Synthesis and Structural/Functional Characterization of Selective M14 Metallocarboxypeptidase Inhibitors Based on Phosphinic Pseudopeptide Scaffold: Implications on the Design of Specific Optical Probes

Article abstract of DOI:10.1021/acs.jmedchem.8b01465

Metallocarboxypeptidases (MCPs) of the M14 family are Zn²⁺-dependent exoproteases present in almost every tissue or fluid in mammals. These enzymes perform a large variety of physiological functions and are involved in several pathologies, such as pancreatic diseases, inflammation, fibrinolysis, and cancer. Here, we describe the synthesis and functional/structural characterization of a series of reversible tight-binding phosphinic pseudopeptide inhibitors that show high specificity and potency toward these proteases. Characterization of their inhibitory potential against a large variety of MCPs, combined with high-resolution crystal structures of three selected candidates in complex with human carboxypeptidase A (CPA)1, allowed to decipher the structural determinants governing selectivity for type-A of the M14A MCP family. Further, the phosphinic pseudopeptide framework was exploited to generate an optical probe selectively targeting human CPAs. The phosphinic pseudopeptides presented here constitute the first example of chemical probes useful to selectively report on type-A MCPs activity in complex media.

Full text of DOI:10.1021/acs.jmedchem.8b01465

Subscriber access provided by MIDWESTERN UNIVERSITY
Article
Synthesis and structural/functional characterization of selective M14
metallo-carboxypeptidase inhibitors based on phosphinic pseudo
peptide scaffold: Implications on the design of specific optical probes
Giovanni Covaleda, Pablo Gallego, Josep Vendrell, Dimitris Georgiadis, Julia
Lorenzo, Vincent Dive, Francesc X. Aviles, David Reverter, and Laurent Devel
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01465 • Publication Date (Web): 28 Jan 2019
Downloaded from http://pubs.acs.org on January 29, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis and structural/functional characterization of selective M14 metallo-
carboxypeptidase inhibitors based on phosphinic pseudo peptide scaffold:
Implications on the design of specific optical probes
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
†
§
Giovanni Covaleda , Pablo Gallego , Josep Vendrell , Dimitris Georgiadis , Julia
†
‡
†
†
‡
Lorenzo Vincent Dive , Francesc Xavier Aviles* , David Reverter* and Laurent Devel*
,
†Institut de Biotecnologia i de Biomedicina and Departament de Bioquímica i de
Biologia Molecular, UniversitatAutònoma de Barcelona, 08193 Bellaterra (Barcelona),
Spain.
‡CEA, Institut des Sciences du Vivant Frédéric Joliot, Service d’Ingénierie Moléculaire
des Protéines (SIMOPRO), Université Paris-Saclay, Gif-sur-Yvette 91190, France.
§Department of Chemistry, Laboratory of Organic Chemistry, University of Athens,
Panepistimiopolis Zografou, 15771, Athens, Greece.
Corresponding authors
Laurent Devel. Phone: +33 169089565; Fax: +33 169089071; E-mail: laurent.devel@cea.fr.
David Reverter. Phone: +34 935868955; Fax: +34 935812011; E-mail: david.reverter@uab.cat.
Francesc Xavier Aviles. Phone +34 606873290; Fax, +34 935812011; E-mail: FrancescXavier.Aviles@uab.es
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
2+
Metallo-carboxypeptidases of the M14 family (MCPs) are Zn dependent exoproteases
present in almost every tissue or fluid in mammals. These enzymes perform a large
variety of physiological functions and are involved in several pathologies such as
pancreatic diseases, inflammation, fibrinolysis and cancer. Here we describe the
synthesis and functional/structural characterization of a series of reversible tight-
binding phosphinic pseudo peptide inhibitors that show high specificity and potency
towards these proteases. Characterization of their inhibitory potential against a large
variety of MCPs, combined with high-resolution crystal structures of three selected
candidates in complex with human CPA1, allowed to decipher the structural
determinants governing selectivity for type-A of the M14A MCP family. Further, the
phosphinic pseudo peptide framework was exploited to generate an optical probe
selectively targeting human CPAs. The phosphinic pseudo peptides presented here
constitute the first example of chemical probes useful to selectively report on type-A
MCPs activity in complex media.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
2

Page 3 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Metallo-carboxypeptidases are exoproteases involved in a great variety of processes,
from digestion to blood coagulation/fibrinolysis, inflammation, pro-hormone and
neuropeptide processing among others, and are also reportedly involved in the progress
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
,2
of certain cancers and other pathologies . M14 metallo-carboxypeptidases (MCPs), one
of the largest families of the metallopeptidase clan 3 including many variants from
archaea, bacteria and eukaryotes, have been divided into three main subfamilies based
on structural similarity and sequence homology. The first one, which includes the
digestive enzymes CPA1, CPA2 and CPB1, mast cell CPA3, as well as CPA4, CPA5,
3
,4
CPA6 and CPO from distinct localizations, has been termed subfamily M14A or A/B .
The second subfamily, including the bioactive peptide-processing or regulatory
5
enzymes CPN, CPE, CPM, CPD among others, has been termed M14B or N/E . In
addition, a novel subfamily termed M14D, Nna-like or CCP comprising enzymes of
larger size with predominant cytosolic location, has also been described 6,7. An
additional classification into three main classes can be established according to substrate
specificity against C-terminal aromatic/hydrophobic residues (A-like), basic residues
(B-like) and acidic residues (O-like) 4,8. With some relevant exceptions due to structural
similarities, this latter classification based on specificities grossly correlates with M14A,
M14B and M14D subfamilies. Overall, differences in classification are also related to
differences in cellular or tissular localization, a fact that is relevant for biotechnological
or biomedical applications.
A consensus sketch of the mutual recognition regions in peptide/protein substrates
and MCPs considers that the essential structures are constituted by five subsites: P , P ,
4
3
P , P and P ’ and the corresponding S , S , S , S and S ’, in substrates and enzymes,
2
1
1
4
3
2
1
1
respectively. Specificity of recognition is mainly governed by the substrate residue
located C-terminally to the scissile peptide bond (P ’ position, Figure 1A), facing the S ’
1
1
specificity pocket of the enzyme. In addition, the substrate unprimed positions (P to P )
1
3
also contribute to the tuning of the selectivity profile towards different members of the
MCP family, but to a lesser extent 8–10
.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Protease inhibitors, both of natural or synthetic origin, are major elements in the
control of peptidase activity 11,12. Besides the wide distribution of the natural ones,
protease inhibitors display a significant variety of structures, sizes and functionalities,
particularly for the serine- cysteine- and metallo-endoproteases 13–16. The relatively
recent attribution of metallo-exoproteases to diseases, as well as the more difficult
chemistry involved, somewhat delayed the finding of inhibitors for this class of
enzymes 1,2. An exception to this has been the development of synthetic small inhibitors
for the angiotensin-converting enzyme (ACE), a dipeptidyl metallo-carboxypeptidase
belonging to the M2 family, in what was an early and successful attempt to generate
drugs to treat hypertension and some types of renal and cardiac dysfunction17,18. More
recent reports on new inhibitors of proteases, including MCPs, describe both synthetic
or natural proteinaceous molecules frequently associated to biomedical interest, of
which hirudin, serpins and cystatins are some of the most relevant examples for the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
latter type 19–21
.
Most of the synthetic metallo-protease inhibitors that have been designed to date are
competitive inhibitors that incorporate a zinc-chelating moiety targeting the zinc ion
within the enzyme active site. Different families of peptides or pseudo peptides
displaying a thiol, carboxylate, hydroxamate or phosphonate function as zinc-binding
group have been described as potent metallo-protease inhibitors 22–24. Fairly similar to
the latter, the phosphinic pseudo peptides with a zinc-chelating phosphinate moiety (-
PO -CH -) represent a unique and valuable class of compounds that share structural
2
2
features with the transition-state intermediates generated during the hydrolysis of the
peptide bond 25 resulting in competitive inhibitors with high affinity for their target.
More importantly, due to their capacity to probe the whole active site across primed and
2
6
unprimed enzyme subsites
the phosphinic pseudo peptides offer further
opportunities for selective interactions compared to inhibitors with a hydroxamate or
carboxylate zinc-binding moiety that probe a limited number of subsites. In addition,
2
7
the phosphinate function itself seems to play a critical role in the inhibitor selectivity .
ACS Paragon Plus Environment
4

Page 5 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Due to their stability and their absence of toxicity, the phosphinic pseudopeptides are
also ideal chemical tools to selectively target metalloproteases activity in vivo 28–32
.
In the frame of this study, we have explored an original series of phosphinic
pseudopeptides enabling to better understand the molecular determinants governing
selectivity between different members of the M14A family, particularly between A-type
and B-type family members. To this purpose, we have designed a small library of 11
tripeptide phosphinic inhibitors differing in residues facing S ’, S and S subsites within
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
2
MCPs catalytic cleft (Figure 1A). The affinity and selectivity profile of these pseudo
peptides have been characterized against relevant recombinant human MCPs (i.e.,
CPA1, 2, 3, 4, B1, TAFI and D). Three pseudo peptides (1, 3 and 8) have been crystallized
in complexes with human CPA1 (hCPA1). In addition, pseudo peptide 8 displaying a
high potency toward hCPA1, 2 and 3 and sparing hCPB1 and D has been declined into
three additional analogues with a N-terminal extension consisting in a PEG linker with a
free amino function, a N-acyl moiety or an amino group conjugated to a fluorescent
reporter (Figure 1B, compounds 9, 10 and 11 respectively). Such fluorescent probe,
keeps a restricted selectivity for A-type M14A MCPs, displaying a null or very weak
affinity for the B- or O- type forms as well as for a variety of proteases from different
clans. Overall, the impact of these structural modifications on the binding properties of
the inhibitors towards human MCPs has been systematically examined in vitro and
structurally characterized.
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Chemical structure of phosphinic pseudo peptides 1 to 11. A) Structure of
pseudo peptide 1 to 7 differing in their P ’ to P residues facing subsites S ’ to S
1
2
1
2
subsites. B) Chemical structure phosphinic pseudo peptides 8 to 11 with a simple acyl
function (8), with a PEG extension (9 and 10) and a PEG extension conjugated to a
fluorescent reporter (11).
ACS Paragon Plus Environment
6

Page 7 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RESULTS AND DISCUSSION
Design and synthesis of phosphinic pseudo peptides 1 to 11: To generate an
exploratory set of phosphinic pseudopeptides we selected basic units with three
residues, covering the S /S /S ’ subsites of MCPs (Figure 1A). It is worth mentioning
2
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
that the nature of the different P , P and P ’ side chains in the phosphinic pseudo
2
1
1
peptides were chosen according to previous knowledge from complexes between MCPs
and natural protein inhibitors, such as the LCI-hCPA2 (PDB code 1DTD), ACI-hCPA1
(PDB code 3FJU), SmCI-hCPA4 (PDB code 4BD9), NvCI-hCPA4 (PDB code 4A94), TCI-
bCPA1 (PDB code 1ZLH) and PCI-bCPA (PDB code 4CPA). These natural compounds,
respectively isolated from leeches, intestinal worms, marine worms, marine snails, tick
1
,33–38
and potato
harbor substrate-like properties with K values in the low
i
nanomolar/sub nanomolar range.
In their P ’ position, the pseudo peptides harbored either a phenylalanine (1) or a
1
homophenylalanine residue (2 to 11), coincident with the preference for hydrophobic
side chains at A-like MCPs S ’ subsite, as derived from studies with synthetic
1
substrates/inhibitors or from proteome derived-peptide libraries 10,39. Among the 11
compounds selected for synthesis, eight were simple pseudo tripeptides with chemical
variations both in their P and P position. Since the P position has only moderate
2
1
2
1
impact on substrate recognition, either a phenylalanine (1 and 2) or an alanine (3 to 11)
was indifferently incorporated. Regarding the P position, the impact of aliphatic (3),
2
acidic (4), basic (5), constrained (7), and polar (6 and 8) side chains on inhibitors binding
was explored. Three additional pseudo tripeptides with either an N-terminal PEG
spacer (9 and 10) or the same extension plus a 6SIDCC fluorophore (11) were also
synthesized. A short PEG spacer was incorporated between the targeting motif and the
bulky fluorescent group to limit steric clashes during probes/MCPs interaction. In the
perspective of cells assay, 6SIDCC fluorophore as a highly negatively charged
fluorescent tag was selected as it displays reduced unspecific binding by comparison
with standard cyanine derivatives 40,41
.
The phosphinic pseudo peptide 1 was obtained as previously described 42. The
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
phosphinic pseudo peptides 2 and 3-11 were synthesized on solid support from Fmoc-
4
3
Phe-POOAd-hPhe-OH
and Fmoc-Ala-POOAd-hPhe-OH 17 building blocks
respectively (Scheme 2). Building block 17 was first synthesized in solution according to
Scheme 1.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1: Chemical pathway to access phosphinic building block 17.
Briefly, diethylmalonate was first converted into the acrylate 12 in three steps (65%
yield), after alkylation in presence of 2-phenylethyl bromide, mono saponification, and
Knoevenagel condensation with formaldehyde. This acrylate was then esterified with
benzyl alcohol to yield protected acrylate 13 (70% yield). This electrophile then
participated in a P-Michael addition reaction with phosphinic acid 15 under mild,
silylating conditions 43 to give compound 16 with a yield of 59%. Compound 15 was
generated in two steps from derivative 14 prepared as previously reported 44,45. Finally,
1
7 was obtained in two steps after protection of the hydroxyphosphinyl moiety with an
adamantyl protecting group followed by a saponification step (85% yield). The
phosphinic building block 17 as a mixture of diastereomers, was thus synthesized in
seven steps from the commercially available diethylmalonate and precursor 15 in an
overall yield of 23%.
ACS Paragon Plus Environment
8

Page 9 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2: Solid phase synthesis of phosphinic pseudo peptides 3 to 11.
4
3
The two Fmoc-phosphinic building blocks Fmoc-Phe-POOAd-hPhe-OH
and
Fmoc-Ala-POOAd-hPhe-OH 17 were first activated in presence of DIC followed by the
addition of DMAP. This solution was immediately added to the resin and the coupling
reaction was carried out under microwave irradiation at 60°C for 10 min. Elongation of
the pseudo peptide sequences was performed using standard Fmoc strategy under
microwave irradiation. For compounds 2-8 and 10, the acetylation reaction was
performed at room temperature in presence of 1-acetylimidazole. In the case of pseudo
peptides 9, 10 and 11, 8-(Fmoc-amino)-3,6-dioxaoctanoic acid was coupled to the
peptide sequence in conditions comparable to those used for amino acid incorporation.
Each resulting pseudo peptide was then cleaved from the solid support and analyzed by
RP-HPLC. Analysis of the crude showed two peaks corresponding to two diastereomers
that possess a S configuration at P , a R configuration at P and a R or S configuration at
2
1
P ’. Regardless of the P position within the pseudo peptide structure, the major peak
1
2
observed in RP-HPLC systematically corresponded to the most potent inhibitor towards
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CPs. This is a known trend in the chemistry of phosphinic peptides, initially reported by
B.P Roques and co-workers 45 and verified by others later on 27,46,47 with no exception
reported to our knowledge. Since in transition-state analogues maximum potency is
directly related to mimicry of natural peptides, we postulated that the most potent
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
isomer possessed a S configuration within P ’ position. Furthermore, given that the
1
major isomer of the starting block (16) leads to the major isomer of the pseudo peptides,
we deduced that the major diastereomer (0.8) of 16 possess the RS configuration and the
minor diastereomer (0.2) the RR configuration. To access fluorescent derivative 11,
48
compound 9 was coupled to 6S-IDCC-NHS ester . The phosphinic pseudo peptides 1
to 11, as well as the building blocks 6 and 8, were stored at −20 °C and protected from
light.
ACS Paragon Plus Environment
1
0

Page 11 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Inhibitory potency and selectivity of phosphinic pseudo peptides. The K values
i
for compounds 1 to 11 were determined on a set of five relevant human MCPs (Table 1).
These phosphinic compounds act as tight-binding inhibitors of MCPs, with Ki values in
the nanomolar to sub nanomolar range. In particular, they displayed the strongest
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
affinity towards hCPA1, with K values 10-to-25-fold lower than those for other A/B-
i
type MCPs (Table 1). The most significant difference appears when P ’ side chain length
1
is increased. Thus, elongation by a single methylene group leads to an inhibitor that
displays almost no affinity for B-type enzymes. Regarding the P position, its
1
simplification (2 with a methylbenzene vs 3 with a methyl group) has no significant
impact on the inhibitor potency and selectivity towards the different A-type MCPs
(
Table 1).
Table 1. Potency of Phosphinic Peptide Inhibitors (K ) toward several MCPs.
i
K
i
(nM)
P
3
P
2
P
1
1
P ’
hCPA hCPA hCPA hCPB hCPD
1
2
4
1
0
.04
±
1.0
±
0.2
0.39
±
0.06
0.12
±
0.01
1
2
NI
NI
CH
CH
3
3
0.01
0.08
±
0.02
0.2
±
0.1
1.9
±
0.2
NI
NI
NI
0
.087
±
0.7
±
0.1
3.3
±
0.3
3
4
CH
CH
3
3
NI
NI
CH
CH
3
3
0
.002
3
.1
.2
4.6
±
0.7
38.6
±
3.2
±
0
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1.2
±
0.2
2.4
±
0.3
25.4
±
1.9
5
6
CH
CH
CH
CH
CH
CH
CH
3
3
3
3
3
3
3
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
CH
CH
CH
CH
3
3
3
3
5
±
0
.9
.4
13.8
±
1.4
60.6
±
7.2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0
.7
±
.1
3.2
±
0.6
7.5
±
0.5
7
0.107
±
0.002
1.7
±
0.3
2.6
±
0.2
8
0
±
0
.3
.1
7.8
±
0.5
5.4
±
0.6
9
0
0
.1287
±
.0004
5.7
±
0.4
3.5
±
0.2
10
0.089
±
0.005
5.3
±
0.7
5.7
±
0.9
1
1
Data are means (n=3) ± SD. K in nM. NI: No inhibitory activity detected for the [I] range from 1
i
0
nM to 1 mM.
Interestingly, when an acidic (4), a basic (5) or a polar side chain (6) is present in the
P2 position a drop of affinity ranging from 10 to 20 is observed toward the three hCPA 1,
2
and 3. Conversely, P aliphatic or aromatic side chains (3, 7 and 8) restore the binding
2
capacity of the inhibitors towards those enzymes. Additionally, none of these
compounds are able to inhibit human CPD, a representative member of the M14B
subfamily. The synthesized series is therefore addressed to the M14A subfamily of
MCPs.
Based on these results, compound 8 displaying a high potency and selectivity
ACS Paragon Plus Environment
1
2

Page 13 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
towards hCPA 1, 2 and 3, was selected as a relevant starting scaffold for the
development of an optical probe 11. Inhibitory activity of 11 as well as that of its
precursors 9 and 10 was assessed on MCPs. Incorporation of the linker (with two
different ends) and fluorophore does not substantially affect the enzymatic inhibitory
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
potency and selectivity of the constructs, as indicated by K values in the nanomolar
i
range for these three molecules (Table 1).
All the above kinetic data was collected in steady state conditions after 2h of pre
incubation with inhibitors 1-11 since in certain cases inhibitors might act as slow
49
binders, besides tight binders . In addition, pre-steady state data was also collected to
derive k and k , and dependent Ki, from the action of compounds 3, 8, 10 and 11 on
on
off
human CPA1. The results are presented in Table 2 (see also Figure S1, S2 and S3 for
details). The derived constants indicate that the four constructs can be considered as a
slow tight binding inhibitors, with values equivalent to those of well-known cases of
5
0
proteases , and with derived pre-steady state Ki values very similar to the ones found
in steady state conditions. These experiments also confirm that the addition of the PEG
bridge and fluorescent head resulting in optical probe 11 does not significantly modify
such constants
Table 2. Kinetic parameters of phosphinic inhibitors on human carboxypeptidase
A1.
Pre-steady-state
Steady-state
koff
kon
Kiapp=k /k
off on
Ki
Ki
Compound
x10-5 s-1)
5
-1 -1
-9
-9
-9
(
(x10 M .s )
(x10 M)
(x10 M)
(x10 M)
3
8
6.0 ± 0.2
7.6 ± 0.3
7.0 ± 0.2
1.27 ± 0.02
0.47 ± 0.02
0.14 ± 0.01
0.16 ± 0.01
0.16 ± 0.01
0.087 ± 0.002
0.107 ± 0.002
0.1287 ± 0.0004
1.44 ± 0.04
1.33 ± 0.04
0.53 ± 0.04
0.53 ± 0.02
1
0
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
11
4.8 ± 0.5
2.03 ± 0.04
0.24 ± 0.03
0.07 ± 0.01
0.089 ± 0.005
[hCPA1] =0.5 nM. 100 mM of AAFP substrate. T=37ºC. Data are means (n=3) ± SD.
Interestingly, probe 11 not only strongly inhibits other A-type M14A-MCPs like
secretory mast cells CPA3, but did not display any inhibitory capacity against other B-
type M14B-MCPs (like blood hCPU/TAFI and hCPN), cytosolic M14D-MCPs (like
hCCP1 and hCCP6), or other peptidases of different catalytic type such as aspartic,
serine or cysteine peptidases (Table 3). Moreover, it displayed only modest inhibitory
activity in the micromolar range against other metallo-peptidases of distinct families,
such as matrix metallo-proteases 9, 10, 11 and 13 (Table 3).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 3. Potency of Phosphinic-based probe (compound 11) (K ) toward several
proteases.
i
Enzyme
K (nM)
i
rat CPA3
0.017 ± 0.008
hCPU (TAFI) NI
hCPN
NI
hCCP1
NI
hCCP6
NI
hMMP9
hMMP10
hMMP12
hMMP13
pPepsin
Papain
2588 ± 560
983 ± 198
1340 ± 195
562 ± 169
NI
NI
bTrypsin
hThrombin
hNElastase
NI
NI
NI
Data are means (n=3) ± SD. K in nM. NI: No inhibitory activity detected for the [I] range from 1 nm
i
0
to 1 mM
Structural characterization of compounds 1, 3 and 8/hCPA1 complexes. In order to
ACS Paragon Plus Environment
1
4

Page 15 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
structurally-functionally analyze the above phosphinic compounds in a detailed and
integrated way, three of them (compounds 1, 3 and 8) were selected and characterized
by protein X-ray crystallography in complex with hCPA1. The crystal structure of 1, 3
and 8 in complex with this enzyme was solved at 2.29 Å, 2.27 Å and 1.72 Å resolution
respectively. The quality of the electron density maps allows the perfect trace of 1, 3 and
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
side chains within the enzyme subsites (Figure 2). This allowed us to confirm the three
stereogenic centers relative configuration of compounds 3 and 8. Indeed, the electron
density maps clearly display only one configuration in P ’ position and no hint of an
1
electron density for another configuration was observed.
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A)
P1
O
O
H
N
N
P
OH
H
HO
O
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
P2
1
P ’
1
B)
P1
O
O
H
N
N
P
OH
H
HO
O
O
P2
3
P ’
1
P2 Y
P2 Y
C)
Y248
P1
R145
Y248
O
O
H
N
R71
N
H
P
OH
R145
I243
P1’ hF
HO
O
R71
P1 A
O
M203
H69
I255
T268 E270
E72
P2
OH
H196
P1’ hF
8
E270
P ’
1
Figure 2. Structural details of the complexes between hCPA1 and the phosphinic compounds 1
(A), 3 (B) and 8 (C). Left column: Chemical structure of compounds 1, 3 and 8. Central column:
2
Fo-Fcomit map (contour sigma level is 1.2) of 1, 3 and 8 in complex with hCPA1 with the
respective following PDB IDs: 4UEZ, 4UEE and 6I6Z. Right column: close-up view of the
primary binding regions within the active site of human hCPA1. The phosphinic compounds are
represented in yellow sticks and hCPA1 residues are shown in cyan. The functionally essential
zinc is visualized as a blue sphere.
Inhibitor 1 displayed smaller B-factors around its P ’ side chain and the phosphinate
1
ACS Paragon Plus Environment
1
6

Page 17 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chelating moiety, indicating a low degree of flexibility around this binding region. The
zinc atom is located at close distance to one oxygen atom of the phosphoryl group (2.19
Å), and also coordinates with the two histidine (at 2.04 and 2.03 Å, respectively) and the
glutamic acid (at 2.05 Å) of the carboxypeptidase zinc-binding residues (Figure 2A).
Additionally, the zinc-coordinating oxygen of the phosphoryl group is also at 2.77 Å to
the active-site Glu270 of the enzyme, the general base residue in the water-mediated
nucleophilic catalysis of metallo-carboxypeptidases. The second oxygen of the
phosphoryl group is at 2.69 Å to the active-site Arg127, which is responsible for the
polarization of the carbonyl group of the scissile peptide bond during the catalysis and
shaping of the tetrahedral intermediate. Thus, the most important residues in the active
site and catalytic machinery of CPA1 are trapped in a net of close interactions with the
inhibitor. The geometry of this complex perfectly resembles the tetrahedral intermediate
formed during the transition state of the catalytic mechanism and, therefore, these
ligands can be considered as transition-state analogues. In the most accepted mechanism
of MCPs activity, the zinc-bound water molecule, polarized by Glu270, is added to the
substrate carbonyl group of the scissile bond, forming an unstable tetrahedral structure
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
that will lead to the cleavage of the peptide bond . In the case of the complexes of other
metallopeptidases with phosphinic peptide inhibitors, the phosphinate-methylene
group of the inhibitor mirrors the tetrahedral intermediate of the transition state 26,27 as
its tetrahedral configuration is shaped by the zinc-coordinating oxygen, next to the
active-site nucleophile (equivalent to Glu270 in MCPs) that mimics the polarized
catalytic water, and by the second oxygen mimicking the carbonyl oxygen polarized by
a positive group (equivalent to Arg127 in MCPs). The observed quasi-planar geometry
of the bonding between the side-chain atoms of Glu270 and Arg127 of hCPA1 with the
zinc atom and with the two oxygen atoms of the compound 1 phosphoryl group
confirms previous hypotheses about the catalytic mechanism of carboxypeptidases 1,8.
The same geometry is observed in the other complex structures between phosphinic
compounds 3 and 8 with hCPA1 (Figure 2B and 2C).
The P ’ subsite of the phosphinic peptide 1 is formed by both the side-chain of a
1
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
phenylalanine and the carboxylate group, which resembles the typical C-terminal
carboxylate binding with Arg145 and Asn144. Tyr248, normally forming a hydrogen
bond with the amide nitrogen of the scissile peptide bond, also displays a “down”
conformation but interacts with one of the two carboxylate oxygens of the P ’ subsite
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Figure 2A). Considering that A-type carboxypeptidases selectively hydrolyze bulky
and aliphatic C-terminal side-chains, and that even tryptophan residues can be
4
accommodated in the large hydrophobic pocket . The presence of a Phenyl residue in
the hydrophobic pocket shaped by Met203, Ile243, Ile247, Ile255, Thr268 and Asn144
simply confirms this preference, which is reinforced by the fact that compounds bearing
a bulkier homophenyl at the P ’ position also behave as potent CPA1 inhibitors (Table 1
1
and Figures 2B-2C). Phosphinic compound 1 displays nano- or sub nanomolar
inhibitory constants against several A-type carboxypeptidases of the M14A family and
is also unexpectedly able to inhibit B-type carboxypeptidases (see Table 1), despite the
preference of the latter for basic C-terminal residues. However, B-type
carboxypeptidases are not inhibited by compounds containing a homophenyl at the P1’
position (as happens with compounds 3 and 8; see Figures 2B-2C). As shown in the
comparison between the S ’ pockets of A-type and B-type carboxypeptidases in Figure 3,
1
hCPA1 is able to accommodate both Phenyl and homophenyl at the P ’ position.
1
However, a model of our phosphinic inhibitors docked into human CPB1 active site
displays steric hindrances within the cavity when homophenyl is located at P ’. This is
1
in accordance with the observed lack of inhibition against B-type carboxypeptidases.
This result reveals the phosphinic inhibitors containing a homophenyl Alanine residue
at the P ’ position as optimal MCPs inhibitors able to distinguish between A-type and B-
1
type classes of M14A metallo-carboxypeptidases.
ACS Paragon Plus Environment
1
8

Page 19 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 3. Comparison of the hydrophobic binding pockets of CPA1 and CPB in complex with
phosphinic compounds 1 and 8. A) close-up stereo view of the surface representation of the
cavity of the substrate binding-site of hCPA1 in complex with phosphinic compounds 1 and 8.
B) close-up stereo view of the surface representation of the cavity of the substrate binding-site of
hCPB modelled in complex with the same compounds. Compounds 1 and 8 are shown in stick
representation in green and yellow, respectively. Major active site residues of hCPA1 and hCPB
are labelled and shown in stick representation.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The electron density maps of the P and P sub-sites of the phosphinic peptides are
1
2
well defined in all structures, and in compound 1 is formed by the side chains of
phenylalanine and leucine, respectively. The peptide bond between the P and the P is
1
2
also forming hydrogen bond interactions, in particular between the P carbonyl oxygen
2
of the inhibitor with Arg71 (at 2.81 Å). The major differences between 1 and 3 are found
in the P subsite, with the substitution of a phenylalanine by alanine in compound 3.
1
However, kinetic analysis indicates that the side chain of the P position does not affect
1
the inhibitory properties of the compound, as shown by comparison of inhibitory
constants of compounds 2 and 3 (Table 1). As commented upon earlier, the P Leucine (1
2
and 3) or Tyrosine residue (8) seems to play a major role in the activity of the inhibitor,
as its substitution by either Glutamate, Lysine, Proline or Serine (compounds 4 to 7)
reduces the inhibitory constant for hCPA1 by two orders of magnitude (Table 1). This
may be explained by a favorable interaction between aromatic or aliphatic P side chains
2
and CPA-active-site Tyr248 at 3.5 Å distance (Figure 2A and C). In solution, such an
interaction may induce the movement of Tyr248 to the “down” orientation described
during catalysis. However, in crystal structures, this favorable interaction could
compete with additional hydrophobic interactions from its neighboring symmetric
molecule and might explain both a weaker electron density map and higher B-factors at
P2 site by comparison to other sites. No particular structural effect due to the
substitution of P Leucine by Tyrosine (compounds 3 and 8, respectively) was observed,
2
coincident with the similarity observed in the inhibitory constants of those compounds
against hCPA1.
ACS Paragon Plus Environment
2
0

Page 21 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Inhibition on a peptide degradative assay. The capability of the new inhibitors in
an orthogonal assay mimicking real conditions was evaluated with construct 8 on the
neurotensin peptide used as model. Samples taken at different intervals from 0-7h
degradative experiment by human CPA1 (20nM) in the absence and presence of
inhibitor 8 at two different concentrations (5 and 500nM), were analyzed by MALDI-
TOF MS. As shown in figure S4, the phosphinic pseudo peptide 8 counteracts the
hCPA1 action over the substrate.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cytotoxicity and cellular uptake. To start evaluating the potential utility of the
herein synthesized phosphinic inhibitors and derived fluorescent probe under biological
conditions, cytotoxicity/cellular viability and initial uptake studies were performed on
both HeLa and HEK293T human cells. Importantly, Phosphinic inhibitor 8 at 1 and 5 µM
applied to cell cultures during 24h showed no sign of cell cytotoxicity using the XTT
assay (Figure S5). A similar behavior was observed by microscopy when the treatment
was performed with fluorescent probe 11 (data not shown). After determining that
inhibitors have no effect on the cell viability, we investigated the probe cellular uptake
in the first cell line type. When using HeLa cells with abundant human CPA3 expression
in cytoplasm (see experimental section), and after a 6h treatment with probe 11 (1 or 5
µM) followed by cell washings, cell analysis by fluorescence confocal microscopy
showed that only a small part of the fluorescent probe internalized into cells (Figure S6),
the rest being washed away. This poor internalization is consistent with the negative
global charge of probe 11. In this respect, a similar behavior has been observed for
another optical probe whose structure was derived from a phosphinic peptide 40.
Interestingly, internalized probe 11 was found to co-localize with human CPA3 as
shown in Figure S6, proving its capacity to detect CPA-type enzymes in biological
assays
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
New chemical tools for studying human CPAs. The pioneer peptide-based phosphonic or
phosphinic metallo-carboxypeptidase inhibitors 23,51,52 were shown to behave as transition state
analogues and to display inhibition constants in the pico-femtomolar range for A-like enzymes.
Detailed crystal structures and enzymatic parameters were obtained in complex with bovine CPA,
but the specificity of those molecules on proteolytic enzymes from the same or related families or
clans was not explored. Later reports related to the present work 26,53 describe peptide phosphinic
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
derivatives addressed to the M2 metallo-carboxypeptidase angiotensin-converting-enzyme (ACE)
42,
to zinc peptidases 54,55 and to TAFI, a circulatory plasma carboxypeptidase B . Here we
56
report an original series of phosphinic pseudo peptides affording for both potency of inhibition
and high selectivity towards human carboxypeptidases. Specifically, we have demonstrated that a
P1’ homo phenylalanine within the phosphinic motif allowed to restrict selectivity to A-type
enzymes.
Importantly, we also showed that such a scaffold could be exploited for the design of optical
probes targeting these enzymes. Therein, the close similarity of Ki values between parent
compound 8, compounds 9-10 and optical probe 11 indicates that additional N-ter extensions,
even combined with a bulky fluorophore, do not compromise the binding strength. Noteworthy,
crystals of probe 11 in complex with human CPA1 were obtained as indicated by the fluorescent
green color of these latter. Despite the flexibility of the PEG fluorophore extension resulting in a
choppy-defined electron density component, the positioning of the phosphinic moiety within the
CPA1 catalytic cleft has been unambiguously confirmed and was similar to that of compound 8.
Overall, compound 11 constitutes to our knowledge the first example of specific optical probe for
human CPA-like enzymes within the M14A subfamily, with no detectable activity against human
M14B and M14D carboxypeptidase subfamily variants and only a very weak one against
metallo-endopeptidases of the MMPs family.
Many of the reports commented upon before had an indirect or direct association with
biomedical goals, like the recent ones on phosphinic inhibitors of the Helicobacter pylori Zn
5
7
carboxypeptidases Csd4 . While only a limited number of CPA-like forms are located within
the acinar cells of the pancreas, in the mast cells granules or in a few cases in the cytoplasm 1,5,14
,
the majority of them act extracellularly and, in some cases, have been related with diseases, like
5
the Duane syndrome and epilepsy . With their negatively charged phosphinate function,
ACS Paragon Plus Environment
2
2

Page 23 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
phosphinic pseudo peptides can be considered as ideal tools to preferentially bind to extracellular
4
0
targets . This property combined with the excellent selectivity displayed by compounds 2, 3, 8
towards the M14A family encourages further work focused on chemical modifications and
structural polishing in order to subsequently validate and adapt these new molecules to in vivo
action for biomedical applications. To this extent, the design of a first fluorescent probe 11 for A-
type MCPs and its validation in a human cell model in culture constitute an important step and
opens possibilities to study these proteases in vivo through fluorescent imaging.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CONCLUSIONS
The combination of the proper size of a pseudo-tripeptide with an uncleavable phosphinic unit
harboring a phenylalanine side chain at the C-terminus/P ’ subsite, results in very tight binding
1
inhibitors for human CPA1-type forms. The developed compounds also show an excellent
inhibitory capability against CPB-like enzymes, which abruptly fades when their P ’ side chain is
1
elongated. Importantly, such a P ’ modification has almost no impact on the inhibitory constants
1
against a number of M14 MCPAs variants, and only a moderate impact when combined with
variations at P and P positions. Additionally, a A-type M14 MCPs selective fluorescent probe
1
2
derived from a phosphinic pseudo peptide with a P ’ homo phenylalanine, a P alanine and a P
2
1
1
tyrosine has been developed and its capacity to co-localize with a CPA enzyme in human cell
cultures has been demonstrated. These results both validate our structure-based design and
suggest that this optical probe, or other ligand-derived probes with a similar phosphinic scaffold,
may be developed for targeting CPA in vivo activity through different imaging modalities.
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
EXPERIMENTAL SECTION
1
1
. Synthetic Chemistry.
.1. Chemical reagent and procedures. All Commercially available reagents and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
solvents were used as received without further purification. Diisopropylcarbodiimide
(DIC) and Trifluoroacetic acid (TFA), 8-(Fmoc-amino)-3,6-dioxaoctanoic acid were
®
purchased from Aldrich (Saint Louis, MO, USA). Wang resin (100-200 Mesh, 0,9
mmol/g) and Fmoc-AA-OH were purchased from Merck (Darmstadt, Germany). 6-
Chloro-1-Hydroxybenzotriazole (ClHOBt) was purchased from Molekula (Newcastle,
United Kingdom). 6S-IDCC-NHS ester was purchased from IC Discovery (Berlin,
Germany). Pseudo peptide syntheses were performed manually in polypropylene
syringe (Supelco) equipped with a polyethylene frit and a stopper. Microwave
experiments were performed with a Discover apparatus (CEM µWave, Matthews, NC,
USA) using the open vessel mode with a SPS kit. Reactions were monitored by thin-
layer chromatography (Merck TLC aluminum sheets coated with silica gel 60F ).
2
54
1
Compounds were purified by Flash chromatography (Silica gel Si 60, 40-43mm). H (200
1
3
31
MHz), C (50 MHz) and P (81 MHz) NMR spectra were recorded on either a Varian
00 MHz Mercury (compounds 12 to 17) or a Bruker 500Mz Avance (compounds 3, 8, 10
2
1
13
and 11). H and C spectra are referenced according to the residual peak of the solvent
5
8 31
based on literature data . P NMR chemical shifts are reported in ppm downfield from
13
31
8
5% H PO (external standard). C and P NMR spectra are fully proton decoupled.
3
4
1
Suggested H NMR peak assignments of reported compounds are based on 2D COSY
experiments. Data are reported as follows: chemical shift, multiplicity (s= singlet, d=
doublet, t= triplet, q= quartet, br= broad, m= multiplet), coupling constants (Hz) and
integration. Optical rotation data were acquired on a Perkin-Elmer 343 polarimeter at
2
1
5oC. DO measurements were performed on a Shimadzu UV spectrophotometer (UV-
800). Analytical RP-HPLC separations were performed on a LC-20AB Prominence
apparatus (Shimadzu Corp., Japan) using a solvent system consisting of (A) 0.1% v/v
aqueous TFA, and (B) 0.1% v/v TFA in CH CN on three different columns :
a
3
ACS Paragon Plus Environment
2
4

Page 25 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
-1
Hypersil C column (4.6 x 250 mm, 5 µm), flow rate=1 ml.min , linear gradient: 0-30
1
8
®
min / 0-100% B : conditions 1, a Kromasil C column (4.6 x 150 mm, 5 µm), flow rate=1
8
-1
mL.min , linear gradient: 0-30 min / 0-100% B : conditions 2, or a XBridge BEH300 C
1
8
-1
column (2.1 x 150 mm, 5 µm), flow rate=1 mL.min , linear gradient: 0-30 min / 0-100%
B: conditions 3.Preparative RP-HPLC separations were performed using the solvent
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
®
-1
system on Kromasil C (10 x 250 mm, 10 µm, flow rate=4 mL.min ), linear gradient: 0-
1
8
®
3
0 min / 0-100% B: conditions 4, or a Kromasil C column (10 x 250 mm, 10 µm), flow
8
-1
rate=4 mL.min , linear gradient: 0-30 min / 0-100% B: conditions 5. Retention times (Rt)
were reported in minutes. Mass spectrometry data were registered using a 4800 MALDI-
TOF mass spectrometer (Applied Biosystems, Foster City, CA, USA) or an ion trap
Esquire HCT spectrometer (Bruker Daltonics, Billerica, MA, USA). Amino acid
compositions were performed on an aminoTac JLC-500/V amino acids analyzer (JEOL,
Tokyo, Japan). The identity and purity (> 95% pure) of each newly synthesized
compound is assessed by analytical RP-HPLC, NMR, and mass spectrometry.
1
. 2. Compound synthesis and characterization.
Solid phase synthesis of pseudo peptides (2-10). Wang resin® (1 equivalent) was first
swelled in DCM for 15 min at room temperature. The appropriate Fmoc-phosphinic
building block (3 equivalents) was dissolved in DMF/DCM: 1/1 and activated with DIC
(3 equivalents, RT, 5 min) followed by the addition of DMAP (0.5 equivalent). The
solution was immediately added to the resin and the coupling reaction was carried out
under microwave irradiation (60°C, 25 W, 10 min). Fmoc removal was performed with
piperidine 20% in DMF under microwave irradiation (3 x 2min, 60°C, 25 W) and the
resin was then washed (2 x DMF, 2 x DCM). Fmoc-AA(OtBu)-OH (0.27 g, 0.64 mmol, 5
eq) was pre-activated during 5 min in presence of DIC (5 equivalents) and Cl-HOBt (5
equivalents) in DMF and added to resin. The coupling reaction was performed under
microwave irradiation (10 min, 60°C, 25 W). To access compounds 2 to 8, after Fmoc
removal and washing steps, their N-terminal extremity was acetylated with 1-
acetylimidizole (100 equivalents) in DMF (45 min, RT). To access compounds 9 and 10
incorporating a PEG linker, an additional elongation step on the N-terminal extremity
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was implemented. In this respect, 8-(Fmoc-amino)-3,6-dioxaoctanoic acid (5 equivalents)
was first activated for 5 minutes in the presence of DIC (10 equivalents) and HOBt (10
equivalents) and then incorporated onto the solid support. The coupling reaction was
run under microwave irradiation (60°C, 25 W, 20 min) followed by a Fmoc removal
step. In the case of compound 10, this step was completed by an acetylation step as
described above. For each individual pseudo peptide, the resin was then washed as
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
described and suspended in TFA/ H O: 95/5 for 45 min at room temperature under
2
gentle stirring. The resin was removed by filtration and washed with TFA/DCM: 1/1
(30 min, RT). This cycle was repeated twice. Combined filtrates were evaporated under
light flow of argon and the crude material purified by RP-HPLC.
(2S)-2-({[(1R)-1-(N-acetyl-L-leucylamido)-2-
phenylethyl](hydroxy)phosphoryl}methyl)-4-phenylbutanoic acid (2). Obtained as a
foamy white solid (14 %) after semi preparative RP-HPLC using conditions 4. R =16.25
t
-
min in conditions 1, R =20.0 min in conditions 2. MS m/z for [C H N O P] , calc. 515.6,
t
27 36
2
6
found 515.3.
2S)-2-({[(1R)-1-(N-acetyl-L-leucylamido)ethyl](hydroxy)phosphoryl}methyl)-4-
(
phenylbutanoic acid (3). Obtained as a foamy white solid (42 %) after RP-HPLC
purification using conditions 4, R =14.2 min in conditions 1, R =13.8 min in conditions 2,
t
t
1
R =19.7 min in conditions 3. H NMR (500 MHz, D O) δ 8.09 (t, J= 6.0 Hz, 1H), 7.62 (d, J=
t
2
1
3
0.4 Hz, 1H), 7.28 – 6.87 (m, 5H), 4.22 – 4.13 (m, 1H), 3.98 – 3.77 (m, 1H), 2.63 –2.47 (m,
H), 1.86 – 1.71 (m, 3H), 1.63 – 1.40 (m, 4H), 1.13 (dd, J= 13.1, 7.6 Hz, 3H), 0.81 (d, J = 6.6
Hz, 3H), 0.77 (d, J = 6.6 Hz, 3H). HRMS m/z for [C H N O P-], calc. 439.2003, found
2
1
32
2
6
4
39.2108.
(2S)-2-({[(1R)-1-(N-acetyl-L-glutamyl-1-amido)ethyl](hydroxy)phosphoryl}methyl)-4-
phenylbutanoic acid (4). Obtained as a foamy white solid (35 %) after semi preparative
RP-HPLC in conditions 4. R =11.3 min in conditions 1, R =14.7 min in conditions 3. MS
t
t
-
m/z for [C H N O P] , calc. 455.4, found 455.2.
2
0
28
2
8
(2S)-2-({[(1R)-1-(Nα-acetyl-L-lysylamido)ethyl](hydroxy)phosphoryl}methyl)-4-
phenylbutanoic acid (5). Obtained as a foamy white solid (28 %) after semi preparative
ACS Paragon Plus Environment
2
6

Page 27 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RP-HPLC in conditions 4. R =10.9 min in conditions 1, R =14.1 min in conditions 3. MS
t
t
-
m/z for [C H N O P] , calc. 454.5, found 454.3.
2
1
33
3
6
(2S)-2-({[(1R)-1-(N-acetyl-L-serylamido)ethyl](hydroxy)phosphoryl}methyl)-4-
phenylbutanoic acid (6). Obtained as a foamy white solid (41 %) after semi preparative
RP-HPLC in conditions 4. R =10.9 min in conditions 1, R =14.5 min in conditions 3. MS
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
t
t
-
m/z for [C H N O P] , calc. 413.3, found 413.2.
1
8
26
2
7
(2S)-2-({[(1R)-1-(N-acetyl-L-prolylamido)ethyl](hydroxy)phosphoryl}methyl)-4-
phenylbutanoic acid (7). Obtained as a foamy white solid (38 %) after semi preparative
RP-HPLC in conditions 4. R =12.3 min in conditions 1, R =16.3 min in conditions 3. MS
t
t
-
m/z for [C H N O P] , calc. 423.4, found 423.2.
2
0
28
2
6
(2S)-2-({[(1R)-1-(N-acetyl-L-tyrosylamido)ethyl](hydroxy)phosphoryl}methyl)-4-
phenylbutanoic acid (8). Obtained as a foamy white solid (42 %) after semi preparative
1
RP-HPLC in conditions 4. R =12.7 min in conditions 2, R =12.5 min in conditions 3. H
t
t
NMR (500 MHz, D O) δ8.04 (d, J= 7.8 Hz, 1H), 7.68 (d, J= 10.4 Hz, 1H), 7.28 – 7.09 (m,
2
5
H), 7.08 – 6.99 (m, 2H), 6.78 – 6.69 (m, 2H), 3.89 – 3.77 (m, 1H), 3.03 – 2.91 (m, 1H), 2.79
– 2.65 (m, 1H), 2.53 – 2.34 (m, 3H), 1.92 – 1.81 (m, 1H), 1.75 – 1.59 (m, 2H), 1.53 – 1.37 (m,
-
1H), 1.14 – 1.06 (m, 3H). HRMS m/z for [C H N O P] , calc. 489.4780, found 489. 4801.
2
4
30
2
7
(2S)-2-({[(1R)-1-(N-{2-[2-(2-aminoethoxy)ethoxy]acetyl}-L-
tyrosylamido)ethyl](hydroxy)phosphoryl}methyl)-4-phenylbutanoic
acid
(9).
Obtained as a foamy white solid (27%) after semi preparative RP-HPLC in conditions 4.
-
R =11.3 min in conditions 2, R =18.3 min in conditions 3. MS m/z for [C H N O P] , calc.
t
t
28 39
3
9
5
92.6, found 592.3.
(2S)-2-({[(1R)-1-(N-{2-[2-(2-acetamidoethoxy)ethoxy]acetyl}-L-
tyrosylamido)ethyl](hydroxy)phosphoryl}methyl)-4-phenylbutanoic
acid
(10).
Obtained as a foamy white solid (57%) after semi preparative RP-HPLC in conditions 4.
1
R =12.6 min in conditions 2, R =19.2 min in conditions 3. H NMR (500 MHz, D O) δ7.94
t
t
2
(d, J= 8.2 Hz, 1H), 7.90 (br s, 1H), 7.80 (d, J= 10.4 Hz, 1H), 7.27 – 7.09 (m, 4H),7.08 – 7.00
(m, 2H), 6.78 – 6.68 (m, 2H), 3.95 – 3.76 (m, 2H), 3.56 (s, 1H), 3.51 – 3.41 (m, 4H), 3.41 –
3
.31 (m, 2H), 3.27 – 3.19 (m, 2H), 3.12 – 3.01 (m, 1H), 2.82 – 2.68 (m, 1H), 2.53 – 2.39 (m,
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
H), 1.86 (s, 3H), 1.96 – 1.81 (m, 1H), 1.79 – 1.61 (m, 2H), 1.54 – 1.42 (m, 1H), 1.17 – 1.07
-
(m, 3H). HRMS m/z for [C H N O P] , calc. 634.6344 and found 634.6389.
3
0
41
3
10
Synthesis of fluorescent probe (11). To a solution of pseudo peptide 9 in DMF (C = 2
mM) were added successively N,N-diisopropylethylamine (10 equivalents) and a
solution of the 6-SIDCC-NHS ester in DMF (1.5 equivalents, C = 3 mM). The resulting
solution was stirred at room temperature and the progress of the reaction was
monitored by RP- HPLC (Supelco Ascentis® Express C column, 0 to 100% B in 10 min).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8
The reaction mixture was quenched with water and the crude material was purified by
RP-HPLC.
2-((1E,3Z,5Z)-3-(4-(2-((2-carboxy-4-phenylbutyl)(hydroxy)phosphoryl)-5-(4-
hydroxybenzyl)-4,7,16,29-tetraoxo-9,12,19,22,25-pentaoxa-3,6,15,28-
tetraazahentriacontan-31-yl)phenyl)-5-(1,1-dimethyl-6,8-disulfo-3-(4-sulfobutyl)-1,3-
dihydro-2H-benzo[e]indol-2-ylidene)penta-1,3-dien-1-yl)-1,1-dimethyl-6,8-disulfo-3-
(4-sulfobutyl)-1H-benzo[e]indol-3-ium (11). Obtained as a foamy blue solid (78 %) after
semi preparative RP-HPLC in conditions 5. R =9.2 min in conditions 1. ε680 nm= 233,200
t
-1
-1 1
M . cm . H NMR (500 MHz, D O/d -DMSO) δ 8.74 – 8.63 (m, 3H), 8.31 – 8.15 (m, 3H),
2
6
7.97 – 6.78 (m, 19H), 6.56 – 6.48 (m, 2H), 5.77 – 5.66 (m, 2H), 3.99 – 2.94 (m, 30H), 2,86 -
2
1
2
.43 (m, overlapping with solvent), 2.36 – 2.18 (m, 9H), 1.90 (s, 12H), 1.79 – 1.43 (m, 9H),
+
.15 – 1.08 (m, 3H). HRMS m/z for [C H N O S P] , calc. 1973.5221, found 1973.5322.
8
7
110
6
32 6
-Methylene-4-phenylbutanoic acid (12). Sodium (2.88 g, 125 mmol) was slowly added
to abs. EtOH (162 mL) over a period of 1 h. To the resulting solution, diethyl malonate
20 g, 125 mmol) was added drop wise over a period of 30 min. The solution was then
(
stirred at room temperature for 30 min and 2-phenylethyl bromide was added drop
wise during 1 h. After the addition was completed, the mixture was refluxed for 6 h, the
solvent was then removed under vacuum. H O (200 mL) and Et O (100 mL) were added
2
2
to the residue. The aqueous layer was extracted with Et O (2 × 100 mL) and the
2
combined organic layers were washed with brine (50 ml), dried over Na SO and
2
4
concentrated. The oily residue was then dissolved in abs. EtOH (50 ml) and a solution of
KOH (28 g, 500 mmol) in abs. EtOH (250 mL) was slowly added over 2 h. After the
ACS Paragon Plus Environment
2
8

Page 29 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
addition was completed, the resulting mixture was stirred at room temperature for 5 h.
After evaporation, the residue was dissolved in H O (100 ml) and Et O (100 mL). The
2
2
aqueous phase was washed with Et O (3 × 50 ml), acidified with 3M HCl and extracted
2
with Et O (3 × 50 ml). The latter extracts were washed with brine (50 mL), dried over
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Na SO and concentrated. The residue (18.8 g) was dissolved in AcOEt (360 mL) and the
2
4
o
mixture was cooled at 0 C. Et NH (10.8 mL, 104 mmol) and paraformaldehyde (3.8 g,
2
1
26 mmol) were then added and the resulting mixture was refluxed for 3 h. After the
completion of the reaction, volatiles were evaporated and the crude residue was
dissolved in Na CO 5% (100 mL) and Et O (100 mL). The aqueous phases were washed
2
3
2
with Et O (2 × 50 mL), acidified with 3M HCl and extracted with Et O (3 × 50 mL). The
2
2
latter extracts were washed with brine (50 mL), dried with Na SO and concentrated
2
4
under vacuum to afford 14.3 g (65% for 3 steps) of compound 12 as a hygroscopic
colorless solid. The obtained spectral data for acid 12 were found identical to literature
5
9
data .
Benzyl 2-methylene-4-phenylbutanoate (13). To an ice-cooled solution of acid 12 (5.0 g,
2
3
8 mmol) in dry CH Cl (140 mL), DIPEA (5.45 ml, 31.2 mmol), benzyl alcohol (3.37 g,
2
2
1.2 mmol), DMAP (346 mg, 2.84 mmol) and EDC.HCl (5.96 g, 31.2 mmol) were
successively added. The resulting solution was stirred at room temperature for 6 h.
After the completion of the reaction, the volatiles were evaporated and the crude
residue was partitioned between H O (20 mL) and Et O (50 mL). The aqueous layer was
2
2
removed and the organic layer were successively washed with 1M HCl (2 × 20 mL), H O
2
(
20 mL), 5% NaHCO (2 × 20 mL) and brine (20 mL), dried over Na SO and
3
2
4
concentrated under vacuum. The oily residue was further purified by silica gel column
chromatography using PE (40-60 ºC)/AcOEt 9:1 as eluent solvent system to afford 5.23 g
1
(70%) of compound 13 as a colorless oil. H NMR (200 MHz, CDCl ) δ 7.46-7.13 (m, 10H),
3
1
3
6.23 (s, 1H), 5.55 (t, J = 1.1 Hz), 5.23 (s, 2H), 2.87-2.76 (m, 2H), 2.71-2.60 (m, 2H); C NMR
(
50 MHz, CDCl ) δ 167.02, 141.45, 139.87, 136.12, 128.69, 128.61, 128.46, 128.32, 128.21,
3
+
+
1
26.08, 125.98, 66.58, 34.98, 34.06. MS m/z for C H O (M+H ), calcd 267.1, found 267.2.
1
8
19
2
(R)-1-[N-(9-fluorenylmethoxycarbonyl)amino]ethyl phosphinic acid (15). A solution of
ACS Paragon Plus Environment
2
9