Synthesis and highly potent anti-inflammatory activity of licofelone- and ketorolac-based 1-arylpyrrolizin-3-ones

Article abstract of DOI:10.1016/j.bmc.2019.115053

Since NSAIDs are commonly used anti-inflammatory agents that produce adverse effects, there have been ongoing efforts to develop more effective and less toxic compounds. Based on the structure of the anti-inflammatory pyrrolizines licofelone and ketorolac

Full text of DOI:10.1016/j.bmc.2019.115053

Journal Pre-proofs
Synthesis and highly potent anti-inflammatory activity of licofelone- and ke-
torolac-based 1-arylpyrrolizin-3-ones
Damian A. Madrigal, Carlos H. Escalante, Gabriel A. Gutiérrez-Rebolledo,
José M. Cristobal-Luna, Omar Gómez-García, Roberto I. Hernández-Benitez,
Ana L. Esquivel-Campos, Salud Pérez-Gutiérrez, Germán A. Chamorro-
Cevallos, Francisco Delgado, Joaquín Tamariz
PII:
S0968-0896(19)30503-6
https://doi.org/10.1016/j.bmc.2019.115053
BMC 115053
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
28 March 2019
18 August 2019
19 August 2019
Please cite this article as: Madrigal, D.A., Escalante, C.H., Gutiérrez-Rebolledo, G.A., Cristobal-Luna, J.M.,
Gómez-García, O., Hernández-Benitez, R.I., Esquivel-Campos, A.L., Pérez-Gutiérrez, S., Chamorro-Cevallos,
G.A., Delgado, F., Tamariz, J., Synthesis and highly potent anti-inflammatory activity of licofelone- and ketorolac-
based 1-arylpyrrolizin-3-ones, Bioorganic & Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.bmc.
2019.115053
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Ltd.

1
Synthesis and highly potent anti-inflammatory activity of
licofelone- and ketorolac-based 1-arylpyrrolizin-3-ones
Damian A. Madrigal,a,b,† Carlos H. Escalante,a,† Gabriel A. Gutiérrez-Rebolledo,b
José M. Cristobal-Luna,b Omar Gómez-García,a Roberto I. Hernández-Benitez,a
Ana L. Esquivel-Campos,c Salud Pérez-Gutiérrez,c Germán A. Chamorro-Cevallos,b,*
Francisco Delgadoa and Joaquín Tamariza,*
a Departamento de Química Orgánica, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional. Prol.
Carpio y Plan de Ayala S/N, 11340 Mexico City, Mexico; b Departamento de Farmacia, Escuela Nacional de Ciencias
Biológicas, Instituto Politécnico Nacional. Av. Wilfrido Massieu S/N, Unidad Adolfo López Mateos, 07738 Mexico City,
Mexico; c Departamento de Sistemas Biológicos, Universidad Autónoma Metropolitana, Unidad Xochimilco. Calzada del
Hueso 1100, Col. Villa Quietud, 04960 Mexico City, Mexico
* Corresponding authors. Tel.: +52 55 57296300/52398 (G.A.C.-C.), +52 55 57296300/62411 (J.T.)
E-mail addresses: gchamcev@yahoo.com.mx (G.A.C.-C.), jtamarizm@gmail.com (J.T.)
† The authors contributed equally to this work.

2
Abstract— Since NSAIDs are commonly used anti-inflammatory agents that produce
adverse effects, there have been ongoing efforts to develop more effective and less toxic
compounds. Based on the structure of the anti-inflammatory pyrrolizines licofelone and
ketorolac, a series of 1-arylpyrrolizin-3-ones was synthesized. Also prepared was a series of
substituted pyrroles, mimicking similar known anti-inflammatory agents. The anti-
inflammatory activity of the test compounds was determined with a phorbol ester (TPA)-
induced murine ear edema protocol. For the most active derivatives, 19b-c/20b-c, the anti-
inflammatory effect was the same as that of the reference compound (indomethacin) and
was dose-dependent. These compounds have an aryl ring at the C-1 position and a
methoxycarbonyl group at the C-2 position of the pyrrolizine framework, which represent
plausible pharmacophore groups with anti-inflammatory activity. The anti-inflammatory
activity of 1-substituted analogs containing a five- or six-membered heterocycles was lower
but still good, while that of the pyrroles was only moderate. Although the docking studies
suggests that the effect of analogs 19a–c/20a–c is associated with the inhibition of
cyclooxygenase-2, experimental assays did not corroborate this idea. Indeed, a significant
inhibition of NO was found experimentally as a plausible mechanism of action.
Keywords: 1-Arylpyrrolizin-3-ones, Substituted pyrroles, Anti-inflammatory activity, Licofelone,
Ketorolac, TPA induced ear edema model.

3
1. Introduction
The inflammatory process is an immune response to a perturbation of the homeostatic
equilibrium of the organism, whether endogenous (e.g., genetic and ageing) or exogenous
(e.g., injury and nutrition).1-3 In addition to providing a defense against challenges to
homeostasis, this process is involved in many acute and chronic diseases such as arthritis,
cancer and diabetes,4,5 as well as cardiovascular6,7 and neurodegenerative disorders.8
Hence, the medications often prescribed for such diseases are steroidal anti-inflammatory
agents.9
The non-steroidal anti-inflammatory drugs (NSAIDs) with a mechanism of action
related to the non-selective inhibition of cyclooxygenase-1 (COX-1), COX-2 and 5-
lipoxygenase (5-LOX) produce serious adverse effects, including peptic ulcers,10–12 renal
failure13 and cardiovascular diseases.14 Consequently, there have been ongoing efforts
during many years to seek effective alternatives with less toxicity.
COX-1, COX-2 and 5-LOX are among the key enzymes involved in the transformation
of leukotrienes to prostaglandins, the latter of which are key inflammatory mediators.15
Indeed, most NSAIDs, including pyrrolizidines,16 were designed to target and inhibit these
enzymes. Licofelone (1)17–19 and the clinically used ketorolac (2a)20 are pyrrolizine-based
NSAIDs that act as potent selective and nonselective inhibitors of COX-1/COX-2/5-LOX21
and COX-1/COX-2 enzymes, respectively (Figure 1).
Its mechanism of action of ketorolac (2a), classified as an NSAIDs with potent
analgesic and moderate anti-inflammatory effects,22,23 is the inhibition of the prostaglandin
synthesis through unspecific inactivation of both COX isoforms (1 and 2). This drug is used
clinically to relieve acute pain associated with inflammation. In research, it is usually
employed as a reference drug for murine models of analgesia rather than acute
inflammation.24 Its pharmacological mechanism is similar to that exhibited by
indomethacin,25 which mainly serves as a reference drug in murine models of acute
inflammation.26–28

4
An abundance of pyrrolizine-based compounds has been synthesized recently for
evaluation as potential anti-inflammatory agents. Many are highly active derivatives and
selective COX-1/COX-2/5-LOX inhibitors, such as compounds 1,29 2b20 and 3-5,30–33
(Figure 1) among other analogs.16,33,34
Figure 1. Licofelone (1), ketorolac (2a), indomethacin and anti-inflammatory pyrrolizine-
based analogs 2b and 3-5.
Pyrrole-based anti-inflammatory agents have also been designed,35–38 such as the
potent derivatives 6a-b39 and 6c-d40 as well as NSAIDs tolmetin (7a) and zomepirac (7b)
(Figure 2).41

5
Figure 2. Anti-inflammatory pyrrole-based analogs 6a-d and 7a-b.
The aim of the current contribution was to synthesize a series of pyrroles 10-15 and
pyrrolizines 19a-g/20a-g in order to assess their anti-inflammatory effect on the formation
of ear edema induced by 12-O-tetradecanoylphorbol-13-acetate (TPA) in male CD1 mice.42
This anti-inflammatory protocol was chosen because it is a reliable model for testing
potential anti-inflammatory agents capable of inhibiting the arachidonic acid biosynthetic
pathway and in this way hampering the formation of prostaglandins, which in turn inhibits
the activity of COX and 5-LOX enzymes.43–45
2. Results and discussion
2.1. Chemistry
Pyrrole-containing agents were presently designed based on the potent anti-
inflammatory activity found with certain N-methyl 5-aroylpyrrole-2-acetic acids.46
Accordingly, a formyl group was introduced at the C-2 position of the pyrrole ring and the
acetic acid equivalent at the nitrogen atom. Since 2-formylpyrrole (8a) can serve as a

6
building block for the divergent synthesis of substituted pyrroles47 and pyrrolizines,48 it was
herein adopted for the preparation of these simple pyrrole derivatives. Hence, the
preparation of the series of substituted pyrroles 10, 11, and 13–15 was achieved by the
readily efficient functionalization of 8a (Scheme 1). The alkylation of 8a with methyl
bromoacetate (9) under basic conditions provided 10 in almost quantitative yield. By
thermal treatment of 10 with N,N-dimethylformamide dimethyl acetal (DMFDMA),
enaminone 11 was obtained in good yield as a single Z isomer (Scheme 1).
Scheme 1. Preparation of 2-formylpyrrole analogs 10 and 11.
The Knoevenagel reaction of 10 with methylene-activated compound 12a under acid
conditions,49 through an organocatalytic condensation carried out by acetic acid ammonium
salts,50 gave the expected condensation product 13 as a single E stereoisomer (Scheme 2).
However, under the usual bromination conditions with NBS, pyrrole 13 was converted into
an inseparable mixture E/Z of diastereoisomers 14/15 (81:19). Despite this unexpected
result, the latter brominated analogs were prepared due to the important anti-inflammatory
activity found for synthetic and natural bromopyrroles,51,52 and the increased lipid solubility
of the brominated derivatives of some NSAIDs, which reduce inflammation better than
bromine-free agents.53
Scheme 2. Preparation of the mixture of pyrrole analogs 14/15.

7
The synthetic approach for the preparation of 1-arylpyrrolizidin-3-ones 19a-g was
envisioned based on a previously reported strategy.54 The series of Knoevenagel derivatives
17a-g was prepared by the condensation of aldehydes 16a-g with dimethyl malonate (12b)
under catalysis with piperidine/AcOH.49 The conjugate addition of pyrrole (8b) to
derivatives 17a-g was carried out with iodine or aluminum chloride as the catalyst, leading
to the corresponding adducts 18a-g in moderate to good yields (Table 1).
Scheme 3. Preparation of 1-arylpyrrolizidin-3-ones 19a-g/20a-g.
The last step of the cyclization of adducts 18a-g to 1-substituted pyrrolizin-3-ones 19a-
g was achieved by employing previously described conditions, with sodium hydride as the
base.54 Most adducts 18a-g were different from those already reported but resulted in low
yields prompting the exploration of other bases. For example, after reacting the substrate
with KOH in anhydrous THF and stirring at room temperature for 12 h, the desired
pyrrolizin-3-ones were obtained in moderate to good yields as inseparable mixtures of
anti/syn diastereoisomers 19a-g/20a-g (Table 2). The anti pyrrolizin-3-ones 19a-g were
afforded as the main isomers in high diastereoisomeric ratio (dr).55 The structure of each
1
13
product was established by H and C NMR assisted by 1D and 2D experiments (NOE,
HMQC and HMBC) and HRMS.

8
Table 1. Reaction conditions, ratios and yields for the preparation of 18a-g and
the mixtures 19a-g/20a-g.a
Entry
17 (Ar)
Lewis acid
18 (%)b
19/20 (ratio)c (%)b
1
2
3
4
5
6
7
17a (C6H4-4-NO2)
17b (C6H4-4-CN)
17c (C6H4-4-Br)
17d (C6H4-3-OMe)
17e (furan-2-yl)
I2
I2
18a (78)
18b (77)
18c (29)
18d (49)
18e (24)
18f (53)
18g (64)
19a/20a (91:9)(78)
19b/20b (92:8)(66)
19c/20c (90:10)(57)
19d/20d (90:10)(57)
19e/20e (92:8)(80)
19f/20f (93:7)(66)
19g/20g (93:7)(70)
I2
I2
AlCl3
I2
17f (thiophen-2-yl)
17g (pyridin-2-yl)
AlCl3
a For the preparation of 18a-g, the reactions were carried out with 17 (1.0 mol equiv), 8b (3.0 mol equiv) and
I2 or AlCl3 (0.5 mol equiv) in anh. MeCN at rt for 12 h. For the preparation of 19a-g/20a-g, the reactions
involved 18 (1.0 mol equiv) and KOH (1.1 mol equiv) in anh. THF at rt for 12 h.
b Determined by 1H NMR.
c After purification by column chromatography.
Owing to the difficulty of separating the 19/20 mixtures by flash column
chromatography and HPLC, base-promoted C-2 epimerization was used to isomerize the
minor isomer to the major one (thermodynamic isomer). Instead of the epimerization,
however, the aromatization of the A ring took place in modest to good yields. Thus, the
19a/20a (91:9) mixture was treated with NaOMe to afford the novel pyrrolizine 21a in 42%
yield (Scheme 4). The yield was slightly improved by treating 19a/20a (91:9) with DDQ as
the oxidant. A better yield of pyrrolizine 21b was obtained by the reaction of 19c/20c
(90:10) with DDQ.

9
Scheme 4. Preparation of pyrrolizines 21a and 21b. Reaction conditions: (a)
MeONa/MeOH (1.0 M) (2.0 mol equiv.), THF, 50 °C, 2 h, 21a (42%). (b) DDQ (2.0 mol
equiv.), CH2Cl2, rt, 12 h; 21a (46%), 21b (70%).
2.2. Anti-inflammatory activity
A single dose (2 mg/ear) of each test compound or indomethacin was applied to a
TPA-induced edema on the mouse ear. Compared to the control animals (subjected to a
TPA-induced edema but not given any anti-inflammatory treatment), there was significant
inhibitory activity generated by functionalized pyrroles 10, 11 and 14/15 (81:19), with
values of 33.73%, 21.37%, and 15.63%, respectively (Table 2). Interestingly, the
structurally simplest compound (10) was the most active, although none of these
compounds were comparable to indomethacin (41.47%). Hence, an assay involving lower
doses was not contemplated. Indomethacin herein served as the positive control because of
being more effective than other drugs (e.g., nimesulide or ketorolac) for reducing TPA-
induced inflammation.56,57 Additionally, it is known that the mechanism of action of
indomethacin is the inhibition of COX-1/COX-2,25 and the interaction between
indomethacin and COX-2 has been corroborated by docking.
The anti-inflammatory effect of the series of pyrrolizines 19/20 was evaluated as well.
The decision to use the mixture of epimers was supported by the docking studies, which
revealed that both had similar interaction energy with the enzymes (vide infra). Thus, each
inseparable mixture of the pyrrolizine series was administered in a single dose of 2 mg/ear,
producing a significant decrease in the development of edema (p < 0.05; vs. the control

10
animals with an untreated TPA-induced edema). Contrary to the functionalized pyrroles, all
the measured effects of the series of pyrrolizines were equivalent to those of the reference
drug (indomethacin). In particular, two mixtures of the pyrrolizines, 19b/20b and 19c/20c,
afforded inhibition of inflammation equivalent to that elicited by the reference drug.
Table 2. Anti-inflammatory activity of pyrrole derivatives 10, 11 and
14/15 (81:19) on a TPA-induced ear edema in male CD1 mice.a
Treatment
Ear edema formation (mg)
Percentage of inhibition
(%)
TPA control
indomethacin
10
17.91±1.20
10.49±0.62b
11.87±0.40bc
14.09±0.45bc
15.11±0.87bc
---
41.47
33.73
21.37
15.63
11
14/15 (81:19)
a Data are expressed as the mean ± standard error of the mean (SEM) (n = 7). All compounds
were administered at the dose of 2 mg/ear. The percentage of inhibition of the edema was
calculated in relation to the TPA control group (considered as 100%), in which a TPA-induced
edema was not treated by any anti-inflammatory agent. Results were processed with two-way
analysis of variance (ANOVA) and the post-hoc Student-Newman-Keuls test (p < 0.05).
b vs. TPA control.
c vs. indomethacin.
As a consequence of the positive results with the pyrrolizines, each element of the
series was tested at three doses to evaluate its anti-inflammatory effect on the formation of
the ear edema induced by TPA in male CD1 mice (Table 3). Due to the dose-dependent
response found for all pyrrolizine compounds, the equation representing the inhibitory
effect vs. dose could be ascertained. The median effective dose (ED50) values were
calculated for each derivative during diverse phases of acute inflammation, thus
establishing initial doses for future experiments in murine models related to chronic
inflammation.58 Whereas the 1.38 mg/ear determined for derivatives 19c/20c and 19b/20b
is similar to the 1.29 mg/ear for indomethacin, the 1.54 mg/ear for 19a/20a was
significantly different than the reference drug. Nevertheless, the latter mixture exerted a
good anti-inflammatory effect with a single topical application at 2 mg/ear (Table 3).
Compared to 19a-c/20a-c, a lower inhibitory activity was shown by derivatives
19e/20e (2.75 mg/ear) and 19g/20g (3.63 mg/ear), which contain a heterocycle at the C-1

11
position. Although the latter two mixtures exhibited significantly lower anti-inflammatory
activity, the result of a single topically-applied dose of 2 mg/ear was noticeable compared
to the control group (Table 3). Derivatives 19d/20d and 19f/20f proved to be moderately
active anti-inflammatory agents. Since all compounds were administered topically,
lipophilicity probably played a key role in the observed effects, which may explain why the
aryl-substituted pyrrolizines produced greater inhibition than the C-1 heterocyclic-
substituted analogs.
Table 3. Percentage of inhibition and median effective doses (ED50) of derivatives
19/20 and 21 on the TPA-induced ear edema in male CD1 mice.a
Treatment
Doses
(mg/ear)
Ear edema
formation (mg)
Percentage of
inhibition (%)
ED50
(mg/ear)
TPA control
indomethacin
-
0.5
1
2
0.5
1
2
0.5
1
2
0.5
1
2
0.5
1
2
0.5
1
2
0.5
1
2
17.61 ± 1.06
11.22 ± 0.35b
9.01 ± 0.25•b
6.99 ± 0.13•*b
11.17 ± 0.40b
9.59 ± 0.21•b
7.17 ± 0.24•*b
11.12 ± 0.24b
9.43 ± 0.17•b
7.23 ± 0.18•*b
11.18 ± 0.59b
9.62 ± 0.35•b
7.77 ± 0.11•*bc
15.19 ± 0.47bc
12.77 ± 0.66•bc
9.25 ± 0.50•*bc
16.74 ± 0.27c
13.46 ± 0.45•bc
9.69 ± 0.37•*bc
16.68 ± 0.10c
14.49 ± 0.35•bc
9.79 ± 0.26•*bc
15.19 ± 0.30bc
13.80 ± 0.44•bc
10.02 ± 0.53•*bc
---
---
32.12
45.49
57.68
32.39
41.97
56.61
32.74
42.94
56.27
32.40
41.77
52.95
13.73
27.40
47.45
4.93
23.52
44.97
5.25
17.70
44.35
13.70
21.61
43.09
1.29 (R2 = 0.99)
19c/20c (90:10)
19b/20b (92:8)
19a/20a (91:9)
19e/20e (92:8)
19d/20d (90:10)
19g/20g (93:7)
19f/20f (93:7)
1.38 (R2 = 0.98)
1.38 (R2 = 0.96)
1.54 (R2 = 0.97)
2.75 (R2 = 0.95)
3.02 (R2 = 0.92)
3.63 (R2 = 0.88)
3.89 (R2 = 0.92)
0.5
1
2
a Data are expressed as the mean ± SEM for each group (n=7). The percentage of inhibition of the
edema was calculated as a percentage of the value of inflammation determined for the control
group (considered as 100%), in which a TPA-induced edema was not treated by any anti-
inflammatory agent. Results were processed with two-way analysis of variance (ANOVA) and
post-hoc Student-Newman-Keuls test (p < 0.05). • vs. 0.5 mg/ear; * vs. 1 mg/ear.
b vs. TPA control.
c vs. indomethacin.

12
Novel pyrrolizines 21a and 21b were also evaluated as potential anti-inflammatory
agents. Following the aforementioned method of TPA-induced ear edema in male CD1
mice, both compounds were tested at three doses, exerting a dose-dependent effect very
similar to that produced by indomethacin (Table 4). Although 21a was less active (ED50 =
1.86 mg/ear) than indomethacin (ED50 = 1.07 mg/ear), it exhibited a greater percentage of
inhibition than 21b (ED50 = 3.23 mg/ear). It is worth noticing that this pattern is contrary to
taht found with the same C-1 aryl substituents in the series of mixtures 19/20, where
19c/20c was more active than 19a/20a.
Table 4. Percentage of inhibition generated by derivatives 21a and 21b on the
TPA-induced ear edema in male CD1 mice, and the corresponding median
effective doses (ED50).a
Treatment
Doses
(mg/ear)
Ear edema
formation (mg)
Percentage of
inhibition (%)
ED50
(mg/ear)
TPA control
-
0.5
19.82 ± 0.69
11.51 ± 0.27b
---
41.94
---
9.44 ± 0.12•b
52.37
61.10
8.52
24.10
53.43
8.46
Indomethacin
1.07 (R2 = 0.99)
1
2
0.5
1
2
0.5
1
2
7.71 ± 0.22•*b
18.13 ± 0.26bc
15.04 ± 0.47•bc
9.23 ± 0.32•*bc
18.14 ± 0.24bc
13.93 ± 0.70•bc
11.72 ± 0.59•*bcd
21a
21b
1.86 (R2 = 0.95)
3.23 (R2 = 0.95)
29.73
40.88
a Data are expressed as the mean ± SEM for each group (n=7). The percentage of inhibition of the
edema was calculated as a percentage of the value of inflammation determined for the control
group (considered as 100%), in which a TPA-induced edema was not treated by any anti-
inflammatory agent. Results were processed with two-way analysis of variance (ANOVA) and
post-hoc Student-Newman-Keuls test (p < 0.05). • vs. 0.5 mg/ear; * vs. 1 mg/ear.
b vs. TPA control.
c vs. indomethacin.
d vs. compound 21a.
Regarding the epimeric compounds 19/20 and considering the difference of the TPA
control values, 21a showed a higher ED50 value (1.86 mg/ear) than 19a/20a (1.54 mg/ear),
but lower than 19e/20e (2.75 mg/ear). Interestingly, the ED50 value was lower for 21b (3.23
mg/ear) than for the mixtures 19f/20f and 19g/20g, indicating the greater activity of the
former.

13
Independently of the role played by the C-1 and C-2 substituents, the current results
suggest that the C-1 and C-2 stereocenters of pyrrolizin-3-ones 19a-g/20a-g do not
represent an exclusive factor for the promotion of anti-inflammatory activity, but are indeed
an essential element in improving such an effect. Consequently, both anti (19a-g) and syn
(20a-g) stereoisomers are probably responsible for inhibiting inflammation (Table 3).
Since indomethacin and the series of pyrrolizines 19/20 demonstrated a comparable
anti-inflammatory effect, they may have the same mode of action involving the inhibition
of COX-1/COX-2.29 Likewise, the epimeric pyrrolizine series shares similar structural
features with licofelone (1) and ketorolac (2a). Therefore, their mechanism of action is
likely to be analogous, meaning that the reduction found in the ear edema could be due to
the inhibition of prostaglandin synthesis by targeting either COX-1/COX-2 or COX-2/5-
LOX.21,29,32,33
2.3. Docking of the analogs 19a-g/20a-g with the COX-2 enzyme
Molecular docking of pyrrolizin-3-ones 19a-g/20a-g was carried out at the active site
59
of the COX-2 enzyme (Tables 5 and S1) on the AutoDock/AutoDockTools program,
based on the crystal structure of this enzyme retrieved from the Protein Data Bank.60 The
docking results for indomethacin are included as a reference. Derivatives 19a-c, having
electron-withdrawing groups (NO2, CN and Br) in para position of the aromatic ring,
showed a better binding energy than 19d (bearing an electron-donor group in the aromatic
ring) and 19e-g (with heteroaromatic rings). The binding energy of 19a was the closest to
that of indomethacin. A similar trend was observed in the binding energy of the syn-
epimers 20a-g, finding better values for 20a-c (-8.07 to -8.82 kcal/mol) than 20d-g (-6.07 to
-7.16 kcal/mol).
The binding mode and the protein-ligand interactions of 19a-g/20a-g are depicted in
Figures 3 and S1. When the binding mode of indomethacin was analyzed at the active site
of the COX-2 enzyme, most of the residues involved in the interaction coincided with those
previously reported from docking studies on this enzyme.61–63 The binding mode of the

14
most active compound, 19a, shares certain key hydrophilic interactions with the reference
drug indomethacin. For example, a conventional hydrogen bond and a carbon hydrogen
bond were formed between the NO2 group of 19a (equivalent to the carboxylic acid moiety
of indomethacin) and the side chain of Arg120 and Tyr355. Moreover, both the phenyl ring
of indomethacin and the pyrrole ring of 19a displayed -alkyl interactions with Val523.
The same -alkyl interaction takes place for almost the entire anti series, while in the syn
series it is seen with the C-1 aryl ring or the C-1 heterocyclic ring. The most active
compound of the syn-series, 20a, exhibits -alkyl type interactions with Val523 and
Ala527, as occurs with indomethacin.
Most interactions between the test compounds and the amino acids of the active site of
the enzyme were hydrophobic. However, hydrophilic interactions were identified between
most pyrrolizine derivatives and the amino acids Tyr385, Tyr355 and Arg120. In the series
of anti-derivatives 19a-g, there were -alkyl type interactions between the C-1 aryl ring
with Val349 and Leu531. These anti-derivatives also showed -alkyl interactions with
Leu352 and Ala527, as occurred with the syn-derivatives, involving the C-1 aryl ring in the
series 19a-d and the heterocyclic ring in 19e-g. Additionally, a -sigma interaction existed
between the Val523 amino acid and the C-1 aryl or heterocyclic ring of all syn-derivatives
20a-g.
The binding orientation of the anti-compounds (Figures 4 and S2) is illustrated by the
overlay of five of the analogs, 19a-d and 19g (Figure 4A), showing their similarity with
each other and with indomethacin (Figure 4B). Thus, a comparable binding region is
observed for the pyrrolizine framework and the C-2 methoxycarbonyl group of the
derivatives and the N-(4-chlorobenzoyl) group of indomethacin (Figure 4B). It is
noteworthy that the polar substituents of the C-1 aryl ring of 19a-d have similar protein-
ligand contacts as the C-3 carboxylic acid of indomethacin (see the interactions with amino
acids Arg120 and/or Tyr355, Figures 3 and S1), which was not the case for the heterocycles
of 19e-g (Figure 4C).
These polar interactions may play a significant role not only in creating the differences
in the interaction energies, but also in the in vivo anti-inflammatory effects. A correlation
was found between the biological activity and molecular docking results of derivatives 19a-

15
c, which have aryl polar substituents. This series of derivatives was the most active,
producing an inhibition of 52.95-56.61%, very close to the value for indomethacin
(57.68%). The compounds containing heterocyclic rings, on the other hand, elicited an
inhibition of 43.09-47.45%.
The anti-series 19a-g and the syn-series 20a-d and 20g had distinct binding
orientations. For the latter series, the C-1 aryl ring occupies the same position as the benzo
moiety of the indole frame of indomethacin (Figure 4D), and the aryl polar substituents are
oriented in the same position as the methoxy group of indomethacin (Figure S2). Hence,
these substituents probably provide an important contribution to the interaction energy of
the ligand with the protein.
Table 5. Docking results of indomethacin and pyrrolizines 19a-c, 19e and 20a at the active
site of COX-2.
Binding
energy
ΔG
Interacting
Residues
Hydrophobic
Interactions
Compound
Polar Interactions
(kcal/mol)
indomethacin
-9.76
His90, Val116, Arg120,
Val349, Leu352, Ser353,
Tyr355, Leu384, Tyr385,
Trp387, Arg513, Phe518,
Met522, Val523, Gly526,
Ala527, Ser530, Leu531.
Arg120 C-H…..O
Arg120 N-H…..O
Tyr355 O-H…..O
Ser530 C-H…..O
-alkyl (Leu352,
Trp387, Val523,
Ala527).
-sigma (Val349).
alkyl (Val349,
Leu384, Met522,
Ala 527, Leu531).
19a
19b
-9.34
-8.62
Arg120, Val344, Ty348,
Val349, Leu352, Tyr355,
Phe381, Leu384, Tyr385,
Trp387, Phe518, Met522,
Val523, Gly526, Ala527,
Ser530, Leu531.
Arg120 C-H…..O
Arg120 N-H…..O
Tyr355 O-H…..O
Tyr385 O-H…..O
-alkyl (Val349,
Met522, Val523).
-sigma (Ala527).
- T-shaped (Trp387).
Arg120, Val344, Tyr348,
Val349, Leu352, Phe381,
Tyr385, Met522, Val523,
Gly526, Ala527, Ser530,
Leu531.
Arg120 N-H…..N
Tyr385 O-H…..O
-alkyl (Leu531,
Leu352, Val523).
-sigma (Val349,
Ala527).
amide -stacked
(Gly526).
19c
-7.95
Arg120, Val344, Tyr348,
Val349, Leu352, Phe381,
Tyr385, Met522, Val523,
Gly526, Ala527, Ser530,
Leu531.
Tyr385 O-H…..O
-alkyl (Leu531,
Val523).
-sigma (Val349,
Ala527).
alkyl (Leu531).

16
19e
20a
-6.52
-8.82
Tyr348, Val349, Leu352,
Tyr355, Phe381, Leu384,
Trp387, Met522, Val523,
Gly526, Ala527, Ser530.
Tyr385 O-H…..O
Tyr385 O-H…..O
-alkyl (Val349,
Met522, Val523).
-sigma (Ala527).
 T-shaped
(Trp387).
His90, Tyr348, Val349,
Leu352, Ser353, Tyr355,
Phe381, Leu384, Tyr385,
Trp387, Arg513, Phe518,
Met522, Val523, Gly526,
Ala527, Ser530.
-alkyl (Leu352,
Ala527).
-sigma (Val523).
amide -stacked
(Gly526).
Indomethacin
19a
19b
19c
19e
20a

17
Figure 3. The predicted binding mode of indomethacin and pyrrolizines 19a-c, 19e and 20a at the active site
of the COX-2 enzyme. The 2D model shows the amino acids involved in such binding. The following
interactions are portrayed with dotted lines: conventional hydrogen bonds (green), carbon-hydrogen (yellow),
-sigma (purple), -alkyl (light pink), amide- stacking (dark pink) and  T-shaped (orange). The basic
amino acids are denoted in blue, hydrophobic in green and polar in cyan.
A
B
C
D

18
Figure 4. A. Overlay of the docking poses of 19a (orange), 19b (purple), 19c (dark blue), 19d (brown) and
19g (cyan). B. Comparison of the binding mode of each of the respective groups of compounds 19a-d and
19g to that of indomethacin (green). C. Overlay of 19e (light green) and 19f (light blue). D. Overlay of 20a
(orange), 20b (purple), 20c (dark blue), 20d (brown) and 20g (cyan), and comparison of the binding mode of
each of the corresponding groups of compounds 20a-d and 20g to that of indomethacin (green).
2.4. COX-2 enzyme and nitric oxide inhibitory activity of 19a-c/20a-c
According to the docking study, the inhibitory effect on COX-2 produced by the series
of pyrrolizin-3-ones 19a-g/20a-g represents a possible mechanism of action for their strong
anti-inflammatory activity on the TPA-induced ear edema. Hence, in vitro COX-2
inhibition assays were explored in relation to these in silico findings. In the cell viability
tests,64 the J774A.1 macrophages (8 x 104 cells/well) were treated with the most effective
mixtures of pyrrolizin-3-ones, 19a-c/20a-c, at concentrations of 6.25 to 200 µg/mL
dissolved in DMSO. Only the mixture 19b/20b showed no toxicity at the concentrations
evaluated (Figure S3). However, 19a/20a and 19c/20c displayed toxicity at concentration
of 50 µg/mL. Therefore, the epimers were used at concentration of 25 µg/mL in subsequent
experiments, resulting in an IC50 of 80.82, ˂200 μg/mL and 74.94 for 19a/20a, 19b/20b
and 19c/20c, respectively. The lack of effect of the three mixtures on the mRNA expression
of COX-2 could be due to their inability to inhibit its production at this concentration.
Moreover, the apparent contradiction between the in vivo inhibition of COX-2 and in vitro
cytotoxicity has been previously observed for an anti-inflammatory drug and potential
anticancer agent (celecoxib), which is explained by the fact that the effect caused in each
model occurs is by independent mechanism of action.65
Lipolisacharide (LPS), a component of the membrane of gram-negative bacteria,
promotes the generation of pro-inflammatory cytokines and nitric oxide (NO) in
macrophages.66 NO is a signaling molecule with several effects, such as vasodilation, host
defense, inflammation and blood clotting.67 In inflammatory processes the overproduction
of NO can activate nuclear factor kappa B (NF-휅B) and trigger the expression of
proinflammatory mediators. Arise in the levels of NO elicited by inducible nitric oxide
synthase (iNOS) can lead to tissue damage.68 Hence, the inhibition of NO production in
macrophages is a potential therapeutic strategy for inflammation.

19
In the present study, the three mixtures 19a/20a–19c/20c significantly lowered the
level of NO in LPS-stimulated J774A.1 macrophages (1 x 106 cells/well), which was
assessed by measuring the presence of nitrites through the Griess reaction.69 Firstly, it was
herein confirmed that the level of nitrites was markedly increased when the macrophages
were exposed to LPS (Figure 5). The value of the nitrite concentration after LPS exposure
was set at 100%, and the basal value (prior to LPS exposure) was found to be 61.19%. At
25 μg/mL, 19a/20a, 19b/20b and 19c/20c generated a very significant inhibition of nitrites
production, with values of 63.63%, 64.75% and 63.52%, respectively. These percentage
nitrite levels were similar to the basal value and to the inhibition obtained with
indomethacin (64.63%) at the same concentration (25 μg/mL).
Figure 5. Effect of 19a/20a, 19b/20b, 19c/20c and indomethacin (Indo) (at 25 μg/mL) on the level of nitrites
in LPS-stimulated macrophages. The basal value was 61.19%. Data represent the mean ± SEM of three
independent experiments. * p <0.05 vs. LPS.
3. Conclusion
A series of pyrroles 10, 11 and 14/15, pyrrolizin-3-ones 19a-g/20a-g and pyrrolizines
21a-b were herein synthesized and tested, finding a moderate to high anti-inflammatory
effect. The median effective doses of some 19a-g/20a-g and 21a derivatives caused over

20
50% inhibition of the ear edema formation. Compounds 19b-c/20b-c proved to be the most
active licofelone- and ketorolac-based analogs, showing an anti-inflammatory effect
comparable to indomethacin. Docking studies performed on the COX-2 enzyme established
multiple molecular modeling interactions between the active site and the different polar and
non-polar functional groups of the derivatives. Plausible pharmacophore groups responsible
for the anti-inflammatory effect of 19b-c/20b-c include the C-1 functionalized aryl ring and
the C-2 methoxycarbonyl group. These pyrrolizine compounds provide good structural
requirements for the design of more active analogs in the future. The C-1 substituted
benzene ring apparently serves a key function as the hydrophobic (benzene ring) and
hydrophilic (polar substituent) moiety, possibly representing a good and versatile scaffold
to promote the formation of valuable active substituents. Due in part to the activity shown
by pyrrolizines 21a-b, it appears that the C-1 and C-2 stereogenic centers are not an
exclusive condition for the anti-inflammatory effect of the derivatives 19a-g/20a-g, but do
indeed improve such an effect.
According to the docking results, the mechanism of action for the reduction of the
TPA-induced ear edema is likely to be the inhibition of COX-2. However, no experimental
evidence corroborated the viability of this mechanistic pathway. On the other hand, the
most active mixtures 19a/20a–19c/20c exhibited a significant inhibition of NO production
(i.e., a decrease in the level of nitrites) in LPS-stimulated macrophages. Whereas iNOS
inhibition can be a plausible mechanism of action, the COX-2 inhibition cannot be
completely ruled out.
The next step by our group in the investigation of the anti-inflammatory effect of these isomers
will be their evaluation in a model of sub-plantar carrageenan-induced mouse paw edema. This of
course implies having previously carried out studies of acute toxicity in mice with the two isomers
orally or intraperitoneally administered.
4. Experimental section
4.1. Chemistry
Melting points were determined on an Electrothermal apparatus and are uncorrected.
1
Infrared spectra (IR) were recorded on a FT-IR 2000 Perkin-Elmer spectrometer. H (300
13
or 500 or 600 MHz) and C (75.4 or 125 or 150 MHz) NMR spectra were obtained on

21
Varian Mercury-300 or Varian VNMR System or Bruker Avance III instruments, with
TMS as internal standard and chemical shifts () are reported in ppm. Mass Spectra were
achieved on a Polaris Q-Trace GC Ultra (Finnigan Co.). High-resolution mass spectra
(HRMS), in electron impact mode, were recorded on Jeol JSM-GCMateII. A Multi-Therm
Benchmark, Model H5000-HC served as a heating and cooling shaker for enzymatic
stability assays. Commercial reagents were used as received from Aldrich and anhydrous
solvents were obtained by a distillation process. Thin layer chromatography was performed
on precoated silica gel plates (Merck 60F254) and column chromatography with silica gel
(230-400 mesh).
4.2.1. Methyl 2-(2-formyl-1H-pyrrol-1-yl)acetate (10)70
After a mixture of 8a (0.300 g, 3.16 mmol) and NaH (0.148 g, 3.7 mmol) in dry DMF
was stirred at 0 °C for 30 min, 9 (0.530 g, 3.4 mmol) was added and stirred at 0 °C for 12 h.
Water (20 mL) was added and the mixture was extracted with hexane/EtOAc (1:1) (2 x 50
mL), the organic layer dried (Na2SO4) and the solvent removed under vacuum. The residue
was purified by column chromatography over silica gel (20 g/g crude, hexane/EtOAc, 8:2)
to afford 10 (0.516 g, 98%) as a yellow oil. Rf 0.53 (hexane/EtOAc, 8:2). IR (film): ῡ 3114,
2955, 2811, 1757, 1659, 1532, 1483, 1410, 1369, 1322, 1217, 1081, 1032, 1001, 766, 698
cm-1. 1H NMR (300 MHz, CDCl3): δ 3.77 (s 3H, CO2CH3), 5.70 (s, 2H, CH2), 6.31 (dd, J =
4.2, 2.7 Hz, 1H, H-4’), 6.92-6.94 (m, 1H, H-5’), 7.00 (dd, J = 4.2, 1.5 Hz, 1H, H-3’), 9.53
13
(d, J = 0.9 Hz, 1H, CHO). C NMR (75.4 MHz, CDCl3): δ 50.1 (CH2), 52.5 (CO2CH3),
110.2 (C-4’), 124.6 (C-3’), 131.6 (C-2’), 132.0 (C-5’), 168.8 (CO2CH3), 179.8 (CHO).
HRMS (EI): m/z [M+] calcd for C8H9NO3: 167.0582; found: 167.0584.
4.2.2. Methyl (E)-3-(Dimethylamino)-2-(2-formyl-1H-pyrrol-1-yl)acrylate (11)
In a threaded ACE glass pressure tube with a sealed Teflon screw cap and magnetic
stirring bar, a mixture of 10 (0.300 g, 1.8 mmol) and DMFDMA (0.66 g, 5.5 mmol) was
stirred at 80 °C for 48 h. The mixture was diluted with CH2Cl2 (5 mL) and the solvent was
removed under vacuum. The residue was purified by column chromatography over silica
gel (20 g/g crude, hexane/EtOAc, 1:1) to furnish 11 (0.287 g, 72%) as a yellow oil. Rf 0.30

22
(hexane/EtOAc, 7:3). IR (film): ῡ 2949, 1694, 1665, 1620, 1528, 1469, 1433, 1390, 1367,
1308, 1215, 1107, 1079 cm-1. 1H NMR (500 MHz, CDCl3): δ 2.67 (br s, 6H, N(CH3)2), 3.64
(s, 3H, CO2CH3), 6.31 (dd, J = 4.3, 2.5 Hz, 1H, H-4’), 6.82-6.84 (m, 1H, H-5’), 7.03 (dd, J
= 4.3, 2.0 Hz, 1H, H-3’), 7.53 (s, 1H, H-3), 9.58 (s, 1H, CHO). 13C NMR (125 MHz,
CDCl3): δ 51.4 (CO2CH3), 97.5 (C-2), 110.3 (C-4’), 121.0 (C-3’), 134.1 (C-5’), 134.8 (C-
2’), 146.2 (C-3), 167.3 (CO2CH3), 179.7 (CHO). HRMS (EI): m/z [M+] calcd for:
C11H14N2O3: 222.1005; found: 222.1005.
4.2.3. Methyl (E)-2-Cyano-3-(1-(2-methoxy-2-oxoethyl)-1H-pyrrol-2-yl)acrylate (13)
In a threaded ACE glass pressure tube with a sealed Teflon screw cap and magnetic
stirring bar, a mixture of 10 (0.10 g, 0.6 mmol), 12a (0.064 g, 0.65 mmol), piperidine
(0.025 g, 0.29 mmol) and glacial AcOH (0.028 g, 0.47 mmol) in CH2Cl2 (5 mL) was stirred
at 70 °C for 24 h. The mixture was diluted CH2Cl2 (50 mL) and washed with water (25 mL)
and a saturated aqueous solution of NaHCO3 until neutral. The organic layer was dried
(Na2SO4) and the solvent removed under vacuum. The residue was purified by column
chromatography over silica gel (20 g/g crude, hexane/EtOAc, 8:2), to give 13 (0.135 g,
91%) as a pale yellow solid. Rf 0.4 (hexane/EtOAc, 7:3); mp 125-126 °C. IR (KBr): ῡ 3113,
2963, 2213, 1741, 1711, 1592, 1477, 1450, 1435, 1413, 1355. 1269, 1201, 1090, 1009, 756,
1
675 cm-1. H NMR (300 MHz, CDCl3): δ 3.80 (s, 3H, CO2CH3), 3.89 (s, 3H, CO2CH3),
4.84 (s, 2H, CH2CO2Me), 6.47 (ddd, J = 4.2, 2.6, 0.6 Hz, 1H, H-4’), 7.05 (dd, J = 2.6, 1.5
13
Hz, 1H, H-5’), 7.80 (dm, J = 4.2 Hz, 1H, H-3’), 7.92 (s, 1H, H-3). C NMR (75.4 MHz,
CDCl3): δ 48.1 (CH2CO2Me), 53.0 (CO2CH3), 53.1 (CO2CH3), 94.0 (C-2), 112.8 (C-4’),
116.7 (CN), 120.3 (C-3’), 127.2 (C-2’), 131.3 (C-5’), 138.9 (C-3), 164.3 (CO2CH3), 167.7
(CO2CH3). HRMS (EI): m/z [M+] calcd for: C12H12N2O4: 248.0797; found: 248.0797.
4.2.4.
Methyl
(E)-2-Cyano-3-(4,5-dibromo-1-(2-methoxy-2-oxoethyl)-1H-pyrrol-2-
yl)acrylate (14). Methyl (Z)-2-Cyano-3-(4,5-dibromo-1-(2-methoxy-2-oxoethyl)-1H-pyrrol-
2-yl)acrylate (15)
A solution of NBS (0.14 g, 0.8 mmol) in dry DMF (3 mL) was added dropwise to a
stirred solution of 13 (0.100 g, 0.40 mmol) in dry DMF (3 mL) at 0 °C. The mixture was
stirred at room temperature for 16 h, then a mixture of hexane/EtOAc/H2O (1:1:0.5) (30

23
mL) was added. The organic layer was dried (Na2SO4) and the solvent removed under
vacuum. The residue was purified by column chromatography over silica gel (20 g/g crude,
hexane/EtOAc, 7:3) to afford a mixture of 14/15 (81:19) (0.137 g, 84%) as a yellow solid.
Rf 0.32 (hexane/EtOAc, 8:3); mp 279-281 °C. IR (KBr): ῡ 2956, 2217, 1749, 1721, 1586,
1438, 1407, 1384, 1259, 1219, 1095, 1043, 770 cm-1. 1H NMR (500 MHz, CDCl3): δ 3.82
(s, 3H, CO2CH3), 3.90 (s, 3H, CO2CH3), 4.93 (s, 2H, CH2CO2CH3), 7.82 (d, J = 0.5 Hz,
1H, H-3’), 7.83 (d, J = 0.5 Hz, 1H, H-3). Signals attributed to the minor isomer 15: δ 3.80
13
(s, CO2CH3), 3.93 (s, CO2CH3), 4.90 (s, CH2CO2CH3), 7.99 (s, H-3 or H-3’). C NMR
(125 MHz, CDCl3): δ 47.6 (CH2CO2CH3), 53.2 (CO2CH3), 53.3 (CO2CH3), 96.2 (C-2),
104.0 (C-4’), 116.0 (C-5’), 116.4 (CN), 121.0 (C-3’), 128.5 (C-2’), 137.5 (C-3), 163.5
(CO2CH3), 166.5 (CO2CH3). Signals attributed to the minor isomer 15: δ 50.4, 53.6, 103.5,
107.0, 114.9, 125.6, 127.4, 138.3, 162.4, 166.4. HRMS (EI): m/z [M+] calcd for:
C12H10Br2N2O4: 403.9008; found: 403.9009.
4.2.5. Dimethyl 2-(4-Nitrobenzylidene)malonate (17a)71
In a threaded ACE glass pressure tube with a sealed Teflon screw cap and magnetic
stirring bar, a mixture of 16a (0.300 g, 1.99 mmol), 12b (0.289 g, 2.19 mmol), piperidine
(0.085 g, 1.00 mmol) and glacial AcOH (0.060 g, 1.00 mmol) in CH2Cl2 (1.0 mL) was
heated to 120 °C for 4 h. The mixture was diluted with CH2Cl2 (40 mL) and washed with
H2O (20 mL) and a saturated aqueous solution of NaHCO3 until neutral. The organic layer
was dried (Na2SO4) and the solvent removed under vacuum. The residue was purified by
column chromatography over silica gel (10 g/g crude, hexane/EtOAc, 95:5) to deliver 17a
(0.475 g, 90%) as a colorless solid. Rf 0.60 (hexane/EtOAc, 7:3); mp 135-137 °C [Lit.71
135-136 °C]. IR (KBr): ῡ 3112, 2959, 1727, 1631, 1601, 1517, 1435, 1346, 1258, 1225,
1199, 1062, 979, 940, 852, 765, 720, 687 cm-1. 1H NMR (500 MHz, CDCl3): δ 3.84 (s, 3H,
CO2CH3), 3.89 (s, 3H, CO2CH3), 7.56-7.60 (m, 2H, H-2”), 7.80 (s, 1H, H-1’), 8.22-8.26
(m, 2H, H-3”). 13C NMR (125 MHz, CDCl3): δ 52.9 (CO2CH3), 53.0 (CO2CH3), 124.0 (C-
3”), 129.3 (C-2), 129.9 (C-2”), 139.1 (C-1”), 139.9 (C-1’), 148.5 (C-4”), 163.7 (CO2CH3),
166.0 (CO2CH3). HRMS (EI): m/z [M+] calcd for C12H11NO6: 265.0586; found: 265.0580.
4.2.6. Dimethyl 2-(4-Cyanobenzylidene)malonate (17b)71

24
Following the method of preparation for 17a, the reaction of 16b (0.300 g, 2.29 mmol),
12a (0.333 g, 2.52 mmol), piperidine (0.098 g, 1.15 mmol) and glacial AcOH (0.069 g,
1.15 mmol) provided 17b (0.453 g, 81%) as a colorless solid. Rf 0.48 (hexane/EtOAc, 7:3);
mp 97-98 °C [Lit.71 97-98 °C]. IR (KBr): ῡ 3038, 2959, 2223, 1739, 1719, 1628, 1440,
1373, 1265, 1223, 1202, 1064, 974, 934, 826 cm-1. 1H NMR (500 MHz, CDCl3): δ 3.84 (s,
3H, CO2CH3), 3.87 (s, 3H, CO2CH3), 7.52 (dm, J = 8.5 Hz, 2H, H-2”), 7.68 (dm, J = 8.5
13
Hz, 2H, H-3”), 7.75 (s, 1H, H-1’). C NMR (125 MHz, CDCl3): δ 52.9 (CO2CH3), 53.0
(CO2CH3), 113.8 (C-4”), 118.1 (CN), 128.7 (C-2), 129.6 (C-2”), 132.5 (C-3”), 137.2 (C-
1”), 140.4 (C-1’), 163.8 (CO2CH3), 166.1 (CO2CH3). HRMS (EI): m/z [M+] calcd for
C13H11NO4: 245.0688; found: 245.0688.
4.2.7. Dimethyl 2-(4-Bromobenzylidene)malonate (17c)71
Following the method of preparation for 17a, the reaction of 16c (0.500 g, 2.70 mmol),
12b (0.392 g, 2.97 mmol), piperidine (0.115 g, 1.35 mmol) and glacial AcOH (0.081 g,
1.35 mmol) produced 17c (0.580 g, 72%) as a colorless oil. Rf 0.55 (hexane/EtOAc, 8:2).
IR (film): ῡ 2953, 1731, 1629, 1588, 1489, 1437, 1374, 1264, 1223, 1072, 1001, 838, 818
1
cm-1. H NMR (500 MHz, CDCl3): δ 3.83 (s, 3H, CO2CH3), 3.84 (s, 3H, CO2CH3), 7.27
(dm, J = 8.5 Hz, 2H, H-2”), 7.50 (dm, J = 8.5 Hz, 2H, H-3”), 7.68 (s, 1H, H-1’). 13C NMR
(125 MHz, CDCl3): δ 52.7 (2CO2CH3), 125.1 (C-4”), 126.1 (C-2), 130.6 (C-2”), 131.6 (C-
1”), 132.1 (C-3”), 141.4 (C-1’), 164.1 (CO2CH3), 166.7 (CO2CH3). HRMS (EI): m/z [M+]
calcd for C12H11BrO4: 297.9841; found: 297.9848.
4.2.8. Dimethyl 2-(3-Methoxybenzylidene)malonate (17d)
Following the method of preparation for 17a, the reaction of 16d (0.500 g, 3.67 mmol),
12b (0.534 g, 4.40 mmol), piperidine (0.157 g, 1.84 mmol) and glacial AcOH (0.110 g,
1.84 mmol) formed 17d (0.579 g, 63%) as a reddish solid. Rf 0.31 (hexane/EtOAc, 8:2); mp
82-83 °C [Lit.71 81-82 °C]. IR (film): ῡ 2954, 1732, 1629, 1579, 1436, 1240, 1069, 785, 690
cm-1. 1H NMR (500 MHz, CDCl3): δ 3.80 (s, 3H, OCH3), 3.84 (s, 6H, 2CO2CH3), 6.95 (dd,
J = 7.5, 2.5 Hz, 1H, H-4”), 6.96 (br s, 1H, H-2”), 7.01 (br d, J = 7.5 Hz, 1H, H-6”), 7.29 (t,
J = 7.5 Hz, 1H, H-5”), 7.74 (s, 1H, H-1’). 13C NMR (125 MHz, CDCl3): δ 52.66 (CO2CH3),
52.67 (CO2CH3), 55.2 (OCH3), 114.2 (C-2”), 116.7 (C-4”), 121.9 (C-6”), 125.7 (C-2),

25
129.9 (C-5”), 134.0 (C-1”), 142.8 (C-1’), 159.8 (C-3”), 164.4 (CO2CH3), 167.1 (CO2CH3).
HRMS (EI): m/z [M+] calcd for C13H14O5: 250.0841; found: 250.0848.
4.2.9. Dimethyl 2-(Furan-2-yl-methylene)malonate (17e)
Following the method of preparation for 17a, the reaction of 16e (0.300 g, 3.13 mmol),
12b (0.454 g, 3.44 mmol), piperidine (0.133 g, 1.57 mmol) and glacial AcOH (0.094 g,
1.57 mmol) resulted in 17e (0.565 g, 86%) as an amber oil. Rf 0.50 (hexane/EtOAc, 7:3). IR
(film): ῡ 3130, 2954, 1730, 1633, 1437, 1257, 1224, 1207, 1083, 1064, 1021, 756 cm-1. 1H
NMR (500 MHz, CDCl3): δ 3.83 (s, 3H, CO2CH3), 3.92 (s, 3H, CO2CH3), 6.50 (dd, J = 3.5,
2.0 Hz, 1H, H-4”), 6.77 (d, J = 3.5 Hz, 1H, H-3”), 7.48 (s, 1H, H-1’), 7.53 (brd, J = 2.0 Hz,
13
1H, H-5”). C NMR (125 MHz, CDCl3): δ 52.4 (CO2CH3), 52.5 (CO2CH3), 112.6 (C-4”),
118.2 (C-3”), 121.2 (C-2), 128.1 (C-1’), 146.3 (C-5”), 148.8 (C-2”), 164.5 (CO2CH3),
166.7 (CO2CH3). HRMS (EI): m/z [M+] calcd for C10H10O5: 210.0528; found: 210.0518.
4.2.10. Dimethyl 2-(Thiophen-2-ylmethylene)malonate (17f)
Following the method of preparation for 17a, the reaction of 16f (0.500 g, 4.46 mmol),
12b (0.648 g, 4.91 mmol), piperidine (0.190 g, 2.23 mmol) and glacial AcOH (0.134 g,
2.23 mmol) yielded 17f (0.728 g, 72%) as a yellow solid. Rf 0.59 (hexane/EtOAc, 7:3) mp
43-44 °C [Lit.71 43-44 °C]. IR (KBr): ῡ 3069, 1721, 1624, 1435, 1259, 1203, 1070, 935,
1
736 cm-1. H NMR (500 MHz, CDCl3): δ 3.84 (s, 3H, CO2CH3), 3.94 (s, 3H, CO2CH3),
7.09 (dd, J = 5.0, 4.0 Hz, 1H, H-4”), 7.38 (dm, J = 4.0 Hz, 1H, H-3”), 7.55 (dt, 5.0, 1.0 Hz,
1H, H 5”), 7.90 (s, 1H, H-1’). 13C NMR (125 MHz, CDCl3): δ 52.6 (CO2CH3), 52.8
(CO2CH3), 121.4 (C-2), 127.8 (C-4”), 131.9 (C-5”), 134.8 (C-3”), 135.5 (C-1’), 135.9 (C-
2”), 164.7 (CO2CH3), 166.6 (CO2CH3). HRMS (EI): m/z [M+] calcd for C10H10O4S:
226.0300; found: 226.0305.
4.2.11. Dimethyl 2-(Pyridin-2-ylmethylene)malonate (17g)
Following the method of preparation for 17a, the reaction of 16g (0.300 g, 2.80 mmol),
12b (0.407 g, 3.08 mmol), piperidine (0.119 g, 1.40 mmol) and glacial AcOH (0.084 g,
1.40 mmol) afforded 17g (0.470 g, 76%) as a reddish solid. Rf 0.38 (hexane/EtOAc, 9:1),
mp 81-83 °C. IR (KBr): ῡ 2952, 1739, 1718, 1631, 1439, 1373, 1278, 1251, 1222, 1070,
1
946, 786 cm-1. H NMR (500 MHz, CDCl3): δ 3.87 (s, 3H, CO2CH3), 3.91 (s, 3H,

26
CO2CH3), 7.26 (ddd, J = 8.0, 5.0, 1.5 Hz, 1H, H-5”), 7.39 (br d, J = 8.0 Hz, 1H, H-3”), 7.67
(s, 1H, H-1’), 7.72 (td, J = 8.0, 2.0 Hz, 1H, H-4”), 8.61 (br dd, J = 5.0, 1.5 Hz, 1H, H-6”).
13C NMR (125 MHz, CDCl3): δ 52.4 (CO2CH3), 52.8 (CO2CH3), 124.5 (C-5”), 126.3 (C-
3”), 128.3 (C-2), 136.7 (C-4”), 139.9 (C-1’), 150.0 (C-6”), 151.0 (C-2”), 164.3 (CO2CH3),
167.0 (CO2CH3). MS (70 eV): m/z 221 (M+, 4), 206 (M+-15, 100), 190 (98), 160 (16), 155
(52), 149 (14), 122 (64), 91 (53). HRMS (EI): m/z [M+] calcd for C11H11NO4: 221.0688;
found: 221.0687.
4.2.12. Dimethyl 2-((4-Nitrophenyl)(1H-pyrrol-2-yl)methyl)malonate (18a)
After stirring a mixture of 17a (0.300 g, 1.13 mmol) and I2 (0.144 g, 0.57 mmol) in dry
MeCN (30 mL) at room temperature for 5 min, 8b (0.227 g, 3.39 mmol) was added, and
stirred at room temperature for 12 h. Water (30 mL) was added, and the mixture extracted
with EtOAc (3 x 40 mL). The organic layer was dried (Na2SO4) and the solvent removed
under vacuum. The residue was purified by column chromatography over silica gel (20 g/g
crude, hexane/EtOAc, 8:2) to obtain 18a (0.295 g, 78%) as a pale yellow solid. Rf 0.38
(hexane/EtOAc, 7:3); mp 174-175 °C [Lit.54 170-171 °C]. IR (KBr): ῡ 3379, 3111, 2959,
1746, 1601, 1515, 1436, 1356, 1307, 1263, 1213, 1151, 861, 732, 703 cm-1. 1H NMR (500
MHz, CDCl3): δ 3.53 (s, 3H, CO2CH3), 3.71 (s, 3H, CO2CH3), 4.21 (d, J = 10.0 Hz, 1H, H-
2), 4.92 (d, J = 10.0 Hz, 1H, H-1’), 5.93-5.96 (m, 1H, H-3”), 6.10 (q, J = 2.5 Hz, 1H, H-
4”), 6.69 (dt, J = 2.5, 1.5 Hz, 1H, H-5”), 7.41-7.46 (m, 2H, H-2’”), 8.13-8.17 (m, 2H, H-
3’”), 8.57 (br s, 1H, NH). 13C NMR (125 MHz, CDCl3): δ 44.0 (C-1’), 52.8 (CO2CH3), 53.1
(CO2CH3), 57.2 (C-2), 107.2 (C-3”), 108.5 (C-4”), 118.3 (C-5”), 123.8 (C-3’”), 129.0 (C-
2”), 129.1 (C-2’”), 147.0 (C-4’”), 147.3 (C-1’”), 167.5 (CO2CH3), 168.6 (CO2CH3). HRMS
(EI): m/z [M+] calcd for C16H16N2O6: 332.1008; found: 332.1010.
4.2.13. Dimethyl 2-((4-Cyanophenyl)(1H-pyrrol-2-yl)methyl)malonate (18b)
Following the method of preparation for 18a, the reaction of 17b (0.300 g, 1.22 mmol),
I2 (0.154 g, 0.61 mmol) and 8b (0.246 g, 3.67 mmol) in MeCN (30 mL) gave 18b (0.295 g,
77%) as a brown solid. Rf 0.31 (hexane/EtOAc, 7:3); mp 166-167 °C. IR (KBr): ῡ 3345,
2959, 2232, 1751, 1606, 1433, 1356, 1301, 1262, 1216, 1177, 1148, 917, 858, 784, 726 cm-
1. 1H NMR (300 MHz, CDCl3): δ 3.52 (s, 3H, CO2CH3), 3.70 (s, 3H, CO2CH3), 4.19 (d, J =
10.4 Hz, 1H, H-2), 4.87 (d, J = 10.4 Hz, 1H, H-1’), 5.92-5.96 (m, 1H, H-3”), 6.09 (q, J =

27
2.7 Hz, 1H, H-4”), 6.68 (dt, J = 2.7, 1.5 Hz, 1H, H-5”), 7.36-7.41 (m, 2H, H-2’”), 7.55-7.60
13
(m, 2H, H-3’”), 8.60 (br s, 1H, NH). C NMR (75.4 MHz, CDCl3): δ 44.2 (C-1’), 52.8
(CO2CH3), 53.1 (CO2CH3), 57.0 (C-2), 106.9 (C-3”), 108.4 (C-4”), 111.1 (C-4’”), 118.2
(C-5”), 118.6 (CN), 128.9 (C-2’”), 129.1 (C-2”), 132.3 (C-3’”), 145.2 (C-1’”), 167.5
(CO2CH3), 168.6 (CO2CH3). HRMS (EI): m/z [M+] calcd for C17H16N2O4: 312.1110;
found: 312.1108.
4.2.14. Dimethyl 2-((4-Bromophenyl)(1H-pyrrol-2-yl)methyl)malonate (18c)
Following the method of preparation for 18a, the reaction of 17c (0.300 g, 1.00 mmol),
I2 (0.127 g, 0.50 mmol) and 8b (0.201 g, 3.00 mmol) in MeCN (30 mL) furnished 18c
(0.108 g, 29%) as a pale yellow solid. Rf 0.30 (hexane/EtOAc, 8:2); mp 155-157 °C [Lit.54
158-160 °C]. IR (film): ῡ 3392, 2953, 1736, 1488, 1434, 1303, 1259, 1149, 1011, 725 cm-1.
1H NMR (500 MHz, CDCl3): δ 3.51 (s, 3H, CO2CH3), 3.69 (s, 3H, CO2CH3), 4.14 (d, J =
10.5 Hz, 1H, H-2), 4.77 (d, J = 10.5 Hz, 1H, H-1’), 5.93 (br s, 1H, H-3”), 6.08 (q, J = 2.8
Hz, 1H, H-4”), 6.65-6.67 (m, 1H, H-5”), 7.13 (br d, J = 8.0, Hz, 2H, H-2’”), 7.41 (br d, J =
8.0 Hz, 2H, H-3’”), 8.46 (br s, 1H, NH). 13C NMR (125 MHz, CDCl3): δ 43.7 (C-1’), 52.6
(CO2CH3), 52.9 (CO2CH3), 57.4 (C-2), 106.6 (C-3”), 108.2 (C-4”), 117.9 (C-5”), 121.2 (C-
4’”), 129.8 (C-2’”), 130.1 (C-2”), 131.7 (C-3’”), 138.8 (C-1’”), 167.7 (CO2CH3), 168.8
(CO2CH3). HRMS (EI): m/z [M+] calcd for C16H16BrNO4: 365.0263; found: 365.0258.
4.2.15. Dimethyl 2-((3-Methoxyphenyl)(1H-pyrrol-2-yl)methyl)malonate (18d)
Following the method of preparation for 18a, the reaction of 17d (0.300 g, 1.20 mmol),
I2 (0.091 g, 0.36 mmol) and 8b (0.241 g, 3.60 mmol) in MeCN (30 mL) delivered 18d
(0.185 g, 49%) as a pale yellow solid. Rf 0.28 (hexane/EtOAc, 8:2); mp 103-105 °C. IR
(film): ῡ 3397, 2954, 1733, 1600, 1586, 1490, 1435, 1259, 1154, 1034, 724 cm-1. 1H NMR
(500 MHz, CDCl3): δ 3.51 (s, 3H, CO2CH3), 3.69 (s, 3H, CO2CH3), 3.76 (s, 3H, OCH3),
4.17 (d, J = 10.5 Hz, 1H, H-2), 4.78 (d, J = 10.5 Hz, 1H, H-1’), 5.95-5.97 (m, 1H, H-3”),
6.07 (q, J = 2.5 Hz, 1H, H-4”), 6.64 (dt, J = 2.5, 1.5 Hz, 1H, H-5”), 6.76 (br dd, J = 8.0, 0.5
Hz, 1H, H-4’”), 6.80 (t, J = 2.5 Hz, 1H, H-2’”), 6.85 (br d, J = 8.0 Hz, 1H, H-6’”), 7.21 (t, J
13
= 8.0 Hz, 1H, H-5’”), 8.38 (br s, 1H, NH). C NMR (125 MHz, CDCl3): δ 44.3 (C-1’),
52.5 (CO2CH3), 52.8 (CO2CH3), 55.1 (OCH3), 57.6 (C-2), 106.4 (C-3”), 108.1 (C-4”),
112.6 (C-4’”), 114.0 (C-2’”), 117.6 (H-5”), 120.3 (C-6’”), 129.6 (C-5’”), 130.6 (C-2”),

28
141.2 (C-1’”), 159.7 (C-3’”), 167.8 (CO2CH3), 168.9 (CO2CH3). HRMS (EI): m/z [M+]
calcd for C17H19NO5: 317.1263; found: 317.1264.
4.2.16. Dimethyl 2-(Furan-2-yl(1H-pyrrol-2-yl)methyl)malonate (18e).
Following the method of preparation for 18a, the reaction of 17e (0.300 g, 1.43 mmol),
AlCl3 (0.096 g, 0.72 mmol) and 8b (0.288 g, 4.30 mmol) in MeCN (30 mL) provided 18e
(0.093 g, 24%) as a brown solid. Rf 0.33 (hexane/EtOAc, 7:3); mp 107-108 °C. IR (KBr): ῡ
3379, 2955, 1754, 1431, 1326, 1285, 1235, 1150, 1012, 913, 731 cm-1. 1H NMR (500 MHz,
CDCl3): δ 3.62 (s, 3H, CO2CH3), 3.65 (s, 3H, CO2CH3), 4.15 (d, J = 10.0 Hz, 1H, H-2),
4.90 (d, J = 10.0 Hz, 1H, H-1’), 6.01-6.02 (m, 1H, H-3’”), 6.08 (q, J = 3.0 Hz, 1H, H-4’”),
6.12 (dm, J = 3.0 Hz, 1H, H-3”), 6.28 (dd, J = 3.0, 1.5 Hz, 1H, H-4”), 6.70 (dt, J = 3.0, 1.5
13
Hz, 1H, H-5’”), 7.34 (dd, J = 1.5, 1.0 Hz, 1H, H-5”), 8.72 (br s, 1H, NH). C NMR (125
MHz, CDCl3): δ 38.2 (C-1’), 52.7 (CO2CH3), 52.8 (CO2CH3), 56.4 (C-2), 107.1 (C-3”),
107.4 (C-3’”), 108.1 (C-4’”), 110.4 (C-4”), 117.9 (C-5’”), 127.5 (C-2’”), 142.0 (C-5”),
152.7 (C-2”), 167.8 (CO2CH3), 168.4 (CO2CH3). HRMS (EI): m/z [M+] calcd for
C14H15NO5: 277.0950; found: 277.0943.
4.2.17. Dimethyl 2-((1H-Pyrrol-2-yl)(thiophen-2-yl)methyl)malonate (18f)
Following the method of preparation for 18a, the reaction of 17f (0.300 g, 1.33 mmol),
I2 (0.170 g, 0.67 mmol) and 8b (0.267 g, 3.99 mmol) in MeCN (30 mL) produced 18f
(0.205 g, 53%) as a brown solid. Rf 0.50 (hexane/EtOAc, 7:3), mp 121-123 °C. IR (KBr): ῡ
3385, 2951, 1752, 1434, 1342, 1277, 1234, 1142, 716 cm-1. 1H NMR (500 MHz, CDCl3): δ
3.59 (s, 3H, CO2CH3), 3.67 (s, 3H, CO2CH3), 4.16 (d, J = 10.0Hz, 1H, H-2), 5.09 (d, J =
10.0 Hz, 1H, H-1’), 6.02-6.04 (m, 1H, H-3’”), 6.09 (q, J = 3.0 Hz, 1H, H-4’”), 6.68 (dt, J =
2.5, 1.5 Hz, 1H, H-5’”), 6.89 (ddd, J = 3.5, 1.5, 1.0 Hz, 1H, H-3”), 6.90 (dd, J = 5.0, 3.5
13
Hz, 1H, H-4’’), 7.16 (dd, J = 5.0, 1.5 Hz, 1H, H-5’’), 8.62 (br s, 1H, NH). C NMR (125
MHz, CDCl3): δ 39.6 (C-1’), 52.7 (CO2CH3), 52.8 (CO2CH3), 58.5 (C-2), 106.8 (C-3’”),
108.1 (C-4’”), 117.8 (C-5’”), 124.7 (C-5”), 125.3 (C-3”), 126.7 (C-4”), 129.9 (C-2’”),
143.3 (C-2”), 167.7 (CO2CH3), 168.5 (CO2CH3). HRMS (EI): m/z [M+] calcd for
C14H15NO4S: 293.0722; found: 293.0730.
4.2.18. Dimethyl 2-(Pyridin-2-yl(1H-pyrrol-2-yl)methyl)malonate (18g)

29
Following the method of preparation for 18a, the reaction of 17g (0.300 g, 1.36 mmol),
AlCl3 (0.091 g, 0.68 mmol) and pyrrole (0.274 g, 4.08 mmol) in MeCN (30 mL) formed
18g (0.248 g, 64%) as a pale yellow solid. Rf 0.36 (hexane/EtOAc, 9:1); mp 112-113 °C. IR
(KBr): ῡ 3377, 2955, 1746, 1709, 1592, 1568, 1473, 1436, 1364, 1308, 1272, 1229, 1178,
1
1146, 923, 801, 723 cm-1. H NMR (500 MHz, CDCl3): δ 3.56 (s, 3H, CO2CH3), 3.60 (s,
3H, CO2CH3), 4.52 (d, J = 10.0 Hz, 1H, H-2), 4.87 (d, J = 10.0 Hz, 1H, H-1’), 6.00-6.06
(m, 2H, H-3’”, H-4’”), 6.69 (br, 1H, H-5’”), 7.12 (td, J = 5.0, 1.5 Hz, 1H, H-5”), 7.25 (br d,
J = 7.5 Hz, 1H, H-3”), 7.58 (t, J = 7.5 Hz, 1H, H-4”), 8.53 (br d, J = 5.0 Hz, 1H, H-6”),
9.19 (br s, 1H, NH). 13C NMR (125 MHz, CDCl3): δ 45.5 (C-1’), 52.4 (CO2CH3), 53.6
(CO2CH3), 57.4 (C-2), 107.6 (C-3’”), 107.8 (C-4’”), 118.3 (C-5’”), 121.9 (C-5”), 124.0 (C-
3”), 128.2 (C-2’”), 136.8 (C-4”), 148.9 (C-6”), 159.6 (C-2”), 168.3 (CO2CH3), 168.9
(CO2CH3). HRMS (EI): m/z [M+] calcd for C15H16N2O4: 288.1110; found: 288.1108.
4.2.19.
Methyl
(1R*,2S*)-1-(4-Nitrophenyl)-3-oxo-2,3-dihydro-1H-pyrrolizine-2-
carboxylate (19a). Methyl (1R*,2R*)-1-(4-Nitrophenyl)-3-oxo-2,3-dihydro-1H-pyrrolizine-
2-carboxylate (20a)54
Compound 18a (0.100 g, 0.30 mmol) was added to a solution of KOH (0.019 g, 0.33
mmol) in anhydrous THF (5 mL) under N2 at 0 ºC, and the mixture was stirred at room
temperature for 12 h. After adding water (30 mL), the mixture was extracted with EtOAc (3
x 20 mL). The organic layer was dried (Na2SO4) and the solvent removed under vacuum.
The residue was purified by column chromatography over silica gel (10 g/g crude,
hexane/EtOAc, 9:1) to generate a mixture of 19a/20a (91:9) (0.070 g, 78%) as a red viscous
oil. Rf 0.64 (hexane/EtOAc, 7:3). IR (film): ῡ 2919, 1760, 1735, 1519, 1402, 1348, 1292,
1243, 1164, 859, 722 cm-1. 1H NMR (500 MHz, CDCl3): δ 3.89 (s, 3H, CO2CH3), 3.94 (d, J
= 5.0 Hz, 1H, H-2), 5.06 (br d, J = 5.0 Hz, 1H, H-1), 6.03-6.05 (m, 1H, H-7), 6.57 (t, J =
3.0 Hz, 1H, H-6), 7.15 (br d, J = 3.0 Hz, 1H, H-5), 7.44-7.48 (m, 2H, H-2’), 8.20-8.24 (m,
2H, H-3’). Signals attributed to the minor isomer 20a: δ 3.26 (s, CO2CH3), 4.51 (d, J = 9.0
Hz, H-1), 5.03 (br d, J = 9.0 Hz, H-1), 6.05-6.06 (m, H-7), 6.60 (t, J = 3.0 Hz, H-6), 7.21
13
(br d, J = 3.0 Hz, H-5), 7.38-7.42 (m, H-2’), 8.15-8.18 (m, H-3’). C NMR (125 MHz,
CDCl3): δ 42.4 (C-1), 53.5 (CO2CH3), 61.6 (C-2), 106.9 (C-7), 112.5 (C-5), 120.2 (C-6),
124.4 (C-3’), 128.4 (C-2’), 138.3 (C-7a), 147.1 (C-1’), 147.6 (C-4’), 164.3 (C-3), 167.2