Discovery and validation of 2-styryl substituted benzoxazin-4-ones as a novel scaffold for rhomboid protease inhibitors

Article abstract of DOI:10.1016/j.bmcl.2018.02.017

Rhomboids are intramembrane serine proteases with diverse physiological functions in organisms ranging from archaea to humans. Crystal structure analysis has provided a detailed understanding of the catalytic mechanism, and rhomboids have been implicated in various disease contexts. Unfortunately, the design of specific rhomboid inhibitors has lagged behind, and previously described small molecule inhibitors displayed insufficient potency and/or selectivity. Using a computer-aided approach, we focused on the discovery of novel scaffolds with reduced liabilities and the possibility for broad structural variations. Docking studies with the E. coli rhomboid GlpG indicated that 2-styryl substituted benzoxazinones might comprise novel rhomboid inhibitors. Protease in vitro assays confirmed activity of 2-styryl substituted benzoxazinones against GlpG but not against the soluble serine protease α-chymotrypsin. Furthermore, mass spectrometry analysis demonstrated covalent modification of the catalytic residue Ser201, corroborating the predicted mechanism of inhibition and the formation of an acyl enzyme intermediate. In conclusion, 2-styryl substituted benzoxazinones are a novel rhomboid inhibitor scaffold with ample opportunity for optimization.

Full text of DOI:10.1016/j.bmcl.2018.02.017

Accepted Manuscript
Discovery and validation of 2-styryl substituted benzoxazin-4-ones as a novel
scaffold for rhomboid protease inhibitors
Parul Goel, Thorsten Jumpertz, Anežka Tichá, Isabella Ogorek, David C.
Mikles, Martin Hubalek, Claus U. Pietrzik, Kvido Strisovsky, Boris Schmidt,
Sascha Weggen
PII:
S0960-894X(18)30106-9
https://doi.org/10.1016/j.bmcl.2018.02.017
BMCL 25613
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
14 November 2017
5 February 2018
8 February 2018
Please cite this article as: Goel, P., Jumpertz, T., Tichá, A., Ogorek, I., Mikles, D.C., Hubalek, M., Pietrzik, C.U.,
Strisovsky, K., Schmidt, B., Weggen, S., Discovery and validation of 2-styryl substituted benzoxazin-4-ones as a
novel scaffold for rhomboid protease inhibitors, Bioorganic & Medicinal Chemistry Letters (2018), doi: https://
doi.org/10.1016/j.bmcl.2018.02.017
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.

Bioorganic & Medicinal Chemistry Letters
Discovery and validation of 2-styryl substituted benzoxazin-4-ones as a novel
scaffold for rhomboid protease inhibitors
a
b
a
b
Parul Goel a,d, Thorsten Jumpertz , Anežka Tichá , Isabella Ogorek , David C. Mikles ,
b
c
b
d,
Martin Hubalek , Claus U. Pietrzik , Kvido Strisovsky , Boris Schmidt *, Sascha Weggen
a,
*
a Department of Neuropathology, Heinrich-Heine University Duesseldorf, Moorenstrasse 5,
40225 Duesseldorf, Germany
b
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech
Republic, Flemingovo n. 2, 166 10, Praha 6, Czech Republic
c Institute for Pathobiochemistry, University Medical Center of the Johannes Gutenberg
University Mainz, Duesbergweg 6, 55128 Mainz, Germany
d Clemens Schoepf Institute for Organic Chemistry and Biochemistry, Technische
Universitaet Darmstadt, Alarich-Weiss-Strasse 4-8, 64287 Darmstadt, Germany
* Corresponding authors: schmidt_boris@t-online.de (B. Schmidt); sweggen@hhu.de (S.
Weggen)
Keywords: rhomboid proteases, intramembrane proteases, benzoxazinones, molecular
docking, inhibition
1

ABSTRACT
Rhomboids are intramembrane serine proteases with diverse physiological functions in
organisms ranging from archaea to humans. Crystal structures analysis has provided a
detailed understanding of the catalytic mechanism, and rhomboids have been implicated in
various disease contexts. Unfortunately, the design of specific rhomboid inhibitors has
lagged behind, and previously described small molecule inhibitors displayed insufficient
potency and/or selectivity. Using a computer-aided approach, we focused on the discovery of
novel scaffolds with reduced liabilities and the possibility for broad structural variations.
Docking studies with the E.coli rhomboid GlpG indicated that 2-styryl substituted
benzoxazinones might comprise novel rhomboid inhibitors. Protease in vitro assays
confirmed activity of 2-styryl substituted benzoxazinones against GlpG but not against the
soluble serine protease -chymotrypsin. Furthermore, mass spectrometry analysis
demonstrated covalent modification of the catalytic residue Ser201, corroborating the
predicted mechanism of inhibition and the formation of an acyl enzyme intermediate. In
conclusion, 2-styryl substituted benzoxazinones are a novel rhomboid inhibitor scaffold with
ample opportunity for optimization.
2

RESULTS AND DISCUSSION
Rhomboids are intramembrane serine proteases present in prokaryotic, archaeal and
eukaryotic organisms 1. In 2001, the first rhomboid was discovered in Drosophila and shown
to perform a critical proteolysis step in EGF-receptor signaling 2, 3. Since then, rhomboids
have been implicated in a wide range of biological processes including bacterial quorum
4
7
sensing , mitochondrial dynamics and integrity 5, 6, and protein quality control . In addition,
8
rhomboids have been identified as putative drug targets in the context of multiple diseases
9
5
such as cancer , diabetes 10, 11, parasitic diseases 12, 13, and Parkinson’s disease . The
crystal structures of rhomboids from E. coli and H. influenzae have been solved and revealed
that rhomboids are serine-histidine dyad proteases composed of 6 core transmembrane
helices, which form a V-shaped cavity and expose the active site to a partially hydrophilic
14,
environment
15. These structures together with numerous biochemical studies have
provided a detailed understanding of the catalytic mechanism of rhomboid proteases 16, but
this has not yet translated into the development of potent, selective and drug-like inhibitors 17.
Through different strategies, from the testing of candidate molecules to rational synthesis to
the screening of small molecule libraries, isocoumarins 3, 18, 19, fluorophosphonates 20, -
22
lactams 21, and -lactones
were found to be effective against rhomboids, but these
inhibitors generally displayed low potency and/or insufficient selectivity 18, 20, 21. Effectively,
inherent liabilities as exemplified by the high reactivity of isocoumarins likely preclude or limit
further development of these compound classes 23, 24
.
Accordingly, using a computer-aided candidate approach, we focused on the discovery of
novel scaffolds with reduced liabilities and the possibility for broad structural variations. One
scaffold we selected was 2-substituted derivatives of 4H-3,1-benzoxazin-4-ones, which were
previously used as heterocyclic acylating agents against serine proteases such as HLE, α-
chymotrypsin, and cathepsin G 23, 25-28. The mechanism of inhibition involves the formation of
an O-acyl enzyme intermediate. The nucleophilic serine reacts with the C-4 carbonyl of the
benzoxazinone, which results in opening of the heterocyclic ring and formation of the O-acyl
enzyme intermediate (Fig. 1A) 23. The enzyme selectivity and potency of acylating agents is
promoted by fast acylation and slow deacylation, which is dependent on the substitution of
the aromatic ring and the C-2 position in case of benzoxazinones 25, 29, 30. A major advantage
of benzoxazin-4-ones is that the core structure consists of two fused aromatic rings, which
allows extensive structural modifications and optimization with respect to the target enzyme.
For initial docking studies into the rhomboid active site, we assembled a molecular database
of thirteen 2-alkyl or 2-aryl substituted benzoxazinones (Fig. 1B).
3

FIGURE 1
In the docking studies, we focused on the initial interactions between the benzoxazinones
and the active site of the rhomboid protease rather than the final reaction product. For
preparation of the docking receptor, we used the co-crystal structure of the E.coli rhomboid
GlpG and the fluorophosphonate inhibitor CAPF (PDB ID: 3UUB), in which the active site
Ser201 is covalently bound to CAPF 31. The molecular modelling experiments were
performed in the molecular operating environment software (MOE) with the DOCK module
and the MMFF94x force field, and scored by London dG and Affinity dG followed by energy
minimization within the enzyme active site cleft 32, 33. The output data was ranked based on
the calculated ligand efficiencies (cLE = docking score / number of heavy atoms) 34, which
revealed that the 2-styryl substituted compound 3 was the most favorable of all 2-substituted
benzoxazinones (cLE=-0.3164). A comparative analysis of the protein/ligand docking results
4

of compound 3 and CAPF indicated that 3 was adequately fitting into the binding pocket of
the enzyme and was not exposed to the external environment (Fig. 2A and B). The core
heterocyclic ring of 3 was oriented towards the S1 subsite while the 2-styryl extension
pointed towards the S2’ subsite of the rhomboid, which had been defined in previous
structures of GlpG in complex with different inhibitors 31, 35. Moreover, close interactions of 3
with the neighbouring residues His254 and Phe245 as shown in the ligand interaction map
were observed and suggested to further explore the scaffold (Fig. 2C).
FIGURE 2
5

To validate the docking results, all derivatives listed in Table 1 were synthesized by
methods shown in schemes 1 and 2 (Fig 3A) 36. The benzoxazinone derivatives were then
evaluated for their inhibitory potency in an established in vitro enzyme activity assay with the
E.coli rhomboid GlpG and the transmembrane domain 2 of the Drosophila protein Gurken as
a substrate 21, 37, 38, 39. Each of the benzoxazinones were pre-incubated with GlpG at a single
concentration of 250 M for 30 min at 37°C. Subsequently, the Gurken substrate was added,
the reaction was continued for another 90 min at 37°C, and the N-terminal Gurken substrate
cleavage fragment was visualized by SDS-PAGE and quantified using ImageJ. Only the 2-
styryl substituted benzoxazinones 3, 5 and 11 showed activity at this concentration. For IC50
40, 41,
determinations, fluorogenic rhomboid substrates were applied as described previously
42. Correlating well with the docking results, compound 3 was a potent rhomboid inhibitor with
an IC50 value of 4.4 ± 1.6 M (Figure 3B). Compound 5 was equally potent (IC50 3.7 ± 1.3
M) while 11 displayed around 10-fold lower activity (IC50 48 ± 14.1 M). Among these three
compounds with a single substitution at the aromatic ring of the styryl substituent, an electron
withdrawing group appeared to increase the potency, which could be further explored in
subsequent studies.
FIGURE 3
Next, we evaluated the active benzoxazinones in a well-established in vitro activity assay
6

for the soluble serine protease -chymotrypsin 43,44. The compounds were pre-incubated with
bovine -chymotrypsin for 30 min at 25°C. Subsequently the substrate N-succinyl-Ala-Ala-
Pro-Phe-p-nitroanilide was added, which under alkaline conditions is turned over by -
chymotrypsin to p-nitroanilline, a yellow compound that can be detected spectroscopically at
410 nm. DCI was used as a positive control and inhibited -chymotrypsin with an IC50 value
of 3.5 µM, comparable to previously reported values 21. In contrast, the benzoxazinones 3
and 5 did not display any inhibitory activity in the -chymotrypsin in vitro activity assay at the
highest concentration of 250 M (data not shown). In addition, we examined the active
benzoxazinones in analogous in vitro protease activity assays for bovine trypsin, human
neutrophil elastase, and human cathepsin G 45. At a concentration of 10 M, none of the
benzoxazinones inhibited trypsin or neutrophil elastase, while DCI almost completely
abolished the activity of both enzymes. At a concentration of 50 M, trypsin activity was
reduced around 50% by compound 3, and neutrophil elastase activity was reduced around
50% by compound 11 (Supplemental Figures 1 and 2). In contrast, the compounds were
more effective against cathepsin G. Both 3 and 5 but not 11 caused significant enzyme
inhibition at 10 M (Supplemental Figure 3). Overall, these results indicated that 2-styryl
substituted benzoxazinones might possess some selectivity for the rhomboid GlpG over
other soluble serine proteases. Notably, in an accompanying paper, Yang, Verhelst and
46
colleagues presented benzoxazinones with a 2-alkoxy substituent as rhomboid inhibitors
.
These derivatives displayed slightly higher potency for GlpG but no apparent selectivity over
bovine chymotrypsin and trypsin, suggesting potential advantages of the docking approach
to optimize the scaffold for this particular target.
The inhibitory activity of 2-styryl substituted benzoxazinones in the GlpG in vitro activity
assay indicated that this scaffold contained active rhomboid inhibitors. However, to ensure
that the reduced GlpG activity was not caused by non-specific effects and that the
mechanism of inhibition conformed to the known reaction mechanism of benzoxazinones
with soluble serine proteases 23, we used mass spectrometry to study the residues in the
rhomboid’s active site that were reacting with the benzoxazinones. GlpG mutants were
generated, expressed and purified, in which Ser201 was exchanged to threonine (S201T) or
in which His150 or His254 were exchanged to alanine. The recombinant proteins were then
reacted in vitro with the 2-styryl substituted benzoxazinones 3 or 5, and binding of the
compounds to the rhomboid was examined by electrospray mass spectrometry. When the 2-
styryl substituted benzoxazinones were reacted with wild type GlpG, two major peaks were
observed in the spectra: the first smaller peak corresponding to the free unbound GlpG and
the second major peak corresponding to the inhibitor-bound enzyme (Fig. 4A). Importantly,
the shifts in the rhomboid’s mass for compounds 3 (283.4 Da) and 5 (248.9 Da) were close
to the theoretical mass differences of 285.73 and 251.29 Da that would occur when the
7

benzoxazinones reacted with the enzyme according to the mechanism shown in Figure 1. In
contrast, no binding and mass shift was seen when the compounds were omitted (vehicle
control) or when the rhomboid was reacted with the inactive benzoxazinone compound 4,
supporting the specificity of the mass spectrometry analysis (Fig. 4A). Further analysis
showed that compound 3 was not able to bind to either the S201T or the H254A mutant (Fig.
4A). This indicated that the benzoxazinone indeed reacted with the catalytic Ser201 to form
an O-acyl enzyme intermediate (Fig. 4B). In the catalytic Ser201-His254 dyad, the histidine is
required to properly activate the serine 16. Hence, in the H254A mutant the serine is likely not
nucleophilic enough to react with the C-4 carbonyl of the benzoxazinone, which could explain
that 3 failed to bind this mutant. However, mutating His150, which contributes to the
oxyanion hole of the active site and is covalently modified by some isocoumarins 19, 47, did
not prevent binding of 3 to the rhomboid as expected (Fig. 4B).
FIGURE 4
8

Formation of the O-acyl enzyme intermediate should be reversible over time (Fig. 1). To
48,49
test this, we used the rapid dilution method
and a previously described fluorogenic
peptidic rhomboid substrate 40. Compound 3 (50 M) was pre-incubated with GlpG for 1 h.
Subsequently, the reaction mixture was rapidly diluted 100-fold and the substrate KSp76 was
added. Cleavage of this substrate by the rhomboid leads to the release of a quencher
peptide and activation of a red fluorophore, which indicates recovery of enzyme activity. This
showed that GlpG activity largely recovered over a time period of 100 min demonstrating
reversibility of the inhibition mechanism (Fig. 4C). Similar results were obtained with the -
lactam L29, a known reversible inhibitor of rhomboids 21. Conversely, pre-incubation with the
isocoumarin JLK-6, an irreversible inhibitor that forms a double-bonded end product with the
rhomboid 19, did not allow recovery of enzyme activity (Fig. 4C).
In summary, by combining molecular docking studies with enzyme activity assays, we
found that 2-styryl substituted benzoxazin-4-ones are a novel template to generate inhibitors
for rhomboid proteases. Mechanistic studies indicated that benzoxazinone inhibitors
covalently modified the catalytic serine residue in the rhomboid active site, analogous to the
reaction mechanism of benzoxazinones with soluble serine proteases. While the identified
active 2-styryl substituted benzoxazinones at present have moderate potency, this new
scaffold allows extensive structural variations with considerable potential to increase potency
and specificity, towards the goal of improved small molecule inhibitors for rhomboid
proteases.
9

ACKNOWLEGEMENTS
We thank Professors Matthew Freeman (University of Oxford) and Ya Ha (Yale School of
Medcine) for bacterial expression constructs. P.G. was supported by a scholarship of the
iGRASPseed graduate school of the Heinrich-Heine-University Duesseldorf. K.S. was a
recipient of the Purkyne Fellowship of the Academy of Sciences of the Czech Republic and
also acknowledges support from EMBO (Installation Grant no. 2329), Ministry of Education,
Youth and Sports of the Czech Republic (projects no. LK11206 and LO1302), Marie Curie
Career Integration Grant (project no. 304154), and the National Subvention for Development
of Research Organisations (RVO: 61388963) to the Institute of Organic Chemistry and
Biochemistry.
CONFLICTS OF INTEREST: none
10

FIGURE LEGENDS
Figure 1. (A) Mechanism of inhibition of soluble serine proteases by 2-substituted
benzoxazin-4-one derivatives 30. (B) Molecular database of 2-alkyl or aryl substituted
benzoxazin-4-ones assembled for docking studies into the rhomboid active site.
Figure 2. Molecular docking model of the rhomboid protease GlpG in complex with a
benzoxazinone-based inhibitor. (A) Front view of GlpG docked with the 2-styryl substituted
compound 3 (yellow) and the known rhomboid inhibitor CAPF (blue). The model indicated
that both compounds occupied a similar space in the rhomboid pocket and were not exposed
to the external surface. (B) Top view of compound 3 and CAPF in the GlpG pocket. (C)
Ligand interaction map of compound 3 in the GlpG pocket. Significant interactions were
observed with neighbouring residues His254 and Phe245. Green: hydrophobic, pink: polar,
red: exposed.
Figure 3. (A) Synthesis of 2-substituted benzoxazin-4-ones. Scheme 1 shows the synthesis
of 2-alkyl/aryl substituted benzoxazinones (compounds 1 and 3-13) obtained by the reaction
of corresponding acyl/benzoyl chlorides with anthranilic acid. Scheme 2 shows the synthesis
of 2-methyl benzoxazin-4-one (compound 2) obtained by a microwave reaction of anthranilic
acid with acetic anhydride. (B) Biological evaluation of 2-alkyl and 2-aryl substituted
benzoxazinones inhibitors in rhomboid in vitro activity assays. All compounds were first
screened in a gel-based activity assay for the E.coli rhomboid GlpG at a single concentration
of 250 M as described 39. Only the 2-aryl substituted benzoxazinones were found to be
active at this concentration and were selected for IC50 determinations. In these assays, the
rhomboid GlpG was pre-incubated with increasing concentrations of the compounds 3, 5 and
11 for 1 h at 37°C. Subsequently, the reaction was started by the addition of fluorogenic
substrates 40. These quenched fluorescent peptides are cleaved by the rhomboid, leading to
the activation of a fluorophore. Fluorescence intensities were normalized to the DMSO
control condition and plotted against log (inhibitor concentration) in GraphPad Prism
software. The figure shows a representative dose-response curve for compound 3. A
40
nonlinear regression curve fit was used to determine apparent IC50 values as described
.
Compounds 3 (4.4 ± 1.6 M) and 5 (3.7 ± 1.3 M) were equally potent while 11 (48 ± 14.1
M) displayed around 10-fold lower activity. Two independent IC50 determinations were
performed for each compound, and IC50 values represent averages ± SD.
Figure 4. Reaction mechanism of 2-styryl substituted benzoxazinone inhibitors. (A) Wild type
GlpG was recombinantly expressed and purified as described previously 40, reacted in vitro
11

with compounds 3 or 5, and binding to the rhomboid was examined by electrospray mass
spectrometry. A second major peak was observed in the spectra corresponding to the
inhibitor-bound enzyme with the expected mass shift compared to the free enzyme. In
contrast, no binding was seen when the compounds were omitted (vehicle control) or when
the rhomboid was reacted with the inactive benzoxazinone 4. The results of one of two
independent experiments are shown. (B) The 2-styryl substituted benzoxazinone 3 was
incubated with different rhomboid mutants and binding was analyzed as in A. Compound 3
was able to bind to wild type GlpG and the H150A mutant with identical mass shifts between
the inhibitor-bound and the free enzyme. However, no binding was observed to either the
S201T or the H254A mutant. This indicated that Ser201 and proper activation of this catalytic
serine by His254 was required for binding of the benzoxazinone to the rhomboid and
formation of the O-acyl enzyme intermediate. The results of one of two independent
experiments are shown. (C) Reversibility of the reaction was tested by the rapid dilution
method. Compound 3 (50 M) was pre-incubated with the rhomboid GlpG for 1 h.
Subsequently, the reaction mixture was rapidly diluted 100-fold with reaction buffer
containing a fluorogenic rhomboid substrate (10 M). GlpG activity largely recovered over a
time period of 120 min demonstrating reversibility of the inhibition mechanism. No recovery
was observed when the reaction mixture was diluted into buffer containing substrate and 50
M compound 3. The known reversible and irreversible rhomboid inhibitors -lactam L29 and
isocoumarin JLK-6 were used as controls and displayed the expected behaviour. The results
of one of two independent experiments are shown.
12

REFERENCES AND NOTES
1.
2.
3.
4.
Freeman M. The rhomboid-like superfamily: molecular mechanisms and biological
roles. Annu Rev Cell Dev Biol 2014, 30, 235-254.
Lee J.R.; Urban S.; Garvey C.F.; Freeman M. Regulated intracellular ligand transport
and proteolysis control EGF signal activation in Drosophila. Cell 2001, 107, 161-171.
Urban S.; Lee J.R.; Freeman M. Drosophila rhomboid-1 defines a family of putative
intramembrane serine proteases. Cell 2001, 107, 173-182.
Clemmer K.M.; Sturgill G.M.; Veenstra A.; Rather P.N. Functional characterization of
Escherichia coli GlpG and additional rhomboid proteins using an aarA mutant of
Providencia stuartii. Journal of bacteriology 2006, 188, 3415-3419.
Chan E.Y.; McQuibban G.A. The mitochondrial rhomboid protease: its rise from
obscurity to the pinnacle of disease-relevant genes. Biochim Biophys Acta 2013,
1828, 2916-2925.
5.
6.
7.
McQuibban G.A.; Saurya S.; Freeman M. Mitochondrial membrane remodelling
regulated by a conserved rhomboid protease. Nature 2003, 423, 537-541.
Fleig L.; Bergbold N.; Sahasrabudhe P.; Geiger B.; Kaltak L.; Lemberg M.K.
Ubiquitin-dependent intramembrane rhomboid protease promotes ERAD of
membrane proteins. Mol Cell 2012, 47, 558-569.
8.
9.
Dusterhoft S.; Kunzel U.; Freeman M. Rhomboid proteases in human disease:
Mechanisms and future prospects. Biochim Biophys Acta 2017, 1864, 2200-2209.
Song W.; Liu W.; Zhao H., et al. Rhomboid domain containing 1 promotes colorectal
cancer growth through activation of the EGFR signalling pathway. Nat Commun
2015, 6, 8022.
10.
Civitarese A.E.; MacLean P.S.; Carling S., et al. Regulation of skeletal muscle
oxidative capacity and insulin signaling by the mitochondrial rhomboid protease
PARL. Cell Metab 2010, 11, 412-426.
11.
12.
Walder K.; Kerr-Bayles L.; Civitarese A., et al. The mitochondrial rhomboid protease
PSARL is a new candidate gene for type 2 diabetes. Diabetologia 2005, 48, 459-468.
Parussini F.; Tang Q.; Moin S.M.; Mital J.; Urban S.; Ward G.E. Intramembrane
proteolysis of Toxoplasma apical membrane antigen 1 facilitates host-cell invasion
but is dispensable for replication. Proc Natl Acad Sci U S A 2012, 109, 7463-7468.
Santos J.M.; Ferguson D.J.; Blackman M.J.; Soldati-Favre D. Intramembrane
cleavage of AMA1 triggers Toxoplasma to switch from an invasive to a replicative
mode. Science 2011, 331, 473-477.
Lemieux M.J.; Fischer S.J.; Cherney M.M.; Bateman K.S.; James M.N. The crystal
structure of the rhomboid peptidase from Haemophilus influenzae provides insight
into intramembrane proteolysis. Proc Natl Acad Sci U S A 2007, 104, 750-754.
Wang Y.; Zhang Y.; Ha Y. Crystal structure of a rhomboid family intramembrane
protease. Nature 2006, 444, 179-180.
Ha Y.; Akiyama Y.; Xue Y. Structure and mechanism of rhomboid protease. J Biol
Chem 2013, 288, 15430-15436.
Strisovsky K. Rhomboid protease inhibitors: Emerging tools and future therapeutics.
Semin Cell Dev Biol 2016, 60, 52-62.
Urban S.; Wolfe M.S. Reconstitution of intramembrane proteolysis in vitro reveals that
pure rhomboid is sufficient for catalysis and specificity. Proc Natl Acad Sci U S A
2005, 102, 1883-1888.
13.
14.
15.
16.
17.
18.
19.
20.
Vosyka O.; Vinothkumar K.R.; Wolf E.V.; Brouwer A.J.; Liskamp R.M.; Verhelst S.H.
Activity-based probes for rhomboid proteases discovered in a mass spectrometry-
based assay. Proc Natl Acad Sci U S A 2013, 110, 2472-2477.
Xue Y.; Ha Y. Catalytic mechanism of rhomboid protease GlpG probed by 3,4-
dichloroisocoumarin and diisopropyl fluorophosphonate. J Biol Chem 2012, 287,
3099-3107.
13

21.
22.
Pierrat O.A.; Strisovsky K.; Christova Y., et al. Monocyclic beta-lactams are selective,
mechanism-based inhibitors of rhomboid intramembrane proteases. ACS Chem Biol
2011, 6, 325-335.
Wolf E.V.; Zeissler A.; Vosyka O.; Zeiler E.; Sieber S.; Verhelst S.H. A new class of
rhomboid protease inhibitors discovered by activity-based fluorescence polarization.
PLoS One 2013, 8, e72307.
23.
24.
Powers J.C.; Asgian J.L.; Ekici O.D.; James K.E. Irreversible inhibitors of serine,
cysteine, and threonine proteases. Chem Rev 2002, 102, 4639-4750.
Powers J.C.; Kam C.M.; Narasimhan L.; Oleksyszyn J.; Hernandez M.A.; Ueda T.
Mechanism-based isocoumarin inhibitors for serine proteases: use of active site
structure and substrate specificity in inhibitor design. J Cell Biochem 1989, 39, 33-46.
Gutschow M.; Kuerschner L.; Neumann U., et al. 2-(diethylamino)thieno1,3oxazin-4-
ones as stable inhibitors of human leukocyte elastase. J Med Chem 1999, 42, 5437-
5447.
25.
26.
27.
Gutschow M.; Neumann U. Inhibition of cathepsin G by 4H-3,1-benzoxazin-4-ones.
Bioorg Med Chem 1997, 5, 1935-1942.
Krantz A.; Spencer R.W.; Tam T.F., et al. Design and synthesis of 4H-3,1-
benzoxazin-4-ones as potent alternate substrate inhibitors of human leukocyte
elastase. J Med Chem 1990, 33, 464-479.
28.
Teshima T.; Griffin J.C.; Powers J.C. A new class of heterocyclic serine protease
inhibitors. Inhibition of human leukocyte elastase, porcine pancreatic elastase,
cathepsin G, and bovine chymotrypsin A alpha with substituted benzoxazinones,
quinazolines, and anthranilates. J Biol Chem 1982, 257, 5085-5091.
Gutschow M.; Schlenk M.; Gab J., et al. Benzothiazinones: a novel class of
adenosine receptor antagonists structurally unrelated to xanthine and adenine
derivatives. J Med Chem 2012, 55, 3331-3341.
Zhong J.; Groutas W.C. Recent developments in the design of mechanism-based and
alternate substrate inhibitors of serine proteases. Curr Top Med Chem 2004, 4, 1203-
1216.
29.
30.
31.
Xue Y.; Chowdhury S.; Liu X.; Akiyama Y.; Ellman J.; Ha Y. Conformational change
in rhomboid protease GlpG induced by inhibitor binding to its S' subsites.
Biochemistry 2012, 51, 3723-3731.
32.
33.
MOE. http://github.com/Yelp/MOE. 2013.
For a detailed description of the docking procedures, please refer to the
Supplementary Material.
34.
35.
Hopkins A.L.; Groom C.R.; Alex A. Ligand efficiency: a useful metric for lead
selection. Drug Discov Today 2004, 9, 430-431.
Vinothkumar K.R.; Pierrat O.A.; Large J.M.; Freeman M. Structure of rhomboid
protease in complex with beta-lactam inhibitors defines the S2' cavity. Structure
(London, England : 1993) 2013, 21, 1051-1058.
36.
37.
For detailed synthesis information and the chemical data, please refer to the
Supplementary Material
Strisovsky K.; Sharpe H.J.; Freeman M. Sequence-specific intramembrane
proteolysis: identification of a recognition motif in rhomboid substrates. Mol Cell 2009,
36, 1048-1059.
38.
Bacterial expression constructs for N-terminally His-tagged GlpG and C-terminally
His-tagged Gurken chimeric substrate (MBP-GurkenTMD-Trx-His6) were kind gifts of
Ya Ha (Yale School of Medicine) and Matthew Freeman (University of Oxford). Both
enzyme and substrate were expressed in E. coli C43(DE3) cells. After induction with
IPTG, GlpG was expressed overnight at 18°C, while Gurken was expressed for 3h at
37°C. The bacterial cells were broken with a nitrogen cavitation bomb, membranes
were isolated by differential centrifugation and proteins were solubilized with 1.5%
dodecyl-beta-maltoside (DDM, Glycon Biochemicals). GlpG and Gurken were purified
via immobilized metal-ion affinity chromatography using a Talon®Crude 1 ml HiTrap
column and an Akta prime plus chromatography system (GE Healthcare). The
recombinant proteins were further concentrated using 10 or 30 kDa molecular weight
14

cut-off Amicon® Ultra Centrifugal filters (Millipore). Protein purity was estimated by
Coomassie Brilliant Blue (Merck) stained SDS-PAGE. Enzyme activity assays were
performed in 20 µl reaction buffer [50 mM HEPES-NaOH (pH 7.5); 5 mM EDTA; 0.4
M NaCl; 10% Glycerol (v/v); 0.05% DDM] containing GlpG (0.35 µM) and Gurken
substrate (1.8 µM). Benzoxazinone inhibitors were pre-incubated with GlpG in
reaction buffer for 30 min at 37°C with gentle shaking. The substrate was added and
the reaction was continued for another 90 min at 37°C. Subsequently, the reaction
mixture was separated by SDS-PAGE and stained with Coomassie Brilliant Blue. The
N-terminal Gurken cleavage fragment was quantified using ImageJ, normalized to the
DMSO vehicle control condition, and the values were plotted in GraphPad Prism
software.
39.
40.
41.
42.
Goel P.; Jumpertz T.; Mikles D.C., et al. Discovery and Biological Evaluation of
Potent and Selective N-Methylene Saccharin-Derived Inhibitors for Rhomboid
Intramembrane Proteases. Biochemistry 2017, 56, 6713-6725.
Ticha A.; Stanchev S.; Skerle J., et al. Sensitive Versatile Fluorogenic
Transmembrane Peptide Substrates for Rhomboid Intramembrane Proteases. J Biol
Chem 2017, 292, 2703-2713.
Ticha A.; Stanchev S.; Vinothkumar K.R., et al. General and Modular Strategy for
Designing Potent, Selective, and Pharmacologically Compliant Inhibitors of Rhomboid
Proteases. Cell chemical biology 2017, 24, 1523-1536.e1524.
Rhomboid inhibition assays were performed in black 96-well plates. Increasing
concentrations of the 2-styryl substituted benzoxazinones 3, 5, or 11 were pre-
incubated with 0.4 μM GlpG in reaction buffer [20 mM HEPES-NaOH (pH 7.4); 150
mM NaCl; and 0.05% (w/v) DDM] for 1 h. Subsequently, the reaction was started by
the addition of 10 μM of the fluorogenic substrates KSp35 (for compounds 5 and 11)
or KSp76 (for compound 3). Both substrates have comparable kinetic properties and
are interchangeable for IC50 determinations. IC50 values were determined as
described previously 40.
43.
44.
DelMar E.G.; Largman C.; Brodrick J.W.; Geokas M.C. A sensitive new substrate for
chymotrypsin. Analytical biochemistry 1979, 99, 316-320.
The ability of the benzoxazinone derivatives to inhibit the classical serine protease -
chymotrypsin was determined in an in vitro activity assay. The compounds were pre-
incubated in concentrations up to 250 M with 4 μM of bovine -chymotrypsin
(PanReac AppliChem) in reaction buffer [50 mM HEPES-NaOH (pH 7.5); 5 mM
EDTA; 0.4 M NaCl; 10 % Glycerol (v/v); 10 % DMSO] for 30 min at 25°C.
Subsequently, 2.5 l of this mixture were added to 100 μl reaction buffer containing
0.12 mM of the substrate N-Succinyl-Ala-Ala-Pro-Phe-p-nitroanilide (Suc-AAPF-pNA,
Merck). Cleavage of the substrate by -chymotrypsin resulted in 4-nitroaniline, which
is yellow under alkaline conditions. A microtiter plate reader (Beckman Coulter) was
used to record the increase in absorbance at 410 nm over 5 min at 25°C. IC50 values
were determined by fitting a nonlinear regression curve to a plot of log (compound
concentration) versus % -chymotrypsin activity using GraphPad Prism software.
Enzyme in vitro activity assays were used to determine the inhibitory activity of the 2-
styryl substituted benzoxazinones against the soluble serine proteases trypsin,
neutrophil elastase and cathepsin G. In the trypsin assay, increasing concentrations
of the benzoxazinones or DMSO vehicle were pre-incubated with 10 nM trypsin from
bovine pancreas (Sigma-Aldrich) for 30 min at 30°C in a total volume of 90 µl assay
buffer (PBS). The reaction was started by the addition of 10 µl substrate solution (1
mM Z-Gly-Gly-Arg-aminomethylcoumarin in PBS, Bachem), and the release of the
fluorescent cleavage product aminomethylcoumarin was monitored at 340 nm
excitation/ 440 nm emission with a microplate reader (Tecan Safire). In the cathepsin
G assay, increasing concentrations of the compounds or DMSO vehicle were pre-
incubated with 200 nM human neutrophil cathepsin G (Enzo Life Sciences) for 30 min
at 37°C in a total volume of 90 µl assay buffer [100 mM HEPES; 500 mM NaCl (pH
7.4)]. The reaction was started by the addition of 10 µl substrate solution (2 mM N-
Succinyl-Ala-Ala-Pro-Phe-p-nitroanilide in assay buffer, Bachem). Enzymatic
45.
15

cleavage of the substrate resulted in the release of the yellow coloured p-nitroaniline,
and absorbance was recorded at 410 nm. To measure the activity of the
benzoxazinones against human neutrophil elastase, the Neutrophil Elastase
Colorimetric Drug Discovery Kit (Enzo Life Sciences) was used according to the
manufacturer`s protocol. In brief, increasing concentrations of the benzoxazinones or
DMSO vehicle were pre-incubated with 0.22 mU neutrophil elastase for 20 min at
37°C in a total volume of 95 µl assay buffer [100 mM HEPES (pH 7.25); 500 mM
NaCl; 0.05 % Tween-20]. The reaction was started by the addition of 5 µl substrate
solution (2 mM N-Methoxysuccinyl-Ala-Ala-Pro-Val-p-nitroanilid in assay buffer), and
the formation of p-nitroaniline was recorded as described for the cathepsin G assay.
In all three assays, changes in absorbance/fluorescence were measured at 1 min
time intervals for 5 min, and the initial reaction slopes were used to determine the
reaction velocity (V). To calculate the % remaining protease activity, the formula (V
inhibitor/ V DMSO control) x 100 was used.
46.
47.
48.
49.
Yang J.; Barniol-Xicota M.; Nguyen M.T.N.; Ticha A.; Strisovsky K.; Verhelst S.H.L.
Benzoxazin-4-ones as novel, easily accessible inhibitors for rhomboid proteases.
Bioorg Med Chem Lett 2017, doi: https://doi.org/10.1016/j.bmcl.2017.1012.1056.
Vinothkumar K.R.; Strisovsky K.; Andreeva A.; Christova Y.; Verhelst S.; Freeman M.
The structural basis for catalysis and substrate specificity of a rhomboid protease.
Embo J 2010, 29, 3797-3809.
Harper J.W.; Hemmi K.; Powers J.C. Reaction of serine proteases with substituted
isocoumarins: discovery of 3,4-dichloroisocoumarin, a new general mechanism based
serine protease inhibitor. Biochemistry 1985, 24, 1831-1841.
Rhomboid activity recovery assays were performed in black 96-well plates.
Compound 3 (100 M), -lactam L29 (1 M) or isocoumarin JLK-6 (10 M) were pre-
incubated for 1 h with 0.4 M GlpG in reaction buffer [20 mM HEPES-NaOH (pH 7.4);
150 mM NaCl; 0.05% (w/v) DDM]. Afterwards, this solution was rapidly diluted 100-
fold into reaction buffer containing 10 M of the fluorogenic substrate KSp76. This
quenched fluorescent peptide is cleaved by the rhomboid leading to activation of a
red fluorophore. Substrate cleavage and enzyme activity recovery were monitored
continuously by measuring fluorescence over a time period of 120 min with excitation
at 553 nm and emission at 583 nm using a microplate reader (Tecan Infinite M1000).
16

17