Development of a novel synthetic process for 2-deoxy-3,5-di-O-p-toluoyl- α-L-ribofuranosyl chloride: A versatile intermediate in the synthesis of 2′-deoxy-L-ribonucleosides

Article abstract of DOI:10.1021/op0500436

A novel synthetic route to 2-deoxy-3,5-di-O-p-toluoyl-α-L- ribofuranosyl chloride (1) from inexpensive D-xylose (3) is described. 1 is a key intermediate in the synthesis of the antiviral agent 1-(2-deoxy-β-L- ribofuranosyl)thymine (β-L-thymidine) (2) and

Full text of DOI:10.1021/op0500436

Organic Process Research & Development 2005, 9, 457−465
Development of a Novel Synthetic Process for
2-Deoxy-3,5-di-O-p-toluoyl-r-L-ribofuranosyl Chloride: A Versatile Intermediate
in the Synthesis of 2′-Deoxy-L-ribonucleosides
Narayan C. Chaudhuri,* Adel Moussa, Alistair Stewart, Jingyang Wang, and Richard Storer
Department of Process Chemistry, Idenix Pharmaceuticals, Inc., 60 Hampshire Street,
Cambridge, Massachusetts 02139, U.S.A.
Abstract:
effective processes for the synthesis of such unnatural
L-nucleosides and their key precursors is essential for
appropriate “cost of goods” considerations. 1-(2-Deoxy-â-
L-ribofuranosyl)thymine or L-dT (2) (Figure 1) is being
developed under the generic name ‘Telbivudine’ by Idenix
Pharmaceuticals, Inc. (formerly Novirio Pharmaceuticals,
Inc.) and Novartis Pharma AG to treat acute and chronic
hepatitis B viral infections.2,4 Currently telbivudine is in phase
III clinical trials. For the synthesis of this compound and
several other 2′-deoxy-L-ribonucleosides of interest, 2-deoxy-
3,5-di-O-p-toluoyl-R-L-ribofuranosyl chloride or L-chloro-
sugar 1 is a key intermediate.10,11 The majority of published
methods10-13 utilize L-arabinose as the starting material to
make this intermediate 1. In 1964 Smejkal and Sorm10
reported the conversion of L-arabinose to L-chlorosugar 1 in
11 steps via diacetyl-L-arabinal as a key intermediate. The
major drawbacks of their process are the extremely long
reaction times of some steps, chromatographic purification
of most intermediates and a very poor overall yield. In 1992
Urata et al.11 described a 9-step synthesis of 1 from
L-arabinose in overall 20% yield but their process would have
difficulties in scaling up as it required several hazardous
reagents. In 1999 Zhang et al.12 described an improvement
over the method of Urata et al.11 by replacing tri-n-butyltin
hydride with diphenylsilane as the reducing agent in a critical
deoxygenation step. More recently Jung and Xu13 prepared
the L-chlorosugar 1 in six steps from L-arabinose in 30%
overall yield. These processes also have limitations for the
industrial manufacture of the L-chlorosugar 1 due to the use
of difficult-to-handle reagents such as carbon disulfide and
ethanethiol.
A novel synthetic route to 2-deoxy-3,5-di-O-p-toluoyl-r-L-
ribofuranosyl chloride (1) from inexpensive
described. 1 is a key intermediate in the synthesis of the antiviral
agent 1-(2-deoxy-â- -ribofuranosyl)thymine (â- -thymidine) (2)
and other 2′-deoxy- -ribonucleosides. This seven-step synthesis
employs a key conversion of the to the configuration of the
sugar moiety (6 to 7 in Scheme 1) using simple reagents and
reaction conditions. The entire process involves only three
isolation steps. The key compound (1) was produced in 11%
overall yield without chromatography.
D-xylose (3) is
L
L
L
D
L
Introduction
L-Nucleosides have attracted great interest as drugs that
possess potent antiviral properties against hepatitis B virus
(HBV)1-7 and human immunodeficiency virus (HIV).1,3,6,8,9
For instance, 2′-deoxy-â-L-cytidine (L-dC), 1-(2-deoxy-â-L-
ribofuranosyl)thymine (â-L-thymidine or L-dT), 2′-deoxy-
â-L-adenosine (L-dA), 2′-fluoro-5-methyl-â-L-arabinofurano-
syluracil (L-FMAU), and 2′,3′-dideoxy-â-L-cytidine (L-ddC)
have shown potent, selective, and specific activity against
HBV replication.2,4,5-7 2′,3′-Dideoxy-â-L-5-fluorocytidine
(L-FddC) and 2′,3′-dideoxy-â-L-adenosine (L-ddA) are found
to be active against both HBV and HIV in cell cultures.6,8,9
Therefore, the development of novel, efficient, and cost-
* Corresponding author. E-mail: chaudhuri.narayan@idenix.com.
(1) Chen, H.; Pai, S. B.; Hurwitz, S. J.; Chu, C. K.; Glazkova, Y.; McClure, H.
M.; Feitelson, M.; Schinazi, R. F. Antimicrob. Agents Chemother. 2003,
47, 1922.
(2) Hernandez-Santiago, B.; Placidi, L.; Cretton-Scott, E.; Faraj, A.; Bridges,
E. G.; Bryant, M. L.; Rodriguez-Orengo. J.; Imbach, J.-L.; Gosselin, G.;
Pierra, C.; Dukhan, D.; Sommadossi, J.-P. Antimicrob. Agents Chemother.
2002, 46, 1728.
(3) Stuyver, L. J.; Lostia, S.; Adams, M.; Mathew, J. S.; Pai, B. S.; Grier, J.;
Tharnish, P. M.; Choi, Y.; Chong, Y.; Choo, H.; Chu, C. K.; Otto, M. J.;
Schinazi, R. F. Antimicrob. Agents Chemother. 2002, 46, 3854.
(4) Bryant, M. L.; Bridges, E. G.; Placidi, L.; Faraj, A.; Loi, A.-G.; Pierra, C.;
Dukhan, D.; Gosselin, G.; Imbach, J.-L.; Hernandez, B.; Juodawlkis, A.;
Tennant, B.; Korba, B.; Cote, P.; Marion, P.; Cretton-Scott, E.; Schinazi,
R. F.; Sommadossi, J.-P. Antimicrob. Agents Chemother. 2001, 45, 229.
(5) Schinazi, R. F.; Gosselin, G.; Faraj, A.; Korba, B. E.; Liotta, D. C.; Chu.
C. K.; Mathe, C.; Imbach, J.-L.; Sommadossi, J.-P. Antimicrob. Agents
Chemother. 1994, 38, 2172.
As our HBV program advanced to phase II and III clinical
trials, the need to investigate alternative synthetic processes
for L-chlorosugar 1 became increasingly important. Herein
we report a novel and efficient process for the synthesis of
2-deoxy-3,5-di-O-p-toluoyl-R-L-ribofuranosyl chloride (1)
from D-xylose (3). This process requires seven chemical
transformations. There are only three steps where the prod-
ucts of reactions are isolated and purified. In all other steps
(9) Laurent, P.; Cretton-Scott, E.; Gosselin, G.; Pierra, C.; Schinazi, R. F.;
Imbach, J.-L.; EL Kouni, M. H.; Sommadossi, J.-P. Antimicrob. Agents
Chemother. 2000, 44, 853.
(6) Lin, T.-S.; Luo, M.-Z.; Liu, M.-C.; Pai, S. B.; Dutschman, G. E.; Cheng,
Y.-C. J. Med. Chem. 1994, 37, 798.
(7) Chu, C. K.; Ma, T.; Shanmuganathan, K.; Wang, C.; Xiang, Y.; Pai, S. B.;
Yao G.-Q.; Sommadossi, J.-P.; Cheng, Y.-C. Antimicrob. Agents Chemother.
1995, 39, 979.
(10) Smejkal, J.; Sorm, F. Collect. Czech. Chem. Commun. 1964, 29, 2809.
(11) Urata, H.; Ogura, E.; Shinohara, K.; Ueda, Y.; Akagi, M. Nucleic Acids
Res. 1992, 20, 3325.
(8) Gosselin, G.; Schinazi, R. F.; Sommadossi, J.-P.; Mathe, C.; Bergogne, M.-
C.; Aubertin, A.-M.; Kirn, A.; Imbach, J.-L. Antimicrob. Agents Chemother.
1994, 38, 1292.
(12) Zhang, W.; Ramasamy, K. S.; Averett, D. R. Nucleosides Nucleotides 1999,
18, 2357.
(13) Jung, M. E.; Xu, Y. Org. Lett. 1999, 1, 1517.
10.1021/op0500436 CCC: $30.25 © 2005 American Chemical Society
Published on Web 06/10/2005
Vol. 9, No. 4, 2005 / Organic Process Research & Development
•
457

a reaction time of approximately 48-60 h was necessary
for the completion of the reaction.
To study the effect of temperature on the above bromi-
nation reaction, we carried out several experiments at 23,
40, 45, 60, 90, and 120 °C,17 employing varied equiva-
lents of HBr. A smooth conversion was observed at 45 °C
with 5 equiv of HBr (as HBA). The reaction time was
greatly reduced under these conditions (see Experimental
Section for detail). Such short reaction time is a signifi-
cant improvement of the process over the existing litera-
ture protocols16,18 that require days to prepare the di-
bromolactones. The crude product of our new process was
triturated with 20% n-heptane in isopropyl ether to get pure
2,5-dibromo-2,5-dideoxy-D-lyxono-1,4- lactone (5) as light
brown crystals. This isolation of 5 was the first of the three
isolation steps of the current process. The overall yield of 5
was approximately 40% for the two steps starting from
D-xylose (3).
Figure 1.
the products are telescoped with minimal processing. The
process avoids the use of any chromatographic purification.
Results and Discussion
C-2 Debromination, Base-Induced Rearrangement,
and Formation of 2-Deoxy-3,5-di-O-p-toluoyl-L-ribono-
1,4-lactone (8). The C-2 debromination of 5 using iodide
ion was examined in detail. This reaction was found to be
sluggish when conducted at room temperature using potas-
sium or sodium iodide and trifluoroacetic acid (TFA) in
acetone,18a-c 2-butanone, or isopropyl acetate (IPAc). When
heated at reflux, the reaction was considerably faster and
required fewer equivalents of the iodide ion to go to
completion. The combination of NaI/TFA/IPAc was found
to be convenient because the use of IPAc (bp 85-91 °C) as
the reaction solvent allowed the reaction to be completed in
2 h at 85 °C, and crude 5-bromo-2,5-dideoxy-D-threo-
pentono-1,4-lactone (6) could be obtained via an IPAc/water
split. Occasionally a trace amount of 5-iodo-2,5-dideoxy-D-
threo-pentono-1,4-lactone (6a) was detected by NMR. There
was, however, no need to separate 6a from 6 as it would
undergo the same conversion in the subsequent step. The
desired monobromolactone 6 was also accessible via the
catalytic hydrogenolysis18a of 5 in excellent yield (see Ex-
perimental Section). Considering the ease of the work-up
and the higher yield and quality of the product, the hy-
drogenolysis method should be preferred to make the
monobromolactone 6 from 5.
The monobromolactone 6 thus obtained was then treated
with 3 mol equiv of potassium hydroxide in water at ambient
temperature. After complete consumption of the starting
material, the mixture was briefly heated to 80 °C and then
cooled to ambient temperature. Thereafter, it was treated with
Amberlite IR-120 (H+) ion-exchange resin to pH 1 to afford
2-deoxy-L-ribono-1,4-lactone (7). This transformation most
likely involves the initial formation of an epoxyaldonic acid
potassium salt 10 (Figure 2) which could ring-open via an
intramolecular nucleophilic attack of the epoxide ring by the
Preparation of 2,5-Dibromo-2,5-dideoxy-D-lyxono-1,4-
lactone (5). D-Xylose (3) was smoothly oxidized to D-xylono-
1,4-lactone (4) by the action of bromine (Scheme 1). Initially
we employed Br2/BaCO3/H2O as the reagent combination
for this oxidation using a slight modification of the procedure
of Isbell.14 At the completion of the reaction, which typically
took several hours at 0 °C followed by gradual warming to
ambient temperature, the excess BaCO3 was removed by
filtration. The clear filtrate containing soluble BaBr2 was
treated with a calculated amount of diluted aqueous H2SO4.
The precipitated BaSO4 was removed by filtration, and the
filtrate was concentrated in vacuo. The residue was treated
with 30% HBr in acetic acid (HBA) in the subsequent step.
During the development of the above bromine-oxidation
step the procedure of Okabe et al.15 employing aqueous
K2CO3 in place of BaCO3 was found to be more practical.
The use of K2CO3 instead of BaCO3 had allowed us to
employ a much higher throughput for this conversion. Since
K2CO3 remained soluble in the reaction mixture, the process-
ing of the product was a lot easier. Thus, an aqueous solution
of D-xylose (3) and K2CO3 was treated with bromine at 0-5
°C. The mixture was gradually warmed to ambient temper-
ature and quenched with formic acid (see Experimental
Section for detail). D-Xylono-1,4-lactone (4) was not isolated
at this stage. Instead the mixture was concentrated in vacuo
at 45-50 °C, and the residual water from it was removed
by coevaporation with glacial acetic acid (HOAc) in vacuo
at 45-50 °C. Thus, a syrupy residue of the crude product 4
was obtained.
The crude product 4 was treated with 30% HBr in acetic
acid (HBA) to get a mixture of 2,5-dibromo-2,5-dideoxy-
D-lyxono-1,4-lactone (5) and its 3-O-acetyl derivative (not
shown in Scheme 1). Therefore, this mixture was quenched
with methanol to convert the 3-O-acetyl derivative of 5 back
to 2,5-dibromo-2,5-dideoxy-D-lyxono-1,4-lactone (5). In the
early stages of our research we allowed lactone 4 to react
with HBA (5-6.5 equivof HBr) at room temperature
according to the procedure of Lundt.16 Under these conditions
(16) For an overview of the available bromodeoxylactones by this method, see:
Lundt, I. Top. Curr. Chem. 1997, 187, 117.
(17) These reactions were conducted utilizing a Barnstead International RS 10
STEM Reaction Block.
(18) (a) Bock, K.; Lundt, I.; Pedersen, C. Carbohydr. Res. 1981, 90, 17. (b)
Bock, K.; Lundt, I.; Pedersen, C. Carbohydr. Res. 1982, 104, 79. (c) Bock,
K.; Lundt, I.; Pedersen, C. Carbohydr. Res. 1979, 68, 313. (d) Pedersen,
C.; Bock, K.; Lundt, I. Pure Appl. Chem. 1978, 50, 1385.
(14) Isbell, H. S. Methods Carbohydr. Res. 1963, 2, 13.
(15) Okabe, M.; Sun, R.-C.; Zenchoff, G. B. J. Org. Chem. 1991, 56, 4392.
458
•
Vol. 9, No. 4, 2005 / Organic Process Research & Development

Scheme 1
carboxylate ion. According to Baldwin’s rule19 the 5-exo
mode would be preferred to the 6-endo mode of opening.
Hence, this would result in an inversion of configuration at
the attacked carbon (C-4), producing the L-configured lactone
7 from its D-configured precursor 10. The lactone 7 thus
formed would be opened by aqueous alkali to the aldonic
acid salt 11, which however, would regenerate the lactone 7
stereochemical assignment of lactone 7 is unequivocal. This
assignment is further substantiated by a comparison of the
optical rotation value of the L-chlorosugar 1 synthesized from
lactone 7 (Scheme 1) with its literature value (see Experi-
mental Section). Such stereospecific D-L interconversions
of aldonolactones activated at the primary position (C-5) via
the base-induced rearrangement are well documented in the
literature,20 and they provide an easy access to the synthesis
of a variety of L-sugars from the D-sugar precursors. After
acidification the reaction mixture was filtered, and the fil-
trate was concentrated at 50 ( 5 °C to a syrup. The residual
water in the syrup was removed by coevaporation with 1,2-
dimethoxyethane (DME), which was the solvent of choice
for the subsequent reaction.
The preparation of the di-O-p-toluate 8 (Scheme 1) from
crude 7 was best accomplished by using 1.8 equiv of
p-toluoyl chloride and 5 equivof pyridine in DME. The crude
product of this reaction was recrystallized from tert-butyl
methyl ether to get pure 8 as a white crystalline material.
This step was the second isolation step of the current process.
The yield of 8 from intermediate 5 was 37-40% for the
three steps. The major by-product in this reaction was p-toluic
anhydride. Its formation was kept minimal when the above
ratio of p-toluoyl chloride was employed for the reaction.
Figure 2.
upon acidification. The stereochemical assignment of lactone
7 was based primarily on the work of Bock et al.18a where
7 was synthesized from D-lyxono-1,4-lactone via a similar
sequence of reactions. The stereochemistry at C-3 and C-4
of D-xylono-1,4-lactone (4) being respectively identical to
that at C-3 and C-4 of D-lyxono-1,4-lactone, the above
(19) Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734.
(20) Lundt, I.; Madsen, R. Top. Curr. Chem. 2001, 215, 177.
Vol. 9, No. 4, 2005 / Organic Process Research & Development
•
459

Scheme 2
Among other bases, we examined triethylamine instead of
pyridine but noticed significant loss of product possibly via
elimination of triethylammonium p-toluate involving the C-2
hydrogen atom of the protected lactone 8.
also led to a number of undesired by-products, thus com-
plicating the reduction process. We needed to develop a
selective and robust protocol to minimize the by-product
formation so that the process could be scaled up easily. A
variety of reducing agents,22 solvents, and conditions such
as Red-Al:2-propanol/toluene, Red-Al:tert-butyl alcohol/
toluene, lithium tri-tert-butoxyaluminohydride/toluene, lithium
tri-tert-butoxyaluminohydride/THF, diisobutylaluminum hy-
dride (DIBALH)/toluene, DIBALH/THF, DIBALH/dichlo-
romethane, DIBALH/1,2-dichloroethane, DIBALH/isopro-
pyl ether, DIBALH/DME, DIBALH/1,2-diethoxyethane,
DIBALH/diglyme, lithium borohydride/THF, sodium boro-
hydride/THF, disiamylborane/THF, disiamylborane/DME,
L-selectride/DME and borane-methyl sulfide complex/THF
were screened for this conversion. These experiments were
carried out at temperatures ranging from -78 °C up to +25
°C, employing varied equivalents of the reducing agents. Of
these, the systems using DIBALH in either DME, THF, or
diglyme appeared promising. When optimized further, it was
found that the reduction of lactone 8 with 1.32 mol equiv of
DIBALH in DME at -60 ( 5 °C provided the highest yield
(>85% yield by HPLC, AUC @ 254 nm) of the lactol 9
Selective Reduction of Lactone 8 to Lactol 9 and
Formation of L-Chlorosugar 1. Recently, we developed an
efficient method21 for the selective reduction of 2,3,5-tri-O-
benzoyl-2-C-methyl-D-ribonic-γ-lactone to 2,3,5-tri-O-ben-
zoyl-2-C-methyl-D-ribofuranose. The reducing agent for this
reduction was generated from the reaction of sodium bis(2-
methoxyethoxy)aluminum hydride (Red-Al) with an equal
molar amount of C2H5OH in toluene. When applied to the
lactone 8 in the present study, this reducing system gave a
poor yield of the desired lactol, 2-deoxy-3,5-di-O-p-toluoyl-
L-ribofuranose (9). The major product of this reaction was
found to be D-threo-1,2,3,5-pentanetetrol 1,3-di-p-toluate (12)
(Scheme 2) due to further reduction of the intermediate lactol
9. The identity of this compound 12 was confirmed by
treating the mixture with p-toluoyl chloride in the presence
of triethylamine and 4-(dimethylamino)pyridine in anhydrous
DME and isolating D-threo-1,2,3,5-pentanetetrol tetra-p-
toluate (15) (Scheme 2). The selective reduction of the
protected lactone 8 to the lactol 9 therefore required a
thorough investigation to minimize the formation of the by-
product 12. In addition to the observed over-reduction, the
cleavage of the labile p-toluoyl ester groups in the lactone 8
(22) For reduction of lactones to lactols, see: (a) Kanazawa, R.; Tokoroyama,
T. Synthesis 1976, 526. (b) Kohn, P.; Samaritano, R. H.; Lerner, L. M. J.
Am. Chem. Soc. 1965, 87, 5475. (c) Okabe, M.; Sun, R.-C.; Tam, S. Y.-K.;
Todaro, L. J.; Coffen, D. L. J. Org. Chem. 1988, 53, 4780. (d) RajanBabu,
T. V.; Nugent, W. A.; Taber, D. F.; Fagan, P. J. J. Am. Chem. Soc. 1988,
110, 7128. (e) Wolf, J.; Jarrige, J.-M.; Florent, J.-C.; Grierson, D. S.;
Monneret, C. Synthesis 1992, 773. (f) Zhang, Q.; Liu, H.-W. J. Am. Chem.
Soc. 2000, 122, 9065.
(21) Storer, R.; Moussa, A.; Chaudhuri, N.; Waligora, F. PCT/U.S. Patent 2003/
039643, WO 2004/052899.
460
•
Vol. 9, No. 4, 2005 / Organic Process Research & Development

with minimal impurity formation. The crude lactol 9 was
processed via an extractive work-up and utilized without
further purification.
Idenix Method 004. The data were reported in area % and
were not adjusted to weight %. Thin layer chromatography
(TLC) was performed on Baker Si-250-F precoated silica
gel plates, and the eluted plates were viewed under UV at
254 nm or stained in appropriate TLC dips as the case might
be.
In the final step the crude lactol 9 was treated with
anhydrous HCl gas in glacial acetic acid/tert-butyl methyl
ether at 0 °C to produce the target molecule 2-deoxy-3,5-
di-O-p-toluoyl-R-L-ribofuranosyl chloride (1) (Scheme 1).
The product crystallized out of the mixture and was collected
easily by vacuum filtration. This constituted the third and
the final step of isolation of goods in the current process.
The yields of 1 were typically over 65% for the two steps
from the protected lactone 8 and were reproducible with
minimal batch-to-batch variations. To our knowledge this is
the first example of the direct conversion of 2-deoxy-3,5-
di-O-p-toluoyl-L-ribofuranose (9) to 2-deoxy-3,5-di-O-p-
toluoyl-R-L-ribofuranosyl chloride (1). Other methods involve
the intermediate formation of the methyl glycoside 13,11,12
or the tri-O-p-toluate 14.13 For the comparison of yields, we
prepared the methyl glycoside 13 and the tri-O-p-toluate 14
from the protected ribose 9 and converted them to the
L-chlorosugar 1 as shown in Scheme 2. Yields via these two
sequences were lower than that of the direct conversion as
developed in the present study. The direct conversion route
thus provides a novel process for the L-chlorosugar 1 from
D-xylose (3) in approximately 11% overall yield.
HPLC conditions: column: Waters Nova-Pak C18 RP,
4 µm, 3.9 mm × 150 mm; flow rate: 1.0 mL/min; mobile
phase for Idenix Method 004: gradient 40-95% acetonitrile,
60-5% water over 13 min followed by isocratic 95%
acetonitrile, 5% water over 7 min and gradient 95-40%
acetonitrile, 5-60% water over the next 10 min; column
temperature: ambient; detection: UV 254 and 272 nm.
D-Xylono-1,4-lactone (4). A 1-L three-necked round-
bottomed flask was equipped with an overhead stirrer
attached to a glass stirring shaft and a Teflon blade, a digital
thermometer connected to a temperature probe, a graduated
addition funnel and an argon line. D-Xylose (3) (100 g, 0.67
mol) and deionized water (270 mL) were charged into the
reaction flask, and the contents were stirred to obtain a clear
solution. It was cooled to about 5 °C using an ice-bath.
Potassium carbonate (113.2 g, 0.82 mol) was added to the
reaction flask in portions, maintaining the temperature below
20 °C, and it was stirred for additional 30 min. When the
internal temperature dropped below 5 °C, bromine (123.0
g, 0.77 mol) was added dropwise at 0-5 °C over a period
of 2 h. The reaction mixture was maintained at 5-10 °C for
a further 30 min and then stirred at room temperature
overnight. It was quenched by dropwise addition of formic
acid (8.0 g, 0.174 mol) over a period of 10 min. The mixture
was stirred at room temperature for a further 40 min. In this
process the bromine color was discharged, and the mixture
reached a pH of approximately 3. The contents were
transferred to a 1-L single-necked round-bottomed flask and
concentrated under reduced pressure at 45-50 °C to a
volume of approximately 50 mL. For NMR analysis an
aliquot was taken to dryness at 45-50 °C under reduced
pressure, and the residue was extracted with hot acetone.
The acetone extract was evaporated to dryness and further
dried under high vacuum to afford a sample of D-xylono-
Conclusions
In summary, we have developed the chemistry of a new
process for the preparation of 2-deoxy-3,5-di-O-p-toluoyl-
R-L-ribofuranosyl chloride (1) which is a versatile intermedi-
ate in the synthesis of many unnatural 2′-deoxy-L-ribonu-
cleosides. With an overall yield of 11%, there are only three
isolation steps in the entire process. In a very critical step,
the process involves an efficient and stereospecific conver-
sion of 5-bromo-2,5-dideoxy-D-threo-pentono-1,4-lactone (6)
to 2-deoxy-L-ribono-1,4-lactone (7). We have also developed
the highly selective DIBALH-reduction of the protected
lactone 8 to lactol 9 in good yield. This DIBALH-reduction
of the current process may be conveniently carried out in
cryogenic reactors for production batches. The process also
involves the novel conversion of the lactol derivative 9
directly to the chlorosugar 1, thus eliminating the need for
going through the glycosidic intermediates 13 or 14. Chro-
matographic purifications are completely avoided in any
stages of the process. The present route thus offers a
promising and potentially scalable process for the large-scale
synthesis of L-chlorosugar 1.
1
1,4-lactone (4); H NMR (DMSO-d6, 400 MHz) δ 6.16-
5.9 (1H, br s, exchangeable with D2O, OH), 5.84-5.6 (1H,
br s, exchangeable with D2O, OH), 5.1-4.66 (1H, br s,
exchangeable with D2O, OH), 4.44-4.36 (1H, m, H-2),
4.28-4.14 (2H, m, H-3 and H-4), 3.72-3.56 (2H, d of an
AB q, ∆υ ) 21.5 Hz, J 12.2, 4.0, and 3.0 Hz, 2 × H-5; 13
C
NMR (CDCl3, 100 MHz) δ 175.7, 80.0, 73.2, 72.3, 58.5.
Glacial acetic acid (200 mL) was added to the main portion
of the crude product and the mixture was concentrated again
at 45-50 °C under reduced pressure to a volume of
approximately 60 mL. This solution of the crude D-xylono-
1,4-lactone (4) in acetic acid was used in the subsequent step
without further purification.
Experimental Section
General. All raw materials including solvents for HPLC
were obtained from commercial sources. Moisture-sensitive
reactions were carried out in an argon atmosphere. Melting
points were measured on a Barnstead/Thermolyne Electro-
thermal Mel-Temp apparatus. NMR spectra were recorded
on a Varian 400 MHz spectrometer in appropriate deuterio-
solvents. Waters 2695 HPLC and 996 PDA detector were
used to analyze all in-process samples and finished products
that were UV-active. The HPLC data were acquired using
2,5-Dibromo-2,5-dideoxy-D-lyxono-1,4-lactone (5). The
solution of the crude D-xylono-1,4-lactone (4) (0.67 mol for
100% conversion) in acetic acid was transferred into a 3-L
three-necked round-bottomed flask whilst warm using ad-
ditional acetic acid (200 mL). To this was added 30% HBr/
Vol. 9, No. 4, 2005 / Organic Process Research & Development
•
461

HOAc (896.3 g, 3.32 mol) at room temperature. The reaction
mixture was stirred at 45 °C for 1 h and then at room
temperature for 2 h. After this time TLC analysis (1:1, ethyl
acetate:hexanes; visualized using KMnO4) indicated two
major products: 2,5-dibromo-2,5-dideoxy-D-lyxono-1,4-lac-
tone (5), Rf 0.5; and 3-O-acetyl derivative of 5, Rf 0.63. The
reaction mixture was cooled to 0 °C using an ice bath, and
methanol (850 mL) was added slowly over a period of 1 h,
whilst maintaining the temperature below 10 °C. It was then
stirred at room temperature overnight. After this time TLC
analysis (1:1 ethyl acetate/hexanes) indicated conversion of
one product (Rf 0.63) to the other product (Rf 0.5). The
mixture was filtered through a Bu¨chner funnel to remove
the KBr salt. It was then concentrated in vacuo and
coevaporated with water (2 × 250 mL). Ethyl acetate (800
mL) and water (250 mL) were added, and the layers were
separated. The aqueous layer was further extracted with ethyl
acetate (2 × 300 mL), and the combined organic extracts
were washed with aqueous saturated sodium hydrogen
carbonate solution (400 mL) followed by water (100 mL).
The EtOAc layer was dried with anhydrous sodium sulfate
(125 g), filtered, and concentrated in vacuo at 50 °C. Before
taking to complete dryness, n-heptane (200 mL) was added
to the thick oil and it was evaporated in vacuo at 50 °C to
give a brown semisolid. A solution of 100 mL of n-heptane
in 400 mL of isopropyl ether was added to it, and the mixture
was stirred at room temperature for 4 h. Filtration of the
slurry and vacuum-drying of the cake at 30-35 °C overnight
yielded pure 2,5-dibromo-2,5-dideoxy-D-lyxono-1,4-lactone
(5) as a light brown solid (71.9 g, yield 40% over 2 steps);
mp 90-93°C; lit.18b mp 92-93°C; 1H NMR (DMSO-d6, 400
MHz) δ 6.37 (1H, d, J 6.1 Hz, exchangeable with D2O,
3-OH), 5.32 (1H, d, J 4.6 Hz, H-2), 4.77-4.7 (1H, m, H-4),
4.43-4.37 (1H, m, H-3), 3.76-3.6 (2H, d of an AB q, ∆υ
extracts were treated with aqueous sodium thiosulfate solu-
tion (48 g of sodium thiosulfate in 160 mL of water). The
aqueous layer was back-extracted with isopropyl acetate (2
× 200 mL), and the combined organic extracts were dried
with sodium sulfate (20 g). It was filtered, and the filtrate
was concentrated in vacuo to yield 5-bromo-2,5-dideoxy-D-
threo-pentono-1,4-lactone (6) (16.38 g, crude) as a brown
thick oil. 6 was found by NMR to contain a trace amount of
the corresponding 5-iodo compound 6a; 1H NMR (D2O, 400
MHz) δ 4.91-4.84 (1H, m, H-4), 4.76-4.71 (1H, m, H-3),
3.7 (2H, d, J 6.9 Hz, 2 × H-5), 3.54-3.4 (not integrated, m,
H-5 for the iodide 6a), 3.1 (1H, dd, J 18.1 and 5.6 Hz, H-2),
2.63 (1H, d, J 18.3 Hz, H-2); 1H NMR (DMSO-d6, 400 MHz)
δ 5.67 (1H, d, J 4.8 Hz, exchangeable with D2O, 3-OH),
4.7-4.62 (1H, m, H-4), 4.46-4.0 (1H, m, H-3), 3.78-3.6
(1H, d of an AB q, ∆υ ) 43.4 Hz, J 10.7, 10.2, 7.8, and 5.3
Hz, 2 × H-5), 3.4-3.28 (0.1H, m, clearly seen in the D2O-
exchanged spectrum, H-5 for the iodide 6a), 2.97 (1H, dd,
J 17.1 Hz, J 6.3 Hz, H-2), 2.36 (1H, d, J 17.5 Hz, H-2); 13
C
NMR (D2O, 100 MHz) δ 179.4, 85.1, 68.4, 39.5, 27.7;
TOFMS-ES- m/z 253 (M + AcOH); elemental analysis of
a purified sample: C5H7BrO3 requires C 30.80, H 3.62, Br
40.97; found C 30.69, H 3.55, Br 41.22.
5-Bromo-2,5-dideoxy-D-threo-pentono-1,4-lactone
(6): Hydrogenation method.18a 2,5-Dibromo-2,5-dideoxy-
D-lyxono-1,4-lactone (5) (7.5 g, 27.4 mmol) was dissolved
in ethyl acetate (120 mL). Triethylamine (2.9 g, 28.7 mmol)
and 5% dry palladium on carbon (1 g) were added with
stirring under a blanket of argon. The argon was replaced
with hydrogen via a vacuum-fill technique using a hydrogen
balloon and a three-way vacuum adapter. The mixture was
then stirred at room temperature under the atmosphere of
hydrogen for 1.5 h. After this time TLC analysis (1:1 ethyl
acetate/hexanes) indicated a new product (Rf 0.2) and residual
starting material (Rf 0.5). Therefore, the reaction mixture was
purged with additional hydrogen (three more times using a
freshly filled balloon) and then stirred under hydrogen for a
further 2 h. After this time TLC analysis indicated only a
trace amount of residual starting material. Therefore, the
reaction mixture was filtered through a Celite pad using ethyl
acetate as the eluent (30 mL). The filtrate was washed with
4 M HCl (30 mL), and the aqueous layer was further
extracted with ethyl acetate (2 × 20 mL). The combined
organic extracts were dried with sodium sulfate (10 g),
filtered, and concentrated in vacuo to a pale-yellow oil which
was dried further in vacuo at 25 °C overnight to yield
5-bromo-2,5-dideoxy-D-threo-pentono-1,4-lactone (6): 5.15
1
) 36.6 Hz, J 10.5, 8.1, and 5.6 Hz, 2 × H-5); H NMR
(CDCl3, 400 MHz) δ 4.8 (1H, d, J 4.4 Hz, H-2), 4.73-4.66
(1H, m, H-4), 4.66-4.61 (1H, m, H-3), 3.77-3.63 (2H,
m, 2 × H-5), 2.66-2.61 (1H, br s, 3-OH); 13C NMR
(DMSO-d6, 100 MHz) δ 171.3, 81.6, 69.2, 48.9, 29.2; 13C
NMR (CDCl3, 100 MHz) δ 169.7, 80.5, 69.0, 48.4, 26.1.
5-Bromo-2,5-dideoxy-D-threo-pentono-1,4-lactone (6):
Sodium iodide method. 2,5-Dibromo-2,5-dideoxy-D-lyxono-
1,4-lactone (5) (35 g, 127.8 mmol) was dissolved in isopropyl
acetate (300 mL). Sodium iodide (76.6 g, 511.0 mmol) and
trifluoroacetic acid (22.0 g, 192.9 mmol) were added to it at
room temperature. With stirring the reaction mixture was
heated at 85 °C (internal temperature) for 2 h. After this
time TLC analysis (1:1, ethyl acetate:hexanes) indicated very
little remaining starting material (Rf 0.5) and a new product
(Rf 0.2) being formed. The reaction mixture was cooled to
room temperature and stirred for 4 h. TLC analysis indicated
no starting material therefore, the reaction mixture was
concentrated in Vacuo to approximately 20 mL to remove
the trifluoroacetic acid. The residue was diluted with fresh
isopropyl acetate (200 mL), washed with saturated aqueous
sodium hydrogen carbonate solution (200 mL) and the layers
were separated. The aqueous layer was further extracted with
isopropyl acetate (3 × 200 mL). The combined organic
1
g; crude yield 96%; H NMR (D2O, 400 MHz) δ 4.91-
4.85 (1H, m, H-4), 4.77-4.72 (1H, m, H-3), 3.72 (2H, d, J
6.7 Hz, 2 × H-5), 3.12 (1H, dd, J 18.3 and 5.8 Hz, H-2),
2.65 (1H, d, J 18.3 Hz, H-2); 1H NMR (DMSO-d6, 400 MHz)
δ 5.68-5.54 (1H, br, 3-OH), 4.68-4.6 (1H, m, H-4), 4.43-
4.36 (1H, m, H-3), 3.76-3.56 (1H, d of an AB q, ∆υ )
46.4 Hz, J 10.7, 10.2, 8.3, 7.8, and 5.3 Hz, 2 × H-5), 2.94
(1H, dd, J 17.1 and 5.4 Hz, H-2), 2.33 (1H, d, J 17.1 Hz,
H-2); 13C NMR (D2O, 100 MHz) δ 179.4, 85.1, 68.4, 39.5,
27.7; 13C NMR (DMSO-d6, 100 MHz) δ 175.3, 83.2, 67.1,
39.5, 29.6.
462
•
Vol. 9, No. 4, 2005 / Organic Process Research & Development

2-Deoxy-L-ribono-1,4-lactone (7). Potassium hydroxide
(85%, 14.9 g, 225.7 mmol) was dissolved in water (125 mL)
at 15 °C. This was added to a stirred solution of the crude
5-bromo-2,5-dideoxy-D-threo-pentono-1,4-lactone (6) (15 g,
76.9 mmol) in water (60 mL) at room temperature. After 3
h TLC analysis (2% methanol in ethyl acetate) indicated
complete consumption of the starting material (Rf 0.65). The
reaction mixture was then heated at 80 °C (internal temper-
ature) for 30 min and cooled to room temperature. The pH
of the reaction mixture was found to be approximately 14.
Amberlite IR-120 (plus) ion-exchange acidic resin (50 g)
was added and the mixture stirred at room temperature for
30 min. At this point the pH was measured to be 3.
Additional resin (40 g) was charged, and the mixture was
stirred at room temperature for a further 30 min. At this point
the mixture registered a pH of 1. It was stirred at room
temperature overnight and checked by TLC that indicated
formation of a new product (Rf 0.25). The resin was then
removed by filtration through a sintered glass funnel and
washed with water (200 mL); the filtrate was concentrated
in vacuo at 50 ( 5 °C. Before taking to complete dryness,
coevaporation with 1,2-dimethoxyethane (2 × 100 mL) was
performed. The residue was combined with fresh 1,2-
dimethoxyethane (200 mL) and anhydrous MgSO4 (10 g).
The mixture was stirred at room temperature for 40 min and
filtered, and the cake was washed with additional 1,2-
dimethoxyethane (75 mL). A small aliquot of the filtrate was
evaporated at 45 ( 5 °C under reduced pressure and dried
further in the high vacuum. Thus, a sample of crude 2-deoxy-
L-ribono-1,4-lactone (7) was obtained for NMR analysis; 1H
NMR (DMSO-d6, 400 MHz) δ 5.0-4.7 (2H, br, 2 × OH),
4.26 (2H, m, H-3 and H-4), 3.54 (1H, dd, J 12.2 and 4.2
Hz, H-5), 3.50 (1H, dd, J 12.2 and 3.9 Hz, H-5), 2.80 (1H,
dd, J 18.1 and 6.3 Hz, H-2), 2.22 (1H, dd, J 17.6 and 2.0
Hz, H-2). The remainder of the filtrate was concentrated by
rotary evaporation at 50 ( 5 °C down to approximately 100
mL and was used in the subsequent reaction without further
purification.
2-Deoxy-3,5-di-O-p-toluoyl-L-ribono-1,4-lactone (8). Py-
ridine (30.4 g, 384.5 mmol) was added to the above solution
of 2-deoxy-L-ribono-1,4-lactone (7) (estimated 76.9 mmol)
in 1,2-DME, and the mixture was cooled between 0 and -5
°C under argon. p-Toluoyl chloride (21.4 g, 138.4 mmol)
was added from an addition funnel over 20 min, maintaining
the internal temperature between 0 and -5 °C. After stirring
at 0 to -5 °C for 3.5 h, TLC analysis (2% methanol in
EtOAc) indicated the formation of two new products (Rf 0.76
and 0.65). No remaining starting material (Rf 0.3) was
detected. HPLC analysis also indicated that the reaction had
reached completion. The reaction mixture, whilst cold
(approximately 5 °C), was quenched with a solution of
sodium hydrogen carbonate (25 g of NaHCO3 dissolved in
300 mL of water). A brown oil separated from the mixture,
and it gradually solidified upon stirring. After 2 h of stirring
at room temperature the solid was collected by vacuum
filtration, washed with water (150 mL), and dried in air
overnight. This crude lactone (25.84 g) was dissolved in
dichloromethane (150 mL), dried with MgSO4 (10 g) for 1
h, and filtered. The filter cake was washed with fresh
dichloromethane (50 mL), and the filtrate was concentrated
at 40 °C to a volume of approximately 50 mL. tert-Butyl
methyl ether (TBME) (100 mL) was added, and the mixture
was concentrated again at 40 °C to approximately 50 mL.
This concentrated solution was then stirred at room temper-
ature to form a slurry. More TBME (50 mL) was added,
and stirring was continued at room temperature for an
additional 2 h. The solid was collected by vacuum filtration,
washed with TBME (50 mL), and dried under vacuum at
30-35 °C overnight to yield 2-deoxy-3,5-di-O-p-toluoyl-L-
ribono-1,4-lactone (8) (10.46 g) as a pale-brown solid; molar
1
yield 37% over three steps from 5; mp 115-118 °C; H
NMR (CDCl3, 400 MHz) δ 7.94-7.85 (4H, 2 × d, J 8.4
Hz, 2 × ArH2, ortho to -CO-); 7.3-7.23 (4H, m, 2 × ArH2,
meta to -CO-), 5.6 (1H, d, J 7.68 Hz, H-3), 4.99-4.93 (1H,
m, H-4), 4.75-4.58 (2H, d of an AB q, ∆υ ) 39.7 Hz, J
12.2, 3.7, 12.5 and 3.3 Hz, 2 × H-5), 3.17 (1H, dd, J 18.7
and 7.3 Hz, H-2), 2.83 (1H, dd, J 18.7 Hz and 1.8 Hz, H-2),
2.43 and 2.42 (6H, 2 × s, 2 × Ar-CH3); 13C NMR (CDCl3,
100 MHz) δ 174.2, 166.1, 166.0, 145.0, 144.6, 130.0, 129.8,
129.5, 129.5, 126.4, 126.0, 82.8, 71.8, 63.9, 35.2, 21.9.
2-Deoxy-3,5-di-O-p-toluoyl-L-ribofuranose (9). A solu-
tion of 2-deoxy-3,5-di-O-p-toluoyl-L-ribono-1,4-lactone (8)
(9.0 g, 24.42 mmol) in 1,2-dimethoxyethane (90 mL) was
cooled to approximately -60 °C under argon with overhead
stirring. A 1 M solution of diisobutylaluminum hydride in
hexanes (32.4 mL, 32.4 mmol) was added dropwise via an
addition funnel over a period of 15 min. The internal
temperature was maintained at -60 ( 5 °C for 1 h, and
HPLC analysis after this period indicated completion of the
reaction. The cold reaction mixture was quenched by addition
of acetone (10 mL) over 2 min followed by 5 N HCl (30
mL) over 5 min. The mixture was stirred at room temperature
for an additional 30 min and concentrated in vacuo at 35
°C. The residual oil was combined with a solution of NaCl
(24 g) in water (60 mL) and extracted with ethyl acetate (3
× 100 mL). The combined organic extracts were dried with
anhydrous sodium sulfate (10 g, 30 min), concentrated to a
volume of 40 mL, and coevaporated with TBME (2 × 125
mL) to yield 2-deoxy-3,5-di-O-p-toluoyl-L-ribofuranose (9)
as a solution in TBME. The second coevaporation above
was stopped when the volume in the distilling flask was
approximately 40 mL. This product was used crude in the
next step. An aliquot of this solution was concentrated in
1
vacuo to dryness for NMR analysis; H NMR (CDCl3, 400
MHz, mixture of R- and â-anomers) δ 7.98-7.88 (4H, m, 2
× ArH2, ortho to -CO- for R- and â-anomers), 7.28-7.18
(4H, m, 2 × ArH2, meta to -CO- for R- and â-anomers),
5.78-5.44 (2H, 3 × m, 1 × d, H-1 and H-3 for R-
and â-anomers), 4.74-4.44 (3H, m, H-4 and 2 × H-5 for
R- and â-anomers), 2.58-2.23 (8H, m, 2 × H-2 and 2 ×
Ar-CH3 for R- and â-anomers).
2-Deoxy-3,5-di-O-p-toluoyl-r-L-ribofuranosyl Chloride
(1): Directly from Lactol 9. A solution of 2-deoxy-3,5-di-
O-p-toluoyl-L-ribofuranose (9) (24.4 mmol approximately)
in TBME (40 mL approximately) as obtained in the previous
step was further diluted with additional TBME (15 mL) and
Vol. 9, No. 4, 2005 / Organic Process Research & Development
•
463

stirred for 20 min at room temperature. Some white
precipitates appeared. Glacial acetic acid (5 mL) was added
with stirring until a clear solution was obtained. This solution
was cooled to 0 °C under argon, and anhydrous HCl gas
(total 20.4 g, 559.5 mmol, measured by difference of weights
of the HCl tank) was passed into it in a steady stream for 25
min. A white solid crystallized out of the solution. After an
additional 10 min of stirring at 0-5 °C an aliquot (0.2 mL)
was quenched with anhydrous ethanol (1.2 mL), and it was
allowed to sit at room temperature with occasional shaking
for 10 min until a clear solution was obtained. HPLC analysis
of this sample detected <1% of unreacted lactol 9. The
reaction mixture was maintained at 0-5 °C for an additional
30 min while being stirred. It was then filtered, and the solid
cake was washed with TBME (50 mL). The product was
dried in a vacuum over KOH for 5 h at room temperature to
yield 2-deoxy-3,5-di-O-p-toluoyl-R-L-ribofuranosyl chloride
(1) as an off-white crystalline solid; yield 6.24 g, 65% over
two steps from the lactone 8; mp 114-116 °C (lit.12 mp
â-anomers by HPLC; yield 480 mg, 98% based on the crude.
An analytical sample of 13 was obtained by flash chroma-
tography on silica gel using 15% EtOAc/n-heptane (v/v) as
1
the eluent; H NMR (CDCl3, 400 MHz, approximately 1:1
mixture of R- and â-anomers) δ 8.0-7.88 (4H, m, 2 × ArH2,
ortho to -CO-), 7.28-7.18 (4H, m, 2 × ArH2, meta to
-CO-), 5.64-5.56 (0.5H, m, H-3 for one anomer), 5.44-
5.38 (0.5H, m, H-3 for the other anomer), 5.23 (0.5H, dd, J
5.4 and 2.0 Hz, H-1 for one anomer), 5.19 (0.5H, d, J 5.4
Hz, H-1 for the other anomer), 4.68-4.36 (3H, m, H-4 and
2 × H-5 for R- and â-anomers), 3.42 (1.5H, s, OCH3 for
one anomer), 3.36 (1.5H, s, OCH3 for the other anomer),
2.6-2.16 (8H, m containing CH3 singlets, 2 × H-2 and 2 ×
Ar-CH3 for R- and â-anomers); 13C NMR (CDCl3, 100 MHz,
mixture of R- and â-anomers) δ 166.7, 166.6, 166.5, 166.3,
144.2, 144.1, 144.0, 143.8, 130.0, 130.0, 129.9, 129.3, 129.3,
127.4, 127.3, 127.2, 127.1, 105.8, 105.2, 82.1, 81.1, 75.6,
74.8, 65.3, 64.5, 55.4, 55.2, 39.5, 21.8.
The above-obtained methyl 2-deoxy-3,5-di-O-p-toluoyl-
L-ribofuranoside (13) (1:1 mixture of R- and â-anomers,
480 mg, 1.25 mmol) was dissolved in TBME (3 mL) and
glacial acetic acid (1 mL). This solution was cooled to 0 °C
under argon. Anhydrous HCl gas was bubbled into this
solution for 15 min, and the reaction mixture was stirred at
0-5 °C for an additional 10 min. HPLC detected about 6%
remaining starting material. This number decreased margin-
ally after the reaction had continued for an additional 45
min. The white solid that crystallized out of the reaction
mixture was collected by vacuum filtration, washed with
TBME (5 mL) and dried under high vacuum at room
temperature for 4 h to get 2-deoxy-3,5-di-O-p-toluoyl-R-L-
ribofuranosyl chloride (1): mp 113-115 °C; yield 228 mg,
25
118-121 °C); [R]D25 ) -117° (c ) 1.0, CHCl3) [lit.23 [R]D
) -125.7° (c ) 0.811, DMF)]; 1H NMR (CDCl3, 400 MHz)
δ 7.99 (2H, d, J 8.4 Hz, ArH2, ortho to -CO-), 7.9 (2H, d, J
8.0 Hz, ArH2, ortho to -CO-), 7.3-7.2 (4H, m, 2 × ArH2,
meta to -CO-), 6.48 (1H, d, J 5.1 Hz, H-1), 5.56 (1H, dd, J
7.3 and 2.6 Hz, H-3), 4.85-4.83 (1H, m, H-4), 4.72-4.56
(2H, d of an AB q, ∆υ ) 36.1 Hz, J 12.1, 4.4 and 3.3 Hz,
2 × H-5), 2.92-2.82 (1H, ddd, J 15.2, 7.5, and 5.3 Hz, H-2),
2.75 (1H, pseudo d, J 15.0 Hz, H-2), 2.42 (3H, s, Ar-CH3),
2.41 (3H, s, Ar-CH3).
2-Deoxy-3,5-di-O-p-toluoyl-R-L-ribofuranosyl Chloride
(1): via the Methyl Glycoside 13. A solution of 1% HCl in
methanol was prepared by the reaction of acetyl chloride
(215 mg, 2.738 mmol) with a measured excess of anhydrous
methanol (10 mL) at 5 °C and stirring the mixture at 5 °C
to room temperature for 40 min. In a separate round-
bottomed flask, 2-deoxy-3,5-di-O-p-toluoyl-L-ribofuranose
(9) (470 mg, 1.27 mmol) was dissolved in anhydrous
methanol (9 mL) and cooled to 10 °C. To this, was added
1% HCl in methanol (1 mL, 0.274 mmol)) as prepared above,
and the reaction mixture was stirred under argon at 10-15
°C for 1.5 h. After this time HPLC analysis detected 18.4%
of unreacted ribofuranose 9. Therefore, another portion of
1% HCl in methanol (1 mL, 0.274 mmol) was added, and
stirring continued at room temperature for a further 1.5 h.
HPLC analysis after this time indicated that the reaction was
close to completion. The reaction mixture was then concen-
trated in vacuo at 30 °C, and any methanol left in it was
coevaporated with TBME (10 mL) at 30 °C. The residue
was taken up in fresh TBME (10 mL), and since some
suspended white solids were observed, ethyl acetate (15 mL)
was added to dissolve the suspension. This solution was dried
with anhydrous sodium sulfate (2 g) and filtered. The filtrate
was concentrated in vacuo at 30 °C and further dried in high
vacuum at room temperature for 2 h to yield a thick oil of
methyl 2-deoxy-3,5-di-O-p-toluoyl-L-ribofuranoside (13).
The product 13 was found to be a 1:1 mixture of R- and
1
47% over three steps from the lactone 8; its H NMR in
CDCl3 was identical to that of the chlorosugar 1 obtained
directly from the lactol 9 above.
2-Deoxy-3,5-di-O-p-toluoyl-r-L-ribofuranosyl Chloride
(1): via the Tri-O-p-toluate 14. For this reaction a batch
of crude 2-deoxy-3,5-di-O-p-toluoyl-L-ribofuranose (9) (550
mg, 1.484 mmol) was prepared via the reduction of the
protected lactone 8 with 2.0 mol equiv of Red-Al:C2H5OH
reagent. To a solution of this crude 2-deoxy-3,5-di-O-p-
toluoyl-L-ribofuranose (9) (550 mg, 1.484 mmol) in anhy-
drous DME (15 mL) were added triethylamine (728 mg,
7.194 mmol) and 4-(dimethylamino)pyridine (18 mg, 0.147
mmol). It was cooled to 0 °C under argon, and p-toluoyl
chloride (445 mg, 2.878 mmol) dissolved in DME (2 mL)
was added over a period of 10 min. The mixture was stirred
at 0-20 °C for 2 h and heated at 45 °C for 4 h. When the
reaction was complete, the mixture was cooled to 5 °C and
quenched with aqueous NaHCO3 (1.25 g dissolved in 20 mL
of water). The crude product was isolated via extractive
work-up using ethyl acetate. This product was found to
contain three main compounds in the ratio of 5:2:15 as
determined by HPLC area %; one of these compounds
originated from an over-reduced by-product 12 generated in
the previous step. Flash chromatography of this mixture on
silica gel using 15% ethyl acetate/hexanes as the eluent
afforded a thick oil (480 mg), which was found to be still a
(23) Fujimori, S.; Iwanami, N.; Hashimoto, Y.; Shudo, K. Nucleosides Nucle-
otides 1992, 11, 341.
464
•
Vol. 9, No. 4, 2005 / Organic Process Research & Development

mixture of the same three compounds in the same ratio. Mass
spectral data of this mixture confirmed these compounds as
2-deoxy-L-ribofuranose 1,3,5-tri-p-toluate (14, R- and â-ano-
mers) and D-threo-1,2,3,5-pentanetetrol tetra-p-toluate (15);
at 30 °C by rotary evaporation under reduced pressure, and
the residue was treated with a solution of NaHCO3 (2.0 g)
in water (20 mL) until effervescence of CO2 ceased. The
mixture was then extracted with EtOAc (2 × 25 mL). The
EtOAc layer was washed with water (20 mL) and brine (20
mL), dried over anhydrous Na2SO4 (5 g), filtered, and
evaporated to dryness. The residue was further dried under
high vacuum at 20 °C overnight to obtain 320 mg (0.52
mmol, 54% recovery) of D-threo-1,2,3,5-pentanetetrol tetra-
p-toluate (15). This compound 15 was found to be 97% pure
by HPLC, tR 14.5 min; 1H NMR (CDCl3, 400 MHz) δ 7.92-
7.82 (8H, m, 4 × ArH2, ortho to -CO-), 7.24-6.09 (8H, m,
4 × ArH2, meta to -CO-), 5.86-5.76 (2H, m, H-2 and H-3),
4.82-4.34 (4H, three complex m, 2 × H-1 and 2 × H-5),
2.44-2.34 (14H, m containing singlets at δ 2.39, 2.38, and
2.36, 2 × H-4, 4 × Ar-CH3); 13C NMR (CDCl3, 100 MHz)
δ 166.6, 166.4, 165.8, 165.8, 144.2, 144.2, 144.0, 143.6,
130.4, 130.0, 130.0, 129.9, 129.8, 129.4, 129.3, 129.1, 127.3,
126.9, 72.4, 70.1, 62.7, 61.1, 30.0, 21.8; TOFMS-ES+ m/z
473.2 (MH+ - p-toluic acid), 609.3 (MH+), 626.3 (M +
+
TOFMS-ES+ m/z 506.4 (M + NH4 for 14), 507.4 (M +
+
+
NH4 + 1 for 14), 626.3 (M + NH4 for 15), 627.3 (M +
+
+
NH4 + 1 for 15) and 628.3 (M + NH4 + 2 for 15).
To a solution of the above compounds (480 mg, 14 and
15) in glacial acetic acid/TBME (4 mL, 1:1, v/v) at 5-10
°C was bubbled an excess of anhydrous HCl gas. After 1 h
at 5-10 °C a white solid crystallized out. The reaction was
continued for additional 2 h at 10-15 °C under argon. The
white solid was collected by vacuum filtration, washed with
TBME (5 mL), and dried in high vacuum at room-
temperature overnight to obtain pure 2-deoxy-3,5-di-O-p-
toluoyl-R-L-ribofuranosyl chloride (1); mp 116-117 °C;
mmp 114-116 °C; yield 40 mg, 7.2% over three steps from
the protected lactone 8; its 1H NMR in CDCl3 was identical
to that of the chlorosugar 1 obtained directly from the lactol
9 (described before); 13C NMR (CDCl3, 100 MHz) δ 166.6,
166.2, 144.5, 144.2, 130.1, 129.8, 129.4, 129.4, 126.9, 126.8,
95.5, 84.9, 73.7, 63.6, 44.7, 21.9, 21.8.
+
+
NH4 ), 627.3 (M + NH4 + 1).
Received for review March 15, 2005.
OP0500436
Recovery of D-threo-1,2,3,5-Pentanetetrol Tetra-p-tolu-
ate (15). The filtrate from the above step was concentrated
Vol. 9, No. 4, 2005 / Organic Process Research & Development
•
465