whereas C-3 possesses a charge of -0.2087 e.u. For this reason, a nucleophile (base) can attack C-4 easily, but not C-3 due to the partial negative charge present on it.

The 300 MHz ¹H NMR spectrum of 12 (in pyridine-d₅) showed the anomeric proton signal at d 5.51 ppm (J 4.2 Hz) as a doublet. The coupling constant between H-1 and H-2 suggests a torsion angle of -130°. The semi-empirical calculations provided a value of -131.64° for the same dihedral angle. The protons H-2 and H-7'' form a multiplet at d 3.79 ppm. Irradiation of the H-1 proton simplified this signal to a four-line pattern. Sharpening of this signal indicated a loss of a small long range coupling between H-1 and H-7''. A doublet at 3.90 ppm (J 4.8 Hz) was ascribed to H-6 and H-6' since irradiation of H-5 at d 4.92 ppm caused its collapse to a singlet. The fact that H-6 and H-6' form a doublet means that the couplings between H-5 and both H-6 and H-6' are similar and implies that the torsion angles between H-5 and H-6, and H-5 and H-6' to be approximately ca. 50° and ca 130°, respectively. We also carried out a variable temperature experiment and found that at 70 °C, H-5 became a doublet of doublet indicating the nonequivalence of H-6 and H6' due to the slow rotation of the C-5 and C-6 bond. In this situation, the coupling constants between H-6 and H-5 and H-6' and H-5 became 6.30 and 2.70 Hz, respectively. Besides this conformational change no other spectral modification was observed during the variable temperature experiment. Irradiation of the protons at d 3.79 ppm (H-2 and H-7'') simplified two signals: (a) H-1 became a singlet, and (b) the doublet at 4.24 ppm belonging to H-7' collapsed to a singlet. H-7' resonates at lower field due to the weak anisotropic effect produced by the p orbital of C=N bond. COSY spectrum confirmed the suggested connectivities. The chemical shifts of the other protons are given in the experimental section.

The ¹³C NMR proton decoupled spectrum (in pyridine-d₅) of compound 12 provided all necessary carbon signals and permitted to assign the chemical shifts of all the carbon atoms. The carbonyl carbon and imine carbon atoms showed signals at d 190.81 and 142.60 ppm respectively. The anomeric carbon is assigned at d 100.35 ppm. That the oxygen atom of C-1 is attached equatorially to the cyclohexyl ring can be inferred from the chemical shift of ¹³C-1', which is at d 74.9 ppm. This is supported by the published values for cyclohexyl derivatives:²³ a carbon atom linked with an equatorial and axial alkoxy group resonate at ca d 71 ppm and ca 65 ppm respectively. Five carbon signals appeared between d 24.2 and 34.1 ppm, respectively. This indicates that C-2' and C-6', C-3' and C-5' carbons are diastereotopic. ²⁴ The chemical shifts of the other carbon atoms of the compound 12 are given in Table 1.

Molecular orbital calculations

The semi-empirical molecular orbital calculations (AM1) of compound 12 gave the expected conformation and relative configurations. The torsion angle (H-1)-(C-1)-(C-2)-(H-2) is -131.64° which clearly shows that the anomeric hydrogen is oriented equatorially. The molecular model shows that five carbon atoms of the pyranose ring are somewhat planar and the ring oxygen atom is above the plane. The other conformation where the anomeric hydrogen and CH₂ group are disposed axially gives the enthalpy of formation 1.69 kcal/mole higher than the first one where the anomeric effect dominates.

The calculations showed that the function O=C¾C=N behaves like an a,b-unsaturated ketone. The carbon and nitrogen atoms of C=N bond have the atomic charges of - 0.2087 and 0.0394 e.u. respectively. That this is true can be seen from the resonance forms given below (Figure 1).

The atomic charges, bond lengths, bond angles, and torsion angles of compound 12 are given in Table 2.

The calculations also showed the dihedral angle (H₈)-(N₈)-(C₇)-(H _7') as 16.86° and (H_7")-(C₇)-(N₈ )-(H₈) as 103.56°. With this, it is presumed that the inversion of the nitrogen N-8 is slow. The torsion angle (O₁₂)-(C₁)-(O₁₁)-(C ₅) of 75.35° also confirms the pseudoaxial disposition of O-12. The structure of the molecule 12 obtained by the semi-empirical molecular orbital calculations (AM1) is given in Figure 2.

Conclusions

We have achieved the syntheses of 12 (four steps and 32.5% overall yield) and 17 (six steps and 46.0% overall yield) starting from 1. The structure of 12 has been established with the help of IR and NMR spectroscopies and elemental analyses. Reduction of 12 as well as 15 did not lead to the expected diamino sugar 18. This difficulty was rationalized in terms of the partial negative charge at C-3, which impedes the approach of the reducing agents. Semi-empirical molecular orbital calculations using AM1 method provided the stable conformation of compound 12 and showed the negative electronic charge at C-3.²⁵

Acknowledgements

The authors thank Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Fundação de Amparo à Ciência e Tecnologia de Pernambuco (FACEPE), and CAPES-COFECUB for financial assistance, to Núcleo de Pesquisas de Produtos Naturais (NPPN), UFRJ, Rio de Janeiro, for recording the 200 MHz ¹H NMR spectra, and to M. L. de Oliveira for her help in the laboratory. We thank Profs. J. Jones Jr., Instituto de Química da UFRJ, RJ, and I. Malvestiti of our Department, for their help in determining the specific rotations. Our thanks are also due to J. B. P. da Silva, H. C.N. Batista and A. R. de O. Cavalcanti for their assistance in computational work.

References

1. Schatz, A.; Bugie, E.; Waksman, S. A.; Proc. Soc. Exptl. Biol. Med. 1944, 55, 66.

2. Kuehl Jr., F. A.; Flynn, E. H.; Holly, F. W.; Mozingo, R.; Folkers, K.; J. Am. Chem. Soc. 1946, 68, 536.

3. Umezawa, H.; Asian Med. J. 1968, 11, 69, and references cited therein.

4. Rinchart Jr., K. L. In The Neomycins and Related Antibiotics. E.R. Squibb Lectures on Chemistry of Microbial Products; John Wiley and Sons, Inc.: New York, 1964, p. 93.

5. Weinstein, M. J.; Ludemann, C. M.; Odem, E. M.; Wagen, G. H.; Rosselet, J. P.; Marquez, J. A.; Coniglio, C. T.; Charney, W.; Herzog, H. L.; Black, J.; J. Med. Chem. 1963, 6, 463.

6. Umezawa, S.; Advances in Carbohydrate Chemistry and Biochemistry; Academic Press: New York, 1974, 30, p. 111.

7. Okubo, S.; Nakamura, N.; Ito, K., Marumo, H; Tanaka, M.; Omura, S.; J. Antibiotics 1979, 32, 347; Okubo, S.; Nakamura, N.; Morimoto, M.; Mineura, K.; Marumo, H.; Omura, S.; J. Antibiotics 1980, 33, 221, 226, 231.

8. Hamada, Y.; Shioiri, T.; Tetrahedron Lett. 1982, 23, 1193.

9. Kusama, T.; Soga, T.; Tohgo, A. (Dauchi Pharmaceutical Co. Ltd.) Eur. Pat. Appl. EP 563, 08 (ClC 07H115/04), 1993; JP Appl. 92/74, 881, 31 Mar 1992; 37pp (CA 121: 9928h).

10. Nishimura, Y. In Studies in Natural Products Chemistry; vol. 10, Stereoselective synthesis (Part F), Atta-ur-Rahman, ed., Elsevier: New York, 1992, p. 495.

11. Papandreou, G.; Tong, M. K.; Ganem, B.; J. Am. Chem. Soc. 1993, 115, 11682.

12. Jain, S.; Surywanshi, S. N.; Bhakuni, D. S.; Ind. J. Chem. 1993, 26b, 866.

13. Dewar, M. J. S.; Zoebisch, E.G.; Heally, E. F.; Stewart, J. J. P.; J. Am. Chem. Soc. 1985, 107, 3902.

14. Stewart, J. J. P.; J. Comp.-Aided Mol. Design 1990, 4, 1.

15. Stewart, J. J. P.; MOPAC 93-00 Manual, Fujitsu Limited: Tokyo, Japan, 1993.

16. Fraser-Reid, B.; McLean, A.; Usherwood, E. W.; Yunker, M.; Can. J. Chem. 1970, 48, 2877.

17. Holder, N. L.; Freser-Reid, B.; Can. J. Chem. 1974, 51, 3357.

18. Fatiadi, A. J.; Synthesis 1976, 65.

19. Fieser, L. F, Fieser, M.; Reagents for Organic Synthesis; vol. 1, John Wiley & Sons, Inc.: New York, N.Y., 1967, p. 191.

20. Srivastava, R.M.; Carthy, B.J.; Fraser-Reid, B.; Tetrahedron Lett. 1974, 2175.

21. Ferrier, R. J.; Prasad, N.; J. Chem. Soc., C 1969, 570.

22. Ferrier, R. J.; Adv. Carbohydr. Chem. 1969, 24, 199.

23. Kalinowski, H.- O.; Berger, S.; Braun, S.; ¹³C-NMR-Spektroskopie; Georg ThiemeVerlag: Stuttgart, N.Y., 1984, p. 240.

24. Eliel, E. L.; Wilen, S. H.; Stereochemistry of Organic Compounds; John Wiley & Sons, Inc.: New York, N.Y., 1994, ch. 8.

25. Taken in part from the M. Sc. Thesis; Ana M. A. de Souza, Universidade Federal de Pernambuco, Brazil, 1986.

Received: October 11, 2000

Published on the web: January 11, 2002

Publication Dates

• Publication in this collection
  25 June 2002
• Date of issue
  2002

History

• Accepted
  11 Jan 2002
• Received
  11 Oct 2000