Synthesis of quinazolines from N -(2-Nitrophenylsulfonyl)iminodiacetate and α-(2-Nitrophenylsulfonyl)amino ketones via 2 H -indazole 1-oxides

Article abstract of DOI:10.1021/jo100784z

(Figure presented) Base-catalyzed rearrangement of 2H-indazoles 1-oxides, prepared by tandem carbon-carbon followed by nitrogen-nitrogen bond formations from easily accessible N-alkyl-2-nitro-N-(2-oxo-2-aryl-ethyl)- benzenesulfonamides using glycine, 2-ni

Full text of DOI:10.1021/jo100784z

pubs.acs.org/joc
Synthesis of Quinazolines from N-(2-Nitrophenylsulfonyl)iminodiacetate
and r-(2-Nitrophenylsulfonyl)amino Ketones via 2H-Indazole 1-Oxides
‡
,†
ꢀ
ꢀ
ꢁ †,§
Sona Krupkova, Greg A. Slough, and Viktor Krchnak*
ꢀꢁ
†Department of Chemistry and Biochemistry, 251 Nieuwland Science Center, University of Notre Dame,
Notre Dame, Indiana 46556, and ‡Chemistry Department, Kalamazoo College, 1200 Academy St.,
Kalamazoo, Michigan 49006. §Present address: Department of Organic Chemistry, Faculty of Science,
University of Palackyꢁ, 771 46 Olomouc, Czech Republic.
vkrchnak@nd.edu
Received April 21, 2010
Base-catalyzed rearrangement of 2H-indazoles 1-oxides, prepared by tandem carbon-carbon
followed by nitrogen-nitrogen bond formations from easily accessible N-alkyl-2-nitro-N-(2-oxo-
2-aryl-ethyl)-benzenesulfonamides using glycine, 2-nitrobenzenesulfonyl chlorides, and bromo
ketones/acetates, yielded high purity quinazolines.
Introduction
form a polymer-supported iminophosphorane intermediate
that was cyclized to 3,4-dihydroquinazolines.7 Lou described
the synthesis starting from 4-bromomethyl-3-nitrobenzoate
or the corresponding amide bound to solid phase. After
substitution of bromide with amine nuclophiles, acylation
with carboxylic acid, and subsequent reduction of nitro group,
the resulting precursors underwent cyclocondensation to qui-
Quinazolines represent a group of pharmacologically
important compounds exhibiting a wide range of bio-
logical activities including antitumoral,1 antibacterial,2 anti-
malarial,3 and anticonvulsant4activities. The FDI-approved
drug Tarceva, a tyrosine kinase inhibitor, is a quinazoline
derivative.5 Significant biological effects were also found in
naturally occurring alkaloids with quinazoline moiety.4
Therefore this class of heterocycles became the center of
attention in many research studies, and a substantial effort
was dedicated to devising synthetic approaches to these
molecules.6 Since combinatorial chemistry has become an
integral part in research and development of new pharma-
ceuticals, particularly in connection with solid-phase syn-
thesis, invention of efficient synthetic methods provided access
to this class of heterocycles on polymer support. Wang and
Hauske developed a method utilizing 2-nitrocinnamic acid to
8
9
€
nazolinone derivatives. Vogtle and Marzinzik and Kamal
and co-workers1 have summarized these and other approaches
to solid-phase synthesis in review articles.
We recently discovered an efficient synthetic route to
indazoles.10 N-Alkylation of polymer-supported 2-nitroben-
zenesulfonyl (2-Nos)-activated/protected amines by bro-
moketones (a variant of the Fukuyama method11) provided
R-(2-nitrobenzenesulfonyl)amino ketones. Treatment of these
2-Nos derivatives with DBU led to tandem carbon-carbon
followed by nitrogen-nitrogen bond formation to produce
indazole oxides of excellent purity. Encouraged by this un-
expected and very clean transformation leading to pharma-
cologically relevant heterocycles, we expanded the original
scope of the tandem reaction to more complex structures.
The target indazoles contained a carbonyl functional group
(1) Kamal, A.; Reddy, K. L.; Devaiah, V.; Shankaraiah, N.; Rao, M. V.
Mini-Rev. Med. Chem. 2006, 6, 71–90.
(2) Kung, P. P.; Casper, M. D.; Cook, K. L.; Wilson-Lingardo, L.; Risen,
L. M.; Vickers, T. A.; Ranken, R.; Blyn, L. B.; Wyatt, J. R.; Cook, P. D.;
Ecker, D. J. J. Med. Chem. 1999, 42, 4705–4713.
(3) Madapa, S.; Tusi, Z.; Mishra, A.; Srivastava, K.; Pandey, S. K.;
Tripathi, R.; Puri, S. K.; Batra, S. Bioorg. Med. Chem. 2009, 17 (1), 222–234.
(4) Witt, A.; Bergman, J. Curr. Org. Synth. 2003, 7, 659–677.
(5) Moyer, J. D.; Barbacci, E. G.; Iwata, K. K.; Arnold, L.; Boman, B.;
Cunningham, A.; DiOrio, C.; Doty, J.; Morin, M. J.; Moyer, M. P.; Neveu,
M.; Pollack, V. A.; Pustilnik, L. R.; Reynolds, M. M.; Sloan, D.; Theleman,
A.; Miller, P. Cancer Res. 1997, 57, 4838–4848.
(7) Wang, F.; Hauske, J. R. Tetrahedron Lett. 1997, 38, 8651–8654.
(8) Zhang, J.; Barker, J.; Lou, B.; Saneii, H. Tetrahedron Lett. 2001, 42,
8405–9408.
(9) Vogtle, M. M.; Marzinzik, A. L. QSAR Comb. Sci. 2004, 23, 440–459.
(10) Bouillon, I.; Zajicek, J.; Pudelova, N.; Krchnak, V. J. Org. Chem.
2008, 73, 9027–9032.
ꢁ
(6) Connolly, D. J.; Cusack, D.; O Sullivan, T. P.; Quiry, P. J. Tetrahedron
2005, 61 (43), 10153–10202.
(11) Fukuyama, T.; Jow, C. K.; Cheung, M. Tetrahedron Lett. 1995, 36,
6377–6374.
4562 J. Org. Chem. 2010, 75, 4562–4566
Published on Web 06/04/2010
DOI: 10.1021/jo100784z
r
2010 American Chemical Society

ꢀ
Krupkova et al.
ꢁ
JOCArticle
FIGURE 1. Time study of the sequential cyclization sulfonamide 1 to quinazoline 3 in the presence of DBU.
SCHEME 1. Quinazolines from Iminodiacetate and
2-Nitrobenzenesulfonylchloridea
attached to the side chain of the substituent in position 2 of
the indazole ring. The carbonyl group enabled subsequent
transformations of indazoles and formation of a fused pyra-
zine ring via an intramolecular nucleophile located on the
R1 side chain, providing a route for pyrazino[1,2-b]indazole
synthesis.12 We also used our indazole chemistry to prepare
2-(2-amino/hydroxyethyl)-1-aryl-3, 4-dihydropyrazino[1,2-b]
indazol-2-iums that cyclized in neutral pH to fused polycyclic
heterocycles13 that unexpectedly rearranged to as yet unex-
plored 2,3-dihydro-1H-imidazo[1,2-b]indazoles.14
While attempting to further expand the scope of this
rearrangement, we discovered a new ring-expansion reaction.
Here, we describe its use for the synthesis of quinazolines and
development of methodology for combinatorial synthesis of
target quinazolines with three diversity positions.
aReagents and conditions: (i) lutidine, DCM, rt, overnight; (ii) DBU,
DMF, rt, overnight.
Results
Diethyl iminodiacetate reacted with 2-nitrobenzene-
sulfonylchloride in the presence of a mild base, lutidine,
and yielded the expected sulfonamide 1. Exposure of the
crude reaction mixture to a strong base, DBU, in DMF
triggered the subsequent transformation to quinazoline 3
(Scheme 1).
Qualitative details associated with the sequential cycliza-
tion of 1 to 3 came from a time-study 1H NMR experiment.
When a 0.11 M solution of 1 in DMSO-d6 was treated with
3 equiv of DBU, an immediate color change ensued and
proton signals (δ 4.16 (s), 4.09 (q)) for sulfonamide 1 rapidly
decreased. An intermediate set of signals (δ 4.39 (q), 4.19 (q))
increased in intensity, and then these signals declined with
production of quinazoline 3 (δ 4.56 (q), 4.45(q)) (seeFigure 1).
Attempts to extract quantitative information for this reaction
were complicated by ester hydrolysis and the presence of water
in the initial reaction mixture. Ester hydrolysis occurred in all
three materials involved in this reaction and accounted for the
low overall yield of 3 (54%). The data plotted in Figure 1
resulted only after thorough drying of solvent and DBU.15,16
(12) Pudelova, N.; Krchnak, V. J. Comb. Chem. 2009, 11, 370–374.
(13) Koci, J.; Krchnak, V. J. Comb. Chem. 2010, 12, 168–175.
(14) Koci, J.; Oliver, A. G.; Krchnak, V. J. Org. Chem. 2010, 75, 502–505.
(15) Burfield, D. R.; Smithers, R. H. J. Org. Chem. 1978, 43, 3966–3968.
(16) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory
Chemicals, 4th ed.; Elsevier: Amsterdam, 1996.
J. Org. Chem. Vol. 75, No. 13, 2010 4563

ꢀ
Krupkova et al.
ꢁ
JOCArticle
SCHEME 2. Synthesis of Quinazolines on Solid Phasea
aReagentsand conditions:(i) piperidine, DMF, rt, 15min;(ii) 2-nitrobenzenesulfonylchloride, lutidine, DCM, rt, overnight;(iii) bromoketone,DMF, rt,
0.5-7h (seeExperimental Section); (iv) DBU, DMF, rt, 30min;(v) DBU, DMF, rt, 10 min toovernight(seeExperimentalSection); (vi) TFA, DCM, rt, 1h.
Rink amide resin acylated with Fmoc-Gly-OH. Later we
also used alkylated BAL resin and ethanolamine-derivatized
Wang resin.
To optimize reaction conditions and to evaluate the effect
of a diverse substitution pattern on 2-nitrobenzenesulfonyl
chlorides and bromoketones on the yield and purity of
products, we prepared a set of model compounds on Rink
amide resin acylated with Fmoc-Gly-OH (Figure 2). Both
2-Nos chlorides and bromoketones were selected to bear
electron-donating as well as electron-withdrawing groups.
Synthesis of acyclic precursors followed from a published
protocol.
The rate of N-alkylation was dependent upon the sub-
stitution pattern of both building blocks (bromoketone and
benzenesulfonyl chloride). The presence of a nitro group on
either bromoketone or 2-Nos chloride markedly accelerated
the reaction; the alkylation was complete in 30 min. By
contrast, when electron-donating substituents were present
on either the ketone or benzenesulfonyl groups, the nitrogen
alkylation reaction required several hours for completion.
LCMS analysis of linear precursors prepared by alkyla-
FIGURE 2. Diversity reagents.
In the presence of water, the consumption of 1 occurred at a
rate similar to that shown in Figure 1, whereas cyclization to 3
can take up to 18 h to achieve 50% conversion.
tion of electron-deficient model compounds already revealed
the presence of indazole-oxides, an indication that these
activated systems cyclized even in the presence of DIEA
used in the alkylation step.
To increase the diversity of compounds accessible by
this ring-expansion transformation, we synthesized combi-
natorial ensembles of linear precursors with three diversity
positions on solid-phase support. N-Alkyl-2-nitro-N-(2-oxo-
2-aryl-ethyl)-benzenesulfonamides 6 were prepared by using
glycine, 2-nitrobenzenesulfonyl chlorides and bromoketones
following a route developed for the synthesis of indazole
oxides.10 When the indazole-oxides 7 were exposed to basic
condition, we observed rearrangement to a six-membered
ring 8 (Scheme 2). Thus, our previously described route for
accessing indazole derivatives was used to acquire quinazo-
lines. First, we evaluated the scope and limitation of this
transformation. Model compounds were synthesized on
Extended exposure to DBU caused further transforma-
tion of indazole-oxides 7 to quinazolines 8. The outcome of
cyclization was highly dependent upon the substitution
pattern. The effect was evaluated on a set of model com-
pounds with particular attention given to the Nos substitu-
tion pattern, 6(1,R1,1). The electron-withdrawing nitro group
accelerated the rearrangement to quinazolines. In compari-
son, nonsubstituted derivative 6(1,1,1) (R2, R3 = H) required
0.2 M DBU overnight to achieve maximum yield of 8(1,1,1),
wheras quinazoline 8(1,4,1) (R2 = NO2, R3 = H) formed in
0.1 M DBU in 10 min. Prolonged reaction time for reactions
involving 6(1,4,1) ledtoproduct decomposition. The electron-
4564 J. Org. Chem. Vol. 75, No. 13, 2010

ꢀ
Krupkova et al.
ꢁ
JOCArticle
TABLE 1. Summary of the Prepared Derivatives
compound
R1-H
R2
R3
method
purity (%)a
yield (%)b
9(1,1,1)
9(1,1,2)
9(1,1,3)
9(1,1,4)
9(1,1,5)
9(1,1,6)
9(1,1,9)
9(1,1,10)
9(1,2,1)
9(1,2,6)
9(1,4,1)
9(1,4,6)
9(2,1,8)
9(3,1,1)
9(3,1,2)
9(3,1,5)
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
H
H
H
H
H
H
H
H
CF3
CF3
NO2
NO2
H
H
H
H
Ph
4-Me-Ph
4-OMe-Ph
4-CN-Ph
4-Cl-Ph
4-NH2-3,5-diCl-Ph
piperidin-1-yl
4-Ph-piperazin-1-yl
Ph
4-NH2-3,5-diCl-Ph
Ph
4-NH2-3,5-diCl-Ph
O-Et
Ph
4-Me-Ph
4-Cl-Ph
A
A
A
B
88
95
90
25
98
90
76
90
91
98
85
92
45
88
79
84
23
12
10
15
11
20
30
41
31
23
33
14
18
48
33
36
A
A
A
A
A
A
C
C
A
A
A
A
NH-CH2-Ph-4-Me
NH-(CH2)2)-OH
NH-(CH2)2)-OH
NH-(CH2)2)-OH
aPurity of crude product before purification. bYield after purification. Method A: 0.2 M DBU in DMF, rt, overnight; method B: 0.2 M DBU in DMF,
rt, 1 day; method C: 0.1 M DBU in DMF, rt, 1.5 h.
SCHEME 3. Diversification of the R3 Substituenta
by DIEA. Indazole oxide was stable in the presence of DIEA
at room temperature. Increased temperature facilitated the
rearrangement to quinazoline, but the purity of crude pro-
duct was low as a result of contamination by unidentified
components.
The diversity of quinazolines was further expanded by
alkylation of the resin 5 with ethyl bromoacetate. The inter-
mediate 6(R3 = OEt) cyclized to indazole oxide 7(1,1,8). At
this stage, the ester 7(1,1,8) was hydrolyzed and the carboxy-
late resin was converted to amides 7(1,1,9-10) (Scheme 3).
Subsequent overnight exposure to DBU converted the inda-
zole oxides 7(1,1,9-10) to quinazolines 8(1,1,9-10). A sum-
mary of quinazoline formation is presented in Table 1.
aReagents and conditions: (i) NaOH in THF/MeOH (1:1), 1 h; (ii)
HOBt, DIC in DMF, 1 h, then piperidine or 1-phenylpiperazine, DMF, 1 h.
donating methoxy groupinmodelcompound6(1,2,1) cyclized
to indazole-oxides only. Attempts to facilitate the conversion
of the indazole-oxide ring to quinazoline by increasing the
temperature led to decomposition, as demonstrated by NMR
analysis of the cleaved resin product.
The structure of bromoketones influenced the conversion
to quinazolines as well. The presence of a nitro group
considerably accelerated the formation of indazole; how-
ever, 6-nitroquinazoline 8(1,1,7) was obtained only in low
yield in a complex mixture of unidentified compounds.
Because we observed partial cyclization during alkylation
in the presence of DIEA, we attempted to replace the DBU
Discussion
Mechanism of Rearrangement. This base-catalyzed rear-
rangement started with scission of the nitrogen-nitrogen bond
caused by the presence of the N-oxide, which was responsible
for electronic activation of the electron-deficient nitrogen and
assisted by the DBU-mediated proton subtraction (Scheme 4).
Formation of the N-hydroxy derivative was important for
subsequent water elimination facilitated by carbon-nitrogen
bond formation. We confirmed the essential role of the
N-oxide by preparing the deoxygenated analog of 7 according
SCHEME 4. Proposed Mechanism of Quinazoline Formation
J. Org. Chem. Vol. 75, No. 13, 2010 4565

ꢀ
Krupkova et al.
ꢁ
JOCArticle
SCHEME 5. Rearrangement of 2,20-(3-Oxo-1H-indazole-
1,2(3H)-diyl)diacetate
semipreparative HPLC. Yield 75.0 mg (25%). ESI-MS m/z =
275, [M þ H]þ. 1H NMR (300 MHz, DMSO-d6): δ 8.49 (d, J =
8.6 Hz, 1 H), 8.29 (d, J = 8.3 Hz, 1 H), 8.22 (t, J = 6.9 Hz, 1 H),
7.99 (t, J = 7.2 Hz, 1 H), 4.56 (q, J = 7.0 Hz, 2 H), 4.45 (q, J =
7.1 Hz, 2 H), 1.40 (q, J = 7.2 Hz, 6 H). 13C NMR (300 MHz,
DMSO): δ 164.7, 163.7, 159.1, 152.5, 151.5, 136.8, 132.1, 130.0,
126.6, 121.7, 63.5, 62.8, 14.8, 14.7 HRMS (FAB) m/z calcd for
C14H15N2O4 275.1026 [M þ H]þ, found 275.1034.
Solid-Phase Synthesis Methods. Synthesis of the resins 5 was
carried out according to published protocol.10
to a recently published protocol.17 After 2 days of exposure
to DBU containing DMF, no change of the indazole was
observed.
No precedent for the transformation of indazole oxides
to quinazolines was found. A distant analogy to the ring
expansion of indazoles can be found in the rearrangement of
diethyl 2,20-(3-oxo-1H-indazole-1,2(3H)-diyl)diacetate 12 to
ethyl 3-(2-ethoxy-2-oxoethyl)-4-oxo-1,2,3,4-tetrahydroqui-
nazoline-2-carboxylate 13 (Scheme 5).18,19
Alkylation with Bromoketones (Resins 6). The polypropylene
fritted syringe was charged with the resin 5 (∼250 mg) and
washed 3 times with DCM and 3 times with DMF. A solution of
bromoketone (1.5 mmol) and DIEA (3 mmol) in 3 mL of DMF
was added and shaken at room temperature for 7 h. Alkylation
of resins 5 prepared from 4-CF3-2-Nos-Cl and 4-NO2-2-Nos-Cl
or 4-CN and 3-NO2 bromoketones was complete in 30 min. The
resin was washed 3 times with DMF and 3 times with DCM.
Synthesis of Amides (Resins 7(1,1,9-10)). Resin 7(1,1,8)
(∼250 mg) was washed with THF and treated with a solution
of 0.5 mL of 10 M NaOH in 10 mL of THF/MeOH (1:1) for 1 h.
The resin was washed with THF, 3% AcOH in THF, THF, and
DCM. A solution of HOBt (2 mmol, 306 mg) and DIC (2 mmol,
312 uL) in 4 mL of DMF was added to the resin and left at
ambient temperature for 1 h. The resin was washed with DMF,
and then a 1 M solution of amine (piperidine and 1-phenyl-
piperazine) in DMF (4 mL) was added and reacted for 1 h. The
resin was washed with DMF and DCM.
Cyclization to Quinazolines (Resins 8). The polypropylene
fritted syringe was charged with the resin 6 (∼250 mg) and
washed 3 times with DCM and 3 times with anhydrous DMF.
Cyclization was carried out at ambient temperature in 5 mL of
DBU solution in DMF. Method A: 0.2 M DBU, overnight.
Method B: 0.1 M DBU, 1.5 h. Method C: 0.1 M DBU, 10 min.
The resin was washed 3 times with DMF, DCM, 5% AcOH in
DCM, DCM, and MeOH.
Conclusion
We have described a convenient preparation of quinazo-
lines via base-catalyzed rearrangement of 2H-indazoles
1-oxides, obtained by tandem carbon-carbon followed by
nitrogen-nitrogen bond formations from easily accessible
N-alkyl-2-nitro-N-(2-oxo-2-aryl-ethyl)-benzenesulfonamides
using glycine, 2-nitrobenzenesulfonyl chlorides, and bromo
ketones/acetates. The transformation tolerated a range of
substituents; however, it was sensitive to their electronic
properties. This synthesis does not require preparation of
dedicated building blocks, rather it was carried out using
commercially available compounds.
Experimental Section
Quinazolines 9. Resin 8 was treated with 50% TFA in DCM
for 1 h. The cleavage cocktail was collected, and the resin was
washed 2 times with 50% TFA in DCM. The washes were
collected and evaporated by a stream of nitrogen. The oily residue
was dissolved in MeOH and purified by semipreparative HPLC.
4-Benzoylquinazoline-2-carboxamide 9(1,1,1). Yield 10 mg
Solid-phase syntheses were conducted on a manually operated
Domino Block synthesizer20 (www.torviq.com) in disposable
polypropylene reaction vessels. Commercially available solvents,
resins, and reagents were used. Wang resin (100-200 mesh, 1%
DVB, 1.0 mmol/g) was obtained from Advanced ChemTech
(Louisville, KY, www.peptide.com). Swelling of resins in DCM
was measured before syntheses, and resins with swollen volume
greater than 7 mL/g of dry resin were used.21 All reactions
occurred at ambient temperature (21 °C) unless stated otherwise.
Quinazoline-2,4-dicarboxylic Acid Diethyl Ester (3). A solu-
tion of diethyl iminodiacetate (1 mmol) in 2 mL of DCM and a
solution of 2-nitrobenzenesulfonyl chloride (1 mmol) in 1 mL of
DCM were combined. Lutidine (1 mmol) was added, and the
reaction mixture was kept at ambient temperature overnight.
The DCM was evaporated by a stream of nitrogen. The residual
material was dissolved in anhydrous DMF, and DBU (4 mmol)
was added. The reaction mixture remained at ambient tempera-
ture overnight. The product was isolated after purification by
1
(23%). ESI-MS m/z = 278, [M þ H]þ. H NMR (300 MHz,
DMSO-d6): δ 8.39-8.25 (m, 2 H), 8.19 (t, J = 7.5 Hz, 1 H), 8.03
(d, J = 8.0 Hz, 1 H), 8.00-7.91 (m, 3 H), 7.87 (t, J = 8.0 Hz,
1 H), 7.79 (t, J = 7.5 Hz, 1 H), 7.60 (t, J = 7.7 Hz, 2 H). 13C
NMR (125 MHz, DMSO-d6): δ 192.8, 165.0, 164.3, 153.7, 150.3,
135.9, 135.1, 134.6, 130.6, 130.5, 129.2, 129.2, 125.6, 121.0.
HRMS m/z calcd for C16H11N3O2 [M þ H]þ 278.0924, found
278.0915
Acknowledgment. The work was supported by the De-
partment of Chemistry and Biochemistry University of
Notre Dame, the NIH (GM079576), Department of Organic
ꢁ
Chemistry University of Palacky, and the Ministry of Edu-
cation, Youth and Sport of the Czech Republic (ME09057).
We appreciate the use of the NMR facility at the University
of Notre Dame.
(17) Morimoto, Y.; Kurihara, H.; Yokoe, C.; Kinoshita, T. Chem. Lett.
1998, No. 8, 829–830.
(18) Bonanomi, M.; Palazzo, G. Farmaco 1977, 32, 490–501.
(19) Baiocchi, L.; Picconi, G. Tetrahedron Lett. 1986, 27, 5255–5256.
(20) Krchnak, V.; Padera, V. Bioorg. Med. Chem. Lett. 1998, 22, 3261–
3264.
Supporting Information Available: Spectroscopic data and
copies of NMR spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
(21) Bouillon, I.; Soural, M.; Miller, M. J.; Krchnak, V. J. Comb. Chem.
2009, 11, 213–215.
4566 J. Org. Chem. Vol. 75, No. 13, 2010