Photoredox-mediated C-H functionalization and coupling of tertiary aliphatic amines with 2-chloroazoles

Article abstract of DOI:10.1021/ol402751j

Herein, conditions for C-H functionalization of tertiary aliphatic amines and their subsequent coupling with a number of 2-chloroazole derivatives are reported. The reaction is facilitated by a catalytic amount of tris-fac-Ir(ppy)₃, with blue light irradiation and takes place under mild and convenient conditions. Most couplings take place with excellent regioselectivity. The reaction is tolerant of a number of functional groups and allows for rapid access to α-azole carbinamines commonly found in post-translationally modified peptides.

Full text of DOI:10.1021/ol402751j

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
Photoredox-Mediated CꢀH
000–000
Functionalization and Coupling of Tertiary
Aliphatic Amines with 2‑Chloroazoles
Anuradha Singh, Amandeep Arora, and Jimmie D. Weaver*
†
†
Department of Chemistry, Oklahoma State University, Stillwater, Oklahoma 74078,
United States
jimmie.weaver@okstate.edu
Received September 23, 2013
ABSTRACT
Herein, conditions for CꢀH functionalization of tertiary aliphatic amines and their subsequent coupling with a number of 2-chloroazole derivatives
are reported. The reaction is facilitated by a catalytic amount of tris-fac-Ir(ppy) , with blue light irradiation and takes place under mild and
convenient conditions. Most couplings take place with excellent regioselectivity. The reaction is tolerant of a number of functional groups and
3
allows for rapid access to R-azole carbinamines commonly found in post-translationally modified peptides.
Aromatic heterocycles are a ubiquitous motif in natural
products, drug molecules, and pesticides. One important
subclass contains tertiary aliphatic amines adjacent to the
heterocycle (Figure 1). Both in nature and in the labratory,
access to this motif is often accomplished via cyclodehy-
dration to form the heterocycle. While cyclodehydration
methodology is routine, it necessitates early incorpora-
tion of substituents, thus, elucidation of structural activity
relationships (SAR) of ring substituents in biological
studies can be a tedious process involving multistep synth-
esis. A more expedient disconnection would involve a late-
stage direct coupling of the ring and the desired carbin-
amine substituent. This would reduce the number of syn-
thetic steps needed to perform proper SAR, thus filling a
real need. However, no general method exists.
1
With the aim of developing a late-stage coupling, we
began to consider how we might access the azole carbin-
amine motif via this disconnection. We hoped to use
abundant tertiary amines which we reasoned could be
†
Authors contributed equally.
(
1) (a) Pe n~ a, S.; Scarone, L.; Manta, E.; Serra, G. Chem. Heterocycl.
3
Compd. 2011, 47, 703–709. (b) Walsh, C. T.; Acker, M. G.; Bowers, A. A.
J. Biol. Chem. 2010, 285, 27525–27531. (c) You, S.-L.; Razavi, H.; Kelly,
J. W. Angew. Chem., Int. Ed. 2003, 42, 83–85.
activated oxidatively. Recently, a number of photoredox
methods have demonstrated R-amino CꢀH functionaliza-
4
5
(
2) (a) Banala, S.; Ensle, P.; S u€ ssmuth, R. D. Angew. Chem., Int. Ed. 2013,
2, 9518–9523. (b) For a list of suppliers for this compound see, http://www.
chemicalbook.com/ProductChemicalPropertiesCB21120099_EN.htm.
c) Kangasmetsa, J.; Hiscock, S.; Johnson, T.; Samuelsson, B.; Tozer, M.;
tions of both isoquinoline and aniline derivatives (eqs 1
and 2, Scheme 1). Unfortunately, the inherent limitation of
these methods, which is that you must desire an isoquino-
line or N-aryl group in your molecule, prevented us from
using them to access our desired motif. Given the numer-
ous methods involving isoquinoline and aniline derivatives
5
(
Grabowska, U. Preparation of amino acid 3-oxohexahydropyrrolo[3,2-
b]pyrrol-4-yl amides as cysteine protease inhibitors. PCT Int. Appl.
WO2007006714A1, 2007. (d) Haupt, A.; Emling, F.; Romerdahl, C. Preparation
of linear dolastatin peptides as antitumor agents. U.S. Patent US5831002A, 1998.
(
e) Zhu, G.-D.; Gandhi, V. B.; Gong, J.; Thomas, S.; Luo, Y.; Liu, X.; Shi, Y.;
Klinghofer, V.; Johnson, E. F.; Frost, D.; Donawho, C.; Jarvis, K.; Bouska, J.;
Marsh, K. C.; Rosenberg, S. H.; Giranda, V. L.; Penning, T. D. Bioorg. Med.
Chem. Lett. 2008, 18, 3955–3958. (f) Duggan, M. E.; Meissner, R. S.;
Perkins, J. J. Naphthyridine derivatives of pyrrolidinylpropionic acid
and analogs useful as integrin receptor antagonists. Patent Appl.
WO9930709A1, 1999, CAPLUS AN 1999:404839. (g) Pioch, R. P. 2,4-
Disubstituted thiazole derivatives. Patent Appl. EP49618A1, 1982, CAPLUS
AN 1982:472354.
(3) There is significant overlap in the photochemical and copper-
mediated processes; for an example and review of copper-mediated
oxidative methods, see: (a) Deb, M. L.; Dey, S. S.; Bento, I.; Barros,
M. T.; Maycock, C. D. Angew. Chem., Int. Ed. 2013, 52, 9791–9795. (b)
Murahashi, S.-I. Angew. Chem., Int. Ed. Engl. 1995, 34, 2443–2465. (c)
Murahashi, S.-I.; Zhang, D. Chem. Soc. Rev. 2008, 37, 1490–1501. (d)
Zhang, C.; Tang, C.; Jiao, N. Chem. Soc. Rev. 2012, 41, 3464–3484.
1
0.1021/ol402751j r XXXX American Chemical Society

Scheme 1. Previous and Current Work
conversion using only 0.75 mol % of catalyst. In all cases,
the major product was the coupled product 3a arising from
coupling with the carbon of the methylene rather than the
0
8
methine carbon (3a ). The photocatalyst was a necessary
component of the reaction, and furthermore, popular
Ru(bpy) Cl did not facilitate the reaction (entry 4). While
initial results indicated (entry 5) that rigorous exclusion of
air was not essential, we found this to be more of the
exception and not universally true. While most reactions
still worked in air, reaction with some amines gave sub-
stantial byproducts. Therefore, we opted to run all of the
substrates under an atmosphere of argon.
3
2
Figure 1. Some examples of natural products, drugs, and pesti-
cides that contain an azole with a carbinamine side chain.
2
and the frequent use of simple aliphatic amines within
photoredox catalysis, it was surprising to find that the
CꢀH functionalization of simpler and far more abundant
5
c,6
aliphatic amines had not been developed.
Thus, we
In an effort to improve the reaction conditions, we
sought to minimize the amount of amine needed. We
envisioned that at least 1 equiv was simply serving as a
base and might be replaced with a convenient alternative.
Indeed, we found that while the reaction would stall out
without the extra equivalents of amine (entry 6), simply
replacing the extra equivalents with pyridine, imidazole,
or carbonates worked well (entries 7ꢀ10) while DBU
decided to investigate whether simple aliphatic amines
and heteroaryl chlorides could be cross-coupled via photo-
7
redox catalysis (eq 3). Herein we report the use of tertiary
aliphatic amines as coupling partners in the photocatalytic
cross-coupling of 2-chlorobenzimidazoles, -benzoxazoles,
-
benzothiazoles, and -thiazoles and related derivatives.
We began our study with 2-chlorobenzothiazole, diiso-
propyl ethyl amine (H u€ nig’s base,) and tris-fac-Ir(ppy) as
3
(entry 11) gave decreased conversion and increased amounts
the photocatalyst (Table 1, entry 1). Initial solvent screen-
ing (entry 2) revealed that dichloromethane, toluene,
tetrahydrofuran, and nitromethane gave trace conversion
at best, while several polar solvents (entry 3) facilitated the
reaction with acetonitrile being optimal and giving smooth
of side products. Again, in an attempt to simplify the
reaction procedure, we looked at the effect of water
(entry 12). We were pleased to find water had little effect
on the reaction, implying that rigorous drying of reagents
and solvents is not necessary for a successful reaction.
Finally, we investigated temperature as a control ele-
ment. We found that the rate of reaction was dependent on
temperature up to a point (entries 13ꢀ15), above which the
conversion dropped severely and was accompanied by a
changein the color of the reaction solution (it turnedblack,
entry 15). However, the regioisomeric ratio (rr) proved to be
relatively insensitive to temperature and thus allowing tem-
perature to be used as a handle for sluggish reactions. Thus,
we ultimately settled on running the reaction under argon, at
45 °C, with 1.2 equiv of amine and 2.0 equiv of imidazole.
With ideal conditions in hand, we next attempted to
explore the scope (Scheme 2). Gratifyingly, we saw that the
(
4) (a) Zou, Y.-Q.; Lu, L.-Q.; Fu, L.; Chang, N.-J.; Rong, J.; Chen,
J.-R.; Xiao, W.-J. Angew. Chem., Int. Ed. 2011, 50, 7171–7175. (b) Liu,
Q.; Li, Y.-N.; Zhang, H.-H.; Chen, B.; Tung, C.-H.; Wu, L.-Z. Chem.;
Eur. J. 2012, 18, 620–627. (c) Ruiz Espelt, L.; Wiensch, E. M.; Yoon,
T. P. J. Org. Chem. 2013, 78, 4107–4114. (d) Freeman, D. B.; Furst, L.;
Condie, A. G.; Stephenson, C. R. J. Org. Lett. 2012, 14, 94–97. (e) Hari,
D. P.; Koenig, B. Org. Lett. 2011, 13, 3852–3855. (f) Kohls, P.; Jadhav,
D.; Pandey, G.; Reiser, O. Org. Lett. 2012, 14, 672–675.
(5) (a) Ju, X.; Li, D.; Li, W.; Yu, W.; Bian, F. Adv. Synth. Catal. 2012,
354, 3561–3567. (b) Zhu, S.; Das, A.; Bui, L.; Zhou, H.; Curran, D. P.;
Rueping, M. J. Am. Chem. Soc. 2013, 135, 1823–1829. (c) McNally, A.;
Prier, C. K.; MacMillan, D. W. C. Science 2011, 334, 1114–1117. (d) Shi,
L.; Xia, W. Chem. Soc. Rev. 2012, 41, 7687–7697.
(6) Ohashia has shown that direct UV photolysis can lead to forma-
tion of the coupled product, and later MacMillan (see ref 5c) showed that
under photocatalysis a single example of a tertiary aliphatic amine
which, contrary to what we have observed, was highly endoselective.
Ohashi, M.; Miyake, K.; Tsujimoto, K. Bull. Chem. Soc. Jpn. 1980, 53,
1
683–1688.
(8) This is not unprecedented; for examples, see: (a) Kotani, S.;
Osakama, K.; Sugiura, M.; Nakajima, M. Org. Lett. 2011, 13, 3968–
3971. (b) Schreiber, S. L. Tetrahedron Lett. 1980, 21, 1027–1030.
(
7) Vachhani, D. D.; Sharma, A.; Van der Eycken, E. Angew. Chem.,
Int. Ed. 2013, 52, 2547–2550.
B
Org. Lett., Vol. XX, No. XX, XXXX

benzothiazole 3a to functionalized benzothiazoles (3c, 3d,
e) suggests that substitution at the 6-position results in
3
Table 1. Identification of the Optimal Conditions
increased selectivity. On the other hand, simplification of
benzothiazole (3a) to simple thiazole (3v) results in dimin-
ished selectivity (11:1 vs 2:1). However, additional sub-
stitution of the thiazole causes the selectivity to return to
more desirable levels. Substitution with an ester or a cyano
group increases the regioselectivity from 2:1 to 4:1 (3v vs 3j
and 3l), while substitution with a chloride or phenyl ring
results in even more dramatic increases from 2:1 to 9:1 (3v
vs 3f and 3i). However, coupling of phenyl-substituted
thiazole 3i proved to be sluggish, giving incomplete con-
version and an isolated yield of only 24%.
GC/
MS
GC/MS
a
0
a
entry modification of experimental conditions conversion 3a/3a
1
2
3
4
5
6
7
8
9
none, 13 h
DCM, CH
DMF, DMSO, or DMA
no photocatalyst or Ru(bpy)
MeCN exposed to air
67%
9:1
b
3
NO
2
, Tol or THF
<6%
29ꢀ44%
0%
57%
21%
nd
nd
nd
b
b
The reaction also worked well with less hindered linear
amines such as tributyl amine (3n, 3o). However, an
increase in the amount of the amine was used to prevent
a second arylation of the product as the reaction neared
completion. Pyrrolidylaminesalsoprovedtobecompetent
coupling partners (3g, 3m). We observed excellent selectiv-
ity (> 20:1 rr) for the acyclic methylene over the cyclic pair
of methylenes. This selectivity is particularly interesting
3
CI
2
9:1
9:1
9:1
13:1
9:1
9:1
9:1
9:1
9:1
12:1
8:1
1.2 equiv of amine at 24 h
2.0 equiv of pyridine and 1.2 equiv of amine 63%
2.0 equiv of imidazole and 1.2 equiv of amine 68%
2.0 equiv of Cs
2
CO
CO
3
and 1.2 equiv of amine
and 1.2 equiv of amine
69%
59%
32%
65%
46%
62%
22%
10 2.0 equiv o f K
2
3
1
1
2.0 equiv of DBU and 1.2 equiv of amine
1
1
1
1
2 up to 15% v/v H
2
O
3 reaction run at 27 °C, 7 h
4 reaction run at 45 °C, 7 h
5 reaction run at 65 °C, 7 h
1
0
since the vast majority of examples of N-oxidation
initiated functionalizations as well as photochemical
1
1
couplings in the literature are highly selective for the
endocyclic position. This finding suggests a different con-
trolling element in this coupling reaction when compared
to these processes which seem related. The reaction also
works well using amines with three different groups (3t)
and highlights the advantage of this coupling method
which provides facile access to a product that would be
difficult to access via other methods. Finally, while sub-
strate 3p derived from a methyl amine offers less complex-
ity than other substrates, it shows that the reaction also
works with simpler methyl amines giving the product as a
single regioisomer.
a
b
Determined by integration of GCMS trace. Not determined.
reaction worked well for benzothiazole (3a, 92%, rr 11:1)
and a number of substituted benzothiazole substrates in-
cluding those substituted with a bromide (3c) and a
chloride (3d), which provides opportunity for further
elaboration of the products via more traditional cross-
coupling methodologies. The fact that neither 6-Br nor
6
-Cl reacted suggests the reaction is highly chemo- and
regioselective for the 2-Cl position. Survival of the aryl
bromide under these conditions is somewhat surprising
considering that reduction of electron-deficient aryl
We suspected that our method would be ideal for rapidly
assessing the SAR of heterocycles substituted with an ali-
phatic carbinamine. Take for instance the drug Nizatidine,
which is an H-2 histamine agonist developed by Eli
9
bromides has been observed under similar conditions.
The reaction also works well for more electron-rich ben-
zothiazole 3e, 61% 13:1 rr, albeit dilution was found to be
necessary to achieve reasonable reaction rates. Impor-
tantly, exploration of other related heterocycles revealed
that the method was not limited to benzothiazole but could
be applied to benzoxazoles (3b, 3h, 3n, 3p, 3s, 3u), Boc-
protected benzimidazole (3k), as well as simple thiazole
derivatives (3f, 3i, 3j, 3l, 3v). In all the heterocycles studied,
similar regioselectivity was observed in which coupling of
the less hindered CꢀH dominated.
2
g,12
Lilly
and sold under the trade names Taxzad and Axid
and was one of the most widely used drugs for a number of
years in combating health issues related to the overproduc-
tion of stomach acid. Inthe development of this drug, SAR
of the carbinamine was determined by variation of the
amine in the acetonitrile component in the initial step of
the reaction sequence (Scheme 3, eq 1). With regard to the
carbinamine side group, Eli Lilly only looked at five
Gratifyingly, the reaction proved tolerant of a number
of functional groups, including a cyano group (3l), ester
3j), halogens (3c, 3d, 3f, 3h, 3n, 3p, 3s, 3u), and methyl
ethers (3e) as well as terminal olefins (3r). Finally, Boc
groups (3k) survived, albeit loss of the Boc group was
initially problematic but was easily circumvented by repla-
cing the imidazole with Cs CO .
The regioselectivity appears to be sensitive, in part, to
further substitution of the aryl chloride. Comparison of
(10) (a) Chen, C. K.;Hortmann, A. G.;Marzabadi, M. R. J. Am. Chem.
Soc. 1988, 110, 4829–4831. (b) Sud, A.; Sureshkumar, D.; Klussmann, M.
Chem. Commun. 2009, 3169–3171. (c) Jovel, I.; Prateeptongkum, S.;
Jackstell, R.; Vogl, N.; Weckbecker, C.; Beller, M. Chem. Commun. 2010,
(
4
6, 1956–1958. (d) Koehler, V.; Bailey, K. R.; Znabet, A.; Raftery, J.;
Helliwell, M.; Turner, N. J. Angew. Chem., Int. Ed. 2010, 49, 2182–2184.
11) (a) Marinkovic, S.; Hoffmann, N. Chem. Commun. 2001, 2001,
(
1576–1578. (b) Yamada, S.; Nakagawa, Y.; Watabiki, O.; Suzuki, S.;
Ohashi, M. Chem. Lett. 1986, 361–364. (c) Barta, M.; Hampl, F.; Liska,
F.; Dedek, V. Collect. Czech. Chem. Commun. 1994, 59, 1820–1832. (d)
Bertrand, S.; Hoffmann, N.; Pete, J.-P. Eur. J. Org. Chem. 2000, 2227–
2
3
2
238.
0
(
12) Pioch, R. P. N-Thiazolylmethylthioalkyl-N -alkylamidines
(
9) Nguyen, J. D.; D’Amato, E. M.; Narayanam, J. M. R.; Stephenson,
and related compounds. Patent Appl. US4382090A, CAPLUS AN
1983:443548(Patent).
C. R. J. Nat. Chem. 2012, 4, 854–859.
Org. Lett., Vol. XX, No. XX, XXXX
C

Scheme 2. Cross-Coupling of Heteroaryl Chlorides with
Aliphatic Amines
Scheme 3. Rapid Synthesis of Previously Unknown Nizatidine
Analogue
by HBr-mediated thiol substitution and finally coupling
with the active pharmacophore to afford 135 mg, 48%
1
4
yield (4a) of the desired analogue as a single regioisomer
with only one purification after the final step. Thus, we
were able to readily access a previously unknown analogue
of Nizatidine differing at the carbinamine side chain from
commercially available components in fewer steps and with
greater structural diversity than previously investigated.
In summary, we highlight the enabling nature of this
reaction as well as the conceptual advance it represents.
First, it provides an alternative disconnection to tradi-
tional cyclodehydration which allows faster and greater
carbinamine diversification. The reaction setup is simple
and utilizes commercially available reagents. It has proven
to be functional-group-tolerant and chemoselective. Sec-
ond, this work highlights the ability of photoredox cata-
lysis to mediate CꢀC bond formation with simple tertiary
aliphatic amines rather than isoquinoline or aryl amine
derivatives. Our future efforts will attempt to under-
a
b
Yields are of isolated products after chromatography. Regioiso-
1
meric ratios were determined on crude material via H NMR. Imidazole
was replaced with 4.0 equiv (total) of amine. Cs CO was used instead of
2 3
1
imidazole. No other possible regioisomer was detected by H NMR.
c
d
e
1
5
stand the mechanism and expand the scope of the
2-chloroazole.
variations of the nitrogen substituent and no substitution
g,12
of the carbinamine methylene was explored.
2
Given the
minimal SAR performed on the carbinamine component,
we thought this would be an ideal opportunity to demon-
strate how we could use our method to rapidly make a
Nizatidine analogue (4a).
Acknowledgment. This work was supported by startup
funds from Oklahoma State University. We are grateful
to J.A.M., J.A.E., and R.J.R. for reading and providing
feedback for this manuscript.
Beginning with commercially available 2-chlorothiazole
carboxylate, we coupled H u€ nig’s base to afford 3j (from
Scheme 2). Taking 200 mg of 3j forward (Scheme 3, eq 2),
Supporting Information Available. Experimental pro-
cedures, analytical data, and further discussion. This
material is available free of charge via the Internet at
http://pubs.acs.org.
1
2,13
we followed the procedure of Eli Lilly.
jected our coupled product to a telescoped synthesis
First, we sub-
beginning with reduction of the ester with LiBH followed
4
(15) For a discussion of the potential mechanism, please see the
Supporting Information.
(
13) Reddy, G. N. C.; Naram, J.; Bandichhor, R.; Reddy, P. P.;
Dubey, P. K.; Bhalerao, D. S. Chem. Biol. Interface 2011, 1, 123–127.
14) The minor regioisomer was removed via chromatography after
(
the inital coupling.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX