A concise, efficient and production-scale synthesis of a protected L-lyxonolactone derivative: An important aldonolactone core

Article abstract of DOI:10.1021/op050222n

A multikilogram-scale synthesis of L-lyxonolactone-2,3-O-isopropylidene is reported. It proceeds efficiently from an optimized, large-scale, aqueous bromine oxidation of D-ribose to D-ribonolactone including a one-pot isopropylidene formation, and subsequent conversion of the D-ribonolactone-2,3- 0-isopropylidene to L-lyxonolactone-2,3-O-isopropylidene via the derived C ₅-mesylate and intramolecular relactonization of the product of aqueous potassium hydroxide cleavage of the D-ribonolactone ring. The inversion of configuration at the C₄-chiral center is understood in terms of an intermediating C₄-C₅-epoxide. The overall process is noteworthy for its operational simplicity, stereochemical integrity, and use of inexpensive chemicals.

Full text of DOI:10.1021/op050222n

Organic Process Research & Development 2006, 10, 484−486
A Concise, Efficient and Production-Scale Synthesis of a Protected
L-Lyxonolactone Derivative: An Important Aldonolactone Core
Hitesh Batra, Robert M. Moriarty, Raju Penmasta, Vijay Sharma, Gabriela Stanciuc, James P. Staszewski,
Sudersan M. Tuladhar, and David A. Walsh*
United Therapeutics Corporation, Research and DeVelopment, 2225 West Harrison Street, Chicago, Illinois 60612, U.S.A.
Satyanarayana Datla and Sunderraj Krishnaswamy
Merrifield Pharma PVt. Ltd., 80B, Kolhar Industrial Area, Bidar-585403, Karnataka, India
Abstract:
A multikilogram-scale synthesis of
propylidene is reported. It proceeds efficiently from an opti-
mized, large-scale, aqueous bromine oxidation of -ribose to
-ribonolactone including a one-pot isopropylidene formation,
L-lyxonolactone-2,3-O-iso-
D
D
and subsequent conversion of the D-ribonolactone-2,3-O-iso-
propylidene to
derived C -mesylate and intramolecular relactonization of the
product of aqueous potassium hydroxide cleavage of the
-ribonolactone ring. The inversion of configuration at the C
chiral center is understood in terms of an intermediating C
-epoxide. The overall process is noteworthy for its operational
L-lyxonolactone-2,3-O-isopropylidene via the
Results and Discussion
5
Our initial route to 1 involved periodate cleavage of
D-gulonolactone-2,3-O-isopropylidene (5) followed by so-
dium cyanoborohydride reduction of the thus formed alde-
D
4
-
-
4a-c
4
hyde (Scheme 1) (3 f 4 f 5 f 6 f 1).
C
5
Although this route provided access to the first few
kilograms of compound 2, it suffered from a number of
limitations:
simplicity, stereochemical integrity, and use of inexpensive
chemicals.
1
. D-Gulonolactone (3) is a very expensive and not a
readily available material.
. All the intermediates required chromatographic puri-
fication.
2
Background
Aldonolactones find wide application in contemporary
organic synthesis because they are inexpensive and serve as
valuable chiral synthons for a range of natural products and
3
. Attempts to scale-up the periodate cleavage step (5 f
6
) failed on a 500-g scale.
4
. The aldehyde formed in step (5 f 6) was unstable
and decomposed rapidly.
. Selective deprotection in step 4 f 5 was difficult to
1
drugs. They are readily accessible from the corresponding
2
aldose by several methods for anomeric oxidation. Activated
aldonolactones have been particularly useful in the synthesis
of iminosugars. In this connection we required substantial
quantities (200-400 kg) of L-lyxonolactone-2,3-O-isopro-
pylidene (1), a key intermediate in our synthesis of UT-231B
5
control on a large scale as the second isopropylidene group
was sensitive to heat and acid.
3
6
. Handling of periodic acid and sodium cyanoborohy-
dride was problematic on a large scale as the workup for
steps 5 f 6 f 1 generated a large amount of toxic
byproducts which raised safety concerns.
(iminosugar, 2), an analogue of deoxynojirimycin. Imino-
sugar 2 is currently in clinical trials as an antiviral agent for
the treatment of hepatitis C.
Because of the need for 1 in the synthesis of 2, a more
reliable synthesis for this critical intermediate was required.
This goal was accomplished using the synthetic route shown
in Scheme 2; namely, oxidation of D-ribose (7) to D-
ribonolactone (8) followed by isopropylidene formation
(8 f 9), mesylate formation (9 f 10), and base hydrolysis
of 10 with intramolecular displacement of the mesylate group
(10 f 1).
*
To whom correspondence should be addressed. Telephone: 001-312-421-
1
819. Fax: 001-312-275-0364. E-mail: dwalsh@unither.com.
(1) (a) Rassu, G.; Zanardi, F.; Battistini, L.; Casiraghi, G. Synlett 1999, 1333.
(b) Lundt, I. Top Curr. Chem. 1997, 187, 117. (c) de Lederkremer, R. M.;
Varela, O. AdV. Carbohydr. Chem. Biochem. 1994, 50, 125. (d) Fleet, G.
W. J. Sugar Latcones as Starting Materials. In Antibiotics and AntiViral
Compounds: Chemical Synthesis and Modifications; Krohn, K., Kirst, H.
A., Maas, H., Eds.; VCH: Weinheim, 1993; p 333.
(
2) Lundt, I. Reaction on C: Oxidation, Reduction and Deoxygenation at the
Anomeric Position. In Glycoscience: Chemistry and Chemical Biology;
Fraser-Reid, B., Tatsuta, K., Thiem, J., Eds.; Springer: Heidelberg, 2000.
3) (a) Lundt, I.; Madsen, R. Iminosugars as Powerful Glycosidase Inhibitorss
Synthetic Approaches from Aldonolactones. In Iminosugars as Glycosidase
Inhibitors-Norjirimycin and Beyond; St u¨ tz, A. E., Ed.; Wiley-VCH: Wein-
heim, 1999; p 93. (b) Withers, S. G.; Namchuuk, M.; Most, R. In
Iminosugars as Glycosidase Inhibitors-Norjirimycin and Beyond; St u¨ tz, A.
E., Ed.; Wiley-VCH: Weinheim, 1999; p 188.
The key transformation 10 f 1 was confirmed on a
laboratory scale prior to scale-up using the literature proce-
(
(4) (a) Fleet, G. W. J.; Ramsden, N. G.; Witty, D. R. Tetrahedron Lett. 1988,
29, 2871. (b) Fleet, G. W. J.; Ramsden, N. G.; Witty, D. R. Tetrahedron
1989, 45, 319. (c) Varela, O.; Zunszain, P. A. J. Org. Chem. 1993, 58,
7860.
4
84
•
Vol. 10, No. 3, 2006 / Organic Process Research & Development
10.1021/op050222n CCC: $33.50 © 2006 American Chemical Society
Published on Web 04/25/2006

Scheme 1a
Scheme 3
a
Conditions: (i) Acetone, H
2
SO
4
(cat.), 2,2-dimethoxypropane. (ii) 80%
, acetic acid.
gave mesylate (10). However, scaling up of this step (9f10)
to a 200-kg scale required several process modifications of
the literature methods. Specifically, it was determined that
the optimal temperature for the reaction was -20 °C for the
mixing of the reagents, and an 8-h reaction time at room
temperature was required. Without purification, the crude
mesylate (10) was subjected to aqueous potassium hydroxide
treatment. After workup and crystallization of crude 1 from
Acetic acid in water. (iii) H
5
6
IO , THF. (iv) NaCNBH
3
Scheme 2a
2
-propanol, an analytically acceptable product was obtained
in 59% yield.
The mechanism of the inversion of the configuration at
the C
envision a process in which an intermediary epoxide 12 of
the retained configuration at C undergoes subsequent
intramolecular ring opening of the epoxide by the carboxylate
nucleophile in a favorable 5-endo-tet process proceeds with
inversion of configuration to yield 1 (Scheme 3).
4
chiral center in 12 to yield 1 deserves comment. We
a
4
Conditions: (i) Br
2
, H . (ii) Acetone, H
2
O, K
2
CO
3
2 4
SO (cat.). (iii) Mesyl
chloride, Et
3
N, CH
2
Cl
2
. (iv) KOH, H O.
2
12
5
dure. The starting point for the synthesis of 1 was the large-
scale production of D-ribonolactone-2,3-O-isopropylidene (9).
There are various methods known in the literature for the
synthesis of 9;
Caperelli et al. showed that none of the methods was suitable
In conclusion, this is the first study of such a large-scale,
practical, and facile synthesis of 1 with an overall yield of
6
-10
6
however, reports by Kaskar et al. and
3
8% starting from D-ribose (7). Compared to our earlier
synthetic process (Scheme 1) this current process (Scheme
) is easily scalable, efficient, and cheaper in terms of use
7
for larger quantities of 1. At this stage we investigated the
conventional methods of oxidations in sugar chemistry, and
2
2
of chemicals involved and control of parameters. We have
prepared more than 400 kg of compound 9 and 200 kg of
compound 1 using this methodology. The procedure used
for synthesis of UT-231B from 2 will be reported in a future
communication.
the oxidation of commercially available D-ribose (7) with
bromine and water worked very well on a 200-kg scale for
the synthesis of D-ribonolactone (8) as shown in Scheme 2.
D-Ribose (7) was oxidized to D-ribonolactone (8) using
bromine/water in the presence of potassium carbonate. In
situ protection of the cis-diol of D-ribonolactone (8) with
acetone in the presence of a catalytic amount of sulfuric acid
afforded D-ribonolactone-2,3-O-isopropylidene (9). This step
was scaled to 200 kg with a yield of 65%. This process is
noteworthy when we consider the key role that D-ribono-
lactone plays as a “chiral cornerstone” in a wide range of
Experimental Section
All chemicals were commercially available. Melting
points were determined on a Fischer Scientific melting point
apparatus and are uncorrected. Optical rotations were mea-
sured on a DigiPol 781 automatic polarimeter (Rudolph
instrument) at a wavelength of 589 nm (sodium D line) in a
1.0-dm cell with a total volume of 1 mL. Specific rotation
11
syntheses of acyclic, cyclopentanones, and bicyclics. Using
a literature procedure, mesylation of 9 with methanesulfonyl
5
chloride in the presence of triethylamine in dichloromethane
[R]
temperature, and concentration (c) was given in grams per
00 mL in the specified solvent. Infrared spectra were
recorded on a Thermo Nicolet AVATAR 360 FT-IR, using
KBr pellets. The NMR spectra were determined on a JEOL
D
is reported in degrees per decimeter at the specified
(
5) Godskesen, M.; Lundt, I.; Madsen, R.; Winchester, B. Bioorg. Med. Chem.
996, 4, 1857.
6) Kaskar, B.; Heise, G. L.; Michalak, R. S.; Vishnuvajjala, B. R. Synthesis
990, 1031.
1
1
(
1
(7) Liu, D.; Caperelli, C. A. Synthesis 1991, 933.
(8) Weimberg, R. Biol. Chem. 1961, 236, 629.
(9) Morgenlie, S. Acta Chem. Scand. 1973, 27 (7), 2607.
3
00 MHz spectrometer, and elemental analyses were per-
formed by Atlantic Microlab, Atlanta, Georgia.
(
10) Pudlo, J. S.; Townsend, L. B. In Nucleic Acid Chemistry; John Wiley &
Sons: New York, 1991; Vol. 4, pp 51-53.
(11) Aldrichimica Acta 1989, 22, 49.
(12) Baldwin, J. E. J Chem. Soc., Chem. Commun. 1976, 734.
Vol. 10, No. 3, 2006 / Organic Process Research & Development
•
485

D-Ribonolactone-2,3-O-isopropylidene (9). D-Ribose (75
kg, 500 mol, 1.0 equiv) was dissolved in water (207 L) at
ambient temperature and cooled to 10-15 °C. To this cold,
clear solution was added solid potassium carbonate (82.5 kg,
was added dichloromethane (750 L) and triethylamine (29.8
kg, 41.0 L, 294 mol, 1.1 equiv) under nitrogen. This mixture
was stirred at ambient temperature until it became clear, and
then it was cooled to -20 °C. To this solution was slowly
added methanesulfonyl chloride (33.3 kg, 22.5 L, 291 mol,
1.1 equiv), and the solution was stirred for 1 h at -20 °C.
The solution was allowed to attain ambient temperature, and
it was maintained at this temperature for 8 h. The reaction
was quenched with water (400 L), and the organic layer was
separated. The aqueous layer was extracted with dichloro-
methane (2 × 200 L). The organic layers were combined,
washed with water (2 × 200 L), and concentrated in vacuo
below 30 °C to obtain the semisolid mesylate (10). To this
crude mesylate was added a solution of potassium hydroxide
(36.5 kg, 651 mol, 2.44 equiv) in 245 L of water maintaining
the temperature below 30 °C. This solution was stirred for
4 h at the same temperature and then adjusted to pH 2.5 to
3.0 by adding 3 M hydrochloric acid (33.5 L). The acidic
solution was concentrated in vacuo maintaining the temper-
ature below 45 °C to afford a solid mass. The solid mass
was triturated with acetone (300 L) and heated to reflux.
The acetone was decanted, dried over sodium sulfate (35
kg), and filtered. The clear filtrate was concentrated in vacuo
below 35 °C to yield the crude product. The product was
crystallized from 2-propanol (100 L) to afford white,
crystalline L-lyxonolactone-2,3-O-isopropylidene (1), 17.1 kg
(43.1%, on a 200-kg scale the yield improved to 59%); mp
597 mol, 1.2 equiv) while maintaining the internal temper-
ature between 10 and 15 °C. The reaction solution was cooled
to 0-5 °C, and bromine (87.4 kg, 547 mol, 1.1 equiv) was
added slowly with stirring over a period of 3-4 h. (Cau-
tion: The temperature was kept below 5 °C during the
addition of bromine as this step was highly exothermic, and
care must be taken while handling bromine. Protective
clothing was used at all times, including full face respirator
equipped with an NIOSH-approved organic vapor-acid gas
canister, laboratory coat, and protective gloves.) After the
addition of bromine was complete, the reaction mixture was
stirred overnight while allowing the temperature to rise to
ambient. The solution was treated slowly with 88% formic
acid (25 L) at room temperature to adjust the pH to 2-3.
The acidic solution was concentrated in vacuo at 50 °C to
yield a brown solid. This brown solid was dissolved by
adding acetone (1500 L, 1185 kg, 20 403 mol, 41 equiv).
To this clear solution was added 4 Å molecular sieves (37.5
kg) and concentrated sulfuric acid (7.5 L). The reaction
mixture was heated at reflux temperature for 4 h under
nitrogen. Then the mixture was cooled to 25-30 °C and
maintained at this temperature for 8 h. The reaction mixture
was adjusted to a pH of 5.5-6.0 by adding sodium hydroxide
flakes (30 kg). The reaction mixture was filtered, and the
filtrate was concentrated in vacuo below 40 °C to afford a
crude, solid mass. This crude solid was dissolved by stirring
in ethyl acetate (250 L) at 40 °C. The solution was filtered,
and the filtrate was concentrated in vacuo to 50% of the
original volume. This solution was cooled to -5 °C to obtain
solid D-ribonolactone-2,3-O-isopropylidene (9) that was
collected by filtration and dried under vacuum at 35 °C. The
yield of 9 was 51.7 kg (55%; on a 200-kg scale the yield
5
25
98-99 °C (lit mp 92-93 °C), [R]
D
-89° (c ) 1.0,
-85.6° (c ) 1.0, acetone)]. IR (KBr):
): δ 1.36 (s,
), 2.72 (brs, 1H, OH), 3.75-
5
20
1
acetone), [lit [R]
3423, 1778 cm . H NMR (300 MHz, CDCl
D
-
1
3
3H, CH
3
), 1.44 (s, 3H, CH
3
3.77 (d, 1H, C-5′H), 3.91-4.01 (m, 2H, C-5H and C-5′H),
4.60-4.64 (m, 1H, C-4H), 4.83-4.89 (m, 2H, C-3H and
C-2H). Anal. Calcd for C
8 12 5
H O : C, 51.06; H, 6.38. Found:
C, 50.96; H, 6.44.
13
improved to 65%); mp 136-138 °C (lit mp 138-139 °C).
Acknowledgment
-
1 1
IR (KBr): 3455, 1762 cm . H NMR (300 MHz, CDCl
3
):
Helpful discussions with Professor G. W. J. Fleet, Dyson
Perrins Laboratory, Oxford University were of substantial
value. The contribution of analytical work by Kunyuan Mao
of United Therapeutics Corporation is gratefully acknowl-
edged.
δ 1.36 (s, 3H, CH ), 1.41 (s, 3H, CH ), 2.95 (brs, 1H, OH),
3
3
3.75-3.80 (dd, 1H, C-5′H), 3.95-3.99 (dd, 1H, C-5H),
4.60-4.62 (m, 1H, C-4H), 4.74-4.78 (d, 1H, C-3H), 4.80-
4.83 (m, 1H, C-2H).
L-Lyxonolactone-2,3-O-isopropylidene (1). To D-ribono-
lactone-2,3-O-isopropylidene (9) (50 kg, 266 mol, 1.0 equiv)
Received for review November 10, 2005.
OP050222N
(13) Hough, L.; Jones, J. K. N.; Mitchell, D. L. Can. J. Chem. 1958, 36, 1720.
486
•
Vol. 10, No. 3, 2006 / Organic Process Research & Development