TCFH-NMI: Direct Access to N-Acyl Imidazoliums for Challenging Amide Bond Formations

Article abstract of DOI:10.1021/acs.orglett.8b01591

Challenging couplings of hindered carboxylic acids with non-nucleophilic amines to form amide bonds can be accomplished in high yields, and in many cases, with complete retention of the adjacent stereogenic centers using the combination of N,N,N′,N′-tetramethylchloroformamidinium hexafluorophosphate (TCFH) and N-methylimidazole (NMI). This method allows for in situ generation of highly reactive acyl imidazolium ions, which have been demonstrated to be intermediates in the reaction. The reagent delivers high reactivity similar to acid chlorides with the ease of use of modern uronium reagents.

Full text of DOI:10.1021/acs.orglett.8b01591

This is an open access article published under an ACS AuthorChoice License, which permits
copying and redistribution of the article or any adaptations for non-commercial purposes.
Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
TCFH−NMI: Direct Access to N‑Acyl Imidazoliums for Challenging
Amide Bond Formations
Gregory L. Beutner,* Ian S. Young,† Merrill L. Davies, Matthew R. Hickey, Hyunsoo Park,
Jason M. Stevens, and Qingmei Ye
Chemical and Synthetic Development, Bristol-Myers Squibb Company, One Squibb Drive, New Brunswick, New Jersey 08903,
United States
S
* Supporting Information
ABSTRACT: Challenging couplings of hindered carboxylic acids with non-nucleophilic amines to form amide bonds can be
accomplished in high yields, and in many cases, with complete retention of the adjacent stereogenic centers using the
combination of N,N,N′,N′-tetramethylchloroformamidinium hexafluorophosphate (TCFH) and N-methylimidazole (NMI).
This method allows for in situ generation of highly reactive acyl imidazolium ions, which have been demonstrated to be
intermediates in the reaction. The reagent delivers high reactivity similar to acid chlorides with the ease of use of modern
uronium reagents.
cyl imidazoliums have been long recognized as highly
reactive acyl transfer agents in the context of amide bond
formation. Although some attention has been given to the
activation of acyl imidazoles with Brønsted acids,1 these species
can be orders of magnitude less reactive than the N-alkylated
analogues, as was noted by Jencks and Lapshin in their kinetic
studies (see Scheme 1).2 Strategies for the synthesis of N-
combines acid chlorides or other active esters with N-
methylimidazole (NMI) has only found limited use, despite
the significant potential of the N-acylimidazolium intermedi-
ate.5
A
As part of ongoing studies that are focused on the large-scale
synthesis of a potential pharmaceutical agent, we discovered
the combination of N,N,N′,N′-tetramethylchloroformamidi-
nium hexafluorophosphate (TCFH) and NMI is a mild
method for in situ generation of highly reactive acyl
imidazoliums, allowing for the formation of the amide in
high yield and with no detectable epimerization. TCFH has
found limited application in the literature, as a reagent for the
in situ generation of acid chlorides6 or, more commonly, as a
precursor in the synthesis of complex uronium-based amide
bond-forming agents.7 Here, the combination of TCFH and
NMI demonstrated advantages in terms of reactivity, but also
practical benefits from the perspective of cost, ease of use, and
the direct access it provides to N-acyl imidazoliums, when
compared to the existing literature methods, which require
either isolation or preformation of some activated ester.5 In
this report, we describe and explore the scope and limitations
of this TCFH-NMI reagent combination for amide bond
formation, which is one of the most common transformations
in pharmaceutical synthesis (see Scheme 2).8 This work
highlights the unique properties of these in situ generated acyl
Scheme 1. N-Acyl Imidazoles and Imidazoliums
methylimidazoliums i have relied on activation of preformed
acyl imidazoles with strong alkylating agents such as
Meerwein’s salt (Me3OBF4), methyl triflate (MeOTf), or
methyl iodide (MeI).3 Work by Rapoport and Batey identified
analogues of carbonyl diimidazole (CDI) for direct access to
highly reactive N-acyl imidazoliums and use in the preparation
of challenging amide bonds (see Scheme 1).4 The need to
isolate these reagents to remove highly reactive alkylating
agents such as Meerwein’s salt, MeOTf, or MeI has limited
their application. A conceptually simpler approach that
Received: May 20, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b01591
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
addition of water, allowing for easy removal of reaction
byproducts such as NMI salts and tetramethyl urea. Finally, the
unique combination of Lewis and Brønsted basicities provided
by NMI11 in this reaction was highlighted by comparison to
other strong basic amines. The weakly Lewis basic amine
DIPEA gave higher levels of anhydride formation, while the
highly Lewis and Brønsted basic amine DBU gave little of the
desired product (Table 1, entries 8 and 9). Re-evaluation of
the other coupling reagents with the optimal solvent and base
highlights the unique properties of TCFH for this trans-
formation, although BEP is comparable (Table 1, entry 7 vs
entries 10−12).12
In order to gain further insight into the high reactivity of this
TCFH-NMI system and confirm the intermediacy of an acyl
imidazolium, a series of spectroscopic studies was undertaken
on the reaction between 1 and 2a. Monitoring the reaction by
in situ IR showed that the reaction was fast when TCFH was
added last, but several distinct reaction intermediates were
visible (see Figure 1). The first short-lived intermediate was a
Scheme 2. A General Strategy for TCFH−NMI Couplings
imidazolium reagents, particularly, for coupling of traditionally
challenging hindered carboxylic acids and poorly nucleophilic
amines.
Our initial studies focused on the coupling between the
hindered carboxylic acid 1 and the electron-deficient aniline
2a. Because of the low nucleophilicity of 2a, common coupling
agents such as HATU, BEP, TFFH, and TCFH with DIPEA as
base in DMF, primarily led to the anhydride 4, along with
significant amounts of the corresponding activated esters (see
Table 1, entries 1−4).9 In combination with TCFH,
Table 1. Screening Amide Bond Formation with 1 and 2a
b
b
conversion
(%)
3a vs 4
(%)
a
entry activator
solvent
base (equiv)
1
2
3
4
5
6
7
8
9
HATU
BEP
DMF
DMF
DMF
DMF
DMF
DIPEA (3.0)
DIPEA (3.0)
DIPEA (3.0)
DIPEA (3.0)
NMI (3.0)
2.6
>99.9
15.5
96.2
85.2
98.5
>99.9
>99.9
24.3
2.7
>99:1
<1:99
<1:99
7.6:92.4
72.6:27.4
>99:1
Figure 1. ReactIR analysis of the reaction of 1 and 2a.
TFFH
TCFH
TCFH
TCFH
TCFH
TCFH
TCFH
HATU
BEP
band at 1691 cm−1, which did not correspond to TCFH itself
(1653 cm−1) or the starting acid 1 (1737 cm−1). This
intermediate was rapidly consumed and replaced by a second
intermediate, distinguished by a strong band at 1765 cm−1.
Pairwise combination of the reagents revealed that the first
intermediate was a complex between TCFH and NMI. NMR
characterization of this intermediate suggested the structure ii,9
which was subsequently crystallized from MeCN/MTBE and
the structure was confirmed by X-ray crystallographic analysis.
The addition of acid 1 to ii led to the rapid formation of the
second intermediate at 1765 cm−1. Again, NMR and MS
characterization of this intermediate was consistent with the
proposed acyl imidazolium iii.13,14
Based on these data, we propose the following mechanism
for the reaction (Figure 2). In the presence of NMI, TCFH is
rapidly converted to adduct ii, which reacts with acid 1 to
generate iii. This highly reactive acyl imidazolium iii then
reacts with the amine to generate the amide, while
simultaneously releasing NMI, which scavenges the proton
released in the C−N bond formation. Because of the rapid
nature of the reaction, we cannot rule out the intermediacy of
an acid chloride.
MeCN NMI (3.0)
MeCN NMI (3.5)
MeCN DIPEA (3.5)
MeCN DBU (3.5)
MeCN NMI (3.5)
MeCN NMI (3.5)
MeCN NMI (3.5)
>99:1
67.8:32.2
6.7:93.3
>99:1
98.6:1.4
92.5:7.5
10
11
12
92.5
38.0
TFFH
a
See the Supporting Information for activator/base structures.
Conversions are calculated based on HPLC area percentage.
b
exchanging the strongly Brønsted basic, weakly Lewis basic
DIPEA (pKa = 11.4)10a for the less Brønsted basic, but highly
Lewis basic amine NMI (pKa = 7.2)10b led to a dramatic
change in the reaction, with the desired amide 3a now being
the major reaction product (Table 1, entries 4 and 5). Running
the reaction in acetonitrile (Table 1, entry 6) minimized
anhydride formation and provided the desired product 3a in
96 HPLC area percent (AP) after 14 h at room temperature.
NMI stoichiometry was examined and it was found that
increasing up to 3.5 equiv suppressed the formation of
anhydride 4 and significantly improved the reaction purity
(Table 1, entries 6 and 7). A wide variety of solvents, including
dichloromethane and toluene gave comparable results to
MeCN, but TCFH had poor solubility and afforded difficult-
to-stir heterogeneous reactions. The homogeneous reaction in
MeCN was preferred since it facilitated isolation by the
This clearer understanding of the reaction mechanism gave
us confidence in the optimal conditions and we moved on to
survey the reaction scope, with respect to the amine, giving
particular focus to poorly reactive anilines (see Figure 3).
Based on the fact that aniline 2a, one of the more difficult
B
DOI: 10.1021/acs.orglett.8b01591
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 2. Conditions for Amide Bond Formation with (S)-5
b
base
(equiv)
conversion
(3 h)
enantiomeric ratio,
er
a
entry activator
1
2
3
4
5
6
7
8
9
TCFH
TCFH
HBTU
HATU
HOTU
BEP
BOP
PyBrOP
DMTMM
NMI (3.5)
NMI (2.1)
NMI (2.1)
NMI (2.1)
NMI (2.1)
NMI (2.1)
NMI (2.1)
NMI (2.1)
NMI (2.1)
>99
>99
6
32
35
>99
11
74
86:14
>99.9:0.1
57:43
57:43
73:27
88:12
60:40
91:9
51:49
Figure 2. Mechanism of the TCFH−NMI amide formation.
81
a
See the Supporting Information for activator/base structures.
Conversions are calculated based on the HPLC area percentage
b
adjusted for UV response factors.
reaction, no difference in yield or enantiopurity of the isolated
amide was observed, regardless of whether TCFH was added
last or if aniline was added last, attesting to the stability of the
acyl imidazolium under these effectively neutral reaction
conditions. A comparison to other standard amide coupling
reagents demonstrates the advantage of this TCFH−NMI
coupling (Table 2, entries 3−9). BEP and PyBrOP, which are
often considered optimal reagents for epimerization prone
substrates,17 along with DMTMM, showed high reactivity but
exhibited significant scrambling of the stereogenic center.
Running the same reactions with DIPEA as base leads to low
conversion and numerous side products, even with TCFH.
These results further demonstrate the importance of Brønsted
base choice in this reaction.
With these improved conditions in hand, a survey of amines
in coupling reactions with 5 showed good reactivity with no
detectable loss in the purity of the adjacent stereogenic center,
even without preactivation, when adding TCFH last (see
Figure 4). Additional experiments with more-nucleophilic
amines and less-hindered acids also performed well, giving
good yields while maintaining the integrity of the stereogenic
center (see Figure 5).
Figure 3. Exploration of substrate scope with acid 1.
examples in previous work,15 quickly and smoothly coupled
with the sterically hindered acid 1 at room temperature using
the TCFH−NMI system (>87 AP product in 2 h), we were
hopeful that other challenging amines would also yield
acceptable results under mild conditions. A series of sterically
and electronically diverse anilines, as well as heteroaryl amines,
all coupled in reasonable reaction times and good yields, even
when the reactions were run under air, without rigorous
anhydrous technique and with no preactivation. As could be
expected, the electronic influence of the 4-position substituent
on the aniline had a large impact on the reaction rate.
Generally, the amide bond formation was complete in less than
24 h at room temperature, and the amides 3a−3g could be
isolated in high yields and >99 AP by the direct addition of
water to the reaction mixture.
The final challenge for this reaction system was the
extension to peptide couplings. Peptide couplings, especially
To further explore the scope of the method and introduce
the additional challenge of preserving a stereogenic center
adjacent to the carboxylic acid, we investigated the coupling of
(S)-phenylpropionic acid 5 with 2a (Table 2).9 Few examples
of couplings with this class of carboxylic acid are known, and
little data on the integrity of adjacent stereocenters is
available.16 Under the optimal conditions developed for 1,
the coupling of 5 with 2a led to high conversion and yield in
less than 30 min, but gave only a 86:14 enantiomeric ratio (er)
of the isolated amide 6a (Table 2, entry 1). Lowering the
charge of NMI to 2.1 equiv restored the integrity of the α-
stereogenic center, allowing for isolation of the desired amide
6a in 93% yield and >99.9:0.1 er. Because of the rapid rate of
Figure 4. Exploration of substrate scope with acid (S)-5.
C
DOI: 10.1021/acs.orglett.8b01591
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
In summary, we have found a simple and robust method to
access acyl imidazoliums and leverage their high reactivity for
the formation of challenging amide bonds under mild
conditions. These conditions also present operational benefits
when compared to other reagents for amide bond formation.
The fact that the reaction can be run without preactivation,
along with the low cost and availability of TCFH makes it a
practical choice, even when used at large scale.19 The
avoidance of additives and byproducts, which can be
challenging to remove or control, such as phosphoramides or
hydroxybenzotriazole derivatives, facilitates handling and
improves the safety of the process.20 Finally, the high water
solubility of all reaction byproducts significantly streamlines
workup and isolation.
In a more general sense, this work also makes clear how
reconsidering our assumptions about Brønsted base choice in a
reaction can have dramatic effects. The pKa of the acidic
hydrogen being removed should always be a consideration,
but, more importantly, the use of a more Lewis basic amine
may lead to unexpected benefits.
Figure 5. Additional substrate scope with TCFH−NMI.
those involving hindered amino acids or residues that are
prone to epimerization, have been a major driver for the
development of new amide bond-forming reagents. Initial
experiments to form simple dipeptides with hindered residues
such as valine, phenylalanine, and protected forms of cysteine
proceeded to high yield with nondetectable amounts of
epimerization, even without preactivation of the acid (see
Figure 6, eqs 1−3). When the reagent was tested in the
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b01591.
1
Experimental procedures, characterization data and H
and 13C NMR data (PDF)
Accession Codes
CCDC 1844135 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, U.K.; fax: +44 1223 336033.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: gregory.beutner@bms.com.
ORCID
Gregory L. Beutner: 0000-0001-8779-1404
Ian S. Young: 0000-0003-0179-9849
Present Address
†Department of Small Molecule Process Chemistry, Genen-
tech, Inc., 1 DNA Way, South San Francisco, CA 94080, USA.
Figure 6. Peptide couplings with TCFH−NMI.
Notes
formation of the Anteunis tripeptide Z-Gly-Phe-Val-OMe 22,18
minor amounts of epimerization were observed (Figure 6, eq
4). Despite the neutral reaction conditions, the high reactivity
of the acyl imidazolium renders it susceptible to intramolecular
attack by a distal amide in a peptide chain, and epimerization is
observed. Still, TCFH-NMI may hold some advantages for
extremely hindered peptide couplings, such as those involving
aminoisobutyric acid (Aib) residues. Under the standard
conditions, 92% yield of the desired dipeptide 24 can be
obtained within <30 min (Figure 6, eq 5).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors would like to thank Albert DelMonte, Matthew
Haley, and Christopher Wilbert (Bristol-Myers Squibb
Company) for their work in the development and scaleup of
this process. The authors would also like to acknowledge
Robert Waltermire (Bristol-Myers Squibb Company) for his
support of this work.
D
DOI: 10.1021/acs.orglett.8b01591
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(20) Wehrstedt, K. D.; Wandrey, P. A.; Heitkamp, D. J. Hazard.
Mater. 2005, 126, 1−7.
REFERENCES
■
(1) (a) Heller, S. T.; Fu, T.; Sarpong, R. Org. Lett. 2012, 14, 1970−
1973. (b) Woodman, E. K.; Chaffey, J. G. K.; Hopes, P. A.; Hose, D.
R. J.; Gilday, J. P. Org. Process Res. Dev. 2009, 13, 106−113.
(2) (a) Lapshin, S. A.; Smirnov, Y. I.; Livinenko, L. M.; Fedorov, V.
V.; Kapkan, L. M.; Lange, R. J. Gen. Chem. USSR 1985, 55, 1385−
1389. (b) Oakenfull, D. G.; Salvesen, K.; Jencks, W. P. J. Am. Chem.
Soc. 1971, 93, 188−194. (c) Wolfenden, R.; Jencks, W. P. J. Am.
Chem. Soc. 1961, 83, 4390−4393.
(3) (a) Vaillard, V. A.; Gonzalez, M.; Perotti, J. P.; Grau, R. J. A.;
Vaillard, S. E. RSC Adv. 2014, 4, 13012−13017. (b) Watkins, B. E.;
Kiely, J. S.; Rapoport, H. J. Am. Chem. Soc. 1982, 104, 5702−5708.
(4) (a) Grzyb, J. A.; Shen, M.; Yoshina-Ishii, C.; Chi, W.; Brown, R.
S.; Batey, R. A. Tetrahedron 2005, 61, 7153−7175. (b) Grzyb, J. A.;
Batey, R. A. Tetrahedron Lett. 2003, 44, 7485−7488. (c) Saha, A. K.;
Schultz, P.; Rapoport, H. J. Am. Chem. Soc. 1989, 111, 4856−4859.
(5) (a) Liu, Y.; Prashad, M.; Ciszewski, L.; Vargas, K.; Repic, O.;
Blacklock, T. J. Org. Process Res. Dev. 2008, 12, 183−191. (b) Ueki,
H.; Ellis, T. K.; Martin, C. H.; Boettiger, T. U.; Bolene, S. B.;
Soloshonok, V. A. J. Org. Chem. 2003, 68, 7104−7107. (c) Ueki, H.;
Ellis, T. K.; Martin, C. H.; Soloshonok, V. A. Eur. J. Org. Chem. 2003,
2003, 1954−1957. (d) Wakasugi, K.; Iida, A.; Misaki, T.; Nishii, Y.;
Tanabe, Y. Adv. Synth. Catal. 2003, 345, 1209−1214.
(6) (a) Tulla-Puche, J.; Torres, A.; Calvo, P.; Royo, M.; Albericio, F.
Bioconjugate Chem. 2008, 19, 1968−1971. (b) Vendrell, M.; Ventura,
R.; Ewenson, A.; Royo, M.; Albericio, F. Tetrahedron Lett. 2005, 46,
5383−5386.
(7) (a) deGruyter, J. N.; et al. Org. Lett. 2017, 19, 6196−6199.
(b) Carpino, L. A.; El-Faham, A. J. Am. Chem. Soc. 1995, 117, 5401−
5402. (c) Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397−4398.
(d) Dourtoglou, V.; Gross, B.; Lambropoulou, U.; Zioudrou, Z.
Synthesis 1984, 1984, 572−574.
(8) Carey, J. S.; Laffan, D.; Thomson, C.; Williams, M. T. Org.
Biomol. Chem. 2006, 4, 2337−2347.
(9) See the Supporting Information for further details.
(10) (a) Fujii, T.; Nishida, H.; Abiru, Y.; Yamamoto, M.; Kise, M.
Chem. Pharm. Bull. 1995, 43, 1872−1877. (b) Gagliano, R. A.;
Knowlton, R. C.; Byers, L. D. J. Org. Chem. 1989, 54, 5247−5250.
(11) (a) Baidya, M.; Brotzel, F.; Mayr, H. Org. Biomol. Chem. 2010,
8, 1929−1935. (b) Bassindale, A. R.; Stout, T. Tetrahedron Lett. 1985,
26, 3403−3406.
(12) Some safety concerns have been identified with BEP; see: Li,
P.; Xu, J. C. Tetrahedron 2000, 56, 8119−8131.
1
(13) H−13C HMBC cross peaks observed from imidazolium peaks
to carbonyl carbon; see the Supporting Information for full
characterization.
(14) (a) Leclercq, L.; Suisse, I.; Agbossou-Niedercorn, F. Eur. J. Org.
Chem. 2010, 2010, 2696−2700. (b) Lapshin, S. A.; Kapkan, L. M.;
Chervinskii, A. Y.; Fedorov, V. V.; Lange, R. Zh. Org. Khim. 1985, 21,
1389−1395.
́
(15) (a) Larrivee-Aboussafy, C.; Jones, B. P.; Price, K. E.; Hardink,
M. A.; McLaughlin, R. W.; Lillie, B. M.; Hawkins, J. M.;
Vaidyanathan, R. Org. Lett. 2010, 12, 324−327. (b) Dunn, P. J.;
Hoffmann, W.; Kang, Y.; Mitchell, J. C.; Snowden, M. J. Org. Process
Res. Dev. 2005, 9, 956−961.
(16) (a) Cheung, Y. Y.; Yu, H.; Xu, K.; Zou, B.; Wu, M.; McManus,
O. B.; Li, M.; Lindsley, C. W.; Hopkins, C. R. J. Med. Chem. 2012, 55,
6975−6979. (b) Li, J.; Subramaniam, K.; Smith, D.; Qiao, J. X.; Li, J.
J.; Qian-Cutrone, J.; Kadow, J. F.; Vite, G. D.; Chen, B. C. Org. Lett.
2012, 14, 214−217. (c) Dunetz, J. R.; Xiang, Y.; Baldwin, A.;
Ringling, J. Org. Lett. 2011, 13, 5048−5051.
(17) (a) Valeur, E.; Bradley, M. Chem. Soc. Rev. 2009, 38, 606−631.
(b) Li, P.; Xu, J. C. J. Org. Chem. 2000, 65, 2951−2958. (c) Han, Y.;
Albericio, F.; Barany, G. J. Org. Chem. 1997, 62, 4307−4312.
(18) van der Auwera, C.; van Damme, S.; Anteunis, M. J. O. Int. J.
Pept. Protein Res. 1987, 29, 464−471.
(19) Pricing for coupling agents at kilogram scale as of March 2016:
TCFH, $56/mol; T3P, $58/mol; EDAC, $93/mol; HATU, $171/
mol.
E
DOI: 10.1021/acs.orglett.8b01591
Org. Lett. XXXX, XXX, XXX−XXX