Relative reactivity of anti- and syn-oximino carbonates and carbamates of 2-pyridylacetic acid esters

Article abstract of DOI:10.1021/ol015737n

(Matrix presented) anti-Oximes of 2-pyridylacetic acid esters are rapidly transformed to pyridine-2-carbonitrile under a variety of conditions while syn-oximes bearing tert-butyl esters can be conveniently deprotected to the corresponding carboxylic acid with subsequent fragmentation to the nitrile.

Full text of DOI:10.1021/ol015737n

ORGANIC
LETTERS
2001
Vol. 3, No. 14
2137-2140
Relative Reactivity of anti- and
syn-Oximino Carbonates and
Carbamates of 2-Pyridylacetic Acid
Esters
Ha Young Kim, Douglas A. Lantrip, and Philip L. Fuchs*
Department of Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907
pfuchs@purdue.edu
Received February 20, 2001 (Revised Manuscript Received June 4, 2001)
ABSTRACT
anti-Oximes of 2-pyridylacetic acid esters are rapidly transformed to pyridine-2-carbonitrile under a variety of conditions while syn-oximes
bearing tert-butyl esters can be conveniently deprotected to the corresponding carboxylic acid with subsequent fragmentation to the nitrile.
In conjunction with our program to utilize the folate receptor
as a cell-specific mediator for the delivery of various
oncolytic agents,1 we required an intermediate capable of
releasing alcohols and amines at pH values in the range of
3-7. On the basis of standard mechanistic considerations,
we elected to investigate the chemistry of pyridinium
carboxylates in hopes of developing a pH-triggered hydroly-
sis which might selectively occur at the more acidic environ-
ment found in the endosomal compartments of cancer
cells.2
based release of drugs bearing alcohols and amines. Scheme
1 depicts our initial hypothesis for the syn-oxime 1s (as will
be shown later, the anti-oxime-carboxylic acid 1a could not
be isolated for parallel study). It was postulated that
Scheme 1
Both syn- and anti-oxime carbonates and carbamates
derived from R-oxocarboxylates 1s and 1a were initially
conceived to offer excellent opportunities for mechanism-
(1) Folate Mediated Drug Delivery. 4. For papers 1, 2, and 3 see: (a)
Wang, S.; Luo J.; Lantrip, D. A.; Waters, D. J.; Mathias, C. J.; Green, M.
A.; Fuchs, P. L.; Low, P. S. Bioconjugate Chem. 1997, 8, 673. (b) Luo,
J.; Smith, M. D.; Lantrip, D. A.; Wang, S.; Fuchs, P. L. J. Am. Chem.
Soc. 1997, 119, 10004. (c) Lee, J. W.; Fuchs, P. L. Org. Lett. 1999, 1,
179.
(2) (a) Tannock, I. F.; Rotin, D. Cancer Res. 1989, 49, 4373-4384. (b)
Jain, R. K. J. Controlled Release 1998, 53, 49-67.
10.1021/ol015737n CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/14/2001

zwitterion 1s would be present at pH values sufficient to
ionize the carboxylic acid but above the anticipated pH of
the protonated pyridine (∼3.6-3.83). Fragmentation of this
species is postulated to involve intermediate 3 formed via
intramolecular acylation of the carbonate (or carbamate)
moiety. Completion of the process would be driven by the
loss of 2 mol of carbon dioxide in concert with formation
of pyridine-2-carbonitrile 4. This fragmentation could be
inhibited at pH 5 where the carboxylic acid is protonated
(2s), or acid catalysis might enhance the intramolecular
acylation sequence (2s′) to increase the rate.
While the 2-pyridyl moiety has been attached to aldoxime,4
glyoxylate oxime,5 hydrazone,6 amides,7 and hydroxy-
iminoacetate,8 the differential reactivity of the anti- and
syn-oxime isomers bearing this functionality have not been
reported.
carboxylic acids 1a or 1s was observed on TLC. Mean-
while syn-isomer 9s with 5% of Pd[PPh3]4 was less reactive
in dichloromethane for 1 h at 25 °C but was converted to
nitrile 4 in an additional 2.5 h of stirring at the same
temperature.
In the latter reaction it appeared that the pyridyl imine
functionality might be operating as a bidentate ligand and
sequestering the palladium[0]. In line with this hypothesis,
a 1:1 mixture of anti/syn-4-pyridyl glyoxylates 10a/10s
resulted in rapid production of pyridine-4-carbonitrile 11,
supporting the postulate that the palladium catalyst was far
more available for effecting the deallylation-fragmentation
reaction (Scheme 3).
Scheme 3
When anti-carbonate allyl esters 8a and 9a (Scheme 2)
Scheme 2a
Since the above allyl ester deprotection strategy did not
provide access to the free carboxylic acids 1s or 1a, we next
examined acid-catalyzed cleavage of tert-butyl esters 13-
15.10 Each of the syn- and anti-allyl and tert-butyl esters
shown in Schemes 2 and 4 were readily prepared from the
parent pyridine acetate oximes via action of the appropriate
chloroformate reagent. Single isomers were isolated using
column chromatography. The structures of syn-oxime allyl
ester 7s and syn-oxime tert-butyl ester 12s were determined
by X-ray diffraction.11
As can be seen in Table 1, proton NMR chemical shift
a Reagents and conditions: (a) Et3N (2.0 equiv), DMAP (0.05),
allyl alcohol (1.5), DCC (1.0), CH2Cl2, reflux, 12 h; (b) acetic acid
(excess), NaNO2 (1.0), water, 0-25 °C, 2 h; (c) Et3N (2.0 equiv),
methyl chloroformate (1.05), CH2Cl2, 25 °C, 0.5 h; (d) (i) trans-
2-methylcyclohexanol (1.0), pyridine (1.33), triphosgene (0.33),
CH2Cl2, reflux, 1.5 h, (ii) Et3N (2.0), oxime starting material (1.0),
25 °C, 0.5 h.
Table 1
anti
syn
no. δH5 (J , Hz) δH2 (J , Hz) no. δH5 (J , Hz) δH2 (J , Hz)
7a
8a
8.58δ
d, 5.9 Hz
8.75δ
8.17δ
d, 8.1 Hz
7.87δ
7s
8.66δ
7.88δ
d, 8.1 Hz
8.14δ
d, 4.9 Hz
8.67δ
8s
d, 4.6 Hz
8.68δ
d, 3.4 Hz
8.57δ
d, 4.9 Hz
8.72δ
d, 4.7 Hz
8.72δ
d, 3.6 Hz
7.90δ
m + H3
7.97δ
m + H3
7.82δ
m + H3
7.83δ
d, 4.7 Hz
8.67δ
d, 4.9 Hz
8.60δ
d, 4.9 Hz
8.69δ
d, 4.7 Hz
8.69δ
d, 7.9 Hz
8.15δ
d, 7.9 Hz
8.00δ
m + H3
8.12δ
d, 8.1 Hz
8.13δ
9a
9s
and syn-carbonate 8s were treated with 5% of Pd[PPh3]4
and 1.5 equiv of phenylsilane9 at 25 °C for 0.5 h, pyridine-
2-carbonitrile 4 was immediately produced in >97% isolated
yield. No evidence of the intermediacy of the requisite
12a
13a
14a
15a
12s
13s
14s
15s
(3) CRC Handbook of Chemistry and Physics values for pyridine-2-
carboxaldehyde and the corresponding aldoxime.
(4) Jose, B.; Sulatha, M. S.; Pillai, P. M.; Prathapan, S. Synth. Commun.
2000, 30, 1509.
d, 4.7 Hz
8.73δ
m + H3
7.97δ
d, 4.1 Hz
8.70δ
d, 8.0 Hz
8.97δ
(5) Kolar, P.; Petric, A.; Tisler, M. J. Heterocycl. Chem. 1991, 28, 1715.
(6) Mathur, N. C.; Goyal, R. N.; Malik, W. U. Indian J. Chem., Sect.
A.: Inorg., Bio-inorg., Phys., Theor. Anal. Chem. 1990. 29, 765.
(7) Calderwood, D. J.; Davies, R. V.; Rafferty, P.; Twigger, H. L.;
Whelan, H. M. Tetrahedron Lett. 1997, 38, 1241.
(8) Kolar, P.; Petric, A.; Tisler, M. J. Heterocycl. Chem. 1993, 30,
1253.
d, 4.2 Hz
d, 8.1 Hz
d, 4.7 Hz
d, 7.7 Hz
and coupling information on the pyridine moiety do not give
a reliable indication of the anti or syn stereochemistry of
2138
Org. Lett., Vol. 3, No. 14, 2001

Scheme 4
the oximes and their derivatives. Once having secured the
geometry of 7s and 12s by X-ray,11 all other derivatives were
related to these standards. As a practical note, we observed
that for all the above anti/syn isomer pairs the syn isomer
was always less polar on silica (1:2 EA/hex), consistent
with the more basic nature of the anti-iminooxime relation-
ship.
intermediates for the synthesis of pyridyl oximino esters. As
shown in Scheme 5, both acids were rapidly transformed
Scheme 5
Reaction of anti-oximino carbonates 13a and 14a furnished
high yields of carbonitrile 4 and only small amounts of
carboxylic acids 16a and 17a, respectively. Oximino car-
bamate 15a underwent similar decarboxylative fragmentation
to nitrile 4 with only a trace of carboxylic acid 18a being
observed by TLC (Scheme 4).
In stark contrast, syn-carbonate 13s furnished 16s in
essentially quantitative yield with only a trace of nitrile 4
being detected. The more sterically demanding carbonate
14s provided acid 17s in 88% yield after 24 h and a trace
of nitrile 4, with 10% of starting 14s recovered. The
deprotection of oximino carbamate 15s proceeded more
slowly than that of the oximino carbonates 13s and 14s but
also delivered the syn-carboxylic acid 18s in very good yield
(80%).
within 5 min at 0 °C to pyridine-2-carbonitrile 4 using
triphosgene. We speculate that intermediate 20a, formed from
anti-oxime 19a, suffers facile deprotonation to 21a followed
by intramolecular acylation of the oximino oxygen to afford
dioxazolidindione 3, a species unknown in the chemical
literature.12 Dioxazolidindione 3 is also the likely intermedi-
ate en route from syn-oxime 19s to nitrile 4. Attempts to
intercept intermediates 20a, 20s, or 3 by rapid quenching
with excess methanol failed, and nitrile 4 was the only
material detected (Scheme 5).
It is interesting to note that NMR samples of 16s, 17s,
and 18s in DMSO-d6 were quantitatively transformed to a
1:1 mixture of nitrile 4 and the corresponding alcohol or
amine over the course of 18 h at 25 °C, demonstrating the
promoting effect of a polar solvent.
In the course of our investigations, we also examined both
syn- and anti-oximino acids 19s and 19a as potential
Reaction of oximino acid 19s with methyl chloroformate
and 2.0 equiv of triethylamine (Scheme 5) results in
elimination of the alcohol moiety of the carbonate with
simultaneous and quantitative production of nitrile 4, pro-
viding credence to the mechanism proposed in Scheme
5.
(9) (a) Thieriet, N.; Alsina, J.; Giralt, E.; Guibe, F.; Albericio, F.
Tetrahedron Lett. 1997, 38, 7275-7278. (b) Dessolin, M.; Guillerez,
M.-G.; Thieriet, N.; Guibe, F.; Loffet, A. Tetrahedron Lett. 1995, 36,
5741-5744.
(10) The oxime esters 12s and 12a were prepared from 5 in parallel
fashion to the synthesis of the allyl series (see Supporting Information).
(11) X-ray data from these two compounds have been forwarded to the
Carbridge Crystalographic Database.
We monitored both the consumption of starting material
(17s and 18s) and the simultaneous production of nitrile 4
(12) Fragmentation of oxime derivatives to benzonitriles has been recently
reported: Coskun, N.; Arikan, N. Tetrahedron 1999, 55, 11943.
Org. Lett., Vol. 3, No. 14, 2001
2139

(pH ) 7), it was observed that a Very slight increase (∼1.2)
in the rate of fragmentation to 4 occurred as the reaction
solution became more acidic (Figure 1). The same experi-
ment conducted on oximino carbamate 18s was substantially
slower, requiring heating at 40 °C to proceed at a reasonable
rate but was also found to be essentially independent of pH
over the range examined (Figure 1). Thus, we can conclude
that the mechanism is in general accord with that shown in
Scheme 1, with the caveat that no significant acceleration is
provided by the acid catalysis at the lower pH conditions.
The ability to hydrolyze carbamate 18s under these mild
conditions attests to the likely intervention of intermediates
1s′ and 3, as direct attack of water or hydroxide ion on the
carbamoyl carbonyl would be expected to produce 19s very
slowly.
In conclusion, the reactivity of functionally differentiated
2-pyridylacetic acid esters bearing anti- and syn-oximino
derivatives have been investigated under palladium-catalyzed,
neutral, acidic, and basic conditions. anti-Oximino sub-
strates were rapidly transformed to pyridine-2-carbonitrile
4, while the syn-oximino carbonate and carbamate of the
tert-butyl ester provided the corresponding acid as a major
product using TFA to cleave the ester. Applications of
variants the syn-pyridyl oximes as prodrugs to deliver
oncolytic agents via the folate receptor are under further
evaluation.
Acknowledgment. We acknowledge Endocyte, Inc. for
financial support of this project. Dr. Iontcho Vlahov is
thanked for insightful discussions.
Supporting Information Available: Experimental pro-
cedures and proton and carbon NMR spectra. This material
is available free of charge via the Internet at http://pubs.acs.org.
Figure 1. pH and temperature effect on hydrolysis rate. Half-lives
are enclosed in brackets.
OL015737N
at three different pH values with HPLC.13 When oximino
carbonate 17s was treated at 25 °C in a solution of formate
buffer (pH ) 3), acetate buffer (pH ) 5), or phosphate buffer
(13) mm × 4.6 mm Econosphere C18, 50/50 acetonitrile/water, 1.0 mL/
min.
2140
Org. Lett., Vol. 3, No. 14, 2001