A practical synthesis of L-valyl-pyrrolidine-(2R)-boronic acid: Efficient recycling of the costly chiral auxiliary (+)-pinanediol

Article abstract of DOI:10.1021/op025587b

A practical synthesis of L-valyl-pyrrolidine-(2R)-boronic acid (1) is detailed. A previously disclosed synthesis of 1 (Snow, R.; Kelly, T. R.; Adams, J.; Coutts, S.; Perry, C. (Boehringer Ingelheim Pharmaceuticals, Inc.). WO 93/10127, 1993) was significantly improved by developing an efficient process for recycling the costly chiral auxiliary (+)- pinanediol.

Full text of DOI:10.1021/op025587b

Organic Process Research & Development 2002, 6, 814−816
A Practical Synthesis of L-Valyl-pyrrolidine-(2R)-boronic Acid: Efficient
Recycling of the Costly Chiral Auxiliary (+)-Pinanediol
Frank S. Gibson,* Ambarish K. Singh, Maxime C. Soumeillant, Percy S. Manchand, Michael Humora, and
David R. Kronenthal
Process Research and DeVelopment, The Bristol-Myers Squibb Pharmaceutical Research Institute, One Squibb DriVe,
New Brunswick, New Jersey 08903, U.S.A.
Abstract:
A practical synthesis of L-valyl-pyrrolidine-(2R)-boronic acid
(1) is detailed. A previously disclosed synthesis of 1 (Snow, R.;
Kelly, T. R.; Adams, J.; Coutts, S.; Perry, C. (Boehringer
Ingelheim Pharmaceuticals, Inc.). WO 93/10127, 1993) was
significantly improved by developing an efficient process for
recycling the costly chiral auxiliary (+)- pinanediol.
Figure 1. L-Valyl-pyrrolidine-(2R)-boronic acid.
which was highly crystalline and exhibited good physico-
chemical properties.
Introduction
Peptides containing boronic acid analogues of amino acids
have recently been investigated for their anticancer and serine
protease inhibiting activities.2 Evidence suggests that these
types of compounds could also be useful as immunosup-
pressants.3 The proline surrogate pyrrolidine-2-boronic acid
imparts significant proteinase inhibitory activity towards
bacteria when present as a C-terminus residue in certain
peptides.4 Our interest was focused on developing a practical
synthesis of L-valyl-pyrrolidine-(2R)-boronic acid 1 (Figure
1). This compound is an immunostimulant and may serve
as a potential adjuvant to standard chemotherapy.
Scale-up of the reported synthesis of 11 was challenging
due to the use of a number of hazardous reagents as well as
inefficient and tedious purification steps. Furthermore,
expensive (+)-pinanediol was utilized as an unrecovered
resolving agent. At a bulk price of nearly $10 000 per kg,
and without an efficient recycle in place, this material initially
comprised >90% of the overall API cost. Because our
production of 1 was projected to be less than 2 kg/yr due to
microgram-dosing levels, it was unlikely the price of bulk
pinanediol could be significantly reduced, as would normally
be the case for typical manufacturing volumes. Additionally,
the HCl salt 1a was not suitable for further development
because of its amorphous and deliquescent nature. An
extensive search for a more suitable salt form eventually led
to the selection of methanesulfonic acid (MSA) salt 1b,
Results and Discussion
The previously disclosed synthesis1 of the key intermedi-
ate 6a was demonstrated on >100-g scale with several
modifications as outlined in Scheme 1. THF was an
acceptable substitute for diethyl ether as the solvent in the
lithiation of 2, and TMEDA was not required as an additive.
With appropriate precautions, we were able to safely
handle sec-butyllithium on multiliter scale. Attempts to
employ a less hazardous reagent for the R-metalation were
unsuccessful.5 The boronic acid 3 was condensed with (+)-
pinanediol, and after removal of the Boc group under acidic
conditions, 6a was obtained as a crystalline solid with
diastereomeric excess (de) of 92%. A single recrystallization
from IPA produced 6a with de >99%.1b Initial supplies of
HCl salt 1a were prepared using the previously reported
synthesis as outlined in Scheme 2. EDAC-mediated coupling
of 6a with Boc-valine gave 8a, which was subsequently
deprotected using anhydrous HCl. The pinanediol auxiliary
was then removed by an exchange process using phenylbo-
ronic acid. Salt-exchange to the more desirable MSA salt
1b was accomplished by the addition of 1 equiv of MSA to
a solution of 1a in acetone-water, resulting in the crystal-
lization of 1b.
While preparing the initial supplies of the API, we
recognized that to devise a safe, scalable, and cost-effective
synthesis of the preferred salt form 1b, we needed to address
the following issues: (1) recycling of the (+)-pinanediol to
reduce the overall cost of goods, (2) elimination of other
costly or environmentally unfriendly reagents and solvents
(e.g., EDAC and dichloromethane), and (3) development of
a process for the direct isolation of 1b without going through
1a.
* Corresponding authors. E-mail: frank.gibson@bms.com.
(1) (a) Snow, R.; Kelly, T. R.; Adams, J.; Coutts, S.; Perry, C. (Boehringer
Ingelheim Pharmaceuticals, Inc.). WO 93/10127, 1993. (b) Kelly, T. A.;
Fuchs, V. U.; Perry, C. W.; Snow, R. J. Tetrahedron 1993, 1009. (c)
Martina, S.; Enkelmen, V.; Wegner, G.; Schluter, A. D. Synthesis 1991,
613.
(2) (a) Matteson, D. S.; Sadhu, H. M.; Lienhard, G. E. J. Am. Chem. Soc.
1981, 5241. (b) Kettner, C. A.; Shenvi, A. B. J. Biol. Chem. 1984, 15106.
(c) Kinder, K. H.; Katzenellenbogen, J. A. J. Med. Chem. 1985, 1917.
(3) Flentke, G. R.; Munoz, E.; Huber, B. T.; Plaut, A. G.; Kettner, C. A.;
Bachovchin, W. W. Proc. Natl. Acad. Sci. U.S.A. 1991, 1556.
(4) Bachovchin, W. W.; Plaut, A. G.; Flentke, G. R.; Lynch, M.; Kettner, C.
A. J. Biol. Chem. 1990, 3738.
(5) (a) Beak, P.; Zajdel, W. J.; Reitz, D. B. Chem. ReV. 1984, 471. (b) Meyers,
A. I. Aldrichimica Acta 1985, 59. (c) Comins, D. L.; Weglarz, M. A. J.
Org. Chem. 1988, 4437. Beak, P.; Lee, W. K. J. Org. Chem. 1990, 55, 8.
814
•
Vol. 6, No. 6, 2002 / Organic Process Research & Development
10.1021/op025587b CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/23/2002

Scheme 1
Scheme 3. Recycling of (+)-pinanediol
Scheme 4
Scheme 2
Recycling of (+)-Pinanediol. We investigated a variety
purification and was successfully used to repeat the sequence
leading to 1b. 1b produced from recycled pinanediol showed
no difference in quality from 1b derived from fresh reagents,
even after three iterations. Trace impurities carried through
with the recycle are effectively purged during the crystal-
lization of 6a.
of hydrolytic and oxidative methods for the removal of the
pinanediol auxiliary6 from 8a and 9a; however, none of these
was as effective as the reported exchange method using
phenylboronic acid in terms of quality and yield. All attempts
to reclaim pinanediol from the resulting phenylboronic acid
pinanediol ester also failed. To recycle the expensive
auxiliary, we eventually envisioned that the exchange of (+)-
pinanediol from 9a would likely be successful with any fairly
organic-soluble boronic acid and recognized that intermediate
3 (in racemic form) might be the perfect choice, for two
important reasons. First, the use of 3 to effect the transes-
terification would eliminate the need for phenylboronic acid
and, of greater importance, would in fact result in an efficient
recycling of the pinanediol, delivering early intermediate 5
as the only byproduct. It was also recognized that 3 would
likely be able to recapture much of the (+)-pinanediol lost
with the mother liquor containing 6b during the crystalliza-
tion of 6a. If successful, almost all of the pinanediol used
could be recycled, dramatically reducing the cost of API.
When examined experimentally, both exchange reactions
proved to be completely successful. Treatment of 9a with 3
in aqueous MTBE generated the desired 1a and 5 in less
than 1 h in near quantitative yield. Furthermore, reaction of
3 with the mother liquor containing 6b rapidly produced 5
and prolineboronic acid 10 (Scheme 3). The efficiency of
these exchanges is due to the fact that 5 is completely
insoluble in water and completely partitions to the MTBE
layer, driving the process to completion. Compound 5
produced in either exchange reaction required no further
N-Trifluoroacetyl-valine Route to 1b. Having developed
a process for the recycling of (+)-pinanediol, we turned our
efforts towards improving the coupling of 6a with a suitable
valine derivative and the subsequent isolation of 1b. From
previous experience, we had learned that simple dipeptide
formation could be easily promoted, with minimal racem-
ization, using the acid chloride of an N-TFA protected amino
acid. If this coupling method succeeded here, removal of
the TFA group would be easily telescoped with isolation of
desired 1b. Consequently, TFA-valine (11)7 was treated with
the Vilsmeier reagent (chloromethylene-dimethylammonium
chloride) in ethyl acetate at 0 °C to generate 12, which was
added to a solution of 6a in aqueous NaHCO3 at 0 °C to
produce 8b in 90% yield and >98% purity after crystalliza-
tion (Scheme 4).
HPLC analysis of the crude product indicated about 1.5%
of the undesired val-(2R) epimer, while the crystallized
material had less than 0.15%.
The coupled product 8b was then hydrolyzed to 9b with
1 N NaOH-MTBE at 50 °C. To the product-rich MTBE
layer, 3 and methanesulfonic acid were each added. The
transesterification was driven to completion due to crystal-
lization of 1b from MTBE as it formed. The desired 1b was
collected by filtration in 90% yield and 99.7% purity. The
(6) (a) Matteson, D. S.; Sadhu, K. M. J. Am. Chem. Soc. 1981, 5241. (b)
Matteson, D. S.; Majumdar, D. J. Am. Chem. Soc. 1980, 7590.
(7) Curphey, T. J. J. Org. Chem. 1979, 44, 5 and references therein.
Vol. 6, No. 6, 2002 / Organic Process Research & Development
•
815

1
pinanediol ester 5 was recovered in 98% yield from the
90% yield and 99.7% purity (HPLC). Mp 134-136 °C; H
MTBE filtrate.
NMR (400 MHz, CDCl3): δ (ppm) 0.8 (s, 3 H), 0.9 (d, 3
H), 1.0 (d, 3 H), 1.27 (s, 3 H), 1.3 (d, 1 H), 1.4 9s, 3 H),
1.7-2.0 (m, 5 H), 2.2-2.1 (m, 4 H), 2.3 (t, 1 H), 3.2 (t, 1
H), 3.5 (q, 1 H), 3.7 (t, 1 H), 4.2 (d, 1 H), 4.6 (q, 1 H), 7.2
(d, 1 H). 13C NMR (400 MHz, CDCl3): δ (ppm) 18.0, 19.8,
24.7, 26.9, 27.7, 27.9, 29.2, 32.3, 36.0, 38.7, 40.0, 45.1, 47.5,
51.6, 56.3, 78.2, 86.2, 116.0 (q), 156.7 (q), 167.4.
Conclusions
The new synthesis of 1b using the TFA-valine route is a
significant improvement over the previously disclosed syn-
thesis.1 The overall cost of goods was reduced to nearly one-
tenth of the initial cost by efficient recycling of pinanediol.
The overall route was made more environmentally friendly
by eliminating phenylboronic acid, dichloromethane, EDAC,
HOBt, diethyl ether, and hexane, which were key reagents
and solvents used in the original synthesis. The cycle time
was improved by elimination of unit operations such as
chromatography and lyophilization.
Valyl-pyrrolidine-(2R)-boronic Acid MSA Salt (1b).
Compound 8b (100 g, 0.225 mol) was dissolved in 1 L of
MTBE. Sodium hydroxide (1 N, 450 mL, 2 equiv) was
added, and the biphasic mixture was stirred vigorously and
heated to 50 °C. After 1 h, HPLC analysis indicated no 8b
remained. The organic layer was reserved, and the aqueous
layer was extracted with one 500 mL portion of MTBE. The
organic layers were combined, and 700 mL of MTBE was
distilled off at atmospheric pressure. N-Boc-pyrrolidine-2-
boronic acid (3) (48.4 g, 1 equiv) was dissolved in 600 mL
of MTBE and added to the previous solution. To this mixture
was added dropwise methanesulfonic acid (14.6 mL, 1
equiv). The resulting clear solution was stirred at ambient
temperature. After 15 min, a white crystalline precipitate
began to form. The reaction mixture was filtered after 4 h,
and 63 g of 1b was isolated in 90% yield and 99.7% HPLC
Experimental Section
The TFA-valine route to 1b and recycling of (+)-
pinanediol are described below. Compounds 2, 3, 5, 6, 8a,
and 9a were prepared according to the previously disclosed
procedures with some modification as described above.
Solvents and reagents were obtained from commercial
sources and used without further purification. Moisture
content was determined by coulometric titration on a
Mitsubishi CA-06 moisture meter on a weight/volume basis.
NMR data was collected on a Bruker 400 MHz spectrometer
at 300 MHz for proton and 75 MHz for carbon in CDCl3.
TFA-valyl-pyrrolidine-(2R)-boronic Acid Pinanediol
Ester (8b). TFA-valine (11) (65.1 g, 1 equiv) was dissolved
in 650 mL of ethyl acetate. The solution was cooled to 0
°C, and Vilsmeier reagent (39 g, 1.3 equiv) was added. The
cold reaction mixture was stirred for about 2 h. In another
flask, amine pinanediol ester hydrochloride 6a, (83 g, 0.29
mol) was combined with water (800 mL) and sodium
bicarbonate (100 g, 4 equiv), and the resulting mixture was
cooled to 0 °C. The acid chloride solution was then
transferred to the aqueous solution, and the resulting biphasic
mixture was stirred vigorously at 0 °C for 1.5 h. The organic
layer was separated and washed with 300 mL of 0.1 M
phosphoric acid, dried over sodium sulfate, and concentrated
to give 8b in 97% yield (125 g), in 98.7% purity (HPLC).
The solid product was dissolved in acetone (375 mL, 3 mL/
g) with warming to 45 °C, and water (375 mL, 3 mL/g) was
added to the resulting solution with stirring. The resulting
slurry was stirred at ambient temperature for 3 h, and the
product was collected by filtration to give 116 g of 8b in
1
purity. The material showed H, 13C NMR, and MS data
identical to those of previously reported material.1
Recycling of (+)-Pinanediol from the Resolution Step.
Compound 3 (254 g, 1.18 mol) was suspended in 2 L of
heptane and 1 L of water. Mother liquor from the resolution
step containing 336 g of mostly 6b and some 6a (total 1
equiv) were added, and the reaction mixture was stirred
vigorously at ambient temperature for 1 h. The heptane layer
was removed from the biphasic mixture, dried over sodium
sulfate, and concentrated to give 400 g (97%) of 5 as a clear
oil, which could be converted to 6a without further purifica-
tion.
Recycling of (+)-Pinanediol from the Final Step. After
the isolation of 1b, the MTBE mother liquor was washed
with 500 mL of water and evaporated to afford 5 (77 g, 98%
yield), which could be converted to 6a without further
purification.
Received for review August 30, 2002.
OP025587B
816
•
Vol. 6, No. 6, 2002 / Organic Process Research & Development