Preparation of Glycals 7. From Glycosyl Bromides

Acetylated glycosyl bromides are the most often used starting materials for the synthesis of a variety of glycals. The original method of preparation,¹⁶⁹ i.e., the reaction of acetobromoglucose with zinc dust in (buffered) aqueous acetic acid at a temperature ranging from - 2 0 °C to room temperature.¹⁷⁰ has undergone countless modifications. A collection of glycals obtained by one of these protocols is presented in Scheme 20 and Table 12.

The starting glycosyl bromide can be isolated before the glycal-forming reaction. However, because aceio-bromosugars are rather sensitive and cannot be stored for long times after isolation, the preparation of glycals has been performed in one continuous operation.¹⁷⁰¹⁷¹ This involves formation of a glycosyl bromide from the free sugar by aceiylation and subsequent acidolysis of the anomeric acetates by hydrobromic acid, yielding a crude mixture which is directly subjected to elimination in the presence of zinc dust. In situ preparation of the starting bromide is also effected from the per-O-acylated or 1-0-acylated sugars. Various methods can be used for enhancing the activity of zinc.¹⁷² Most frequently zinc—copper¹⁷⁰ or zinc-platinum^{1 7}' couples are pre-pared by adding copper(II) sulfate or platinic chloride, respectively, to the reaction mixture. The reaction medium (acetic acid and water in various ratios ranging from 1:1 to 9:1) is usually buffered bv sodium or ammonium acetate. Illustrative results obtained by these modifications are shown under method A in Table 12. Method B is a recent update of A with respect to the formation of tlie glycosyl bromide from a free sugar in the one-pot procedure.¹⁷¹ Method A was also applied for the preparation of D-glucals 191.¹⁷¹ 192.¹⁷⁵ 193.¹⁷⁶ and 194.¹⁷⁷ n-Allal 199 was obtained from the ri-aliropyrnnosyl¹'⁸ as well as from the D-allopyranosyl bromide. '⁰ and 0-gulal^1V8 200 was obtained from n-idopyranosyl bromide. i.-Rham-nal 206.¹⁶⁰ glycals of uronic acid esters 208¹⁸¹ and 209.¹⁸² as well as of 5<ihio-i)-glucose 210¹⁸' were also made in this way. It is seen that depending on the sugar configuration yields of the glycals vary from excellent to poor. A reason for the low yields can be the ready solvolysis of the reactive glycosyl bromide in the applied medium; therefore, replacement of water with teirahydrofurari¹⁸⁴ (method C) or the use of vitamin B-12 as a special mediator in methanol¹⁸¹ (method D) providing essentially neutral conditions was suggested. Acetic acid could similarly be omitted, and high-yieldingglycai formation could be achieved in aprotic media with highly reactive zinc—silver/

graphite¹⁸⁶ (method E) or with simply activated zinc in the presence of a /V-base¹⁸⁷ (method F). 1-Methyl-imidazole (MIM) or 4-methyl-pyridine (4-Pic) proved to be the best /V-bases, and besides ethyl acetate benzene, tetrahydrofuran, acetone, and dichloro-methane were also applicable as solvents. 5-Thio-xylal¹⁸⁸ 211 and 5-thio ribal¹⁸⁸" 211a, used for the synthesis of oral antithrombotic agents, as well as 216.¹⁸³ an intermediate in die synthesis of fiuorinated ligands to probe binding of antigenic determinants

Carbanionic Reactivity of the Anorectic Center in Carbohydrates S c h e m e 20

108 Chemical Reviews. 2001, Vol. 101, No. 1 Somsak

ft"

tor details of conditions see labia 12 or text

<RO)„ OR • O

("OL S

195 196 197

162 R¹ = R² = R³ = OAc ^{A c}?

= R² = R³ = OBz ^{A C}° ' f\ D

183 R'

= R² = R³ = OAc ^{A c}?

= R² = R³ = OBz ^{A C}° ' f

184 R¹ = R² = R³ = OPiv (90 % H) A c O - 4 185 R¹ - R² - R³ « OMe (96 % G)

186 R¹ = R² = R³ = OBn (96 % G) 198 187 R¹ = OBz. R² = R³ = OAc (96 % G)

188 R ' = OTs, R² - R³ = OAc 189 R' = OBn. R² = R³ = OAc (95 % G) 190 R' - OTBDPS, R² = R³ = OAc (88 % G) 191 R = SAc. R² = R³ = OAc (87 % A)

192 R ' - R² - OAc, R³ - OBn A c o 4R ?

193 R¹ = R² = OAc. R³ = OMe (51 % A)

194 R¹ = R³ = OBz. R² = OMe (62 % A) 201

R¹ = h ,R^! = Ph. R³ = Ac (91 % G) R¹ = H, R² = Me, R³ = Ac (84 % G) R¹ = R² = Me. R³ = H (90 % G)

AcO AcO \ a A C O -K a

P

AcO AcO

199 200

(12 % A from D-a/iro-Bf) ( 1 1 % A from (76 % A from o-elfo-Br) o-ido-Br)

AcO

.0

AcO

XXV

A c 0 OAc

OR² 204 R¹ - R² - Ac 205 R ' = R² = Bz 20« R¹ - Me, R²- A c

( 6 8 % A)

207

R'COjMe

208 R' = H,

R² = OAc (72 % A) 209 R¹ = OAc.

R² « H (51 % A) AcO AcO

" e r - r V i

5 - X j y 210 R - CHjOAc

( 6 9 % A) 211 R = H

( 8 5 % F)

A c O - p

AcO

211«

(95 % F )

R = Bz R = Ms or Ts

I OBz

BzO \

(79 % F} 212 R = H. R² = OBz 2 1 4 ( 1 1 % F ) CWc (51 % via R = Ms Al (80 % F) 213 R = 0 B Z R²- H 215 ( 9 % Fl

AcO X ^ f

of Vibrio cholcrae. were also prepared by this last procedure.

Ketohexoses can provide exo- or endo-glycals as shown for the D-fructals 212 and 214 as well as L-sorbals 213 and 215 prepared by method F. The elimination favors formation of the exocyclic double bond also with method A giving 212 and 214 in 7:1 ratio.¹⁹⁰ To direct the elimination toward the

endo-position, the leaving ability of the 3-O-substituent had to be enhanced by introducing 3-O-ntesyl or 3-0-tosyl groups in diacetone-fructose. Subsequent ex-change of protecting groups, formation of the ano-meric bromide, and subjecting it to the conditions of method A gave 214 in 51% overall yield for the whole sequence, while the acetylated analogue was obtained in 53%.¹⁹⁰

Attempts to get furanoid glycals by method A were less successful. The desired products could be ob-tained in low yields, and the main products resulted from the hydrolysis of the starting furanosyl bro-mides.¹⁹² The acid-sensitive furanoid glycals could be acquired front furanosyl bromides by introducing a good leaving group in the 2-position and using sodium iodide under neutral conditions as shown in

eq 6.^;'" ¹⁹¹ The method was also used for the directed preparation of endo-^t9> (e.g., 2 1 4 ) and e.voglycals¹⁹"

of keioses (cf. compound 232 in Scheme 22).

R O - RO

Br Nat

a c e t o n e * RO 03SC6H4NOjp

> <

⁶

>

RO R = Bz (pBrf 7 2 % R = p M e O C0H4C O 6 8 %

After early speculations.¹⁹⁶ detailed mechanistic studies on glycal formation were carried out with methods A and F in the past decade. Gas chromato-graphic analysis and identification of byproducts (in addition to solvolytic ones) formed in reactions of several acetobromopyranoses performed with method A revealed the intermediacy of glycosvlium ions.¹⁹⁷ On the other hand, similar analysis showed the presence of /?-D-glucose pentaacetate as the sole very minor byproduct formed from acetobromoglucose under the conditions of method F, and inhibition and trapping experiments suggested the involvement of glycosyl radicals in such reactions.¹⁹⁸ The most

X <

Table 12. Preparation of Pyranoid Glycals from Glycopyranosyl Bromides (see Scheme 20)

method (yield [%| (configut •atIon of tile starting compound))

A I 8 U . I 7 0 Q U I ( I K I Dt«r, ¡ 7 I K 1 1 p I K 7 ( ; 2 0 0 , 2 I W , | 2 . M ¡ 2 1 1 7

Zn, ArOH-HzO Zn. ArOH-H/O Zn, THF—AcOH 9:1. Zn. NHiCI, MeOH. Z.n-Ag/C. THF. Z.n. M1M, (CpzTi"'CI)z Cr"KDTA Al-Hg. THF, glycal (classic) r.l. (Improved) 0 "C lo r. i. Vlt. B 12. r.l. 20 "C to r.l. KtOAc, reflux THF. r.l. DMF—H20, r.l. 0 °C to r.l.

182 60-70'⁷"" 98 80-90 94 [n-gtucdi

95 (t) iruirind) 90" 95 187

g¡)20O.2OH 91¹⁹⁹ in situ"

(D-glued)

942(10.202

8 7 ' " In situ"(n-/»anno)

87 85 (D-gluco) IS (n monno)

183 100'( 10U>0°C) 100'( 10U>0°C) 80 90 89"" 88 60 (D-gluco)

65 (D-manno)

188 8 7 " ( - 5 t o 0 ° C ) 81¹⁸⁷

198 6 6^m P f r . t . 59' Cu (-10 to 0 °C)

58 80-90 93 ^{8 7}i»i 95¹⁸⁷ 8g200.202

77'"" in situ"

71 60

201 61*

35"

80-90 92"® i-k 8g?00. 202 ⁶⁶

202 60' (0- 10°C) 51 80 90 IJ

203 60' (0- 10°C)

80-90 00iw, B5'^B7 4-Plc/Phl I'™ ₇Q200. 202 90

204 8() 170,,

80"

93"

71 4 5 84"® In situ" 82

205 79" 80

207 40"

30'

83 60' s

72' Py/PhH 82¹⁹⁹ In situ

" From the free sugar at - 1 0 to 20 °C. "With in situ prepared reagent. ' Lundt I.; Pederscn. C. Acta Chem. Scant!. 1970, 24. 240-246. ''Nagabhushan, T. L. Can. J. Client 1970.

48. 257-261. ' Using plutinic chloride activation. 'Using copper(II) sulfate activation: Rosenthal. A.: Read. D. Mcth. Carbohydr. Chem. 1963, 2. 457-402. * From the free sugar at - 10 °C: Wcygand, 1\ Methods Carbohydr. Cbem. 1962. /, 182-185. " From the tetraacetate at 0 "C: Banner, P.: Boxler. D. L.: Brambilla, R.; Mallams. A K.: Morton, J. B.; Relchert, P.; Sanclllo, I'. D.; Surprenant. H.; Toinalesky. G.: Lukacs, G.: Oleskei. A.; Tliung. I I : Vulenle, I..: Omura, S. J. Chern. Sue.. Perk In Trans. 1 1979. 1000 1622. 'Yield unknown:

Marco-Contclles, J.: Ruiz. J. Tetrahedron Lett. 1998. 39. 6393 6394. 'Yield unknown: Pérez-Pérez, M. .).: Doboszewski. B.: Rozenski, .).: Herdewijn, P. Tetrahedron: Asymmetry 1995, 6. 973-984. ⁴ Yield unknown: Doboszewski, B.; Blaton. N.: Herdewijn. P. ./. Org. Chem. 1995, 60. 7909-7919. 'Humoller, F. L. Methods Carbohydr. Chem. 1962, 1. 83-86.

"' By using 4-mcthylpyrlriine in refluxlng benzene. " From I he i el rant el ale at - 10 C Isolln. B.: Relchstein, T. Heir. Chlm. Acta 1944. 27, 1146-1149. " Hadfleld. A. F.; Cunningham.

I. ; Sartorclll. A. C. Carbohydr. Res. 1979, 72. 93-104. "Frum the free sugar at 0 to - 1 0 "C: F.I Khadem, H. S.: Swailz, D. L.; Nelson, J. K.: Betty, 1.. A. Carbohydr. Res. 1977. 58.

230 234. ' From the tetraacetate at 10 °C: Iselln. B.; Relchstein. I. Ilelv. (him Ana 1944, 27. 1200 1203. ' Yield unknown: Pudlo, P.; Thiern, J.; VIH, V. Chem. Ber. 1990. 123.

1129-1135. ¹ Yield unknown: Ludewig, M.: Thlein, J. Lin J. Org. Chem. 1998. 1 189- 1191. 'By using pyridine In refluxlng benzene: Bajza, I. Personal communication.

£ 3

Carbanionic Reactivity of the Anomeric Center in Carbohydrates S c h e m e 21

T V ^ N ^ R N

-( A C 0QW ^ O A c ^{, A i}%A^Xc

UAC protic ^U apmtic L ^UAC

R = OH. OAc I ^{r e ' /}

T Y / '

F O I - 0 - I Ï O - O *

(AcOL ®OAc (ACO Î T Î I IFACOIN (AcOy, "OA

l 2 e

OAc

2e'\

r o i r —

F « » ¿ « j

0

-AcO

O Q G

(AcOln.. OAc (AcO)n OAc (AcOk,

S c h e m e 22

R'CL n R²0 OR¹

2 1 7 R ' = It. R² = R³ = OAc 2 2 5 R ' - R³ = OBz, R² « IV 218 R¹ = IU R² = R³ = OAc 226 R¹ = R² = OAc, R³ = IV 219 R¹ - R³ - OAc. R² - II

220 R ' = R³ = OAc R² = I 221 R¹ = R³ = OAc. R² * IV 222 R ' = R³ = OAc R² = V 223 R ' - R² = OAc, R³ = I 224 R¹ = R² = OAc R³ = II

OAc

OAc

II "*>

A S

AcO -OAc AcO .OAc

bonds 227 e p R = CI CH,R 228 e f> R = H 229 a 8 R = O T s f 230 e a R = O T s

a c e t o n e 231 R¹ = R² = Bz 232 R ' = Ms. R² = Ac

important mechanistic conclusions are summarized for both the protic (method A) and the aprotic

(method F) reaction in Scheme 21.

Formation of the glycosyiium ion may be ascribed to the highly polar, ionizing, and dissociating solvent (average dielectric constant for AcOH—H20 1:1 : e ~ 42) used in method A. This ion can be reduced or rearranged before reduction to give carbanionic

spe-108 Chemical Reviews. 2001, Vol. 101, No. 1 Somsak

cies both at C-l and C-2, respectively. The anomeric anion can be protonated to give the dehalogenated product or it can eliminate an acetate to give the glycal. The C-2 anion can be protonated, resulting in a 2-deoxy derivative, or eliminate an acetate, giving the glycal or a 2,3-unsaturated sugar. In the nonsolvolytic and apolar medium of method F (for the solvents used 2 < e < 7), a (possibly dissociative) electron transfer results in the formation of glycosyl radicals which rearrange rather slowly. Therefore, a second electron cransfei gives the anomeric anion exclusively. This can only get stabilized by hie elimination of in acetate because protonation is impossible in the aproLic medium. It was suggested that these processes lead to a chain reaction. This ran explain the usually very high purity (>95%) of the raw product from transformations by method F.

It is not to be excluded that the radical pathway is operative under protic conditions as well. The con-siderations on solvent polarity are corroborated by the results obtained with method C (for the applied solvent mixture « ~ 7). The excellent yields afforded by method D may reveal that vitamin B-12 directs the reaction to proceed via radical intermediates either by an electron transfer to the substrate from a Co(I) species or by the formation of a readily homolyzable glycosyl—cobalt intermediate.

The titanium(III) species (Cp2TiCl)2 (also generated in situ by manganese)¹⁹⁹ is an excellent one-electron transfer agent, and thus, it can direct the reaction to proceed via radicals (method G). The aprotic and neutral conditions allow the use of various protecting groups in highly reactive glvcosyl bromides as il-lustrated by the high-vielding preparation ofglucais 185-187, 189. 190. and 195-197.²"" ²⁰² A 2-deox\

glycosyl titanium(IV) compound was isolated as an analogue of the intermediate in lite above reactions An authoritative survey was published on : he use oi Ti(III) reagents in carbohydrate chemist t\ d"²'¹ The methylated glucal 185 was also obtained from the corresponding bromide with sodium naphthalenide in tetrahydrofuran in 41% vield.²⁰³

Chromium(II) is known to generate radicals from alkvl halides by an inner-sphere electron transfer.

In aqueous medium the chromium(II)aqua complex does not react with acetobromoglucose. I lowever, the reactivity of this ion can be enhanced (method 11) by complexation, especially with ethylenediaminetetra-acetic acid (EDTA) to produce glycals in excellent yields even in highly polar aqueous medium²⁰⁴ (t ~ 58 for a D M F - H20 1:1 mixture). This reflects that glycal formation can be the main pathway even in a solvolytic milieu if the radical-forming step is faster than solvolysis. The intermediate glycosyl chromium-(III) species have rather long lifetimes as shown by UV—vis spectroscopy.²⁰⁵ The reaction of acetobromo-glucose with Cr(II)TMEDA generated in situ from chromium(III) chloride with manganese was also investigated.²⁰⁶

Aluminum amalgam proved a good alternative to the zinc-based reagents illustrated under method I.²⁰⁷ Electrolysis of acetobromoglucose with mercury pool electrodes in acetonitrile gave 182 quantitatively.²⁰⁸ The reaction of acetobromoglucose with samarium

XXV

yield.²⁰⁹²¹⁰

Disaccharide glycals (Scheme 22) such as gentiobial 217. melibial 218. cellobial 219, maltal 220, and lactal 221 prepared by method A were surveyed.³⁴ . Newer examples for the application of method A with X X V ^{d l}" ^{a n d} ""¡saccharides are 2 2 2^{2 1 1} (93%), 2 2 3^{2 1 2} (65%).

laminaribial 224 (54%,²¹³ 70%²¹⁴), a pyruvated lac-tal²¹⁵ (98%). the galactosylated galactais ^{2 2 52 1 6} (88%) and 2 2 6^{2 1 7} (92%). the 6,6'-difunctionalized cellobi-als²¹⁸ 227 (98%) and 228 (92%), lactal 229 (65%) and maltal 230 (60%) derivatives,²¹⁹ and per-O-acetylated maltotrial²²⁰ (42%). Method B gave per-O-acetylated maltal 220 (86%). lactal 221 (61%). and maltotrial (50%) from the appropriate free sugars.¹⁷¹ From the corresponding acetobromo disaccharides were pre-pared cellobial 219 by methods E (83%),¹⁸⁶ F (75%).¹⁸⁷ G (91%),¹⁹⁹ and H (95%)²⁰⁵ and maltal 220 (94%) by method G.¹⁹⁹ Method E gave the exocyclic glycal of leucrose 231 (42%), but this compound was obtained in 65% yield by method A. The analogous 232 was prepared by the sodium iodide protocol in 50%

yield.²²¹ Method F was also used for the obtention of lactal²²² and maltal but yields were not indicated.²²³ 2. From Glycosyl Chlorides

Glycosyl chlorides are less reactive than bromides, and from among the methods listed in the previous section, zinc-silver/graphite- and Cr(II)-based re-agents are applicable/have been tested for the con-version of these starting materials into glycals.

Acetvlated pvranoid derivatives (Scheme 20) 182 (93%), 201 (82%). and 203 (97%) were obtained with a Cr(II)EN complex in DMF after reacetylation²²⁴ and 182 (79%), 198 (91%), 201 (72%). 203 (90%), and 204 (61%) with Cr(II)EDTA in water-DMF²⁰⁴-²⁰⁵ from the corresponding glvcopyranosyl chlorides.

As it was seen, synthesis of furanoid glycals required special glycosyl bromides as substrates and iheir acid sensitivity excluded the classical Fischer—

Zach procedure from the useful methods. These difficulties could be overcome by using the highly reactive zinc—silver/graphite which gave 235 and 253 (Scheme 23) in 86% and 81% yields, respectively, from easily available furanosyl chlorides.¹⁸⁶ Similarly activated potassium (method I in Table 13) also gave

PGO OPG PG = alkyl

RO 0 0

R • PG, H. alkyl

""9/1-0

|R*

2 3 3 R¹ = R² = H

( 0^S

MeO OH

235 R ' = MOMOCHj. R² = H 236 R ' = MOMOCHj. R² - Bn 237 R¹ = MOMOCH2. R² MOM 238 R' = MOMOCHj, R² = S i M e P h j 239 R¹ = MEMOCH2. R² = H 240 R¹ = TMSOCHj. R² » TMS 2 4 1 R ' = TBDMSOCH2, R² = H 242 R¹ = TIPSOCHj. R² = H 2 4 3 R¹ = A, R² = H 244 R¹ = A, R² = C

Ù

= K }

₂₄₇

P

_OH

:X

248

O

J P

f ott 2 OMOM 250 R¹ = BnOCH2, R² = Bn

251 R ' " MOMOCH2. R² « MOM 252 R¹ = MOMOtCHjfe, R² = H 253 R¹ = A. R² = H 254 R¹ = A. R² = Me 255 R¹ = A. R² = Bn 256 R¹ = A. R² = MOM 257 R¹ = A, R² = SiMePh-258 R¹ - A. R² * SitBuPh-259 R¹ = A. R² = B

260 H,C

X Q ^{A V} l?0.?

Û

o 0

c

A

excellent yields of furanoid glycals protected aL will in the 3-position. and the pvranoid compound 186 (Scheme 20) was made in 88% yield from the

per-O-Table 13. P r e p a r a t i o n o f F u r a n o i d Glycals f r o m Glycosyl Chlorides (see Scheme 23)

m e t h o d g l y c a l s p r e p a r e d (%)

2 3 5 (92). 2 3 6 (86). 2 3 7 (80). 2 3 8 (90). 2 5 3 ( 9 6 ) . 2 5 5 ( 8 4 ) . 2 5 6 (77).

2 5 7 (90). 2 5 8 (86)

2 4 3 ( 5 4 ) .2 0 3 2 2 8 e n a n t i o m e c - 2 4 3 ( y i e l d u n k n o w n ) . " 2 5 3 ( 5 9 ) .^{2 0 3} 2 5 3 ( 8 2 . - 3 5 ° C ) .^{2 2}' 2 5 4 ( 6 0 ) ^

2 4 4 (69). 2 5 9 (63)

2 3 3 ( 5 1 ) .^{2 2 9} 2 3 4 ( 7 5 ) . " 2 3 5 (80).^{2 2 6}-^{2 2 9} 2 3 9 ( 6 5 - 8 2 ) . ' 2 4 0 ( 6 0 ) .^J2 4 1 (65) 2 4 2 ( 6 5 - 8 2 ) . - 2 4 7 ( 6 8 ) . ' e n a n t i o m e r - 2 4 7 ( 6 1 ) . ' 2 4 8 ( 7 6 ) . ' 2 4 9 ( 8 2 ) . « ' 2 5 0 ( y i e l d u n k n o w n ) . ' 2 5 1 ( 5 6 ) . ' 2 5 2 ( 7 0 ) . 2 5 3 ( 7 5 ) .^{2 2 C i} 2 6 0 ( 7 9 ) '

I^{2 2 5} C8K . T H F . 0 °C.

t h e n e l e c t r o p h i l e (R²X) I I X a - n a p h t h a l e n i d e . T H F .

r . t . . t h e n e l e c t r o p h i l e (R²X) I I I2 0 3 2 2 8 N a o r K . T H F . r . t . .

t h e n e l e c t r o p h i l e (R²X) I V L i . N H3( 1 ) . - 7 8 " C .

t h e n e l e c t r o p h i l e (R²X)

" B e r t r a n d . P . : G e s s o n . J . - P . : R e n o u x . B.: T r a n o y . L. Tetrahedron Lett. 1 9 9 5 . 36. 4 0 7 3 - 4 0 7 6 '' I r e l a n d . R . E.: W i l c o x . C . S : T h a i s r i v o n g s . S : V a n i e r , N . R . Can. J. Chem. 1 9 7 9 '.57. 1 7 4 3 - 1 7 4 5 ' C h e n g . J C . - Y . H a c k s e l l , L'.: D a v e s J r . G . D. J. Org. Chem.

1 9 8 5 50. 2 7 7 8 - 2 7 8 0 . « ' A b r a m s k i . W . C h m i e l e w s k i M J. Carbohydr. Cliem 1 9 9 4 13. 1 2 5 - 1 2 8 . ' D u s h i n . R . G . : D a n i s h e f s k y . S . J . J. Am. Chem Soc. 1 9 9 2 . 114. 6 5 5 - 6 5 9 . ' H a c k s e l l L \ : D a v e s j i G . D J. Org. Chem. 1 9 8 3 . 48. 2 8 7 0 - 2 8 7 6 . « I r e l a n d . R E : V e v e r t . J . - P . Can. J. Chem. 1 9 8 1 . 59. 5 7 2 - 5 8 3 . ' ' I r e l a n d . R E ; V e v e r t . J . - P . J. Org. Chem. 1 9 8 0 45. 4 2 5 9 - 4 2 6 0 ' E l - L a g d a c h . A : D i a z . Y.: C a s t i l l o n . S . Tetrahedron Lett. 1993. 34. 2 8 2 1 - 2 8 2 2 . ' O b a y a s h i . M : S c h l o s s e r . M Chem. Lett. 1 9 8 5 1 7 1 5 - 1 7 1 8

* G h o s h . A . K.. M c K e e . S . P . : T h o m p s o n . W . J . J. Org. Chem. 1 9 9 1 . 56. 6 5 0 0 - 6 5 0 3 ' P i n u n g . M . C . : L e e . Y. R . J. Am Chem. Sor.

1 9 9 5 117. 4 8 1 4 - 4 8 2 1 .

Carbanionic Reactivity of the Anomeric Center in Carbohydrates Chemical Reviews, 2001, Vol. 101, No. 1 135 S c h e m e 24

OR¹ 262 R ' = R² = Bn, R³ = SPh 263 R¹ = R² = Ac,

R³ = S 02P h OR² 264 R¹ = Bn. R² = AHyl,

R³ - S O j P h

j // ^ ) °^{n P h} j P G O OPG

PG = alkyl

Q - Q

R5 n = 0,2 6 o

R = PG, H

v ? 0 „ P h

X X V

Pfr^/q

OR¹

266 R¹ = Bz. R² = SPh 266 R' = Bn. R² = SPh 267 R' = Ac. R² = S O?P h

182 R = AC R O - L ^³ 186 R = Bn

Ph

i o

X

268 R' 269 R¹ R³ 270 R¹ 'OR² 271 R¹

= R² = Bn. R³ = SPh ?

= R² = Ac. R³ = S 02P h R²Q°

= R²= B n . R³ = S O j P h R^O

= R²= Bn, R³ = H

„ OR'

OR²

273 R' 274 R¹ 275 R¹ 276 R¹ 277 R¹ 278 R' 279 R*

280 R¹ 281 R¹

OTMS

X:

TMSO S P h

V

O . O S P h

O

= R² = Ac, R³ = SPh

= R² = Bz. R³ = SPh

= R² = Me, R³ = SPh

= R² = Bn. R³ = SPh

= H. R² = Bn. R³ = SPh - R²^ Ac. R³ = S O j P h - R² = Bz R³ = S 02P h

= R² = M « R³- S C ^ P h

= R² = Bn. R³ = S O , P h R³

4

^{. . O OR}

283 R' = R³ « H. R² « Ph 284 R' = H, R² = Ph. R³ = Ac 285 R ' - H . R²- P h , R³' = B n 286 R¹ = R² = Me. R³ = TMS

287 R¹ = H. R² = R³ = Me 288 R¹ - R² - H. R³ - Ph 289 R¹ = Bn R² = H, R³ = Ph 290 R ' = Ac. R² = H. R³ = Ph

291 R = H 292 R = Ac 293 R " Me 2 9 4 R = Bn

AcO

benzylated glucopyranosyl chloride.²²" Sodium sand in toluene at 70 °C gave 253 in l 1% yield together with 9% of 259.²⁰³ Lithium, magnesium, zinc, or zinc—copper couple were unreactive toward acetal-protected glycofuranosyl chlorides, and transnteta-lation with organolithiunts proved also

unsuccess-ful.²⁰³²²⁶ Samarium diiodide was reported to give

none of the targeted reductive elimination product 253 from the corresponding furanosyl chloride.²²⁷ However, sodium naphthalenide (method II) gave moderate yields of the expected glycals.^{203 228} which could be enhanced by lowering the reaction temper-ature.²²⁷ Sodium or potassium metals (method III) furnished the disaccharide glycals 244 and 2 5 9.^{203 228} Further trials to optimize obtention of 244 from 2,3:

5,6-di-O-isopropylidene-a-D-mannofuranosyl chloride showed lithium, sodium, and potassium in liquid ammonia-THF (10:1 mixture) at - 7 8 °C to give the target glycal and the product of reductive dehaloge-nation (1.4-anhydro-2.3:5,6-di-0-isopropylidene-D-mannitol) in 8:1. 11:1. and 15:1 ratios, respectively, in 75—80% yields. The use of sodium trimesitylbo-rane in THF at —20 °C raised the above ratio to

>50:1 (70%), and with LiDBB the yield was also increased to 94%. This reagent was applied to obtain 261 in 81% yield as an intermediate in the synthesis of subunits of monensin polyether antibiotics.²²⁷ By far the most widely applied is method IV. which gives

the desired furanoid glycals in good yields sometimes accompanied by small amounts (up to 20%) of reduc lively dehalogenated byproducts.²²⁶²²¹' The latter because of their rather unreactive nature, generalh do not disturb further transformations of the main product glycals; therefore, separation can be effected in later stages of the synthesis. No byproduct was formed with the pvranoid derivatives 245 and 246 (90% yield for both). A one-pot procedure was devel oped starting with an acetalated 1-OH-unprotected glycose (in certain cases obtained by DIBAL reduc-tion of the corresponding lactone as the first step in a continuous operation) to be converted into the chloride by hexamethylphosphorus triamide and carbon tetrachloride, which was then subjected to the reductive elimination conditions. Yields in Table 13 method IV mostly refer to such one-pot procedures.

3. From 7-Thioglycosides and Their S-Oxides

Because of the rather sensitive nature of glycosyl halides, the much more stable 1-thio-, 1-sulfinyl-. and 1-sulfonylglycosyl derivatives were also investigated as precursors of glycals. Preparation of these starting materials is as easy as that of the halides.²³⁰

Reaction of phenyl 2,3:5,6-di-0-isopropylidene-1 thio-/S-D-mannofuranoside (282a Scheme 24) with potassium graphite followed by in situ silylation gave 258 (96%) (Scheme 23), and the per-O-benzylateri 262

In document their 1-cyano and Cl (Pldal 194-200)