Hetero Diels - Alder-biocatalysis approach for the synthesis of (S)-3-[2-{(methylsulfonyl)oxy}ethoxy]-4-(triphenylmethoxy)-1-butanol methanesulfonate, a key intermediate for the synthesis of the PKC inhibitor LY333531

Article abstract of DOI:10.1021/op020202a

A cost-effective and easily scaled-up process has been developed for the synthesis of (S)-3-[2-{(methylsulfonyl)oxy}ethoxy]-4-(triphenylmethoxy)-1-butanol methanesulfonate, a key intermediate used in the synthesis of a protein kinase C inhibitor drug through a combination of hetero Diels - Alder and biocatalytic reactions. The Diels - Alder reaction between ethyl glyoxylate and bntadiene was used to make racemic 2-ethoxy-carbonyl-3,6-dihydro-2H-pyran. Treatment of the racemic ester with Bacillus lentus protease resulted in the selective hydrolysis of the R-enantiomer and yielded S-2-ethoxycarbonyl-3,6-dihydro-2H-pyran in excellent optical purity, which was reduced to S-3,6-dihdro-2H-pyran-2-yl methanol. Tritylation of this alcohol, followed by reductive ozonolysis and mesylation afforded the product in 10-15% overall yield and with >99% ee and chemical purity. Details of the process development work done on each step are given.

Full text of DOI:10.1021/op020202a

Organic Process Research & Development 2002, 6, 471−476
Hetero Diels-Alder-Biocatalysis Approach for the Synthesis of
(S)-3-[2-{(Methylsulfonyl) oxy}ethoxy]-4-(triphenylmethoxy)-1-butanol
Methanesulfonate, a Key Intermediate for the Synthesis of the PKC Inhibitor
LY3335311
Jean-Claude Caille*
PPG-SIPSY, Z. I. La Croix Cadeau B. P. 79, 49242, AVrille Cedex, France
C. K. Govindan* and Heiko Junga
PPG Industries Inc, 440 College Park Dr., MonroeVille, PennsylVania 15146, U.S.A.
Jim Lalonde and Yiming Yao
Altus Biologics Inc., 625 Putnam AVenue, Cambridge, Massachusetts 02139, U.S.A.
Abstract:
amenable to scale-up since yields in individual steps are high
and key intermediates are solids that can be easily purified
by crystallization. However, the method uses expensive
starting materials and reagents such as (R)-chloro-2,3-
propanediol or R-glycidol, vinylmagnesium bromide and
potassium tert-butoxide.
Two moles of ozone are required to cleave the double
bonds in the diallyl intermediate 3. Overall yield of 1
obtained by this method was 45-52%. An evaluation of the
Lilly method indicated to us that a lower-cost synthesis of 1
could be developed by the use alternate raw materials.
The Hetero Diels-Alder Strategy. A retro-synthetic
analysis of 1 as shown in Scheme 3 indicates that 3,6-
dihydro-2H-pyran derivatives could serve as key intermedi-
ates in the synthesis of 1 and that hetero Diels-Alder
reactions could be used to make these 3,6-dihydropyran
intermediates.
A cost-effective and easily scaled-up process has been developed
for the synthesis of (S)-3-[2-{(methylsulfonyl)oxy}ethoxy]-4-
(triphenylmethoxy)-1-butanol methanesulfonate, a key inter-
mediate used in the synthesis of a protein kinase C inhibitor
drug through a combination of hetero Diels-Alder and bio-
catalytic reactions. The Diels-Alder reaction between ethyl
glyoxylate and butadiene was used to make racemic 2-ethoxy-
carbonyl-3,6-dihydro-2H-pyran. Treatment of the racemic ester
with Bacillus lentus protease resulted in the selective hydrolysis
of the R-enantiomer and yielded S-2-ethoxycarbonyl-3,6-dihy-
dro-2H-pyran in excellent optical purity, which was reduced
to S-3,6-dihdro-2H-pyran-2-yl methanol. Tritylation of this
alcohol, followed by reductive ozonolysis and mesylation af-
forded the product in 10-15% overall yield and with >99%
ee and chemical purity. Details of the process development work
done on each step are given.
Hetero Diels-Alder reactions have been successfully
employed to prepare complex, highly functionalized mol-
ecules, including natural products.3 More recently, asym-
metric versions of this powerful reaction have been devel-
oped either by incorporating chiral auxiliaries in the dienophile
or by the use of asymmetric catalysis.4 Incorporation of the
glyoxyloyl group into an optically active molecule like (2R)-
bornane-10,2-sultam followed by reaction with 1-methoxy-
1,3-butadiene in the presence of europium catalysts has been
reported to produce diastereoisomerically pure 3,6-dihydro-
pyran derivatives.5 However, the reaction of this sultam with
butadiene did not occur readily and required pressures of
up to 10 kbar.6 Therefore, this reaction cannot be readily
Introduction
The title compound 1 (Scheme 1) is one of the two key
intermediates used by Eli Lilly and Co in the synthesis of 2,
a protein kinase C inhibitor. Compound 2, a staurosporine
analogue is in clinical trials for the treatment of retinopathy
and nephropathy in patients with diabetes mellitus. A recent
paper from Lilly describes2 the development of a process to
make 1 and 2 in multikilogram quantities. In this process,
an acyclic bis-indolylmaleimide is reacted with 1, and the
product obtained is converted to 2 in several steps as shown
in Scheme 1.
(3) For a review on hetero Diels-Alder reactions see: Bednarski. M. D.; J. P.
Lissikatos. J. P. In ComprehensiVe Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon Press: Oxford, U.K.; 1991; Vol.2, p 661.
(4) (a) For a review on asymmetric hetero Diels-Alder reactions see: Waldman,
H. Synthesis 1994, 535-551. (b) Mikami, K.; Motoyama, Y.; Terada, M.
J. Am. Chem. Soc. 1994, 116, 2812 and references therein.
(5) (a) Bauer, T.; Chapuis, C.; Kozak, J.; Jurczak, J. HelV. Chim. Acta 1989,
72, 482. (b) Bauer, T.; Chapuis, C.; Kozak, J.; Jurczak, J. J. Chem. Soc.,
Chem. Commun. 1990, 1178.
The method chosen by Eli Lilly and Co to prepare 1 in
large quantities is shown in Scheme 2. The synthesis is
* Corresponding authors.
(1) Part of this work has been described before. Caille, J.-C. U.S. Patent 6,-
300,106 B1, 2001.
(2) Faul, M. M.; Winneroski, L. L.; Krumrich, C. A.; Sullivan, K. A.; Gillig,
J. R.; Neel, D. A.; Rito, C. J.; Jirousek, M. R. J. Org. Chem. 1998, 63,
1961-1973.
(6) Jurczak, J. Private communication.
10.1021/op020202a CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/16/2002
Vol. 6, No. 4, 2002 / Organic Process Research & Development
•
471

Scheme 1
Scheme 2
a hetero Diels-Alder-biocatalysis approach is shown in
Scheme 4. In this study, we report the details of the
development of a commercial process to make 1 based on
this synthetic scheme.
Results and Discussion
Synthesis of 2-Carboethoxy-3,6-dihydro-2H-pyran, R,S-
4. We chose to develop the reaction between butadiene and
ethyl glyoxylate because these two raw materials are com-
mercially available and relatively inexpensive. It has been
reported that the reaction between butadiene and ethyl
glyoxylate ethyl hemiacetal affords 2-carboethoxy-3,6-di-
hydro-2H-pyran in 46% yield.9 The reaction between butyl
glyoxylate and butadiene has been reported to give 41% yield
of the corresponding 3,6-dihydropyran ester.9
scaled up. Jorgensen and Johannsen7have reported that
butadiene reacts slowly with isopropyl glyoxylate in the
presence of 2,2′-isopropylidene-bis[(4S)-4-tertbutyl-2-ox-
azoline] and copper triflate to produce S-2-isopropyloxycar-
bonyl-3,6-dihdro-2H-pyran in 55% yield and 87% ee. This
reaction is not considered to be of commercial importance
since the catalyst is very expensive and used in relatively
large quantities, the reaction is very slow, and the ee of the
product is not high enough to obtain an optically pure
product.
Of the two potential Diels-Alder reactions shown in
Scheme 3, the reaction between trityloxyacetaldehyde and
butadiene looks very attractive. This reaction would be highly
atom economic and if it can be done with a chiral catalyst,
would lead directly to 2-S-trityloxymethyl-3,6-dihydro-2H-
pyran 6 (Scheme 4) that could be used in place of 3. There
is precedent in the literature for such a hetero Diels-Alder
reaction. Ghosh and co-workers8 have reported that benzy-
loxyacetaldehyde reacts with Danishefsky’s diene in the
presence of Cu(II)-bis(oxazoline) complexes to yield the
corresponding dihydropyranone in 75% yield and 85% ee.
A complex mixture was obtained as the product when
butadiene was allowed to react with trityloxyacetaldehyde
in the presence of metal complexes.6 Therefore, we concen-
trated our efforts on developing a process based on the
reaction of butadiene with a glyoxylic acid ester. We
expected that it would be possible to resolve 2-carboalkoxy-
3,6-dihydropyrans through selective enzymatic hydrolysis of
one of the enantiomers. Our proposed synthesis of 1 using
Ethyl glyoxylate is commercially available as a 50%
solution in toluene, and all our experiments were carried out
with this solution. The presence of a solvent during the
reaction is beneficial since it helps solubilize the oligomers
and polymers formed during the reaction and makes the
reaction mixture less viscous and easily stirred. To carry out
the Diels-Alder reaction, ethyl glyoxylate solution and
hydroquinone were charged to an autoclave and heated to
150-170 °C. An excess of butadiene was then pumped into
the reaction mixture over 1-2 h. The vapor pressure of
butadiene at 160 °C is ∼40 bar. Due to the presence of
solvents, the maximum pressure observed during the reaction
was of the order of 15-16 bar. A pressure drop was observed
as butadiene was consumed and the pressure stabilized at
10-11 bar. Temperature was a critical factor in obtaining
good conversions. Below 155 °C, conversion of ethyl
glyoxylate to the product was lower, and up to 20% of ethyl
glyoxylate was recovered even when an excess of butadiene
was used. Under similar conditions <5% of ethyl glyoxylate
remained unreacted at 160-170 °C. The reaction was not
carried out at higher temperatures, since DSC data indicated
that ethyl glyoxylate is not very stable above 220 °C. Ethyl
glyoxylate that remains unreacted can be recovered and
recycled if desired. The reaction has also been carried out
by charging all the reactants together at the beginning, but
this procedure offered no improvement in yield or purity.
The addition of butadiene gradually to the reaction mixture
(7) Johanson M.; Jorgensen, K. A. J. Org. Chem. 1995, 60, 573.
(8) Gosh, A. K.; Mathivanan, P.; Cappiello, J. Tetrahedron Lett. 1997, 38,
2427-2430.
(9) Abuzov, A. Y.; Klimov, E. M.; Klimova, E. I. Dokl. Akad. Nauk SSSR
1962, 142, 341-343.
472
•
Vol. 6, No. 4, 2002 / Organic Process Research & Development

Scheme 3
Scheme 4
hydrolyzed acid was also determined after conversion of the
acid to its methyl ester. Several commercial formulations of
B. lentus protease and ChiroCLEC-BL (cross-linked enzyme
crystals from Altus Biologics) exhibited E-values > 50.
Aqueous formulations of B. lentus protease were selected
for further studies because of their lower cost. The low cost
of the enzyme also eliminated the need for its recovery and
reuse.
Reaction conditions for the resolution of 4 were optimized
using a ∼5% solution of B. lentus protease. Optimization
studies were performed using 0.3 mL of protease formulation,
1 g of 4, and 1 mL of 0.3 M phosphate buffer. Results of
our optimization studies can be summarized as follows: (1)
raising the temperature of the reaction from 25 to 45 °C
increased the activity of the enzyme (conversions of 42.5
and 52%, respectively, in 1 h) but reduced the selectivity
slightly (from an E value of 36.7 to 31.4). (2) The presence
of cosolvents such as toluene or acetonitrile reduced the
activity of the enzyme. (3) At higher substrate concentration,
relatively higher enzyme levels were required to get the same
rate conversion. (4) The pH of the reaction mixture was an
important factor in obtaining good selectivity and conversion.
At pH 7, selectivity was high (E ) 36.6), but conversion
was lower (39.3%) after 16 min. At pH 8, higher conversion
was obtained (50.5%) over the same period at the expense
of selectivity (E ) 31.5). At higher pH (>9), nonselective
base-catalyzed hydrolysis of the ester was observed. (5) The
rate of the reaction was slower when buffer concentration
was high. Thus, the conversion obtained in 1 M phosphate
buffer was only about 25% of that obtained in 0.1 and 0.3
M phosphate buffers under similar conditions (7 vs 27%).
Even on large scale, the resolution of 4 was carried out
easily by stirring the ester, buffer, and enzyme together at
20-30 °C. The pH during the reaction was maintained at
7-8 by the addition of sodium hydroxide. The reaction was
followed by chiral GC for the ee of S-4 in the organic phase
and was stopped when the desired ee was achieved. It was
possible to obtain 35-40% recovery of S-4 with ee g99.5%.
After the reaction, a solvent was added, and S-4 was
extracted out and recovered. As can be seen from the
Experimental Section, unlike that in many other enzyme-
catalyzed processes, the throughput in our process is high,
and work-up of the product is simple. As has been mentioned
before, the low cost of the enzyme also eliminated the need
for its recovery and reuse, thus making this resolution an
easy process to scale up.
is a safer process, since it minimizes the potential for the
exothermic polymerization of butadiene at higher tempera-
tures.
The crude product obtained in the Diels-Alder reaction
contained several impurities including unreacted ethyl gly-
oxylate, toluene, ethyl glyoxylate diethyl acetal, and buta-
diene oligomers and polymers. Most of these impurities were
separated from the product by fractional distillation. How-
ever, complete removal of ethyl glyoxylate diethyl acetal by
distillation was difficult without significant yield loss since
its boiling point was close to that of the product. Product
after distillation was 90-95% pure and it was used without
further purification in the next step. The isolated yield of 4
based on the amount of ethyl glyoxylate consumed ranged
from 50 to 60%.
Enzymatic Resolution of R,S-4. The resolution of
racemic esters by selective enzymatic hydrolysis of one of
the isomers is a well-known process. Many types of
hydrolases work in this application.10 To be economical, the
enzyme should be inexpensive, commercially available, and
highly active. It is also preferable to find an enzyme to
selectively hydrolyze R-4 to the acid, which can then be
removed easily by base extraction. S-4 can then be easily
recovered from the reaction mixture in high optical purity
even if the enzyme is only moderately selective.
Twenty-eight commercially available hydrolases (Chiro-
screen EH, Altus Biologics) were screened to identify an
enzyme that has the above characteristics. Initial screenings
were performed on small scale in a phosphate buffer (1 mL)
at pH 7 using 20 µL of ester and 100 mg of the enzyme.
Aliquots of the ester were analyzed by GC to determine the
extent of hydrolysis and selectivity. To determine the
enantioselectivity factor E,11 the optical purity of the
(10) Bornscheuer, U. T.; Kazlaauskas R. J. Hydrolases in Organic Synthesis;
Wiley-VCH: Weinheim, 1999.
(11) Chen, C. S.; Sih, C. J.; Girdukas, G. J. Am. Chem. Soc. 1982, 104, 7294.
We have demonstrated that R-3,6-dihydro-2H-pyran-2-
carboxylic acid can be recovered from the aqueous phase
Vol. 6, No. 4, 2002 / Organic Process Research & Development
•
473

by acidification and extraction. In principle, the acid can be
converted to an ester, reequilibrated to a racemic mixture,
and recycled through the process although the economic
viability of this approach has not been determined.
Reduction of S-4. The reduction of S-4 to (S)-3,6-
dihydro-2H-pyran-2-yl methanol 5 was carried out easily by
adding the ester slowly to solutions of lithium aluminum
hydride (LAH) in THF or adding LAH solution to the ester.
The reaction was done at 0-10 °C to reduce racemization.
The ee of the alcohol 5 obtained was generally close to that
of the starting ester. Purification of 5 can be carried out by
vacuum distillation, but we have demonstrated that the crude
product can be used as such in the next step. The yield of
the product in this step was 85-90%.
We have also demonstrated that this reduction can be done
using sodium borohyride (SBH) and methanol. However, the
reaction is difficult to control since SBH does not react
without methanol. The reaction between methanol and SBH
can become uncontrollable on large scale. Our previous
experience with similar reductions is that temperature control
is difficult in such reactions and racemization can occur at
higher temperatures. Therefore, LAH was chosen for scale-
up.
Tritylation of 5. Tritylation of R-glycidol was done in
dichloromethane by the Lilly group. Due to waste disposal
issues associated with this solvent, we chose to do the
tritylation of 5 in pyridine.12 Reaction mixtures were worked
up by water quench followed by filtration and recrystalli-
zation of the crude product. Recrystallization of the crude
product was done from 2-propanol containing small amounts
of triethylamine. The addition of triethylamine was necessary
to avoid the formation of trityl isopropyl ether by solvolysis
of the product when heated in 2-propanol. The yield of the
product in this step was 75-80%. Impurities such as ethyl
glyoxylate diethyl acetal carried over from previous steps
were removed during the isolation of 6. Crystallization of 6
may also offer an opportunity for enantiomeric enrichment
if it is required.
Reductive Ozonolysis of (S)-2-Trityloxymethyl-3,6-
dihydro-2H-pyran, 6. Ozone is one of the best reagents that
can be used to cleave a CdC bond.13 The reaction of ozone
with olefins is extremely fast. However, in absence of a
reactive solvent, ozonides and peroxides that are often very
unstable are produced as major products. These products
decompose on warming to give complex mixtures. The
mechanism of ozonolysis of olefins has been studied
extensively, and the Criegee mechanism is well accepted.
Ozonolysis of an olefin in the presence of methanol
produces a methoxy hydroperoxide and a carbonyl compound
as products. It is well known that these intermediates can
be reduced with sodium borohydride to produce alcohols.
Although 6 could give intermediates different from 3 during
ozonolysis, we anticipated that the intermediates formed from
6 would be reduced by sodium borohydride to yield 7.
Ozonolysis of 6 was studied at temperatures ranging from
-78 to -0 °C in mixtures of dichloromethane and methanol.
To determine the stability of the intermediates formed, carbon
and proton NMR spectra of the reaction mixtures were
recorded at low temperatures in CD2Cl2/CD3OD mixtures.
The intermediates formed were found to be stable below -30
°C. Diol 7 was obtained in high yields when the reaction
mixture was quenched with aqueous sodium borohydride.
If the reaction mixture was warmed to room temperature
before the quench, a complex mixture of products formed
as indicated by NMR spectrum. Similarly, only polymeric
products were formed when the ozonolysis was carried out
without methanol even at -78 °C. When ozonolysis was
done at -5 to 0 °C in the presence of methanol, formation
of trityl methyl ether was noticed, a result similar to which
has been observed before.2 Diol 7 was obtained in near
quantitative crude yield when ozonolysis was done below
-30 °C, followed by the addition of the reaction mixture to
cold, aqueous sodium borohydride solution. The product was
used without further purification in the next step.
Preparation of 1. The conversion of 7 to 1 in high yields
has been described by the Lilly group. The reaction of the
diol with methane sulfonyl chloride was carried out in
dichloromethane in the presence of triethylamine. The crude
product was then recrystallized from a mixture of ethyl
acetate and heptane. The reaction can also be done in ethyl
acetate, eliminating the use of dichloromethane. Care should
be exercised to remove all traces of acid from 1 since the
trityl group can cleave in the presence of acids, causing the
decomposition of product. The yield of the product in this
step was 85-90%. The chemical purity was >99% and ee
was >99.5%.
Summary
We have developed a commercial process for the syn-
thesis of 1 through a combination of chemical and biocata-
lytic reactions. Our overall yields are lower than that reported
in the previous synthesis.2 However, cost estimates indicate
that our process is very cost-competitive since we have been
able to eliminate the use of several expensive reagents and
simplify the overall process. The first four steps of the
process have been successfully scaled up to make several
hundred kilograms of 6, closely matching laboratory yields,
purity, and ee. Further efforts to reduce the cost of
manufacturing 1 will be concentrated on the recovery and
recycle of R-3,6-dihydro-2H-pyran2-carboxylic acid. Chiral
hetero Diels-Alder reactions have found applications in the
synthesis of other pharmaceutical intermediates such as
compactin and mevinolin.4,5 The strategy described in this
work should be applicable to these and similar compounds.
Experimental Section
Raw Materials. All raw materials used in this study were
purchased from commercial sources and used without further
purification.
(12) Green T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic Synthesis; John
Wiley & Sons: 1999, p 102-104.
(13) For a review see: Bailey, P. S. Ozonation in Organic Chemistry; Academic
Press: New York, 1978.
Analytical. HPLC analyses were performed on Varian
9000 series or HP1100 series instruments using reverse-phase
mode. A Zorbax Rx or XDB C8 column was used for the
474
•
Vol. 6, No. 4, 2002 / Organic Process Research & Development

analyses of 1, 7, and 8. Detection wavelength was 210 nm.
Gradient elution of solvents from 60% water (modified with
0.02% phosphoric acid) and 40% acetonitrile to 40% water
over 20 min, followed by holding for 20 min gave satisfac-
tory separations of most components. Solvent flow rate was
1.5 mL min-1. GC analyses were performed on a HP-6890
gas chromatograph with FID detector. A DB-1 column from
J&W Scientific was used for analyses of reaction mixtures
and products in the first three steps. GC conditions: 80 °C-
3min-10 °C/min-250 °C-10min, He flow at 1 mL/min.
A â-Dex column from Supelco or Cyclodex-B column from
J&W Scientific was used for the chiral purity analysis of 5.
GC conditions: 100 °C-10min-1 °C/min-115 °C-5 min.
NMR spectra were recorded on a Bruker EM-500 spectrom-
eter in CDCl3 unless otherwise noted.
chloride, and filtered through Celite. Toluene was removed
in vacuo to obtain 58.0 g of S-4 as a yellow oil (38%
recovery). Purity of the crude product was >97%, and ee
was 99.5% as determined by GC.
Preparation of (S)-3,6-Dihydro-2H-pyran-2-yl metha-
nol 5. A 1-L, four-neck round-bottom flask was equipped
with a mechanical stirrer, thermometer, and 500-mL addition
funnel.
The flask was purged with nitrogen and charged with 500
mL of 1 M lithium aluminum hydride solution in THF. The
solution was cooled to 0-5 °C. A solution of 150 g of S-4
in 140 mL of tetrahydrofuran was added at 0-8 °C over 80
min. After the addition, the reaction mixture was stirred at
6-12 °C for 2 h. The reaction mixture was quenched at
5-10 °C by careful addition of 19 mL of water. To the
mixture was then added 19 g of 15% sodium hydroxide
solution and 58 mL of water gradually. The reaction mixture
was stirred at room temperature for 2 h. The precipitated
aluminum hydroxide was removed by filtration. The pre-
cipitate was washed four times with 100 mL portions of
warm THF. The filtrates were combined and THF was
stripped in vacuo. To the residue was added 300 mL of
toluene and the solvent was again removed in vacuo to obtain
103.2 g (94%) of yellow oil. Purity of the product was 97%
as determined by GC analysis.
Preparation of 2-carboethoxy 3,6-Dihydro-2H-pyran
4: General Procedure. A high-pressure autoclave was
charged with a 50% solution of ethyl glyoxylate in toluene
and BHT (1 wt % based on the weight of ethyl glyoxylate
solution). The mixture was heated to 150-170 °C and
butadiene (1.2-1.4 mol/mol of ethyl glyoxylate) was added
over a period of 1-1.5 h. After the addition, the mixture
was stirred at the same temperature until a GC analysis of
the reaction mixture showed no further decrease in the level
ethyl glyoxylate (8-10 h). Excess butadiene was then purged
out with nitrogen. The solvent was removed in vacuo on a
rotary evaporator, and the residue was distilled fractionally
under vacuum. The fraction distilling at 85-95 °C and 10-
20 millibar was collected. The yield of 4 after accounting
for recovered ethyl glyoxylate varied from 50 to 60%,
depending on reaction conditions. The product thus obtained
was 90-95% pure on the basis of area % GC analysis. In
some experiments, volatile products were first separated from
the oligomers and polymers by flash vacuum distillation, and
the distillate obtained was subjected to fractional distillation.
The purity of the product obtained in this case was ∼97%.
The structure of the product was confirmed by GC-MS
Preparation of (S)-2-Trityloxymethyl-3,6-dihydro-2H-
pyran 6. A 3-L, four-neck round-bottom flask equipped with
a mechanical stirrer, thermocouple, and 125-mL addition
funnel was charged with 269 g of trityl chloride, 5.4 g of
DMAP and 600 mL of pyridine. A clear, yellow solution
resulted after 5 min of stirring. The temperature dropped from
23 to 14 °C. The mixture was heated to 20 °C over 15 min,
and 100 g of 5 was added over 20 min. An exotherm from
20 to 30 °C was observed. The reaction mixture was heated
to 50 °C over 20 min and stirred overnight. The reaction
mixture (brown slurry) was cooled to 20 °C. Water (1 L)
was added over 20 min. An exotherm from 20 to 26 °C was
noted, and a fine solid separated out. The mixture was stirred
at 5-10 °C for 1 h. The solids were collected by filtration
and washed with 0.7 L water. The solids were washed on
the filter with 0.5 L of heptane and sucked dry on the filter
for 30 min. The crude product was recrystallized from 1.2
L of a mixture of 9:1 2-propanol and triethylamine. The
crystals were filtered and washed with 200 mL of heptane
and vacuum-dried. The yield of the product was 240.7 g
1
analysis and NMR spectra. H NMR (500 MHz) δ 1.32 (t,
3H), 2.38 (m, 2H), 4.22 (m, 1H), 4.26 (q, 3H), 4.3 (dd, 2H),
5.74 (m, 1H), 5.84 (m, 1H). 13C NMR (500 MHz) δ 14.08,
27.61, 60.95, 65.41, 71.99, 122.81, 125.96, 171.27.
Resolution of 2-Carboethoxy-3,6-dihydropyran R,S-
4. A 1-L reactor was charged with 360 mL of a pH 7
phosphate buffer (made by dissolving 39.2 g KH2PO4 and
10.37 g of NaH2PO4 in 1 L water) and 150 g of 2-carbo-
ethoxy-3,6-dihydropyran. The pH of the mixture was re-
adjusted to 7.0 by the addition of 50% sodium hydroxide. A
commercial formulation of B. lentus protease (90 mL, ∼5%
solution of the enzyme) was then added, and the mixture
was stirred at room temperature for 8 h. The pH of the
reaction mixture was maintained between 7 and 8 by the
addition of 50% sodium hydroxide solution with an automatic
pH controller. Samples of the organic phase were analyzed
periodically by GC to determine the extent of reaction. After
8 h, the ee of S-ester in organic phase was >99.5%. The pH
of the mixture was adjusted to 8, and the mixture was
extracted twice with 100 mL each of toluene. Organic phases
were combined, washed with 100 mL of saturated sodium
1
(77.2%). Mp 120-123 °C. H NMR δ 1.74 (m, 2H) 2.96
(br s 1H) 3.07 (br m 1H) 3.4-3.7 (m, 3H) 4.64-4.82 (m,
2H) 7.19 (m, 3H) 7.26 (m, 6H) 7.43 (m, 6H).
Ozonolysis: Low-Temperature NMR Studies. (1) A
solution of 0.2 g of 6 in 0.4 mL of CD2Cl2 was cooled in a
dry ice-acetone bath, and ozonated air was bubbled through
until the solution turned blue. The solution was then purged
with nitrogen to remove excess ozone. Proton NMR spectrum
of the solution was recorded at -40 °C and showed only
very broad peaks, indicating the presence of polymeric
products. A white precipitate formed when the solution was
diluted with acetone. (2) A solution of 0.2 g of 6 was
prepared in a mixture of 1.5 mL of CD2Cl2 and 1 mL of
Vol. 6, No. 4, 2002 / Organic Process Research & Development
•
475

CD3OD.The solution was cooled in a dry ice-acetone bath.
Ozonolysis was carried out as above, and the resulting
solution was split into two portions. CMR and PMR spectra
of first portion were recorded at -40 °C. The second portion
was warmed to room temperature, and NMR spectra were
recorded after 1.5 h. The spectra of the latter sample were
very different from that of the former and showed the
presence of a complex mixture.
cooled to 0-5 °C, and a solution of 43.5 mL of methane
sulfonyl chloride in 50 mL of dichloromethane was added
at 0-5 °C over 80 min. The reaction mixture was the stirred
for 1 h and then diluted with 500 mL of dichloromethane. It
was then extracted twice with 315 mL of water and once
with 315 mL of saturated sodium bicarbonate solution. The
organic phase was dried over magnesium sulfate and filtered.
The solution was evaporated to dryness on rotary evaporator
to obtain 111.1 g of crude product. The crude product was
dissolved in 345 mL of ethyl acetate at 35-40 °C and
filtered. The filtrate was charged to a 2-L reactor, and 690
mL of heptane was added dropwise over 3.5 h. The mixture
was stirred for 90 min and then filtered in a pressure filter.
The filter cake was reslurried in 300 mL of heptane and
filtered, washing the crystals with additional heptane. The
crystals were dried in vacuo to obtain 96.7 g (84.5%) of 1.
HPLC analyses indicated the product to have ee and chemical
purity >99%.
Preparation of (S)-3-(2-Hydroxyethoxy)-4-(triphenyl-
methoxy)-1-butanol 7. A jacketed reactor equipped with a
fritted gas inlet, thermocouple, and dry ice condenser were
charged with 71.5 g (0.2 mol) of 6, 350 mL of dichlo-
romethane, and 250 mL of methanol. The clear, yellow
solution was cooled to -40 °C, and ozonated air (generated
by a PCI ozone generator) was bubbled through the solution
at -40 to -25 °C until the solution turned light blue (∼85
min). Nitrogen was bubbled through the reaction mixture at
-45 to -50 °C for 20 min until the blue color disappeared.
A 1-L flask was charged with 17.5 g of sodium boro-
hydride and 500 mL of 0.02 N caustic and cooled to -5 °C
with stirring. The cold reaction mixture obtained after
ozonolysis was added to the sodium borohydride solution
over 10 min. The mixture was then allowed to slowly warm
to room temperature and stirred overnight. The phases were
separated, and the aqueous phase was back-extracted with
50 mL of dichloromethane. The combined organic phases
were evaporated to dryness on a rotary evaporator to obtain
81.9 g (100%) of viscous oil that was used in the following
experiment.
Acknowledgment
We express our appreciation to PPG Industries Inc. for
permission to publish this work. We acknowledge the help
of Dr. Tim Flood in recording and interpreting NMR spectra
and of Dr. Felicia Tang with HPLC analyses. We also thank
Professor Janusz Jurczak (Polish Academy of Sciences,
Warszawa, Poland) for preliminary work.
Note Added after ASAP: There were errors in the
Preparation of (S)-2-Trityloxymethyl-3,6-dihydro-2H-
pyran 6 published in the version posted ASAP March 16,
2002. The corrected version was posted ASAP April 26,
2002.
Preparation of (S)-3-[2-{(Methylsulfonyl)oxy}ethoxy]-
4-(triphenylmethoxy)-1-butanol Methanesulfonate 1. A
2-L reactor was charged with solution of 81.9 g (0.21 mol)
of (S)-3-(2-hydroxyethoxy)-4-(triphenylmethoxy)-1-butanol
obtained in the previous experiment and 700 mL of dichlo-
romethane and 87 mL of triethylamine. The mixture was
Received for review January 2, 2002.
OP020202A
476
•
Vol. 6, No. 4, 2002 / Organic Process Research & Development