A new method using 1,3,5-triazine as an umpolung hydrogen cyanide equivalent toward the syntheses of isoquinolinone and 2-pyridone derivatives

Article abstract of DOI:10.1016/j.tetlet.2018.09.029

An investigation of Hermecz and Hartenstein's cyclization methods was conducted. Overcoming the limited substrate scope allowed the expansion of the synthetic application of 1,3,5-triazine as an umpolung hydrogen cyanide equivalent. The reaction proceeded under mild conditions to provide various isoquinolinone and 2-pyridone derivatives with excellent yields.

Full text of DOI:10.1016/j.tetlet.2018.09.029

Accepted Manuscript
A new method using 1,3,5-triazine as an umpolung hydrogen cyanide equivalent
toward the syntheses of isoquinolinone and 2-pyridone derivatives
Joji Hayashida, Shinya Yoshida
PII:
S0040-4039(18)31110-9
https://doi.org/10.1016/j.tetlet.2018.09.029
TETL 50268
DOI:
Reference:
To appear in:
Tetrahedron Letters
Received Date:
Revised Date:
Accepted Date:
1 August 2018
6 September 2018
11 September 2018
Please cite this article as: Hayashida, J., Yoshida, S., A new method using 1,3,5-triazine as an umpolung hydrogen
cyanide equivalent toward the syntheses of isoquinolinone and 2-pyridone derivatives, Tetrahedron Letters (2018),
doi: https://doi.org/10.1016/j.tetlet.2018.09.029
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A new method using 1,3,5-triazine as an umpolung hydrogen cyanide
equivalent toward the syntheses of isoquinolinone and 2-pyridone
derivatives.
Joji Hayashida* and Shinya Yoshida
Process Chemistry Labs, Astellas Pharma Inc., 160-2 Akahama, Takahagi, Ibaraki 318-0001,
Japan
*
Corresponding authors:
Joji Hayashida: joji.hayashida@astellas.com
1

Abstract
An investigation of Hermecz and Hartenstein’s cyclization methods was conducted.
Overcoming the limited substrate scope allowed the expansion of the synthetic application of
,3,5-triazine as an umpolung hydrogen cyanide equivalent. The reaction proceeded under
1
mild conditions to provide various isoquinolinone and 2-pyridone derivatives with excellent
yields.
Graphical Abstract
2

Keywords
triazine
cyclization
umpolung
isoquinolinone
2
-pyridone
3

1
,3,5-Triazines are widely known as common building blocks in various functional materials
1
such as melamine resins, dyes, and bioactive compounds. As synthetic tools, triazine
2
derivatives are often utilized in halogenation and peptide coupling reactions. For instance,
Kunishima’s 4-(4,6-Dimethoxy-1,3,5-triazine-2-yl)-4-methylmorpholinium chloride
(
DMTMM) is recognized as an effective and durable coupling reagent in aqueous media.3
However, most of these reactions are categorized as either complete or no carbon frame
incorporation. An example of partial structural transformation is the case of Gattermann-Koch
2
,4
formylation : 1,3,5-triazine (1) is used as a -CH=O source under acidic conditions. Although
triazine 1 is also considered as a trimer of hydrogen cyanide, its application as a –CH=N- unit
is limited. HCN tends to form a cyanide anion and act as a nucleophile due to a low pKa of
approximately 9. Therefore, if triazine 1 behaves as an umpolung hydrogen cyanide synthetic
equivalent, it would be an interesting reagent for the community.
During our course of the study on ASP3700 (4), an osteoarthritis drug discovered by
5
Astellas Pharma Inc., we encountered an opportunity to explore an effective and robust
method to synthesize the key intermediate 3 (Figure 1). As described in our previous paper,
the use of 1,3,5-triazine and NaOMe with diesters 2 allowed us to achieve the sequential
transesterification, olefin isomerization, 2-pyridone ring formation, followed by hydrolysis to
afford the methanoisoquinoline 3 in 63.3% yield without silica gel column chromatography.6
Here, we describe our investigation on this new cyclization strategy in comparison with
4

7
8
related reports by Hermecz and Hartenstein and its application to various isoquinolinones
and 2-pyridones syntheses. In addition, our novel approach toward the amino-isoquinoline
derivatives is discussed.
Figure 1. 1,3,5-triazine promoted cyclization for the synthesis of the key intermediate of
ASP3700 (4).
In 1980, Hermecz et al. discovered that the reaction of ethyl acetoacetate with 1 and NaOEt
in EtOH afforded 11.8% of the pyrimidopyridinone as a major byproduct along with the
desired 4-pyridone. Later, the reaction was applied to the synthesis of naphthyridinones from
7
2
,6-dimethyl pyridines-3,5-dicarboxylate derivatives (Figure 2). In their reports, the reaction
proceeded smoothly to give various aza-isoquinolinones in 21-97% yield. However, the
substrate scope was quite limited at the 4-position. On the other hand, Hartenstein et al.
8
showed that Hantzsch esters and their derivatives showed similar reactivity. Although a
stronger base and high temperature (NaH/DMF/100C) were required, a number of 2- and 4-
5

substituted cyclic products were obtained in 11-85% yield. Curiously, both research groups
utilized rather sophisticated substrates but not simpler starting materials such as methyl 2-
methylbenzoate (5) or methyl 2-methylnicotinate (6). Therefore, we began investigating the
cyclization with these compounds.
Figure 2. Hermecz and Hartenstein’s previous works
Following the Hermercz report, ester 5 was treated with alkoxides in alcohol (Table 1, Entry
1
, condition A and B). Surprisingly, the reaction did not proceed at all. Next, we tried the
much harsher Hartenstein’s method, NaH/DMF/100C, as well as NaHMDS/THF (condition
C and D). However, no desired product was observed. Based on their reports, sodium cation
coordination to the nitrogen lone pair seemed to play an important role in the deprotonation
stage. Therefore, we switched the substrate to 2-methylnicotinate 6. Unfortunately, the
6

reaction still did not provide the target compound (Entry 2). Furthermore, to increase the
acidity and stabilize the generated anion at the benzylic position, the compounds bearing nitro
7
and phenyl 8 groups were tested (Entries 3 and 4). Nevertheless, the desired cyclization did
not take place. These results indicate that the key for this reaction is balancing the acidity to
form an anion intermediate and sufficient nucleophilicity to attack triazine 1. Therefore, along
6
with our success in ASP3700 synthesis, CO
2
Me attached ester 9 was investigated. To our
delight, the reaction proceeded smoothly to afford the desired isoquinolinone 10 in
quantitative yield.
Table 1. Hermecz and Hartenstein’s method with simple substrates.
Entry
1
Substrate
Conditions
Yield %
condition A: 1(5.0 equiv.), NaOMe(10.0 equiv.), MeOH, 24 h, reflux
condition B: 1(5.0 equiv.), NaOEt(10.0 equiv.), EtOH, 24 h, reflux
condition C: 1(5.0 equiv.), NaH(5.0 equiv.), DMF, 24 h, 110C
condition D: 1(5.0 equiv.), NaHMDS(1.3 equiv.), THF, 24 h, RT
No reaction
No cycloadduct
Complex mixture
Complex mixture
2
3
4
condition B
condition C
No cycloadduct
Complex mixture
condition A
condition A
Complex mixture
No reaction
5
condition E: 1(5.0 equiv.), NaOMe(10.0 equiv.), MeOH, 1 h, RT
Quantitative
7

Next, the optimization of the reaction was conducted (Table 2). The use of triazine 1 was
able to be reduced to 1.2 equivalents (Entries 1 and 2). Interestingly, despite that one
molecule of triazine 1 could provide three -HC=N- sources, the reaction with 0.4 equivalent
formed only 34.8% of the cycloadduct (Entry 3). As described in the reaction mechanism for
6
our ASP3700 synthesis , the sequential addition to triazine 1 followed by amide bond
formation is the key for this transformation. This result implied that the potential formamidine
source is not the -HC=N- donor. Next, base selection showed that inorganic salts such as
Na
2
CO
3
, K
2
CO and NaOAc can also lead to the desired cyclization but with insufficient yield
3
(
Entries 4-6). These poor results could be caused by the lower solubility of these inorganic
bases. Although using an extended reaction time, a strong organic base such as DBU
promoted the reaction as effectively as NaOMe with 99.3% yield (Entry 8). Solvent screening
revealed that aprotic solvents such as MeCN and toluene are also applicable in this reaction
(
Entries 9-14). Not surprisingly, proper tuning of the base and solvent seemed to be vital for
the reaction behavior (Entry 11 vs. 14). Reducing the amount of base to 5 equivalents affected
the reaction efficiency less so using NaOMe/MeOH than the DBU/MeOH system (Entry 15
vs 16). Further sodium methoxide reduction showed a significant yield drop of 83.6% (Entry
1
7). This phenomenon indicates that the quick elimination of the formamidine unit in
6
intermediates IV described in our proposed mechanism seemed a crucial factor for their
8

reaction presumably due to the instability. Considering reaction time and robustness, the
reaction condition in entry 15 was selected as our primary condition.
9

Table 2. Optimization of the cyclization of diester 9.
Entry 1 amount
Base
Solvent
(0.2 M)
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeCN
TBME
DMF
Time
Temp.C
Yield %a
(
equiv.)
(10 equiv.)
NaOMe
NaOMe
NaOMe
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
5.0
1.2
0.4
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1 h
RT
Quant.
99%
35%
47%b
46%b
15%b
15%b
99%
81%
77%
35%
80%
92%
90%
98%
77%
84%
1 h
RT
24 h
24 h
24 h
24 h
24 h
24 h
9 h
RT
Na
2
CO
3
50C
50C
50C
60C
RT
K
2
CO
3
NaOAc
DIPEA
DBU
NaOMe
NaOMe
NaOMe
NaOMe
DBU
RT
0
7 h
RT
1
3 h
RT
2
Toluene
TBME
DMF
3 h
RT
3
24 h
7 h
RT
4
DBU
RT
5c
6c
7d
NaOMe
DBU
MeOH
MeOH
MeOH
1 h
RT
20 h
2 h
RT
NaOMe
RT
a
b
c
d
HPLC yield.
The incomplete reactions after 24 hours.
5
2
equivalents of base were used.
.5 equivalents of NaOMe were used.
With the optimal reaction condition in hand, we examined the substrate scope in 1,3,5-
triazine mediated isoquinolinone synthesis (Figure 3). The compounds bearing various
functional groups, such as NO , Me, OMe, Cl, and F, converted smoothly to give the
2
corresponding isoquinolinones in excellent yields. (83.8-98.4%). Particularly, in the
pharmaceutical industry, fluoride attached isoquinolinone 15 seemed to be useful for future
drug discovery. These results indicated the reaction is feasible with both electron-withdrawing
1
0

and electron-donating groups. Interestingly, methyl group attached diesters afforded the
desired product 12 in 69.5% yield under NaOMe/MeOH conditions. On the other hand, the
DBU/MeOH system showed a much better result of 84.4%. With these favorable results on
diester derivatives, we assumed other alpha carbon activating functionalities such as ketone,
cyanide, sulfone, and phosphorus ester would also facilitate the cyclization. As expected,
benzoyl, cyano, and benzenesulfonyl, which substituted isoquinolinones 16-18, were also
9
obtained in excellent yields. Based on these results, activating the benzylic position to
enhance nucleophilicity with proper functional groups seems to be a more crucial factor than
the electron density in the aromatic ring.
1
1

Figure 3. The substrate scope.
Next, the reaction was applied to aliphatic substrates (Scheme 1). The compounds
containing 5-membered and 6-membered rings proceeded smoothly to afford the desired 2-
pyridones in 75.3-87.1% yield. Furthermore, the linear substrate 24 also gave 25 with 58.0%
1
0
yield , although the reaction efficacy dropped presumably due to the loss of the Thorpe-
1
1
Ingold effect. Overall, our 1,3,5-triazine mediated cyclization well suits 2-pyridone
synthesis, as well.
1
2

Scheme 1. The reaction with aliphatic substrates.
Finally, we attempted to switch the ester moiety in the substrate for other functional groups.
6
In principle, the proposed anionic intermediate could be trapped with various intramolecular
1
2
electrophiles. Inspired by classical Kreutzberger’s aminoquinazoline synthesis, 2-
cyanophenylacetonitrile (26) was treated with triazine 1 and NaOMe in MeOH. (Scheme 2).
To our delight, the cyclization proceeded smoothly to give a desired amino-isoquinoline 27 of
8
1.6%. Moreover, compound 28 bearing a benzenesulfonyl group also underwent cyclization
with excellent yield. These preliminary results imply that our 1,3,5-triazine mediated
cyclization strategy would be applicable for various types of heteroaromatic compound
syntheses by switching intramolecular anion acceptors.
1
3

Scheme 2. Application to amino-isoquinoline synthesis.
In conclusion, we have investigated the Hermecz and Hartenstein’s reports on the use of
1
,3,5-triazine as an umpolung hydrogen cyanide equivalent. Overcoming the limited substrate
scope allowed the discovery of a new method for synthesizing various isoquinolinone and 2-
pyridone derivatives in an efficient manner. Finally, converting the intramolecular
electrophile to nitrile showed further potential applications toward the syntheses of various
heteroaromatic compounds.
1
4

Funding
This research did not receive any specific grant from a funding agency in the public,
commercial, or not-for-profit sectors.
Declaration of interest
Conflicts of interest: none
Acknowledgements
The authors thank Kohei Miyamoto and Hana Kaneumi for analytical support and Atsushi
Ohigashi and Minoru Okada for project management advice.
References and notes
1
. For general reviews for trizaines, see: a) Neunhoeffer, H.; Wiley, P. F. The Chemistry of
Heterocyclic Compounds, Chemistry of 1,2,3-Triazines and 1,2,4-Triazines, Tetrazines,
and Pentazin, John Wiley & Sons, 2009. b) Blotny, G. Tetrahedron 2006, 62, 9507-9522.
. For a general review of 1,3,5-trizaine in organic synthesis, see: Aksenov, A.V.;
Aksenova, I.V. Chem. Heterocycl. Compd. 2009, 45, 130-150.
2
3
. a) Kunishima, M.; Kawachi, C.; Iwasaki, F.; Terao, K.; Tani, S. Tetrahedron Lett. 1999,
4
0, 5327–5330. b) Kunishima, M.; Kawachi, C.; Morita, J.; Terao, K.; Iwasaki,F.; Tani,
S. Tetrahedron 1999, 55, 13159–13170. c) Kunishima, M.; Morita, J.; Kawachi, C.;
Iwasaki, F.; Terao, K.; Tani, S. Synlett 1999, 1255–1256. Kunishima’s other triazine
based reagents: d) Yamada, K.; Igarashi, Y.; Betsuyaku, T.; Kitamura, M.; Hirata, K.;
Hioki, K.; Kunishima, M. Org. Lett. 2018, 20, 2015-2019. e) Yamada, K.; Tsukada, Y.;
Karuo, Y.; Kitamura, M.; Kunishima, M. Chem. Eur. J. 2014, 20, 12274–12278. f)
Yamada, K.; Fujita, H.; Kunishima, M. Org. Lett. 2012, 14, 5026-5029.
4
. a) Gattermann, L.; Koch, J. A. Ber. 1897, 30, 1622-1624. b) Kreutzberger, A., Angew.
1
5

Chem. 1967, 79, 978.
5
. a) Matsushima, Y.; Kameda, M.; Honda, S.; Sugano, Y.; Usuda, H.; Terasaka, T.;
Terasawa, T.; Nishigaki, F., “Preparation of condensed ring pyridine compounds as
agonists of cannabinoid receptor type 2 (CB2 receptor)” WO 2012014963 A1, Feb.02,
2
012. b) Matsushima, Y.; Kameda, M.; Honda, S.; Sugano, Y.; Usuda, H.; Terasaka, T.;
Terasawa, T.; Nishigaki, F., “Preparation of fused ring pyridine compounds as agonists of
cannabinoid receptor for treating inflammatory diseases and pain” US 20130143874 A1,
June 06, 2013.
6
7
. For proposed reaction mechanism, please see in supporting information or Oka, H.;
Hayashida, J.; Yoshida, S.; Kido, Y.; Ono, H.; Sugimori, T.; Matsushima, Y. Chem Lett.
2
018, in preparation. We examined the cyclization conditions with formaldehyde and also
formamidine sources. Unfortunately, no desired product was observed.
. a) Balogh, M.; Hermecz,I.; Mészáros, Z.; Simon, K.; Pusztay, L.; Horváth, G.; Dvortsák,
P. J. Heterocycl. Chem.1980, 17, 359. b) Balogh, M.; Hermecz,I.; Naray-Szabo, G.;
Simon, K., and Z. Meszaros, J. Chem. Soc., Perkin Trans. 1 1986, 753.c) Balogh, M.;
Hermecz, I.; Simon, K.; Pusztay, L. J. Heterocycl. Chem.1989, 26, 1755.
8
9
. Kleinschroth, J.; Mannhardt, K.; Hartenstein, J.; Satzinger, G. Synthesis 1986, 859.
. Phosphorus ester attached substrate was also found to be applicable for the cyclization.
Due to the purification issue in the starting material, this example was not listed in figure
3
.
1
1
1
0. The yield was determined by HPLC.
1. Beesley, R. M.; Ingold, C. K.; Thorpe, J. F. J. Chem. Soc. Trans. 1915, 107, 1080.
2. Kreutzberger, A.; Stevens, M. F. G. J. Chem. Soc. (C), 1969, 1282.
1
6

Highlight



An efficient and simple approach to access various isoquinolinone and 2-pyridones
derivatives.
A useful application of 1,3,5-trizaine as an umpolung hydrogen cyanide synthetic
equivalent.
A novel cyclization for the synthesis of amino-isoquinoline derivatives.
1
7