One-pot oxidation and rearrangement of propargylamines and in situ pyrazole synthesis

Article abstract of DOI:10.1021/ol501842b

Reported here are procedures for a one-pot oxidation and rearrangement of propargylamines to synthesize enaminones, with supporting mechanistic studies. Also reported are the extended one-pot syntheses of pyrazoles, including celecoxib and various heterocyclic compounds.

Full text of DOI:10.1021/ol501842b

Letter
pubs.acs.org/OrgLett
One-Pot Oxidation and Rearrangement of Propargylamines and in
Situ Pyrazole Synthesis
Jinshan Chen,* Roberta Properzi, Daniel P. Uccello, Jennifer A. Young, Russell G. Dushin,
and Jeremy T. Starr
Pfizer Worldwide R&D, Eastern Point Road, Groton, Connecticut 06340, United States
S
* Supporting Information
ABSTRACT: Reported here are procedures for a one-pot oxidation and
rearrangement of propargylamines to synthesize enaminones, with
supporting mechanistic studies. Also reported are the extended one-pot
syntheses of pyrazoles, including celecoxib and various heterocyclic
compounds.
xidation of propargylamines 1 and subsequent trans-
formation of the isolated propargylamine N-oxides 2 to
cycles in situ starting from the corresponding propargylamine
precursors.
O
enaminones 3 has been reported (Scheme 1).1 The literature
We sought a protocol to perform the oxidation step in a
protic solvent and then carry out the thermal rearrangement in
situ without the need for a solvent switch or isolation of the
intermediate N-oxide. This would circumvent the issue of
spontaneous decomposition of the N-oxide during isolation, in
addition to simplifying the overall transformation. We
anticipated the oxidation would require a protic solvent
because an acidic proton is required for the thermal
fragmentation of isoxazolinium intermediate 4. Thus, a number
of oxidizing agents and protic solvent systems were evaluated
with 5 as a model substrate in reactions run at 20 °C for 16 h
for the oxidation step and 85 °C for 5 h for the rearrangement.
This led to the observation of an encouraging 59% conversion
to the enaminone 6 when using m-CPBA in 95% ethanol and
5% water.6 Upon further optimization, the oxidation was found
to be complete within 30 min at room temperature7 and
rearrangement at 85 °C took only an additional 30 min in
absolute ethanol. With shorter reaction times, high molecular
weight side products were suppressed8 and the conversion of 5
to 6 was clean, with an isolated yield of approximately 60%
upon column chromatography purification (Scheme 2).
Peracetic acid was found to perform similarly to m-CPBA in
this process.9
Scheme 1. Enaminones from Propargylamines
protocols involve two discrete steps: initial oxidation, typically
carried out with m-CPBA in dichloromethane at 0 °C, and a
subsequent thermally induced formation of enaminones in a
protic solvent, such as methanol at refluxing temperature for 6
h. Isolation of the propargylamine N-oxide intermediate can be
complicated as a result of several transformations that lead to
spontaneous decomposition at ambient temperature, partic-
ularly on silica gel.2 This may have contributed to the lack of
literature reports using this two-step sequence for the synthesis
of enaminones beyond the initial report.3
A plausible mechanism of this enaminone formation involves
formation of isoxazolinium intermediate 4, generated by
addition of the N-oxide oxygen to the carbon−carbon triple
bond and subsequent proton transfer from a protic solvent,
such as methanol. The newly formed methoxide then
deprotonates the methylene of isoxazolinium 4, which triggers
fragmentation and isomerization to form enaminone 3.1
Herein we report an in situ oxidation and rearrangement
protocol to prepare enaminones in one pot from propargyl-
amines,4 along with experimental evidence that supports the
formation of enaminones via isoxazolinium intermediate 4. In
addition, we also describe an extended one-pot process to
synthesize pyrazoles, including celecoxib,5 and other hetero-
Scheme 2. In Situ Propargylamine Oxidation and
Rearrangement
Received: June 25, 2014
Published: July 28, 2014
© 2014 American Chemical Society
4146
dx.doi.org/10.1021/ol501842b | Org. Lett. 2014, 16, 4146−4149

Organic Letters
Letter
We then turned our attention to mechanistic understanding
of the transformation. Literature precedents of isoxazolinium
ions are rare. The most relevant examples were reported by
Chiacchio and co-workers,10a−c involving an isoxazolinium ring
formed by [3 + 2]-cycloaddition followed by quaternization of
nitrogen with methyl triflate. Compound 7, a 3,3-disubstituted
isoxazolinium triflate with an electron-withdrawing ester
functional group attached to the C4 position, underwent
base-induced thermal rearrangement to furnish enone 8 as the
major product in methanol (Scheme 3).10a,b The other two
Scheme 4. Isoxazolinium Ion as a Viable Precursor to
Enaminones
Scheme 3. Literature Isoxazolinium Chemistry
from propargylamines would be a useful extension of this
oxidation rearrangement protocol. To that effect, we developed
viable conditions to carry out a three-step sequence in one pot
to synthesize pyrazoles. For example, upon oxidation of 1 using
m-CPBA or peracetic acid, with the introduction of hydrazine
or various substituted hydrazines, either before or after thermal
rearrangement, we obtained pyrazoles in 36% to 91% isolated
or combined yields of the regioisomeric products (Table 1).
This protocol was applied to propargylamines with aryl
isoxazolinium salts Chiacchio studied, 9 and 10, both bear an
electron-withdrawing group at C3 and a second electron-
withdrawing group attached to C4.10c When 9 and 10 were
subjected to a base at elevated temperature, a mixture of enone
11 and enaminone 12 was obtained. Chiacchio et al.10c
rationalized the outcome via proton transfer at the methyl-H vs
the C3−H during the rearrangement, leading to 11 and 12
respectively (Scheme 3).
Table 1. Extended One-Pot Pyrazole Formation
Realizing the difference in the electronic nature of the
isolated literature isoxazolinium ions 7, 9, and 10, compared
with intermediate 4, we set out to prepare an isoxazolinium ion
with electronic and regiochemical properties more in line with
4 (R = Ph), via a synthetic pathway independent of
propargylamine oxidation and rearrangement. Such an
intermediate, if proven viable as a precursor to enaminones,
would provide experimental evidence to support the mecha-
nism outlined in Scheme 1.
Thus, N-hydroxypropargylamine 13 was prepared and
alkylated with methyl triflate at 20 °C under neutral conditions
to furnish propargylamine N-oxide 14A (Scheme 4). When
treated with Na2CO3 in ethanol at 85 °C, compound 14A
transformed into enaminone 15, as expected. On the other
hand, 13 was also converted into isoxazoline 16 via a cyclization
mediated by NaAuCl4.11 From 16, we synthesized and isolated
isoxazolinium 17 via methylation using methyltriflate in the
absence of base. Compound 17 was stable at neutral pH and
did not spontaneously rearrange. However, in the presence of
Na2CO3 in ethanol, 17 underwent a transformation to
enaminone 15 at room temperature.12 Furthermore, 14B was
also synthesized and isolated via alkylation of N,N-dimethyl
hydroxylamine with phenyl propargyl bromide. Compound
14B was then subjected to Na2CO3 in ethanol at 85 °C, with
the outcome being the expected enaminone 6. These
observations are consistent with the oxidation and rearrange-
ment mechanism via the isoxazolinium intermediate as outlined
in Scheme 1.13
a
Enaminones are versatile intermediates to access important
heterocycles such as pyrazoles, triazoles, pyrimidinones, and
pyrimidines. An in situ synthesis of these heteroaryls directly
Compound 18e was synthesized starting from compound 15.
b
c
Regioisomers were separated by column chromatography. Re-
gioisomers could not be separated by column chromatography.
4147
dx.doi.org/10.1021/ol501842b | Org. Lett. 2014, 16, 4146−4149

Organic Letters
Letter
substitution on the terminal acetylene C3 position (18a to
18e14 42 to 91% yield),15 as well as alkyl substitutions (18f to
18j 36% to 46% yield). Aryl and alkyl substitutions on the C1
position of propargylamines were also tolerated (18e, 48%
yield). In addition, hydrazine, alkyl hydrazines, and aryl
hydrazines appeared to participate in this transformation
without issue (R4 = H, Me, iPr, Ph). Although the isolated
yields are not consistently high, the method has the potential to
rapidly deliver a wide variety of functionalized heterocycles
from a single intermediate, something highly valued in the drug
discovery synthesis environment. For example, literature
syntheses of α-hydroxylated pyrazoles involve metalation of
protected pyrazoles followed by trapping the carbanion with a
carbonyl electrophile,16 with the limitations of carbanion
chemistry in addition to the requirement of a pyrazole nitrogen
protecting group for N-unsubstituted pyrazoles. The propargyl-
amine strategy has no such prerequisite conditions; thus,
compound 18h was prepared using the current strategy in 36%
isolated yield in one reaction vessel.
which provided 26 in an isolated yield of 83% (Scheme 6). This
approach was inspired by the zinc acetylide aldehyde addition
pioneered by Carreira and Frantz.22
Scheme 6. Zn(OTf)2 Catalyzed Propargylamine Synthesis
Our initial observation appeared to suggest that the oxidation
of 26 proceeded unusually slowly, with a substantial amount of
26 remaining upon treatment with m-CPBA for 0.5 h.
Extending the reaction time or increasing the reaction
temperature did not improve conversion. Lowering the
oxidation temperature to 0 °C, however, improved the
conversion substantially. This may be due to spontaneous
hydrolysis of the enaminone to the diketo intermediate in this
trifluoromethyl activated species, thus generating Me2NH
which competed for m-CPBA. At lower temperature, the
hydrolysis was sufficiently suppressed to minimize the
interruption of the intended oxidation. When hydrazine was
introduced for the formation of pyrazole 27, two major side
products were observed: the hydrated species (M + 18), which
was suppressed by introduction of the aryl hydrazine as an
HCl/dioxane suspension, and a partially reduced species (M +
2), presumably 28, formed via the enone pathway similar to the
formation of compounds 8 and 11 in Scheme 3. This side
product was suppressed when additional water (10% volume)
was introduced after the oxidation.
This strategy was expanded to one-pot syntheses of other
heterocyclic systems (Scheme 5). For example, isoxazole 19
Scheme 5. Additional One-Pot Processes
We proceeded with the synthesis of celecoxib 29 using
propargylamine 26 and 4-hydrazinylbenzenesulfonamide. The
desired product 29 was indeed formed in 37% isolated yield
after column purification, in line with the overall yield of all
previous syntheses except in the industrial process5b (Scheme
7). Importantly, the undesired pyrazole regioismer product
Scheme 7. One-Pot, Three-Step in Situ Celecoxib Formation
was synthesized from propargylamine 1 in 53% isolated yield by
substitution of hydroxylamine for hydrazine.17 Triazole 20 was
prepared in 33% yield in one pot by cycloaddition of
phenylazide and elimination of dibutylamine after the thermal
rearrangement.18 Pyrimidine 21 was similarly obtained in 36%
isolated yield by condensation with benzamidine.19 When α-
hydroxylated propargylamine 22 was used in the process and p-
TSA was introduced after the thermal rearrangement,
spirocyclic ether 23 was obtained in 44% yield.20
Given the success of the enaminones for the synthesis of
pyrazoles, we turned our attention toward developing a novel
synthesis of celecoxib5 using the propargylamine oxidation
rearrangement strategy. Trifluoromethylated propargylamine
26 has previously been synthesized via coupling of 24 with 1,1-
bis(dimethylamino)-2,2,2-trifluoroethane (25) using CuI or
ZnI2. In both cases, 1 equiv of the metal iodide was used and
the isolated yields of 26 were <60% after an extended reaction
time at elevated temperature.21 We developed a new and more
efficient protocol to synthesize compound 26 where coupling
of 24 and 25 was effected by 0.2 equiv of Zn(OTf)2, without
the use of any solvent or other additive, at 75 °C for only 0.5 h,
formed is 2%−3% of the desired regioisomer product 29,
comparable to the ratio observed in the second generation
manufacturing process.2
Finally, a one-pot, four-step synthesis of celecoxib 29 was
realized by incorporating the synthesis of propargylamine 26
into the one-pot, three-step process; thus 29 was prepared
directly from 24 in one pot with a 39% yield after
chromatography (Scheme 8).
Scheme 8. One-Pot, Four-Step Celecoxib Formation
4148
dx.doi.org/10.1021/ol501842b | Org. Lett. 2014, 16, 4146−4149

Organic Letters
Letter
phenyl-1-propyne and triethylamine was subjected to the oxidation
conditions employed in our process and the result was again oxidation
of the amine instead of 1-phenyl-1-propyne.
(10) (a) Chiacchio, U.; Rescifina, A.; Chiacchio, M. A.; Romeo, G.;
Romeo, R. J. Org. Chem. 2003, 68, 3718−3720. (b) Chiacchio, U.;
Liguori, A.; Rescifina, A.; Romeo, G.; Rossano, F.; Sindona, G.;
Uccella, N. Tetrahedron 1992, 48, 123−132. (c) Chiacchio, U.;
Casuscelli, F.; Liguori, A.; Rescifina, A.; Romeo, G.; Sindona, G.;
Uccella, N. Heterocycles 1993, 36, 585−600.
(11) Debleds, O.; Dal Zootio, C.; Vrancken, E.; Campagne, J. M.;
Retailleau, P. Adv. Synth. Catal. 2009, 351, 1991−1998.
(12) A small quantity of the enone byproduct (M + 1 = 209.07) was
observed by LCMS but not isolated. Compound 13 was directly
transformed into 18e in a one-pot three-step process in 36% yield.
Similarly, 14b was converted to 18a in a one-pot process in 41% yield.
(13) Propargylamine N-oxides have been reported to rearrange to o-
allenyl ethers; see: Szabo, A.; Galambos-Farago, A.; Mucsi, Z.; Timari,
G.; Vasvari-Debreczy, L.; Hermecz, I. Eur. J. Org. Chem. 2004, 687.
Yet, under our reaction conditions, the o-allenyl converts to an
unidentified product that is not the enamine.
In summary, we have established a one-pot protocol to
synthesize enaminones directly from propargylamines via
oxidation and rearrangement. We demonstrated experimentally
that isoxazolinium ions are viable precursors for the formation
of enaminones, which is consistent with the proposed oxidation
rearrangement mechanism. One-pot, three-step synthetic
protocols for functionalized pyrazoles and other heterocycles
were also presented. Although the yields were mostly moderate,
this method can provide rapid access to functionalized and
structurally diverse heterocycles. As a demonstration of this, we
developed a one-pot, four-step synthesis of celecoxib starting
from commercially available reagents. Considering the ease of
synthetic access of propargylamines, this method constitutes a
useful alternative to existing methods to access these hetero-
cycles.
ASSOCIATED CONTENT
* Supporting Information
■
S
(14) Compound 18e was synthesized starting from compound 15.
(15) Browne, D. L.; Taylor, J. B.; Plant, A.; Harrity, J. P. A. J. Org.
Chem. 2010, 75, 984−987.
(16) Katritzky, A. R.; Rewcastle, G. W.; Fan, W.-Q. J. Org. Chem.
1988, 53, 5685−5689.
Experimental details and full spectroscopic data for all
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
(17) Compound 19 was previously prepared in a two-step sequence
starting from acetophenone in 57% isolated yield. See: Nakamura, T.;
Sato, M.; Kakinuma, H.; Miyata, N.; Taniguchi, K.; Bando, K.; Koda,
A.; Kameo, K. J. Med. Chem. 2003, 46, 5416−5427.
■
*E-mail: michael.j.chen@pfizer.com.
Notes
(18) Structure of 20 was confirmed by 2D NMR.
(19) Zheng, X.; Song, B.; Xu, B. Eur. J. Org. Chem. 2010, 23, 4376−
4380.
The authors declare no competing financial interest.
(20) Saimoto, H.; Shinoda, M.; Matsubara, S.; Oshima, K.; Hiyama,
T.; Nozaki, H. Bull. Chem. Soc. Jpn. 1983, 56, 3088−3092.
(21) Xu, Y.; Dolbier, W. R., Jr. J. Org. Chem. 2000, 65, 2134−2137.
ACKNOWLEDGMENTS
■
We acknowledge Professor E. J. Corey (Department of
Chemistry and Chemical Biology, Harvard University) for
helpful chemistry discussions as well as Xidong Feng and Jason
Ramsay (Pfizer Global Research and Development, Groton,
CT) for analytical chemistry support.
(22) Frantz, D.; Fassler, R.; Carreira, E. M. J. Am. Chem. Soc.. 1999,
̈
121, 11245−11246.
(23) (a) A ratio of ∼2:98 was observed based on ELSD detection
using Waters Acquity HSS T3, 2.1 mm × 50 mm, C18, 1.7 micron
column with a column temperature of 60 °C and gradient method of
5% B for 0.1 min followed by a linear ramp to 95% B 0.1 to 2.6 min,
hold at 95% B 2.6 to 2.95 min, and finally return to initial conditions
2.95 to 3.0 min at a flow rate of 1.25 mL/min. (b) A ratio of ∼3:100
was observed based on 19F NMR. See ELSD chromatogram and 19F
NMR in the Supporting Information.
REFERENCES
■
(1) (a) Dushin, R. G.; Trybulski, E. J. WO 99/06351. (b) Szabo, A.;
Hermecz, I. J. Org. Chem. 2001, 66, 7219−7222.
(2) Bernier, D.; Wefelscheid, U. K.; Woodward, S. Org. Prep. Proced.
Int. 2009, 41, 173−210.
(3) No reports of enaminone synthesis using propargylamines have
emerged since the initial disclosures of ref 1.
(4) Uccello, D. P.; Dushin, R. G.; Starr, J. T.; Chen, J. One-pot
Oxidation and Rearrangement of Propargylamines, American Chemical
society 244th National Meeting, Philadelphia, PA, Abstract ORGN 361
(oral presentation), August 19−23, 2012.
(5) (a) McCormack, P. L. Drugs 2011, 11, 2457−2489. (b) Letendre,
L. J.; McGhee, W. D.; Snoddy, C.; Klemm, G.; Gaud, H. T. WO
2003099794. (c) Reddy, M.; Ramana, V.; Bell, S. C. WO2003024958.
(d) Reddy, M.; Ramana, V.; Bell, S. C. WO 2003024400. (e) Oh, L. M.
Tetrahedron Lett. 2006, 47, 7943−7946. (f) Gaulier, S. M.; McKay, R.;
Swain, N. A. Tetrahedron Lett. 2011, 52, 6000−6002. (g) Hu, J.; Chen,
S.; Sun, Y. l.; Rao, Y. Org. Lett. 2012, 14, 5030−5033.
(6) Oxone was among the oxidizing agents evaluated initially. Water
was included as a cosolvent to facilitate solubilization.
(7) In most cases, N-oxidation of propargylamines was completed
within minutes at 20 °C.
(8) Enaminones are known to self-condense when subjected to an
elevated temperature. Enaminone 6 undergoes a trimeric condensation
to form 1,3,5-trisubstituted benzene at 110 °C in high yield. See for
example: Al-Zaydi, K. M.; Nhari, L. M.; Borik, R. M.; Elnagdi, M. H.
Green Chem. Lett. Rev. 2010, 3, 93−99.
(9) Throughout our studies, no oxidation of the proparyl sp1 carbons
was observed under our conditions. In addition, a 1:1 mixture of 1-
4149
dx.doi.org/10.1021/ol501842b | Org. Lett. 2014, 16, 4146−4149