A simple and versatile protocol for the preparation of 1,3-functionalized heterocycles utilizing benzoylpyruvates

Article abstract of DOI:10.1002/jhet.448

(Chemical Equation Presented) Additional Supporting Information may be found in the online version of this article. (Chemical Equation Presented) Acid-mediated condensation between benzoylpyruvates and various dinucleophiles in alcoholic solvent furnished the heterocyclic imprint in moderate to good yield. Combining a range of symmetric as well as nonsymmetric nitrogen/nitrogen or nitrogen/carbon centered dinucleophiles resulted in excellent regioselectivity. γ-Difunctionalized fused pyrimidines, pyridazines, and pyridines were produced in this manner. The protocol was designed to obviate chromatographic purification.

Full text of DOI:10.1002/jhet.448

878
A Simple and Versatile Protocol for the Preparation of
1,3-Functionalized Heterocycles Utilizing Benzoylpyruvates
Vol 47
a
Jens M. J. Nolsoe, Anne Ertan, Mats Svensson, and Dirk Weigelt *
b
a
a
¨
a
¨ ¨
Local Discovery Research Area CNS & Pain Control, AstraZeneca R&D Sodertalje,
SE-151 85 So¨derta¨lje, Sweden
Early Development, Pharmaceutical and Analytical R&D, AstraZeneca R&D Sodertalje,
b
¨
¨
¨
¨
Sodertalje SE-151 85, Sweden
*E-mail: dirk.weigelt@astrazeneca.com
Additional Supporting Information may be found in the online version of this article.
Received December 18, 2009
DOI 10.1002/jhet.448
Published online 10 June 2010 in Wiley InterScience (www.interscience.wiley.com).
Acid-mediated condensation between benzoylpyruvates and various dinucleophiles in alcoholic sol-
vent furnished the heterocyclic imprint in moderate to good yield. Combining a range of symmetric as
well as nonsymmetric nitrogen/nitrogen or nitrogen/carbon centered dinucleophiles resulted in excellent
regioselectivity. c-Difunctionalized fused pyrimidines, pyridazines, and pyridines were produced in this
manner. The protocol was designed to obviate chromatographic purification.
J. Heterocyclic Chem., 47, 878 (2010).
INTRODUCTION
(Scheme 1). Considering the number of different ben-
zoylpyruvates 1 readily available from commercial ace-
tophenones 10, this protocol can be seen as an alterna-
tive to arylation of the parent heterocyle via cross-cou-
pling reactions. Indeed, the preparation of prerequisite
substrate for the corresponding cross-coupling may be a
nontrivial matter. In contrast, merited by the ease of
execution, the delineated condensation strategy is ame-
nable for parallel synthesis.
Addition of ambivalent nitrogenous nucleophiles to b-
diketones offers an attractive entry to the preparation of
aromatic heterocyclic structures carrying distinct func-
tional motifs. Acting as a template, the structural fea-
tures of the b-diketone provide the central framework to
be resonated in the resulting aza-heterocycle. We have
in particular directed our attention towards the applica-
tion of benzoylpyruvates 1 as the means to prepare pyri-
dine and pyrimidine derivatives incorporating a c-related
aryl ester pattern.
RESULTS AND DISCUSSION
From a pharmaceutical point of view, the spatial
arrangement of substituents conferred on heterocycles
via benzoylpyruvate chemistry is highly interesting, as it
can give rise to pronounced biological activity. For
example, pyrazoles, isoxazoles, pyrimidines, and pyri-
dines carrying a c-aryl carboxy motif are associated
with such diverse effects as GPCR antagonism [1–4],
ion-channel modulation [5], and kinase inhibition [6,7].
Herein, we want to report a convenient route for the
preparation of novel c-arylated aza-heterocyclic esters
by highly chemo- and regioselective condensation of
lithio benzoylpyruvate 1 with aromatic amines 2–5
As is expected for b-diketones lacking a mirror plane,
the ambiphilic nature of benzoylpyruvates poses a
potential regiochemical problem. However, in the latter
case, the reactivity gets further complicated by the pres-
ence of the ester functionality. Although the ester func-
tionality imparts the adjacent carbonyl with an aug-
mented electron deficiency, it might itself undergo
nucleophilic attack, thereby giving rise to a potential
chemoselective problem as well (Fig. 1) [8].
In general, the stipulated benzoylpyruvate 1 can read-
ily be accessed by reacting the metal enolate of the
C
V
2010 HeteroCorporation

July 2010
A Simple and Versatile Protocol for the Preparation of 1,3-Functionalized
Heterocycles Utilizing Benzoylpyruvates
879
Scheme 1
Scheme 2
Gratifyingly, when first having treated Li-1 with two
equivalents of hydrochloric acid, subsequent cyclodehy-
dration could be realized for a range of aromatic amines
2–5 vide infra. Furthermore, the reaction displayed a
distinct regiochemical preference in each case.
In accordance with the delineated protocol, a selection
of five-membered aza-heterocyclic amines 2 was reacted
with Li-1 to furnish the corresponding [a]-fused pyrimi-
dines 6 (Scheme 3). However, the influence exerted by
the individual N,N-dinucleophiles on the success of
cyclodehydration implicated a dependency on the pKa of
the protonated parent heterocyclic system (Table 1). To
expand upon the latter point, the pKa-values of pyrazole,
1,3,4-triazole, and tetrazole are 2.5, 2.2, and –3.0,
respectively [12,13], whereas the pKa of imidazole is
7.0 [12]. This may provide an explanation for the failure
of imidazole 2b and benzimidazole 2f to react under the
given conditions.
corresponding acetophenone 10 with diethyl oxalate 11
[9–11]. In particular, owing to the excellent chelating
ability of b-diketones, the en route lithium salt of ben-
zoylpyruvate 1 can usually be isolated as a shelf-stabile
solid by simple filtration [11]. Thus, sequential treatment
of 10 with LDA and diethyl oxalate 11 was the pre-
ferred method to prepare the needed starting material
(Scheme 2). Initially, we opted for the p-Br-substituted
benzoylpyruvate Li-1 as a model compound, expecting
it to be conducive in terms of product identification,
based on its isotopic fingerprint (LC-MS) and crystallo-
graphic properties (X-ray). However, in some cases we
found that the parent benzoylpyruvate Li-1 (R ¼ H)
proved superior for crystallographic purposes.
By the means of single-crystal diffraction technique,
it was subsequently possible to make an unambiguous
regiochemical assignment of 6b (Fig. 2), resulting from
cyclodehydration with pyrazole 2a. Curiously, a reversal
of regiochemistry takes place when interchanging the
ester moiety in Li-1 for a CF3-group, i.e., turning the
system into a conventional b-diketone. Thus, applying a
similar protocol as ours, Filyakova et al. have reported
on the preparation of an analogous 5-phenyl-7-(trifluoro-
methyl)pyrazolo[1,5-a]pyrimidine, which was verified
by crystallography [14]. The two opposing results indi-
cate the ability of the ester to act as regiochemical han-
dle (vide infra).
Although the binding of lithium was beneficial with
regard to the isolation of 1, the chelated form proved to
have an adverse effect on the cyclodehydration. Without
in situ quenching of Li-1, concomitant retro-aldol reac-
tion was observed on the application of any prospect
dinucleophile in refluxing ethanol. Consequently, the ini-
tial attempts only returned acetophenone 10 as the
breakdown product.
With respect to regiochemistry, the fused system
obtained through condensation of tetrazole 2d posed the
intriguing possibility of valence tautomery [15]. At the
outset, it was expected that the initially formed tetra-
zolo[1,5-a]pyrimidines 6f and 6g could undergo
Scheme 3
Figure 1. Tautomery in benzoylpyruvates. [Color figure can be viewed
in the online issue, which is available at www.interscience.wiley.com.]
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

¨
J. M. J. Nolsoe, A. Ertan, M. Svensson, and D. Weigelt
880
Vol 47
Table 1
Synthesis of [a]-fused pyrimidines 6.a
Entry
Product
R
Aryl amine
Time (h)c
Yield (%)d
1
2
6a
6b
6c
6d
6e
6fb
6gb
6h
6i
p-Br
H
1H-Pyrazol-5-yl 2a
1H-Pyrazol-5-yl 2a
13
13
96e.
48
48
13
13
13
13
96e.
75
69
n.r.
73
70
79
68
85
87
n.r.
3
4
p-Br
p-Br
H
1H-Imidazol-5-yl 2b
1H-1,2,4-Triazol-5-yl 2c
1H-1,2,4-Triazol-5-yl 2c
1H-Tetrazol-5-yl 2d
1H-Tetrazol-5-yl 2d
2H-Indazol-3-yl 2e
5
6
7
p-Br
H
8
9
10
p-Br
H
2H-Indazol-3-yl 2e
1H-Benzimidazol-2-yl 2f
6j
p-Br
a The reactions were performed with equimolar proportions of Li-1 and amine 2 in the presence of 2 equiv. of HCl (aq).
b Interconverts to the corresponding azide 12.
c Time to achieve full conversion according to HPLC and LC-MS.
d Isolated yield.
e Experiment was terminated at the time indicated as no reaction had occurred.
equilibration to 2-azidopyrimidines 12f, g (Fig. 3). How-
ever, the ‘‘pseudo symmetry’’ of 12f, g would then ob-
literate the existing regiochemical preference via the
erosive interconversion to 13f, g. Subsequently, it was
indeed demonstrated by the application of X-ray crystal-
lography that in the solid state the tentatively assigned
structure of fused tetrazole 6f actually was 2-azidopyri-
midine 12f (Fig. 4).
tional C,N-dinucleophiles via blocking or replacing the
active ring-nitrogen. Hence, it was envisioned that the
‘‘aniline-type’’ aryl amines 4 would furnish [b]-fused
pyridines. Although the primary projection was correct,
the regiochemistry was at variance with the preceding
examples. Thus, when Li-1 was reacted with C,N-dinu-
cleo-philes 4, the resultant cyclodehydration provided
the [b]-fused pyridine 8 (Scheme 5).
Having established the feasibility of cyclodehydration
on Li-1 utilizing selected N,N-dinucleophiles, the focus of
the protocol was aimed at fusion with other types of five-
membered aza-heterocylic amines, i.e., prospective C,N-
dinucleophiles. Following this line of reasoning, Li-1 was
allowed to react with N-amino heterocycles 3 to provide
[b]-fused pyridazines 7 (Scheme 4). In the selected cases,
the observed regiochemistry corresponded to the results
obtained with aryl amines 2. Here, however, the degree of
aza-functionalization had a dramatic influence on the effi-
ciency, with regard to product formation (Table 2).
Applied on the product originating from condensation
between Li-1 (R ¼ H) and ‘‘enamine’’ 4a, X-ray crys-
tallography provided the conclusive structural informa-
tion, showing 8b to be the resulting regioisomer (Fig. 5).
Though, seemingly contradictory, with regard to ben-
zoylpyruvate 1, the sequence of addition leading to
the different fused systems could well be the same (vide
supra). Instead, it was surmised that the origin of regio-
chemical divergence resided on the nature of the applied
dinucleophile. Depending on the hard/soft character of
the aryl amine, in analogy with enamines, either the
interconnected nitrogen or carbon could take precedence
in the cyclodehydration. However, it also became clear
Five-membered aza-heterocycles with an encased eth-
ylene amine motif could in principle be rendered opera-
Figure 2. Crystal structure of compound 6b determined by single-crys-
tal diffraction technique at 200 K. [Color figure can be viewed in the
online issue, which is available at www.interscience.wiley.com.]
Figure 3. Valence tautomery in tetrazolo[1,5-a]pyrimidine 6f and 6g.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

July 2010
A Simple and Versatile Protocol for the Preparation of 1,3-Functionalized
Heterocycles Utilizing Benzoylpyruvates
881
Figure 4. ORTEP presentation of compound 12f.
that the choice of C,N-dinucleophiles 4 was rather lim-
ited (Table 3).
By extending the ensemble of C,N-dinucleophiles to
six-membered aryl amines 5, i.e., bona fide anilines, it
was anticipated that the regiochemistry would realign
with the initially observed preference to provide [b]-
fused pyridines 9 (Scheme 6). However, if the underly-
ing assumption regarding the regiochemical preference
of benzoylpyruvate 1 proved correct, the resulting qui-
nolines 9 would have the reversed structure compared
with the Skraup–Doebner–von Miller protocol [15].
Upon reacting Li-1 with anilines 5, it became evident
that cyclodehydration could only be achieved when
strongly electron donating substituents were present on
the aryl moiety of the prospective C,N-dinucleophile
(Table 4). The interplay of electronic properties is how-
ever more subtle than what might initially be gleaned
from the reaction scheme. Thus, when Li-1 was treated
Figure 5. Crystal structure of compound 8b determined by single-crys-
tal diffraction technique at 200 K. [Color figure can be viewed in the
online issue, which is available at www.interscience.wiley.com.]
with 4-amino-1H-indole 5d, cyclization proceeded
smoothly (entry 8 and 9). While, in contrast, the reac-
tion with 6-amino-1H-indole 5e was exceedingly slug-
gish (entry 11). Yet in both instances, the position ortho
to the aniline functionality is activated by an indole-
nitrogen (Fig. 6).
Another aspect of the 4-amino-1H-indole derivatives 9f
and 9g is the comprehensive prototropic tautomery avail-
1
able to the fused rings. Thus, in the pertinent case, H
NMR indicates that the extended 14 p-electron system
Scheme 4
Scheme 5
Table 2
Synthesis of [b]-fused pyridazines 7.a
Table 3
Synthesis of [b]-fused pyridines 8.a
Time Yield
(%)c
Time Yield
(h)b (%)c
Entry Product
R
Aryl amine
(h)b
Entry Product
R
Aryl amine
1
2
3
4
7a
7b
7c
7d
p-Br
H
Pyrrol-1-yl 3a
Pyrrol-1-yl 3a
0.5
0.5
85
79
71
65
1
2
3
4
8a
8b
8c
8d
p-Br 2-Methyl-2H-pyrazol-3-yl 4a 120 70
2-Methyl-2H-pyrazol-3-yl 4a 120 70
p-Br 5-Methylisox-azole-3-yl 4b 120 n.r.
p-Br 3-Methylisothi-azol-5-yl 4c 120 n.r.
H
p-Br 1,2,4-Triazol-4-yl 3b 120
1,2,4-Triazol-4-yl 3b 120
H
a The reactions were performed with equimolar proportions of Li-1 and
amine 3 in the presence of 2 equiv. of HCl (aq).
b Time to achieve full conversion HPLC or LC-MS.
c Isolated yield.
a The reactions were performed with equimolar proportions of Li-1 and
amine 4 in the presence of 2 equiv. of HCl (aq).
b Time to achieve full conversion.
c Isolated yield.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

¨
J. M. J. Nolsoe, A. Ertan, M. Svensson, and D. Weigelt
882
Vol 47
Scheme 6
Table 4
Synthesis of [b]-fused pyridines 9.a
Time Yield
(h)b (%)c
Entry Product
R
Aryl amine
1
2
3
4
5
6
7
8
9
9a
9b
9c
9d
9e
9f
9g
9h
9i
p-Br 3-Methoxy-phenyl 5a
p-Br 3,5-Dimethoxy-phenyl 5b
120 n.r.
13
13
82
73
90
72
H
3,5-Dimethoxy-phenyl 5b
p-Br 3-(Dimethylamino)-phenyl 5c 13
H
3-(Dimethylamino)-phenyl 5c 13
p-Br 1H-Indol-4-yl 5d
13 25d
13 23d
H
1H-Indol-4-yl 5d
p-Br 1H-Indol-6-yl 5e
1H-Indol-6-yl 5e
n.d. n.d.
120 28d
Figure 6. Tricycles resulting from condensation with 4-amino-1H-
indole 5e and 6-amino-1H-indole 5f.
H
a The reactions were performed with equimolar proportions of Li-1 and
amine 5 in the presence of 2 equiv. of HCl (aq).
b Time to achieve full conversion or no further reaction occurred by
HPLC or LC-MS.
diketones [14], we wanted to make a side-by-side com-
parison with benzoylpyruvate 1, applying high-level QM
calculations to gain insight into the origin of the
observed regio-divergence.
c Isolated yield.
d Preparative HPLC due to instability.
Discounting the reactive nature of the ester moiety, a
cursory inspection of benzoylpyruvate 1 might pin down
the system as merely a benzoylacetone with an electron
withdrawing group append to its terminus. Indeed from
behaves less aromatic than anticipated. The repositioning
of the active proton also involves carbon prototropes and
the only specie observed by NMR-spectroscopy was attrib-
uted to the 4H-tautomer (Fig. 7). This behavior probably
reflects the ‘‘quinone-like’’ nature of the compound.
By comparison, the linear isomer 9i resulting from con-
densation of Li-1 (R ¼ H) with 6-amino-1H-indole 5e
behaves more like an aromatic 14 p-electron system and
prototropic tautomery is less pronounced. However, in
general the quinolines 9 were less stabile than the previ-
ous classes of cyclodehydration products 6, 7, and 8.
With regard to the identity of the resulting com-
pounds, we once again resolved to address the issue of
regiochemical preference by capitalizing on X-ray crys-
tallography. Applied to the cyclodehydration of Li-1 (R
¼ Br) with 3-(N,N-dimethylamino)aniline 5d, the struc-
ture was shown to be 9d (Fig. 8). Thus, the condensa-
tion between benzoylpyruvates 1 and anilines 5 provide
verily regioisomeric quinolines not attainable via the
classic Skraup–Doebner–von Miller procedure [15].
CALCULATIONS
In the light of some contrasting findings published by
Filyakova et al. regarding condensation on selected b-
Figure 7. Tautomery in pyrrolo[2,3-h]quinolines 9f and 9g.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

July 2010
A Simple and Versatile Protocol for the Preparation of 1,3-Functionalized
Heterocycles Utilizing Benzoylpyruvates
883
Figure 8. Crystal structure of compound 9d determined by single-crys-
tal diffraction technique at 200 K. [Color figure can be viewed in the
online issue, which is available at www.interscience.wiley.com.]
Figure 10. Calculated relative energies of protonated model benzoyl-
pyruvate 15 based on B3LYP/6-31G**. [Color figure can be viewed in
the online issue, which is available at www.interscience.wiley.com.]
this vantage point, benzoylpyruvate 1 compares well
with the fluorinated b-diketone 14 used by Filyakova
et al. [14]. Thus, in both cases, the keto-functionalities
might be projected to exhibit ‘‘pseudo-degenerate’’
behavior (Fig. 9). As an initial conjecture, one would
consequently expect cyclodehydration on benzoylpyru-
vate 1 and fluorinated b-diketone 14 to give rise to simi-
lar regioselectivity. Indeed, our calculations on the rela-
tive energies of protonated b-diketone 14 only differ
with 0.3 kcal/mol for the two positions. This is within
the error margin of the applied method [16].
CRYSTALLOGRAPHY
Data
collection,
structure
solution,
and
refinement. Diffraction data for (6b), (8b), (9f), and
(12a) were collected at 200(2) K using either a Nonius
Kappa-CCD or a Bruker APEX-II CCD diffractometer
with graphite-monochromated Mo Ka radiation (k ¼
˚
0.71073 A). The structures were solved by direct meth-
ods and refined with F2 against all reflections. All but
one had one molecule in the asymmetric unit, which for
8b contained two crystallographically unique molecules.
Figures 2–5 and 8 show the molecular conformation,
with the atom-labeling schemes. Compounds 6b, 8b, 9d,
and 12a do not contain H-bond donor atoms and conse-
quently do not form classical H-bonds [18].
Taking into account, the participation of the ester
moiety in benzoylpyruvate 1 alters the picture dramati-
cally: The ability of the ester function to act as a hydro-
gen bond acceptor facilitates formation of a five-mem-
bered chelate with the adjacent carbonyl. According to
the performed calculations, using methyl derivative 15
as a model compound, the individual protonated forms
of the b-diketo system alone are energetically compara-
ble. However, when the protonated forms involving a
five-membered chelate are introduced, the energy of the
system is markedly lower by 2.6 and 3.2 kcal/mol,
respectively (Fig. 10) [16,17]. Thus, these calculations
lend support to the hypothesis that the ester moiety in
benzoylpyruvate 1 may serve as a regiochemical handle.
CONCLUSIONS
This article outlines a simple and versatile approach
to furnish fused heterocycles grafted on the benzoylpyr-
uvate backbone. Depending on the nature of the applied
dinucleophilic class, an alternation of regiochemistry
could be observed. However, in each case the process
was highly regioselective. At the basis of the observed
selectivity, we postulate the participation of the ester
portion in benzoylpyruvates as a decisive factor.
EXPERIMENTAL
NMR spectra were recorded either on a Bruker DPX400
NMR spectrometer operating at 400 MHz for 1H, and 101
MHz for 13C equipped with a four-nucleus probe-head with Z-
gradients, or a Bruker Avance III 500 NMR spectrometer,
1
Figure 9. Pseudo-degenerate nature of the b-diketo system.
operating at 500 MHz for H and 126 MHz for 13C equipped
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

¨
J. M. J. Nolsoe, A. Ertan, M. Svensson, and D. Weigelt
884
Vol 47
with a 5mm TCI cryo probe-head with Z-gradients. Chemical
shifts are given in ppm down- and upfield from tetramethylsi-
lane (0.00 ppm). DMSO-d6 dH 2.49; dC 39.51 and CDCl3 dH
7.27; dC 77.00 were used as reference signals. All experiments
were performed at a sample temperature of 26ꢀC 6 2ꢀC. LC-
MS analyses were performed on a LC-MS system consisting
of a Waters Alliance 2795 HPLC, a Waters PDA 2996 diode
array detector, a Sedex 75 ELS detector, and a ZQ 2000 single
quadrupole mass spectrometer. The mass spectrometer was
equipped with an electrospray ion source (ES) operated in pos-
itive and negative ion mode. The capillary voltage was set to
3.3 kV and the cone voltage to 28 V, respectively. The mass
spectrometer was scanned between m/z 100–700 with a scan
time of 0.3 s. The diode array detector scanned from 200–400
nm. The temperature of the ELS detector was adjusted to
40ꢀC and the pressure was set to 1.9 bar. Separation was per-
formed on Gemini C18 3.0 ꢁ 50, 3 lm (Phenomenex) run at a
flow rate of 1 mL/min. A linear gradient was applied starting
at 100% A (A: 10 mM ammonium acetate in 5% acetonitrile)
ending at 100% B (B: acetonitrile) in 4 min followed by 100%
B until 5.5 min. The column oven temperature was set to
40ꢀC. HPLC analyses were performed on an Agilent HP1100
system consisting of a G1322A Micro Vacuum Degasser, a
G1311A Quaternary Pump, a G1367 Well-Plate Autosampler,
a G1316A Thermostated Column Compartment and a G1315A
Diode Array Detector. The diode array detector was scanned
from 200 to 400 nm, step and peak width were set to 2 nm
and 0.01 min, respectively. The column used was Gemini C18,
resulting homogeneous mixture was heated and refluxed for
the times given (vide supra). After heating, the reaction mix-
ture was allowed to slowly cool to ambient temperature. In
case of precipitation, the deposited material was collected,
washed with ethanol, and the mother liquor was evaporated in
vacuo. Upon evaporation, the remnant was recrystallized from
ethanol and further cropped.
If no initial precipitation occurred, the reaction mixture
was evaporated in vacuo, whereupon the remnant was dis-
solved in chloroform and washed with dilute aqueous sodium
hydroxide. The organic phase was subsequently evaporated in
vacuo and the remnant was dissolved in hot methanol, etha-
nol, or extracted with boiling hexane. Subsequently, upon
cooling crystallization ensued and the deposited material was
cropped.
Ethyl 4-(4-bromophenyl)-1H-pyrazolo[3,4-b]pyridine-6-car-
boxylate (6a). The title compound was obtained by the reac-
tion of lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid ethyl
ester (Li-1, R ¼ Br) and 3-aminopyrazole (2a). Yield: 0.259 g
1
(75%). H NMR (CDCl3): d 1.49 (t, 3H, J ¼ 7.2 Hz), 4.56 (q,
2H, J ¼ 7.1 Hz), 7.05 (d, 1H, J ¼ 2.3 Hz), 7.70 (s, 1H), 7.7–
7.8 (m, 2H), 8.0–8.1 (m, 2H), 8.28 (d, 1H, J ¼ 2.3 Hz). 13C
NMR (CDCl3): d 14.3, 62.8, 100.0, 106.7, 126.1, 129.4, 130.8,
132.0, 145.98, 146.01, 146.3, 148.9, 164.0. HRMS: Found
346.0198, Calcd. for C15H13BrN3O2: 346.0191.
Ethyl
4-phenyl-1H-pyrazolo[3,4-b]pyridine-6-carboxylate
(6b). The title compound was obtained by the reaction of lithio
2,4-dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H) and
˚
3.0 ꢁ 50 mm, 3.0 lm, 110 A run at a flow rate of 1.0 mL/
3-aminopyrazole (2a). Yield: 0.185
g
(69%). 1H NMR
min. The column oven temperature was set to 40ꢀC. A linear
gradient was applied, starting at 100% A (A: 10 mM ammo-
nium acetate in 5% acetonitrile) and ending at 95% B (B: ace-
tonitrile) after 6.5 min then 95% B for 0.5 min. High resolu-
tion mass spectra (HRMS) were recorded on a Micromass Q-
Tof micro mass spectrometer equipped with a LockSpray
source and connected to an Acquity UPLC system with a PDA
detector. All analyses were acquired using positive mode elec-
trospray ionisation (ESIþ) in full scan and Leucine Enkephalin
(Sigma) was used as the lock mass (m/z 556.2771) at a con-
centration of 0.9 pmol/lL and a flow rate of 100 lL/min with
a 1:10 split, ion source:waste. Cone Voltage was set to 54 to
achieve ꢂ 200 counts for Leucine Enkephalin. Nitrogen was
used as the nebulizing and desolvation gas, with the source
operated at 120ꢀC and the desolvation gas at 300ꢀC. Capillary
voltage was about 3000 V and cone voltage was about 30 V.
Argon was used as the collision gas. Chromatographic separa-
tion for HRMS was achieved with a 2.3 min linear gradient
from 95% A (A: 10 mM ammonium acetate in MilliQ water þ
5% acetonitrile) to 95% B (B: acetonitrile) over an ACQUITY
UPLC BEH C18 1.7 lm, 2.1 ꢁ 50 mm column maintained at
65ꢀC and run at a flow rate of 0.7 mL/min with a 1:5 split,
ion source:waste. Analytes were diluted in H2O:ACN (50:50)
until suitable concentration for the LC-MS analysis.
(CDCl3): d 1.50 (t, 3H, J ¼ 7.1 Hz), 4.57 (q, 2H, J ¼ 7.1 Hz),
7.05 (d, 1H, J ¼ 2.5 Hz), 7.6–7.7 (m, 3H), 7.71 (s, 1H), 8.1–
8.2 (m, 2H), 8.29 (d, 1H, J ¼ 2.3 Hz). 13C NMR (CDCl3): d
14.3, 62.7, 99.8, 106.9, 128.8, 129.3, 130.6, 131.4, 146.0,
146.4, 147.2, 149.0, 164.1. HRMS: Found 268.1097, Calcd.
for C15H14N3O2: 268.1086.
Ethyl 7-(4-bromophenyl)-[1,2,4]triazolo[1,5-a]pyrimidine-5-
carboxylate (6d). The title compound was obtained by the
reaction of lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid
ethyl ester (Li-1, R ¼ Br) and 3-amino-1,2,4-triazole (2c).
1
Yield: 0.255 g (73%). H NMR (CDCl3): d 1.50. (t, 3H, J ¼
7.2 Hz), 4.57 (q, 2H, J ¼ 7.1 Hz), 7.78 (d, 2H, J ¼ 8.6 Hz),
8.03 (s, 1H), 8.11 (d, 2H, J ¼ 8.60 Hz), 8.70 (s, 1H). 13C
NMR (CDCl3): d 14.2, 63.2, 108.6, 127.4, 128.0, 130.9, 132.4,
147.8, 151.9, 155.7, 157.6, 163.5. HRMS: Found 347.0151,
Calcd. for C14H12BrN4O2: 347.0144.
Ethyl 7-phenyl-[1,2,4]triazolo[1,5-a]pyrimidine-5-carboxy-
late (6e). The title compound was obtained by the reaction of
lithio 2,4-dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H)
and 3-amino-1,2,4-triazole (2c). Yield: 0.187 g (70%). 1H
NMR (CDCl3): d 1.40 (t, 3H, J ¼ 7.1 Hz), 4.57 (q, 2H, J ¼
7.1 Hz), 7.6–7.7 (m, 3H), 8.04 (s, 1H), 8.1–8.2 (m, 2H), 8.70
(s, 1H). 13C NMR (CDCl3): d 14.2, 63.1, 108.8, 129.1, 129.3,
129.5, 132.4, 149.0, 151.8, 155.7, 157.6, 163.5. HRMS: Found
269.1039, Calcd. for C14H13N4O2: 269.1039.
Ethyl 2-azido-6-(4-bromophenyl)pyrimidine-4-carboxylate
(12f). The title compound was obtained by the reaction of
lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid ethyl ester (Li-
1, R ¼ Br) and 5-aminotetrazole (2d). Yield: 0.277 g (79%).
1H NMR (CDCl3): d 1.46 (t, 3H, J ¼ 7.1 Hz), 4.51 (q, 2H, J
¼ 7.2 Hz), 7.6–7.7 (m, 2H), 8.0–8.1 (m, 2H), 8.11 (s, 1H). 13C
General methods for the condensation of benzoylpyru-
vates (Li-1, R ¼ H, Br) with dinucleophiles (2–5). Lithio
2,4-dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H)
(0.226 g, 1.00 mM) or lithio 2,4-dioxo-4-(40-bromophenyl)bu-
tyric acid ethyl ester (Li-1, R ¼ Br) (0.305 g, 1.00 mM) and
the appropriate dinucleophile (2–5) (1.00 mM) were mixed to-
gether, suspended in ethanol (5.0 mL) and conc. aqueous hy-
drochloric acid (166 lL, 12.0M, 2.00 mM) was added. The
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

July 2010
A Simple and Versatile Protocol for the Preparation of 1,3-Functionalized
Heterocycles Utilizing Benzoylpyruvates
885
NMR (CDCl3): d 14.2, 62.9, 111.9, 127.2, 128.9, 132.4, 133.9,
158.2, 163.0, 163.7, 167.0. HRMS: Found 348.0099, Calcd.
for C13H11BrN5O2: 348.0096.
1H), 8.3–8.4 (m, 2H), 9.33 (s, 1H). 13C NMR (CDCl3): d 14.2,
63.5, 115.3, 126.8, 130.1, 130.7, 132.5, 136.5, 139.9, 142.9
146.3, 162.2. HRMS: Found 347.0151, Calcd. for
C14H12BrN4O2: 347.0144.
Ethyl 8-phenyl-[1,2,4]triazolo[3,4-f]pyridazine-6-carboxy-
late (7d). The title compound was obtained by the reaction of
lithio 2,4-dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H)
and 4-amino-1,2,4-triazole (3b). Yield: 0.174 g (65%). 1H
NMR (CDCl3): d 1.51 (t, 3H, J ¼ 7.1 Hz), 4.58 (q, 2H, J ¼
7.2 Hz), 7.5–7.6 (m, 3H), 8.03 (s, 1H), 8.3–8.4 (m, 2H), 9.33
(s, 1H). 13C NMR (CDCl3): d 14.2, 63.4, 115.5, 129.2, 129.3,
131.4, 131.8, 137.8, 139.8, 143.2 146.3, 162.3. HRMS: Found
269.1043, Calcd. for C14H13N4O2: 269.1039.
Ethyl 2-azido-6-phenyl-pyrimidine-4-carboxylate (12g). The
title compound was obtained by the reaction of lithio 2,4-
dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H) and 5-
1
aminotetrazole (2d). Yield: 0.182 g (68%). H NMR (CDCl3):
d 1.47 (t, 3H, J ¼ 7.1 Hz), 4.51 (q, 2H, J ¼ 7.1 Hz), 7.5–7.6
(m, 3H), 8.15 (s, 1H), 8.2 (m, 2H). 13C NMR (CDCl3): d 14.1,
62.8, 112.2, 127.5, 129.1, 132.2, 134.5, 157.9, 162.9, 163.8,
168.2. HRMS: Found 270.0993, Calcd. for C13H12N5O2:
270.0991.
Ethyl 4-(4-bromophenyl)pyrimido[1,2-b]indazole-2-carboxy-
late (6h). The title compound was obtained by the reaction of
lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid ethyl ester (Li-
1, R ¼ Br) and 3-aminoindazole (2e). Yield: 0.339 g (85%).
1H NMR (CDCl3): d 1.54 (t, 3H, J ¼ 7.1 Hz), 4.60 (q, 2H, J
¼ 7.2 Hz), 7.4–7.5 (m, 1H), 7.7–7.8 (m, 1H), 7.8–7.9 (m, 2H),
7.9–8.0 (m, 1H), 8.09 (s, 1H), 8.1–8.2 (m, 2H), 8.5–8.6 (m,
1H). 13C NMR (CDCl3): d 14.4, 62.7, 110.9, 114.7, 117.0,
121.6, 122.4, 126.0, 129.7, 130.6, 131.0, 132.2, 141.7, 143.5,
144.4, 151.7, 164.1. HRMS: Found 396.0349, Calcd. for
C19H15BrN3O2: 396.0348.
Ethyl 6-(4-bromophenyl)-1-methyl-pyrazolo[3,4-b]pyridine-
4-carboxylate (8a). The title compound was obtained by the
reaction of lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid
ethyl ester (Li-1, R ¼ Br) and 3-amino-2-methyl-2H-pyrazole
1
(4a). Yield: 0.253 g (70%). H NMR (CDCl3): d 1.53 (t, 3H, J
¼ 7.1 Hz), 4.25 (s, 3H), 4.56 (q, 2H, J ¼ 7.1 Hz), 7.6–7.7 (m,
2H), 8.0–8.1 (m, 2H), 8.21 (s, 1H), 8.39 (s, 1H). 13C NMR
(CDCl3): d 14.3, 34.0, 62.0, 111.9, 114.6, 124.3, 128.9, 131.89,
131.93, 132.5, 137.2, 151.5, 155.3, 165.1. HRMS: Found
360.0342, Calcd. for C16H15BrN3O2: 360.0348.
Ethyl 1-methyl-6-phenyl-pyrazolo[3,4-b]pyridine-4-carboxy-
late (8b). The title compound was obtained by the reaction of
lithio 2,4-dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H)
and 3-amino-2-methyl-2H-pyrazole (4a). Yield: 0.196 g (70%).
1H NMR (CDCl3): d 1.52 (t, 3H, J ¼ 7.1 Hz), 4.25 (s, 3H),
4.54 (q, 2H, J ¼ 7.1 Hz), 7.5–7.6 (m, 3H), 8.2 (m, 2H), 8.23
(s, 1H), 8.38 (s, 1H). 13C NMR (CDCl3): d 14.3, 34.1, 61.9,
111.7, 115.2, 127.5, 128.9, 129.7, 128.9, 131.8, 132.5, 138.5,
151.7, 156.7, 165.3. HRMS: Found 282.1249, Calcd. for
C16H16N3O2: 282.1243.
Ethyl 4-phenylpyrimido[1,2-b]indazole-2-carboxylate (6i). The
title compound was obtained by the reaction of lithio 2,4-
dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H) and 3-
1
aminoindazole (2e). Yield: 0.277 g (87%). H NMR (CDCl3):
d 1.54 (t, 3H, J ¼ 7.1 Hz), 4.60 (q, 2H, J ¼ 7.1 Hz), 7.4–7.5
(m, 1H), 7.6–7.8 (m, 4H), 7.9–8.0 (m, 1H), 8.11 (s, 1H), 8.2–
8.3 (m, 2H), 8.4–8.6 (m, 1H). 13C NMR (CDCl3): d 14.4,
62.6, 111.2, 114.7, 117.0, 121.6, 122.2, 128.9, 129.5, 130.4,
131.0, 131.4, 141.8, 144.4, 144.7, 151.8, 164.2. HRMS: Found
318.1250, Calcd. for C19H16N3O2: 318.1243.
Ethyl 4-(4-bromophenyl)-5,7-dimethoxy-quinoline-2-car-
boxylate (9b). The title compound was obtained by the reac-
tion of lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid ethyl
ester (Li-1, R ¼ H) and 3,5-dimethoxyaniline (5b). Yield:
Ethyl 4-(4-bromophenyl)pyrrolo[2,1-f]pyridazine-2-carboxy-
late (7a). The title compound was obtained by the reaction of
lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid ethyl ester (Li-
1
1, R ¼ Br) and 1-aminopyrrole (3a). Yield: 0.292 g (85%). H
1
0.342 g (82%). H NMR (CDCl3): d 1.46 (t, 3H, J ¼ 7.1 Hz),
NMR (CDCl3): d 1.48 (t, 3H, J ¼ 7.1 Hz), 4.54 (q, 2H, J ¼
7.1 Hz), 6.72 (dd, 1H, J ¼ 4.3 Hz, 1.5 Hz), 7.06 (dd, 1H, J ¼
4.4 Hz, 2.9 Hz), 7,28 (s, 1H) 7.6–7.8 (m, 4H), 8.02 (dd, 1H, J
¼ 2.8 Hz, 1.5 Hz). 13C NMR (CDCl3): d 14.3, 62.3, 100.5,
108.0, 115.7, 119.1, 124.0, 125.4, 129.6, 132.1, 135.0, 139.6,
142.2, 163.9. HRMS: Found 345.0248, Calcd. for
C16H14BrN2O2: 345.0239.
3.53 (s, 3H), 3.96 (s, 3H), 4.53 (q, 2H, J ¼ 7.1 Hz), 6.55 (d,
1H, J ¼ 2.3 Hz), 7.1–7.2 (m, 2H), 7.38 (d, 1H, J ¼ 2.3 Hz),
7.5–7.6 (m, 2H), 7.76 (s, 1H). 13C NMR (CDCl3): d 14.3,
55.3, 55.8, 62.3, 101.2, 101.3, 115.7, 121.0,121.3, 129.8,
130.2, 140.8, 147.2, 148.2, 150.9, 156.6, 161.6, 165.0. HRMS:
Found 416.0509, Calcd. for C20H19BrNO4: 416.0497.
Ethyl 5,7-dimethoxy-4-phenyl-quinoline-2-carboxylate (9c). The
title compound was obtained by the reaction of lithio 2,4-
dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H) and 3,5-
Ethyl 4-phenylpyrrolo[2,1-f]pyridazine-2-carboxylate (7b). The
title compound was obtained by the reaction of lithio 2,4-
dioxo-4-(40-bromophenyl)butyric acid ethyl ester (Li-1, R ¼
H) and 1-aminopyrrole (3a). Yield: 0.211 g (79%). 1H NMR
(CDCl3): d 1.48 (t, 3H, J ¼ 7.2 Hz), 4.54 (q, 2H, J ¼ 7.1 Hz),
6.77 (dd, 1H, J ¼ 4.4 Hz, 1.4 Hz), 7.05 (dd, 1H, J ¼ 4.3 Hz,
2.8 Hz), 7.32 (s, 1H), 7.5–7.6 (m, 3H), 7.7–7.8 (m, 2H), 8.02
(dd, 1H, J ¼ 2.8 Hz, 1.3 Hz). 13C NMR (CDCl3): d 14.3,
62.3, 100.6, 108.1, 115.6, 118.9, 125.8, 128.1, 128.9, 129.7,
136.1, 140.9, 142.2, 164.1. HRMS: Found 267.1133, Calcd.
for C16H15N2O2: 267.1134.
dimethoxyaniline (5b). Yield: 0.248
g
(73%). 1H NMR
(CDCl3): d 1.47 (t, 3H, J ¼ 7.2 Hz), 3.51 (s, 3H), 3.97 (s,
3H), 4.54 (q, 2H, J ¼ 7.1 Hz), 6.55 (d, 1H, J ¼ 2.3 Hz), 7.3
(m, 2H), 7.35 (d, 1H, J ¼ 2.0 Hz), 7.4 (m, 3H), 7.83 (s, 1H).
13C NMR (CDCl3): d 14.4, 55.3, 55.8, 62.2, 101.1, 101.5,
116.0, 121.3, 127.0, 127.1, 128.1, 142.0, 147.4, 149.3, 151.1,
156.9, 161.3, 165.3. HRMS: Found 338.1389, Calcd. for
C20H20NO4: 338.1392.
Ethyl
4-(4-bromophenyl)-7-(dimethylamino)quinoline-2-
Ethyl 8-(4-bromophenyl)-[1,2,4]triazolo[3,4-f]pyridazine-6-
carboxylate (7c). The title compound was obtained by the
reaction of lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid
ethyl ester (Li-1, R ¼ Br) and N-4-amino-1,2,4-triazole (3b).
Yield: 0.248 g (71%). 1H NMR (CDCl3): d 1.52 (t, 3H, J ¼
7.1 Hz), 4.59 (q, 2H, J ¼ 7.2 Hz), 7.7–7.8 (m, 2H), 8.03 (s,
carboxylate (9d). The title compound was obtained by the
reaction of lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid
ethyl ester (Li-1, R ¼ H) and 3-(N,N-dimethylamino)aniline
1
(5c). Yield: 0.360 g (90%). H NMR (CDCl3): d 1.48 (t, 3H, J
¼ 7.1 Hz), 3.13 (s, 6H), 4.54 (q, 2H, J ¼ 7.1 Hz), 7.21 (dd,
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

¨
J. M. J. Nolsoe, A. Ertan, M. Svensson, and D. Weigelt
886
Vol 47
1H, J ¼ 9.4 Hz, 2.8 Hz), 7.4 (m, 2H), 7.42 (br. s, 1H), 7.6 (m,
2H), 7.72 (d, 1H, J ¼ 9.4 Hz), 7.80 (s, 1H). 13C NMR
(CDCl3): d 14.4, 40.4, 62.1, 108.0, 117.3, 118.4, 119.8, 122.8,
125.8, 131.1, 131.7, 137.0, 147.8, 150.0, 151.3, 165.7. HRMS:
Found 399.0711, Calcd. for C20H20BrN2O2: 399.0708.
Acknowledgments. The authors are much indebted to Dr. Zara
Sands for initial discussions on the calculations, to Dr. Alexandra
Bernlind for generously providing her expertise on NMR, and to
Mrs. Fanny Bjarnemark for performing the HRMS analysis con-
tained within this article.
Ethyl 7-(dimethylamino)-4-phenyl-quinoline-2-carboxylate
(9e). The title compound was obtained by the reaction of lithio
2,4-dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H) and
3-(N,N-dimethylamino)aniline (5c). Yield: 0.231 g (72%). 1H
NMR (CDCl3): d 1.49 (t, 3H, J ¼ 7.1 Hz), 3.13 (s, 6H), 4.56
(q, 2H, J ¼ 7.2 Hz), 7.21 (dd, 1H, J ¼ 9.5 Hz, 2.7 Hz), 7.45
(br. s, 1H), 7.5–7.6 (m, 5H), 7.80 (d, 1H, J ¼ 9.4 Hz), 7.85 (s,
1H). 13C NMR (CDCl3): d 14.4, 40.4, 62.1, 107.8, 117.6,
118.3, 120.3, 126.3, 128.46, 128.53, 129.5, 138.1, 147.6,
149.3, 149.9, 151.3, 165.7. HRMS: Found 321.1612, Calcd.
for C20H21N2O2: 321.1603.
Ethyl 4-(4-bromophenyl)-7H-pyrrolo[2,3-h]quinoline-2-car-
boxylate (9f). The title compound was obtained by the reaction
of lithio 2,4-dioxo-4-(40-bromophenyl)butyric acid ethyl ester
(Li-1, R ¼ H) and 4-amino-1H-indole (5d). Yield (based on
prep. HPLC of 0.080 g crude material): 0.020 g (25%). 1H
NMR (DMSO-d6): d 1.21 (t, 3H, J ¼ 7.1 Hz), 4.11 (q, 2H, J
¼ 7.2 Hz), 5.08 (s, 1H), 6.2–6.3 (m, 1H), 6.3–6.4 (m, 1H),
6.6–6.7 (m, 2H), 7.3–7.5 (m, 2H), 7.5–7.7 (m, 2H) 8.31 (s,
1H), 11.02 (d, 1H, J ¼ 2.0 Hz). 13C NMR (DMSO-d6): d 14.0,
61.3, 79.2, 113.2, 116.7, 117.5, 120.6, 122.1, 124.0, 125.2,
127.3, 128.1, 131.8, 138.5, 139.6, 144.5, 152.0, 156.0, 165.3.
HRMS: Found 395.0403, Calcd. for C20H16BrN2O2: 395.0395.
REFERENCES AND NOTES
[1] Astles, P. C.; Harper, M. F.; Harris, N. V.; McLay, I.M.;
Walsh, R. J. A.; Lewis, R. A.; Smith, C.; Porter, B.; McCarthy, C.
PCT Int Appl 1995, WO 95/13262 A1, 191.
[2] Barth, F.; Casellas, P.; Congy, C.; Martinez, S.; Rinaldi,
M.; Anne-Archard, G. U.S. Pat. Appl 1997, US 5624941 A, 45.
[3] Ghosh, S.; Elder, A. M.; Carson, K. G.; Sprott, K.; Harri-
son, S. PCT Int Appl 2004, WO 2004/032848 A2, 257.
[4] Goodfellow, V.; Rowbottom, M.; Dyck, B. P.; Tamiya, J.;
Zhang, M.; Grey, J.; Vickers, T.; Kiankarimi, M.; Wade, W.; Hudson,
S.C. PCT Int Appl 2004, WO 2004/080411 A2, 86.
[5] Makings, L.R.; Grootenhuis, P.; Hurley, D.J.; Tung, R.D.;
Termin, A.P. PCT Int Appl 2004, WO 2004/108133 A2, 79.
[6] Lee, J.; Kim, H. J.; Choi, S.; Choi, H. G.; Yoon, S.; Kim,
J.-H.; Jo, K.; Kim, S.; Koo, S.-Y.; Kim, M.-H.; Kim, J. I.; Hong, S.-
Y.; Kim, M. S.; Ahn, S.; Yoon, H.-S.; Cho, H.-S. PCT Int Appl 2004,
WO 2004/080979 A1, 106.
[7] Anderson, D. R.; Stehle, N. W.; Kolodziej, S. A.; Reinhard,
E. J. PCT Int Appl 2004, WO 2004/055015 A1, 223.
¨
[8] Nolsoe, J. M. J.; Weigelt, D. J Heterocyclic Chem 2009, 46, 1.
[9] Freri, M. Gazz Chim Ital 1938, 68, 612.
Ethyl
4-phenyl-7H-pyrrolo[2,3-h]quinoline-2-carboxylate
[10] Butenandt, A.; Hallmann, G.; Beckmann, R. Chem Ber
(9g). The title compound was obtained by the reaction of lithio
2,4-dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H) and
4-amino-1H-indole (5d). Yield (based on prep. HPLC of 0.080
1957, 90, 1120.
[11] Murray, W. V.; Wachter, M. P. J. Heterocyclic Chem 1989,
26, 1389.
1
[12] Grimmett, M. R. In Comprehensive Organic Chemistry;
Barton, D. H. R., Ollis, W. D., Sammes, P. G., Eds.; Pergamon: New
York, 1979; Vol. 4, p 357.
g crude material): 0.018 g (23%). H NMR (CDCl3): d 1.30 (t,
3H, J ¼7.2 Hz), 4.23 (q, 2H, J ¼ 7.2 Hz), 5.45 (s, 1H), 6.25
(s, 1H), 6.48 (d, 1H, J ¼ 6.8 Hz), 6.6–6.8 (m, 2H), 7.3–7.5
(m, 5H), (br. s, 1H). 13C NMR (CDCl3): d 14.0, 62.1, 112.6,
117.9, 119.9, 121.7, 121.8, 125.6, 126.1, 127.7, 128.6, 129.1,
139.3, 139.9, 144.8, 153.3, 156.3,165.9. HRMS: Found
317.1282, Calcd. for C20H17N2O2: 317.1290.
[13] Sokololova, M. M.; Ostrovskii, V. A.; Koldobskii, G. I.;
Mel’nikov, V. V.; Gidaspov, B. V. Zh Org Khim 1974, 10, 1085.
[14] Filyakova, V. I.; Kuznetsova, O. A.; Ulomskii, E. N.; Ryba-
lova, T. V.; Gatilov, Yu. V.; Kodess, M. I.; Rusinov, V. L.; Pashke-
vich, K. I. Russ Chem Bull Int Ed 2002, 51, 332.
Ethyl
5-phenyl-1H-pyrrolo[3,2-g]quinoline-7-carboxylate
[15] Wu, Y.-C.; Liu, L.; Li, H.-J.; Wang, D.; Chen, Y.-J. J Org
Chem 2006, 71, 6592 and references therein.
(9i). The title compound was obtained by the reaction of lithio
2,4-dioxo-4-phenylbutyric acid ethyl ester (Li-1, R ¼ H) and
6-amino-1H-indole (5e). Yield (based on prep. HPLC of 0.080
[16] Calculations where performed using the Jaguar program
(v7.5) from Schrodinger Inc using the B3LYP functional and the
LACVP** basis set. The relative energies includes solvent effects
using the PBF approximation.
1
g crude material): 0.022 g (28%). H NMR (CDCl3): d 1.50 (t,
3H, J ¼ 7.2 Hz), 4.57 (q, 2H, J ¼ 7.2 Hz), 6.67 (dd, 1H, J ¼
3.0 Hz, 2.3 Hz), 7.08 (t, 1H, J ¼ 2.9 Hz), 7.5–7.6 (m, 2H),
7.6–7.7 (m, 3H), 7.91 (br. s, 1H), 8.0–8.1 (m, 2H), 8.10
(s,1H). 13C NMR (CDCl3): d 14.2, 62.1, 104.0, 117.0, 121.7,
123.7, 124.3, 125.4, 126.4, 128.5, 128.6, 129.3, 129.5, 139.8,
141.4, 144.4, 144.8, 146.5, 165.7. HRMS: Found 317.1299,
Calcd. for C20H17N2O2: 317.1290.
[17] For structure 15b, the C-O bond of the carbonyl proximal
to the ester moiety is more elongated compared to the benzylic car-
˚
˚
bonyl (1.33A vs. 1.31A) and therefore more electrophilic. This is in
line with the observed regiochemical preference in this reaction.
[18] Crystallographic details are available in the supporting
information.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet