Palladium-Catalyzed Site-Selective Amidation of Dichloroazines

Article abstract of DOI:10.1021/acs.orglett.8b01483

A highly site-selective amidation reaction of substituted 2,4-dichloroazines is reported. Palladium acetate/1,1′-bis(diphenylphosphino)ferrocene (dppf) was identified as the optimal catalyst system, producing >99:1 C-2/C-4 selectivity for most examples. The generality of this transformation was demonstrated through a survey of a diverse amide/substituted 2,4-dichloroazine scope, leading to the preparation of the desired C-2 amidated products in good to excellent yields.

Full text of DOI:10.1021/acs.orglett.8b01483

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Palladium-Catalyzed Site-Selective Amidation of Dichloroazines
Ian S. Young,* Anna-Lena Glass, Theresa Cravillion, Chong Han,* Haiming Zhang,
and Francis Gosselin
Department of Small Molecule Process Chemistry, Genentech, Inc., 1 DNA Way, South San Francisco, California 94080,
United States
S
* Supporting Information
ABSTRACT: A highly site-selective amidation reaction of sub-
stituted 2,4-dichloroazines is reported. Palladium acetate/1,1′-
bis(diphenylphosphino)ferrocene (dppf) was identified as the
optimal catalyst system, producing >99:1 C-2/C-4 selectivity
for most examples. The generality of this transformation was
demonstrated through a survey of a diverse amide/substituted
2,4-dichloroazine scope, leading to the preparation of the desired
C-2 amidated products in good to excellent yields.
ighly functionalized heterocycles are valuable pharma-
cophores in drug discovery.1 Chemoselective or regiose-
Initial screening of a series of palladium catalysts to evaluate
the regioselectivity of the amidation was conducted for the
coupling of 2,4-dichloropyridine (1a) and 2-pyrrolidinone (2a)
using 5 mol % of palladium catalyst, potassium carbonate as
base, and toluene as the solvent (Figure 1). Among all of the
H
lective formation of a C−N bond2 on polyhalogenated hetero-
cyclic scaffolds3 has been an intriguing challenge for synthetic
chemists, and the development of selective processes could allow
for the construction of such diverse targets through sequential
transformations. The need for predictability and reproducibility
is especially important in a pharmaceutical setting where strin-
gent requirements on the acceptable levels of the minor isomer
in the active pharmaceutical ingredient (API) are enforced by
tight specifications.4
Pioneering work by Houk and co-workers through a com-
putational approach indicated that bond-dissociation energies
(BDEs) could be one determining factor in achieving selective
couplings on polyhalogenated heterocycles under palladium
catalysis.5 Although reports exist detailing selective transition-
metal-catalyzed C−N aminations on 2,4-dichloropyridine
derivatives,6 examples where the C−N bond is forged via an
amidation reaction are scarce.7 Morgentin and co-workers
reported a Pd-catalyzed site-selective coupling of a series of
N-acetylanilines with the C-6 position of 4,6-dichloronicotinoni-
trile in moderate yields (27−85%) using Xantphos as a ligand.8
In this study, the acetyl group served as a protecting group to
minimize the bis-arylation side product commonly observed in
Pd-catalyzed amination reactions involving primary amines and
was subsequently removed to yield a net amination. Examples
also exist for Pd-catalyzed C-2 selective ureidations of
2,4-dichloropyridines using Xantphos as the ligand, although
the reported substrate scope was limited to only a few examples.9
In the course of chemical development efforts on an internal
program in our laboratories, we were tasked with developing
a highly site-selective amidation protocol with substituted
2,4-dichloropyridines. Due to the lack of literature guidance
with regard to this challenge, we hoped to develop a general
solution to this problem and relied on high-throughput experi-
mentation10 to rapidly determine a starting point for further
optimization.
Figure 1. Ligand screening for regioselective amidation of 1a with 2a.
literature precedents for Pd-catalyzed amidation of aryl halides,11
a catalyst system consisting of a bidentate ligand such as
Xantphos, Josiphos, and dppf proved to afford the best selec-
tivity of the desired C-2 coupled product 3a.12 Dppf13 was
chosen over Josiphos as the optimal ligand for further opti-
mizations because of its more affordable cost. Following this
lead, potassium phosphate tribasic was identified as an optimal
Received: May 10, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b01483
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
base, affording higher conversion using only 1 mol % of
Pd(OAc)2 (Table 1, entries 2 and 3). Subsequent examination
Scheme 1. Scope of Amides in Pd-Catalyzed Coupling
Reactions with 2,4-Dichloropyridine (1a)
Table 1. Condition Optimizations for the Pd-Catalyzed
a
Coupling Reaction of 1a and 2a
Pd(OAc)2/dppf
(mol %)
conv/
b
bc
,
entry
base
solvent
3a
3a/3a′
1
2
3
4
5
6
7
8
5/10
1/2
1/2
1/0
1/1
1/3
0.5/1.5
1/3
K2CO3 toluene
86/84
98/97
94/93
0/0
56/54
100/99
82/81
97/95
98.5:1.5
98.6:1.4
>99:1
N/A
>99:1
>99:1
98.8:1.2
98.8:1.2
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
toluene
dioxane
dioxane
dioxane
dioxane
dioxane
t-AmOH
a
Reaction conditions: 1a (3 mmol), 2a (100 mol %), Pd(OAc)2
(0.5−5 mol %), ligand (1.5−10 mol %), base (200 mol %), solvent
b
(0.5 M). Determined by HPLC analysis of crude reaction mixture.
a
b
c
Reactions performed on 3 mmol scale. Isolated yields.
<1% 3a′′ observed in all cases.
acidity of the aromatic amide N−H led to partial formation of
the product potassium salt, leading to 40−50% of the material
being removed during filtration. Incorporation of an aqueous
workup for these two examples restored the yields to levels in
line with the other substrates. It should be noted that although
we did not prepare the C-4 markers for each example in
Scheme 1, in all cases no impurities were observed over 1 area
% by HPLC in the crude reaction mixture by LC−MS analysis,
indicating a >99:1 selectivity for the C-2 position. The positional
selectivity was confirmed by single-crystal X-ray crystallographic
analysis of 3b and 3k.16
We next examined how additional substituents on the
2,4-dichloropyridine scaffold impacted coupling efficiency with
2-pyrrolidinone (2a, Scheme 2). Both electron-donating and
electron-withdrawing substituents were tolerated at the 3-position
(4b−g), although substrates bearing electron-donating groups
required extended reaction times (−Me, 4f) or increased
catalyst loading (−OMe, 4g). No reaction was observed when
preparation of 4o bearing a free amino group was attempted,
which can be attributed to either chelation of the catalyst or
the increased electron density inhibiting oxidative addition.
The yield for 4d was the lowest reported, resulting from the
generation of significant double-addition product. Additional
substitution patterns (4h/4i) as well as other heterocycle
derivatives such as pyridazines (4j), quinolines (4k), and
7-azaindoles (4l) all resulted in good to excellent yields of
products.15 The reaction to produce free phenol containing 4m
did proceed cleanly, although under the standard conditions
the conversion did not proceed past 70%. The discrepancy
between observed conversion and isolated yield is the result
of loss during purification. Attempts to synthesize 4n led to
complicated reactions profiles. The site-selectivity was again
confirmed by obtaining single-crystal X-ray structures for com-
pounds 4d, 4h, and 4i.17
of the Pd precatalyst to ligand ratio revealed that the optimal
rate could be achieved using 1:3 Pd(OAc)2/dppf (entries
3−6). An attempt to lower the palladium loading to 0.5 mol %
based on the optimal 1:3 Pd/ligand ratio gave an incom-
plete 82% conversion after 1 h at 90 °C (entry 7). Though
1,4-dioxane was found to be the best solvent for this transforma-
tion, providing the highest C-2/C-4 product ratio and fast
reaction rate, we identified tert-amyl alcohol (t-AmOH) as a
greener alternative, giving only a slightly slower reaction rate
compared to the reaction in 1,4-dioxane (entries 8 vs 6).14 The
utility of the t-AmOH conditions was demonstrated on 60 mmol
scale (100 mol % of 2a utilized) to produce a crude reaction
profile of 99.0 area % of 3a, 0.6 area % of 3a′, and 0.4 area %
of 3a′′ by HPLC analysis and a 96% isolated yield after puri-
fication by column chromatography.
Upon identification of a C-2-selective catalytic system
(Table 1, entry 6), we surveyed the reaction generality with
respect to amide. For analogue preparation, the amide was
limited to 100 mol % to prevent unwanted conversion of the
desired product to the bis-amidated impurity. Even with-
out an excess of either coupling partner, all reactions with
2,4-dichloropyridine (1a) reached >60% conversion in 1 h and
>98% conversion after overnight heating (Scheme 1).15 The
amide scope was found to be broad, with primary aliphatic
amides of varying steric bulk yielding product in good to
excellent yields (3b−e). Secondary cyclic amides such as
ε-caprolactam (3j) and other nucleophiles such as oxazolidi-
nones (3i) and cyclic ureas (3k) also performed well in the
coupling. For the examples mentioned above, a simple workup
consisting of filtration to remove the inorganics, concentration
of the filtrate, and purification of the crude residue by column
chromatography was utilized. When the products derived from
aromatic amides (3f/3h) were subjected to a similar isolation
protocol, lower yields (45−55%) were encountered in multiple
attempts. This discrepancy was surprising, as the reaction pro-
files showed that product was formed in >95 area % as indi-
cated by HPLC. It was subsequently discovered that the higher
One discrepancy observed with these more advanced pyridine
substrates was that higher levels (1−4 area % by HPLC
analysis) of a minor impurity were observed in some of the
crude reaction mixtures in Scheme 2 (vs Scheme 1). HPLC−MS
B
DOI: 10.1021/acs.orglett.8b01483
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Scope of Dichloroazines in Pd-Catalyzed
Coupling Reactions with 2-Pyrrolidinone (2a)
Table 2. Preliminary Mechanistic Study To Probe the
Selectivity-Determining Step
a
a
a
conv
(%)
C-2 product
(A %)
C-4 product
(A %)
substrate
C-2/C-4
1a
1p
1q
100
86.3
50
99.4
6.4
nd
0.4
79.9
50
99.6/0.4
7.4/92.6
0/100
a
Determined by HPLC analysis of crude reaction mixture.
observed as the major one with a 7:93 C-2/C-4 product ratio.
These results are consistent with the first scenario, suggesting
oxidative addition is likely to be the selectivity determining
step that is in accord with the BDE proposal outlined by Houk
and co-workers (86.3 kcal/mol for C2−Cl bond vs 89.5 kcal/mol
for C4−Cl bond using B3LYP in the case of 2,4-dichloropyr-
idine).5
The ability to further functionalize the C-2-amidated
products at the C-4-position could extend the utility of this
method. Based on the lack of observed bis-amidated product
with Pd(OAc)2/dppf catalyst system (Schemes 1 and 2), we
increased both the amide stoichiometry (150 mmol %) and
catalyst loading (2 mol % Pd(OAc)2/6 mol % dppf) for the
initial attempts at installing the second amide. Gratifyingly, the
second amidation was rapid, reaching >99% conversion in 2 h
for two examples (5a/5b, Scheme 3). This strategy for the
a
Reactions performed on 3 mmol scale unless otherwise indicated in
b
c
the Supporting Information. Isolated yields. 5 mol % of Pd(OAc)2
and 15 mol % of dppf used.
analysis indicated that this new impurity had the same mass as
the desired product, leading to the conclusion that it was likely
the C-4 regioisomer. Closer examination indicated that the
levels of the C-4 products were the highest in pyridine deriva-
tives containing an electron-withdrawing substituent. This led
us to question whether this was the result of reduced catalyst
selectivity, or instead due to background SNAr reactivity that
favored the C-4 position. Performing the appropriate control
reactions [Pd(OAc)2 and dppf ligand omitted] led to signif-
icant amounts of both the C-2/C-4 products for 4b, 4c, 4d, 4h,
and 4i (7−41% conversions compared to 73−100% conversions
in Pd-catalyzed reactions after 1 h).18 Fortunately, even for these
electron-deficient substrates, the Pd-catalyzed C-2 amidation
process was found to outcompete the nonselective background
SNAr reaction favoring the C-4 amidation and provide excellent
selectivity (>96:4 C-2/C-4).
Scheme 3. Synthesis of Differentially Substituted Bis-
amidated Pyridines
There are two most probable scenarios to account for the
observed selectivity: (1) The product ratio is determined by a
rate-limiting oxidative addition step to form the corresponding
LPd(Het-Ar)(X) species which would undergo fast amide−
halide exchange and reductive elimination steps subsequently
to provide the C−N coupled products resembling the ratio of
the corresponding oxidative addition adducts. (2) The product
ratio is largely dependent on the rate of amide−halide exchange
step with the corresponding C-2 and C-4 oxidative addition
adducts being in a fast equilibrium under the Curtin−Hammett
principle.19 In the latter scenario, we envisioned that the reaction
would still favor the C-2 product even with a faster oxida-
tive addition to the C-4 position induced by a weaker C−X
bond. The coupling of 2-chloro-4-iodopyridine (1q) with
2-pyrrolidinone (2a) under the standard reaction conditions
afforded complete C-4-coupled product shown in Table 2,
albeit with a slower reaction rate after 1 h (50% vs complete
conversion using 2,4-dichloropyridine 1a); while in the case of
2-chloro-4-bromopyridine (1p), C-4 coupled product was still
preparation of positional isomers 5a/5b offered the advantage
that development and optimization of a second catalyst system
was not necessary to access diversely functionalized pyridine
derivatives in high overall yield. Based on these findings, we
hypothesized that it should be possible to arrive at the same
double adduct by a sequential addition of two different amides
to a single reaction flask. Proof of concept was obtained by
preparation of 5a via this strategy. The ease at which diversely
functionalized bis-amidated pyridines can be prepared via this
methodology should allow medicinal chemists to quickly
generate analogs. Although this process is initially selective for
C
DOI: 10.1021/acs.orglett.8b01483
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(3) For a recent review on site-selective functionalization of
halogenated heterocycles using Pd catalysis, see: (a) Fairlamb, I. J.
S. Chem. Soc. Rev. 2007, 36, 1036−1045. For other representative C−
N coupling examples, see: (b) Peng, Z.-H.; Journet, M.; Humphrey,
G. Org. Lett. 2006, 8, 395−398. (c) Keylor, M. H.; Niemeyer, Z. L.;
Sigman, M. S.; Tan, K. L. J. Am. Chem. Soc. 2017, 139, 10613−10616.
(4) ICH Harmonised Tripartite Guideline: Impurities in New Drug
Substances Q3A (R2). http://www.ich.org/fileadmin/Public_Web_
Site/ICH_Products/Guidelines/Quality/Q3A_R2/Step4/Q3A_
R2__Guideline.pdf (accessed 04/29/2018).
(5) (a) Legault, C. Y.; Garcia, Y.; Merlic, C. A.; Houk, K. N. J. Am.
Chem. Soc. 2007, 129, 12664−12665. (b) Garcia, Y.; Schoenebeck, F.;
Legault, C. Y.; Merlic, C. A.; Houk, K. N. J. Am. Chem. Soc. 2009, 131,
6632−6639.
(6) For selected examples, see: (a) Morgentin, R.; Barlaam, B.;
Foote, K.; Hassall, L.; Hawkins, J.; Jones, C. D.; Le Griffon, A.; Peru,
C-2 amidation, tailoring the order of which the amide reactants
are added allows for indirect selectivity at the 4-position
(compare 5a/b). It should be noted that development of a
highly C-4 selective Pd-catalyzed process on 2,4-dichloropyr-
idine remains an unsolved problem.20
In summary, we have identified a catalyst system that
amidates dichloroazines such as substituted 2,4-dichloropyr-
idines with a high level of C-2 selectivity. The substrate scope
with respect to both amide and dichloroazine is broad, allowing
for the preparation of a diverse series of functionalized hetero-
cycles in good to excellent yields (70−97%). The utility of this
method was further expanded through development of sequential
amidation protocols to form differentially substituted bis-
amidated pyridines. The high level of site-selectivity imparted
by this catalyst system is intriguing from a mechanistic per-
spective, and additional studies will be reported in due course.
́
A.; Ple, P. Synth. Commun. 2012, 42, 8−24. (b) Burton, R. J.;
Crowther, M. L.; Fazakerley, N. J.; Fillery, S. M.; Hayter, B. M.;
Kettle, J. G.; McMillan, C. A.; Perkins, P.; Robins, P.; Smith, P. M.;
Williams, E. J.; Wrigley, G. L. Tetrahedron Lett. 2013, 54, 6900−6904.
(7) Examples from the patent literature utilized Xantphos or t-
BuXPhos derivative ligands which were shown to be suboptimal in
our reaction development: (a) Brown, M.; Chen, Y.; Cushing, T. D.;
Gonzalez Lopez De Turiso, F.; He, X.; Kohn, T. J.; Lohman, J. W.;
Pattaropong, V.; Seganish, J.; Shin, Y.; Simard, J. L. PCT Int. Patent
Appl. WO 2010151737 A2, Dec 29, 2010. (b) Jung, P. J. M.; Lachia,
M. D.; Leipner, J.; Brocklehurst, D.; De Mesmaeker, A.; Wendeborn,
S. V. PCT Int. Patent Appl. WO 2014131732 A2, Sep 4, 2014. (c)
Bronson, J. J.; Chen, L.; Ditta, J. L.; Dzierba, C. D.; Jalagam, P. R.;
Luo, G.; Macor, J. E.; Maishal, T. K.; Nara, S. J.; Rajamani, R.; Sistla,
R. K.; Thangavel, S. PCT Int. Patent Appl. WO 2017059085 A1, Apr
6, 2017.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.8b01483.
■
S
General experimental procedures, characterization
1
details, and H, 13C, and 19F NMR (PDF)
Single-crystal X-ray reports (PDF)
Accession Codes
CCDC 1842393−1842398 contain the supplementary crys-
tallographic 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
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(8) Delvare, C.; Koza, P.; Morgentin, R. Synthesis 2011, 2011,
2431−2436.
́
(9) (a) Abad, A.; Agullo, C.; Cunat, A. C.; Vilanova, C. Synthesis
̃
2005, 2005, 915−924. (b) Yule, I. A.; Czaplewski, L. G.; Pommier, S.;
Davies, D. T.; Narramore, S. K.; Fishwick, C. W. G. Eur. J. Med. Chem.
2014, 86, 31−38.
AUTHOR INFORMATION
Corresponding Authors
■
(10) Gordillo, A.; Titlbach, S.; Futter, C.; Lejkowski, M. L.; Prasetyo,
E.; Rupflin, L. T. A.; Emmert, T.; Schunk, S. A. High-Throughput
Experimentation in Catalysis and Materials Science. In Ullmann’s
Encyclopedia of Industrial Chemistry; Wiley-VCH: Weinheim, 2014; pp
1−19.
*E-mail: young.ian@gene.com.
*E-mail: han.chong@gene.com.
ORCID
(11) For representative examples, see: (a) Hartwig, J. F.; Kawatsura,
M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar-Roman, L. M. J. Org.
Chem. 1999, 64, 5575−5580. (b) Yin, J.; Buchwald, S. L. J. Am. Chem.
Soc. 2002, 124, 6043−6048. (c) Huang, X.; Anderson, K. W.; Zim, D.;
Jiang, L.; Klapars, A.; Buchwald, S. L. J. Am. Chem. Soc. 2003, 125,
6653−6655. (d) Klapars, A.; Campos, K. R.; Chen, C.; Volante, R. P.
Org. Lett. 2005, 7, 1185−1188. (e) Hicks, J. D.; Hyde, A. M.; Cuezva,
A. M.; Buchwald, S. L. J. Am. Chem. Soc. 2009, 131, 16720−16734.
(12) Product site selectivities were unambiguously assigned by
obtaining the single-crystal X-ray structure of 3a′ (C-4 product).
(13) Sollot, G. P.; Snead, J. L.; Portnoy, S.; Peterson, W. R., Jr.;
Mertwoy, H. E. Chem. Abstr. 1965, 63, 18147.
(14) (a) Prat, D.; Pardigon, O.; Flemming, H.-W.; Letestu, S.;
Ducandas, V.; Isnard, P.; Guntrum, E.; Senac, T.; Ruisseau, S.;
Cruciani, P.; Hosek, P. Org. Process Res. Dev. 2013, 17, 1517−1525.
(b) Prat, D.; Hayler, J.; Wells, A. Green Chem. 2014, 16, 4546−4551.
(15) In most cases, the reactions were heated overnight to ensure
full consumption of 1. See the Supporting Information for relative
rates of reaction comparison based on amide/substituted dichlor-
oazine.
Ian S. Young: 0000-0003-0179-9849
Chong Han: 0000-0002-2863-3921
Haiming Zhang: 0000-0002-2139-2598
Francis Gosselin: 0000-0001-9812-4180
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Professor Stephen Buchwald (Massachusetts
Institute of Technology) for helpful discussions, Dr. Antonio
DiPasquale for single-crystal X-ray analysis, and Dr. Kenji
Kurita for collecting HRMS data (Genentech, Inc.).
REFERENCES
■
(1) For a recent review, see: Taylor, A. P.; Robinson, R. P.; Fobian,
Y. M.; Blakemore, D. C.; Jones, L. H.; Fadeyi, O. Org. Biomol. Chem.
2016, 14, 6611−6637.
(16) See the Supporting Information for more details. NOESY
analysis of compounds 3i, 3j, and 3k could not conclusively
distinguish between C-2 vs C-4 amidation.
(17) See the Supporting Information for more details. NOESY
analysis of compounds 4h, 4j, and 4m could not conclusively
distinguish between C-2 vs C-4 amidation.
(2) For reviews on Pd- or Cu-catalyzed C−N couplings, see:
(a) Evano, G.; Blanchard, N.; Toumi, M. Chem. Rev. 2008, 108,
3054−3131. (b) Hartwig, J. F. Acc. Chem. Res. 2008, 41, 1534−1544.
(c) Surry, D. S.; Buchwald, S. L. Chem. Sci. 2010, 1, 13−31. (d) Surry,
D. S.; Buchwald, S. L. Chem. Sci. 2011, 2, 27−50. (e) Ruiz-Castillo, P.;
Buchwald, S. L. Chem. Rev. 2016, 116, 12564−12649.
D
DOI: 10.1021/acs.orglett.8b01483
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(18) See the Supporting Information for more details. Running the
control reactions in a polar aprotic solvent such as DMSO to mimic
standard SNAr conditions led to higher ratios of the C-4 product.
Similar competing background reactivity was observed during the
selective Pd-catalyzed amination of 2,4-dichloropyrimidines; see ref
3b.
(19) Seeman, J. I. Chem. Rev. 1983, 83, 83−134.
(20) Dai, X.; Chen, Y.; Garrell, S.; Liu, H.; Zhang, L.-K.; Palani, A.;
Hughes, G.; Nargund, R. J. Org. Chem. 2013, 78, 7758−7763.
E
DOI: 10.1021/acs.orglett.8b01483
Org. Lett. XXXX, XXX, XXX−XXX