OCTOBER 1967
NMRSTUDIESON ACETYLATED ~-THIOALDOPYRANOSE DERIVATIVES
133'; "::A: 266 mp (e 11,750); A::,",' br, bonded absorption 3.04.5, 5.92, 6.2-6.3 (br), 7.2,and 7.4 P; nmr in deuterioacetone T 5.04 (s) -CHn, T 8.28 (s) -CHs; positive ferric chloride test. A n a l . Calcd for CsH602Br2:C, 26.69; H, 2.24; Br, 59.21. Found: C, 27.45; H, 2.55; Br, 58.79.
3077
Registry No.+ 899-79-6; 5 , 13865-85-5; 6, 1386586-6; 7, 13765-87-7; 8 , 13865-88-8; 8a, 517-0643; g (X = Br), 13865-90-2; 9 (X = Cl), 13865-91-3; 9 (X = FSOJ, 13865-92-4; 10, 13865-93-5.
Nuclear Magnetic Resonance Studies on Acetylated 1-Thioaldopyranose Derivatives''2 CHARLES v. HOLLAND, DEREKHORTON,3 MARTHA J. MILLER,' Department of Chemistry, The Ohio Stale L'niversity, Columbus, Ohio @ZiO AND
NORMAN S. BHACCA
Department of Chemistry, Louisiana State Unitersity, Baton Rouge, Louisiana
70803
Received '4pril 10,1967 The nmr spectra of the fully acetylated I-thioaldopyranoses having the configurations P-D-ZY~O ( I ) , a-L-arabino (2), 0-D-rib0 (4), p-D-gluco ( S ) , and p-D-galacto ( 6 ) were determined in chloroform-d, acetone-de, and benzene-ds.
5 to higher field than its position in the 1-oxygenated The H-1 signal in these derivatives appears ~ 0 . 3 ppm analogs. The relative chemical shifts of the various ring protons were sufficiently different in the three solvents to permit useful conformational and configurational information to be derived by partial first-order analysis of spectra measured a t 60 MHz. Spectral measurements a t 100 MHz were required for first-order analysis of the spectra of the hexose derivatives 5 and 6 . First-order analysis of the signals of the methine protons, in the WLarabino derivative 2 (or its D enantiomorph, 3), was not possible at 100 .MHz with any of the three solvents; complete first-order analysis was, however, possible when the spectrum was measured a t 220 MHz in chloroform-d. Comparative spectral data are recorded for a series of S-substituted analogs (7-10) of substance 6 , and 4,6-di-0acetyl-l-S-acetyl-2,3-dideoxy-l-thio-a-~-erythro-hex-2-enopyranose (11) is shown to adopt the Hi conformation.
Investigations in this laboratory on the reactions of thio sugar derivatives with halogens have shown2 that the progress of the reactions can be followed conveniently by nmr spectroscopy. Acetylated 1thioaldose derivatives react with bromine to give acetylated glycosylsulfenyl bromide^,^^^^ acetylated glycosyl bromides, and other products, according t o the conditions. The nmr spectra of a range of acetylated aldopyranosyl bromides have been analyzed6 in terms of configurational and conformational factors, and it has been shown2 that these products are readily detected in the mixtures of substances formed when 1-thioaldose derivatives are treated with bromine under various conditions. The present report describes a comparative analysis of the nmr spectra of a series of acetylated l-thioaldopyranoses and some related derivatives, a t 60, 100, and, in some cases, a t 220 hIHz. The results illustrate the use of solvent effects as an aid in spectral analysis, for reducing the signals of methine and methylene protons to patterns that are amenable t o first-order interpretation. Such interpretation is particularly facile for a proton (or protons) attached to C-5 of the pyranoid ring, and the derived coupling constants are especially useful for providing information on conformation and configuration. (1) Supported in part by the Agricultural Research Service, U. S. Department of Agriculture, Grant No. 12-14-100-7208 (71) (The Ohio State University Research Foundation Project 1827) administered by the Northern Utilization Research and Development Division, Peoria, Ill. The 60-MHz nmr spectrometer was provided through a grant from the National Science Foundation. (2) Preliminary reports of parts of this work have been given: (a) C. V. Holland, D. Horton, M. J. Miller, and W. N. Turner, Abstracts, 149th National Meeting of the American Chemical Society, Detroit, Mich., April 1965,p 1D; (b) D. Horton and M. J. Miller, Carbohydrate Rea., 1,335 (1965). (3) To whom inquiries should be addressed. (4) Undergraduate Research Participant, 1964-1965. (5) R. H.Bell, D. Horton, and M. J. Miller, t o be published. ( 6 ) D. Horton and W. N. Turner, J. Or8, Chem., 80, 3387 (1965).
Materials
1-Thio-0-D-glucopyranose pentaacetate (5) was prepared from tetra-0-acetyl-a-D-glucopyranosyl bromide and potassium thio1acetate?v8 1-Thio-P-D-galactopyranose pentaacetate (6), 1-thio-0-D-xylopyranose tetraacetateg (l), 1-thio-a-L-arabinopyranose tetraacetate (2), and the a-Danalog (3) of 2 were prepared similarly. Low yields of product were obtained when this procedure was used t o prepare 1-thio-p-D-ribopyranose tetraacetate (4) from tri-0-acetyl-p-D-ribopyranosyl bromide, and it was difficult to remove an accompanying side product. Condensation of tri-0-acetyl-8-D-ribopyranosyl bromide with thiourea, to give 2-(2,3,4tri-0-acetyl-P-~-~~~bopyranosyl)-2-thiopseudourea hydrobromide, followed by cleavage ,Of the S-amidino group by the general procedure of Cernf, VrkoE, and Stan;k,lo with subsequent reacetylation, gave pure 4. The same route was also used to prepare 2 (an 3'0) by way of 2-(2,3,4-tri-O-acetyl-a-~-arabinopyranosyl)2-thiopseudourea (and its D enantiomorphlO). The enantiomorphs 2 and 3 had the anticipated opposite signs of rotation, and were each obtained in two dimorphous forms, one melting at 39" and the other a t 81.5-82". Melting points of 79" and 80-81" lo have been reported for substance 3. The anomeric configurations assigned to the products are those anticipated t o result from attack by the sulfur nucleophile on an intermediate, 1,2 cyclic acetoxonium ion during the condensation step. Nmr and optical rotatory data provide firm support for the anomeric assignments. Each of the products 1-6 showed absorption a t 5.855.90 pm in its infrared spectrum, characteristic* of the (7) J. F. Danielli, M. Danielli, J. B. Fraser, P. D. Mitchell, L. N. Owen, and G. Shaw, Biochem. J . , 41, 325 (1947). (8) D. Horton and M. L. Wolfrom, J.
[email protected]., 97, 1794 (1062). (9) M. Gehrke and W. Kohler, Ber., 64,2696 (1931). (10) M.bern?, J. Vrkoz, and J. Stanzk, Collection Czech. Chem. Commun.. 14,134 (1959).
HOLLAND, HORTON, MILLER, AND BHACCA
3078
...
..
1
A
.
o
AcO H
H
H
OAc
4
3
H
H ~
h
skAcO Ac 0
H
2
AcO
,
H
U
I OAc
H
VOL.32
A SAC
H
5
AcO
C
b
AcO
H
6
o
H
AcO
7
R - z CH2Ph
8 9
R-Bz
H
H
11
R = CSOEt
10 R = W - B U S-acetyl group, in addition t o the 0-acetyl absorption near 5.75 pm. Benzyl 2,3,4,6-tetra-O-acety~-l-thio-~-~-glucopyrano&del1 (7) was obtained in high yield by treatment of tetra-0-acetyl-a-D-glucopyranosyl bromide with potassium a-toluenethioxide in ethanol, followed by reacetylation of the product, which was partially saponified during the condensation. This general procedure, satisfacttory for preparation of 2,3,4,6-tetra-O-acetyl-lS-benzoyl-l-thio-P-D-glucopyranose12v1a ( 8 ) and tetra-0acetyl-P-D-glucopyranosyl ethyl~anthate'~ (9) without the acetylation step, was unsatisfactory as a preparative route to t-butyl 2,3,4,6-tetra-O-acetyl-l-thio-P-~glucopyranoside (10) when sodium 2-methyl-2-propanethioxide was used even when the product was Figure 1.-The low-field portion of the 100-MHz spectra of reacetylated. Mercaptolysis of P-D-glucopyranose pen1-thio-8-D-xylopyranose tetraacetate (1 ) in chloroformd, acetonetaacetate with 2-methyl-2-propanethiol and zinc chlode, and benzeneds. ride, by the procedure of Fletcher and co-workersl6 gave 10 in good yield. 4,6-Di-O-acetyl-l-S-acetyl-2,3I dideoxy-l-thio-a-~-erythro-hex-2-enopyranose (11) was prepared from D-glucal triacetate and thiolacetic acid, as described by Tejima and co-workers.16
Results and Discussion CDC13
The nmr spectra of the acetylated 1-thioaldopyranoses l d were measured a t room temperature in chloroforma!, acetone-&, and benzene-&. Substances 2 and 3 gave, as anticipated, identical spectra. All spectra were determined a t 60 and also a t 100 MHz. The low-field portion of the 100-MHz spectrum for each of (11) W. Schneider, J. Sepp, and 0. Stiehler, Ber., 61, 220 (1918). (12) J. Kocourek, Collection Czech. Chem. Commun., PS, 316 (1964). (13) W. Schneider and A. Bansa, Ber., 64, 1321 (1931). (14) W. Schneider, R. Gille, and K . EiBfeld, i b i d . , 61, 1244 (1928). (15) H. B. Wood, Jr., B. Coxon, H. W. Diehl, and H. G . Fletcher, Jr., J . 078. Chem., 29, 461 (1964). (16) T. Maki, H. Nakamura, 8.Tejima, and M. Akagi, Chem. Pharm. BUZZ. (Tokyo), 18, 764 (1966).
v-L
.
J
4.5
5.0
5.5
6.0
6.5
t
Figure 2.-The low-field portion of the 220-MHz spectrum of 1-thio-a-Larabinopyranose tetraacetate (2) in chloroformd.
the three solvents, is shown for 1 (Figure l), 3 (Figure 3), 4 (Figure 4),5 (Figure 5 ) , and 6 (Figure 6). The spectrum of 2 in chloroform-a! was also measured a t 220 MHz, and the low-field portion of this spectrum is shown in Figure 2. The lowest field portions of the 220-MHz spectra of 1 and 5, in acetone-de, are shown in Figure 7, and Figure 8 shows the 220-MHz spectrum
NMRSTUDIES ON ACETYLATED 1-THIOALDOPYRANOSE DERIVATIVES
OCTOBER 1967
II
cm5
5.0
a0
65 5
Figure 6.-The low-field portion of the 100-MHz spectrum of l-thio-@-D-galactopyranose pentaacetate ( 6 ) in chloroformd, acetone&, and benzene&.
I
4.5
u
5n
4.5
5).
3079
6.51
6.0
5.5
Figure 3.-The low-field portion of the 100-MHz spectrum of 1-thio-a-D-arabinopyranose tetraacetate (3) in chlofororm-d, acetone-&, and benzene-&.
1 I
U
100 MHz
H
SAC
OAc
H
I
45
4.6
4.7
4.0
4.9
5.0
53
5.2 T.
Figure 7.-The signals of H-1, 2, 3, and 4 of l-thio-@-D-xylopyranose tetraacetate (1) and 1-thio-8-D-glucopyranose pentaacetate ( 5 ) in acetone-& a t 220 MHz. 45
50
u
80
6.3 T
Figure 4.-The low-field portion of the 100-MHz spectrum of 1-thio-&D-ribopyranose tetraacetate (4) in chlofororm-d, acetoneds, and benzene-ds.
' Y / / k , ' . ' ' ' ' low-field portion of the 220-MHz spectrum of 4,6-di-0-acetyl-l-S-acetyl-2,3-dideoxy-l-thio-a - D erythro- hex- 2The scale divisions correspond enopyranose (11) in acetone& to 10 Hz. "
Figure %-The
Figure 5.-The low-field portion of the 100-MHz spectrum of 1-thio-8-D-glucopyranose pentaacetate ( 5 ) in chlofororm-d, acetonedo, and benzene-&.
of the unsaturated derivative 11, in acetone-da. Chemical shift data for the methine and methylene protons, taken from the 100- and 220-MHz spectra, are given in Table I. First-order coupling constants are given in Table 11, and chemical-shift data for the acetyl methyl protons are given in Table 111.
-
Signals of the Acetyl Methyl Groups.-In all of the spectra measured in chloroform-d or acetone&, a characteristic2 three-proton singlet, assigned to the (equatorial) S-acetyl group, is observed near T 7.65. Signals of the acetoxy groups? in these two solvents, showed resonances in the T 7.93-8.11 range; in the case of substances 2 (in chloroform-d), 4, and 6, a threeproton singlet at somewhat lower field was also observed. The latter signal may be attributed to an axial acetoxy group, 17-1@ although such an assignment can be regarded as tentative onlyi9 and does not constitute unambiguous evidence, such as would be provided by synthesis of specifically trideuterioacetylated (17) R . U.Lemieux, R. K. Kullnig, H. J. Bernstein, and W, G . Schneider, J . A m . Chem. SOC.,80, 6098 (1958). (18) L. D. Hall, Adoan. Carbohydrate Chem., 19, 51 (1964). (19) R . U. Lemieux and J. D. Stevens, Can. J . Chem., 48, 2059 (1965).
HOLLAND, HORTON, ~LIILLER, AND BHACCA
3080
TABLEI CHEMIC.4L SHIFTS O F METHINEAND METHYLENE PROTONS
O F PERACETYLATED
7
Compd
Confign
Solvent CDCli {EgabCO
1
@-D-Z$O
a
a-barabino (or a-n-arabino)
H-1 a
H-2 a
4.62 5.00 4.60 5.04 4.49 4 . 8 1
CDCla
(or a)
CDCls
4
@-D-Tibo
CDCli [g:a)zCO
8
8-D-OlUCO
CDClr {gghCO
8
&D-walacto
{,@C:iCO
c H - 3 e a
4.65' 4.81'
7-H-4e
4.78 4.75 4.63 4.57-4.95c 4.89' 4.464.98' 4.40-4. 7SC
4.74'
CDClr
4.55-4.76' 4.55 4.33
4.93 4.93 4.99
4.59-5. ooc 4.63 4.4&4.93C
4.62 4.97
a 5.08 5.11 5.02
4.38 4.94 4.51 4.41 4.96 4.52 4.18 4.80 4.39
4.52-4.94' -4.76 4.73
~
1
@-D-X$O
2
a-L-arabino
4
p-D-ribo
5
@-D-glUCO
6
8-D-galacto
8.6d 8.1 e 7.0d e e 7.7 7.6 7.9 e 10.0
8.2' 7.9 e 8.0d
e e 3.0 3.1 3.1 e 9.1
5.84(5.86)) 5.90(5.91)) 6.03(6.03))
6.46(6.45)) 6.40(6.39)b 6.73(6.72))
6. 04d (13.05)~
6.26d (8.25))
6 , 04d
6.2Bd
6.05d 6 . 2Zd (6.24)'
e1.19~ 6 . 60d (6. 59)b
6 . 0 2 (6.O6lb 6.03(6.05)b 6.15(6.19)b
6 . 1 7 (6.13)b 6.20(6.18)b 6.28(6.23)b 6.12 -5.98 6.37
5.68 6.28
-4.54 4.50
--H-6-
-H-6'-
5 . 7 1 (5.73)) 6.74 (5.76)' 5 . 7 5 (6.77)'
5.92 5.95 (5.92)* 5.88(5.96)'
-5.8905 . 8 7 (5.91)) 5 . 9 9 (5.95)' 5 . 8 3 (5.84)* 5.97 (5.95))
Shifts given in parentheses were calculated by ABX The two H-5 signals are not unambiguously differenti-
4.gd 8.6d 11.9d 8.8 (8.8)c 11.8 4.9(5.0)" 4 . Y (4.6)" 2.4, (2.0)" 12.2 4.7dJ (4.6)ezd 2.3d4f (2.4)Ccd 1 2 . j d 3.0f 12.6 3.6f 2.0(1.7)c 12.4 3.9(4.1)" 7.5(9.0)" 11.8 4.7(3.1)c 12.0 4.6(4.4)" 7 . 7 (8.4)" 7.5(10.1)" 11.7 5.2(2.5)" 9.4 4 . 4 (4.9)" 2.2 (2.2)" 9.3 5.0 (5.6)c ~2 (1.7)"
(CD3)zCO )C6D6 CDCli
7.66 (3)
CDC18 (CDI)&O CsD6 'CDC13 (CDa)&O CsDe 'CDC13 (CD3)zCO CsDs ,CClab
7.58 (3) 7.60 (3)
