Synthesis of Nitrogen-Containing Heterocycles through Catalytic Dehydrative Cyclization Reactions

Article abstract of DOI:10.1021/acs.orglett.0c03868

Re2O7 in hexafluoroisopropyl alcohol provides access to cationic intermediates from alcohols through the intermediacy of perrhenate esters. This manuscript describes the application of the system to the formation of a number of weakly basic heterocyclic systems through dehydration reactions and intramolecular nucleophilic addition. The influence of the substrate structure on the reaction rates and stereocontrol is discussed with respect to intermediate ion pairs.

Full text of DOI:10.1021/acs.orglett.0c03868

pubs.acs.org/OrgLett
Letter
Synthesis of Nitrogen-Containing Heterocycles through Catalytic
Dehydrative Cyclization Reactions
Freddy O. Rodriguez del Rey and Paul E. Floreancig*
Cite This: https://dx.doi.org/10.1021/acs.orglett.0c03868
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Re2O7 in hexafluoroisopropyl alcohol provides access to
cationic intermediates from alcohols through the intermediacy of perrhenate
esters. This manuscript describes the application of the system to the
formation of a number of weakly basic heterocyclic systems through
dehydration reactions and intramolecular nucleophilic addition. The influence
of the substrate structure on the reaction rates and stereocontrol is discussed
with respect to intermediate ion pairs.
eterocycles that contain a carbon−nitrogen double bond
mediated allylic cation formation to the synthesis of oxazolines
and related structures (Scheme 1). Our previous studies
H
are commonly employed as bioisosteres for carbonyl
groups in medicinal chemistry.1 Consequently, these groups
are present in a high percentage of the top-selling
pharmaceutical agents.2 These groups are often aromatic, but
the benefits of increasing the percentage of sp3-hybridized
carbons in drug discovery3 justify efforts to generate
nonaromatic heterocycles with carbon−nitrogen double
bonds. Biologically active natural products with oxazoline
and thiazoline subunits provide validation for this assertion.4
Additionally, oxazolines are also common subunits in ligands
for asymmetric catalysis5 and serve as aromatization substrates
through mild oxidative protocols.6
Scheme 1. Proposed Heterocycle Synthesis and Application
to the Beckmann Rearrangement
showed13 the importance of maintaining an acidic environment
for these processes, presumably due to the requirement of the
protonated perrhenate ester as an ionization intermediate.14
This concern was mitigated by our observation that oxime 4
undergoes a smooth Re2O7-catalyzed Beckmann rearrange-
ment to form amide 5 in hexafluoroisopropyl alcohol (HFIP).
This transformation, while efficient and proceeding through
mild conditions, was not explored further due to similarities to
previously reported studies.15 This success, however, showed
that basic groups do not necessarily inhibit Re2O7-mediated
transformations. We postulate that the exceptional capabilities
of HFIP as a hydrogen bond donor mitigate basicity in these
reactions.16
A common method for accessing oxazolines employs the
dehydration of β-hydroxy amides whereby the hydroxy group
is converted to a leaving group with a stoichiometric
dehydrating agent, followed by the displacement by the
carbonyl group.7 Catalytic protocols for oxazoline synthesis
through mechanistically distinct pathways have also been
reported.8 Our objective in this study is to develop a catalytic
dehydrative protocol9 to generate oxazolines and related
structures from simple substrates whereby the product retains
an alkene as a potential precursor to other functional groups.10
We hypothesized that heterocycles could be prepared
through the Re2O7-catalyzed11 dehydrative cyclization protocol
that we developed for tetrahydropyran synthesis, where allylic
alcohols serve as allylic cation precursors.12 We report that this
strategy is effective with amide, thioamide, and oxime
nucleophiles, leading to the formation of oxazolines, thiazo-
lines, oxazines, thiazines, and isoxazolines. The process can be
stereoselective, with reaction rates and stereochemical out-
comes implicating the importance of ion-paired intermediates.
The potential for catalytic cycle inhibition due to the
generation of a basic product was a concern in applying Re2O7-
Received: November 22, 2020
https://dx.doi.org/10.1021/acs.orglett.0c03868
© XXXX American Chemical Society
Org. Lett. XXXX, XXX, XXX−XXX
A

Organic Letters
pubs.acs.org/OrgLett
Letter
a
We initially studied oxazoline formation through the
conversion of 6 to 7 (Scheme 2). Exposing 6 to Re2O7 (5
Table 1. Scope Studies
Scheme 2. Demonstration of Concept
mol %) in HFIP indeed resulted in the formation of 7 in a 78%
yield within 5 h at 60 °C. We reasoned that HOReO3 is the
actual catalyst for the process. The conversion of 6 to 7 in the
presence of 2 mol % HOReO3 proceeded under the same
conditions in a nearly identical yield despite the lower rhenium
loading. This is significant because HOReO3 is substantially
less expensive than Re2O7 and easier to handle. The unique
merits of the oxorhenium catalysts are evident from the
attempt to conduct the reaction with 10 mol % p-TsOH as the
catalyst, which led to negligible product formation and showed
that this process does not proceed through simple Brønsted
acid catalysis. Solvent changes resulted in a lower yield with
significant byproduct generation (see the Supporting Informa-
tion for details), demonstrating that the capacity to stabilize
cationic intermediates, sequester water, and mitigate product
basicity16 make HFIP an ideal solvent for these trans-
formations. The low conversion of hydroxy amide 8 to 9
even under forcing conditions disqualified the mechanistic
alternative of a nucleophilic attack by the hydroxy group to an
activated carbonyl group.
Several substrates were exposed to the reaction conditions to
determine the scope of the process (Table 1). Further
experimentation showed that the reactions were complete
within 4 h at 45 °C, and the entries in the table show the
results under these conditions unless otherwise noted. The
substitution pattern of the alkene has a significant impact of
the reaction rate. 1,1-Disubstituted alkene substrates react at
approximately the same rate as terminal alkene substrates
(entry 2), while (E)-1,2-disubstituted alkene substrates react
much faster than terminal alkene substrates (entry 4); this
correlates with intermediate carbocation stabilities. Conduct-
ing the reaction with a mixture of alkene geometrical isomers
resulted in the formation of the (E)-alkene as the exclusive
product (entry 3) due to the low rotational barrier in the allylic
cation intermediate.17 Stopping this reaction at a partial
conversion showed an enrichment of the (Z)-substrate,
demonstrating that (E)-allylic alcohols ionize faster than (Z)-
allylic alcohols due to the steric greater allylic repulsion in the
intermediate cation.18 The transposed allylic alcohol isomer is
an effective cation precursor, as shown by the cyclization of 16
in entry 5. Benzylic alcohols can be used to prepare aryl-
substituted oxazolines (entry 6). Benzamide derivatives are not
essential, as shown by the successful cyclization of 19 (entry
7). Thioamides can be used to prepare sulfur-containing
heterocycles (entries 8 and 10). Increasing the tether length
allows for oxazine and thiazine formation, although this
a
Reactions proceeded at 45 °C for 4 h with 5 mol % Re2O7 in HFIP.
b
See the Supporting Information for details. Schemes for substrate
syntheses can be found in the Supporting Information. Isolated yields
of pure products. Reaction conducted at rt for 2 h. Reaction
conducted at 80 °C for 48−72 h.
c
d
e
reaction is slower than the formation of oxazolines and
thiazolines (entries 9 and 10).
We prepared allylic alcohol 27 as a mixture of diastereomers
to address the potential for stereocontrol. Exposing 27 to
Re2O7 (Scheme 3A) provided a 73% yield of oxazoline 28 as a
single stereoisomer with prolonged heating. We noticed that
one stereoisomer of the substrate was consumed significantly
faster than the other. Thus, we separated the stereoisomers and
subjected them to identical reaction conditions. Exposing 27-
syn to Re2O7 for 1 h at 60 °C yielded 28 in a 81% yield, while
27-anti provided only a 40% yield of 28. The latter reaction
produced primary allylic alcohol 29 faster than the reaction of
28, indicating that transposition is faster than cyclization
(Supporting Information). The buildup of 29 shows that its
cyclization is slow relative to that of 27-syn. This was verified
by exposing 29 to Re2O7 and observing the slow formation of
28. The relative rates of product formation from the isomers
are shown in Scheme 3B. Running the reactions to complete
conversion showed that 28 was formed in a 91% yield from 27-
B
https://dx.doi.org/10.1021/acs.orglett.0c03868
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
a
relationship between the substituents on the oxazoline is
generated through a backside attack. anti-Perrhenate ester 32
initially forms the tight ion pair 33, whereby a direct backside
attack would form the thermodynamically disfavored cis-isomer
34. The steric interaction sufficiently inhibits cyclization,
making the transposition to 29 favorable. The accumulation of
29 in preference to product formation is consistent with the
cyclization of 27-anti requiring an initial transposition to 29.
The low reactivity of 29 was somewhat surprising in
consideration of the similar reactivities of isomers 6 and 16
(Table 1). We attribute this to the energetic benefit derived
Scheme 3. Rate Differences between Isomeric Substrates
−
from the interaction of ReO4 with both termini of the allylic
cation.19 Steric destabilization arises in the conversion of 35 to
31. Thus, the successful generation of 31 from 30 results from
the fact that the steric interactions in 31 are also present in 30.
These results indicate that the perrhenate ester formation is
rapid and not rate-determining in these reactions. Ionization is
a slower process. Therefore, 30 reacts faster than 35 to form 31
because it forms readily but exhibits the steric destabilization of
the cation prior to ionization. Stereoselectivity was also
observed in the slower cyclizations that form oxazines
(Supporting Information).
The ability to control the stereochemical outcome of the
cyclization provides the opportunity to use chiral amides for
the generation of stereochemically rich products. An example
of this is shown in Scheme 5. Substrate 36 undergoes
Scheme 5. Preparation of a Product with an Exocyclic
Stereocenter
a
Panel A shows the reaction schemes, and panel B shows the roduct
formation over 1 h. Each time point is the average of three trials.
cyclization in the presence of Re2O7 to yield oxazoline 37 as a
7.2:1 mixture of diastereomers in a 53% yield after 36 h.
Several elements of this process distinguish it from the
conversion of 27 to 28. Mosher ester analysis of single
stereoisomers of the amido alcohol portion of the substrate
showed that epimerization did not occur in its preparation.16
NMR analysis indicated that the minor product diastereomer is
the cis-disubstituted oxazoline product, which is consistent
syn after 4 h at 45 °C and in a 68% yield from 27-anti after 24
h at 65 °C.
The difference in the cyclization rates of these isomeric
substrates can be attributed to the fates of the tight ion pairs
that form immediately upon the ionization of the intermediate
perrhenate esters (Scheme 4). syn-Perrhenate ester 30 ionizes
to form the tight ion pair 31 that is properly aligned to deliver
28 directly, whereby the thermodynamically favored trans-
1
with prior reports of the H NMR chemical shifts of the ring
hydrogens.8c We were unable to establish the stereochemical
orientation of the side chain carbamate unambiguously, so the
possibility of quantitative epimerization at this site, while
unlikely, cannot be discarded. The diminished thermodynamic
preference for the trans-isomer arises from the steric
interaction between the isopropyl group in the side chain
and the alkenyl group on the ring (see the Supporting
Information for details). The reaction was noticeably slower
than the reaction of 27 to 28. This can be attributed to the
presence of additional Lewis basic groups, the inductive
deactivation of the nucleophilic carbonyl group, and the
development of the steric interactions in the transition state.
These factors resulted in the cyclization being slower than the
transposition, leading to the syn- and anti-diastereomers
reacting with a similar efficiency. Despite the challenges of
this transformation, the product was generated with a useful
yield and stereocontrol. The resulting oxazoline provides an
Scheme 4. Ion Pairing in the Allylic Alcohol Isomers
C
https://dx.doi.org/10.1021/acs.orglett.0c03868
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
intriguing scaffold for library design based on the placement of
multiple substituents in disparate spatial orientations.
are core elements in numerous pharmaceutical agents and
biologically active natural products.
We then studied whether the scope of the method could be
expanded to prepare isoxazolines by using oxime nucleophiles.
While we showed that oximes undergo Beckmann rearrange-
ments under these conditions (Scheme 1), isoxazoline
formation should be viable if allylic cation formation and
cyclization is faster than oxime activation and migration. This
was demonstrated through the efficient conversion of oxime 38
to isoxazoline 39 (Scheme 6). The scope of this process was
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.0c03868.
■
sı
Schemes for substrate syntheses, cyclization protocols,
stereochemical determinations, 1H and 13C NMR
spectra, and crystallographic data for compound S4-syn
(PDF)
FAIR data, including the primary NMR FID files, for
compounds 5−27, 27-anti, 27-syn, 28, 29, 36−43, S1,
S2, S3, S4-syn, S4-anti, S5, and S6 (ZIP)
Scheme 6. Dehydrative Route to Isoxazolines
Accession Codes
CCDC 2027024 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: + 44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
■
Paul E. Floreancig − Department of Chemistry, University of
Pittsburgh, Pittsburgh, Pennsylvania 15260, United States;
orcid.org/0000-0001-9626-3187; Email: florean@
pitt.edu
tested by the exposing 40 to provide 41 in a good yield, albeit
requiring a longer reaction time compared to that of the
cyclization of 38 due to the greater difficulty in forming the
intermediate cation. Notably, this reaction can be conducted at
a 1 mmol scale without a reduction in efficiency, showing that
preparative-scale transformations are possible with this
method. The high yield of this transformation, where 40 was
employed as a 1.3:1 mixture of oxime geometrical isomers,
showed that the cyclization is slower than the E−Z-oxime
isomerization. Diastereocontrol can be achieved in this process
when a stereocenter is incorporated into the substrate, as seen
in the conversion of 42 (prepared as a 3:1 mixture; the isomers
were not determined) to isoxazoline 43 as a 7.5:1 mixture of
diastereomers. These processes in which the substrates can be
accessed through aldol reactions, followed by oxime formation,
provide a convenient alternative to cycloaddition-based
approaches to isoxazolines20 and proceed under milder
conditions than routes that employ the hydroxy group of the
oxime as a nucleophile in additions to oxidatively generated
allylic cations.21
We have demonstrated that Re2O7 and HOReO3 catalyze
dehydrative cyclizations to form nitrogen-containing hetero-
cycles from allylic alcohols. These transformations proceed
efficiently despite the mild basicity of the products. HFIP is the
optimal solvent for these transformations due to its ability to
stabilize the cationic intermediates, sequester water, and
mitigate the effects of product basicity through hydrogen
bonding. A wide range of structural variations are tolerated by
the protocol. Reaction rates are determined by the ease of the
cation formation and the kinetics of the ring closure.
Diastereocontrol was observed in several substrates, with rate
differences being attributed to tight ion-pair intermediates in
the cyclization transition states. The method was applied to the
synthesis of oxazolines, thiazolines, oxazines, thiazines, and
isoxazolines. This work vastly expands the scope of products
that are available through experimentally facile oxorhenium-
catalyzed dehydrative cyclizations to include heterocycles that
Author
Freddy O. Rodriguez del Rey − Department of Chemistry,
University of Pittsburgh, Pittsburgh, Pennsylvania 15260,
United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.orglett.0c03868
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National Institutes of Health (R01-GM103886)
for generous funding of this project. We thank Qi Qin
(University of Pittsburgh) for useful discussions on the design
of NMR experiments and Aidan Cole (University of
Pittsburgh) for conducting the initial studies on the Beckmann
rearrangement.
REFERENCES
■
(1) Meanwell, N. A. J. Med. Chem. 2011, 54, 2529−2691.
(2) Vitaku, E.; Smith, D. T.; Njardarson, J. T. J. Med. Chem. 2014,
57, 10257−10274.
(3) (a) Lovering, F.; Bikker, J.; Humblet, C. J. Med. Chem. 2009, 52,
6752−6756. (b) Lovering, F. MedChemComm 2013, 4, 515−519.
(4) Roy, R. S.; Gehring, A. M.; Milne, J. C.; Belshaw, P. J.; Walsh, C.
T.; Roy, R. S.; Gehring, A. M.; Milne, J. C.; Belshaw, P. J.; Walsh, C.
T. Nat. Prod. Rep. 1999, 16, 249−263.
(5) (a) Desimoni, G.; Faita, G.; Jørgensen, K. A. Chem. Rev. 2006,
106, 3561−3651. (b) Desimoni, G.; Faita, G.; Jørgensen, K. A. Chem.
Rev. 2011, 111, PR284−PR437.
(6) (a) Evans, D. L.; Minster, D. K.; Jordis, U.; Hecht, S. M.; Mazzu,
A. L.; Meyers, A. I. J. Org. Chem. 1979, 44, 497−501. (b) Williams, D.
R.; Lowder, P. D.; Gu, Y.-G.; Brooks, D. A. Tetrahedron Lett. 1997,
D
https://dx.doi.org/10.1021/acs.orglett.0c03868
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
38, 331−334. (c) Liu, L.; Floreancig, P. E. Org. Lett. 2010, 12, 4686−
4689. (d) Xu, Z.; Zhang, C.; Jiao, N. Angew. Chem., Int. Ed. 2012, 51,
11367−11370.
(7) (a) Wipf, P.; Venkatraman, S. Tetrahedron Lett. 1996, 37, 4659−
4662. (b) Burrell, G.; Evans, J. M.; Jones, G. E.; Stemp, G.
Tetrahedron Lett. 1990, 31, 3649−3652. (c) Crosignani, S.; Young, A.
C.; Linclau, B. Tetrahedron Lett. 2004, 45, 9611−9615. (d) Vor-
bruggen, H.; Krolikiewicz, K. Tetrahedron 1993, 49, 9353−9372.
̈
(e) Xu, Q.; Li, Z. Tetrahedron Lett. 2009, 50, 6838−6840.
(8) (a) Reddy, L. R.; Saravanan, P.; Corey, E. J. J. Am. Chem. Soc.
2004, 126, 6230−6231. (b) Sakakura, A.; Kondo, R.; Ishihara, K. Org.
Lett. 2005, 7, 1971−1974. (c) Morse, P. D.; Nicewicz, D. A. Chem.
Sci. 2015, 6, 270−274. (d) Huang, H.; Yang, W.; Chen, Z.; Lai, Z.;
Sun, J. Chem. Sci. 2019, 10, 9586−9590.
(9) For reviews on catalytic dehydrative processes, see: (a) Dry-
zhakov, M.; Richmond, E.; Moran, J. Synthesis 2016, 48, 935−959.
(b) Chen, L.; Yin, A.-P.; Wang, C.-H.; Zhou, J. Org. Biomol. Chem.
́
2014, 12, 6033−6048. (c) Baeza, A.; Najera, C. Synthesis 2013, 46,
25−34. (d) Ishihara, K. Tetrahedron 2009, 65, 1085−1109.
(10) For an oxidative approach to similar heterocycles, see:
Lemercier, B. C.; Pierce, J. G. Org. Lett. 2015, 17, 4542−4545.
(11) For reviews of Re2O7-mediated transformations, see:
(a) Volchkov, I.; Lee, D. Chem. Soc. Rev. 2014, 43, 4381−4394.
(b) Bellemin-Laponnaz, S. ChemCatChem 2009, 1, 357−362.
(12) Rohrs, T. M.; Qin, Q.; Floreancig, P. E. Angew. Chem., Int. Ed.
2017, 56, 10900−10904.
(13) (a) Jung, H. H.; Seiders, J. R., II; Floreancig, P. E. Angew.
Chem., Int. Ed. 2007, 46, 8464−8467. (b) Xie, Y.; Floreancig, P. E.
Chem. Sci. 2011, 2, 2423−2427. (c) Xie, Y.; Floreancig, P. E. Angew.
Chem., Int. Ed. 2013, 52, 625−628. (d) Xie, Y.; Floreancig, P. E.
Angew. Chem., Int. Ed. 2014, 53, 4926−4929. (e) Afeke, C.; Xie, Y.;
Floreancig, P. E. Org. Lett. 2019, 21, 5064−5067.
(14) Qin, Q.; Xie, Y.; Floreancig, P. E. Chem. Sci. 2018, 9, 8528−
8534.
(15) (a) Kusama, H.; Yamashita, Y.; Narasaka, K. Bull. Chem. Soc.
Jpn. 1995, 68, 373−377. (b) Furuya, Y.; Ishihara, K.; Yamamoto, H.
Bull. Chem. Soc. Jpn. 2007, 80, 400−406. (c) Mo, X.; Morgan, T. D.
R.; Ang, H. T.; Hall, D. G. J. J. Am. Chem. Soc. 2018, 140, 5264−5271.
(16) (a) Pozhydaiev, V.; Power, M.; Gandon, V.; Moran, J.; Lebœuf,
D. Chem. Commun. 2020, 56, 11548−11564. (b) Colomer, I.;
Chamberlain, A. E. R.; Haughey, M. B.; Donohoe, T. Nat. Rev. Chem.
̈
2017, 1, 0088. (c) Shuklov, I. A.; Dubrovina, N. V.; Borner, A.
Synthesis 2007, 2007, 2925.
̈
(17) Mayr, H.; Forner, W.; Schleyer, P. v. R. J. Am. Chem. Soc. 1979,
101, 6032−6040.
(18) Morales-Rivera, C. A.; Floreancig, P. E.; Liu, P. J. Am. Chem.
Soc. 2017, 139, 17935−17944.
(19) (a) Bellemin-Laponnaz, S.; Gisie, H.; Le Ny, J. P.; Osborn, J. A.
Angew. Chem., Int. Ed. Engl. 1997, 36, 976−978. (b) Bellemin-
Laponnaz, S.; Le Ny, J. P.; Dedieu, A. Chem. - Eur. J. 1999, 5, 57−64.
(20) Kiss, L.; Nonn, M.; Fulop, F. Synthesis 2012, 44, 1951−1963.
(21) (a) Zhu, X.; Wang, Y.-F.; Ren, W.; Zhang, F.-L.; Chiba, S. Org.
Lett. 2013, 15, 3214−3217. (b) Chen, F.; Yang, X.-L.; Wu, Z.-W.;
Han, B. J. Org. Chem. 2016, 81, 3042−3050.
E
https://dx.doi.org/10.1021/acs.orglett.0c03868
Org. Lett. XXXX, XXX, XXX−XXX