An efficient one-pot synthesis of tethered cyclohexadiene enaminonitriles from methyl-ketones: An effective route to quinazolines

Article abstract of DOI:10.1016/j.bmcl.2007.10.048

The quinazolines represent a useful natural product scaffold with demonstrated activities against disorders such as high blood pressure and benign prostatic hyperplasia. Here we report on the synthesis and biological activity of a series of quinazolines t

Full text of DOI:10.1016/j.bmcl.2007.10.048

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 3126–3130
An efficient one-pot synthesis of tethered cyclohexadiene
enaminonitriles from methyl-ketones:
An effective route to quinazolines
Bennie J. Bench, Victor H. Suarez and Coran M. H. Watanabe*
Department of Chemistry, Texas A&M University, College Station, TX 77843, USA
Received 1 September 2007; revised 12 October 2007; accepted 15 October 2007
Available online 18 October 2007
Abstract—The quinazolines represent a useful natural product scaffold with demonstrated activities against disorders such as high
blood pressure and benign prostatic hyperplasia. Here we report on the synthesis and biological activity of a series of quinazolines
that were prepared by a one-pot synthesis of substituted cyclohexadiene enaminonitriles from methyl-ketones. The approach, which
employs NaH, complements published procedures where LDA is utilized. While the NaH catalyzed reaction generates the cyclo-
hexadiene enaminonitriles in high yields with heterocyclic substrates, the reaction fails to promote product formation of aliphatic
alkyl substrates. On the contrary, the LDA mediated synthesis favors the long chain alkyl substituents while reactions involving the
aromatic substrates result in low yields. The final conversion to the quinazolines is also a modification on literature protocols.
In cellular assays, the quinazolines showed the most promising activity against Jurkat with CC50 values in the low micromolar
range. Weak activity was observed against microbial strains (Bacillus subtilis, Escherichia coli, and Saccharomyces cerevisiae). The
substituted enaminonitrile intermediates also exhibited weak anti-microbial activity and cytotoxicity against human T-cell
leukemia.
Ó 2007 Elsevier Ltd. All rights reserved.
Heterocyclic systems have achieved prominence as phar-
maceuticals, fungicides, and sovatochromatic dyes.1–4
The quinazolines (Fig. 1), for example, have been imple-
mented to treat high blood pressure (Doxazosin 1 and
Prazosin 2) and benign prostatic hyperplasia (Alfuzosin
3).5–12 Likewise, enaminonitriles, such as 4 and its ana-
logues 5–7 (Fig. 2), have shown biological activity (4,
18 lM; 5, 35 lM; 6, 19 lM; 7, 16 lM) against scrapie-
infected mouse neuroblastoma cells (ScN2a) inhibiting
PrPSc,13 an isoform of the cell-surface glycosylphosphat-
idylinositol (GPI)-anchored protein PrPc.14
tion of a,b-unsaturated nitriles can be invoked by
treatment with lithium diisopropylamide (LDA)15 or
other strong bases to give the enaminonitriles (Fig. 4)
in moderate yields.16,17 We additionally report on the
biological activity of these compounds including their
cytotoxic effects against bacteria (E. coli and B. subtilis),
yeast, and human T-cell leukemia (Jurkat). Interest-
ingly, members of a functionalized cyclohexadienal li-
brary have shown their ability to stimulate neurite
outgrowth, while others have demonstrated anti-cancer
activity.18,19
In this investigation, we report on the synthesis of qui-
nazolines generated from the a cyclohexadiene enamino-
nitrile scaffold (Fig. 3) that was in turn prepared by a
one-pot strategy involving treatment of methyl-ketones
with excess NaH. We compare this synthetic methodol-
ogy with that reported previously, where self-dimeriza-
Enaminonitriles can be generated in a one-pot format by
implementation of
a Horner–Wadsworth–Emmons
reaction (Fig. 5). When methyl-ketones are treated with
an ylide solution generated by treatment of diethyl
(cyanomethyl)-phosphonate with one equivalent of so-
dium hydride, conversion to the a,b-unsaturated nitrile
is observed.21 However, in the presence of excess sodium
hydride, the reaction proceeds to give the
enaminonitrile.
Keywords: Enaminonitrile; Quinozoline; Horner–Wadsworth–Emmo-
ns; Cell-based assays.
To optimize the synthesis of our enaminonitriles, we uti-
lized 8 as a model substrate, where it was found that six
*
Corresponding author. Tel.: +1 979 458 8094; fax: +1 979 458
8095; e-mail: watanabe@mail.chem.tamu.edu
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.10.048

B. J. Bench et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3126–3130
O
3127
O
O
O
N
N
N
O
N
N
O
O
N
O
O
N
N
NH2
NH2
1
2
O
H
N
H
N
N
O
O
N
O
NH2
3
Figure 1. Quinazolines used as therapeutic agents.
Cl
Cl
Cl
S
NH2
S
N
NH2
CN
S
N
NH2
CN
S
N
NH2
CN
N
NH2
CN
NC
NC
NC
NC
O
O
S
4
5
6
7
Figure 2. Enaminonitriles active against mice ScN2a cells.
NH2
R
R
O
CN
One Pot Synthesis
N
C6H12NO3P
excess NaH
R
R
NH2
R
N
Figure 3. Quinazalines derived from a cyclohexadiene enaminonitrile scaffold.
excess equivalents of NaH were required to facilitate the
deprotonation of the nitrile (b-methyl group, Fig. 6) to
give an 89% yield of the enaminonitrile after 4 days.
When fewer equivalents were used, 0.5, 1.0, 1.25, and
2.0 equiv, no product was obtained while 3.0 and
4.0 equiv gave yields of 7% and 43%, respectively. At
higher concentrations, 8.0 and 10.0 equiv of NaH, 41%
and 10% of product were isolated. Presumably, the high
concentration of base drives the majority of the nitrile to
the deprotonated or b-methylenic form, diminishing
product formation (Fig. 6). Time course analysis also re-
vealed that 4 days was optimal as yields diminished be-
fore and after that point.
With these optimized conditions in hand, various sub-
strates were tested to investigate the scope of the reac-
tion as shown in Table 1. Product yields ranged from
79% to 96% for substrates containing small aromatics
and bulky groups while aliphatic alkyl substrates gave
no product. Thiophene 8 gave a yield of 89%, while
the bromo-substituted thiophene 9 was generated in al-
most quantitative yield, 96%. With phenyl and its substi-
tuted derivatives, bromo-phenyl 11 exhibited the lowest
yield, 79% as compared to 88% for both the phenyl 9
and fluoro-phenyl 10 substituents. Two of the higher
yielding substrates selected were consisting of bulky sub-
strates, biphenyl 12 and naphthalene 13, giving product
yields of 93% and 90%, respectively.
The primary limitation of this one-pot strategy is that it
does not facilitate the self-dimerization of aliphatic alkyl
substrates. When either b-ionone 15 or 2-octone 16 was
employed, we found that no dimer product was seen in
the crude 1H NMR with the only reaction product being
the nitrile intermediate itself. This could be due to reduced
stabilization of the resulting b-methyl anion, requiring a
stronger base to deprotonate the methyl group.
R
D : R =
E : R =
CN
A : R = CH3
B : R = C2H5
C : R = C6H5
R
NH2
Figure 4. Known cyclohexadiene enaminonitriles.20

3128
B. J. Bench et al. / Bioorg. Med. Chem. Lett. 18 (2007) 3126–3130
O
R
O
CN
P
O
CN
1 equiv.
NaH
excess
NaH
O
+
CN
R
R
R
NH2
Figure 5. Horner–Wadsworth–Emmons reaction; 1.1 equiv C6H12NO3P, 1.1 or 7.1 equiv NaH.
Table 1. Substrate effects with excess NaH as compared to LDA
CN
R
R group
Compound
NaH yield
(%)
LDA yield
(%)
R
CN
CN
CN
CN
S
R
R
8
9
89
96
88
88
79
28
24
18
20
20
R
S
Br
R
R
CN
NH2
Figure 6. Proposed mechanism of reaction.
CN
NH
10
11
12
R
R
F
To evaluate the effectiveness of our NaH catalyzed reac-
tion, we analyzed the formation of enaminonitriles from
a,b-unsaturated nitriles using LDA in tetrahydrofuran
(THF) (Fig. 7).15 In this case, the nitrile substrates are
generated from their corresponding methyl-ketones, iso-
lated, and subsequently reacted with LDA, thus produc-
ing the enaminonitriles in two steps. We saw an inverse
correlation with LDA as compared to the NaH reaction
(Table 1). The NaH reaction readily dimerizes aromatics
and bulky substrates in high yields, while LDA gives
minimal yields, presumably reflecting steric effects, the
lack of full dissociation of the catalyst prior to self-con-
densation. Conversely, reactions catalyzed with LDA
seemed to favor long chain alkyl substituents. Previous
reports of LDA dimerization with alkylidene malono-
nitriles including 3-methyl-2-butenenitrile and 3-
methyl-2-petenenitrile gave high to moderate yields of
90% and 60%, respectively.15
Br
13
14
93
90
12
13
15
16
—
—
36
31
R
CN
On the other hand, we demonstrate that aromatic sub-
strates give yields ranging from 12% to 28%. The ali-
phatic alkyl substrates exhibited the highest yields of
all compounds tested with 15 giving a 36% yield and
16 produced in 31% yield. This of course is an improve-
ment on the NaH reaction, which in these instances gave
no product. However, the reverse was observed in the
case of small aromatics where product yields were low
with LDA as opposed to NaH. For instance, thiophene
(8) gave a yield of 28% with LDA, while the NaH pro-
moted reaction gave an overall yield of 89%. The effect
was more pronounced with the bulky substrates. Biphe-
nyl (13) gave a 12% yield with LDA while a 93% yield
was obtained with NaH.
O
ii
i
CN
R
R
R
NH2
Figure 7. Reagents and conditions: (i) 1.1 equiv C6H12NO3P, 1.1 equiv
NaH; (ii) LDA.
that the reaction took 1 h to go to completion, we were un-
able to observe product formation under these condi-
tions.15 We, therefore, took to further optimizing this
reaction. With formic acid having a low boiling point, a
reflux setup was implemented. After 1 h, very little prod-
uct was observed in the crude 1H NMR using 8 as a model
substrate. The reaction time was thus extended to 12 and
24 h, respectively. At 12 h, about half of the starting mate-
rial was converted, giving a 37% yield, and at 24 h the
reaction went to completion, affording a 73% yield.
The synthesized enaminonitriles, with their 1,2-relation-
ship of the amino and nitrile groups, were subsequently
exploited in the synthesis of their quinozoline derivatives
via the heating in the presence of formic acid and formam-
ide at 200 °C (Fig. 8). While it was previously reported
Utilizing these optimized conditions, we converted each
of the enaminonitriles to their corresponding quinazo-

B. J. Bench et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3126–3130
3129
Table 3. Cytotoxic activities of enaminonitriles and quinazolines
against Jurkat
NH2
N
R
R
CN
i
Compound
IC50 (lg/ml)
lM
R
NH2
R
N
8
9
51.485
6.424
4.342
6.257
9.213
8.478
7.35
172.15
13.93
15.13
19.37
20.52
19.31
18.99
12.65
46.75
15.36
9.51
Figure 8. Reagents and conditions: (i) CHOOH, H2NCHO, 200 °C,
24 h.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
lines (Table 2). Overall yields for each substrate were
moderate with biphenyl 22 giving the highest at 78%
and bromo-thiophene 18 resulting in the lowest at 60%
yield. Phenyl 19 with its fluoro 20 and bromo 21 ana-
logues gave moderate yields of 74%, 67%, and 70%,
respectively. The bulky naphthalene 23 substrate was
comparable to biphenyl 23 with an overall yield of
74%. The alkyl containing b-ionyl 24 and octyl 25 con-
taining substrates were also generated in moderate yields
(yet lower than the aromatics), 54% and 57%,
respectively.
5.452
14.166
5.008
4.586
5.522
1.133
5.962
4.848
3.881
5.849
4.148
17.59
3.24
12.69
10.4
9.37
12.8
12.61
The biological activities of the synthesized enaminonitr-
iles and quinazolines were evaluated in cell-based assays
including their anti-proliferative/cytotoxic effects against
Jurkat (a leukemic T-cell line), bacterial strains (E. coli
and B. subtilis), and yeast (S. cerevisiae). Cytotoxicity
against T-cells was measured by staining the nuclei of
the cells with propidium iodide and measuring its fluo-
rescence, while anti-microbial activity was assessed by
measuring cell density at OD600. The most pronounced
effects were observed against Jurkat, where in all cases
the quinazolines were more active than their corre-
sponding enaminonitriles (Table 3). The most effective
compounds 20 (CC50 3.24 lM), 23 (CC50 9.37 lM),
and 18 (CC50 9.51 lM) were heterocyclic and in many
instances, halogenated. With respect to the anti-micro-
bial assays, all compounds displayed moderate activi-
ties. MIC values were P163 lM in B. subtilis,
P189 lM in E. coli, and P197 lM against S. cerevisiae.
Table 2. Quinozoline product yields
We report on a one-pot strategy in the preparation of
cyclohexadiene enaminonitriles. The reaction is high
yielding with heterocyclic systems and other bulky sub-
strates thus complementing the previously reported syn-
theses. The cyclohexadiene enaminonitriles were
additionally implemented as precursors in the synthesis
of quinazolines, a scaffold with demonstrated therapeu-
tic activities. In this regard, we optimized reaction con-
ditions reported in the literature, which generated each
quinozoline in moderate yields.
R group
Compound
Yield
73
S
17
S
Br
18
19
20
21
60
74
67
70
In cell-based assays, of all compounds tested, the qui-
nazolines gave the most promising effects, where CC50
values in the low micromolar range were measured
against a human leukemic cell-line. The identification
of the cellular targets of these lead compounds might di-
rect the synthesis of functionally and structurally related
analogues with more potent biological activities.
F
Br
22
23
78
74
Supplementary data including experimental details,
NMR spectra, and biological data are available online
free of charge.
Acknowledgments
24
25
54
57
We thank the Welch Foundation (A-1587), Texas A&M
University, and the NIH CBI Training Program (T32
GM008523) for support of this work. We also thank
Ms. Vanessa Santiago and Dr. Shane Tichy for perform-
ing MS analyses.

3130
B. J. Bench et al. / Bioorg. Med. Chem. Lett. 18 (2007) 3126–3130
11. Civantos Calzada, B.; Aleixandre de Artinano, A. Phar-
macol. Res. 2001, 44, 195.
Supplementary data
12. Docherty, J. R. Pharmacol. Ther. 1989, 44, 241.
13. Perrier, V.; Wallace, A. C.; Kaneko, K.; Safar, J.;
Prusiner, S. B.; Cohen, F. E. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 6073.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2007.10.048.
14. Pan, K. M.; Baldwin, M.; Nguyen, J.; Gasset, M.; Serban,
A.; Groth, D.; Mehlhorn, I.; Huang, Z. W.; Fletterick, R.
J.; Cohen, F. E.; Prusiner, S. B. Proc. Natl. Acad. Sci.
U.S.A. 1993, 90, 10962.
15. Tucker, H.; Golding, G.; Purvis, S. R. Tetrahedron Lett.
1981, 38, 1373.
16. Mirek, J.; Adamczyk, M.; Mokrosz, M. Synthesis 1980,
296.
17. Shabtai, J.; Ney-Igner, E.; Pines, H. J. Chem. Soc., Perkin
I 1973, 2230.
18. Bench, B. J.; Liu, C.; Evett, C. R.; Watanabe, C. M. H. J.
Org. Chem. 2006, 71, 9458.
19. Bench, B. J.; Tichy, S. E.; Perez, L. M.; Benson, J.;
Watanabe, C. M. H., submitted for publication.
20. Asato, A. E.; Watanabe, C.; Li, X.-Y.; Liu, R. S. H.
Tetrahedron Lett. 1992, 33, 3105.
References and notes
1. Hickmott, P. W. Tetrahedron 1982, 38, 1975.
2. Blaha, K.; Cervinka, O. Adv. Org. Chem. 1963, 4, 1.
3. Albert, A. Adv. Heterocycl. Chem. 1982, 32, 1.
4. Wamhoff, H. Adv. Heterocycl. Chem. 1985, 32, 1.
5. Wahba, A. E. Chem. Rev. 1993, 93, 1991.
6. Ashton, W. T.; Walker, F. C., Jr.; Hynes, J. B. J. Med.
Chem. 1973, 16, 694.
7. Jen, T.; Dienel, B.; Bowman, H.; Petta, J.; Helt, A.; Love,
B. J. Med. Chem. 1972, 15, 727.
8. Chern, J. W.; Tao, P. L.; Yen, M. H.; Lu, G. Y.; Shaiu, C.
Y.; Lai, Y. J.; Chien, S. L.; Chan, C. H. J. Med. Chem.
1993, 36, 2196.
9. Chern, J. W.; Shiau, C. Y.; Lu, G. Y. Bioorg. Med. Chem.
Lett. 1991, 1, 571.
10. Campbell, S. F.; Michael, M. J.; Hardstone, J. D.; Lewis,
B. N.; Palmer, M. J. J. Med. Chem. 1987, 30, 49.
21. Uchikawa, O.; Fukatsu, K.; Tokunoh, R.; Kawada, M.;
Matsumoto, K.; Imai, Y.; Hinuma, S.; Kato, K.; Nishik-
awa, H.; Hirai, K.; Miyamoto, M.; Ohkawa, S. J. Med.
Chem. 2002, 45, 4222.