Synthesis of pyridodiazepinediones by using the ugi multicomponent reaction

Article abstract of DOI:10.1002/ejoc.201000549

Benzodiazepines show a broad spectrum of biological activities. In an ongoing effort to extend molecular diversity in this type of systems, we developed a strategy for synthesizing 3,4-dihydro-1H-pyrido[2,3-e][1,4] diazepine-2,5-dione compounds starting from 2-hydroxynicotinic acid and by using an Ugi reaction as a key step in the synthesis. We opted to use 2isocyanophenyl benzoate instead of Armstrong's convertible isocyanide in this multicomponent reaction.

Full text of DOI:10.1002/ejoc.201000549

FULL PAPER
DOI: 10.1002/ejoc.201000549
Synthesis of Pyridodiazepinediones by Using the Ugi Multicomponent Reaction
An M. Van den Bogaert,[a] Jo Nelissen,[a] Margriet Ovaere,[b] Luc Van Meervelt,[b]
Frans Compernolle,[a] and Wim M. De Borggraeve*[a]
Keywords: Ugi reaction / Pyridodiazepinediones / NMR spectroscopy / Multicomponent reactions
Benzodiazepines show a broad spectrum of biological activi-
ties. In an ongoing effort to extend molecular diversity in this
type of systems, we developed a strategy for synthesizing
pounds starting from 2-hydroxynicotinic acid and by using an
Ugi reaction as a key step in the synthesis. We opted to use 2-
isocyanophenyl benzoate instead of Armstrong’s convertible
3,4-dihydro-1H-pyrido[2,3-e][1,4]diazepine-2,5-dione
com- isocyanide in this multicomponent reaction.
Introduction
Benzodiazepines show a broad spectrum of biological ac-
tivities.[1] They are widely used as antianxiety drugs that
help to relieve nervousness, tension and other symptoms by
slowing the central nervous system. Valium (Diazepam) and
Xanax (Alprazolam) are two of the most used and best
known benzodiazepines (Figure 1).
Figure 2. Conversion of 2-hydroxynicotinic acid to 3,4-dihydro-1H-
pyrido[2,3-e][1,4]diazepine-2,5-dione compounds.
A high degree of functionalization was one of the goals
we set ourselves when developing this synthesis, together
with easy access of starting materials.
In our synthetic sequence the commercially available 2-
hydroxynicotinic acid was converted to target compounds 1
(Figure 2) bearing various substituents R1 to R4 (see
Table 1). The planned key step in the synthesis was an Ugi
reaction,[3] which is widely used to synthesize benzodiazep-
ine-type compounds.[4]
Figure 1. Valium and Xanax.
Table 1. Yields for Ugi and Suzuki reactions.
Product Starting material
R1
R2
R3
R4
Yield [%]
1a
1b
1c
1d
1e
1f
1g
2a
2c
2d
2e
2f
2b
2c
2f
2f
2f
Ph
Ph
Ph
Me
Bu
Bn
Pr
Pr
Pr
Pr
Pr
Pr
Ph
Pr
Pr
Pr
Ph
Pr
Ph
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bu
Bn
Bn
Bn
81
83
66
69
48
61
65
22
24
28
21
48
19
Results and Discussion
Large numbers of both natural and synthetic benzodi-
azepines have been reported. In an attempt to extend the
molecular diversity in this type of systems, we developed
a general strategy for preparing 3,4-dihydro-1H-pyrido[2,3-
e][1,4]diazepine-2,5-dione compounds 1 (Figure 2), since
only a few of these derivatives with a pyridine ring are
known to date.[2]
4-MeOC6H4 Bn
3-FC6H4
3-ClC6H4
Bn
Bn
2f
2e
4a
4c
4c
4b
4b
4c
3-O2NC6H4 Bn
Br
Br
Br
Br
Br
Br
H
Bu
Bu
Bn
Bn
Bu
2g
[a] Molecular Design and Synthesis Department of Chemistry,
Katholieke Universiteit Leuven,
Celestijnenlaan 200F – box 2404, 3001 Heverlee, Belgium
[b] Biochemistry, Molecular and Structural Biology Section,
Department of Chemistry, Katholieke Universiteit Leuven,
Celestijnenlaan 200F – box 2404, 3001 Heverlee, Belgium
Fax: +32-16327990
A general retrosynthetic analysis of the pyridodiazepine-
diones 1 is presented in Scheme 1. The R1-substituted pyr-
idodiazepinediones 1 could be prepared by a Suzuki cross-
coupling reaction on bromo derivatives 2. An Ugi multi-
component reaction on nicotinic acid derivatives 4 by using
2-isocyanophenyl benzoate as the isocyanide can yield in-
View this journal online at
E-mail: wim.deborggraeve@chem.kuleuven.be
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000549.
wileyonlinelibrary.com
Eur. J. Org. Chem. 2010, 5397–5401
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5397

W. M. De Borggraeve et al.
FULL PAPER
termediate 3, which cyclises into the desired 3,4-dihydro- carried out by using benzyl- or butylamine, respectively, as
1H-pyrido[2,3-e][1,4]diazepine-2,5-dione core structure 2. the amine component and benzaldehyde or butanal as the
When R2 in precursor 2 is a hydrogen atom, an extra substi- aldehyde component. For the desired cyclization step, a
tution step is required to introduce the corresponding N- convertible isocyanide is required in the procedure. Nor-
substituent. Compound 4 in turn can be prepared from 2- mally, 1-isocyanocyclohexene (Armstrong’s convertible iso-
hydroxynicotinic acid in three steps.
cyanide) is used for the synthesis of benzodiazepines.[6]
However, due to reported synthetic and stability problems
with this compound (also occurring in our hands), other
classes of convertible isocyanides were also applied.[7] We
finally opted to use 2-isocyanophenyl benzoate (7), which
was prepared from benzoxazole according to the method
reported by Pirrung and Ghorai (Scheme 3).[7b]
Scheme 3. Synthesis of 2-isocyanophenyl benzoate.
Intermediate 3, the initial product of the Ugi reaction,
could not be purified. Hence, the crude mixture was dried
and heated with KOtBu in dry THF[7a] to provide the pyri-
dodiazepinedione core structure 2 (Scheme 4). The yields
over the two steps were very moderate [due to side products
and extensive chromatography (in some cases up to three
purifications)]; however, it was possible to obtain the de-
sired compounds in a pure form. The poor yield of the Ugi
reaction is probably due to the use of 4 without a protective
group, such as Boc or Fmoc.[8]
Scheme 1. Retrosynthetic analysis.
The first step in the synthesis is the bromination of 2-
hydroxynicotinic acid,[5] which according to a literature pro-
cedure[5a] produced 5-bromo derivative 5 in yields of about
70% (Scheme 2). This halogen is needed for the final func-
tionalization of the pyridine core.
Scheme 4. (a) R4NH2, R3CHO, MeOH, 2-isocyanophenyl benz-
oate; (b) KOtBu, THF; (c) R1B(OH)2, 5 mol-% Pd[P(Ph3)]4, 2 
Na2CO3.
Scheme 2. Synthesis of intermediates 4; compound 4a (R2 = H) is
commercially available.
When R2 is a butyl or benzyl group, only one extra step
Substitution of the 2-hydroxy group in pyridine com- involving a Suzuki cross-coupling reaction was needed to
pound 5 by reaction with thionyl chloride afforded the cor- yield the final products 1. This conversion was achieved by
responding 2-chloro derivative 6.[5b,5c] Next, compound 6 heating a mixture of compound 2 with 1.1 equiv. of an aryl-
was heated with benzylamine or butylamine to provide the boronic acid, a 2  Na2CO3 solution and a catalytic
corresponding 2-aminopyridine derivatives 4b and 4c. amount of [Pd(PPh3)4] at 80 °C in toluene. Products 1 were
Compound 4a with an unsubstituted 2-amino group is obtained in yields ranging from 48 to 83% (Scheme 4).
commercially available.
For 2a, an extra step is needed, which involves methyl-
In the next step, a multicomponent Ugi reaction was ap- ation of the NH group by using NaH and MeI. In a final
plied to compounds 4b and 4c to introduce both R3 and R4 step, a Suzuki reaction is applied to 2b to afford the tetra-
groups, and the resulting crude intermediates 3 are submit- substituted pyridodiazepinedione structure 1a (Scheme 4,
ted to base-catalyzed ring closure. The Ugi reaction was Table 1).
5398
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5397–5401

Synthesis of Pyridodiazepinediones by Using the Ugi Multicomponent Reaction
Transparent plate crystals of 2a were obtained. A small
From Hyperchem MM+ molecular mechanics and AM1
single crystal was measured by X-ray crystallography (Fig- semiempirical calculations, it appears that compounds 1
ure 3). The crystal structure clearly shows one conforma- and 2 can exist as two conformers, in which the 3-substitu-
tion for compound 2a with the propyl side chain disposi- ent has either an equatorial or axial orientation. In Fig-
tioned in an axial fashion.
ure 5, this is illustrated for the 3-Prax and 3-Preq conformers
corresponding to the (3S) enantiomer of the 3-propyl-sub-
stituted compound 2f. Interconversion of the two conform-
ers, in which the two planar amide bonds remain fixed, can
be effected by a coordinated rotation about the four single
bonds of the seven-membered ring.
Figure 5. Conformational equilibrium for the (3S) enantiomer of
2f.
The existence of two conformers was confirmed by 2D
NMR spectroscopic experiments (ratio ca. 1.1:1). The Hy-
perchem AM1 calculations for compound 2f (Figure 6) in-
deed revealed about equal energies for the 3-Prax and 3-Preq
conformers, whereas a somewhat larger energy difference
was calculated for compound 2e, favouring the 3-Phax con-
former.
Figure 3. X-ray diffraction measurement of 2a. CCDC-692196 con-
tains the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
The NMR spectra of the pyridodiazepinediones 1 and
2 revealed some interesting conformational characteristics,
which were reported also for benzodiazepines.[7a] Indeed,
except for 1g, 2a and 2e, the spectra of compounds 1 and
2, dissolved in CDCl3 or [D6]DMSO, at 298 K all exhibit
split signal patterns. This finding reveals the existence of
two conformers, which slowly interconvert on the NMR
spectroscopic time scale. Cooling of compound 2a below
270 K also results in a splitting of signals, again showing
the existence of two interconverting conformers (Figure 4).
However, even when cooling the 3-phenyl-substituted com-
pound 2e to 250 K, no separation of signals was observed.
Figure 6. Hyperchem presentation for the 3-Preq and 3-Prax con-
formers of the (3S) enantiomers of 2f.
Heating of a solution of compound 2f in CDCl3 did not
produce changes in the 1H NMR spectrum. However, when
the sample was dissolved in [D6]DMSO, changes were ob-
served starting from 333 K. Coalescence of all proton sig-
nals occurred at 353 K. This is illustrated (Figure 7) for the
methyl group located in the propyl side chain of 2f.
Figure 7. Proton signal observed for the methyl group located in
the propyl side chain of 2f at different temperatures.
Conclusions
The synthesis of 3,4-dihydro-1H-pyrido[2,3-e][1,4]diazep-
ine-2,5-dione compounds 1 has been reported. By using the
Ugi reaction as a key step, structural variation was achieved
Figure 4. NMR spectroscopic variable-temperature experiment
with compound 2a.
Eur. J. Org. Chem. 2010, 5397–5401
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5399

W. M. De Borggraeve et al.
FULL PAPER
one (1b; R1 = Ph, R2 = Bu, R3 = Pr, R4 = Bn): See the general
procedure for the Suzuki reaction: 2b (61 mg, 0.14 mmol), phen-
ylboronic acid (50 mg), Na2CO3 solution (1.0 mL). Column
chromatography: silica gel; (1) 100 DCM Ǟ MeOH/DCM, 1:99;
(2) EtOAc/heptane, 10:90. A final purification was performed by
HPLC (80 Ǟ 100 MeOH/water + 1% HCO2H; 20 min; flow rate
at three positions, that is, 1, 3 and 4. A fourth variation was
accomplished by a Suzuki cross-coupling reaction on the 7-
bromo position of precursor 2. The NMR spectra of com-
pounds 1 and 2 revealed the existence of a slow equilibrium
between two conformers with equatorial and axial orienta-
tion of the 3-substituent.
0.2 mL/min). Yield: 83%. Oil. IR (NaCl): ν = 1686 (s, CO amide),
˜
1643 (s, CO amide), 1600 (s, phenyl) cm–1. Compound 1b exists in
2 conformations in a ratio of 1:1 at 298 K. 1H NMR (400 MHz,
CDCl3): δ = 8.81 [t (2ϫ d), Jmeta = 3 Hz, 2 H, CH], 8.51 + 8.49
[2ϫ (d, Jmeta = 3 Hz, 1 H, CH)], 7.68 (m, J = 7 Hz, 4 H, CH), 7.50
(m, J = 7 Hz, 4 H, CH), 7.44–7.26 (m, 12 H, H arom), 4.99 (d, J
= 14 Hz, 1 H, CH2), 4.84–4.74 (m, 3 H, CH2), 4.28–4.19 + 4.14–
4.05 (m, 6 H, CH + CH2), 2.27–2.19 (m, 1 H, CH2), 1.79–1.70 (m,
1 H, CH2), 1.69–1.00 (m, 14 H, CH2), 0.90–0.83 [quint (2ϫ t), 9
H, CH3 + CH3], 0.65 (t, J = 7 Hz, 3 H, CH3) ppm. 13C NMR
(100 MHz, CDCl3): δ = 169.3 + 168.3 (C), 167.7 + 165.2 (C), 150.0
+ 149.1 (C), 149.5 + 149.2 (CH), 138.3 + 138.2 (CH), 137.7 + 136.3
+ 136.2 + 136.1 (Cipso), 133.8 + 133.4 (C), 129.6 + 129.4 (2) + 128.9
(2) + 128.6 (2) + 128.2 + 127.9 + 127.6 + 127.0 (2, C arom), 124.7
+ 124.3 (C), 65.3 + 57.0 (CH), 54.2 + 46.6 (CH2), 45.8 + 45.0
(CH2), 32.5 + 29.3 (CH2), 30.5 + 30.3 (CH2), 20.2 + 20.1 (2) +
19.7 (CH2 + CH2), 14.0 + 13.9 (2) + 13.4 (CH3 + CH3) ppm.
EIMS: m/z (%) = 441 (94) [M]+, 398 (5) [M – C3H7]+, 385 (10)
[M – C4H8]+, 350 (10) [M – C7H7]+. HRMS: calcd. for C28H31N3O2
441.2416; found 441.2426.
Experimental Section
General: NMR spectra were recorded with a Bruker Avance 300
(300 MHz), a Bruker Avance 400 (400 MHz) or a Bruker Avance
600 II+ (600 MHz) spectrometer. HPLC was performed with a
Waters Delta 600 analytical/preparative HPLC system equipped
with a Waters 996 PDA detector (Alltech C18 Prevail column,
5 µm, 22ϫ150 mm and Phenomenex Luna C18 column, 5 µm,
21.2ϫ150 mm). Melting points (uncorrected) were recorded with a
Reichert-Jung Thermovar or an Electrothermal 9200 melting point
apparatus.
General Procedures. Ugi Reaction: The aldehyde (1 equiv.) and the
amine (1.25 equiv.) were dissolved in methanol. This solution was
stirred at room temperature under argon for 1 h until the imine
had formed (checked by mass spectrometry). Then, the starting
material and the isocyanide (1 equiv.) were added. The mixture was
stirred until the reaction was complete (TLC monitoring). Suzuki
Reaction: [Pd(PPh3)4] (5 mol-%) and degassed 2  Na2CO3 solution
(6.8 equiv.) were added to a mixture of starting material and
boronic acid (1.1 equiv.) in toluene (degassed). The mixture was
stirred under argon at 80 °C until disappearance of the starting
material according to TLC. The solution was concentrated under
reduced pressure. Water was added, and the aqueous layer was ex-
tracted with CH2Cl2 (3ϫ50 mL). After drying the pooled CH2Cl2
layers with MgSO4 and concentration under reduced pressure, the
residue was purified by column chromatography.
Supporting Information (see footnote on the first page of this arti-
cle): X-ray data of 2a, experimental procedures and full spectro-
scopic data for all new compounds.
Acknowledgments
We thank Professor S. Toppet and K. Duerinckx for their assist-
ance with NMR spectroscopic analysis, Ir. B. Demarsin for HRMS
measurements and D. Henot for preparative HPLC. A. M. V. D. B.
(Research Assistant of the FWO) thanks the Fonds voor Wetensch-
appelijk Onderzoek (FWO-Vlaanderen) for financial support. J. N.
thanks the Institute for the Promotion of Innovation through Sci-
ence and Technology in Flanders (Belgium, I. W. T.) for financial
support.
Representative Experimental Procedure for 4-Benzyl-7-bromo-3-
propyl-3,4-dihydro-1H-pyrido[2,3-e][1,4]diazepine-2,5-dione (2a; R1
= Br, R2 = H, R3 = Pr, R4 = Bn): The Ugi product was synthesized
according to the general procedure for Ugi reactions by using but-
anal (0.21 mL), benzylamine (0.31 mL), 2-amino-5-bromonicotinic
acid (500 mg, 2.3 mmol) and 2-isocyanophenyl benzoate (0.51 g).
After addition of 4a and the isocyanide to the imine, the mixture
was stirred for 2 h. After this time, the mixture was concentrated
under reduced pressure and dried under vacuum. Subsequently, dry
THF and KOtBu (0.517 g, 2 equiv.) were added to the crude prod-
uct. The solution was then refluxed for 16 h, after which time the
THF was removed under reduced pressure. The residue was puri-
fied by column chromatography (silica gel, heptane/EtOAc, 80:20).
A last purification was performed by HPLC (60 Ǟ 100 MeOH/
water + 1% HCO2H; 20 min; flow rate 0.2 mL/min). Yield: 22%.
[1] Recent review: N. A. Meanwell, M.A. Walker, in Comprehen-
sive Heterocyclic Chemistry III (Eds.: A. R. Katritzky, C. A.
Ramsden, E. F. V. Scriven, R. J. K. Taylor), Elsevier Ltd, Kid-
lington, Oxford, 2008, vol. 13, p. 183.
[2] a) C. G. Boojamra, K. M. Burow, L. A. Thompson, J. A. Ell-
man, J. Org. Chem. 1997, 62, 1240–1256; b) A. Kamal, K. L.
Reddy, V. Devaiah, N. Shankaraiah, G. S. K. Reddy, S.
Raghavan, J. Comb. Chem. 2007, 9, 29–42.
[3] a) I. Ugi, Isocyanide Chemistry, Academic Press, New York,
USA, 1971; b) I. Ugi, Angew. Chem. 1962, 74, 9–22; c) I. Ugi,
C. Steinbrückner, Chem. Ber. 1961, 94, 734–742; d) I. Ugi, C.
Steinbrückner, Chem. Ber. 1961, 94, 2802–2814.
M.p. 176.1–176.9 °C. IR (KBr): ν = 1676 (s, CO amide), 1632 (s,
˜
CO amide) cm–1. Compound 2a appears as a single conformer at
1
298 K. H NMR (600 MHz, CDCl3): δ = 9.46 (br. s, 1 H), 8.61 (s,
1 H), 8.54 (s, 1 H), 7.36–7.32 (m, 5 H), 4.92 (d, 1 H), 4.77 (d, 1
H), 4.12 (br. s, 1 H), 1.50–1.43 (m, 1 H), 1.28–1.21 (m, 1 H), 1.20–
1.09 (m, 2 H), 0.70 (br. s, 3 H) ppm. 13C NMR (150 MHz, CDCl3):
δ = 170.6 (C), 163.6 (C), 152.8 (CH), 146.2 (C), 143.7 (CH), 136.0
(C), 129.0 (CH), 128.8 (CH), 128.4 (CH), 122.2 (C), 115.3 (C), 63.9
(CH), 54.7 (CH2), 32.7 (CH2), 19.5 (CH2), 13.4 (CH3) ppm. EIMS:
m/z (%) = 387 (100) [M]+, 345 (7) [M – C3H6]+, 296 (37) [M –
C7H7]+. HRMS: calcd. for C18H18BrN3O2 387.0582; found
387.0586.
[4] a) A. Shaabani, A. Maleki, H. Mofakham, J. Comb. Chem.
2008, 10, 595–598; b) R. A. De Silva, S. Santra, P. R. Andre-
ana, Org. Lett. 2008, 10, 4541–4544; c) A. P. Ilyn, A. S. Trifilen-
kov, J. A. Kuzovkova, S. A. Kutepov, A. V. Nikitin, A. V. Iv-
achtchenko, J. Org. Chem. 2005, 70, 1478–1481; d) A. Liu, H.
Zhou, G. Su, W. Zhang, B. Yan, J. Comb. Chem. 2009, 11,
1083–1093; e) M. Sanudo, M. Garcia-Valverde, S. Marcaccini,
J. J. Delgado, J. Rojo, T. Torroba, J. Org. Chem. 2009, 74, 2189–
2192; f) A. Domling, Chem. Rev. 2006, 106, 17–89; g) H. Zhou,
W. Zhang, B. Yan, J. Comb. Chem. 2010, 12, 206–214; h) C.
Hulme, J. Peng, G. Morton, J. M. Salvino, T. Herpin, R. La-
boudiniere, Tetrahedron Lett. 1998, 39, 7227–7230.
Representative Experimental Procedure for 4-Benzyl-1-butyl-7-
phenyl-3-propyl-3,4-dihydro-1H-pyrido[2,3-e][1,4]diazepine-2,5-di-
5400
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5397–5401

Synthesis of Pyridodiazepinediones by Using the Ugi Multicomponent Reaction
[5] a) B. Haché, J. S. Duceppe, P. L. Beaulieu, Synthesis 2002, 528;
b) T. W. Gero, L. W. Jacques, R. P. Mays, D. H. Reid, D. A.
Shamblee, Y. S. Lo, Synth. Commun. 1989, 19, 553–559; c)
A. D. Cale Jr., T. W. Gero, K. R. Walker, Y. S. Lo, W. J.
Welstead Jr., L. W. Jacques, A. F. Johnson, C. A. Leonard, J. C.
Nolan, D. N. Johnson, J. Med. Chem. 1989, 32, 2178–2199.
[6] a) T. A. Keating, R. W. Armstrong, J. Org. Chem. 1996, 61,
8935–8939; b) T. A. Keating, R. W. Armstrong, J. Am. Chem.
Soc. 1996, 118, 2574–2583; c) C. Hulme, J. Peng, S.-Y. Tang,
C. J. Burns, I. Morize, R. Labaudiniere, J. Org. Chem. 1998,
63, 8021–8023.
[7] a) T. Lindhorst, H. Bock, I. Ugi, Tetrahedron 1999, 55, 7411–
7420; b) M. C. Pirrung, S. Ghorai, J. Am. Chem. Soc. 2006,
128, 11772–11773.
[8] A. Domling, I. Ugi, Angew. Chem. Int. Ed. 2000, 39, 3168–
3210.
Received: June 20, 2010
Published Online: August 16, 2010
Eur. J. Org. Chem. 2010, 5397–5401
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5401