Identification of a novel orally bioavailable ERK5 inhibitor with selectivity over p38α and BRD4

Article abstract of DOI:10.1016/j.ejmech.2019.05.057

Extracellular regulated kinase 5 (ERK5) signalling has been implicated in driving a number of cellular phenotypes including endothelial cell angiogenesis and tumour cell motility. Novel ERK5 inhibitors were identified using high throughput screening, with

Full text of DOI:10.1016/j.ejmech.2019.05.057

Accepted Manuscript
Identification of a novel orally bioavailable ERK5 inhibitor with selectivity over p38α
and BRD4
Stephanie Myers, Duncan C. Miller, Lauren Molyneux, Mercedes Arasta, Ruth H.
Bawn, Timothy Blackburn, Simon J. Cook, Noel Edwards, Jane A. Endicott, Bernard
T. Golding, Roger J. Griffin, Tim Hammonds, Ian R. Hardcastle, Suzannah J. Harnor,
Amy Heptinstall, Pamela Lochhead, Mathew P. Martin, Nick C. Martin, David R.
Newell, Paul J. Owen, Leon C. Pang, Tristan Reuillon, Laurent J.M. Rigoreau, Huw
Thomas, Julie A. Tucker, Lan-Zhen Wang, Ai-Ching Wong, Martin E.M. Noble,
Stephen R. Wedge, Celine Cano
PII:
S0223-5234(19)30467-2
DOI:
https://doi.org/10.1016/j.ejmech.2019.05.057
EJMECH 11367
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 25 February 2019
Revised Date: 30 April 2019
Accepted Date: 20 May 2019
Please cite this article as: S. Myers, D.C. Miller, L. Molyneux, M. Arasta, R.H. Bawn, T. Blackburn, S.J.
Cook, N. Edwards, J.A. Endicott, B.T. Golding, R.J. Griffin, T. Hammonds, I.R. Hardcastle, S.J. Harnor,
A. Heptinstall, P. Lochhead, M.P. Martin, N.C. Martin, D.R. Newell, P.J. Owen, L.C. Pang, T. Reuillon,
L.J.M. Rigoreau, H. Thomas, J.A. Tucker, L.-Z. Wang, A.-C. Wong, M.E.M. Noble, S.R. Wedge, C.
Cano, Identification of a novel orally bioavailable ERK5 inhibitor with selectivity over p38α and BRD4,
European Journal of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.05.057.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

ACCEPTED MANUSCRIPT
Graphical Abstract
Identification of a Novel Orally Bioavailable ERK5 Inhibitor with Selectivity over p38α
and BRD4.
An HTS hit was optimized to give a sub-micromolar ERK5 inhibitor, selective over p38α and
BRD4, suitable for use as an in vivo tool compound.
1

ACCEPTED MANUSCRIPT
Identification of a Novel Orally Bioavailable ERK5
Inhibitor with Selectivity over p38α and BRD4.
Stephanie Myers, ◊ Duncan C. Miller, Lauren Molyneux, Mercedes Arasta, Ruth H. Bawn,
a
a
◊
a
◊
b
a
a
c
b
b
a
Timothy Blackburn, Simon J. Cook, Noel Edwards, Jane A. Endicott, Bernard T. Golding,
a
Z
d
a
a
a
Roger J. Griffin, Tim Hammonds, Ian R. Hardcastle, Suzannah J. Harnor, Amy Heptinstall,
c
b
a
b
d
Pamela Lochhead, Mathew P. Martin, Nick C. Martin, David R. Newell, Paul J. Owen, Leon
d
a
e
b
b
C. Pang, Tristan Reuillon, Laurent J. M. Rigoreau, Huw Thomas, Julie A. Tucker, Lan-Zhen
b
d
b†
b†
a†
Wang, Ai-Ching Wong, Martin E. M. Noble, Stephen R. Wedge and Celine Cano *
a.
Newcastle Drug Discovery, Northern Institute for Cancer Research, School of Chemistry,
Bedson Building, Newcastle University, Newcastle upon Tyne, NE1 7RU, UK.
b.
Newcastle Drug Discovery, Northern Institute for Cancer Research, Paul O’Gorman
Building, Medical School, Framlington Place, Newcastle upon Tyne, NE2 4HH, UK.
c.
Signalling Laboratory, The Babraham Institute, Babraham Research Campus, Cambridge,
CB22 3AT, UK
d.
Cancer Research UK Therapeutic Discovery Laboratories, London Bioscience Innovation
Centre, 2 Royal College Street, London NW1 0NH, UK.
e.
Cancer Research UK Therapeutic Discovery Laboratories, Jonas Webb Building,
Babraham Campus, Babraham, Cambridgeshire CB22 3AT, UK
1

ACCEPTED MANUSCRIPT
◊
These authors contributed equally to the work.
†
To whom enquiries should be addressed. For enquiries concerning chemistry, contact C.C. Tel
(44) 191 208 7060. E-mail: celine.cano@ncl.ac.uk; for enquiries concerning structural biology,
contact M.E.M.N. Tel. (44) 191 208 4422. E-mail: martin.noble@ncl.ac.uk; for enquiries
concerning bioscience, contact S.R.W. Tel. (44) 191 208 4418. E-mail: steve.wedge@ncl.ac.uk.
Z
Deceased 24 September 2014
ABSTRACT
Extracellular regulated kinase 5 (ERK5) signalling has been implicated in driving a number of
cellular phenotypes including endothelial cell angiogenesis and tumour cell motility. Novel
ERK5 inhibitors were identified using high throughput screening, with a series of pyrrole-2-
carboxamides substituted at the 4-position with an aroyl group being found to exhibit IC values
5
0
in the micromolar range, but having no selectivity against p38α MAP kinase. Truncation of the
N-substituent marginally enhanced potency (~ 3-fold) against ERK5, but importantly attenuated
inhibition of p38α. Systematic variation of the substituents on the aroyl group led to the selective
inhibitor 4-(2-bromo-6-fluorobenzoyl)-N-(pyridin-3-yl)-1H-pyrrole-2-carboxamide (IC50 0.82
µM for ERK5; IC >120 µM for p38α). The crystal structure (PDB 5O7I) of this compound in
5
0
complex with ERK5 has been solved. This compound was orally bioavailable and inhibited
bFGF-driven Matrigel plug angiogenesis and tumour xenograft growth. The selective ERK5
inhibitor described herein provides a lead for further development into a tool compound for more
extensive studies seeking to examine the role of ERK5 signalling in cancer and other diseases.
2

ACCEPTED MANUSCRIPT
KEYWORDS
ERK5; Extracellular regulated kinase 5; BMK1; kinase; bioavailable; pyrrole carboxamide
INTRODUCTION
Extracellular regulated kinase 5 (ERK5, BMK1, MAPK7) is the terminal kinase in one of the
four major mitogen activated protein (MAP) kinase pathways [1]. MAP kinase pathways are
central to cancer, as well as numerous other pathologies, and recent clinical success with B-RAF
and MEK1/2 inhibitors, in melanoma in particular, has demonstrated that these kinases are
amenable to therapeutic exploitation [2]. Correlative clinical data have implicated ERK5 as a
potential target in a number of tumour types, including prostate and breast cancer, and
mechanistic studies have demonstrated potential roles for ERK5 in cell migration and
angiogenesis [3, 4]. Targeting the non-canonical MAPK pathway through inhibition of ERK5
has generated a lot of interest in the last decade. The oxindole-based BIX02189 (Figure 1; 1) was
one of the first potent dual ERK5-MEK5 inhibitors reported [5]. Subsequently, pre-clinical data
generated with a pyrimidodiazepinone small molecule inhibitor of ERK5 kinase activity (2;
XMD8-92) suggested that perturbation of this signalling pathway can impede the growth of
human tumours in vivo [6]. A more potent and selective ERK5 inhibitor from the
pyrimidodiazepinone series, (3; AX15836) was subsequently disclosed as a useful tool for the
study of ERK5 inhibition [7]. Recently, BAY-885 (4) was reported as a potent ERK5 chemical
probe [8].
3

ACCEPTED MANUSCRIPT
Figure 1. Structures and pharmacological profiles of reported ERK5 inhibitors.
A role for ERK5 in cell proliferation was first indicated by studies using dominant-negative
ERK5 constructs that inhibited epidermal growth factor-stimulated proliferation of HeLa cells
[9]. However, more extensive recent studies [10], using both MEK5/ERK5 inhibition and ERK5
knockdown, do not support a role for ERK5 in cellular proliferation. In contrast, there are
compelling data from a number of studies [11-16] demonstrating that ERK5 signalling promotes
tumour cell migration and invasion, a property that is consistent with the clinical relationship
between tumour ERK5 over-expression and metastatic disease. For example, in an orthotopic
model of prostate cancer, Leung and colleagues [17] demonstrated that ERK5 promotes the
4

ACCEPTED MANUSCRIPT
formation of metastases. Similarly, in an in vivo orthotopic breast cancer model, ERK5
knockdown resulted in a marked reduction in lymph node metastasis [18].
A third aspect of tumour biology that may be exploited by ERK5 inhibitors is inhibition of
tumour angiogenesis. A substantial body of data based initially on constitutive and conditional
knockout mouse models shows that ERK5 is required for endothelial cell survival, in particular
maintenance of vascular integrity and endothelial cell migration. MEK5, the obligate kinase
required for ERK5 activation, and ERK5 knockout mouse embryos die at approximately day
1
0.5 from tissue failure that includes impaired cardiac development, which may result from an
inability to permit maturation of the supporting vasculature [19, 20]. Importantly, in a
conditional ERK5 knockout adult mouse model, ERK5 depletion following Cre recombinase
induction disrupts vascular integrity and results in death from multiple organ haemorrhage within
two to four weeks [21], consistent with a key role for ERK5 in endothelial cell biology [22].
Further studies by Hayashi and colleagues [23] unequivocally link ERK5 signalling to tumour
angiogenesis in two in vivo mouse tumour models, and in a growth-factor induced angiogenesis
model. Consequently, treatment with an ERK5 inhibitor may be anticipated to have a therapeutic
effect on tumour vasculature in solid cancers.
Whilst the prototype ERK5 kinase inhibitor, XMD8-92 (2) [6], has been used to explore
inhibition of ERK5 signalling in vivo, a recent study [24] has demonstrated that XMD8-92 (2), in
common with some other kinase inhibitors [25], can also inhibit bromodomain (BRD) proteins,
and that this property may underlie some of the cellular effects of the compound.
Following a high-throughput screening campaign a pyrrole series of ERK5 inhibitors was
identified, as exemplified by 5 [26]. Encouragingly, compound 5 passed pan-assay interference
5

ACCEPTED MANUSCRIPT
(PAINS) and aggregator potential assessment [27]. The current paper describes the hit-to-lead
development of potent ERK5 inhibitors in this pyrrole series.
CHEMISTRY
Structure-activity relationships (SARs) were developed for the ketone and amide substituents
of lead compound 5. Regioselective Friedel-Crafts acylation of commercially available methyl
pyrrole 2-carboxylate (6) using the appropriate acid chloride and aluminium trichloride gave
methyl esters 7-18 (Scheme 1).
Hydrolysis of esters 7-18 provided carboxylic acids 19-30, and subsequent amide coupling
mediated by either CDI or PCl furnished inhibitors 31-46. The PCl -mediated coupling
3
3
conditions proved beneficial with electron-poor amines, where CDI coupling was generally
unsuccessful.
Scheme 1. Reagents and conditions; (i) AlCl , appropriate acid chloride, DCM, 0 °C-RT, 16 h,
3
2
2
2-82%; (ii) LiOH, THF/H O, 60 °C, 16 h, 40-99%; (iii) CDI, THF, 70 °C, 3 h; appropriate
amine, 50 °C, 3 h then RT, 16 h, 35-95%, or PCl , appropriate amine, MeCN, 150 °C, MW, 5-30
3
min, 48-76%; R groups are defined in Tables 1 and 2.
RESULTS AND DISCUSSION
6

ACCEPTED MANUSCRIPT
Following successful hit validation [26], attention initially focussed on developing SARs for
substitution of the aroyl ring of 5 in an attempt to improve potency against ERK5, while
retaining the most potent 4-picolylamide substituent identified in the initial hit validation
campaign [26]. Fluoro substitution at the 2-position (32) was beneficial, as was chloro
substitution at the 2- or 3-positions (35, 36). Substitution at the 4-position (34, 37) led to loss of
ERK5 inhibition. 2,3- or 2,6-disubstitution (38, 40) conferred a modest further improvement in
potency.
Due to the similarity between published p38α MAP kinase inhibitors [28] and our hit series, a
counter-screen against p38α was employed to determine the selectivity of the series. Compounds
were generally inversely selective for p38α, with the exception of 40 which showed modest 4-
fold selectivity for ERK5.
Table 1. Biological evaluation of compounds 31-40.
1
a
b
ID
R
ERK5; IC50 (µM)
40 ± 2.6
p38α; IC50 (µM)
31
32
33
Ph
-
2-F-phenyl
3-F-phenyl
5.9 ± 5.9
0.36 ± 0.14
-
>120
7

ACCEPTED MANUSCRIPT
34
35
36
37
38
39
40
4-F-phenyl
>120
-
2-Cl-phenyl
3-Cl-phenyl
15 ± 2.8
6.3 ± 0.1
>120
-
6.5 ± 1.9
-
4-Cl-phenyl
2,3-diF-phenyl
2,6-diF-phenyl
2,6-diCl-phenyl
7.0 ± 2.5
30 ± 32
4.4 ± 2.1
2.4 ± 0.75
0.58 ± 0.27
12.3 ± 4.7
a
ERK5 IC50s determined using an IMAP FP Progressive Binding System kit (Molecular Devices
b
#
R8127). p38α IC50s determined against full length p38 protein (Millipore #14-251) in a
LANCE assay format using a Ulight-MBP peptide substrate. Determinations ± standard
deviation (mean of n ≥ 4).
Attention turned to the amide substituent. Truncated amide groups, lacking the methylene
spacer, were synthesised with the aim of improving potency against ERK5 (Table 2). 4-
Pyridylamide (41) was around 30-fold more potent than its methylene spaced analogue (39),
exhibiting sub-micromolar ERK5 inhibition. Gratifyingly, truncation of the amide substituent led
to a considerable improvement in selectivity over p38α. The 3-pyridylamine regioisomer (42)
maintained similar ERK5 potency and selectivity. Replacing one of the fluoro substituents with a
larger halogen atom, while retaining the second fluoro substituent proved beneficial in terms of
both potency and selectivity (43-46). Five compounds from this series were tested in a cell-based
ERK5-dependent MEF2D reporter gene assay. Cellular potency was within 5- to 10-fold of the
8

ACCEPTED MANUSCRIPT
cell-free IC50 values (Table 2), with compounds 43, 45 and 46 having similar profiles. Due to its
sub-micromolar potency, high selectivity for ERK5 over p38α in cell-free assays, and less than
4
-fold difference beween isolated enzyme and cell-free assays, compound 46 was selected as a
representative of the pyrrole carboxmide series for further profiling. Growth inhibition in PC3
cells was assessed for compound 46, which exhibited a GI of 44 ± 2.8 µM.
5
0
Table 2. Biological Evaluation of compounds 41-46.
ERK5
a
b
ERK5; IC50
µM)
p38α; IC50
1
2
c
ID
R
R
MEF2D; IC50
(
(µM)
c
(µM)
4
4
4
4
4
4
1
2
3
4
5
6
2,6-diF
2,6-diF
0.90 ± 0.15
1.1 ± 0.27
100 ± 35
35 ± 0.90
>120
8.0 ± 10
n.d.
2-Cl, 6-F
2-Cl, 6-F
2-Br, 6-F
2-Br, 6-F
0.60 ± 0.30
0.60 ± 0.20
0.70 ± 0.20
0.82 ± 0.10
2.3 ± 1.5
4.0 ± 1.4
4.0 ± 2.8
3.0 ± 1.2
93.2 ± 12.5
>120
>120
9

ACCEPTED MANUSCRIPT
a
ERK5 IC50s determined using an IMAP FP Progressive Binding System kit (Molecular Devices
b
#
R8127). p38α IC s determined against full length p38 protein (Millipore #14-251) in a
5
0
LANCE assay format using a Ulight-MBP peptide substrate. Determinations ± standard
c
deviation (mean of n ≥ 4). Cell-based activity against ERK5 was determined using a
transcriptional reporter assay in which mutationally activated MEK5 (MEK5D) and ERK5 were
co-expressed in HEK293 cells and the transactivation domain of the transcription factor MEF2D
(an ERK5 substrate) fused to the DNA binding domain of the yeast GAL4 protein. ERK5-
dependent activity of GAL4-MEF2D was assayed using a co-expressed luciferase reporter.
Determinations ± standard deviation (mean of n = 2).
A crystal structure of the kinase domain of ERK5 in complex with 46 was determined at 2.4 Å
resolution, revealing the binding mode for the pyrrole series and rationalising the SAR results
(
Figure 2, PDB ID 5O7I). As expected, 46 binds in the ERK5 ATP-binding site. The pyrrole NH
and amide carbonyl form hydrogen bonds with the backbone carbonyl of Asp138 and amide of
Met140 from the ERK5 hinge-region, respectively. The 3-pyridinyl substituent lies toward the
mouth of the ATP-binding pocket, and is twisted 35° out-of-plane with respect to the amide.
Electron density for the pyridyl ring is less well defined than that for the 4-(2-bromo-6-
fluorobenzoyl)-1H-pyrrole-2-carboxamide moiety, suggesting a degree of rotational freedom
about the N-C bond. The ketone carbonyl lies in plane with the pyrrole ring, and is oriented
toward the back of the ATP-binding pocket where it makes a water-mediated hydrogen bond to
the backbone amide of Phe201. The 2-bromo-6-fluorophenyl moiety lies in an orthogonal plane
to the pyrrole ring, and occupies a hydrophobic pocket formed by the aliphatic portions of the
side-chains of Tyr66, Val69, Lys84, Ile115, Leu137, Leu189 and Asp200. This pocket is highly
10

ACCEPTED MANUSCRIPT
complementary in shape to the 2-bromo-6-fluorophenyl moiety, whilst the orthogonal bioactive
conformation is favoured by ortho-substitution, accounting for the observed SAR around the
aroyl ring. The side-chain of Tyr66 from the tip of the glycine-rich P-loop that covers the ATP-
binding site re-orients to stack edge-to-face against the phenyl ring, explaining the loss of
potency for 4-substituted compounds, 34 and 37.
Figure 2. Compound 46 bound at the ERK5 ATP-binding site (PDB ID 5O7I). (a) Crystal
structure of the ERK5-46 complex determined at 2.4 Å resolution, shown in coral ribbon with
key active-site residues labelled and shown as sticks, and 46 with carbon atoms in green. The
2
Fo-Fc omit electron density, contoured at 1.5 σ around 46, is displayed as a blue mesh. (b) As
a) with a transparent electrostatic surface drawn over the protein atoms. (c) Top down view of
6 bound within the ERK5 ATP-binding site highlighting the potential hydrogen bonding
(
4
network formed with the hinge region (D138-M140), and the potential water mediated hydrogen
bond formed with D200 of the DFG.
Compopund 46 progressed into further pharmacokinetic and pharmacological profiling. In
both human and mouse microsomal stability assays 46 exhibited low intrinsic clearance (HLM
Cl <5 µL/min/mg protein; MLM Cl 26 µL/min/mg protein) and had high flux and a low
int
int
-6
efflux ratio (ER) in a caco-2 cell permeability assay (BA >10 × 10 cm/s, ER 0.82). Solubility,
as assessed in a turbidometric solubility assay (Cyprotex), estimated the precipitation
u
concentration of 46 to be >100 µM, (Table 3). 46 had moderate plasma protein binding (F 0.06).
11

ACCEPTED MANUSCRIPT
Compound 46 had an IC50 of >5 µM against a panel comprising four cytochrome P450 isoforms
(2C19, 2C9, 2D6, 3A4).
When screened at a concentration of 10 µM against a panel of 456 kinases (DiscoverX), in
addition to ERK5, 46 inhibited only seven other kinases by ≥ 90% (DCAMKL3, JAK1, SLK,
MAP3K15, TYK2, JAK2, MST2 and DCAMKL1; full kinome panel inhibition data can be
found in the supplementary information). Overall, 46 showed good drug-like properties and was
progressed into in vivo drug metabolism pharmacokinetic (DMPK) studies in mice.
A 10 mg/kg dose of 46 was administered through both oral and intravenous routes to female
CD1 mice at 8-10 weeks of age (3 mice per time point). The plasma of these mice was collected
at eight time points following administration (Table 3). The terminal plasma half-life was 38
min, with a plasma clearance of 27 mL/min/kg, and oral bioavailability of 68%.
Table 3. In vivo pharmacokinetic parameters for 46 dosed in female CD1 mice (Charles River,
Kent UK) 10 mg/kg either i.v. or p.o.
Administration
Dose (mg/kg)
AUC (µg/mL.min)
Cmax (µM)
PO
10
254
6
IV
10
372
25
1.2
-
V (L/kg)
d
-
T
max (min)
15
12

ACCEPTED MANUSCRIPT
Cl
p
(mL/min/kg)
-
27
38
-
t (min)
1
/2
F (%)
68
In vivo efficacy was assessed using a Matrigel plug assay in female CD1 mice (8-10 weeks
old) and an A2780 human ovarian carcinoma xenograft model in female CD1 nude (nu/nu) mice
(8-10 weeks old). In both studies, the efficacy of 46 was compared to that of the most advanced
ERK5 literature ERK5 inhibitor at the time, XMD8-92 (2). The Matrigel plug assay is a measure
of angiogenesis and was performed by subcutaneously inoculating mice with 500 µL of Matrigel
containing basic fibroblast growth factor (bFGF; 500 ng/mL). This induces formation of a
Matrigel plug, which can be permeated by host cells and leads to formation of new blood vessels.
In the presence of an anti-angiogenic agent the extent of new blood vessels formed would be
reduced. Mice were randomised 24 h following Matrigel inoculation, to receive either; vehicle,
XMD8-92 (2) (50 mg/kg, i.p.) or 46 (100 mg/kg, p.o.) twice-daily for 7 days (5 mice per group).
On day 7 the mice were humanely sacrificed and the Matrigel plugs removed and snap frozen in
liquid nitrogen. The concentration of haemoglobin present could then be determined as a
surrogate measure for the extent of blood vessel formation. An anti-angiogenic effect was
apparent with Matrigel plugs from animals treated with both, XMD8-92 (2) (two tailed unpaired
t-test, n = 5, p = 0.042) and compound 46 (two tailed unpaired t test n = 5 p = 0.0009) having
lower concentrations of haemoglobin than those from control animals receiving vehicle (Figure
3
). There was no significant difference between the two treatments (two tailed unpaired t-test, n
5, p = 0.367).
=
13

ACCEPTED MANUSCRIPT
p= 0.0009
p= 0.042
2000
1500
p= 0.367
1000
500
0
Figure 3. Haemoglobin levels determined in a bFGF Matrigel plug assay. Female CD1 mice, 5
per group, were treated with XMD8-92 (2) (50 mg/kg, i.p.) or 46 (100 mg/kg, p.o.) twice-daily
for 7 days and p values determined using an unpaired two-tailed t-test.
In a human tumour xenograft model, mice were inoculated subcutaneously on the right flank
7
with 1 × 10 A2780 human ovarian carcinoma cells, and tumours left to grow for 7 days. The
mice (9 per group) were then randomised to receive either vehicle, XMD8-92 (2) (50 mg/kg, i.p.)
or 46 (100 mg/kg, p.o.) twice-daily for 10 days. Tumour volumes were significantly reduced (as
assessed by an unpaired two-tailed t-test) in mice treated with either XMD8-92 (2) (62%
inhibition; p = 0.001) or with 46 (57% inhibition; p = 0.002) when compared to the control group
(
Figure 4a). Nadir weight was used as an indicator of drug toxicity. None of the animals lost
more than 4% of their starting body weight with the nadir body weights being 96% for control,
8% for XMD8-92-treated and 99% for animals in the 46 treatment group (9 mice per group)
9
14

ACCEPTED MANUSCRIPT
(
Figure 4b). These initial in vivo studies suggest that an ERK5 inhibitor could have potential as
an anti-angiogenic agent and thereby constrain solid tumour growth.
Vehicle Control
XMD8-92
Control
XMD8-92
46
6
4
2
00
00
00
0
120
46
1
1
10
00
9
0
80
0
5
10
0
5
10
Day
15
20
Days of treatment
Figure 4. a). Treatment of female CD1 nude mice bearing A2780 human ovarian carcinoma
xenografts with vehicle control, XMD8-92 (2) (50 mg/kg, ip) or 46 (100 mg/kg, po) (9 per
group). Mice were treated twice-daily for 10 days at the end of treatment mean tumour volumes
were significantly smaller in mice treated with either XMD8-92 (2) (p = 0.001) or with 46 (p =
0
.002) when compared to the control group (unpaired two-tailed t-test) b). Percent body weight
change over the course of the study.
During the course of the study, we became aware of the recently reported activity of the
selective ERK5 inhibitor XMD8-92 (2) towards BRD4.[24] Thus, the potential of 46 to interact
with the first bromodomain of BRD4 was assessed using Surface Plasmon Resonance (SPR). In
contrast to the positive control XMD8-92 (2), no binding was observed from the sensorgram for
compound 46 (Figure 5), confirming that effects seen in the in vivo efficacy models were not due
to inhibition of BRD4.
15

ACCEPTED MANUSCRIPT
Figure 5. Characterisation of the binding of 46 and XMD8-92 (2) to BRD4 by SPR. 1:1 steady-
state binding model of 2 and 46 against BRD4, resulting in a K of 1.64 µM and no binding,
d
respectively.
CONCLUSIONS
An ERK5 inhibitor with sub-micromolar potency in a cell-free assay has been developed based
on a pyrrole carboxamide scaffold through structure-based optimisation of the aryl ketone and
amide substituents. Greater than one-hundred fold selectivity over p38α MAP kinase was
attained through truncation of the amide substituent. Compound 46 does not interact with the
BRD4 bromodomain. Compound 46 exhibited suitable pharmacokinetics in mouse to permit in
vivo studies following oral dosing and does not interact with the BRD4 bromodomain.
Compound 46 suppressed both tumour xenograft growth and bFGF-driven Matrigel plug
angiogenesis. Further development of the pyrrole carboxamide series of ERK5 inhibitors to
provide compounds with single digit nanomolar ERK5 inhibition in the cell-free ERK5 assay,
and sub-micromolar activity in cell-based assays will be the subject of a future publication [29].
16

ACCEPTED MANUSCRIPT
EXPERIMENTAL METHODS
Biological Assay Protocols
ERK5 protein production
Cloning of ERK5. The cDNA encoding the catalytic domain of human ERK5 (2-409aa) (NCBI
Accession No. BC009963) was cloned into pFastBac-HTb (Invitrogen), which contains an N-
terminal 6 x His tag and a TEV cleavage site.
Cloning of
Accession No. BC008838) (MEK5DD) was generated by PCR-based site-directed mutagenesis
QuikChange II Site-Directed Mutagenesis kit, Stratagene). The dual phosphorylation sites,
∆N11MEK5DD. Constitutively active mutant form of human MEK5 (NCBI
(

serine 311 and threonine 315, were replaced with aspartic acid. Truncated MEK5DD lacking the
N-terminal 11 residues (13-448 aa), termed ∆N11MEK5DD, was cloned into pFastBacGST (a
modified pFastbac-1 vector that encodes an N-terminal GST tag and a thrombin cleavage site,
created in-house).
Baculovirus Production and Protein Expression. Cell Culture: Sf9 suspension cultures grown
in serum free SF900II medium (Invitrogen) were maintained in shaker flasks at 28
continuous shaking at 150 rpm.
°C with
GST-∆N11MEK5DD and ERK5 recombinant baculoviruses were generated utilizing the Bac-to-
Bac baculovirus expression system, as suggested by the manufacturer (Invitrogen).
ERK5 protein was co-expressed with GST-∆N11MEK5DD by co-infecting Sf9 cells, seeded at a
6
density of 1x10 cells/mL, with both ERK5 and GST-∆N11MEK5DD viruses at MOIs of 5. 72 h
17

ACCEPTED MANUSCRIPT
after infection, the cells were harvested by centrifugation at 1500 rpm for 10 min. The cell pellet
was washed with ice-cold PBS and the cells centrifuged as above. The cell pellet was flash
frozen in liquid nitrogen and stored at -80 °C until processing.
Protein Purification. 500 mL of harvested SF9 culture lysed by resuspending in 10 mL lysis
buffer (50 mM HEPES pH 7.87, 150 mM NaCl, 1% Triton X-100, 50 mM imidazole, 1 mini
EDTA-free complete protease inhibitor tablet (Roche), 1% phosphatase inhibitor cocktail 2
(
Sigma)) and rolling at 4 °C for 45 minutes. Insoluble material removed by centrifugation and
clarified supernatant applied to a 1 mL HisTrap HP column (GE) pre-equilibrated in Buffer A
50 mM HEPES pH 7.87, 150 mM NaCl, 50 mM imidazole). Column washed with 10 CV's
(
Buffer A. Bound protein eluted with a 20 CV linear gradient from Buffer A to Buffer B (50 mM
HEPES pH 7.87, 150 mM NaCl, 1 M imidazole); all steps run at 0.5 mL/min.
Fractions containing ERK5/MEK dialysed overnight against Buffer C (50 mM HEPES pH 7.87,
1
50 mM NaCl) using slidealyzer dialysis membrane, at 4 °C.
Dialysed protein loaded onto a 1 mL CaptoQ trap column (GE), washed with 10 CV's of Buffer
C. Bound protein eluted using an 80 CV linear gradient from Buffer C to Buffer D (50 mM
HEPES pH 7.87, 1 M NaCl), All steps carried out at 0.5 mL/min flow rates.
Fractions containing ERK5/MEK pooled and dialysed against Buffer E (50 mM HEPES pH
7
8
.87, 150 mM NaCl, 25% glycerol, 5 mM DTT), filter sterilised and 250 µL aliquots frozen at -
0 °C.
Batch variable yield from 75 – 240 µg/ml (440 µg – 1069 µg total protein) of ERK5/MEK
complex.
18

ACCEPTED MANUSCRIPT
Biochemical ERK5 activity assay. Compound IC50s were determined against ERK5 using
FAM-EGFR-derived peptide substrate (LVEPLTPSGEAPNQ(K-5FAM)-COOH, Molecular
Devices #R7129) and an IMAP FP Progressive Binding System kit (Molecular Devices
#
R8127). Compounds were serially diluted 3-fold to generate 10-point curves. Assays were
incubated for 2 h at 37 °C in a 10 µL reaction volume containing ERK5 (5-10 nM depending on
batch activity), 1x IMAP reaction buffer, 1mM DTT, 250 nM peptide substrate, 300 µM ATP
and 4% DMSO. Detection of phosphorylated product was performed by transferring 1 µL of the
enzyme reaction above with 9 µL of 1x IMAP reaction buffer and 30 µL of IMAP Binding
Solution, prepared as per the manufacturer’s instructions, and incubating for 2 h at RT.
Fluorescence polarisation was measured on a Pherastar (BMG) using an FP 485-520-520 optic
module.
Biochemical p38 activity assay. Compound IC s were determined against full length p38
50
protein (Millipore #14-251) in a LANCE assay format using a Ulight-MBP peptide substrate
PerkinElmer #TRF0109). Compounds were serially diluted as for the ERK5 biochemical assay.
(
Assays were incubated for 1 h at 37 °C in a 10 µL reaction volume containing p38 (1-3 nM
depending on batch activity), 50 mM Tris pH 7.5, 5mM MgCl , 0.5 mM EGTA, 2 mM DTT,
2
0
.01% Triton X100, 50 nM peptide substrate, 350 µM ATP and 4% DMSO. Quenching of the
reaction and detection of phosphorylated product were performed by addition of 10 µL of a
mixture of EDTA (5 mM final concentration) and LANCE Ultra Europium-anti-phospho-MBP
antibody (PerkinElmer #TRF0201, 2 nM final concentration), and incubating for 2 h at RT,
protected from light. FRET was measured on a Pherastar (BMG) using an HTRF 337-665-620
optic module.
19

ACCEPTED MANUSCRIPT
ERK5-Dependent Dual-Luciferase Reporter Assay. Given the paucity of validated ERK5
substrates we assessed cell-based activity against ERK5 using a transcriptional reporter assay in
which mutationally activated MEK5 (MEK5D) was co-expressed in cells together with ERK5
and the transactivation domain of the transcription factor MEF2D (an ERK5 substrate) fused to
the DNA binding domain of the yeast GAL4 protein. ERK5-dependent activity of GAL4-
MEF2D was assayed using a co-expressed luciferase reporter as previously described [30].
Plasmids. The HA-ERK5 and EGFP-MEK5D constructs have been described previously[30] and
were derived from constructs provided by J.D. Lee. GAL4-MEF2D was a kind gift from J.M.
McDermott and GAL4 LUC and CMV-Renilla reporters from D. Gillispie, Universidad de La
Laguna, Spain.
Cell culture. HEK293 were cultured in DMEM supplemented with 10% FBS at 37
atmosphere of 5% CO and 95% humidity as described previously [30].
°C in an
2
Luciferase assays. HEK293 were transfected using Lipofectamine 2000 (LifeTechnologies,
https://www.lifetechnologies.com/uk/en/home/brands/product-
brand/lipofectamine/lipofectamine-2000.html), according to the manufacturer’s instructions.
ERK5 inhibitors was added to a final concentrations of 30 µM, 10 µM, 3 µM, 1 µM and 0.3 µM
4
h post transfection. After a further 20 h cells were harvested and processed for firefly and
renilla luciferase activity using the Dual Luciferase Reporter (Promega,
http://www.promega.co.uk/products/reporter-assays-and-transfection/reporter-
assays/dual_luciferase-reporter-assay-system/) assay according to the manufacturer’s
instructions.
20

ACCEPTED MANUSCRIPT
Quantification of Cellular Inhibition of ERK5. Raw data was processed using Microsoft Excel
and GraphPad Prism 6.02, enabling generation of IC values. Each data point is the mean of 3
5
0
values ± standard deviation.
In vivo Experiments
All in vivo experiments were reviewed and approved by the Newcastle University Animal
Welfare and ethical review body, and performed according to the Guidelines for the Welfare set
out by an ad hoc committee of the National Cancer Research Institute (Workman et al, 2010) and
national law.
Matrigel Plug Assay Protocol. Matrigel containing basic fibroblast growth factor (bFGF),
which induces angiogenesis, was injected sub-cutanously into 15 8-10 week old female CD1
mice (Charles River, Kent UK). 24 hours post Matrigel inoculation, mice were randomised to
receive either vehicle, 50 mg/kg XMD8-92 i.p. twice daily for 7 days or 46 p.o. twice daily for 7
days. On day 7, mice were killed; Matrigel plugs were removed and snap frozen in liquid
nitrogen. Haemoglobin was measured using the Drabkin’s method. Drabkin’s reagent
(containing sodium bicarbonate, potassium ferricyanide and potassium cyanide) allowed
quantitative colorimetric determination of haemoglobin concentrations in blood.
Human Tumour Xenograft Model Protocol. Female CD1 nude (nu/nu) mice at 8-10 week old
(Charles River Kent UK) were inoculated with A2780 cells (human ovarian carcinoma) cells
subcutaneously on the right flank, and tumour volume was allowed to develop for 7 days to an
3
average volume of 72 mm . Mice were randomized (9 per group) to receive either vehicle, 50
mg/kg XMD8-92 i.p. or 100 mg/kg 46 p.o. twice daily for 10 days. Tumour volumes were then
measured using digital calipers three times per week. Statistical analysis of mean tumour
21

ACCEPTED MANUSCRIPT
volume, at the end of treatment, was performed using an unpaired two tailed t-test. Body weight
was monitored as an assessment of drug toxicity.
Pharmacokinetic (PK) Studies. Female CD1 mice 8-10 week old (Charles River, Kent UK)
were treated with 10 mg/kg of 46 either i.v. or p.o., and were bled by cardiac puncture under
terminal anaesthesia at selected time points post-treatment (0.083-24 hours, 3 mice/time point).
Blood was collected into heparinised tubes, and plasma was separated and stored at –20 °C until
analysed. Drug concentrations were determined by LCMS calibrated with standards prepared in
control plasma and pharmacokinetic parameters determined using a non-compartmental
(trapezoidal) method. Terminal half-life was determined using log linear regression of the final
2
six time points (r > 0.96).
Crystallographic Protocols. Preparation of ERK5-46 complex crystals. Purified
unphosphorylated ERK5 kinase domain (residues 46-402) was purchased from Proteros
Biostructures GmbH and co-crystals with compound were prepared in a similar manner as
-1
described [31]. ERK5 (46-402) at 11.5 mg mL in storage buffer (50 mM HEPES (pH 6.5), 150
mM NaCl, 10 % (v/v) glycerol, 2 mM DTT) was mixed with 46 (100 mM in 100 % DMSO) to
give a final concentration of 1 mM 46 and 1 % (v/v) DMSO. Complex formation was allowed to
proceed for two hours on ice. The sample was then clarified by centrifugation (5 min, 16 000g, 4
ºC) immediately before use in crystallisation. Crystals were grown by sitting drop vapour
diffusion at 20 ºC in 96-well MRC plates, by mixing the protein: compound complex with
crystallisation buffer comprising 5 % (v/v) PEG 6000, 0.1 M MES (pH 6.0), 5 mM DTT in a 1:1
ratio to give a 0.8 µL drop. Drops were immediately streak seeded with a seed stock prepared
from crystals of ERK5 with an indazole ERK5 inhibitor (in house series - unpublished structure).
The seed stock was prepared by looping two crystals of the ERK5: indazole complex into 2 µL
22

ACCEPTED MANUSCRIPT
crystallisation buffer (5 % (v/v) PEG 6000, 0.1 M MES (pH 6.0), 5 mM DTT). The buffer and
crystals were transferred into a micro-centrifuge tube containing a stabilisation buffer (20 µL 5
%
(v/v) PEG 6000, 0.1 M MES (pH 6.0), 5 mM DTT plus 80 µL 6 % (v/v) PEG 6000, 0.1 M
MES (pH 6.25), 5 mM DTT) and the crystals were crushed by vortexing with a Teflon bead.
The seed stock was aliquoted into cryo-tubes, flash frozen in liquid nitrogen, and stored at -80 ºC
until use.
X-ray diffraction data collection, structure solution and refinement for the complex of
ERK5 with 46. Crystals were passed briefly through a cryoprotectant solution comprising 4.9 %
(v/v) PEG 6000, 70 mM MES (pH 6.0), 3.5 mM DTT, 30 % (v/v) glycerol, 1 % (v/v) DMSO and
1
0 mM 46 before flash cooling in liquid nitrogen. Data were collected at 100 K on beamline I04
at the DIAMOND Light Source (Oxford, UK). Data were processed using XDS [32], and
programs from the CCP4 suite [33] as implemented within autoPROC [34]. The structure was
solved by molecular replacement using Molrep [35] and a high resolution structure of ERK5
(PDB ID 5BYZ [31]) as the search model. Rounds of manual model completion in Coot [36]
were interspersed with maximum-likelihood TLS refinement in Buster (v. 2.10.2) [37]. Ligand
libraries and coordinates were generated from the corresponding SMILES string using grade
(v1.2.9, Global Phasing Ltd., Cambridge, UK) and cross-validated against the CSD [38] using
Mogul [39] after fitting the ligand into difference density and reciprocal space refinement.
Crystallographic data statistics are provided in Table S1.
23

ACCEPTED MANUSCRIPT
BRD4 Expression, Purification and Surface Plasmon Resonance Protocols
Expression and purification of recombinant BRD4 bromodomain 1. Harvested bacterial cells
were resuspended in lysis buffer comprising 50 mM HEPES (pH 7.4), 200 mM NaCl, 10 mM
-
1
-1
imidazole, 0.5 mg mL lysosyme, and 0.2 mg mL DNAse at 4 ºC for 1 h. After sonication and
2
+
centrifugation (1 h at 35,000 x g), the supernatant was purified by immobilized Ni ion affinity
chromatography. The peak fractions were pooled and incubated with GST tagged HRV 3C
protease (50:1) at 4 ºC overnight. The cleaved His-tag was separated from BRD4 by size
exclusion chromatography using a Superdex 75 (26/60) column (GE Healthcare), equilibrated
and run in 50 mM HEPES (pH 7.4), 200 mM NaCl and 1 mM DTT. All purification steps were
performed using an ÄKTA Pure (GE Healthcare) at 4 ºC.
Surface Plasmon Resonance. SPR-based ligand binding assays were performed using a
BIAcore S200 (GE Healthcare) at 25 ºC using single cycle affinity. Immobilisation of BRD4 was
achieved using standard amine coupling on a CM5 chip surface. The surface was prepared
-
1
through activation with EDC/NHS, followed by injection of 10 µg mL BRD4 until a target
level of 8000 RU was reached. The surface was then quenched using 1 M ethanolamine and
washed with running buffer (10 mM HEPES, 150 mM NaCl, 0.01 % (v/v) TWEEN20, 0.5 mM
-1
TCEP and 1 % (v/v) DMSO) at a flow rate of 30 µL min . XMD8-92 and 46 were injected in a
dose-response manner (nine points ranging from 0-20 µM) with contact time 30sec and
dissociation time 160sec in series across the reference and BRD4-immobilised flow cells using
solvent correction to account for bulk refractive index changes. The reference channel response
was subtracted from the BRD4-immobilised channel response and dose-response data were fitted
using an affinity steady-state 1:1 binding model to determine the Kd.
24

ACCEPTED MANUSCRIPT
Accession codes
The coordinates for the ERK5-46 complex has been deposited in the wwPDB under accession
number 5O7I.
Summary of Generic Analytical and Chromatographic Conditions. All commercial reagents
were purchased from Sigma-Aldrich Chemical Company, Alfa Aesar, Apollo Scientific or
Tokyo Chemical Industry UK Ltd. The chemicals were of the highest available purity. Unless
otherwise stated, chemicals were used as supplied without further purification. Anhydrous
solvents were obtained from AcroSealTM or Aldrich SureSeal™ bottles and were stored under
nitrogen. Petrol refers to the fraction with a boiling point between 40 and 60 °C. Thin layer
chromatography utilised to monitor reaction progress was conducted on plates pre-coated with
2
silica gel Merck 60F254 or Merck NH F254S. The eluent was as stated (where this consisted of
more than one solvent, the ratio is stated as volume:volume) and visualisation was either by short
wave (254 nm) ultraviolet light, or by treatment with the visualisation reagent stated followed by
heating. ‘Flash’ medium pressure liquid chromatography (MPLC) was carried out either on a
Biotage SP4 automated purification system or a Varian 971-FP automated purification system,
using pre-packed Varian or Grace silica or amino-bonded silica cartridges. All reactions carried
out in a microwave were performed in a Biotage Initiator with Sixty robot. Melting points were
1
13
19
determined using a VWR Stuart SMP40 apparatus and are uncorrected. H, C and F nuclear
magnetic resonance (NMR) spectra were obtained as either CDCl , MeOD or DMSO-d
solutions and recorded at 500 MHz, 126 MHz and 471 MHz, respectively, on a Bruker Avance
3
6
1
3
III 500 spectrometer. Where C NMR data are not quoted, insufficient material was available or
problems obtaining high resolution spectra were encountered. Chemical shifts are quoted in parts
25

ACCEPTED MANUSCRIPT
per million (δ) referenced to the appropriate deuterated solvent employed. Multiplicities are
indicated by s (singlet), d (doublet), t (triplet), q (quartet), quin (quintet), m (multiplet), br
(broad) or combinations thereof. Coupling constant values are given in Hz. Homonuclear and
heteronuclear two dimensional NMR experiments were used where appropriate to facilitate
assignment of chemical shifts. LC-MS was carried out on a Waters Acquity UPLC system with
PDA and ELSD employing positive or negative electrospray modes as appropriate to the
individual compound. High resolution mass spectrometry (HRMS) was performed by the EPSRC
UK National Mass Spectrometry Facility, Swansea University, Singleton Park, Swansea, SA2
8
PP. FTIR spectra were recorded on either a Bio-Rad FTS 3000MX diamond ATR or an Agilent
Cary 630 FTIR as a neat sample. UV spectra were obtained using a U-2001 Hitachi
Spectrophotometer with the sample dissolved in ethanol. The purity of compounds was assessed
by high performance liquid chromatography (HPLC). All tested compounds were ≥95% purity.
The HPLC instrument was an Agilent 1200 equipped with a photodiode array detector (190-400
nm). Sample temperature, ambient; injection volume, 5 µL; flow rate, 1mL/min. 5% to 100%
MeCN gradient over 9 min and an isocratic hold at 100% MeCN for 2.5 min, before returning to
initial conditions. Mobile phase A = 0.1% ammonia in water or 0.1% formic acid in water,
mobile phase B = MeCN. Column: Waters XSELECT CSH C18, 5 µm, 4.6 mm × 150 mm.
Column maintained at ambient temperature.
General Synthetic Procedures
General procedure A: To a suspension of AlCl (2.5 eq.) in DCM (1 mL/mmol AlCl ) at 0 °C
3
3
was added the relevant acid chloride (2 eq.) followed by methyl 1H-pyrrole-2-carboxylate (1
eq.). The resulting mixture was allowed to reach RT and stirred for 16 h. The reaction was
quenched at 0 °C with a 1 M aqueous solution of HCl (20 mL). The product was extracted with
26

ACCEPTED MANUSCRIPT
3
DCM (3 × 100 mL), washed with an aqueous saturated solution of NaHCO (2 × 100 mL) and
brine (100 mL). Combined organic layers were dried over Na SO and concentrated in vacuo.
2
4
General procedure B: To a solution of pyrrole ester (1.0 eq.) in THF (8 mL/mmol) was added
LiOH (20 eq.) in water (13 mL/mmol). The resulting mixture was heated at 60 °C for 18 h,
cooled to RT and acidified to pH 4-5 with aq. HCl (1.0 M). The product was extracted into
EtOAc (100 mL/mmol), washed with water (100 mL/mmol), brine (100 mL/mmol) and dried
2 4
over Na SO . The solvent was removed under vacuum to obtain the product.
General procedure C: A solution of carbonyldiimidazole (2.0 eq.) and the required carboxylic
acid (1.0 eq.) in THF (5 mL/mmol) was heated to 70 ºC for 3 h. The appropriate amine (2.5 eq.)
was added and the mixture heated at 50 °C for 3 h then at RT for 16 h. The product was extracted
into EtOAc (50 mL/mmol), washed with water (50 mL/mmol), brine (50 mL/mmol) and dried
over Na SO . The solvent was removed under vacuum to give the crude product.
2
4
General procedure D: The appropriate carboxylic acid (1.0 equiv.) was dissolved in MeCN (5
mL/mmol pyrrole) before the relevant amine (2.5 equiv.) was added followed by phosphorus
trichloride (1.0 equiv.). The mixture was heated using microwave irradiation at 150 °C for 5 min.
The reaction was quenched with a few drops of H O and the solvent removed in vacuo. The
2
residue was re-dissolved in EtOAc (50 mL/mmol pyrrole) and washed with a saturated aqueous
solution of NaHCO (50 mL/mmol pyrrole) before being extracted with EtOAc (3 × 30
3
2 4
mL/mmol pyrrole). The combined organic extracts were dried over Na SO , and the filtrate
concentrated in vacuo to afford the crude product.
27

ACCEPTED MANUSCRIPT
Compound Syntheses
Methyl-4-benzoyl-1
H-pyrrole-2-carboxylate (7) Prepared according to general procedure A
using methyl-2-pyrrole carboxylate (112 mg, 0.90 mmol), benzoyl chloride (205 µL, 1.79 mmol)
and aluminium chloride (298 mg, 2.23 mmol) in DCM (2.2 mL). Purification via MPLC (silica;
0
-40% EtOAc in petroleum ether) gave 7 as a yellow solid (199 mg, 96%); R 0.40 (50% EtOAc
f
-1
in petroleum ether); mp 148-149 °C; λmax (EtOH)/nm 285, 238; IR υmax/cm 3293, 1715, 1620,
1
1
7
596, 1555; H NMR (300 MHz, DMSO-d
6
)
δ
3.81 (3H, s, OCH
3
), 7.15 (1H, s, H-pyrrole), 7.48-
1
3
.67 (4H, m, 3 × H-Ar and H-pyrrole), 7.73-7.83 (2H, m, 2 x H-Ar), 12.76 (1H, s, NH); C
30.1 (OCH ), 115.5 (CH-pyrrole), 123.2 (C), 124.2 (C), 128.0 (4 x
NMR (75 MHz, DMSO-d6)
δ
3
C-Ar), 128.7 (CH-pyrrole), 131.3 (C-Ar), 138.6 (C-Ar), 160.1 (COOMe), 188.6 (CO); HRMS
+
+
(
ES ) calcd for C H O N [M+H] 230.0812, found 230.0811; Anal. calcd for C H NO : C,
13 12 3 13 11 3
6
8.11%, H, 4.84%, N, 6.11%, found: C, 68.33%, H, 4.78%, N, 6.21%.
-pyrrole-2-carboxylate (8) Prepared according to general
procedure A using methyl-2-pyrrole carboxylate (3.0 g, 24.0 mmol), 2-fluorobenzoyl chloride
5.7 mL, 48.0 mmol) and aluminium chloride (5.52 g, 60.0 mmol) in DCM (60 mL). Purification
via precipitation (EtOAc in petroleum ether) gave 8 as a yellow solid (4.16 g, 70%); mp 129-130
Methyl 4-(2-fluorobenzoyl)-1
H
(
-
1
1
°
C; λmax (EtOH)/nm 284, 234; IR υmax/cm 3291, 1722, 1632, 1553; H NMR (300 MHz,
DMSO-d6) 3.79 (3H, s, OCH ), 7.06 (1H, s, H-pyrrole), 7.27-7.37 (2H, m, H-pyrrole and H-
Ar), 7.40-7.66 (3H, m, 3 x H-Ar), 12.79 (1H, s, NH); C NMR (75 MHz, DMSO-d
OCH ), 114.8 (CH-pyrrole), 115.8 (d, J = 21.5 Hz, C-Ar), 123.6 (C), 124.1 (d, JC-F = 3.3 Hz,
δ
3
1
3
6
) δ 51.1
(
3
C-F
C-Ar), 125.2 (C), 127.7 (d, JC-F = 15.3 Hz, C), 129.1 (CH-pyrrole), 129.2 (d, JC-F = 3.1 Hz, C-
Ar), 132.2 (d, JC-F = 8 Hz, ArH), 158.5 (d, JC-F = 247.3 Hz, C-F), 160.0 (COOMe), 185.6 (CO);
28