Convergent Synthesis of 2-Oxazolone-4-carboxylates Esters by Reaction of Aldehydes with Ambivalent N-Cbz-α-Tosylglycinate Ester

Article abstract of DOI:10.1021/acs.orglett.0c01703

N-Cbz-α-tosylglycinate ester was combined with aldehydes in a redox-neutral sequence leading to 2-oxazolone-4-carboxylates with high functional group tolerance. While the scope of the method was delineated to primary and secondary aliphatic aldehydes as well as aromatics, no racemization occurred with chiral aldehydes such as Garner's. Hitherto unknown, this process relies on the ambivalent role of N-Cbz-α-tosylglycinate ester acting as a pronucleophile.

Full text of DOI:10.1021/acs.orglett.0c01703

pubs.acs.org/OrgLett
Letter
Convergent Synthesis of 2‑Oxazolone-4-carboxylates Esters by
Reaction of Aldehydes with Ambivalent N‑Cbz-α-Tosylglycinate
Ester
̈
Masahiro Abe, Baptiste Picard, and Michael De Paolis*
Cite This: https://dx.doi.org/10.1021/acs.orglett.0c01703
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: N-Cbz-α-tosylglycinate ester was combined with aldehydes in a redox-neutral sequence leading to 2-oxazolone-4-
carboxylates with high functional group tolerance. While the scope of the method was delineated to primary and secondary aliphatic
aldehydes as well as aromatics, no racemization occurred with chiral aldehydes such as Garner’s. Hitherto unknown, this process
relies on the ambivalent role of N-Cbz-α-tosylglycinate ester acting as a pronucleophile.
ue to the combination of nitrogen and oxygen in a five-
membered aromatic ring, 2-(3H)oxazolones (2-oxo-2,3-
dihydrooxazoles) are pertinent molecules in the fields of
medicinal chemistry and agrochemistry,1 as illustrated with I
and II (Scheme 1A) with respectively antibacterial1a and
First, α-ketoesters are oxidized into α-nosyloxy or α-bromo-
β-ketoesters, and then the nitrogen nucleus is introduced with
methyl carbamate by nucleophilic addition in harsh conditions,
upon activation with p-toluenesulfonic acid or silver triflate, in
refluxing toluene.3 Within this strategy, alternative routes were
described by combining either α-amino-β-ketoesters with
triphosgene1b or α-diazo-β-ketoesters with methyl carbamate.1c
Besides the limited functional group tolerance displayed by
these methods, the assemblage and/or oxidation of the acyclic
compounds remain impediments to broader molecular
diversity.
D
Scheme 1
At the inception of this study was an attempt to promote the
Mannich coupling between dihydrocinamaldehyde 1a and N-
Cbz-α-tosylglycinate ester 2 which inadvertently led to 2-
(3H)oxazolone 3a in 25% yield (eq 1).
herbicidal activities.1b,2 As planar and aromatic derivatives of
amino acids such as serine and threonine, electron-poor 2-
(3H)oxazolone-4-carboxylate esters have much promise in the
aforementioned fields (Scheme 1B). However, the known
strategy toward 5-substituted-2-(3H)oxazolone-4-carboxylates
requires two limiting steps.
Received: May 19, 2020
https://dx.doi.org/10.1021/acs.orglett.0c01703
© XXXX American Chemical Society
Org. Lett. XXXX, XXX, XXX−XXX
A

Organic Letters
pubs.acs.org/OrgLett
Letter
Scheme 2. Scope of the Method
Simply promoted by nBu4NF, the redox-neutral and
convergent domino sequence gave 5-alkyl and 5-aryl-2-
(3H)oxazolone-4-carboxylates from 2 and functionalized,
aliphatic and aromatic, aldehydes (Scheme 1C). The role of
2 in this sequence was puzzling, reacting usually as a
proelectrophile.4,5 Interestingly, this study illustrates the
ambivalent reactivity of 2 as a pronucleophile.
From the initial experiment (eq 1), we set out to diminish
the amount of 4,6 and to that end, the temperature of reaction
was decreased (Supporting Information (SI) contains the
detailed optimization). Pleasingly, operating at −60 °C gave 3a
in 65% yield by reacting 1a (1.5 equiv) with 2 (1 equiv) in the
presence of nBu4NF (2.5 equiv). This result was improved by
conducting the experiment with molecular sieves (4 Å) and
nBu4NF (3 equiv), affording 3a in 74% yield (Scheme 2).
Switching to a less polar solvent (toluene) or modulating the
ratio 1a/2 negatively or only marginally affected the efficiency
of the sequence. As for the promotor, instead of nBu4NF, a
stronger base such as LiHMDS caused the decomposition of
the reagents, whereas tetrabutylammonium difluorotriphenyl-
silicate (TBAT) left the starting materials unchanged.7
coupling compared to dihydrocinnamaldehyde (74% vs 54%
yield). Further, oxazolone 3d featuring an acid-sensitive acetal
functional group was prepared in 61% yield from the
monoprotected glutaraldehyde derivative. Since no oxidant is
required to form the heterocycle, electron-rich functional
groups are compatible as demonstrated with the preparation of
3e (46%) from (S)-citronellal, containing a prenyl appendage.
While 4-pentynal was found to be less compatible, the
corresponding oxazolone 3f was isolated in 21% yield;8
electron-rich alkynylaryls such as 5-(p-tolyl)pent-4-ynal and
the alkynylindole were efficiently combined with 2 into 3g
(82%) and 3h (81%). In spite of the higher steric hindrance of
secondary aldehydes, they still reacted with 2 as demonstrated
with isopropyl and cyclopropyl substituted 2-(3H)oxazolones
3i (47%) and 3j (30%). The more functionalized N-Boc-4-
piperidinecarboxaldehyde was converted into 3h in 44% yield,
an alkaloid of pharmaceutical pertinence.1c A limit was reached
with tertiary aldehyde, as pivalaldehyde was found to be
unreactive and, instead of the heterocycle, the sole production
of 4 was noted. On the other hand, aromatic aldehydes reacted
well with 2 as observed with benzaldehyde and 4-
cyanobenzaldehyde giving 3i (42%) and 3j (58%).9 Interest-
ingly, since the motif is easily hydrolyzed in acidic conditions,
α-carboxaldehyde-α′-methoxy-γ-pyrone10 was efficiently com-
bined with 2 into 3n (62%), a heterocycle sharing structure
similarity with phenoxan, a natural product with anti-HIV and
With a protocol in hand, we examined the scope of the
method starting with primary aldehydes such as nonaldehyde
and phenylacetaldehyde which were converted into 3b (64%)
and 3c (54%). For the latter, it is noteworthy that the acidity of
the aldehyde only moderately impacted the efficiency of the
B
https://dx.doi.org/10.1021/acs.orglett.0c01703
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
fungicide properties.11 Moreover, a basic aromatic ring such as
pyridine was smoothly combined with 2 into 3o (65%).
However, hindered and electronically rich pyridines were less
reactive and 3p was assembled in 13% yield from 2 and the
corresponding aldehyde.8
Scheme 4. Postulated Mechanism
Eventually, α-chiral aldehydes were examined to assess the
risk of racemization induced by the conditions of reaction. To
that end, Garner’s aldehyde was selected due to its tendency to
racemize in the presence of base. Pleasingly, 2-(3H)oxazolone
3q was isolated without racemization (er = 99:1) in 22% yield
after exposure of Garner’s aldehyde to 2 in the standard
conditions.8 Although low, the unoptimized yield remains
synthetically useful and reflects the challenge of forming C−C
bonds from sterically hindered reagents. A less hindered chiral
aldehyde, deriving from the Roche’s ester, was more easily
combined with 2 since 2-(3H)oxazolone 3r was isolated in
37% yield. Incidentally, the bulky tert-butyldiphenylsilyl
protecting group (TBDPS) was not removed in these
conditions thus proving its compatibility with the method.
To gain insight into the mechanism, 1a was exposed to
nBu4NF in the absence of 2 (Scheme 3, eq 2), which led to
when the Cbz group of 2 was replaced by the more hindered
Boc group gives credit to this assumption.13 As an illustration
of the competition between the paths A and B, a small amount
of 4 was observed when primary aldehydes were reacted while
the production of 4 was significant with secondary aldehydes
and predominate when a tertiary aldehyde was engaged.
Simpler and advanced synthetic processes are urgently and
continuously sought, and in this context, a convergent
synthesis of various 5-alkyl and 5-aryl-2-(3H)oxazolone-4-
carboxylates was developed. Within the competition between
C- or N-deprotonation of 2 with nBu4NF, the unprecedented
sequence begins with the aldolization reaction of 2′ which,
despite being sterically hindered, reacted with a good variety of
aldehydes, the main limitation of the method being the steric
hindrance of the electrophile. Since no oxidation step or acidic
conditions are needed, aldehydes with electron-rich, acid-
sensitive, or pyridinyl appendages were successfully converted
into 2-(3H)oxazolones from 2, which can be prepared in large
scale and in one step. Furthermore, because 2-(3H)oxazolones
are known intermediates en route to oxazoles14 or
oxazolidinones,3a this method may have a bearing on the
synthesis of these motifs.
Scheme 3. Mechanistic Perspective
enal 5 resulting from the self-aldolization/crotonisation
process. Since the α-hydroxyketone 6 was not observed, an
umpolung mechanism in which 1a would act as a nucleophile
was thus ruled out. Moreover, exposure of 2 to these
conditions (eq 3) led to 4 (44% yield, 55% conversion)
demonstrating the ambivalent reactivity of 2. To examine the
possibility of a radical mechanism, we performed the coupling
of 1a with 2 (eq 4) in the presence of nBu4NF (3 equiv) and of
2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO, 3 equiv), a
scavenger of radical species. Since the outcome was not
significantly impacted (65% vs 56% yield in 3a), we surmised
that an anionic mechanism was likely.
Accordingly, 2 would act as a nucleophile after carbon
deprotonation into 2′, a process in competition with the usual
nitrogen deprotonation of 2 leading to imine 7 (Scheme 4). As
observed in the presence of tertiary aldehyde, the coupling
between 2′ and 7 becomes the major pathway (path A) leading
to 4 via the Mannich adduct III. In the cases of primary and
secondary aldehydes (path B), an aldolization occurs with 2′
forming the alcoholate IV, which after p-tolylsulfinate
elimination and intramolecular addition of the alcoholate to
the carbonyl of the Cbz group would give 2-(5H)oxazolone V.
Subsequent isomerization of V would follow, ultimately
affording 2-(3H)oxazolone 3. The success of the whole
sequence probably relies on the addition of the alcoholate to
the carbamate shifting the series of equilibrium to V, this step
being facilitated by the noncoordinating nature of the
quaternary ammonium.12 The fact that the process failed
ASSOCIATED CONTENT
* Supporting Information
■
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.0c01703.
General procedure and characterizations of each
compounds, as well as optimization of the procedure,
1
variations of the reagents and copies of H, 13C NMR
spectra and HPLC, GC traces (PDF)
FAIR data, including the primary NMR FID files, for
compounds 3a−r and 4b (ZIP)
AUTHOR INFORMATION
Corresponding Author
■
̈
Michael De Paolis − Normandie Univ, 76000 Rouen, France;
orcid.org/0000-0001-8139-3544;
Email: Michael.depaolis@univ-rouen.fr
Authors
Masahiro Abe − Normandie Univ, 76000 Rouen, France
Baptiste Picard − Normandie Univ, 76000 Rouen, France
Complete contact information is available at:
C
https://dx.doi.org/10.1021/acs.orglett.0c01703
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
the reactivity of anionic species such as 2′ and alcoholate IV is also
enhanced by noncoordinating organic cations such as nBu4N+.
(13) Other pronucleophiles with modulation of substituents were
tested without success; see SI for details.
(14) Vintonyak, V. V.; Kunze, B.; Sasse, F.; Maier, M. E. Chem. - Eur.
J. 2008, 14, 11132−11140.
https://pubs.acs.org/10.1021/acs.orglett.0c01703
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are very grateful to Drs. Julien Legros and Jacques
Maddaluno (Normandie University) for helpful discussions
and to the Fondation des Substances Naturelles for a
fellowship to M.A. We acknowledge the financial support
from ISCE-CHEM (Interreg IV, European Program), Interreg
FCE LabFact (European Program and cofunded by ERDF)
and Labex SynOrg (ANR-11-LABX-0029).
REFERENCES
■
(1) (a) Surivet, J.-P.; Panchaud, P.; Specklin, J.-L.; Diethelm, S.;
Blumstein, A.-C.; Gauvin, J.-C.; Jacob, L.; Masse, F.; Mathieu, G.;
Mirre, A.; et al. J. Med. Chem. 2020, 63, 66−87. (b) Smalley, T. L.;
Boggs, S.; Caravella, J. A.; Chen, L.; Creech, K. L.; Deaton, D. N.;
Kaldor, I.; Parks, D. J. Bioorg. Med. Chem. Lett. 2015, 25, 280−284.
(c) Yamashita, M.; Lee, S.-H.; Koch, G.; Zimmermann, J.; Clapham,
B.; Janda, K. D. Tetrahedron Lett. 2005, 46, 5495−5498. (d) Nam, N.-
H.; Kim, Y.; You, Y.-J.; Hong, D.-H.; Kim, H.-M.; Ahn, B.-Z. Bioorg.
Med. Chem. Lett. 2001, 11, 3073−3076. (e) Kudo, N.; Taniguchi, M.;
Furuta, S.; Sato, K.; Endo, T.; Honma, T. J. J. Agric. Food Chem. 1998,
46, 5305−5312.
(2) Synthesis of electron-rich 2-(3H)oxazolones: (a) Verlee, A.;
Heugebaert, T.; van der Meer, T.; Kerchev, P.; Van Hecke, K.; Van
Breusegem, F.; Stevens, C. V. ACS Catal. 2019, 9, 7862−7869.
(b) Habert, L.; Sallio, R.; Durandetti, M.; Gosmini, C.; Gillaizeau, I.
Eur. J. Org. Chem. 2019, 2019, 5175−5179. (c) Lu, Z.; Cui, W.; Xia,
S.; Bai, Y.; Luo, F.; Zhu, G. J. Org. Chem. 2012, 77, 9871−9877.
(d) Istrate, F. M.; Buzas, A. K.; Jurberg, I. D.; Odabachian, Y.; Gagosz,
́
F. Org. Lett. 2008, 10, 925−928. (e) Hashmi, A. S.; Salathe, R.; Frey,
W. Synlett 2007, 2007, 1763−1766.
(3) (a) Okonya, J. F.; Hoffman, R. V.; Johnson, M. C. J. Org. Chem.
2002, 67 (4), 1102−1108. (b) Hoffman, R. V.; Johnson, M. C.;
Okonya, J. F. Tetrahedron Lett. 1998, 39, 1283−1286. Oxidative
approach from N-carboxy α-dehydroamino acid anhydrides: (c) Yo-
nezawa, Y.; Shin, S.-G.; Ohtsu, A.; Yoshimura, J. Chem. Lett. 1982, 11,
1171−1172.
(4) Engberts, J. B. F. N.; Strating, J. Recl. Trav. Chim. Pays-Bas 1965,
84, 942−950.
(5) Review on α-amido sulfones: Petrini, M. Chem. Rev. 2005, 105,
3949−3976.
̈
(6) Schumann, S.; Zeitler, K.; Jager, M.; Polborn, K.; Steglich, W.
Tetrahedron 2000, 56, 4187−4195.
(7) KHMDS, DMAP, DABCO, and NaH were also tested as a
promoter with various outcomes. See SI for details.
(8) Unoptimized yield. The conversion of 2 was total, and
compound 4 was the major product of the reaction.
(9) Cinnamaldehyde, trans-2-bromo-3-(2-nitrophenyl)acryl alde-
hyde, and 2-nitrobenzaldehyde were found unreactive, as well as
PhCOCF3.
(10) De Paolis, M.; Rosso, H.; Henrot, M.; Prandi, C.; d’Herouville,
F.; Maddaluno, J. Chem. - Eur. J. 2010, 16, 11229−11232.
(11) (a) Ishibashi, Y.; Ohba, S.; Nishiyama, S.; Yamamura, S.
Tetrahedron Lett. 1996, 37, 2997−3000. (b) Garey, D.; Ramirez, M.;
Gonzales, S.; Wertsching, A.; Tith, S.; Keefe, K.; Pena, M. R. J. Org.
̃
Chem. 1996, 61, 4853−4856. (c) Kunze, B.; Jansen, R.; Pridzun, L.;
̈
Jurkiewicz, E.; Hunsmann, G.; Hofle, G.; Reichenbach, H. J. J.
Antibiot. 1992, 45, 1549−1552. (d) Modern Crop Protection
̈
Compounds, 2nd ed.; Kramer, W., Schirmer, U., Jeschke, P.,
Witschel, M., Eds.; Wiley-VCH Verlag GmbH & Co: 2012.
(12) Why nBu4NF gave the best results in the series of bases was not
investigated, but combined effects may explain the superiority of the
promotor. Hence, while the mild basicity of the reagent is important,
D
https://dx.doi.org/10.1021/acs.orglett.0c01703
Org. Lett. XXXX, XXX, XXX−XXX